MO RTA L AN D DI VIN E IN E ARLY GREEK

EPISTEMOLOGY

This book demonstrates that we need not choose between seeing so-called
Presocratic thinkers as rational philosophers or as religious sages. In particular,
it rethinks fundamentally the emergence of systematic epistemology and reflec-
tion on speculative inquiry in Hesiod, Xenophanes and Parmenides. Shaul Tor
argues that different forms of reasoning, and different models of divine disclo-
sure, play equally integral, harmonious and mutually illuminating roles in early
Greek epistemology. Throughout, the book relates these thinkers to their reli-
gious, literary and historical surroundings. It is thus also, and inseparably, a study
of poetic inspiration, divination, mystery initiation, metempsychosis and other
early Greek attitudes to the relations and interactions between mortal and divine.
The engagements of early philosophers with such religious attitudes present us
with complex combinations of criticisms and creative appropriations. Indeed, the
early milestones of philosophical epistemology studied here themselves reflect
an essentially theological enterprise and, as such, one aspect of Greek religion.

shaul tor is Lecturer in Ancient Philosophy in the Departments of Classics

and Philosophy, King’s College London.
 CAMBRIDGE CLASSICAL STUDIES

General Editors
R . G . O S B O R N E , W. M . B E A R D , G . B E T E G H ,
J . P. T . C L A C K S O N , R . L . H U N T E R , M . J . M I L L E T T ,
S . P. O A K L E Y, T . J . G . W H I T M A R S H
MO RTA L AN D DI VIN E IN E ARLY GREEK
EPISTEMOLOGY

A Study of Hesiod, Xenophanes and Parmenides

SH AUL TOR
King’s College London
 University Printing House, Cambridge cb2 8bs, United Kingdom
One Liberty Plaza, 20th Floor, New York, ny 10006, USA
477 Williamstown Road, Port Melbourne, vic 3207, Australia
4843/24, 2nd Floor, Ansari Road, Daryaganj, Delhi – 110002, India
79 Anson Road, #06–04/06, Singapore 079906

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781107028166
doi: 10.1017/9781139235747
© Faculty of Classics, University of Cambridge 2017
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2017
Printed in the United Kingdom by Clays, St Ives plc
A catalogue record for this publication is available from the British Library.
Library of Congress Cataloging-in-Publication Data
names: Tor, Shaul, author.
title: Mortal and divine in early Greek epistemology : a study of Hesiod,
Xenophanes, and Parmenides / Shaul Tor, King’s College London.
description: New York : Cambridge University Press, 2017. | Series: Cambridge
classical studies | Includes bibliographical references and index.
identifiers: lccn 2017039153 | isbn 9781107028166 (alk. paper)
subjects: lcsh: Pre-Socratic philosophers. | Knowledge, Theory of. | Hesiod. |
Xenophanes, approximately 570 b.c.-approximately 478 b.c. | Parmenides.
classification: lcc b200.k45 t67 2017 | ddc 182–dc23
LC record available at https://lccn.loc.gov/2017039153
isbn 978-1-107-02816-6 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
 ‫לאמי מורתי‬
To my mother and teacher
 CONT ENTS

Preface and Acknowledgements page ix

List of Abbreviations xii

Introduction 1
1 Rationality and Irrationality, Philosophy and Religion 10
1.1 Rationality and Irrationality 10
1.2 Some Promising Candidates? Milesians, Hippocratics and
Myth-Critics 19
1.3 Philosophy and Religion 36
1.4 Rationality and Irrationality: Religious Belief 48
1.5 Hesiod and Philosophy 52
2 Hesiodic Epistemology 61
2.1 Semantic Overview: pseudea, alêthea, etyma, homoia 65
2.2 The Muses’ Address and Divine Dispensation in Hesiod 72
2.3 The Theogony: Conclusions 90
2.4 Epistemological Optimism in the Works and Days and Final
Remarks 95
3 Xenophanes on Divine Disclosure and Mortal Inquiry 104
3.1 Divination 108
3.2 Against a Notion of Disclosure 116
3.3 Setting the Limits 128
3.4 Divine Disclosure and Mortal Inquiry 133
3.5 A52 Revisited: A Clean Sweep? 144
Introduction to the Chapters on Parmenides 155
4 Why Did Parmenides Write Doxa? 163
4.1 Approaches to the Aetiological Question 163
4.2 Parmenides’ Theory of Human Cognition 169
4.3 Cognition, krisis, Sense-Perception 183
4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa 196
4.5 Conclusions: Why Did Parmenides Write Doxa? 215

vii
 Contents
5 How Could Parmenides Have Written Alêtheia? 222
5.1 Recap and Introduction: Some Resolutions of the
Paradoxical Implications of Parmenides’ Theory of Human
Cognition 222
5.2 Parmenides on the Soul: Hot Metempsychosis and
the Physiology of Divinisation 227
5.3 The Proem in Its Time and Place 250
5.4 Conclusions: How Could Parmenides Have Written
Alêtheia? 277
5.5 The Ontological Question; Being, Intelligence and
Intelligibility in Alêtheia 285
6 Retrospect and Prospect 309
6.1 Interrelations 309
6.2 The daimôn and Muse of Empedocles 318
6.3 Final Remarks: Reason and Revelation, Philosophy
and Religion Again 339

Appendix 347
Bibliography 360
Index Locorum 387
General Index 399

viii
 P R E FA C E A N D A C K N O W L E D G E M E N T S

The roots of this book are in a Cambridge University doctoral

thesis, and my first and foremost thanks go to my supervisors,
Malcolm Schofield and Robert Wardy. They guided my efforts
throughout that treacherous leg of my career with humbling wis-
dom and knowledge, devoted care, critical but always open-
minded insight and unfailing encouragement and patience. In the
years since, I was extraordinarily fortunate to receive the same
kind and generous support also from David Sedley and Gábor
Betegh, who went far beyond the call of duty for doctoral exam-
iners and who likewise were always there to comment on drafts,
give a word of advice or puzzle over some ancient texts. I am
profoundly grateful to them all.
It will be impossible to detail my debt to all the others who have
read or heard my work and to the many conversations from which
it has benefited. I must mention George Boys-Stones, István
Bodnár, Patricia Curd, Harvey Lederman, James Lesher,
Geoffrey Lloyd, Catherine Rowett and James Warren, all of
whom kindly read and helpfully commented on earlier drafts of
different chapters. In particular, Robin Osborne, Hannah Willey
and (again) Gábor Betegh read through the entire manuscript and
made invaluable suggestions. None of these people should of
course be held responsible for the results. In one form or another,
I presented parts of the book at conferences and seminars in
Cambridge, London, Glasgow, Durham, St Andrews, Budapest,
Prague and Leiden. I owe much to comments from the audiences
on those occasions. I am very grateful also to Michael Sharp from
Cambridge University Press, for his seemingly inexhaustible
reserves of patience and expert advice.
I could not ask for better or more thought-provoking teachers,
colleagues and students than the ones I have had in Cambridge (as
a student in St John’s College and a Junior Research Fellow in
ix
 Preface and Acknowledgements
Jesus College) and London (as a lecturer in King’s College
London). Above all, the ancient philosophy seminars in both
institutions – and the ancient philosophy communities in both
cities more broadly – have provided a most stimulating and nour-
ishing working environment, and one which I hold dear. I owe also
a particular and longstanding debt of gratitude to David Butterfield
for his ever generous help, advice and friendship, and for his
inspiring example, throughout our years in Cambridge together.
The earliest (doctoral) phases of this study were funded by a St
John’s College Henry Arthur Thomas Scholarship, an Overseas
Research Studentship and an Overseas Trust Bursary. Later stages
of the project were made possible by Jesus College Cambridge and
King’s College London. A period of sabbatical leave granted by
the latter was instrumental. I am grateful to those bodies for
making my research possible.
I endeavoured to make this book as welcoming as possible to
non-expert and Greek-less readers. This was easier to achieve
fully in some sections than in others, where the argument is
unavoidably more technical. Even in those places, though, I have
tried to make the discussion as accessible as possible. I have
restricted to the notes issues that some readers may legitimately
wish to skip but others perhaps will not, and which would other-
wise have disrupted the flow of the text. I have transliterated into
English characters important individual Greek words or even brief
phrases, when I felt that this might help the Greek-less reader to
assess the argument or be interesting for them.
Translations from Greek and Latin sources are in general my
own or modified, but I regularly consulted and made liberal use of
existing translations. In particular, translations of Hesiod generally
follow Most (2006), translations of Xenophanes follow Lesher
(1992), translations of Parmenides follow Coxon (2009), Palmer
(2009) and Graham (2010) and translations of Empedocles follow
Inwood (2001). For these and other pre-Socratics, I also consulted
and drew on the translations in KRS (1983) and Curd and
McKirahan (2011).
An earlier and shorter version of Chapter 3 was previously
published as Tor (2013a). The most central ideas in Chapters 4

x
 Preface and Acknowledgements
and 5 were published in much abridged form in Tor (2015).
Everything has been expanded, elaborated, rewritten and revised.
During the long process of writing this book, my family gave
me the same unconditional and indispensable love and encourage-
ment that they always have. My mother, Michal Arbell-Tor, has
been, throughout those years and for as long as I can remember, an
unfailingly interested, exercising and exciting person with whom
to think and argue about the intellectual questions that preoccupied
me. There is no question that it is because of her (and not just for
the obvious reasons) that I have ended up doing what I do and in
the way that I do it. Finally and most importantly, I thank again
Hannah, this time not as the inimitable reader and interlocutor that
she is, but as the wonderful partner that she is. Without her love
and support I could not imagine this book or my life, nor would
I care to.

xi
 ABBR EVIATIONS

Further to the conventions and works listed below, I follow the

abbreviations in the Oxford Classical Dictionary (4th edn).
Alexander Alexander, P. J. (1967) The Oracle of Baalbek:
the Tiburtine Sibyl in Greek Dress.
Washington, DC.
Beckby Beckby, H. (1957–1958) Anthologia Graeca,
2nd edn, 4 vols. Munich.
Bernabé Bernabé, A. (2004–2007) Poetae Epici Graeci:
Testimonia et Fragmenta. Pars II. Orphicorum
et Orphicis Similium Testimonia et Fragmenta.
Munich.
Campbell Campbell, D. A. (1982) Greek Lyric I: Sappho
and Alcaeus. Cambridge, MA.
Chilton Chilton, C. W. (1967) Diogensis
Oenoandensis Fragmenta. Leipzig.
DG Diels, H. (1879) Doxographi Graeci. Berlin.
Drachmann Drachmann, A. B. (1903–1927) Scholia Vetera
in Pindari Carmina, 3 vols. Leipzig.
Dübner Dübner, F. (1969) Scholia Graeca in
Aristophanem. Hildesheim.
Düring Düring, I. (1961) Aristotle’s Protrepticus.
Stockholm.
Gaisford Gaisford, T. (1962) Etymologicum Magnum.
Amsterdam.
GJ Graf, F. and Johnston, S. I. (2013) Ritual Texts
for the Afterlife: Orpheus and the Bacchic
Gold Tablets, 2nd edn, London.
Gregorio Di Gregorio, L. (1975) Scholia Vetera in
Hesiodi Theogoniam. Milan.
Hilgard Hilgard, A. (1965) Grammatici Graeci, vol.
IV.1–2. Hildesheim.
xii
 List of Abbreviations
Irigoin Irigoin, J. (1993) Bacchylide. Dithyrambes,
Épinicies, Fragments. Paris.
Isnardi Parente Isnardi Parente, M. (1982) Senocrate –
Ermodoro: frammenti. Napoli.
Kindstrand Kindstrand, J. F. (1990) [Plutarchi] De
Homero. Leipzig.
Latte Latte, K. (1953) Hesychii Alexandrini
Lexicon, vol. I. Copenhagen.
Lentz Lentz, A. (1965) Grammatici Graeci, vol. III.
1–2. Hildesheim.
M Maehler, H. (2001) Pindari Carmina cum
Fragmentis. Pars II. Fragmenta. Leipzig.
Müller Müller, K. (1965) Geographi Graeci Minores,
vol. II. Hildesheim.
Papageorgius Papageorgius, P. N. (1888) Scholia in
Sophoclis Tragoedias Vetera. Leipzig.
Parke Parke, H. W. (1967) The Oracles of Zeus:
Dodona, Olympia, Ammon. Oxford.
Pertusi Pertusi, A. (1955) Scholia Vetera in Hesiodi
Opera et Dies. Milan.
Pfeiffer Pfeiffer, R. (1949) Callimachus, vol.
I. Oxford.
PW Parke, H. W. and Wormell, D. E. W. (1956)
The Delphic Oracle, vol. II: The Oracular
Responses. Oxford.
Rabe Rabe, H. (1906) Scholia in Lucianum. Leipzig.
Rose Rose, V. (1967) Aristotelis Qui Ferebantur
Librorum Fragmenta. Stuttgart.
Sandbach Sandbach, F. H. (1967) Plutarchi Moralia,
vol. VII. Leipzig.
Scheer Scheer, E. (1958) Lycophronis Alexandra, vol.
II. Berlin.
Smith Smith, M. F. (1993) Diogenes of Oenoanda: the
Epicurean Inscription. Naples.
Usener Usener, H. (1963) Epicurea. Rome.
Wendel Wendel, K. (1974) Scholia in Apollonium
Rhodium Vetera. Berlin.

xiii
 I N T R O D U C T I ON

This book is first and foremost a study of the verses of Hesiod,

Xenophanes and Parmenides. It examines these thinkers as key
figures in the emergence of systematic epistemology and systema-
tic reflection on the nature of speculative inquiry. I submit that, in
each of the three thinkers, novel forms of critical and reflective
thought coexist with novel positions concerning the relation and
interactions between gods and mortals. Indeed, in the case of each
author, critical thinking on the one hand, and reflections about
the interactions between mortal and divine on the other hand,
play complex, harmonious and equally integral roles, which can
be understood fully only in relation to one another. The thread
running throughout the book is the thesis that, for Hesiod,
Xenophanes and Parmenides alike, theology and ‘anthropology’
are logically prior to epistemology.1 More specifically, their diver-
gent views on the cognitive capacities and limitations of mortals
are, and can only be properly understood as, a corollary of their
correspondingly divergent views on (i) the nature of the divine,
(ii) the nature of the mortal and (iii) the nature of the relation and
interactions between them. The book aims not merely to argue for
this thesis, but also – and in particular – to demonstrate and
explore its usefulness as a fresh perspective on a range of often
long-standing interpretative problems.
The book falls into six chapters. In Chapter 1, I situate the
inquiries pursued in the subsequent chapters in their proper rela-
tion to broader fundamental questions concerning rationality and
irrationality, and philosophy and religion. This opening chapter
aims to bring to the fore the bigger issues at stake in the subsequent
1
I use the term ‘anthropology’ here in an etymological sense, referring to conceptions of
the human.

1
 Introduction
investigations into interrelations between theology and epistemol-
ogy. In doing so, it clarifies the insight which the results of those
investigations afford for our understanding of early Greek philo-
sophy and religion more generally. Historically, we have asso-
ciated under the single term ‘rational’ very distinct intellectual
phenomena. On the one hand, we describe as ‘rational’ coherent,
critical, inferential, questioning and explicative thinking. On the
other hand, we often identify as ‘rational’ human inquiries that
proceed without any appeal to divine interference or aid, as well
as, more generally, secularising moves away from god-centred
patterns of thought and explanation. In Chapter 1, I consider the
problematic influence that these deep-seated associations have
exerted and continue to exert on the business of interpreting
early Greek philosophy. I submit that the difficult challenge of
extricating ourselves from their long shadow lies largely ahead of
us. Similarly, we will examine the entrenched expectation that
philosophy should operate more or less independently from tradi-
tional religion and pull in opposite directions from it, as well as the
nexus of assumptions that underpins this expectation. We will ask
in what ways, within the context of Greek polytheism, some
philosophers can indeed be seen to come into conflict with some
traditional religious attitudes and practices and to what extent such
critiques were or were not perceived as a religious problem or
a social threat. To be sure, philosophical critiques are important
and should not be marginalised. But the engagements of philoso-
phers with traditional or non-philosophical religious attitudes are
hardly limited to criticisms. Furthermore, philosophical criticisms
can sometimes be inextricably combined with positive and crea-
tive appropriations, even of those very same aspects of traditional
religion that are being criticised. Ultimately, philosophical theol-
ogies constitute one aspect of the flexible and inclusive mass of
beliefs, representations and practices that was Greek religion.
The studies of Xenophanes and Parmenides in this book offer
two extended illustrations of these principles.
In Chapter 2, we turn to Hesiod and, in particular, to the striking
and enigmatic way in which his Muses articulate their relation to
the poet in lines 27–8 of the Theogony: ‘We know how to speak
many falsehoods which are like verities, and we know, whenever
2
 Introduction
we wish, how to utter truths.’ We will consider this address both in
its immediate context in the Theogony and against the broader
background of Hesiod’s reflections on the mortal and the divine,
and the male and the female. I will argue that, in the Theogony,
Hesiod decisively and consistently encourages a cautious and
destabilising stance in response to the Muses’ address: the
Muses leave it uncertain – and no mortal poet could himself
ascertain – whether the verses which they inspire comprise truths,
falsehoods or some combination of the two. Hesiod’s understand-
ing of his relation to the Muses, moreover, forms one poetic-
epistemological aspect of a coherent and holistic conception of
the human condition as a whole. At one juncture elsewhere in the
Hesiodic corpus, however, we encounter a competing and more
optimistic reinterpretation and revaluation of the Muses’ address
(Works and Days 646–62). We will tentatively consider certain
theological developments in Hesiod’s thought, which could under-
lie and explain this divergence between Hesiod’s epistemological
stance in the Theogony and at this moment in the Works and Days.
More importantly, we will see that a synoptic consideration of the
poet’s voice, as it emerges from the Hesiodic corpus as a whole,
produces a picture of epistemological and theological ambiva-
lence. Ultimately, the primary thrust of the Muses’ address to the
poet is to raise, but leave unresolved, the question of the proper
way to interpret it. Put differently, Hesiod’s Muses crystallise, not
an epistemological position, but an epistemological framework.
Within this framework, the problem of epistemology becomes –
for Hesiod as for the philosophers who followed his lead – the
problem of understanding the nature of the interactions between
mortal and divine.2
Chapter 3 addresses Xenophanes’ reflections on the nature of
divine disclosure. By contrast with the common view, Xenophanes
does not deny categorically the reality of divine disclosure.
Nor, however, does he acquiesce in traditional assumptions of
disclosure. Rather, Xenophanes specifically rejects traditional
2
The reasons for according such focused treatment and pride of place to Hesiod are
elaborated more fully in Ch. 1.5, which discusses further Hesiod’s intrinsic interest and
his formative relation to later philosophical thought (or, if one is so inclined: to later,
philosophical thought).

3
 Introduction
conceptions of divine disclosure as theologically faulty. He sup-
plants those traditional conceptions with his own, alternative
understanding of what divine disclosure amounts to and how it
works. Xenophanes’ novel conception of divine disclosure
grounds his novel views concerning the possibilities and limita-
tions of mortal beliefs and speculative inquiry. It forms, moreover,
one coherent aspect of his overall re-conceptualisation of divinity
and of his social and moral world view. Xenophanes, then, does
not simply reject traditional ideas about divine disclosure without
a trace. Rather, he transforms those traditional ideas in radical
ways. Xenophanes remained profoundly influenced by what he
rejected.
Chapters 4 and 5 are devoted to the difficult and complex case of
Parmenides. In Chapter 4, we will see that Parmenides advances
a physiological theory of human cognition. According to this
theory, humans qua humans must, as a matter of physiological
necessity, experience and form beliefs about multiple, heteroge-
neous, mobile and differentiated things and processes. They
cannot but experience and think in terms of such sensory contrasts
as light and dark, hot and cold, rare and dense, etc. Famously,
however, Parmenides thought that Being or ‘what-is’ had very
different features. What-is is ungenerated, imperishable, indivisi-
ble, homogeneous and immobile. Why, then, did Parmenides think
that mortals must continue to reflect about and strive to understand
the natures of generated, multiple, heterogeneous and mobile
things and processes, even after they came to realise that the
ultimate reality – what-is – involves no generation, multiplicity,
heterogeneity or motion? I will argue that Parmenides’ theory of
human cognition best positions us to answer that much-debated
question. To think of and in terms of generated, heterogeneous and
mobile things and processes is a necessary and even appropriate
aspect of what it is to think and live as a mortal. If, however, we
explain in this way Parmenides’ abiding interest in cosmological
accounts of change and differentiation, then a new problem arises.
If humans are hardwired to think in terms of sensory contrasts and
about differentiated and heterogeneous objects, then how was
Parmenides – a human – also able to sustain the qualitatively
different kind of thought that is necessary for conceiving of the
4
 Introduction
undifferentiated and homogeneous what-is? If humans must, by
physiological necessity, think in terms of multiplicity and hetero-
geneous differentiation, then how was Parmenides also able to
think otherwise? In Chapter 5, I argue that the human agent for
Parmenides is not simply and strictly human. The mortal also
possesses a divine part or aspect: his fiery and aethereal soul.
The mortal is capable of sustaining a higher-than-mortal type of
thinking by momentarily coming to think with – or as – his divine
soul. This is, moreover, the fundamental reason for which
Parmenides begins his poem by describing his journey to
a goddess, who proceeds to disclose the truth of things to him.
The goddess, through her disclosure and guidance, enables the
mortal to come to think with or as his divine soul and to sustain the
higher-than-human thought which is required for the cognition of
what-is. It is only through the goddess’s initiation, therefore, that
Parmenides was able to master the system of argumentation that is
developed in the poem, and so to comprehend, evaluate and accept
for himself the truth of the doctrines which the goddess revealed.
In this way, we can do justice to the emphatic prominence in
Parmenides’ poem of both divine disclosure and argumentative
reasoning. Furthermore, as we develop this interpretation, we will
see Parmenides drawing in positive and appropriative ways on
a variety of contemporary and traditional religious models, includ-
ing poetic inspiration, divinatory oracles, mystery initiations and
metempsychosis.
These discussions of Hesiod, Xenophanes and Parmenides are
offered as essentially self-standing studies that, in conjunction,
disclose instructively divergent yet related approaches to episte-
mology. In Chapter 6.1, we will consider more directly the critical
and formative engagements by Xenophanes with Hesiod and by
Parmenides with both Hesiod and Xenophanes. Our discussion of
the interrelations between the three thinkers will shed further light
on, and will itself in turn be illuminated by, the individual studies
of them in the previous chapters.
As I indicated above, we begin in Chapter 1 by considering
critically certain historical and still-influential notions of ration-
ality and irrationality. Positive accounts of rationality, which seek
to identify some of what rationality includes and excludes, will not
5
 Introduction
be a starting point for this book but – within the confines of its
particular scope – will be one of its outcomes. In Chapter 6.3,
I recapitulate certain, more or less implicit ideas of rationality
which I find to be operative in Hesiod, Xenophanes, Parmenides
and Empedocles, as well as in some other models of the interac-
tions between gods and mortals (such as divination). Within the
intervening chapters themselves, we will not be helping ourselves
to the terms ‘rational’ and ‘irrational’ as interpretative tools.
This book does not pretend to offer a wholesale reconsideration
of early Greek philosophy nor, indeed, to exhaust the immensely
rich and challenging question of the connections between theology
and epistemology in early Greek philosophy. I aim to offer here
a new analysis of these connections and of their significance in
some key episodes in the emergence in archaic Greece of systema-
tic reflection on the nature of speculative inquiry. I by no means
wish to suggest that the story ends there. On the contrary, it is my
hope that the interpretative approaches developed and pursued
here could serve as useful starting points for considerations of
other and later developments in philosophy, theology and episte-
mology. In Chapter 6.2, we will take one such forward look by
considering (in a focused and circumscribed manner) one espe-
cially important and illuminating later case: the epistemological
significance of the daimôn and Muse in the thought of
Empedocles. We will find that Empedocles too couches his epis-
temological reflections within a broader theological framework.
Furthermore, Empedocles too posits his own version of what I will
refer to in this book as ‘epistemically significant interactions’.
By this term, I mean interactions between mortal and divine agents
that enable the mortal to attain knowledge, or to come by poten-
tially true beliefs and views, which he could not have attained or
come by independently of those interactions.3 I will use the terms

3
Heraclitus would offer another productive test-case for the interrelations between theol-
ogy and epistemology (cf. Ch. 1.3, n.98). Heraclitus, however, does not predicate his
inquiry on a notion of epistemically significant interactions readily comparable to
Hesiod’s Muses, Xenophanes’ model of divine disclosure, Parmenides’ goddess or
Empedocles’ Muse. Inclusion of him in this study would have required not only
a significantly longer book, but also one with a somewhat looser thematic focus.
As I do not claim to have exhausted the story, so too I do not pretend to have uncovered

6
 Introduction
‘divine revelation’ and ‘divine disclosure’ to refer to the same type
of interactions.
As we shall presently see, there exists an artificial schism in the
scholarship between conceptions of the early Greek philosophers
as systematic, rational thinkers and as poets, mystics and religious
figures. This schism also helpfully brings out a methodological
divergence. Although we must eschew oversimplifying general-
isations here, we can fairly say that, by and large, scholarship in
the analytical tradition tends to reconstruct philosophical positions
and arguments more through an internal examination pursued
independently of advancing claims about their cultural, historical
and literary context.4 Historical reconstructions of dialectical con-
text, moreover, tend to privilege a philosopher’s formative reac-
tions to the theories of those conventionally classified as his earlier
philosophical colleagues. By contrast, what has come to be called
the ‘anthropological’ approach seeks to re-contextualise texts that
a long philosophical tradition de-contextualised.5 At its most
radical, however, this approach dismisses the study of theoretical
and philosophical reflection in the textual output of those we call
early philosophers as a failure to recognise that this output was
fundamentally shaped by the agonistic cultural and pragmatic
circumstances in which it was produced.6
The following investigations into Hesiod, Xenophanes and
Parmenides draw essentially and throughout on a consideration
of their complex engagements – competitive, polemical, appro-
priative, critical and creative – with a range of culturally prevalent
paradigms of theology and epistemology. Philosophical texts are
thus examined in the light of, but are not thereby reduced or
assimilated to, their religious, literary and historical contexts.
What follows is by focus and structure a study of Hesiodic,
Xenophanean and Parmenidean epistemology. But it is also, and

any sort of master key for (early) Greek epistemology. For example, I cannot see that
a comparable inquiry into the connections between epistemology and notions of the
interactions between mortal and divine would be an especially productive way to
approach Democritus’ reflections about cognition, perception and speculative inquiry.
On Democritus’ epistemology, see Lee (2005) 181–250.
4
Barnes (1982) xii makes the point explicit (‘[p]hilosophy lives a supracelestial life etc’);
see also Curd (2002) 133–4; Wedin (2014) 5.
5 6
As Laks (2003) 20 puts it. Explicitly in Gemelli Marciano (2002) 92, 96.

7
 Introduction
inseparably, a study of poetic inspiration, divination, mystery
initiation, metempsychosis and, to put it most generally, a range
of early Greek attitudes to the relation and interactions between
mortal and divine. Homeric material, in particular, figures promi-
nently throughout. This means that we will be encountering in this
book what we might call different sorts or modes of ‘theology’.
When discussing Hesiod, Xenophanes, Parmenides and
Empedocles, we will generally be dealing with more or less self-
conscious, systematic and elaborated reflections about the divine.
But we will find that expressions and representations of divination
and mystery initiations, for example, can also convey certain
conceptions of divinity, albeit in a more flexible and implicit – if
not sometimes underdetermined and vague – manner.7
I do not wish to stake a universal methodological or theoretical
claim. Different interpretative projects require and will reward
different interpretative approaches. My contention is that, specifi-
cally with regard to the business of analysing the emergence of
philosophical epistemology in archaic Greece, methodological
purism of either stripe has led, and will inevitably lead, to reduc-
tive and distortive portrayals. Here, the analysis of systematic,
critical reflection and the contextualisation of philosophical texts
in their religious, literary and historical surroundings must,
I believe, be pursued in relation to each other and illuminate one
another. Logical and philological analysis, cultural and religious
history and literary criticism are all indispensable tools. Walter
Burkert’s diagnosis of the state of Pythagorean scholarship in 1962
7
On ‘theology’ as an interpretative category and the need to modify its application in
different contexts and in relation to different materials, see further Eidinow, Kindt,
Osborne and Tor (2016). A parallel point can be made about our use of the category of
‘anthropology’ in reference to ancient conceptions of the human. I will not attempt in this
book to circumscribe a sharply defined category of ‘myth’ or ‘philosophical myth’,
which I will then apply to, or tease out of, all the texts under discussion. This book
defends particular contentions concerning the roles of, for example, Hesiod’s Muses and
Parmenides’ goddess. If these contentions contravene any global conception of (philo-
sophy’s engagement with) myth, then, from the perspective of the present inquiry, so
much the worse for that global conception. According to Morgan (2000) 1–37, for
example, philosophers, even as they reintegrate myth, universally retain a polemical
attitude towards it (e.g. 16–17, 34–5, 290–1). My own account of Parmenides DK28 B1
registers no tension between ‘mythical’ and ‘philosophical’ elements in his thought;
contrast Morgan (2000) 5, 11, 67–87. As Parker (2011) 23 cautions: ‘“Greek myths” are
not a unified category about which we have any reason to expect that general statements
can be made.’

8
 Introduction
seems urgently relevant for current attitudes to early epistemol-
ogy: ‘The very thing that might seem rash, in view of the funda-
mental differences of interpretation, is what the nature of the
situation demands: as many-sided a treatment of the problem as is
possible.’8

8
Burkert (1972; first published in German 1962) 12. ‘Many-sided’ syntheses of analytical,
historical, religious and literary perspectives on early Greek philosophy are pursued by,
for example, Cornford (1952); Lloyd (1966), (1979) and (1987); Mourelatos (2008a; 1st
ed. 1970); Betegh (2004) and Bryan (2012).

9
 1

R AT I O N A L I T Y A N D I R R AT I O N A L I T Y,
PHILOSOPHY AND RELIGION

We must not be misled by the use of the word theos in the remains
that have come down to us.
John Burnet

1.1 Rationality and Irrationality

We may usefully begin with Parmenides’ goddess. Parmenides
was active in the city of Elea in South Italy in the early fifth century
BCE and composed a poem in Hexametric verse. In the opening
lines of his poem, Parmenides describes, in stunning and complex
detail, the chariot ride of a youth – a kouros – to an encounter with
a goddess (DK28 B1).1 Once the youth reaches the goddess, she
welcomes him (DK28 B1.24) and, from that point onwards,
remains the sole speaker. The rest of the poem was, in its entirety,
an address by the goddess to the kouros. The goddess proceeds to
issue two rather distinct accounts, one of ‘what-is’ or ‘the unsha-
ken heart of well-rounded reality’ (DK28 B1.29), the other
a comprehensive cosmology. Famously, the goddess’ account of
what-is is developed through a sustained and sophisticated system
of deductive argumentation, the first of its kind in extant Western
thought. Now, just how we should think about the relation between
these different components or aspects of Parmenides’ poem – the
kouros’ journey to the goddess and her two subsequent accounts –
will be a central preoccupation in the second half of this book.
To begin, however, we may step back and ask a broader question:
how do we respond to a philosophical poem, which puts its claims
in the mouth of a goddess, but which also gives pride of place to
a pioneering argumentative method of substantiating those same
claims? How do we respond to what strikes us as a fundamental

1
The text is cited and translated in Ch. 5.3.

10
 1.1 Rationality and Irrationality
tension between divine revelation and rational argumentation in
a text in which both things are emphatically central and which,
nonetheless, does not itself seem to register or gesture towards any
tension between them?
We can broadly demarcate two central ways in which modern
scholars have historically responded to Parmenides’ goddess, two
contrasting types of response which, to a significant extent, have
framed the modern study of Parmenides.
Most scholars, heeding Burnet’s call ‘not to be misled’ by the use
of theological language in our evidence, have viewed the goddess
as inessential to Parmenides’ philosophy, most often reducing her
to (dispensable) poetic imagery on the grounds that the philosopher
recommends his truth to us, not because it is divinely sanctioned,
but because it is reasoned and logically sound.2 Those who con-
sider the goddess indispensable, or at least allow that she might
play a more substantial and non-metaphorical role of some sort,
generally either implausibly downplay the prominence of
Parmenides’ own reasoning and argumentation3 or do not proceed
to develop an account of how precisely reasoning and divine
disclosure might cohere and cooperate, often still considering
them dissonant and opposing tendencies.4
2
For example, Reinhardt (1974) 302; Vlastos (1970a) 49, cf. (1946) 76; Dolin (1962)
97–8; Tarán (1965) 28; Burkert (1969) 29; (2000) 145 (on ‘rational’ and ‘religious’
tendencies, cf. also Burkert (1972) 280 and (1985) 292); Coxon (2009) 17–18, 286;
Barnes (1982) 155ff; Mackenzie (1982) 2; Cosgrove (1974) 87–8; Blank (1982) 177;
Curd (1998) 104, cf. (2002) 124, 133–5; Sedley (1999a) 114; Mogyoródi (2006) 150–1,
157; Lesher (2008) 472–6; Kahn (2009) 215; Granger (2008) 17; (2010) esp. 30–1 (with
n.37), 33–5 (at 35–6, Granger allows the possibility of compatabilism between what he
calls ‘mystical insight’, which generates certain beliefs, and reason, which justifies
them); Bollack (2011) 20 (‘meditation as opposed to revelation’); Clay (2015) 124–6.
3
Jaeger (1947) 96–8 (‘a mere instrument’); Kingsley (1999); Gemelli Marciano (2008);
(2013) 45–105; Shaw (2004); cf. Heitsch (1966) 201.
4
E.g. Diels (2003) 9; Verdenius (1949) 120–1, 128; Cornford (1952) 118–20; Guthrie
(1965) 10–12; Miller (1968); Roloff (1970) 199; Pfeiffer (1975) 190; Wiesner (1997) 21;
Hussey (1990) 29; Lesher (1994a) 16–19; Most (1999a) 353–5; Göbel (2002), esp. 159,
167. There are, of course, stimulating exceptions. Bodnár (1989) 65–6 suggests that the
goddess discloses to Parmenides propositions that he could not otherwise discover but
which he accepts only upon judging them to be true. Bodnár’s view is attractive but
insufficient. He does not elucidate why the mortal cannot ‘reach by himself’ propositions
whose truth-value he can himself determine (he does not address DK28 B16 or the
relation between Alêtheia and Doxa). Mansfeld (1964) 61–2, 260–1 argues that the
goddess guarantees the truth of the otherwise indemonstrable premises of Parmenides’
syllogisms, which are therefore accepted on trust (cf. Burkert (1969) 27, n.60). I argue in
Ch. 5.1 and Ch. 5.4 that this interpretation too, whether tenable or not, fails to appreciate

11
 Rationality and Irrationality, Philosophy and Religion
Here, as often elsewhere, the more extreme positions can offer
especially clear and self-conscious articulations of tensions and
presuppositions which, in more muted forms, pervade the scholarly
discussion quite broadly, and which it remains difficult for us truly
to move beyond. The opposing treatments of Parmenides in James
Lesher (2008) and Peter Kingsley (1999), in particular, throw these
tensions and presuppositions into sharper relief. For Lesher (2008),
the view that the goddess is essential to Parmenides’ attainment of
knowledge would mark the poem as ‘a set of truths revealed to him
by a divine power’, where these truths are accepted unreflectively
and because they are divinely revealed.5 But, since Parmenides
employs ‘carefully reasoned arguments’ involving an ‘active effort’
on his part as on ours, Lesher advocates ‘a more “rationalist” or
“humanistic” reading’: divine disclosure is mere imagery, ‘trap-
pings of the older outlook’, while Parmenides in fact prescribes,
by contrast, ‘inquiry and reflection’.6 Kingsley (1999) decries such
reductive readings and avers that, since Parmenides’ mystic wisdom
is manifestly attained by disclosure, it could in no way stem from
‘thinking or reasoning,’7 a position that will not withstand scrutiny
of the system of argumentation developed in the poem (especially
DK28 B8).8
What is most instructive here are the assumptions on which
Lesher and Kingsley agree. For both scholars, reasoning and dis-
closure are incompatible categories. The former considers divine
disclosure as the poetic trappings subsequently attached to truths
that were attained independently and through reasoning. The latter
construes disclosure as genuinely prior and essential to such attain-
ment and dismisses talk of reasoning in Parmenides as an anachro-
nistic imposition on our part. This scholarly dilemma between
reasoning and disclosure thus has close affinity with the debate

the complexity of the problem resolved by Parmenides’ interaction with the goddess and
the fundamental significance of this interaction.
5
Lesher (2008) 473.
6
Lesher (2008) 473–6; cf. similarly Vlastos (1947) 49. Lesher (1994a) 18–19, conversely,
suggests that Parmenides credited his achievements to ‘a combination of mortal and
divine efforts’. He does not elaborate how the two might relate to each other.
7
Kingsley (1999) 144, cf. 121.
8
As Granger (2010) insists against Gemelli Marciano (2008) and Mourelatos (2013)
insists against Gemelli Marciano (2013).

12
 1.1 Rationality and Irrationality
over the status of Parmenides’ journey: is it a genuine religious
experience or mere (perhaps allegorical) imagery?9 André Laks
(2003, 2013) – bringing to the surface attitudes which often operate
implicitly – gives voice to a similar connection between the ques-
tion of rationality and the status of religious imagery and discusses it
in terms of a general theoretical model. He distinguishes (i) ‘phe-
nomena’ and (ii) ‘references’ as (i) the manifestations of certain
traditional religious phenomena (claims, patterns of thought,
experiences) and (ii) subsequent distanced and figurative or allego-
rical references to those religious phenomena. He asks, further,
whether we can (at least sometimes) legitimately speak of historical
processes of ‘rationalisation’, which lead from the former to the
latter, and defends the view that we can indeed see Parmenides (and,
to some degree, Empedocles) as exemplifying such a process of
rationalisation. Laks’ contrast between ‘phenomena’ and ‘refer-
ences’, and his notion of ‘rationalisation’ help to accentuate the
terms in which the dilemma has long been felt and approached: are
we to reduce Parmenides’ encounter with the goddess to a figurative
reference to related religious phenomena on the basis of his ration-
ality? Or should we rather reject such reductions as anachronistic, at
the cost of downplaying his rationality?
The common ground in the dispute surrounding Parmenides’
goddess reflects prevalent and quite general attitudes towards
‘rationality’, ‘irrationality’, ‘philosophy’ and ‘religion’ in
approaches to early Greek thought and culture. There is no
question of pursuing here a thorough investigation into these
categories, their histories and the sheer diversity of their uses.
Such an investigation would be impracticable within the confines
of a single volume (let alone this chapter) and insufficiently
focused for the particular preoccupations of the study at hand.
I wish, rather, to probe a little further and to identify some basic
aspects of common usage, which underpin certain tensions and
expectations that have exerted a formative influence on the

9
Genuine experience: e.g. Jaeger (1947) 94–6; Cornford (1952) 11; Mansfeld (1964)
222–73, esp. 223, 251, 259; Göbel (2002) 159; Kingsley (1999) 50; Gemelli-Marciano
(2008). (Allegorical) imagery: in addition to the citations given above (n.2), note Fränkel
(1975a) 1ff; Bowra (1953c) 39–40; Tarán (1965) 31, 230; Kirk (1970) 250; Granger
(2008).

13
 Rationality and Irrationality, Philosophy and Religion
development of modern scholarly attitudes, and whose elucidation
is indispensable for framing this study.
First, ideas of ‘rationalisation’ typically stem from and reflect
a deep-seated association of the rational with secularising tenden-
cies and the irrational with the religious and god-centred. E.R.
Dodds’ monumental The Greeks and the Irrational (1951) epito-
mises this same association. Precisely because Dodds never
pauses to circumscribe the last word in his title, his study com-
prises an instructive (and in many ways still up to date) index of
phenomena implicitly classified under that rubric. Almost every
page of Dodds’ work proves to be dominated by diverse manifes-
tations of thought and behaviour preoccupied with the divine and
especially with various postulations of divine interferences to
explain the functioning of the world and human experiences
within it (including, for example, direct encounters between
gods and mortals, divination, inspiration, god-sent dreams).
Correspondingly, it is familiar practice to identify the ‘rationality’
of early Greek philosophy with the extent (however one construes
this extent) to which it emerges by moving beyond such traditional
religious, god-centred and mythical modes of explanation.10
Often, early philosophical accounts are ‘rational’ insofar as they
are ‘natural’, i.e. insofar as they progress beyond postulations of
divine agency or an ordering intelligence11 or advance a purified
or deflated conception of the divine as an impersonal, non-
anthropomorphic part or aspect of nature.12 Again, we regularly
describe as ‘rationalising’, ‘naturalising’ and ‘demythologising’
either the removal of the divine from explanatory models and
narratives13 or transitions towards a figurative construal of it.14

10
For some illustrative examples, see e.g. Guthrie (1962) 29; (1965) 389; KRS (1983)
72–5; Laks (2002) 34–5; Humphreys (2004); Granger (2007) 419–22; Kahn (2013) 1–3.
11
E.g. Naddaf (2006) 164–5, with n.15; cf. (2005) 163.
12
E.g. Curd (2002) 121; Taub (2003) 74. Cf. more generally Humphreys (2004) 51
(‘secularization and rationalization’); also Martin (2009) 292 on the Athenian assem-
blies: ‘When making political decisions, the Athenians thus tended to act in a way we
may call rational, not only regarding the logical coherence of their arguments but also in
the limitation of these arguments to the human sphere.’
13
A prominent case is the ‘rationalisation’ of a myth when this refers to the removal of
divine agency from the story, as for example in Griffiths (1999) 181; Stern (1996) 12;
Hawes (2014) 12.
14
E.g. Cornford (1950) 42; Kirk (1970) 240–1; Bremmer (1999) 75–6; Laks (2003).

14
 1.1 Rationality and Irrationality
Second, rationality is associated with the activity of reasoning.
Here Parmenides’ arguments mark only the limiting case for early
Greek rationality. The markers of rationality extend to preoccupa-
tions with inference and evidence, questioning, explanation and
systematisation broadly construed, a concern with and the capacity
for logical coherence and consistency and the fundamental pro-
pensity to reflect and speculate through such means about the
world and the foundations of one’s understanding of it.15 In this
other usage, then, the term ‘rationality’ is used to pick out aspects
of a certain type of thinking: coherent, critical, inferential, ques-
tioning and explicative. There is one especially important sense in
which this aspect of rationality is conventionally taken – as we saw
in the case of Parmenides – to be in tension with god-centred
thought. Rational inquiry connotes an active human inquiry
which progresses without recourse to notions of interaction with
the divine, whereas, correspondingly, divine interferences in
inquiry are often taken to comprise a transmission of truths to
unreasoning, unreflective and altogether passive recipients.16
It bears emphasising, moreover, that it was for a long time
a familiar practice in scholarly literature to employ these loaded
categories – ‘rational’ and ‘irrational’ – casually and as more or
less clear terms which require no glossing. Scholars would identify
as irrational such belief-systems or practices as, say, divination17
or astrology18 without pausing to specify in each case what it is
about them which merits the label and is picked out by it.
The significant point here is that acquiescence in cavalier applica-
tions of these labels itself allowed ‘rational’ and ‘irrational’ to
sustain broad and flexible as well as unexamined (not to mention
vague) clusters of associations.

15
See e.g. Lloyd (1979) esp. 62–143 for a study of early Greek philosophical and scientific
reasoning. These aspects of rationality are the focus of Barnes (1982) who, indeed,
cautions against uncritically linking them with complete repudiations of all things
divine (4–5; nonetheless, Barnes retains the expectation that this kind of rationality
and god-centred thought should pull in opposite directions, e.g. 98: ‘uneasy bedfel-
lows’). Van der Eijk (2005) 9, n.17 calls for a similar disambiguation in the study of
ancient medicine.
16
A telling illustration: KRS (1983) 73 describe Odysseus’ rational-analytical capacities
as the recognisable toolkit of philosophy and, in the same breath, downplay his relation
to divinities as almost incidental to his conduct: ‘he is a rational man’.
17
Burkert (1992) 195, n.2. 18 Dodds (1951) 245–6.

15
 Rationality and Irrationality, Philosophy and Religion
During the nineteenth and twentieth centuries, these clusters of
associations fuelled a developmental narrative, nourished by
attitudes characteristic of the European Enlightenment, which
depicted the emergence of Greek thought from god-centred, tradi-
tional religious and mythical propensities to the secular and
critical illumination of reason. John Burnet’s Early Greek
Philosophy (first published in 1892) gives strong expression to
this picture of the first Greek philosophers, but it was Wilhelm
Nestle’s Vom Mythos zum Logos (1942) which most influentially
popularised it as a model for the trajectory of the Greek mentality
as such.19 Some of the subsequent challenges to Burnet’s and
Nestle’s models are especially seminal for this study. Dodds
(1951) has rendered it impossible to ignore the prevalence in
every attested period of the type of (god-centred) phenomena
which he investigates as ‘the irrational’.20 Werner Jaeger (1947),
already prescribing a middle course between one-sided percep-
tions of early philosophers as either Enlightened secularists or
mystical, developed a powerful account of the prominence of
theological problems and theories in pre-Socratic speculations.21
Francis Cornford (1952) argued for the ongoing, formative and
often appropriative relation of early philosophy to older and con-
temporaneous religious thought and culture. He identified the
early philosopher as one of the differentiated types emerging
from an older prophet-poet-sage complex and still much indebted
to it,22 while he also dated certain tensions between this differ-
entiated type and the other ones to as early as the first Milesian
philosophers.23 Geoffrey Lloyd (e.g. 1979; 1987) developed
nuanced and painstaking explorations of both discontinuities and
overlaps in different domains between emerging ‘rational’

19
On Nestle’s abiding influence, see Buxton (1999) 1–4; Most (1999b) 26–31. For diverse,
more or less critical revaluations of Nestle, see Buxton et al. (1999); Morgan (2000)
30–7; Fowler (2011).
20
Cf. Lloyd (1979) 4–5; Harrison (2006) 124.
21
Jaeger (1947) note v, 6; followed by Guthrie (1952); cf. also Broadie (1999).
22
Cornford (1952) 62–126. Most (2003) 317–19 sees the Graeco-Roman philosopher as
a species of the genus ‘divine man’ (theios anêr), typically (but not invariably) expres-
sing his wisdom and godliness also through rational argument and investigation. Most
(1999a) and (2007) shows the positive and appropriative engagements of early philo-
sophers with the epic traditions.
23
Cornford (1952) 127–55.

16
 1.1 Rationality and Irrationality
philosophical and scientific thought and traditional ‘irrational’
religious and mythical thought. He highlights, for example, the
compresence in many Hippocratic texts of notions of illnesses as
caused, diagnosed and treated naturally or by divine agency.24
In the latter half of the twentieth century, the discovery of the
Derveni papyrus (probably dating to about 340 BCE) offered new
possible insights. Gábor Betegh (2004) has made a strong case for
the now-prevalent view that the author of this text is concerned to
explicate Orphic theology, eschatology and initiation within
a cosmological framework.25 Thus, for example, the Derveni
author identifies Orpheus’ Zeus with Mind and intelligent Air
(PDerv. col.14–17, 19, 23). In the case of Empedocles (fifth
century BCE), the publication in 1999 of the Strasbourg papyrus
conclusively put to rest already languishing attempts to divorce as
two entirely unrelated aspects of his thought his so-called scien-
tific or philosophical speculations and his so-called religious
preoccupations.26
These important landmarks notwithstanding, there is no ques-
tion that the deep-seated notion that philosophical (rational)
thought and religious or god-centred (irrational) thought pull in
opposite directions continues to pose a fundamental challenge for
the development of interpretative models that seek genuinely to
move beyond such oppositions and to integrate the different
aspects of our evidence for the early Greek philosophers. Once
again, the explicit assertions made by the more recent successors
of the traditional narrative and its most radical recent debunkers
illustrate the tensions at play. Lesher’s (2008) model is global: the
rational, upbeat and secularised philosophical epistemology of
‘the Presocratics’ (exemplified by Xenophanes, Alcmaeon,
Heraclitus, Parmenides and even Pythagoras and Empedocles),
in which the mortal argues and is active, succeeds and overthrows
the older, defeatist poetic epistemology for which interactions
with the divine were central and in which the mortal was passively

24
See Lloyd (1979) 18–56. On Regimen offers one clear illustration of such compresence
(see Ch. 1.2).
25
See Betegh (2004) esp. 283, 354–9; cf. Most (2007) 277–9; Bernabé (2013) esp. 27–8.
26
See further Ch. 6.2. The papyrus did not, though, put an end to deflationary and
deliteralised interpretations of Empedocles’ daimonology: Ch. 6.2, n.29.

17
 Rationality and Irrationality, Philosophy and Religion
receptive.27 Patricia Curd (2002) similarly poses as competing the
conception of the pre-Socratics as ‘systematic thinkers’
committed to ‘rational inquiry and naturalistic explanations’ inde-
pendently of divine interferences and as religious thinkers, poets,
prophets, healers and law-givers.28 She ascribes to the pre-
Socratics a ‘rational’ rather than ‘poetic, mystical, and religious’
intellectual project, and globally defends this traditional interpre-
tation against what she calls ‘hypo-rational’ objections to it.29
Correspondingly, Kingsley (1995; 1999; 2003) presupposes
throughout his work that the centrality of associations with the
divine for the attainment of wisdom in Pythagoras, Parmenides
and Empedocles categorically dissociates them from anything
like rational, philosophical inquiry or argument. Laura Gemelli
Marciano (2002; 2008; 2013) offers a mirror image of the tradi-
tional narrative as expressed by Curd: Xenophanes, Heraclitus,
Parmenides, Empedocles et al. are practically-minded religious
sages, poets, diviners, healers and initiators, formed by engaging
with their cultural surroundings, rather than systematic thinkers
engaged in rational, philosophical inquiry.
To be sure, many interpreters will not be happy to align them-
selves with either party.30 But the problem which such interpreters
are facing – and I count myself among them – is not as easy as the
decision to grant or deny their assent to one camp or the other.
The scholars cited in the previous paragraph give voice to concep-
tions of philosophy, religion, rationality and irrationality from
which we cannot extricate ourselves by simple fiat. The painstak-
ing and challenging work of developing thick interpretative
accounts of particular thinkers, texts and movements which in
fact do justice to both aspects of our evidence, which do not
suppress one in the face of the other and, most importantly,
27
See similarly Curd (2013a); Granger (2013b) 165–7. 28 Curd (2002) n.b. 115–16.
29
Curd (2002) 119; see similarly Curd (2013a) 222 (‘a new “secular philosophical”
outlook’). Similar, more or less generalised or qualified attitudes are reflected, for
instance, by Humphreys (2004) e.g. 55 (‘[t]he new secular model of the cosmos’);
Naddaf (2005) n.b. 163; Kahn (2013); Gregory (2013).
30
For example, Benitez and Tarrant (2015) express dissatisfaction with the ‘standard
dichotomy’ (221) between philosophers as followers of reason, and poets and other
contemporary figures as followers of religion. Cf. the discontent expressed in Sassi
(2009) 225–9 with both ‘hyper-rational’ and ‘hypo-rational’ interpretations of
Parmenides.

18
 1.2 Some Promising Candidates?
which integrate – and show how we can integrate – the insights
afforded by both, remains largely ahead of us.
An important first step, then, is to disentangle clearly what are
distinct intellectual phenomena, which had traditionally been
lumped together under the heading of ‘rationality’: on the one
hand, coherent, critical, inferential, questioning and explicative
thinking; on the other hand, secularising moves away from pat-
terns of thought in which gods, interactions with gods and divine
interferences are central or accommodated. We should not expect
from the get-go that the former will always align with the latter or,
indeed, that it should tend to do so.

1.2 Some Promising Candidates? Milesians,

Hippocratics and Myth-Critics
I do not wish to reject categorically any qualified variation or
application of the ‘myth to reason’ model nor, therefore, the
possibility that, in some cases, both aspects of ‘rationality’ do
indeed align in a given thinker or text. Nonetheless, we will
examine in this section some of the most important and, at first
sight, promising candidates for such an alignment, and this exam-
ination will illustrate the difficulty of the question of ‘rationality’
and ‘irrationality’ in archaic and classical Greece. It will also give
us a better understanding of pertinent aspects of the broad intel-
lectual world which thinkers like Xenophanes, Parmenides and
Empedocles inhabited.
Let us turn first to the Ionian cosmologists, Thales,
Anaximander and Anaximenes, who worked in Miletus in succes-
sive generations in the late seventh and sixth centuries BCE.
According to a traditional and entirely plausible view, the
Milesians not only manifest innovative forms of coherent, critical,
questioning and explicative thinking – showing an emerging
preoccupation with empirical observations and with theories
which account for them coherently and economically (‘abductive’
reasoning) – but also break with what preceded them insofar as
they strive towards less mythological and anthropomorphising
patterns of cosmological explanation. So, for example, when
Anaximander maintains that thunder and lightning occur when
19
 Rationality and Irrationality, Philosophy and Religion
rare and light wind breaks out violently from surrounding thick
clouds (DK12 A23), or when Thales traces earthquakes to move-
ments in the water on which the earth rests (DK11 A15; Hippol.
Haer. 1.1.2), they are markedly and even pointedly forgoing
explanatory appeals to the capricious and all-too-human divine
persons in terms of whom these phenomena had previously been
described.31
At the same time, however, our evidence – scarce as it is –
suggests that the Milesians move towards a distinctively rather
than less god-centred world view. Thales famously advanced the
slogan ‘all things are full of gods’ (πάντα πλήρη θεῶν, Aristotle,
de An. 411a7–8 = DK11 A22). Some later writers associate him
directly with the idea of divine governance, although we should
not put much weight on these sources. So, Philoponus reports what
appears to have been a general view (φασί) according to which
Thales had stated (ἔλεγεν) that ‘providence extends to the extremes
and nothing escapes its notice, not even the smallest thing’ (in de
An. 86.30–1).32 This report, though, could represent an interpre-
tative inference from the ‘all things are full of gods’ slogan itself.
Other sources ascribe to Thales the idea that a divine cosmic
intelligence uses water to effect and maintain the world order,
but they smack of Stoicising influences.33 We have better evidence
for intelligent divine governance in Anaximander. Aristotle tells
us that Anaximander presented his favoured cosmological princi-
ple, ‘the unbounded’ (apeiron), as a deathless and indestructible
power which surrounds (περιέχειν) and is steering (κυβερνᾶν) all
things (Phys. 203b10–15 = DK12 A15). Aristotle’s term for ‘steer-
ing’ literally and standardly refers to the helmsman’s direction of
his ship. Our lone fragment of Anaximander (DK12 B1), more-
over, indicates that a pattern of cosmic justice characterises the
balanced interplay between competing cosmic forces (presumably
the opposites), and our evidence from Aristotle (as well as from
31
See further Gregory (2013) 43–67. This does not mean that the Milesians were not
influenced in their cosmogonic and cosmological speculations by earlier Greek and
Near-Eastern mythologies; see Burkert (2008) esp. 68–72; cf. (2000) 137–9.
32
Cf. Sedley (2007) 7, n.21.
33
See KRS (1983) 97, n.1. Aëtius 1.7.11 = DK11 A23 presents this cosmic mind as an
immanent power which itself suffuses water; cf. similarly but less clearly Cicero, ND
1.25.2–7 (2–5 = DK11 A23).

20
 1.2 Some Promising Candidates?
Simplicius (in Ph. 9.24.13–21), who quotes the fragment) suggests
that it is the apeiron which governs or ‘steers’ these well-ordered
processes as a kind of cosmic helmsman. Concerning
Anaximenes, Aëtius reports as follows: ‘Just as, he says, our
soul, which is air, controls us (συγκρατεῖ ἡμᾶς), so too breath
and air surround (περιέχει) the whole cosmos’ (1.3.4 = DK13
B2). Now, it is difficult to say just how much we can infer from
this analogy between cosmic air or ‘breath’ (pneuma) and the
human soul, even if we assume that the substance of the analogy
does indeed go back to Anaximenes.34 Strictly speaking, Aëtius
only reports that Anaximenes said that air ‘surrounds’ the cosmos.
And yet, in this particular respect the point of contact between the
analogues is unclear (does the human’s soul ‘surround’ them?).
The analogy could instead, then, convey the idea that, in encom-
passing the world order, the cosmic air relates to it in the same way
in which our microcosmic human soul (our ‘air’) relates to us and
controls us: that is, in an animate and (at least largely) intentional
manner. Even independently of this difficult and uncertain ana-
logy, however, the fact that Anaximenes appealed to the agency of
air to explain, for example, the regular and orderly courses or
‘turnings’ (tropai) of the heavenly bodies (DK13 A15) plausibly
suggests that his air not only surrounds the world order like
Anaximander’s apeiron, but also – again, like Anaximander’s
apeiron – exerts an intelligent directive and steering influence on
this well-ordered world.35 Indeed, in the face of a predilection to
think of ‘pre-Socratic’ philosophers as non-teleological thinkers –
a predilection which goes back to Plato – it has been persuasively
argued that the idea that the world is governed by intelligent divine
power was a pervasive and long unquestioned assumption in the
sixth and early fifth centuries.36
34
Alt (1973) esp. 157–60 finds in B2 an anachronistic retrojection onto Anaximenes of the later
views of Diogenes of Apollonia. This could certainly be true, although Alt dismisses too
easily the alternative possibility, that Anaximenes had influenced Diogenes’ views (162–4).
35
Cf. Alt (1973) 132. On the divinity of Anaximenes’ air, see further DK13 A10; cf. Pl. Phlb.
30a3–c7, with KRS (1983) 150–1; cf. Sedley (2007) 6–7. In the light of these signs of
directive control and guidance, we should not consider Anaximander’s and Anaximenes’
divinities an ultimately atheistic metaphor for nature, contra Whitmarsh (2015) 58–9.
36
Sedley (2007) 1–8; Guthrie (1952) esp. 90–6; Palmer (1998) 11, 33–4; cf. Warden (1971)
12; Drozdek (2004) 147; Long (1996) 140. In addition to the (patchier) evidence for the
Milesians considered here, see Xenophanes: DK21 B25 (see Ch. 3.4); Heraclitus: DK22

21
 Rationality and Irrationality, Philosophy and Religion
Importantly, then, when the Milesians explain thunder and
earthquakes in the manner described above, they do not thereby
implicitly reject the view that these phenomena should be under-
stood as the products of divine activity.37 By giving a detailed
physical specification of some event (say, my hand flexing) we
have in no way demonstrated that this event did not constitute
a cognisant or intentional action. What little pertinent evidence we
have points in the other direction. Anaximander, for example, can
take it that thunder and lightning happen because wind breaks out
from thick clouds and, equally, as a result of the general steering of
all things by the apeiron (πάντα κυβερνᾶν). The Milesians do
forgo accounts of natural phenomena in terms of the actions of
erratic and all-too-human divine agents. They do not, however,
forgo accounts of natural phenomena in terms of the actions of
divine agents.
What is more, we have no reason to think that the speculations
of the Milesians brought them either into a practical conflict with
traditional ritual or, indeed, into a logical conflict with general
theological assumptions that might plausibly be taken to under-
pin such ritual. It is often taken as read that the Milesians’
cosmology and theology left no room for the traditional gods of
cult and prayer, gods who may interfere in human affairs or be
responsive to human actions and words. Indeed, scholars often
take it that the Milesians sought to limit the divine strictly to the
cosmological functions of their cosmological principles.38 This
is all far from clear. The fact that their central cosmic principles
are accorded a divine status – or, indeed, are conceived as the

B41 (ἐκυβέρνησε πάντα); B64 (πάντα οἰακίζει); Parmenides: DK28 B12 (πάντα
κυβερνᾷ); Diogenes of Apollonia: DK64 B5; cf. Thgn. 373; Soph. El. 175–6; Tra. 127.
Anaxagoras, while avoiding traditional theological vocabulary, describes the cosmic
Mind as a supremely cognisant and judicious power (γνώμην γε περὶ παντὸς πᾶσαν
ἴσχει . . . πάντα ἔγνω) which rules over things (κρατεῖ) and which placed in order past,
present and future states of affairs (διεκόσμησε, DK59 B12); with Sedley (2007) 7–30.
We will return in the next section to Plato’s representation of the early philosophers as
non-teleological thinkers. In the Philebus, Plato portrays rather differently the agreed-
upon tenets of ‘our ancient predecessors’: ‘mind and a certain wondrous understanding
arrange and steer’ all things (διακυβερνᾶν, 28d7–9); ‘mind always rules the all’ (30d7–8);
‘mind is our king over both heaven and earth’ (28c6–8).
37
Contrast e.g. Whitmarsh (2015) 57; cf. Granger (2013b) 171–2, 176.
38
E.g. Gerson (1994) 1–32, esp. 28; Algra (1999) 53; Gregory (2013) 43–67, cf. 1–22;
Granger (2013b); Whitmarsh (2015) 53, 65–6.

22
 1.2 Some Promising Candidates?
chief divinity – does not imply that these principles, in their
austere cosmological functions, exhaust the sphere of divinity.
The fact that air is divine need not mean that nothing else in the
world can be meaningfully called a divinity. In order to draw this
inference, we would have to assume that the Milesians were
monotheists. Our evidence here is even scarcer than elsewhere
but, again, what little we have does not substantiate this
assumption.
Thales, as we noted, was from early on associated with the
slogan ‘all things are full of gods’. It is often taken that ‘gods’
could refer either to souls (to which Thales appears to have
appealed as originators of motion: Aristotle, de An. 405a19–21 =
DK11 A22) or to water (his central cosmological principle).39
We can also, however, entertain the view that ‘gods’ involves
here something closer to the usual sense of the word.
The statement may convey the broader theological thought that
the world as a whole is widely inhabited by immortal, animate and
sentient agents (perhaps of divergent degrees of power and impor-
tance), at least some of whom can impact on us and be responsive
to us.40 The set of ‘gods’ on such a view could include (promi-
nently) water and/or motion-generating souls, but it need not be
exhausted by either or both. The thought that the world confronts
us at every turn with divine powers and agents agrees in funda-
mental ways with attitudes which we find in the verses of Homer
and Hesiod.
Augustine expressly states that Anaximenes acknowledged
and discussed gods: ‘nor did he deny the gods or pass them
over in silence (nec deos negauit aut tacuit); he did not, how-
ever, believe that air was fashioned by them, but that they

39
See Pinto (2016), who mounts an interesting case for the second alternative. Aristotle
suggests a version of the first one, but he overtly presents this as a speculative inter-
pretation (ὅθεν ἴσως, de An. 411a7–8) and is perhaps influenced by Plato’s appropriation
of the dictum at Lg. 10.899b.
40
‘Widely inhabited’: Pinto (2016) 252 observes that the slogan can but need not
indicate that ‘gods’ are literally omnipresent in the world. He compares Gorgias’
ἅπαντα γὰρ πλήρη φυλακω ͂ ν (Pal. 12), which does not mean that guards are literally
pervading everything but that there are a lot of them around. Equally, it is possible
that Thales espoused literal divine omnipresence, if water is (partly) what ‘gods’
refers to, and if water is omnipresent. These are both plausible but not obvious
premises.

23
 Rationality and Irrationality, Philosophy and Religion
originated from air’ (Civ. Dei 8.2 = DK13 A10). This statement
fits with Hippolytus’ report that, according to Anaximenes, air is
the source ‘from which the things that are and were and will be
and gods and divine things come to be, and the rest from the
offspring of this’ (ἐξ οὗ τὰ γινόμενα καὶ τὰ γεγονότα καὶ τὰ
ἐσόμενα καὶ θεοὺς καὶ θεῖα γίνεσθαι, τὰ δὲ λοιπὰ ἐκ τῶν τούτου
ἀπογόνων, Haer. 1.7.1 = DK13 A7.1). These remarks come in
the context of our most detailed (and seemingly informed)
doxographic account of Anaximenes, and there are inconclusive
reasons to think that this particular sentence may go back to
Anaximenes himself.41 It is, again, at least likely that the word
‘gods’ signifies immortal, animate and sentient agents who may
be responsive to human words and actions. This is the usual
sense of the term and, furthermore, the way in which Augustine
puts his point – that Anaximenes did not deny the existence of
gods but did make them the offspring of air – suggests the
qualified accommodation of existing conceptions of gods within
a cosmological framework anchored by divine air. It has been
suggested that the phrase ‘gods and divine things’ signifies
exclusively other constituents (or states-of-air) which go into
the makeup of the ordered world: fire, wind, earth, stone, etc.42
This is likely enough as an interpretation of ‘divine things’ but it
is an arbitrary restriction on the meaning of the masculine plural
word for ‘gods’. Furthermore, as a construal of the conjunction
‘gods and divine things’ as a whole, this view has the unattrac-
tive consequence of reducing both conjuncts into a single cate-
gory. The list which Hippolytus records is carefully structured.
Yes, the final pair, ‘gods and divine things’, forms a bipartite
conceptual unit distinguished from the opening trio. At the same
time, each constituent in the list is marked off from the others in
a uniform manner (καί . . . καί) and a simple reduction into
a single category would not be possible in the case of any
41
See Granger (2013b) 181–2. Granger observes the assonance and rhetorical accumula-
tion of the sentence (which opens Hippolytus’ account of Anaximenes) and interestingly
interprets the high-sounding language as Anaximenes’ pious praise for divine air in its
comprehensive responsibility for all there is, comparing similar language in Il. 1.70;
Hes. Th. 38 (cf. 32); Anaxag. DK59 B12 (καὶ ὁποῖα . . . καὶ ὁποῖα ἔσται); Emped. DK31
B21.9.
42
Granger (2013b) 181.

24
 1.2 Some Promising Candidates?
other two constituents.43 We should not baulk at the suggestion
that Anaximenes’ divine air is capable of generating animate,
sentient and responsive agents. Air is, after all, the ultimate
generator of all things, including humans.
If Thales and Anaximenes accommodated animate and sentient
gods, then nothing in their cosmological speculations would con-
flict with those gods taking all kinds of effective action in the
world to the benefit of individuals or groups of people – much as
humans successfully and coherently do, but on a larger scale or in
more powerful and diverse ways – in response to prayer and
sacrifice.44
The Milesians display pioneering preoccupations with specula-
tive inquiry, systematic explanation, observation and abductive
reasoning. Relatedly, they critically and pointedly distance them-
selves from the anthropomorphising patterns of explanation that
typify the mythological tradition. And yet, divinity and the intel-
ligent guidance of the world appear to retain a prominent place in
their systems. There is, moreover, no evidence for a general logical
or theological tension with the assumptions that could plausibly be
taken to underpin traditional religious ritual. Greek religion, it
must be remembered, was not a theological unity.45 If the
Milesians rejected the anthropomorphising portrayals of the gods
familiar from the mythological traditions, it does not follow that
they repudiated the basic idea of a multiplicity of sentient and
responsive gods, or the practice of traditional cult, which was
never concerned to prescribe to the worshipper any particular

43
Note that, on any interpretation, DK13 A7 and A10 show that Anaximenes did not think
that the divine sphere was exhausted by air. Some products of air (‘gods and divine
things’) can be meaningfully designated ‘divine’ by contradistinction with other pro-
ducts of air (e.g. animals), which cannot be so designated.
44
The closest that our evidence for any of the Milesians comes to conceivably excluding
the gods of cult and prayer is Aristotle’s remark that Anaximander understood the
apeiron as ‘the divine’ (τὸ θεῖον). At a push, one could take the article to suggest that
the apeiron exhausts the sphere of divinity. Interpreted in this monotheistic way,
however, Aristotle’s remark would be nothing more than his own excessively strong
inference from the fact that Anaximander saw the apeiron as divine and ascribed
divine attributes to it (ἀθάνατον γὰρ καὶ ἀνώλεθρον ὥσπερ φησὶν Ἀναξίμανδρος, Phys.
203b13–15 = DK12 A15). It is difficult to assess the possible religious import of our
reports (DK12 A17) that Anaximander considered as gods ‘the innumerable worlds’
(Cic. ND 1.25) or ‘the boundless heavens’ (Aët. 1.7.12).
45
We will return to this point in the next section.

25
 Rationality and Irrationality, Philosophy and Religion
theological conception of the gods. Indeed, our sources suggest
that Anaximenes expressly accommodated the gods within his
own novel cosmological system.
None of this is to deny that it was possible, at least for a casual
observer, to see in Milesian-style cosmological accounts of natural
phenomena an attempt to write the divine out of the equation
altogether and, therefore, a general threat to traditional religiosity.
Such suspicious responses to cosmological speculation (especially
as represented by Anaxagoras) did emerge later on in Classical
Athens (Chapter 1.3). This makes it all the more notable, however,
that our evidence for the Milesians offers nothing comparable.
We have no sign that they or their contemporaries took their
inquiries to compete in a general way for the same logical or
conceptual territory with traditional religious attitudes or to sup-
plant them in their own appropriate contexts. Indeed, Plato relates
the tradition that Thales and the other members of the Seven Sages
offered the first fruits of their wisdom to Apollo by inscribing the
famous sayings in his temple at Delphi (ἀπαρχήν . . . ἀνέθεσαν,
Protag. 343a1–b3). I do not wish to affirm that the Milesians did,
in fact, admit gods who may respond to prayer and sacrifice. Our
evidence is too precarious to speak with confidence. But the
evidence certainly does not support the common view that their
cosmological accounts, or the very distinctive divinities which
constituted their key principles, clashed with the traditional pos-
tulation of such gods. The ease with which this view is nonetheless
habitually asserted instructively reflects problematic expectations
concerning the orientation of rational natural philosophy.46

46
Do the Milesian accounts of natural phenomena conflict with their status as omens?
Conceivably, but only if and when the account maintains that the phenomenon in
question results from periodic and entirely predictable processes and occurs on
a schedule which cannot be traced to human affairs. Eclipses are one interesting case
(Thal. DK11 A5, A17; Hdt. 1.74; Anaximand. DK12 A21–22; cf. Thuc. 7.50.4). This
consideration will not, however, affect other omens (birds, entrails, oracles; see Ch. 3.1).
Plutarch does present a disagreement between Anaxagoras and the seer Lampon over
the explication of a one-horned ram, although the triumphant ease with which he goes on
to expose the false dichotomy and conclude that Anaxagoras correctly explicated the
‘how’ of the phenomenon and Lampon the ‘why’ makes the anecdote seem like some-
thing of a Platonic foil (Peric. 6; with Hershbell (1982) 141–3, and Ch. 1.3 for Plato’s
attitude to Anaxagoras). The sources which report Xenophanes’ rejection of divination
indicate that he was (together with Epicurus) alone in doing so (DK21 A52). We noted
that Plato associated Thales with Delphi.

26
 1.2 Some Promising Candidates?
At first sight, the Hippocratic treatise On the Sacred Disease
(probably written in the late fifth century BCE) offers a champion
example of a text in which both senses of ‘rationality’ coincide.
The treatise attacks the view that epilepsy is the result of divine
punishment meted out by angry gods and treatable by ritual
purifications, incantations and an assortment of other measures
such as avoiding black clothes or goatskin (Morb. Sacr. 1–2).
Epilepsy is, instead, a hereditary condition caused by an excess
of phlegm, which blocks passages to and from the brain (Morb.
Sacr. 5–16). Furthermore, however false his own theory may be,
however ungrounded many of his speculations and however inef-
fectual his proposed treatments, the author does argue for his
views by appeal to empirical evidence and abductive inferences.
So, for example, he ingeniously explains that the convulsive limb
movements characteristic of epileptic seizures are caused by the
air which moves violently up and down the body as it tries but fails
to escape through vessels blocked by phlegm, and he similarly
traces the loss of voice and cognitive functions to the disruption
caused to the supply of air to the brain (Morb. Sacr. 10). Equally,
the author reasons against the purifiers. He observes that they are
vulnerable to counter-evidence. If goatskin causes epilepsy, then
the inhabitants of Libya should suffer from the disease universally
(Morb. Sacr. 2). If epilepsy constitutes divine punishment, then
why does it affect animals no less than humans (Morb. Sacr. 14)?
And do these different statements about the causes of epilepsy not
constitute an internally incoherent jumble (Morb. Sacr. 2)?
And yet, the author does not move from a religious to
a secularised world view. Far from removing the divine from his
analysis, the author both begins and ends his work by advancing
and emphasising an arresting universal theory of disease: ‘all are
divine and all human’ (πάντα θεῖα καὶ πάντα ἀνθρώπινα, Morb.
Sacr. 21, cf. 1).47 The fact that the author never does more than hint
tantalisingly at what either term (or their conjunction) amounts to
in his view, and leaves this as to some degree an open matter for his
readers’ own elaborations, in no way detracts from the prominence

47
Similar assertions open Ch. 5 and conclude Ch. 16, prominently bookending the
author’s positive account of epilepsy.

27
 Rationality and Irrationality, Philosophy and Religion
which he accords this thesis. Whatever else, moreover, his theol-
ogy does involve some form of divine intervention. In Chapter 4,
the author writes that his antagonists purify with blood and other
such materials those who suffer from epilepsy, as though they
were polluted or impious offenders, ‘they whom they should
have treated in the opposite ways; they should have sacrificed
and prayed and brought them to the sanctuaries to supplicate the
gods’ (Morb. Sacr. 4). Now, it is not clear whether the author is
saying that those who suffer from epilepsy should have been
brought into the temples or only those who are in fact in the
conditions in which the purifiers falsely believe the epileptic to
be (the polluted and the impious).48 The author leaves it open for
us to read the sentence in the former, more inclusive way. Quite
possibly, then, he prescribes prayer and sacrifice as one proper
response to illness. At a minimum, however, he prescribes such
temple ritual as the proper response to pollution and impiety.
Indeed, the author proceeds to advance the asymmetrical theolo-
gical view that the divine purifies but does not pollute: ‘Well,
I myself do not hold that a man’s body is polluted by a god, the
basest thing by the holiest . . . Indeed, for the greatest and most
unholy of our transgressions it is the divine that purifies and
sanctifies us and becomes our cleanser’ (Morb. Sacr. 4).
On the Sacred Disease, in short, exhibits a conflict not between
religious and secular world views but between competing theolo-
gical conceptions of health, disease, pollution and purification.49
The treatise argues against some conceptions of divine interven-
tion while accommodating others.50 It at least leaves open the
possibility of cooperation between human and divine agencies in
the treatment of diseases. The Hippocratic On Regimen articulates
this attitude more explicitly. The author not only advances certain
dietary measures but also mentions Sun, Heavenly Zeus, Zeus
Protector of Home, Athena Protector of Home, Hermes, Apollo,
48
That is, it is not clear whether οὕς looks back to τοὺς ἐχομένους as its antecedent or only
to the other masculine plurals after ὥσπερ: καθαίρουσι γὰρ τοὺς ἐχομένους τῇ νούσῳ
αἵματί τε καὶ ἄλλοισι τοιούτοις ὥσπερ μίασμά τι ἔχοντας, ἢ ἀλάστορας . . . οὓς ἐχρῆν
τἀναντία τούτων ποιεῖν κτλ.
49
The author even levels a charge of impiety against the purifiers, Morb. Sacr. 3–4.
50
The author is particularly scandalised by what he represents as the purifiers’ claims to be
able to coerce the gods to act, Morb. Sacr. 4.

28
 1.2 Some Promising Candidates?
Earth and the Heroes as healers who may be invoked to avert
diseases in certain circumstances (4.89; cf. also 90) and prescribes
a combination of human and divine efforts: ‘Prayer is fitting and
surely good, but it is while also lending a hand oneself that one
must call upon the gods’ (4.87; cf. 89–90). We would be reading
against the grain if we interpreted this act, not as an intercessory or
petitionary prayer, but as an aspirational self-address or a morale-
boosting measure for patients who hold beliefs which the author
himself lacks.51 Such a deflationary construal renders it peculiar
that the author should identify so specifically particular gods who
are suitable for particular circumstances. His remark that the
mortal too should contribute (καὶ αὐτὸν ξυλλαμβάνοντα) indicates
a combination of human and divine agencies.52
The historiographical tradition of what we call the ‘rationalisa-
tion’ of myths looks particularly promising at first sight. The
genealogist, geographer and ethnographer Hecataeus of Miletus
(born around the middle of the sixth century BCE) began his
genealogical work with these words (BNJ 1 F1):
῾Εκαταῖος Μιλήσιος ὧδε μυθεῖται· τάδε γράφω, ὥς μοι δοκεῖ ἀληθέα εἶναι· οἱ γὰρ
῾Ελλήνων λόγοι πολλοί τε καὶ γελοῖοι, ὡς ἐμοὶ φαίνονται, εἰσίν.

Hecataeus of Miletus speaks as follows. I write these things, as they seem to me to

be true. For the accounts of the Greeks are many53 and laughable, as they appear
to me.

Hecataeus opens this prose work not with an invocation of the

Muses or any other sign of divine disclosure, but with a polemical
appeal to implicit standards of plausibility and respectability
grounded in his own authority (ὥς μοι δοκεῖ . . . ὡς ἐμοὶ
φαίνονται). But what is laughable for Hecataeus and what is not
laughable? Hecataeus certainly removes from some myths what

51
Gregory (2013) 92–4.
52
Nutton (2013) 111–15 dismisses the traditional notion of a general tension or hostility
between so-called secular and religious medicine or human and divine healers, and
discusses some aspects of the close associations between them in fifth- and fourth-
century Greece. For On the Sacred Disease, see further van der Eijk (2005) 5, 19–21,
45–73; Nutton (2013) 113–144. Airs, Waters, Places (22) offers a similar universal
conception of disease (πάντα ὅμοια καὶ πάντα θεῖα).
53
Hecataeus perhaps hints here both at unsystematic proliferation and internal
inconsistency.

29
 Rationality and Irrationality, Philosophy and Religion
we would call their fantastical features and develops alternative
explanations for the stories’ origins which, while entirely spec-
ulative, accord better with observed phenomena. So, for example,
Heracles was not dispatched to fetch Hades’ dog, Cerberus.
Rather, he was sent to fetch a snake, whose swift-acting and deadly
poison earned him the name ‘the Hound of Hades’, whence the
confusion arose and the story developed (BNJ 1 F27; cf. F19, F26,
F35a). On the other hand, the same Hecataeus, who opened his
genealogical work in the manner cited above, was also able to
trace his own ancestry as far back as a god in the sixteenth
generation (BNJ 1 T4, F300 = Hdt. 2.143) and to explore more
generally questions of divine descent, ascribing to Achelous, for
example, a different divine parentage (Sun and Earth) than the one
he is accorded in Alcaeus (fr. 450 Campbell = BNJ 1 F35b) or in
Hesiod (Th. 337–40; Hecataeus: BNJ 1 F35b, with Pownall
(BNJ) ad loc.).54
Like his forerunner Hecataeus, Herodotus (writing in the
middle of the fifth century BCE) similarly eschews anything like
invocations of the Muses and participates sometimes in the same
tradition of myth criticism.55 The types of direct and personal
interactions between gods and humans that we encounter
constantly in Homer and Hesiod are missing from Herodotus’
own accounts of the past. A famous and notable example is the
complete absence of the gods in Herodotus’ opening discussion of
the successive and historically consequential abductions of certain
prominent women, including Io, Europe and Helen (1.1–5).
Indeed, Herodotus can sometimes express overt scepticism con-
cerning assertions by his contemporaries of direct divine
54
In her commentary on BNJ 1 F300 = Hdt. 2.143, Pownall convincingly rejects as
ungrounded and unmotivated earlier attempts to suggest that Hecateaus’ claims for
divine ancestry were somehow not truly meant to be taken as such. Moyer (2002)
responds to West (1991), who argued that Herodotus had himself fabricated the entire
story of Hecataeus’ and his own encounters with the Egyptian priests. Even West,
though, allows that the details of Hecataeus’ genealogy adumbrated in this episode
are unlikely to be Herodotus’ invention (152). Aëtius ascribes to Hecataeus (along with
Heraclitus) a conception of the Sun (one of Achelous’ parents, in his account) as an
intelligent agent with an ignited mass for a body (ἄναμμα νοερόν, BNJ 1 F302d = Aët.
2.20.6).
55
E.g. 2.55–7 (the oracle at Dodona was not founded by a speaking dove but by a foreign
priestess referred to as a ‘dove’), cf. 4.8–11; see further Hawes (2014) 8–11; Moyer
(2002) 84–7.

30
 1.2 Some Promising Candidates?
visitations. He doubts, for example, the Chaldeans’ claims that, in
the temple of Baal, the god himself visits the shrine and rests upon
the couch (ἐμοὶ μὲν οὐ πιστὰ λέγοντες, 1.182; cf. 1.60). And yet, the
gods dominate The Histories and Herodotus clearly considers it
a central aspect of his brief as an inquisitive historian to consider
questions concerning the (often indirect) ways in which divine
interventions guide and at least co-determine the progression of
human affairs. Oracles, omens and dreams play major roles in his
historical narratives (e.g. 1.6–91; 7.12–19). At one juncture,
Herodotus surveys several competing explanations for
Cleomenes’ fatal madness – including, for example, the Spartans’
opinion that Cleomenes simply drank excessive amounts of neat
wine – and expressly settles on the view that Cleomenes was
divinely punished for his foul play in dealing with the Pythia (6.75,
84).56 At some junctures, the observation of fitting events which
would otherwise constitute extraordinary coincidences leads
Herodotus – in a perfectly respectable if no doubt vulnerable exam-
ple of abductive reasoning – to infer divine guidance (δῆλον ὦν μοι
ὅτι θεῖον ἐγένετο τὸ πρῆγμα, 7.137; πολλοῖσι τεκμηρίοισι, 9.100).57
To take, finally, a later but no less illuminating example,
Palaephatus’ On Unbelievable Tales (probably written some-
time in the second half of the fourth century BCE) represents
the best-known specimen of the ‘rationalist’ tradition of myth
criticism.58 Palaephatus pursues a systematic campaign to elim-
inate from traditional myths their ‘unbelievable’ features, such
as monsters and cross-species metamorphoses. He sometimes
argues against such features on broadly biological grounds
(centaurs, for instance, are an impossibility because a human
cannot swallow what a horse digests, de Incred. 1). He always
proposes an alternative chain of events, which accords more
easily with our experiences, and which explicates how the
story originated even as it deflates it (the Centaurs were merely

56
For divine punishment as a historical factor, cf. e.g. 4.205; 6.91, 139; 7.134–7; 8.129,
with Harrison (2000) 102–121; Fowler (2010) 329–30.
57
With Fowler (2010) 333–4; Parker (2011) 7. On the centrality of the gods in Herodotus’
historiography, see further Harrison (2000); Fowler (2010); cf. Kindt (2006) 45–6.
58
On Palaephatus, see Stern (1996); (1999); Hawes (2014) 37–91. On the uncertain dating
of the text, see Hawes (2014) 227–38.

31
 Rationality and Irrationality, Philosophy and Religion
the first people to ride horseback, seen from a distance: de
Incred. 1).59 Palaephatus’ critiques and historical reconstruc-
tions, moreover, are not merely floating in argumentative
space. They are integrated into a thought-out theoretical frame-
work, which posits as an organisational principle the regularity
of natural processes and phenomena and, therefore, a basic
physical and biological continuity between past and present
(de Incred. pref. 9–12). Furthermore, they are informed – at
least purportedly – by first-hand investigations into the indigen-
ous traditions and conditions of the location of each tale (de
Incred. pref. 22–4). But what is unbelievable for Palaephatus
and what is not unbelievable? What sort of thing does
Palaephatus target for reasoned criticism and what is he in fact
happy to accommodate? Notably, Palaephatus has no trouble
speaking of Zeus as a god who might well, if he so wished,
intervene in human affairs by effecting the transportation of
a mortal from one location to another in some way, although –
surely – he would not do so by having a maiden ride a bull (de
Incred. 15). Indeed, Palaephatus even has no trouble allowing
that Artemis could have transformed Actaeon into a deer: ‘But it
seems to me that Artemis has the ability to do whatever she
wishes’ (ἐμοὶ δὲ δοκεῖ Ἄρτεμιν μὲν δύνασθαι ὅ τι θέλοι ποιῆσαι, de
Incred. 6). Yes, nature displays an orderly and unchanging
regularity (de Incred. pref. 9–12), but, to judge by this revealing
statement, this is because the gods, who could do otherwise,
choose to preserve this regularity and can be reliably expected to
continue to do so. The laws of nature are unchanging but not
immutable.60 Interestingly, Palaephatus’ rationalisations quite
generally focus on eliminating monsters and heroes from tradi-
tional tales: he shows little interest in approaching the gods in

59
On Palaephatus’ biological attitudes and their Aristotelian background, see Hawes
(2014) 55–8. As is often observed, Palaephatus pursues much the same project of
systematic myth-criticism which Plato’s Socrates describes and derides at Phdr.
229d–e.
60
Pace Stern (1996) 10, therefore, Palaephatus is not struggling with a conflicted attrac-
tion to the principles of (i) divine omnipotence and (ii) immutable natural laws.
Principle (ii) has no textual basis in Palaephatus. Pref. 9–12 does not use modal
language. Stern (1996) 10 ultimately asserts – incompatibly with the statement about
Artemis in de Incred. 6 – that ‘Palaephatus imagines the power of the gods as limited.’

32
 1.2 Some Promising Candidates?
a similar fashion.61 Palaephatus’ particular brand of critical
reasoning shows no tension with the postulation of interven-
tionist and even omnipotent gods and does not bring him into
conflict with the traditional pantheon (although it does seem to
exclude any story which ascribes to the gods capricious and
unpredictable conduct). It would be tendentious to insist that
such tension and conflict would have been ‘next in order’ on
Palaephatus’ agenda, had he only persevered sufficiently in his
project of rationalisation.
To approach the question from another direction, can we point
towards cases of irrationality, in which the marked absence of
characteristic features of rational thinking coincides with distinc-
tively god-centred attitudes? In fact, Henk Versnel’s thought-
provoking study, Coping with the Gods (2011), is devoted to
drawing out a pertinent aspect of Greek religion. Versnel develops
a rich argument for the view that Greek religion is typified by what
he calls ‘luxurious multiplicity’. On this view, incompatible stor-
ies, representations and beliefs about gods coexist side-by-side in
different contexts in a peaceful and serene manner, without gen-
erating anxiety or concern over their incompatibility. The case of
divine cult names (epikleseis) represents one of Versnel’s more
convincing examples. When we encounter (as we constantly do)
gods with the same ‘first-name’ and different cult names, are we
dealing with multiple, locally worshiped gods or with multiple
designations for one single god? Is Zeus Olympios the same god as
Zeus Ephestios or Zeus Meilichios, or are these all distinct divine
agents? Versnel mounts a strong argument for the view that this
was not a question which the Greeks in general experienced as
a problem to be addressed and definitively resolved. In different
contexts, different attitudes to this issue may (or may not) come to
the fore, without generating general concerns regarding the incon-
sistency between them.62
61
This thematic point is often noted, e.g. Stern (1996) 9–10, 24. Contrast, for example,
Hecataeus BNJ 1 F35a: Pluto was actually the first mortal to have spoken of post-
mortem punishment as a reason to live justly.
62
See Versnel (2011) 60–87, offering a wide-ranging discussion of the evidence. In different
passages, Callimachus instructively reflects both a unifying take (Ap. 69–71) and
a separatist one (fr. 200a Pfeiffer); cf. Versnel (2011) 82. In X. Symp. 8.9, Socrates
suspends judgement in the case of Aphrodite Ourania and Aphrodite Pandemos.

33
 Rationality and Irrationality, Philosophy and Religion
A preoccupation with logical coherence is undoubtedly a key
feature of what we identify and discuss as rational thinking.
Insofar as Versnel successfully demonstrates the phenomenon of
luxurious multiplicity in Greek religion, his thesis, I think, has
interesting implications for the question of rationality and irration-
ality. Aspects and instances of Greek theology which display
comfort with inconsistency – which are characterised by the
peaceful coexistence of logically competing conceptions of the
gods – are, to that extent, irrational. We need not be using that label
pejoratively here. The flexible appeal to different theological con-
ceptions and explanatory principles in response to different con-
texts – as each situation demands – could be thought of as
a natural, useful and even intellectually honest response to the
ultimate unknowability of the divine.63 There are, however, at
least two important types of counter-example to Versnel’s thesis.
First, we do encounter texts that express coherent theological
positions and polemically exclude alternatives. The opening lines
of the first Homeric Hymn to Dionysus offer one clear example.
The poet surveys the various traditions regarding the birthplace of
Dionysus and rejects them as falsehoods (ψευδόμενοι) in favour of
his own view: the god was, in fact, born on Mt. Nyssa (1.1–9).
We will find much more complex instances of coherent and occa-
sionally polemical approaches to certain questions of theology in
Hesiod, Xenophanes, Parmenides and Empedocles.64
Second, we also find texts in which competing theological
views are indeed juxtaposed, but where their co-presence is any-
thing but peaceful or serene. A passage in Hesiod’s Works and
Days, which we will have occasion to cite and discuss
(Chapter 2.2), offers one powerful example. Within the space of
a few lines (267–73), Hesiod vacillates between competing con-
ceptions of Zeus as a reliable, just dispenser of punishments and
rewards and as an inconstant governing force, which operates
according to ultimately unforeseeable whims. This juxtaposition
is a source of fraught consternation and anxiety. Hesiod is all too
aware that this conception of Zeus as a capricious force, which he

63
Versnel (2011) 213 makes similar remarks (not using the term ‘irrational’).
64
On Hesiod, see also Ch. 1.5.

34
 1.2 Some Promising Candidates?
momentarily entertains here, would be disastrously incompatible
with his overarching ethical campaign in the Works and Days to
promote the life of justice (270–2). I suggested that we can mean-
ingfully describe as irrational the peaceful juxtapositions of com-
peting theological conceptions and representations that are
characteristic of what Versnel calls luxurious multiplicity. By the
same token, when Hesiod hesitates here between competing theo-
logical alternatives, while precisely making an issue of their
mutual incompatibility and divergent ethical ramifications, he
exemplifies Greek theological rationality.65
Our consideration in this section of selective but important and
revealing candidates for ‘rationality’ and ‘irrationality’ throws up
two key points. First, the connection between the two broad senses
of ‘rationality’ which were distinguished above – coherent, critical
and argumentative thinking on the one hand and secularising
tendencies on the other hand – may not be arbitrary, but it is
certainly contingent rather than necessary. Second, it is imperative
to consider each particular thinker or text in their own right (which
does not mean, of course, to ignore their explicit or implicit
connections with other thinkers or texts). We cannot infer or
suspect a tendency towards secularisation from the presence of
reasoned argumentation or on the basis of general expectations
concerning the direction in which ‘rational’ thinking should tend.
We must ask in each case where, if anywhere, the evidence allows
for a distinction between rationality and irrationality66 to be drawn
and what precisely can be seen to fall on either side of the divide.
It is only on the basis of such particular, context-sensitive and
bottom-up analyses that patterns (of whatever breadth) can even-
tually be argued for and discussed.

65
Compare the argument in Gagné (2013) 238 n.160 against Versnel (2011) 201–12 that
Solon’s competing appeals, within the same hymn, to retributive theodicy (fr. 13.1–32
W) and the arbitrary forces of fate or chance (33–70 W) do not peacefully coexist but are
pointedly opposed to one another. Interestingly, Plutarch deleted lines 267–73 of the
Works and Days ‘on the grounds that they were unworthy of Hesiod’s judgement
concerning justice and injustice’ (so Proc. Σ Op. 270–3 Pertusi). We will see in
Ch. 2.2 that these lines do connect profoundly with Hesiod’s reflections elsewhere
about divine dispensations.
66
Or, if and when these Greek words are not only used but also contrasted with one another
(as at Pl. Gorg. 523a), between ‘logos’ and ‘mythos’ (cf. n.19 above).

35
 Rationality and Irrationality, Philosophy and Religion
1.3 Philosophy and Religion
Our reconsideration in this book of the relation between so-called
‘rational’ and ‘irrational’ attitudes embodies also a reconsidera-
tion, within the particular remit of this study, of the expectation
that ‘philosophy’ and ‘religion’ should come into conflict and of
the ways in which they can be seen to do so.
In his City of God (6.5), Augustine discusses a trichotomy,
expressed by the famous first-century BCE Roman scholar
Varro, between three different types of theology: first, a ‘mythical’
or ‘fantastic’ theology deployed by the poets, second, a ‘natural’
theology deployed by the philosophers and third, a ‘civic’ theol-
ogy deployed by the city and its religious officials in the admin-
istration of state cults. Varro concludes: ‘The first theology is
especially adapted (accommodata) to the theatre, the second to
the world, the third to the city.’67 Few historians today would
accept that the Greeks in general operated with a clear and widely
recognised division between poetic and civic gods.68 By contrast,
the Varronian idea of ‘natural theology’ – whatever its original
intentions – has long offered a theoretical framework which was
welcome and attractive for scholars who saw the Greeks’ philo-
sophical theology as operating on a qualitatively different plane to
traditional and popular religious attitudes and practices. Scholars
have repeatedly given voice to the perception that philosophical
theology always or at least very predominantly related to such
traditional attitudes and practices – if it related to them at all – in
the manner of explicit polemic or implicit dismissal. The inter-
pretative model of ‘natural theology’, in short, continues to exert
a real and deep-seated influence on our expectations concerning
the orientation of early Greek philosophical reflections about the

67
Augustine presents Varro as distinguishing competing theological world views and
competing gods, and as favouring ‘natural theology’. But the scheme could rather
delineate different and severally legitimate theological registers through which our
relation to the gods can be articulated and understood (cf. Most (2007) 276–7). This is
what Posidonius (Aët. 1.6.9) seems to have done when he demarcated three forms
through which reverence concerning the gods has been set out for us (διὰ τριῶν . . .
εἰδῶν): of nature (taught by philosophers), of poetic myth and of civic laws and customs
(nomoi).
68
Mikalson (e.g. 2010 esp. 16–19) is an exception; for some objections, see Tor
(2012).

36
 1.3 Philosophy and Religion
gods.69 As an interpretative category, ‘natural theology’ is also
closely connected to the ideal of an independent, ‘rational’ human
inquiry, which proceeds without any appeal to divine revelation or
intervention.70
We find in Plato a distinct but instructively related representa-
tion of early Greek philosophy, which was to exercise enormous
influence on later conceptions of the so-called ‘pre-Socratics’.
Plato’s Socrates, in his intellectual autobiography in the Phaedo,
portrays the ‘inquiry into nature’ (peri physeôs historia), which
preceded his own novel philosophical turns, as a search for non-
teleological, mechanistic and material explanations for the con-
stitution and character of all things. A recitation from Anaxagoras’
book (the only earlier thinker mentioned by name), who made
Mind a central principle of his philosophy, raised hopes, which
were quickly dashed: Anaxagoras failed to use his Mind and did
not in fact advance teleological explanations for natural phenom-
ena (Phd. 96a5–99d2). Aristotle’s favoured terms for his philoso-
phical predecessors are ‘physiologoi’ and ‘physikoi’, and his own
canonical account of the beginnings of natural philosophy in
Metaphysics Alpha is clearly influenced by Plato’s Phaedo.71
The interesting historical question for Aristotle here is not in the
first instance one about divine intelligence or teleology. For
Aristotle, Anaxagoras’ value lay in the progress which he made
towards demarcating the material and efficient causes from one

69
Vlastos (1970b) is a classic statement of this approach. Jaeger (1947) also begins from
and operates within Varro’s framework of ‘natural theology’ (1–4). The book-length
accounts of ancient Greek philosophy and religion by Gerson (1994) and Drozdek
(2007) adopt a broadly Varronian approach (Gerson features ‘natural theology’ in his
title). On early Greek philosophy in particular, see further e.g. Curd (2013a); (2015);
Gregory (2013); Kahn (2013); Whitmarsh (2015) 52–68; also (with some distinctive
modulations) Graham (2013b).
70
See e.g. Granger (2013b) 165, who labels the earliest Greek philosophers ‘natural
theologians’ on the grounds that ‘they call upon no Muse and depend on no divine
revelation or sacred text’. Granger (2007, 2008, 2010, 2013a, 2013b) works within the
interpretative tradition of natural theology, although he also cautions that the early
philosophers remained influenced by theological assumptions in the religious tradition,
which they nonetheless always sought to criticise, naturalise and purify.
71
On this canonical account, see the chapters in Steel (2012). On Aristotle’s discussion of
the early natural philosophers in Metaph. Α.3–4, see esp. Barney (2012) and Betegh
(2012). The latter examines Aristotle’s account of Anaxagoras and his inheritance from
Plato’s Phaedo. For echoes of the Phaedo in Metaphysics Alpha, cf. also Menn (2012),
211–16; Hülsz (2013) 188 n.3.

37
 Rationality and Irrationality, Philosophy and Religion
another, a breakthrough which required the insight that matter
could not organise itself. Nonetheless, the way in which
Aristotle articulates this breakthrough echoes the Phaedo’s narra-
tive. It was unsatisfactory to commit the good and beautiful state
of things to spontaneity and chance (τῷ αὐτομάτῳ καὶ τύχῃ); and
so, ‘when someone said that mind is present in nature, just as it is
in animals, as the cause of the order and of the entire arrangement,
he seemed like a sober man by comparison with the random
statements of his predecessors’ (984b11–19). Aristotle echoes
Plato, moreover, in his complaint that Anaxagoras failed to make
proper explanatory use of this cosmic principle, merely wheeling
it out when at a loss for some other explanation (985a10–21).72
The implication of all this concerning still earlier thought is clear.
The pre-Anaxagorean physiologoi – the drunkards to Anaxagoras’
sobriety – failed to achieve even such an ultimately disappointing
recognition of the promise of a universal divine intelligence for the
inquiry into nature.73
In what was to become a very influential passage, then, Plato
had portrayed the ‘pre-Socratic’ philosophers as thinkers who
operated with strictly mechanistic and non-teleological cosmogo-
nies and cosmologies. To be sure, neither Plato nor Aristotle
suggests that the pre-Socratics denied the existence of gods in
general. Still, Plato’s narrative offers us a strikingly secularising
portrayal of the workings of early cosmology itself. On this
portrayal, the first philosophers completely excluded divine will
and intelligence from their accounts of natural phenomena.
Aristotle’s own secularising attitude towards the physiologoi is
indirectly reflected in the label which he uses for their own

72
Betegh (2012) 133–4, though, argues that the primary focus of Aristotle’s own criticism
is more that Anaxagoras appealed to Mind unsystematically and haphazardly than that
he always unduly diminished its causal role.
73
Aristotle allows that talk of love or desire in Hesiod, Parmenides and Empedocles could
convey related, if even more embryonic and implicit, preoccupations with efficient
causation and with the good and fine order of things, 984b22–985a10, cf. 988a14–17,
33–4. In de An. 1.2, Aristotle discusses earlier thinkers who, he says, made soul a first
principle. In the cases of Anaxagoras and Diogenes of Apollonia, Aristotle specifies that
their principle was one of both knowledge and motion (τό τε γινώσκειν καὶ τὸ κινεῖν, 405a
18, 23). In the cases of Thales, Heraclitus and Alcmaeon, he only explicitly relates the
soul to motion (405a19–21, 25–405b1), although concerning Heraclitus he adds ellipti-
cally that what is in motion is known by what is in motion (405a27–8).

38
 1.3 Philosophy and Religion
predecessors – the earlier and more obscure ‘theologoi’ – with
whom, at least within the domain of natural philosophy, he favour-
ably contrasts the physiologoi.74 Plato’s representation of
Socrates’ predecessors in the Phaedo differs from the conception
of them as ‘natural theologians’.75 And yet, the Platonic narrative
strikes some attractive notes for scholars who wish to follow
certain versions of the natural theology model. It fits well with
the idea that reflection on the divine played at most a subordinate
role in early Greek natural philosophy and was far removed from
preoccupations with the gods elsewhere in Greek culture. It also
gives comfort to the position that the early philosophers retained
only a radically attenuated and non-interventionist notion of god,
something closer in fact to a principle or law of nature.
We have already seen reason to believe that the Platonic
narrative is deeply problematic. When discussing the
Milesians in the previous section, we noted that, in fact, our
evidence for many of the early Greek philosophers undercuts
Plato’s suggestion that Socrates’ philosophical predecessors had
failed to consider or operate with a notion of intelligent divine
governance.76 Furthermore, what we highlighted and criticised
when we considered modern treatments of the Milesians was
one important illustration of the ‘natural theology’ approach to
early Greek philosophy. The standard claim, that the Milesians
restricted the divine to the austere cosmological functions of
their key principles and, by doing so, clashed directly and quite
generally with traditional religious attitudes, runs far ahead of
our evidence (if not in a different direction from it). That
particular interpretative claim draws its plausibility chiefly
from the implicit expectation that philosophical, natural theol-
ogy should operate on a separate and qualitatively different
plane to traditional religion.

74
E.g. Metaph. 1000a9–10; 1071b26–8; 1075b24–7. On these Aristotelian classifications,
see further Palmer (2000). On secularising tendencies in Aristotle and the Peripatetic
doxographic tradition more generally, see Mansfeld (2013).
75
See Heidel (1910) for a classic modern statement of the Platonic picture; cf. e.g. Naddaf
(2005) 163 (the pre-Socratics understood physis as ‘blind necessity (anankê), without
any recourse to intentional cause’) and Kahn (2013; but contrast Kahn (1979) 272).
76
Ch. 1.2, with n.36. Note the citations there of Plato’s own conflicting remarks in the
Philebus on the consensual views of earlier thinkers.

39
 Rationality and Irrationality, Philosophy and Religion
Here as before, it will be helpful to locate ourselves between
two extremes. At the opposite end to the natural theology
approach, it has been argued that, since Greek religion itself
comprised a variety of practices and ‘a way of referring to the
world’, rather than a particular body of beliefs, we can never
meaningfully speak of early philosophy as objecting to traditional
religion.77
The denial that we can speak of particular belief-components in
relation to Greek religious attitudes and practices is a radical
inference from certain important features of Greek (and Roman)
religion which, in a sufficiently moderate and flexible formulation,
can be described uncontroversially.78 Greek religion involved
organised cults but nothing like fixed dogma, authoritative scrip-
tures or an established Church. The Greeks routinely regulated
ritual and publicly scrutinised its performance, but they made no
attempt to prescribe particular theological interpretations of ritual
or to police religious beliefs in general. They bore a broadly loose
and relaxed attitude towards different ways of thinking about the
divine. Indeed, contrasting conceptions and representations of the
gods were ubiquitous and coexisted without tension.79
The resulting tendency among modern historians to marginalise
the historical and interpretative significance of religious beliefs
has, however, come under instructive criticism. Yes, the Greeks
did not themselves typically isolate beliefs about the gods as
a matter for religious anxiety or scrutiny. Nonetheless, if we – as
interpreters of Greek religion – wish to give plausible or some-
times even coherent accounts of certain religious discussions,
expressions, representations and practices, then we must register
the operation of certain fluctuating and indeterminate sets or
systems of beliefs which are reflected in those religious phenom-
ena or which best account for them. It is this principle which we

77
Boys-Stones (2009) 1–3 and passim.
78
The radical position is also advanced by Giordano-Zecharya (2005) and refuted by
Versnel (2011) 539–59. The latter shows, among other things, that the expressions
theous nomizein and theous hêgeisthai are standardly used with both cognitive (‘to
believe in gods’) and ritualistic (‘to worship gods’) force, with both readings often being
simultaneously possible.
79
On this point, see esp. Versnel (2011), see further Ch. 1.2; cf. Gould (1985) 7–8; Most
(2003) 301–4; Betegh (2006a) 627–8; Kearns (2006) 324; Kindt (2012) 19–25.

40
 1.3 Philosophy and Religion
see at work in, for example, Parker’s and Versnel’s contentions
that the continued practice and perceived efficacy of such activ-
ities as prayer, divination and reciprocal gift-giving are intelligible
only in relation to widespread beliefs in the existence of interven-
tionist and communicative gods,80 in Harrison’s explorations of
the textual articulations, the applications, the sustainability and the
explanatory significance of different and not infrequently compet-
ing propositions about the gods,81 or in Gould’s well-known con-
ception of Greek religion as a system concerned prominently with
framing lived experience and rendering it explicable.82
As Harrison puts it: ‘To seek to describe Greek religion by
means of a stark opposition of ritual and dogma is little more
than to offer a choice of two caricatures.’83
The overall situation is more complex and less tidy than is
suggested either by the ‘natural theology’ model or by the catego-
rical denial that early philosophical theology ever clashed with
traditional religious attitudes.
Xenophanes (working in the late sixth and early fifth centuries
BCE) is one important thinker whom later tradition identified as
a philosophical ancestor and who exposes and questions some
central traditional religious beliefs and even (in divination) prac-
tices. Indeed, Xenophanes expressly circumscribes a category of
belief (dokos, DK21 B34.4) and uses this category to refer to
certain religious attitudes which he presents as both misguided
and universal (‘but the mortals believe (dokeousi) that gods are
born etc . . . ’, DK21 B14; note A12, where the affirmation of

80
Parker (2011) 1–39; Versnel (2011) esp. 552–3. See also Osborne (2016) on the
theological attitudes implicit in sacrificial regulations and, on the interrelations of
beliefs and ritual practices, Kearns (2006) 322–4 and Kowalzig (2007) 1–3 (discussing
Plato’s reflections at Lg. 10.887c-e on how the belief in the existence of gods underlies,
and is strengthened by, ritual activities). See further the discussions in this book of, for
example, divination (Ch. 3.1) and mystery initiations (Ch. 5.3).
81
Harrison (2000) esp. 18–22; (2006) esp. 135–6; (2007) esp. 376, 380–4; (2015b).
82
Gould (1985) esp. 4–5, 14, 32–3. Note that none of the interpretative projects listed here
require us to render the mental experiences of individual religious practitioners the
object of our inquiry (on this point, cf. Feeney (1998) 15, 30–1). But see Harrison
(2015a) on the upsurge in studies which strive to do just that. See, for example, the
universalising cognitive approach pursued by Ustinova (2009), for some reservations
about which see Tor (2010). For more general reservations about some cognitive
approaches, see Kindt (2012) 44.
83
Harrison (2000) 20.

41
 Rationality and Irrationality, Philosophy and Religion
divine births is branded impious).84 We can go even further. What
Xenophanes targets for questioning is a pervasive system of inter-
related beliefs, which comprises the mutually reinforcing biologi-
cal, visual, cognitive and moral views that the gods are
fundamentally like us in how they come to be, what they look
like, how they think and how they behave and misbehave (DK21
B10–16, B23). In Chapter 3, we explore at length Xenophanes’
polemic against traditional models of divine disclosure.
When we look to Classical Athens, moreover, we can point to
some real evidence of tension and confrontation. Here too, though,
we must be careful. Not all of our evidence is as reliable as it
appears at first sight. In particular, while some later biographical
sources report that a number of intellectual figures in fifth-century
Athens (including Anaxagoras, Protagoras and Euripides) were
subjected to legal and other persecutions, historians have raised
substantial and, in some cases, compelling challenges against
these claims. We hear, for example, that the Athenians wished to
burn Protagoras’ books because of his agnosticism (Timon apud
Sextus Empiricus, M 9.57) – or actually did so (Diogenes Laertius,
9.52) – and that they condemned him to death and he was forced to
flee the city (Sextus Empiricus, M 9.56). But we can scarcely
square these later claims with Socrates’ assertion to Anytus in
Plato’s Meno (91d–e) that Protagoras consistently enjoyed an
excellent reputation (εὐδοκιμῶν οὐδὲν πέπαυται), both in Athens
and all over Greece, throughout his 40-year career as well as after
his death.85 In his Pericles (32.2), Plutarch writes that Diopeithes,
in an effort to direct suspicion against Pericles through his asso-
ciation with Anaxagoras, introduced a decree which provided for
state prosecution (eisangelia) against ‘those who do not

84
Xenophanes does not thereby express a dismissive view of ‘beliefs’ as such. In B35, he
appears to present his own views as instances of belief. See further Bryan (2012) 6–57
and below: Ch. 3.3–5.
85
See further Dover (1988), Wallace (1994) and Gagné (2009). Among other things, they
highlight internal inconsistencies in our reports, the influence exerted by the lionised
model of Socrates on later ideals of the civic intellectual, and the striking readiness of
ancient biographers to fabricate and elaborate stories, sometimes even on the basis of
comedic sources (e.g. Satyrus – just after asserting that Euripides was prosecuted for
impiety – affirms that he was attacked by women during the Thesmophoria, clearly
treating Aristophanes’ Thesmophoriazusae as the record of a historical event: vit. Eur.
col.10).

42
 1.3 Philosophy and Religion
acknowledge divine matters (τοὺς τὰ θεῖα μὴ νομίζοντας) or teach
accounts (λόγους) about celestial objects’. We certainly cannot
exclude that this (or some such) decree was passed in the 430s,
although the failure of any other source to mention it, even when
doing so would have been highly apposite, should give us pause.86
At any rate, we have no evidence of prosecutions under
Diopeithes’ decree and foreign intellectuals did not hesitate to
visit Athens during the latter decades of the fifth century.
We do not, however, need to rely on such problematic evidence
in order to recognise that there did emerge in Classical Athens
a strain of suspicion against philosophical speculation, especially
in connection with its perceived religious implications. In 399
BCE, Socrates was accused of – and eventually executed for –
corrupting the youth, refusing to acknowledge the gods acknowl-
edged by the city (θεούς . . . οὐ νομίζοντα) and introducing new and
different divinities (δαιμόνια καινά, Plato, Apol. 24b; Xenophon,
Mem. 1.1.1; Diogenes Laertius, 2.40). Socrates’ accusers may
have been spurred in part by political motivations, but the fact
that they successfully pursued such a religious charge is itself
instructive.87 In Plato’s Apology, Socrates worries that the same
people who credit the rumours that he studies the things in the sky
and under the earth (like the natural philosophers) and makes the
weaker argument stronger (like the sophists) also ‘believe that
those who pursue such inquiries do not even acknowledge gods’
(18b–c). Later on, he remarks similarly that his detractors have
simply charged him with those shopworn stereotypes, among them
‘not acknowledging gods’ (θεοὺς μὴ νομίζειν), which are readily
attached to all philosophers (τὰ κατὰ πάντων τῶν φιλοσοφούντων
πρόχειρα, 23d). With these remarks, Socrates scarcely suggests
that the natural philosophers have merited this reputation for

86
E.g. Σ Aristoph. Av. 988; Σ Eq. 1085; Σ Vesp. 380 Dübner (these scholia collect
information about the fervent Diopeithes in religious contexts); Ephor. BNJ 70 F196 =
Diod. Sic. 12.39.2; X. Mem. 4.7.6. We may add Plato’s Apology (see anon). For
scepticism concerning the historicity of Diopeithes’ decree and speculations about
likely origins for the report in Plutarch, see further Dover (1988) 146–7, Wallace (1994)
137–8.
87
See Ober (2010) for the specificity of Socrates’ trial to its very particular historical and
political context in an Athens that was reeling from the Peloponnesian war, the bloody
rule of the Thirty and the tumultuous restoration of democracy in 403 BCE.

43
 Rationality and Irrationality, Philosophy and Religion
a general denial of gods, but he does bear witness to the existence
of the stereotype. Later on, when Meletus claims that Socrates
does not accept the divinity of Sun and Moon but says that the
former is a rock and the latter earth, Socrates retorts that Meletus
appears to have confused him with Anaxagoras (26d).88 Socrates
also suggests that his caricaturised portrayal in Aristophanes’
comedy had contributed to popular misconceptions of him (19c).
And indeed, Aristophanes’ Clouds (first staged in 423 BCE) not
only presents Socrates as a crackpot natural philosopher but also
makes the same connections, to which Socrates alludes rather
critically in Plato’s Apology, between the cosmological specula-
tions of natural philosophers and general denials of the gods. In the
play, Socrates will not recognise any gods other than the Clouds
(365) – he later adds Chaos and Tongue (423–4) – and leaves no
room for anything that could be called Zeus (οὐδ’ ἐστὶ Ζεύς, 367).
His new disciple, Strepsiades, promises that he will no longer
sacrifice to any other gods, or even speak with one if he happened
to bump into them (425–6). A chorus in Euripides draws similar
connections, aligning the person who acknowledges god (θεὸν
ἡ[γε]ῖσθαι) with the one who disavows ‘the crooked deceits of
those who talk about the heavens’ (μετεωρολόγων . . . ἀπάτας,
TrGF 5.913).
Whatever the intentions of early cosmologists, then, it was at
least possible to see in their distinctively demythologised and non-
anthropomorphic accounts of natural phenomena a global threat to
traditional religious beliefs and worship. This is the case, even if
this perception was found particularly among people who did not
suffer from excessive familiarity with the strange intellectual
88
Socrates does not thereby suggest that Anaxagoras was an atheist. In his last dialogue,
the Laws, Plato sketches a complex position which does combine a naturalistic and non-
teleological cosmogony and cosmology with the atheistic thesis that the gods were
invented as a device for social and legal control (10.889a–890a). Sedley (2013a, 2013b)
interestingly but speculatively argues that what Plato describes is not his own concoc-
tion of different positions (this is the common view, e.g. Mayhew (2008) 76–89) but an
actual underground and anonymous movement of atheists in late fifth- and early fourth-
century Athens. An atheistic reduction of the gods to a device for social control was
expressed in the famous Sisyphus fragment (S.E. M 9.54), albeit through the mouth of
the villainous Sisyphus (Aët. 1.7.2). It is much less clear that Prodicus’ theory, that
humans first formed their notion of the gods by deifying such beneficial things as corn
and water (S.E. M 9.17–18), amounts to a denial of the existence of gods; see further
Sedley (2013b) 141.

44
 1.3 Philosophy and Religion
pursuits of which they were suspicious (as Plato’s Socrates sug-
gests), and even if Aristophanes and Euripides play on and exag-
gerate these suspicions rather than substantiate them. Be that as it
may, for us to privilege as paradigmatic such hostile responses to
philosophy, and whatever legal persecutions took place in the very
specific political and intellectual climate in Athens around the late
fifth and early fourth centuries, would be to distort what was
overall a relaxed relationship in pre-Christian antiquity between
philosophical speculation and the religious life of the cities and
their inhabitants.89
Xenophanes, by his own account, spent almost seven decades
travelling from city to city throughout Greece, flogging his
wisdom and his views (DK21 B8, B45) as a thing of great civic
value (B2).90 Xenophanes, it appears, was welcome wherever he
went. He gives no hint of hesitation on his part or of blowback
against his critiques of traditional religious beliefs. Indeed, the
attacks against Socrates were a dramatic and exceptional event
even within the spatial and chronological confines of late fifth- and
early fourth-century Athens. Protagoras staked an agnostic
position (DK80 B4) and, to judge by Plato’s near-contemporary
testimony, this did not harm his appeal and reputation in
Athens. Year after year, moreover, the Athenians attended and
celebrated productions of Euripides’ plays, which featured not
only expressions of piety, but also such radical theological chal-
lenges as Bellerophon’s derisive assertion of atheism: ‘Does
someone say that there are indeed gods in heaven? There are not,
there are not, if a man is willing not to give credence to an ancient
account like a fool’ (TrGF 5.286).91 According to Xenophon,
Aristodemus the Small, with whom Socrates argued about this,
openly and publicly denied the reality of divine providence and,
therefore, avoided and even mocked the practice of sacrifice and
divination (Mem. 1.4.2–18). If we look towards later periods, the
Epicureans openly rejected the existence of providential gods who
may respond to prayer and sacrifice (Sent. 1; Ep. ad Menoec.
123–4), although they did not object to ritual practice itself
89
See similarly Most (2003) 305–7; Betegh (2006a) 625. 90 See further Ch. 3.5.
91
Compare the scathing denial of theodicy in Eur. TrGF 5.506 and contrast the assertion of
both theism and theodicy in (Eur.?) TrGF 2.624.

45
 Rationality and Irrationality, Philosophy and Religion
(Philodemus, Piet. 737–51; Lucretius, DRN 6.68–79). The Stoic
Zeno faced no persecution when he denied the sanctity and appro-
priateness of temples (SVF 1.264). The Academic and Pyrrhonian
sceptics developed a broad battery of arguments that questioned
the existence and nature of the gods from numerous theological
perspectives (Cicero, ND 3.29–52; Sextus Empiricus, PH 3.2–12;
M 9.137–90).92
Nonetheless, we should also not exaggerate the general toler-
ance of the Greeks by imagining that the public airing of radical
theological and religious views carried no social risk and never
faced what Parker calls the informal ‘social censorship of
speech’.93 The evidence which we reviewed from Plato,
Aristophanes and Euripides demonstrates this social risk, and the
charges brought against Socrates were its highly extraordinary
limiting case. Diogenes Laertius records the anecdote that, when
the Megarian philosopher Stilpo was asked whether the gods felt
gratitude for prayers and adorations, he replied that such questions
should not be discussed in the street but in private (2.117).94
Equally, however, there is a great gap between recognising the
reality of such social risk (at least at some junctures of Greek
history) and maintaining that philosophical theology was always
or largely thought to clash in a global and general way with
traditional religious attitudes and practices. Our sources do not
bridge this gap. Indeed, the absence of any archaic evidence
comparable to the Athenian evidence discussed here should give
us pause.
The Greeks’ general tolerance towards theological speculations
and criticisms reflects something important about Greek religion.
Quite apart from philosophical theology, Greek religion presents
us with ‘a jostling mass of competing beliefs and values and
interpretations and uncertainties’.95 As we noted above, while
ritual practice was regulated and publicly scrutinised, no attempt
was made to regulate its interpretations or religious beliefs quite
generally. This theological flexibility and the Greeks’ general
comfort with contrasting conceptions and representations of the
92
On the religious provocations perpetrated by (or ascribed to) the Cynics, see Goulet-Cazé
(1996).
93
Parker (2011) 38. 94 Cf. also S.E. PH 3.2; M 9.49. 95 Parker (1997) 148.

46
 1.3 Philosophy and Religion
gods reflect in part a widespread recognition of the ultimate
unknowability of the divine.96 It is hardly clear precisely what
aspects of the flexible ‘jostling mass’ which Greek polytheism
comprises must be excluded by early inquiry into nature. Indeed,
insofar as the project of early natural philosophy was to explain the
emergence and current structure of the cosmos and of the living
beings which inhabit it (a set which can easily include gods as well
as humans and animals, as we saw for example in the case of
Anaximenes in Chapter 1.2), there is no obvious reason that this
particular project should encroach on, or clash with, religious
ideas of cultural and political importance about the involvement
of the gods in human civic institutions and endeavours or in the
affairs of individual worshippers. Furthermore, once we stop
thinking of ‘religion’ (a word without parallel in ancient Greek)
as a unified whole, then what philosophical criticisms of particular
attitudes, interpretations or (more rarely) practices we do find
convey not an external, anti-religious outlook but internal reli-
gious modifications.97
The engagements of philosophers with traditional or non-
philosophical religious attitudes are by no means limited to criti-
cisms and, even when the philosophers do criticise, their critiques
do not always involve a traceless rejection of what they criticise.
We will explore in later chapters two extended test cases by
examining how Xenophanes (Chapters 3 and 6.1) and
Parmenides (Chapters 5 and 6.1) engaged with traditional and
current ideas of divinity, poetic inspiration and divination as well
as (in Parmenides’ case) mystery initiation and metempsychosis.
We will find in those thinkers complex and often inextricable
combinations of criticism and creative appropriations.
Xenophanes was both scandalised and influenced by traditional

96
Herodotus famously states that ‘all humans have an equal measure of knowledge about
[the gods]’ (2.3). For expressions of the unknowability of the gods, cf. e.g. Od. 23.81–2;
Xenoph. DK21 B34; Pind. fr. 61 M; Aesch. Ag. 160–6; Eur. Hel. 711–12, 1137–43; Or.
418; And. 1.139; Isoc. 1.50. Protagoras’ agnosticism (DK80 B4) takes this theological
attitude one step further. On divine unknowability as a factor in Greek religion, see
Parker (2011) 33; Harrison (2000) 191 (with nn.31, 33), 258 (with n.29).
97
For this point, cf. similarly Most (2003) 307–8; Harrison (2006) 132; (2007) 382;
Betegh (2006a) 626; Boys-Stones (2009) 3; Kindt (2015) 40; contrast Dodds
(1951) 191.

47
 Rationality and Irrationality, Philosophy and Religion
ideas of divine disclosure. Parmenides appropriates positively and
puts to novel philosophical use ideas of deification and metemp-
sychosis, as well as models of interactions between mortals and
personal gods adumbrated in the traditions of mystery initiations.
Our study of these test cases will support Lloyd’s seminal conten-
tion that, when analysing the emergence of the ‘philosophical’
from the ‘traditional’, we must never invoke ‘any talk of
a different mentality, a different logic, or a totally different con-
ceptual framework’.98 Most importantly, if the theses defended in
this book are on the right lines, then the pioneering strides in
philosophical epistemology taken by Hesiod, Xenophanes,
Parmenides and Empedocles represent an essentially theological
enterprise. As such, this enterprise was not just profoundly influ-
enced by certain aspects of Greek religion: it constituted part of the
‘jostling mass’ which was Greek religion.99

1.4 Rationality and Irrationality: Religious Belief

It has been observed that, to dismiss or marginalise any talk of
belief on the grounds that the Greeks did not operate with
a monotheistic and scriptural category of belief, is precisely to
over-privilege the monotheistic and scriptural perspective.100
Pausing on this insight briefly will enable us to get clearer on the
98
Lloyd (1979) 265 and passim. Heraclitus’ remarks on religious attitudes and practices
(DK22 B5, B14–15, B32, B92–3, B96) would offer another productive test case, but an
adequate discussion of these difficult texts would require a longer book with a different
thematic focus. Heraclitus has often been pegged as an enemy of traditional religion,
e.g. Kahn (1979) 263, 266–7; Burkert (1985) 309; Wildberg (2011) 215; Gregory
(2013) 112–13, 125. Other scholars argue that Heraclitus in fact criticises misguided
and misleading attitudes towards religious practices and their associated patterns of
thought rather than those practices and patterns of thought themselves; see C. Osborne
(1997) esp. on B5 and B15; Adomenas (1999); cf. Boys-Stones (2009) 4. Most (2013)
esp. 164–6 reasonably questions the notion that Heraclitus champions a systematic
approach to religion. He sees Heraclitus as deeply preoccupied with religious and
theological questions and as, in different fragments, accommodating, reinterpreting,
rejecting and raising open-ended questions about different Greek religious attitudes
and practices. In Tor (2016), I argue that in DK22 B93 Heraclitus both devotes involved
and creative theological reflection to Delphic Apollo and appropriates his divination as
a framework within which to understand the nature of his own dialectical practice.
99
Most (2003) 308, 319 characterises in similar terms Graeco-Roman philosophy as
a whole. Cf. Kindt (2015) 40: ‘we should not exclude . . . “personal” views from the
way in which we define and delimit religion’.
100
Harrison (2000) 20.

48
 1.4 Religious Belief
peculiar conditions which frame our peculiar habit of associating
the one word ‘rational’ with both the reasoning and the secular.
Tertullian, a Christian thinker writing in the second and third
centuries CE, lists in his On the Flesh of Christ such assertions as
‘the Son of God died’ and ‘having been buried, he rose again’, and
comments with a pointedly provocative paradox: ‘it is by all means
to be believed, because it is absurd . . . it is certain, because it is
impossible’ (prorsus credibile est, quia ineptum est . . . certum est,
quia impossibile, 5.4). It is only by accepting these claims, which
would of course appear as folly if subjected to the ordinary mea-
sures of wisdom ‘in the world’ (in saeculo), that one will become
truly wise (5.1). In a manifesto of Young Earth creationism pub-
lished in 1999, Kurt Wise writes: ‘ . . . if all the evidence in the
universe turns against creationism, I would be the first to admit it,
but I would still be a creationist because that is what the Word of
God seems to indicate.’101 These statements are provocative but not
incomprehensible to us only because we know to recognise ‘reli-
gious belief’ as a special and distinct kind of belief. We must not
depict a consensus where there is anything but one, but for many, at
least, as for Tertullian and Wise, religious belief is a question of
faith, where this means that it is self-consciously insulated from
what we ordinarily recognise as procedures of criticism and evalua-
tion and typically involves some concomitant form of self-
commitment and trust. It is neither possible nor necessary to convey
here how pervasive this notion of faith is in monotheistic Western
thought. We find it championed, challenged and reflected on in
thinkers as varied as, say, Martin Luther, who pronounces reason
‘the greatest enemy that faith has’, Pascal, who insists on the super-
iority of that religious conviction which is not acquired or preserved
through reasoning, Wittgenstein, whose reflections on what he calls
‘an extraordinary use of the word “believe”’ revolve throughout
around this same conception of a special, religious kind of belief
which, properly understood, does not pretend to be a matter of
reasonableness, and Alvin Plantinga, who seeks rather to demon-
strate the ‘rational acceptability of Christian belief’ and to challenge
the widespread view that such belief ‘lacks justification, rationality,

101
Wise (1999) 332; quoted by Dawkins (2007) 323.

49
 Rationality and Irrationality, Philosophy and Religion
or warrant’.102 Gregory of Nyssa (fourth century CE) nicely illus-
trates how the familiar monotheistic readiness to approach religious
belief as a special kind of belief is historically connected with
scripture when, before attempting to establish and elucidate the
doctrine of the Trinity on the basis of reasoning (logos), he
stipulates that, even should this venture fail, we will retain our
determined faith unchanged (τὴν πίστιν ἀμετάθετον ἕξομεν, ad
Ablab. 3.1.38.19–39.7).
We do not, to my knowledge, encounter such qualitative isola-
tions of ‘religious belief’ in the non-scriptural context of pre-
Christian Graeco-Roman sources. A partial parallel in Cicero (ND
3.5–6, 9–10) is instructive precisely for its fundamental dissimila-
rities. Cotta advances the claim that, qua Roman statesman and
priest (n.b. ND 2.168.3–4; 3.5.4–6, 3.6.1–5), he ought (debeo, 3.6.4)
to accept, on the authority of tradition and the ranking religious
officials, and independently of argument, the propriety of Roman
state cult and, correspondingly, certain rudimentary propositions
(e.g. deos esse, 3.9.11). Equally, however, Cotta himself stipulates
that such acceptance would be improper for Balbus, who, qua
philosopher (a te enim philosopho, 3.6.3–4), must rationally ground
his belief-system (in the domain of religion just as in any other).
What the Academic Cotta delineates, then, are the divergent com-
mitments of civic priest and philosophical theologian, not the
inherently distinct nature of ‘religious belief’ as a special kind of
belief. We find nothing like Gregory’s insulations in, say, Plato’s
theological reasoning in Republic 2–3 or Laws 10, a point which
Eusebius, for one, observes and criticises (PE 13.14.1.1–3.1).103

102
Martin Luther (1995) no.353; Pascal (2008) noo.110, 131; Wittgenstein (1966) 58–9,
and throughout 53–72; Plantinga (2000) vii, xiii. See Evans (2005) for a brief survey of
diverse attitudes, both historical and contemporary, to the relation between faith and
reason, ranging from the ‘rationalist’ to the ‘fideist’ extremes. On fideism, see also
Penelhum (1983).
103
For scripture as a new perspective in emerging competition with philosophical reason,
cf. Boys-Stones (2009) 21. On the Cicero passage, cf. Graver (2009) 128 (‘[t]his is not
the fideism of later ages’). Plato’s Timaeus 40d6-e4 offers another interesting but,
again, partial, significantly dissimilar and by no means straightforward parallel. First,
the tone of the passage, and of Timaeus’ assertion that, despite their failure to produce
likely or compelling proofs, we should still believe (πειστέον, 40d7) the theogonies
composed by the supposed offspring of the gods in question, is not at all obvious.
Timaeus (or Plato?) keeps a pointed distance from the poets’ claims for divine ancestry

50
 1.4 Religious Belief
In another treatise, Tertullian asks ‘what has Athens to do with
Jerusalem?’ and maintains that Christianity had supplanted philo-
sophy: ‘We have no need of curiosity after Christ, no need of
inquiry after the gospel’ (de Praesc. Haeret. 7.9–13). In the intro-
duction to his doctoral thesis, Karl Marx finds in Prometheus’
assertion ‘I hate all the gods’ (Aeschylus, Pr. 975) the universal
motto of philosophy itself – a cry for mortal liberation and inde-
pendence from religion which emblematises philosophy’s ‘abso-
lutely free heart’.104 There is, of course, a great deal between
heaven and earth other than Tertullian and Marx. But the strange
agreement between their programmatic remarks reflects well how
our historically contingent and peculiar insulation of faith and
religious authority from human reasoning reinforces our expecta-
tion that philosophical (‘rational’) and religious (‘irrational’)
thought should pull in opposite directions.
In polytheistic Greece, we encounter neither in philosophical
nor in other contexts such isolations of ‘religious beliefs’ as
special in kind, as independent from, and incommensurable with,
beliefs in other domains, and as insulated from and contrasted with
the activity of reasoning.105 We will find in Hesiod, Xenophanes,
Parmenides and Empedocles different models of divine disclosure
and divine intervention which, in different ways, facilitate or even
necessitate critical human reflection and human inquiry.

(ὡς ἔφασαν, 40d8; φασκόντων, 40e2), and one could plausibly find a measure of irony
in, among other things, his deflationary description of the bygone births of gods as mere
‘household matters’ for the poets (οἰκεῖα, 40e2; cf. the reserved που at 40d9). Plato is, at
any rate, perfectly willing elsewhere to have Socrates criticise such poetic authorities,
not least on matters which concern the familial affairs of their claimed divine ancestry
(e.g. Euth. 6a-c). Second, Timaeus’ remark is, on any reading, a circumscribed one. He
does not refer in a general way to beliefs about the gods or the divine, but specifically to
the issue of birth (τὴν γένεσιν, 40d6–7) and specifically in relation to only one particular
class of generated gods as distinct from another (note 40e4; see 41a3–5, 40a2–40d5).
Though both the Plato and Cicero passages offer fascinating early expressions of
thought-patterns which are not unrelated to later monotheistic fideist attitudes, neither
passage can be assimilated to such later attitudes and neither isolates religious belief as
a special, distinct kind of belief.
104
Marx (2008) 13–15.
105
Parker (2011) 1–13 interestingly identifies the operation of what he calls ‘evidences’
for ancient Greek religious beliefs, expressions (mostly casual) of which are found in
archaic and classical sources (e.g. the apparent efficacy of worship and divination,
fitting coincidences, the appearance of design).

51
 Rationality and Irrationality, Philosophy and Religion
1.5 Hesiod and Philosophy
I noted in the Introduction that this book pursues a self-
consciously circumscribed investigation. It does not aim to
exhaust what there is to be said about its guiding themes. On the
contrary, it is hoped that the interpretative approaches developed
here could offer starting points for other similar investigations (see
further Chapter 6.2). Nonetheless, in the light of the very fact that
the type of inquiries pursued here could be extended almost open-
endedly in relation to further materials, the chosen structure of the
subsequent chapters and especially the focused treatment of
Hesiod require a word of justification. This discussion will also
lead us helpfully into our examination of Hesiod’s epistemology in
the next chapter.
First, Hesiod is intrinsically interesting. He is, I submit, our
earliest extant thinker who critically isolates the question of the
conditions of human speculation, articulates a coherent framework
within which to consider it and integrates it into a broader con-
ception of the human condition. In short, we must not say that
‘Before [Xenophanes] there is no reflective discussion about
knowledge.’106 We have long thought of a questioning and pro-
blematising response to unreflective and overly blithe stances as
a hallmark of the pre-Socratic philosophers. When it comes to the
nature of divine disclosure and its implications for the epistemic
predicament of mortals – if not elsewhere as well – Hesiod man-
ifests these characteristics in full.107 Second, in order to analyse
Hesiod’s critique, we will need to consider how he relates to
a pervasive poetic and epistemological tradition, which presents
us both with related (if less sharply defined) worries and misgiv-
ings and with claims and assumptions which are vulnerable to
those worries and misgivings. Thus, our study of Hesiod (in
conjunction with our discussion of divination in Chapter 3.1)

106
Graham (2010) 132.
107
The view that Hesiod sets out to problematise unreflective stances concerning disclo-
sure does not require us to postulate an overly neat and naïve evolutionary model of
early Greek poetics, such that Hesiod is the first to put a question mark over poetic
truth-claims. Halliwell (2011) 19 rightly criticises such evolutionary models, noting
that the Homeric epics contain their own subtle negotiations of the relation between
truth and poetry.

52
 1.5 Hesiod and Philosophy
will best equip us with an understanding of the most dominant
epistemological attitudes and strategies in relation to which
Xenophanes, Parmenides and Empedocles also position them-
selves. Third, in critically isolating the question of epistemology
Hesiod clearly casts it in particular theological terms. For Hesiod,
to understand the epistemic limitations of mortals and the (proble-
matic) prospect of epistemic betterment is, manifestly, to under-
stand the nature of (i) the mortal, (ii) the divine and (iii) the
relation and the interactions between them. Xenophanes,
Parmenides and Empedocles, on my view, develop their own
radically divergent approaches to these same epistemological pro-
blems through this same general theological strategy. Insofar as
these (central) test cases go, I submit, it is not merely the case that
Hesiod first demarcates the problem of epistemology: Hesiod
crystallises the major contours of an emerging religious enterprise
of systematic reflection on the conditions of speculative inquiry.
We will find in Xenophanes, Parmenides and Empedocles illumi-
nating polemical responses to Hesiod, which are nonetheless
influenced by, and essentially continuous with, this epistemologi-
cal breakthrough. Indeed, it may be plausibly argued that
Xenophanes and Parmenides pitch their polemical responses as
reactions specifically to the way in which Hesiod first articulates
and introduces the Muses’ relation to him (Chapter 6.1). It is
a truism that Hesiod bore a formative relation to philosophical
cosmogony and cosmology. We will find in him a key initiator of
philosophical epistemology.108

108
It would be difficult to exaggerate and impossible even to sketch here the depth and
scope of Hesiod’s influence on later Greek thought and culture (Hdt. 2.53 and Herac.
DK22 B57 (‘Hesiod is the teacher of the most people’) are famous early testaments).
Koning (2010) is an extensive study; on philosophy, see 189–233. Koning (2010)
200–17 identifies Hesiod’s importance also for early epistemologists, but our
accounts bear little resemblance. Indeed, Koning sees Hesiod’s epistemological
stance simply as blithe reliance on divine disclosure (200). Most (1999a) explores
the foundational and fundamental influence exercised by Homer and Hesiod on the
style, subject matter and substance of early Greek philosophy. For Hesiod’s influence
on Ionian cosmogony and cosmology, see e.g. Cornford (1952) 187–201; Stokes
(1962) and (1963); Lloyd (1966) 192ff; KRS (1983) 34; Rowe (1983) 125, n.14;
Sedley (2007) 2ff; Koning (2010) 190–200. For Hesiod’s impact on Plato, see Boys-
Stones, Haubold et al. (2010). I refer to Aristotle presently. For later enlistments of
Hesiod as a philosophical ally, see e.g. SVF 1.167 (Zeno); 1.539 (Cleanthes); 2.1077,
p.316, 12–15 (Chrysippus); Plot. 4.3.14.78–80.

53
 Rationality and Irrationality, Philosophy and Religion
This is not the place to pursue a general discussion of the
peculiarities of the label ‘pre-Socratic’ or of the extent to which
we can or cannot isolate early Greek philosophy in general as
a distinctive intellectual phenomenon and demarcate it sharply
from other contemporary or later intellectual developments.109
It is, however, important that we highlight and dissociate our-
selves from two connotations, which have sometimes attached
to talk of ‘pre-Socratic philosophy’ and ‘early Greek philoso-
phy’. First, the interest of early philosophers in cosmology is
doubtlessly important and central, but these were breathtakingly
ambitious and holistic thinkers. Historically, scholars have often
over-privileged as primary or even exclusive the cosmological
projects of the pre-Socratics to the suppression of their
(typically undifferentiated) theological, epistemological, meth-
odological, psychological, ethical, political and ontological
preoccupations.110 Second, the thinkers who have been conven-
tionally classified as pre-Socratic philosophers certainly
engaged with other and earlier thinkers who have also been so
classified, and these engagements are doubtlessly important and
central. But we have sometimes tended to imagine that early
philosophers engaged with other, earlier philosophers in
a primary or consciously distinctive way, and we have often
marginalised their engagements with figures and attitudes not
conventionally classified as ‘philosophical’, as though the early
philosophers knew when they were and when they were not
responding to fellow philosophical colleagues.111 Xenophanes’
intellectual interlocutors were epic poets and diviners no less
than Ionian cosmologists. Parmenides was engaged in
a reflective dialogue with, for example, the ideas underpinning
the initiatory Gold Tablets no less than with Xenophanes. It is
a significant point that those pre-Socratic thinkers whom later
Greek writers identify as earlier philosophers did not already
distinguish themselves with that label.

109
For different views on these matters, see Long (1999) 5–13; Laks (2002); Sassi (2002)
esp. 71–3, 77; Naddaf (2006); Lloyd (2002); Gemelli Marciano (2002) esp. 112.
110
I follow Long (1999) 5–7, 11–12; cf. also Curd (2002) 121; Sassi (2002) 72.
111
For the observation, cf. Robbiano (2006) 16. For the historiographical tendency, see
e.g. Naddaf (2006) 165; Kahn (2013) 5.

54
 1.5 Hesiod and Philosophy
Notably, Heraclitus does not demarcate different kinds of rival
sages when he states: ‘Much-learning does not teach how to have
understanding; otherwise, it would have taught Hesiod and
Pythagoras, and again Xenophanes and Hecataeus’ (DK22
B40).112 We cannot, however, leave at that the question of
Hesiod’s relation to early philosophical inquiry. Antiquity already
offers diverse attitudes.113 Hesiod can be straightforwardly
included in or excluded from the business of philosophising.
Diogenes Laertius writes that Parmenides ‘philosophises through
verses, just like Hesiod and Xenophanes and Empedocles’
(φιλοσοφεῖ, 9.22.11–12).114 Sextus, by contrast, easily helps him-
self to the view that Hesiod is no competent judge of philosophical
matters (τῶν κατὰ φιλοσοφίαν, PH 3.123.1–2).115 Aristotle sees
Hesiod as sufficiently involved in explaining natural phenomena
to engage with him as a proto-cosmologist.116 Hesiod’s employ-
ment of ‘Eros’ as a moving and attracting force wins him recogni-
tion alongside Parmenides and Empedocles as someone who
gestures towards a final and efficient cause, albeit in an embryonic
and obscure manner.117 At another juncture, though, Aristotle
complains that ‘those who follow Hesiod and all the
theologians . . . talk over our heads’ and discourages serious
engagement with ‘those who deal in mythological sophistries’
(τῶν μυθικῶς σοφιζομένων, Metaph. Β.4 1000a5–19).118 So, we
should not overlook the qualification ‘in a way’ (πώς) in
Aristotle’s nonetheless striking recognition of mythological
speculative curiosity: ‘and the lover-of-myth is also in a way
112
Cf. Gemelli Marciano (2002) 112.
113
Koning (2010) 106, 191 oversimplifies when he portrays an ancient consensus that
Hesiod was a philosopher.
114
Cf. less unequivocally Hippol. Haer. 1.26.1–2 = DG 574–5 (citing Th. 108–39; note
περὶ φύσεως . . . λέγει); Σ Op. 130–1, 130a Pertusi.
115
Recall also Posidonius’ demarcation between philosophical ‘natural’ and poetic
‘mythical’ registers of theology, Aët. 1.6.9 (cf. n.67 above). Hesiod is mentioned as
an example of mythical theology at 1.6.14–15.
116
Metaph. A.8 989a8–12; Cael. 3.1 298b28; Phys. 4.1 208b29–33. Aristotle is motivated
in part by his cyclical view of human civilization and his notion that vestiges of
knowledge from the previous pre-cataclysmic period are preserved in the form of
myths, e.g. Metaph. Λ.8 1074b10–13, with Johansen (1999) 285–91; Palmer (2000)
192–203.
117
Metaph. Α.4 984b23–985a7; cf. Palmer (2000) 188.
118
See Lloyd (1999) 155; Johansen (1999) 289. Cf. Metaph. N.4 1091b4–10, with Ross
(1966) ad loc.

55
 Rationality and Irrationality, Philosophy and Religion
a lover-of-wisdom’ (καὶ ὁ φιλόμυθος φιλόσοφός πώς ἐστιν, Metaph.
Α.2 982b18–19).119 Other ancient passages accentuate what is
anyway implicit in the rich reception of Hesiod in later Greek
philosophy, namely that Hesiod is good to think with for philoso-
phical inquirers. In one story, for example, the schoolboy Epicurus
embarks on his philosophical career after Hesiod’s assertion, ‘first
of all Chasm came to be’ (Th. 116), sparks his curiosity and makes
him wonder what, in that case, Hesiod thinks Chasm came to be
from (Sextus Empiricus, M 10.18–19; Diogenes Laertius, 10.2).120
Hesiod’s complex and systematic theogony is simultaneously
a cosmogony and a cosmology. His system of developing (familial)
relations between different divine and (simultaneously) natural,
societal and psychological powers amounts to a universal explana-
tory account of the world’s principles of organisation and
operation.121 On those sorts of grounds, scholars have sometimes
pronounced Hesiod philosophical or semi-philosophical.122 If most
scholars nonetheless remain disinclined to consider Hesiod
a philosopher, this seems to do above all with the irremediably
mythological nature of his theogonic-cosmogonic accounts123 and
his apparent disinterest in disciplining those accounts by explicit or
implicit argumentation.124 Long disqualifies Hesiod from the com-
pany of the early philosophers also for not being ‘critical and
unconventional’ and for not aiming to have an ‘educationally pro-
vocative’ and ‘transformative’ impact on his audience.125 But those
charges are, on my view of Hesiod, unwarranted.
A different reason for not counting Hesiod a philosopher was
adduced by Rowe (1983), who ascribes to him what he labels the

119
Guthrie (1970) 250 highlights the qualification. On Aristotle’s complex attitude
towards the theologoi, see Palmer (2000).
120
See also e.g. Pl. Apol. 41a6–8; Tht. 155d (with Hes. Th. 265–6); cf. Crat. 406a3–5.
121
Following KRS (1983) 73; see also Clay (2003) e.g. 14–20.
122
E.g. Fränkel (1975b) 102–7, 515; Gigon (1945) 13, 22ff; Wade-Gery (1949) 81, 86;
more recently, Koning (2010) 233–5; cf. also Clay (2003) 2; see further Rowe (1983)
124; Kirk (1970) 238.
123
KRS (1983) 73; cf. Vlastos (1970b) 100–1; less subtly, Burnet (1930) 6. Some scholars
are open to the idea that myths at least manifest their own ‘rationality’, insofar as they
display their own speculative content and often highly abstract strategies for organising
and representing experience; see Burkert (1999); Sassi (2002) 65–6; KRS (1983)
loc. cit.
124
Long (1999) 12–13; Naddaf (2006) 166, n.20; cf. Lloyd (1979) 233.
125
Long (1999) loc. cit.

56
 1.5 Hesiod and Philosophy
‘multiple approaches’ phenomenon: Hesiod may juxtapose differ-
ent and sometimes seemingly or actually incompatible accounts or
explanations of the same thing without considering such combina-
tions problematic.126 Hesiod’s juxtaposition in the Works and
Days of the story of Prometheus and Pandora (42–105) and his
account of the succession of human races (106–201) is by far the
clearest example.127 Both episodes explain the deterioration of the
human condition, but they do so through logically incompatible
narratives (positing a two-stage human history and a gradual
decline respectively). Hesiod neither indicates a preference nor
attempts somehow to force these accounts into consistency.
Indeed, he issues the following programmatic remark when he
embarks on the Races episode: ‘But, if you like, I will relate128 to
you another account’ (ἕτερον . . . λόγον, Op. 106). The words
‘another account’ (or: ‘a different account’) highlight the juxtapo-
sition of the two episodes and, thereby, the absence of any attempt
to effect a logical reconciliation between them. This juxtaposition
and this programmatic statement could conceivably reflect episte-
mological modesty concerning the early events of mankind.
Significantly, however, the rhetorical conceit ‘if you like’ accords
the reader a hypothetical choice, which suggests that Hesiod takes
the accounts’ worth to reside in their suitability for, and impact on,
their audience (nominally, Perses).129 At this juncture, the two
narratives serve Hesiod’s purposes regardless of their incompat-
ibility (or ultimate truth-value).
The Theogony, though, affords nothing like such a clear or
central example. The fact that the Moirai are inconsistently born
of both Night (Th. 217) and of Themis and Zeus (Th. 904)130 is
a marginal and seemingly unique case, which strikes me as the
exception that proves the rule, throwing into sharp relief the over-
all coherence and consistency of Hesiod’s vast, complex and grand
theogonic-cosmogonic system. More difficult to assess is the
import of the fact that, throughout the Theogony, Hesiod (like
Homer), may represent the same divinity both as an
126
Recall Versnel’s ‘luxurious multiplicity’; see Ch. 1.2. 127 Rowe (1983) 132–3.
128
The precise force of ἐκκορυφώσω is unclear, see West (1978) ad loc.; Haubold (2010)
27, n.61; Canevaro (2015) 198. Rowe (1983) 133 rightly focuses on ἕτερον . . . λόγον.
129
Haubold (2010) 26. 130 Rowe (1983) 128.

57
 Rationality and Irrationality, Philosophy and Religion
anthropomorphised person and as what we would demarcate as
natural phenomena (Ouranos), impulses (Eros, Eris), sensations
(Pain, Hunger) or societal qualities (Lawlessness). So, for exam-
ple, Might both sits by Zeus’ throne (Th. 385–8) and is something
in which Zeus is preeminent (Th. 49).131 The perception that such
representations embody competing conceptions could well aggra-
vate our pre-existing inclination to consider anthropomorphism
intrinsically unphilosophical. But the unapologetic ubiquity and
unremarked juxtapositions of such dualities leave no doubt that
Hesiod and Homer themselves, at any rate, treat these representa-
tions as complementary reflections of the same phenomenon under
different aspects.132
What we should take from Rowe is the point that, prominently
in the Works and Days and possibly in the Theogony (although we
can cite only one isolated incident), Hesiod occasionally employs
logically incompatible narratives to express insights which are
themselves compatible and complementary. Importantly, how-
ever, throughout Hesiod’s corpus these complementary insights
themselves involve not merely prescriptive or moralising injunc-
tions, but also certain descriptive theological and anthropological
positions.133 In fact, the Prometheus-Pandora and Races narratives
cohere not only in extolling the virtues of work and justice respec-
tively, but also in grounding the status of work and justice as
inexorable necessities of the human condition by reference to
Zeus’ rule over the cosmos and its human inhabitants.134
What merits and excludes the label ‘philosophical’ is inevitably
a contested and elusive question. Ultimately, it is not in itself an
important question for our purposes. Far more significant is the
growing recognition that Hesiod warrants sympathetic considera-
tion as a serious, sophisticated and reflective thinker.135 Beyond
that, and beyond the general coherence of his theogonic-
cosmogonic, theological, anthropological and moral world view,
it is Hesiod’s epistemological self-awareness which is crucial for

131
See further Lloyd (1966) 199–202; Rowe (1983) 125–9.
132 133
Lloyd (1966) 202; Rowe (1983) 130. Contrast Rowe (1983) 134.
134
Following Clay (2003) 38, 126; cf. Canevaro (2015) 150–1.
135
See esp. Clay (2003); cf. Haubold (2010); Koning (2010) 190–1; Canevaro (2015);
Iribarren (forthcoming).

58
 1.5 Hesiod and Philosophy
us and which importantly relates him to subsequent philosophical
epistemology.136 If nothing else, Hesiod, I believe, shall have
earned his place as a key initiator of philosophical epistemology
if Lloyd is right that ‘[p]hilosophy and science can only begin
when a set of questions is substituted for a set of vaguely assumed
certainties.’137
Nonetheless, we can afford to be specific on two pertinent
respects in which Hesiod differs from Xenophanes, Parmenides
and Empedocles. First, Hesiod’s epistemology accommodates
nothing like critical or argumentative reasoning of the prominence
and sophistication of those later thinkers. Second, the epistemolo-
gical framework which Hesiod crystallises remains, on my view,
sufficiently indeterminate to accommodate ambivalent vacilla-
tions between competing elaborations of its fixed principles
(Chapter 2.4). By marked contrast, Xenophanes, Parmenides and
Empedocles all resolve this indeterminacy into definite epistemo-
logical positions. It is another matter, though, whether this latter
divergence must mark Hesiod as more or less philosophical than
his successors or is altogether orthogonal to that question.
I am not, in sum, concerned to reject the generic classification of
Hesiod as a ‘poet’ or to insist that he has been wrongly refused the
title of ‘philosopher’. I do not deny that we can draw broad
distinctions between philosophical and non-philosophical writers
in archaic Greece, either by retrojecting the classifications
expressed in the later Greek philosophical tradition or by import-
ing our own modern criteria of what counts as philosophy, even
though, on both accounts, such distinctions would have to be
flexible and messy at their edges, when it comes to figures like
Hesiod or Pherecydes.
Be that as it may, it is important that we remain critical of the
motivations and preoccupations which are commonly attributed
and denied to Hesiod on the basis of our generic classification of
him as a poet. The Muses’ address to Hesiod, which will be our
primary focus in Chapter 2, provides one important illustration of
this point. It is sometimes assumed, by interpreters of otherwise
136
Boys-Stones (2010) 32 observes Hesiod’s epistemological self-awareness; contrast
Hussey (1990) 19; Lesher (2008) 463; Graham (2010) 132.
137
Lloyd (1979) 266.

59
 Rationality and Irrationality, Philosophy and Religion
diverse views, that, in order to understand why the Muses address
Hesiod as they do in Theogony 26–8, we must explain why it
should be in Hesiod’s interest to have the Muses address him in
this way: ‘why should Hesiod choose to attribute a defect to his
own poetry?’138 It is no doubt important to consider how Hesiod
constructs his poetic authority at different junctures. But I see no
more reason to expect that Hesiod’s epistemological reflections
should be explicable in such reductively utilitarian terms than, say,
Xenophanes’ own disavowal of knowledge, which he likewise
expresses through poetic hexametric verses (DK21 B34;
Chapter 3.3). Hesiod knows nothing of an ‘ancient disagreement
between philosophy and poetry’ (Rep. 10.607b1-6).

138
Ledbetter (2003) 43, 60 (my emphases); cf. e.g. Murray (1981) 91; Bergren (1983) 82.

60
 2

H E S I O DI C E P I S T E M O L O G Y

Alongside the Homeric poems, Hesiod offers some of the earliest

surviving and most consequential writings of ancient Greece. Two
great poems have come down to us, the Theogony and the Works
and Days. The former treats principally of the births of gods and
the divine structure (physical as well as moral) of the cosmos and
its history from its beginnings and up to the current age of Zeus’
reign. The latter takes the form of a protreptic addressed to
Hesiod’s wayward brother Perses, and covers a broad range of
issues ranging from general ethical principles and the nature and
implications of Zeus’ just government of the world, to more
particular observations of proper measures and times in the con-
duct of one’s daily life and work.
The Theogony begins with the Muses on top of Mt. Helicon.
The Muses dance on soft feet down the mountain, singing of the
holy race of the immortal gods (Th. 1–21). On one occasion (αἵ νύ
ποθ’), they came upon Hesiod as he was herding his sheep at the
foot of the mountain and taught him beautiful song (καλὴν ἐδίδαξαν
ἀοιδήν, Th. 22–3). In the following lines, the Muses address Hesiod
in their first and last unambiguously direct speech (Th. 24–8):
τόνδε δέ με πρώτιστα θεαὶ πρὸς μῦθον ἔειπον, 24
Μοῦσαι Ὀλυμπιάδες, κοῦραι Διὸς αἰγιόχοιο· 25
ποιμένες ἄγραυλοι, κάκ’ ἐλέγχεα, γαστέρες οἶον,
ἴδμεν ψεύδεα πολλὰ λέγειν ἐτύμοισιν ὁμοῖα,
ἴδμεν δ’ εὖτ’ ἐθέλωμεν ἀληθέα γηρύσασθαι. 28

And this speech the goddesses first of all spoke to me

The Olympian Muses, daughters of Aegis-bearing Zeus:
‘Shepherds of the field, base, shameful things, mere bellies:
We know how to speak many falsehoods which are like verities,1
And we know, whenever we wish, how to utter truths.’

1
Or: ‘realities’.

61
 Hesiodic Epistemology
These words, which must serve as the starting and focal point for
any account of Hesiod’s epistemological preoccupations, have
been described without exaggeration as ‘one of the most enigmatic
statements about poetry to be found in Greek literature’.2
The Muses precisely do not explicate the implications of their
address. Its interpretation thus becomes a challenge of reading the
subtext.
We can usefully outline two primary schools of interpretation.
The ‘truths-only’ interpretation finds in the Muses’ address the
assurance that, while the Muses sometimes tell falsehoods to other
poets, Hesiod’s Theogony will consist purely of truths. The con-
trast between the Muses’ two types of discourse is traditionally
recast as anti-Homeric polemic: ‘Lügengesang ist homerisch,
Wahrsang hesiodeisch.’3 Other truths-only interpreters reject4 or
do not commit themselves to this specific identification.5
Before we turn to the second primary approach, we must clearly
distinguish (i) the truths-only interpretation, according to which
the Muses imply that they communicate only truths to Hesiod,
from (ii) the compatible but distinct view that the poet acquires
knowledge from the Muses (let us call this the ‘divine knowledge’
interpretation). The two claims are often conflated.6 But (i) with-
out (ii) comprises an entirely different understanding of the
epistemological nature of Hesiod’s relationship with the Muses
than (i) and (ii). The Muses make a pointed first-person knowledge
claim (‘we know . . . we know’), which contrasts emphatically with
the abuse they direct at their mortal addressee. We cannot take it,

2
Pucci (1977) 8.
3
Nietzsche (1922) 98; cf. e.g. Göttling (1843) 7; Paley (1861) ad loc.; Snell (1953) 138–9;
Heitsch (1966) 198–9; Diller (1966) 690–1; Roloff (1970) 119, n.22; Murray (1981)
90–1; Bergren (1983) 82; Koning (2010) 304.
4
Cornford (1952) 104–5; West (1966) 162.
5
Jaeger (1947) 11; Pellikaan-Engel (1978) 11; Nagy (1990) 44–7; Hussey (1990) 37; Bowie
(1993) 22; Morgan (2000) 21; Katz and Volk (2000) 122–3; Most (2006) xxii-xxiii, n.7;
Torgerson (2006) 36; Granger (2007) 408–9; Ferella (2013) 44; Pucci (2009) 43; cf. also
Pucci (1977) who, however, argues that the doubleness of the Muses’ speech ‘creeps into
the text’ despite its author (esp. 1, 12, 25, 27, 31). Canevaro (2015) 43 expresses a truths-
only view but allows (n.41) that Hesiod is conscious that he could not guarantee the truth of
his account in the Theogony.
6
For casual ascriptions of knowledge-claims to Hesiod, see e.g. Walcot (1963) 6; Roloff
(1970) 120; Murray (1981) 91; Bergren (1983) 70–1; Stoddard (2004) 42, 61; Granger
(2008) 15; (2013b) 167; Ustinova (2009) 169.

62
 Hesiodic Epistemology
then, that Hesiod confuses (i) the prospect of receiving truths from
the Muses through divine favour with (ii) the prospect of acquiring
the same power and knowledge to which they lay claim. I will
argue in what follows that the divine knowledge interpretation is
nowhere a viable construal of Hesiodic or, for that matter, Homeric
epistemology. Homer and Hesiod deploy the vocabulary of knowl-
edge deliberately (ἴδμεν, Th. 27–8; ἴστε, Il. 2.485). They are careful
to dissociate themselves from claims for divine knowledge.
Neither resorts to such claims in constructing their poetic author-
ity. The mortal poet does not pretend to acquire the knowledge
which characterises the Muses.
The alternative to the truths-only interpretation is best defined
negatively as denying that Hesiod makes the categorical truth-
claim posited by the truths-only interpretation. We can then further
distinguish two different versions of the alternative view. One
version adds the positive claim that the Muses imply that the
Theogony does in fact involve falsehoods.7 Another version of
the alternative view – call it the ‘ambiguous’ interpretation – takes
the Muses to leave it uncertain whether the Theogony comprises
truths, falsehoods which are like verities, or some combination of
both.8
We will first survey the semantic range in epic poetry of the
constitutive terms of the Muses’ address (Chapter 2.1). We will
thus establish some important preliminary points about Hesiod’s
usage of these terms in Theogony 27–8 and defend the (traditional)
translation given above. Scholars too often discuss the Muses’
address in isolation from the Theogony as a whole. In
Chapter 2.2, we will consider these verses both in their immediate
context and against the broader background of Hesiod’s concep-
tion of mortal and divine and male and female, and in relation to
Homeric (including Hymnic) notions of poetic inspiration. I will
conclude that the Theogony decisively supports what I called the

7
Wilamowitz (1916) 473. A combination of this strong view with the divine knowledge
interpretation is neither incoherent nor uninstantiated: a knowledgeable Hesiod could
elect to include in his poetry falsehoods which are like verities. This seems to be the
upshot of Stroh (1976) 96–8.
8
Clay (1988) 328; (2003) 63; Ledbetter (2003) 40–4; Halliwell (2011) 18, n.37; cf. Wade-
Gery (1949) 86; Thalmann (1984) 151–2.

63
 Hesiodic Epistemology
‘ambiguous’ interpretation and develop (Chapter 2.3) a particular
elaboration of it. In Chapter 2.4, however, we will address
a divergent epistemological stance expressed at one juncture in
the Works and Days (646–62) and examine its implications for
our understanding of the Muses’ address and of Hesiod as
a theological and epistemological thinker.
Hesiod is the earliest surviving Greek author to name him-
self within his own text (Th. 22). But scholars have grown
suspicious of attempts to find in the depiction of ‘Hesiod’ and
his life-story a transparent autobiography and tend to view this
depiction as designed to serve certain rhetorical or didactic
purposes.9 At no juncture will we estimate somehow the
extent to which the narrating poet and his autobiography, as
constructed in our text, may or may not have corresponded to
the life of a historic personality.10 The question for us is how
the prominent persona of the mortal poet, as the text discloses
him to us, construes and represents his relation to the divine
and the epistemological ramifications of this relation. Again,
we need not insist that Hesiod’s encounter with the Muses
necessarily refers to the historic event of a perceived epiphany
(although it certainly could) nor, a fortiori, assume that in the
Muses’ address Hesiod transcribes verbatim what he believed
to be words once spoken to him at the foot of Mt. Helicon.11
The essential point is that, in his narration of his encounter
with the Muses, the mortal poet articulates his understanding
of what he construes and represents as an initial and
ongoing12 interaction with divine agents, which alone renders
his poetry possible.

9
E.g. Stoddard (2004) esp. 1–33; Nagy (1990) 47–82; Haubold (2010) 11 and passim;
contrast Wade-Gery (1949) 84–5; Fränkel (1975b) 94–6.
10
Cf. Haubold (2010) 22, n.45; Clay (2003) 3. I will, though, resist attempts to regard
Hesiod as ‘literary fiction’.
11
Dodds (1951) 117, for example, postulates a historic epiphany and offers parallels; cf.
Wade-Gery (1949) 85–7; Most (2006) xiii-xiv; contrast e.g. Bowie (1993) 20; Stoddard
(2004) 7, 67–8. Cf. West (1966) 158–61. On the ubiquity and immense diversity of
divine epiphanies throughout Greek (and Roman) antiquity, see Versnel (1987);
Petridou (2015).
12
Hesiod continues to invoke the Muses (‘tell me’, Th. 114–15).

64
 2.1 Semantic Overview: pseudea, alêthea, etyma, homoia
2.1 Semantic Overview: pseudea, alêthea, etyma, homoia
The Muses’ address turns on four pivotal concepts: pseudea (‘fal-
sehoods’), alêthea (‘truths’), etyma (‘verities’ or ‘realities’) and
homoia (‘like’ or ‘equivalent to’). An overview of these terms will
enable us to outline the kinds of domains encompassed by the
Muses’ address and to acquire a first insight into the interpretative
predicament in which its open-endedness leaves its audience.
It will also enable us to dispel contentions that this epic vocabulary
was not used, like our term ‘true’ (or, mutatis mutandis, ‘false’), to
pick out statements which state ‘p’ when it is the case that p.13
The term pseudos and its cognates standardly signify that an
account of a present, past or future state-of-affairs is false.
In Odyssey 14.379–87, for example, Eumaeus recounts how an
Aetolian deceived him (ἐξήπαφε, 379) by stating that he had seen
Odysseus in Crete and that the latter will arrive later that year and
enjoins Odysseus not to curry favour through such falsehoods
(ψεύδεσσι, 387).14 The Iliad employs this vocabulary to identify
false prophecies, i.e. divine deception (Il. 21.276; Il. 2.81; cf. Il.
2.348–9; also h.Hom. 4.562–3). Notably, we also find it predicated
of genealogical and even theogonic accounts. In a Homeric hymn,
the poet denounces competing traditions surrounding the location of
Dionysus’ birth (Drakanon, Ikaros, etc.) as falsehoods (σε λέγουσι
γενέσθαι | ψευδόμενοι, 5–6), insisting instead on Mt. Nysa.15
The closest parallel to the Muses’ address is Homer’s concluding
remark at Odyssey 19.203 on Odysseus’ fabrication of a Cretan
autobiography (165–202): ‘Speaking many falsehoods he made
them like verities’ (ἴσκε ψεύδεα πολλὰ λέγων ἐτύμοισιν ὁμοῖα).
Odysseus’ fabrications here clearly involve biographical and gen-
ealogical falsehoods (ἐμὸν γόνον, 166). The term pseudea can also
possess a normative and evaluative force. In Iliad 4.404–5, for
example, we have an assessment of martial prowess and, perhaps,

13
Morgan (2000) 21–2, for example, claims that early Greek poetry recognises no criteria
of truth and falsehood except insofar as ‘true’ meant ‘valid’ and ‘memorable’ (mis-
representing Cole (1983), as we shall presently see).
14
Cf. similarly Od. 14.124–5, 365; Od. 3.20, 328 (with Od. 3.16, 79–101; Od. 4.316–31);
h.Hom. 4.369 (with 379).
15
Cf. Il. 5.635–6 on Sarpedon’s descent (ψευδόμενοι δέ σέ φασι Διὸς γόνον αἰγιόχοιο |
εἶναι).

65
 Hesiodic Epistemology
moral character: Stheneleus rejects as lies (μὴ ψεύδε’ ἐπιστάμενος
σάφα εἰπεῖν) Agamemnon’s contention that Diomedes is inferior to
his father in battle (370–400).16 We also find deceitful statements of
favourable intentions, as when the disguised Odysseus tells
Eumaeus that a Phoenician con man had convinced him to travel
to Libya under false pretences (ψεύδεα βουλεύσας, Od. 14.296).17
Iris’ communication, which Priam insists was not false (ψεῦδος, Il.
24.222), included both descriptive predictions (182–7) and injunc-
tions in accordance with Zeus’ will (171–81).
In Iliad 4.235, Agamemnon expresses his conviction that Zeus
will not give his aid to lies or, perhaps, liars (ἐπὶ ψευδέσσι).
The point here is that the Trojans acted against the Greeks in
defiance of their oaths. Hesiod similarly employs this terminology
when delineating the divine penalties for mortal (Op. 282–5) and
divine (Th. 782–806, note 783) perjury. Correspondingly, the
genealogy in the Theogony of the gods Falsehoods – Pseudea –
underscores their status as a threatening power. The offspring of
Strife, they are siblings of, for example, Toil, Forgetfulness, Hunger,
Pains, Battles, Murders and Lawlessness (Th. 226–32).18 The kind,
just and harmless god Nereus is without-falsehood and true
(ἀψευδέα καὶ ἀληθέα, Th. 233–6). The calamitous Pandora, con-
versely, acquires from Hermes deceits (ψεύδεα), devious discourses
(λόγους) and a thievish character (Op. 78). The phrase ‘to give
a lying grace with one’s tongue’ (ψεύδεσθαι γλώσσης χάριν,
Op. 709) signifies making a false pretence of favour, something
which is closely associated with inflicting harm (κακὸν ἔρξαι,
Op. 708; cf. Thgn. 389–90). These passages are perfectly consistent
with the statement in Works and Days 788–9 (whatever the author-
ship of these lines) that the sixth day is excellent for bearing men,
since they will excel in deceits (ψεύδεα): like Odysseus, Hermes,
Pandora and the Muses, such men will be on the informed side of
the interaction between deceiver and deceived.

16
Again, when Agamemnon concedes at Il. 9.115 that Nestor’s description was not false
(ψεῦδος), he endorses both the factual aspects of the latter’s account of the feud
surrounding Briseïs and his allocation of blame (Il. 9.96–113); cf. Il. 23.576, with
570–85, 402–41.
17
False prophecies may overlap with this sense, as in Il. 2.348–9.
18
The consequences of perjury explain the relation to Oath (Th. 231–2).

66
 2.1 Semantic Overview: pseudea, alêthea, etyma, homoia
The term pseudea, then, activates a cluster of associations
centred on the notion of threat and deception. The Hesiodic (and
Homeric) usage of this vocabulary thus already excludes the idea
that, in their address to Hesiod, the Muses are merely delineating
what are, for the audience, two different but entirely unproble-
matic and equally valenced kinds or genres of poetic discourse.19
The term alêthês involves more, and has somewhat different
emphases, than our word ‘true’, as Cole (1983) has shown. In epic
poetry, the term refers almost exclusively to the informational
content of communication. It marks communication as free from
forgetfulness or inattentiveness (lêthê) and, therefore, as conscien-
tiously comprehensive and devoid of irrelevant or misleading
accretions.20 It is clear and entirely consistent with Cole’s analy-
sis, however, that, to qualify as ‘alêthês’, an account must exhibit
also what we would call the logical property of truth.21 When
Hector’s steward affirms that she speaks truths (ἀληθέα) when
she reports that Andromache is present at the City Walls, she
affirms, in primis, that Andromache is present at the City Walls
(Il. 6.383–9).22 The term also standardly highlights the veracity
and comprehensiveness of accounts of past events.23 References
to the future are typically found only insofar as speakers truly
report their favourable or hostile intentions,24 although a Homeric
poet describes quite generally the favourable conditions under
which the divine and divinatory Thriai speak truth (ἀληθείην) to
their consultants (h.Hom. 4.560–1). As with the epic vocabulary of
falsehood, this vocabulary of truth also extends to the didactic
sphere of normative prescription and ethical evaluation. Eurycleia
truly identifies for Odysseus the twelve of the fifty servant girls
19
The baggage and connotations of the term pseudea also invalidate the view that Th. 27
merely articulates something as comparatively safe and circumscribed as our modern
category of literary fiction and that Hesiod can be classified as such (Bowie (1993)
esp. 21, 37; Stoddard (2004) passim, esp. 40–2; Halliwell (2011) 14–15).
20
Cole (1983) 7–21; cf. Palmer (2009) 89, n.107; Clay (1988) 327; (2003) 61, n.50.
Hesiod himself highlights this etymologising of alêthês (Th. 233–6: Νηρέα . . .
ἀληθέα . . . οὐδὲ θεμίστων λήθεται).
21
Pace Morgan (2000) 21–2; Palmer, loc. cit. 22 Cf. Il. 24.407–9.
23
h.Hom. 2.433 (ἀληθέα πάντ’, cf. 406: ἐρέω νημερτέα πάντα, 416: πάντα διίξομαι);
Od. 7.297 with 241ff; Od. 16.226ff; also qualifying accounts of past conversations:
Od. 17.122 (πᾶσαν ἀληθείην); Od. 16.61–6 (ἀληθέα πάντ’).
24
Od. 21.212–16 (ὡς ἔσεταί περ, ἀληθείην καταλέξω); Od. 18.342; Od. 17.15; cf. Cole
(1983) 15–17.

67
 Hesiodic Epistemology
guilty of shamelessness and disrespect (Od. 22.420–5). Nestor,
after accepting Telemachus’ request that he speaks the truth con-
cerning the events of Agamemnon’s murder (ἀληθές . . . ἀληθέα
πάντ᾽, Od. 3.247–54), articulates (i) a counterfactual scenario
(256–61), (ii) an account of the events (262–312) and (iii) an
expressly prescriptive moral (313–16).25
Outside the Muses’ address, the only other occurrence of
alêthês in the Theogony is also one of just two instances in
Homer and Hesiod where the term qualifies, not speech, but
a person: Nereus, we saw, is himself characterised as ‘without
falsehood and true’ and, concomitantly, as a kind and benign god
(ἤπιος . . . ἤπια δήνεα οἶδεν, 233–6).26
By contrast with alêthês, the term etymos (or, interchangeably,
etêtymos) can unambiguously qualify either speech as true27 or
states-of-affairs and objects as real or genuine.28 Often, though,
the term could admit both readings. In Odyssey 19.203, we could
be told that Odysseus made his many falsehoods like ‘verities’ or
like ‘realities’ (ἐτύμοισιν ὁμοῖα).29 We need not dwell on this
dilemma with regard to the Muses’ address. The Muses may
produce falsehoods which are like ‘truths’ (in which case the
etyma of line 27 would be equivalent to the ‘truths’ of line 28) or
like ‘realities’, i.e. like all those aspects of the world which the
Muses narrate truly and faithfully when they speak truths (Th. 28)
and speciously when they speak falsehoods (Th. 27).30
Again, the term etymos demonstrates a broad and flexible range
of applications. The term can signal the truth of such a proposition
as ‘this land is Ithaca’ (Od. 24.258–9; cf. Od. 13.232–5; h.Hom.
3.467–8) or of the statement of the number of figs in a fig tree
25
Cf. also Od. 11.507 (πᾶσαν ἀληθείην) with 510–12 (where Odysseus evaluates
Neoptolemus’ capacity as a deliberator).
26
Cf. Il. 12.433: γυνὴ χερνῆτις ἀληθής, interpreted variously as ‘honest’ and ‘conscien-
tious’ (Hainsworth (1993) 362) or ‘careful’ (Willcock (1978) 320); for these connota-
tions, cf. also Il. 23.358–61. There are, finally, two obscure occurrences in the Days,
signifying, it seems, the truths that certain days are favourable to certain actions: Op.
768 (by divine dispensation: ἐκ Διόθεν, 765); Op. 814–18.
27
μῦθος ἐτήτυμος, Od. 23.62; cf. Il. 23.440.
28
‘Odysseus really did come’ (ἔτυμόν τοι | ἦλθ’), Od. 23.26; cf. Il. 13.111; h.Hom. 3.64–5;
note ἐτήτυμος ἄγγελος (h.Hom. 2.46; Il. 22.437–9) for a ‘true’ messenger, i.e. one who
communicates truths (ἐτήτυμα, h.Hom. 2.44).
29
Cf. Il. 10.534; Od. 4.140; h.Hom. 2.44; h.Hom. 3.176; cf. Pucci (1977) 9, 35 n.8.
30
Cf. Ferrari (1988) 57.

68
 2.1 Semantic Overview: pseudea, alêthea, etyma, homoia
(ἀριθμὸς ἐτήτυμος, Hesiod, fr. 278.5 M-W). The word regularly
characterises accounts of past events as true (Od. 14.186–90;
Od. 23.62; Od. 4.645–7), but Homeric and Hesiodic poetry exhibits
no unambiguous applications of the term to (accounts of) future
states-of-affairs.31 With etymos too we can point to contexts where
specifically genealogical or biographical truths or realities are in
question (κείνου μέν τοι ὅδ’ υἱὸς ἐτήτυμον, Od. 4.157; cf. Od. 14.
186–7; Od. 24.297–8; Od. 1.174–6 and, again, Od. 19.203). And
this term too may qualify as true such clearly prescriptive and
evaluative judgements as ‘it is no base thing to ward off destruction
from one’s companions when they are oppressed’ (Il. 18.128–9;
cf. Il. 13.111) or ‘Antilochus is sensible and wise’ (οὔ σ’ ἔτυμόν γε
φάμεν πεπνῦσθαι, Il. 23.440). Outside the Muses’ address, the word
etymos recurs in Hesiod’s two extant poems only in a programmatic
statement at the end of the proem to the Works and Days: ‘I would
tell etêtyma to Perses’ (Op. 10).32 Hesiod does not circumscribe or
elucidate how much or what aspects of the subsequent poem he
hopes will fall under the category of ‘truths’ or ‘realities’. Thus, the
scope and emphases of the term etêtyma here are as open-ended as
they are in the Muses’ address in the Theogony and equally amen-
able to be reassessed in the light of the scope and emphases of
Hesiod’s ensuing preoccupations.33
In some passages, the term homoios requires or suggests a sense
of equivalence or identity. Achilles is angered whenever one man
robs his (social?) equal (τὸν ὁμοῖον . . . ἀμέρσαι, Il. 16.53) and
recognises that both Patroclus and himself were fated to redden the
selfsame earth (i.e. Troy: ὁμοίην γαῖαν, Il. 18.329).34 We also
encounter, however, a weaker and looser sense: Hera and Athena
do not possess the same gait as doves but a like one (τρήρωσι
πελειάσιν ἴθμαθ’ ὁμοῖαι, Il. 5.778; cf. h.Hom. 3.114). Here two
31
Penelope’s theory of oneiromancy at Od. 19.562–7 (ἔτυμα κραίνουσι) is a possible but
ambiguous case.
32
For a discussion of this line, see Ch. 2.4.
33
Neither Homeric nor Hesiodic Greek exhibits a shred of evidence for the view that the
word alêthea (Th. 28) indicates divine truths and etyma (Th. 27) human truths (e.g.
Stoddard (2004) 80–5; Tsagalis (2009) 133–4). Among numerous difficulties, this
arbitrary reading generates the irresolvable problem of satisfactorily explicating why
the Muses should claim or mention only the lopsided abilities to articulate divine truths
and falsehoods like human truths.
34
Cf. also Op. 114; Od. 2.276–7; Il. 5.440–2 (note ἶσ’).

69
 Hesiodic Epistemology
elements are somehow alike or similar, rather than equal or iden-
tical, with regard to their respective possession of a certain
quality.35 Hesiod’s Works and Days offers similar if more open-
ended comparisons. The warlike Bronze race is, not simply
unequal or non-identical with, but indeed nothing like (οὐδὲν
ὁμοῖον, 144) the infantile Silver race (127–55). Again, Hesiod’s
apocalyptic vision for the Iron race is one in which (i) a father is no
longer ‘like’ (ὁμοίιος) his children, nor (ii) vice versa nor
(iii) a guest like his host nor (iv) a companion like his companion
(Op. 182–3). No notion of equivalence, equality or identity could
make sense of all four cases. The gloss ‘like-minded’36 captures
best the envisaged breakdown of familial and social affinity
(cf. ὁμοφροσύνην, Od. 6.180–4). When, however, the adjective
can coherently support either the translation ‘like’ or ‘same’, and
even in many passages which do suggest a certain emphasis, we
cannot assume one sense to the exclusion of the other.37 To that
extent, the term homoios ambiguously blurs the distinction
between (i) likeness and (ii) equivalence or identity. In their
address to Hesiod (ψεύδεα . . . ἐτύμοισιν ὁμοῖα), the Muses may
have in mind falsehoods that are somehow like truths, or false-
hoods which are partially38 equivalent to or overlapping with
truths with respect to their possession of some unspecified quality
or qualities, and they may convey these senses indistinctly.
The notion of deceptive speciousness is not intrinsic to the term
homoios but it is conveyed by the Muses’ surrounding contrast
between truths (Th. 28) and falsehoods which are like and/or

35
Cf. also Il. 10.437; h.Hom. 2.177–8 (χαῖται . . . ἄνθει ὁμοῖαι); Od. 6.230–1; Od. 23.
157–8; Il. 17.51–2. Heiden (2007), arguing that homoios only ever signifies equiva-
lence, blankly denies the possibility that two things may be like or similar to each other
with respect to their possession of some quality, restricting the notions of ‘likeness’ and
‘similarity’ to outward appearance (esp. 156 with n.11; 158 with n.15).
36
Most (2006) ad loc.
37
For example, the uninitiated do not enjoy the same or similar things in the afterlife
(ὁμοίων | αἶσαν, h.Hom. 2.481–2), cf. e.g. Od. 2.121; Il. 18.120. Both likeness and
equality arguably function in the formula δέμας ἀθανάτοισιν ὁμοῖος (vel sim.), Od. 3.468;
Od. 23.163; Od. 8.14; Od. 6.16; Hes. frr. 10a.33; 10a.49; 291.1 M-W. In many passages
which indicate an emphasis on equality, a connotation of likeness seems plausible, as in
Il. 2.553–4 (τῷ δ’ οὔ πώ τις ὁμοῖος . . . κοσμῆσαι ἵππους).
38
As with Odysseus’ falsehoods at Od. 19.203 (cf. Clay (2003) 62). The word homoia
hardly implies that the Muses’ falsehoods (pseudea) are wholly equivalent to truths,
pace Heiden (2007) 153, 169–72.

70
 2.1 Semantic Overview: pseudea, alêthea, etyma, homoia
partially equivalent to truths (Th. 27), combined with the absence
of signposts or criteria with which we might distinguish the
Muses’ lovely, melodious truths from their lovely, melodious
falsehoods. The term homoios similarly acquires this force
through its context when Odysseus constructs for Penelope
a false autobiography which often overlaps with and resembles
the truth (Od. 19.203) and when Aphrodite renders herself ‘like
(ὁμοίη) in stature and bodily form to an unwed mortal maiden’
(h.Hom. 5.82).39 By contrast with this remark about Aphrodite, but
like Homer’s remark about Odysseus’ verisimilitudinous fabrica-
tions, Hesiod’s Muses leave entirely unspecified and open-ended
the respects in which one item (‘falsehoods’) is like and/or par-
tially equivalent to another (‘verities’ or ‘realities’).40
The foregoing survey brings out, in embryonic form, core
aspects of the dynamic between poet and Muses. The Muses
issue an underdetermined programmatic statement, articulated
through several underdetermined key concepts, and – on the
semantic level as on the epistemological level – leave Hesiod,
and us, to assess for ourselves all that they might mean by saying
what they say.41 Furthermore, the usage of the terms pseudea and
alêthea elsewhere and especially in Hesiod already situates the
Muses’ dispensation of truths and falsehoods within the broader
theological framework of the divine dispensation of good and bad.
In issuing truths, the Muses recall the agency of a ‘true’ and benign
Nereus (Th. 233–6). In issuing baneful (Th. 226–32) and specious
falsehoods, they are more akin to a cunning Hermes – or, signifi-
cantly, a Pandora (Op. 78).
The Muses’ refusal to clarify or delimit the terminology of their
programmatic statement invites us to reassess continually its scope
and emphases as we explore the scope and emphases of the
Theogony. Theogonic and cosmogonic states-of-affairs, the ten-
sions and (familial) relations between the powers which structure
and govern the cosmos, past and present interactions between
male and female and between gods and mortals – all these domains

39
Cf. Bryan (2012) 15–16. 40 On this point, cf. Heiden (2007) 168–9.
41
Ledbetter (2003) 45 argues that attempts to pinpoint the Muses’ falsehoods violate the
rhetorical force of their unrestricted mastery of deception.

71
 Hesiodic Epistemology
are encompassed in the prospect of comprehensive truths and
specious deceptions.
I submit further, as a hypothesis the plausibility of which will be
substantiated by the next two sections, that we should not overlook
the attested normative and evaluative force of this vocabulary of
truth and falsehood. We will see that the mortal poet and his
audience do not suffer the same epistemic paralysis in assessing
and accepting the didactic and prescriptive import of the
Theogony’s accounts of, for example, the interactions between
gods and mortals and between male and female agents.
Furthermore, these prescriptive insights, which we can establish
on the basis of Hesiod’s inspired narratives (whatever their
descriptive truth-value), inform our evaluation of the theological
and epistemological implications of the Muses’ address itself.
In other words, we can find some daylight in the Theogony
between prescriptive or evaluative insight and the assurance of
complete, literal descriptive truth (just as we found in Hesiod’s
juxtaposition of the Prometheus-Pandora and Races episodes in
the Works and Days; see Chapter 1.5).42 Our survey in this section
showed that the threatening spectre of falsehoods is one important
aspect of the Muses’ address. But perhaps, then, by open-endedly
signalling that their falsehoods are ‘like verities’ or ‘realities’, the
Muses (and Hesiod) also create space for descriptive falsehoods
which, with respect to their possession of some unspecified quality
or qualities, are nonetheless partially but importantly like, identi-
cal with or equal to genuine theogonic or theological states-of-
affairs. Some falsehoods (pseudea), then, may share with some
corresponding verities or realities (etyma) the quality of offering
evaluative and prescriptive insight.

2.2 The Muses’ Address and Divine Dispensation in Hesiod

We cannot over emphasise the textual fact that at no point do the
Muses say that to Hesiod they will dispense only truths and never
42
Mutatis mutandis, one could draw a comparison on this point with Plato’s ‘Noble Lie’ at
Rep. 2.382aff. (We might even find an allusion to Th. 27–8 in 382d2–3: ἀφομοιοῦντες τῷ
ἀληθεῖ τὸ ψεῦδος ὅτι μάλιστα. Cf. Belfiore (1985); Van Noorden (2015) 135–6.) On the
Noble Lie, see the discussion in Rowett (2016).

72
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
verisimilitudinous falsehoods.43 Before narrating his encounter
with them, Hesiod does advertise the aesthetic and, I suggest,
didactic quality of the ‘fine’ song which the Muses taught him,
but notably refrains from making any truth-claims (καλήν, Th. 22).
Hesiod precedes his theogonic account itself with an extended
disquisition on the value of the Muses’ song and of the mortal
poet’s Muse-inspired song (Th. 36–115). Throughout, Hesiod
focuses on the aesthetic qualities and the soothing and therapeutic
effects of song. So, for example, when ‘a poet, a servant of the
Muses’ (ἀοιδὸς | Μουσάων θεράπων, Th. 99–100) sings of gods
and men, then a distraught auditor ‘forgets his sorrows at once and
does not remember his anxieties; for swiftly the gifts of the god-
desses have turned them aside’ (102–3).44 Conspicuously – in the
light of the Muses’ opening address – not once does Hesiod stake
a truth-claim or use veridical terminology in this sustained reflec-
tion on the nature and benefits of inspired song. In order to achieve
their therapeutic and mind-transporting qualities, the poet’s
inspired verses do not need to be true or believed.45 It should be
emphasised, though, that the independence of the song’s therapeu-
tic value from its truth-value does not mean that the Muses’
dramatic contrast between their verisimilitudinous falsehoods
and their truths loses all significance. As we shall see in further
detail, this contrast activates an epistemic anxiety which fits into
a broader cluster of anxieties that frames the human condition. Put
simply, there is a great deal in Hesiod’s world and world view from
which we need the soothing distractions of inspired song, and this
includes our weakness and paralysis when it comes to assessing
the truth-value of inspired narratives about the history and struc-
ture of this world we inhabit (Th. 26–8).
The goddesses descend from Zeus’ altar (Th. 4) and topmost
Helicon (7) to meet Hesiod at the foot of the mountain (23).
The trajectory underscores their respective divine and mortal
43
The question-begging paraphrase of Th. 28 in West (1966) 162 is symptomatic of many
truths-only interpreters: ‘when we choose, we can reveal the truth, and we are going to
reveal it to you’. The last clause is not in the Greek. Cf. Clay (2003) 59–60.
44
Cf. Th. 37 (τέρπουσι), 39–40 (αὐδή . . . ἡδεῖα), 40–1 (γελᾷ . . . ὀπὶ λειριοέσσῃ), 51
(τέρπουσι), 55 (λησμοσύνην τε κακῶν κτλ), 65 (ἐρατήν . . . ὄσσαν), 68 (ὀπὶ καλῇ), 96–7
(γλυκερή . . . αὐδή), 104 (ἱμερόεσσαν ἀοιδήν).
45
See further Ledbetter (2003) 40–60; Halliwell (2011) 16–17.

73
 Hesiodic Epistemology
statuses.46 If the Muses first sang among themselves of other gods
(Th. 11–21), they now sing to Hesiod of Hesiod (26) and of
themselves (27–8). First of all (πρώτιστα, Th. 24) the Muses
disclose and determine the nature of the poet’s relation to them.
Nowhere else in Hesiod can we parallel his clear demarcation of
himself here from the goddesses as the auditor of their verses:
‘And this speech the goddesses first of all spoke to me’ (Th. 24).
The Muses’ address – their first and last unambiguously direct
speech – thus outlines a framework of poetics and epistemology to
which everything else in the Theogony is internal. Clay interest-
ingly associates this ‘first of all’ (Th. 24) with the only other
occurrence of this phrase in the Theogony: ‘first of all Chasm
came into being’ (116). Hesiod’s engagement with the question
of cosmic origins requires a prior engagement with the question of
epistemic origins.47 The Muses, then, first of all articulate the
epistemological caveats and limitations of their mediation
(Th. 26–8) only through which can the cosmogonic ‘first of all’
be made available to mortals. Hesiod’s deictic singular (τόνδε δέ
με, Th. 24) contrasts with the Muses’ vocative plurals: ‘Shepherds
of the field, base, shameful things, mere bellies’ (Th. 26). In
addressing and characterising Hesiod, the Muses address and
characterise a group which he represents.48 The designation ‘shep-
herds’, and the encounter with Hesiod as he tends his sheep, situate
the human addressee between beasts and goddesses.49 This posi-
tioning appropriately introduces and complements the contrast
between mortal and divine capacities and statuses. The same
hierarchy of god-man-beast prominently frames the Prometheus
episode, where Zeus determines the relations between gods and
men as he establishes the norms of sacrifice.50 The Muses’ abuse
and Hesiod’s encounter with them as a whole negotiate, further-
more, the prospect for mortals to edge some way towards either
extreme. The insult ‘mere bellies’ rhetorically reduces the

46
Following Σ Th. 23 Gregorio. 47 Building on Clay (2003) 52, 57.
48
See similarly Demeter’s words about humans to Metaneira in h.Hom. 2.256–7 (νήϊδες
ἄνθρωποι κτλ); cf. Clay (2003) 57, n.37; Stoddard (2004) 72, n.30; Halliwell (2011)
13–14. Demeter precisely berates Metaneira’s human inability to demarcate divine
dispensations of ‘good’ (cf. Th. 28) and ‘bad’ (cf. Th. 27).
49
See similarly Stoddard (2004) 75, 79. 50 Cf. R. Osborne (1997) 18.

74
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
shepherds to the bestial status of the sheep they tend.51 The Muses
also, however, enable Hesiod to produce narratives which, in their
beauty and subject matter, transcend a human’s independent artis-
tic and, since they could be true (Th. 28), epistemic reach.
The Muses’ assertion that they know how to inspire both truths
and specious falsehoods is followed by these lines (Th. 29–34):
ὣς ἔφασαν κοῦραι μεγάλου Διὸς ἀρτιέπειαι, 29
καί μοι σκῆπτρον ἔδον δάφνης ἐριθηλέος ὄζον 30
δρέψασθαι,52 θηητόν· ἐνέπνευσαν δέ μοι αὐδὴν
θέσπιν, ἵνα κλείοιμι τά τ’ ἐσσόμενα πρό τ’ ἐόντα,
καί μ’ ἐκέλονθ’ ὑμνεῖν μακάρων γένος αἰὲν ἐόντων,
σφᾶς δ’ αὐτὰς πρῶτόν τε καὶ ὕστατον αἰὲν ἀείδειν. 34

So spoke great Zeus’ daughters, ready-of-speech,

And they granted me to pluck for myself a staff, a branch of luxuriant laurel,
A thing of wonder. And they breathed into me a voice
Inspired, so that I would celebrate the things that will be and were before,
And they bade me to sing of the race of the blessed ones who always are,
But always to chant of themselves first and last.
In its only Homeric occurrence, the description ‘ready-of-speech’
(Th. 29) signifies treacherous verbal trickery (‘ready of speech and
a thief with words’, Il. 22.281). In emphasising the Muses’ mas-
tery of words, Hesiod cautions against a blithe lack of circumspec-
tion in assessing their statement and all its possible ramifications.
The Muses either present Hesiod with a laurel staff or grant him
to pluck one for himself (Th. 30–1).53 Hesiod’s staff does not
decide or even plausibly pertain to the question of the
Theogony’s truth-value.54 In the Iliad, Agamemnon is deceived
by Zeus while holding the very ancestral and imperishable staff
(Il. 2.46, cf. 1–38, 79–81, 85–6, 348–9) which he received through
Zeus’ favour (Il. 2.100–8). Hesiod’s staff does underscore the
communal and didactic authority of his poetry.55 In addition to

51
For the bestialising force of γαστέρες οἶον, cf. also the parallels in West (1966) ad loc.; Σ
Th. 26 Gregorio; Hesych. sv. γ.190 Latte, and Epimenides’ variation: DK3 B1
(Κρῆτες . . . θηρία, γαστέρες ἀργαί), cf. Paley (1861) ad loc.; Clay (2003) 57; Stoddard
(2004) 74, 78; Tsagalis (2009) 133.
52
Or: δρέψασαι. The dilemma is insoluble, although δρέψασθαι is the lectio difficilior. Cf.
Il. 23.21.
53
Depending on whether we read δρέψασαι or δρέψασθαι, see the preceding note.
54
Pace e.g. Griffith (1983) 48–9 with n.50.
55
Griffith (1983) 50; cf. Segal (1992) 28; Stoddard (2004) 185–6.

75
 Hesiodic Epistemology
kings (Il. 1.279) and, more generally, arbitrators (Il. 1.234–9),
a staff is elsewhere wielded by priests (Il. 1.14–15, 28) and
diviners (Od. 11.90–1; Aeschylus, Ag. 1265) as well as heralds
(Il. 7.277–8) and, temporarily, anyone speaking in the assembly
(Il. 1.245–6; Il. 2.278–9, etc).56 The staff would seem, then, to
align Hesiod with the Muse-favoured king who, with soothing,
persuasive and gifted speech, effects communal harmony and
issues just decrees (Th. 80–92).57 But, as we saw, our text belies
the notion that Hesiod anchors this claim for societal and didactic
value and authority by asserting a speculative truth-claim for the
descriptive content of the Theogony. The proem nowhere uses
veridical terminology to describe the welcome effects of poet,
king or Muses.
The Muses inspire Hesiod so that he may glorify ‘the things that
will be and were before’ (Th. 31–2). The phrase adopts the markers
of divination. With the omission of ‘things-that-are’, it is identical to
the scope of the Muses’ own songs in Theogony 38 and of the Iliad’s
description of the seer Calchas, who ‘knew the things that are and
the things that will be and were before’ (ᾔδη τά τ’ ἐόντα τά τ’
ἐσσόμενα πρό τ’ ἐόντα, 1.69–70). This echo of divination reinforces
the impression, engendered by the Muses’ address, that Hesiod is
preoccupied here with epistemological concerns: what sort of
access does his continued interaction with epistemically superior
divinities allow the poet (and his audience) into matters which, as
with divination, lie beyond human experience and independent
human cognition?58 But the use of mantic terminology offers noth-
ing like unambiguous support for the truths-only interpretation.59
As we shall see in detail (Chapter 3.1), while we occasionally
encounter mantic knowledge claims (as of Calchas in Il. 1.69–70),
we also find the widespread recognition that, precisely because
knowledgeable divine communication must always be processed
by the mortal diviner and/or consultant, mortal inferences from it
are always conjectural and fallible (as of Calchas in Il. 2.299–300).

56
West (1966) ad Th. 30.
57
Cf. Clay (1988) 333; (2003) 73–6; Ledbetter (2003) 48; Stoddard (2004) 184–6.
58
If nothing else, the omission in Th. 32 of ‘things-that-are’ accentuates Hesiod’s focus in
the Theogony on matters external to human experience.
59
Pace Katz and Volk (2000) 122.

76
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
And, of course, the communication itself may be false. In language
highly reminiscent of the Muses’ address, the Homeric Hymn to
Hermes speaks of divinatory bee-maidens, the Thriai, who either
wish to speak the truth or deceive when in favourable or unfavour-
able moods.60 Conspicuously – and even pointedly – Hesiod’s
appropriation of the divinatory formula in Theogony 32 does not
include a knowledge-claim or truth-claim, such as the claim made
of Calchas’ knowledge in the Iliad.61
One might identify in the use of the verb ‘to be’ in Theogony 32
a veridical expression: Hesiod is to celebrate the things that will be
and were before truly. Finding such a veridical nuance in the line
is, however, hardly obvious. The participial use of the verb ‘to be’
(τὰ ὄντα) for ‘the truth’ or ‘the facts’ is all but unparalleled in
Homer and Hesiod. In fact, the only clear candidate is the remark
about Calchas quoted above.62 Even here, however, the verb is
used primarily in an existential sense. Calchas has knowledge of
the things that exist, will exist and existed in the past. If the line
does convey a veridical lexical nuance – Calchas knows the things
that are as they truly are, etc. – then this is because the verb ‘knew’
(ᾔδη) is an epistemic success term. The force of the verb ‘to be’ in
line 32 is likewise existential. The subject matter of Hesiod’s poem
are genuinely existing objects, which will be in the future and were
before. In fact, since the Theogony contains no prophecies about
future states-of-affairs, it has been persuasively argued that the
conjunction ‘the things that will be and were before’ most likely
signifies everlasting entities: i.e. the gods themselves.63 The next
line approaches the same point in a different way, by saying that
the Muses bade Hesiod to sing of (ὑμνεῖν) the race of those who
always exist (γένος αἰὲν ἐόντων, Th. 33). Hesiod is inspired so that
he may sing of and celebrate the gods. Again, if Iliad 1.70 conveys
a veridical nuance, then this is because ‘knew’ conveys epistemic
success. The idea of ‘telling of’ or ‘celebrating’ or ‘glorifying’
60
h.Hom. 4.560–4. I return to this passage below.
61
For the connections between poetry and divination, see more generally Chadwick
(1942); Dodds (1951) 80–2; Cornford (1952) 76–9, 102; West (1966) ad Th. 32;
Ustinova (2009) 168–9; cf. PW xxxiii-xxxiv; Detienne (1996) 23; Marquardt (1982)
9, n.2; Katz and Volk (2000) 124; Flower (2008) 78 (with qualifications).
62
Cf. Kahn (1973) 350–1; see 332–3 on the general rarity of the veridical usage of εἰμί.
63
Clay (2003) 65–7.

77
 Hesiodic Epistemology
something cannot, however, shoulder the same load. First, even
before we consider the all-important context of line 32 of the
Theogony, there is no clear reason to think that the verb must
signify for Hesiod something with veridical implications over
and above the treating of a subject matter in the context of song.
‘Celebrating’ something can seem almost functionally equivalent
to ‘singing’ of it and indicates the activity of bards in quite
a general way (κλέα φωτῶν | ᾄσομαι ἡμιθέων ὧν κλείουσ’ ἔργματ’
ἀοιδοί, h.Hom. 32.18–19).64 Second, and most importantly,
Hesiod is of course aware – and he himself indicates later on –
that songs and celebrations of gods and heroes are the bread and
butter of every inspired ‘poet, servant of the Muses’ (Th. 99–101).
Crucially, however, whether or not any given Muse-inspired cel-
ebration articulates such things that are or were as they truly are or
were is precisely what Hesiod’s Muses have called into question.
The existing conventions of epic discourse afforded Hesiod a rich
array of unmistakably veridical terms with which he could have
staked clear and explicit truth-claims.65 It is difficult to find in
Theogony 32 what would be – after the destablishing declarations
in lines 26–8 – a remarkably muted one.
One might still feel unease with the idea that the Muses could
inspire a poet, only to have him celebrate the gods and even the
Muses themselves through (at least in part) verisimilitudinous
falsehoods. Again, however, given that celebrations of the gods
are ubiquitous in Muse-inspired poetry (Th. 99–101), the Muses’
address effectively tells us that they do so. After all, we can hardly
imagine that, in Theogony 27, the Muses are informing us about
a capacity which they possess – to articulate verisimilitudinous
falsehoods – but which never manifests itself when they inspire
their poets. Even aside from these textual points, however, we
should not presume that it is on the topic of the gods, of all things,
that the Muses can be expected to divulge complete and

64
Halliwell (2011) 36–92 shows that the Homeric poems are not devoid of unsettled and
critical reflections concerning the reliability and verifiability of the veracity of poetic
discourse and the ‘renown’ or ‘reputation’ (kleos) it conveys. In Il. 2.486, the epistemic
status of kleos is contrasted unfavourably with that of sure knowledge (‘but we hear only
report (kleos) and do not know anything’). See further Ch. 2.3.
65
See e.g. the veridical language in Od. 3.254; Od. 21.212; Il. 24.373.

78
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
transparent truths. The gods can disclose or dissimulate them-
selves as they please, when they inspire mortals no less than
when they encounter them (we may think here of the disguised
Demeter and Dionysus in the Homeric Hymns which celebrate
those gods), and the inability of humans to be certain of their
footing is itself a testament to the power and glory of the gods.
We must not project back onto Hesiod a preoccupation with the
idea, which we find in later philosophical thinkers, that speaking
falsehoods about the gods is itself a matter of impiety, although
even there the concern is with statements which are not only false
but also offensive.66 Does it follow, then, that Hesiod is encoura-
ging us to doubt that, for example, Zeus really did triumph over his
enemies and now rules among the gods? We do not need to go that
far. Although, in principle, the epistemological worries which
Hesiod articulates concerning divine inspiration could ultimately
encourage one to develop global questions concerning the very
basis of any of our beliefs about the gods, Hesiod may well be
taking as read such central facets of divine history and politics and
never imagines that his audience will do otherwise. Importantly,
however, we are certainly not obliged to ascribe to Hesiod this
kind of thoroughgoing theological scepticism in order to maintain
the interpretation that the Muses’ address has the effect of leaving
the poet uncertain concerning the truth-value of his verses. After
all, by setting aside as presupposed these very basic facets of
divine history and politics, we have hardly validated in a general
way the truth of Hesiod’s vast theogonic-cosmogonic and theolo-
gical-cosmological accounts, let alone of some given statement or
detail within his poem. We have by no means guaranteed that all
that this inspired poet tells us about the history and structure of our
world are, quite simply and unproblematically, truths.
In the shadow of the Muses’ address, the fact of inspiration
(Th. 31–2) precisely does not correspond to a guarantee of the
communication of truths. Hesiod’s word for ‘voice’ (αὐδήν) indi-
cates a specifically human voice, while the qualifier ‘inspired’ or
‘divine’ (θέσπιν), conversely, signals the divine origin and status

66
E.g. Xenoph. DK21 A12; Pl. Lg. 10.899d4–900b6; see further Mikalson (2010) 145–6,
158–9.

79
 Hesiodic Epistemology
of this voice.67 Hesiod’s ‘divine human-voice’ comprises, I think,
a highly apposite paradox. It is only due to the divine origin of his
poetry (θέσπις) that the subsequent account could indeed comprise
truths. That, however, Hesiod’s own voice – his own authority –
remains human (αὐδή) coheres with his lack of means to exclude
the prospect of falsehoods. Hesiod’s transformation through the
Muses’ inspiration does not amount to the acquisition of the
Muses’ own capacity to articulate, and to discriminate between,
truths and falsehoods which lie beyond the reach of human
cognition.
Hesiod’s divine human-voice also raises the intractable question
of the nature of the collaboration between poet and Muses involved
in inspiration.68 We would be wrong to ask here for a clear and
precise demarcation between divine influence and human
agency. A distinctive feature of the interaction between gods and
mortals in Homer, traditionally termed ‘double motivation’ or
‘over-determination’, is that the same actions and impulses may
be construed as caused simultaneously by both divine and human
agencies.69 Often, mortal heroes can perform their actions only
through continuous divine interaction or ‘inspiration’ (‘gleaming-
eyed Athena breathed (ἔμπνευσε) might into him’, Il. 10.482), but
they are not thereby the gods’ inactive, mindless puppets. In the
Odyssey, the bard Phemius strikingly instantiates over-
determination: ‘I am self-taught (αὐτοδίδακτος) and a god breathed
into my mind all kinds of lays’ (Od. 22.347–8). We find no warrant
in these words to defuse what puzzles us about them by arbitrarily
isolating, say, the artistic aspects of poetic composition from its
informational content and restricting inspiration to the latter.70
Inspiration obviously extends also to the former.71 If anything,

67
For αὐδή, see e.g. Od. 5.333–5, with Clay (1974) passim (131 on SVF 2.144); cf. Nagler
(1996) 143. For θέσπις, e.g. Od. 8.498. The ‘divine human-voice’ paradox of Th. 31–2 is
observed by Clay (2003) 65 with n.65; Stoddard (2004) 86 n.55, 182.
68
Mutatis mutandis, this issue will be central in our discussions of Xenophanes,
Parmenides and Empedocles.
69
The Greeks are routed ‘by Hector and father Zeus’ (Il. 15.636–7, cf. 693–5; Il. 20.94;
etc.); see e.g. Janko (1992) 3–4; Kearns (2004) 59, n.2 and, more generally, Dodds
(1951) 1–27.
70
Following Murray (1981) 97; Bowie (1993) 17; Segal (1992) 26–7.
71
E.g. Th. 22, 104; Od. 8.44–5; Od. 17.518–20; Archil. fr. 1.2 W; Thgn. 250; etc.; cf.
Lesher (1994a) 17; Murray (1981) 95.

80
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
the second clause seems to expound the first. Phemius, presumably
to indicate that he does not simply mimic the poems of other
minstrels, identifies himself and, inextricably, his association with
the divine as the source of his poetry. Plato influentially resolved
this problem of poetic agency by recasting the inspired poet as an
entirely passive, possessed mouthpiece, lacking any understanding
of what he voices.72 Importantly, however, no clear sign of Plato’s
model can be found in any pre-Classical source.73 Archaic por-
trayals of inspiration do not register a tension between divine
influence and the poet’s active agency in, say, shaping the topic,
wording and internal structure of his narrative or in leading us
implicitly or explicitly to reflect on those narratives. As often, we
mistake at our own peril Plato’s prescriptive portrayals (here, of
poetic inspiration) for descriptive accounts of the common view.
Nonetheless, a widespread and important aspect of inspiration
is the poet’s epistemic dependence on the content provided by the
Muses. Poetic invocations typically recall the structure of mantic
consultations. The mortal poet sets the terms of an inquiry (‘tell
which one of them first came to be’, Th. 115) and thereby delimits
the possible range of the subsequent, unverifiable reply (‘first of all
Chasm came into being’, 116).74 With the response to the invoca-
tion (as in Th. 116 and following), the text offers us no means to
distinguish the poet’s voice from the Muses’. Odysseus rightly
concludes that Demodocus sings with divine inspiration, on the
grounds that he narrated events in Troy accurately, as though he
was himself there or heard from a witness (Od. 8.487–91).75
72
Ion, passim, esp. 533e5–534d4; Ap. 22b-c3; Men. 99c2-d5; Phdr. 245a; cf. Lg. 4.719c1-
d1, 3.682a1–6.
73
As Murray (1981) showed; cf. Dodds (1951) 82, 101 n.121–2; Bowie (1993) 18;
Halliwell (2011) 57–9. Scholars persist in the anachronism of projecting backwards
Plato’s resolution, e.g. Ledbetter (2003) 8–60, esp. 53–8; Katz and Volk (2000) 127–9;
Stoddard (2004) 63–7; Robbiano (2006) 68; Palmer (2013) 318.
74
Cf. Il. 1.1–9 (esp. 8–9); Il. 2.760–70; Il. 11.218–22; Il. 14.508–12; Il. 16.112ff; Od.
1.1–10; h.Hom. 5.1; h.Hom. 9.1; h.Hom. 14.1–2; h.Hom. 17.1; h.Hom. 19.1; h.Hom.
20.1; h.Hom. 31.1–2; h.Hom. 32.1–2; h.Hom. 33.1; also Il. 5.703ff; Il. 8.273–6; Il.
11.299ff; Il. 16.692ff.
75
In the light of Odysseus’ remark that Demodocus recounts the events like an eyewitness,
the phrase ‘in order’ (κατὰ κόσμον, Od. 8.489; cf. similarly κατὰ μοῖραν, Od. 8.496)
could indicate, in this particular context, a descriptively true as well as a well-structured
and proper account, Murray (1981) 94; Walsh (1984) 8–9; Goldhill (1991) 57–8, 68;
Ledbetter (2003) 16. Still, we can agree with Halliwell (2011) 84–8 that accuracy need
not be the focus of this terminology.

81
 Hesiodic Epistemology
Odysseus’ position as auditor, however, is highly unusual. His
ability to check Demodocus’ account against his own first-hand
experience and knowledge of its subject matter is the exception
that proves the rule. In his invocation to the Muses in Iliad
2.484–93, the Homeric poet expressly contrasts the Muses’ inde-
pendent knowledge – ‘for you are goddesses and you are present
and know all things’ (πάρεστέ τε ἴστέ τε πάντα, 485) – with the
ignorance of mortals: ‘We hear only report (κλέος) and do not
know anything’ (οὐδέ τι ἴδμεν, 486).76 In one sense, the Homeric
poet places himself on an epistemological par with his audience
(‘we know nothing’).77 In another sense, however, this common
ignorance procures for him an authority which is difficult to
contest: we are all ignorant, but his inquiries are answered by
knowledgeable goddesses.78 The Muses’ address to Hesiod thus
relates to and destabilises a traditional theological and epistemo-
logical attitude. By introducing and underscoring the possibility of
false inspired poetry, Hesiod puts precise pressure on the tension
(already present but more muted in Homer’s invocation in Iliad 2)
between a poet’s incorrigible ignorance concerning matters which
he can narrate only through inspiration and his acquiescence in the
truth of his narratives.79

76
One could devise different stories about the source of the knowledge of Hesiod’s Muses.
In the Hymns, for example, Hermes appears to be born with innate knowledge of
theogonic events which precede his birth, see h.Hom. 4.54–9, 427–33, cf. 440–2 (ἐκ
γενετῆς); note κατὰ κόσμον (433), with the preceding note above. In Hesiod himself that
epistemological question is never clearly raised. I see no reason to burden Hesiod with
the premise that the Muses know a state-of-affairs if and only if they directly experi-
enced it (contra Heitsch (1966) 197; Clay (2003) 64, drawing on Il. 2.485). It would
follow that the Muses’ knowledge claim (Th. 27–8), insofar as it pertains to events
preceding their own birth, is itself false. But nowhere in the Hesiodic corpus does this
prospect play a discernible role. Is Hesiod, then, simply inconsistent in ascribing
knowledge to the Muses? To avoid inconsistency, it is best in the first place not to
import and generalise as a universal theory of (divine) knowledge the apparent associa-
tion in Il. 2.485 between the Muses’ presence in Troy and their own knowledge of the
events of Troy.
77
The ignorance of ‘we mortals’ and our helpless dependence (general and epistemic) on
gods are themes that pervade archaic poetry, e.g. Archil. fr. 131 W; Semon. fr. 1.1–10
W (note line 4: ἃ δὴ βοτὰ ζόουσιν, οὐδὲν εἰδότες, cf. Th. 26); Thgn. 133–42; Simon. PMG
527; Sol. fr. 13.17–76; frr. 16–17 W; Pind. O 7.24–6; N 6.4–7; N 11.43–6; cf. Lesher
(2008) 476–7 with n.1; Torgerson (2006) 32–6; Bodnár (1989) 61–3.
78
Cf. Stoddard (2004) 89; Ledbetter (2003) 17.
79
See along similar lines Clay (2003) 63–4.

82
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
The truths-only interpretation has been criticised for suppres-
sing the ‘unsettling threat’ of the Muses’ address.80 The Muses’
specification ‘we know, whenever we wish (εὖτ’ ἐθέλωμεν), how to
utter truths’ compellingly vindicates this objection. The qualifica-
tion underscores the unforeseeable capriciousness of the Muses’
dispensation of truths and falsehoods.81 Hesiod’s catalogue of
Hecate’s dispensations of success and failure to men of different
professions (kings, warriors, athletes, horsemen, seafarers,
cowherds, goatherds, shepherds) repeatedly employs the same
vocabulary (ᾧ δ’ ἐθέλῃ, Th. 429; ὅν κ’ ἐθέλῃσιν, Th. 430; οἷς κ’
ἐθέλῃσι(ν), Th. 432, 439; θυμῷ γ’ ἐθέλουσα, Th. 446). Hecate’s
‘easy exercise of arbitrary power’82 is illustrated by her attitude to
fishermen: ‘Easily the glorious goddess bestows a big haul, and
easily she takes it away once it appeared, if she so wishes in her
heart’ (ἐθέλουσά γε θυμῷ, Th. 442–3). In a sense, divine caprice is
even more pronounced in the Muses’ ‘whenever we wish’ than
with Hecate: though not exclusively, Hecate is at least typically
described as granting or denying favour to certain mortals rather
than at certain times (Th. 419, 429–30, 432–3, 439). In Hesiod’s
animal fable, the Eagle says to the Nightingale: ‘I shall make
dinner of you, if I wish (αἴ κ’ ἐθέλω), or I shall let you go’
(Op. 208–9). Whatever else we make of this ainos, Hesiod has
clearly aligned himself, qua poet, with the Nightingale, whom he
describes as a ‘bard’ (ἀοιδόν, Op. 208). Recall, finally, the account
in the Homeric Hymn to Hermes, reminiscent of the Muses’
address, of the mantic Thriai, who willingly speak the truth
when they have feasted on honey but issue falsehoods when
deprived of it (ἐθέλουσιν ἀληθείην ἀγορεύειν . . . ψεύδονται,
h.Hom. 4.560–4). The dispensation of good and bad in accordance
with the fluctuating volition of goddesses, the qualification ‘when-
ever’ (εὖτε) and Hesiod’s inclination to picture himself in the
clutches of such capricious higher powers (Op. 208–9) all unmis-
takably point to a threat and an anxiety in relation to the Muses’

80
Ledbetter (2003) 44.
81
Thalmann (1984) 149; Clay (1988) 328. Such qualifications typically act as ‘let-out
clauses’, explicating why gods do not always exercise their capacities, West (1966) 163.
82
Clay (1984) 34 and passim; cf. (2003) 134–6.

83
 Hesiodic Epistemology
address of helpless dependence on the unforeseeability and fickle-
ness of (female) divine whim.
Hecate also provides a parallel as a female divinity whose
powers to affect mortal men both beneficially and malignantly
are rooted in her exceptionally close association with Zeus
(Th. 411–28, 448–52). Zeus’ ascension and subsequent distribu-
tion of provinces, powers and honours (τιμαί) throughout the
divine sphere is the Theogony’s announced theme (Th. 112–13)
and the climax of its narrative (881–5). The proem recapitulates
the same theme conjointly with the Muses’ procession to Zeus
(νισομένων πατέρ’ εἰς ὅν· ὁ κτλ, Th. 71–4). We need hardly labour
the point that, within Zeus’ world order, the Muses, like Hecate,
can possess and exercise their powers only through the sanction of
Zeus, with whom they similarly enjoy an exceptionally close
association.83
The deities of archaic poetry are, quite generally, the dispensers
of both good and bad. This role is most pronounced for Zeus and
for those female divinities whose powers are especially associated
with Zeus’ sanction, such as Hecate, the Moirai and the Muses.84
It is within this broader theological framework that Hesiod’s
Muses declare their ability to dispense both truths and falsehoods.
Hesiod occasionally resolves the doubleness of divine dispensa-
tion by isolating different divinities hitherto identified and tracing
invariably beneficent effects to one side and invariably pernicious
effects to the other.85 Like the Homeric Sirens and the Thriai,
however, the nine Muses comprise a unified divine force of the
same mind (ὁμόφρονας, Th. 60) and would resist any attempt to
isolate Muses of truth (28) from Muses of falsehood (27).
Of course, not all divine dispensations are capricious in the
Hesiodic corpus. It is a central theme and contention of the
Works and Days that Zeus dispenses personal and civic goods
and evils in response to just and unjust behaviour. First, however,

83
See esp. Th. 1–4; 25 (introducing Th. 26–8), 71–6; cf. 36–7, 51.
84
Numerous examples could be cited. See, most famously, Zeus’ two jars of good and bad
gifts: Il. 24.527–33; cf. e.g. Op. 3–8, 166–73, 667–9. Moirai: Th. 904–6; cf. Sol. fr.
13. 63–4 W. Muses: Od. 8.63; cf. West (1966) 408.
85
Op. 11–26 (on the two Strifes); Th. 869–80 (dividing beneficent winds from arbitrary
(872), variable and deleterious winds; contrast Op. 643–5).

84
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
the Theogony and Works and Days differ pertinently on this point.
The Theogony does reflect on divine dispensations of good and
bad to mortals (as with Hecate) and yet shows little or no trace of
this meritocratic aspect of divine dispensation. Indeed, explicit
examples of the dispensation of good in response to just behaviour
occur only in the Works and Days.86 Gods other than Zeus play
only a strikingly attenuated role in the Works and Days, which
focuses almost to their exclusion on Zeus’ justice-based interven-
tions in human life.87 We hear nothing, for example, of Hecate’s
dispensations.88 Second, the theodicy of the Works and Days itself
falters in a very instructive way when Hesiod moves from
a universal statement of Zeus’ all-seeing vigilance to reflect on
his own personal predicament and relationship with the divine
(Op. 267–73):
πάντα ἰδὼν Διὸς ὀφθαλμὸς καὶ πάντα νοήσας 267
καί νυ τάδ’, αἴ κ’ ἐθέλῃσ’, ἐπιδέρκεται, οὐδέ ἑ λήθει
οἵην δὴ καὶ τήνδε δίκην πόλις ἐντὸς ἐέργει.
νῦν δὴ ἐγὼ μήτ’ αὐτὸς ἐν ἀνθρώποισι δίκαιος 270
εἴην μήτ’ ἐμὸς υἱός, ἐπεὶ κακὸν ἄνδρα δίκαιον
ἔμμεναι, εἰ μείζω γε δίκην ἀδικώτερος ἕξει.
ἀλλὰ τά γ’ οὔπω ἔολπα τελεῖν Δία μητιόεντα. 273

The eye of Zeus sees all things and marks all things,
And now these things too, if he so wishes, he perceives, nor does it escape him,
What kind of justice this city contains.
But, as things are, may I not myself be just among people,
Nor my son, for it is a bad thing for a man to be just,
If the more unjust is going to have greater justice.
But I do not yet fear89 that Zeus the Counsellor will bring these things to fruition.
The resonant conditional ‘if he so wishes’ suddenly renders Zeus’
perception and subsequent treatment of justice and injustice
dependent on the unforeseeable and inscrutable condition of the
god’s mood (Op. 268–9). What follows leaves no doubt that the
qualification reflects a crack in the theological apparatus that
86
This poem too gives far more weight to the punishment of injustice, e.g. Op. 180–201,
218–24, 238–66, 282–5, 325, 333–4, etc.; in the Theogony: Th. 220–2, 231–2. But note
Op. 280–1 and Zeus’ rewards to the just city in Op. 225–37.
87
See Clay (2003) 142–3.
88
In Ch. 2.4, we will return to these divergences (at least in their emphases) between the
two poems.
89
Or: ‘anticipate’.

85
 Hesiodic Epistemology
underlies Hesiod’s campaign to promote the just life against the
unjust. Hesiod expresses a wish not to be just himself, nor to have
a just son, since justice is itself an evil if the more unjust will
receive greater justice (Op. 270–2; contrast 218). His concluding
remark, that he does not yet fear or, perhaps, anticipate that Zeus
will bring this scenario to ultimate fruition (Op. 273), does nothing
to dispel the preceding crisis of confidence and, in fact, only
further corroborates the overall impression of theological ambiva-
lence. Most interestingly for us, Hesiod’s loss of heart itself is
expressed with the same ominous vocabulary that qualifies the
Muses’ dispensations of truths and falsehoods (‘if he so wishes’,
Op. 268; ‘whenever we wish’, Th. 28).90
Hesiod describes the Muses’ influence as a ‘gift’ (ἱερὴ δόσις,
Th. 93–5; cf. 103–4). This description further implicates the
Muses in the unsettling doubleness of divine dispensation.91
The Olympians are formulaically ‘gods, givers of good things’
(δωτῆρες ἐάων, e.g. Th. 46, 111, 633, 664; cf. Od. 8.325), and
divine gifts are often providential (Op. 320, 549; cf. Th. 958; Op.
126). As before, however, this is only one side of a variable and
irresistible dispensation. In the Iliad, Paris states that no mortal
can reject the gods’ gifts even though none would accept them
willingly (Il. 3.65–6). In the Homeric Hymn to Demeter, divine
gifts constitute the inexorable, grievous yoke (cf. Th. 26) on mortal
necks which is the human condition (h.Hom. 2.216–17, 147–8).92
Again, the closer we come to Hesiod’s personal predicament, the
more pronounced the harsh duality of divine dispensation
becomes. To our own Iron Race, the gods ‘give’ grievous cares
(Op. 178). Our consolation – the best we can hope for – is the
admixture of good things with bad (Op. 179). Again, Hesiod traces
his father’s arrival to the place from which he himself now sings
(τύιδ’, Op. 635) to Zeus’ ‘gift’ of poverty (637–8, cf. 717–18).
The gods’ most interesting gift, however, is ‘Woman’
(Theogony) or ‘Pandora’ (Works and Days). By considering
Pandora, and the issues surrounding her fabrication by the gods,
90
Hesiod is, of course, not alone in reflecting on the tension between theodicy and Zeus’
actual dispensations; see e.g. Theognis 373–82, 731–52.
91
Following the lead of Pucci (1977) 3, 5–6 n.1, 28.
92
Cf. h.Hom. 3.189–93; Thgn. 444–6; Eur. Alc. 1071; Sol. fr. 13.63–4 W.

86
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
we will be better positioned to pull together the strands of this
analysis of the Theogony.
The Theogony’s divine politics revolve around the threat of
goddesses who engineer the downfall of male patriarchs through
cunning intelligence (mêtis). Exploiting his sexual desire
(Th. 177), Gaia deprives Ouranos of his patriarchy and manhood
through a concealed trap, using Kronos as her agent (κρύψασα
λόχῳ, 174, δόλον, 175). The fickleness of female cunning is high-
lighted when Kronos’ own overthrow is similarly credited to
Gaia’s trickery (Th. 494). Rhea, relying on Gaia’s and (the emas-
culated) Ouranos’ mêtis (Th. 471), conceals Zeus just as Gaia had
concealed Kronos (Th. 482, cf. 174) and presents Kronos with
a rock in swaddling (485). Rhea, in other words, passes off
a verisimilitudinous yet specious counterfeit for the genuine arti-
cle and Kronos’ downfall is expressly traced to his inability to
suspect or apprehend the difference between the two (Th. 488–91).
Zeus stops the process by out-tricking, ingesting and appropriating
the pregnant Mêtis herself, with whom he then devises ‘both good
and bad’ (Th. 886–900). His success in retaining power is repeat-
edly traced to his cognitive superiority and vigilant apprehension
(Th. 653–7, 730–1, 837–8). As an established ruler, Zeus enlists
the goddess Styx to curb, through the consequences of perjury, the
destabilising danger of falsehoods (ψεύδεα) within the divine
sphere (Th. 783–805). Following his ascension, moreover, Zeus
fixes under Parnasus, in ‘sacred Pytho’, the very same stone with
which Kronos was deceived ‘to be a sign (σῆμα) hereafter, a marvel
(θαῦμα) for mortal human beings’ (Th. 498–500). Zeus’ gesture
encapsulates what will imminently become central with Pandora.
Having appropriated and neutralised divine and especially female
deceptiveness within the divine sphere, Zeus transfers it to the
mortal sphere. Delphi is a place of oracles (note σῆμα, Th. 500).93
The stone is itself the instrument and symbol of female divine
deception. It is itself a verisimilitudinous but specious counterfeit,
which was passed off as the genuine article. We cannot but recall
the Muses’ own verisimilitudinous falsehoods. And this stone is

93
For ‘sacred Pytho’ and Apolline divination, cf. Od. 8.79–81; h.Hom. 24.1–2; with West
(1966) 303.

87
 Hesiodic Epistemology
what Zeus presents to mortal men as the epitome of his cosmic rule
and of the dynamic which governs the communication between
mortal and divine.
The stories of Prometheus and Pandora form a single episode
(Th. 535–616; Op. 42–105).94 A feast in Mekone marked some
type of foundational separation between gods and men: ‘when
gods and mortal men were separated (ἐκρίνοντο) in Mekone . . . ’
(Th. 535–6).95 Prometheus attempts to trick Zeus into accepting an
inferior portion of the food. The ensuing action determines the
norms and strictures of human life: sacrifice, work, sex and mar-
riage, and a mixture of good and evil. Zeus’ choice of the worse
portion directly explains our current sacrificial relationship with
the gods (Th. 557). Indirectly, it spells also the inexorable neces-
sity of work by bringing about his later denial of fire to mortals
(δόλου μεμνημένος αἰεί, Th. 562) and his concealment of the means
of effortless livelihood (βίον, Op. 42–50). Most importantly, it is in
retaliation to Prometheus’ subsequent theft of fire that Zeus
introduces Woman into human life, dispensing ‘a beautiful evil
in exchange for the good’ (Th. 585, 570, 602). Zeus’ encounter
with the Titan Prometheus and the subsequent fabrication of
Pandora thus frame the establishment of the relationship between
mortal and divine. As an insightful scholiast observes: ‘In Mekone
it was decided (ἐκρίνετο) what is a god and what a human being.’96
Prometheus’ trickery consists again in attempting to pass off
a verisimilitudinous but specious counterfeit for the genuine
article, concealing the bare bones with rich fat and the rich
innards with skin (Διὸς νόον ἐξαπαφίσκων . . . καλύψας . . .
καλύψας, Th. 537–41). Lines 550–2 of the Theogony insist that
Zeus sees through the trick and chooses the worse portion delib-
erately. This insistence is not obviously compatible with Zeus’
subsequent wrath upon seeing the trick (ὡς ἴδεν, Th. 554–5,
cf. 558).97 His position as always the master and never the victim

94
On shifts in emphasis and the complementarity between the accounts in the two poems,
see Canevaro (2015) 108–14.
95
Clay (2003) 101–4; Vernant (1982); ἐκρίνοντο may but need not carry a legal sense, West
(1966) ad loc.
96
Σ Th. 535.1–2 Gregorio.
97
Cf. West (1966) 321. On the inconsistencies, see Loney (2014).

88
 2.2 The Muses’ Address and Divine Dispensation in Hesiod
of mêtis is, at any rate, momentarily compromised when
Prometheus successfully conceals fire in a hollow fennel-stalk
(λαθὼν Δία, Op. 52, 55; cf. Th. 565–9) but it is confidently
vindicated both by Prometheus’ ultimate punishment (Th. 521–5,
613–16) and, more importantly, by the mortal inheritance of their
divine struggle.
Pandora ‘incarnates the problem of imitation’.98 This new
Woman is specious verisimilitude, the deceptive likeness of a
chaste maiden (παρθένῳ αἰδοίῃ ἴκελον, Th. 572; Op. 72). Mortal
men take pleasure (τέρπωνται) in unwittingly embracing their
own evil (Op. 58). Her unmistakable affiliation with mêtis is
most explicit in the Works and Days, where, in language reminis-
cent of the Muses (Th. 27), Hermes endows her with ‘deceits
(pseudea), devious discourses and a thievish character’
(Op. 78).99 Pandora’s attributes and adornments are throughout
erotically charged.100 Correspondingly, the gods explicitly
involved in her making are predominantly those most closely
associated with mêtis or with erotic power: the Theogony specifies
Hephaestus and Athena, while the Works and Days adds
Aphrodite, Hermes, the Graces, Peithô and the Hôrai. Notably,
the Muses are associated with the same erotic imagery as Pandora,
Aphrodite and other such threateningly female figures.101
Zeus’ initiative and oversight is emphasised (Th. 570–2, 580;
Op. 69–71, 79, 99). The Works and Days thus only renders explicit
the already implicit status of Woman in the Theogony as the quin-
tessential divine gift (Πανδώρην κτλ, Op. 81–2).102 Hephaestus
98
Bergren (1983) 75.
99
Pandora is initiated into clever weaving (Op. 64), a female art rich with connotations of
cunning intelligence, cf. Od. 13.303 (μῆτιν ὑφήνω) etc., with Murnaghan (1995) 64;
Slatkin (1996) 234–5.
100
Th. 570–84; Op. 60–82 (καλὸν εἶδος ἐπήρατον . . . χάριν . . . ὅρμους χρυσείους (cf. h.
Hom. 6.11, of Aphrodite) κτλ).
101
The Muses ‘dance with slender feet’ (Th. 3; cf. Th. 195 of Aphrodite) and bathe their
‘tender skin’ (Th. 5; cf. ἁπαλόχροος, Op. 519). Their golden frontlets (Th. 916) recall
Pandora (στεφάνην χρυσέην, Th. 578) and ‘golden Aphrodite’ (Th. 822; h.Hom. 6.7–8).
The erotic Graces (cf. Th. 907–11) and Desire (Himeros, cf. Th. 132, 177) neighbour
the Muses (Th. 64; cf. Th. 7–8, 70) and attend Aphrodite (Th. 201–2; h.Hom. 5.61; Od.
8.364). The provinces of both Aphrodite (Th. 206) and the Muses encompass ‘pleasure’
(τέρψις ἀοιδῆς, Th. 917, cf. Th. 37, 78). See further Marquardt (1982) 1–3, 8–9 and, for
the close association between Muses and Nymphs, Marquardt (1982) 9–10; West
(1966) ad Th. 7, 25.
102
Cf. Op. 57, 85–9 (δῶρον . . . Ζηνός . . . κακὸν θνητοῖσι); Th. 510–14.

89
 Hesiodic Epistemology
unveils Woman in a congregation of all gods and men but, con-
spicuously, she is pronounced ‘intractable’ specifically for mortal
men (ἀμήχανον ἀνθρώποισιν, Th. 588–9). She incarnates for mortal
men the very kind of trick (πῆμ’ ἀνδράσιν . . . δόλον, Op. 82–3)
which was the undoing of an Ouranos or a Kronos (δόλον, Th. 175;
δολωθείς, Th. 494). This intractableness of Woman is explicitly
linked to her status as the origin of the ‘destructive race’ of mortal
women (Th. 590–1) and, concomitantly, as the introducer of all
types of evil and suffering into human life (Th. 592–9; Op. 90–104).
Thus, even though Hesiod does not think of dissimulating cunning
intelligence as the prerogative of female divinities and women,
there is no question that he considers it importantly characteristic of
them.103 By ingesting Mêtis herself, Zeus appropriated and tamed
the unsettling power of cunning, female deceptiveness. By creating
Woman, he transfers this unsettling power to the mortal sphere as
a necessary evil for mortal men, who are constrained by the need for
assistance in their old age and for patrilineal continuity (‘high-
thundering Zeus set up women as an evil for mortal men etc.’,
Th. 600–12). Through this dual process, Zeus attained and stabilised
his patriarchal, cosmic rule and, in the wake of Mekone, established
the contrast and the relations between gods and men – or ‘what is
a god and what a human being’.

2.3 The Theogony: Conclusions

It is a textual fact that reassurances on the Muses’ part nowhere
expressly accompany their address at Theogony 26–8. Our analy-
sis of these lines and their immediate context invalidated the
prevailing dismissal of this textual fact as mere ellipsis.
The Muses produce no criteria to demarcate their true (Th. 28)
from their specious (27) melodious articulations. Such markers as
the bestialising abuse which they direct against Hesiod (Th. 26)
and Hesiod’s paradoxical ‘divine human-voice’ (Th. 31–2) accent-
uate the poet’s abiding inability to demarcate them independently.
In turn, Hesiod’s very recognition of both true and false inspired

103
One interesting exception and special case is Nêmertês (‘Unerring’), who ‘has her
father’s mind’ (Th. 262), the kind, truthful and ‘unerring’ Nereus (Th. 233–6).

90
 2.3 The Theogony: Conclusions
poetry (Th. 27–8), along with such resonant markers as ‘whenever
we wish’ (28) and ‘ready-of-speech’ (29), put pressure on any
inspired poet who would combine an admission of personal ignor-
ance with a simultaneous truth-claim for his verses.
We acquire a deeper understanding of the Muses’ address,
moreover, by analysing it within the Theogony’s broader theolo-
gical conception of the relation between (i) mortal and divine and
(ii) male and female. The intractable threat of deceit at the hands of
female cunning intelligence anchors Hesiod’s account of what is
a god and what a human being. It forms a central part of the divine
dispensation of good and bad gifts – and, ultimately, of what it
means to live as a man – within Zeus’ world order. Our inability to
identify when the Muses grant truths and when specious counter-
feits, therefore, is only one, poetic-epistemological aspect of
a coherent and holistic conception of the human condition.
The intractability and indispensability of the specious counterfeit
which is Woman is another, parallel and similarly Zeus-sanctioned
aspect of that same condition.
The Theogony’s narrative, by associating Gaia, Rhea and Woman
with the same cunning intelligence which the Muses master, or by
explaining as it does the stone under Delphi, invites us to revisit our
understanding of what the Muses said to Hesiod ‘first of all’. How
are men to deal with Muses or, indeed, with Woman? They cannot
emulate Zeus by ingesting the goddess Mêtis or enlisting Styx. But
they can at least avoid the gullible unawareness of an Ouranos or
a Kronos. The Theogony traces the downfall of male gods to their
blithe acquiescence in the deceptively specious favours of cunning
female goddesses. It is not credible that as sophisticated, reflective
and frequently anxious a poet as Hesiod framed this very same
poem by first of all acquiescing blithely in the gifts of cunning
female goddesses, which they themselves caution may be true but
may be deceptively specious. The upshot of this, however, is that the
enterprise of pushing our understanding of the origins and govern-
ing principles of the world beyond the narrow limits of our immedi-
ate experience – of transcending, epistemically, our status as ‘mere
bellies’ – is significantly and incorrigibly compromised. The divine
origin of Hesiod’s verses means that they could be true. But he
makes no guarantees.
91
 Hesiodic Epistemology
The Muses’ address articulates a theological-epistemological
framework to which the ensuing narrative is internal. That
narrative, however, continually provides us with exegetical
and prescriptive markers with which to revisit and enrich our
understanding of their address. Importantly, Hesiod and his
audience would not seem to suffer the same epistemic paralysis
when drawing such exegetical and prescriptive inferences from
accounts whose descriptive truth-value they are powerless to
evaluate. The story of Epimetheus, for example, says that he
became an evil for mortal men by unthinkingly accepting from
Zeus the gift of Woman. There is an intrinsic judiciousness in
the principle that one should not presuppose the kindness of
a gift the nature of which one is unable to assess and the favour
of whose giver one must not presume (Th. 511b-14; cf.
Op. 85–9). Nor, correspondingly, should one acquiesce in the
truth of the Muses’ gifts, whose truth-value and the truthful-
ness of whose dispensers at any given moment one is unable to
assess. This prescriptive and didactic insight, that we must
approach the Muses’ gifts with caution, is one which we can
evaluate and determine for ourselves on the basis of their
address. After all, the Muses could not deceitfully misrepresent
themselves to us as the type of divinities who may deceitfully
misrepresent things to us while, in fact, never deceitfully mis-
representing things to us.
Further considerations show the explanatory appeal of Hesiod’s
view of divine dispensations and help explain why he should
construe himself in the first place as the recipient of this divine
address and approach its evaluation as he does. Given the assump-
tion of divine management, Hesiod’s theology of the often incon-
stant dispensation of good and bad explicates and is corroborated
by (perceived) experience. We may outline a rough inferential
move (without, of course, ascribing it to Hesiod as a conscious
argument):
1. The gods govern the world along with its human inhabitants.
2. If a human succeeds and experiences benefits, some god(s) dispensed
good to him. If a human fails and experiences harms, some god(s)
dispensed bad to him.

92
 2.3 The Theogony: Conclusions
3. Different and often the same humans both succeed and fail and
experience both benefit and harm.
4. Therefore, either different or the same gods dispense both good
and bad.
This universal, theological framework, in turn, supports the
notion that the Muses inconstantly dispense both truths and
falsehoods as a special case of the general scheme. Since, more-
over, the Muses (like Hecate) give little hint that their dispensa-
tions of good and bad respond systematically to mortal merit and
demerit, divine inspiration inevitably assumed the spectre of
capriciousness (‘whenever we wish’). The gendered aspect of
the erotically charged Muses and of divine politics, in turn, trades
on Hesiod’s complementary perception of mortal women as
worthy of suspicion, as well as on the already traditional notion
of Muses – and Sirens – as female deities.104 Finally, we may
further accentuate the explanatory force of the Muses’ address by
granting to Hesiod the observation that different poetic accounts
of the same, unknowable matters are sometimes inconsistent.105
Given this point and, moreover, the premise that poets who sing
about matters of which they are ignorant do so through
inspiration,106 it follows that not all inspired poetic narratives
are true and, therefore, that the relevant gods dispense both truths
and falsehoods (Th. 27–8).
We saw that the Muses’ address destabilises certain tradi-
tional theological and epistemological attitudes towards Muse-
given inspiration. Hesiod’s critique did not, however, emerge ex
nihilo. We find in the epic tradition parallel (if more partial and
embryonic) attitudes towards theology, epistemology, poetry
and (female) cunning intelligence. Homer’s Sirens, for example,
echo the Muses’ anaphora ‘we know . . . we know’ (ἴδμεν, Od.
12.189–91). Furthermore, while their statement in line 186 may
be true (‘no one has yet sailed past here with his black ship’),
104
Note the cunningly dissimulating, Pandora-like woman of Op. 373–4 who motivates
a universal, prescriptive call for suspicion and wariness (Op. 375), with Canevaro
(2015) 115–17; cf. e.g. Th. 590–612; Op. 586.
105
This observation is most probably conveyed by Op. 11 (‘so there was not just one birth
of Strife after all’: this corrects Hesiod’s own previous account of Strife in Th. 225–32)
and certainly (on the subject of Dionysus’ birth-place) by h.Hom. 1.2–9 (ψευδόμενοι).
106
See Ch. 2.2.

93
 Hesiodic Epistemology
line 187 (‘before, that is, he has heard etc.’) slides into specious
deceptiveness.107 Menelaus recounts how Helen stood outside
the Wooden Horse, mimicking the voices of good Greek wives
(ἴσκουσ’, Od. 4.278–9). His narrative provokes us to ask
whether Helen’s own preceding story, which antithetically
touted her loyalty upon recognising the disguised Odysseus
(Od. 4.238–64), might itself comprise a specious counterfeit.108
In a manner reminiscent of Helen, Aphrodite renders herself
‘like’ a chaste maiden (ὁμοίη, h.Hom. 5.82). And we may also
recall the Delian maidens, who, in their enchanting songs, know
how to mimic (μιμεῖσθ’ ἴσασιν) the voices of all humans (h.Hom.
3.156–64). In the Homeric Hymn to Hermes, we find in Hermes
a master of deceptive cunning intelligence and, simultaneously,
theogonic poetry.109 Odysseus too is both a champion of
specious counterfeits and a poet-like figure.110 Indeed, it is in
reference to one of his Cretan Lies that Eumaeus expressly
compares Odysseus to an enchanting poet (Od. 17.514–21)
and in reference to another that his specious, verbal dissimula-
tions are styled ‘falsehoods . . . like verities’ (ψεύδεα . . .
ἐτύμοισιν ὁμοῖα, Od. 19.203).111 Even as he avows that he is
personally confident that Odysseus narrated his descent
into the Underworld truly – and in the manner of a bard –
Alcinous reflects on how the recipients of certain false
narratives are powerless to assess their truth-value (ψεύδεα,
Od. 11.363–9).112 The Odyssey too thus gestures towards desta-
bilising questions concerning the status of (inspired but) unver-
ifiable narrations.113

107
On affinities between the Sirens and Hesiod’s Muses, cf. Doherty (1995) 83; Clay
(2003) 60; Ledbetter (2003) 27. Both Calypso and Circe are first introduced singing and
weaving (Od. 5.61–2; Od. 10.221–3); cf. Nagler (1996) 152.
108
Cf. Goldhill (1991) 62–3. Helen’s first word: ἴδμεν (interrogatively, Od. 4.138).
109
Theogony: h.Hom. 4.54–9, 425–33, 450–5; mêtis: passim, note 368–86.
110
Od. 11.367–9; Od. 21.406–9, cf. Moulton (1977) 153. Walsh (1984) 129 lists Th. 27–8
among later conceptions of poetry influenced by the image of deceitful Odysseus as
bard-like; cf. Goldhill (1991) 67–8.
111
The formula recurs only of Nestor (Thgn. 713–14), who delivers an extended enco-
mium of mêtis in Il. 23.304–50. For the classic treatment of mêtis and its cultural
currency, see Detienne and Vernant (1978).
112
Cf. Schein (1996) 18; Slatkin (1996) 229; Walsh (1984) 20; Goldhill (1991) 47–8.
113
On Homeric poetics and the question of truth in poetry, see further Halliwell (2011)
36–92.

94
 2.4 Epistemological Optimism in the Works and Days
2.4 Epistemological Optimism in the Works
and Days and Final Remarks
The proem to the Works and Days is worth citing in full:
Μοῦσαι Πιερίηθεν ἀοιδῇσι κλείουσαι, 1
δεῦτε Δί’ ἐννέπετε, σφέτερον πατέρ’ ὑμνείουσαι,
ὅν τε διὰ βροτοὶ ἄνδρες ὁμῶς ἄφατοί τε φατοί τε,
ῥητοί τ’ ἄρρητοί τε Διὸς μεγάλοιο ἕκητι.
ῥέα μὲν γὰρ βριάει, ῥέα δὲ βριάοντα χαλέπτει, 5
ῥεῖα δ’ ἀρίζηλον μινύθει καὶ ἄδηλον ἀέξει,
ῥεῖα δέ τ’ ἰθύνει σκολιὸν καὶ ἀγήνορα κάρφει
Ζεὺς ὑψιβρεμέτης, ὃς ὑπέρτατα δώματα ναίει.
κλῦθι ἰδὼν ἀίων τε, δίκῃ δ’ ἴθυνε θέμιστας
τύνη· ἐγὼ δέ κε Πέρσῃ ἐτήτυμα μυθησαίμην. 10

Muses from Pieria who glorify in songs

Come here and speak of Zeus, singing hymns of your father,
Through whom mortal men are alike unfamed and famed,
Unspoken of and spoken of by the will of great Zeus,
For easily he strengthens, and easily he crushes the strong,
And easily he diminishes the conspicuous and increases the inconspicuous,
And easily he straightens the crooked and withers the proud,
Zeus the high-thunderer, who dwells in the highest halls.
Hearken, watching and listening, and straighten the verdicts with justice
Yourself; but as for me, I would speak truths to Perses.114
Several scholars maintain that, in these lines, Hesiod denies the
Muses a role, or at least a significant role, in the Works and Days.
By contrast with the Theogony, the argument goes, the Works and
Days addresses human affairs, which Hesiod can grasp and relay
independently of divine mediation. In the proem, then, Hesiod
establishes a division of labour. He (i) invokes the Muses to sing of
Zeus (Op. 1–8), (ii) prays to Zeus to uphold justice (9–10) and
(iii) affirms that he will independently tell truths to Perses (10).115
While we can appreciate its underlying insights, this interpreta-
tion is, as it stands, problematic. First, the verb in Hesiod’s con-
cluding statement (κε . . . μυθησαίμην) is not a future indicative but
an optative. The precise force of the verb is, indeed, difficult to
determine. It does not, however, commit Hesiod to a clear,

114
Alternatively: ‘let me speak truths to Perses’.
115
Clay (2003) 76–80; (2015) 114–15; Stoddard (2004) 83–4, 146–7 n.24, 190–1;
Haubold (2010) 15–16, 29–30; Palmer (2013) 319; Canevaro (2015) 10, 101.

95
 Hesiodic Epistemology
straightforward truth-claim (‘I will speak truths to Perses’).
The (aorist) optative with κε suggests, and certainly accommodates,
a softer and more tentative assertion: ‘and I would speak . . . ’ or
‘and I should like to speak . . . ’116 This line, moreover, culminates
an extended prayer, encompassing a sustained set of requests direc-
ted at both the Muses and Zeus (δεῦτε . . . ἐννέπετε . . . ὑμνείουσαι,
Op. 2; κλῦθι . . . ἴθυνε, Op. 9).117 The context of κε . . . μυθησαίμην,
therefore, evokes also the optative’s hortatory sense: ‘but as for me,
may I speak . . . ’ or ‘let me speak . . . ’.118 Hesiod’s exhortation is
reflexive (he is exhorting himself), but its context – an invocation of
the Muses to come and sing, and a prayer to Zeus to hearken and
watch over – suggests also a call for the divine favour which will
facilitate his enterprise.
It remains the case, however, that, by contrast with the
Theogony, line 10 of the Works and Days lays conspicuously
more weight on the prospect that the ensuing narrative may com-
prise truths. The pronoun ‘I’ (egô, Op. 10), moreover, does place
novel emphasis on the mortal poet as distinct (but not thereby
separated) from the Muses. Furthermore, the Works and Days does
indeed explore prominently certain human affairs the prescriptive
and descriptive ‘truths’ of which Hesiod can work out and grasp
independently.119 The primarily didactic aspect of the ‘truths’
(etêtyma) in line 10 is accentuated by the juxtaposition with
Zeus’ just governing in line 9.120 Throughout the poem, however,
Hesiod explores and explains the human sphere through its deter-
minative relationship with the unknowable, divine sphere. It would
be an oversimplification to say that, whereas the Theogony deals
with gods and divine history, the Works and Days addresses the
116
Smyth (1972) noo.1824, 1826. Even if a straightforward truth-claim introduced the
Works and Days, this would only have highlighted the absence of one introducing the
Theogony, contra West (1966) 162.
117
κλῦθι is standardly used for ‘hear my prayer’, e.g. Il. 1.37; Il. 10.278; Od. 4.762, etc.
118
Smyth (1972) no.1830.
119
This is perhaps reflected in Hesiod’s ideal of the ‘most excellent man (πανάριστος),
who thinks of everything for himself, considering what will be better later and in the
end’ (Op. 293–4). The second best thing is to obey good advice (Op. 295). Canevaro
(2015) emphasises the importance of this ideal for Hesiod. Still, this intellectual and
ethical call for critical reflection and deliberation does not amount to a categorical cry
of liberation from the need for divine aid or guidance (as Canevaro (2015) 130 herself
allows). Recall also Phemius’ words at Od. 22.347–8.
120
Cf. Op. 286 (ἐσθλὰ . . . ἐρέω . . . Πέρση).

96
 2.4 Epistemological Optimism in the Works and Days
conditions of human existence and mundane affairs. While we can
speak of a shift in emphasis from one to the other, both poems are
interested in both gods and humans, and always treat of one with
an eye to the other.121 Accordingly, Hesiod invokes the Muses to
sing, not simply of divine affairs, but precisely of how Zeus’
dispensations of success and failure determine human life and
endeavour (n.b. Op. 2–4).122 That is, far from expelling or margin-
alising the Muses, Hesiod invokes them to approach (δεῦτε . . .
ἐννέπετε . . . ὑμνείουσαι, Op. 2) and articulate a central aspect of the
central preoccupations of the ensuing poem.123 Hesiod indeed
differentiates between Zeus’ activity of just governing and his
own activity of singing (τύνη· ἐγώ, Op. 9–10). He does not,
however, dissociate his singing from divine influence or favour.
Hesiod’s advice to Perses on the subject of seafaring (Op. 646–93)
demonstrates the Muses’ abiding significance in matters external to
the poet’s experience. More importantly, these lines, which are
surprisingly often ignored in the literature on Theogony 27–8, con-
vey a striking take on the Muses’ address to Hesiod.
Hesiod will reveal to Perses ‘the measures of the sea’ (δείξω, 648)
despite having no expertise in seafaring or ships (649). As though it
was a tangential qualification, Hesiod recounts as the single, negli-
gible exception to his ignorance of sailing his journey across the short
distance between Aulis and Euboea to Amphidamas’ funeral games.
But the story of Hesiod’s brief voyage serves as a foil for a swift
epistemological argument (656–62): ‘ . . . there I declare that I gained
victory with a song, and carried off a tripod with handles. This
I dedicated to the Heliconian Muses, where they first set me on the
path of clear-toned song (τὸ πρῶτον . . . ἐπέβησαν ἀοιδῆς). (i) Such,
then, is the extent of my experience of many-bolted ships (τόσσον . . .
πεπείρημαι); (ii) yet even so I shall speak forth the mind of aegis-
bearing Zeus (Ζηνὸς νόον), (iii) for (γάρ) the Muses taught me how to
sing a wondrous song (ἐδίδαξαν ἀθέσφατον ὕμνον ἀείδειν).’

121
See similarly Most (2006) xliii. 122 Contra Stoddard (2004) 146–7, n.24.
123
The subject matter of the song which the Muses are invited to sing about Zeus is not,
therefore, merely ‘tangential’ to the poem’s central themes of work and justice, contra
Canevaro (2015) 101. Zeus’ cosmic governance and determination of the relations
between gods and men are what anchors the necessity and status of work and justice
(Ch. 2.2–3).

97
 Hesiodic Epistemology
What resurfaces with this argument, in a more sophisticated
elaboration, is the poet’s old admission of ignorance and simulta-
neous truth-claim. Hesiod (i) admits inexperience in seafaring and,
therefore, ignorance of the facts pertinent to an informed account
of it. He then rehabilitates his account by (ii) reformulating the true
object of the account as the ‘mind’ or ‘plan’ (noos) of Zeus, rather
than seafaring itself, and by (iii) justifying (‘for’: γάρ) his claim
that he will relay Zeus’ mind by tracing his ability to sing to the
Muses. (iii) supports (ii) only if the Muses’ teaching enabled
Hesiod, inter alia, to relay Zeus’ mind. (ii) finally rehabilitates
Hesiod’s account only if ‘the mind of Zeus’ encompasses, under
a different description, the same propositions and injunctions
which would make up a proper account of ‘the measures of the
sea’. Specifically, ‘the mind of Zeus’ must presumably encompass
the propositions and injunctions which make up Hesiod’s own
ensuing account which spans, for example, both descriptions of
relatively safe and dangerous seasons (e.g. 663–5, 670–7) and
such prescriptions as the advice to load a boat in springtime only
with the smaller share of one’s possessions (689–93).
This argument clearly embodies a reading, within the Hesiodic
corpus – although not in the Theogony – of the Muses’ initial
address. Hesiod dedicated his prize where the Muses first trans-
formed him (τὸ πρῶτον κτλ, Op. 658–9), i.e. where they first
taught him how to sing a wondrous song (Op. 662). His remark
refers to the same first encounter narrated in Theogony 22–34.
If from that occasion onwards the Muses have been enabling
Hesiod to give consistently true accounts of any matter external
to his experience, then they would appear to have been infusing
him with truths (Th. 28).124
The passage strikes a triumphant tone. While Hesiod’s journey
only underscores his lack of expertise in seafaring (οὔτε . . .
σεσοφισμένος, Op. 649), his victory in that journey exemplifies

124
Op. 662 (ἐδίδαξαν . . . ἀείδειν) in no way indicates that the Muses endowed Hesiod once
and for all with an ability to sing, which he now retains independently of their continued
favour and inspiration. The same aorist occurs in Th. 22 (Ἡσίοδον . . . ἐδίδαξαν ἀοιδήν)
where the noun, moreover, is functionally equivalent to the infinitive used in Op. 662;
cf. Th. 34–5. Both in the Theogony (104–15) and, I argued, in the Works and Days, the
Muses are still invoked to initiate the account.

98
 2.4 Epistemological Optimism in the Works and Days
the poetic expertise which compensates for his ignorance.125
Hesiod’s victory and subsequent pious dedication portrays
a happy, reciprocal and cyclical relationship of divine favour
followed by mortal offering. Above all, Hesiod uses unqualified
future indicatives to affirm that he will reveal the measures of the
sea and, put differently, speak the mind of Zeus (δείξω, 648; ἐρέω,
661). This epistemological triumphalism diverges from what we
have seen strong evidence to maintain is the Theogony’s consid-
ered and more circumspect epistemological stance.
We should not explain away the break in this passage of the
Works and Days with the Theogony’s position. One might insist
that Hesiod nowhere makes an express truth-claim for his account
of seafaring. And, indeed, he need not claim that the account
comprises truths and nothing but truths. Still, the conversational
implicatures of Hesiod’s future indicatives, and of the overall tone
of the passage, plainly amount at least to a broad truth-claim.
Alternatively, one might contend that we cannot infer a global
epistemological stance from Hesiod’s claims about the specific
topic of seafaring. But steps (i)–(iii) give a strong impression of
articulating a general, programmatic statement about the poet.
And why should seafaring warrant privileged optimism?
The fatal stakes and cultural baggage of seafaring (one recalls
Odysseus’ encounters with god-sent storms) render Hesiod’s self-
assurance here all the more provocative and pronounced. Finally,
no independent evidence supports isolating the excursus from the
rest of the Hesiodic corpus as an interpolation, ancient excisions
notwithstanding.126
If, then, one traces the mortal poet’s construal of his relation to
the Muses as it emerges from a synoptic consideration of the
Hesiodic corpus as a whole, the resulting picture is one neither
of unreflective self-assurance nor of a consistent scepticism but of
epistemological ambivalence. Notably, parallel ambivalent vacil-
lations between optimism and pessimism are visible both else-
where in the Works and Days and, in fact, within the account of
125
For this double reading of σεσοφισμένος, see Rosen (1990) 101–2 with n.14.
126
Having inspired the Contest between Homer and Hesiod, the excursus was rejected by
some commentators who saw in it an anachronistic reference to that later tradition, see
West (1978) 319.

99
 Hesiodic Epistemology
seafaring itself. We saw how, immediately after he affirms Zeus’
all-seeing vigilance, Hesiod conveys a momentary but telling
crisis of confidence in Zeus’ dispensations, and so in the just life
(‘if he so wishes etc.’, Op. 267–73). The intractable variability of
Zeus’ agency resurfaces again in Works and Days 483–4: ‘But the
mind of Aegis-bearing Zeus is different at different times (ἄλλοτε
δ’ ἀλλοῖος), and it is difficult for mortal men to grasp it.’ This
comment qualifies the reliability of Hesiod’s surrounding agricul-
tural advice.127 A scholiast wonders how it coheres with Hesiod’s
self-assured claim in line 661 that he will articulate the mind of
Zeus.128 And the scholiast is quickly vindicated. Having identified
a span of fifty days in which sailing is safe, Hesiod qualifies:
‘unless Poseidon, the earth-shaker, or Zeus, king of the immortals,
wish to destroy (εἰ δὴ μή . . . ἐθέλῃσιν ὀλέσσαι), for in them is the
fulfilment of good and bad alike’ (Op. 665–9). We are familiar by
now with the theological and epistemological resonances in
Hesiod of such conditions as ‘unless they wish’ when they qualify
divine dispensations of ‘good and bad alike’. With seafaring too,
then, Hesiod’s brief moment of self-assured optimism is swiftly
subjected to the inscrutability and unforeseeability of divine
volition.
We noted above that the Theogony shows little preoccupation
with the idea of gods dispensing good and bad in accordance with
mortal justice and injustice respectively. This theodicy is, how-
ever, a central theme and contention of the Works and Days.
(We cannot, moreover, dismiss this theological divergence
between the two poems as simply a matter of their dealing with
different themes: at Theogony 429–47, for example, the catalogue
of Hecate’s dispensations of successes and failures of all kinds
offered an obvious opportunity to trace them to mortal merit and
demerit.) If Hesiod’s break in the seafaring passage with the
Theogony’s epistemological stance should be explained by any-
thing more than fluctuations in the poet’s hopes concerning the
gods’ disposition towards him, I can see no better place to look to
than this theological difference between the two poems. It is
possible that, by promoting and systematising the notion of just

127
Cf. West (1978) ad loc; Canevaro (2015) 198. 128
Σ Op. 661a Pertusi.

100
 2.4 Epistemological Optimism in the Works and Days
dispensations of good and bad, Hesiod finally generated
a theological environment which rendered at least viable the
optimistic (and self-congratulatory) revaluation that to him the
Muses, always operating with Zeus’ sanction, dispense truths
(i.e. good). After all, Hesiod’s diatribes about justice and injustice
leave no doubt that he views his own conduct as just and worthy.
A connection between Zeus’ just dispensations and Hesiod’s epis-
temic predicament is arguably suggested by the close juxtaposi-
tion in the proem between (i) Hesiod’s prayer to Zeus that he
governs justly (Op. 9–10) and (ii) his hope – or implicit request –
that he would speak truths to Perses (Op. 10). The epistemological
tension between Hesiod’s self-assured promise to relay ‘the mea-
sures of the sea’ or ‘Zeus’ mind’ (646–62) and his subsequent
swift reversion to the intractable unforeseeability of Zeus’ storms
(665–9) would then map onto what we saw is a broader theological
tension which the Works and Days engenders by venturing sys-
tematically into problems of morality. For, even as the Works and
Days pursues the notion that divine dispensations of good and bad
correspond to mortal justice and injustice, it displays at telling
junctures abiding recognitions of the variability and inscrutability
of divine agency and the difficulty of maintaining this idea of
theodicy in the face of experience.
Any such explanation of the seafaring passage will be neces-
sarily speculative. And the excursus on seafaring, whether or not
we deem it explicable in some such manner (and independently
even of authorial considerations), is interesting for our purposes
above all for its very divergence from the epistemological stance
implicit in the Theogony. Considered side by side, the compet-
ing interpretations of the Muses’ initial address to Hesiod which
we encounter in the Hesiodic corpus show something important
about it. They demonstrate that the primary thrust of the Muses’
laconic and enigmatic address is precisely to raise but leave
unresolved the urgent and treacherous question of how the
poet and his audience should interpret it. In Theogony 26–8,
the Muses articulate the fixed epistemological framework within
which the poet operates, a framework in which his capacities
and limitations are determined by the nature of (i) the (episte-
mically superior) divine, (ii) the (epistemically inferior) mortal
101
 Hesiodic Epistemology
and (iii) the interactions between them. It is, however, essential
to the Muses’ address that it is indeterminate and amenable to
conflicting construals. As usual with divine communication, the
Muses do not linger to converse with Hesiod and elucidate their
words. Hesiod is left to assess for himself the ramifications of
their statement for his own case. The Theogony (along with
much of the Works and Days) implicitly but decisively develops
a cautious assessment and inscribes Hesiod’s interaction with
the Muses within a theological cosmos in which the goddesses’
address prescribes for the male poet wary circumspection rather
than blithe assurance. In Works and Days 646–62, conversely,
we find a strikingly optimistic revaluation of the Muses’
address, albeit a fleeting one, which quickly betrays a kind of
theological tension and uncertainty visible also elsewhere in the
poem. When we take a synoptic view of the Hesiodic corpus as a
whole, we find neither an unconcerned dogmatism nor a strictly
consistent scepticism, but ambivalence between competing epis-
temological stances. In the final analysis, what the Muses articu-
late for Hesiod is not an epistemological position but an
epistemological framework, which identifies the problem of
epistemology as the problem of understanding the nature of
the interactions between mortal and divine, but which is suffi-
ciently indeterminate to accommodate competing elaborations
of its fixed principles.
The only secure sense in which the Muses render Hesiod
epistemically privileged, therefore, is in enabling him to recog-
nise in an informed and disillusioned manner the complexity of
the poet’s epistemic predicament and the fixed factors which
determine the truth-value of his output. Crucially, far from cun-
ningly dissimulating it, the Muses forthrightly disclose to Hesiod
their capacity to deceive.129 Any subsequent evaluation and
revaluation of the ramifications of their words for his own case,
however optimistic or pessimistic, will thus always be informed
of – and, ultimately, haunted by – the prospect of the divine
dispensation of falsehoods.

129
Contrast the dissimulating Hermes: ‘I will tell the truth (ἀληθείην ἀγορεύσω) . . . I do
not know how to speak falsehoods’ (οὐκ οἶδα ψεύδεσθαι, h.Hom. 4.368–9).

102
 2.4 Epistemological Optimism in the Works and Days
We saw that, by introducing this prospect as he does, Hesiod
brings certain traditional and epistemological attitudes to a crisis
point (which, moreover, the short-lived triumphalism of the sea-
faring passage only accentuates). We will find in Xenophanes,
Parmenides and Empedocles radically divergent ways of conver-
sing with and moving beyond the epistemological challenge first
clearly isolated by Hesiod.

103
 3

XENOPHANES ON DIVINE DISCLOSURE

A N D M O RTAL I N Q U I RY

Xenophanes of Colophon wrote his verses in the late sixth and

early fifth centuries BCE. By his own account (DK21 B8, B45),
Xenophanes spent most of his adult life travelling all around
Greece, moving from city to city, performing his works and offer-
ing his wide-ranging reflections. Our own focus in this chapter will
be his thoughts on the question of divine disclosure. We will
ultimately see, though, how Xenophanes aims to locate the mortal
as an epistemic as well as moral agent within a holistic and
coherently conceived world order.
Cicero speaks of Xenophanes as the only one among the most
ancient philosophers who, while asserting the existence of the
gods, ‘did away with divination from its very foundation’
(diuinationem funditus sustulit). With the further exception of
Epicurus, divination has been otherwise accepted universally
(Div. 1.3.5 = A52). Aëtius similarly writes: ‘Xenophanes and
Epicurus did away with divination’ (Ξενοφάνης καὶ Ἐπίκουρος
ἀναιροῦσι τὴν μαντικήν, 5.1.1 = A52). Lesher (1992), who affirms
his acceptance of these reports of Xenophanes, cites favourably
Dodds’ evaluation of the implications of such acceptance: ‘If this
is true, it means that, almost alone among classical Greek thinkers,
he [Xenophanes] swept aside not only the pseudo-science of
reading omens but the whole deep-seated complex of ideas about
inspiration.’1 Both earlier and later commentators on Xenophanes’
relation to traditional theology similarly echo Cicero’s talk of
a traceless rejection.2
1
Dodds (1951) 181 (cf. Dodds (1973) 6), cited at Lesher (1992) 141, cf. 96, 114, 118,
123, 154.
2
‘A clean sweep of all the elements of popular belief which were hostile to his higher
standard’ (Gomperz (1906) 163), ‘Xenophanes schaltet die Götter im Einklang mit seiner
Theologie völlig aus’ (Kleingünther (1933) 41), ‘the entire framework . . . is here swept
away’ (Hussey (1990) 19), cf. Schrödinger (1954) 68 (‘clear away’); on divination: ‘par

104
 Xenophanes on Divine Disclosure and Mortal Inquiry
We may assume that Cicero and Aëtius are drawing on a
common source. Cicero’s diuinatio translates the Greek mantikê
(Div. 1.1.1), the two doxographers use parallel verbs (sustulit and
ἀναιροῦσι), and the association of Xenophanes and Epicurus
recurs in both.3 The extant fragments, however, contain no cog-
nate of mantikê and we cannot be certain that Xenophanes himself
used this vocabulary of divination. If not, the reports would most
probably derive from a reading of B18 (quoted and translated in
Chapter 3.2) as expressing a wholesale denial of divine disclosure.
And if our common source pounced on B18 in doxographic
eagerness to identify an exception to the otherwise near-
universal acceptance of divination,4 most modern scholars were
all too happy to inherit from antiquity a Xenophanes who cham-
pions independent mortal inquiry against divine interferences.5
The minority of scholars who object to this reading of B18 do
not ask what the role of divine disclosure in mortal inquiry might
be, or consider the implications of this role for our understanding
of Xenophanes’ reaction to divination.
Cicero’s and Aëtius’ reports invite a wholesale reconsideration
of the relation between mortal and divine in Xenophanes’ episte-
mology. We will engage with these reports most fruitfully by
analysing Xenophanes’ reactions to the theological and epistemo-
logical suppositions that underlie the concept of mantikê.
Xenophanes, we will see, indeed rejected divination along with
other traditional models of divine disclosure such as poetic
inspiration. But he was profoundly influenced by what he rejected.

la base’ (Bouché-Leclercq (1879–1882) 33); ‘altogether’ (Flower (2008) 8) (my

emphases).
3
The two passages give parallel accounts also of the Stoic and Peripatetic views. On the
Stoics: τὰ πλεῖστα μέρη τῆς μαντικῆς ἐγκρίνουσι (5.1.1); omnia fere illa defenderent (Div.
1.3.6); on Aristotle and Dicaearchus or ‘Dicaearchus the Peripatetic’: τὸ κατ᾽
ἐνθουσιασμόν . . . τοὺς ὀνείρους (5.1.4); somniorum et furoris reliquit (Div. 1.3.5).
4
Ancient doxographers are generally eager to oppose antithetical views, cf. Mansfeld
(1999b) 27, 30–1; Betegh (2010) 36–7. In both Cicero and ps.-Plutarch, the rejection of
divination by Xenophanes and Epicurus is juxtaposed to its (variously qualified) accep-
tance by Plato, the Stoics, Aristotle and Dicaearchus (Div. 1.3.5–6; DG 5.1.1.4).
5
Tellingly, Loenen (1956) 136 styles Xenophanes an ‘enlightened critic’ when defending
the traditional reading of B18 (cf. Gomperz (1906) 163; Dodds (1973) 4–5), while
Shorey (1911) 89, arguing against it, speculates that scholars have been ‘misled by . . .
partiality for the pre-Socratics’; more recently, cf. Curd (2002) 120–9; Lesher (2008)
468–9; Gregory (2013) 104.

105
 Xenophanes on Divine Disclosure and Mortal Inquiry
What does it mean to say that Xenophanes did away with
mantikê? What is the ‘whole deep-seated complex of ideas’ against
which the doxographers’ judgement is to be evaluated? We must not
overlook the possibility of a conceptual gap between Xenophanes
and his doxographers.6 In Chapter 3.1, I delineate the central,
pertinent suppositions underlying divination. While taking account
of later and ‘historical’ evidence, we will focus primarily on the
closest conceptual relatives of the later term mantikê in the epic
tradition, a tradition against which Xenophanes situated himself so
explicitly.7 This analysis undermines the anachronistic and mislead-
ing misconception, pervasive in the literature on Xenophanes, that
Greek disclosure and divination typically involved the direct trans-
mission of truths and knowledge to passive mortal recipients.
To understand Xenophanes’ reactions to it, we must appreciate the
role of active mortal agency and reasoning in the Greek discourse of
disclosure and divination.
In the first instance, this chapter offers a new interpretation of
the logic of Xenophanes’ remarks about divine disclosure in
B18.1. In Chapter 3.2, I argue that Xenophanes neither denies
categorically the reality of divine disclosure nor acquiesces in
traditional attitudes towards it. Rather, Xenophanes categorically
rejects the traditional notions of disclosure and expressly sup-
plants them with a novel alternative. In Chapter 3.3, I develop
Lesher’s (1983) convincing and neglected argument that
Xenophanes arrives at his scepticism, i.e. the negative aspects of
his epistemology, by rejecting traditional divinatory assumptions
(B34). But Lesher tells only half the story. In Chapter 3.4, I argue
that Xenophanes’ alternative notion of disclosure underlies his
positive reflections regarding what does lie within the scope of
mortal epistemology. More speculatively, I develop two alterna-
tive interpretations of the precise notion of purposiveness under-
lying Xenophanean disclosure. Xenophanes most probably
6
Not least given Cicero’s preoccupation with narrowly Stoic definitions of divination
(Div. 1.1.1, 1.5.9), and because the doxographic tradition which Cicero and Aëtius
reproduce and instantiate is notoriously prone to reduce earlier thought to later
categories.
7
See especially the criticisms of Homer and Hesiod in B10–12; see also B1.21–3; B14–16;
B26; B32; D.L. 9.18.20 = A1; D.L. 2.46 = A19; Timon apud D.L. 9.19.14 = A1; Sextus
PH 1.224 = A35; Plut. Reg. Apophth. 175c = A11.

106
 Xenophanes on Divine Disclosure and Mortal Inquiry
reconceptualises the notion of divine disclosure radically as the
view that the divine purposively facilitates all mortal experience
and belief-formation as part of its intelligent direction of the
cosmos and its inhabitants. It remains possible, though less likely,
however, that he maintains less idiosyncratically that the divine
guides particular mortals in particular circumstances.
Finally, in Chapter 3.5 we will ask how these conclusions about
Xenophanes can lend further nuance to our understanding of the
complexity of his engagement with the culturally and theologically
dominant models of divine disclosure manifested in divination and
poetic inspiration.8 Xenophanes reacts to divinatory models of
sagehood in a polemical but also appropriative manner. This reac-
tion casts light on Xenophanes’ own conception and representation
of himself as a sage and on the theological, epistemological and
social world view that he champions against his rivals.9
For Xenophanes too, as with Hesiod, understanding epistemol-
ogy means understanding the nature of, and the relations between,
mortal and divine. At the heart of this enterprise lies the question
of disclosure. At the outset, we must not let the fact that divination
inhabits the religious and intellectual margins of modern Western
culture blind us to its prominence in archaic and classical Greek
beliefs and practices.10 At the end, we should resist our temptation
to extend Xenophanes’ iconoclastic criticisms of what might strike
us as unpalatable superstitions to a wholesale denial that the divine
exerts an active influence on the epistemological equation.
Xenophanes’ polemic against theologically dominant models of
divine disclosure has indeed exercised a formative influence on his
epistemological thinking. But his response to ‘the whole deep-
seated complex of ideas’ surrounding disclosure is not a complete
‘sweeping-aside’ from the very foundation (funditus) but, in some
important respects, a radical and subversive transformation.

8
We will discuss further Xenophanes’ responses to Hesiod in Ch. 6.1.
9
In her interesting analysis of Xenophanes’ opponents and addressees, Gemelli Marciano
(2002) 92, n.19 surprisingly dismisses offhand the possibility of an engagement with
diviners.
10
See Flower (2008) 1–103; Johnston (2008) 1–17; cf. Burkert (2005) 29–30; Flacelière
(1976) 4. For some ancient testimonies to the universality and centrality of divination in
the Graeco-Roman world, see X. Apol. 13; Cic. Div. 1.1.1–1.3.5, 1.6.11–12; S.E.
M 9.132; Oenomaus apud Eus. PE 5.27.5.

107
 Xenophanes on Divine Disclosure and Mortal Inquiry
3.1 Divination
A few key points will preoccupy us here. (i) Divination presupposes
the pervasive contrast between divine epistemic superiority (often,
omniscience) and mortal epistemic limitations. More precisely,
communication between mortal and divine presupposes an episte-
mological world view in which the epistemic abilities and limita-
tions of mortals are conceived by contrast with and in reference to
the superior epistemic capacities of the gods. (ii)a Diviners most
typically form propositions concerning non-experienced states of
affairs on the basis of experienced events or objects intentionally
disclosed by a divinity for that purpose (omens, oracles, dreams).
(ii)b Since mortals can attain knowledge or insight concerning these
non-experienced states only on the basis of omens (or oracles or
dreams), those who have the ability and the expertise to infer from
them appropriately are taken to transcend the epistemic limitations
of all other mortals and to approximate the epistemic state of the
gods in particular and momentary respects. (iii) Such epistemic
approximation to the divine is associated with a more general
approximation to the divine.
Mantic disclosure typically consists in the following: the gods,
knowing p, cause the mortal agent to perceive an object or event
with the intention that an ideal observer will interpret that object or
event to mean, and therefore come to believe, p. The gods thus
communicate p. In Gricean terms, omens have ‘non-natural mean-
ing’. To cite Denyer, ‘they mean the facts they signify, not in the
way clouds mean rain, but in the way that a road sign means
a bridge ahead.’11 Underlying divine communication is the view
that the gods not only facilitate human existence by maintaining
our world order (e.g. celestial and seasonal processes), but are also
aware of and concern themselves with particular dilemmas haunt-
ing particular mortals.12 We may distinguish at least three senses
in which the objects of divination may be non-experienced.
P could refer to (1) an event in the past or future (‘Who will win
the Trojan war?’, ‘Were Agis’ blankets and pillows stolen?’),

11
Denyer (1985) 5.
12
Xenophon formulates the principle explicitly: Mem. 1.4.15–18; Cyr. 1.6.46; Symp. 4.
47–9; cf. Flower (2008) 105–6.

108
 3.1 Divination
(2) current facts inscrutable due to various practical limitations
(‘How many figs are in this fig tree?’, ‘Is Lysanias not the father of
Annyla’s unborn child?’), (3) questions entirely not amenable to
autopsy (‘What does Zeus will?’, ‘What actions incurred divine
wrath and which would allay it?’, ‘Was the first-fruit offering to
Apollo satisfactory?’).13
Already in Homer we find the indispensability of mantic
consultations ‘in both public and private’ (Cicero, Div. 1.2.3).
For Achilles, divination is required to discover what course of
action would end the pestilence (Il. 1.62–3), while Circe sends
Odysseus to Teiresias, the seer (mantis), to learn what path he must
follow to reach Ithaca (Od. 10.538–40). The goddess Circe uses
the same words to describe the inquiry about his journey which
Odysseus must put to the seer Teiresias as those which the goddess
Eidothea uses to describe the inquiry about his own journey which
Menelaus must put to the god Proteus.14 The fact that the mortal
Teiresias fulfils a parallel role to that of the god Proteus perhaps
brings out the particularly liminal status of the dead yet fully
cognitive Teiresias (Od. 10.490–5). But it also gestures at
a similar and more general liminality, characteristic of the diviner
as such. Achilles begins a reply to his omniscient mother by noting
its redundancy: ‘You know. Why, then, should I recount these
things to you when you know all?’ (Il. 1.365). Menelaus addresses
Eidothea (the Knowing Goddess) in similar terms: ‘but the gods
know all things’ (Od. 4.379). While he reserves omniscience for
the divine, however, Achilles accords Calchas sure knowledge
(μάντις εὖ εἰδώς, Il. 1.384–5) and so a certain approximation and
access to this privileged state. The poet similarly ascribes to
Calchas, qua augur, knowledge of ‘the things that are and the
things that will be and were before’ (Il. 1.69–70). Later sources
retain the notion that knowledgeable or omniscient gods willingly
provide signs only on the basis of which mortals can gain an

13
(1) Il. 2.303–30; Parke 272 no.27 (Dodona); (2) Hes. fr. 278 M-W; Parke 266 no.11
(Dodona); (3) h.Hom. 3.131–2; Il. 1.92–100; Hdt. 8.122. Chaerephon could ask the
Delphic oracle: ‘is any man wiser than Socrates?’ (Pl. Apol. 21a; cf. later in Philostratus’
biography of Apollonius of Tyana: ‘which is the most perfect philosophy?’, VA 8.19.
40–4).
14
ὅς . . . ἐλεύσεαι, Od. 10.539–40; Od. 4.389–90.

109
 Xenophanes on Divine Disclosure and Mortal Inquiry
insight into non-experienced states for which there is otherwise no
evidence.15 Hence, Sophocles’ chorus speaks of Teiresias as ‘the
mantis in whom alone of mortals truth is implanted’ (OT 298–9).
Side by side with the vocabulary of knowledge, however, the
role of mortal agency in the extrapolation of propositions from
omens precluded the construal of divination as an infallible trans-
mission of truths. Calchas infers from the portent of a serpent
devouring eight young sparrows and then their mother that
the Greeks shall war for as many years and in the tenth take
Troy (Il. 2.303–30). Odysseus stresses to the Greeks that they all
witnessed the omen in question (301–2) and urges them to stay in
Troy to discover whether or not Calchas divined (i.e. interpreted it)
correctly (ὄφρα δαῶμεν | ἢ ἐτεσν Κάλχας μαντεύεται ἦε καὶ οὐκί,
299–300). Here Calchas’ own consultants and supporters repre-
sent his divination as a conjecture, to be verified only if and when
the state of affairs in question comes within the scope of their
collective experience. Hence, the same chorus in Sophocles can
say of Teiresias, ‘the wise augur’, that, while Zeus and Apollo are
knowledgeable, no sure criterion (κρίσις . . . ἀληθής) can determine
whether a mantis carries more weight than any other mortal
(OT 484, 496–501).
We cannot determine to what extent divinations were deduced
from systematic exegetic rules. It appears that interpretations of
divine messages typically started from basic semiotic principles
(e.g. lightning on the right is a favourable sign (Il. 2.353; cf. Arat.
Phain. 1.5–6), a lobeless liver unfavourable) and were further
elaborated in the light of the nature of the inquiry or
circumstances.16 In this respect, divination was practicable by
any informed mortal who recognised an omen as such. In
Xenophon’s Cyropaedia, Cyrus’ father imparts to him a working
knowledge of omen-divination, so that he would not be entirely
dependent on potentially deceitful manteis, or at a complete loss if
he should ever find himself without one (1.6.2). Xenophon himself

15
X. Hipp. 9.9; Cyr. 1.6.46. Xenophon’s Socrates warns against putting to the oracle
questions that can be ascertained by mortal means (Mem. 1.1.9; cf. Mem. 1.4.18: περὶ
τῶν ἀδήλων ἀνθρώποις); cf. Oenomaus apud Eus. PE 5.28.2 (οὐ δίχα θεοῦ . . .
εὑρίσκεται).
16
Flower (2008) 117: ‘context informs interpretation’; cf. Trampedach (2008) 223.

110
 3.1 Divination
learned much from observing professionals at work, and speaks
once of a liver-omen so clearly unfavourable that any layman
would have recognised that the gods objected to the proposed
plan.17 As Flower writes: ‘Xenophon clearly believes that divina-
tion is a teachable craft, and that any intelligent person can learn
it.’18 We need not be surprised, then, that, when asked to define the
art of divination, the sophist, and probably diviner, Antiphon
replied: ‘the conjecture of an intelligent person’.19
Polydamas infers from an eagle flying on the left, and carrying
a snake which it fails to deliver to its nest, that the Trojans shall
likewise make some military advances but fail to finish the task.
Though not a diviner, he avers that he interpreted as a diviner
would have (ὧδέ χ’ ὑποκρίναιτο θεοπρόπος, Il. 12.211–29).20
In lieu of a mantis, Helen similarly volunteers her own augury
(μαντεύσομαι), premising that the right-hand position of the eagle
calls for a favourable interpretation (Od. 15.160–78). Perhaps less
surprisingly, we find lay practitioners also in the interpretation of
oracles. Themistocles argues correctly against the expert chres-
mologues’ exegesis of the Wooden Walls oracle that, had Apollo
indicated a coming defeat at Salamis, he would not have styled the
island ‘divine’ (Hdt. 7.142.3–143.1).21 Lysander similarly accuses
the chresmologue Diopeithes of misinterpreting an oracle (οὐκ
ὀρθῶς . . . ὑπολαμβάνειν, Plutarch, Lys. 22.5–6), and argues for
an alternative reading.22 Such passages presuppose the distinction
between experienced professionals, well-versed in oracle exegesis
or the semiotic principles underlying mantic interpretation, and
typically less able lay practitioners.

17
Anab. 5.6.29; 6.1.31; cf. 6.4.15. The lay Aegisthus recognises an unfavourable liver sign
at Eur. El. 826–32.
18
Flower (2008) 129.
19
ἀνθρώπου φρονίμου εἰκασμός, DK87 A9; cf. Pendrick (2002) 243. For Antiphon as
a diviner and dream-interpreter, see e.g. Cic. Div. 2.70.144; Diog. Oen. fr. 24 Smith, with
Pendrick (2002) 49–53; Flower (2008) 125–6; but see contra Gagarin (2002) 99–101; cf.
Woodruff (2004) 330–1. The seer Ophioneus practised a method of divination which
emphasised learning all the facts relevant to each case (Paus. 4.10.6).
20
Il. 13.730–733 notwithstanding, see Flower (2008) 120, n.30 against Dillery (2005)
172–3.
21
See Dillery (2005) 210–12; Johnston (2005a) 17; Harrison (2006) 140.
22
X. Hell. 3.3.3; Plut. Lys. 22.5–6; Ages. 3.3–5.

111
 Xenophanes on Divine Disclosure and Mortal Inquiry
In stressing mortal agency, I do not claim that divination always or
even usually involved the picturesque exegetical reasoning instan-
tiated by Calchas, Helen and Polydamas.23 Indeed, Homeric divina-
tion is not invariably inferential.24 Nor do I follow those scholars
who exclude entirely the elusive notion of inspiration from Greek
omen-divination or the extent to which the gods may not only
disclose omens but also somehow guide certain favoured individuals
in the act of interpreting them.25 We should neither ignore Plato’s
and Cicero’s influential dichotomy between uninspired technical
divination (e.g. augury, extispicy) and inspired natural divination
(dreams, ecstasy) nor mistake it for a straightforward exposition of
the general view.26 To contract work in omen-divination, Deiphonus
claimed to be the son of Euenius of Apollonia, who ‘acquired
instantaneously an innate capacity for divination’ (ἔμφυτον αὐτίκα
μαντικὴν εἶχε, Herodotus, 9.92–4). Quite generally, the continual
prominence of certain mantic families going back to a (mythical)
eponymous ancestor who acquired his mantic capacities as a divine
gift or by other supernatural means, and the widespread practice
among itinerant diviners of advertising themselves by claiming
descent from those families, suggest the notion that, through divine
favour, some mortals are more capable mantic interpreters.27
Compared with ecstatic possession, the decoding of divine commu-
nication primarily on the basis of observed semiotic principles is less
reliant on the notion of inspiration, but does not exclude it.28

23
Though note Plutarch, who, citing Theopompus (fourth century), records five such
examples preceding a single confrontation (Dion. 24; cf. Timol. 8).
24
At Il. 7.44–53, the seer Helenus (cf. Il. 6.76) somehow ‘overhears’ divine deliberations;
cf. the seer Theoclymenus’ vision at Od. 20.351–62. Both events are otherwise unpar-
alleled in Homer. Theoclymenus augurs standardly at Od. 15.525–34.
25
Bremmer (2008) 150 styles Greek seers ‘technicians’.
26
Pl. Phdr. 244b6-d5; Cic. Div. 1.6.11. Plato’s terminology, dictated by his particular
agenda, is certainly idiosyncratic: he restricts mantikê to ‘natural’ divination.
Olympiodorus expounds the Platonic classification, indicating that it was not readily
recognisable (διττή ἐστιν ἡ μαντικὴ παρὰ Πλάτωνι, in Alc. 69.21–70.4). Cicero too
presents the classification as a position undertaken by some theorists of divination (qui
duo genera diuinationum esse dixerunt, 1.18.34), perhaps thinking especially of the
Phaedrus. For Cicero’s interest in the Phaedrus, see Yunis (2011) 27.
27
On mantic families, see Flower (2008) 37–50, 130; Dillery (2005) 174, 192; Burkert
(1992) 43–6; Bremmer (2008) 134.
28
I broadly agree with the qualified reactions to the Platonic-Ciceronian classification in
Dillery (2005) 171–2 and Flower (2008) 26, 84–90. Manetti (1993) 22 adopts the
classification wholesale.

112
 3.1 Divination
Exegetical reasoning, however, reflects a broader point. Far
from an infallible transmission of truths to passively recipient
mortals, our sources, as early as Homer, often present divination
as involving or even requiring the application of human interpre-
tative judgement and reasoning to encoded evidence communi-
cated by the gods.29 The gods do not restrict their communications
to a few chosen individuals. Both mantic reasoning and auxiliary
inspiration, working in tandem, could extend to any lay mortal
sufficiently informed of the basic semiotic principles. Helen intro-
duces the argument for her augury as an inspired conjecture: ‘I will
divine, as the immortals cast it in my heart and as I think it will
come to pass’ (μαντεύσομαι, ὡς ἐνὶ θυμῷ | ἀθάνατοὶ βάλλουσι καὶ
ὡς τελέεσθαι ὀΐω, Od. 15.172–3). She reasons about the omen as
the gods inspire her to reason. Like any discipline such as poetry,
medicine or masonry (listed alongside divination as ‘public-crafts’
at Od. 17.383–4), divination could be performed badly or by
amateurs. The very fact that texts from Homer to Sophocles are
anxious to represent dismissals of particular instances of divina-
tions (rather than of divination as such) as impious and deleterious
demonstrates that such dismissals were hardly inconceivable
within the divinatory belief system.30
Cicero’s character Quintus rightly replies to the objection that
mantic predictions do not invariably come true that the same can
be said of any conjectural discipline (Div. 1.14.24), reflecting
Odysseus’ construal of Calchas’ divinations as conjectural.
Although he speaks of ‘technical’ omen-divination, Quintus’
insight can be extended also to the reception of oracles through
fallible, interpretative reasoning, a principle analysed by Plato
(Tim. 72a1ff) and dramatised by Euripides (Ion 532ff). Greek
tradition abounds with the didactic tragedies of consultants who
mistook oracular responses for self-evident, self-sufficient
answers, assuming naively an unreflective reading. Most
famously, Herodotus relates how the Lydian king, Croesus, was
told by Delphic Apollo that if he marched he would destroy a great
29
See along similar lines Vernant (1991b) 308; Harrison (2000) 245; Flower (2008)
13–14; Hollmann (2011) 111, n.145; Kindt (2015) 40.
30
E.g. Od. 2.178–80; Od. 20.351ff (ἐπ’ αὐτῷ ἡδὺ γέλασσαν, 358); see further Griffith
(1993) 101; Flower (2008) 17, 119–22.

113
 Xenophanes on Divine Disclosure and Mortal Inquiry
kingdom and mistakenly inferred that he would destroy the
Persians’ kingdom rather than his own (1.53). Tisamenus of Elis
began his illustrious mantic career by misinterpreting a Delphic
oracle that he would ‘win the five greatest contests’ as referring to
athletic competitions (ἁμαρτὼν τοῦ χρηστηρίου); the Spartans
correctly interpreted ‘contests of war’ and employed his services
(Λακεδαιμόνιοι δὲ μαθόντες, Hdt. 9.33).31 As Robert Parker
observes, the Wooden Walls oracle reshaped rather than resolved
public debate: ‘discussion resumed, though in appearance at
a different level: no longer a problem of tactics or politics, but of
philology.’32 The integral and fallible role of mortal interpretation
in the derivation of beliefs from oracles informs Heraclitus’ asser-
tion that ‘the lord whose oracle is the one in Delphi neither says
nor conceals but gives a sign’ (DK22 B93). The god does not
articulate the answers to our queries, but provides a starting point
from which we can work towards answers.33 Mutatis mutandis,
Heraclitus’ insight can be generalised to all forms of divination.34
Self-representations and representations of the diviner as
knowledgeable probably derive from the supposition that the
omen itself encodes a divinely held and thus a true proposition,
coupled with avowed professional confidence or confidence in the
profession. A diviner certain of the correctness of his analysis of
an omen or oracle, and also perhaps of divine favour or inspiration,
will be certain of the truth of the derived proposition. The phe-
nomenon of favoured mantic individuals and families illustrates
the association between epistemic and a more general approxima-
tion to the divine. The epistemic superiority of Teiresias (tradi-
tionally the father of Manto, mother of Mopsus) extended to
a uniquely substantial survival in the afterlife. Sophocles’ chorus,

31
For other such Delphic traditions in Herodotus, see 1.55 (with 1.91); 1.165–7; 4.163–4;
5.43–5; 6.76–80; see further Kindt (2006); Barker (2006); Maurizio (2013) 111–13; Tor
(2016).
32
Parker (2000) 80. Fontenrose (1978) portrays the historical Delphic oracle as essentially
a Yes/No answering service (see 233–5). Parker (2000) 80 n.14–15 counters effectively.
Cultural representations of divination are anyway at least as (if not more) pertinent for
our investigation.
33
See further Tor (2016).
34
Even lot and dice divination, see Div. 1.18.34; Graf (2005) 62. Cf. dream-interpretation
(ὀνειροπόλον, Il. 1.63), practised amateurly (Od. 19.535ff), and fallibly (Hdt. 1.209–10);
see also Od. 19.559ff; Aesch. Pr. 485–6.

114
 3.1 Divination
when styling Teiresias the only mortal in possession of truth,
describes the diviner as himself divine (τὸν θεῖον . . . μάντιν, OT
298). The mythical diviner Amphiaraus of the Melampodidae
later became a god and was worshipped as such throughout
Greece (Paus. 1.34.4). Eperastus, a historical mantis of the
Melampodidae, described his clan in an inscription as ‘equal to
the divine’ (ἰσοθέων, Paus. 6.17.6). The Homeric lineage of
Theoclymenus illustrates the point. Mantios sired Cleitus and
Polypheides. Dawn snatched away the former to live among the
immortals (ἵν’ ἀθανάτοισι μετείη, Od. 15.251), while Apollo made
the latter, Theoclymenus’ father, far the best mantis (Od.
15.252–3).35 Like Teiresias, Theoclymenus’ father and paternal
uncle combine to express the affinity between the seer’s transcen-
dence of mortal epistemic limitations and a more general
transcendence.
Cicero’s formal definition of diuinatio as foreknowledge of
future events (Div. 1.1.1; cf. 1.5.9) betrays the narrow focus of
certain Hellenistic debates. But Cicero at once reveals the broader
significance of the term when he describes divination as ‘a
thing . . . by which human nature can most nearly approach the
power of the gods’ (res . . . quaque proxime ad deorum uim natura
mortalis possit accedere, 1.1.1).36
There is no question that Xenophanes expresses a contrast
between divine epistemic superiority and mortal epistemic limita-
tions, and he clearly conceives of the latter by reference to the
former. In B34 (quoted and translated in Chapter 3.3), he almost
certainly identifies mortal epistemic limitations by an implicit con-
trast with the divine, as suggested by the specification that we
mortals cannot have knowledge ‘about gods’, and as the ancient
and modern readers of the fragment agree.37 Xenophanes’ views on
the cognitive dissimilarity between mortal and divine (B23) coexist

35
On Theoclymenus’ lineage, see Dillery (2005) 173–4; Flower (2008) 42.
36
Cf. Quintus: interpretes . . . proxime ad eorum, quos interpretantur, diuinationem
uidentur accedere (1.18.34).
37
So Varro and Arius Didymus: hominis est enim haec opinari, dei scire (apud Aug. Civ.
Dei 7.17); ὡς ἄρα θεὸς μὲν ο῏ιδε τὴν ἀλήθειαν (apud Stob. 2.1.17 = A24); cf. Fränkel
(1974) 128; Lesher (1992) 164; Barnes (1982) 139. Alcmaeon opened his treatise with
a contrast between divine knowledge and human conjecture concerning (at least) non-
evident matters (σαφήνειαν μὲν θεοὶ ἔχοντι, ὡς δὲ ἀνθρώποις τεκμαίρεσθαι, DK24 B1).

115
 Xenophanes on Divine Disclosure and Mortal Inquiry
with the notion that the latter is cognitively superior to the former.38
In the remainder of this chapter, I will evaluate Xenophanes’ reac-
tions to the other facets of divination – (ii)a, (ii)b and (iii)39 – which
we analysed here.

3.2 Against a Notion of Disclosure

Xenophanes writes (B18):
οὔτοι ἀπ’ ἀρχῆς πάντα θεοὶ θνητοῖσ’ ὑπέδειξαν,
ἀλλὰ χρόνῳ ζητοῦντες ἐφευρίσκουσιν ἄμεινον.
Indeed gods did not from the beginning intimate40 all things to mortals,
But as they search in time they discover better.
The first line of this fragment negates a claim about divine dis-
closure. Most scholars take it that Xenophanes categorically
rejects disclosure: at no point have the gods revealed anything to
mortals. Call this the ‘majority’ interpretation. Others think that,
through the expressions ‘from the beginning’ and ‘all things’,
Xenophanes allows that at some times the gods did reveal some
things (at least once something). This is the ‘minority’ interpreta-
tion. Scholars agree, then, that the negation in B18.1 displays two
markers which could be seen to restrict an otherwise unrestricted
remark about divine disclosure, one concerning the temporal
dynamic of disclosure (‘from the beginning’) and another con-
cerning the quantity of disclosure (‘all things’). The majority
interpreters insist that these qualifiers do not, in fact, restrict the
remark about divine disclosure.41
38
Cognitive superiority seems to be the focus of B24, see Drozdek (2004) 146; Granger
(2013a) 248. B25 perhaps implies omniscience (god’s mind engages with ‘all things’);
Simplicius cites the fragment to this effect (A31.9); cf. McCoy (1989) 237–8; Curd
(2013a) 230. See also McCoy (1989) 236 and Barnes (1982) 139 on B18 and KRS
(1983) 180 on B38.
39
For the statements which correspond to these numerals, see the opening paragraph of
this section.
40
Adopting the rendering in Lesher (1992) ad loc. We will presently discuss ὑπέδειξαν in
detail.
41
See the exposition in Lesher (1991) 230 of the positions that have been and can be taken.
Majority interpreters: Gomperz (1906) 162; Kleingünther (1933) 41; Loenen (1956);
Dodds (1973) 4; Lesher (1983; 1991; 1992; for further details, see Tor (2013a) 259 with
n.41); McKirahan (1994) 68–9; Curd (2002) 129; Mogyoródi (2006) 126–7; Bryan
(2012) 52–5; Granger (2013a) 262. Minority interpreters: Shorey (1911); Verdenius
(1955); Barnes (1982) 140; Tulin (1993) 133–5; Robinson (2008) 489.

116
 3.2 Against a Notion of Disclosure
This set-up is misleading in failing to recognise that, in addition
to those temporal and extensional qualifiers, Xenophanes also
qualifies the manner in which, according to the view he rejects,
the gods disclose things. The verb which Xenophanes uses –
‘intimate’ (hypodeiknymi) – is not a simple and bland term for
disclosure. The three qualifications in B18.1, I will argue, do not
restrict a rejection of divine disclosure as such. Rather, they
combine to express a particular, traditional and theologically
faulty notion of the nature of divine disclosure, which
Xenophanes rejects. That is, I argue, the gods never disclosed
anything in the manner in which the view negated in B18.1 con-
ceives of disclosure. Contrary to the minority interpretation, then,
Xenophanes does not allow that the gods did ‘intimate’ (hype-
deixan) some things at some time. Contrary to the majority inter-
pretation, he does not deny divine disclosure categorically. Rather,
Xenophanes supplants the traditional understanding of disclosure
with his own, alternative notion of it.
In epic poetry, the verb deiknymi (‘show’) can signify divine
disclosure, but never with the prefix hypo.42 In archaic and classi-
cal Greek, hypodeiknymi only very rarely describes divine actions
towards mortals. Xenophanes’ use of the verb to signify divine
disclosure is, to my knowledge, unique in these periods.
We should, therefore, heed the prefix in hypedeixan, which indi-
cates disclosure in an indirect, cryptic and even underhand
manner.43 Xenophon (Mem. 4.3.13) says that the gods indirectly
imply (hypodeiknyousin) a certain precept about proper worship
(that mortals should honour them even though they cannot see
them) through the discreet ways in which they quite generally
benefit mortals without revealing themselves openly. Herodotus
writes that god gives many mortals a glimpse (hypodexas) of
blessedness, before ruining them utterly (1.32). The ‘showing’ in
question here consists simply in the mortal’s happiness prior to
42
Il. 13.243–4; Od. 3.173–4; cf. Op. 448–51; Pind. fr. 131b M; also, Hdt. 1.209; 7.37
(προ-).
43
On these connotations of the prefix, cf. Lesher (1991) 237, n.19; (1992) 153. Lesher
himself, however, downplays the pointedness of Xenophanes’ terminology, listing
purported parallels. But, other than Mem. 4.3.13 (addressed presently), none of the
other texts listed contain the verb hypodeiknymi; see further Tor (2013a) 251–2
with n.12.

117
 Xenophanes on Divine Disclosure and Mortal Inquiry
calamity. Neither passage uses the verb in reference to divine
communication. Xenophon, however, supports the identification
of a notion of indirectness in hypedeixan, while Herodotus demon-
strates the sinister undertones of the verb and shows that it can be
naturally used to convey suspicion or even criticism in descrip-
tions of divine behaviour.44 Though not invariably, such connota-
tions, as well as those of cryptic secretiveness, often characterise
the prefix and other occurrences of the compound.45 More gen-
erally, both Herodotus and Xenophon use hypodeiknymi in
a pointed way in their descriptions of divine behaviour. They do
not use the verb as a mere synonym for deiknymi.
Crucially, his terminological novelty notwithstanding,
Xenophanes does employ this exceptional vocabulary to signify
a notion of divine disclosure. The issue at stake is how mortals
come to discover what they discover. Xenophanes regards the kind
of divine intimation that he rejects as mutually exclusive with his
prescription of temporally protracted inquiries (B18.2). Whatever
else, such inquiries clearly extend for Xenophanes to the business
of forming and assessing beliefs. Thus, for example, Xenophanes’
observations (whether first-hand or not) about marine fossils
found inland support his novel theory of the earth’s past submer-
sion (A32–3). Correspondingly, B18.1 must articulate a notion of
disclosure which, whatever else, purports to offer an instant access
to truths, which conflicts with the call to search for them through
such temporally protracted inquiries.
The negated view conceives of disclosure as an indirect, secre-
tive and cryptic affair. Furthermore, in this manner the gods dis-
closed everything (panta) from the beginning (ap’ archês). Given
Xenophanes’ qualification of the manner of disclosure (hype-
deixan), the markers ‘from the beginning’ and ‘all things’ do not
restrict a rejection of disclosure simpliciter, but further qualify the
particular notion of indirect, cryptic disclosure, which
44
Cf. ‘feigning virtue’, Thuc. 4.86.5.
45
LSJ, sv. ὑ πό, F, III; Smyth (1972), no.1698.4; cf. e.g. Il. 21.44. In a scholion on
Lycophron (Σ 344 Scheer), φρυκτὸν ὑποδείξας signifies Sinon’s secret disclosure of
a signal to the Greeks (cf. Lesher (1992) 153). In literary criticism, hypodeiknymi came
to signify that an inexplicit poetic text indirectly implies distinctions or insights (Plut.
Mor. 23e1; 645a5) or even elaborate philosophical doctrines (e.g. ps.-Plut. de Hom. 2.
1063–74, 1298–308 Kindstrand).

118
 3.2 Against a Notion of Disclosure
Xenophanes rejects. That is, Xenophanes does not severally reject
the isolated claims that the gods disclosed (i) from the beginning,
(ii) everything and (iii) cryptically. Rather, the qualifications
(i)–(iii) combine to express a unified view of how disclosure
works, i.e. the view that from the beginning gods cryptically
disclosed everything, which Xenophanes rejects.
What is the target of Xenophanes’ attack? Lesher suggests that
Xenophanes particularly targets omen-divination.46 We need not
restrict the notion of divine disclosure negated in B18.1 to omen-
divination or even to divination generally. Yet, elsewhere too
Xenophanes’ theological criticisms confront pervasive and
authoritative traditional suppositions (B10–14) and it is very likely
that B18.1 conveys – among other things – a polemical attitude to
divination, the most dominant form of divine disclosure.47 Indeed,
we saw that divination involves the signification of propositions
indirectly through their encryption in diverse objects or events (or,
for oracles, in enigmatic pronouncements). The verb ‘intimate’
(hypedeixan), signalling a cryptic, enigmatic signification, is
therefore highly apposite as a critical representation of mantic
communication.
Xenophanes’ cosmology coheres with a polemic against divina-
tion in his remarks about divine disclosure in B18.1. In the epic
tradition, Iris the messenger (Il. 15.158–9; Th. 780–1) is standardly
represented as an omen sent by Zeus (Il. 11.27–8; Il. 17.547–8).
Xenophanes’ assertion, ‘she whom they call Iris, this thing too is by
nature a cloud’ (B32), is thus plausibly read as a deflationary
reduction of what is commonly regarded as a portentous deity to
the fleeting phenomenon of exhalation.48 In the light of this, similar
reductions of a variety of atmospheric phenomena, several related to
omen-divination, could also relate to a denial that such phenomena
constitute encoded divine communications.49 There are also our

46
Lesher (1991) 237–40.
47
Xenophanes’ use of mantic terminology in his denial of knowledge-claims in B34 (see
Ch. 3.3) fits with a preoccupation with divination in B18.1.
48
Dodds (1951) 196, n.7; Lesher (1991) 241; (1992) 139–4.
49
Clouds in the ordinary sense (A46, B30), lightning (A45), comets, shooting stars (A44),
the moon (A43), the stars (A38), St. Elmo’s fire (‘also those whom some call the
Dioskouroi are clouds’, A39); See Lesher, loc. cit.; KRS (1983) 174; Hussey (1990)
24; Mourelatos (1989) 282–5; (2008b) 135–7, 149–50; Gregory (2013) 103–4.

119
 Xenophanes on Divine Disclosure and Mortal Inquiry
express reports that Xenophanes rejected divination (A52), unless
they themselves derive from B18.1.
The open-ended ‘all things’ (panta, B18.1) is also instructive.
The force of ‘all things’ or ‘everything’ in Xenophanes is else-
where context-sensitive. At different junctures, ‘panta’ may refer
to all things (B27, if authentic) or all members of a subset of things
(B29) which undergo natural processes, to everything (or perhaps
every sort of thing) censured among mortals (B11.150) and to
every item in a foregoing catalogue of public honours (B2.10).51
The gods’ disclosure of ‘everything’ to mortals could certainly
encompass divinely inspired accounts of the world which are
universal in scope, such as Hesiod’s Theogony.52 But the indirect,
cryptic disclosure of ‘everything’, and especially the emphatic
contrast with inquiries over time, again point also to
a preoccupation with divination. For Xenophanes, mortals must
conduct protracted inquiries in the hope only for gradual, hard-
won advances. The notion of disclosure negated in B18.1 com-
petes with Xenophanes’ call for such inquiries in B18.2 because, on
this notion, gods made everything instantly available to humans,
and so every object of human inquiry which was not already
independently available to them.53 ‘All things’ aptly conveys the
essentially open-ended range of questions – encompassing any
truth – concerning which, traditionally, the gods always commu-
nicated to mortals.
It is an important point here that the Greek phrase ‘ap’ archês’
signifies, not ‘(once) at the outset’, but ‘from the outset
onwards’.54 The symmetry of the contrast between this phrase

50
Xenophanes’ catalogue of blameworthy actions is certainly not exhaustive; Lesher
(1992) ad loc renders ‘all sorts of things’.
51
On περὶ πάντων in B34.2, see Ch. 3.3. 52 Koning (2010) 204.
53
After all, even if the Greeks could often emphasise the effort and fallibility of inter-
pretation involved in divination, it remains the case (i) that the response to one’s
inquiries is attained promptly (certainly by contrast with the types of prolonged, even
decades-long (cf. B8) inquiries which Xenophanes prescribes in B18.2) and (ii) that
individual diviners readily advanced or were ascribed confident knowledge-claims on
the basis of such swift disclosures (as we saw in the previous section and will see again
in the next one). In B18, Xenophanes legitimately picks up on these aspects of received
notions of disclosure.
54
As Lesher (1991) 232–3 shows; see LSJ. sv. ἀπό II; cf. Hes. Th. 425; Hdt. 2.50, 104,
113; Gorg. B11a.29. Classen (1996) demonstrates that Homer and Hesiod consistently
use ἀρχή, especially with the prepositions ἐξ or ἀπό, in this same sense.

120
 3.2 Against a Notion of Disclosure
and Xenophanes’ assertion that mortals discover better by inquir-
ing ‘over time’ (χρόνῳ, B18.2) suggests that (as widely assumed)
the phrase ‘from the beginning onwards’ qualifies the verb, ‘inti-
mate’: the gods did not always (‘from the beginning’) intimate
everything to mortals (ap’ archês . . . hypedeixan). Conceivably,
though, we could instead – or also – take ‘from the beginning’ to
qualify ‘all things’. On this reading, B18.1 says that the gods did
not intimate to mortals every truth about the world from its very
beginnings onwards (ap’ archês panta). Nonetheless, the entirely
open-ended and universalising phrase ‘everything from the begin-
ning onwards’ would still suggest that gods still now intimate to
mortals everything, up to and including present-day and, indeed,
future truths.
On either reading, the aorist tense of ‘intimate’ (hypedeixan)
does not necessitate the strange view that, on some single, unique
occasion, gods just once made a one-off revelation of everything to
mortals, but then ceased making such revelations to mortals.55
Rather, gods rendered and still render instantly available to mor-
tals ‘everything’, including any truth that they wish to attain. The
specification ‘from the beginning onwards’ most likely stresses
that the gods always did so. Alternatively, it stresses that the scope
of divine disclosure encompasses comprehensively any matter
from the beginnings of cosmic history and onwards continually.
Either way, the negated view, that gods intimated everything to
mortals (B18.1), conflicts with Xenophanes’ call for temporally
protracted inquiries (B18.2) because, on this view, gods still now
render instantly available to mortals any object of inquiry which is
not independently instantly available to them.56
55
Even if one did (implausibly, to my mind, and in a manner that would be difficult to
square with the attested meaning of the phrase ἀπ’ ἀρχῆς) maintain such a view of
B18.1, this would require us only to modulate some details in my argument that, in this
fragment, Xenophanes articulates and rejects a highly qualified and very particular
conception of divine disclosure, which he supplants with his own alternative concep-
tion. The further point, that Xenophanes bore a critical attitude towards traditional,
divinatory notions of divine disclosure, is, moreover, reflected independently also else-
where in our evidence, as discussed in this section and the next one.
56
We may usefully compare Xenophanes’ use of the aorist in B38 (for the text, see
Ch. 3.4). It would be implausible to read into this statement an elaborate and confusingly
compressed deist thesis that, on some single, primordial occasion, god established
mechanisms which now cause the generation of yellow honey without any further
divine ministration. The fragment much more naturally conveys the view that, for as

121
 Xenophanes on Divine Disclosure and Mortal Inquiry
In the light of these considerations, we may perhaps best think
about Xenophanes’ use of the aorist in B18.1 (and B38) as having
something of a gnomic flavour. The gnomic aorist (usually trans-
lated with a present tense) expresses a general truth, on the under-
standing that a past occurrence can stand in for a universal and
recurrent pattern, as in the following examples: ‘He who reveres
the daughters of Zeus who issue their voice, him they benefit
(ὤνησαν) greatly and hear (ἔκλυον) as he prays’ (Il. 9.508–9);
‘but the fool realises (ἔγνω) once he has suffered’ (Hesiod,
Op. 218), ‘just as a noble horse, even if he is old, does not lose
heart (θυμὸν οὐκ ἀπώλεσεν) amidst dangers . . . ’ (Sophocles,
El. 25–6).57 We may similarly find a gnomic nuance in
Xenophanes B38: ‘Had it not been the case that god makes
(ἔφυσε) yellow honey, they would say that figs are much
sweeter.’58 In B18.1, Xenophanes also puts in his sights
a general claim (which is most likely emphasised to have obtained
always: ap’ archês): ‘gods intimate everything to mortals’. He
subjects this claim to a denial: ‘it is not the case that . . . ’ (οὔτοι).
A close parallel to Xenophanes’ language, which also suggests the
aorist’s gnomic potential, demonstrates that the claim which
Xenophanes negates can naturally represent a general purported
truth. A poet of the Hymns speaks of ‘the race of semi-divine
heroes whose deeds gods showed [or: show] to mortals’ (θεοὶ
θνητοῖσιν ἔδειξαν, h.Hom. 31.19). The poet puts forward the
general truth, that gods disclose the subject matter of heroic poetry
to mortals, with an almost verbatim parallel to Xenophanes’
phrase ‘gods intimated [or: intimate] to mortals’ (θεοὶ θνητοῖσ’
ὑπέδειξαν, B18.1).
This close echo also reminds us, however, that we need not
delimit the target of Xenophanes’ criticism in B18.1 to mantic

long as there had been yellow honey, god caused the generation of yellow honey, and
still now causes it. Confirmation of this (independently more plausible) interpretation is
found in B25, which shows that god’s intelligent and purposeful ministration of all
things is an on-going affair (Ch. 3.4). In Hdt. 1.32, the aorist participle ὑποδέξας conveys
that god gave a show of blessedness to doomed mortals, not simply once, but right up to
destroying them.
57
For the gnomic aorist, see further Smyth (1972) noo.1931–2; Goodwin (1889) noo.
154–9.
58
Cf. n.56 above.

122
 3.2 Against a Notion of Disclosure
communication. The Homeric poet uses the usually mantic dei-
knymi to express a notion of poetic inspiration. Xenophanes, in
employing the very same phrase, with the conspicuous addition of
the prefix hypo, may thus be reacting critically also to the elusive
notion of poetic inspiration, itself, as here and as we saw in
Chapter 2.2, often closely associated with divination. We may
conclude that, however wide a net Xenophanes casts in B18.1,
the line’s language, its opposition to the alternative of temporally
protracted inquiries and Xenophanes’ polemical cosmology and
reported rejection of divination all indicate that his invective
against a notion of divine disclosure here conveys a polemic
against traditional, authoritative paradigms of disclosure, which
prominently included mantic communication and poetic
inspiration.59
Let us take stock. Xenophanes is thinking in B18.1 of
a particular manner of divine disclosure. The line evokes tradi-
tional paradigms of disclosure to which (as we shall see further in
the next section) Xenophanes bore a polemical attitude. We may
thus follow the majority interpretation to an extent. Xenophanes is
unlikely to have allowed that the gods do disclose cryptically and
indirectly (hypedeixan) some things at some points in their inter-
action with mortals, agreeing with the negated view about the
manner of disclosure, while disagreeing only about its temporal
and quantitative scope. Equally, however, the central insight of the
minority interpretation has been vindicated. Xenophanes carefully
formulates a highly qualified view and it is implausible to return to
him the same view divested of those qualifications. Rather than
following standard terminology (say, θεοὶ θνητοῖσιν ἔδειξαν),
Xenophanes instead chose a pointedly exceptional term for dis-
closure, which highlights the notions of indirect, cryptic and
secretive communication. Furthermore, ‘from the beginning’ and
‘all things’ are indeed emphatic qualifications. Rejecting the view
that gods disclosed everything from the beginning onwards in an
indirect and cryptic manner, while a pointed critical representation
59
Indeed, Xenophanes could conceivably also be implying that the gods did not instantly
bestow whatever cultural and material prosperity we have (B18.1) and that we progress
culturally and materially over time (B18.2). Lesher rejects these connotations plausibly
but not conclusively, see Ch. 3.5.

123
 Xenophanes on Divine Disclosure and Mortal Inquiry
of traditional conceptions of disclosure, would be a remarkably
peculiar way of saying that the gods never and in no way disclosed
anything to mortals. Fortunately, we have evidence, I believe, that
Xenophanes did in fact formulate an alternative (B36):
ὁππόσα δὴ θνητοῖσι πεφήνασιν εἰσοράασθαι . . .60
It is in fact difficult to determine whether we should translate the
verb here transitively – ‘however many things they have disclosed
for mortals to look upon’61 – or intransitively: ‘however many
things have appeared for mortals to look upon’.62 I submit that, on
either grammatical construal, these words convey the alternative
notion of divine disclosure which B18.1 independently leads us to
expect. Xenophanes’ critique there calls for an alternative, which
is not tainted by the characteristics of the highly qualified notion of
disclosure that he rejects. The account of mortal experiences in
B36 fits the bill especially because (as we shall see momentarily) it
precisely inverts the account rejected in B18.1. With this last
consideration in mind, we would still plausibly maintain that

60
Herodian cites B36 for its instantiation of a linguistic peculiarity: shortened penultimate
syllables in -σι verbs (περὶ Διχρόνων 3.2 16.17–29; de Pros. Cathol. 3.1 535.28–35
Lentz). Although editions of Xenophanes do not reflect this fact, B36 is also cited by the
much later Georgius Choeroboscus (in Theod. 4.2 88.27–33 Hilgard). Georgius’ exam-
ples and discussion of the same phenomenon are, however, clearly reliant on and closely
echo Herodian’s (or an intermediary or common source). The slightly corrupted reading
πεφήνασθ’ in Ms. O of Georgius leaves no question whatever that he too wrote
πεφήνασιν. Mss. CV carry πεφύκασιν. The banalisation of πέφηνεν to πέφυκεν can be
paralleled, see Soph. El. 646 and, possibly, Eur. Phoen. 916 (with Mastronarde (1994)
411–12). In this case, the banalisation was perhaps specifically occasioned. Georgius
retains (but reorders chronologically) all of Herodian’s examples, except Od. 7.114: ἀμφὶ
δὲ δένδρεα μακρὰ πεφύκασι. A scribe ultimately responsible for the CV reading and
familiar with the original discussion in Herodian – or simply highly habituated to Homer
(in whom πεφήνασιν never occurs) – may, then, have reintroduced πεφύκασι under the
influence of the omitted Homeric line.
61
Similarly, Lesher (1992) 39; Barnes (1982) 140; Benitez and Tarrant (2015) 215.
62
Similarly, DK ad loc.; Burnet (1930) 121; Edmonds (1931) ad loc.; Guthrie (1962) 397.
Although we would otherwise expect the second perfect of φαίνω to carry an intransitive
sense, this is at least balanced by the consideration that such a reading of B36 would
leave us with ὁππόσα as a neuter nominative taking a plural verb, a very peculiar usage
of ὁπ(π)όσα for which there is, to my knowledge, no parallel in extant archaic and
classical Greek. A TLG search of all occurrences of ὁπ(π)όσα in texts up to and
including Aristotle indicates that the word is far more often used in the accusative (as
in Il. 24.7; Od. 14.47) and, when nominative, invariably takes, as one would expect,
a singular verb (e.g. h.Hom. 2.365; Hes. fr. 204.113 M-W). In B18.1, moreover, θνητοῖσ’
ὑπέδειξαν is transitive, which perhaps suggests a correspondingly transitive sense for the
structurally parallel θνητοῖσι πεφήνασιν.

124
 3.2 Against a Notion of Disclosure
Xenophanes is referring implicitly to the agency which underlies
mortal experiences even if we followed the intransitive translation.
Herodotus (9.120) offers an instructive parallel.63 We could read
the statement about Protesilaus here intransitively – ‘Athenian
stranger, do not fear this portent, for [sc. it] has not appeared to
you (πέφηνε), but it is to me that Protesilaus of Elaeus signifies
that . . . ’ – or transitively: ‘ . . . for Protesilaus of Elaeus has not
shown [sc. it] to you (πέφηνε), but it is to me that he signifies
that . . . ’. On either construal, the passage would still demonstrate
how naturally this terminology can signify that certain appear-
ances (ὁππόσα, B36; τὸ τέρας τοῦτο, Hdt. 9.120) have appeared to
their recipients (θνητοῖσι, B36; ἐμοί, Hdt. 9.120) at the instigation
of a divine power.
If we follow the transitive translation, then ‘gods’ is the only
candidate for the missing subject. Who else could make things
evident to ‘mortals’ (θεοὶ θνητοῖσ’, B18.1)?64 We may further
support the impression that this text expresses a notion of disclo-
sure even on the intransitive translation by raising and addressing
the following question: should we read B36 as a whole as the
fragment of a statement made in propria persona? First, we have
no reason not to take B36 in this way. The polytheistic language
(on the transitive translation) is certainly no obstacle, since
Xenophanes undeniably uses such language positively and, any-
way, our evidence does not support ascriptions of a strict mono-
theism to him.65 Second, and more importantly, we do have reason
to read B36 positively, for, I argue, the positive account of mortal
63
ξεῖνε Ἀθηναῖε, μηδὲν φοβέο τὸ τέρας τοῦτο· οὐ γὰρ σοὶ πέφηνε, ἀλλ’ ἐμοὶ σημαίνει ὁ ἐν
Ἐλαιοῦντι Πρωτεσίλεως ὅτι κτλ.
64
Cf. Lesher (1992) 176–7.
65
Monotheistic construals of B23 are highly improbable syntactically, see Stokes (1971)
76–8. I take no issue with the suggestion that Xenophanes’ theology is incipiently
monotheistic insofar as it may imply a tendency towards reconceptualising the divine
in the image of the ‘greatest god’ (B23–6, see Schofield (1997) 72–3). But – and this is
the important point for us – Xenophanes uses both polytheistic and monotheistic
language when speaking positively about the divine (esp. B1.24; cf. B34.2 and, more
contentiously, B18, B11–12, B14–16, B23). This fact alone problematises ascriptions of
full-fledged monotheism, and this is especially true of B23 itself: if Xenophanes wanted
to convey strict monotheism there, then his simultaneous talk of ‘gods’ in the same line
will have all but guaranteed that his point would be lost on an audience for whom strict
monotheism was hardly already a familiar and easily identifiable concept. Xenophanes
himself most probably remained vague on the numerical question. See further Tor
(2013a) 258 n.36; Granger (2013a) 237–8.

125
 Xenophanes on Divine Disclosure and Mortal Inquiry
experiences in B36 is a reversal of the view negated in the polem-
ical B18.1. The verb which Xenophanes uses here (phainô) is
a standard, bland term for divine disclosure and conveys none of
the pointed undertones of the language of B18.1 (hypedeixan).66
The denial that the gods intimated everything (panta . . . thnêtois’)
is fittingly balanced by a statement concerning the limited scope of
what the gods have disclosed (hopposa dê thnêtoisi). Finally, if in
B18.1 Xenophanes rejects the view that the gods have always
already made instantly available any and all truths about the
world (ap’ archês), his talk in B36 of however many things the
gods have disclosed allows that the current set of disclosed things
may vary. We shall see the significance of this last point when we
consider Xenophanes’ engagement with the possibility that our
available body of evidence may change and that counter-evidence
to our beliefs may be discovered (Chapter 3.4).
In B18.1, Xenophanes rejects only a very particular notion of
divine disclosure with carefully qualified characteristics. In B36,
he articulates a reversal of this notion of disclosure. This latter
fragment, then, most likely either implies (on the intransitive
translation) or refers to (on the transitive translation) an alter-
native notion of disclosure, which is precisely innocent of the
characteristics of the rejected one. The notion of disclosure
advanced in B36 (thnêtoisi pephênasin) supplants the one
rejected in B18.1 (thnêtois’ hypedeixan). These fragments illu-
minate one another. Xenophanes not only admitted divine influ-
ence over mortal inquiry but also explicitly represented that
influence as a form of divine disclosure.67 We must, then, employ
more nuanced vocabulary. We cannot speak of Xenophanes’
position on disclosure simpliciter. Rather, Xenophanes replaces
what he takes to be the traditional view with his own, alternative
conception.

66
E.g. Il. 2.308, 318, 324, 353; Il. 4.381; Od. 3.173–4; Od. 21.413; cf. Hdt. 9.120 (πέφηνε).
67
Scholars have disregarded B36 when discussing B18 and Xenophanes’ views on divine
disclosure. A stimulating exception: Barnes (1982) 140 observes, without further com-
ment, that B18 is complemented by B36. The presumption that disclosure ‘is not the sort
of thing Xenophanes’ god . . . does’ (McKirahan (1994) 68–9) perhaps discouraged
recognition that B36 may express an alternative, novel notion of disclosure. See further
Tor (2013a) 259 with n.41 on Lesher (1983; 1991; 1992).

126
 3.2 Against a Notion of Disclosure
B18.1 is not ‘a firmly negative comment . . . on the question of
divine agency’.68 It is a firm rejection of one particular conception
of divine agency. It has been claimed that Xenophanes’ talk of
inquiries (zêtountes, B18.2) ‘explicitly contrasts . . . with divine
revelation’.69 The false assumption that divine disclosure must
conflict with mortal agency is instructively belied by the subse-
quent history of this very terminology of inquiry. The term zêtêsis
and its cognates can express the instigation and facilitation of
mortal inquiry through divination, sometimes signifying the
consultation itself.70 Indeed, the cognate reduplicated form of
the verb (dizêmai) – the verb which Parmenides employs to
describe his own philosophical inquiry (which involves divine
disclosure prominently) – is a standard technical term in Delphic
oracular responses, expressing the act of consulting Apollo.71
Since zêtêsis can signify both specifically philosophical inquiry
and divination,72 it is not surprising that divine disclosure was
sometimes thought to instigate and guide philosophical zêtêsis, as
with Socrates’ inquiries following Apollo’s response to
Chaerephon.73
Finally, the later grammarians offer an insight which harmo-
nises profoundly with our earliest testimonies to Greek concep-
tions of divine disclosure. An etymology of mantis popular among
them traces it to matô or mô, a verb which, they explain, is
synonymous with ‘to inquire’ (zêtein). For the diviner is inquisi-
tive (zêtêtikos): he is ‘one who inquires into unseen and

68
Lesher (1991) 233.
69
Mogyoródi (2006) 126–7, 151; cf. Gomperz (1906) 162; Kleingünther (1933) 41; Dodds
(1973) 4; McKirahan (1994) 68; Wardle (2006) 107; Bryan (2012) 53–4; Lesher (2013)
86–7. Shorey (1911) 89 observes that glosses of ζητοῦντες as ‘searching by themselves’
supply an αὐτοί Xenophanes never used.
70
Note the chorus’ expectation at Soph. OT 277–9; cf. Aesch. Pr. 775–6; for omen-
divination: Di. Hal. Ant. Rom. 3.70.3; Plut. Mor. 340c7; Paus. 3.22.12; signifying
consultations: Di. Sic. Lib. 17.51; Di. Hal. Ars Rhet. 9.7.49–50; Ant. Rom. 12.9.1; cf.
the late Oracula Tiburtina 58 Alexander.
71
See Ch. 5.3, n.113.
72
E.g. Pl. Phd. 66d (philosophical inquiry, cf. Lg. 1.631a); Phdr. 244c (divination).
73
Pl. Apol. 22a (ζητοῦντι κατὰ τὸν θεόν), 23b, 29c9-d4; cf. Crat. 406a; Ti. 47a. Socrates’
sign (μοι μαντική, Apol. 40a4–7; cf. X. Mem. 1.1.2–4; Pl. Phdr. 242c3–4) informs his
argument at Apol. 40b3-c3; cf. X. Apol. 8. Plutarch (channelling Ammonius) traces
philosophy to inquiry (τὸ ζητεῖν), inquiry to puzzlement, and puzzlement to Delphic
Apollo (Mor. 385c); cf. Iambl. apud Stob. Anth. 2.2.5.11–16.

127
 Xenophanes on Divine Disclosure and Mortal Inquiry
non-evident things’ (ὁ τὰ ἀφανῆ καὶ ἄδηλα ζητῶν).74 Not for noth-
ing, ancient doxographers employ identical vocabulary to describe
the domains concerning which, according to Xenophanes, we cannot
have knowledge: ‘of unseen things’ (τῶν ἀφανέων), ‘in non-evident
matters’ (ἐν τοῖς ἀδήλοις).75 Before addressing the role of divine
disclosure in Xenophanes’ own inquiries into unseen and non-
evident matters, we must first see how he arrives at the view, that
we cannot have knowledge about such things, by criticising tradi-
tional notions of disclosure.

3.3 Setting the Limits

What will Xenophanean disclosure not be? Divine disclosure will
not render ‘everything’ instantly available to mortals. It will not
amount to mantic communication or poetic inspiration. Since,
furthermore, Xenophanes maintains that, concerning certain
matters, mortals cannot attain clear and certain knowledge,
Xenophanean disclosure will not guarantee such knowl-
edge (B34):
καὶ τὸ μὲν οὖν σαφὲς οὔτις ἀνὴρ ἴδεν οὐδέ τις ἔσται 1
εἰδὼς ἀμφὶ θεῶν τε καὶ ἅσσα λέγω περὶ πάντων·
εἰ γὰρ καὶ τὰ μάλιστα τύχοι τετελεσμένον εἰπών,
αὐτὸς ὅμως οὐκ οἶδε· δόκος δ’ ἐπὶ πᾶσι τέτυκται. 4

And that which is clear and certain no man has seen nor will there be anyone
Who knows about gods and what I say about all things;
For even if, in the best case, someone succeeded in speaking what has been
fulfilled
Still he himself does not know; but belief is fashioned for all.
‘That which is clear and certain’ (to saphes) indicates veracity,
clarity and certitude.76 Numerous, generically heterogeneous texts
associate this terminology with the knowledge, pronouncements
or, once, person of the mantis.77 Xenophanes’ term for the object

74
Etymol. Mag. 574.69–75 Gaisford; Geor. Choer. in Theod. 4.1 200.3–5 Hilgard; cf.
Hesychius’ entry for ἐρέων: μαντευσόμενος. ἐρωτήσων. ζητήσων, sv. ε.5770 Latte; also
Olymp. in Alc. 70.1. For μαντεύομαι as ‘hunt down’ or ‘trace out’, cf. Theoc. 21.45.
75
Epiphanius and Sextus: below, n.83.
76
See Lesher (1992) 156–7; Tor (2013b) 10 with n.23; S.E. M 7.50; cf. Aesch. Ag. 1369–71.
77
Lesher (1983) 31 cites Il. 12.228–9 (σάφα θυμῶ ͂ν
ͅ εἰδείη); Od. 1.202 (μάντις . . . οἰωνω
σάφα εἰδώς) and 17.153 (οὐ σάφα οἶδεν, Theoclymenus contrasting his own divinatory

128
 3.3 Setting the Limits
of true statements and beliefs, ‘tetelesmenon’, conveys the sense
‘what has come to completion’ or ‘fulfilment’ and has its prove-
nance in Homeric divination.78 Diviners inquire above all into the
disposition and will of the gods (e.g. h.Hom. 3.131–2, 539a), the
fruition of whose designs is standardly expressed as their arrival at
their fulfilment (telos).79 The same terminology is used formulai-
cally for the formation and fulfilment of mantic predictions.80
In the Iliad and Odyssey, ‘what has been fulfilled’ is invariably
associated with statements about future states-of-affairs.81 It can
have the sense ‘what is to be fulfilled’.82 The formula ‘ah, stranger,
would that this word of yours might be fulfilled (τετελεσμένον εἴη)’
expresses the desire to see a prediction come to pass. It is
addressed twice to the seer Theoclymenus (Od. 15.536; Od.
17.163, where Theoclymenus just claimed sure, mantic knowl-
edge (σάφα): 153–4) and once to Odysseus upon his ‘prediction’
of his own imminent arrival (Od. 19.309). In Odyssey 19.547, the
eagle-Odysseus, undertaking the role of dream-diviner, employs
the term when divining from Penelope’s dream of the slaughtered
geese (ὅ τοι τετελεσμένον ἔσται). Xenophanes appropriates and
employs this Homeric notion of ‘what has been fulfilled’ more
broadly as the object of all unknowledgeable true statements and
beliefs ‘about the gods and what I say about all things’.
This last phrase specifies the scope of the matters concerning
which Xenophanes denies us knowledge. At this juncture, then,
when formulating the essential limitations of mortal epistemology,

knowledge with Telemachus’ ignorance). See further Soph. Philoc. 1338; OT 285–6,
390 (of the diviner himself); Eur. Hipp. 346; TrGF 5.483; Hdt. 7.228; playfully at Pl.
Phdr. 242c3–6; Rep. 7.523a8 (cf. also Phd. 69d4–6; Ti. 72b7–c1). The rejection of
divinatory knowledge-claims ‘about gods’ in Eur. TrGF 5.795 is especially reminiscent
of B34 (θάκοις μαντικοῖς ἐνήμενοι | σαφῶς διόμνυσθ᾽ εἰδέναι τὰ δαιμόνων . . . θεῶν
ἐπίστασθαι πέρι).
78
As convincingly argued by Lesher (1983) 29–30; (1992) 158.
79
Διὸς δ’ ἐτελείετο βουλή, Il. 1.5; cf. Th. 402–3; Op. 83; h.Hom. 4.10; h.Hom. 2.323.
80
μαντεύσομαι . . . ὡς τελέεσθαι ὀΐω (Od. 1.200–2; Od. 15.172–3); τὰ δὴ νῦν πάντα τελεῖται
(Il. 2.330; Od. 2.176; Od. 5.302; Od. 13.178; cf. Il. 14.48; Od. 18.271); cf. Zeus’
portentous eagle (τελειότατον πετεηνω ͂ ν, Il. 8.247; Il. 24.315); note τετελεσμένον as
divinely ‘appointed’, h.Hom. 4.572; cf. Op. 799.
81
The most common formula (14 occurrences) is ἐρέω, τὸ δὲ καὶ τετελεσμένον ἔσται (vel
sim.), used especially for threats and promises, e.g. Il. 1.212; Il. 8.401; Od. 2.187; cf. Il.
9.310.
82
As in the formula τελέσαι δέ με θυμὸς ἂνωγεν, εἰ δύναμαι τελέσαι γε καὶ εἰ τετελεσμένον
ἐστίν, Il. 14.195–6; Il. 18.426–7; Od. 5.89–90.

129
 Xenophanes on Divine Disclosure and Mortal Inquiry
Xenophanes is concerned in particular with mortal statements
about non-everyday, non-pedestrian, non-experienced matters.83
The phrase ‘about gods’ reflects this emphasis in itself.
Furthermore, ‘what I say about all things’ is unlikely to signify
all of Xenophanes’ statements unqualifiedly, since, if ‘all things’
(πάντων) includes the gods, their separate mention becomes cur-
ious. The phrase thus more likely signifies cosmological universal
generalisations84 or, since ‘all things’ can refer to more than just
cosmic processes (as we noted in Chapter 3.2), perhaps universal
generalisations as such. Xenophanes’ insistence that no mortal has
‘seen’ the clear and certain truth, and that even the one who
possesses true belief does not himself know (αὐτός), suggests
a concern with the unavailability of personal experience and
cognition. Our inability to speak knowledgeably about such mat-
ters is related to our inability, in these cases, to consider directly
and fully the objects of our statements or the entire body of
evidence which bears on them.85
Mantic disclosure was a, if not the, culturally dominant para-
digm of attaining knowledge concerning non-experienced
matters86 and, in particular, about matters involving the gods.87
When he points out that even the hypothetical best-case-scenario
of a true statement does not imply sure, personal knowledge,
Xenophanes undermines what could seem the strongest case for

83
Cf. Lesher (1991) 236; (2013) 85. Both Epiphanius and Sextus observe this emphasis:
μάλιστα τω ͂ ν ἀφανέων (Advers. Haeres. 2.2.9 = DG 590); τό γε ἐν τοῖς ἀδήλοις (M 7.51).
Alcmaeon DK24 B1 and the Hippocratic On Ancient Medicine (1) emphasise in terms
that recall B34 our inability to have knowledge concerning non-evident things; see Sassi
(2013) 297–8; Tor (2013b) 5, n.11.
84
Gomperz (1906) 164; Barnes (1982) 139; Lesher (1992) 167–8; Mogyoródi (2006)
132–3; cf. B29 and (if authentic) B27.
85
It does not follow that ἴδεν, εἰδώς and οὐκ οἶδε are expressions of knowledge restricted
specifically to perceptual cognition, see further Tor (2013a) 262 n.53. Since τὸ σαφές
expresses not an instance of truth but a generic concept of ‘the truth’, it is difficult to
minimise the non-sensory connotations of ἴδεν, see Lesher (1983) 37; Yonezawa (1989)
433; cf. Classen (1989) 100. We need not determine whether the elliptical οἶδε refers
back to τὸ σαφές (alongside ἴδεν and εἰδώς) or, less likely, introduces a second-order
clause, because knowing the truth about x and knowing that one spoke or believes truly
about x imply each other, as Hussey (1990) 18, n.21 observes.
86
Interestingly, when Anytus claims to know what sort of people sophists are despite
having no experience of them (ἄπειρος αὐτῶν), Socrates jokingly identifies mantic
means as the only possible explanation, Pl. Men. 92c4–7.
87
As in Eur. TrGF 5.795, n.77 above.

130
 3.3 Setting the Limits
mantic knowledge-claims, namely divination’s purported track
record (cf. Plato, Euth. 3c (περὶ τῶν θείων); Cicero, Div.
1.13.23).88 Xenophanes’ criticism is further supported by the
tension, inherent in the discourse of divination itself, between
mantic knowledge-claims and the widespread recognition of the
conjectural status of individual mantic statements (Chapter 3.1).
Nonetheless, and despite Xenophanes’ pointed adoption of mantic
terminology in B34, here too poetic inspiration is likely also high
on his agenda. Homer famously invokes Muses who witnessed,
and are thus knowledgeable, to guide the narrative of the ignorant
poet (Il. 2.485–6). Again, Hesiod’s Muses facilitate his disquisi-
tion on seafaring precisely despite his own inexperience
(Op. 660–2).
In B18, Xenophanes moves directly from rejecting traditional,
optimistic and naive notions of disclosure (B18.1) to describing
the actual epistemic predicament of mortals (B18.2). The two
couplets constituting B34 display, I believe, the same structure.
Xenophanes highlights matters external to mortal experience as
those concerning which mortals lack knowledge, and employs
mantic terminology for the objects of knowledge (to saphes) and
true beliefs (tetelesmenon). These considerations (along with the
parallel structure of B18) suggest that, in B34 too, Xenophanes
progresses from denying a traditional and overly optimistic con-
ception of disclosure (B34.1–2) to describing the actual predica-
ment of mortals (B34.3–4). Xenophanes derives the essential
epistemic limitations of mortals, their inability to attain knowl-
edge concerning non-experienced matters, from his rejection of
the notions of disclosure and mantic communication traditionally
thought to facilitate such knowledge and to enable mortals to
transcend such limitations.
The statement that no man will ever have knowledge about gods,
etc. (B34.1–2) indicates a modal thesis regarding what is possible
and impossible for mortals. Lesher (1983) himself recognises that

88
Following Lesher (1983) 33. τὰ μάλιστα marks a hypothetical best-case-scenario what-
ever its syntax, see Lesher (1992) 158. τύχοι could suggest an element of chance (contra
Yonezawa (1989) 433) but does not exclude volition pursued methodically (Fränkel
(1974) 126). Cf. h.Hom. 4.566, describing a successful mantic consultation (αἴ κε
τύχῃσι).

131
 Xenophanes on Divine Disclosure and Mortal Inquiry
his interpretation of Xenophanes’ scepticism leaves him with an
apparently invalid inference: why should the argument that mortals
cannot attain knowledge through divine disclosure merit the con-
clusion that they cannot attain it simpliciter? Lesher offers two
explanations. First, diviners enjoyed the status of paradigm cases.
Since the conditions of knowledge cannot be met in the most
promising cases, they cannot be met. Second, though he denies
divine disclosure categorically, Xenophanes retains the traditional
premise that only through divine disclosure could mortals attain
knowledge.89 Xenophanes is, I think, unlikely to have continued
to consider diviners the most promising candidates for knowledge
after systematically exposing their divination as an empty illusion.
Lesher’s second explanation initially looks more promising. Here
too, however, it is difficult to accept that Xenophanes uncritically
and unreflectively took on the assumptions operative in divination
concerning the conditions for the possibility of knowledge while
disagreeing only about the possibility of fulfilling those conditions.
Was Xenophanes unable to approach the question of knowledge
except through a conceptual-theological framework of divine
disclosure, which, according to Lesher, he himself repudiated
categorically as fundamentally ill-conceived? These unpalatable
ramifications stem from the oversimplifying assumption that
Xenophanes discards disclosure altogether. Xenophanes, I submit,
identifies what mortals can or cannot know through divine disclo-
sure with what they can or cannot know simpliciter because he
shares with those he criticises the fundamental premise that the
nature of divine disclosure determines the manner in which, and
certainty with which, mortals form propositions concerning non-
experienced matters. Although he indeed arrives at his view of
mortal limitations by rejecting traditional notions of disclosure, it
is no accident that Xenophanes positively employs divinatory termi-
nology (to saphes, tetelesmenon). The thrust of B34 is that, contrary
to traditional theological beliefs, divine disclosure is not such as to
make knowledge about non-experienced matters possible.
Disabused of the illusion that divine disclosure consists in the
indirect intimation of everything from the beginning (B18.1), and

89
Lesher (1983) 32–4.

132
 3.4 Divine Disclosure and Mortal Inquiry
is such as to bring sure and clear knowledge within mortal grasp
(B34.1–2), we can now address the role of disclosure in the
facilitation of mortal inquiry (B18.2) and the formation of con-
jectural beliefs (B34.3–4).

3.4 Divine Disclosure and Mortal Inquiry

In itself, B36 only speaks of a set of things which mortals
experience: ‘however many things they have disclosed [or:
have appeared] for mortals to look upon’. In all likelihood,
however, Xenophanes is interested in these objects of experience
as objects which the gods disclosed as evidence and which
mortals employed as such when forming beliefs and
conjectures.90 B18.1 rejects a misguided conception of disclo-
sure and B18.2 advocates in its stead temporally protracted
inquiries: ‘but searching in time they discover better’. Since,
then, Xenophanes advances in B36 his own conception of dis-
closure as a reversal of and an alternative to the one rejected in
B18.1, his conception of disclosure most probably relates to the
inquiries advocated in B18.2 and, therefore, to the way in which
mortals form judgements and beliefs through such inquiries.91
Xenophanes, furthermore, advances some notion of mortal-
oriented divine action (thnêtoisi pephênasin, B36) to replace
the one he rejects (thnêtois’ hypedeixan, B18.1). The echo
between the two phrases suggests that, in Xenophanean disclo-
sure, the gods still in some sense act purposively towards
mortals. Xenophanes’ use of the final infinitive ‘to look upon’
(εἰσοράασθαι) corroborates this impression and goes some way
towards sharpening the purpose in question. Standardly, such
final infinitives, not only imply volition in the performance of
the action signified by the finite verb, but also clarify the purpose
for which that action was performed. Regularly, x does some-
thing to or for y so that y performs another action signified by the
infinitive. Consider Odyssey 5.196–7: ‘the nymph laid out before
90
Scholars often see in B36 a connection between perception and judgement-formation,
e.g. Eisenstadt (1974) 149; Lesher (1992) 178–9.
91
Note, moreover, that φαίνω is a standard term for divine disclosure influencing mortal
belief-formation; see the passages listed above, n.42.

133
 Xenophanes on Divine Disclosure and Mortal Inquiry
him all kinds of nourishment to eat and drink (ἔσθειν καὶ
πίνειν)’.92 As the nymph lays out food before her guest in order
that he will eat and drink it, so too in B36 the gods show things or
(if we translate intransitively) things appear (through divine
agency) for mortals to look upon, i.e. in order that they will
look upon them.
The final infinitive ‘to look upon’ thus also supports further the
impression that Xenophanean disclosure concerns belief-
formation. Elsewhere in the extant fragments, we find
Xenophanes employing verbs of perception to express the import
of the perception in question for the formation, retention or revi-
sion of a belief or judgement: ‘purple, red and greenish-yellow to
behold’ (ἰδέσθαι, B32), concerning the nature of the rainbow as
cloud; ‘the upper limit of the earth is seen (ὁρᾶται) here at our feet’
(B28), concerning the finitude of the earth on this end; and, less
literally, ‘he would be more glorious to look upon (προσορᾶν) for
his townsmen’ (B2.6), concerning the perceived status of
a victorious athlete. Retaining also the perceptual connotations
of εἰσοράασθαι we may gloss B36 as ‘however many perceptible
things the gods disclosed for mortals to consider’.
As a thinker who is preoccupied with our inability to consider
directly all the evidence which bears on matters which exceed our
experience, Xenophanes is naturally concerned with the ways in
which our repertoire of experiences, so to speak, influences our
formation of beliefs. B38 instructively reflects this concern and
offers independent evidence for divine influence on mortal per-
ceptual experience and belief-formation:
εἰ μὴ χλωρὸν ἔφυσε θεὸς μέλι, πολλὸν ἔφασκον
γλύσσονα σῦκα πέλεσθαι.
If god had not made yellow honey, they would say
That figs are much sweeter.
Xenophanes’ counter-factual thought-experiment shows that, if
honey had not been part of the evidence available for us to con-
sider, we would have judged differently concerning figs. In the
first instance, B38 seems to caution that even some statements that

92
Cf. e.g. Il. 5.775–7; Th. 218–19; see further Smyth (1972) noo.2008–11.

134
 3.4 Divine Disclosure and Mortal Inquiry
concern what we experience (say, ‘honey is the sweetest food’ or
‘honey is very sweet’), and that appear to be grounded in a direct
inspection of the entire relevant body of evidence and not to
involve implications concerning what we cannot experience, do
in fact involve such implications, and are therefore corrigible
given the possibility of currently unknown counter-evidence.93
Unless mortals can exclude the existence of unknown members
of a class whose discovery would require a revision of current
judgements concerning other members of that class (as, for exam-
ple, the discovery of a substance sweeter than honey would require
us to revise our current judgements concerning the relative
sweetness of honey), those judgements can be asserted only
conjecturally.94 Many of our beliefs thus commit us to more than
what is empirically guaranteed by our experience.95 In this
respect, B38 implies further, I think, an a fortiori argument: the
same corrigibility would of course characterise statements about
non-experienced matters, such as the statement that all growing
things consist of earth and water (B29) on the basis of our obser-
vation of growing things within our experiential repertoire.
The central point for us is the following: to express a scenario in
which yellow honey was never available to us as evidence,
Xenophanes constructs a scenario in which god has never made
yellow honey available to us as evidence. Xenophanes’ point is
that if god had not enabled us to taste honey, we would have
formed different beliefs concerning figs and concerning sweet-
ness: his contention is not that we might have judged differently
had god not made honey in such times and places that we could not
taste it even if he had made it. Some scholars suggest that we can
ignore the theological language of B38 and that ‘if god had not
made yellow honey’ is merely a conventional way of saying ‘if
honey had never existed’.96 But they offer no parallels for this

93
Some, not all: we can say with certainty that all the figs we have tasted thus far were less
sweet than all the honey we have tasted. Still, B38 excludes the possibility that knowl-
edge is secured for mortals concerning everything they perceive directly. Aëtius reports
that Xenophanes traced the common but mistaken belief in the sun’s circular orbit to
misleading perceptual appearances (A41a); cf. Guthrie (1962) 397–8.
94
Cf. Mogyoródi (2006) 139.
95
Following McCoy (1989) 237; Mogyoródi (2006) 138.
96
Guthrie (1962) 376; Granger (2013a) 266; (2013b) 172, n.13; Curd (2013a) 231, n.32.

135
 Xenophanes on Divine Disclosure and Mortal Inquiry
purportedly conventional way of speaking. The assertion that god
caused the growth (ἔφυσε) of honey seems, on the contrary, rather
distinctive. At any rate, our discussions of other reflections in
Xenophanes (B18, B36, B34) indicate that we should take the
theological language of this fragment seriously. In B38,
Xenophanes is not simply observing that judgements are
constrained by the available evidence. The fragment reflects
a broader point concerning the relation between the range of
experiences with which the divine presents us and the beliefs
and conjectures we form on the basis of those experiences. Had
god not facilitated for any mortal the particular experiences he did,
or had he facilitated different experiences in addition, that mortal’s
judgements might have turned out otherwise. B38 and B36 thus
illuminate one another. B38 demonstrates the same preoccupation
as B36 with divine facilitation of mortal perceptual experience,
but, unlike the truncated B36, explicitly connects this facilitation
with the formation of judgement. Again, taken in isolation, B38
shows only the influence of divine action on mortal perceptual
experience and so belief-formation. B36 demonstrates that
Xenophanes could conceive of this sort of divine influence as, in
some sense, purposive (θνητοῖσι πεφήνασιν εἰσοράασθαι).
Any attempt to determine more precisely what kind of volition
the gods display towards us would be necessarily speculative.
I would like to outline two possible alternatives. On the first
alternative, which I tentatively favour, Xenophanes radically
reconceptualises the notion of divine disclosure as the notion
that the gods have purposively enabled us to perceive and consider
everything that we perceive and consider (call this ‘universal
disclosure’). On the second, only some of the things that we
experience and consider have been brought to our attention and
disclosed for our consideration by the gods (‘particular disclo-
sures’). In other words, when Xenophanes speaks of ‘however
many things’ (ὁππόσα) have been disclosed to us, is he referring
to the totality (universal disclosure) or only to a subset (particular
disclosures) of the perceptual experiences of mortals?
Let me first clarify what I mean by ‘universal disclosure’. Since,
as mortal agents, we are essentially limited both spatially and
temporally, the range of things we can perceive and consider is
136
 3.4 Divine Disclosure and Mortal Inquiry
also necessarily limited. Furthermore, it is on the basis of this
limited range of things that we form beliefs and conjectures about
any matter, whether internal or external to our experience.
According to universal disclosure, the divine determines the
scope and content of the experiences included within the necessa-
rily limited experiential repertoire of any mortal agent, intending
(among whatever other purposes it has, of course) to facilitate the
general discursive engagement of mortals with their surroundings
and so their formation of beliefs on the basis of their experiences
(πεφήνασιν εἰσοράασθαι). B25 is pertinent here:
ἀλλ’ ἁπάνευθε πόνοιο νόου φρενὶ πάντα κραδαίνει.
But without toil he shakes all things by the thought97 of his mind.
Xenophanes’ specification that his supreme god possesses the
psychic organs ‘thought’ (phrên) and ‘mind’ (noos), and exerts
his cosmic influence through them, strongly indicates a notion of
intelligent and purposeful cosmic governance.98 The verb noein
signifies the cognition of a situation, analogous in its non-
inferential operation to sense-perception, as with Xenophanes’
god at B24: ‘Whole he sees, whole he cognises (noei), whole he
hears.’99 Already in our earliest sources, however, noein desig-
nates further a volitional reaction to the situation cognised, where-
fore both the verbal and substantive forms came to signify the
activity or product of planning and, from similar volitional reac-
tions to similar situations, the disposition and character of the
cognising and planning agent.100 The ever-observant mind of
97
Or: ‘by the volition’.
98
Cf. Drozdek (2004) 147–8; von Fritz (1974) 33–4; Warden (1971) 10; Lesher (1992) 104,
107–9. I cannot see how the issue is influenced either way by Xenophanes’ image of
cosmic ‘shaking’, from which some scholars infer that divine influence falls short of
intelligent direction, e.g. Hussey (1990) 27, n.44; Granger (2013a) 257, 266–7. Attempts
to emend κραδαίνει are arbitrary and unnecessary, see Palmer (1998) 10–11 with n.17; cf.
Lesher (1992) 107. The precise relation between god’s phrên and noos is unclear. φρενί is
most plausibly an instrumental-locative dative, signifying the psychic organ with and in
which god’s mind operates; see Darcus (1978) 26; Tor (2013a) 267 n.66. We could
conceivably relate νόου also back to πόνοιο (‘without toil of mind’), but the juxtaposition
of the two psychic terms would still strongly encourage us to take νόου together with
φρενί as well. There is certainly no scope for divorcing somehow god’s phrên from his
noos. Both terms convey the same connotations of intelligence, will and purpose. In B24,
οὖλος δὲ νοεῖ encapsulates god’s unified mental life quite generally.
99
Cf. Il. 3.396ff; Od. 16.160.
100
Plan: Il. 4.308–9; Od. 2.122; disposition: Il.16.34–5; Od. 1.3; Od. 9.175–6.

137
 Xenophanes on Divine Disclosure and Mortal Inquiry
Zeus (Il. 15.461) is frequently his will or plan.101 In this respect,
Zeus’ superior mind not only comprehends but also moulds and
fashions the pattern of events, sometimes directly affecting or
determining mortal actions.102 The phrên of Zeus can similarly
signify his volition.103 In a close parallel to B25, Aeschylus
describes how god, remaining motionless (cf. B26), realises with-
out toil his ‘will’ or ‘purpose’ (phronêma, Supp. 100–3). Most
often, phrên is the location or instrument of cognition, emotion,
deliberation and planning.104
Xenophanes posits a perfect correspondence between god’s
intelligent volition and its effortless realisation in states of affairs.
Notably, Zeus’ Olympus-shaking nod to Thetis in the Iliad
(1.528–30), an oft-cited parallel to B25, indicates his considered
adoption of a course of action, which will now progress to its
inexorable conclusion.105 As we saw, the fruition of divine plans is
standardly expressed as their arrival at their telos.106 In the light of
B25, Xenophanes’ talk of the object of true statements as ‘what
has been fulfilled’ (tetelesmenon, B34.3) stems from a theological
world-view that construes states-of-affairs as the fulfilment of
divine volition. Our evidence belies Lesher’s (1983) assimilation
of Xenophanes’ implicit view to the explicit Epicurean one, that
celestial events take place without the ministration of an immortal
being.107 His statement (1992) that Xenophanes’ cosmology
banishes the traditional gods ‘to the explanatory sidelines’ is
misleading.108 Xenophanes forgoes entirely rather than side-
lines the traditional, anthropomorphic gods, but his conception
of divinity retains cosmological explanatory prominence. Nor can

101
Il. 8.143; Th. 1002; h.Hom. 4.10.
102
Il. 15.242; Il. 16.103; Od. 24.164; cf. Pind. P 5.122–3. For these features of noos and its
cognates, and for further parallels, see von Fritz (1974) esp. 23–4, 33; Warden (1971)
3–9; Darcus (1978) 26; (1994) 107–8, 119ff; (1995) 37, 44ff.
103
Il. 10.45–6; Il. 12.173; Il. 15.194; of agents other than Zeus: Soph. Ant. 993 (φρενός);
OC 1182 (φρενί); cf. von Fritz (1974) 34.
104
Zeus’ deliberation and planning: Il. 2.3; Il. 16.435; of other agents: Il. 5.671, Il. 10.4, Il.
16.83; Il. 21.19; Od. 11.146, 204, 474; h.Hom. 4.66; h.Hom. 3.257. See further Darcus
(1978) 26; (1994) 109–11; (1995) 39.
105
Cf. Mogyoródi (2002) 283, n.140.
106
Ch. 3.3. Note Th. 1002; h.Hom. 4.10 (μεγάλοιο Διὸς νόος ἐξετελεῖτο); h.Hom. 23.2; h.
Hom. 2.323; Thgn. 142 (θεοὶ δὲ κατὰ σφέτερον πάντα τελοῦσι νόον).
107
Lesher (1983) 28; cf. Barnes (1982) 96; Granger (2013a) 267.
108
Lesher (1992) 137; also Classen (1989) 96.

138
 3.4 Divine Disclosure and Mortal Inquiry
I agree with Mogyoródi that Xenophanes maintains ‘a categorical
distinction between nature and the divine’.109 First, god’s dissim-
ilarity cannot be categorical. For one thing, if god’s physical
dissimilarity (οὔτι δέμας θνητοῖσιν ὁμοίιος) meant that he lacked
a physical body, by parity of reasoning he would also be left
without a mind (οὐδὲ νόημα, B23.2).110 Second, whatever the
spatial relation between god and the natural world (if
Xenophanes ever addressed this question), B38 guarantees that
god facilitates directly even minute processes occurring in that
world, while B25 indicates that such facilitation is exercised
intelligently, purposively and universally.
God’s ‘growing’ of honey (to render ἔφυσε, B38) is, then, one
instance of the causal facilitation of such ‘growing’ in general.
Furthermore, the class of things that ‘become and grow’ (ὅσα
γίνοντ’ ἠδὲ φύονται) consist of earth and water (B29), and we
mortals too ‘come into being from earth and water’ (ἐκγενόμεσθα,
B33). Thus, the continual emergence and preservation of particu-
lar mortal lives are also themselves part of god’s intelligent cosmic
governance of all things. In disclosing things for mortals to con-
sider, the divine could display towards them the same sort of
cosmic, intelligent volition which it displays in directing all nat-
ural processes, however large-scale or minute. Making possible
and realising (i) the scope and content of any mortal’s experiential
repertoire, (ii) mortals themselves as perceptive and cognitive
agents and (iii) the discursive engagement of mortals with the
world around them (εἰσοράασθαι, B36) would thus be part of the
divine’s cosmic plan. On this interpretation, ‘however many
things’ (ὁππόσα, B36) signifies the totality of mortal experiences.
God’s making of honey represents a genuine instance of disclo-
sure, while mortal belief-formation and conjectural reasoning are
indeed informed and constrained by the evidence which the divine
resolved to disclose (B38). Xenophanes, on the universal inter-
pretation, discards the mantic idea that a subset of our experiences
was occasioned by the divine to encourage the formation of
particular judgements. Rather, he counters, everything we see,

109
Mogyoródi (2002) 264; (2006) 144–5, 150.
110
Cleve (1965) 10; Broadie (1999) 211; cf. KRS (1983) 172; Lesher (1992) 94.

139
 Xenophanes on Divine Disclosure and Mortal Inquiry
everything we experience has appeared to us (in part) for us to look
upon and consider.
Importantly, it does not follow from this interpretation that the
divine is especially concerned with mortals. B36 reflects rather
that Xenophanes is especially concerned with this particular aspect
of intelligent and purposeful divine actions, namely, the sense in
which some of them purposively facilitate, among innumerable
other things of course, mortal experience, perception and belief-
formation (θνητοῖσι πεφήνασιν εἰσοράασθαι). By maintaining that
the divine enables us to experience, and to engage with our
experiences discursively, Xenophanes would of course not be
suggesting that it runs the cosmos simply or especially so that
mortals will have something to think about.111
Also pertinent here is Xenophanes’ statement that ‘belief is fash-
ioned (τέτυκται) for all’ (B34.4b). In Homer, the term ‘fashioned’
commonly signifies things fashioned for mortals by gods.112
Indeed, it can indicate the lot of mortals (and also immortals) as
determined by their place within a world order regulated by divine
agency only through which things are (made to be) what they are.
For example, Achilles must, qua mortal and as even Heracles did,
accept his death whenever Zeus and the other immortals fulfil it
(τελέσαι), if such a lot is ‘fashioned’ for him (τέτυκται, Il.
18.115–21).113 ‘From Zeus all things are fashioned’ (Διὸς δ’ ἐκ
πάντα τέτυκται), an oft-cited Orphic slogan as old as the Derveni
theogony, occurs in the context of hymns celebrating Zeus as the
ruler and preserver of both cosmic events and specifically mortal
existence.114 Universal disclosure specifies a meaningful sense in
which the divine indeed fashions opinion or opining (dokos) to all.
According to the alternative, ‘particular disclosures’ interpreta-
tion, when Xenophanes speaks of ‘however many things’ have

111
I thank Patricia Curd for pressing me to clarify this point.
112
As Lesher (1992) 159 observes.
113
Cf. Il. 3.101; h.Hom. 5.29–32; h.Hom. 2.269; Il. 4.84 = Il. 19.224; Il.14.246; Il. 21.191;
Hes. Op. 744–5, 752. Cf. the foundation of our world-order by Zeus, Poseidon and
Hades: h.Hom. 2.86 (ἐτύχθη).
114
PDerv. col.17.12 (cf. col.19.10); Pl. Lg. 4.715e8–716a1 with the scholiast; ps.-Ar. Mu.
401a25ff (ἀρχὸς ἁπάντων); Porph. apud Eus. PE 3.9.2 (μέγας ἀρχὸς ἁπάντων); for
these passages, cf. Palmer (1998) 26–31 (not discussing B34.4). Classen (1989) 100
mentions the parallel.

140
 3.4 Divine Disclosure and Mortal Inquiry
been disclosed to us, he is referring only to a subset of our
perceptual experiences. On this reading, the divine brings parti-
cular things to the consideration of some mortals (εἰσοράασθαι) in
some circumstances. This will not mean that the gods communi-
cate messages. Rather, they guide those mortals in their formation
of inquisitive (ζητοῦντες, B18.2) beliefs, and perhaps also every-
day ones.115 Xenophanes could have been concerned to insist that,
given the essential spatio-temporal limitations of mortals, even if
some conjectures are based on such disclosures, they cannot sup-
port knowledge-claims (B34). We should not exclude offhand the
possibility that Xenophanean deities could show such concern for
individual mortals. Xenophanes asks us to pray for the capacity to
act justly (B1.13–16), possibly signalling that the divine may aid
those whose prayers it heeds to do so. His criticisms of imputations
of moral transgressions to the gods (B11–12) could imply that
such predications constitute category errors, but could equally
suggest rather an insistence on the moral goodness of the gods
for whom we should always have regard (B1.24) and to whom we
should pray for the capacity to act justly.116 The latter alternative
is perhaps supported by Xenophanes’ denunciations of some false
theological views as impious and unseemly: it is not immediately
intelligible why predications of attributes to categorically amoral
and impersonal things should occasion this sense of religious
offence.117 Again, Xenophanes reportedly observed that an oath
is not an equal challenge for pious and impious men, plausibly
because the latter do not fear divine retribution.118
Nonetheless, Xenophanes emphasises the radical cognitive
dissimilarity of the greatest god (B23) and questions in general

115
God’s making of honey in B38 would reflect a preoccupation with divine influence on
mortal beliefs without necessarily constituting an instance of the disclosure spoken of
in B36.
116
Category mistake: Mogyoródi (2002) 273–4; Reiche (1971) 97–8. Moral goodness:
Heidel (1943) 258; Barnes (1982) 93–4; Granger (2013a) 241.
117
οὐδέ . . . ἐπιπρέπει (B26, with Lesher (1992) 112); ἀσεβοῦσιν (Ar. Rhet. 1399b6–9 =
A12); οὐ γὰρ ὅσιον (ps.-Plut. A32).
118
Ar. Rhet. 1377a19 = A14. See the analysis of this report in Lesher (1992) 201, n.15.
Particular disclosures would still be embedded within a broader system of intelligent
and purposive cosmic governance through which things are generated and preserved
(as discussed above). Xenophon interestingly reflects on providence on both the
particular and universal levels, Mem. 1.4.2–18; 4.3.11–13.

141
 Xenophanes on Divine Disclosure and Mortal Inquiry
the beliefs that gods are like us in their nature, conduct and
appearance (B10–16). One could reasonably insist that,
Xenophanes’ talk of prayers for a moral disposition notwith-
standing, the cognitive dissimilarity of divinity should involve
a lack of particular awareness of, or concern for, particular mortal
affairs, and that the sort of god Xenophanes is comfortable
postulating is consequently a more removed being than particular
disclosures require. When he affirms that it is not proper for god
‘to travel to different places at different times’ (μετέρχεσθαι,
B26), Xenophanes uses a verb of motion which connotes direct
divine interference in particular mortal affairs.119 Indeed, if
Xenophanes held that gods disclose particular things to particular
mortals in particular circumstances, it becomes more difficult to
see what he found so objectionable about mantic communication
in the first place.
Arguably, then, universal disclosure is the more probable inter-
pretation of the notion of divine, mortal-oriented intentionality
operative in Xenophanes’ conception of divine disclosure.
Universal disclosure would, and particular disclosures could,
influence the formation of mortal beliefs, not just concerning the
natural world, but whenever a discursive engagement with observed
experience plays a formative role. Xenophanes’ critically digested
experience of the fall of Colophon, for example, supports his con-
tention that luxury is socially deleterious (B3). But this raises an
interesting albeit irremediably speculative question: does
Xenophanean disclosure influence the formation of beliefs about
gods? Our evidence for Xenophanes’ theological methodology is
scarce. Some scholars ascribe to Xenophanes aprioristic theological
reasoning and even deny that conjectures based on sense-
experience play any part whatsoever.120 Xenophanes comments
that Ethiopians and Thracians fashion gods in their own divergent
images (B16) and that, if horses, oxen or lions could draw, they
would draw gods looking like horses, oxen or lions (B15). These

119
E.g. Il. 5.456, 461.
120
Barnes (1982) 82–99; Mogyoródi (2002) 265; (2006) 145–6. For theological reasoning
in Xenophanes, cf. also McKirahan (1994) 61–2; Warren (2007) 46, 54; Granger
(2013a) 235–6. Others suggest divine inspiration (Cleve (1965) 28–30) or mystic
intuition (Jaeger (1947) 49; Nietzsche (1974) 120–1).

142
 3.4 Divine Disclosure and Mortal Inquiry
remarks arguably suggest suspicion towards naïve, uncritical deri-
vations of beliefs about gods from our parochial experience of
ourselves as other sentient beings. In B26, moreover, Xenophanes
appears to draw inferences from some criterion of theological
propriety: it is not ‘proper’ for god to travel (οὐδέ . . . ἐπιπρέπει).
But we may wish to soften the assertion that theology proceeds
without recourse to observed experience. First, positive cosmologi-
cal propositions can constitute negative theological propositions.
She whom they call Iris is in fact a cloud, ‘purple, red and greenish-
yellow to behold’ (ἰδέσθαι, B32; cf. εἰσοράασθαι, B36; see also
A38, A43).121 Again, Xenophanes’ use of an ethnographic insight
concerning the divergent representations of the gods among the
Ethiopians and Thracians (B16) shows that such empirical observa-
tions can inform his critique of anthropomorphism. Second, I see no
Xenophanean reason to exclude the possibility that (purported)
observed experience may positively influence our conjectures
about unseen gods. For example, Xenophanes maintained that
suns are generated and quenched anew each day (A33, A41).
Such cosmic regularities can plausibly be taken to corroborate the
belief that a divine intelligence directs those processes (B25).122
Finally, since many of Xenophanes’ theological propositions take
the via negativa (emphasising what the gods are not like), the
conceptual framework of his theology is intelligible only in refer-
ence to those attributes whose ascription to the gods he rejects.
The gods, for example, do not wear clothing (B14) or commit
adultery (B11–12). In this weak but important sense, observed
experience is the basic condition of all discursive inquiry.123

121
Broadie (1999) 209 argues similarly.
122
Aristotle writes that Xenophanes formed the view that ‘the One is the god’ (or that ‘the
god is the One’) ‘having looked towards the whole heaven᾽ (εἰς τὸν ὅλον οὐρανὸν
ἀποβλέψας, Metaph. Α.5 986b21–7 = A30). Although the view itself could hardly be
Xenophanes’, Aristotle may be reflecting the fact that empirical observation played
a role in his theology. Palmer (1998) 5–7 persuasively argues that ‘heaven’ signifies
here, not what Xenophanes referred to when he stated that ‘the One is the god’, but what
he considered before arriving at that view. Cf. Aristotle apud S.E. M 9.22, tracing the
belief in divine cosmic regulation to observations of well-ordered celestial motions; see
also Cic. ND 2.37.
123
Gábor Betegh suggests to me yet another possible interpretation of the logic of B18.1,
which would also take Xenophanes to reject traditional notions of divine disclosure and
to supplant them with his own alternative notion of disclosure, as I have reconstructed it

143
 Xenophanes on Divine Disclosure and Mortal Inquiry
3.5 A52 Revisited: A Clean Sweep?
What can we say, then, about Xenophanes’ engagements with
culturally and theologically dominant models of divine disclo-
sure? As we saw in Chapter 3.1:
(ii)a diviners form propositions concerning non-experienced states of affairs on
the basis of experienced events or objects intentionally disclosed by a divinity for
that purpose.

Diviners form fallible, conjectural beliefs by interpreting proposi-

tions that gods encode in a subset of the objects of mortal percep-
tion (omens, oracles). Not generally restricted to chosen
individuals, mantic interpretation is open to any mortal informed
of the fundamental semiotic or exegetical principles and capable
of interpretative reasoning. Xenophanes rejected the view that the
gods cryptically communicate any and every truth. He retained,
however, the core principle that, by disclosing objects of experi-
ence to them, the divine purposefully enables mortals to reason out
conjectures about states of affairs external to their experience, the
world around them and their place within it. Xenophanes most
probably reconceptualised disclosure as the purposive facilitation
of mortal judgement through the facilitation of mortal experience
as a whole (‘universal disclosure’). If so, then divine disclosure

in Ch. 3.2–4. Perhaps with the prefix hypo in the exceptional and pointed term
‘intimate’ (hypedeixan) Xenophanes reflects his own notion of divine disclosure as
opposed to traditional notions and not, as I argued in Ch. 3.2, vice versa. Might
hypedeixan reflect the indirectness which arguably characterises divine disclosure on
Xenophanes’ own view? After all, for Xenophanes, as I reconstructed his position, the
divine does not communicate truths to mortals, but, rather, purposively renders mortals
able to form true beliefs about matters external to their experience by facilitating their
discursive engagement with their surroundings. In B18.1, Xenophanes could be stipu-
lating that not even in this indirect manner did the gods disclose to mortals everything
from the beginning. This interpretation is equally and similarly non-identical with the
‘minority’ interpretation because it takes Xenophanes not simply to restrict the tem-
poral and quantitative scope of divine disclosure but also to qualify the manner of
disclosure (hypedeixan). Now, many of the observations in Ch. 3.2–4 concerning the
language and rhetoric of B18.1, and concerning its apparent relation to other extant
fragments and to certain prominent contemporary theological and epistemological
attitudes, favour the particular analysis of the line’s logic which I advanced above
and according to which the term hypedeixan, in the context of B18.1, gestures towards
traditional models of divine disclosure. This remains my preferred reading. But I can
see nothing to exclude conclusively Betegh’s ingenious alternative suggestion, which
would amount to the same overall view of Xenophanes’ epistemology and theology as
the one I have defended in this chapter.

144
 3.5 A Clean Sweep?
becomes the essential condition of all mortal belief-formation and
inquiry. Less likely, the gods disclose to mortals a subset of the
perceptual objects which they experience in order to guide the
formation of particular beliefs in life and inquiry (‘particular
disclosures’).
A case in point: on the basis of marine fossils found inland,
Xenophanes conjectured that the earth was submerged in the
distant past and will be again in the future (A32–3).124
Herodotus shows that even here Xenophanes advances on
what would fall naturally within the epistemic reach of mantic
communication. Onomacritus, he writes, would have succeeded
in falsely ascribing to the ancient Musaeus an oracle stating that
the islands lying off Lemnos would disappear under the sea had
he not been caught red-handed in the act of interpolation (7.6).
Within the framework of his theology, Xenophanes’ conjectures
about the earth are in effect propositions about the content of
divine volition, which will later be realised in states-of-affairs
(‘by the thought of his mind’, B25; ‘what has been fulfilled’,
B34). If in B36 Xenophanes was gesturing at the more familiar
idea that the gods may guide certain mortal inquirers, then the
discovery of marine fossils would be a prime candidate for an
instance of such disclosure. More probably, however, it is in the
sense that a purposive divine intelligence enabled and governed
his existence as a perceptive and discursive agent, and deter-
mined the scope and content of his experiential repertoire,
that Xenophanes would maintain that the fossils he observed,
like the honey he tasted, were disclosed for his consideration by
the gods.
Xenophanes does not endorse some modified form of the art of
deciphering divine messages. Rather than messages, the gods
disclose evidence. Xenophanes advances his conception of divine
disclosure against the traditional one and we should not downplay
the differences between the two. Although divination never mar-
ginalised mortal agency and reasoning, Xenophanes lays revolu-
tionary emphasis on our role in expanding the available body of

124
For a rigorous and sympathetic account of Xenophanes as a serious natural philoso-
pher, see Mourelatos (2008d).

145
 Xenophanes on Divine Disclosure and Mortal Inquiry
evidence by pursuing temporally protracted inquiries, rather than
by decoding entrails or birds.
(ii)b Since mortals can attain knowledge or insight concerning these non-
experienced states only on the basis of omens, expert and able diviners are
taken to transcend the epistemic limitations of all other mortals and to approx-
imate the epistemic state of the gods in particular and momentary respects.

Like his opponents, Xenophanes takes it that the nature of divine

disclosure determines the epistemic potential of mortals. But,
whereas divination and poetic inspiration enable those individuals
who experience them to transcend the limitations of other mortals,
Xenophanes most probably rendered disclosure the fundamental
condition of all mortal discursive activity (universal disclosure).
On any interpretation, Xenophanes recognises that like any man
(ἀνήρ), past or future, divine disclosure does not place him in
a position to claim knowledge (B34).125 B35 appears to offer us
an insight into Xenophanes’ reflections on his own epistemic
predicament:
ταῦτα δεδόξασθαι μὲν ἐοικότα τοῖς ἐτύμοισι . . .
Let these things be believed as like the truths . . .

Since Plutarch (Mor. 746b) cites the fragment when exhorting

a bashful companion to state his opinions outright, adding that
Ammonius habitually cited it to the same effect, ‘these things’
(ταῦτα) is best read as referring to views which Xenophanes
himself endorses as instances of belief (dokos, B34.4).126
The subject matter of these views is presumably coextensive
with the range of things about which mortals are restricted to
opining (‘about gods etc’, B34.2).127 The crux lies in assessing
the further sense in which Xenophanes qualifies his endorsement
of his views with the loaded phrase ‘as like the truths’ (ἐοικότα τοῖς

125
We certainly cannot take B34 to identify the limitations of every mortal, past or future
(!), except Xenophanes, pace Snell (1953) 140–5; Cleve (1965) 27–30; Yonezawa
(1989) 433; Wiesner (1997) 22; Drozdek (2004) 152; Gemelli Marciano (2002) 93–4.
Xenophanes traces the epistemic limitations of mortals to their essential spatio-
temporal limitations (Ch. 3.3). He indicates no exceptions to the principle that ‘belief’
is allotted to all (ἐπὶ πᾶσι, B34.4); see further Lesher (1992) 167; Mogyoródi (2006)
150, n.88.
126
Cf. Bryan (2012) 16–17.
127
Barnes (1982) 140; Lesher (1992) 175; Bryan (2012) 17.

146
 3.5 A Clean Sweep?
ἐτύμοισι). Most persuasively, Bryan argues that Xenophanes is
endorsing his beliefs as apparently similar to truths, where this
apparent similarity could mean that his beliefs are true,128 but
qualifies his endorsement with the warning that this similarity
could also be specious.129 After all, Xenophanes allows that mor-
tal beliefs can be completely true (B34.3). In B35, then, he should
accommodate the possibility that his views are true. Excluding this
possibility, and determining in advance that his views can only
ever merely resemble the truth but fall short of it, would be to
imply equally unwarranted conviction about their truth-value as
making a knowledge-claim. But Xenophanes offers a qualified
endorsement. His views are an instance of corrigible belief.
It was precisely its qualified nature that rendered the phrase so
appealing for Ammonius as an encouragement to one’s bashful
interlocutors to do all a mortal can, that is, to recommend one’s
speculative views forthrightly as fallible candidates for truth.
Side-by-side with this epistemological self-awareness, however,
Xenophanes is not bashful about contrasting himself favourably
with his peers. Xenophanes questions beliefs which mortals hold
quite generally: ‘but the mortals (οἱ βροτοί) believe that gods are
born etc’ (B14). The definite article creates the impression that this
belief is held universally among mortals, excepting Xenophanes
himself.130 Xenophanes decries how ‘all (πάντες) have learned
according to Homer’ (B10), again suggesting his unique exclusion
from a universal consensus. Quite generally, the Xenophanes who
emerges from our sources is one urgently aware of the preferability
of his own views to the deeply and disastrously misguided autho-
rities, beliefs and values of his contemporaries. In the surviving
evidence alone, his invectives cover anthropomorphism, divination,
knowledge-claims, poetic myths (‘the fabrications of older genera-
tions’, B1.22), the social veneration of athletes (B2), luxury (B3),
stinginess (B21) and metempsychosis (B7); his targets include

128
Cf. Il. 3.170: Agamemnon looks like a king (ἔοικε).
129
Cf. Il. 3.219: Odysseus looks like a fool (ἐοικώς). See further Bryan (2012) 6–57. For
the translation ‘truths’, see Bryan (2012) 25–6. We could, mutatis mutandis, interpret
similarly the programmatic thrust of B35 while glossing ἐτύμοισι as ‘realities’. On the
term etymos, see Ch. 2.1.
130
Yonezawa (1989) 435–6; cf. also Aikin (2016) 173–4.

147
 Xenophanes on Divine Disclosure and Mortal Inquiry
mortals quite generally (B14; cf. B16 on Ethiopians, Thracians and
(implicitly) Greeks), Homer and Hesiod (B11–12), Olympic victors
(B2), Colophonians (B3), Simonides (B21), Pythagoras (B7) and
Epimenides (A1).131 Xenophanes is hardly coy about advertising
his intellectual services and highlighting his subversive iconoclasm:
‘Our wisdom is better than the strength of men and horses; but these
customs are quite arbitrary; nor is it just to prefer strength to this
good wisdom’ (B2.11–14).
Xenophanes’ sense of superiority most probably derives at least
in part from his revolutionary recognition of the epistemological
limitations all mortals share and from a lifetime of inquiry
informed by this recognition. Disabused of theological fantasies
and informed of the pitfalls haunting mortal belief-formation,
Xenophanes developed his cosmological, theological and socio-
moral world view through a lifetime of intellectual inquiry to
which he himself bears witness: ‘already there are seven and
sixty years tossing about my thought throughout the land of
Greece’ (B8). The Odyssey features prominently the idea that
learning and wisdom are expanded through wandering. This idea
became a central methodological tenet among proponents of
Ionian ‘inquiry’ (historiê).132 Xenophanes’ principle ‘as they
search in time they discover better’ (B18.2) is instantiated most
dramatically by his observations of marine fossils found inland,
which support his novel conjectures about the earth’s past and
future submersions (A32–3) and, perhaps, the preferability of
these conjectures to Anaximander’s theory of the earth’s desicca-
tion (DK12 A27).133 The critical expansion of our experiential
repertoire may conclusively falsify a previously held belief (e.g.
‘figs are the sweetest substance’) or inconclusively justify the
formation or retention of conjectures by assessing them against

131
We should possibly include also Bacchanals (B17) and Thales (D.L. 9.18.11 = A1), cf.
Yonezawa (1989) 434–5; Gemelli Marciano (2002) 90–6.
132
Esp. Od. 1.1–4. Ionian inquiry: e.g. Hdt. 1.30 (. . . φιλοσοφέων γῆν πολλὴν); 4.76 (. . .
γῆν πολλήν . . . σοφίην πολλήν); Hecataeus is styled ‘a well-travelled man’ (ἀνὴρ
πολυπλανής, BNJ 1 T12a); cf. Democ. DK68 B299.6–8; B68; D.L. 1.43–4; see further
Montiglio (2000) 88–90; (2005) 100–1, 123–46.
133
For the contrast with Anaximander, cf. Mogyoródi (2006) 134, n.40; cf. KRS (1983)
177–8.

148
 3.5 A Clean Sweep?
an ever-growing body of evidence.134 Mortals who reason from
divinely encoded messages which are not there, who acquiesce in
delusions of poetic inspiration, or who form beliefs on the basis of
their parochial experiential repertoire blithely and uncritically, are
much more liable to go astray.
It is sometimes thought that, by saying that mortals discover
better as they search in time, Xenophanes indicates (in part) an
idea of gradual social and cultural progress. Lesher argues plau-
sibly, though not conclusively, against this interpretation (he
observes that Xenophanes is acutely aware of socially destructive
practices, developments and authorities: B1–3; B10–12).135
Be that as it may, Xenophanes does claim the skill (sophiê) to
engender lawfulness (eunomiê) and economic prosperity in the
city (B2), i.e. to effect socio-political improvement. This claim
does not rest on one isolated branch of his teaching. Xenophanes’
ethical reflections on Colophon’s fall, for example, support his
contention that luxury is socially deleterious (B3). In turn, his
cosmological and theological world view exposes well-heeded
authorities, from Homer and Hesiod to diviners and other self-
styled religious experts, as charlatans who advocate often deleter-
ious counsel backed by ill-conceived notions of divine disclosure.
Xenophanes’ critiques of poetic representations of divinities as
creatures liable to stasis (B1.21–3) and lawlessness (B12) are one
reflection of his encouragement of civic lawfulness.136
Conceptions of the gods as human-like and, therefore, lawless
are apt to engender lawlessness in the mortal sphere.137 We are
left with Xenophanes as a holistic reformer, who engenders civic
lawfulness and prosperity by spreading a particular world view,

134
Cf. Reiche (1971) 88; Darcus (1978) 39, n.58; Warren (2007) 53. Lesher’s reading of
ἐφευρίσκουσιν ἂμεινον as a better method of discovering (i.e. Ionian historiê: Lesher
(1991) 244–6; (1992) 153) highlights a likely connotation, but it is implausible to
exclude a corresponding notion of improved results.
135
Lesher (1991) 231–7; (1992) 151–2; cf. Schäfer (1996) 124–7.
136
Note B1.23: ‘in these [sc. narratives] there is nothing beneficial (χρηστόν)’. See further
Lesher (1992) 61; cf. Bowra (1953a) 10–11. Aristotle reports that Xenophanes criti-
cised (unidentified) poetic narratives as neither true nor beneficial, Poet. 1460b10–11,
1460b35–1461a1 (οὔτε βέλτιον οὕτω λέγειν οὔτ᾽ ἀληθῆ).
137
See Warren (2013) on Xenophanes’ attitude to the ethical consequences of pious and
impious beliefs about gods; cf. also Broadie (1999) 209–10; Mogyoródi (2002) 277–8;
Hedrick (2007) 286; Aikin (2014) 7.

149
 Xenophanes on Divine Disclosure and Mortal Inquiry
which spans theology, epistemology, cosmology and socio-moral
reflection, and which locates the mortal as a moral and discursive
agent within a divinely governed and disclosed world order.
The diviner transcends mortal epistemic limitations.
Xenophanes’ self-conscious intellectual superiority favours the
conclusion that, for him, the gulf between mortal belief and divine
knowledge is qualitatively unbridgeable but quantitatively fluid.
Our conjectural beliefs about the world around us and our place
within it are forever corrigible. Still, continual empirical inquiry,
informed by a recognition of the true nature of divine disclosure,
allows mortals to evaluate increasingly grounded conjectures
against an ever-growing body of evidence. God is epistemically
superior to us at least in part because he cognises all things directly
and synoptically (B25, B34; cf. B24). Mortal inquiries, then, bring
us ever closer (but never close) to divine knowledge.
In Xenophanes’ own prosaic example, the more sweet substances
we experience, the closer our conjectures approach a true judge-
ment which is informed by the direct cognition of all sweet
substances.
Philo carelessly assimilates Xenophanes to Parmenides and
Empedocles as mortals who claimed to be ‘divine men’ (diuini
uiri). Rather, says Philo, by devoting their lives to the contempla-
tion of nature and pious praise of the gods, these three philoso-
phers attained the height of mortal excellence, but nothing higher
(optimi quidem uiri, A26). Philo, I conclude, prescribes for
Xenophanes the very conception of his intellectual superiority
that Xenophanes himself in fact developed. Philo’s false ascription
to Xenophanes of a claim for divine status brings us to one last
feature of divination:
(iii) Such epistemic approximation to the divine is associated with a more general
approximation to the divine.

Xenophanes’ insistence on an irreducible dissimilarity between

mortal and divine (B23) implies a categorical rejection of ideas of
deification and god-like men, as exemplified in Teiresias’ substan-
tial survival in the afterlife or the seer Theoclymenus’ family-tree
(Chapter 3.1). In all four versions of a floating anecdote,
Xenophanes urges people who consult him about a question of
150
 3.5 A Clean Sweep?
cult practice to decide whether they believe that a liminal figure
(Leucothea, Osiris) is mortal or divine, and to regulate their ritual
accordingly: they should either mourn for or sacrifice to this figure,
but not both (A13). Whatever the historicity of this tradition, it
reflects something of the thinker with whom it became associated.
Xenophanes rejects the malleability which characterises the bound-
ary between mortal and divine in Greek religious thought.138
There is, I believe, some evidence that Xenophanes explicitly
reacted to this phenomenon. Greek tradition accorded extraordin-
ary longevity to Epimenides, an itinerant religious expert who, as
a divine man, performed mantic services for cities.139 Diogenes
remarks that Xenophanes mentioned the rumour that Epimenides
lived to be 154 years old (1.111 = B20). Elsewhere, Diogenes
writes that Xenophanes criticised Epimenides, and then immedi-
ately proceeds to speak of Xenophanes’ own, less fantastic long-
evity (καθάψασθαι δὲ καὶ Ἐπιμενίδου, μακροβιώτατός τε γέγονεν),
citing the philosopher’s own estimation of his age at 92 (9.18–19 =
B8). The order of Diogenes’ exposition suggests that Xenophanes’
remark about Epimenides’ reported longevity was incredulous,
and, indeed, may indicate that Xenophanes presented his own
careful estimation of his age (‘if indeed I know how to speak
truly concerning these things’, B8.4) as a reaction against the
hyperbolic traditions surrounding Epimenides. Empedocles
speaks of a sage who can recall ten and twenty lifetimes (DK31
B129). Porphyry, citing the fragment, identifies the anonymous
sage as Pythagoras (VP 30). Whether or not we accept this identi-
fication, it is plausible that Xenophanes’ derision of Pythagorean
metempsychosis (B7) reflects a critical reaction to the same con-
ception of sagehood, which Empedocles later adopts and devel-
ops, and which associated wisdom with the ability to transcend the
limits of a solitary, observed mortal life.140

138
See also Broadie (1999) 211; Warren (2013): 310–11.
139
For the label ἀνὴρ θεῖος, see Pl. Lg. 1.642d5. Epimenides’ purifications (ps.-Ar. Ath. 1;
Plut. Sol. 12; D.L. 1.110) fall squarely within traditional mantic expertise, see Flower
(2008) 27; Dillery (2005) 181–2; Burkert (1992) 42–3. Epimenides is ascribed prog-
nostications (D.L. 1.114–15), verse oracles (Paul Ep.Tit. 1.12) and the label ‘chresmo-
logue’ (Σ Luc. 25.6.1 Rabe).
140
Later tradition has Epimenides too recall his many incarnations (D.L. 1.114).
On Empedocles, see Ch. 6.2.

151
 Xenophanes on Divine Disclosure and Mortal Inquiry
Xenophanes had excellent epistemological reasons to retain the
basic and ubiquitous association of sagehood with longevity, as
exemplified most iconically by the old and wise Nestor and as
reflected in the formula ‘I was born earlier and I know more.’141
Xenophanes’ preoccupation with his longevity is reflected also in
the ancient testimonia.142 Unsurprisingly, then, Xenophanes
(a) comments on his own advanced old age (B8), (b) reflects on
the physical weakness of old men (B9, B1.17–18) and (c) criticises
the conventional preference of physical strength to his wisdom
(B2.14). Most importantly, Xenophanes embeds his representation
of himself as an itinerant sage in a calculation of his longevity,
setting the number at 92, with 67 years of active inquiry. Both the
relatively sober number (in comparison, at any rate, with the
tradition about Epimenides to which he may be reacting), and
Xenophanes’ careful qualification of his estimation (‘if indeed
I know how to speak truly concerning these things’), are not
emphasised despite his claims for sagehood. Rather, Xenophanes
illustrates the disillusioned nature of his sagehood – a sagehood to
which he lays claim not least on the basis of his disillusioned
rethinking of divine disclosure and mortal epistemology.
As Lloyd has argued, the emergence of the polis was accompa-
nied by an explosion of diverse ‘wise men’ figures in the seventh
and sixth centuries, answering a growing need for ‘both what we
should call political, and religious and intellectual, leadership’.143
It is against this background that we should evaluate Xenophanes’
conception and representation of his own sagehood. The elegy
which contained his satirical derision of Pythagorean metempsy-
chosis began with the words: ‘Now I will come to yet another
account, and I will show the way’ (νῦν αὖτ’ ἄλλον ἔπειμι λόγον,
δείξω δὲ κέλευθον, Diogenes Laertius, 8.36 = B7). This magisterial
and, indeed, divinatory programmatic statement may be parodying

141
πρότερος γενόμην καὶ πλείονα οἶδα (vel sim.), Il. 21.440; Il. 13.355; Il. 19.219; also, Il.
3.108–10; cf. Haubold (2010) 16, n.18.
142
In addition to Diogenes Laertius (9.18–19), Phalerum in his On Old Age and Panaetius
both recorded a tradition that Xenophanes, like Anaxagoras, outlived his sons (apud
D.L. 9.20.9). Ps.-Lucian, perhaps miscalculating B8, puts his death at 91 (A6),
Censorinus at more than 100 (A7). Timon traces Xenophanes’ failure to achieve full-
fledged scepticism to his senility (πρεσβυγενὴς ἔτ’ ἐών, A35).
143
Lloyd (1979) 249; cf. Dillery (2005) 183.

152
 3.5 A Clean Sweep?
the rhetoric of the thinkers and doctrines subsequently criticised.144
But, however satirically, Xenophanes appropriates the authority of
those claimants for sagehood whom he lampoons and with whom he
must vie for the same social and intellectual territory.
The most familiar incarnation of divination for Xenophanes
would undoubtedly have been those itinerant diviners, ‘migrant
charismatic specialists’ as Burkert styles them, who travelled the
Greek world from South Italy to Asia Minor.145 Never simply
relieving academic curiosity about the future, they invariably
furnished counsel about any matter public and private.146
Moving from city to city, providing an autonomous authority
external to local civic institutions, they forever remained outsiders
for the communities they counselled.147 The influence of these
practitioners, representing ‘the intellectual elite of the time’,148
extended, as Plato later complained, from individual consultants to
‘entire households and cities’ (ὅλας οἰκίας καὶ πόλεις, Lg. 10.909b;
Rep. 2.364e). As Dillery puts it, ‘[t]he independent religious
expert . . . was always precisely that, independent . . . he seems
always positioned outside the political structure of the state,
essential to but also separate from the governance of the
polis’.149 Xenophanes spent a lifetime ‘tossing about’ his
thoughts, counsels and authority (φροντίς) throughout Greece
(B8). His journeys, he emphasises, take him ‘from city to city’
(B45) and his wisdom benefits the community of the polis as
a whole (πόλις . . . πόλει . . . πόλιος, B2.19–22). Xenophanes was
an itinerant sage counselling cities on anything from the use of
perfume (B3) or wine (B5), to the nature of the gods and divine

144
φράζεο δ᾽ ἀνέρα μάντιν ὑφηγητῆρα κελεύθου, Anth. Graec. 14.114.7 Beckby; for
diviners or the divine showing the ‘way’ see also Od. 10.538–9; Od. 4.389–90; PW
218.5; PW 374.9; Ennius apud Cic. Div. 1.58.132; cf. also Op. 648; Parm. DK28 B2.
1–2; Anth. Graec. 9.80 Beckby; PW 380.3; PW 517.2; Od. 5.237–41.
145
Burkert (1992) 42; Flower (2008) 29–37.
146
Halliday (1913) 42; Parker (2000) 77; Flower (2008) 75–6. Greek terminology for
divination exemplifies its purported usefulness: χράω, χρῆσις, χρῆμα (Emp. DK31
B115.1), χρηστήριος, χρήστης κτλ. Cf. X. Mem. 4.3.12 (τὰ συμφέροντα . . . ᾗ ἂν
ἄριστα γίγνοιτο); h.Hom. 3.287–93 (νημερτέα βουλήν . . . χρέων).
147
Eumaeus includes the diviner (mantis) in a list of ‘public workers’ whom one would
invite to one’s house though a complete stranger, Od. 17.382–5; cf. Od. 1.415–16; Od.
9.508–10; Aesch. Ch. 32. See further Dillery (2005) esp. 176–8, 223; Vernant
(1991b) 305.
148
Burkert (1992) 42. 149 Dillery (2005) 184, 224–5; Flower (2008) 58–9.

153
 Xenophanes on Divine Disclosure and Mortal Inquiry
disclosure, attempting to engender prosperity and lawfulness
(B2.19–22).150 Less immediate factors too made a social and
intellectual confrontation with divination and its models of sage-
hood inevitable. The various types of divination practised in
Greece were rather recent Eastern importations.151 Xenophanes
was haunted by the fall of his Ionian city to ‘the Mede’ (B22), and
associated it with the corrupting cultural influence of the East (B3
with Theopompus apud Athenaeus 12.526c). By appropriating
and radically transforming the discourse of divination,
Xenophanes instantiates a broad phenomenon in Greek interac-
tions with Eastern paradigms, practices, crafts and ideas.152
Furthermore, Greek tradition, consistently and as early as the
sixth century, located the archetypal mantic contest between
Mopsus and Teiresias at Xenophanes’ native Colophon.153
Indeed, local traditions considered Mopsus a founding figure.154
We can only wonder whether and how Xenophanes treated these
traditions in a poem about the founding of Colophon (Diogenes
Laertius, 9.20 = A1). We may conclude, however, that the socio-
political context of his philosophical activity all but constrained
this subversive iconoclast to form his conception of mortal epis-
temology in general, and of his own sagehood in particular, in
response to such divine men and their divination. Whatever else,
Xenophanes’ response was not a clean sweep from the very
foundation.

150
Engendering lawfulness (eunomiê): later Greek tradition has Delphic Apollo impart
this skill to Lycurgus, PW 216.
151
Burkert (1992) 41–87 (dating the ‘orientalising period’ to 750–650); Flower (2008)
24–5 (between the eighth and sixth centuries); cf. West (1997) 46–51.
152
Burkert (1992) passim; R. Osborne (1997) 11–16.
153
See ps.-Proclus’ argumentum to the Nostoi, Chr. 277ff; cf. Hes. fr. 278 M-W (note
Strabo: de Colophone disserens); Lycoph. 424–30 with Σ Lycoph. 424–7 Scheer
(ἐρίζοντα περὶ τῆς μαντικῆς); Callisthenes (or Callinos) apud Strab. 14.4.3; Apollod.
Ep. 6.2–4 (περὶ μαντικῆς ἤρισε); Σ Dion. Perieg. 850 Müller (μαντικῇ νικηθείς).
The tradition and the evidence are discussed by Bremmer (2006) 140, with n.37.
154
Paus. 7.3.2.

154
 I N T R O D U C T I O N TO T H E CH A P T E R S O N
PA R M E N ID E S

Our evidence for Parmenides all derives from a single Hexametric

poem. As we noted in Chapter 1.1, Parmenides’ poem opened with
a proem, which narrated the chariot ride of a youth – a kouros –
towards a goddess, who acts as the sole speaker for the remainder
of the poem. After welcoming the kouros, the goddess describes
the programme of study to follow (DK28 B1.28b-30):
χρεὼ δέ σε πάντα πυθέσθαι 28
ἠμὲν ἀληθείης εὐκυκλέος ἀτρεμὲς ἦτορ
ἠδὲ βροτῶν δόξας, ταῖς οὐκ ἔνι πίστις ἀληθής. 30

And it is right for you to learn all things,

Both the unshaken heart of well-rounded reality,1
And the beliefs of mortals, in which there is no real trust.
These two aspects of the kouros’ programme of study correspond to
two rather distinct accounts that the goddess proceeds to issue, and
to which scholars typically refer as the two parts of Parmenides’
poem. In the first part, the goddess expounds the nature of ‘what-is’
(to eon). We learn that what-is is ungenerated, imperishable, chan-
geless, motionless, unified and homogeneous. The goddess glosses
the subject matter of this component of the kouros’ programme of
study as ‘the unshaken heart of reality (alêtheiês)’. We will follow
the practice of referring to this part of the poem as ‘Alêtheia’
(B2-B8.51a). After concluding her account of what-is as ungener-
ated, changeless, homogeneous, etc., the goddess – surprisingly –
proceeds to develop a comprehensive and systematic cosmology,
based on the two opposite principles ‘Light’ and ‘Night’. Here the
goddess expounded the origin and nature of the ordered world, of
celestial bodies, of gods and humans. This second part of the poem
corresponds to the ‘beliefs (doxas) of mortals’ in the kouros’

1
Alternatively: ‘trustworthy’ or ‘persuasive’ reality (εὐπειθέος). Palmer (2009) 378–80
convincingly defends the reading εὐκυκλέος.

155
 Introduction to the Chapters on Parmenides
programme of study. We will follow the standard practice of refer-
ring to it as ‘Doxa’ (B8.51b-B19). By a ‘Doxastic thing’, I will refer
to any item which is characterised by any of the attributes denied to
what-is in B8, such as mobility or having-come-into-being, and
which theories advanced in Doxa take as their subject matter
(e.g. stars, people, embryos). Why this second part is there, and
what its status is, are two of the central and most difficult questions
with which Parmenides’ poem confronts us.
In the first part of the poem, then, Parmenides aligns what-is
with alêtheiê, and this term is understood as something like
ultimate or fundamental reality. The goddess, as we saw, describes
the subject matter of the first part as ‘the unshaken heart of well-
rounded alêtheiê’ (B1.29). This gloss both foreshadows the align-
ment of what-is with alêtheiê and already indicates that
Parmenides employs the term to signify, not so much a logical
property of thought or speech (‘truth’), but a core, ultimate or
fundamental reality: what-is is the unshaken heart of reality
(alêtheiê). The goddess’s words in B8.50–1 confirm both points:
ἐν τῷ σοι παύω πιστὸν λόγον ἠδὲ νόημα
ἀμφὶς ἀληθείης.
Here I conclude for you the trustworthy account and understanding
Concerning true reality.
The goddess here uses alêtheiê to refer to the subject matter of
her immediately preceding account and contemplation, i.e. to
what-is and the properties deduced for it in B8. She does not
use this terminology to qualify the account and contemplation
themselves as ‘true’ (although we can allow, as secondary con-
notations of this terminology, ‘truth’ and ‘what is true’ insofar as
these typify true accounts of the true nature of ultimate reality).2
In another indication that he is aligning or even identifying what-
is and alêtheiê, Parmenides later describes also what-is as
‘unshaken’ (ἀτρεμές, B8.4), just as he had earlier described the
heart of alêtheiê (B1.29).

2
Note further B1.30; B8.28 (πίστις ἀληθής: ‘true conviction’, i.e. real or genuine convic-
tion); again, B8.17–18 (‘for it is not a real (ἀληθής) way’). On the semantics of ἀληθείη
and ἀληθής in Parmenides, I follow Palmer (2009) 89–93; Coxon (2009) 282–3; Cole
(1983) 25–6.

156
 Introduction to the Chapters on Parmenides
Since Parmenides himself advertises that Alêtheia provides
knowledge of ‘the heart of reality’, and given also the revolu-
tionary nature and subsequent influence of the ontology of
Alêtheia, it is unsurprising that commentators have in the past
tended to privilege Alêtheia while marginalising the poem’s
other aspects. Much recent work on Parmenides, however, has
sought to correct this traditional attitude and to insist on the
centrality and significance of Doxa (Chapter 4.1). And indeed,
we must recognise the limitations of the tendency to privilege
Alêtheia at the expense of the poem’s other aspects for one who
wonders how Parmenides might have recommended his philoso-
phy to his audience as viable. The Epicurean polemicist Colotes
contended that Parmenides’ ontology left no room for our own
lives to take place. Although Plutarch rightly rebukes Colotes for
ignoring Parmenides’ scientific inquiries (Chapter 4.1), it is
difficult to fault Colotes for complaining that Parmenides did
not himself clarify just how Alêtheia’s account of true reality as
homogeneous and unchanging coheres with the heterogeneous
multiplicity which surrounds us and which we ourselves com-
prise (apud Plutarch, Mor. 1113–14). We must neither ignore
Alêtheia’s ontology when considering other aspects of
Parmenides’ thought nor wholly subject them to it. We should
neither marginalise Doxa nor suppress the fact that Parmenides’
abiding preoccupation with Doxastic things – whose qualities are
incompatible with those which are deduced for ‘what-is’ (to eon)
or ‘reality’ (alêtheiê) – and with beliefs in which, in some sense,
there is no ‘real trust’, is a problem which requires an explana-
tion. Parmenides does not himself affirm, deny or explicate in an
express or direct way the consistency between Alêtheia’s con-
tention that what-is is homogeneous and unchanging and the
presence of its human audience, which comprises differentiated,
mortal beings. Even if Alêtheia does not, pace Colotes, exclude
the presence of its audience, and even if Parmenides never
thought otherwise, it seems plausible that the viability of
Parmenides’ philosophy for such a heterogeneous audience was
significantly anchored, as the proem suggests, in the epistemic
transformation promised for the mortal subject as related to an
encounter with the divine, and not predominantly, let alone
157
 Introduction to the Chapters on Parmenides
exclusively, in the description itself of what-is as homogeneous
and unchanging.3
The bias that philosophers before Socrates tended to ignore the
philosophising subject informed the development of the very cate-
gory ‘pre-Socratic’. In 1996, Anthony Long argued that work on the
early Greek philosophers ‘has yet to take proper measure of their
attention to philosophical methodology, second-order inquiry,
mind, and the relation of the knower to the known’.4 In particular,
Long highlights three features of Peripatetic doxography which still
exert a strong influence over modern scholarship: 1. the suppression
of statements about the divine; 2. the assimilation of thinking or
mind to sense-perception (aisthêsis); 3. the assumption that early
Greek philosophers marginalised second-order, methodological
questions and construed their project as describing a world external
to themselves which did not include the observer in the materials
investigated.5 We still have a way to go in redressing the blind spots
which Long had diagnosed. The following study of Parmenides will
address throughout all three points. It will prove important, how-
ever, to maintain a clear distinction between different (though not,
of course, unrelated) lines of inquiry and interpretation. When we
explore below the epistemological and religious aspects of
Parmenides’ poem, and attempt to shed some new light through
this investigation on the poem’s structure and appeal, we will not
expect thereby to have disposed of the ontological paradox rightly
highlighted by Colotes.
In a way, the question ‘how can mortals cognise what-is?’ is both
inherently paradoxical (at least prima facie) and textually justified.
As we shall see, the account in Alêtheia relies on the distinction – or
‘krisis’ – between ‘is’ and ‘is not’. It is through a very different kind
of krisis that the account in Doxa is structured, one between Light
and Night as two opposite elements. Alêtheia’s concept of what-is

3
In regarding Parmenides’ text as performative and transformative I agree with Kingsley
(1999) e.g. 122, 144, 163; Gemelli Marciano (2008) and (2013); Robbiano (2006), and
Ranzato (2015); cf. Long (1999) 13; contrast Laks (2003) 29 (‘une connaissance pure-
ment théorique’). Cf. more generally Hadot (1995) 64: ‘the work [of the ancient
philosopher] . . . is written not so much to inform the reader of a doctrinal content but
to form him.’
4
Long (1996) 127–8; followed by Robbiano (2006) 17–18; cf. Curd (1998) 35.
5
Long (1996) 128–9.

158
 Introduction to the Chapters on Parmenides
cannot be conveyed through the contrast that underpins Doxa.
The contrast between Light and Night, each of which both is (itself)
and is not (its counterpart), may account for stars and embryos, but
precisely cannot account for the homogeneous what-is, which
strictly is and in no way is not. Equally, Alêtheia’s structuring
contrast cannot express items like ‘mortals’. Alêtheia, by privileging
exclusively the krisis between ‘is’ and ‘is not’ as the contrast that
should inform our understanding of ultimate reality (B8.15–16a),
while retaining ‘is’ and discarding ‘is not’, points us towards
a demarcation that can convey only the homogeneous, unchanging
and immobile what-is. Mortals are generated, perishable, mobile
and heterogeneous: they are, essentially, Doxastic things. The ques-
tion ‘how can mortals cognise what-is?’ thus straddles the concep-
tual frameworks that underpin the two parts of the poem. It is itself
inexpressible both in terms of the structuring contrast of Alêtheia
(which could not by itself articulate the idea of ‘mortals’) and within
that of Doxa (which could not articulate ‘what-is’).
At the same time, it is a question that Parmenides himself raises.
The programmatic description of the ensuing poem with which we
started (B1.28b-30) takes the form of a conversation between
a goddess and, emphatically, a mortal. We could not articulate the
notion of a mortal, or the distinction between mortals and a goddess,
in terms of Alêtheia’s structuring contrast between ‘is’ and ‘is not’.
But Parmenides’ narrative framework makes it very clear that
Doxastic opinions are specifically those of mortals (βροτῶν δόξας,
B1.30). The goddess, conversely, possesses knowledge of Alêtheia.
And yet, the mortal agent, to whom the goddess speaks, must also
attain this knowledge (B1.29). In a sense, Parmenides could hardly
avoid asking how a mortal could cognise what-is. He could not have
formulated his second-order, programmatic reflections from within
the conceptual framework of Alêtheia’s contrast between ‘is’ and ‘is
not’. Within this conceptual framework, which is capable only of
articulating the homogeneous what-is and (perhaps) its self-
knowledge, we can no longer discuss the efforts of mortal agents,
like Parmenides and ourselves, to come to know this thing.6

6
The echoes (if not perfect continuities) in imagery between Doxa and the proem, convin-
cingly analysed by Laks (2003), underscore further that the introductory encounter

159
 Introduction to the Chapters on Parmenides
Unless we maintain either that there is not even a prima facie
tension between Alêtheia’s account of ultimate reality and the
manifest presence of Doxastic things or that Doxastic things can
be coherently explained away as ‘illusions’ (I will reject both of
those positions), we must recognise that Alêtheia’s ontology does
confront us with a puzzle concerning the precise ontological status
of Doxastic things. This does not mean, however, that any inquiry
into any aspect of the relation between the two parts of Parmenides’
poem will have to begin from this puzzle, frame its interpretation
around it or rely on any one particular resolution of it.
We may distinguish three different questions concerning the
two parts of Parmenides’ poem. First, an ‘aetiological question’:
Why did Parmenides write and include Doxa? Why is the second
part there? Second, an ‘epistemological question’: what are the
ways in which the mortal agent, to whom the goddess speaks, can
think and what are the ways in which he must think? To clarify,
Alêtheia and Doxa discuss different kinds of objects, what-is and
Doxastic things (humans, stars, etc.) respectively. Since
Parmenides included both parts, he must consider thoughts about
both kinds of objects ultimately possible for the mortal agent. But
is thought concerning any kind of object unavoidably necessary
for the mortal? Does the mortal’s ability to cognise a different kind
of object then become a problem which requires an explanation?
Third, an ‘ontological question’: given the doctrine of Alêtheia,
what precisely is the status of Doxastic things? What is the nature
of the relation between what-is and Doxastic things?
Those commentators who address the aetiological question gen-
erally consider one’s view on it a consequence of one’s view on the
ontological question (Chapter 4.1). In the following chapters, we
will explore a different route with a different focus, addressing first
and foremost the aetiological and epistemological questions in
relation to one another. We will consider as interrelated matters
Parmenides’ impetus and rationale for developing and including

between the goddess and the kouros takes place without the framework of Alêtheia.
In particular, by locating the journey where the recurrent succession of Day and Night is
regulated (B1.11–14), Parmenides recalls the dualist, cosmological conception of Light
and Night as the two equal constitutive elements (B9.3–4). See Laks (2003) 17–18; cf.
Mourelatos (2008a) 13, 40; Coxon (2009) 15.

160
 Introduction to the Chapters on Parmenides
Doxa, his conception of the mortal epistemic agent in relation both
to the investigations in Doxa and to those in Alêtheia, and the role of
the relation between mortal and divine in his poem. In this way, we
will pursue a complementary but different perspective on the pro-
blem of the relation between the two parts of Parmenides’ poem.7
This is not to suggest, of course, that considerations of Alêtheia’s
account of what-is – or ‘the heart of reality’ – should or could be
ignored when pursuing either the epistemological or the aetiological
questions. My suggestion will be, rather, that we can usefully
elucidate and investigate those questions in relation to one another,
while postponing a focused discussion of what I labelled the ‘onto-
logical question’ until a later stage in the argument, and while
insisting only on certain limited but important points in relation to
this latter question and otherwise retaining a, to some degree, open-
ended and flexible attitude towards it.
In Chapter 4, we will see that, according to Parmenides, the type
of thinking which underpins Doxa is an ineluctable and even
appropriate aspect of mortal life. This conclusion will best posi-
tion us to answer the aetiological question and explain why
Parmenides wrote and included Doxa. In Chapter 5, however, we
will find that the mortal agent is nonetheless capable of sustaining
the qualitatively different thinking of Alêtheia by momentarily
coming to think with – or as – his divine (fiery, aethereal) soul.8

7
I will disagree, then, with the view of Bryan (2012) 100–4 that Parmenides’ epistemology
is object-centred and not subject-centred. On my account, preoccupations with both the
subjects and the objects of knowledge and belief play important roles in his epistemology.
8
My conclusions in Ch. 5 concerning the conception and function of soul in Parmenides
run counter to the claim, which one sometimes encounters (e.g. Sassi (2016) 453), that
notions of the soul as the principle not only of life but also of knowledge and under-
standing were only finally formed in the second half of the fifth century. In fact, however,
we do not need Parmenides to establish this historical point. His rough contemporary
Heraclitus describes the ideal soul as ‘wisest’ (DK22 B118; see further Schofield (1991);
Betegh (2007); (2009), and cf. Sassi herself (2016) 453, n.5). Indeed, this idea did not
emerge ex nihilo. The conception of soul as the locus of emotional and cognitive as well
as vital functions is arguably suggested by the still earlier Pythagorean notion that the
soul preserves across successive incarnations a person’s essential identity (ψυχή,
Xenoph. DK21 B7), and, in some special cases, their memories (Emped. DK31 B129,
with Burkert (1972) 138–9 for further sources). In fact, the idea that, after death, the soul
(ψυχή) retains some consciousness, memories and even an awareness of events in the
world of the living is evident already in Homer (Homeric assertions about the mindless
and powerless dead notwithstanding), e.g. Il. 23.64–107 (cf. Il. 24.592–5); Od. 11.
543–64, with Edmonds (2013) 252–67.

161
 Introduction to the Chapters on Parmenides
This will explain how Parmenides was also able to sustain the non-
mortal, divine thought, which underpins Alêtheia, and to come to
understand what-is. In Chapter 5.5, we will return, in the light of
those foregoing discussions, to the ‘ontological question’.
It is Parmenides’ goddess who, through her disclosure and
guidance, enables the mortal kouros to come to think with or as
his divine soul, and to sustain a higher-than-human, divine
thought. In 1945, Louis Gernet offered an important insight,
which has often been obscured in later scholarship: a broad spec-
trum extends between taking Parmenides to offer the ‘literal’
report of an actual chariot ride to a goddess and reducing his
account to ‘mere imagery’, divested of any reference to an encoun-
ter with the divine.9 What follows will aim to elucidate the sense in
which Parmenides’ postulation of an epistemically significant
interaction10 with a divine power (but not thereby an actual chariot
ride) is integral to his understanding of his acquisition of knowl-
edge and required by his epistemology. On the interpretation
defended below, argumentative reasoning and divine disclosure
play complementary and equally indispensable roles in
Parmenides’ thought.11

9
Gernet (1981; orig. 1945) 354.
10
For this terminology, see the introduction to this book.
11
Esp. Ch. 5.4. On the primary scholarly responses to Parmenides’ goddess, see Ch. 1.1.

162
 4

W H Y D I D PA R M E N I D E S W R I T E D O X A?

4.1 Approaches to the Aetiological Question

Those who addressed the aetiological question (why is Doxa there?)
have generally considered one’s view of it a consequence of one’s
position on the ontological question (what is the status of Doxastic
things?). In the past, scholars tended to relegate Doxastic things to
the status of mere appearances or illusions and, consequently,
typically minimised the significance of Parmenides’ cosmology,
most often reducing Doxa to a dialectical exercise. Owen (1960),
for example, had influentially explained Doxa as a ‘wholly dialec-
tical’ device, which aimed only to analyse and caution against the
flawed presuppositions of all cosmological theories.1 Doxa thus
became merely an ‘exemplar . . . of all erroneous systems’2 or ‘a
case-study in self-deception, indecisiveness and confusion’.3 Many
commentators maintain that the goddess develops and exposes as
false the best or most plausible cosmology in order to equip
Parmenides with an exemplar against which all other cosmologies
are, a fortiori, flawed. Doxa thus inoculates Parmenides against the
temptations of cosmology as such.4
On the assumption, which I share, that the goddess represents
Doxa as the best possible account of Doxastic things, she indeed
implies that even the best cosmology could never constitute an
1
Owen (1960) 89; cf. Tarán (1965) 227–8, 267; Furth (1974) 249; Cosgrove (2014) 15–18,
25–6.
2
Long (1975) 83.
3
Mourelatos (2008a) 260, reflecting his views in 1970. Mourelatos (2008a) xxxvii–xlii
criticises his Route (1970) for being ‘too dismissive of the scientific content’ of Doxa.
4
Long (1975) 83, 96–7; de Rijk (1983) 46–7; Gallop (1984) 23, Barnes (1982) 157;
Inwood (2001) 25; Trépanier (2004) 131–7; Warren (2007) 100–1; Miller (2011) 56;
Bryan (2012) 110–13; Wedin (2014) 63, n.89. Gregory (2014) speculates that the
dialectical and inoculative (23, 26, 45) point of Doxa is rather that there can be no
sufficient reason to prefer one complete cosmology over another.

163
 Why Did Parmenides Write Doxa?
account of the unshaken heart of ultimate reality. Nonetheless,
the scope and nature of Parmenides’ cosmological investigations
undermine these dialectical responses to the aetiological ques-
tion. The goddess had concluded in Alêtheia her critical demon-
strations that processes like coming-to-be and change do not
typify what-is. Both direct and indirect evidence indicates that
what followed in Doxa was an extended and detailed exposition,
thoroughly positive in tone, of diverse scientific theories, span-
ning, among other things, universal cosmology (DK28 B9, B12;
A37), cosmogony (B10–11), astronomy (B10–11; B14–15;
A40a), geography (A44a; B15a), theogony (B13), anthropogony
(Diogenes Laertius, 9.22, A53), embryology (B18; A53–4) and
human physiology and cognition (A46 = B16, A46a-b, A52).
Parmenides’ very general description of the subject matter of
Doxa as ‘the beliefs of mortals’ (B1.30, B8.51) bespeaks the
broad scope of the cosmology. Indeed, the goddess’s programme
of study for the youth encompasses ‘all things’ (B1.28b).
An indication of purported exhaustiveness also seems implicit
in her promise to Parmenides that she will issue to him such an
account of the ordering of things (diakosmos) that, henceforth, no
judgement of mortals will ever surpass him (B8.60–1).
The conclusion to Doxa (B19, quoted in Chapter 4.4) may give
the impression that the cosmology has now (directly or indir-
ectly) accounted for all that humans have marked out with a name
and which involves the perceptible processes of becoming, per-
sisting for a time and then perishing.5 Plutarch instructively
rebukes Colotes for implausibly marginalising the sheer scope
of Parmenides’ scientific inquiries and the thoroughness with
which he pursued them (Mor. 1114b-c):
For he [sc. Parmenides] also said many things about the earth and about the heavens
and about the sun and about the moon and about the stars, and he related the
generation of humans; and, as an ancient man working in natural philosophy and
putting together his own treatise rather than plundering another’s, he left nothing
unspoken of all the principle topics (οὐδὲν ἄρρητον . . . τῶν κυρίων παρῆκεν).

5
Coxon (2009) 387 takes τάδε (B19.1) to refer to all natural substances (except the two
basic elements) but adds that ἔφυ and τραφέντα (B19.1–2) suggest a primary concern with
living things. This may be right, at least of this concluding juncture in Doxa, although
Parmenides uses ἔφυ also of the heavens (B10.5–6).

164
 4.1 Approaches to the Aetiological Question
Recent work on Parmenides has, moreover, given us a better appre-
ciation of the involved, sophisticated, innovative and insightful nature
of his scientific inquiries, most notably (but not only) his astronomy.6
It strains credulity to suppose that, in developing such systematic,
ambitious and specific cosmological accounts, Parmenides’ sole or
even primary motivation was simply to construct ‘an exemplar of all
erroneous systems’ or merely to exemplify an entirely general meta-
physical confusion.7 Nor is it clear what the dialectical or inoculative
benefit consists in. In Doxa, the goddess presented particular, con-
crete theories concerning, say, the source of the moon’s light
(B14–15; A42) or the embryological processes by which children
become similar to one parent or the other (A54). By elaborating these
theories, and adding thereby nothing to Alêtheia’s remarks about such
processes as motion and coming-to-be, the goddess perhaps deters
Parmenides from rival astronomical and embryological accounts, but
she does not thereby inoculate him against astronomical and embry-
ological accounts as such, nor, again, does she thereby display an
entirely general critique of cosmology as such.8
Other commentators have sought to identify a more strictly
‘pragmatic’ motivation for Doxa, seeing the cosmology as an
attempt to produce a somehow ‘useful’ account of our heteroge-
neous surroundings. It has been suggested that Parmenides’ moti-
vation for devising Doxa’s theories is of the same sort as the
motivation for designing a chariot axle that did not screech or
for growing superior olives.9 The pragmatic interpretation is,
6
Graham (2006) 179–82; (2013a) 85–108; Mourelatos (2011); Cerri (2011); Casertano
(2011) 21–49; Journée (2012); Mansfeld (2015); cf. Kraus (2013) 489–96. Kurfess (2016)
31–7 underplays both the strength and the diversity of our evidence for a wide-ranging
cosmological ambition on Parmenides’ part. That readers like Colotes could still interpret
Parmenides as denying wholesale the world of change and differentiation, and criticise him
for doing so, is hardly surprising in the light of his rejection of such things with respect to
‘what-is’ in Alêtheia and of his own pejorative remarks (discussed anon) about Doxa.
7
For the observation that Doxa’s scope itself renders dialectical readings implausible, cf.
Clark (1969) 24; Nehamas (2002) 57, n.46. Although Curd (1998) makes this point
(12–13), she is vulnerable to her own objection. While she identifies the error of Doxa as
its initial postulation of enantiomorphic opposites (110), she denies that Parmenides
thought such an error necessary for cosmology (6, 100). Why, then, did Parmenides not
produce a cosmology free of this error? Cf. Nehamas (2002) 61.
8
Contra Cosgrove (2014) 17–18, 21.
9
Cosgrove (2014) 20–1, 25–6: Doxa’s theories are ‘worth their pragmatic value or utility
to mortals, but . . . that is all’. For Cosgrove, Doxa’s accounts have no cosmological,
scientific or philosophical status whatsoever.

165
 Why Did Parmenides Write Doxa?
however, implausible on this narrow construal of utility. Doxa is
hardly limited to resolving or addressing everyday or concrete
difficulties. Parmenides was evidently motivated to develop in
Doxa numerous and various speculative theories, without feeling
the need to specify the pragmatic value of those theories or to
identify practical applications for them, even when it is far from
clear that there could be any such practical applications or what
they might be. In B10, moreover, the goddess herself presents
Doxa’s astronomy as the pursuit of a kind of knowledge:
εἴσῃ δ’ αἰθερίαν τε φύσιν τά τ’ ἐν αἰθέρι πάντα 1
σήματα καὶ καθαρᾶς εὐαγέος ἠελίοιο
λαμπάδος ἔργ’ ἀίδηλα καὶ ὁππόθεν ἐξεγένοντο,
ἔργα τε κύκλωπος πεύσῃ περίφοιτα σελήνης
καὶ φύσιν, εἰδήσεις δὲ καὶ οὐρανὸν ἀμφὶς ἔχοντα 5
ἔνθεν ἔφυ τε καὶ ὥς μιν ἄγουσ’ ἐπέδησεν Ἀνάγκη
πείρατ’ ἔχειν ἄστρων. 7

You shall know the nature of aether and all the signs in the aether
And, of the pure torch of the shining sun,
Its unseen10 works and whence they came to be,
And you shall find out about the wandering works of the round-eyed moon
And her nature, and you shall know also the surrounding heaven
Whence it grew and how Necessity led it and fettered it in shackles
To hold the limits of the stars.11
On a broader understanding of utility, however, the pragmatic
interpretation would be in itself insufficient and incomplete. If it
is not just a matter of utility narrowly construed, then why should
we desire or benefit from even the best possible version of a type of
inquiry which Parmenides frames in unmistakably pejorative
terms as, in some sense, without ‘real trust’ (B1.30) and ‘decep-
tive’ (B8.52)? Why are we not better off simply abandoning
Doxa’s cosmological speculations about the origin and nature of
the heterogeneous surroundings in which our mortal lives take
place for ontological contemplation of the nature of what-is, even
after and in the light of Alêtheia?12 Biographical explanations,

10
Or: ‘unseen-making’, with Mourelatos (2011) 170–4.
11
Contrast Cosgrove (2014) 20–1, 24: Doxa is not a cognitive enterprise.
12
For another expression of the pragmatic approach, see Minar (1949) 44, 49–50, 55; cf.
similar suggestions in Inwood (2001) 25, n.50; Coxon (2009) 352; Hussey (1990) 30;
Robbiano (2006) 210; also Detienne (1996) 133.

166
 4.1 Approaches to the Aetiological Question
such as Nietzsche’s ascription of Alêtheia and Doxa to different
stages in Parmenides’ life, are fanciful and fail to explain the
inclusion of both parts in the ‘final edition’ (B1.28b-30).13
Doxa has enjoyed a resurgence in recent years. There is a growing
interest in the substance of Doxa’s theories and a greater awareness of
the important position that Parmenides’ cosmology occupied in his
poem. An increasingly dominant view maintains that Doxa’s theories
represent Parmenides’ own views of Doxastic things and are intended
to describe such things correctly.14 I will follow and defend this
position below (in particular, Chapter 4.4). It would be
a fundamental mistake, however, to assume that, once we take it
that Parmenides allows that Doxa accounts correctly for Doxastic
things, or once we respond to the ontological question with our
particular preferred way of explicating the precise status of
Doxastic things and the manner in which they relate to what-is,
then the need to explain why Doxa is there in the first place vanishes.
Crucially, a resolution to the ontological question will not of itself
constitute a resolution to the aetiological question. The reason for this
is that Parmenides himself emphatically frames Doxa in a pejorative
way. Even as the goddess decrees that Parmenides must also learn the
beliefs of mortals (βροτῶν δόξας), she indicates that – in some
sense – those beliefs lack ‘real trust’ (πίστις ἀληθής, B1.30). When
the goddess introduces her cosmology, she indicates that – in some
sense – her account, or the ‘order’ of her words, is ‘deceptive’
(ἀπατηλόν, B8.51b-2). Furthermore, Parmenides, we noted, aligns
what-is, as the subject matter of the first part of the poem, with the
‘unshaken heart of well-rounded reality’ (alêtheiê).

13
Nietzsche (1974) 114–15; for a similar criticism, see Cosgrove (2014) 10–11.
Mackenzie (1982) argues that the two parts jointly form a dialectical dilemma:
Alêtheia preserves reason but rules out the existence of reasoners, whereas Doxa
preserves them but irrationally (7–9). It is difficult to square this view of Parmenides
as a self-consciously ‘working philosopher’ (9) with the clear axiological asymmetry
between the two parts and with Parmenides’ conclusive dogmatism (πάντα, B1.28).
Furthermore, if Parmenides is aware that he is both premising and refuting the existence
of multiple reasoners in Alêtheia (1–3, 8), should he still consider its argumentation
rationally sound?
14
See (with reflections on recent developments) Palmer (2009) 162; Bredlow (2011)
221–2; Mansfeld (2015). Nearly every contribution in Cordero (2011a) adopts this
position. For some earlier expressions of it, see Nehamas (2002) 56–60; Coxon
(2009) 342; Cornford (1933) 97–8, 110–11. See further Kraus (2013) 481–2 for
a survey of modern attitudes towards Doxa.

167
 Why Did Parmenides Write Doxa?
Now, it is by no means obvious what the deceptiveness of the
goddess’s account amounts to. Significantly, she never describes the
account as false. As we shall see in Chapter 4.4, Doxa’s theories
are not deceptive in the particular ways in which they account for
the origins, natures and behaviours of their particular objects of
inquiry (e.g. in the way in which Doxa’s embryology explains
embryos). Indeed, we already saw how, in an important set of
programmatic remarks, the goddess promises the kouros knowledge
of celestial bodies and processes (B10). Doxa’s deceptiveness, I will
argue, relates rather to the way in which the ignorant mortals, who
are not privy to the account of what-is in Alêtheia, are prone to
mistake the cosmology for an account that gets at and discloses, not
merely the natures of Doxastic things and processes, but the nature
of ultimate reality. The kouros, who learns Doxa after and in the
light of Alêtheia, is therefore in a privileged position to recognise
Doxa’s deceptiveness and to avoid this mistake.
In sum, then, even though Parmenides never pronounces Doxa’s
theories false of Doxastic things, he nonetheless emphatically rele-
gates Doxa’s investigations to a decidedly inferior status. Doxa’s
theories precisely do not and could not capture what-is: ‘the unsha-
ken heart of reality’, the core, ultimate and fundamental nature of
reality, to which the first part of the poem is devoted. They lack ‘real
trust’. Our text itself, then, pointedly raises the aetiological ques-
tion. If knowledge of the unshaken heart of reality is available to
Parmenides, then why is it right for him and incumbent upon him
(χρεὼ δέ σε, B1.28) to learn also these pejoratively framed mortal
beliefs (B1.29–30)? Even if Doxa is the best possible account of
Doxastic things, why should Parmenides pursue all these systematic
and painstaking investigations into the nature of Doxastic things –
investigations which he himself so emphatically frames as inferior –
after attaining the knowledge of the heart of reality in Alêtheia?
As Sedley writes, ‘Why join in the game?’15
15
Sedley (1999a) 123. Even if, then, we read ὡς at B8.61 as final rather than causal, this
will hardly resolve the aetiological question, since we would still ask ourselves why
Parmenides should wish to join this mortal race in the first place; see further Ch. 4.5.
We could only explain away the aetiological question by trying to deny the (undeniable)
point that the goddess’s pejorative comments do in fact refer to Parmenides’ cosmology.
Against attempts to do so in Thanassas (2006), (2011) 300–1, and Cordero (2011b), see
Tor (2015) 7–8, n.16.

168
 4.2 Parmenides’ Theory of Human Cognition
We will approach the aetiological question from a different
direction. In order to do so, we will begin by considering
Parmenides’ theory of human cognition.

4.2 Parmenides’ Theory of Human Cognition

The key texts for us in this section will be Parmenides B16 and its
doxographic context. We will see that Parmenides advanced
a particular theory of human cognition. The basic features of this
theory will be the following. Our bodies consist of fluctuating
mixtures of the two fundamental, opposite elements, Hot and
Cold (or ‘Light’ and ‘Night’, ‘Rare’ and ‘Dense’, etc.). By the
Hot in us we cognise Hot, and by the Cold in us we cognise Cold.
Our cognition is determined at each moment by the fluctuating
combinations of these elements of which we consist. What our
body cognises at any moment are its own or other instances of
these elements.
We owe our knowledge of B16 above all to Aristotle’s follower,
Theophrastus, who, in his On the Senses, cites the lines and offers
an invaluable commentary. Theophrastus begins his treatise with
the suggestion that, concerning sense-perception, previous philo-
sophers can be divided into two camps (Sens. 1–2). The first group
(Parmenides, Empedocles, Plato) subscribed to the principle that
perception is of like by like (τῷ ὁμοίῳ, e.g. hot by hot). The second
group (Anaxagoras, Heraclitus) maintained, by contrast, that
unlike perceives unlike (τῷ ἐναντίῳ). Here is the beginning of
Theophrastus’ exposition of the like-by-like camp, and his discus-
sion of Parmenides B16 in its entirety (Sens. 2.5–4.8 = A46):
περὶ ἑκάστης δὲ τῶν κατὰ μέρος οἱ μὲν ἄλλοι σχεδὸν ἀπολείπουσιν, Ἐμπεδοκλῆς δὲ
πειρᾶται καὶ ταύτας ἀνάγειν εἰς τὴν ὁμοιότητα. (3) Παρμενίδης μὲν γὰρ ὅλως οὐδὲν
ἀφώρικεν ἀλλὰ μόνον ὅτι δυοῖν ὄντοιν στοιχείοιν κατὰ τὸ ὑπερβάλλον ἐστὶν ἡ
γνῶσις. ἐὰν γὰρ ὑπεραίρῃ τὸ θερμὸν ἢ τὸ ψυχρόν, ἄλλην γίνεσθαι τὴν διάνοιαν·
βελτίω δὲ καὶ καθαρωτέραν τὴν διὰ τὸ θερμόν· οὐ μὴν ἀλλὰ καὶ ταύτην δεῖσθαί τινος
συμμετρίας·
[Parmenides B16: cited below]
(4) τὸ γὰρ αἰσθάνεσθαι καὶ τὸ φρονεῖν ὡς ταὐτὸ λέγει· διὸ καὶ τὴν μνήμην καὶ τὴν
λήθην ἀπὸ τούτων γίνεσθαι διὰ τῆς κράσεως· ἂν δ’ ἰσάζωσι τῇ μίξει, πότερον ἔσται
φρονεῖν ἢ οὔ, καὶ τίς ἡ διάθεσις, οὐδὲν ἔτι διώρικεν. ὅτι δὲ καὶ τῷ ἐναντίῳ καθ’ αὑτὸ
ποιεῖ τὴν αἴσθησιν, φανερὸν ἐν οἷς φησι τὸν νεκρὸν φωτὸς μὲν καὶ θερμοῦ καὶ φωνῆς
169
 Why Did Parmenides Write Doxa?
οὐκ αἰσθάνεσθαι διὰ τὴν ἔκλειψιν τοῦ πυρός, ψυχροῦ δὲ καὶ σιωπῆς καὶ τῶν
ἐναντίων αἰσθάνεσθαι. καὶ ὅλως δὲ πᾶν τὸ ὂν ἔχειν τινὰ γνῶσιν. οὕτω μὲν οὖν
αὐτὸς ἔοικεν ἀποτέμνεσθαι τῇ φάσει τὰ συμβαίνοντα δυσχερῆ διὰ τὴν ὑπόληψιν.

Concerning each of the individual [senses], the others more or less omit [discus-
sion], but Empedocles attempts to refer these too to likeness. (3) Parmenides
determined nothing, saying only that there are two elements and our knowledge
depends on that which exceeds. Thought varies according to whether the hot or
the cold predominates; but a better and purer thought is the one through the hot;
yet even such thought requires a certain fit.
[Parmenides B16]
(4) For he speaks of perceiving and cognising as the same thing. Consequently,
both remembering and forgetting also result from the mixture of these. But if there
should occur an exact equality in the mixture, he does not determine whether or not
there would be cognition or what the general state would be. But it is clear that he
attributes perception also to the opposite [sc. element] by itself from that passage in
which he says that a corpse, because of the departure of the fire, does not perceive
light, heat and sound, but does perceive cold and silence and the other opposites.
And, in general, everything that is has some knowledge. So, in this way, he seems to
cut off by mere assertion the unattractive consequences of his conception.

Throughout this passage, Theophrastus uses cognitive and perceptual

language interchangeably. Indeed, he seems to try and justify this
practice by asserting that Parmenides himself ‘speaks of perceiving
and cognising as the same thing’. We will discuss in Chapter 4.3 the
relation between the cognitive and perceptual language in B16 and in
Theophrastus’ commentary. For now, I will follow Theophrastus’
practice and use both sorts of terms interchangeably.
Who are ‘the others’ in the opening sentence (οἱ μὲν ἄλλοι, Sens.
2.6) and what, precisely, did Parmenides fail to determine?
Theophrastus is very unlikely to be stating that all his predeces-
sors, with the single exception of Empedocles, failed to treat the
specific mechanisms of the different senses (περὶ ἑκάστης κτλ).
If we look at his own subsequent accounts in On the Senses, this
would be false of Alcmaeon, Anaxagoras, Diogenes and
Democritus. ‘The others’, therefore, must refer only to the other
members of the like-by-like party, i.e. Plato and Parmenides. This
reading is further supported by the comment that Empedocles
attempts to refer the specific mechanisms of different senses ‘to
likeness’ (εἰς τὴν ὁμοιότητα), which indicates that, having intro-
duced both the like-by-like and the unlike-by-unlike parties,

170
 4.2 Parmenides’ Theory of Human Cognition
Theophrastus has now already embarked on his discussion of the
former. This interpretation also makes good sense of the qualifica-
tion that ‘the others’ omitted discussion ‘more or less’ (σχεδὸν
ἀπολείπουσιν) since Plato, on Theophrastus’ account, did address
some specific sensory mechanisms, but only those of hearing and
sight (Sens. 5ff). Our reading of this opening sentence concerning
the like-by-like party determines in turn our construal of the next
sentence (linked to it with γάρ). What Parmenides failed to deter-
mine are the specific mechanisms of the different senses.
The following qualification, ‘saying only that . . . ’ (ἀλλὰ μόνον
ὅτι), thus represents what Theophrastus took to be the closest that
Parmenides did come, despite failing to address the specific sen-
sory mechanisms, to accounting for the diversity of our sensory
experiences: there are two elements and our cognition (γνῶσις)
depends on that which exceeds.16
That Theophrastus indeed takes this feature of Parmenides’
theory to explain perceptual-cognitive variations is confirmed by
statement (i) in the next three, difficult statements (Sens. 3.3–5): (i)
‘Thought varies according to whether the hot or the cold predo-
minates; (ii) but a better and purer thought is the one through the
hot; (iii) yet even such thought requires a certain symmetria.’
The key question in interpreting these remarks is whether symme-
tria signifies here a ‘proportion’ (between Hot and Cold) or rather
an ‘appropriate measure’ or ‘fit’ (between Hot as subject of per-
ception and Hot as object of perception). We may distinguish two
alternative interpretations. If we translate ‘symmetria’ as ‘propor-
tion’, the thrust of the remarks will be as follows. (i) Thought
varies according to whether the Hot or the Cold predominates, and
(ii) a thought in which Hot predominates is better than one in
which Cold predominates; (iii) nonetheless, even a thought in
which Hot predominates requires some proportion of Cold.
The implication would be that, for there to be any thought at all,
there must be a mixture of both Hot and Cold.17 On the translation

16
This analysis of Sens. 2.5–3.3 follows the lead of Laks (1990) 10–12 and Stratton (1917)
157, n.4.
17
E.g. Tarán (1965) 257; for symmetria as ‘proportion’, cf. also Verdenius (1964) 10;
Hershbell (1983) 46; Andriopoulos (1975) 558; KRS (1983) 261; Coxon (2009) 377,
380–1.

171
 Why Did Parmenides Write Doxa?
‘fit’, the thought will run as follows. (i) In every instance of
a thought in which both elements are involved, thought varies
according to whether the Hot or the Cold predominates; (ii)
thought through Hot is better than thought through Cold. That is,
perception of Hot through Hot is a better kind of perceptual-
cognitive act than the perception of Cold through Cold; (iii) none-
theless, even thought through Hot, although superior in various
ways, still requires a certain fit or appropriate measure between
the subject and object of perception. That is, perception through
Hot is similarly limited in its perceptual objects only to what
matches the Hot, i.e. what is like the Hot, i.e. Hot.18
Several factors speak strongly in favour of the construal ‘fit’ or
‘appropriate measure’ and the interpretation based on this con-
strual. First, symmetria is a recurrent technical term in On the
Senses and invariably signifies, not a proportion between the
constitutive elements of a mixture, but that an organ of perception
is commensurable with its object and therefore adequately recep-
tive of it (as, for example, in Empedocles’ theory of perception
through pores).19 Second, it is difficult to see how the words ‘the
thought through the Hot’ (τὴν διὰ τὸ θερμόν, Sens. 3.4) could
conceivably describe a thought in which Hot predominates over
Cold but in which both elements are involved. Conversely, this
expression naturally signifies the thought of Hot through Hot.
Most conclusively, Theophrastus offers his remarks on the percep-
tion of corpses explicitly in order to make the point that each
element perceives by itself and independently of the other: ‘But
it is clear that he attributes perception also to the opposite [sc.
element] by itself from that passage in which he says that a corpse,

18
So Fränkel (1975a) 17; Laks (1990) 14–16. As Laks (1990) 14 observes, the two
readings imply slightly different punctuations. The reading based on the construal
‘proportion’, would closely connect (i) and (ii), separating διάνοιαν from βελτίω with
a comma. Conversely, the reading based on the translation ‘fit’, construes (i) and (ii) as
less immediately connected, separating them with a raised comma. But this latter
reading does not implausibly dissociate (i) the remarks about variation from (ii) the
superiority of Hot. The proposition that the thought of Hot through Hot is better than the
thought of Cold through Cold implies the superiority of a compound cognitive act,
which involves both elements but in which the perception of Hot through Hot prepon-
derates, to compound cognitive acts in which the thought of Cold predominates.
19
See Sens. 12.4; 12.7; 13.3; 14.5; 15.5, 8; 32.6; 35.8; 46.3; similarly, Pl. Meno 76d4 (on
Empedocles). Cf. Stratton (1917) 157–8, n.5; Fränkel (1975a) 17; Laks (1990) 14–16.

172
 4.2 Parmenides’ Theory of Human Cognition
because of the departure of the fire, does not perceive light, heat
and sound, but does perceive cold and silence and the other
opposites’ (Sens. 4.4–6). The Cold corpse can only perceive Cold
things.20 The words ‘also to the opposite’ (καὶ τῷ ἐναντίῳ) indi-
cate that Hot too is, of course, likewise perceived through the Hot
by itself. When, therefore, Theophrastus writes that even the
thought through the Hot requires a certain symmetria, he is not
stating – in direct contradiction to his own remarks about the
corpse passage – that, for any cognition to occur, there must be
some proportion of both Hot and Cold in the subject. Rather, he is
stating that even the superior thought of Hot through Hot is limited
by the basic like-by-like requirement for a kind of match or fit
between subject and objects of perceptual reception. A mixture of
both elements, therefore, is not necessary for any sort of percep-
tion to occur. Nonetheless, although mixtures of both elements are
not essential for perception as such, they do characterise the
cognition of living humans, who comprise both elements. While
by the Hot in us we independently perceive Hot and by the Cold
we independently perceive Cold, different mixtures of both ele-
ments in the human agent produce different compound acts, which
involve both the perception of Hot (through Hot) and the percep-
tion of Cold (through Cold). So, when it comes to our perception
of a compound object which itself comprises some combination
of both elements (this tree, this human), it is the simultaneous

20
Laks takes B9.3–4 to indicate that every Doxastic thing consists of both elements. He
infers that, while corpses illustrate the principle that each element perceives indepen-
dently of the other, they do not provide an actual instance of pure Cold perceiving pure
Cold, but rather a mixture in which Cold very strongly outweighs Hot; see Laks (1990)
12–13; cf. Andriopoulos (1975) 558. Laks’ conclusion could conceivably be right, but
his premise is false. B9.3–4 indicates that the totality of things is full of Light and Night
in equal measure. πᾶν cannot signify each Doxastic thing, for then each Doxastic thing
would be full of Light and Night in equal measure. But in many things the two are mixed
unequally; cf. Tarán (1965) 162. The term ἔκλειψις seems to suggest a simple rather than
an almost complete departure (cf. e.g. Hdt. 6.25; 7.37; Strab. 9.5.12). Corpses thus most
probably offer an actual instance of pure Cold perceiving pure Cold, as assumed by
Mansfeld (1996) 173–4; Finkelberg (1986) 405–6. It might be argued on phenomenal
grounds that some vestiges of Light must linger in and around the corpse, since it is not,
after all, a supremely cold and heavy (let alone dark) object. But this objection may not
have occurred to Parmenides. At any rate, the significant point is indeed the illustrative
one: the Cold in the corpse is perceptive of Cold things, and to whatever extent –
whether totally or very largely – the corpse is now devoid of Hot, it is no longer
perceptive of Hot things.

173
 Why Did Parmenides Write Doxa?
occurrence in our frame of our perception of Hot in it through the
Hot in us, and of Cold in it through the Cold in us, that yields our
complex sensory experience of the object.21
If, then, we cognise through each element by itself and inde-
pendently of the other, and if every Doxastic thing comprises Hot,
Cold or some combination of both, then it would seem to follow
that each thing possesses the means of cognition. Theophrastus in
fact records this corollary: ‘And, in general, everything that is
has some knowledge’ (Sens. 4.6–7). It is impossible to determine
whether this is a view which Theophrastus found expressed in
Parmenides (as, indeed, it was in Empedocles: DK31 B110.10),
or his own interpretative inference. At any rate, the inference
seems warranted. And so, if we were initially struck by the view
that the corpse remains to a certain degree perceptive and cogni-
sant, we can now see that such a view is not so surprising in
a world in which everything possesses some form or measure of
cognition.
I follow, then, the construal of symmetria as ‘fit’ or ‘appropriate
measure’ and will proceed below on the assumption that humans
are taken to perceive through each element by itself and indepen-
dently of the other. This point, though, is in fact less crucial for our

21
Theophrastus’ remark that Parmenides did not determine whether, in the case of an
equal mixture, there will be cognition of any kind or what the (or its) general
disposition would be (Sens. 4.2–3) is the strongest prima facie evidence in favour
of the competing construal. If cognition does not necessitate a proportion of both
elements, then why should the prospect of an equal mixture pose any puzzles?
The objection is not, however, compelling. While each element cognises by itself,
every case of human cognition is nonetheless characterised by different mixtures of
both elements. Parmenides traced variations in human cognition to different rela-
tions of predominance (Sens. 3.2–3) and drew hierarchical distinctions between acts
of cognition in which Hot predominates and those in which Cold predominates.
On this basis, Theophrastus could legitimately wonder what kind of cognition or
condition, if any, an equal mixture of Hot and Cold would produce in living humans,
who essentially consist of both elements. Theophrastus voices this query immedi-
ately after commenting on Parmenides’ view that memory and forgetfulness derive
from mixtures of the elements (διὰ τῆς κράσεως, Sens. 4.2), memory most probably
involving a preponderance of Hot and forgetfulness one of Cold (especially given
the reports that Parmenides traced sleep and old age to a deficiency in fire, A46a–b).
Theophrastus’ puzzle, then, offers no compelling evidence for the construal of
symmetria as ‘proportion’ (against which we listed a number of weighty considera-
tions) and for the consequent competing interpretation of his statements in Sens. 3.
3–5, which would conflict directly with his comments on the perception of corpses
(as discussed above).

174
 4.2 Parmenides’ Theory of Human Cognition
purposes and, mutatis mutandis, our discussion below of the
cognition of living humans could be formulated in terms of the
competing interpretation of symmetria. Far more important for
us is the fact that Theophrastus’ paraphrase of the lost passage
about the perception of corpses confirms two of his assertions.
First, it confirms that Parmenides adopted the principle that like
perceives like.22 Second, it confirms that Parmenides regarded
the cosmological elements as that by which we perceive.
Theophrastus appears to work from a first-hand acquaintance
with the original passage (ἐν οἷς φησι, Sens. 4.4) and, whatever
general view we hold of his credentials as a doxographer, could
not have possibly fabricated the very specific position that Cold
corpses perceive or cognise Cold things but not Hot things.
The systoichia of light, heat and sound as contrasted with cold,
silence ‘and the opposites’ betrays the misleadingly oversimpli-
fied nature of Theophrastus’ terminology of Hot and Cold
(retained below for convenience), familiar from the Peripatetic
doxography of Parmenides elsewhere. Parmenides, it seems,
employs different adjectives and labels to bring out different
perceptual aspects of the opposition between the two Doxastic
elements (Light/Night, Light/Heavy, Rare/Dense, etc., see B8.
56–9, cited in Chapter 4.3). The catalogue of what the Cold corpse
can and cannot perceive indicates that, for Parmenides, through
each Doxastic element we perceive other instantiations of that
element in all its perceptual aspects (Light: bright/light/hot/rare,
etc.; Night: dark/heavy/cold/dense, etc.).
With this background in mind, we can turn to B16 itself.
We may usefully divide the fragment into three components:

22
Bredlow (2011), 235–6 indirectly supports the historical plausibility of Theophrastus’
ascription of a like-by-like position to Parmenides by noting various earlier and near-
contemporary parallels. The eyes not only perceive light but are themselves ‘brilliant’ or
‘sparkling’ (ὄσσε φαεινώ, Il. 13.3, 7; φαεινά, Il. 13.435) and ‘fiery’ (Il. 12.466; Il. 13.474;
Od. 4.662; Od. 19.446; cf. Emped. DK31 B84). Helios is at once the cosmic source of
light and himself all-seeing, Il. 3.277; Od. 11.109; Od. 12.323; καταδέρκεται ἀκτίνεσσιν,
Od. 11.16; cf. h.Hom. 2.70; φάος εἰσοράασθαι (‘a light for seeing’), Il. 14.345.
Empedocles adopts the like-by-like principle of perception (DK31 B109). We also
encounter the notion that like attracts to like explaining, for example, the body’s
absorption of food or drugs (Nat. Hom. 6; Morb. 4.2–3), the behaviour of fire (Emped.
DK31 B62.6) and the movement of atoms (D.L. 9.31 = DK67 A1; DK68 A128). Cf.
more generally Od. 17.218 (ὡς αἰεὶ τὸν ὁμοῖον ἄγει θεὸς ἐς τὸν ὁμοῖον).

175
 Why Did Parmenides Write Doxa?
(i) ὡς γὰρ ἑκάστοτ’23 ἔχει κρᾶσις24 μελέων πολυπλάγκτων,25
τὼς νόος ἀνθρώποισι παρέστηκεν·26 (ii) τὸ γὰρ αὐτό
ἔστιν ὅπερ φρονέει μελέων φύσις ἀνθρώποισιν
καὶ πᾶσιν καὶ παντί· (iii) τὸ γὰρ πλέον ἐστὶ νόημα.
(i) For as on each occasion is the mixture of much-wandering
limbs, so is mind present to humans; (ii) for, in all humans and in
each, that which cognises is the same thing, namely the nature of
the limbs;27 (iii) for the full is thought.
(i) This statement expresses a physiological theory of human cogni-
tion. Different mixtures in our frame determine in each instance the
state of our mind (ὡς . . . τὼς νόος ἀνθρώποισι). The term ‘limbs’
(μελέων), especially since in (ii) Parmenides speaks of the human
frame (μελέων φύσις ἀνθρώποισι), probably refers to the human
frame as a whole rather than directly to the two cosmic
elements.28 Nonetheless, since the doxographic context, as analysed
above, renders it beyond doubt that the mixture in question is of the
two elements (δυοῖν ὄντοιν στοιχείοιν κτλ, Sens. 3.2), the ‘mixture
of our limbs’ (or ‘frame’) would still indirectly signify the particular
compound of Hot and Cold of which our frame consists at any given
moment.29 Again, different mixtures of the Hot in us (by which we
perceive Hot) and the Cold in us (by which we perceive Cold)
produce different compound cognitive acts.
(iii) I have translated pleon as ‘full’. An alternative translation is
‘more’. Does (iii), then, state that ‘the full’ or ‘the more’ is thought?

23
Preserved by Theophrastus and the two oldest mss. of Aristotle’s Metaphysics. ἑκάστῳ
(Ab) and ἕκαστος (E2) cannot be conclusively excluded, although, as Coxon (2009) 378
argues, the former seems a corruption of ἑκάστοτ’ and the latter a later correction; cf.
Heitsch (1974) 191; Verdenius (1964) 6; Tarán (1965) 169.
24
Retaining the manuscripts’ κρᾶσιν and postulating some implicit subject (e.g. νόος or τις)
would yield the same overall sense (as Heitsch observes (1974) 191) but involve an unusual
use of ἔχει (Mourelatos (2008a) 253; cf. Fränkel (1975) 16; Verdenius (1964) 14).
25
Over Aristotle’s all but unparalleled and awkward πολυκάμπτων, following Mansfeld
(1996) 162–3, n.17; Coxon (2009) 378; Verdenius (1964) 7–8; Tarán (1965) 170.
26
Over Aristotle’s unmetrical παρίσταται, prompted by his immediately preceding cita-
tion of Empedocles’ παρίσταται (Metaph. Γ.5 1009b20, 23), cf. Fränkel (1975) 16; Tarán
(1965) 170; Heitsch (1974) 191–2; Coxon (2009) 379.
27
Or: ‘The nature of the limbs is the same as that which it cognises.’
28
Verdenius (1964) 7; (1949) 126; cf. Fränkel (1975a) 16; Tarán (1965) 170; Coxon
(2009) 378; Sider and Johnstone (1986) 22; Vlastos (1946) 66, n.4; contra: Laks (1990)
6, n.15. For μέλεα or γυῖα signifying the living mortal’s frame as a whole, see Il. 17.211;
Od. 6.140; Od. 10.363; Od. 13.432; Pind. P 4.48; fr. 131.3 M; Aesch. Pers. 991; cf.
Fränkel (1975a) 16; Tarán (1965) 170.
29
Cf. Verdenius (1964) 20.

176
 4.2 Parmenides’ Theory of Human Cognition
There is a simple but compelling reason to reject ‘more’.
By identifying thought with whichever element exceeds,
Parmenides would be denying the lesser element any influence
whatsoever on any given mortal thought (noêma), even when the
elements are almost evenly mixed in the frame.30 But in (i) it is
particular mixtures of both elements which determine the state of our
mind (noos).31 ‘The full’, then, is thought. B9.3–4 further supports
this reading and suggests that the ‘fullness’ in question is, as implied
independently by (i), of the two elements: ‘All is full (pleon) of light
and unseen night together, of both in equal measure.’ The totality of
Doxastic things is ‘full’ of both elements in equal measure. It seems
impossible to determine to what extent, if any, Parmenides clarified
further the nature of the ‘fullness’ to which he refers in (iii).
The thought may have gone no further than the following. By Hot
and Cold we cognise Hot and Cold respectively; furthermore, as
living mortals we comprise both elements; thus, our thought is ‘the
full’, i.e. the sum of Hot and Cold of which we consist as it
encounters, receives and so cognises Hot (cf. Sens. 4.5–6: ‘light,
heat, sound’, etc.) and Cold (‘cold, silence and the opposites’) from
without. There is, then, no reason to ascribe to Parmenides anything
like the Empedoclean theory of perception-through-pores in order to
make sense of the translation ‘full’.32 Nonetheless, an interpretation
of to pleon as ‘the full’ may well underlie Theophrastus’ talk of ‘fit’
or ‘commensurability’ (symmetria) between the Parmenidean sub-
ject and objects of perception.33 Theophrastus repeatedly uses
30
This last point renders it implausible to defend ‘more’ by importing into (iii) – as Hussey
(2006) 28 suggests – the Anaxagorean principle (DK59 B12) that the predominating
elements in a mixture determine its appearance. In the context of Parmenides’ statement
in (iii) – ‘the πλέον is thought’ – the implication would be that, even when the ratio is
very close (e.g. when there is just a little bit more Hot than Cold in the frame), the
thought would be Hot and of Hot and in no way Cold or of Cold.
31
Following Tarán (1965) 256–7; see further the arguments in Mansfeld (1964) 191–2;
Laks (1990) 7–8; Kraus (2013), 494–5. Opting for ‘more’: Heitsch (1974) 198; Coxon
(2009) 380–1; Andriopoulos (1975) 555; Long (1996) 147, n.35. As Laks (1990) 7–8
n.19 argues, νόημα most probably signifies a concrete instance of thinking; cf. also Tarán
(1965) 254.
32
Pace Bredlow (2011) 241–2.
33
Given the interpretation of Sens. 1–3 and of symmetria followed above, Theophrastus’
words κατὰ τὸ ὑπερβάλλον . . . τὴν διάνοιαν (Sens. 3.2–4) are not a gloss of B16.4
implying that he read πλέον as ‘more’; following Laks (1990) 6; contra e.g. Verdenius
(1964) 17, n.3; Tarán (1965) 257–8, 262; Coxon (2009) 380–1); Kraus (2013) 494; cf.
n.21 above.

177
 Why Did Parmenides Write Doxa?
symmetria in reference to Empedocles’ theory of perception through
pores (Sens. 12.4; 12.7; 13.3; 14.5; 15.5, 8). His application of the
term here suggests that he saw Parmenides as anticipating, in a more
obscure and embryonic manner, Empedocles’ conception of cogni-
tion as a state in which like is filled by like.34 If Theophrastus did
entertain such a view about the relation between Parmenides’ and
Empedocles’ theories, it would seem to be a hermeneutically and
historically plausible one. Empedocles’ theory of pores explicitly
postulates the perception of like-by-like (‘by earth we see earth; by
water – water etc.’, DK31 B109) and could have aimed in part to
explicate Parmenides’ less developed and murkier notion of ‘the
full’.
These considerations notwithstanding, the word pleon itself
is indeed ambiguous. The sense ‘more’ could play a role as an
underlying, secondary connotation. While the primary thrust
of the term would be to indicate a fullness and a filling in the
sense analysed above, the word could perhaps also serve to
put us in mind of the general point that the predominance of
one element or the other shapes the state and quality of our
thoughts (which are themselves conceptualised as a process of
being filled).35
(ii) The expression ‘the nature of the limbs’ or ‘of the
frame’ (μελέων φύσις) in (ii) recalls the ‘mixture of limbs’
(κρᾶσις μελέων) in (i). This parallel indicates that ‘nature’
(φύσις) in (ii) should be construed as the frame’s constitution
or character consisting, as in (i), in a combination of Hot and
Cold.36 We may distinguish three primary, grammatically
viable readings of (ii):

34
Cf. Laks (1990) 16–18, who observes also that Aëtius goes as far as ascribing to
Parmenides (alongside Empedocles, Anaxagoras and Democritus) a full-fledged
theory of pores (παρὰ τὰς συμμετρίας τῶν πόρων, A47), which offers precisely the
sort of account of different particular perceptions which (the no doubt better
informed) Theophrastus had complained Parmenides failed to offer.
35
Hussey (2006) 29 suggests that both senses, ‘the full’ and ‘the more’, could be at play in
B16.4.
36
Following Mourelatos (2008a) 254; cf. Hershbell (1983) 49. The sense is attested as
early as Od. 10.302–3 (πόρε φάρμακον . . . καί μοι φύσιν αὐτοῦ ἔδειξε) and paralleled in
B10.1 (αἰθερίαν τε φύσιν) and B10.4–5 (ἔργα . . . σελήνης καὶ φύσιν); cf. Verdenius (1964)
15, n.5; Curd (1998) 42ff.

178
 4.2 Parmenides’ Theory of Human Cognition
(ii)a For, in all humans and in each, that which cognises is the same
thing, namely the nature of the limbs (construing ὅπερ as subject of
ἔστιν and φρονέει, μελέων φύσις as in apposition with it)
(ii)b . . . the nature of the limbs is the same as that which it cognises
(μελέων φύσις as the subject of both ἔστιν and φρονέει with αὐτό as its
predicate, ὅπερ as accusative)
(ii)c . . . it is the same thing which the nature of the limbs cognises (αὐτό
as subject of ἔστιν, ὅπερ as accusative, μελέων φύσις as subject of
φρονέει).37
Our primary criterion for choosing between (ii)a-c must be how
well each reading explicates the logical structure of the three
statements which constitute B16. Statement (ii) begins with ‘for’
(γάρ). Thus, (ii) should ideally justify or imply (i), i.e. the propo-
sition that different mixtures of Hot and Cold in our frame deter-
mine the state of our cognition. Statement (iii) likewise begins
with ‘for’. Thus, (ii) in turn should ideally be justified or implied
by the more obscure, and therefore less helpful, (iii): ‘the full is
thought’.38
(i) indeed follows from (ii)a. The proposition that mixtures of
Hot and Cold in our frame determine at any moment the state of
our cognition follows from the proposition that the nature of our
frame is what cognises, given that the nature of our frame is itself
nothing more than its current mixture of Hot and Cold (by which
we cognise Hot and Cold respectively). We might worry that (ii)a
comes closer to repeating than to justifying (i).39 But (ii)a does
seem to explain (i), in a manner which sufficiently accounts for
what may well be a rather cavalier ‘for’. We can think of (ii)a as
articulating the constant principle that underlies all particular
instances of cognition. The reason that, as (i) states, any particular
instance of human cognition (ἑκάστοτε) is determined by the
current mixture of Hot and Cold in our frame is, as (ii)a explains,
that, invariably, the current constitution of our frame is what
cognises. Furthermore, we can see how (ii)a would follow from
37
E.g. (ii)a: DK, ad loc; KRS (1983) 261; (ii)b: Fränkel (1975a) 42, n.53, cf. 16; Vlastos
(1946) 66; Kahn (1969) 722; Laks (1990) 7; (ii)c: Tarán (1965) 256; Mourelatos (2008a)
253; Hershbell (1983) 49. On the syntactic proposals in Bredlow (2011) 243–5, see Tor
(2015) 10 (n.21), 13 (n.31).
38
Minimally, the three statements must cohere: (ii) cannot conflict with (i); contra
Robbiano (2006) 131, n.356.
39
So Verdenius (1964) 14.

179
 Why Did Parmenides Write Doxa?
(iii). If thought is identified with the state in which the Hot and
Cold of which we consist within are filled by, and so cognise, Hot
and Cold from without, then the sum of Hot and Cold within us,
i.e. the nature of our frame, will indeed be what cognises.
(ii)b explicitly formulates the principle that like perceives like.
The term ‘the same’ (τὸ αὐτό) will presumably signify here
qualitative rather than numerical identity, identifying the subject
and object of perception as the same kind of thing (e.g. Hot).40
Again, we can see how (ii)b could imply (i). If the nature of our
frame is qualitatively the same as the objects which it perceives,
then different mixtures of Hot and Cold in our frame will indeed
produce different cognitions (e.g. the more Hot we have the more
or better able we will be to perceive things rich in Hot). And again,
(ii)b could be taken to follow from (iii). If, as (iii) states, thought is
a state in which like is filled by like, then, as (ii)b states, the Hot
and Cold of which our frame consists are indeed qualitatively the
same as what our frame thinks.41
Reading (ii)c is untenable. Proposition (i) could not follow from
proposition (ii)c. The proposition that different mixtures of Hot
and Cold in our frame determine the state of our cognition could
not follow from the proposition that each and every mortal cog-
nises the same thing. Indeed, the two statements are inconsistent.
Since different mixtures produce different cognitions, and since
our mixtures fluctuate – ‘on each occasion’ (ἑκάστοτ’, B16.1), ‘the
thought varies’ (ἄλλην γίνεσθαι τὴν διάνοιαν, Sens. 3.3–4) – it is
not the case that each and every mortal cognises the same thing.42

40
Vlastos (1946) 66–7 implies the qualitative reading. Verdenius (1964) 18 insists on
numerical identity, but this undermines the capacity of any Doxastic thing to cognise any
other Doxastic thing and, indeed, any distinction between subjects and objects of
cognition, such as living humans and the sounds they perceive (Sens. 4.4–6).
41
One worry (though by no means a decisive one) is that (ii)b requires a slightly
unusual application of the formula ὁ αὐτὸς ὅσπερ, which is typically used to
express different statements about one and the same object, as in (ii)a and (ii)c,
rather than to identify distinct objects or concepts with each other, as in (ii)b. See
e.g. Od. 8.107: ἦρχε δὲ τῷ αὐτὴν ὁδὸν ἥν περ οἱ ἄλλοι. See further Heitsch (1974)
197; Coxon (2009) 380.
42
Tarán (1965) 258, 261 explains that, whenever a given mixture is present in the body, the
same thought would result. All that (ii)c states, however, is ‘all mortals think the same
thing’, not ‘whenever mortals have the same mixture, then they all think the same thing’.
The qualification ‘whenever . . . mixture’ assumed by Tarán jars with the universalising
καὶ πᾶσιν καὶ παντί and is simply not found in our text (it certainly cannot be supplied

180
 4.2 Parmenides’ Theory of Human Cognition
Readings (ii)a and (ii)b, then, are grammatically and logically
viable. It is difficult to choose between them. Indeed, perhaps we
should not choose. It may well be the case that the notable syntac-
tic ambiguity in B16.2a-4b (is ὅπερ subject or object?) is deliberate
and that both senses are in play.43 (ii)a tells us: it is the same thing
that thinks in all humans, i.e. the mixture of Light and Night which
is the nature of the limbs. (ii)b tells us: this thinking mechanism is
itself the same as what it thinks, i.e. other tokens and compounds
of Light and Night.
Indeed, whether we opt for translation (ii)a or translation (ii)b,
both of those propositions will in fact be true according to
Parmenides’ theory of human cognition, as we reconstructed this
theory before we raised the question of the interpretation of these
particular lines (B16.2b-4a). Let us recapitulate the key features of
this theory of cognition. We consist of fluctuating mixtures of Hot
and Cold. By the Hot and Cold within us we cognise Hot and Cold
respectively. Our cognition is determined at each moment by the
combinations of Hot and Cold of which our mortal bodies consist.
What the mortal body cognises at any moment are its own or other
instances of these elements (its cognitive engagement with which
is described as a state of fullness).44

from (i), which itself needs to follow from (ii) rather than imply it). Hershbell (1983) esp.
50 (followed by Gallop (1984) 87; Robbiano (2006) 131–2) accepts reading (ii)c,
interprets πλέον as a reference to what-is and locates B16 in Alêtheia. Both
Theophrastus’ commentary (note esp. δυοῖν ὄντοιν στοιχείοιν (Sens. 3.2) and his dis-
cussion of corpses), and the references to mixture and the human body within B16 itself,
secure its Doxastic provenance. For a detailed refutation of Hershbell’s proposal, see
Andriopoulos (1975) esp. 556–7; cf. Laks (1990) 3, n.9. Although ‘the full’, like so
many other aspects of Doxa, echoes descriptions of what-is in Alêtheia (as discussed by
Mourelatos (2008a) 254–9), it cannot refer to what-is in any straightforward manner.
43
Cf. Hussey (2006) 29 for the suggestion that the syntactic ambiguity of B16.2b-4a is
deliberate.
44
Hussey (2006) advances the novel suggestion that B16 describes, not our mind’s
direct cognitive-and-perceptual contact with external objects, but rather its engage-
ment with scale-models which are internal to the mind, made out of the same
materials (Light and Night) and (at least in the case of perception) causally
generated by the outside world, which they represent. Hussey’s arguments for
this intriguing view are not compelling, but, at any rate, nothing much hangs for
us on whether or not we entertain it. In either case, our physiological makeup
determines what things, and what sort of things, we perceive-and-cognise (recall
the corpse passage: light, heat, sound, darkness, cold, silence, etc.), whether our
cognitive engagement with such things is direct or mediated via internal scale-
models.

181
 Why Did Parmenides Write Doxa?
Three points bear further emphasis. First, B16 pointedly describes
the cognition of humans (ἀνθρώποισι . . . ἀνθρώποισι, B16.2–3).
As Karl Reinhardt remarked: ‘Coming from the goddess’s lips, how
studied is the emphasis on human knowledge!’45 Second, B16 is
a universalising account of human cognition (‘as on each
occasion . . . in all humans and in each . . . ’). It nowhere indicates
any room for exceptions. It describes all, not some, acts of human
cognition. Third, and relatedly, human cognition of Hot and Cold is,
on this theory, passively determined by the deliverances of physiolo-
gical mechanisms. As on each occasion is the mixture of their limbs,
so is mind simply present to humans (ὡς . . . παρέστηκεν, B16.1–2a).
The poetic models which B16 famously echoes46 underscore this
point:
τοῖος γὰρ νόος ἐστὶν ἐπιχθονίων ἀνθρώπων,
οἷον ἐπ’ ἦμαρ ἄγῃσι πατὴρ ἀνδρῶν τε θεῶν τε.
For such is the mind of humans upon the earth
As the day which the father of gods and humans brings upon them.
(Od. 18.136–7)
τοῖος ἀνθρώποισι θυμός . . .
γίνεται θνητοῖς, ὁποίην Ζεὺς ἐφ’ ἡμέρην ἄγῃ.
Such . . . for mortal humans the mind
Becomes, as the day which Zeus brings upon them.
(Archilochus, fr. 131 W)
These passages, like B16.1–2a, establish a causal relation between
certain conditioning factors and the human mind as conditioned by
them.47 Although Parmenides diverges from the other poets by
conspicuously omitting any mention of divine ministration, he
similarly underscores the passive determination of, again, specifi-
cally human thought.48 According to B16, for the human mind
(noos), cognition (noos) of Hot and Cold, and by Hot and Cold, is
something that just happens.

45
Reinhardt (1974; orig. 1916) 307.
46
See Fränkel (1975) 15; cf. e.g. Verdenius (1964) 27; Tarán (1965) 253; Heitsch (1974)
194–5; Coxon (2009) 95, 378–9; Kraus (2013) 494.
47
Heitsch (1974) 195.
48
On the point concerning the passive determination of thinking expressed in B16, cf.
Coxon (2009) 380 (‘a function of physical temperament’); Mourelatos (2008a) 225–6;
Heitsch (1974) 194; KRS (1983) 262; Laks (1990) 12.

182
 4.3 Cognition, krisis, Sense-Perception
4.3 Cognition, krisis, Sense-Perception
In this section, we will consider how the type of thinking which
Parmenides identifies as a physiological necessity for the human
mind relates to the type of thinking which underpins the cosmol-
ogy of Doxa.
The cognitive language which Parmenides uses in B16 (noos,
phroneei, noêma) indicates that the fragment expounds the phy-
siological mechanisms that determine the kinds of things of which
we are aware, as well as our thoughts and judgements with regard
to such things. The term noein and its cognates signify, not merely
‘thought’ without involving what we would call a belief compo-
nent, but rather the apprehension of a situation, object or person in
a way that implies the formation of a judgement or judgements
with regard to them. To give two of countless possible illustra-
tions, in the Iliad Helen recognises that the neck of the disguised
Aphrodite is that of a deity (ἐνόησε θεᾶς . . . δειρήν, Il. 3.396ff),
while Talthybius judges that the man before him is Machaon (τὸν
δὲ νόησεν κτλ, Il. 4.200).49 The term phroneei in B16.3 must
signify the function of the mind (noos) and is here, as often, almost
synonymous with the Greek verb gignôskein (‘cognise’, ‘recog-
nise’; cf. ἡ γνῶσις, Sens. 3.3; τινὰ γνῶσιν, Sens. 4.7), as in the
Homeric formula: ‘I understand (ginôskô), I comprehend
(phroneô); you bid these things to someone who realises
(noeonti)’, where the three verbs serve as functional
equivalents.50 Now, by Parmenides’ time, an erring mind (noos)
was by no means an incoherent notion. Hesiod, for example, can
speak of profit or women deceiving one’s mind (Op. 323, 373),
warn against actions which people undertake ‘in the ignorance of
their mind’ (ἀιδρείῃσι νόοιο, Op. 685) and even narrate
Prometheus’ deception of Zeus’ mind (Th. 537).51 Aristotle’s off-
hand ascription to Parmenides on the basis of B16 (and to various
other thinkers on the basis of equally uncompelling citations) of

49
See further the discussion of noos and noein in Ch. 3.4; note also von Fritz (1974) esp.
23–4, 33; Warden (1971) 3–4; Darcus-Sullivan (1994) 108.
50
Od. 16.136; cf. Verdenius (1964) 16.
51
See further von Fritz (1974) 29; cf. Mourelatos (2008a) 175–6. On the use of noein and
its cognates with regard to what-is as compared to their use in B16 and of erring mortals,
see Ch. 5.5.2.

183
 Why Did Parmenides Write Doxa?
the view that all our impressions must be true (Metaph. Γ.5 1009b
13–15) is thus quite unmotivated. The important point for us,
however, is that noos is always a faculty which, whether success-
fully or not, attempts to cognise correctly a given object or state of
affairs (rather than merely ‘think’ or ‘perceive’). In sum, what we
get in B16 is a theory, not merely of sensory perceptions that
involve no belief component, but, again, of the factors which
determine the kinds of things of which mortals are aware, as
well as their thoughts and judgements with regard to those things.
This is what I mean by saying that B16 is a theory of (human)
cognition.52
Throughout the poem, Parmenides makes pivotal and calcu-
lated use of the term krisis and its cognates. The term encom-
passes both dichotomies and judgements made on their basis.53
Whereas Alêtheia is framed by one kind of dichotomy, between
‘is’ and ‘is not’ or ‘what-is’ and ‘what-is-not’, Doxa is framed by
a very different one, between ‘Light and Night’ (or: ‘Hot’ and
‘Cold’, ‘Rare’ and ‘Dense’, etc.). Parmenides’ use of the term
krisis, and especially his contrast between the krisis of Alêtheia
and the krisis of Doxa, are important for our purposes.
By considering how the goddess introduces the Doxastic krisis,
and how it relates to the krisis of Alêtheia, we may both sharpen
our account of the kind of cognition ascribed to the mortal
mind in B16 and further elucidate the fragment’s import for the
epistemological question. After the goddess criticises the mental
habits of the wayward mortals (B7.3–5a), she issues a positive
injunction to the kouros (B7.5–6):
κρῖναι δὲ λόγῳ πολύδηριν ἔλεγχον
ἐξ ἐμέθεν ῥηθέντα.
But judge by reasoning the much-contested examination
Spoken by me.
The judgement or discrimination – krinai – which the goddess
prescribes here can only align with the contrast, the krisis, between

52
Parmenides’ use of this cognitive language in B16 thus invalidates Vlastos’ claim,
(1946) 67, that B16 expresses a theory strictly of non-cognitive sense-perception.
53
Separation and selection: e.g. Il. 2.362; Il. 6.188; Il. 9.521; Il. 11.697; Il. 5.501.
Judgements: e.g. Il. 16.387; cf. Kahn (1973) 382.

184
 4.3 Cognition, krisis, Sense-Perception
‘is’ and ‘is not’, which expressly frames the argumentation about
what-is in Alêtheia (B8.15–16):
ἡ δὲ κρίσις περὶ τούτων ἐν τῷδ’ ἔστιν·
ἔστιν ἢ οὐκ ἔστιν.
But the discrimination concerning these matters lies in this:
[It] is or [it] is not.
The ‘account’ or ‘reasoning’ (logos) which the goddess prescribes for
the kouros thus presumably looks forward to the tightly structured
system of deductive arguments in B8, which are based on this krisis
and through which the goddess concludes that what-is is ungener-
ated, imperishable, changeless, motionless and homogeneous.
When she introduces Doxa, the goddess traces mortal beliefs to
a pointedly different krisis, one between two elemental opposites
with respect to their perceptual form (B8.50–9):
ἐν τῷ σοι παύω πιστὸν λόγον ἠδὲ νόημα 50
ἀμφὶς ἀληθείης· δόξας δ’ ἀπὸ τοῦδε βροτείας
μάνθανε κόσμον ἐμῶν ἐπέων ἀπατηλὸν ἀκούων.
μορφὰς γὰρ κατέθεντο δύο γνώμας ὀνομάζειν·
τῶν μίαν οὐ χρεών ἐστιν – ἐν ᾧ πεπλανημένοι εἰσίν –
ἀντία δ’ ἐκρίναντο δέμας καὶ σήματ’ ἔθεντο 55
χωρὶς ἀπ’ ἀλλήλων, τῇ μὲν φλογὸς αἰθέριον πῦρ,
ἤπιον ὄν, μέγ’ ἐλαφρόν, ἑωυτῷ πάντοσε τωὐτόν,
τῷ δ’ ἑτέρῳ μὴ τωὐτόν· ἀτὰρ κἀκεῖνο κατ’ αὐτό
τἀντία νύκτ’ ἀδαῆ, πυκινὸν δέμας ἐμβριθές τε. 59

Here I conclude for you the trustworthy account and understanding

Concerning true reality; from this point on learn mortal opinions,
Hearing the deceptive order of my words.
For they resolved to name two forms
Of which just one may not be named – this is where they have gone astray –
And they distinguished opposites in accordance with their body,54 and they set
signs
Apart from each other. To one, the aethereal fire of flame:
It is gentle, very light-weight, everywhere the same as itself,
But not the same as the other. And that other again is in itself
Just the opposite: dark55 night, a dense body and heavy.

54
δέμας is used of living bodies or bodies regarded as alive (Coxon (2009) 346).
It indicates a thing’s concrete, perceptual aspects: the daughter of Chryses is not inferior
to Clytaemnestra οὐ δέμας οὐδὲ φυήν (Il. 1.115).
55
The immediate context calls for this translation, but the term ἀδαῆ also suggests
ignorance. Night is ‘ignorant’ inasmuch as it allows only the much inferior and dull

185
 Why Did Parmenides Write Doxa?
We discuss the sense in which the goddess’s account of Doxa is
‘deceptive’ in Chapter 4.4. Our focus here is on the nature of
Doxa’s krisis and how the demarcation between Light and Night
relates and compares to the demarcation between ‘is’ and ‘is not’.
Aristotle aligns (τάττει, Metaph. Α.5 986b33–987a2) and even
identifies (GC 1.3 318b6–7) Light with what-is and Night with
what-is-not. Identifying either element with what-is is untenable
because both possess properties which are incompatible with the
properties of what-is.56 Light, for example, is mobile (B12.2)
while what-is is immobile (B8.26, cf. B8.41). The mixtures in
which both elements participate (e.g. B16.1) are quite foreign to
the static homogeneity of what-is. Nor can we even align either
Doxastic element with what-is or what-is-not. Parmenides empha-
tically highlights their equality (all is full of both elements ‘in
equal measure’, B9.4).57 There is nothing negative or privative
about Night as contrasted with Light (or vice versa). Night, for
example, is heavier and denser than Light, which is also, for its
part, characterised by some measure of weight and density (B8.
57–9). Both elements, therefore, constitute equally positive things,
and neither can be asymmetrically aligned with what-is-not.58
This, in turn, raises a more significant point. All such align-
ments obscure the qualitative difference between the distinction
made by Alêtheia’s krisis (is / is not, what-is/what-is-not) and the
distinction made by Doxa’s krisis (Light/Night, or Hot/Cold,
etc.).59 On any reading, Alêtheia does not establish what-is-not
as an equally positive counterpart to what-is. Parmenides stresses
the equality of what-is with itself and the stable self-sufficiency of
its identity: ‘for it is equal with itself in every direction’ (B8.49),
‘remaining the same and in the same by itself it lies’ (B8.29).
By contrast, the two Doxastic elements are precisely equal with

cognition of Cold through Cold, and its preponderance brings about a general deteriora-
tion in one’s mental faculties; see further Ch. 5.2, with n.69.
56
Clark (1969) 26.
57
See similarly Long (1975) 90; Kraus (2013) 487–8; Verdenius (1964) 62; Minar (1949)
52; Cornford (1933) 108.
58
The superior vitality and mental vigour of Hot and, in general, the axiological nature of
the contrast between the two elements may explain Aristotle’s alignment, but they do
not warrant it.
59
In highlighting the qualitative difference between the two kriseis, I follow and develop
Mansfeld (1964) 86–91, 133; cf. also Kraus (2013) 489.

186
 4.3 Cognition, krisis, Sense-Perception
each other and the identity of each with itself is pointedly qualified
by its demarcation from its counterpart: ‘everywhere the same as
itself, but not the same as the other’ (B8.57–8).60 Parmenides’
juxtapositions of contrary attributes (Light/Night, Light/Heavy,
Rare/Dense, etc., B8.56–9) characterise the elements of Doxa as
enantiomorphic opposites, each one being what the other is not.61
Light, therefore, is hot, light, rarefied etc. and is not cold, dark,
heavy, dense etc. The opposite is the case for Night. The krisis of
Alêtheia, therefore, maintains a distinction between ‘is’ and ‘is
not’, and aligns the homogeneous what-is entirely with ‘is’ and in
no way with ‘is not’. Conversely, the Doxastic krisis transgresses
the distinction between ‘is’ and ‘is not’ by positing two elements,
each of which simultaneously is (itself) and is not (the other):
‘being and not-being’ (εἶναί τε καὶ οὐχί, B8.40), and again: ‘every-
where the same as itself, but not the same as the other’ (τωὐτόν . . .
μὴ τωὐτόν, B8.57–8). Doxa’s krisis demarcates two equal, ele-
mental opposites. It precisely fails to sustain the separation
between ‘is’ and ‘is not’. We see here, importantly, that
Alêtheia’s krisis and Doxa’s krisis are competing in the sense
that we could not maintain, and think in terms of, both sorts of
demarcations at the same time. The Doxastic contrast between
Light and Night – both of which are retained in the cosmology and
each of which is itself and is not the other – transgresses Alêtheia’s
krisis, which separates ‘is’ from ‘is not’, only one of which is
retained (‘is’) and characterises what-is. When we think in terms
of Doxa’s krisis, we cannot at the same time think in terms of
Alêtheia’s krisis.
The context of Simplicius’ citation of B8.53–9 is instructive (in
Ph. 9.30.20–31.7 = Coxon (2009) Testim. 204). Simplicius reports
that as the ‘elemental principles of things that are generated’
Parmenides posited an opposition (antithesis), which was glossed
interchangeably with the pairs Light/Darkness, Fire/Earth, Rare/
Dense, the Same/Other (9.30.20–2). Simplicius cites in addition
a prose notation which he found in his copy of the poem: ‘And
60
This phrase clearly applies symmetrically to both Doxastic elements (n.b. ἀτὰρ κἀκεῖνο
κτλ, B8.58–9), and cannot be seen as attributing specifically to Light one of the proper-
ties of what-is, pace Kahn (2009) 216.
61
Cf. Curd (1998) 107–8; Nehamas (2002) 55, n.43.

187
 Why Did Parmenides Write Doxa?
indeed in between the verses a short passage in prose is inserted as
though by Parmenides himself. It reads as follows: “To this are
applied the rare and the hot and the light and the soft and the light-
weight, but to the dense are named the cold and darkness and the
hard and the heavy. For these things are demarcated in either way
each from the other”’ (ταῦτα γὰρ ἀπεκρίθη ἑκατέρως ἑκάτερα,
9.31.3–7). This catalogue of opposites is closely reminiscent of
Parmenides’ own juxtapositions of contrary attributes in B8.56–9.
Whether or not the specific terms that the author of the prose
notation highlights all derive from Parmenides himself, he
expresses an incisive insight into the conceptual basis of Doxa’s
elemental cosmology. As Simplicius writes, Parmenides ‘clearly
posited two antithetical elements’ (ἀντίθετα δύο στοιχεῖα, 9.31.7)
and, as the annotator concludes, ‘these things are demarcated
(ἀπεκρίθη) in either way each from the other (ἑκατέρως
ἑκάτερα)’. The essence of Doxa does not lie in any one aspect of
the opposition between the two elements (‘Light/Night’, or ‘Hot/
Cold’, etc.). It lies, rather, in the nature of the Doxastic krisis itself.
That is, it lies in the demarcation between the two equal enantio-
morphic elements – on the basis of (all) their sensible aspects
(δέμας, B8.55) – each of which elements essentially both is (itself)
and is not (the other).
With this account of Doxa’s krisis in place, we can take a closer
look at the difficult lines B8.53–5 (cited above). We will return in
Chapter 4.4 to the voluntariness which Parmenides associates with
mortal naming (‘they resolved to name two forms’). Our focus
here is on Parmenides’ subsequent remarks on this action.
The translation of B8.54a is controversial. Most commentators
translate ‘one of which must not be named’ and take the implied
subtext to be ‘[sc. while it is right to name the other]’. This
construal draws on Aristotle’s alignment of Light with what-is
and Night with what-is-not. Thus the name ‘Night’ is to be dis-
carded and ‘Light’ retained.62 We saw, however, that this
Aristotelian alignment is fundamentally ill-conceived.

62
E.g. Patin (1899) 591–4; Vlastos (1946) 74; Woodbury (1986) 11 n.26; Kahn (2009)
216–17; cf. Mourelatos (2008a) 85–7. Others align Night with what-is: Popper (1992)
12–16; Sedley (1999a) 124. Against the attempt in Long (1975) 90–1 to reject the

188
 4.3 Cognition, krisis, Sense-Perception
An alternative rendering, then, is: ‘of which just one may not be
named [sc. without the other]’. We might worry that this construal
will be incoherent as a criticism of the mortals, who do name both
elements and not just one without the other.63 Coxon (2009)
follows this translation, but avoids the incoherence by construing
‘this is where they have gone astray’ as the mortals’ own criticism
(endorsed by Parmenides) of cosmologies which are based on
a single element.64 Coxon’s interpretation entails, however, that
while ‘they resolved’ in B8.53 and ‘they distinguished’ in B8.55
have the same subject (the mortals), the words ‘they have gone
astray’ in B8.54 suddenly refer to a different ‘they’ (the monistic
cosmologists whom the mortals criticise). But there is no textual
support for such an abrupt and confusing shift in subject.65
Philosophical denunciations, moreover, are for the goddess to
utter, not the mortals. I would retain this construal but interpret
the phrase, not as itself criticising the mortals, but as an analysis of
what we found above (independently of this line) to be the essen-
tial relation between the two opposites of Doxa’s krisis (each of
which essentially both is itself and is not the other), as contrasted
with the krisis of Alêtheia. By contrast with the krisis between ‘is’
and ‘is not’, in which one element is retained (‘is’) and the other
rejected (‘is not’), one cannot postulate one Doxastic element but
deny its counterpart. If there is Light, there must also be Night, and
vice versa. ‘Doxastic monism’ is a contradiction in terms.
The phrase ‘of which just one may not be named [sc. without the
other]’ is the goddess’s analysis of the nature of the Doxastic
krisis. The words ‘this is where they have gone astray’ is her
parenthetical denunciation of the mortals, which is evoked by
that analysis but pertains to the entire sequence as a whole: they
resolved to name two forms, of which just one (without the other)
may not be named (unlike ‘is’ and ‘is not’, one of which must be
retained (‘is’) and the other discarded (‘is not’): naming opposites

Aristotelian alignment but retain the translation, see Guthrie (1965) 53, n.3; Mourelatos
(2008a) 220, with n.71.
63
So Schofield, see KRS (1983) 256, n.1; contrast KR, ad loc.; for the construal, see e.g.
Verdenius (1964) 62; Clark (1969) 26.
64
Coxon (2009) 80, 344–5; cf. Verdenius (1964) 62; Nehamas (2002) 62.
65
Following Mourelatos (2008a) 82ff.

189
 Why Did Parmenides Write Doxa?
in accordance with this other kind of krisis is where they went
wrong) and they distinguished opposites in accordance with their
perceptual form (ἐκρίναντο δέμας).66
It is because the mortals fail in this manner to sustain the
discrimination in Alêtheia between ‘is’ and ‘is not’, only by
means of which can one come to understand the true nature of
what-is, that the goddess describes them as ‘undiscriminating
hordes’ (ἄκριτα φῦλα, B6.7). There is an unmistakable hierarchy
between the krisis of Alêtheia and the krisis of Doxa.
As Alexander Mourelatos has emphasised, the modal language
which Parmenides uses here and elsewhere (‘may not’, οὐ χρεών)
suggests a normative necessity, a ‘right, due or proper necessity’,
rather than blind inexorability.67 In this particular case, of course,
the substantive term χρεών applies to the articulation of the
Doxastic forms itself rather than directly to intelligent agents
who may choose to act in accordance with or in defiance of such
a normative injunction.68 As such, the term indicates here an
intrasystematic normative necessity. Given the essential nature
of the Doxastic krisis, and given the interdependence of the two
enantiomorphic elements Light and Night, the attempt to posit
only one element of the distinction without the other would be an
inappropriate normative error as well as a conceptual confusion.
What, finally, of the relation between sensory perception on the
one hand and Doxa’s krisis and the theory of human cognition in
B16 on the other hand? After Theophrastus cites B16, his first
remark – apparently in reference to the fragment as a whole – is:
‘For he [sc. Parmenides] speaks of perceiving and cognising as the
same thing’ (τὸ γὰρ αἰσθάνεσθαι καὶ τὸ φρονεῖν ὡς ταὐτὸ λέγει,
Sens. 4.1). Must we assent to the assessment that ‘on any inter-
pretation this last remark is thick-headed?’69
Aristotle’s treatment of Parmenides B16 in Metaphysics
Gamma (5 1009b12–1010a3) no doubt underlies Theophrastus’
66
On this reading, then, τῶν . . . εἰσίν smoothly links μορφάς . . . ὀνομάζειν with ἀντία . . .
δέμας as a coherent progression. This, coupled with its suitability as an analysis of the
Doxastic krisis, justifies its adoption over Cornford’s ‘not so much as one of which
should be named’, (1933) 109, which, as Clark (1969) 26 observes, renders the remark
‘this is where they have gone astray’ awkwardly redundant.
67
Mourelatos (2008a) xxxi, 277–8; cf. Robbiano (2006) 74–5; Fränkel (1975a) 27.
68
For the distinction, cf. Robbiano (2006) 278. 69 Vlastos (1946) 71, n.38.

190
 4.3 Cognition, krisis, Sense-Perception
statement and could well be what motivated him to include B16 in
his On the Senses in the first place. In this part of Metaphysics
Gamma, Aristotle is surveying attitudes which, he claims, misled
past thinkers into questioning or undermining the law of non-
contradiction. In this passage (1009b12–1010a7), Aristotle criti-
cises what he represents as a widespread, ancient conception of
sense-perception. Here is the key statement (1009b12–15):
ὅλως δὲ διὰ τὸ ὑπολαμβάνειν φρόνησιν μὲν τὴν αἴσθησιν, ταύτην δ’ εἶναι ἀλλοίωσιν,
τὸ φαινόμενον κατὰ τὴν αἴσθησιν ἐξ ἀνάγκης ἀληθὲς εἶναί φασιν.

In general, it is because they assume that sense-perception is cognition, and that

it is an alteration, that they say that what appears to our senses must be true.

For Aristotle, then, this ancient conception took sense-perception

to be knowledge and drew the objectionable conclusion that all
impressions derived from sense-perception must be true. Aristotle
proceeds to cite or paraphrase statements from Empedocles,
Parmenides, Anaxagoras and Homer in order to implicate them
in this alleged conception. None of the statements which Aristotle
mobilises, however, contain any language which clearly signifies
sense-perception. Aristotle cites Empedocles B106 and B108 (for
the texts, see Chapter 6.2) to show that, according to Empedocles,
changes in our physical disposition cause changes in our knowl-
edge (1009b17–21). He then cites Parmenides B16, remarking
only: ‘Parmenides too proclaims in the same way’ (1009b21–5).
A Homeric description of a concussed Hector as ‘other-thinking’
or ‘other-cognising’ (ἀλλοφρονέοντα) is tenuously taken to imply
the view that even the delirious (literally, ‘those-who-cognise-
besides’ or ‘think-besides’: τοὺς παραφρονοῦντας) still cognise
(φρονοῦντας) and, therefore, to imply the same conception of
sense-perception. And, if Homer thinks that both the delirious
and the non-delirious cognise, then, according to him, things
must be both one way and the other way simultaneously (1009b
28–33), so that Homer too undermines the law of non-
contradiction. Anaxagoras is implicated in this same conception
simply on the basis of a reported remark to his companions
that things would be for them such as they take them to be
(1009b25–8).

191
 Why Did Parmenides Write Doxa?
Most strikingly, Aristotle identifies the assumption that ‘the
things-which-are’ (ta onta) are, exclusively, perceptible (aisthêta)
as what propelled all these thinkers to adopt the conception of sense-
perception which he criticises (1010a1–3). In the passage cited
above, Aristotle only states that these early thinkers assumed that
sense perception is cognition (the definite article in τὴν αἴσθησιν
indicates that φρόνησιν is the predicate). And yet, Aristotle’s diag-
nosis here, that these thinkers recognised only perceptible objects in
their ontology, suggests that they recognised no form of knowledge
or thinking other than sense-perception and, therefore, committed
themselves to the view that all knowledge is sensory knowledge and
that all impressions are true, thereby leaving no room for error. And
indeed, when Aristotle reprises and develops this discussion in
On the Soul (3.3), he has the early thinkers identify cognition or
knowledge and sense-perception: ‘Indeed the ancients say that
cognising and perceiving are the same thing’ (οἵ γε ἀρχαῖοι τὸ
φρονεῖν καὶ τὸ αἰσθάνεσθαι ταὐτὸν εἶναί φασιν, 427a21–2). His
concern here is precisely that ‘the ancients’ thus left no conceptual
space for error: ‘yet they should have accounted at the same time
also for being mistaken’ (427a29−b1). Aristotle himself, after all,
maintains that ‘sense-perception of the appropriate objects of sense
is always true’ (ἡ μὲν γὰρ αἴσθησις τῶν ἰδίων ἀεὶ ἀληθής, 427b
11–12). It is in part for this reason that he is concerned to demarcate
perception from other mental activities, in which error can reside
(427b6–11):
ὅτι μὲν οὖν οὐ ταὐτόν ἐστι τὸ αἰσθάνεσθαι καὶ τὸ φρονεῖν, φανερόν· τοῦ μὲν γὰρ
πᾶσι μέτεστι, τοῦ δὲ ὀλίγοις τῶν ζῴων. ἀλλ’ οὐδὲ τὸ νοεῖν, ἐν ᾧ ἐστι τὸ ὀρθῶς καὶ
τὸ μὴ ὀρθῶς – τὸ μὲν ὀρθῶς φρόνησις καὶ ἐπιστήμη καὶ δόξα ἀληθής, τὸ δὲ μὴ ὀρθῶς
τἀναντία τούτων – οὐδὲ τοῦτό ἐστι ταὐτὸ τῷ αἰσθάνεσθαι.

It is evident that perceiving and cognising are not the same thing – for all
animals have a share of the former, but only a few of the latter. And neither is
the activity of the mind, in which there is ‘correctly’ and ‘incorrectly’ –
‘correctly’ being cognition and knowledge and true opinion, and ‘incor-
rectly’ being the opposite of these things – neither is this the same thing as
perceiving.

Aristotle’s contention in Metaphysics Gamma 5 and On the Soul

3.3, then, is that his early predecessors failed to make this sort of

192
 4.3 Cognition, krisis, Sense-Perception
demarcation. He aligns Parmenides, on the basis of B16, with an
alleged ancient failure to think about cognition and objects of
cognition other than in terms of sense-perception, so that our
cognitive engagement with the world becomes nothing more
than the direct, inerrant function of changes in our physical dis-
position. It is especially difficult to square Aristotle’s ascription to
Parmenides of the view that the things-which-are are exclusively
perceptible with his own earlier distinction between, on the one
hand, Parmenides’ concept of what-is (to on) as One and, on the
other hand, Parmenides’ concept of plurality as based on sense-
perception (τὸ ἓν μὲν κατὰ τὸν λόγον πλείω δὲ κατὰ τὴν αἴσθησιν,
Metaph. Α.5 986b26–34). Elsewhere, Aristotle even appears to
associate Parmenides with those who ‘going beyond sense-
perception (aisthêsis) and disregarding it on the grounds that one
should follow reason (logos), declared that the All is one and
unmoved’ (GC 1.8 325a13ff).70 Perhaps, then, when Aristotle
includes Parmenides among those who admitted only perceptible
objects in their ontology, he has in mind strictly the author of
Doxa.
Parmenides, as we noted, uses cognitive rather than perceptual
language in B16. Were Aristotle and Theophrastus, then, simply
wrong to see him as assimilating sense-perception and cognition?
Provided that, again, when interpreting B16 they mean by
‘Parmenides’ the author of Doxa – and both Aristotle and
Theophrastus were sensitive to the distinction between the two
parts of the poem71 – the answer would seem to be more compli-
cated. First, the lost passage that dealt with the ability of corpses to
experience cold and silence, and their inability to experience sound,
heat and warmth, indicates that the theory covers also what falls
under ‘sense-perception’. Second, the two sensible elements of
Doxa (Light/Night, Hot/Cold, Rare/Dense, Lightweight/Heavy,
etc.) – discriminated in accordance with their sensible aspects
(ἐκρίναντο δέμας, B8.55) – are both that by which we humans
cognise and what we cognise. The things by which and of which
humans qua humans are fundamentally aware are therefore, in
70
Laks (1999) 257–8 observes these tensions. On the related Peripatetic statements – or
complaints – that Parmenides ‘identified soul and mind’, see Ch. 5.2.
71
See Ch. 4.5; cf. Mansfeld (1996) 166, 174–5; Coxon (2009) 381.

193
 Why Did Parmenides Write Doxa?
a way, perceptual. Finally, the goddess associates an uncritical use
of sensory experiences with the mental habits of the mortals whom
she criticises in Alêtheia (B7.3–6):
μηδέ σ’ ἔθος πολύπειρον ὁδὸν κατὰ τήνδε βιάσθω, 3
νωμᾶν ἄσκοπον ὄμμα καὶ ἠχήεσσαν ἀκουήν
καὶ γλῶσσαν, κρῖναι δὲ λόγῳ πολύδηριν ἔλεγχον 5
ἐξ ἐμέθεν ῥηθέντα.
Nor let habit born of long experience force you along this way,
To wield an aimless eye and echoing ear
And tongue, but judge by reasoning the much-contested examination
Spoken by me.
The references to ‘eye’, ‘ear’ and ‘tongue’ clearly connote sensory
experiences. Since the goddess’s warnings here elaborate her injunc-
tion to withstand the mortal ‘habit’ (ἔθος), however, she may well
target more broadly also the assumptions and categories which
structure the mortal world view. Her reference to the ‘tongue’
(γλῶσσαν), in particular, could suggest a language, which is struc-
tured by terms such as ‘perish’ and ‘locomote’ (we will discuss the
importance of mortal ‘naming’ in Chapter 4.4).72 The goddess
associates an uncritical and unreflective acquiescence in sensory
experiences with the mortals’ misguided world view. Her description
of the wayward mortals as ‘at once deaf and blind’ (κωφοὶ ὁμῶς
τυφλοί τε, B6.7) plays with this theme, recasting the mortals’ unre-
flective use of sight and hearing as a kind of figurative blindness and
deafness. Parmenides creatively appropriates here a common Greek
topos of associating the loss of sensory capacities with the attainment
of extraordinary wisdom or skill and of recasting literal blindness
as figurative sight. We may think of the blind poet Demodocus
(the Muse ‘deprived him of his eyes, but gave him sweet song’,
Od. 8.63–4) and of the blind Teiresias whose mind (phrenes) and
understanding (noos) were steadfast even in death (Od. 10.493; cf.
Od. 12.267). In her description of the wayward mortals as deaf and
blind, the goddess gives us the other side of the coin, recasting literal
sight and hearing as figurative blindness and deafness. We find
similar rhetorical moves in Epicharmus’ adage that ‘the mind
72
See Mansfeld (1999), 331–3. But I cannot accept Mansfeld’s negative contention –
followed by Robbiano (2006) 97–8; Bredlow (2011) 232–3 – that with ‘eye’, ‘ear’ and
‘tongue’ Parmenides in no way connotes sensory experiences.

194
 4.3 Cognition, krisis, Sense-Perception
(nous) sees and the mind hears; everything else is deaf and blind’
(κωφὰ καὶ τυφλά, DK23 B12) and in Prometheus’ remarks on the
state of mortals prior to his blessings: ‘seeing they saw in vein,
hearing they did not hearken’ (βλέποντες ἔβλεπον μάτην, κλύοντες
οὐκ ἤκουον, Aeschylus, Pr. 447–8).73
Now, none of these points mean that Parmenides categorically
denigrates sensory experiences. Perceptual observations permeate
Doxa’s accounts and are essential for our understanding of the
origins, natures and behaviours of Doxastic things and processes.
But perceptual observations are not the means by which one arrives
at knowledge of what-is, of the unshaken heart of reality.74 If we
acquiesce in our sensory experiences uncritically and unreflectively
(if we apply an ‘aimless eye’ or an ‘echoing ear’), then such
experiences will fortify further the entrenched habits of mortal
thinking and lead us away from such knowledge. Again, none of
the points made above imply that Parmenides himself elaborated
a theory, or even clearly isolated the category, of ‘sense-perception’
as such.75 We need not even interpret Theophrastus (or Aristotle, for
that matter) as asserting otherwise, the title of his On the Senses
notwithstanding. Theophrastus’ remark, ‘for he speaks of perceiv-
ing and cognising as the same thing’ (Sens. 4.1), can be taken to
indicate his own exegetical analysis of B16: Parmenides does not
demarcate here what we, as Peripatetics, would distinguish as (i)
the perceptual activities of the soul and (ii) the cognitive activities of
that distinct part of the soul, the mind.76
Let us take stock. Parmenides identifies as the essence and con-
ceptual basis of Doxa’s cosmology a demarcation, or krisis, between

73
See also Herac. DK22 B34. Sophocles’ Oedipus Tyrranus is the locus classicus for this
conceptual dynamic of sight and blindness, see Goldhill (1986). Note also the blind seer
Euenius of Apollonia in Hdt. 9.92–4. For blind seers and poets, see further Ustinova
(2009) 168–76; Petridou (2015) 214–16. On the complex and diverse associations
between sight, blindness, insight, ignorance, transgression and punishment in Greek
literature and art, see Coo (2016). For divination as figurative ‘sight’, cf. also Aesch. Pr.
499 (‘I opened their eyes’: ἐξωμμάτωσα). For figurative ‘blindness’ as the inability to
divine, see e.g. Plato Tim. 72b9-c1 (τυφλόν) and, echoing Epicharmus’ adage, Plut. de
Soll. Animal. 975b11 (κωφὰ πάντα καὶ τυφλὰ τῆς προνοίας).
74
On this point, see similarly Frère (2011), 140, 144–5; cf. Laks (1999) 261–2.
75
Something which Mansfeld (1999) 342–4 rightly questions.
76
For λέγειν ὡς in the sense of ‘regard as’, cf. Aesch. Ag. 672 (λέγουσιν ἡμᾶς ὡς ὀλωλότας).
See along similar lines Verdenius (1964) 16; cf. also Mansfeld (1964) 172; Laks
(1999) 256.

195
 Why Did Parmenides Write Doxa?
two enantiomorphic elements with respect to their perceptual features
(δέμας). By a brute, physiological necessity, human cognition func-
tions in accordance with this very same krisis. Conversely, Alêtheia’s
argumentation and its concept of the homogeneous what-is is framed
by the qualitatively different krisis between ‘is’ and ‘is not’. We could
not maintain and think in terms of both sorts of demarcations (kriseis)
at one and the same time.
The human mind is analysed in terms of mixtures between Hot
and Cold and in accordance with the principle of like-by-like.
Cognition of, and in terms of, the two enantiomorphic elements
occurs in the human mind passively and ineluctably. The tendency
of mortals to think in accordance with the Doxastic krisis, to
recognise Doxastic things and form first-order judgements with
respect to them (e.g. ‘this person moves’, ‘this corpse is cold’), is
a function of the very constitution of their cognitive apparatus.
Consisting of Hot and Cold, we think Hot and Cold and in terms of
Hot and Cold (Light and Night, Rare and Dense, Lightweight and
Heavy, etc.). ‘Perception’ in Doxa is thus not merely of sensible
qualities narrowly construed, but of cosmic principles and ele-
ments. In perceiving trees and humans, what we perceive are the
elements of which they consist; one does not simply feel warm,
one perceives mixtures of Hot and Cold.77
In B16, Parmenides thus identifies, and couches in a cognitive
theory, what at bottom may seem a rather plausible insight: the
human mind cannot avoid experiencing, recognising and generating
judgements with regard to a multiplicity of heterogeneous things,
which display such perceptual processes as generation and locomo-
tion. Parmenides could no more simply discard such judgements
than he could permanently switch off such experiences.

4.4 Ineluctability and Volition in the Mortal’s

Relation to Doxa
It is, then, an ineluctable necessity for the human mind to think in
terms of the same contrasts between sensory opposites which
underpin also the cosmology of Doxa. But what follows from

77
For this last point, cf. von Fritz (1974) 50–1.

196
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
this? What precisely is up to the mortal and what is not up to him?
No simple inference can be made from the unavoidability of
Doxastic cognition for the human mind to Parmenides’ motiva-
tions for the composition and inclusion of Doxa. Our evidence
presents us with a complex mixture of implications. The goddess
not only emphasises the epistemic ineluctability of Doxastic
thought for all humans, but also encourages the kouros precisely
to keep his guard from such thought. I do not suggest that
Parmenides’ conception of the ineluctability of Doxastic cognition
for the mortal qua mortal led him simply to acquiesce in such
cognition uncritically and unqualifiedly. I suggest, rather, that this
conception best explains why and how Parmenides both develops
and advances his theories of Doxastic things as accounts of
Doxastic things and keeps those theories at arm’s length. Put
differently, it best explains why the kouros must learn also mortal
beliefs (βροτῶν δόξας), but after and in the light of Alêtheia.
To establish this explanation, we must explore further the ways
in which the mortal’s relation to Doxastic thought is and is not up
to him.
The recurrent image of wandering in Parmenides further accent-
uates the problem. The human mind is associated with
a ‘wandering’ frame: ‘As on each occasion is the mixture of much-
wandering limbs (μελέων πολυπλάγκτων), so is mind present to
humans’ (B16.1–2). This image closely recalls the wandering
mind of the wandering mortals berated by the goddess (B6.4–6):
ἣν δὴ βροτοὶ εἰδότες οὐδὲν
πλάζονται, δίκρανοι· ἀμηχανίη γὰρ ἐν αὐτῶν 5
στήθεσιν ἰθύνει πλαγκτὸν νόον.
[The way] on which indeed mortals, knowing nothing,
Wander, double-headed; for helplessness in their
Breasts directs their wandering mind.
We may recall also the goddess’s remark that the mortals have
‘gone astray’ (πεπλανημένοι εἰσίν, B8.54).78 As Mourelatos
(2008a) has taught us, we must not dismiss such verbal echoes
78
Cf. Od. 20.195: ‘the gods bring to misery far-wandering men’ (πολυπλάγκτους
ἀνθρώπους). The connection between the images of wandering in B6, B8.54 and B16
has long been observed, see e.g. Reinhardt (1974) 307; Fränkel (1975a) 17–18; Vlastos
(1946) 69; Verdenius (1964) 7–8; Long (1996) 147; Torgerson (2006) 28.

197
 Why Did Parmenides Write Doxa?
as accidental. In this case, the recurrent image of wandering impels
us to ask how, and in what sense, the kouros can avoid the mortal
wandering criticised by the goddess. After all, it is precisely the
same kind of inconstant, helpless wandering cognition, which B16
pronounces a physiological fact for the human mind, that char-
acterises in Alêtheia also the human mind whose example the
goddess exhorts the kouros, and vicariously his audience, to
avoid. The goddess nowhere qualifies her recurrent references to
‘mortals’ (B6.4; B1.30; B8.39, 51–5; B16.2–3; B19.3). Her
remarks in B6.4–6 refer to mortals in quite a general way and
not to some very particular, small subset of mortals. This is con-
firmed by her emphasis in B16 that her analysis of the human
mind, as determined by the varying mixtures of its much-
wandering limbs, pertains universally to every human agent on
every occasion of human thinking: ‘as on each occasion (ὡς γὰρ
ἑκάστοτ’) . . . in all humans and in each’ (καὶ πᾶσιν καὶ παντί,
B16.1, 4).79 In what sense, then, can the kouros avoid the error of
human ‘wandering’? To what extent and in what respect is the
mortal’s relation to Doxastic cognition up to him?
In Alêtheia, the goddess denies that such features as coming-to-
be, perishing, locomotion and heterogeneous differentiation char-
acterise what-is, the unshaken heart of reality (B8.3–7, 21
(γένεσις . . . ὄλεθρος), 26–8, 41). But such phenomena and pro-
cesses are built into the very idea of ‘mortals’. The notion of
mortals involves processes and properties like coming-to-be and
heterogeneous differentiation no less than any supposition about
such processes and properties that may be affirmed by mortals.
One thing which we cannot say, then, is that the mortals make
a deliberate and avoidable choice to experience and recognise
a heterogeneous plurality of things, which come into being and
perish. The processes of differentiation, coming-into-being and

79
Parmenides’ language in B6 echoes remarks in Theognis, again on mortals in general
(139–41): ‘ . . . the constraints of grievous helplessness (πείρατ’ ἀμηχανίης) hold one
back. But we humans have vein thoughts, knowing nothing (μάταια νομίζομεν, εἰδότες
οὐδέν)’; for the reminiscence, see Torgerson (2006) 33. For the view that B6 refers to
mortals in a general way, see further Reinhardt (1974) 301; Jaeger (1947) 101; Guthrie
(1965) 23–4; Montiglio (2000) 149; Nehamas (2002) 56 (arguing persuasively against
identifying Heraclitus as the goddess’s target), contra e.g. Nietzsche (1974) 122; Coxon
(2009) 300, 381–2; Dueso (2011) 282; Frère (2011); Miller (2011) 45.

198
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
perishing are logically prior to the very possibility of mortals
choosing to do anything at all. Nor, moreover, could the mortals’
deliberate and avoidable error be a choice to think in terms of
Doxa’s principles or to form first-order judgements over others
(e.g. ‘this corpse is cold and not warm’) with respect to the many
things which they experience and recognise. As we saw,
Parmenides’ theory of human cognition renders not only the
experience and recognition of heterogeneous plural things, but
also the adoption and generation of judgements with respect to
them, a passive function of a physiological apparatus that char-
acterises each and every mortal agent.
We must address at this point an objection which would no
doubt be raised against this last claim. The objection runs as
follows. Since the theory of human cognition adumbrated in B16
comes from Doxa, it is deceptive and therefore should not worry
us in the first place regarding its implication that mortal cognitive
agents must think in terms of Doxa’s principles and generate and
maintain judgements with regard to heterogeneous things.80
We can begin to address this objection by taking a closer look at
the goddess’s characterisation of Doxa as ‘deceptive’ and at some
of her other programmatic comments with regard to the cosmol-
ogy. The goddess introduces Doxa pointedly after, and in the light
of, the conclusion of her ontological arguments in B8 concerning
the properties of what-is (B8.50–2):
ἐν τῷ σοι παύω πιστὸν λόγον ἠδὲ νόημα 50
ἀμφὶς ἀληθείης· δόξας δ’ ἀπὸ τοῦδε βροτείας
μάνθανε κόσμον ἐμῶν ἐπέων ἀπατηλὸν ἀκούων. 52

Here I conclude for you the trustworthy account and understanding

Concerning true reality; from this point on learn mortal opinions,
Hearing the deceptive order of my words.
The deceptiveness of the goddess’s cosmological account
(ἀπατηλόν), then, is contrasted, not with alternative accounts of
the same things (mortals, stars, etc.), but with the account of what-
is in Alêtheia. Through this contrast, the goddess implies that,
having learned the account in Alêtheia, the kouros is now in
a position to recognise the ensuing deceptiveness. This again
80
For examples of such attitudes to B16, see Ch. 5.1.

199
 Why Did Parmenides Write Doxa?
indicates that the following cosmological theories are deceptive,
not in their internal details, e.g. in the way in which Doxa’s
embryology describes and explains embryos – for why should
the kouros be placed any better to recognise that sort of deceptive-
ness? – but rather in their potential to mislead wayward mortals,
who are ignorant of the account of what-is adumbrated in Alêtheia,
into mistaking them for accounts which get at and disclose the
nature of ultimate reality. As we shall presently see in detail,
mortals are misled in just this way when, in their ignorance of
‘the unshaken heart of reality’, they think that categories like
coming-to-be, perishing and locomotion capture reality
(πεποιθότες εἶναι ἀληθῆ, B8.39). As Nehamas puts it: ‘The Doxa
is deceptive not because of its specific contents but because of the
way its contents have been (and are likely to be) interpreted.’81
Again, it is because Doxa’s theories are essentially incapable of
disclosing the unshaken heart of reality (alêtheiês, B1.29) – i.e.
what-is and its properties – that they can be said to lack true or real
trust (pistis alêthês, B1.30). In relation to this lack of ‘real trust’ in
Doxa’s accounts, we may also advance a further consideration.
The changeable and untidy things and processes which constitute
the proper objects of the cosmological inquiries in Doxa simply do
not allow for the same sort of firm and steadfast apprehension
which is afforded by the markedly distinct, deductive arguments in
Alêtheia concerning the nature and properties of the unshaken
heart of reality. The cosmological inquiries into the origins, nat-
ures and behaviours of, for example, stars and mortals do not lend
themselves in the same way to the sheer, pure reliability and
steadfastness of apprehension to which the deductive study of
what-is does lend itself.82
Neither in her remark on the deceptiveness of Doxa nor in her
comment on ‘real trust’ does the goddess undermine the status of
Doxa’s cosmological accounts as accounts of Doxastic things and
processes. In fact, the goddess confirms this status in another
important, programmatic statement. As we already had occasion
to note (Chapter 4.1), the goddess promises the kouros, thrice
81
Nehamas (2002) 59; cf. similarly Johansen (2016) 20.
82
With this last point concerning the absence of πίστις ἀληθής in Doxa’s accounts, I follow
Palmer (2009) 92, 167–75.

200
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
within five lines, knowledge of Doxastic things and processes:
‘you will know . . . you will learn . . . you will know’ (εἴσῃ . . .
πεύσῃ . . . εἰδήσεις, B10). The goddess tells the kouros that he will
come to know and understand the nature of aether, the celestial
bodies, the functions of the sun and their source, the nature,
motions and functions of the moon, the source of the heaven,
and how necessity constrained it to hold the limits of the stars.83
The programmatic employment of epistemic vocabulary with
respect to Doxastic things reflects the assumption that accounts
pertaining to such things can describe them correctly or incorrectly
and, furthermore, that the accounts advanced in Doxa do so
correctly.84 The cosmology in the second part of the poem is
intended to expound correctly the origins, natures and behaviours
of the multiple and heterogeneous things and processes that serve
as its objects of inquiry by tracing those things and processes to
(interactions between) the two constitutive principles which
underlie them. Thus, for example, the proposition that the
moon’s light is a reflection of the sun’s (B14–15) accounts for
the moon and the source of its light correctly, in a way that
competing accounts, e.g. the proposition that the moon is a ring
of fire (Anaximander, DK12 A11.5) or an incandescent disk of air
(Anaximenes, DK13 A7.4), do not. B10 thus corroborates the
foregoing interpretation of the deceptiveness of Doxa and con-
firms what is already suggested independently by the comprehen-
sive scope, the systematic nature and the detail of Parmenides’
cosmological inquiries (see Chapter 4.1) and what is generally
assumed in the ancient doxography: Doxa represents Parmenides’
own views of Doxastic things. As the late ancient commentator
John Philoponus rightly insists: ‘for also in the sections pertaining
to opinion (ἐν τοῖς πρὸς δόξαν), he wrote what seemed the case to
him’ (τὰ αὐτῷ δοκοῦντα ἔγραφεν, in Ph. 16.22.6–7 = Coxon
(2009) Testim. 193).
Let us consider, lastly, the goddess’s final introductory remarks
on Doxa (B8.60–1):
83
From DK onwards, most editors consider B10 part of the introduction to Doxa. Laks
(2003) 17, n.32 rightly observes that, equally, it could have followed B1 as part of the
goddess’s initial programmatic introduction of both parts of the poem.
84
See similarly Vlastos (2008) 376–7; cf. Guthrie (1965) 51; Palmer (2009) 180.

201
 Why Did Parmenides Write Doxa?
τόν σοι ἐγὼ διάκοσμον ἐοικότα πάντα φατίζω,
ὡς οὐ μή ποτέ τίς σε βροτῶν γνώμη παρελάσσῃ.
This ordering of things, eoikota and entire [or: entirely eoikota], I relate to you,
So that no judgement of mortals will ever overtake you.
The term which I rendered ‘overtake’ (παρελάσσῃ) derives from the
Homeric vocabulary for chariot races. It signifies, not merely the
notion of passing by, but of driving past one in a race.85
The goddess promises the kouros that no mortal judgement will
‘outstrip’ him after he learns the cosmology which follows.86
Doxa’s cosmology endows the kouros with a systematic knowl-
edge and understanding of every aspect of the changing and
heterogeneous plurality in which mortal life takes place,
spanning, in its comprehensive scope, universal cosmology, cos-
mogony, astronomy, geography, theogony, anthropogony, repro-
duction, embryology, cognition, etc. (recall Chapter 4.1). It is the
best possible, unsurpassable account of all such subjects of
inquiry. More difficult is the goddess’s description of Doxa’s
cosmic order (διάκοσμον) as ‘eoikota’ or ‘eoikota in its entirety’
(ἐοικότα πάντα). The adjective is tantalisingly ambiguous and
has been variously taken to mean, among other things, ‘accu-
rate’, ‘appropriate’, ‘fitting’, ‘plausible’ and ‘specious’. After
a comprehensive discussion, Bryan concludes that, in our analy-
sis of this term, we must allow for the interrelation of different
senses on different levels of interpretation.87 We can accommo-
date this semantic insight concerning eoikota and relate the term
to the interpretation of Doxa which we established above on the
basis of other evidence. Doxa’s cosmology can be entirely accu-
rate and appropriate, insofar as it correctly expounds the natures
of Doxastic things and affords one an understanding of such
things. But it is also misleading or treacherous in its appearance
precisely in the same respect that it is ‘deceptive’, i.e. insofar as it
is liable to be mistaken by ignorant mortals for an account which
gets at and discloses the nature of fundamental reality.

85
Cf. Il. 23.382 (παρέλασσ’), 638 (παρήλασαν).
86
I will suggest in Ch. 4.5 that, by ‘judgement of mortals’, the goddess refers here, not to
any judgement formed by a mortal, but to a kind of judgement which is distinctive of
mortal cognition.
87
Bryan (2012) 58–113.

202
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
In sum, we can meet and defuse the objection that, since the
theory of human cognition in B16 comes from Doxa, it is ‘decep-
tive’ and, therefore, should have no bearing on our understanding
of Parmenides as a mortal cognitive agent. This theory of human
cognition is, in fact, one component of a comprehensive and
systematic cosmology, which is intended to endow the kouros
with knowledge and understanding of the origins, natures and
behaviours of Doxastic things and processes. Above all, however,
the objection is vacuous because the theory of mortal cognition
advanced in B16 is no more or less ‘deceptive’ than the very
notion of Parmenides as a mortal cognitive agent (and, therefore,
as a thing which essentially involves coming-into-being, perish-
ing, locomotion and internal differentiation), to whom the goddess
speaks as a mortal cognitive agent, and whose knowledge and
beliefs we are trying to explain.
We thus return to the question with which we began this section:
in what sense can the kouros avoid the error of human wandering
and human cognition? The human activity of ‘naming’ discloses
an illuminating sense in which the relation of the criticised mortals
to Doxa involves an active and avoidable act of assent over and
above the inexorable deliverances of physiological mechanisms.
When introducing Doxa, the goddess states that the mortals
resolved to name the two Doxastic elements (κατέθεντο . . .
ὀνομάζειν) and she represents this action as an avoidable error
(‘this is where they have gone astray’, B8.53–4). Furthermore,
she concludes Doxa (see Simplicius, in Cael. 7.558.8) with the
following words (B19):
οὕτω τοι κατὰ δόξαν ἔφυ τάδε καί νυν ἔασι 1
καὶ μετέπειτ’ ἀπὸ τοῦδε τελευτήσουσι τραφέντα·
τοῖς δ’ ὄνομ’ ἄνθρωποι κατέθεντ’ ἐπίσημον ἑκάστῳ. 3

In this way, then, in accordance with belief, all these things grew and now are
And hereafter from this point they will end once they received their sustenance;
And upon them humans laid down a name, a mark to each one.
The goddess concludes the cosmology with a summary of suppo-
sitions, which are held with regard to Doxastic things, and with the
closely related assertion that mortals have laid down a name
distinguishing each one. ‘Naming’ the two Doxastic elements

203
 Why Did Parmenides Write Doxa?
and each Doxastic thing is, then, identified as the fundamental
mortal error (B8.54) and associated with an index of Doxastic
suppositions (B19). It is clear, therefore, that with the term ‘nam-
ing’ (onomazein) Parmenides indicates not the noncommittal
labelling of an item, but something stronger and more technical.
When glossing Parmenides’ word onomazein with the English
term ‘naming’ we must thus always have this stronger sense in
mind – whatever precisely it turns out to be. We should note,
moreover, that, while Parmenides’ peculiar notion of naming is
distinctive, the very strength of his usage is not itself idiosyncratic.
Greek authors often approach names as involved and ideally
illuminating descriptions of their referents, rather than mere
labels. Hesiod, for example, explains that the Titans were so
named because they strained or exerted (titainontas) themselves
in their wickedness (Th. 207–10; cf. also Th. 144–5, 234–5,
281–3). Heraclitus also makes much of names. For him, names
are always both significant and deficient: if properly interpreted,
they yield some insight into the natures of their referents, but they
never properly manage to express the complex and unified natures
of the things they name. Thus, for example, the name of the bow
gives us an insight only into one of the two opposites whose
collaboration and unity is manifest in the bow: ‘The name of the
bow is life (bios), but its work is death’ (DK22 B48; cf. B32,
B67).88
What, then, are the particular implications of mortal ‘naming’
for Parmenides? The following passage from Alêtheia helps elu-
cidate the point (B8.36b-41):
οὐδὲν γὰρ <ἢ> ἔστιν ἢ ἔσται 36
ἄλλο πάρεξ89 τοῦ ἐόντος, ἐπεὶ τό γε Μοῖρ’ ἐπέδησεν
οὖλον ἀκίνητόν τ’ ἔμεναι· τῷ πάντ’ ὄνομ’ ἔσται,

88
For the general Greek conception of a name as ‘an appropriate and illuminating
description’, cf. Mourelatos (2008a) 183–7. Mourelatos (2008a) 221 observes that
Parmenides’ goddess employs ‘to name’ repeatedly ‘with interpretive-epistemic
force’. On the significance of names for Hesiod, see also Clay (2003) 160–1.
On Heraclitus, see further Kirk (1962) 116–22; Rowett (2013) 180–7; Tor (2016) 103–4.
89
‘Besides’, ‘except’; see LSJ. sv. παρέκ. I know of no parallel for πάρεξ in the sense
‘(different) from’ affirmed for this passage by de Rijk (1983) 43. Tarán (1965) 128–9
compellingly defends this text of B8.36–7 against Coxon’s οὐδὲ χρόνος ἔστιν κτλ, which
is intrinsically baffling and incongruous to the surrounding context; cf. also Palmer
(2009) 385.

204
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
ὅσσα βροτοὶ κατέθεντο πεποιθότες εἶναι ἀληθῆ,
γίγνεσθαί τε καὶ ὄλλυσθαι, εἶναί τε καὶ οὐχί, 40
καὶ τόπον ἀλλάσσειν διά τε χρόα φανὸν ἀμείβειν.
For nothing else either is or will be
Beside what-is, for Fate shackled it
Τo be whole and unchanging. All things will be its name,
Αs many as mortals laid down, believing them to be real,
Both coming to be and perishing, both being and not-being,90
And altering place and exchanging bright colour.
We will return to this passage, and ask how precisely we might
interpret the goddess’s statement that ‘nothing else either is or will
be beside what-is’, when we come to explore the ontological
question in Chapter 5.5. We can already temper in passing, how-
ever, any inclination to infer from this statement that Doxastic
things – including mortals – are categorically non-existent illu-
sions. Such an inference would render the passage incoherent.
If Doxastic things were simply non-existent, then there would be
nothing there to generate or experience the ‘illusion’ of such
processes as changing place or perishing, let alone to name such
things. Our focus at this juncture, however, is on the mortal
activity of ‘naming’ and on the way in which the goddess relates
it here to the true nature of ultimate reality.
Scholars traditionally assumed that the reading of the Greek text
of B8.38b given here (τῷ πάντ’ ὄνομ’ ἔσται) required the transla-
tion: ‘therefore, all those things will be a (mere) name’ (with τῷ as
an inferential particle, as in B8.25). It was then rightly objected
that this construal requires the supplement ‘mere’, for which the
Greek affords no indication.91 A much more natural rendering of
this Greek text, however, would be: ‘all things will be its name’
(interpreting τῷ as a demonstrative pronoun referring to the
90
Since both pairs in B8.40 are linked with τε καί, each pair represents a compound name,
whose two components are taken (by the mortals who are laying down these names) to
be interdependent; cf. similarly Vlastos (2008) 368, n.4; Fränkel (1975a) 34. At any rate,
εἶναί τε καὶ οὐχί must signify the compound name ‘both being and not-being’. ‘Not-
being’ alone is ‘without name’ (ἀνώνυμον, B8.17) and ‘being’ (εἶναι) could hardly have
been isolated for naming by the ignorant mortals, who, given Parmenides’ strong notion
of ‘naming’, would have thus successfully recognised what-is; cf. similarly Furley
(1973) 7, n.23.
91
E.g. Woodbury (1971) 145; Furley (1973) 7, n.22; Long (1975) 88; Mourelatos (2008a)
181 (his views in 1970); Graham (2010) 238; cf. Reinhardt (1974) 303; Cornford (1933)
101; Minar (1949) 42.

205
 Why Did Parmenides Write Doxa?
nominee, i.e. what-is; cf. τό in B8.37). This would be in keeping
with a familiar grammatical construction, as in the Homeric line:
‘let his name be Odysseus’ (τῷ δ’ Ὀδυσεὺς ὄνομ’ ἔστω, Od.
19.409; cf. h.Hom. 5.198).92 Given the translation ‘all things will
be its name’, the philosophical divergence between this construal
and the alternative and less likely reading of the Greek – τῷ πάντ’
ὀνόμασται (‘to it all things will be named’) – narrows consider-
ably. On both texts, the goddess identifies what-is as the true
referent of mortal names.93 But why is she doing so?
Importantly, the goddess’s epexegetic remark, ‘believing them to
be real’ (πεποιθότες εἶναι ἀληθῆ), renders explicit what is implicit
in her comments on mortal naming in B8.54 and B19.3.
The mortal resolution to establish names (ὄνομ’ ἔσται, | ὅσσα
βροτοὶ κατέθεντο) for such phenomena as both coming-to-be and
perishing, both being and not-being, locomotion and chromatic
change constitutes far more than noncommittal labelling.
By resolving to establish such names as real, true or genuine
(πεποιθότες εἶναι ἀληθῆ), the mortals reflect the conviction that,
e.g. ‘Hot’, ‘coming-to-be’ and ‘locomotion’ are to be embraced as
categories which can get at, reflect and disclose the true nature of
ultimate reality (alêtheiê). The mortals’ resolve to name Doxastic
things and processes in this way thus rests on and betrays ignor-
ance of what-is, which problematises such naming. Once one
recognises the homogeneous and unchanging what-is as exclu-
sively what the unshaken heart of reality comprises, one also
recognises the deceptiveness of the conviction that the sensory
contrast anchoring Doxa’s cosmology could underpin an account
which gets at and expounds, not merely the origins, natures and
behaviours of those things which constitute its proper objects
(stars, people, embryos, etc.), but the fundamental, core nature of
92
Following Kingsley (2003) 190–1; Mourelatos (2008c) 386–7.
93
The reading τῷ πάντ’ ὀνόμασται, defended by Woodbury (1971) and Vlastos (2008), is
philologically implausible. ὀνομάζειν nowhere takes a dative as the nominee without the
use of ἐπί: Woodbury’s alleged parallels (149) all involve (as conceded by Vlastos
(2008) 379; Furley (1973) 7, n.22) either ἐπί or the noun ὄνομα coupled with τίθεσθαι
identifying the dative as the nominee. Cf. the criticisms in Mourelatos (2008c) 386;
Meijer (1997) 177, n.941. More damningly, coupling the dative τῷ with a passive
perfect (ὀνόμασται) suggests a dative of agent, which – as Mourelatos (2008c) 386
observes – would yield the absurd result: ‘by it [sc. what-is] all those [sc. names] have
been named’.

206
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
reality (what-is). The goddess tells us that, in fact, what-is is the
only possible referent for categories which purport to get at and
capture the nature of ultimate reality.
Consistently, therefore, the mortal activity of naming (onoma-
zein) involves not merely the noncommittal labelling of the two
Doxastic elements (B8.54), Doxastic things (B19.3) and Doxastic
processes (B8.40–1). Rather, it involves the active, objectionable
and avoidable resolve (κατέθεντο, B8.39, 53; B19.3) to embrace
such labels as categories which, when applied in the context of
a cosmological account, get at, reflect and disclose the fundamen-
tal nature of reality (πεποιθότες εἶναι ἀληθῆ). In B9 we read:
αὐτὰρ ἐπειδὴ πάντα φάος καὶ νὺξ ὀνόμασται 1
καὶ τὰ κατὰ σφετέρας δυνάμεις ἐπὶ τοῖσί τε καὶ τοῖς,
πᾶν πλέον ἐστὶν ὁμοῦ φάεος καὶ νυκτὸς ἀφάντου. 3

But since all things have been named light and night,
And the things corresponding to their powers have
been given as names to these things and to those,
All is full of light and unseen night together.
We could not coherently take this passage to mean that, by naming
the two elements, mortals have literally generated the heteroge-
neous plurality experienced by them. It is only given and within
a context of heterogeneous plurality that mortals (essentially het-
erogeneous and differentiated things that they are) could choose to
do anything at all.94 Rather, the goddess’s dense statements here
seem to summarise her earlier, more expansive disquisitions con-
cerning mortal naming. The first line (‘But since . . . light and
night’) recapitulates her remarks on the mortals’ foundational
resolve to name the two basic, enantiomorphic elements (B8.
53–9). The second line (‘And the things . . . and to those’) recapi-
tulates her remarks on the mortals’ resolve to name such further
processes as ‘coming-to-be’ and ‘changing place’ (B8.38b-41)
and makes explicit that such processes – and indeed all events
and processes on the level of everyday phenomena – are traced
to, and explained in terms of, interactions between the two

94
See similarly Cornford (1933) 110; contrast Reinhardt (1974) 296–7, 308; Mansfeld
(1964) 215–21; Woodbury (1986) 8.

207
 Why Did Parmenides Write Doxa?
fundamental elements.95 The goddess’s remarks in B9.1–2 thus
recall and cohere with her strong use of ‘naming’ elsewhere in the
poem. The third line can be read from two perspectives. For the
informed kouros, who approaches Doxa after and in the light of
Alêtheia, the appropriate consequence to draw is that the totality of
Doxastic things – the entire heterogeneous multiplicity which
exhibits change and internal differentiations – comprises and is
traced to the two elements. For the wayward mortals, who are
ignorant of what-is and who mistake the principles and categories
of the cosmology for the principles and categories of an account
that gives us the nature of ultimate reality, the ‘all’ (pan) which is
traced to Light and Night would be truly exhaustive of all things
which are there for one to learn and to inquire about.96
In sum, Parmenides, as a mortal, is constrained to experience and
to recognise a plurality of changing and heterogeneous things.
Equally, he is also constrained to think through, and in terms of, the
ubiquitous contrast between the two fundamental sensory opposites
in all their perceptual aspects (Light/Night, Hot/Cold, etc.) and to
generate first-order judgements with regard to multiple, changing,
heterogeneous things. At the same time, the kouros is encouraged to
learn mortal beliefs precisely after and in the light of Alêtheia. To be

95
In this interpretation of how B9.1–2 relates to B8.53–9 and to B8.38b-41, I follow
Palmer (2009) 171–4.
96
Cf. B1.28b-30: ‘but it is right for you to learn [or: ‘inquire about’] all things (πάντα
πυθέσθαι), both . . . and . . . ’; B10: ‘you will know . . . you will learn . . . you will know’.
Since Parmenides describes not-being, as opposed to being, as ‘without name’
(ἀνώνυμον, B8.17), he could not have maintained a categorically pejorative attitude
towards naming as such, as noted by Vlastos (2008) 377–8, contra e.g. Loenen (1959)
39–40 with n.68; Verdenius (1949) 130; Diels (1887) 251, n.1. This point, however, in
no way undercuts the goddess’s consistently critical attitude towards mortal naming.
In B8.40, the mortals name both ‘being’ and ‘not-being’: they do not manage to name
‘being’ alone: see n.90 above; contrast Vlastos (2008) 378, inconsistently with 368, n.4.
The so-called ‘Cornford fragment’, ‘it is sole, unchanging; the All has the name
“being”’ (οἶον, ἀκίνητον τελέθει· τῷ παντὶ ὄνομ’ εἶναι, Pl. Tht. 180e1, with Cornford
(1935) 122), thus coheres with my discussion of the goddess’s attitude to mortal naming.
That said, there is no strong reason to ascribe the statement to Parmenides himself.
Plato’s follow-up comment on the statement, ‘and all the other things which Melissuses
and Parmenideses affirm in their opposition to all these people [sc. Heraclitean flux
thinkers]’, does not, given its generality and especially Plato’s ordering of the names,
suggest a verbatim quotation of Parmenides, although it could certainly explain how
Simplicius wrongly came to present the statement as such (in Ph. 9.29.18, 143.10).
Arguments for authenticity are necessarily speculative: Cornford (1935) 122–3; (1934)
94, n.1; Woodbury (1971) 148–9; Mourelatos (2008a) 185–8.

208
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
sure, by hearing the cosmology which makes up the second part of
the poem, the kouros will gain knowledge and understanding of the
origins, natures and behaviours of Doxastic things and processes
(such as, among many other things, the nature and workings of
human cognition). But, aware as he is of the true nature of ‘the
unshaken heart of reality’, the kouros will also be in a position to
recognise the ‘deceptiveness’ of the goddess’s account of Doxa. That
is, he will be able to avoid the mortals’ error of mistaking this sort of
cosmological account for an account which gets at and discloses the
nature of what-is, of ultimate reality.
Before we proceed to weave together the strands of this chapter
into an answer to the aetiological question (Chapter 4.5), we
should address the goddess’s final couplet in the proem as it has
come down to us. These two lines are especially beset by ambi-
guities and difficulties, and no translation or interpretation of them
could be remotely certain. Nonetheless, a consideration of some
ways in which we may interpret these concluding remarks will
pertain to, and help to underscore further, some of the themes
which we discussed above. After the goddess informs the kouros
that he must learn both the heart of reality and mortal beliefs, she
adds (B1.31–2):
ἀλλ’ ἔμπης καὶ ταῦτα μαθήσεαι, ὡς τὰ δοκοῦντα
χρῆν δοκίμως εἶναι διὰ παντὸς πάντα περῶντα.
But, nonetheless, you shall learn these things too,
how for the things that seem and are accepted
It was [or: would have been] right to be accep-
tably, all of them traversing across everything.
Several aspects of this translation are tentative and uncertain, and
several others require some argumentation and elucidation. What
is the reference of ‘these things too’ (καὶ ταῦτα)? How exactly
should we interpret and translate the cognate words ta dokounta
and dokimôs (here rendered respectively ‘the things that seem and
are accepted’ and ‘acceptably’)? Should we construe chrên as
a past obligation (‘how it was right’) or a past counterfactual
(‘how it would have been right’)?97

97
Also, should we indeed read περῶντα (‘all of them traversing across everything’) or περ
ὄντα (‘all of them altogether indeed being’)?

209
 Why Did Parmenides Write Doxa?
The words ‘but, nonetheless, you shall learn these things too’
come immediately after the goddess declares that Parmenides
must learn mortal beliefs ‘in which there is no real trust’. This
progression clearly indicates that ‘these things’ refers back to
mortal beliefs, which the mortal shall learn despite their lack of
‘real trust’.98
Mourelatos highlights the emphatic figura etymologica in
dokounta . . . dokimôs, and the prominence of the notion of
acceptance in dok- words (whose stem is related to the δεκ- of
δεκ(/χ)ομαι). He thus construes the dokounta as ‘accepted’ things,
rejecting entirely the sense ‘appearing’.99 Such a surgical distinc-
tion seems too neat. As Mourelatos himself concedes, ‘the two
expressions δοκεῖ μοι and φαίνεταί μοι have a tendency to converge
and mimic one another.’100 The dokounta are perhaps best thought
of as signifying, more inclusively and in a unified manner, the
things which seem to mortals and are accepted by them.101
The ‘things that seem and are accepted’ would presumably be
the things which comprise the subject matter of ‘the beliefs of
mortals’, i.e. Doxastic things and processes quite generally (stars,
people, coming-to-be, locomotion, etc.). Importantly, however,
the connotation of ‘appearing’ in the term dokounta in no way
suggests that, through this designation, Doxastic things are some-
how reduced to categorically non-existent ‘illusions’. Rather, the
term underscores that mortals (who, we must remember, them-
selves essentially comprise Doxastic things no less than anything
with which they interact) experience, register and form judgements
98
Contra Mourelatos (2008a) 209, therefore, we cannot take B1.31–2 to express the
injunction to learn some third item over and above what-is and mortal beliefs. In his
gloss, ‘you shall learn about truth and about opinions; but, nevertheless, you shall learn
this also . . . ’, Mourelatos omits the pertinent interjection ‘in which there is no real
trust’. On this point, cf. Cordero (2011) 107–8. It is independently implausible to
interpret ταῦτα as a further item to what has already been described as an exhaustive
programme of study (πάντα, B1.28b). ταῦτα can take δόξας as its antecedent.
The gender discrepancy is no obstacle, see LSJ. sv. οὗτος BII.3. This need imply
neither that ὡς . . . περῶντα expresses mortal beliefs as articulated by the mortals
themselves (as Owen (1960) 88–9 implies, and Clark (1969) 21 rightly denies) nor
that in ὡς . . . περῶντα the goddess is speaking from within the framework of Doxa
rather than commenting on it from an external perspective (as Mourelatos (2008a)
209–10, n.46 assumes when criticising Owen).
99
Mourelatos (2008a) 194–200. 100 Mourelatos (2008a) 199; cf. Kahn (1973) 393–4.
101
Cf. Clark (1969) 19–20. I am not suggesting a ‘fusion of the two senses’ (questioned by
Mourelatos (2008a) 195, n.5), but the absence of a strict demarcation between them.

210
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
about all the things and processes that appear to them, i.e. all those
things that simply present themselves to the human mind.
Furthermore, we saw above that it is by perceptual things and of
perceptual things that humans qua humans are, by physiological
necessity, aware. In that respect, we can also say that ‘the things
which seem’ appear to us in perceptual ways: they comprise sensory
objects and processes (ultimately, interactions of Light and Night)
and are registered through sensory cognitive mechanisms. This is
not to say, however, that the term dokounta itself serves to highlight
the sensory nature of our experience of ‘the things which seem and
are accepted’.
The kouros shall learn how, for the things which seem and are
accepted, ‘it was right [or: it would have been right] to be accep-
tably (dokimôs)’. The adverb dokimôs is cognate with the term
dokounta and literally means ‘acceptably’. But it bears no hint of
the concession or compromise felt in that English word and
suggests, rather, success in meeting a fixed standard applied rig-
orously. The adjective dokimos generally bears the senses ‘accep-
table’ or ‘accepted’ and hence ‘approved’, ‘reliable’, ‘notable’.
In its two classical occurrences, the adverb seems to indicate that
something was done or took place ‘genuinely’, ‘actually’ or ‘prop-
erly’, i.e. in a fully ‘acceptable’ manner: ‘I too sustain the genu-
inely (dokimôs) lamentable fate of the departed’ (Aeschylus, Pers.
547); ‘ . . . so that he himself might properly (dokimôs) become
fine and good’ (Xenophon, Cyr. 1.6.7).102
We may next distinguish two possible interpretations, depend-
ing on what we think the ‘acceptance’ amounts to here. On the first
interpretation, we will take it that ‘to be acceptably’ or ‘properly’
indicates a mode of being which is in line with the strong and
uncompromising application of the verb ‘to be’ in Alêtheia.
On this view, ‘to be acceptably’ means to be the true and funda-
mental reality. Call this the ‘absolute’ interpretation of dokimôs.
Within the absolute interpretation, there are then two further viable

102
The syntax is ambiguous in both cases. Alternatively, δοκίμως could be qualifying, with
the same force, the verb in the Aeschylus passage (‘properly sustain’) and the adjec-
tives in the Xenophon passage (‘properly fine and good’). On δοκίμως, see further
Owen (1960) 86; Tarán (1965) 212–13; Mourelatos (2008a) 200, 204; Coxon (2009)
285; Lesher (1984) 19; Curd (1998) 21–2.

211
 Why Did Parmenides Write Doxa?
options, which amount to much the same philosophical result.
The statement concerning the ‘things that seem and are accepted’
could be either a counterfactual spoken by the goddess from out-
side the framework of Doxa (‘how for the things that seem and are
accepted it would have been right to be acceptably’), or a past
obligation spoken by the goddess from within the perspective of
Doxa (‘how [sc. from the perspective of mortals ignorant of
Alêtheia] it was in fact right for the things that seem and are
accepted to be acceptably’).103 The goddess cannot, on the abso-
lute interpretation of dokimôs, be stating a past obligation from
outside the framework of Doxa.104 That is, she cannot be stating
that – even from the perspective of one who had acquired the
knowledge of Alêtheia – it was in fact right for Doxastic things to
be, in the full and unqualified sense that what-is is. This would
amount to confusing Doxastic things with what-is and beliefs
about them with knowledge of it, incompatibly with B1.30 (‘in
which there is no real trust’) and quite generally the deductions of
the properties of what-is in B8. It is essential to humans (trees,
pebbles, etc.) that they are generated, perishable, mobile and
heterogeneous. Such things, then, precisely do not instantiate the
mode of being of what-is.105 Nor can the goddess’s statement be
a counterfactual spoken from within Doxa, for this would imply
that the ignorant mortals are in a position to deny, correctly, that
Doxastic things can be said to be in the same way that what-is is.
In sum, if the goddess’s statement is a counterfactual spoken from
outside Doxa, the kouros will learn how (sc. were it not for the true
nature of what-is as established in Alêtheia), it would have been
right for the things that appear to mortals and are accepted by them
to be genuinely. If it is a past obligation spoken from within Doxa,
the kouros will learn how (sc. from the perspective of mortals
ignorant of Alêtheia) it was in fact right for the things that appear
to mortals and are accepted by them to be genuinely. In either case,
the goddess’s statement (‘how . . . acceptably’) specifies, from the
goddess’s perspective, how one ignorant of Alêtheia would be

103
Nehamas (2002) 58, n.49 identifies and allows both of these options.
104
This appears to be how Thanassas (2011) 294, n.8 reads these lines.
105
See further Ch. 5.5.1.

212
 4.4 Ineluctability and Volition in the Mortal’s Relation to Doxa
inclined to interpret Doxa, confusing the things that appear to and
are accepted by mortals with what genuinely is.
On an alternative reading of dokimôs – call it the ‘intrasyste-
matic’ interpretation – we may indeed take the goddess to express in
propria persona a past obligation (‘how it was right’). On this
interpretation, we would, after all, take the expression ‘to be accep-
tably’ to indicate, not the mode of being instantiated by what is (in
line with the strong application of ‘to be’ in Alêtheia), but a distinct
and inferior mode of being typical of Doxastic things. Now, it will
not be the adverb dokimôs by itself which would generate this sense
of concession or qualification, but the context within which this
adverb, and the goddess’s statement as a whole, are configured.
The goddess’s gloss of this aspect of the kouros’ programme of
study (‘how . . . everything’) comes immediately after her stipula-
tion that mortal beliefs lack ‘real trust’ and that the mortal shall learn
them despite this fact (‘but, nonetheless . . . ’, B1.31). Within this
pejorative framework, it is possible to interpret the words ‘how . . .
to be acceptably’ as stipulating a standard, which is indeed met as
fully and successfully as possible by the things which seem to
mortals and are accepted by them, but which has already itself
been qualified as something that lacks real trust. That is, Doxastic
things instantiate their own distinctive mode of being in an (as far as
this mode of being goes) acceptable and proper way, but this mode
of being itself falls short of the mode of being associated with what-
is in Alêtheia. The intrasystematic interpretation has the great
advantage of emphasising and accounting for the marked figura
etymologica: ‘ta dokounta . . . dokimôs’. The things which seem and
are accepted (ta dokounta) have a mode of being which is proper
quite specifically for them (dokimôs). The acceptability which
Doxastic things can legitimately be ascribed (dokimôs) is qualified
as the sort of acceptability which is proper and approved for things
of that nature (ta dokounta).106 The kouros, then, will also learn in
what way even those things, beliefs about which are without real
trust (because they do not reflect the nature of fundamental reality),
are after all there in a way which is distinct and proper specifically

106
Cf. similarly Johansen (2016) 6: ‘We should allow dokimōs to indicate reliability of the
sort appropriate to what it qualifies.’

213
 Why Did Parmenides Write Doxa?
for them. The final specification – ‘ . . . to be acceptably, all of them
traversing (perônta) across everything’ – accentuates how the mode
of being which is proper to such things differs from the unshaken
heart of reality. Such things traverse across everything: they con-
stantly exhibit motion and change.107 Similarly to the absolute
interpretation, the kouros is again in a superior position to the
ignorant mortals. Unlike them, he will be informed of the ways in
which Doxastic things and their mode of being diverge from, and
fall short of, what-is. Unlike them, he would not mistake Doxastic
things for the fundamental reality.
I highlighted at the outset that any translation and interpretation
of B1.31–2 must be uncertain. The arguments concerning the
epistemological and aetiological questions advanced in this chap-
ter do not depend on the foregoing analysis of this difficult couplet.
Nonetheless, let us tentatively highlight the general conclusions,
which have emerged from this analysis.
First, it is quite possible that Parmenides conveys in these lines
the notion that Doxastic things possess a mode of being which falls
short of the mode of being instantiated by what-is. On the absolute
interpretation of ‘to be properly’, this idea would be more implicit
and indirect. Nonetheless, the insistence that only what-is properly
or genuinely is could be thought to create room for the view that
Doxastic things possess something else, which falls short of such
genuine being (δοκίμως εἶναι). On the intrasystematic interpreta-
tion, Parmenides directly and explicitly ascribes a distinct mode of
being to Doxastic things. We will revisit these possibilities and B1.
31–2 in Chapter 5.5.1.
Secondly, on both the absolute and intrasystematic interpreta-
tions (although in slightly different ways), these lines reflect the
same sense in which Doxa is deceptive, and the same sense in
which the kouros will be wary of this deceptiveness, which we

107
This last point relies on the reading περῶντα, favoured by most editors (see Palmer
(2009) 380), over περ ὄντα. The latter is arguably a lectio facilior¸ but could not be
conclusively excluded. Mourelatos (2008a) xxxii, 212–14 champions περ ὄντα. Contra
Mourelatos, however, the idea in περῶντα of entering a space and then leaving it behind
is not inappropriate: the image represents Doxastic things as constantly traversing from
one location and then onto the next through everything (διὰ παντός), like ships
continuously ranging every which way across a kind of global sea (cf. Od. 6.272,
Od. 24.118).

214
 4.5 Conclusions
found before. The mistake that one who is ignorant of the doctrine
of Alêtheia will make is to confuse the things that appear to
mortals and are accepted by them with ultimate reality and to
construe accounts of such things as if they were accounts of
ultimate reality. Once again, Doxa is ‘deceptive’ (B8.52) and
lacks ‘real trust’ (B1.30), not because it misdescribes its objects
(mortals, stars, embryos), but because it may mislead one into
believing that these objects, or the two elements of which they
consist, are in fact the fundamental nature of reality, which would
be inconsistent with the nature of what-is as established in
Alêtheia.

4.5 Conclusions: Why Did Parmenides Write Doxa?

We are now in a position to draw an inference from our analysis of
the epistemological question to the aetiological one and to offer an
interpretation of B1.28b-30 (for the text and translation, see the
introduction to the chapters on Parmenides).
By accepting Alêtheia’s ontology and its notion of what-is,
Parmenides can and should come to attain a critical and informed
attitude towards his experience and recognition of such Doxastic
processes as locomotion, and of such things as stars, as well as
towards his own and all other first-order beliefs with regard to
such processes and things. The fundamental qualities of Doxastic
things and processes are incompatible with the fundamental qua-
lities of the unshaken heart of reality. Correct accounts of Doxastic
things are liable to (but must not) be mistaken for correct accounts
of reality, of what-is. Significantly, however, the very notion of
mortals, as distinct from each other and from other things such as
stars and embryos, diverges from what Alêtheia describes as
reality no less than any other Doxastic thing concerning which
mortals may hold or reject first-order beliefs, or than those beliefs
themselves. I find no evidence that Parmenides attempted to
square the apparent presence of mortals (stars, embryos . . .) with
the incompatibility of their properties with the properties of what-
is, or that he saw the need to do so.108 The important point for us is

108
See further Ch. 5.5.1.

215
 Why Did Parmenides Write Doxa?
that, insofar as we can speak of such Doxastic things as mortals
(and the fact that the goddess speaks of and to mortals shows that
there is an extent to which we can do so), we can also construe and
describe them correctly and incorrectly. The theories advanced in
Doxa do so correctly. They represent Parmenides’ own views of
the origins, natures and behaviours of Doxastic things and are
taken to endow their student with knowledge of such things
(B10). The views that the moon’s light is a reflection of the
sun’s, and that mortals comprise Hot and Cold, account respec-
tively for the moon and for mortals correctly, in a way that
competing accounts do not. The theory of mortal cognition
advanced in B16, therefore, bears directly on our question:
‘What are the ways in which the mortal agent, to whom the god-
dess speaks, must think?’
In B16, Parmenides advances a theory of human cognition. This
theory analyses the same ‘wandering’, Doxastic cognition which
characterises the ignorant mortals in Alêtheia (B6.4–6) as an
ineluctable and passively determined function of the physiologi-
cal, cognitive apparatus which characterises all mortals univer-
sally. As we noted (Chapter 4.3), the fragment thus reflects
Parmenides’ recognition that he can no more simply discard first-
order beliefs with regard to heterogeneous, plural things and
processes than he can permanently switch off his experience and
recognition of such things and processes. All mortals comprise
themselves Hot and Cold; perceive a heterogeneous plurality of
things as mixtures of Hot and Cold; recognise such things, and
form first-order judgements with respect to them.
By brute, physiological necessity, therefore, mortals judge
things in accordance with precisely the same contrast between
two, sensible enantiomorphic elements, each one of which both
is (itself) and is not (the other), which constitutes also the essence
and basis of the cosmology developed in Doxa (ἐκρίναντο δέμας,
B8.55). The individual theories advanced in Doxa comprise very
specific and sophisticated accounts of particular Doxastic things
and processes. But the fundamental sensory oppositions, which
underpin all those theories, are constitutive of mortal thinking as
such. Again, mortals perforce perceive, and form first-order
beliefs about, such sensible processes as change (‘this man
216
 4.5 Conclusions
grows older’), which diverge from what Alêtheia shows to be the
unshaken heart of reality.
Parmenides pursued and included Doxa’s cosmology because
he held (i) that thinking through contrasts between two sensible
opposites, and experiencing, recognising and forming judgements
with regard to a multiplicity of heterogeneous things such as we
encounter in our sensory experiences, is an inescapable aspect of
what it is to think and live as a mortal, and (ii) that the theories
advanced in Doxa – even as they fail to capture the homogeneous
and unchanging nature of the unshaken heart of reality – account
for such heterogeneous and multiple things correctly.
This is why the mortal should learn all things (χρεὼ δέ σε πάντα
πυθέσθαι), not only the true nature of reality, of what-is (ἠμὲν
ἀληθείης . . . ἦτορ), but also mortal beliefs (ἠδὲ βροτῶν δόξας),
while recognising, although he could never simply discard them
qua first-order beliefs, that all such beliefs lack real trust (ταῖς οὐκ
ἔνι πίστις ἀληθής). This is why, despite its failure to capture the
unshaken heart of reality, the mortal should nonetheless learn
Doxa too (ἀλλ’ ἔμπης καὶ ταῦτα μαθήσεαι, B1.31), and attain
knowledge and understanding of such things as stars, mortals
and embryos (B10).
I am not arguing that B16 is a programmatic text. I do not
suggest that, in it, Parmenides explicitly or implicitly communi-
cates why he wrote Doxa. I am arguing, rather, that B16 articulates
a doctrine which, given further considerations, best explains why
Parmenides wrote Doxa. It explains, that is, why learning Doxa
makes good epistemological sense also for the mortal who has
attained knowledge of the true nature of what-is.
After she introduces the key principles of Doxa, the goddess
announces that she will now embark on her account of the cosmol-
ogy proper, ‘so that’ (ὡς) no ‘judgement of mortals’ (βροτῶν
γνώμη) shall outstrip Parmenides in the race (B8.61).
The ambiguity in the English as to whether ‘so that’ introduces
a result or final clause tracks the ambiguity in the Greek. It is
unclear whether the goddess is saying that she will be narrating
Doxa with the result that, or with the intention that, no mortal
judgement shall outstrip Parmenides. At any rate, even if we read
a final clause, this line hardly expresses by itself an answer to the
217
 Why Did Parmenides Write Doxa?
aetiological question since, were it not for the ineluctability of
Doxastic cognition as analysed above, we would still ask our-
selves why Parmenides should wish to join the mortal race in the
first place. However we read this clause, the arguments advanced
in this chapter indicate that, by a ‘judgement of mortals’ (βροτῶν
γνώμη, B8.61) and the ‘beliefs of mortals’ (βροτῶν δόξας, B1.30),
the goddess is referring, not to beliefs and judgements which
happen to be held by agents who happen to be mortal, but to
beliefs and judgements which essentially belong to a specifically
mortal kind of cognition.109 With regard to B8.61, this seems
independently plausible. If a ‘judgement of mortals’ indicated
any kind of judgement held by a mortal, rather than judgements
specific to mortal cognition, then Alêtheia itself would presumably
have sufficed to secure the kouros against being outstripped by any
other judgement.
The goddess’s statement that ‘it is right’ for Parmenides to learn
mortal beliefs or that he ‘must’ do so (χρεὼ δέ σε, B1.28b) high-
lights that learning Doxa is appropriate for Parmenides. In uses of
this vocabulary, the nature, status or identity of an agent typically
renders it incumbent upon him to conduct himself in a certain way.
So, for example, Sarpedon states that champions, who receive the
highest social honours, must fight in the foremost ranks (χρή,
Il. 12.315).110 Doxastic thought and cognition are not only an
ineluctable but also an appropriate aspect of what it is to think
and live as a mortal.
At the same time, we saw that Parmenides’ attitude to Doxa,
unlike that of the ignorant mortals, is informed by his knowledge
of Alêtheia. While Parmenides accepts Doxa’s theories qua
accounts of Doxastic things, he recognises their inadmissibility
as accounts of what-is. Parmenides will not, therefore, be bullied
(μηδέ σ’ . . . βιάσθω, B7.3) into sharing in the avoidable resolve of
mortals (κατέθεντο, B8.39, 53; B19.3) to embrace their mortal
language (‘tongue’, B7.5), and the appellations of Doxastic

109
Cherubin (2005) 21 suggests similarly, although without pursuing the point in the ways
defended here or relating it to B16, that βροτῶν δόξας are ‘a set of beliefs and a way of
thinking that reflect something of what it is to be a mortal’.
110
For the point that the term χρεών (and its cognates) signifies normative necessity, see
Mourelatos (2008a) xxxi, 277–8.

218
 4.5 Conclusions
elements (B8.54), things (B19.3) and processes (B8.40–1), as
categories which actually reflect and disclose the fundamental
nature of reality (B8.39; B1.30). So too, he will not heedlessly
acquiesce in his sensory experiences as trustworthy reflections of
what-is (recall ‘an aimless eye and echoing ear’, B7.4).111
In B6, does the goddess state that she withholds the kouros from
(εἴργω), or that she will later commence with (ἄρξω), the way of
inquiry which the ignorant mortals wander? The Greek is incom-
plete and these or any other supplements will be uncertain.112
Reading ‘I withhold’ (εἴργω), and taking the line to mean simply
that the goddess categorically bars the kouros from pursuing
mortal inquiries, flies in the face of all the evidence which we
considered in this chapter and the basic point that Parmenides
does, in fact – whatever one makes of this fact – go on to pursue
a mortal way of inquiry in some detail. One possibility, however, is
to keep the traditional supplement εἴργω but to interpret the god-
dess as barring the kouros, not from any way of mortal inquiry
whatsoever, but from the particular way of mortal inquiry which
the ignorant mortals wander (ὁδοῦ ταύτης διζήσιος <εἴργω> |
αὐτὰρ ἔπειτ’ ἀπὸ τῆς, ἣν δὴ βροτοὶ εἰδότες οὐδὲν | πλάζονται, B6.
4–6). That is, the goddess bars the kouros from pursuing, not any
inquiry based on mortal thinking or any cosmological inquiry, but
the manner of inquiry which the uninformed mortals exhibit.
Indeed, the following lines focus on this point. The ignorant mor-
tals pursue their way of inquiry in a daze of aimless confusion.
The goddess’s own exposition of the natures of Doxastic things
(B8.60 and following) will be a different affair. Alternatively, we
can read ‘I will commence’ (ἄρξω), or a verb to some such effect,
and take it that the goddess’s own exposition of the way of mortal
inquiry will lack the flaws exhibited by the ignorant mortals, who
wander about this same way aimlessly and helplessly. Either way,
since the kouros will pursue mortal inquiries in the informed
manner which we described above, and since the theories which
the goddess expounds in Doxa account for Doxastic things and
111
In addition to ‘aimless’, ἄσκοπος can also signify ‘heedless’, ‘un-regardful’, see Il.
24.157; Aesch. Ag. 462.
112
εἴργω: e.g. Wedin (2014) 53–71. ἄρξω: e.g. Nehamas (1981); Palmer (2009) 159–60,
367.

219
 Why Did Parmenides Write Doxa?
processes correctly, the cosmology will endow the kouros with
knowledge and understanding of its proper subject matter (B10),
so that he will never be outstripped by any other judgement of
mortals (B8.61). By contrast, the goddess describes the mortals
who pursue mortal inquiries without comparable guidance, and
without the ability to situate such inquiries in relation to an
awareness of what-is, as doing so blindly and helplessly, and as
remaining confused and ignorant (εἰδότες οὐδέν, B6.4).
Our conclusion, that Parmenides accepts Doxa as an account of
Doxastic things, but rejects it as an account of the unshaken heart of
reality (of what-is), gives us a clear explication of the sense in which
he both developed and advanced Doxa’s theories and kept those
theories at arm’s length (‘in which there is no real trust’, B1.30,
etc.). Ancient commentators were often sensitive to the point that
Parmenides does represent Doxa as part of his philosophy, but does
not do so unqualifiedly. Theophrastus (apud Alexander, in Metaph.
31.9–14 = Coxon (2009) Testim. 40) writes that Parmenides both
argued that the all is one and ungenerated, and attempted to expound
the coming-to-be of the many things. He qualifies: ‘not holding
beliefs in the same way about both, but, on the one hand, in
accordance with truth [or “reality”: alêtheia] . . . and, on the other
hand, in accordance with belief (doxa)’. Simplicius observes that,
unlike Melissus who denies coming-to-be categorically (ὅλως),
Parmenides does allow it, but ‘not with respect to truth (alêtheia),
but with respect to belief’ (doxa, in Cael. 7.556.12–14 = Coxon
(2009) Testim. 203). Aristotle names Parmenides as the only monist
to appreciate the efficient cause, but qualifies that he too did so only
‘to the extent that he posits not only one but in a way also two
principles’ (κατὰ τοσοῦτον ὅσον οὐ μόνον ἓν ἀλλὰ καὶ δύο πως
τίθησιν αἰτίας εἶναι, Metaph. Α.3 984b3–4; cf. Α.5 986b27–34).
Doxa is sufficiently Parmenides’ for Aristotle to ascribe to him
the postulation of two causes as his own thesis, but it is not so
unqualifiedly (‘to the extent that . . . in a way’).113
Adopting Parmenides’ conception of the mortal cognitive agent
as our starting point for thinking about the relation between the two
parts of the poem is defensible not least because Parmenides himself

113
See also Philop. in Ph. 16.22.6–7, quoted in Ch. 4.4.

220
 4.5 Conclusions
does much the same. The programmatic statement that both
Alêtheia and Doxa are to be learned (ἠμέν . . . ἠδέ, B1.29–30) is
embedded in a conversation between a goddess and, emphatically,
a mortal. Our ensuing analysis of the aetiological problem, how-
ever, gives rise to another. If we previously asked, ‘why did
Parmenides write Doxa?’, we will now ask, ‘how could
Parmenides have written Alêtheia?’ If, being a mortal, Parmenides
must think in Doxa’s structuring categories, i.e. through the krisis
between Light and Night, then how could he have also thought in
the competing structuring categories of Alêtheia, i.e. through the
krisis between ‘is’ and ‘is not’? If, being a mortal, it is appropriate
for Parmenides to learn mortal beliefs, then why is it also appro-
priate for him to learn Alêtheia (‘it is right for you to learn all
things’, B1.28)?
There is, then, an apparent tension in Parmenides’ thought.
Doxastic cognition characterises the human mind ineluctably
(‘ . . . so is mind present to humans’, B16.2). But, when encounter-
ing the goddess, Parmenides is encouraged to resist the mental
habits of mortals (B7) and to apply his mind rather to what-is: ‘but
gaze . . . with your mind’ (λεῦσσε . . . νόῳ, B4.1). Exploring this
tension will help us not only to address our new problem – ‘how
could Parmenides have written Alêtheia?’ – but also to deepen our
understanding of the relation and interaction between mortal and
divine in Parmenides’ poem.

221
 5

HO W CO U L D PA R M E N I D E S H AV E W R I T T E N
ALÊTHEIA?

5.1 Recap and Introduction: Some Resolutions

of the Paradoxical Implications of Parmenides’
Theory of Human Cognition
Our central concern in this chapter will be to ask why Parmenides
thinks that the mortal is capable of attaining the knowledge of
reality proffered in Alêtheia, despite the fact that, insofar as he is a
mortal, he is constrained to think and to experience in human
ways; that is, in ways which could not sustain an understanding
of Alêtheia. It will help, first, to review how some previous readers
have responded to the apparently puzzling implications of
Parmenides’ account of human thinking in DK28 B16, and to
sharpen further the challenge that confronts us.
It has long been appreciated that the idea that the human mind is
passively determined in the way that B16 suggests appears to jar,
prima facie, with the mortal’s ability to acquire knowledge of what-
is. Some scholars pre-empt any paradoxical implications by stating
that Doxa’s theories – its account of human thinking in B16
included – simply do not represent Parmenides’ own views.1 As
we saw in the previous chapter (especially Chapter 4.4), however,
Doxa’s theories do represent Parmenides’ own views of the natures,
origins and behaviours of Doxastic things (such as human cognitive
agents). Scholars also regularly take it that the mortals’ habit of
thinking like mortals and their possession of first-order Doxastic
beliefs, as analysed in B16, is a regrettable, avoidable and reversible
choice.2 Such descriptions are, as they stand, oversimplifications.

1
Lesher (1984) 29, n.39; Trépanier (2004) 164; Torgerson (2006) 37, n.28; cf. Gallop
(1984) 120–1.
2
E.g. Woodbury (1986) 8ff (‘a deliberate, though mistaken policy’); Curd (1998) 114 (‘a
bad habit’); (2011) 121, 131; (2015).

222
 5.1 Recap and Introduction
As we saw (Chapter 4.2), Parmenides’ theory of human cognition is
a universalising account, which describes all, not some, acts of
human cognition. But the concern behind these descriptions is
clear enough. The ineluctability of Doxastic cognition to the mortal
(reflected in B16) must somehow coexist with the mortal’s evident
capacity to think in accordance with Alêtheia’s krisis, and to come
thereby to cognise what-is.
Anthony Long takes it that in B16 Parmenides describes the
‘normal (i.e. delusory) human νοῦς’.3 In doing so, however, Long
illicitly limits (n.b. ‘normal’) the unqualified and universalising
account in B16 of the human mind (‘as on each occasion . . . in all
humans and in each . . . ’). But if the human mind can never know
what-is, then how can the human kouros know it?
Jaap Mansfeld (1964) argued that, once we posit divine revela-
tion as the source of Parmenides’ knowledge, further questions
become superfluous.4 According to Mansfeld, only the goddess’s
intervention, which he takes to comprise her assurance of the truth
of the otherwise indemonstrable premises of Parmenides’ argu-
mentation, could explain how the mixture of Light and Night
which is the mortal mind could have left its humanity behind,
come to know what-is, and been initiated into Alêtheia’s logic.5
Mansfeld is, I believe, entirely correct that, if we are to account for
the mortal’s capacity to know what-is, then we must see
Parmenides as positing non-figuratively an epistemically signifi-
cant interaction with the divine. Parmenides describes to us an
interaction with a divine power through which he was able to
attain knowledge which he could not have attained independently
of that interaction.
As it stands, however, Mansfeld does not yet confront the
difficulty posed by Parmenides’ account of human thinking in
B16. The problem runs deeper than is implied by Mansfeld’s
assumption that the bare postulation of genuine divine revelation
satisfactorily and finally explains the mortal’s capacity to sustain

3
Long (1996) 147; cf. similarly Curd (2015) noo.15–17.
4
Mansfeld (1964) 259 (‘nehmen wir aber an, daß Parmenides seine Erkenntnis der Gnade
und der Offenbarung verdankt, so sind weitere Fragen überflüssig und das alte Problem
der Beziehung zwischen Fr. 16 und Fr.1 f ällt fort’).
5
Mansfeld (1964) 260–1.

223
 How Could Parmenides Have Written Alêtheia?
Alêtheia’s argumentation and to acquire knowledge of what-is.
The question is not simply, ‘how can the mortal come to find out
and ascertain the truth of Alêtheia’s premises and conclusions?’,
but rather, ‘how can the mortal even come to recognise, under-
stand and contemplate the argumentation and tenets of Alêtheia
and, thereby, the homogeneous what-is, if Parmenides maintains
that such understanding is possible for the divine but not for the
mortal mind?’ How can the goddess reveal to the mortal’s mind
something which the mortal mind cannot cognise? My own
answer to these questions ascribes to the goddess a similarly
essential role, but it is not that of acting as the ‘guarantor of
truth’ of otherwise indemonstrable propositions, which are there-
fore accepted on trust.6 Even if the goddess also provided for
Parmenides an assurance of the truth of his accounts,7 this will
not explain how Parmenides can maintain both (i) that the human
mind, unlike the goddess, cannot know what-is, and (ii) that,
through divine favour, the initiated human, like the goddess,
does know what-is.8 The knower himself must have in some
sense become divine and acquired a higher-than-mortal, divine
mind.
In an important article, Gregory Vlastos (1946) suggested that
the mortal will know what-is when he thinks pure Light as some-
thing changeless and self-identical, with a mind completely free of
darkness. ‘So immovable a thought’, writes Vlastos, ‘could only
have Being as its object: not the “seeming” light of the senses,
whose apparent qualities depend on the felt contrast with dark-
ness.’ The mortal, Vlastos concludes, would thus ‘think light as
pure Being’.9
6
‘Bürgin der Wahrheit’, Mansfeld (1964) 261.
7
I return to this question (noncommittally) in Ch. 6.3.
8
The insightful reflections in Ranzato (2015) 205–7 also do not explain this same puzzle:
how and why does Parmenides think that the kouros, a mortal creature of Light and Night,
can through the goddess’s favour resist the thinking of Light and Night and transcend it to
something different and higher? Every human mind must think of and through sensory
oppositions. Drozdek (2007) 46-8 interestingly posits logos as something divine in the
mortal actuated by the goddess but does not support this idea.
9
Vlastos (1946) 71–2, 67–8; followed by Kahn (2009) 217 with n.10. When it comes to the
question of the prima facie paradoxical implications of B16 itself concerning mortal
thinking, Vlastos (1946) 66–7 simply does away with the problem by dismissing B16 as a
doctrine only of non-cognitive sense-perception. As we saw in Ch. 4.3, the cognitive
language of B16 renders such a view of the fragment untenable.

224
 5.1 Recap and Introduction
Our inquiries in this chapter into Parmenides’ ideas of soul,
eschatology, initiation and revelation will ultimately lead us to
agree with Vlastos on his core point that, when the mortal knower
thinks what-is, he is thinking with a mind that is pure light and
divested entirely of darkness. Vlastos himself, though, nowhere
asks exactly how the mortal might come by this ability to divest his
mind of darkness and to think with pure light – a question which
will preoccupy us throughout this chapter – nor puts this idea to the
interpretative uses suggested below. Furthermore, it will be useful
to highlight here certain aspects of Vlastos’ account with which we
should not agree, and to delineate further the interpretative pro-
blems which will be addressed in the following sections.
It is difficult to see how Vlastos can designate the very same
object of knowledge as (i) Being rather than ‘the light of the
senses’, and (ii) Light, which the mortal thinks as pure Being.
Vlastos eventually resolves this difficulty in his account by pos-
tulating that ‘the positive self-identity of fire [sc. rather than
fire simpliciter] . . . is Being’. Vlastos, then, adopts Aristotle’s
ontological alignment of Hot with Being and Cold with not-
Being. Night – unlike Light – cannot be and is somehow only
apparently ‘positive’.10 We saw in Chapter 4.3 that Aristotle’s
ontological identification or even alignment of Light with what-
is is ill-conceived. There is no sense in which Parmenides con-
siders Night (heavier and denser than Light as it is) a less positive
thing than Light. Indeed, since both elements symmetrically pos-
sess the property of self-identity, it is arbitrary to identify what-is
with Light’s property of self-identity but not Night’s.11 We saw
that Parmenides is in fact at pains to emphasise that, unlike the
self-identity of what-is (B8.49, 29), the self-identity of each
Doxastic element is inextricably related to its contradistinction
from the self-identity of the other (B8.57–8). We cannot, therefore,
identify what-is with Light, or with some quality of Light, and
conclude that Light (or a quality of it) and what-is are the same
ontological item and, therefore, the very same object of knowl-
edge. What-is cannot itself be described as Light or in terms of a

10
Vlastos (1946) 73–4 with n.45 (my emphasis), 76; cf. Kahn (2009) 217.
11
Vlastos (1946) himself (at 72, 76) describes self-identity as a property of both opposites.

225
 How Could Parmenides Have Written Alêtheia?
certain relation between Light and Night. It cannot be the case that
thinking Light (or Night, or some proportion of Light and Night) is
itself nothing other than thinking what-is.
Nonetheless, it can still be the case that the mortal’s capacity to
sustain knowledge of what-is is concomitant with and even depen-
dent on his acquisition of a very particular physiological state, viz.
his acquisition of an exclusively Hot (= Light) mind. Indeed, I will
strive to establish this view and to explore its significance in the
course of the following sections. At this point, at any rate, we can
introduce in principle the mere interpretative possibility that our
ability to think what-is is concomitant with and dependent on a
state in which our mind is pure Light and divested entirely of
Night. What Parmenides cannot do, however, is to conflate Light
and what-is as items in the world, or to posit that thinking Light
(some of the fundamental properties of which are incompatible
with some of the fundamental properties of what-is: Chapter 4.3)
just is, in itself, nothing other than thinking what-is.
Indeed, although he conflates Light and Being as objects of
cognition in this way, Vlastos also insightfully recognises a diffi-
culty inherent in the very attempt to articulate a physical formula
for knowledge of what-is:
No such formula could be given without translating Being into terms of
Becoming. We are here face to face with the central paradox of Parmenides’
theory of knowledge. The mortal frame, qua mortal, cannot think Being. Yet the
“knowing man” can and does think it . . . To resolve the paradox is impossible, for
it is only the epistemological counterpart of the ontological dualism of Being and
Becoming.12

We must disentangle here clearly different interpretative pro-

blems, which belong to different interpretative orders. In the first
instance, I highlighted in the introduction to these chapters on
Parmenides that a certain paradox inheres in attempting to explain
how anything other than what-is (e.g. mortals, the goddess) can
cognise what-is. In a sense, this attempt straddles the conceptual
frameworks of both parts of the poem. It is itself inexpressible both
in terms of the structuring contrast of Alêtheia (the krisis between
‘is’ and ‘is not’ allows for the expression of what-is but not of
12
Vlastos (1946) 71–2.

226
 5.2 Parmenides on the Soul
things like mortals) and in terms of the structuring contrast of
Doxa (the krisis between Light and Night (or Hot and Cold, etc.)
allows for the expression of mortals but not of what-is). This
paradox indeed seems to be the epistemological aspect of an
ontological tension. If what-is comprises reality, then what exactly
is the status of things like mortals, and how can they know what-
is? Indeed, how can they do anything at all for that matter? This
ontological tension will not be our focus in this chapter, although
in Chapter 5.5.1 I will advance my own suggestions about it
(suggestions on which the central arguments of this chapter will
not depend). The question on which we will be focusing is more
limited. Parmenides describes knowledge of what-is as divine
knowledge, i.e. as attainable by the goddess, but unattainable by
the human mind. The human mind passively and ineluctably
perceives-and-cognises Doxastic things and operates in terms
of the Doxastic krisis (Light/Night) rather than Alêtheia’s krisis
(‘is’/‘is not’). But if this is so, then how can the knowing man
nonetheless think what-is? It is with this paradox in mind that I
wish to address the question – with which Parmenides’ engage-
ment seems textually undeniable – ‘how can mortals cognise
what-is?’ My proposed resolution of this problem will not by itself
help with the prima facie, ontological paradox which haunts any
talk of anything other than what-is knowing what-is (or doing
anything at all). Nor, however, will my arguments further exacer-
bate the ontological difficulty by attempting somehow to align or
to identify what-is with Light or to argue that knowledge of what-
is is in fact nothing other than knowledge of Light (or of Night, or
of some proportion between the two).

5.2 Parmenides on the Soul: Hot Metempsychosis

and the Physiology of Divinisation
We will see in this section that Parmenides understood the mortal
agent to be a complex thing. The mortal agent comprises not only a
mortal but also a divine part – his soul. The soul consists of the
aethereal element Hot, undergoes eschatological migrations and is
possessed of a cognitive disposition which is superior and more
pure. Our findings in this section will thus help lay the ground for
227
 How Could Parmenides Have Written Alêtheia?
our discussion in Chapter 5.3 of the proem (B1), and for the
broader argument advanced in this chapter that a sort of episte-
mological becoming-like-god is central to Parmenides’ system. In
this section, we will prepare for and support that argument by
exploring the cosmological, physiological, psychological13 and
eschatological aspects of the presence and function of a divine
part within the mortal.
Various contemporary and near contemporary sources attest
(different, broadly complementary aspects of) what appears to
have been a familiar view, according to which the mortal com-
prises both a mortal, earthy element and a divine, higher, aethereal
element. The aethereal element is identified with the soul, and both
elements return to their source at the mortal’s death. Now, we
cannot reconstruct with any confidence Parmenides’ views – he
does appear to have held some – concerning the topography of the
soul’s origin and destination. I offer some aporetic remarks con-
cerning this matter in the Appendix. Nonetheless, by considering
our evidence for Parmenides’ attitude to Hot, aether, metempsy-
chosis and especially the soul and its divinity, we will be able to
establish that Parmenides advanced what may reasonably be con-
sidered a version of this contemporary view of the soul. Before we
delve into our evidence for Parmenides, therefore, it will be useful
first to lay out some reflections of this general attitude.
Epicharmus, writing in the late sixth or fifth century, describes
the process of death: ‘It was combined and it was separated, and it
went back whence it came, earth to earth, and breath upwards’
(συνεκρίθη καὶ διεκρίθη κἀπῆλθεν ὅθεν ἦλθεν πάλιν, | γᾶ μὲν εἰς γᾶν,
πνεῦμα δ’ ἄνω, DK23 B9). Elsewhere, Epicharmus suggests that
the quality of one’s experiences following this separation may
depend on the nature of one’s disposition prior to it: ‘If you are
by nature pious in mind, you would not suffer any evil when you
die; your breath will persist above in heaven’ (ἄνω τὸ πνεῦμα
διαμενεῖ κατ’ οὐρανόν, DK23 B22). An epigram for the fallen in
Potidaea (432 BCE) states that ‘aether received their souls, and
earth their bodies’ (αἰθὲρ μὲμ φσυχὰς ὑπεδέχσατο, σόμ[ατα δὲ
χθόν], IG I2 945). In Euripides’ Erechtheus, Athena proclaims

13
By Parmenides’ ‘psychology’ I mean his conception of the nature of the soul.

228
 5.2 Parmenides on the Soul
concerning Erechtheus’ daughters: ‘Their souls, then, did not
depart to Hades, but I have settled their breath in the aether’
(ψυχαὶ μὲν οὖν τῶνδ’ οὐ βεβᾶσ’ [Ἅιδ]η̣ν πάρα, | εἰς δ’ αἰθέρ’
αὐτῶν πνεῦμ’ ἐγὼ [κ]α̣τῴκισα, TrGF 5.370.71–2). Athena’s state-
ment has the ring of an eschatological divine reward about it, like
Epicharmus’ comment concerning the pious man’s breath and,
perhaps, like the Potidaea inscription.14 In Euripides’ Suppliants,
Theseus articulates what sounds more like a cyclical cosmological
principle (reminiscent of Epicharmus DK23 B9), although one
whose orderly and smooth implementation perhaps depends on
the fulfilment of the proper burial practices: ‘Let the dead now be
buried in the earth, and, whence each thing came to the light, there
let it return, the breath to the aether, and the body to the earth’
(πνεῦμα μὲν πρὸς αἰθέρα, | τὸ σῶμα δ’ ἐς γῆν, Supp. 531–4).
Euripides engages with aethereal eschatology twice more. In his
Helen, we hear that not only the living but also the dead can
undergo punishment, since ‘the mind of the dead does not live,
but it does have immortal thought, as it falls into immortal aether’
(ὁ νοῦς | τῶν κατθανόντων ζῇ μὲν οὔ, γνώμην δ’ ἔχει | ἀθάνατον εἰς
ἀθάνατον αἰθέρ’ ἐμπεσών, 1013–16). In his lost Chrysippus,
Euripides spoke of the ‘aether of Zeus’ as the ‘progenitor of both
men and gods’, and again articulated a cyclical principle: ‘the
things that are grown from the earth return to the earth, and the
things that sprouted from the seed of aether went back again (πάλιν
ἦλθε) to the heavenly vault, and nothing of what comes to be dies’
(θνῄσκει δ’ οὐδὲν τῶν γιγνομένων, TrGF 5.839.1–2, 9–12 =
Anaxagoras, DK59 A112). A similar eschatology is expressed
on some fourth- and third-century Attic tombstones: ‘The earth
raised the body to the light, and the earth conceals it; but aether has
again taken the breath, as it previously gave it’ (γαῖα μὲν εἰς φάος
ἦρε . . . γαῖα δὲ κεύθει | σῶμα, πνοὴν δὲ αἰθὴρ ἔλαβεν πάλιν, ὅσπερ
ἔδωκεν, IG II/III2 12599), and again: ‘Eurymachus’ soul and his
lordly thoughts moist aether holds, but his body this tomb holds’
(Εὐρυμάχου ψυχὴν καὶ ὑπερφιάλος διανοίας | αἰθὴρ ὑγρὸς ἔχει,
σῶμα δὲ τύμβος ὅδε, IG II/III2 11466). This last inscription is
almost identical to an epigram in the Aristotelian Peplos

14
Rohde (1925) 437 finds retributive eschatology in the Potidaean epigram.

229
 How Could Parmenides Have Written Alêtheia?
(fr. 641.7–8 Rose: θυμὸν δὴ Κύκνου . . . ὅδε), where aether is,
however, not ‘moist’ but ‘bright’ (λαμπρός). The similarity sug-
gests a common formula.15
We should not try to force a homogeneous or overly worked-out
theology and eschatology on such texts. Nonetheless, we may note
that they all divide the human person into two fundamental compo-
nents: the earthy human body and the aethereal soul or breath.
Furthermore, a cyclical principle is at work, either invariably or as
a matter of divine discretion. At the time of the person’s death, each
component part returns to its proper place of origin. The earthy body
descends back into the earth, whereas the aethereal soul or breath
ascends back to the aether, understood as a higher and brighter region
of the world. Some notion of the persistence and, indeed, immortality
of the aethereal soul is clearly implicit in these passages. The earthy
corpse may decay in the ground, but our aethereal soul or breath –
including, at least in some articulations of this conception, our
thoughts and our judgement – carries on and persists in the aether.
For the pious man of Epicharmus, this is something to look forward
to. Equally, aether has received the souls of those who fell in
Potidaea, and the souls of the daughters of Erechtheus found a new
home ([κ]α̣τῴκισα) in the aether (rather than Hades). In his Helen,
Euripides expressly speaks of our mind’s retention of ‘immortal’
thought or judgement (γνώμην . . . ἀθάνατον).
Let us turn now to Parmenides and see how a version of this
general conception – a distinctive version, to be sure – emerges
from our evidence.
Parmenides positions ‘common aether’ (αἰθήρ . . . ξυνός, B11.2)
in the midst of a catalogue of such celestial bodies as the sun, the
moon, the Milky Way and the heavens. In B10, the goddess pro-
mises the kouros knowledge of the nature of aether and of the
celestial bodies positioned in it (αἰθερίαν τε φύσιν τά τ’ ἐν αἰθέρι
πάντα | σήματα), as well as of the sun, the moon and the heavens
(σήματα, then, denotes, as often, celestial bodies). For Parmenides,

15
For discussions of these passages, see Rohde (1925) 435–8; Gomperz (1929) 24 with
n.18; Dodds (1951) 174, n.112; Burkert (1972) 361; (2004) 112–13; Gagné (2007) 13
with n.39; Mihai (2010). Vlastos (1946) 75 associates Parmenides with such passages
(although without further elaboration or argument) for the basic idea that the human
comprises both aethereal and earthy parts. On Vlastos (1946), see Ch. 5.1.

230
 5.2 Parmenides on the Soul
therefore, ‘aether’ signifies primarily a hot, celestial region (note
also his reference to ‘the hot might of starts’, B11.3).16 But
Parmenides also characterises the element Light, which is the
same as the Hot (see Chapter 4.2–3), quite generally as ‘the aether-
eal fire of flame’ (φλογὸς αἰθέριον πῦρ, B8.56). Thus, the element
Fire or Hot – including the Fire or Hot within us – is itself called
‘aethereal’, perhaps insofar as the element Hot is concentrated
chiefly in the celestial region of aether, and insofar as this region
is closely associated with, and, it seems, consists especially of, the
element Hot.17 Although it is never explicitly paralleled in the
extant fragments, moreover, the common doxographic employment
of ‘earth’ for Parmenides’ Night18 is supported by that element’s
characterisation as solid, heavy and dark (B8.59).
In our analysis in Chapter 4.2 of Theophrastus’ account of
Parmenides in his On the Senses, we saw that Theophrastus
gives us conclusive evidence that Parmenides considered corpses
to be Cold things and took death to involve ‘the departure of the
fire’ (τὴν ἔκλειψιν τοῦ πυρός, Sens. 4.5–6) from the Cold body.19
We may note here further the congruent (but, in themselves, less
compelling) reports that Parmenides also explained both ageing
(Aëtius, A46a) and sleeping (Tertullian, A46b) by reference to a
more partial decrease in Hot.20 Also congruently, Diogenes

16
In his notoriously difficult account of Parmenides’ cosmology (A37), Aëtius describes
aether as the uppermost region of the universe (ἀνωτάτω πάντων τοῦ αἰθέρος), which
orders the fiery heavens (ὑπ’ αὐτῷ τὸ πυρῶδες ὑποταγῆναι τοῦθ’ ὅπερ κεκλήκαμεν
οὐρανόν), and Cicero speaks of a ring of fire (continentem ardorem lucis orbem) which
encircles the heaven (ND 1.28 = A37). For discussions of this evidence, see Morrison
(1955) 60–5; Coxon (2009) 362–9.
17
See similarly Coxon (2009) 353; cf. Kahn (1960) 148. This formulation does not
commit us to the view that Parmenides equates aether and fire, an idea criticised by
Kingsley (1995) 18. The association of aether with celestial regions is, however, clearly
not a later development, contra Kingsley (1995) 16–17. Kahn (1960) 140ff shows that,
already in epic, ‘aether’ signifies most properly ‘celestial light’, and, moreover, that epic
texts often do not strictly demarcate light as an active force, the state it produces
(celestial brightness or translucence), and the celestial region it inhabits. Kahn (1960)
148 observes also that the etymological connection of αἰθήρ with αἴθω (‘kindle’, ‘blaze’)
aided its later association with fire.
18
See Coxon (2009) Testim. 26; 32; 40–1; 53; 88; 90; 193; 195–6; 204; 207.
19
Note Ch. 4.2, n.20: the departure of fire from the corpse at death leaves it either entirely
devoid of that element or at least very largely so.
20
Tertullian is speaking en masse also of Empedocles, who is accorded also in DK31 A85
the views that sleep results from the partial, and death from the complete, departure
of Hot.

231
 How Could Parmenides Have Written Alêtheia?
Laertius reports that, in his anthropogony, Parmenides maintained
that ‘the generation of humans arose first from the sun’ (γένεσίν τε
ἀνθρώπων ἐξ ἡλίου πρῶτον γενέσθαι, A1.9–10), perhaps suggest-
ing that Parmenides conceived of the sun as the initial source of
Hot and, therefore, of life in mortals.21 As we can confidently
maintain on the basis of our evidence from Theophrastus alone,
then, Parmenides clearly subscribed to the widespread association
of Hot with one’s life-force and vitality.22
Now, direct doxographic reports concerning Parmenides’ view
of the soul are scanty, schematic and variable. Aëtius indiscrimi-
nately ascribes to Parmenides and Hippasos the view that the soul
is fiery (πυρώδη, A45). According to Macrobius, Parmenides
maintained that it was constituted from earth and fire (ex terra et
igne, A45). Theophrastus, according to Diogenes Laertius, said
that Parmenides identified soul (psychê) and mind (nous, A1.11–
12). In view of this scarcity and variance, it is not surprising that
the question of Parmenides’ psychology has been largely ignored
in modern scholarship.23 And indeed, it is not at all clear that
Parmenides explicitly discussed the soul’s physical composition.
That is not to say, however, that Parmenides did not use a term in
reference to souls nor, moreover, that he did not do so in a manner
that implied a certain integration of it into his cosmological

21
In retaining the manuscript reading ἐξ ἡλίου, I follow the avoidance of unnecessary
emendations in Tarán (1965) 251–2 (contrast Verdenius (1947) 287–9). As Tarán
suggests, αὐτόν in the following clause is most probably elliptical for αὐτὸν τὸν
ἄνθρωπον, to be supplied from the immediately preceding ἀνθρώπων: ‘and the human
himself (αὐτόν) comprises the Hot and the Cold, from which all things combine’. The
clause thus indicates that the human, as finally produced, comprises both Hot and Cold,
in keeping with Censorinus’ report (A51) that Parmenides’ (like Empedocles’) anthro-
pogony involved the mixture of two elements (he gives ignis and umor). If so, then the
sun would play for Parmenides a primary but non-exhaustive role in the first emergence
of living mortals. Compare the causal role which Anaximander accorded the sun in the
generation of ζῷα (DK12 A11.6, with KRS (1983) 141–2); cf. also PDerv. col.25.9–10.
For the generative role of Hot and fire, see also Archel. DK60 A1; A4.5; Emped. DK31
B62. For divine aether as generative, cf. Eur. TrGF 5.877; 839.1–5 = Anaxag. DK59
A112.
22
The representation of the vital force in humans as a kind of fiery heat is at least as old as
Homer, see Il. 21.463–6 (βροτῶν . . . οἳ . . . ἄλλοτε μέν τε | ζαφλεγέες τελέθουσιν . . . ἄλλοτε
δὲ φθινύθουσιν ἀκήριοι), and cf. Il. 24.54 for the corpse as ‘un-sensing earth’ (for
Parmenides, as we saw, the fireless corpse is perceptive and cognisant in its own
way). On vital heat in the Hippocratics, Aristotle and the Stoics, see Solmsen (1957).
23
See e.g. Casertano (2011) 49, n.111: ‘we do not have any clue to judge on this point’.

232
 5.2 Parmenides on the Soul
system. In fact, Simplicius offers us a far more reliable report of
what shows every sign of being a theory of metempsychosis. A
close consideration of this report will support Aëtius’ account of
Parmenides’ conception of the soul against Macrobius’ –
Parmenides did indeed take the soul to be a fiery thing – while
Theophrastus’ reported comment will be exposed as a formulaic
and misleadingly oversimplified Peripatetic complaint.24
When expounding Parmenides’ account of Doxa’s goddess,
Simplicius writes (in Ph. 9.39.19–21 = Coxon (2009) Testim. 207):
καὶ τὰς ψυχὰς πέμπειν ποτὲ μὲν ἐκ τοῦ ἐμφανοῦς εἰς τὸ ἀειδές, ποτὲ δὲ ἀνάπαλίν
φησιν. ἀλλὰ ταῦτα μὲν διὰ τὴν πολλὴν νῦν ἄγνοιαν τῶν παλαιῶν γραμμάτων
μηκύνειν ἀναγκάζομαι.

And he says that she conveys the souls, now from the visible to the invisible,25 and
then back again. I am compelled to go on about these issues at length due to
current ignorance of the ancient texts.

Admittedly, the passage does not confirm beyond doubt that ‘the
souls’ (tas psychas) reflects Parmenides’ own phraseology, although
this is entirely plausible. The important point to recognise, however,
is that Simplicius works here very closely indeed from Parmenides’
text. In the previous sentences (in Ph. 9.38.30–9.39.18), Simplicius
24
Relying on Theophrastus’ reported statement, Tarán (1965) 252 with n.59 reinterprets
Macrobius as referring to man as a whole, and dismisses Aëtius as Stoicising. Coxon
(2009) 382 integrates all three reports: the soul is fiery (Aëtius), but contains also a
portion of earth (Macrobius), since it is identified with mind as explicated in B16
(Theophrastus). Vlastos (1946) 70 assumes that both fire and its opposite are present
in the Parmenidean soul. The conception of the soul as fiery can be paralleled elsewhere
in early Greek thought. Betegh (2007) 13–24 has demonstrated that the Heraclitean soul
comprises exhalations, ranging from atmospheric air to (superior) celestial fire. In its
ideal state, then, the soul is fiery (B118; contrast the drunkard’s ‘moist’ soul at B117); cf.
also Schofield (1991) 20–1, 28–30. Aristotle ascribes to Democritus and Leucippus the
view that the soul is ‘a kind of fire and a hot thing’ (πῦρ τι καὶ θερμόν, de An. 404a1–6).
Conceptions of the soul as immaterial are not clearly attested before Plato: Burkert
(1985) 323; Vernant (1991a) 29–30.
25
The contrast with ἐκ τοῦ ἐμφανοῦς by itself guarantees that ‘invisible’ must be the
meaning of ἀειδές here (as universally assumed). With τὸ ἀειδές, moreover, Simplicius
clearly preserves an allusion to ‘Hades’, whatever the original Parmenidean word
(ἀϊδές? ἀειδές?). For Hades and invisibility, see already Il. 5.845. LSJ in fact do not
offer ‘unseen’ or ‘invisible’ as a possible sense of ἀειδής, but this is an oversight. For just
a handful of numerous possible examples, see Plut. Mor. 382F5, 1130A7; Phil. de Abr.
75.5; Corn. ND 74.6; Clem. Strom. 5.14.93.5.2–4; Greg. Nyss. de An. et Res. 46.48.38;
Eus. PE 11.27.6.3–27.8 (citing Pl. Phd. 79a6-b16). Mss. for Pl. Gorg. 493b4–5 and Phd.
79a6-b16, 83b4 carry forms of both ἀειδής and ἀϊδής, although editors typically (but not
invariably) prefer the latter.

233
 How Could Parmenides Have Written Alêtheia?
cites verbatim a series of fragments (B13; B12.1–3 and B8.50–61)
and, soon after (9.39.27–9.40.6), he cites verbatim another series of
fragments (B8.26–8 and B8.30–3). In the midst of these citations, he
introduces this sentence with ‘he says’ (φησιν). Most conclusively, in
the immediately following sentence, Simplicius apologises that he
must expound Parmenides at such length and in such detail due to
current ignorance of the ancient texts (9.39.20). Elsewhere
(9.144.28), Simplicius similarly explains that he is citing at length
from Parmenides’ treatise because of its rarity. Simplicius thus
provides strong and under-discussed evidence that, in a now lost
passage, the goddess was explicitly endowed with the eschatological
role of cyclically receiving and dispatching ‘souls’ or, minimally,
what Simplicius referred to as such in the context of a very close
paraphrase. But how should we interpret the terms which Simplicius
employs, and can we relate Doxa’s eschatological doctrine of soul-
conveyance to its elemental cosmology?
Crucially, we know – as we just recalled – that Parmenides
considered Hot the mortal’s life-force and death a ‘departure of
the fire’ from the Cold corpse. The conveyance of souls reported
by Simplicius, therefore, corresponded to a physiological account
of the inception and cessation of life. The evidence indicates, then,
a cyclical process (ποτὲ μέν . . . ποτὲ δὲ ἀνάπαλιν) whereby the
goddess dispatches Hot from one location to the other, where she
incarnates it in a Cold body (so that the living mortal comprises a
mixture of both Hot and Cold), and dispatches it back to the other
location when separating it from the Cold.26
26
Deichgräber (1959) 716–19 assumes (without mentioning the Cold corpse) that the
goddess conveys pure fire-souls, but contrasts these with the ‘inkarnierten Seelen’ of
living mortals, which comprise fluctuating proportions of both elements. Mansfeld
(1964) 170 rightly objects that we cannot extend ‘soul’ also to the Night-element of
the body if we maintain (as I follow Deichgräber in doing) that Night (or Cold) takes no
part in the metempsychosis. Mansfeld’s own view (171–4), that Simplicius is merely
paraphrasing B16, is impossible to accept. B16 makes no mention of divine ministration
and Simplicius nowhere uses psychê as a synonym for nous when discussing
Parmenides. I cannot see how any reader of Simplicius’ paraphrase could suspect
anything like the doctrine of B16 from what would be an inexplicably elliptical and
idiosyncratic exposition of it. Similarly unmotivated and untenable, I think, is the idea
(which was put to me) that Simplicius might merely be paraphrasing the proem. This
would, again, constitute an impenetrably compressed and misleading gloss – this time of
the detailed journey to the goddess and her opening words in B1 – which could scarcely
be squared with Simplicius’ immediately following remark that he was forced to go on
at length about these aspects of the poem due to current ignorance of the text. In B1,

234
 5.2 Parmenides on the Soul
This framework raises a question concerning conceptual prior-
ity. Did Parmenides identify the mortal’s soul as a particular token
of Hot, or the Hot itself as living or ensouled stuff? On the first
alternative, the soul would be identified with the stable, core Hot
element, which animates the mortal continuously from birth to
death and persists across successive incarnations. The goddess’s
conveyance of Hot to and from the mortal would represent two
delimited actions at both extremes of the mortal’s life. If, as seems
possible but less likely, Hot is itself a living substance, so that
one’s soul was simply the sum of Hot in him at any given moment,
then the goddess would seem to convey Hot continuously and
gradually. An adult mortal life as a whole would be characterised
by a gradual decline in living substance, with the cause of ageing
being a partial (A46a) decline in Hot (the mortal would lose his
soul-stuff bit by bit). But there will also be fluctuations in the
presence of living substance in the mortal, with Hot decreasing, for
example, whenever the mortal is asleep (A46b). Simplicius uses
the plural and substantive term ‘the souls’ (tas psychas). If this
reflects Parmenidean usage (recall how closely Simplicius works
here from Parmenides’ text), then it supports the former alterna-
tive, that Parmenides explained the soul as a particular token of
Hot.27 More generally, we will see momentarily that what
Simplicius reports is almost certainly a doctrine of metempsycho-
sis. If so, then unless Parmenides advances a very idiosyncratic
subversion of notions of metempsychosis, this in itself suggests
that they are the same souls, i.e. for Parmenides, the same unified,
individuated tokens of Hot, which the goddess conveys to and fro
at birth and death, as opposed to ever-changing batches of Hot,

moreover, it is not the goddess but the Heliades who ‘convey’ the kouros (B1.8) and
nothing there parallels the notion, which takes centre stage in Simplicius’ brief state-
ment, of a cyclical conveyance. There is, finally, the context (in Ph. 9.38.18–39.21):
Simplicius is illustrating the cosmological functions of the goddess (whom he identifies
as the cosmic efficient cause: 39.13) within that section or ‘account’ (τὸν λόγον, 39.10)
of the poem which belonged in the domain of ‘belief’ (δόξαν, δοκοῦν, κτλ, 38.20–6,
39.10–12), which came after the account of the intelligible in Alêtheia (συμπληρώσας
κτλ, 38.28–9) and which was introduced along with the two physical elements in B8.50–
61 (cited here by Simplicius) – i.e. within Doxa.
27
As does, for what it is worth, Aëtius’ report (A45) that Parmenides described the soul as
fiery.

235
 How Could Parmenides Have Written Alêtheia?
whose unity and identity are not retained through successive
incarnations. Admittedly, however, Parmenides need not have
indicated a clear position on this particular question. It is also
possible that Parmenides did identify Hot as ensouled stuff but still
spoke of particular ‘souls’ as enduringly individuated tokens of
this ensouled stuff.28
On either reading, Parmenides exemplifies what Gábor Betegh
has described as the widespread ‘portion model’ of the relation
between soul and cosmos, in which the soul comprises one or more
of the specific stuffs which also have cosmic functions (in this
case, ‘Hot’ or ‘Light’ or ‘fire’).29 Notably, Betegh highlights that
the portion model can express and address eschatological concerns
insofar as the soul’s stuff(s) – and, therefore, its origin and desti-
nation – are standardly taken to be superior and, often if not
invariably, divine.30 Indeed, in his Helen (1013–16, cited above),
Euripides employs something very much like the portion model to
explain that both the living and the dead can undergo punishment,
since the mind of the deceased retains immortal cognition as it
falls into immortal aether.31 Quite generally, in the passages
expressing aethereal eschatology which we considered in this
section, the eventual return of souls to the aether constituted a
happy fate (and, in some cases, a divine reward) for those parti-
cular souls. Nothing prevents the portion model from accommo-
dating a notion of the persistent unity of the soul. Individual souls
can comprise particular, stable tokens of some stuff(s), as is most

28
Betegh (2007) 6–13, 24–7 shows that Heraclitus uses psychê both as a mass term
denoting a kind of cosmic stuff (ψυχή, B36), commensurable with ‘water’ and ‘earth’,
and as a count noun (ψυχῇσιν, B36; cf. Simplicius’ plural τὰς ψυχάς). The Hippocratic
treatise de Hebdomadibus offers a partial parallel to what I reconstructed above: at
death, ‘the breath of the hot (τοῦ θερμοῦ) departs as a whole back to the whole’, a process
glossed, apparently interchangeably, as the soul (ἡ ψυχή) leaving behind the cold, mortal
corpse (τὸ ψυχρὸν καὶ τὸ θνητόν), Hebd. 52; cf. also Nat. Hom. 3. Heat causes life, and
its departure death, also in Lucretius, DRN 3.121–9.
29
Betegh (2006b), not discussing Parmenides. The proem (B1), though, evinces
what Betegh calls the ‘journey model’ of the relation between soul and cosmos,
see Ch. 5.3. Betegh (2006b) 43–6 observes that the two models are juxtaposed in both
Empedocles and the Derveni papyrus. Cf. also Betegh (2004) 346.
30
Betegh (2006b) esp. 32, 36–8, discussing in particular frr. 226 and 228a Kern = 437 and
436 Bernabé.
31
The other Euripidean passages cited above leave no doubt that, in Hel. 1013–16 too, the
mind is implied to be aethereal also in origin.

236
 5.2 Parmenides on the Soul
probably the case with Parmenides’ soul, and as is indeed the case,
mutatis mutandis, with Euripides’ mind in his Helen.
We know that Parmenides developed a naturalistic embryolo-
gical theory in his Doxa (B17–18; A53–4). One might wonder,
then, whether Parmenides’ doctrine of metempsychosis and soul-
conveyance could fit with this embryology. Did Parmenides main-
tain a coherent and consistent account of the formation of living
mortals? In fact, the two theories hang together closely, and the
connections between them are illuminating. Above all, it is again
the governing goddess who administers not only the conveyance
of souls but also the processes of copulation and birth (B12.3–6):
δαίμων ἣ πάντα κυβερνᾷ· 3
πάντα γὰρ <ἣ> στυγεροῖο τόκου καὶ μίξιος ἄρχει
πέμπουσ’ ἄρσενι θῆλυ μιγῆν τό τ’ ἐναντίον αὖτις 5
ἄρσεν θηλυτέρῷ
. . . The divinity who steers all things.
For she rules over the hateful birth and mixture of all things,
Conveying the female to mingle with the male and again contrariwise
The male to mingle with the female.
Parmenides’ striking description of birth as ‘hateful’ has been
convincingly related to current discourses of metempsychosis.32
Indeed, the term for ‘hateful’ here (στυγερός) is a common and
traditional epithet for death and Hades.33 Parmenides’ pointedly
inverted portrayal of birth as ‘hateful’ accords with eschatological
ideas that our incarnate life is flanked by a superior prenatal and
post-mortem existence. A fourth-century Gold Tablet from Thurii
in South Italy holds out to the initiate the prospect that he may ‘fly
out of the painful, grievous circle’ (κύκλō δ’ ἐξέπταν βαρυπενθέος
ἀργαλέοιο, GJ 5.5). The phrase is plausibly taken to express final
release from the burden of cyclical reincarnations. Proclus cites
the ritual Orphic prayer ‘to cease from the circle and to have
32
Burkert (1972) 284; Coxon (2009) 372.
33
στυγερὸν θάνατον, Od. 24.414; Aesch. Ch. 1008; Eur. Med. 994; πάντες . . . στυγεροὶ
θάνατοι, Od. 12.341; στυγερὸς Ἀίδης, Il. 8.368 (cf. h.Hom. 2.395; Il. 20.61–5); κήρ . . .
στυγερή, Il. 23.78–9 (cf. Th. 211–12) Cf. Il. 23.48 for the funeral feast as στυγερή.
Seeing especially that the traditional ‘hateful death’ (στυγερὸν θάνατον) occurs at 994,
Medea’s description of life as hateful at 145–7 (βιοτὰν στυγεράν) is a pointedly para-
doxical inversion (occasioned by her unique circumstances), reminiscent of the inver-
sion in Parmenides B12.4. Parmenides’ description of birth as ‘hateful’ could of course
be overdetermined, explicable in terms of both eschatology and physical pain.

237
 How Could Parmenides Have Written Alêtheia?
reprieve of evil’ (κύκλου τ’ ἂν λήξαι καὶ ἀναπνεύσαι κακότητος) as
the aim of the mysteries of Dionysus and Korê (fr. 229 Kern = 348
Bernabé).34 A Pythagorean acusma, which probably predates
Aristotle, pronounces pains good and pleasures bad because mor-
tals enter this life as a punishment (Iamblichus, VP 85).35 A
generation after Parmenides, Empedocles recounts his own
entrance into human life as a traumatic and lamentable event: ‘I
wept and I wailed when I saw the unfamiliar place’ (DK31
B118).36
Notably, Simplicius, when describing the goddess’s convey-
ance of souls, employs the very same verb (πέμπειν, in Ph.
9.39.19) which Parmenides uses for her ministration of copulation
(πέμπουσ’, B12.5). Parmenides also uses the same verb in the
proem for the conveyance of the youth by the Daughters of the Sun
(πέμπειν, B1.8; cf. προὔπεμπε, B1.26). As we will see in detail in
the next section, the proem figures the youth’s journey to the
goddess as that of the discarnate, post-mortem soul. In short, it is
the same terminology that describes the goddess’s ministration of
the different processes (transmigratory and embryological) which
govern the formation and dissolution of particular mortal lives.
(The recurrence of the term in B1.8, 26 and B12.5 makes it likely
that Simplicius’ use of it tracks Parmenides’ own language.)
Finally, in Simplicius’ discussion, his citation of the first three
lines of the fragment whose second half we just cited and exam-
ined (B12.1–3), and his description of Parmenides’ theory of soul-
conveyance, are separated only by his citation of B13: ‘first of all
the gods, then, she devised Eros’ (πρώτιστον μὲν Ἔρωτα θεῶν
μητίσσατο πάντων).37 Now, this fragment is chiefly concerned
with Doxa’s lost theogony (Simplicius introduces it with the

34
Cf. Johnston (2013b) 127; Bremmer (2014) 75–6.
35
Burkert (1972) 166ff (note 168, n.14) convincingly argues that Iamb. VP 82–6 derives
from Aristotle, who, indeed, very nearly cites this acusma at Protr. fr. 106 Düring (apud
Iamb. Protr. 47.25); cf. Dillon and Hershbell (1991) 107, n.3; Huffman (2009) 36, n.43;
cf. also Philol. DK44 B14.
36
With Clem. Strom. 3.3.14; Wright (1995), ad loc.; Burkert (1972) 284.
37
In Ph. 9.39.14–19. Simplicius then adds an ‘etc.’ (καὶ τὰ ἑξῆς), and immediately reports
the theory of soul-conveyance (19–20). The proximity in Simplicius’ account may well
reflect proximity in Parmenides’ poem. By comparison, Simplicius then follows a
citation of B8.26–8 immediately with a citation of B8.30–3 (9.39.27–9.40.6).
Simplicius cites B12.2–6 some pages earlier (in Ph. 9.31.13–17).

238
 5.2 Parmenides on the Soul
words: ‘and he says that she is also the cause of gods, saying . . . ’,
in Ph. 9.39.17–18), and it was no doubt chosen for quotation by
Simplicius partly under the influence of Plato’s interest in it (Symp.
178b). But this statement about Eros could also plausibly have
related to Parmenides’ account of copulation.38 Some lines from
Parmenides’ embryology, rendered into Latin by Caelius
Aurelianus, speak of ‘the buds of Venus’ (Veneris . . . germina,
B18). Caelius, therefore, shows that Parmenides associated inse-
mination with the very same divine milieu to which Eros
belonged, regardless of precisely which Greek appellation under-
lies his Venus.39
In sum, Parmenides’ embryology and his doctrine of soul-con-
veyance are not only compatible but hang together closely. The
goddess’s direct administration of intercourse and birth (ἄρχει |
πέμπουσ’, B12.4–5), and perhaps also her indirect facilitation of
these processes through the creation of Eros (μητίσσατο, B13),
cohere with her eschatological role of conveying souls (καὶ τὰς
ψυχὰς πέμπειν κτλ, in Ph. 9.39.19–20). These are constituents of a
unified account, which explicates the formation and dissolution of
living mortals through divine agency.
Erwin Rohde, and some subsequent scholars, thought that the
coexistence of the theory of metempsychosis reported by
Simplicius with Parmenides’ account of human cognition in B16
poses special difficulties.40 But there is no special problem in
accounting, on the basis of these two texts, for the relation in
Parmenides between soul (psychê, or whatever the operative
term in Parmenides’ eschatology was) and mind or cognition

38
As Coxon (2009) 372 suggests.
39
Caelius’ stipulation latinos enim ut potui simili modo composui, ne linguarum ratio
misceretur (B18) should caution against, rather than encourage, drawing seemingly self-
evident inferences. ‘Aphrodite’ is of course possible but not obvious. There is no clear
evidence that Parmenides ever made direct reference to any member of the Olympian
pantheon. While Plutarch interprets the maker of Eros in Parmenides as Aphrodite (Mor.
756e10–11 = B13), Aëtius refers to the goddess as Necessity (Ἀνάγκη), Fate
(Εἱμαρμένη), Justice (Δίκη) and Providence (Πρόνοια; A32, A37; cf. Coxon (2009) 280).
40
Rohde (1925) 373 assumes the equivalence of soul and mind in Parmenides and reasons
that, given B16, the theory of metempsychosis is an inconsistent adage, which
Parmenides, ‘not speaking as a physiologist, but as an adherent of the Orphic-
Pythagorean theosophy’, imported without modification. Deichgräber (1959) 716 and
Mansfeld (1964) 174, whose alternatives I criticised above (n.26), allow Rohde’s view
as a possibility.

239
 How Could Parmenides Have Written Alêtheia?
(noos). As we saw in Chapter 4.2, we register things, according to
Parmenides, through both elements. By the Hot in us we cognise
Hot things, and by the Cold in us we cognise Cold things. In the
living mortal, who comprises mixtures of both elements, mind or
cognition (noos) will thus be a function of the activity of both
elements. The soul, therefore, either as a particular, delimited
token of Hot or as the sum of ensouled Hot in the mortal at any
given moment, will play a non-exhaustive role in the living mor-
tal’s mind.
But what of Theophrastus’ remark that, according to Parmenides,
‘the soul and the mind are one and the same’ (τὴν ψυχὴν καὶ τὸν
νοῦν ταὐτὸν εἶναι, A1.11–12)? Parallels from Peripatetic doxogra-
phy elsewhere make clear that this remark is merely criticising
Parmenides in formulaic terms for failing to draw a particular
Peripatetic conceptual distinction.41 Aristotle ascribes the very
same identification to Democritus (ταὐτὸν ψυχὴν καὶ νοῦν) simply
on the alleged basis (γάρ) that Democritus regarded what appears as
true (de An. 1.2. 404a27–9). In doing so, Democritus failed to isolate
mind as a particular capacity in relation to truth (404a30–1). We saw
in Chapter 4.3 that, in Metaphysics Gamma, Aristotle cites B16 to
assimilate Parmenides to an alleged ancient consensus, which
included also Democritus among others, according to which what
appears is true, and sense-perception is ‘cognition’ or ‘knowledge’
(φρόνησις, 1009b12–1010a3). As we noted, when Aristotle later
reprises and develops this same discussion in On the Soul (3.3
427a17-b14), he asserts more explicitly the stronger view that the
ancients identified knowing (τὸ φρονεῖν) and perceiving (τὸ
αἰσθάνεσθαι) as one and the same thing. The ancients thus failed
to isolate the perceptual activities of the soul from the mind’s
distinctive activity (τὸ νοεῖν), in which there is both ‘correctly’
and ‘incorrectly’ (τὸ ὀρθῶς καὶ τὸ μὴ ὀρθῶς), with the former
including cognition (φρόνησις, 427b6–11; cf. 427a21–8).42 In sum,
an established Peripatetic-doxographic tradition cited Parmenides’
theory of human cognition in B16 to implicate him in the view that
all impressions are true, and identified that same view with the
41
Mansfeld (1964) 172 interprets similarly; cf. Tarán (1965) 252, n.59.
42
For Aristotle himself on the relation between nous and psychê, see de An. 429a10–430b32;
cf. 415a11–12.

240
 5.2 Parmenides on the Soul
failure to distinguish soul and mind. Theophrastus’ reported remark
that Parmenides identified soul and mind conveys this same,
Peripatetic criticism of Parmenides. We saw in Chapter 4.3 that
his statement in On the Senses 4.1, that Parmenides speaks of
perceiving and cognising as the same thing (τὸ γάρ . . . ὡς ταὐτὸ
λέγει), seems heavily indebted to the same passage in Aristotle’s
Metaphysics Gamma, and is best read as the criticism that
Parmenides failed to isolate what a Peripatetic would distinguish
as sense-perception and cognition. Indeed, this sentence from On
the Senses could be, at some remove, what Diogenes Laertius ends
up reporting as Theophrastus’ claim that Parmenides identified soul
and mind.43
Is there any truth at all, then, in the remark that Parmenides
assimilated soul and mind? The remark is perhaps partially true to
the extent that the soul of the living mortal is (or is part of) the Hot
in him, and the living mortal not only perceives but also cognises
with the Hot in him. As we saw in Chapter 4.3, Parmenides’ theory
in B16 expounds not merely the perceptual experiences of humans
but, more broadly, the mechanisms which determine the sorts of
things of which we are aware as well as our beliefs and judgements
concerning those things. It explains how mind (noos) is present to
mortals and how the human frame cognises (phroneei). At the
same time, Theophrastus’ reported remark is unjustified insofar as
Parmenides does not reduce mind, which at least in the living
mortal involves both Hot and Cold, to soul. As we know from
Theophrastus’ own report in On the Senses (4.4–6), the corpse, at
any rate, has no Hot soul in it, but nonetheless perceives-and-
cognises (Cold things). And if the notion that something can be
lifeless but perceptive strikes us as excessively counterintuitive,
we should recall Euripides’ explicit expression of the idea that ‘the
mind of the dead does not live, but does have immortal thought’
(Hel. 1014–16). We would be viewing Parmenides (and Euripides)
through our own, etic category of ‘life’ – which would not in itself
be a hermeneutic sin, of course – if we insisted that, if the corpse is
perceptive, then it must retain some sort of life. At any rate, for
43
A later (and no doubt derivative) expression of this same doxographic tradition, Aëtius
A45 indiscriminately lumps Parmenides together with Democritus – as well as
Empedocles – as thinkers who identified soul and mind.

241
 How Could Parmenides Have Written Alêtheia?
Parmenides, the Cold corpse is perceptive but not ensouled (fol-
lowing ‘the departure of fire’).
Simplicius couches the process of metempsychosis within some
sort of topography: ‘from the visible to the invisible, and then back
again’. Any elaboration of this topography, however, would be
necessarily speculative. The invisible (to aeides) is an obvious
allusion to Hades (the etymological association with invisibility is
as old as Il. 5.845), but the topographical implications of this
allusion are unclear. Neither here nor in our discussion of the
proem will we rely on inevitably speculative and inconclusive
reconstructions of Parmenides’ eschatological topography.44
Now, whether or not Parmenides made aether the source or
destination of the prenatal and post-mortem soul, we can conclude
that he developed what is recognisably a version of the familiar,
contemporary eschatological conception of the soul as aethereal in
nature, and as incarnated at birth in bodies, which comprise also
another, axiologically lower (earthy) element. Parmenides con-
ceived of the element Light or Hot as, in itself and regardless of its
spatial location, ‘aethereal’, inasmuch as it is the element espe-
cially characteristic of the aether, understood as the higher and
Hotter celestial region of the universe. We have no direct evidence
that he described the lower element as ‘earthy’, although there is a
widespread assumption to this effect in the doxographic tradition,
and he certainly described it as cold, heavy and dark. He took the
Hot to be the cause of life, separated from the Cold at death. He
conceived of the soul as Hot and, therefore, aethereal (or, possibly
but less likely, of the aethereal Hot as itself ensouled stuff).
Most importantly, we observed above that the divinity of the
soul is clearly implicit, and sometimes explicit, in the contempor-
ary eschatological conceptions of the soul as aethereal.45 We can
now turn more directly to the eschatological migrations and divi-
nity of the soul in Parmenides’ system.
That the cyclical conveyance of souls to and fro by a goddess
indeed expresses a doctrine of metempsychosis is overwhelmingly
likely. The cyclicality which Simplicius highlights (τὰς ψυχάς . . .
44
See the Appendix for a discussion of these issues.
45
One could have read the divinity of aether already into its association in epic with the
overdetermined adjective δῖος (‘heavenly’, ‘noble’, ‘divine’), e.g. Il. 16.365; Th. 697.

242
 5.2 Parmenides on the Soul
ποτὲ μέν . . . ποτὲ δὲ ἀνάπαλιν) indicates that the goddess conveys
from life to death ‘and then back again’ the same souls46 or,
minimally, different amalgamations of the same batches of enso-
uled Hot. Parallel accounts of cyclical metempsychosis strongly
favour the same conclusion. We already noted in this section the
fourth-century Gold Tablet from South Italy which speaks of
escape from the ‘painful, grievous circle’ (GJ 5.5; cf. κύκλου, fr.
229 Kern = 348 Bernabé). Pindar uses cyclical language very
reminiscent of Simplicius’ report when expressing a process of
metempsychosis which is administered by a female goddess
(Φερσεφόνα . . . ψυχὰς πάλιν, fr. 133 M apud Men. 81b8–10). So
does Plato’s Socrates when, before citing these lines from Pindar,
he comments on this doctrine (τοτὲ μὲν τελευτᾶν . . . τοτὲ δὲ πάλιν
γίγνεσθαι, Men. 81b4–5).
Parmenides’ theory of metempsychosis itself, like any such
theory, presupposes some notion of the soul’s divinity. As
Burkert puts it, by denying that the soul is subject to death, by
making it athanatos, metempsychosis renders ‘the epithet which
since Homer had characterised the gods in distinction from
men . . . the essential mark of the human person’.47 Indeed, even
if, for whatever reason, one was disinclined to allow that
Parmenides’ theory of soul-conveyance should be thought of as
a doctrine of metempsychosis, there is no question that, in this
psychological scheme, our souls precede our birth and survive our
death. The soul cannot, therefore, be mortal in the same sense that
living humans are. Indeed, Simplicius makes no suggestion that
the cyclical eschatological process (ποτέ . . . ποτὲ δὲ ἀνάπαλιν)
was delimited in any way.48
We may further support the plausibility of this conclusion, that
Parmenides considered the transmigrating, aethereal soul divine,
46
Pace Tarán (1965) 248–9, n.51, rightly criticised on this point by Kurfess (2016)
41, n.63.
47
Burkert (1985) 300; cf. Bremmer (1983) 71–2. Metempsychosis and the soul’s divinity
are intimately connected e.g. at Hdt. 2.123 (ψυχὴ ἀθάνατός ἐστι) and Pl. Men. 81b3-6
(ψυχήν . . . ἀθάνατον), where Socrates ascribes these views to unnamed priests and
priestesses and to inspired poets. This connection, especially in contemporary South
Italy, is explored further in this and the following section.
48
My argument that Parmenides considered souls divine requires only this point, that they
precede the mortal’s birth and survive his death in this way. I will continue, however, to
refer to Parmenides’ doctrine of metempsychosis.

243
 How Could Parmenides Have Written Alêtheia?
by observing further instructive parallels for the divinity of the
soul, or of (hot) aether, or of the hot or aethereal soul.
Anaximenes may have already related our soul to his divine air,
which controls the cosmos (DK13 B2).49 Diogenes of Apollonia
(fifth century), who also adopts air as his key principle, describes
our soul as a portion of god (μόριον τοῦ θεοῦ, DK64 A19.42).
Alcmaeon of Croton, probably writing in the first half of the fifth
century, argued for the immortality of the soul, and for its similar-
ity to the divine celestial bodies, on the basis of its capacity for
everlasting self-motion (DK24 A12).50 Pherecydes of Syros, writ-
ing in the sixth century, is reported by Apponius to have divided
the soul into two spirits, heavenly (de caelo spiritum) and earthy
(terrenis seminibus comparatum, F86b Schibli = DK7 A5).
Scholars have prudently questioned the historicity of this report.51
And yet, Apponius’ account may well be a Neoplatonic and
Christian52 misinterpretation of the genuine view that the mortal
comprises both a divine, aethereal element and a mortal, earthy
one, the soul being identified with the former.53 This interpretation
would agree with the recurrent insistence in our testimonia that
Pherecydes (first) maintained both metempsychosis and the divi-
nity and immortality of the soul (as well as with the archaic
conception of the soul as aethereal).54
That Pythagoras and early Pythagoreans in South Italy
advanced notions of metempsychosis and, concomitantly, the
divinity of the soul is perhaps the only doctrinal statement that
can be made about early Pythagoreanism without reservation,
even if there is little prospect for crystallising a sharply defined
theory.55 In several Gold Tablets, the initiate’s soul exclaims to
49
See further Ch. 1.2.
50
Alc. DK24 B2 might express a notion of metempsychosis, see KRS (1983) 347–8; cf.
Burkert (1972) 296 with n.97–8; but contrast Guthrie (1962) 353–6.
51
See Schibli (1990) 109–12. Gomperz (1929) 24–5 accepts it.
52
See Schibli, loc. cit. 53 Vlastos (1970b) 110 with n.60 argues similarly.
54
Divinity: Cicero (perhaps drawing on Posidonius: see Schibli (1990) 104), F7 Schibli =
DK7 A5; cf. F48; F85a-b Schibli; metempsychosis: F2 Schibli = DK7 A2; F88 Schibli =
DK7 B6, with Schibli (1990) 105–6, 109, 112–13, 117–20, 123–5. Doxographers
regularly make Pherecydes Pythagoras’ teacher, F2; F11; F20; F48–9; F55; F84; F85b
Schibli.
55
See esp. Xenoph. DK21 B7; Ion DK36 B4; Hdt. 2.123; Ar. de An. 407b20–3; Theophr.
apud Por. Abst. 3.26.1–4; with Burkert (1972) 121–36; cf. Schofield (1991) 25–7;
Vlastos (1970b) 110–11, n.62; Kahn (2001) 18; Huffman (2009) 21; Lloyd (2014)

244
 5.2 Parmenides on the Soul
eschatological divinities ‘I am a child of Earth and starry Sky
(Γῆς . . . καὶ Οὐρανοῦ), but my race is heavenly (γένος οὐράνιον)’,
and, in others, ‘For I also claim to be of your blessed race (ὑμῶν
γένος).’56 In the next section, we will see that there are significant
affinities between Parmenides’ proem and the Tablets, and that
these formulae in the Tablets, spoken by the initiate’s soul, concern
his divinisation.
Pindar, reflecting on both the gulf and the affinity between
mortal and divine, writes: ‘from a single mother we both draw
our breath’ (N 6.1–2). Elsewhere, Pindar speaks of an image of
life, which survives the mortal’s death and exercises divinatory
powers in his sleep, since it alone comes from the gods (fr. 131 M).
Plato, as we saw, cites Pindar’s description of a retributive
metempsychosis when himself expounding metempsychosis and
the soul’s immortality.57 The Hippocratic author of On Fleshes
understands the Hot to be an immortal substance, which appre-
hends and knows all things and which, he says, is what ‘the
ancients’ (οἱ παλαιοί) named ‘aether’ (Carn. 2 = DK64 C3).
Aristotle is following in this broad tradition when he maintains
that the power of every soul shares in a divine Hot, which is
analogous to the element of the stars (i.e. aether).58
Numerous further examples could be cited (Stoic psychology
being an obvious if later case). But we can conclude quite gen-
erally that the divinity of the soul, especially in the context of
metempsychosis and in South Italy, was already maintained before
Parmenides’ time, and that the more specific conception of the

28–32, 43. Zhmud (2012) 221–38 maintains that Pythagoras took metempsychosis from
the Orphic traditions. As Schofield (loc. cit.) and Huffman (2009) 34–5 argue, the
evidence from Xenophanes, Herodotus and Ion of Chios gives every reason to believe
that ‘psychê’ was the operative term in Pythagorean metempsychosis. As far as our
evidence from Simplicius goes, this is most probably the case also in Parmenides’
doctrine of metempsychosis. I return in the next section to Parmenides’ relation to the
early Pythagoreans. Wilamowitz (1959) vol.2 209, when rejecting the possibility of
Pythagorean influence, wrongly asserts: ‘von der Seele, gar von ihrer Unsterblichkeit
und Wanderung, weiß Parmenides nichts’.
56
See nn.139, 141 below.
57
On Pindar’s eschatology, see further Johnston (2013b) 100ff.
58
GA 736b29–38, with Solmsen (1957) 121–2 on the peculiarities of the passage and the
possible motivations underlying it; cf. Burkert (1972) 362, n.60. Aristotle, at least as
paraphrased by Sextus at M 9.20–1, even argues that our notion of the divine was partly
derived from observations about the soul.

245
 How Could Parmenides Have Written Alêtheia?
(hot) aethereal soul as divine was current both during and after his
generation, if not espoused already by Pherecydes precisely in the
context of metempsychosis.
Before we move on from this survey, though, we might consider
finally the possibility of a dualistic conception of the mortal in the
archaic Orphic tradition. An Orphic anthropogony seems to sug-
gest some sort of Titanic provenance for, or element in, mortals by
tracing the origin of mankind to the ashes of the Titans, who were
struck by Zeus’ thunderbolt in punishment for tearing apart and
devouring Persephone’s son, the infant Dionysus. For
Olympiodorus, the sixth-century CE commentator who first relays
the myth in something like a complete form, the story even
indicates a Dionysiac element in mortals: ‘. . . our body is
Dionysiac; for we are a part of him, if indeed we consist of the
soot from the Titans who tasted of his flesh’ (in Phd. 1.3). In the
verses of Pindar which Plato cites in his Meno (81b8-c4 = fr. 133
M), the transmigrations of mortal souls are related to some process
of offence and recompense: ‘Persephone will receive requital for
an ancient grief.’ Plato ascribes these same ideas of an immortal
soul and metempsychosis to ‘priests and priestesses who are con-
cerned to be able to give an account (logos) of their practices’
(Men. 81a10–12). Plato’s citation from Pindar leaves the cause of
Persephone’s grief unspecified, and it is at least possible (though
hardly obvious) that it refers to some version of the Dionysus
myth.59 If so, then an Orphic conception of the human as compris-
ing in part a Titanic or divine component may go back to the
archaic period. It remains and will remain debated, however,
whether this Orphic anthropogony in fact belongs to the archaic
stratum of Orphism or was a (much) later invention.60

59
See Johnston (2013a) 69
60
Johnston (2013a) esp. 66–9 argues for a late sixth- or early fifth-century BCE date for
the myth’s debut; cf. also Bremmer (2014) 62–3, 76. For the view that the Dionysus
myth cannot be traced back to early Orphic traditions, that Persephone’s grief in Pindar
refers rather to her abduction by Hades and that Olympiodorus himself largely concocts
the myth which he relays, see Edmonds (2013) 296–391. Some Orphic fragments (226
and 228a Kern = 437 and 436 Bernabé) express the view that the soul comes from,
consists of, and returns to aether; see Betegh (2006b) 36–8. For a survey of Greek
notions of an afterlife among the stars, or ‘astral immortality’, see Burkert (1972)
357–68.

246
 5.2 Parmenides on the Soul
Having already argued that Parmenides considered the soul
divine and subject to eschatological migrations, I turn to some
intriguing remarks by the Platonist philosopher Numenius, which
Porphyry records in his On the Cave of the Nymphs (22.6–23.7 =
Coxon (2009) Testim. 133):
Now the theologians posited Cancer and Capricorn as these two gates, while
Plato spoke of two mouths. Of these, Cancer is the one through which souls
descend, and Capricorn is the one through which they ascend. Cancer is northerly
and is used for descent, while Capricorn is southerly and is used for ascent. The
northerly parts are for souls descending to birth (ψυχῶν εἰς γένεσιν κατιουσῶν),
and it is right that the gates of the cave [sc. in Homer] towards the north are the
places where humans descend, while the southerly ones are not for the gods but
for those ascending to the gods (τῶν εἰς θεοὺς ἀνιουσῶν). For the same reason, he
[sc. Homer] did not say that it is a way of gods, but of immortals (ἀθανάτων) –
which is a term used in common also for souls (ἐπὶ ψυχῶν), which are immortal
either per se or by their essence. These two gates he [sc. Numenius] says that
Parmenides too recalled in his Physics and also the Romans and Egyptians.

Numenius aligns Parmenides with an alleged tradition that posits

two celestial gates by which souls descend ‘to birth’ (εἰς γένεσιν)
and ascend ‘to gods’ (εἰς θεούς). Now, unlike Simplicius’ report of
Parmenides’ doctrine of soul-conveyance, Numenius’ report is not
independently reliable. Numenius indiscriminately situates
Parmenides within what he dubiously styles a widespread tradition
spanning also the ‘theologoi’, Plato, Homer and ‘the Romans and
Egyptians’. Indeed, Numenius enlists Homer in this tradition
simply on the basis of his distinction between two sets of gates
which lead up to the Nymph’s Cave, northern gates designed for
mortal usage (καταιβαταὶ ἀνθρώποισιν) and ‘more divine’ south-
ern ones (θεώτεραι, Od. 13.110–12).61 First, however, the very
eclecticism of Numenius’ list renders it unlikely for the choice of
Parmenides to be simply arbitrary. More importantly, our evidence
from Simplicius, as analysed above, does corroborate Numenius’
ascription to Parmenides of the two distinct psychic journeys. The
first journey mentioned by Numenius is explicitly ‘to birth’, while
the latter (‘to gods’) no doubt follows the mortal’s death (as the
reference to Plato’s Myth of Er confirms). Even the vocabulary of

61
The ‘two mouths’ (δύο στόμια) refer to Plato’s ‘openings’ (χάσματα) in the Myth of Er
(Rep. 10.614c2); we get στομίου at 10.615d5.

247
 How Could Parmenides Have Written Alêtheia?
gates (although not their identification as northern and southern, or
their portrayal as axiologically asymmetrical) may be supported
or, perhaps, motivated by the proem’s two aethereal gates of night
and day (πύλαι . . . αἰθέριαι, B1.11–13; cf. θυρέτροις, B1.13).62
This is especially the case since the proem’s imagery echoes
Doxa’s cosmological framework and would rightly be styled
eschatological, insofar as the journey of the kouros is represented
in terms of the journey of the post-mortem, discarnate soul.63 In
the light of all of this, it is noteworthy that Numenius emphatically
accentuates the divinity of the migrating soul, which precedes our
birth and survives our death, praising Homer for describing the
way upwards with terms that are not restricted to gods, but extend
also to other divine things, such as souls.64
Can we say anything about the physiological composition of the
goddess herself? It has been suggested that the goddess should be
identified with the aether, and some evidence inconclusively
favours this tempting identification.65 But we will not assume
this hypothesis here.66 Nonetheless, it is worth observing the
conjunction of two points in the light of, and in further support
of, our conclusion that Parmenides associated the Hot with life and
with the divine. First, Parmenides strikingly associates the divine
with the female. Aside from the goddess herself, the proem boasts

62
Cf. Coxon (1986) 163, 241–2.
63
For both points, see Ch. 5.3. For the point that the proem’s journey takes place within the
framework of Doxa, see the introduction to the chapters on Parmenides, n.6.
64
Homer in fact employs the intriguing comparative ‘more divine’ (θεώτεραι).
65
According to Cicero, Parmenides posited a ring of light which confined heat and girdled
the sky (caelum) and called it a god (A37). According to Aëtius’ fuller summary at A37,
the fiery sky (τὸ πυρῶδες . . . οὐρανόν) was placed under and governed by the aether,
which should thus presumably be identified with Cicero’s divine ring of light. Both
accounts echo Parmenides’ description of the celestial ‘goddess who governs all things’
(B12.3); see further Coxon (2009) 14–15, 280–1, 363–70.
66
Especially on the assumption that the revelatory goddess of B1 and the governing
goddess of Doxa are one and the same (by far the most economical and natural view,
although not one on which, as it happens, any of the arguments advanced in these
chapters depend), one might raise an epistemological challenge to the identification of
the goddess with aether. If the goddess’s constitution involves purely Light and in no
way any Night, then, given the principle of the perceptual cognition of like by like, how
can she govern, or disclose the nature of, Night and objects which involve that element?
But I can see no strong reason to think that Parmenides was ever struck by the question
of the physiological composition of the goddess herself or of her own knowledge of
things, any more than Hesiod addressed the question of his Muses’ source of knowledge
(cf. Ch. 2.2, n.76).

248
 5.2 Parmenides on the Soul
an elaborate, all-female divine cast (including most prominently
the Daughters of the Sun), which conveys the kouros to the god-
dess. Second, Aristotle reports that, according to Parmenides, the
female’s constitution is hotter than the male’s (A52; cf. A53).67
We saw in the previous chapter that Parmenides emphatically
represents knowledge of what-is as divine knowledge. He con-
siders it unattainable by the human mind, which is limited to the
qualitatively different sort of cognition that underpins the cosmol-
ogy of Doxa. But we saw in this section that Parmenides identifies
a divine constituent within the mortal – his soul – which comprises
only Hot, is detached at death from the Cold, mortal coil and
persists across successive incarnations. We may also recall here
Theophrastus’ report that the thought of Hot through Hot is a
better and purer kind of cognitive act than the thought of Cold
through Cold (Sens. 3.4). Recall also that Parmenides traced, not
only death, but also old age and forgetfulness to a deficiency in
Hot, while, correspondingly, not only vitality but also memory
involved its preponderance.68 Theophrastus’ statement, then, sug-
gests that a hotter constitution is characterised by a superior mental
disposition and superior mental activity. Hot thought constitutes a
more vital, healthy, acute and effective kind of cognitive act and,
consequently, a preponderance of Hot means mental activity of
this nature, while a preponderance of Cold, the cause of forgetful-
ness and old age (senility?), constitutes a general deterioration in
one’s mental faculties.69
As we saw in Chapter 5.1, we cannot identify what-is with Light
or some quality of Light, and it cannot be the case that thinking
Light is itself nothing other than thinking what-is. The axiological
nature of the contrast between Light and Night in Doxa does not
support Aristotle’s mistaken alignment of Light with what-is and

67
On the associations within Parmenides’ cosmology between the female, Light, vitality
and the divine, see further Journée (2012).
68
Ch. 4.2, with n.21.
69
Night can be characterised as ‘ignorant’ (ἀδαῆ, B8.59), then, insofar as it brings about
such cognitive and mental incapacity, and allows only the inferior cognition of Cold
through Cold, and so not of all that ensouled things register: light, heat, sound, etc.
(Sens. 4.4–6). Vlastos (1946) 68, 72 similarly sees Hot as the ‘active aspect of the mortal
mind, memory and judgement’; cf. also Tarán (1965) 259; Laks (1990) 14, n.40; Journée
(2012) 308–10.

249
 How Could Parmenides Have Written Alêtheia?
Night with what-is-not, although it may certainly explain how
Aristotle came by this view. Nonetheless, it is pertinent and
important that, in his cosmology, psychology, physiology and
eschatology, Parmenides thinks of the mortal agent as a complex
thing that comprises not only a mortal but also a divine part, which
is independent of and prior to the living mortal and is, furthermore,
associated with a superior cognitive and mental disposition. If the
living mortal could somehow promote the divine part within
himself against the mortal and attain (however momentarily) the
cognitive and mental condition of the discarnate, post-mortem,
pure Hot soul, then he would be attaining the cognitive and mental
condition of a divine thing.
The proem to Parmenides’ poem will further substantiate the
conclusions at which we have arrived in this section concerning
the soul’s physiological makeup and its divinity. Furthermore,
we will find that it characterises the kouros’ journey as the
eschatological journey of the discarnate, post-mortem soul, and
highlights his unusual privilege in undertaking that journey
while still alive. It represents, moreover, the kouros’ attainment
of the disposition which is necessary for acquiring knowledge
of what-is as a cognitive identification with his divine soul. The
proem also offers some insight into how Parmenides thought
such cognitive divinisation was possible for the living mortal,
and will best prepare us for identifying more precisely what
this divinisation involves. It is to the proem, then, that we shall
now turn.

5.3 The Proem in Its Time and Place

Notions of becoming like god (homoiôsis theôi) as the end of the
ideal philosophical life play significant roles in many explorations
of theology, epistemology and ethics in ancient Greece. Plato
prescribes assimilation to the divine on several occasions and in
different contexts. In the most sustained and instructive passage,
he maintains that the philosopher, who constantly exercises the
divine, rational soul-part within him (to which Plato refers as his
internal daimôn, Tim. 90c4–6), will attain divine as opposed to
mortal thoughts and cognition (φρονεῖν μὲν ἀθάνατα καὶ θεῖα, Tim.
250
 5.3 The Proem in Its Time and Place
90c1; cf. δόγματα, Tim. 90b3), and so partake of divinity insofar as
this is possible for a human.70 Aristotle similarly enjoins that ‘we
must not heed those who say that a human must think and cognise
(φρονεῖν) human things, and a mortal, mortal things. Rather, we
should become immortal so far as possible (ἐφ’ ὅσον ἐνδέχεται
ἀθανατίζειν)’, adding: ‘it may even be held that this is what each
one is, inasmuch as it is the better and authoritative part’ (EN 10.7
1177b31–1178a3).71 Similar (and simultaneously divergent) des-
cendants may be cited from a wide array of later Greek and
monotheistic philosophies.72 For all their differences, these mod-
els all take up and develop a pre-existing, pervasive and deep-
seated theology of approximation – as I will refer to it – which
construes the boundary between mortal and divine as a malleable
and negotiable one, which understands the height of mortal excel-
lence as an approximation to, or attainment of, divine qualities or
capacities, and which already dominates the Iliad. They do so
(with the idiosyncratic exception of the Epicurean model) by
adopting the fundamental principle that the human being com-
prises a mixture of divine and non-divine parts and, correspond-
ingly, capacities. The more one promotes the divine aspects of
one’s complex identity, the closer one assimilates oneself to divi-
nity in one’s cognition and/or character.

70
Correspondingly, by exercising our mortal, non-rational soul-parts we acquire mortal
cognition and become as mortal as a mortal can, Tim. 90b1-c6; cf. Tht. 176a8-b2
(ὁμοίωσις θεῷ κατὰ τὸ δυνατόν); Rep. 6.500c8-d1; see further Sedley (1999b); Annas
(1999) esp. 52–65; Armstrong (2004).
71
Cf. also fr. 61 Rose. See Sedley (1999b) 324–8. For the theology and morality which
Aristotle targets here, see Epicharm. DK23 B20; cf. also Pind. I 5.16; Eur. Alc. 799.
72
For later Platonists, see e.g. Alcin. Didasc. 28.3–4; Plotin. Enn. 1.2.5–7; Apul. de Plat.
2.23.30–2; cf. Xenoc. fr. 236 Isnardi Parente, with Annas (1999) esp. 59, 63–71; Sedley
(1999b) 322–3. For Stoics, see e.g. Sen. Ep. 92.3–4; Epic. Disc. 2.14.12–13 (with Long
(2002) 145–7, 156–8, 170–2); cf. Plut. Mor. 1076a. For the more idiosyncratic,
Epicurean take, e.g. Sent. Vat. 33; Men. 135.5–9; Lucr. DRN 5.8 (cf. 3.322);
Philodem. fr. 386 Usener; Diog. Oen. fr. 52 col.4 Chilton, with Warren (2000) esp.
260–1; Erler (2002); cf. Most (2003) 314; Kenney (1971) 1–3. For ancient Jewish and
Christian variants, see e.g. Philo Fug. 62–4 (Annas (1999) 65–6, n.37 cites further
passages); Clem. Strom. 2.19; 2.22; 5.14.94.4–95.2, with Russell (2006). On these and
such passages, cf. also Dodds (1951) 238, 255–6, n.6, 257, n.19; Most (2003) 313–16;
Betegh (2006a) 636–7 with n.17; Lenz (2007); Boys-Stones (2009) 15. For an illus-
trative example of the numerous medieval descendants, see the discussion in Kreisel
(1994) of Maimonides and his Greek and Islamic sources. Cases could be made also for
Empedocles (cf. Ch. 6.2) and Heraclitus, see Roloff (1970) 180–6, 192–7; Broadie
(1999) 219–20; Betegh, loc. cit.

251
 How Could Parmenides Have Written Alêtheia?
Conceptions of the human as a complex of mortal and divine
parts themselves emerge alongside metempsychosis already dur-
ing the archaic period (as we saw in the previous section). They
offer a new psychological and eschatological framework in which
to accommodate the widespread theology of approximation pro-
minent already in Homer. Xenophanes strikingly renounced these
fundamental aspects of Greek religious thought (see Chapter 3.5).
That Parmenides, conversely, understood the human to be a com-
plex of mortal and divine we already saw in our discussion of his
conception of the soul. In this section, I will argue further that
Parmenides espouses what we will rightly call his own model of
homoiôsis theôi in a manner that bears comparison with Plato and
some of his philosophical descendants. To attain divine knowl-
edge, the knower must himself be (qualifiedly, momentarily) divi-
nised by identifying himself with his divine soul. Parmenides will
emerge as a significant early exponent of the philosophical tradi-
tion of homoiôsis theôi.73
Parmenides’ poem opened as follows (B1):
ἵπποι ταί με φέρουσιν, ὅσον τ’ ἐπὶ θυμὸς ἱκάνοι, 1
πέμπον, ἐπεί μ’ ἐς ὁδὸν βῆσαν πολύφημον ἄγουσαι
δαίμονος, ἣ κατὰ πάντ’ ἄ<σ>τη74 φέρει εἰδότα φῶτα·
τῇ φερόμην· τῇ γάρ με πολύφραστοι φέρον ἵπποι
ἅρμα τιταίνουσαι, κοῦραι δ’ ὁδὸν ἡγεμόνευον. 5
ἄξων δ’ ἐν χνοίῃσιν ἵ<ει> σύριγγος ἀυτήν
αἰθόμενος (δοιοῖς γὰρ ἐπείγετο δινωτοῖσιν
κύκλοις ἀμφοτέρωθεν), ὅτε σπερχοίατο πέμπειν
Ἡλιάδες κοῦραι, προλιποῦσαι δώματα νυκτός
εἰς φάος, ὠσάμεναι κράτων ἄπο χερσὶ καλύπτρας. 10
ἔνθα πύλαι νυκτός τε καὶ ἤματός εἰσι κελεύθων,
καί σφας ὑπέρθυρον ἀμφὶς ἔχει καὶ λάινος οὐδός·
αὐταὶ δ’ αἰθέριαι πλῆνται μεγάλοισι θυρέτροις·
τῶν δὲ δίκη πολύποινος ἔχει κληῖδας ἀμοιβούς.
τὴν δὴ παρφάμεναι κοῦραι μαλακοῖσι λόγοισιν 15

73
The essential notion that the mortal’s achievement of the rational cognition of Alêtheia
involves his divinisation in some way has of course been suggested: Cornford (1952)
120; Roloff (1970) 171–7; Hussey (1990) 37; Göbel (2002) 158–66; Drozdek (2007)
46–8; Miller (2011) 43–57. These discussions do not relate this notion to the questions
about and the links between Parmenides’ epistemology, psychology and eschatology
addressed here, or develop it in the ways suggested below.
74
The corruption κατὰ †πάντ’ ἄτη† (B1.3) cannot be emended with confidence. I follow
noncommittally the traditional emendation ἄστη.

252
 5.3 The Proem in Its Time and Place
πεῖσαν ἐπιφραδέως, ὥς σφιν βαλανωτὸν ὀχῆα
ἀπτερέως ὤσειε πυλέων ἄπο· ταὶ δὲ θυρέτρων
χάσμ’ ἀχανὲς ποίησαν ἀναπτάμεναι πολυχάλκους
ἄξονας ἐν σύριγξιν ἀμοιβαδὸν εἰλίξασαι
γόμφοις καὶ περόνῃσιν ἀρηρότε· τῇ ῥα δι’ αὐτέων 20
ἰθὺς ἔχον κοῦραι κατ’ ἀμαξιτὸν ἅρμα καὶ ἵππους.
καί με θεὰ πρόφρων ὑπεδέξατο, χεῖρα δὲ χειρί
δεξιτερὴν ἕλεν, ὧδε δ’ ἔπος φάτο καί με προσηύδα·
ὦ κοῦρ’ ἀθανάτοισι συνάορος ἡνιόχοισιν,
ἵπποις ταί σε φέρουσιν ἱκάνων ἡμέτερον δῶ, 25
χαῖρ’, ἐπεὶ οὔτι σε μοῖρα κακὴ προὔπεμπε νέεσθαι
τήνδ’ ὁδόν (ἦ γὰρ ἀπ’ ἀνθρώπων ἐκτὸς πάτου ἐστίν),
ἀλλὰ θέμις τε δίκη τε. χρεὼ δέ σε πάντα πυθέσθαι
ἠμὲν ἀληθείης εὐκυκλέος ἀτρεμὲς ἦτορ
ἠδὲ βροτῶν δόξας, ταῖς οὐκ ἔνι πίστις ἀληθής. 30

The mares, which bear me as far as my heart might reach,

Conveyed me, when they brought me to the far-famed way
Of the divinity, that bears over all cities the man who knows.
By this way I was borne, for by this way the much-attending mares bore me
As they strained the chariot, and the maidens led the way.
And the axle in the naves sent forth the screech of a pipe
As it blazed, for it was driven by two whirling
Wheels on both sides, when the maidens, the Daughters of the Sun,
Made haste to convey me, after they had left behind the House of Night
For the Light, and had pushed back their veils from their faces with their
hands.
There stand the gates of the paths of Night and Day,
And a lintel encloses them and a stone threshold;
And the aethereal gates themselves fit into great doors.
And much-avenging Justice holds their alternating keys.
The maidens blandished her with gentle words
And cleverly convinced her that she should push away for them
The bolted bar quickly from the gates. And the gates made
A yawning gap in the doorway as they swung open, and caused the bronze
Hinges to rotate in turn in their sockets,
Fitted with pins and rivets; straight through them
The maidens held the chariot and the mares down the road.
And the goddess received me kindly, and took my right hand in hers
And spoke these words to me and addressed me as follows:
‘O youth, companion to immortal charioteers
And to the mares which bear you, you who arrive at my house,
Welcome, since it was not an evil fate which sent you forth to travel
This way, for indeed it is far from the track of humans,

253
 How Could Parmenides Have Written Alêtheia?
But it was Law and Justice who sent you. And it is right for you to learn all
things,
Both the unshaken heart of well-rounded reality,
And the beliefs of mortals, in which there is no real trust.’75
It has been well observed that the proem is ‘intrinsically vague’.76
We will not, I hope, lose sight of this salutary caveat. Our aim in
this section will by no means be to demystify the proem’s rich
mysteriousness and, indeed, obscurity, much less to subject – and
thereby render explicable – every detail of its elaborate imagery to
an exhaustive explanatory framework. The discussion that follows
will focus only on some aspects of the proem, analysing them both
internally and in the light of previously established conclusions
and certain external parallels and models, which constitute part of
the theological and conceptual framework within and against
which the kouros’ journey is described.
The topographical markers in the proem are sufficiently ambig-
uous, and the cultural models which it evokes so varied, that the
trajectory of the kouros’ journey may be viably interpreted as,
among other things, a descent (katabasis), an ascent (anabasis) or
inherently blurred. A systematic but inconclusive treatment of this
question will be reserved for the Appendix. I will not here pursue
or rely on any such inevitably speculative and unsettled recon-
struction. Instead, I will highlight several features of the kouros’
journey, which any tenable reconstruction of its trajectory would
have to take into account. We will emphasise and explore the
significance of two points in particular. First, Parmenides repre-
sents his journey as an eschatological one, the journey of the post-
mortem, discarnate soul, and, furthermore, evokes the imagery
and terminology of mystery initiations. Second, the kouros’ jour-
ney is made possible by divine powers, which are identified as
agents of Light, Heat and the higher, celestial and aethereal region,
and with which he himself is intimately connected.
The journey’s topography famously appropriates elements of
the Hesiodic Underworld (Th. 736ff). Parmenides’ ‘House of
Night’ (δώματα νυκτός, B1.9) recalls Hesiod’s Underworldly
75
For B1.31–2, see Ch. 4.4. This translation of the proem draws in particular on Graham
(2010) ad loc.
76
Mourelatos (2008a) 14.

254
 5.3 The Proem in Its Time and Place
‘abode of Night’ (Νυκτός . . . οἰκία, 744). In the proem, the kouros
passes through the gates of the paths of Night and Day (B1.11)
whose keys are described as ‘alternating’ (κληῖδας ἀμοιβούς,
B1.14; cf. ἀμοιβαδόν, B1.19). In Hesiod, Night and Day greet
one another as they alternately traverse a great threshold
(ἀμειβόμεναι μέγαν οὐδόν, 749; cf. Parmenides’ οὐδός at B1.12 in
the same metrical position), one descending into the House of
Night while the other ascends from it (748–54). Again, Hesiod’s
description of Tartarus as a ‘great chasm’ (χάσμα μέγ’, 740)
located beyond the infernal gates (πυλέων ἔντοσθε, 741) is echoed
in B1 in the yawning chasm, which is generated by the gates as
they swing open (πυλέων . . . χάσμ’ ἀχανές, B1.17–18).77 This
clear appropriation of Hesiod’s Beyond already indicates that the
kouros travels on an eschatological path to the setting of the
afterlife. Relatedly and significantly, in describing his path as
that of the ‘knowing man’ (εἰδότα φῶτα, B1.3), and in having
the goddess address him as kouros (B1.24), Parmenides is adopt-
ing the technical vocabulary of mystery initiations.78 We will
return below to mysteries and to initiatory eschatological journeys
(in particular to the journey of the initiate’s soul in the Gold
Tablets). For now, I note only that the language of mystery initia-
tion further supports the impression that the kouros’ journey is
assimilated to that of the post-mortem soul. This impression is
finally confirmed by the goddess’s reassurance that it is not an ‘evil
fate’ (μοῖρα κακή) that has conveyed Parmenides on this way
(τήνδ’ ὁδόν, B1.26–8). The phrase ‘evil fate’, as scholarly
77
The topological marker ‘there’ (ἔνθα, B1.11) structures throughout Hesiod’s account of
the Underworld: Th. 729, 734, 736, 758, 767, 775, 807, 811. For the parallels between
B1 and Th. 733–66, see Morrison (1955) 59–60; Burkert (1969) 8, 11–13; Dolin (1962)
96; Mourelatos (2008a) 15; Miller (2006) 7–8; Palmer (2009) 54–5; Kraus (2013) 454;
and esp. Pellikaan-Engel (1978) 8–10 for a painstaking exposition of the verbal echoes.
78
As Burkert (1969) 5 with n.11 observes, there are two fundamental cases in which εἰδώς
is used without an object: (i) when linked to a verbum dicendi (e.g. πρὸς εἰδότας λέγειν)
and (ii), of which (i) is an extension: in reference to the initiate; cf. Diels (2003) 49;
Jaeger (1947) 98; Bowra (1953c) 50–1; Coxon (2009) 272; Kingsley (1999) 62; Palmer
(2009) 58; Kraus (2013) 455; Ranzato (2015) 93. See [Eur.] Rhes. 971–3 (σεμνὸς τοῖσιν
εἰδόσιν θεός; cf. also Eur. Bacch. 72–4: ὦ μάκαρ, ὅστις εὐδαίμων τελετὰς θεῶν εἰδώς);
Aristoph. Nub. 1241 (parodying: Ζεὺς γελοῖος ὀμνύμενος τοῖς εἰδόσιν); Andoc. Myst.
30.10 (τοῖς εἰδόσιν); cf. Eur. TrGF 5.781.11–13; Pind. O 2.85 (φωνάεντα συνετοῖσιν, note
the eschatological theme; cf. Bowra, loc. cit.) and, perhaps more playfully, Pl. Symp.
199a1–2. For the kouros motif as initiatory, see Burkert (1969) 14, n.32; Kingsley
(1999) 71–4; cf. Rohde (1925) 601–3; also Verdenius (1947) 285.

255
 How Could Parmenides Have Written Alêtheia?
consensus rightly recognises, is a formulaic designation for death.79
The goddess, then, impresses upon Parmenides that he has travelled
while still alive the way which mortals (that is, their souls) travel
after death and that his special privilege in doing so is proper and
divinely sanctioned (οὔτι σε μοῖρα κακή . . . ἀλλὰ θέμις τε δίκη τε,
B1.26–8).80
The Daughters of the Sun evoke the myth of Phaethon’s disas-
trous celestial journey in Sun’s chariot, which is probably at least
as old as Hesiod (fr. 311 M-W; cf. Th. 987) and which formed the
subject of Aeschylus’ Heliades (TrGF 3.68–73) and Euripides’
Phaethon (TrGF 5.771–86).81 That the gates through which the
chariot travels are ‘aethereal’ (B1.13), whatever topographical
inferences we may or may not draw from this description, further
strengthens this association. The identification of his guides as
Sun’s Daughters thus recasts Parmenides as a new, successful
Phaethon, guided safely by the appropriate charioteers (B1.24),
and secures the identification both of the chariot with its blazing
axle (αἰθόμενος, B1.7), and of the ‘much-prompted’ or ‘much-
attending’ horses (B1.4), as those of the Sun.82 In line 3, we see
that the kouros’ road is that of some unspecified divinity (ὁδόν . . .
δαίμονος). A reader left puzzled by this indeterminate remark, who
then learned that the kouros’ charioteers were the Daughters of the
Sun (B1.5, 9, cf. 24) and, therefore, that his chariot and horses
were those of the Sun (B1.9), and, furthermore, that the kouros was
journeying on the paths of Night and Day (B1.11), would most
79
Ganschinietz (1919) 2409; Burkert (1969) 14; Mourelatos (2008a) 15; Pellikaan-Engel
(1978) 60–1; Coxon (2009) 10, 17, 282; Kingsley (1999) 61; Robbiano (2006) 73;
Ranzato (2015) 70. See Il. 13.602; cf. e.g. Il. 3.101; 6.488–9; Od. 11.560; similarly,
Quint. Posthom. 13.440. Note also the image of passing Hades’ gates, Il. 5.646; Il.
23.71, with Robbiano (2006) 154, n.405.
80
That the term νέεσθαι (B1.26) connotes the sense ‘return’ (e.g. Il. 23.229; Il. 12.32–3; Il.
6.189) is appropriate given Parmenides’ theory of metempsychosis: the kouros returns
to his soul’s prenatal habitat.
81
Cf. also κοῦραι | Ἡλιάδες in Apollonius’ Argonautica 4.603–4. For these and further
references, see Bowra (1953c) 44–5; Burkert (1969) 6–7; Palmer (2009) 56 with n.16.
82
Cf. Burkert (1969) 7; Owens (1979) 17; Palmer (2009) 56. αἰθόμενος (B1.7) relates the
chariot to the fiery element Light (cf. Coxon (2009) 273) and recalls later descriptions of
Helios’ chariot in the context of Phaethon’s myth, as in Nonnus’ Dionysica 38.192
(ἔμπυρον ἅρμα) and Lucian, DDeor. 24.2 (ἐπιβὰς τοσούτου πυρός), see further Bowra
(1953c) 45. For the iconic chariot and horses of the Sun, cf. h.Hom. 2.62–3; h.Hom.
4.68–9; h.Hom. 28.13–14; h.Hom. 31.8–9. For πολύφραστοι (B1.4), see Mourelatos
(1965) 262; (2008) 22.

256
 5.3 The Proem in Its Time and Place
naturally infer that the divinity in question was the Sun.83 That the
Sun’s path takes him daily to the Beyond (traditionally, the
Underworld) is a standard Greek topos (as in Hesiod’s Theogony
760–1 (οὐρανόθεν καταβαίνων) and Stesichorus, PMG 185).
Indeed, in the Odyssey, Hermes the psychopomp leads souls to
Hades by ‘Sun’s gates’ (Ἠελίοιο πύλας, Od. 24.12).
Parmenides, then, represents the kouros as an initiate who journeys
into a scenery which is obviously modelled on Hesiod’s Underworld.
Furthermore, he represents him as positioned, in a manner reminis-
cent of Phaethon, in the seat of the ultimate manifestation of celestial
Light in a journey through gates which are themselves described as
‘aethereal’. Any tenable reconstruction of the proem’s trajectory
would have to allow for both of these points. We can turn now to
develop further the significance of these conclusions (without com-
mitting ourselves to a particular spatial trajectory).
The most important point for us to recognise is that the mortal
kouros is travelling towards his divinisation. This is already sug-
gested by the description of his way as that of a divinity (ὁδόν . . .
δαίμονος, B1.3). With this notion of a divine way, we may compare
especially Homer’s Cave of the Nymphs, with its celebrated
demarcation of a mortal entrance from that reserved for divine
travellers: ‘There are two sets of doors to the cave. One set, facing
North, is for humans to go down by (καταιβαταὶ ἀνθρώποισιν), and
the other in turn, facing South, is more divine; nor do men enter by
there, but it is the way of immortals’ (ἀλλ’ ἀθανάτων ὁδός ἐστιν,
Od. 13.109–12; cf. θεῶν . . . κελεύθου, Il. 3.406). Pindar speaks of
the ‘way of Zeus’ (Διὸς ὁδός), by which souls journey to the island
of the blessed (O 2.70), while an Orphic hymn describes the rising
and setting of celestial bodies as the ‘ways of heavenly gods’ (θεῶν
ὁδοὶ οὐρανιώνων, fr. 168.15 Kern = 243.15 Bernabé).84
83
For the view that ὁδόν rather than δαίμονος is the antecedent of ἥ, see Coxon (2009) 272;
Lesher (1994a) 1, and, for the further identification of the daimôn with the Sun,
Cornford (1952) 118, n.1; Guthrie (1965) 7; Burkert (1969) 4; Palmer (2009) 56–7,
with nn.17–19. Alternatively, though, the antecedent could be the daimôn, which would
then be feminine and, presumably, refer to the narrating goddess, see Bowra (1953c) 50;
Mansfeld (1964) 225. Both readings are syntactically possible, but the immediate
context favours an identification of the daimôn as Helios, whereas the goddess is not
clearly introduced until B1.22, cf. Burkert, loc. cit.
84
These passages are noted as parallels to B1.3 by Diels (2003) 46–7; Bowra (1953c) 50;
Burkert (1969) 4.

257
 How Could Parmenides Have Written Alêtheia?
Later on, the goddess emphatically states that this path is ‘far
from the track of humans’ (ἀνθρώπων, B1.27).85 This statement
itself immediately follows, and further explains and amplifies (ἦ
γάρ), the identification of the way which the kouros travels as that
which the soul usually travels after death (μοῖρα κακή κτλ, B1.26–
7). Within Parmenides’ psychological and eschatological system,
then, we would expect the kouros’ journey to be that of the fiery,
Hot, Light, aethereal soul. And indeed, he is conveyed by the
Daughters of the Sun, agents emblematic of fiery Heat, Light
and the celestial region of aether.86 Again, his journey is that of
the divine soul and, fittingly, he is closely related and even assimi-
lated to these divine guides. The appellation ‘youth’ (κοῦρε,
B1.24) places the traveller on an equal footing with the aforemen-
tioned maidens (κοῦραι, B1.9) as their brotherly counterpart.87
This impression is further supported by the portrayal of the kouros
as a new and successful Phaethon – another child of Helios – and,
significantly, by the goddess’s immediately following (B1.24)
affirmation that the kouros is closely linked to his divine guides:
‘O youth, companion to immortal charioteers’. The term translated
as ‘companion’ here (συνάορος) is a fairly rare word, which
suggests an appropriate and strong connection.88
Since the kouros is associated with the Daughters of the Sun and
travels in the manner of the Sun, it is tempting to find in
Parmenides’ concept of the εἰδὼς φώς a punning play on φῶς: the
knowing ‘man’ (φώς) is a knowing ‘light’ (φῶς). When describing
the moon In B14, Parmenides makes virtually the same pun by
substituting the phrase ‘a foreign light’ (ἀλλότριον φῶς) for the
familiar Homeric phrase ‘a foreign man’ (ἀλλότριος φώς, e.g. Il.
85
Recall how in B16 the goddess underscores that she is describing here the thinking of
humans (ἀνθρώποισι, B16.2–3).
86
In B10 and B11, the aether and the sun, together with the other celestial bodies, are
linked in the context of Parmenides’ astronomy.
87
As Burkert (1969) 14, n.32 observes.
88
Homer uses the term to describe the lyre as the companion of a rich feast (δαιτὶ συνήορος
[sc. φόρμιγξ], Od. 8.99), and Pindar to describe praise as the companion of the lyre
(εὐλογία φόρμιγγι συνάορος, N 4.5). A scholion glosses συνάορος in the latter passage as
‘brought together with’ or ‘corresponding with’ (συμφερομένη) and ‘being in common
with’ (κοινωνοῦσα), Σ N4.9.1 Drachmann. Plato cites Pindar for the view that hope
accompanies (συναορεῖ) the just old man, Rep. 1.331a6–9. By Classical times, συνάορος
signifies close intimacy and often wedlock, e.g. Eur. Or. 654, 1136. Coxon (2009) 52
renders ‘consort’; cf. Kingsley (1999) 75.

258
 5.3 The Proem in Its Time and Place
5.214).89 Our analysis of Parmenides’ conception of the soul in the
previous section enables us to understand the significance of this
pun in the wider framework of his psychology and eschatology.
Under the guidance of the Daughters of the Sun, the kouros travels
the way of the knowing man, or the knowing light: the kouros
journeys to the goddess with – or as – his Light, Hot, fiery,
aethereal soul.
The goddess’s kindly reception of the kouros (πρόφρων
ὑπεδέξατο, B1.22) speaks comforting reassurance.90 If the gates
of the afterlife are proverbially hateful to mortals (Il. 9.312), his
own arrival through them, conversely, calls for a joyful welcome,
since (χαῖρ’, ἐπεί) it was not the work of death (μοῖρα κακή) but of
Themis and Dikê (B1.26–8). Epic texts standardly speak of what is
the ‘law’ (themis) or ‘justice’ (dikê) of certain groups given the
nature, status or identity of their members (e.g. as gods, mortals,
kings, slaves, guests, suppliants, women).91 By saying that it was
sanctioned and right for him to travel this divine path, far from the
track of humans, the goddess accepts the kouros as the sort of
agent for whom such a journey is appropriate. The vocabulary of
obligation which the goddess uses in her next breath (χρεὼ δέ σε,
B1.28b) similarly expresses what is normatively necessary for
certain agents, given their nature, status or identity.92 And what
is necessary for the kouros is to learn, not only the beliefs of
mortals, but also the heart of reality (B1.28b-30).
A confluence of internal evidence, then, indicates that the
kouros’ journey involves and culminates in some sort of divinisa-
tion, and, in particular, in some sort of identification with the
divine, Light soul within him. Now, in progressing along this
journey, the kouros is repeatedly characterised as the dependent
object of acts of guidance and transportation, the control over
89
For the pun, see Torgerson (2006) 41–2, cf. 28–9; Kahn (2009) 215; Cosgrove (2011)
29. ἰσόθεος φώς is one of the common Homeric formulae for a godlike man (see below).
90
As does her hand-clasp (B1.22–3), see Coxon (2009) 10. Mansfeld (2005) 554–5
compares similar scenes of Heracles or Theseus with Athena on archaic vases, and
interestingly suggests that the gesture signals some sort of enhancement in the kouros’
status.
91
Themis: Il. 9.134, 276; Il. 11.779; Il. 19.177; Od. 9.268; Od. 14.130; h.Hom. 3.541; cf. Il.
16.796; Od. 10.73; Od. 14.56. dikê: Od. 4.691; Od. 11.218; Od. 14.59; Od. 19.43; Od.
24.255; h.Hom. 3.458; Scut. 85.
92
See Ch. 4.5.

259
 How Could Parmenides Have Written Alêtheia?
which is attributed to his divine guides (με φέρουσιν . . . μ’ ἐς ὁδὸν
βῆσαν . . . φερόμην . . . φέρον . . . κοῦραι δ’ ὁδὸν ἡγεμόνευον, B1.1–
5; cf. ἥ . . . φέρει, B1.3).93 The kouros is borne along the way
because that is where the mares are carrying him (B1.4). It is the
kourai, not the kouros, who persuade Dikê with ‘gentle words’ to
unbolt the gates (B1.15–17). We noted that, by describing the
kouros as a ‘companion to’ (συνάορος) immortal charioteers,
the goddess is highlighting the close link between the kouros and
the kourai. But the literal sense of the term is ‘bound’ or ‘yoked
together’. The literal equestrian image, especially given the eques-
trian context of this statement itself, further underscores the
kouros’ dependence on the guidance of the kourai: the kouros is
himself ‘yoked’ to his divine charioteers.94 The journey, we saw,
identifies the mortal kouros with his discarnate, divine soul and
culminates in some sort of sanctioned divinisation. Furthermore, it
takes place at, or at least in keeping with, the kouros’ own volition
(‘as far as my heart might reach’, B1.1).95 At the same time,
however, it is an important point, which Parmenides takes pains
to emphasise, that the mortal’s ability to undergo this journey is
dependent on his privileged access to these divine conveyors and
guides, who bring him to the goddess, who then issues to him her
revelation.
So far, we have pursued a mostly internal examination of the
proem. In the remainder of this section, we will consider further
some cultural frameworks, both pan-Hellenic and more local,
which the proem evokes. These parallels will corroborate the
conclusions at which we arrived largely on the basis of the internal
evidence and, by contextualising them, help us make better sense
of them. They will enable us to grasp more clearly how the proem
constructs the mortal’s attainment of the disposition which is
93
Cf. Jacobs (1999) 187–9; Lesher (1984) 24–5.
94
In Homer, the literal equestrian sense is in evidence in uses of the cognate verb, e.g. Il.
10.499 (σὺν δ’ ἤειρεν ἱμᾶσι); Il. 15.680 (συναείρεται ἵππους).
95
θυμός suggests above all a conative faculty, prompting action through volition; cf. Od.
15.339, with Coxon (2009) 270. Through divine gift, Demodocus sings ‘in whatever
way his heart (θυμός) urges him to sing’ (Od. 8.44–5). θυμός could also indicate the
instrument of thought or understanding (e.g. Il. 2.409), see Lesher (1994a) 9–10.
Mourelatos (2008a) 17, n.21 (cf. also Primavesi (2013) 55, n.59) conclusively refutes
the tenuous claim in Tarán (1965) 9, 13, 16, 27, 30 that the optatives ἱκάνοι (B1.1) and
σπερχοίατο (B1.8b) suggest that the journey occurs repeatedly.

260
 5.3 The Proem in Its Time and Place
anterior to his reception of the goddess’s revelation and his acqui-
sition of the knowledge of what-is.
We will best begin to approach the theological models and
patterns of thought which underlie the kouros’ divinisation by
first stepping back to a fundamental observation. Already our
earliest sources do not typically conceive of human agents as
agents who are simply and strictly human. Archaic texts are pre-
occupied with the gulf which separates mortal and divine, but we
also find in them an unmistakable measure of fluidity between
these two categories. Most obviously, sexual unions between gods
and mortals produce divine, semi-divine or godlike offspring.
Hesiod, for example, stresses that it was the mortal Semele who
bore the immortal son Dionysus (τέκε . . . ἀθάνατον θνητή, Th.
940–2). Indeed, Hesiod invokes the Muses quite generally to sing
of the liaisons between goddesses and men and of the resulting
‘children like gods’ (θεοῖς ἐπιείκελα τέκνα, Th. 965–8).96 The
Muses oblige (969–1018), and the offspring of gods and women
permeated the Ehoiai (see Th. 1019–22). In the Iliad, Zeus simi-
larly lists his liaisons with various women and the resultant god-
like and divine offspring (Il. 14.313–25). In the Homeric Hymn to
Aphrodite, the poet explains Aeneas’ exceedingly godlike nature
(μάλα . . . θεοείκελος, 279) as the son of the mortal Anchises and the
goddess Aphrodite, and as a nursling of nymphs who are them-
selves intermediate between gods and men (h.Hom. 5.255–79).
Even certain groups of mortals, such as the Phaeacians, may be
close(r) to the divine by birth (ἀγχίθεοι γεγάασιν, Od. 5.35).97
But such intermingling between gods and mortals is only the
starkest illustration of the deep-seated and more general notion
that the mortal and the divine inhabit the same, commensurable
plane, and that the former can approximate the latter. Even in the
absence of semi-divine parentage, and even prior to and indepen-
dently of the appearance of notions of the soul’s divinity and
metempsychosis, we find that some mortal agents possess also
divine features and capabilities, and have the potential to promote
them.
96
Mortal ‘godlikeness’ always expresses an approximation to the divine in some features
or accomplishments; note that, in h.Hom. 31.7, ἐπιείκελον ἀθανάτοισιν is used of a god.
97
Cf. Od. 7.201–6 (with Vernant (1991a) 45, n.27); h.Hom. 5.200–1.

261
 How Could Parmenides Have Written Alêtheia?
Almost every page of Homer illustrates the principle that, by
virtue of a privileged affinity with the divine, whether through
semi-divine parentage or through interactions and associations
with the divine, mortals are capable of what would otherwise
have been impossible for them. Mortals who attained such heights
of excellence are standardly and ubiquitously represented as ‘god-
like’, ‘equal to the gods’ or, simply, ‘divine’. Homer also describes
a son of two mortals, like Odysseus, as ‘divine’ (Ὀδυσσῆος θείοιο,
Il. 2.335), and refers to the heroes fallen at Troy quite generally as
‘the race of half-divine men’ (ἡμιθέων γένος ἀνδρῶν, Il. 12.23).
Epithets which describe the best mortals as godlike or equal to the
gods, either in general or in particular respects, permeate the
verses of Homer.98 As the passages presently cited (and numerous
others) exemplify, the frequency and formulaic nature of some
Homeric deifying epithets do not imply that these epithets are by
now vestiges of an obsolete theology but, on the contrary, reflect
the vibrant ubiquity of the notion of (partial, temporary) mortal
approximation to the divine.99 We observed before that Homer
regularly construes the same actions and impulses as caused
simultaneously by both human and divine agencies (‘double
motivation’).100 The actions or features which exemplify approx-
imation to divinity are not achieved by the gods for mortals,
conceived as mere passive instruments. Rather, it is (only) through
divine interaction, communication or parentage that mortals are
capable of certain accomplishments. To offer one illustration, if
98
For example, θεῖος, e.g. Il. 16.798–9 (ἀνδρὸς θείοιο); Od. 2.259; ἡμίθεος, h.Hom. 31.19;
h.Hom. 32.18–19; cf. Simon. PMG 523; ἰσόθεος φώς, Il. 2.565; Od. 1.324; δαίμονι ἶσος,
Il. 20.493; Il. 21.18; h.Hom. 2.235; ἀντίθεος, Il. 5.629; Od. 1.21; Διὶ μῆτιν ἀτάλαντος, Il.
2.636; θεόφιν μήστωρ ἀτάλαντος, Il. 7.366; Od. 3.110; ἀτάλαντος Ἄρηϊ, Il. 13.295; ἶσος
Ἄρηϊ, Il. 11.295; θεῷ ἐναλίγκιος αὐδήν, Il. 19.250; θεοείκελος, Il. 1.131; Od. 3.416 (cf.
θεοῖς ἐπιείκελ’, Il. 9.485; ἐπιείκελον ἀθανάτοισιν, Il. 11.60; θεοῖσι γὰρ ἄντα ἐῴκει, Il.
24.630); θεοειδής, Il. 24.217; Od. 1.113; γυνὴ ἐϊκυῖα θεῇσιν, Il. 11.638; ἰκέλη χρυσέῃ
Ἀφροδίτῃ, Il. 19.282; δέμας ἀθανάτοισιν ἐοικώς, h.Hom. 5.55; cf. Od. 8.14. Note also the
overdetermined δῖος (‘heavenly’, ‘noble’, ‘divine’), e.g. Il. 15.652. For epic godlike-
ness, see Roloff (1970) 3–83 (cf. Griffin (1980) 81–102) and, for later archaic and
classical literature, Roloff (1970) 102–26.
99
Centuries later, Plato (Men. 99c7-d9) remarks that women and Spartans describe good
men as ‘divine’. (Plato appeals to this custom in support of the idea that the true beliefs
of ignorant but inspired and so ‘divine’ mortals may have a divine origin.) As Homer
teaches us, we should not assume that such encomiastic terminology did not correspond
to some widespread, implicit theology of approximation.
100
Ch. 2.2.

262
 5.3 The Proem in Its Time and Place
Diomedes can contend against mortal adversaries independently
of divine interaction, it is only after Athena granted him might
(menos), and in conjunction with her active aid, that he – equal to a
god (δαίμονι ἶσος, Il. 5.438, 459, 884) – can successfully challenge
Ares (Il. 5.1–3, 121–32, 815–61).
Right after the poet describes the ambitious Diomedes as equal
to a god (δαίμονι ἶσος, 438), Apollo warns the hero to recall and
respect the disparity between mortal and divine: ‘consider, son of
Tydeus, and draw back, and do not wish to think on a par with the
gods’ (μηδὲ θεοῖσιν | ἶσ’ ἔθελε φρονέειν, Il. 5.440–2).101 That the
extent to which mortals can and should approximate divinity is a
constant issue of theological and moral reflection demonstrates the
openness of Greek religious thought to the possibility of such
approximation. Full-blown deification is the limiting case of
approximation to the divine. The prospect of deification is mark-
edly absent from the Iliad, but it is in plain sight in the Odyssey,
Hesiod and the Homeric Hymns. The conflicting epic traditions
concerning Heracles instructively reflect on-going engagements
with the question of where and whether an impassable line should
be drawn. The Iliad insists that even Heracles, the best-case-
scenario of mortal excellence and divine favour (φίλτατος), was
constrained by the inexorable fact of his mortality (Il. 18.115–21).
Conversely, Hesiod’s Theogony and the eponymous Homeric
Hymn render him a deified god dwelling on Olympus (Th. 950–
5; h.Hom. 15.7–8). The poet of Odyssey 11, in a gesture of con-
ciliatory synthesis,102 recounts that Heracles himself (autos)
rejoices among the immortals, while his image or phantom
(eidôlon) remains in Hades (601–4).103 Dodds has, therefore,
101
In the same book, Dione describes the dire consequences of mortal aggression against
gods, Il. 5.382–415; cf. Il. 6.128–43. On the slippery slope from godlikeness to hybris,
see further Roloff (1970) 77–80; Griffin (1980) 168–9.
102
Following Bremmer (1983) 81.
103
For the challenges posed by Heracles, the Dioskouroi and Asclepius to the boundary
between mortal and divine, see further Burkert (1985) 208–15. Odysseus must refuse
precisely an offer of full-blown deification in order for the narrative of nostos to take
place (as presently discussed). But the goddess Leucothea, previously a mortal woman,
exemplifies the road not taken: Od. 5.334–5. In Hesiod, Semele and Ariadne similarly
become ageless goddesses, Th. 940–2, 949. Abduction by the gods may lead to
deification, h.Hom. 5.202–14; Th. 987–91; cf. the more vague Od. 15.250–1; Il.
20.232–5. Hesiod’s myth of Races recognises various middle-ground positions
between mortal and divine, see Ch. 6.1.

263
 How Could Parmenides Have Written Alêtheia?
oversimplified our evidence when he wrote: ‘Men may share at
times in the divine attribute of power, but in Homer there is
nevertheless no real blurring of the sharp line which separates
humanity from deity.’104 In fact, the prevalent phenomenon of
mortal participation in divine capacities, manifested in the ‘god-
like’ or ‘divine’ man, is throughout concurrent with a constant
negotiation of the malleable line which separates humanity from
deity, a negotiation in which ‘Homer’ does not adopt a consistent
position.
Parmenides’ language is fundamentally Homeric. Epic phra-
seology, resonances, narrative conventions and imagery permeate
the extant lines, which contain on average only one non-Homeric
word for every three verses, while most of these exceptions are
themselves compounded from Homeric words.105 As Mourelatos
maintains, we should not dismiss Parmenides’ epic language as
mere superficial garb. Parmenides, rather, ‘uses old words, old
motifs, old themes, and old images precisely in order to think new
thoughts in and through them’.106 On a rudimentary level, then, by
immersing his poem and its proem in Homeric language and
imagery, Parmenides relates the kouros’ novel approach to his
goddess to the approach of a Homeric protagonist to his patron
divinity.
More specifically, however, scholars have plausibly identified
in Parmenides a particular engagement with the Odyssey. The
same language of journeying (ὁδός, κέλευθος, πέμπω, ἡγεμονεύω,
etc.) dominates both poems. Furthermore, if the Daughters of the
Sun guide the kouros in his own journey to the setting of the
afterlife, it is Circe, another daughter of the Sun (Od. 10.138),
who directs Odysseus to Hades, where he is to learn from the seer
Teiresias (Od. 10.537–40; Od. 11.50, 89).107 While, however, both
protagonists travel far beyond the familiar track into eschatologi-
cal locations, their journeys also diametrically diverge. Odysseus

104
Dodds (1951) 10.
105
Coxon (2009) 7–12 (Coxon prints epic parallels under each page of text); Mourelatos
(2008a) 1–46.
106
Mourelatos (2008a) 6, 39; cf. Dolin (1962) 93.
107
For further echoes, see Havelock (1958) 137–40; Mourelatos (2008a) 17–25;
Montiglio (2005) 150.

264
 5.3 The Proem in Its Time and Place
is repeatedly made to wander astray, sometimes through malignant
divine agency, and it is through his wandering that he amasses
knowledge and experiences (μάλα πολλὰ | πλάγχθη κτλ, Od. 1.1–
4). Conversely, the kouros’ divine guides escort him directly to his
goal (ἰθύς, B1.21) and precisely prevent him from undergoing the
wandering which the poem associates throughout with error and
ignorance.108 We may sharpen one overlooked aspect of this
pointed divergence. It is fundamental to the Odyssey that, for the
narrative of nostos to take place, Odysseus must reject the offers of
divinisation which are proffered to him by his female divine host
Calypso (Od. 5.135–6; Od. 7.255–8; Od. 23.334–7) and choose
instead the life of a wandering mortal (ἄνδρα . . . πλάγχθη, Od.
1.1–2).109 Parmenides’ female divine host, by contrast, precisely
separates the willing (B1.1b) traveller from those to whom she
refers as ignorant mortals and whom she repeatedly describes as
wandering (βροτοὶ εἰδότες οὐδὲν | πλάζονται . . . πλαγκτὸν νόον,
B6.4–5; cf. B8.54; B16.1). The kouros, then, pointedly outdoes
Homer’s Odysseus in willingly accepting divinisation.110
That Parmenides employs the metre, language and imagery of
inspired poetry and oracles is a fact so fundamental that it is often
ignored. In addition to Parmenides’ language, the chariot and road of
song are conventional idioms of inspired poetry.111 Parke and
Wormell, in the introduction to their edition of Delphic oracles
(1956), observed that Parmenides, like Empedocles and Heraclitus,
displays ‘affiliation in subject matter, tone, and style with the Pythian
responses’.112 Indeed, Parmenides’ goddess echoes, I think, the
Pythia’s terminology. The term which Parmenides repeatedly uses
to refer to his philosophical inquiry is, in its verbal form (dizêmai), a

108
Mansfeld (1964) 229–31; Lesher (1994a) 4–6; Montiglio (2005) 150.
109
See Vernant (1996).
110
See Ch. 6.1 on related divergences between Parmenides and Hesiod.
111
B1 displays especially close verbal and imagistic echoes with Pindar’s chariot, which is
led by female mules on a way towards the gates of song (O 6.22–7). For the chariot,
path and journey of poetry, see also Od. 8.479–81 (οἴμας Μοῦσ’ ἐδίδαξε); Pind. O 9.80–
1; P 10.65; I 2.1–2; 8.61; N 1.7; N 6.53–4; Bacch. Epin. 5.176–8; Paean. fr. 2 Irigoin;
Choeril. EGF 1.4–5; Emped. DK31 B3.5; B35.1; Timoth. PMG 791 col.5.225–6. See
further Fränkel (1975a) 1–2; Bowra (1953c) 41–3; Burkert (1969) 3, n.6, 10, n.21;
Ranzato (2015) 128, and, for a dedicated exploration of the markers of poetic inspira-
tion in B1, Lesher (1994a) 8–16.
112
PW xxxiii–xxxiv.

265
 How Could Parmenides Have Written Alêtheia?
standard technical term in Delphic oracular responses, which
expresses the act of consulting Apollo.113 More broadly, both the
Pythia and Parmenides’ goddess characteristically issue their disclo-
sure to their mortal audience with numerous second person pronouns
and, typically, imperatives,114 and both formulaically bid their audi-
ence to ‘ponder’ (phrazesthai) the content of their disclosure.115 The
self-asserting rhetoric of both inclines to acerbic contrasts between
their own epistemic superiority and mortal ignorance,116 and both
have always been noted for the studied and laconic obscurity of their
hexameters.117
These observations – and, of course, the fundamental fact that
Parmenides frames his poem as divine disclosure – are significant
for our inquiry above all because they support the thesis that
Parmenides postulates an epistemically significant interaction
with the divine only through which can he attain the knowledge
he comes to attain.118 As we saw (Chapter 5.1), however, the mere
postulation of divine disclosure is not by itself sufficient to resolve
our difficulty. Human cognition could not grasp and sustain the
knowledge of Alêtheia even if it were communicated to it. Here

113
PW 111.2; 202; 216.5; 317.1; 379.4; 406.3; 517.2. The usage is at least as old as
Odyssey 11 (προσηύδα μάντις . . . νόστον δίζηαι κτλ, 99–100; see similarly Apollo’s
reply to Alcmaeon: νόστον δίζηαι, PW 202; cf. Thuc. 2.102). In Herodotus, the phrase
διζημένων τὸ μαντήιον indicates the attempt to interpret an oracular response (7.142;
compared by Guthrie (1962) 418–19 with Herac. DK22 B101: ἐδιζησάμην ἐμεωυτόν).
Compare Parmenides’ ὁδοὶ διζήσιος: B2.2; B6.3; B7.2. Parmenides also uses the verb:
B8.6 (διζήσεαι). Robbiano (2006) 125–6 takes the references to ‘signs’ (σήματα) in
B8.2 to evoke the mantic framework of a divinity issuing signs; cf. also Robbiano
(2006) 170, n.451 (on πεφατισμένον, B8.35).
114
On the Pythia: see PW xxv; cf. B1.28b; B2.1; B7.2, 5–6; B8.8, 50–2.
115
B6.2 (τά σ’ ἐγὼ φράζεσθαι ἄνωγα) cf. PW 7 (ταῦτά νυν εὖ φράζεσθε), 112, 181, 225,
301, 374, 408, 487 (φράζεσθ’ ἐξ ἐμέθεν χρησμὸν θεοῦ, cf. B7.5–6: κρῖναι . . . ἔλεγχον | ἐξ
ἐμέθεν ῥηθέντα).
116
On the Pythia: e.g. PW 41; 52; 55, with PW xxv; Maurizio (2013) 116–17. Cf. B6; B7.
On this point, cf. Ranzato (2015) 89–91.
117
On the Delphic oracle, note most famously Herac. DK22 B93; see further Ch. 3.1. On
Parmenides, see by way of illustration Plutarch A16; Proclus A17; KRS (1983) 241;
Barnes (1982) 155. On ‘systematic equivocity’ in Parmenides, see Mourelatos (2008a)
222ff, esp. 260–3.
118
In Ch. 2 (esp. 2.2), we discussed the role of epistemically significant interactions with
the divine in poetic inspiration. For divination, see Ch. 3.1. Mansfeld (1964) 273
explained satisfactorily, in my view, why Parmenides wrote in verse: ‘Die Dichtung
ist in der griechischen Tradition das Wort der Gottheit. Weil des Parmenides Weg zur
Philosophie der Weg der Offenbarung war, mußte er ein Dichter werden’; cf. also Most
(1999a) 353–5.

266
 5.3 The Proem in Its Time and Place
too, then, it is worth emphasising that the higher epistemic status
of the inspired poet or diviner (however precisely we are to under-
stand this status in different cases) was also sometimes accompa-
nied by the more general and less clearly demarcated perception of
the poet or diviner himself as godlike or divine.119
Parmenides’ association of the acquisition of divine knowledge
with the divinisation of the knower (homoiôsis theôi) is thrown
into sharper relief by his appropriation of the imagery and termi-
nology of mystery initiations. A variety of mystery cults spreads
throughout the Greek world from the end of the seventh century
onwards, including Eleusinian, Dionysiac, Samothracian and
Meter cults, as well as different clan and family-based initiations.120
We saw that Parmenides pointedly adopts the vocabulary of mys-
tery initiation by describing the traveller as a ‘knowing man’ (εἰδότα
φῶτα, B1.3) and a ‘youth’ (κοῦρ’, B1.24). We can now explore
further the significance of this point.
The kouros, rightly and properly initiated (B1.26–8), is through-
out characterised by contradistinction with the ignorant ‘mortals’.
The contrast between the initiated and the uninitiated is similarly
central to the rhetoric of the mysteries. Representations of the
blessedness (makarismos) of the former, to whom alone the secret
of the mysteries is lawfully divulged (e.g. Lysias, 6.51; Isocrates,
4.28.6–7), always picture it in contradistinction with the ignorance,
exclusion and sorry state and prospects of the latter.121 The kouros,
like mystery initiates, is permitted to progress only once the appro-
priate words – or passwords – are spoken.122 His eschatological

119
We explored this aspect of divination in Ch. 3.1. For poetic inspiration, consider for
example the common term ‘divine minstrel’ (θεῖος ἀοιδός), Od. 1.336; Od. 4.17; Od.
8.87 (etc.); Marg. fr. 1.1 W; Theocr. Id. 16.44; cf. Bacchyl. Ep. 9.3, with Roloff (1970)
117–19.
120
On the mystery initiations of ancient Greece, see Burkert (1985) 276–304; (1987);
Bowden (2010); Bremmer (2014).
121
E.g. h.Hom. 2.480–2; Soph. TrGF 4.837.3–4. One popular eschatological image has the
uninitiated lying in mire, Pl. Phd. 69c; D.L. 6.39; Plut. fr. 178.16–20 Sandbach; Aristid.
22.10 Keil. In another, they carry water in broken pitchers, Paus. 10.31.9; cf. Pl. Gorg.
493b. An Italian lekythos from about 500 BCE depicts a similar scene, Cook (1940)
vol.3, pl.36. We might compare the futile (endlessly?) back-turning path of the mor-
tals (B6.9).
122
μαλακοῖσι λόγοισιν, B1.15. Compare the Eleusinian ‘password’ (σύνθημα), Clem.
Protr. 2.21.2, with Burkert (1983) 269–73; (1985) 285–7; Richardson (1974) 22–3.
A similar parallel obtains with the Gold Tablets, as we shall see presently.

267
 How Could Parmenides Have Written Alêtheia?
journey culminates in a direct encounter with a female deity, who
remains unnamed, and is identified only as ‘goddess’ (θεά,
B1.22).123 References to Underworld and initiatory divinities stan-
dardly leave them unnamed.124 Even before availing ourselves of
the Gold Tablets, we may explore more tentatively and generally the
eschatological and divinising facets of initiations, focusing perforce
on the best but by no means well-attested initiations at Eleusis.
As early as the Homeric Hymn to Demeter, our texts trace the
initiate’s blessedness to an act of directly seeing what is revealed
in the mysteries, and allude to the ‘appearances’ (phasmata) in
question.125 We of course do not precisely know what the ‘inef-
fable apparitions’ (Aelius Aristides, Or. 22.3 Keil) were, and
‘precisely’ could be a misleading expectation. Our sources do,
however, regularly speak of the fear which the ritual first provokes
and then dispels in the initiate. Aelius Aristides writes that ‘of all
that is divine to humans (θεῖα ἀνθρώποις)’ Eleusis is at once ‘the
most frightening and joyful’.126 Plutarch suggests an eschatologi-
cal understanding of this fear, comparing the soul’s experiences at
death to the experiences of ‘those celebrating great initiations’
(fr. 178.5–6 Sandbach). After first wandering through frightening
dark paths and undergoing panic, shivering, sweat and amazement
(7–10), some wonderful light (φῶς τι θαυμάσιον) meets the initiate
along with pure meadows, dances, sacred words and holy views
(10–13). Now fully initiated and free, he celebrates the rites with
his fellow initiates. This is why, not only what they signify, but
also the very signifiers ‘dying’ (teleutan) and ‘being initiated’

123
The goddess is neither Justice nor Night. See the Appendix, with n.25.
124
Referring only to θεός, θεά, θεαῖ, τὼ θεώ, etc., e.g. Soph. OC 1548 (ἥ τε νερτέρα θεός);
Th. 767 (θεοῦ χθονίου); LSCG 4; 5; 8; 20 (Eleusis, fifth century); Syll.3 736.33
(Μεγάλοις θεοῖς, Andania). See further references and discussion in Rohde (1925)
185, n.19; West (1966) 369–70; Richardson (1974) 27; Burkert (1969) 13–14 with
n.31. The latter draws the connection with Parmenides’ nameless goddess; cf. also
Kingsley (1999) 96; Bonnechere (2003) 186, n.43.
125
h.Hom. 2.480 (ὄπωπεν); Soph. TrGF 4.837 (δερχθέντες); Pl. Phdr. 250b6–8 (μακαρίαν
ὄψιν τε καὶ θέαν . . . εἶδόν τε καὶ ἐτελοῦντο κτλ); Aristid. Or. 22.3 Keil (cf. 12); also Eur.
Bacch. 73–5. For the φάσματα, see further the sources cited in Burkert (1983) 288, n.64
(referring also to the mysteries of Dionysus and Sabazios), cf. also 275 with n.3.
126
Aristid. 22.2 Keil; cf. Aesch. TrGF 3.387 (ἔφριξ’· ἐρῶ δὲ τοῦδε μυστικοῦ τέλους); Plut.
Mor. 47a; 81d–e; cf. Pl. Phdr. 251a3–4 (πρῶτον . . . ἔφριξεν κτλ); see further Richardson
(1974) 306–7; Burkert (1987) 92–3; (1985) 277 (here associating this process with the
anticipation and overcoming of death); Bremmer (2014) 13.

268
 5.3 The Proem in Its Time and Place
(teleisthai) are similar (6–7). In Apuleius’ Metamorphoses, the
priest of Isis describes the initiation as ‘like a voluntary death and
salvation by grace’ (Met. 11.21.26–7). Apuleius then narrates our
only extant first-person account of a mystery initiation. After
playfully regretting that he cannot lawfully sate the reader’s desire
to know all the details (11.23.21–7), Apuleius nonetheless con-
cedes the following (11.23.28–31, tr. after Burkert (1987) 97):
I approached the frontier of death (accessi confinium mortis), I set foot on the
threshold (limine) of Persephone, I was borne through all the elements and came
back, I saw at midnight the sun sparkling in white light (solem . . . lumine), I
approached close to the gods of the upper and the nether world and reverenced
them from near at hand (deos . . . accessi coram et adoraui de proxumo).

Neither Plutarch nor Apuleius can be taken at face value. Both have
their own agenda and may be to some degree creatively theorising
and Platonising. Equally, though, neither should be written off as
mere idiosyncratic fiction. It is itself noteworthy that Plato invokes
the mysteries when he recasts philosophy as the business of practis-
ing dying during one’s life and, thereby, ensuring post-mortem bliss
(Phd. 80e-81a; 69c). Pindar more directly supports the association
between initiation and the experience of death when he states that,
having seen the mysteries (ἰδὼν κεῖν’), the initiate knows the end of
life (οἶδε μὲν βίου τελευτάν), as well as a Zeus-granted beginning (fr.
137 M). This association between initiation and the experience of
death is highly reminiscent of Parmenides’ eschatological proem, in
which the kouros journeys to the setting of the afterlife, as are the
interplay of light and darkness, Plutarch’s analogy with the journey
of the post-mortem soul and Apuleius’ emphasis on his unmediated
encounter with the divine.127 Interestingly, Apuleius comments in
particular about seeing the sun amidst darkness. On the following
day, he was stood in front of the goddess’s statue clothed ‘in the
likeness of the Sun’ (ad instar Solis, 11.24.1–18). As we saw,
Parmenides too adopts the Sun’s role in the proem.128 Most

127
On the importance and interrelation of light and darkness imagery and the experience
of death in Greek representations and appropriations of mystery initiations, see more
generally Seaford (2005) and (2010); see also Ustinova (2009) 229–55; Patera (2010);
Paleothodoros (2010).
128
In Aristophanes’ Frogs, it is only the initiates in the Underworld who enjoy the sun and
light (454–6, cf. 445). The Stoic Cleanthes pronounces the gods mystic forms, the sun a

269
 How Could Parmenides Have Written Alêtheia?
importantly, the association of the experience of death with the
initiate’s initial terror and subsequent bliss instructively coheres
with the mysteries’ perceived eschatological function. Our
Eleusinian evidence advertises above all what is prominent also in
the Gold Tablets: initiations were regularly perceived to ensure a
blessed afterlife and so, as a hierophant’s funerary inscription
records, dispel the fear of death.129
Now, Burkert emphasises that immortality is never mentioned
in connection with the Eleusinian initiations, in which death
remains an ineluctable reality, although not an absolute end.130
While lifetime deification is admittedly not at issue, Burkert’s
emphasis is nonetheless misleading. The eternal and blissful exis-
tence after death, which the mysteries were regularly taken to
guarantee, amounts to post-mortem divinisation, and has demon-
strably been understood as such. The initiates alone can be said to
live after death (τοῖσδε γὰρ μόνοις ἐκεῖ ζῆν ἔστι, Sophocles, TrGF
4.837.3–4). In Pindar, again, the initiate knows, not just the end of
life, but, crucially, also a divinely granted beginning (διόσδοτον
ἀρχάν, fr. 137 M). His hopes concern simultaneously the end of
life and an eternal life (τοῦ σύμπαντος αἰῶνος, Isocrates, 4.28.10–
12). Plato writes that, according to the founders of the mysteries
and those who participate in them, initiation aims to ensure that the
initiate, i.e. his post-mortem soul, spends eternity dwelling
‘among the gods’ (μετὰ θεῶν, Phd. 69c3–7, 81a8–9). In the
Homeric Hymn to Demeter, which at least in part constitutes an
aition for the Eleusinian mysteries, the queen Metaneira sees the
disguised Demeter placing her son, Demophon, in a great fire, and

torchbearer and the cosmos a mystery, SVF 1.538. On various associations of the sun
with mysteries, see further Seaford (2005) 602–4.
129
H.Hom. 2.480–2; Pind. fr. 137 M; Soph. TrGF 4.837; Isoc. 4.28.10–12; Cic. Leg. 2.36;
Crinagoras Anth. Pal. 11.42; Aristid. Or. 22.10 Keil; D.L. 6.39. Funerary inscription:
IG II/III2 3661.5–6; cf. Plut. fr. 178.19 Sandbach (φόβῳ δὲ θανάτου). Cf. Burkert
(1985) 277, 289; (1983) 269; further on Dionysiac mysteries: Burkert (1985) 294–5.
A similar eschatological purpose may well have been associated with other Greek
mysteries, but we could not presuppose it across the board. Samothracian mysteries
appear to have aimed at least initially to ensure safe naval voyages, Σ Ap. Rh. 77.15–16
Wendel; see further Burkert (1985) 284. Bremmer (2014) 18–20 plausibly suggests that
agricultural fertility was at least another important focus of the Eleusinian rites
themselves, but he downplays the degree to which our evidence repeatedly emphasises
the mysteries’ eschatological aspect.
130
Burkert (1983) 294–5; (1985) 289.

270
 5.3 The Proem in Its Time and Place
misinterprets the goddess’s good intention to deify him.
Demophon already grew to be godlike (235, 241) and, were it
not for Metaneira’s intervention, would have become ‘ageless and
immortal’ (242), but now death must remain ineluctable for him
(260–3). Nonetheless, the mysteries that Demeter proceeds to
establish in Eleusis and their attendant rewards of a blessed after-
life (480–2) may be said, in keeping with the other texts consid-
ered here, to substitute her initial intention to bestow lifetime
deification. It is unsurprising and instructive that Heracles and
the Dioskouroi, who epitomise the prospect of scaling the bound-
ary between mortal and divine, were emphatically numbered
among the initiates.131
Put simply, the promise of the mysteries to remove the fear of
death is recurrently linked with a promise of divinisation. Insofar as
the former is central to the mysteries, so is the latter. The perceived
eschatological function of the mysteries relies on and fosters the
notion of a divine potential within the mortal. Through initiation,
this potential is actualised. The mortal’s initiation involves a trans-
formation which is marked by a new relation to the divine, as well as
by his integration into a new group of initiates, circumscribed by
contradistinction with the uninitiated. The initiate’s transformation
means the attainment of a new and higher status, which – according
to a dominant conception of the mysteries in our sources – will
subsequently result in everlasting, post-mortem bliss.
The Gold Tablets, some of which are very close to Parmenides
in time, place, language and imagery, offer a refreshingly tangible
corroboration.132 The extant specimens date from the late fifth
century BCE to the second or third century CE (references here are
to BCE unless otherwise stated). These little scraps of gold foil are
found in tombs in South Italy (predominantly), Sicily, Rome,
Crete, some parts of Mainland Greece (particularly Thessaly)
and Macedonia. Despite some regional variations, they display
131
For literary and pictorial attestations as early as Pindar, see Burkert (1983) 267 with
n.11; (1985) 213. Note X. Hell. 6.3.6 (πρώτοις ξένοις); Aristid. Or. 22.4 Keil (ξένων
πρώτους).
132
On the Tablets in general, see Graf and Johnston (2013a); Bernabé and Cristóbal
(2008); Edmonds (2011a). Whether or not we should label the Tablets ‘Orphic’ has
no bearing on my argument. For competing answers, see Bernabé and Christóbal
(2011) and Edmonds (2011b).

271
 How Could Parmenides Have Written Alêtheia?
remarkable formulaic continuity across different eras and
locations.133 The Tablets range from extended narratives, which
instruct the initiate’s post-mortem soul as to what path it should
travel and what it should do and say in the afterlife, to briefer
‘proxies’, which themselves speak the appropriate words on
behalf of the initiate’s soul, to mere records of names or names
plus ‘initiate’ or simply ‘initiate’.134
Much like the kouros, the initiate’s soul in the Tablets journeys to
the setting of the afterlife where it encounters guardians who, once it
speaks the appropriate words (or passwords), admit it to an audi-
ence with an eschatological goddess, the ‘Queen of the Chthonian
ones’.135 The goddess is sometimes left unnamed and sometimes
identified explicitly as Persephone. Just like Parmenides’ eschato-
logical goddess, she comports herself ‘kindly’ to the initiate
(πρόφρων).136 There are unmistakable affinities, then, between the
eschatological ‘way’ of the post-mortem soul which the kouros
travels in Parmenides’ proem (ὁδόν . . . τήνδ’ ὁδόν, B1.2, 5, 27)
and what our earliest and longest Gold Tablet from Hipponion
describes as the ‘sacred way (ὁδόν . . . hιεράν) on which also other
glorious initiates and bacchoi travel’.137
The initiation to the goddess, as elsewhere, guarantees for the
initiate, who is again ‘blessed’ (ὄλβιος) or ‘thrice-blessed’
(τρισόλβιος), a better lot in the afterlife.138 Most importantly for
133
Bernabé and Cristóbal (2008) 5.
134
Extended narratives: GJ 1–3; 8; 25. Proxies: GJ 10–14, 29. Records: GJ 20–4a; 31–6.
135
See GJ 5–7 (South Italy (Thurii), fourth cent.); 8 (Sicily, circa third cent.). See also GJ 1
(South Italy (Hipponion) late fifth cent.), where the same scene culminates in an
admission to a Chthonian King (καί . . . βασιλε͂ϊ). Some editors read βασιλεί<αι>
(Bernabé and Cristóbal (2008) 48–9) in conformity with the Tablets elsewhere, but
this is perhaps overly homogenising. For passwords, cf. also the list of Bacchic
σύμβολα in GJ 27.
136
Unnamed: e.g. GJ 5; 8–9. ‘Persephone’: e.g. GJ 6–7; 26a–b. It is neither warranted nor
illuminating to identify therefore Parmenides’ goddess as Persephone, contra Kingsley
(1999) 94. πρόφρων: GJ 6.7; 7.7 (South Italy (Thurii), fourth cent.); B1.22 (πρόφρων
ὑπεδέξατο); cf. Kingsley (1999) 64. Floyd (1992) 252–4 observes that epic combina-
tions of πρόφρων and ὑποδέχεσθαι suggest a relation between agents of unequal
stations. Note esp. Demeter on Demophon at h.Hom. 2.226 (πρόφρων ὑποδέξομαι).
137
GJ 1.15–16 (South Italy (Hipponion), late fifth cent.). Feyerabend (1984) observes the
parallel in the (cultic and poetic) image of the road in B1 and GJ 1. For further echoes in
language and imagery between Parmenides and the Tablets, see Ranzato (2015) 66–70.
138
GJ 3.5; 6.7; 7.7; cf. GJ 2.11. ὄλβιος: e.g. GJ 5.9; τρισόλβιος: GJ 26a-b.1. Sophocles also
describes initiates as ‘thrice-blessed’, TrGF 4.837.1. In at least one case, the initiate’s
blessed state seems to involve breaking free of the cycle of metempsychosis, the notion

272
 5.3 The Proem in Its Time and Place
us, in several of the Tablets the soul’s eschatological, initiatory
journey culminates in an explicit articulation of its divinisation, an
idea which I argued is, in some form, often important if more
implicit in Greek mysteries elsewhere. At the end of his journey,
the initiate is now in a position to insist on his rightful divine status
and he either has been or will presently be divinised. First, the
initiate’s soul formulaically states to the guardians: ‘I am a child of
Earth and starry Sky (γῆς . . . καὶ οὐρανοῦ), but my race is heavenly
(αὐτάρ . . . οὐράνιον).’139 Hesiod speaks of ‘the sacred race of
immortals . . . who were born from Earth and starry Sky’ (Th.
105–6), and the kinship specifically of initiates with the divine
can be paralleled elsewhere.140 Second, and more explicitly, in
several fourth-century South Italian Tablets, the initiate finally
states to the unnamed goddess: ‘For I also claim to be of your
blessed race (ὑμῶν γένος).’141 In one case, the initiate is answered
expressly: ‘ . . . you will be a god instead of a mortal (θεὸς δ’ ἔσηι
ἀντὶ βροτοῖο).’142 In another South Italian Tablet, the initiate is
similarly addressed: ‘You have become a god instead of a human
(θεὸς ἐγένου ἐξ ἀνθρώπου).’143

of which further relates the Tablets to Parmenides’ eschatology as we reconstructed it

in Ch. 5.2: see GJ 5.5, cited in Ch. 5.2 with fr. 229 Kern = 348 Bernabé. On these texts,
see further Johnston (2013b) 118–19, 127, 132–3. Several other sources link mystery
initiation with notions of metempsychosis (e.g. Pl. Lg. 9.870d5-e3), but this connection
should not be exaggerated. For both further sources and qualifications, see Burkert
(1987) 87–8.
139
GJ 2.6 (South Italy (Petelia), fourth cent.); 8.12, 15 (Sicily, circa third cent.); 29.3–4
(Thessaly, fourth cent.); uel sim.: GJ 25.8–9 (Thessaly, 350–300); without the last
clause: GJ 1.10; 10–14.3; 16.3; 18.3.
140
For the comparison with Th. 105–6, cf. Johnston (2013b) 114–16, 124. An inscription
at the entrance to a Meter sanctuary in Crete welcomes initiates but warns ‘those who
transgress into the race of the gods’ (θιῶν γένος, fr. 32.iv.1–3 Kern = 568.1–3 Bernabé).
(More specifically, the goddess welcomes the initiates οἳ γονεὰν ὑπέχονται (2b); the
sense is uncertain, but ‘who guarantee their lineage’ is one plausible possibility, see
Burkert (1987) 157, n.58.) In the ps.-Platonic Axiochus, a dying man in need of
consolation is termed ‘kin to the gods’ (γεννήτης τῶν θεῶν) by virtue of the
Eleusinian mysteries, 371d5-e4 with Rohde (1925) 602–3; see further Burkert (1987)
76–7; (1972) 359.
141
GJ 5.3; 6.3; 7.3 (South Italy (Thurii) fourth cent.).
142
GJ 5.9 (South Italy (Thurii), fourth cent.). One immediately recalls Emped. DK31
B112.4 (see Ch. 6.2, n.51).
143
GJ 3.4 (South Italy (Thurii), fourth cent.), cf. the later GJ 9.2–4 (Rome, second or third
cent. CE). Burkert (1985) 295 writes that these statements go beyond everything else
known from Classical Greek mysteries. But, if the strong language of apotheosis
(theos) is indeed unique, the initiate’s immortalisation nonetheless takes place only

273
 How Could Parmenides Have Written Alêtheia?
Now, unlike the speaker in the Tablets, the kouros is, pointedly,
not yet dead (B1.26). Any statement about views which
Parmenides may have held concerning the post-mortem fate of
the initiated soul will be pure speculation, and extrinsic to a study
of the poem. We have seen throughout that Parmenides appropri-
ates traditional religious models of this kind in a decidedly crea-
tive manner. By contrast with the interpretative approach that
Peter Kingsley (1999) has made famous (or notorious), analysing
the religious context of Parmenides’ thought, and recognising the
interpretative exigency of such analyses, does not warrant us
simply to reduce or assimilate Parmenides to those religious
models which he appropriates. Our own analysis in this section
of Parmenides’ appropriation of the language and imagery of
mystery initiations, and especially the comparison with the initia-
tory Gold Tablets of South Italy, further substantiates the same
conclusion – at which we first arrived through an internal evalua-
tion of the evidence – that Parmenides frames the kouros’ escha-
tological journey to the goddess as one which takes its culmination
to be a transformative divinisation of the mortal through an iden-
tification with his divine soul. Our interpretation of Parmenides’
application of this initiatory model of divinisation – of the nature
and role of divinisation in his thought – must be context-sensitive.
Before detailing that conclusion (Chapter 5.4), we will briefly
address the association in our sources of Parmenides with the
Pythagoreans. An association of some kind is eminently plausible.
The Pythagorean communities, lying close to Elea, were an impor-
tant social and political force in contemporary South Italy.144
Some of our reports, furthermore, do not sound like mere formu-
laic, doxographic systematisation. Notably, Sotion records the
very specific and distinctive testimony that Parmenides estab-
lished a hero shrine for the deceased Pythagorean Ameinias,
who first led him ‘into quietude’ (εἰς ἡσυχίαν), a familiar term

after death and is as such equivalent to what, as I argued above, we also get in our
evidence for the Eleusinian mysteries. Bernabé and Cristóbal (2008) 178 reasonably
suggest that these statements indicate the acquisition of superhuman status, but not
thereby equality with the major gods (i.e. something like Heracles’ full-blown
deification).
144
On our evidence for the society and political activities of early Pythagorean commu-
nities in South Italy, see Burkert (1972) 115–20; Horky (2013) 85–124; Rowett (2014).

274
 5.3 The Proem in Its Time and Place
associated with the Pythagorean life.145 The Tabula of Cebes (of
uncertain date) tantalisingly speaks about an old wise man ‘who was
emulating the Pythagorean and Parmenidean life (βίον) in both
word and deed’ (2.2.2–5).146 The early Pythagoreans advanced
notions of metempsychosis and the divinity of the soul.147 Our
identification of these same notions in Parmenides (Chapter 5.2)
lends further support to reports of some sort of relation with the
Pythagoreans, although our evidence allows for nothing like
detailed elaborations.
Aristotle reports that, in their ‘very secret doctrines’ (ἐν τοῖς
πάνυ ἀπορρήτοις), the Pythagoreans observed a taxonomy of
rational beings, according to which ‘one kind is a god, one is a
human, and one is like Pythagoras’ (τὸ μέν ἐστι θεός, τὸ δ’
ἄνθρωπος, τὸ δὲ οἷον Πυθαγόρας, fr. 192 Rose).148 Whether or
not the early Pythagoreans put the point in this or even a proximate
way, the hagiographic traditions surrounding Pythagoras support
tracing to them some such conception and fit with Aristotle’s
remark. To name only three among numerous other reported
feats, Pythagoras commands a unique ability to recall facts about
his earlier incarnations (as probably reflected in Empedocles
DK31 B129), is observed in different places simultaneously, and
intuitively issues miraculous predictions.149 In the most extreme
case, Pythagoras was identified wholesale with Hyperborean
145
D.L. 9.21. For ‘quietude’ as a Pythagorean observance, see D.L. 8.7, 10 (cf. 8.31–2);
Luc. Auct. 3; Isoc. Bus. 29.6–8; Iambl. VP 95–7, 114, 197, cf. 10, 64–5; Hippol. Haer.
1.2.18, with Burnet (1930) 94–5; Burkert (1969) 28, n.62; (1972) 178–9, 199; Coxon
(2009) 40–1; Kingsley (1999) 179–87 (caveat lector); Gemelli Marciano (2014) 144–
5. Hero cult itself relates to the theology of approximation: it is regularly concomitant
with a conception of the worshipped hero as a former mortal, who now enjoys a higher-
than-mortal status; see e.g. on the heroised Archilochus, E1 II.50–1 apud Clay (2004)
106 ([ἀθ]ά̣νατος . . . [ἔ]σται), and Cleomedes, Paus. 6.9.8 (μηκέτι θνητόν); cf. Currie
(2002) on Euthymus. In GJ 2.11 (fourth century, Petelia), the initiate is promised he
will ‘rule among the heroes’. On heroisations of philosophers after death, see Grau
(2013).
146
For further testimonia linking Parmenides with the Pythagoreans, see Iamblichus,
Proclus and Photius in A4; Anatolius A44; Iamblichus VP 267.19 = Coxon (2009)
Testim. 154; with Burkert (1969) 28; (1972) 280; Kingsley (1999) 223–4; cf. also
Robbiano (2006) 135–6.
147
Ch. 5.2.
148
Ps.-Apollonius counts Pythagoras ‘in the race of daimones’, Ep. 50.1–2; cf. Iambl. VP
10–11.
149
Burkert (1972) 136–47 comprehensively surveys the hagiographic tradition; cf. Dodds
(1951) 144–5 (with notes).

275
 How Could Parmenides Have Written Alêtheia?
Apollo.150 Barring this limiting case of full-blown apotheosis,
however, Pythagoras emerges from these traditions as the tertium
quid which Aristotle describes: as a mortal who is decidedly closer
to the divine or more divine.151
Most interesting for us is the fact that our sources, indirect and
satirical as they are, also link Pythagoras’ exceptionally divine
status to the tradition that he undertook a descent to the
Underworld (a katabasis). In Hermippus’ parodying version of
this tradition, Pythagoras in fact hides in an underground chamber,
to which his mother relays ‘what happened . . . and the time’ (τὰ
γινόμενα . . . καὶ τὸν χρόνον), presumably, that is, who died and
when. An emaciated Pythagoras then re-emerges, allegedly from
an eschatological katabasis, and dupes his audience into believing
that he narrates to them his experiences from such a journey and,
therefore, that he is ‘some divine person’ (θεῖόν τινα, apud
Diogenes Laertius, 8.41).152 The prospect of an approach to a
higher, divine status in journeys like Pythagoras’ eschatological
katabasis corresponds to traditions, prominent in South Italy, of
parallel journeys undertaken by the likes of Heracles and
Orpheus.153 The link between such journeys and a heightened
status was sufficiently familiar for Sophocles’ Orestes to state
without further elaboration: ‘For often before now I have seen

150
Ar. apud Ael. VH 2.26; cf. Iambl. VP 30.14–18. See further Burkert (1972) 141–2, 149
for sources and discussion.
151
Iamblichus expressly reports inferences from Pythagoras’ feats to his super-human
status, VP 143 (κρείττονος . . . οὐχὶ ἀνθρώπου).
152
Diogenes (8.21) outlines a similarly snide katabasis of Pythagoras also by Hieronymus
of Rhodes. Hermippus’ satire is reflected e.g. in Σ Soph. El. 62 Papageorgius.
Herodotus (4.94–6) records a similar story about Zalmoxis, presenting him as a former
slave of Pythagoras and linking the ‘immortalising Getai’ (Γέται ἀθανατίζοντες) to the
Pythagoreans. Concerning Hermippus’ satire, Burkert brilliantly argues that the strange
detail about Pythagoras’ ‘mother’ suggests a parallel role, in the Pythagorean traditions
which he satirises, for a divine Meter, or Demeter, to whom Pythagoras owed his
astonishing wisdom, see Burkert (1969) 23, 26; (1972) 159; Bonnechere (2003) 172.
For Demeter as ‘Meter’, see e.g. Andoc. Myst. 124.7. Pythagoras’ house in Croton was
converted into a temple for Demeter, Tim. BNJ 566 F131. If this is right, then
Pythagoras too encounters in his eschatological journey a female deity, like the initiate
of the Tablets, and like Parmenides’ kouros.
153
On these katabaseis, especially in South Italy, see Burkert (1969) 5, 17, 25 n.57; (1985)
195, 296; (1972) 130, n.57; Kingsley (1999) 61–3; Graf and Johnston (2013b) 173–6;
Bremmer (2014) 59–61; for Orpheus, also frr. 978–1002, 1087 Bernabé. Plato puta-
tively ascribes an initiatory myth about the soul in Hades to ‘some clever man, perhaps
some Sicilian or Italian’, Gorg. 493a1–c3, with Dodds (1951) 225, n.5.

276
 5.4 Conclusions
clever men die falsely in words; then, when they return home, they
are held in greater honour’ (ἐκτετίμηνται πλέον, El. 62–4).154 We
may finally note in passing the anecdotes surrounding such ‘divine
men’ figures, frequently associated with South Italy, as Aithalides,
Aristeas, Abaris, Hermotimus, Epimenides, Phormio and Leonymus,
who are linked in our sources with Pythagoras, and who are accorded
parallel, super-human feats. In particular, the traditions that these
divine men performed miraculous journeys, undertaken at least in
some versions with their souls and while their bodies themselves
remained stationary, have long elicited comparisons with the extra-
ordinary journey described in Parmenides’ proem.155 The kouros,
then, in undertaking as a living mortal an eschatological journey –
and perhaps especially in journeying with or as his soul – would seem
to evoke also the Pythagorean model and these current and local
models of the exceptionally divine man.156

5.4 Conclusions: How Could Parmenides Have Written

Alêtheia?
Parmenides describes the traveller in the proem as a ‘knowing
man’ (εἰδότα φῶτα, B1.3). But, as in comparable passages which
identify initiates as ‘those who know’, he does not elaborate what

154
Proclus reports that ‘many among the ancients’, including Democritus, wrote about
those who have been thought to die and subsequently returned, in R. 2.113.6–9 = DK68
B1.1–4.
155
For psychic journeys in these traditions, see e.g. Pliny NH 7.174–5, cf. Hdt. 4.15; Σ
Apoll. Rhod. 1.645 = Pher. DK7 B8. For the moniker theios anêr itself, see Pl. Lg.
1.642d5. The traditions and the sources are discussed in Bremmer (1983) 24–53; Dodds
(1951) 141–2 (with notes); Burkert (1972) 147–53; on Aristeas, see further Bolton
(1962). For broad comparisons of these traditions with Parmenides B1, see e.g. Guthrie
(1965) 11–12; Mourelatos (2008a) 41–5. Aristotle reportedly hesitated as to whether
Hermotimus or Anaxagoras first stated that ‘mind (nous) is the god in us . . . and that
mortal life contains a portion of some god’ (fr. 61 Rose; cf. Metaph. A.3 984b15–22);
cf. Betegh (2006a) 636; (2012) 115–18.
156
A first-century CE inscription, found in Elea on the base of Parmenides’ bust, reads:
‘Parmeneides, son of Pyres, Ouliadês, physikos’. Several scholars speculatively but not
implausibly argue that, along with the titles oulis and phôlarchos on related finds, this
inscription points to an Apolline medical association, which practised incubation and
recognised Parmenides as its founder, see Kingsley (1999) esp. 77–86, 139ff; Ustinova
(2004); cf. Burkert (1969) 22. Pace Kingsley, however, these late inscriptions hardly
afford the key for the decipherment of Parmenides’ fragments, as Palmer (2009) 61,
n.37 rightly objects. At most, they interestingly disclose a local reception of
Parmenides as a figure of religious authority.

277
 How Could Parmenides Have Written Alêtheia?
it is that the knower knows. The phrase could signal that the kouros
should be thought of as (like) an initiate without alluding further to
knowledge of Parmenides’ philosophy. Alternatively, that the way,
or the goddess, ‘bears the man who knows’ may signify that the
kouros has now been set on the path which ultimately leads the
mortal to knowledge of reality.157 Either way, the proem’s ‘way’
(ὁδός) cannot be identified with the ‘way’ of inquiry of what-is
(‘that [it] is and cannot not be’) which the goddess announces
later – in B2 – because this will imply that, in B2, the goddess is
telling Parmenides what way he is to follow after he has already
traversed it. Rather, as Malcolm Schofield puts it: ‘The implicit
question tackled in Fr. 1 is: “What puts someone in the position to
raise and understand the goddess’s questions of Fr. 2?”’158
The foregoing study of Parmenides allows us to frame this
question in the context of a more specific problem. Parmenides
identifies as the essence and conceptual basis of the cosmology
developed in Doxa the very same krisis between the two opposite
sensory elements in accordance with which specifically mortal
cognition functions by brute, physiological necessity. It is the
qualitatively different krisis between ‘is’ and ‘is not’ which struc-
tures the argumentation of Alêtheia and frames its concept of
what-is. I argued in Chapter 4 that Parmenides wrote Doxa in
part because thinking in accordance with the Doxastic krisis, and
experiencing, recognising and forming first-order beliefs with
regard to a multiplicity of heterogeneous things, is an inescapable
aspect of what it is to think and live as a mortal. This analysis,
however, left us with a new difficulty. If, being a mortal,
Parmenides must think in Doxa’s structuring categories, i.e.

157
Cf. Mansfeld (1964) 227; Roloff (1970) 172; Lesher (1994a) 7. The scholastic argu-
ment that the words ‘knowing man’ imply that the kouros is knowledgeable already in
B1.3, and so prior to and independently of meeting the goddess (e.g. Torgerson (2006)
25, 30, 41), is wholly unpersuasive. That the way on which the mares set the kouros is
‘the way which carries the knower’ (the argument presupposes that ὁδός rather than
δαίμων is the antecedent of ἥ) simply does not imply that the kouros must already be a
knower. If the phrase alludes to knowledge of Parmenides’ philosophy, then the
comment must be proleptic for the kouros: indeed, the comment is itself part (recall
the mares) of the very dramatic setting which, even if nothing else, manifestly portrays
the goddess’s disclosure as anterior to the kouros’ acquisition of knowledge (B1.28b-
30). Mansfeld (1964) 225–9 offers different objections.
158
Schofield (1987) 357; see also Coxon (2009) 12–13.

278
 5.4 Conclusions
through the krisis between Light and Night, then how could he
have also thought in the principal categories of Alêtheia, i.e.
through the qualitatively different krisis between ‘is’ and ‘is
not’? If, being a mortal, it is appropriate for Parmenides to learn
mortal beliefs, then why is it also appropriate for him to learn the
unshaken heart of well-rounded reality (χρεὼ δέ σε πάντα
πυθέσθαι, B1.28b)?
Schofield outlines the following answer to the question of the
proem as he posed it: ‘The divine represents a perspective opposed
to our own human perspective. The way of the goddess represents . . .
the abandonment of a merely mortal way of looking at things – of
βροτῶν δόξαι, in short.’159 We are now, again, in a position to
integrate and to develop this attractive but embryonic idea within a
fuller and more particular epistemological and theological frame-
work. In Chapter 5.3, we analysed the proem in itself, and in the
light of Parmenides’ own eschatology and psychology (as recon-
structed in Chapter 5.2) and of a variety of religious and literary
paradigms, both local and pan-Hellenic, with which it engages den-
sely and creatively. Our analysis recommends the following conclu-
sions. In keeping with Greek religious thought quite generally (but
unlike Xenophanes), Parmenides does not conceive of the human
agent as an agent who is simply and strictly human. The mortal, for
Parmenides, is a complex hybrid, which comprises both divine and
mortal aspects and parts. In attaining the disposition which is neces-
sary for acquiring knowledge of what-is, the living kouros is cogni-
tively identified with the divine within him – i.e. with his Hot,
aethereal soul – and, as such, undergoes a process of epistemic
homoiôsis theôi.
Greek initiation is always and everywhere transformative, and it
is the divine which facilitates the initiate’s transformation. The
epistemological thrust of the encounter between the kouros and the
goddess is, I suggest, the following. With her injunctions and
arguments throughout Alêtheia, the goddess repeatedly enjoins
the kouros to think and to attain a certain understanding in new
ways, and through an unfamiliar kind of opposition (‘is’/‘is not’),
which his own mortal mind could not have identified or

159
Schofield, loc. cit.

279
 How Could Parmenides Have Written Alêtheia?
recognised, and she guides him in doing so. The mortal is able to
heed the goddess’s injunctions and follow her guidance only
because there is in him a non-mortal, divine part – a pure Hot and,
as such, supremely vital and cognitively acute and active part –
which can be responsive to such injunctions and guidance. The
process of following the goddess’s injunctions and arguments is a
process of becoming cognitively identified with one’s divine soul to
the temporary suspension of the mortal parts and aspects of one’s
composite mortal mind. It is in gradually acquiring the ability to
register and evaluate Alêtheia’s qualitatively different kind of
thought that the kouros thus undergoes a process of cognitive
homoiôsis theôi. Importantly, the kouros could not have undergone
this process in the absence of his encounter with the goddess. The
human mind, a prisoner to passively determined, uncontrollable
physiological processes, could never have pulled itself up by its
own bootstraps. This point follows from Parmenides’ theory of
human cognition. It is confirmed and reflected in his repeated
characterisations of the kouros as dependent on the guidance and
transportation of his divine guides and, above all, in the goddess’s
divine disclosure with which his journey culminates.
As we saw in Chapter 5.1, what-is cannot itself be described as
Light or in terms of a certain proportion or relation between Light
and Night. The thinking which is reflected in Alêtheia, framed by
the krisis between ‘is’ and ‘is not’, is qualitatively different from
the thinking which underpins Doxa, framed by the krisis between
Light and Night. Notably, the interplay of Light and Darkness
dominates the proem but disappears entirely when the goddess
embarks on her exposition of the way ‘that [it] is’. Parmenides,
then, is able to explain how a mortal can sustain a higher-than-
mortal thought by thinking with his divine Light soul, but this
does not mean that Light and what-is can be identified as the
same item in the world nor, therefore, as the very same object of
knowledge. Nonetheless, on the basis of our discussion of
Alêtheia’s krisis and Doxa’s krisis (Chapter 4.3), we can suggest
one sense in which the knowing mortal’s pure-Light cognitive
constitution is indeed structurally suited to thinking in terms of
Alêtheia’s krisis. Alêtheia’s krisis involves retaining one side
(‘is’) and rejecting the other (‘is not’). Doxa’s krisis, conversely,
280
 5.4 Conclusions
involves retaining both elements (Light and Night) and essen-
tially relating each one to its enantiomorphic counterpart. When
we think purely with our Light soul, however, we are in a position
to cognise pure Light and no longer any Night. We thus adopt a
physiological frame-of-mind which is suited to thinking and
being aware in terms of only one thing, without an equally
positive and equally reified enantiomorphic counterpart.160
Thus, the Light that we perceive is not itself what-is. And yet,
when we temporarily acquire a pure-Light cognitive constitution,
we are in a position to apply our mind to the goddess’s exposition
of what-is as something that is wholly homogeneous and unified,
something that is precisely free of those clashes – between
enantiomorphic and equally reified opposites – that constantly
dominate human experience and human thought.
Both Plato and Aristotle, for whom homoiôsis theôi similarly
means identifying ourselves with what is divine in us, relate such
divinisation to our capacity for reasoning and contemplation. In
this respect, as in many others, they are the descendants of
Parmenides. The deductive argumentation that permeates
Alêtheia constitutes a distinctive and pioneering sort of rational
thought. By momentarily suspending thought through the krisis
‘Light/Night’, and being liberated to recognise instead the quali-
tatively different krisis ‘is / is not’, the kouros can take the further
step of selecting ‘is’ and rejecting ‘is not’. Having thus formed the
notion of ‘what-is’, the kouros can then follow the tightly devel-
oped arguments of B8, which throughout deduce the properties of
what-is on the basis of the contrast between ‘is’ and ‘is not’, and
grasp what this thing must be like. He recognises that, to be what
strictly is and what has no share whatsoever in what-is-not, it must
be perfected, ungenerated, imperishable, indivisible, homoge-
neous, immobile, etc.161 It is through this process that the goddess
enjoins the kouros to persevere when she instructs him: ‘judge

160
I thank Gábor Betegh for this observation.
161
It is beyond the scope of this study to offer reconstructions of the different arguments in
B8 for these properties of what-is. For such reconstructions, see e.g. McKirahan
(2008). That the primary mainspring of these arguments is the contrast ‘is / is not’,
and the determination to exclude categorically ‘is not’ from what-is, is clear and
expressly stated (B8.15–16).

281
 How Could Parmenides Have Written Alêtheia?
(krinai) by reasoning (logos) the much-contested examination
spoken by me’ (B7.5–6). With the injunction to judge or discri-
minate, the goddess invokes Alêtheia’s krisis between ‘is’ and ‘is
not’. With the term logos, she is looking forward to the deductions
which are argued for on this basis in her examination (cf. B8.50).
For Parmenides, when we apply our mind to this argumentative
reasoning, and to the krisis between ‘is’ and ‘is not’ which frames
it, this amounts to suspending the operation of our mortal percep-
tual-cognitive apparatus and thinking instead with – or as – our
divine souls. For the Parmenides who emerged here, a human’s
mind can reason about ‘is’ and ‘is not’, but it is then a divinised
and no longer a human mind, thinking in non-human, divine ways
and attaining non-human, divine knowledge.162
The interpretation of the indispensable function of the goddess
advanced here thus fully accommodates the undeniable promi-
nence of argumentative reasoning in the poem. Indeed, it is even
compatible with the view (although it does not require it) that the
logic of Alêtheia is designed to be self-sufficient and that none of
its premises are in need of external divine ratification.163 Far from
demanding passive acceptance, the goddess bombards the kouros
with injunctions to engage his mind in an evaluation of her
disclosure.164 Indeed, the tenets of Alêtheia could not have been
communicated to a piously obedient but uncritical mind. It is only
by mastering Alêtheia’s argumentative reasoning, by actively
applying his liberated understanding to the krisis ‘is / is not’, that
the kouros can register, evaluate and accept for himself its account
of the unshaken heart of reality.
None of the above implies that Parmenides is deified pure and
simple so that he can simply dispense with human thinking.

162
Lesher (2008) 473–6 is, on my view, precisely wrong to claim that Parmenides
championed the reasoning of ‘the human mind’ and sought to ‘humanize knowledge’
(my emphases). Specifically ‘human’ cognition is analysed in B16 (ἀνθρώποισι . . .
ἀνθρώποισιν). It is the very Doxastic cognition beyond which we must go (Ch. 4.2–3).
Equally, therefore, we cannot accept the postulation in Curd (2015) that, while
Parmenides maintains that ‘mortal’ thinking (βροτῶν δόξας, B1.30) cannot grasp
what-is, he simultaneously champions the capacity of ‘human’ thinking to do so.
Note also B1.27: the kouros has travelled far from the track of ‘humans’ (and cf.
ἄνθρωποι at B19.3).
163
On this question, see further Ch. 6.3.
164
See B4.1; B6.2; B7.2, 5–6; cf. B8.7–8, 52.

282
 5.4 Conclusions
Despite momentary achievements of such cognitive homoiôsis
theôi, the kouros remains a living mortal. He remains constrained
in the course of his life to experience, and to think and form first-
order beliefs about, multiple, mobile, differentiated things in
accordance with contrasts between sensory opposites. The ineluct-
ability and appropriateness of Doxastic cognition to the mortal and
to mortal life is no less part of the teachings of the goddess than
Alêtheia. As Plato and Aristotle will later prescribe, Parmenides
strove to become divine insofar as this is possible for a mortal. To
use an imperfect analogy: for Parmenides, thinking rationally
(where this means the type of thinking which underpins
Alêtheia) is like holding your breath, whereas thinking like a
mortal is like breathing. Breathing is an inexorable and unavoid-
able part of human life. And yet you can, with effort, suspend that
human necessity for stretches of time, even if you cannot lead a
human life in this way.
Significantly, the goddess also renders the kouros a prophet of
sorts and enjoins him to convey her disclosure (B2.1): ‘Come now,
I will tell you, and do you convey the discourse once you have
heard it’ (κόμισαι δὲ σὺ μῦθον ἀκούσας).165 The goddess’s many
second-person addresses and injunctions have the effect of placing
the reader in the kouros’ role.166 Indeed, if Parmenides is to render
coherent not only his own but also his readers’ ability to acquire
knowledge of what-is, then his text – like all initiatory texts – must
be thought of as performative and transformative. In those who
read, or, rather, in those who gradually acquire the ability to
understand Alêtheia and its argumentation, the poem must effect
the same initiation and the same cognitive transformation which
the kouros’ unmediated, initiatory encounter with the goddess
effected in him. For us, no less than for Parmenides himself, the
process of acquiring the ability to apply our minds to thinking
about ‘is’ and ‘is not’ is the process of identifying ourselves,
however momentarily, with our divine soul and, therefore, of

165
For this force of κομίζω, see Mourelatos (2008a) 17 with n.20; cf. e.g. Isoc. Bus. 28.5–7
(. . . ἄλλην φιλοσοφίαν . . . εἰς τοὺς Ἕλληνας ἐκόμισεν).
166
Mourelatos (2008a) 16 suggests that Parmenides leaves the kouros unnamed precisely
in order to allow his audience to identify themselves with the protagonist.

283
 How Could Parmenides Have Written Alêtheia?
undergoing a cognitive homoiôsis theôi. For us, no less than
for Parmenides himself, applying our mind to such thought is
possible only thanks to his own first encounter with the goddess.
By guiding him to a cognitive homoiôsis theôi, the goddess
enabled Parmenides to produce a text, which, by reprising this
encounter, is to effect the same cognitive transformation in its
audience.167
Even if Parmenides would have allowed that one could forma-
lise the arguments of Alêtheia without making reference to any
deities, he would, on my view, insist that we cannot explain any
mortal’s ability to register and evaluate such arguments without
making reference to that mortal’s divine soul. Furthermore, we
cannot explain how a mortal – Parmenides – was first able to
register and evaluate such arguments without making reference
to that mortal’s epistemically significant interaction with the
divine. By asking, ‘Why did Parmenides choose to represent his
philosophy as acquired by him through an interaction with a god-
dess?’, we wrongly presuppose that Parmenides conceived of his
philosophy independently of conceiving of it as acquired by him
through an interaction with a goddess. The poem subverts our
pervasive and deep-seated expectation that reasoning and divine
disclosure must constitute dissonant and opposing tendencies.
Argumentative reasoning and an epistemically significant interac-
tion with the divine play harmonious and equally indispensable
roles in Parmenides’ philosophy.

167
We might usefully compare the cognitive function of initiation in Parmenides, while
never losing sight of its distinctiveness, with associations elsewhere in our sources for
mystery initiations with the inducement in the initiate of what scholars broadly term
‘altered states of consciousness’. On these associations, see Burkert (1987) 89–91,
112–14 (with notes); Ustinova (2009) 226–55; cf. Bonnechere (2003) on katabasis,
initiation and altered consciousness in the oracle of Trophonius. A fragment of
Aristotle tantalisingly states that ‘those who are being initiated do not need to learn
something but to undergo something and to be put into a certain condition; that is, to
become fit-for-purpose’ (οὐ μαθεῖν τι δεῖν ἀλλὰ παθεῖν καὶ διατεθῆναι, δηλονότι
γενομένους ἐπιτηδείους, fr. 15 Rose). (At first sight, Aristotle’s statement may seem to
suggest that initiations were generally not thought to involve new understanding, but
the context in Synesius indicates rather a distinction between different forms and stages
of the acquisition of new understanding; see further Burkert (1987) 69–70; Graf and
Johnston (2013a) 248 n.78. At any rate, the Derveni author takes it as read that would-
be initiates hope in part to attain knowledge (εἰδήσειν), col.20; cf. Chrys. SVF 2.42; and
see further Burkert (1987), loc. cit.)

284
 5.5 Being and Intelligence in Alêtheia
5.5 The Ontological Question; Being, Intelligence
and Intelligibility in Alêtheia
The foregoing study of Parmenides could cohere with different
elaborations of what I called the ‘ontological question’,168 and of
the relation between being, intelligence and intelligibility in
Alêtheia. This section addresses both issues. My suggestions
here will be offered in a more tentative vein. Our aim is to consider
how we might best think about these much-disputed problems,
with our foregoing conclusions in mind, so as to achieve a more
holistic understanding of all the different aspects of Parmenides’
philosophy and see how this unified philosophy can be viewed as a
viable and even attractive one. But an important concern will be to
contend that the preceding analyses of the religious and epistemo-
logical aspects of the relation between the two parts of the poem do
not depend on a prior resolution of the problems addressed in this
section, and to gather together and systematise certain limited but
pertinent points which were already established in the course of
our discussions above.

5.5.1 The Ontological Question

In the light of the goddess’s account of ‘reality’, what is the status
of Doxastic things? How precisely do Doxastic things relate to
what-is? Parmenides’ ontology and cosmology nowhere explicate
this relation. Parmenides nowhere directly elucidates how the
ontology of Alêtheia coheres with the manifest availability of
Doxastic things for experience, discussion and belief. There is
no reason to assume that Parmenides did, despite his silence on
the matter, develop a fully worked-out approach to these ques-
tions. The project of isolating for explicit analysis and discussing
the relation between what unqualifiedly is (such-and-such) and the
multiple population of everyday experience waited to become a
focal point of Plato’s metaphysics.169

168
See the introduction to the chapters on Parmenides.
169
What Plato’s ‘Parmenides’ first critiques in the Parmenides (130e5–134e8) is specifi-
cally the coherence of Socrates’ novel notion of the relation of participation between
Forms and particulars.

285
 How Could Parmenides Have Written Alêtheia?
Answers to the ontological question, I suggest, offer philoso-
phical and interpretative elaborations for an issue which our text
itself raises, but never itself elaborates or addresses. I will criti-
cally explore here what strike me as the most instructive or viable
types of approach, and defend one philosophical and interpretative
elaboration which is, I think, on balance most faithful to
Parmenides’ text and to the tenor of his thought, and which is
expressed by some of his ancient readers. There is no question of
attempting here a comprehensive survey of ontological models.
Nor do I wish to exclude that there could be other viable elabora-
tions. First, however, let us recapitulate certain pertinent points,
which were established in the previous sections on Parmenides.
Doxastic things, we saw, are manifestly available for experience,
discussion and belief. The goddess speaks of, and to, Doxastic
things (mortals). Furthermore, we can expound such things cor-
rectly and incorrectly. Indeed, one can acquire knowledge of the
nature of Doxastic things and processes (B10). Doxa describes such
things correctly. It correctly expounds their origins, natures and
behaviours by tracing them to (interactions between) the two con-
stitutive principles which underlie them. It correctly explains, for
example, the human mind (B16) and procreation (B17–18).
Whatever else we say in response to the ontological question,
then, Parmenides clearly takes it that there are present and available
such things as mortals, who can experience and be experienced, err
and know, think and be thought about, speak and be spoken to and
about, and be expounded correctly or incorrectly. At the same time,
what-is is aligned with alêtheiê, understood in some sense as the
core or ultimate reality, and the kouros’ attitude towards Doxa is
informed by his knowledge of Alêtheia. He accepts Doxa’s theories
as accounts of Doxastic things and processes. But he also recognises
that beliefs about and accounts of Doxastic things form no part of an
account of what-is. Indeed, the mortals’ error consists not least in
mistaking Doxastic accounts for accounts of the nature of the core
or ultimate reality, and Doxa’s cosmology is ‘deceptive’ insofar as it
is liable to be misinterpreted in just this way.
I do not restate these points so that I could assume them as
premises in the following criticisms. I will not use them in this
way, even though I do believe that any viable answer to the
286
 5.5 Being and Intelligence in Alêtheia
ontological question will have to take them into account. Rather, I
recapitulate these conclusions in order to clarify what I take to be
the only relevant points on which the evidence permits us to insist.
Beyond these limited points, moreover, the arguments of the
previous sections regarding the religious and epistemological
aspects of Parmenides’ poem do not rely on any particular story
concerning the ontological status of Doxastic things or the nature
of the relation between what-is and Doxastic things.
The view that Alêtheia’s ontology implies the non-existence of
Doxastic things – call this ‘non-existentialism’ – was for a long time
the standard interpretation, but appears to have become a minority
position.170 Non-existentialism has its mainspring above all in the
fact that, when deducing the properties of what-is in B8, the goddess
does not issue assertions of the form ‘what is changeless is F’, but
rather ‘what-is is changeless’. At least prima facie, then, by pre-
dicating attributes to what-is (to eon), Parmenides does not seem to
delineate the attributes of a subset of the things which are, but to
state unqualifiedly that ‘what-is’ did not come to be and is imper-
ishable (B8.6–21), is homogeneous and indivisible (B8.22–5) and is
changeless and immobile (B8.26–31, 38, 41).171 But how could
Parmenides, then, still allow that there are some things, such as
mortals and stars, which are (what they are), and which do come to
be, change, move and perish? How could Parmenides reject gen-
eration and perishing (B8.21–2, 27–8) only with respect to what-is
as contrasted with other things (mortals, etc.), rather than with
respect to what-is and, therefore, categorically?172
Such considerations raise the challenging question of how
Parmenides could avoid committing himself to non-existentialism.
But, whatever our ultimate answer may be, ascribing to Parmenides
the unqualified view that Doxastic things (mortals, trees, stars, etc.)
170
But see e.g. Wedin (2014).
171
Even if one viewed esti in B2 as a paradigm of predication (‘ . . . is (. . .)’) without
existential force, the deductions of B8 undoubtedly specify the properties of a certain
entity, what-is. Some interpreters object for this reason to formal readings of B2, Gallop
(1979) 63; Barnes (1982) 161.
172
In framing the issue in this way, I sidestep the question of whether the extant fragments
converge on the thesis that what-is is unique (numerical monism). Melissus explicitly
infers some form of numerical monism, although from an anti-Parmenidean premise
(B6 with KRS (1983) 395). After Plato, numerical monism often crops up as the
doxographic hallmark of Parmenides.

287
 How Could Parmenides Have Written Alêtheia?
do not exist as a conscious and unapologetic philosophical tenet is in
many ways a non-starter. The untenable ramifications of this tradi-
tional view are not generally spelled out. If we maintain non-
existentialism, for example, we cannot also consistently maintain
that Doxastic things appear to mortals. If there are no such things as
Doxastic things or, therefore, mortals, then there is nothing there
either to give rise to or to experience such appearances. No one, I
take it, would suggest that it is the unchanging and immobile what-
is which either generates or registers the appearance that this
person is walking. Scholars often describe Doxastic things and
processes as ‘illusions’ or ‘hallucinations’.173 But, at least as a
description of non-existentialism, this is an incoherent articula-
tion of an interpretation which cannot be coherently articulated.
We experience an illusion, presumably, when we observe one
state-of-affairs and are under the impression that we are obser-
ving another state-of-affairs. We hallucinate when we mentally
conjure up a state-of-affairs.174 But all such scenarios require,
incompatibly with non-existentialism, that we are there to
observe or mentally conjure up states-of-affairs. It is surely not
what-is, as described in B8, which both creates and registers
illusions or hallucinations!175
The objection here is not merely that Parmenides neglects to
explain how, given his commitment to non-existentialism, Doxastic
things appear to mortals, but that he renders this matter inexplicable.
He closes off in advance any possible avenue of explanation. After
all, there is, ex hypothesi, only one candidate answer for what causes
and registers these appearances or illusions – what-is – and this
candidate is wholly unsuitable.
We can appreciate the difficulty from another direction by
asking whether the non-existentialist will classify Doxastic things

173
Nietzsche (1974) 122–4; Rohde (1925) 372; Reinhardt (1974) 310; Jaeger (1947) 105;
Burnet (1930) 182; Vlastos (2008) 376, n.29 (cf. (1946) 76, n.59; (1973) 65–6); Owen
(1960) 89; Tarán (1965) 230; Guthrie (1965) 75 (‘a hallucination or dream’); Furth
(1974) 268; Gallop (1984) 22–3; Lloyd (1987) 61; Sedley (1999a) 117; Gemelli
Marciano (2002) 87; Sisko (2003) 102; Trépanier (2004) 152.
174
Cf. Austin (1962) 22–5.
175
A more specific consequence of non-existentialism is that Parmenides himself does not
exist (Palmer (2009) 181–3 emphasises this point). A fortiori, it is not the case that
Parmenides wrote a poem, which argues for non-existentialism.

288
 5.5 Being and Intelligence in Alêtheia
as what-is or as what-is-not. Non-existentialism draws its appeal
in large part from the veneer of uncompromising rigour in its
insistence that Parmenides permits us a choice only between
these two alternatives: (it) is or (it) is not. Ex hypothesi, Doxastic
things could not be identified as what-is or classified under its
rubric as things-that-are. There is hardly a better prospect, how-
ever, for identifying them as what-is-not or classifying them as so
many things-that-are-not. What-is-not cannot be described,
pointed out or thought about (B2.7–8; B8.7–9). This is patently
false of Doxastic things. In fact, the goddess asserts that the way of
what-is-not cannot be inquired into (παναπευθέα, B2.6), but she
enjoins the kouros to inquire into Doxastic things (πυθέσθαι,
B1.28b; πεύσῃ, B10.4). For non-existentialists, therefore,
Doxastic things prove unclassifiable within what they take to be
an exhaustive framework. Indeed, this is why non-existentialists
fall back on the incoherent measure of describing Doxastic
things as ‘illusions’ (a term which lacks even a remote parallel
in the text).
Non-existentialism does not merely impute to Parmenides an
unbelievable thesis, but also imports into his text an internal
incoherence and inconsistency. By writing that mortals name
things, and by developing a theory of, say, embryology,
Parmenides commits himself, minimally, to the recognition that
it is possible that it should appear as if mortals name things and
such theories are developed. But this possibility is directly incon-
sistent with non-existentialism, unless the unchanging what-is is
what gives those names, develops such theories and registers the
appearances that doing so is possible.176 If Parmenides was a non-
existentialist, then he was also an extraordinarily confused and
inconsistent thinker.
This is not to say, however, that clear explanations of how
Parmenides could reconcile the view that Doxastic things exist –
‘existentialism’ – with the ontology of what-is are easily forth-
coming. One way of salvaging existentialism would be to maintain
that Alêtheia and Doxa describe two separate worlds, where the
176
Sedley (1999a) 125 (cf. also Nietzsche (1974) 132–3) questions how what-is could
‘find room to misconceive itself’, but does not pursue this insight to spell out the
interpretative nightmares attendant on unvarnished non-existentialism.

289
 How Could Parmenides Have Written Alêtheia?
goddess’s ontological strictures concerning the former do not
exclude the existence of Doxastic things in the latter.177 But the
fragments offer no textual support for such a two-world model.178
More damningly, if certain a priori logical and semantic consid-
erations deductively led us to accept the existence only of what-is
and, ex hypothesi, to exclude the existence of Doxastic things in
the first world, then why should those same considerations not
force us to the same conclusion concerning the population of the
second (third, fourth . . .) world?179
According to a different existentialist model, what-is and
Doxastic things are consubstantial. This model sees Parmenides
as a material monist. Parmenides takes what-is to be the single
underlying substance of which all things in the changeable cosmos
consist, much as (on one interpretation) Anaximenes posits that all
things are made of air. Light and Night, in turn, are ‘forms’ or
‘phase-states’ of what-is.180
One difficulty with this view is that differences between differ-
ent items and qualities in the world would have to be causally

177
E.g. Nehamas (2002) 62–3; cf. Clark (1969) 27.
178
Cf. de Rijk (1983) 30, 52; Palmer (2009) 183.
179
In the absence of a two-world solution, though, the spatially extended nature of what-is
as a uniformly extended, uninterrupted, immobile and delimited entity (B8.22–31, 42–
9) throws up yet another puzzle: how can the same world accommodate, not only
logically but also spatially, both the uninterrupted what-is and Doxastic things? In fact,
the same puzzle emerges concerning the apparent colocation of certain privileged
entities with all other things with regard to Xenophanes’ god (possibly), Anaxagoras’
Mind (n.b. DK59 B14: ‘most assuredly even now it is where also all the rest is’) and
Empedocles’ Love and Strife. One response is to consider the problem items incorpor-
eal, Curd (2010); (2013b). Parmenides’ spatial descriptions of what-is have often been
interpreted metaphorically, although there is no obvious warrant for doing so (e.g.
Owen (1960) 95–9; Mourelatos (2008a) 124, xxxv–xxxvii; contrast Schofield (1970)
131–4; Sedley (1999a) 117–22). Alternatively, Betegh (2016) 415–20 interestingly
suggests that it was the privileged and divine status of the special entities which led
Xenophanes, Empedocles and Anaxagoras to acquiesce in positing their colocation
with all other things without attempting to discover or determine the underlying
physical mechanisms, and compares similar phenomena in ancient Greek religious
attitudes to the physical relationship between gods and things. What-is is unquestion-
ably a special and privileged sort of entity, and a good case can be made that it enjoys
divine status, albeit in a pointedly non-traditional and refined sense; see Ch. 5.5.2 and
Ch. 6.1. Conversely, Palmer (2009) 180–8 equips Parmenides with sophisticated
explications of the physical mechanisms underpinning the spatial relation between
what-is and other things, but his solutions are not always clear; for criticisms, see
Wedin (2014) 243–5; Sisko and Weiss (2015) 47–8.
180
Sisko and Weiss (2015). Whatever its drawbacks, this model does have the advantage
of dispelling altogether any puzzle about spatial colocation.

290
 5.5 Being and Intelligence in Alêtheia
traceable to differences in how they relate to what-is, understood
as the one and only substance which underlies all things. By
comparison, this is why Anaximenes posits the processes of rar-
efaction and condensation, which explicate how the single under-
lying substance, air, issues in a plurality of different items like
clouds (from more rarefied air) and rocks (from more condensed
air). But no comparable account is possible in Parmenides’ case.
The unchanging what-is does not undergo qualitative modulations
(such as rarefaction and condensation) and Parmenides empha-
sises its thoroughgoing homogeneity and even distribution: ‘nor is
it divisible, since it is all alike; nor is it any more there . . . nor any
less’ (B8.22–5), and again: ‘for it is not right for it to be any greater
nor any smaller here or there . . . nor is what-is such that there
would be of what-is more here and less there’ (B8.44b-8). But
if what-is is the single underlying material of all things, and if,
for example, a certain stretch of Night is denser and heavier
than an equivalent stretch of Light (B8.59), then will we not
have to say that the Night stretch has a higher concentration of
what-is than the Light stretch? In general, it is difficult to see how
the heterogeneity of different items and processes in the world
could be explicable even in principle if the single underlying
substance of which they all consist is evenly distributed in a
completely uniform and unchanging way and can undergo no
qualitative changes.181
Another problem with this view stems from Parmenides’ asser-
tion that what-is abides without motion in the same place (B8.26–
31, 38, 41). There is no conceptual difficulty with the thought that
the ultimate substance of which a human or a flying arrow
consist is ungenerated and imperishable even though the
human and flying arrow are not. We need only think of
Democritean atoms. But we could not say in the same vein that

181
A similar objection applies to the non-existentialist hypothesis in Mourelatos (2008a)
xliv–xlviii and (2011) 184–9 that the relation between Alêtheia and Doxa is akin to the
relation between Sellars’ Scientific Image (SCI) and Manifest Image (MI), where the
theoretical objects of SCI exist and the observational ones of MI do not. It is clear how
the constituents and processes in SCI (e.g. atoms) could causally underlie and, indeed,
explain the (only apparent) constituents and processes of MI (e.g. humans). But it is not
clear how what-is could be related to Doxastic things in a comparable way. Cosgrove
(2014) 15 objects similarly.

291
 How Could Parmenides Have Written Alêtheia?
the fundamental material of which they are composed always stays
put even though the human being and flying arrow do not.182
Parmenides’ remarks concerning the indivisibility, homogene-
ity and even distribution of what-is seem to warn precisely against
thinking of what-is as consubstantial with the rest of the cosmos’
population. After all, this would subject it to the internal differ-
entiations which characterise this population. The same concern
recurs in B4:
οὐ γὰρ ἀποτμήξει τὸ ἐὸν τοῦ ἐόντος ἔχεσθαι 1
οὔτε σκιδνάμενον πάντῃ πάντως κατὰ κόσμον
οὔτε συνιστάμενον. 3

For you will not cut off what-is from holding fast to what-is,
Neither dispersing everywhere every way in a world-order
Nor drawing together.183
Parmenides, then, distances himself from the idea that what-is and
Doxastic things are consubstantial. He precisely discourages us
from construing what-is as a cosmogonic or cosmological starting
point in the Ionian vein. Overall, Parmenides displays a complex
attitude towards the Ionian project, at once critical and appropria-
tive. On the one hand, Light and Night are Ionian-style cosmolo-
gical principles, to which all the phenomena of change and
differentiation are traced and reduced in an explanatory account.
On the other hand, Parmenides denies that what we find in those
cosmological principles is the ultimate and fundamental reality.
Indeed, mistaking them for the ultimate reality is a central facet of
the mortals’ error (Chapter 4.4–5).

182
Sisko and Weiss (2015) 55 with n.35 are aware of the latter difficulty. They solve it by
maintaining that the property of immobility applies to the totality of what-is (it does not
egress its bounds) and positing that Parmenides’ physics involved what they call
‘antiperistasis’, i.e. motions of cyclical replacement within a voidless plenum, and so
allowed relative changes of location within the spatial limits of what-is. It is not clear
precisely what sort of internal motions Sisko and Weiss have in mind or how they are
meant to ground the complex and heterogeneous motions involved in Parmenides’
cosmogony and cosmology. At any rate, what-is is not only a voidless plenum but also
an entirely homogeneous and evenly distributed one, with no distinctions or diver-
gences between different areas or parts. A plurality of different motions within it would
mark off the differently moving areas from one another. A simple rotation of the sphere
on its axis would not raise this problem but could certainly not exhaust all the intra-
cosmic motions of Parmenides’ cosmology.
183
On this point, cf. Palmer (2009) 184–5; for kosmos as ‘world order’, see Palmer (2009)
184, n.65 (comparing Herac. DK22 B30).

292
 5.5 Being and Intelligence in Alêtheia
A more metaphysical version of the idea that what-is and
Doxastic things are consubstantial takes what-is to be a steady and
unchanging substance in all things or the true essential nature of
each and every thing (but not thereby their material substrate).184
But the problem of immobility recurs. It is not easy to see what it
might mean to say that the essential nature or substance of the arrow
abides without motion in the same place even as that same arrow
itself flies. This model thus ultimately encourages us to think of
what-is as a property (beingness185) rather than as an entity. But it is
at least doubtful that we can square such a deflationary construal of
what-is with the deductions and specifications in B8 of the different
properties which it bears (homogeneity, indivisibility, boundedness,
sphericity). It is worth stressing that all these problems with the
various attempts to make what-is and Doxastic things consubstan-
tial do not undermine the idea that they are compresent. As far as
these problems go, we can still think of the same spherical plenum
which is the world as inhabited at one and the same time both by
what-is and by the many different things. Insofar as what-is is
concerned, the world is entirely homogeneous: what-is is the same
everywhere without any internal articulation. The same space occu-
pied by this human and this tree – and, indeed, by all other such
discrete items – is also occupied by an uninterrupted and continuous
stretch of being. Nothing ‘happens’ to this plenum of being when
this human moves from here to there. We ran up against our
conceptual problems (and against B4 and B8) only once we tried
to think of the unchanging, continuous and homogeneous what-is as
itself constituting the being of this tree, or the being of this human,
either as their material substrate or as their abiding essence or
nature.186
The most radical existentialist proposal is that we should think
of Doxastic things themselves (humans, trees, etc.) as so many
beings or ‘things-that-are’.187 The difficulty here is more direct. If
Doxastic things are things-that-are (in the loaded sense of
Alêtheia), then they must have the properties of what-is as
184 185
De Rijk (1983). As de Rijk (1983) himself puts it.
186
The compresence of what-is with Doxastic things does, however, involve the phenom-
enon of ‘colocation’; see n.179 above.
187
E.g. Thanassas (2011) 291–4; Pulpito (2011) 203–6.

293
 How Could Parmenides Have Written Alêtheia?
explicated in B8. But it is essential to humans qua humans (or
pebbles qua pebbles, etc.) that they are generated, perishable,
mobile and not homogeneous. If the idea is that it is not qua
human that some particular thing is, but qua entity, then we are
no longer talking about the same thing or about the being or
existence of humans. How can we say that a particular human
(or tree, etc.) never came to be and does not move, insofar as it –
i.e. that same, particular human – is an entity?
The final existentialist model I will consider, before proceeding to
my own suggestions, is the hypothesis developed by John Palmer
(2009) that Parmenides’ metaphysics is founded on the modal demar-
cation of necessary and contingent being. Palmer highlights the
modal clauses in Parmenides’ initial articulation of Alêtheia’s krisis
in B2: ‘that [it] is and that [it] cannot not be . . . that [it] is not and that
[it] must not be’. If B8 refers, not simply to that which is, but, more
restrictedly, to what-is-and-must-be, then its rejection of generation,
perishing, change and mobility with regard to necessary being
need not extend also to contingent things, like mortals and trees,
which are (what they are), but not necessarily. The ontology of
what-is, which always signifies what-is-and-must-be, is thus
compatible with the (contingent, non-necessary) existence of
Doxastic things.188 Palmer ingeniously and painstakingly recon-
structs a coherent and attractive philosophical position. The
difficulties with it are textual.
Palmer observes that nothing whatsoever suggests that the sig-
nificance of the modal clauses (B2.3b, B2.5b) is already somehow
present in the assertoric clauses (B2.3a, B2.5a).189 It is thus all the
more noteworthy, however, that Parmenides himself omits the
modal clauses, and retains only the assertoric ones, when he
reprises Alêtheia’s krisis as that between ‘[it] is’ and ‘[it] is not’
(ἡ δὲ κρίσις . . . ἔστιν ἢ οὐκ ἔστιν, B8.15–16), and describes the first
way as the way ‘that [it] is’ (μῦθος ὁδοῖο . . . ὡς ἔστιν, B8.1–2).
Indeed, throughout B8, Parmenides repeatedly refers to ‘what-is’
(to eon, B8.19, 32, 35, 37; cf. 25, 47–8), but nowhere speaks in
modal terms of ‘what-must-be’. The modal clauses in B2 do

188
Palmer (2009) 45–188. Palmer too continues (with Parmenides) to refer to ‘What Is’.
189
Palmer (2009) 83.

294
 5.5 Being and Intelligence in Alêtheia
indeed indicate that what-is has the modal property that it must be.
What is difficult to square with our text is the stronger claim that
this modal property means that the subject of the deductions in B8
is, not that which is (which has this modal property), but, more
restrictedly, what-is-and-must-be. On Palmer’s reading, it is in fact
false to state that what-is is ungenerated, indivisible, etc., unless
we mentally supply the qualification, which Parmenides himself
consistently and misleadingly fails to highlight: ‘what-is-and-
must-be’. If the modal fact, that what-is must be, plays this crucial
logical role, then the complete disappearance of the modal clauses
from Parmenides’ extended deductions of the attributes of ‘what-
is’ in B8 becomes mystifying. Parmenides’ persistent, unqualified
talk of ‘what-is’ suggests rather that this modal property of what-
is, that it must be, does not license us to speak of the subject of the
deductions in B8 in terms of the restricting conjunction ‘what-is-
and-must-be’. The attributes belong, not restrictedly to what-is-
and-must-be, nor restrictedly to what-is-and-is-unchanging, but to
what-is, which must be, and is unchanging, and is ungenerated,
etc. If this is right, however, then the modal fact, that what-is must
be, has no bearing on whether the existence of what-is is logically
compatible with the existence of Doxastic things. This brings us
back to square one of the ontological question.
At this point, we should take a step back and reassess the
adequacy of starting from or focusing on the category of ‘exis-
tence’ in the first place. Parmenides himself speaks in terms of
‘being’ (einai, to eon) and ‘reality’ (alêtheiê, alêthês). Interpreters
of Parmenides once attempted to establish stark demarcations
between existential, predicative and veridical uses of the verb ‘to
be’ and to determine which sense was operative in Parmenides’
arguments. As a growing scholarly consensus recognises, how-
ever, we cannot plausibly impose such stark demarcations on our
text and all three senses are involved (often inextricably) in
Parmenides’ use of the verb.190 There is, furthermore, a syntactic

190
See esp. Brown (1986) 54–7, 69 and passim; (1994) 216–20; cf. Kahn (2002) 85–90
(and, already, Kahn (1969) 712; (1973) 401–2); de Rijk (1983) 50; Long (1996) 144;
Sedley (1999a) 115; Miller (2006) 42, n.63, 44–5; Robbiano (2006) 80. To take one
example, the claim that what-is-not cannot be indicated or thought about suggests a
confluence of senses of the verb ‘to be’. We can indicate and think about something

295
 How Could Parmenides Have Written Alêtheia?
and semantic divergence between the English verb ‘exist’ and the
Greek ‘is’ (esti). Whereas the former is always complete, the latter
can at least in principle receive a complement (is such-and-such).191
Corresponding to this divergence is a logical and conceptual one.
Whereas existence is ‘all-or-nothing’ and does not admit of degrees,
the notion of being (such and such) allows, at least in principle, of
being more or less, unqualifiedly or qualifiedly, properly or some-
how catachrestically.192
Parmenides employs the terms alêtheiê and alêthês to indicate
reality and what is real or genuine. What-is, as the subject matter
of Alêtheia (recall especially B1.29; B8.50–1), is thus aligned with
reality.193 By contrast, then, Doxastic things (the subject matter of
Doxa) fall outside and fall short of the rubric of reality and what is
real. One could leap from this point to non-existentialism and say
that, in dissociating Doxastic things from reality, Alêtheia cate-
gorically identifies what-is as the only item in the world. We saw,
however, that insurmountable objections face this position.
Alêtheia may rather suggest the more inclusive and philosophi-
cally viable view, not that only what-is exists, but that it is only of
what-is that we can use the term ‘being’ in a strict and uncompro-
mising application of the verb. Only what-is really, unqualifiedly
and properly speaking is (and is, therefore, ungenerated, imper-
ishable, etc.). On this line of thought, what-is is, properly speak-
ing, the only truly real entity. Since Doxastic things cannot be said
to involve only ‘is’ to the complete exclusion of ‘is not’ (nor, of
course, only ‘is not’ to the exclusion of ‘is’), one could only say

which does not exist, or is false or is not something-or-other. Parmenides seems to have
in mind rather what is quite simply nothing (cf. μηδέν, B6.2). One could still con-
ceivably privilege the predicative sense and construe what-is-not – or ‘nothing’ – as
what has no properties, i.e. what is not anything at all. But this construal has immediate
existential and veridical implications, which Parmenides himself does not demarcate as
such from the predicative issue: such a thing could not exist, and nothing could be true
of it.
191
In a classic paper, Brown (1994) argues that the absence of a sharp distinction between
complete (existential, non-predicative) and incomplete (predicative, non-existential)
senses of esti is a quite general feature of ancient Greek, which is duly reflected in the
philosophies of Parmenides, Plato and Aristotle; see also Brown (1986). Leigh (2008)
questions the universality of this thesis in relation to the distinctive case of some of the
arguments in Plato’s Sophist.
192
On this conceptual issue, see further Vlastos (1973) 65 and passim; Kahn (2002) 88.
193
See further the introduction to the chapters on Parmenides.

296
 5.5 Being and Intelligence in Alêtheia
that they are by using the verb in a qualified, imperfect and
imprecise or catachrestic way. It is for this reason that Doxastic
things cannot be considered real or genuine entities. But, however
qualifiedly, imperfectly, mutably and contingently, they still are
(what they are) and, therefore, they are still there to be experi-
enced, spoken of and understood. They are not, then, implied to be
non-existent.194
An urgent requirement for this interpretation is to explicate
clearly what this statement amounts to, that only what-is really
is, whereas Doxastic things fall short of being real entities.
Parmenides’ deductions of the predicates of what-is do give us
much to work with here. What-is did not come to be and will not
perish or change, it is homogeneous and indivisible and it does not
locomote. These predicates all reflect different respects in which
what-is involves strictly ‘is’ and in no way ‘is not’: there is no time
in which it was not or will not be (what it is), and there are no
places or areas in which it is not, or in which it is to some lesser
degree. By contrast, a Doxastic thing like a human or a pebble is
(what it is) now but not earlier or later, and here but not there
(B8.36b-41). Furthermore, what-is possesses all of its properties in
a perfect and absolute way (it is entirely and superlatively homo-
geneous, etc.). Indeed, what-is is only ascribed properties which
can be possessed in this way. The characterisation of Light and
Night is pertinently different. For example, Night is heavy and
dense while Light, for its part, also possesses some measure – a
smaller measure – of weight and density: it is rare and light-weight
(B8.56–9). Quite generally, the characterisation of Light and Night
in relation to one another as enantiomorphic opposites involves
them essentially with ‘is not’ as well as ‘is’: Light is hot, light-
weight and rarefied and is not cold, dark, heavy and dense, while

194
The approach sketched here and elaborated below has affinities with that for which
Johansen (2016) argues from another direction. Johansen suggests that the partial
extent to which the cosmos meets Parmenides’ criteria of intelligibility (i.e. the
‘signs’ (B8.2) set out in Alêtheia) is also the extent to which it is. The cosmos thus
enjoys a degree of being, corresponding to the extent to which it satisfies these criteria
of intelligibility and, therefore, to which it is like the being (what-is) which satisfies
them entirely. At some junctures, Palmer (2009) also indicates that Parmenides associ-
ates what-is with reality as a mode of being of which contingent things fall short; thus,
‘only What Is only or really is’ (165, 179).

297
 How Could Parmenides Have Written Alêtheia?
the opposite is the case for Night (see further Chapter 4.3). But
Night is not perfectly or absolutely heavy. Indeed, it is unclear that
anything could be. Conversely, what-is possesses all of its attri-
butes (homogeneity, indivisibility, etc.) in a perfect and absolute
manner.
In sum, Doxastic things are involved with ‘is not’ as well as
with ‘is’ (εἶναί τε καὶ οὐχί, B8.40) in a number of ways. They are
now but not later or earlier, and they are here but not there. They
possess properties mutably and not perfectly or absolutely (they
admit of being (such-and-such) more or less). They comprise two
opposite elements each of which essentially is what the other is
not. It is for these reasons and in these respects, then, that Doxastic
things fall short of being real entities. It is for these reasons that the
verb ‘to be’ could not be used of them in the strict and uncompro-
mising manner which is on display in Alêtheia and in which it is
used of what-is. Rather, such things are but also are not, and, as
such, they could only be said to be in what would amount to a
compromising and imprecise use of the term. Correspondingly, it
is the nature of the properties of what-is, and the manner in which
it possesses those properties, which render it the only truly real
entity and the only thing which could truly and strictly speaking be
said to be, since it alone does not involve ‘is not’ in any way.
‘Reality’, for Parmenides, comprises that which instantiates a
mode of being which excludes ‘is not’ categorically.
It is worth noting that the poem indeed confronts us with both
technical uses of the verb ‘to be’, which restrict it to what-is, and
looser uses, which deploy the verb more broadly also of Doxastic
things and processes. For example, the goddess’s introduction of
the way ‘that [it] is’ (ἔστιν, B2.3) instantiates the technical use (cf.
e.g. B8.15–16). But contrast her statement that humans are in a
state of having erred (εἰσίν, B8.54) or that, within the human
constitution, the full is thought (ἐστί, B16.4).195 Such statements
reflect the broader and looser use of the verb, on which humans can
be said to be (now, here, in a state of having erred, etc.), even
though their involvement with ‘is not’ means that the verb could
195
For the broader usage, cf. also B1.11, 27; B6.9; B9.3; B16.3; B19. De Rijk (1983) 49
draws a distinction between technical and non-technical uses of εἶναι in Parmenides and
attempts to classify every occurrence of the verb accordingly.

298
 5.5 Being and Intelligence in Alêtheia
not be used of them in the strict and uncompromising way in which
it is used of what-is.
The goddess concludes the proem with the difficult assertion that
the mortal shall learn ‘how for the things that seem and are accepted
(ta dokounta) it was [or: would have been] right to be acceptably
(dokimôs), all of them traversing across everything’ (B1.31–2). In
our analysis of these lines at the end of Chapter 4.4, we identified
two possible interpretations. On the first (‘absolute’) interpretation,
the phrase ‘to be acceptably’ or ‘properly’ signals true, genuine
being, alluding to the uncompromising attitude towards the verb ‘to
be’ in Alêtheia and to the status of what-is as a real entity. On
the second (‘intrasystematic’) interpretation, ‘to be acceptably
(dokimôs)’ indicates, not the pure mode of being instantiated by
what-is, but a distinct and inferior mode of being which is typical of
Doxastic things and elements (ta dokounta) and which is reflected in
part in their mutability and constant motions (‘all of them traversing
across everything’). In fact, on either interpretation Parmenides
could be taken to convey here the notion that Doxastic things
possess their own mode of being which falls short of the mode of
being instantiated by what-is. On the intrasystematic interpretation,
Parmenides would be making this point directly and explicitly. On
the absolute interpretation, the idea would be more implicit. Still,
the insistence that only what-is properly or genuinely is can be seen
to create room for the possession by Doxastic things of something
else, which falls short of such genuine being (dokimôs einai). On
either reading, the creation of ontological room for Doxastic things
would indeed answer an urgent need for Parmenides, insofar as
Doxastic things could not be reduced simply to what-is-not. On the
other hand, such ontological space notwithstanding, it is still what-is
alone which is identified with what Parmenides calls ‘reality’,
insofar as it alone strictly is, in line with the uncompromising
application of the verb ‘to be’ in Alêtheia, which excludes any
involvement with ‘is not’. It would be question-begging to
reason that Doxastic things cannot have a qualified and imperfect
kind of being because Alêtheia rejects the notion of qualified and
imperfect being with regard to what-is.196 That Doxastic things can

196
See e.g. Long (1975) 95; Robinson (1975) 631.

299
 How Could Parmenides Have Written Alêtheia?
and what-is cannot be so qualified could be a central facet of their
dissimilarity.
The goddess’s remarks on mortal naming in B8.36b-41 offer a
helpful test-case for the evaluation of the ontological question (for
the text and translation, see Chapter 4.4). One might conclude that
only what-is exists from the statement that ‘nothing else either is or
will be except what-is’ and the following inference that mortal
names, like ‘coming-to-be-and-perishing’ and ‘both-being-and-
not-being’, in fact have what-is as their true referent. But this read-
ing would starkly exemplify the internal, textual and philosophical
incoherence of non-existentialism. If only what-is exists, then mor-
tals do not exist and, a fortiori, cannot name anything. Alternatively,
however, this progression could underscore that only what is strictly
or really is, in line with the uncompromising or technical applica-
tion of the verb ‘to be’ in Alêtheia. The goddess emphasises in this
connection that the mortals take their names to be true reflections of
reality (πεποιθότες εἶναι ἀληθῆ, B8.39). Mortal naming purports to
capture reality truly (Chapter 4.4). The whole and unchanging what-
is is the only genuinely real entity and, as such, coincides with
reality. On a de re rather than de dicto level of analysis, therefore,
what-is is in fact the only possible referent for mortal names.
When the goddess reprises the krisis between ‘is’ and ‘is not’ in
B8, she specifies that this is the appropriate discrimination ‘concern-
ing these matters’ (περὶ τούτων, B8.15). When she completes her
deductions of the properties of what-is, she notes that she has now
concluded her account ‘about reality’ (ἀμφὶς ἀληθείης, B8.51).
Parmenides thinks of Alêtheia as one, privileged domain of inquiry,
which treats of a certain, privileged subject matter – ‘what-is’ or
‘reality’. In his cosmology, Parmenides recognises a different domain
of inquiry, which treats of a different subject matter – Doxastic things,
which are but also are not. It is the most plausible response to the
ontological question, I suggest, that Parmenides does – as he must –
allow room in his ontology also for these mutable and heterogeneous
items, even though their involvement with ‘is not’ means that they,
unlike what-is, are denied the status of ‘real’ entities.
What, finally, of the goddess’s description of what-is as ‘the
unshaken heart of well-rounded reality’ (B1.29)? There is little
prospect for interpreting the words ‘of reality’ as a possessive
300
 5.5 Being and Intelligence in Alêtheia
genitive. Well-rounded reality is itself, for Parmenides, identical
with the indivisible what-is (recall B8.50–1), which does not have
a heart as opposed to other parts. Far more likely, this is the so-
called genitive of material or contents, as in the Homeric expres-
sion ‘barrier of teeth’ (ἕρκος ὀδόντων, Il. 4.350), where the barrier
comprises the teeth.197 The unshaken heart itself comprises the
well-rounded reality. Does this characterisation of reality as a
steady heart help towards our evaluation of the ontological ques-
tion? In a different context, this expression could have naturally
represented what-is as a cosmological agent, which animates and
acts on all other things. Nothing else in the fragments, however,
remotely suggests such a conception of what-is. Indeed, we saw
that Parmenides pointedly resists treating what-is as an Ionian-
style cosmological principle. Most likely, the characterisation of
what-is as a stable heart primarily serves to reflect its own vitality
and (perhaps) intelligence (see Chapter 5.5.2). Nonetheless,
although what-is is not a causally efficient cosmological agent,
its characterisation as an unshaken heart (B1.29) still aptly reflects
its status as the core and ultimate reality in the world, which abides
stably even as it is concealed from human understanding by the
many and heterogeneous things which constantly traverse this
same world in every direction (B1.31–2).
Far from being a novel interpretation, the basic thrust of my
proposals can be found among some of Parmenides’ ancient read-
ers, who discuss his notion of what-is in terms of the Platonic
concept of ‘what really (or: truly) is’ (to ontôs on). So, for example,
Plutarch rejects Colotes’ contention that Parmenides simply did
away with such things as fire, water, rocks and, indeed, all the cities
of Europe and Asia. Rather, Plutarch explains, Parmenides reserved
the term ‘being’ to what really is since it alone abides in being.
Correspondingly, he judged that the other things, which now are and
now are not, require a designation other than ‘being’.198 Plutarch is
admittedly construing Parmenides here in terms of his own, later
conceptual framework, although he scarcely differs in this respect
197
See further Smyth (1972) noo.1323–4.
198
Plut. Mor. 1114d-e (esp. e6–10; d5–7) = Coxon (2009) Testim. 113; similarly, Simp. in
Cael. 7.557.20–558.17 = Coxon (2009) Testim. 203; cf. also Philop. in Ph. 21.30–
22.15 = Coxon (2009) Testim. 193.

301
 How Could Parmenides Have Written Alêtheia?
from Parmenides’ modern interpreters. At any rate, we would be
rash to dismiss Plutarch’s remarks as uninsightful on those grounds.
In a passage famous for its Parmenidean echoes, Plato will
expressly represent the objects of belief as things which both
are and are not and, therefore, as ontological intermediates: ‘in
between . . . what purely is and what altogether is not’ (μεταξύ . . .
τοῦ εἰλικρινῶς ὄντος τε καὶ τοῦ πάντως μὴ ὄντος, Rep. 5.478d5–7).
There is no question of reading Plato back into Parmenides. Rather,
by observing what was conceptually possible for Plato, we can
lend further, extrinsic support to what I suggested above (without
appealing to this echo) is on balance the most viable approach to
the earlier and more obscure text of Parmenides. The later concepts
of ‘what purely is’ (to eilikrinôs on) and ‘what really is’ (to ontôs
on), I suggest, reflect an ontological attitude which is already at
work more embryonically in Parmenides himself. Indeed,
Parmenides too modifies the verb ‘to be’ with an adverb – ‘to be
acceptably’ (dokimôs einai, B1.32) – either in reference to the
genuine being of what-is or to a distinct mode of being possessed
by Doxastic things.
Nowadays, one runs the risk of being diagnosed with
‘Platonitis’ by offering an interpretation of Parmenides which is
deemed excessively close to later Platonic attitudes.199
Nonetheless, I conclude that there is a significant affinity between
Parmenides and the later tradition of Platonic philosophy, over
which he has clearly exerted a profound influence. There is much
interpretative insight and potential in the basic interpretation of the
ontological question which Plutarch adopts in his response to
Colotes.
This is not to claim, however, that Parmenides articulated or
systematised his attitude to the ontological question in terms
which were any more explicit or direct than the evidence which
we considered in this section. Parmenides stands at the very
beginning of a long philosophical tradition, which will later
issue in clearer and systematic classificatory demarcations
between what purely or really is, what altogether is not, and the

199
Cordero (2011b) 100. My interpretation, though, does not involve thinking of Doxa as a
realm of mere appearances, which is the main symptom Cordero has in mind.

302
 5.5 Being and Intelligence in Alêtheia
ontological intermediates, the objects of perceptual experience,
which both are and are not. With the benefit of historical hindsight,
these clear and worked-out articulations are familiar to us in a way
that they could not have been to Parmenides himself.200 As I noted
at the outset, I put forward in this section what I take to be the
philosophical and interpretative elaboration which is, on balance,
most faithful to the obscure text of Parmenides and to the direction
in which his thought tends. Parmenides himself does not recognise
a special need to explicate how those things (stars, mortals, etc.),
whose properties diverge from the properties of what-is, are mani-
festly available for experience, discussion, belief and explanation.
If nothing else, however, his poem points towards at least one
viable ontological avenue for meeting this need coherently.
Simplicius offers a sobering insight: ‘when he said that what-is
is one, Parmenides was perfectly aware that he himself was born
and had two feet’.201 Much like Plutarch, Simplicius is arguing
that Parmenides did not reject generation and perishing categori-
cally, but with respect to ‘what really is’ (ἐπ’ ἐκείνου, 7.559.14–
27). I suggested in this section that there is much to be said for this
conception of Parmenides’ ontology (nor should we reject this
conception simply on the grounds that Plutarch and Simplicius
themselves may well have been attracted to it by the impulse to see
Parmenides in a Platonic light). Nonetheless, the textual, episte-
mological and theological interpretations of Parmenides which
were developed over the last two chapters do not require this
conception. Insofar as those interpretations are concerned, we
need only agree with Simplicius that Parmenides clearly recog-
nises that, in some sense, Doxastic things are there to be experi-
enced by mortals and to occasion views in them. He clearly
recognises that, in some sense, those mortals, and he himself, are
there to form correct and incorrect views about multiple and
heterogeneous things (such as they themselves and the number
of their feet) and about reality.

200
Cosgrove (2014) 10 remarks similarly on Parmenides’ relation to the later Platonic
tradition, although he does not ascribe to Parmenides the attitude to the ontological
question which I suggest here.
201
In Cael. 7.559.27–560.1 = Coxon (2009) Testim. 203.

303
 How Could Parmenides Have Written Alêtheia?
5.5.2 Being, Intelligence and Intelligibility in Alêtheia
A number of notoriously difficult verses in Alêtheia address the
relation of cognition or understanding (noein) to what-is (B3;
B6.1–2a; B8.34–6a; cf. B8.7–9).202 The syntax of these verses is
ineradicably ambiguous, and their translation is highly uncertain.
As such, they are best considered in the light of already established
readings, rather than adopted as interpretative starting points.203
B3 reads:
τὸ γὰρ αὐτὸ νοεῖν ἐστίν τε καὶ εἶναι.

This fragment has been analysed both as B3(i), ‘for it is the same
thing to understand and to be’, and as B3(ii), ‘for the same thing is
(there) for understanding and for being’.204 B3(i) asserts the iden-
tity of what-is and understanding or cognition. B3(ii) indicates
more weakly that what-is is somehow especially an object of
understanding. This dilemma could not be reasonably determined
on the basis of the words of B3 alone.
B6.1–2a reads:
χρὴ τὸ λέγειν τε νοεῖν τ’ ἐὸν ἔμμεναι· ἔστι γὰρ εἶναι,
μηδὲν δ’ οὐκ ἔστιν.
Syntactically, perhaps the most sober and straightforward transla-
tion is: ‘It is necessary205 to say and understand that what-is is; for
it is to be, but nothing is not.’206 It is syntactically conceivable,
however, to take ἐόν as a predicate and ἔμμεναι as the copula, and
translate: ‘It is necessary that speaking and cognising be real-and-
true (ἐὸν ἔμμεναι).’ On this construal, B6.1–2a could also, like B3(i),
identify being and understanding (and speaking).207
202
For the point that noein and its cognates indicate, not merely thought without a belief
component, but, minimally, an attempt to cognise or apprehend a situation, object or
person, see Ch. 3.4, Ch. 4.3. These particular verses invite the translation ‘under-
standing’ inasmuch as they use noein as a success term in relation to the successful
cognition of what-is (as discussed below).
203
Cf. Mourelatos (2008a) xv.
204
B3(i): Long (1996) 132–8; Sedley (1999a) 120; Robbiano (2006) 57–8. B3(ii): Coxon
(2003); Palmer (2009) 118–22; Wedin (2014) 202–29.
205
χρή, signifying ‘normative necessity’.
206
Alternatively, ‘ . . . nothing [it] is not’, Palmer (2009) 113–14.
207
See Kahn (1988) 260–1 (cf. Kahn (2002) 86, n.13). Kahn lists several possible
construals of B6.1, and convincingly argues that the translation ‘what can be spoken
and thought must be’ is not remotely plausible.

304
 5.5 Being and Intelligence in Alêtheia
B8.34 reads:
ταὐτὸν δ’ ἐστὶ νοεῖν τε καὶ οὕνεκεν ἔστι νόημα.

We may distinguish three possible construals:

B8.34(i): understanding and the understanding that [it] is are
one and the same
B8.34(ii)a: understanding and that because of which there is
understanding [sc. what-is] are the same thing
B8.34(ii)b: the same thing is both for understanding and that
because of which there is understanding.
One key question here is how to interpret οὕνεκεν. If we translate
οὕνεκεν as ‘that’208, then we get B8.34(i).209 If, however, οὕνεκεν
means ‘because of which’,210 then we have two options. We could
read B8.34(ii)a, which identifies understanding and being.211 But
the translation ‘because of which’ does not necessarily imply that
identification. Alternatively, we can construe νοεῖν as a datival
infinitive (a construal which is also plausible for νοεῖν in B3 – as
in B3(ii) – and for νοῆσαι in B2.2), meaning ‘for understanding’.
This would yield B8.34(ii)b.212
The immediately following words (B8.35–6a) are no less diffi-
cult and uncertain:
οὐ γὰρ ἄνευ τοῦ ἐόντος, ἐν ᾧ πεφατισμένον ἐστιν,
εὑρήσεις τὸ νοεῖν.
We can again distinguish alternative possible translations:

208
Cf. e.g. Od. 5.216. 209 Cornford (1939) 34; Robinson (1975) 630.
210
E.g. Il. 9.505; cf. B8.13, 32.
211
Phillips (1955) 552–3; Long (1996) 136; Sedley (1999a) 120; Crystal (2002) 217.
212
Palmer (2009) 164; cf. KRS (1983) 252; Wedin (2014) 211. The goddess states in B2.2
that the ways ‘that [it] is’ and ‘that [it] is not’ are the only two ways of inquiry available
for understanding (νοῆσαι). Taking this statement to mean that these are the only
conceivable sorts of inquiry (e.g. by rendering νοῆσαι more weakly as ‘for thinking’
or ‘to be thought of’) would introduce a glaring inconsistency with the later emergence
of the cosmology. However else one interprets its status, the cosmology is certainly not
inconceivable, nor is it identical with either of the two ways introduced in B2.2. Palmer
(2009) 63–73 plausibly argues that νοῆσαι is used here in a strong sense, and that it
indicates the sort of understanding which facilitates and is achieved in a comprehension
of reality, i.e. (once the second way is eliminated in lines 6–8) of what-is. Alternatively,
Wedin (2014) 65, n.95 suggests that the later emergence of the cosmology does not
contravene the statement made in B2.2, because the cosmology proceeds by the
combination of the two ways of inquiry first identified there.

305
 How Could Parmenides Have Written Alêtheia?
B8.35–6a(i): for not without what-is, depending upon which [sc.
what-is], it [sc. understanding] has been expressed,
will you find understanding
a(ii)
B8.35–6 : for not without what-is, in what has been expressed
[sc. the preceding arguments], will you find
understanding
B8.35–6a(iii): for not without what-is, in that in which it [sc.
understanding] has been expressed, will you find
understanding.
What is the subject of the verb ‘has been expressed’ (πεφατισμένον
ἐστιν) and what is the antecedent of ‘in which’ or ‘upon which’
(ἐν ᾧ)? An obvious candidate for the subject of ‘has been expressed’
is ‘understanding’ (τὸ νοεῖν) at the end of the sentence. Taken by
itself, moreover, the word-order most naturally suggests that what-
is is the antecedent of ‘in which’ (τοῦ ἐόντος, ἐν ᾧ). But it is difficult
to make sense of the notion that understanding has been expressed
in what-is. One possibility, then, is to interpret the clause as signify-
ing a relation of dependence rather than location: ‘depending upon
which’. This gives us B8.35–6a(i).213 Alternatively, we may ques-
tion whether what-is must, after all, be the antecedent, and take the
‘in which’ clause to be making an implicit reference to the preced-
ing verses and arguments as its antecedent. In this case, the subject
of ‘has been expressed’ could be impersonal: literally, ‘in that in
which it has been expressed’. This gives us B8.35–6a(ii).214 Or else,
we can combine this interpretation of the antecedent with the
assumption that ‘understanding’ is the subject of the clause. This
yields B8.35–6a(iii).215
Where do these considerations leave us? No syntactical con-
strual of these verses could be remotely compelling or definitive.
On some syntactically viable (although hardly obvious) readings,
they assert the identity of what-is and understanding or cognition
(B3(i); B8.34(ii)a). They do minimally, however, express some sort
of close connection between what-is and understanding.
Minimally, these statements seem to converge on the notions
that what-is (and, perhaps, the proposition ‘[it] is’) is somehow
213
Palmer (2009) 164, n.39 (offering parallels for this sense of ἐν with the dative).
214
Sedley (1999a) 120. 215 Robbiano (2006) 169–70.

306
 5.5 Being and Intelligence in Alêtheia
especially an object of understanding (B3(ii)) and, perhaps, some-
how a necessary condition for understanding (B8.35-6a(i)).
Understanding is either occasioned by what-is (B8.34(ii)a-b), or
else it means cognition of what-is (B8.34(i)).
We must, however, qualify and nuance even this minimal
impression by recognising a textual fact. However we read these
verses, it is simply not the case that Parmenides invariably uses his
cognate terminology for mind and understanding (noos, noein) as
success words, where success means cognition or knowledge of
what-is. In Alêtheia, the goddess explains that the mortals fail to
cognise what-is, and that helplessness guides their wandering
mind (πλάζονται . . . πλαγκτὸν νόον, B6.5–6). Again, her analysis
of the human mind in Doxa allows it wandering cognition, but not
cognition of what-is (πολυπλάγκτων . . . νόος . . . φρονέει . . .
νόημα, B16). One’s mind (noos) can fail to grasp what-is and, as
we have seen throughout Chapter 4, can grasp and even know
(B10) other things. Mourelatos consequently suggests that we
should construe the affinity which is expressed in the verses
which we considered in this subsection as an ideal rather than a
fact, styling it a ‘normative necessity’.216 Put differently, we must
take it that these statements are being made from within the frame-
work of Alêtheia’s account of reality. In these verses, ‘understand-
ing’ is used almost technically as a clear success term, where
success means cognition and understanding of reality as deli-
neated in Alêtheia. The affinity expressed in Alêtheia between
what-is and understanding refers to that cognition and understand-
ing (noein) which is based on the krisis between ‘is’ and ‘is not’,
and which comprises the selection of the former and rejection of
the latter. It refers to that purely reliable and uniquely steadfast
apprehension of what-is, the unshaken heart of reality.
The relation between being and intelligence in Alêtheia also
raises interesting questions concerning the nature of what-is. If,
within Alêtheia’s account of reality, being is identified with cogni-
tion, then what-is must be intelligent as well as intelligible.217 That
what-is is intelligent and self-knowing would of course be

216
Mourelatos (2008a) 175–6; cf. Curd (1998) 49.
217
Long (1996); cf. Trépanier (2010) 292–5.

307
 How Could Parmenides Have Written Alêtheia?
consistent also with the more minimal readings of these verses.
Indeed, what-is is the stable heart of reality (B1.29). This term
(êtor) is never used, in any period of ancient Greek, except of the
life, emotion or thought of a human or divine person. The goddess,
it seems, promises to teach Parmenides the stable heart of living
reality.218
The possible intelligence and apparent vitality of what-is raise
the question of whether, and, if so, in what way, we should think of
what-is as divine. We will return to this question when discussing
the similarities between Xenophanes’ supreme god and
Parmenides’ what-is (Chapter 6.1). Here I will only highlight an
important qualification. Noticeably, and even pointedly,
Parmenides withholds traditional divine epithets from what-is.219
If there is a sense in which we may nonetheless think of what-is as
divine, it must be a highly refined, nuanced, pointedly non-tradi-
tional and arguably specifically Xenophanean sense. The divinity
of what-is (if that is a legitimate way of speaking) is certainly
not of the same type as, or on a par with, the more traditional
divinity of the goddess, who initiates the kouros and issues her
revelation to him (thea, B1.22). It is this divinity of the goddess
with which the mortality of the kouros is contrasted, from the
proem onwards, in a complex dynamic which lies at the heart of
the epistemology of Parmenides’ poem.

218
Long (1996) 142–3 emphasises this semantic point in support of this conclusion; cf.
also Coxon (2009) 282–3.
219
As noted by Mourelatos (2008a) 44; cf. Cornford (1939) 29, 43; Granger (2008) 5;
(2010) 35.

308
 6

RETROSPECT AND PROSPECT

In this final chapter, we will take some opportunities to look both

backwards and forwards. We will first (Chapter 6.1) consider how
Xenophanes relates himself to Hesiod and how Parmenides relates
himself both to Xenophanes and to Hesiod. We will ask how the
interrelations between these thinkers might further illuminate and,
in turn, be illuminated by our self-standing discussions of them in
the previous chapters. Next (Chapter 6.2), we will consider how
Empedocles fits into the broad tradition of theology and epistemol-
ogy that has emerged from the foregoing studies of Hesiod,
Xenophanes and Parmenides. Empedocles will thus provide one
(important) example of how the interpretative approaches devel-
oped above might be deployed also in relation to other and later
developments in theology and epistemology. Finally (Chapter 6.3),
we will revisit – in the light of the conclusions defended in this
book – the questions of rationality and irrationality, and philosophy
and religion, which were our focus in Chapter 1.

6.1 Interrelations
We saw in the previous chapters how Hesiod, Xenophanes and
Parmenides inscribe and pursue questions of epistemology within
a theological framework. We explored the particular ways in
which their divergent views on the cognitive capacities and limita-
tions of mortals flow from their correspondingly divergent views
on (i) the nature of the divine, (ii) the nature of the mortal and (iii)
the nature of the relation and interactions between them.
With their dramatic opening address, ‘we know how to speak
many falsehoods which are like verities (ψεύδεα . . . ἐτύμοισιν
ὁμοῖα), and we know, whenever we wish, how to utter truths’

309
 Retrospect and Prospect
(ἀληθέα, Th. 27–8), Hesiod’s Muses articulate a framework in
which the problem of epistemology becomes the problem of under-
standing the interactions between mortal and divine. The Theogony
(along with much of the Works and Days) prescribes a wary and
circumspect stance within this framework. The goddesses leave it
uncertain, and the mortal poet is helpless to determine, whether the
verses they inspire comprise truths, verisimilitudinous falsehoods or
a mixture of the two. The framework that the Muses articulate is,
however, sufficiently indeterminate to allow competing elabora-
tions of its fixed principles. At one juncture in the Works and
Days (646–62), we encounter a competing and more optimistic,
albeit fleeting, stance in response to their address.
We find in Hesiod, then, the following principles: (i) The problem
of understanding the epistemic capacities and limitations of mortals
is the problem of understanding correctly what mortals and gods are
like, how they relate to one another and how they interact; (ii) it is
only through a special and privileged interaction with the divine
that the mortal poet can produce potentially true (since divinely
disclosed) accounts of matters that lie beyond the grasp of human
cognition; (iii) the mortal cannot himself ascertain the truth-value of
those accounts. Principle (i) frames Xenophanes’ and Parmenides’
own otherwise radically divergent approaches to epistemology.
In this respect, Hesiod not only isolates for consideration the epis-
temological question of the conditions of human speculation: he
crystallises the major contours of an emerging enterprise of theolo-
gical epistemology. Xenophanes can be said to deny (ii) and
develop his own, very different version of (iii). Mortals form for-
ever corrigible and conjectural beliefs and judgements about states
of affairs external to their experience on the basis of a necessarily
limited range of experiences, which (or, minimally, some of which)
the divine purposively enables them to consider. No mortal can
himself know or ascertain the truth-value of his corrigible accounts
(cf. iii). The scope of divine disclosure most probably encompasses
all that any mortal experiences (‘universal disclosure’). At any rate,
Xenophanes stakes no claims for a special and privileged relation
with the divine (contra ii). Parmenides can be said to deny (iii), but
develop his own, very different version of (ii). The human mind
perforce thinks of, and in terms of, sensory contrasts between Light
310
 6.1 Interrelations
and Dark, Hot and Cold, etc. It perforce experiences and forms
beliefs about multiple, heterogeneous, mobile and differentiated
things and processes. But, through his encounter with the goddess
(cf. ii), the kouros attains the ability to think rather through the
competing categories of Alêtheia (is / is not), to master its argu-
mentation and to ascertain for himself the true homogeneous and
unchanging nature of ultimate reality (contra iii).
We can only tentatively trace deliberate and precise engage-
ments by Xenophanes with Hesiod, and by Parmenides with
Hesiod and Xenophanes. Yet, if nothing else, this exercise will
help us to delineate further the continuities and discontinuities
between the three thinkers, and will afford a new perspective on
familiar textual echoes.
It is often observed that Xenophanes’ programmatic remark in
DK21 B35 – ‘let these things be believed as like the truths’
(ἐοικότα τοῖς ἐτύμοισι) – recalls the Muses’ address in Theogony
27–8 (ἐτύμοισιν ὁμοῖα).1 I interpreted the phrase ‘like the truths’ as
a qualified endorsement. By stating that his views are like truths,
Xenophanes signals that they may be true, but he cautions that this
similarity could also be specious.2 Xenophanes thus encapsulates
in a single line both sides of Hesiod’s couplet – both the prospect
of truths (ἀληθέα) and that of verisimilitudinous falsehoods
(ψεύδεα . . . ἐτύμοισιν ὁμοῖα). Xenophanes, we saw, advances his
own alternative notion of divine disclosure against the traditional
notions that he rejects. This conclusion places this textual echo in
a new light. Both Hesiod and Xenophanes faced the same question
which, we saw (Chapter 3.1), also exercised Greek reflections on
divination: why are statements formed on the basis of divine
disclosure sometimes false? Hesiod appeals to the possibility of
divine deception and the unforeseeable arbitrariness of divine
caprice. By positioning himself in DK21 B35 against the Muses’
address to Hesiod in Theogony 27–8, Xenophanes underscores
1
See Bryan (2012) 12–16; cf. Guthrie (1962) 396, n.2; Heitsch (1966) 232–3; Hussey
(1990): 35; Koning (2010) 205–6; Iribarren (forthcoming). Xenophanes names Hesiod
when criticising his theology: DK21 B11; cf. related critiques: B12, B1.21–3; B14–16;
B26; also D.L. 9.18.8–10 = A1; D.L. 2.46.8–11 = A19. In Rep. 2.382d2–3, Plato seems to
allude to Th. 27–8 while criticising (in a manner reminiscent of Xenophanes) Hesiodic
theology among other things; see Ch. 2.1, n.42.
2
Following Bryan (2012) 6–57; see Ch. 3.5.

311
 Retrospect and Prospect
that, in keeping with his theological objections to the makeup of
Hesiod’s gods (DK21 B11–12, B14, etc.), his own second-order
reflections do not involve the gods in deception or caprice. Rather,
we can explain the corrigibility of belief and the possibility of
error once we realise that divine revelation in fact consists in the
disclosure of a necessarily limited repertoire of experiences to
spatially and temporally limited mortals.
Again, scholars have long observed that Parmenides’ goddess
recalls the Muses’ address when she issues the programmatic
statement which was our focus in Chapter 4: ‘And it is right for
you to learn all things, both the unshaken heart of well-rounded
reality (ἀληθείης . . . ἦτορ), and the beliefs of mortals, in which
there is no real trust’ (βροτῶν δόξας, ταῖς οὐκ ἔνι πίστις ἀληθής,
DK28 B1.28b-30).3 Like Hesiod, and unlike Homer who remains
more in the background, Parmenides makes central the figure of
the mortal agent who is identified with the poetic voice.4 Both
poets narrate in extensive detail a first-person encounter with an
all-female divine apparatus. It is from goddesses that the ensuing
poem is said to emanate and they avow that they issue (or, in
Hesiod’s case, that they can issue) two types of discourse: one
trustworthy and concerned with ‘truths’ (alêthea, Th. 28) or ‘true
reality’ (alêtheiê, DK28 B1.29),5 while the other is, in some sense,
not so. In both cases, the interaction between mortal and divine
presupposes and reflects the epistemic superiority of the latter.
And yet, the goddess is not ‘an exact counterpart’ of Hesiod’s
Muses.6 The Muses do not clarify the implications of their words
for their relationship with Hesiod, or whether the Theogony itself
will comprise truths, deceptive falsehoods or a mixture.
By contrast, Parmenides’ goddess expressly states that the kouros
will inquire both into ultimate reality and mortal beliefs devoid of
real trust. Indeed, she later signposts what parts of her speech fall
under which category (DK28 B8.50–2), and flags up that her own

3
Jaeger (1947) 94; Dolin (1962) 94ff; Mourelatos (2008a) 33, 219; Pellikaan-Engel
(1978) 6–7; Robbiano (2006) 41; Torgerson (2006) 35; Iribarren (forthcoming); cf.
Heitsch (1966) 201; KRS (1983) 262; Schofield (2007) 4.
4
Jaeger (1947) 98.
5
On Parmenides’ use of alêtheiê, see Introduction to the Chapters on Parmenides.
6
Pace Jaeger (1947) 94.

312
 6.1 Interrelations
account of mortal beliefs is ‘deceptive’ (ἀπατηλόν) insofar as it is
liable to be mistaken for an account which gets at and discloses the
nature of ultimate reality (what-is). Most importantly, the goddess
directs her mortal addressee to judge for himself her disclosure
through Alêtheia’s argumentation (DK28 B7.5–6), and so to iden-
tify and evaluate for himself the distinction between her accounts
of reality and her accounts of mortal beliefs.7 When it comes to her
relation to the kouros, Parmenides removes from his kind goddess
the spectre of a mocking and almost predatory deceptiveness that
marks female divine wisdom in Hesiod. The Muses open their
address with scornful mockery (‘shepherds of the field, base,
shameful things, mere bellies,’ Th. 26). The goddess is, by con-
trast, all gentle reassurance (‘and the goddess received me kindly,
and took my right hand in hers etc.’, DK28 B1.22–8).8
The divinisation of the kouros signals a related contrast.
The Muses descend to the mortal poet, while the goddess wel-
comes the kouros to her domain.9 If the Muses begin by rhetori-
cally reducing the shepherd to the level of the beasts that he tends
(‘mere bellies’), the goddess aligns the kouros, conversely, with
his divine guides and the divine milieu into which he is welcomed
as an appropriate member (DK28 B1.24–8a; see Chapter 5.3).
Hesiod is irremediably unable to discriminate truths from specious
falsehoods. This inability corresponds to his inexorably mortal
status. By contrast, the kouros, through his capacity for reasoning,
attains the ability to discriminate accounts of true reality from
accounts of mortal beliefs in which there is no real trust. This
attainment corresponds to his divinisation and his alignment with
his divine escort and with the goddess.
7
For this point, cf. Morgan (2000) 75; Torgerson (2006) 36; Koning (2010) 212; Benzi
(2016).
8
One might go even further. We find in Parmenides a nexus of associations between the
female, Light, vitality, better and purer thinking and the divine; in particular, within his
dualistic system, the element which is bright rather than dark and hot rather than cold also
appears to be the element which typifies the nature of the female rather than the male (see
Ch. 5.2, with nn.67, 69). Thus, insofar as Parmenides approaches the goddess through
that within him which is hot, aethereal and bright (his soul), he is also arguably
approaching her through that within him which is more female and less male.
We might say, then, that Parmenides not only rehabilitates female divine wisdom, but
also aligns the knowing mortal himself with those aspects of his constitution which are
brighter, hotter, more intelligent and more vital and – as such – more female.
9
Koning (2010) 212 observes the divergence in the direction of travel.

313
 Retrospect and Prospect
The framework that Hesiod’s Muses articulate can accommo-
date ambivalent vacillations on the part of the poet between more
or less optimistic stances (Chapter 2.4). By contrast, Xenophanes
and Parmenides both resolve the indeterminacy of the Muses’
address to Hesiod, and of the framework which they articulate,
into definite epistemological positions.
The tradition that Parmenides studied under Xenophanes is
highly suspect. The statements in Aristotle (Metaph. Α.5 986b
21–2) and Theophrastus (apud Diogenes Laertius, 9.21.1–3; cf.
Simplicius, in Ph. 9.22.26–31) very possibly derive from Plato’s
offhand and vague reference to ‘the Eleatic tribe (ἔθνος), which
begins with Xenophanes and even earlier’ (Soph. 242d4–6).10
Xenophanes does appear to have spent some of the latter part of
his life in Sicily (Diogenes Laertius, 9.18.5–6) and could concei-
vably have travelled to South Italy. No clearly dependable evi-
dence, however, links him to Elea. We could not rely on the
historicity of the tempting report that Xenophanes composed
a poem on Elea’s colonisation (Diogenes Laertius, 9.20.6–7).11
Nor could we press Aristotle’s anecdote, which has Xenophanes
issuing advice about religious practice to the people of Elea, since
it is only one (albeit much the earliest) among other versions found
in Plutarch, which place the story in Egypt (DK21 A13).12
Much stronger evidence that Parmenides encountered and
engaged with Xenophanes’ thought is found in some striking echoes
between Parmenides’ descriptions of what-is and Xenophanes’
descriptions of his greatest god. These echoes might themselves
have motivated the doxographic traditions that linked Xenophanes
and Parmenides.13 The immobility and invariability of the supreme
god and what-is are expressed in markedly similar language: ‘but he
always abides in the same place not changing at all’ (ἐν ταὐτῷ μίμνει
10
Plato ascribes to ‘the Eleatic tribe’ numerical monism. By comparison, consider Tht.
179e3–4 (also 152d–e; 160d5–8 (φῦλον)), where ‘Heraclitean’ flux is ‘Homeric and still
more ancient’. See further KRS (1983) 163–6; Coxon (2009) 18–19. Diogenes (9.21.
3–8) immediately proceeds to Sotion’s statement that Parmenides followed, not
Xenophanes, but the Pythagorean Ameinias.
11
See the scepticism in KRS (1983) 164–6.
12
The version in [ps.-?]Plutarch (Mor. 228e) has Lycurgus addressing the Thebans.
13
The echoes are observed in Coxon (2009) 18–19, 326–9; Long (1996) 143, 148; Palmer
(2009) 329–30; Bryan (2012) 97–100. Xenophanes was known to Heraclitus
(DK22 B40).

314
 6.1 Interrelations
κινούμενος οὐδέν, Xenophanes, DK21 B26); ‘unchanging . . . abid-
ing in the same place by itself it lies (ἀκίνητον . . . ταὐτόν τ’ ἐν
ταὐτῷ τε μένον καθ’ ἑαυτό τε κεῖται), and so it remains there
steadfast’ (Parmenides, DK28 B8.26–30). The ‘wholeness’ of
what-is recalls how the supreme god sees, thinks and hears as
a unified ‘whole’ (οὖλον, Parmenides, DK28 B8.4; οὖλος,
Xenophanes, DK21 B24). Xenophanes berates ‘mortals’ for ascrib-
ing coming-into-being to the gods (οἱ βροτοί . . . γεννᾶσθαι, DK21
B14.1; cf. A12). Parmenides berates them for ascribing it to what-is
(ὅσσα βροτοὶ κατέθεντο . . . γίγνεσθαι, DK28 B8.39; cf. B8.21).
We may also recall the apparent vitality of what-is (ἦτορ, DK28
B1.29) and the possibility (it is nothing more) that Parmenides
expressly accorded it intelligence and (self-)cognition.14
Parmenides, then, seems to appropriate some features of
Xenophanes’ supreme god in his novel account of reality in
Alêtheia. This appropriation must not, however, obscure the fact
that these two thinkers are theologically worlds apart. Parmenides
markedly avoids ascribing traditional divine epithets to what-is.15
He uses the language of divinity rather of the Daughters of the Sun,
who convey the mortal kouros, and of the goddess, who discloses
everything to him (θεά . . . ἀθανάτοισι, DK28 B1.22, 24; cf.
δαίμονος, B1.3). Again, it is Doxa’s goddess, not what-is, who
assumes the traditional governing and directive capacity of the
divine: ‘the goddess, who steers all things’ (DK28 B12.3; compare
Xenophanes’ god who ‘shakes all things by the thought of his
mind’, DK21 B25).16 It is this decidedly non-Xenophanean god-
dess whose divinity Parmenides contrasts with the mortals whose
thought and beliefs the kouros will transcend by encountering the
goddess and being himself divinised. If Parmenides considers
reality vital and might have expressly accorded it intelligence,
and if he appropriates some features of Xenophanes’ supreme
god in his account of ultimate reality, then – even though

14
Long (1996) 143, 148 emphasises these points. The extant fragments do not support
later reports that Xenophanes’ god is spherical (DK21 A1; A28.7), which may be further
assimilating his god to what-is (Parm. DK28 B8.43).
15
Ch. 5.5.2.
16
For this point, cf. Palmer (2009) 329. The goddess’s administration of metempsychosis,
copulation and birth was discussed in Ch. 5.2.

315
 Retrospect and Prospect
Parmenides himself never speaks in such terms – we may none-
theless think of what-is as divine in a nuanced and non-traditional,
Xenophanean sense. It is, however, a very different and, in many
ways, more traditional theological framework within which
Parmenides reflects about the relation between what he explicitly
calls ‘mortal’ and ‘divine’ and about the epistemological ramifica-
tions of this relation.
Xenophanes rejects the pervasive tendency in Greek religious
thought to see the boundary between mortal and divine as
a malleable and negotiable one, and to allow that some human
agents may approach or attain divine qualities or capacities.
Parmenides displays a distinctive version of this tendency.17
Again, Xenophanes, unlike Parmenides, admits no notion of
a privileged interaction with the divine, through which a mortal
can attain knowledge. Xenophanes’ perspective is made possible,
and remains constrained, by what the divine enables him to con-
sider: Xenophanes does not consider himself specially privileged
in these respects. This disillusioned attitude raises an interesting
point, which Xenophanes himself does not seem to address.
Xenophanes’ account of mortal epistemology draws on certain
views concerning what the gods are and are not like. But all such
views can never amount to more than corrigible beliefs (ἀμφὶ θεῶν,
DK21 B34). Xenophanes’ account of mortal epistemology, then,
itself depends on certain corrigible beliefs.18 Xenophanes identi-
fies the epistemic capacities and limitations of mortals from
within the mortal perspective. His account of mortal epistemology
is thus itself subject to those same limitations. Parmenides, by
contrast, describes the capacities and limitations of mortals from
an external, divine perspective. The kouros recognises that the
human mind operates perforce in accordance with sensory con-
trasts (Light/Night, etc.) only after he himself mastered the quali-
tatively different thought that underpins Alêtheia (is / is not) and

17
For Xenophanes, esp. Ch. 3.5; for Parmenides, esp. Ch. 5.3–4. Xenophanes attacks ideas
of metempsychosis (DK21 B7), while Parmenides advances one (Ch. 5.2).
18
This point is not peculiar to my interpretation of Xenophanes in Chapter 3. It will apply
to any interpretation which accepts that, for Xenophanes, what the gods are and are not
like bears on the epistemic predicament of mortals (n.b. esp. DK21 B18) and yet cannot
be ascertained by mortals conclusively (esp. DK21 B34).

316
 6.1 Interrelations
attained divine knowledge about the nature of ultimate reality.
The kouros, then, learns what specifically human cognition is
like only after he himself transcends it to sustain divine thought
with, or as, his divine soul, and only after and in the light of
Alêtheia.
Hesiod’s Muses grant him access to an extra-human, divine
perspective on the cosmos.19 In this respect, Hesiod sides with
Parmenides against Xenophanes. For Hesiod, however, the mortal
poet can never himself ascertain the truth-value of accounts dis-
closed from this divine perspective. In this respect, Hesiod’s epis-
temological attitude is closer to Xenophanes’ disillusioned denial
that mortals can themselves ascertain the truth-value of accounts
about matters external to their direct cognition or experience.
A similar point can be made about Hesiod’s attitude to the
boundary between mortal and divine. We noted that Hesiod
describes the divine or godlike products of sexual unions between
gods and mortals, and that these accounts starkly illustrate the
notion that mortal and divine inhabit the same, commensurable
plane, and that the former can approximate the latter.20 Hesiod’s
anthropogony in his narrative of the Races (Op. 106–201) is instruc-
tive here. The narrative enigmatically promises to relate the com-
mon origin (or, perhaps, the initial shared conditions) of gods and
mortals.21 It expounds humanity’s gradual decline (even if it is not
the case that each race is inferior to all the ones that preceded it) to
the current condition of the Iron Age through various middle-
ground positions between the extremes of mortality and divinity.
Thus, the ‘race of humans’ (γένος . . . ἀνθρώπων, Op. 109) who
made up the Golden Race ‘lived as gods’ (Op. 112), without ageing
(Op. 113–14) and died as in sleep (Op. 116). They persist as benign
divinities (daimones, Op. 121–6). Again, the Heroic Race is styled
‘divine’ and is represented by ‘half-gods’ (ἀνδρῶν ἡρώων θεῖον
γένος . . . ἡμίθεοι, Op. 159–60). Of these, some died, while others
still live, ‘apart from humans’ (δίχ’ ἀνθρώπων) in the Islands of the
Blessed (Op. 166–73).22 The general direction of this graded
19 20
Clay (2003) esp. 72, 182 emphasises this point. Ch. 5.3.
21
‘How from the same (ὁμόθεν) gods and mortal humans came to be’, Op. 108, with West
(1978) ad loc.
22
Cf. Parmenides’ ‘far from the track of humans’, DK28 B1.27.

317
 Retrospect and Prospect
process of anthropogony, then, is one of decline, away from divi-
nity, and towards our Iron Age. Like Parmenides, and unlike
Xenophanes, Hesiod accepts the theological phenomenon of mortal
godlikeness and, more generally, a measure of malleability and
negotiability in the boundary between mortal and divine. Unlike
Parmenides, however, Hesiod recognises no prospect for us to
ascend back to godlikeness.23 This theological divergence between
Hesiod and Parmenides corresponds to the divergence between
their respectively disillusioned and optimistic epistemologies.
Unlike Homer and other archaic sources, Hesiod never describes
the mortal poet as divine or godlike (θεῖος ἀοιδός, Od. 4.17, etc.).24
For Hesiod, the poet is inexorably, even wretchedly, mortal. In these
respects, his theology and epistemology are, again, closer to
Xenophanes than to Parmenides.
We saw how Hesiod displays ambivalent theological vacilla-
tions of a kind we find neither in Xenophanes nor in Parmenides,
and how the epistemologies of both alike diverge from the inde-
terminacy of his framework. We can now add, however, that, in
certain other aspects of their theology and epistemology,
Xenophanes and Parmenides share with Hesiod more than they
share with one another. All three thinkers reconsider the nature of
the divine and reconceptualise its role in human reflection and
inquiry. In doing so, they show both fundamental continuities and
instructive discontinuities with one another.

6.2 The daimôn and Muse of Empedocles

As noted in the Introduction, this book does not aim to exhaust the
question of the connections between theology and epistemology in
early Greek philosophy. The aim was to offer a new analysis of the
significance of these connections in certain key episodes in the

23
In On the Obsolescence of Oracles (415b–c), Plutarch underscores this aspect of
Hesiod’s text. He sees Hesiod as the first to demarcate clearly between four classes:
gods, daimones, heroes and mortals. He then immediately proceeds to specify an
alternative model, which preserves those same categories but, unlike Hesiod, introduces
the prospect of upward mobility from one to the other.
24
For the poet as divine or godlike, see further Ch. 5.3, with n.119. Contrast Th. 95,
99–100; Op. 26, 208; cf. Stoddard (2004) 90. On Hesiod’s ‘divine human-voice’ (αὐδὴν
θέσπιν, Th. 31–2), see Ch. 2.2.

318
 6.2 The daimôn and Muse of Empedocles
emergence in Greece of systematic reflection about speculative
inquiry, and to put forward interpretative approaches which could
serve as starting points for considerations of other and later devel-
opments. In other words, the story does not end here, and I have
not tried to tell all of it. But one later episode in the story that
allows for a more circumscribed consideration and which will,
I believe, uniquely enrich the overall picture emerging from this
book, is that of the daimôn and Muse of Empedocles.
It is very likely that Empedocles (working in Sicily, around the
middle of the fifth century BCE) was aware of and engaged with all
three of the thinkers who were our primary focus in this book. It is
difficult not to see in his Muse ‘Calliopeia’ (DK31 B131) an echo of
Hesiod, who makes Calliopeia chief among the Muses (Th. 79).25 He
criticises the view that the earth is unbounded as one held by people
who have seen little of the whole (B39). Aristotle (Cael. 294a21–8)
relates this criticism to Xenophanes, who indeed held this view
(DK21 B28). Elsewhere, Empedocles speaks of a holy mind (φρὴν
ἱερή) and denies that it has human form (B134), echoing
Xenophanes’ remarks on his greatest god (DK21 B23–5).26 Finally,
echoes of Parmenides’ language and thought in Empedocles abound
and make it clear that Empedocles was familiar with his work.
At one juncture, Empedocles invites us to hear the ‘undeceptive
expedition’ of his account (λόγου στόλον οὐκ ἀπατηλόν, B17.26).
Commentators have traditionally and plausibly found here an unre-
served rehabilitation of the enterprise of cosmology in the face of the
statement of Parmenides’ goddess that her account of Doxa’s cos-
mology is ‘deceptive’ (ἀπατηλόν, DK28 B8.51b-2; Chapter 4.4).27
Empedocles elaborated a cosmology which postulated four
elements (earth, fire, air and water) and two guiding cosmic

25
On the strong and unmistakable echoes of Hesiod in Empedocles, see Most (2007)
284–92. When Aristotle observes that Hesiod, Parmenides (note DK28 B13) and
Empedocles all recognise in love or desire a cosmic force (Metaph. Α.4 984b22–985a
10), he arguably touches on a genuine connection between the three.
26
Picot and Berg (2013) reasonably contend that this echo is less than simple and stress the
theological divergences between the two thinkers.
27
On Empedocles’ engagement with Parmenidean ideas and his use of Eleatic-style
argumentation, see e.g. Graham (1999); Inwood (2001) 22–30; McKirahan (2005).
Osborne (2006) and Palmer (2009) 225–349 question the orthodoxy that Parmenides
exerted a profoundly formative influence over the following generation of cosmologists.

319
 Retrospect and Prospect
powers (Love and Strife). He also narrated his own career as
a daimôn, who was exiled from the company of the gods for his
crimes and is now undergoing a series of transmigrations, leading
up to his imminent return to the company of the gods. Earlier
attempts to separate these two aspects of Empedocles’ thought as
consciously unrelated and divorced were already coming under
considerable strain when the publication in 1999 of the Strasbourg
papyrus – in which both appear side by side – finally put them to
rest.28 The question of the relation between these aspects of
Empedocles’ thought remains, however, a central topic of
debate.29 Discussions have centred predominantly on the connec-
tions of Empedocles’ theology, anthropology and eschatology
with his cosmology and ethics. We will focus here on the connec-
tions of his theology, anthropology and eschatology with his
epistemology. In particular, how do Empedocles’ daimonology,
theory of metempsychosis and invocations of the Muse fit
together? How do they relate to his conception of cosmological
speculative inquiry?30

28
In particular, ensemble d of the papyrus (which includes B139) treats within the same
passage both the cosmological topic of zoogony and Empedocles’ crimes in a past
lifetime; see further Trépanier (2003) 14–19; cf. Sassi (2009) 190. For earlier critiques
of the separatist policy, see Guthrie (1952) 102–3; Kahn (1971); Osborne (1987b).
Nothing in my discussion will turn on whether our fragments derive from two poems or
only one.
29
Indeed, the papyrus did not put an end to deflationary approaches to Empedocles’
daimonology. Primavesi himself – one of the original editors – highlights the echoes
and broad structural correspondences between the cosmic and daimonic cycles and
restricts Empedocles’ talk of transmigration to a purely mythological and evaluative
level of expression that should not be regarded as positive doctrine; see Primavesi
(2008) 262–8. More likely, the echoes between the different cycles reflect the point, not
that the daimôn’s biography is merely a symbolic representation of cosmic history, but
that it is the same two antagonistic powers, Love and Strife – powers which exert moral
as well as physical influence (B115.14: ‘trusting in mad Strife’; cf. B128) – that guide
not only macrocosmic processes in the world but also the microcosmic life of the
daimôn who is Empedocles.
30
One central preoccupation in this section will be the connection in Empedocles
between (i) the prolongation of existence and the expansion of experiences across
successive incarnations and (ii) the acquisition of greater authority and wisdom. This
fundamental connection has been recognised (Trépanier (2004) 169; Sassi (2009)
193–4, 234–5; Palmer (2009) 276–8; Ferella (2013) 43–7; Ranzato (2015) 99–105;
Clay (2015) 128–34), but much more remains to be asked and explored concerning the
light it sheds on Empedocles’ conceptions of wisdom, understanding, speculative
inquiry and self-identity, as well as the broader role and import of certain texts (such
as B106 and B108).

320
 6.2 The daimôn and Muse of Empedocles
Empedocles tells us about his exile from the gods (B115):
ἔστιν ἀνάγκης χρῆμα, θεῶν ψήφισμα παλαιόν, 1
ἀίδιον, πλατέεσσι κατεσφρηγισμένον ὅρκοις·
εὖτέ τις ἀμπλακίῃσι φόνῳ φίλα γυῖα μιήνῃ,
†ὅς καὶ† ἐπίορκον ἁμαρτήσας ἐπομώσει,
δαίμονες οἵτε μακραίωνος λελάχασι βίοιο, 5
τρίς μιν μυρίας ὧρας ἀπὸ μακάρων ἀλάλησθαι,
φυόμενον παντοῖα διὰ χρόνου εἴδεα θνητῶν
ἀργαλέας βιότοιο μεταλλάσσοντα κελεύθους.
αἰθέριον μὲν γάρ σφε μένος πόντονδε διώκει,
πόντος δ’ ἐς χθονὸς οὖδας ἀπέπτυσε, γαῖα δ’ ἐς αὐγὰς 10
ἠελίου φαέθοντος, ὁ δ’ αἰθέρος ἔμβαλε δίνῃς·
ἄλλος δ’ ἐξ ἄλλου δέχεται, στυγέουσι δὲ πάντες.
τῶν καὶ ἐγὼ νῦν εἰμι, φυγὰς θεόθεν καὶ ἀλήτης,
νείκεϊ μαινομένῳ πίσυνος. 14

There is an oracle of necessity, an ancient decree of the gods,

Everlasting, sealed with broad oaths:
Whenever one in his crimes stains his own limbs with blood . . .
committing misdeeds, swears falsely,
One of the daimones who have obtained as their lot a long life,31
He is to wonder for thrice ten thousand seasons away from the blessed ones,
Growing to be all sorts of forms of mortal things through time,
Interchanging the hard paths of life.
For the force of aether drives him into the sea,
And the sea spits him out onto the surface of the earth, and the earth into the
beams
Of the shining sun, and it throws him into the eddies of aether;
And one receives him from another, but all abhor him.
I too am now one of them, an exile from the gods and a wanderer,
Trusting in mad strife.
Having committed some crime such as bloodshed or perjury,
Empedocles must wander the cosmos away from his fellow
gods, being successively reincarnated in ‘all sorts of forms of
mortal things’. In B117, we learn that Empedocles already became
‘a boy and a girl and a bush and a bird and a fish travelling in the
sea’. In B127, he refers to daimones who are reincarnated as lions
among beasts and laurels among trees. Here too Empedocles is
likely thinking of past lives that he experienced first-hand.
Elsewhere, he laments for crimes which he himself had committed
31
These long-lived daimones are most likely included among the ‘long-lived gods’ of
B21.12; note φυγὰς θεόθεν (B115.13); cf. Sedley (2005) 346.

321
 Retrospect and Prospect
as some such wild beast: ‘Woe! That the pitiless day did not
destroy me before I devised with my claws terrible deeds for the
sake of food’ (Strasb. d 5–6 = B139). In B146, he describes the
penultimate stop before the final return to godhood: ‘And finally
they become prophets and singers of hymns and doctors and
leaders among men who dwell on earth. From there, they blossom
as gods, foremost in honours.’ As the last stop before the final
return to godhood, Empedocles noticeably lists activities which
typify himself, suggesting that his own return to the company of
the gods is imminent. He describes this happy destination as an
idyllic gathering (B147).
Empedocles’ special awareness of his true identity as a daimôn
and his past lives clearly has dramatic ethical ramifications.
It follows, for example, that we must forgo meat and animal
sacrifice: Empedocles envisages a chilling scenario in which
a father sacrifices and consumes an animal which in fact houses
the reincarnated daimôn of his dead son (B137; cf. B128; B136).
But what are the epistemological ramifications of this special
awareness of past lives?
We should consider this question in relation to the traditional
association of wisdom with longevity and experience – a tradition
which we discussed in relation to Xenophanes. In Homer, Nestor
emblematises old age and wisdom. Old men in general ‘see both
backwards and forwards’ (Il. 3.108–10). That is, they possess both
a more informed and a more careful perspective. The Odyssean
notion of expanding one’s wisdom and learning by expanding
one’s range of experiences becomes a methodological tenet
among proponents of Ionian historiê. Thus, Xenophanes asserts
that ‘as [mortals] search in time they discover better’ (DK21
B18.2). Indeed, we found reason to think that Xenophanes may
have put forward his estimation of his own age at 92 as a reaction
against the hyperbolic tradition that Epimenides lived to be 154.
At any rate, Xenophanes manifests the widely prevalent associa-
tions between (i) old age, (ii) an extended repertoire of experiences
and (iii) the premium placed on one’s authority as a thinker and
speaker. Heraclitus reacts in part against this cluster of associa-
tions when he says: ‘Much-learning does not teach how to have

322
 6.2 The daimôn and Muse of Empedocles
understanding; otherwise, it would have taught Hesiod and
Pythagoras, and again Xenophanes and Hecataeus’ (DK22 B40).32
It is with this background in mind that we should read
Empedocles’ description of an extraordinary sage: ‘And there
was among them a man of surpassing knowledge (περιώσια
εἰδώς), who indeed obtained the greatest wealth (πλοῦτον) in his
breast, master of all kinds of particularly wise deeds; for whenever
he reached out with all his breast he easily beheld each of all the
sorts of things which are (ῥεῖ’ ὅ γε τῶν ὄντων πάντων λεύσσεσκεν
ἕκαστον) in ten or twenty lifetimes of humans’ (B129). Porphyry,
who quotes the fragment (VP 30), identifies the sage as
Pythagoras, traditions about whom do indeed have him remem-
bering his past lives.33 The important point for us is that the
anonymous sage serves for Empedocles as a kind of epistemolo-
gical hero. His surpassing knowledge is related to his ability to
behold – that is, to conjure mentally and so to draw on – a range of
experiences spread over multiple lifetimes of humans. The image
of ‘wealth’ stored up in his thinking organs again reflects the
association between his extraordinary knowledge and the extra-
ordinarily rich repertoire of experiences of which he is aware, and
on which he can draw.34 In B132, we find the general principle:
‘blessed is he who acquired wealth in his divine breast’ (θείων
πραπίδων . . . πλοῦτον).
In keeping with his positive portrayal of the sage, Empedocles
also presents us with epistemological villains (B2):
στεινωποὶ μὲν γὰρ παλάμαι κατὰ γυῖα κέχυνται· 1
πολλὰ δὲ δείλ’ ἔμπαια, τά τ’ ἀμβλύνουσι μέριμνας.
παῦρον δ’ ἐν ζωῇσι βίου μέρος ἀθρήσαντες
ὠκύμοροι καπνοῖο δίκην ἀρθέντες ἀπέπταν
αὐτὸ μόνον πεισθέντες, ὅτῳ προσέκυρσεν ἕκαστος, 5

32
On these issues, see further Ch. 3.5.
33
See Burkert (1972) 138–9. According to Diogenes, some thought that the sage was
rather Parmenides (8.54). Empedocles himself, then, probably left him unnamed.
34
The sage’s acquisition of wealth in his breast and ability to behold objects from multiple
‘lifetimes’ strongly suggests a recollection of past lives, pace Macris and Skarsouli
(2012) 362. This is corroborated by the connections elsewhere in Empedocles –
explored below – between metempsychosis, experience, memory, wisdom and author-
ity. The sage is not ascribed a complete view of the life of the world or of all objects and
events (contra Macris and Skarsouli (2012) 365–6, 371), but one restricted to ‘ten or
twenty’ human lifetimes (B129.6).

323
 Retrospect and Prospect
πάντοσ’ ἐλαυνόμενοι, τὸ δ’ ὅλον <πᾶς> εὔχεται εὑρεῖν·
οὕτως οὔτ’ ἐπιδερκτὰ τάδ’ ἀνδράσιν οὔτ’ ἐπακουστά
οὔτε νόῳ περιληπτά. 8

For narrow devices are spread throughout their limbs;

And many wretched things strike in, and they blunt their reflections.
And after they had seen only a small portion of life in their lifespans
Swift to their doom they sore and fly off in the manner of smoke,
Each one believing only that thing which he has chanced to meet,
As they are driven in every direction, but <each> boasts that he has seen the
whole;
In this way these things are neither seen nor heard by men
Nor grasped with the understanding.
These thinkers operate on the basis of a narrow and myopic
repertoire of experiences and so with a narrow and myopic
perspective.35 Furthermore, they are unaware that their perspec-
tive is limited. Having observed only a small slice of world, they
mistake it for all there is to be aware of and extrapolate carelessly
from their immediate and narrow spatial and temporal surround-
ings. By juxtaposing Empedocles’ positive description of the
sage (B129) with his account of these fools, we can see that B2
does not describe a cognitive predicament which typifies all
mortal agents equally and to the same extent.36 The epistemic
superiority of the sage was related to his mental retention of
experiences from multiple lifetimes. Similarly, the fools’ episte-
mic inferiority is traced to some kind of transience. Rather than
storing up experiences from this lifespan and adding them to
another, ‘swift to their doom they sour and fly off in the manner
of smoke’. Rather than persisting across lives, they are scattered:
‘driven away in every direction’. This transience could be strictly
cognitive or quite general. These expressions could indicate only
that the failed thinkers are unaware of their past transmigrations
and past lifetimes. They are thus cognitively impermanent and
cognitively limited at any moment to their present lifetimes.
Alternatively, Empedocles could indeed be saying that, unlike
the anonymous sage, unlike Empedocles himself or (presum-
ably) unlike his addressee Pausanias, these people are not
35
Cf. similarly B39 (‘having seen little of the whole’).
36
Contrast the tenor of Palmer (2013) 316, 319, 323–4 (different to Palmer (2009) 276–8).
οὕτως (B2.7) refers to the faulty methods of the ὠκύμοροι, cf. Panagiotou (1983) 282–3.

324
 6.2 The daimôn and Muse of Empedocles
reincarnated daimones, but mere humans, limited in every sense
to a single lifetime.37
Either way, the fools’ epistemic inferiority has an obscure
physiological manifestation. Their possession of ‘narrow devices
throughout their limbs’ (B2.1) is correlated with their narrow
perspective and faulty manner of reflecting, as they are distracted
by and, perhaps, overreact to and over-generalise from the many
wretched things they encounter. Empedocles draws a general
moral: ‘In this way these things are neither seen nor heard by
men nor grasped with the understanding.’ It is not obvious what
‘these things’ (τάδε) are. Empedocles could mean quite generally
the objects of his inquiries: the physical and moral nature and
order of all things. Alternatively – or in addition – τάδε could
gesture towards our immediate vicinity (‘these things here’): peo-
ple cannot hope to apprehend truly even the things immediately
around them if they view them only in the context of a narrow
perspective. On either reading, their faulty dataset and penchant to
over-generalise from parochial experiences mean that the fools
cannot hope to achieve proper understanding by either perceptual
or non-perceptual means. The fools, then, possess ‘narrow
devices’ in their ‘limbs’. Correspondingly, Empedocles says of
Pausanias that ‘these things’ (ταῦτα: the same question concern-
ing the precise reference of the term recurs), if Pausanias was able
to store them in his breast with a proper effort of mind, would be
present for him stably throughout his life (δι’ αἰῶνος38
παρέσονται), and his physical cognitive apparatus would literally
and materially grow (αὔξει, B110; cf. B17.14; B129.2, 4).39

37
Sedley (2005) 345–7 interestingly identifies in Empedocles a distinction between one
race of (Love-generated) daimones and another of (Strife-generated) humans who do
not transmigrate.
38
This language (αἰών) could plausibly convey here, not a brief period of time limited to
Pausanias’ current incarnation (contrast the fools’ fleeting ‘lifespans’ and paltry ‘portion
of life’: παῦρον δ’ ἐν ζωῇσι βίου μέρος, B2.3), but the span of his existence quite
generally, as it does in relation to the daimones in B115.5 (μακραίωνος, cf. θεοὶ
δολιχαίωνες, B21.12; B23.8).
39
On B110, see further Macris and Skarsouli (2012) 367–8. Is the idea in Empedocles that,
through such cognitive and material ‘growth’, Pausanias (and Empedocles’ ideal read-
ers?) will also come to be aware of and recall his own past lives? Or does Empedocles
only want us to recognise his own inimitable authority on the basis (in part) of his own
inimitable awareness and memories? Our evidence does not give a clear answer.

325
 Retrospect and Prospect
In B115 (cited above), Empedocles identifies himself as
a daimôn who has won by lot a long-lasting life (5) and grown
into all manner of mortal life forms through time (διὰ χρόνου, 7).
He recounts how his journey through successive life forms takes
him to different areas of the cosmos, dominated by different
elements: from the aether to the sea, from the sea to the earth,
from the earth to the sun and then back into the eddies of aether
(9–12). Certainly, these lines are a self-lamentation. The daimôn is
not permitted to linger anywhere and is jostled from one incarna-
tion to the next (12). Still, in the light of Empedocles’ description
of the anonymous sage (B129), there is also an element of epis-
temic self-promotion here. Empedocles’ lamentable journey
resulted in an extraordinarily extensive range of personal encoun-
ters with large-scale manifestations of the world’s different areas
and basic, elemental building blocks.
Again, that Empedocles was ‘a boy and a girl and a bush and
a bird and a fish travelling in the sea’ (B117) reflects his surpass-
ingly rich repertoire of experiences. Xenophanes asks us to ima-
gine how not only humans but also cows and horses would depict
the gods (DK21 B15). Heraclitus asks us to reflect on the nature of
seawater by considering its impact not only on humans but also on
fish (DK22 B61). Empedocles takes to an eccentric new level the
epistemological worry about myopically limiting ourselves only
to the human perspective on things. He experiences not only
different areas of the world and in different times, but also in
diverse ways and from diverse perspectives. Empedocles struc-
tures B117 in such a way as to underscore that these perspectives
range over both genders (‘a boy and a girl’) and include organisms
belonging to the land, the air and the sea.
B11 offers one concrete case in which Empedocles’ extraordin-
ary repertoire of experiences makes some epistemological differ-
ence: ‘Fools – for their reflections do not have a far-reaching
intelligence (δολιχόφρονες) – are those who expect that what pre-
viously was not comes to be or that anything dies and is utterly
destroyed.’ A person aware of his own transmigratory career (like
Empedocles and the anonymous sage) would be aware that
a thing’s generation need not mean its absolute popping into
existence (rather than its construction from pre-existing
326
 6.2 The daimôn and Muse of Empedocles
constituents), and that its dissolution need not mean its categorical
disappearance (rather than its transformation into something else).
Again, Empedocles says that a wise man would not divine that,
before we are ‘formed as mortals’ (πάγεν τε βροτοί) or after we
die, we are nothing, or that it is only during ‘what people call life’
(i.e. our current incarnation), that we are subject to evil and good
(δειλὰ καὶ ἐσθλά, B15).40 Of course, Empedocles’ eschatological
experiences do not directly yield his thesis that birth is only the
combination, and death the separation, of persisting elements. But
his experiences do cohere well with this cosmological view about
birth and death. The fools fail to have a ‘far-reaching intelligence’.
The imagery is again that of an excessively limited and myopic
perspective. In another text, Empedocles begins his account of
a zoogony, which appears to be influenced by Strife, and which
certainly takes place in an earlier period than the present one, with
the words: ‘But come now! Hear these things about how separat-
ing fire brought up the nocturnal shoots of men and women, full of
lamentations. For the account is neither wide of the mark nor
uninformed’ (ἀδαήμων, B62.1–3). The last term, adaêmôn, sug-
gests that one is uninformed by personal experience in some
matter.41 With the word adaêmôn, Empedocles plays on his term
of art daimôn. His account is not uninformed: it is given by
a daimôn possessed of a long life and extensive experiences.
In his Cratylus (398b5–7), Plato presents the same etymology
for ‘daimôn’, deriving it from the word daêmôn (‘knowing’,
‘experienced’).42
In sum, part of what underlies Empedocles’ claims for episte-
mic authority is his ability to draw on a surpassingly rich repertoire
of experiences, spanning spatially and chronologically diverse
parts of cosmic history, as well as the diverse perspectives of
different organisms. Empedocles not only wanders the cosmos
as a daimôn, he also pursues cosmological inquiry as a daimôn.

40
Empedocles’ figurative use of the verb ‘divine’ (μαντεύσαιτο) draws on the role in
divination of the formation of views on the basis of experiences, but also evokes the
association of divination with transcending a narrow mortal view of things. For both
points, see Ch. 3.1.
41
Il. 5.634; Il. 13.811; Od. 12.208; Od. 17.283; cf. Od. 24.244 (ἀδαημονίη).
42
For δαήμων, cf. Il. 15.411; Il. 23.671.

327
 Retrospect and Prospect
Is Empedocles, then, unwary of the Heraclitean objection that
much-learning does not generate understanding, certainly not by
itself? Does he imagine naively that the accumulation of experi-
ences automatically yields a proportionate accumulation of wis-
dom? The answer must be an emphatic ‘no’. In fact, Empedocles is
clear that one must know how to use and how not to use perceptual
experiences (B3.9–13). He cautions us against setting more store
by any one of our perceptual faculties than any other. He urges us
to consider each object in all of the perceptual ways in which it is
manifest. Most interestingly, Empedocles stresses in this connec-
tion that our perceptual organs constitute ‘a passage for under-
standing’ or ‘for insight’ (πόρος . . . νοῆσαι, B3.12). Empedocles,
then, puts some daylight between perceptual experience and
understanding. Our perceptual faculties are instruments that can
facilitate but also obstruct understanding and insight, depending
on how we use them.
B17 gives further indications of the proper role and limitations
of perceptual experiences, and goes some way towards clarifying
how they might offer ‘a passage for understanding’ in the highest
sense. When he touches on the operation of Love among the
elements, Empedocles introduces a brief methodological digres-
sion: ‘ . . . and love among them, equal in length and breadth. And
you, gaze on her with your mind (νόῳ δέρκευ) and do not sit with
stunned eyes (ὄμμασιν). For she is recognised even by mortals
(νομίζεται), inborn in their joints, and by her they think loving
thoughts and accomplish works of unity, calling her by the names
Joy and Aphrodite. Her no mortal has perceived (δεδάηκε) whir-
ling among them. But you, hear (ἄκουε) the undeceptive expedi-
tion of my account’ (B17.20–6). Some important objects of
inquiry, like Love, cannot be registered through sensory percep-
tion, however rich our repertoire of perceptual experiences.
Empedocles cleverly underscores this point. The digression starts
with the injunction to look at Love with the mind (21) and ends
with the injunction to listen to Empedocles’ account. Empedocles
uses sensory terminology to bookend his core point that sensory
perception will not get us the cognition we are looking for here.
You precisely cannot see or hear Love. On a first reading, then, we
might conclude that personal, perceptual experience is irrelevant
328
 6.2 The daimôn and Muse of Empedocles
for the enterprise of coming to recognise and understand Love. But
the situation is not so simple. There is a meaningful sense in which
we are indeed asked to ‘see’ and ‘hear’. In the middle of the
digression (‘for she is recognised . . . Aphrodite’, 22–4),
Empedocles appeals to familiar manifestations of Love’s influ-
ence, which we can experience directly. He asks us to recognise
Love as that power which, residing within us, makes us think
loving thoughts and effect unity (as opposed to strife and division).
Our direct and personal awareness of the operation of Love within
us can then bring us closer to recognising the work of this same
unifying power in the world at large. In sum, we cannot perceive
Love; and yet, our direct awareness of Love’s effects on us can
indirectly help us to recognise and understand the nature of the
power which causes those effects.43
In the continuation of the same passage in the Strasbourg
papyrus, Empedocles describes the generation of animals, humans
and plants under the influence of Love (and perhaps also Strife).
Empedocles invites Pausanias to hear his true account (Strasb.
a (ii) 22), but remarks that he will also demonstrate his point to
Pausanias through his eyes ([δεί]ξω σοι καὶ ἀν’ ὄσσ’, 23; cf. 21:
‘not through ears alone’). Empedocles then describes the natures
of some zoogonic products, specimens of which are still extant
(νῦν ἔτι λοιπά, 25), and encourages Pausanias to deploy his own
observations of them as evidence: ‘from these things carry to your
mind true proofs of words (ἐκ τῶν ἀψευδῆ κόμισαι φρενὶ δείγματα
μ[ύθων]), for you will see (ὄψει) the coming together and the
unfolding of the offspring’ (29–30). Lines 24–8 are fragmentary,
and we cannot determine the precise details of the supposed visual
evidence. We find a related point, though, in Empedocles’ asser-
tion that ‘hairs and leaves and the dense feathers on birds are the
same (ταὐτά), and the scales on stout limbs’ (B82). These see-
mingly disparate objects display a functional similarity –
a protective and preserving effect, which points to the hand of

43
On the importance of ‘analogical reasoning’ in Empedocles (and in B17), whereby we
grasp and corroborate universal cosmological principles by reflecting on instances of
their operation within the limited sphere of our own experiences, see Kamtekar (2009);
Palmer (2013) 322–7. The former (226–31) also emphasises the distinctions in
Empedocles between thinking and perceiving.

329
 Retrospect and Prospect
Love.44 At any rate, Empedocles’ general methodological attitude
here is similar to what we found before. We cannot directly
perceive Love, but we can directly perceive some of her products
and effects. Empedocles, in sum, has good reasons for thinking
that we become better equipped epistemically as our repertoire of
experiences grows richer. The more direct experiences one has of
the products and effects of Love and Strife, the more one has to
work with in coming to recognise and understand these universal
powers.
Let us turn to two further fragments which, I think, are very
relevant for us:
πρὸς παρεὸν γὰρ μῆτις ἀέξεται ἀνθρώποισιν.
For people’s cunning intelligence is expanded in relation to what is present.
(B106)
ὅσσον <δ’> ἀλλοῖοι μετέφυν, τόσον ἄρ σφισιν αἰεί
καὶ τὸ φρονεῖν ἀλλοῖα παρίσταται.
And insofar as they change over to become of a different sort, so far it is always
Present to them also to cognise different sorts of things.
(B108)
Aristotle cites these lines to implicate Empedocles in an alleged
ancient consensus, which identified sense-perception as cognition
and failed to demarcate it from any other sort of thought or
cognition.45 On Aristotle’s portrayal, the point of these lines is
that one’s thinking is always nothing more than unwilled, uncon-
trolled and veridical reactions to one’s current physiological state
and immediate surroundings. But could this really be the import of
Empedocles’ statements? Indeed, we saw that Empedocles not
only distinguished between (i) uncontrolled perceptual cognition
and experience and (ii) deliberate, non-perceptual understanding,
but also developed sophisticated methodological reflections about
the proper relation between the two. He also clearly thought that
false opinions are possible (B2, B39). Finally, by contrast with
Aristotle’s concluding remark (Metaph. Γ.5 1010a1–3), that what

44
Note also B84 on Love’s construction of the pupil, which is protected by the eye like
a fire inside a lantern. Cf. Kamtekar (2009) 232–3.
45
See Ch. 4.3 on Aristotle’s doxographic discussions in Metaph. Γ.5 1009b12–1010a3 and
de An. 3.3 427a17-b14.

330
 6.2 The daimôn and Muse of Empedocles
underpinned these views about sense-perception was the assump-
tion that the things-that-are are exclusively perceptible (aisthêta),
Empedocles demonstrably thought that some objects of inquiry
and understanding (like Love) are not perceptible.46 We can more
plausibly relate these fragments to the themes that we have been
charting.
B108 conveys rather directly the idea that, by becoming differ-
ent sorts of things, one gets the opportunity to experience different
sorts of awareness and cognition. To be sure, the point is not
limited to metempsychosis. The fragment expresses the more
general thought that changes in our constitution lead to changes
in the sorts of awareness and cognition that are available to us.
Thus, when we become ill or healthy or old or young or a man or
a woman or a lion or a bird, the nature of our awareness and our
cognition changes along with our constitution. Changes in our
constitution quite generally lead to changes in what is available
for us to experience and register. Nonetheless, metempsychosis
will be one especially significant and radical type of change,
which – if, like Empedocles, one can retain one’s recollections
of past incarnations – leads to a quantitatively and qualitatively
richer repertoire of experiences.
B106 tells us that people’s mêtis – their cunning intelligence,
their resourcefulness – increases in proportion to what is present to
them. I suggest that at least one point this line is making is that the
richer your repertoire of experiences is – the more things are
present to you – the greater your cunning intelligence will be.
We saw that Empedocles does not hold the naïve view that an
accumulation of experiences automatically yields greater wisdom
and understanding. In this fragment he does, however, indicate
that a richer repertoire of experiences will allow people an
increased ability to reflect thoughtfully, carefully and resource-
fully. The thought is not so implausible: the more experience you
gain, the more things you become aware of, the more careful and
shrewd you get in the ways in which you act and think on the basis
of your experiences. The anonymous sage, who can behold
(λεύσσεσκεν) with his mind each of all the sorts of things which

46
Cf. also B133.

331
 Retrospect and Prospect
are in ten or twenty lifetimes (B129), is precisely a figure to whom
an enormous array of things are present whenever he wishes.
The more that is currently present in the form of memories to
humans – from their current and past lifetimes – the more they
have to work with in expanding their intelligence.
We can thus integrate these fragments into Empedocles’ overall
epistemological framework. B106 and B108 reflect the epistemo-
logical significance in Empedocles of experience and recollection,
as well as of metempsychosis.
Let us sharpen further one important sense in which
Empedocles does not merely advocate a simple, quantitative accu-
mulation of raw perceptual experiences across several lifetimes.
Empedocles is attuned to the point that a daimôn incarnated in one
sort of body may experience and respond to the world in
a radically different way to a daimôn incarnated in a very different
sort of body. Different organisms have different perspectives on
things. In his current incarnation, moreover, Empedocles is able to
develop a second-order, synoptic and critical perspective on all
those other perspectives which he occupied in earlier incarnations.
Empedocles regrets the crimes of bloodshed which he committed
for the sake of food as a wild beast (Strasb. d 5–6 = B139, quoted
above). Clearly, when he was incarnated as a wild beast,
Empedocles was not able, as he is now able (at least to a greater
degree), to choose his actions in a reflective and deliberative way.
(Nor presumably was he then aware, as he is now, of his own
position within a series of incarnations.) Otherwise, he would not
have acted as he did. As a wild beast, Empedocles was simply
driven to tear flesh for the sake of food. Now, however, he is able to
analyse and lament his actions as a beast in terms which were not
available to him as a beast. In hindsight, he can situate and criticise
those actions within a universal cosmological and ethical
framework.
It is not, then, merely raw perceptual experiences and cogni-
tions, from different parts of cosmic history, which Empedocles
carries over to be analysed in his current incarnation. Empedocles
also has an insight into what we might grandly call different sorts
of consciousness. Empedocles carries over a recollection of what
it is like to be a wild beast (or a girl, a fish, a laurel tree, etc.).
332
 6.2 The daimôn and Muse of Empedocles
Empedocles thus has some insight into the very different ways in
which Love and Strife, as universal cosmological as well as moral
powers, operate within very different creatures and organisms.
By becoming different sorts of things, we register and experience
different sorts of things (B108). In his current incarnation,
Empedocles is able to develop a second-order perspective con-
cerning those earlier perspectives and experiences. He can, for
example, reflect on the (cosmological and moral) nature of Strife
as it dominated his experiences and actions as a wild beast and
situate his current position, in relation to his earlier ones, as the
latest link in a series of reincarnations.47
Now, the universal and second-order perspective of which
Empedocles is capable in his current incarnation also allows him
an informed recognition of the epistemic and ethical limitations of
that incarnation. Indeed, Empedocles expressly tells us that he is
still now trusting in mad Strife (νῦν εἰμι φυγάς . . . νείκεϊ μαινομένῳ
πίσυνος, B115.13–14). This brings us to Empedocles’ requests for
divine aid:
καὶ σέ, πολυμνήστη λευκώλενε παρθένε Μοῦσα,
ἄντομαι, ὧν θέμις ἐστὶν ἐφημερίοισιν ἀκούειν,
πέμπε παρ’ εὐσεβίης ἐλάουσ’ εὐήνιον ἅρμα.
And you, Muse, much-remembering white-armed maiden,
I beseech you: what is right for ephemeral creatures to hear,
Send to me, driving your well-reined chariot issuing from piety.
(B3.3–5)

47
One might reasonably ask Empedocles how our experiences in and of one sort of
consciousness (a wild beast, a bird) could be transferred into and preserved as
memories in a very different one (a human). We should remember that, at least on
rudimentary, perceptual levels of cognition, all things are aware and experience in
fundamentally the same manner within Empedocles’ system. Sensory cognition oper-
ates on the principle of like-by-like: by earth we see (ὀπώπαμεν) earth, by water –
water, by aether – aether, by fire – fire (B109). (It is true that, in B109.3 – ‘and Love by
Love, and Strife by baneful Strife’ – Empedocles segues into a more figurative
application of the like-by-like principle (analogical reasoning); but we should not on
those grounds completely divest the term ὀπώπαμεν of its literal perceptual signifi-
cance with regard to the four elements in B109.1–2; contra Kamtekar (2009).)
Indeed, all things have some measure of cognition and awareness (φρόνησιν . . .
νώματος αἶσαν, B110.10). In other words, the elements are themselves independently
cognisant. Earth is aware of earth, etc. Thus, some elemental compounds (say, a bush
or a fish) will have relatively simpler and more primitive cognitive experiences.
Others (a human) will have more complex ones. But these experiences will be broadly
commensurable.

333
 Retrospect and Prospect
εἰ γὰρ ἐφημερίων ἕνεκέν τινος, ἄμβροτε Μοῦσα,
ἡμετέρας μελέτας <ἅδε τοι> διὰ φροντίδος ἐλθεῖν,
εὐχομένῳ νῦν αὖτε παρίστασο, Καλλιόπεια,
ἀμφὶ θεῶν μακάρων ἀγαθὸν λόγον ἐμφαίνοντι.
For if for the sake of any ephemeral creature, immortal Muse,
<It has pleased you> to let our concerns pass through your
thought,
Be present to me again now as I pray, Calliopeia,
As I display a good discourse about the blessed gods. (B131)48
Why does Empedocles feel an intellectual or dialectical need to
appeal to divine disclosure and to ‘the assurances issuing from our
Muse’ (παρ’ ἡμετέρης . . . πιστώματα Μούσης, B4)? If he bases his
claims for epistemic authority in no small measure on his posses-
sion of a superhuman and surpassingly rich repertoire of experi-
ences, then why does he also need the Muse? One simple but
viable answer is that Empedocles’ rich repertoire of experiences
only renders him more acutely aware of all that still lies outside his
range of personal experiences. However extensive, his experi-
ences as an exiled and wandering daimôn still cover only
a relatively small slice of cosmic history. Empedocles is very
aware of the pitfalls of mistaking whatever one chances to meet
for the whole. This awareness sets him apart from the reckless
fools (B2). Empedocles’ cosmological story ranges from the uni-
fication of all things in the homogeneous Sphere of Love to their
complete segregation under the total rule of Strife (and back
again). It certainly extends to areas and events which lie well
outside his repertoire of personal experiences. We need not posit
a neat and tidy classification of cosmological tenets, such that
Empedocles asserts on the Muse’s authority whatever pertains to
events and processes that lie outside the range of his experiences,
and on his own authority whatever he encountered first-hand.
I suggest only that we can see why Empedocles felt the need to
appeal to divine aid and authority so as to go beyond what he could
say about cosmic history independently of it. Within the frame-
work of divine revelation, our ordinary perceptual experiences,
and Empedocles’ extraordinarily rich and extensive repertoire of
48
‘About the blessed gods’ does not delimit the subject matter or exclude cosmology.
The elements are themselves gods (‘Zeus’, ‘Hera’, ‘Aidoneus’ and ‘Nestis’, B6).

334
 6.2 The daimôn and Muse of Empedocles
perceptual experiences, serve to support and illuminate universal
cosmological and ethical principles and ideas, which Empedocles
could only have received from the goddess.49
Empedocles’ appeals to the Muse raise another important ques-
tion. If Empedocles is a daimôn with a long-lasting life, if he is, as he
says, ‘better than much-destroyed mortal humans’ (θνητῶν περίειμι
πολυφθερέων ἀνθρώπων, B113), then why does he also describe
himself as an ephemeral creature?50 Could we say that Empedocles is
describing only his audience as ephemeral creatures? This solution is
implausible. Empedocles prays: ‘if for the sake of any ephemeral
creature, immortal Muse, it has pleased you to let our concerns
(ἡμετέρας μελέτας) pass through your thought . . . ’ (B131.1–2).
These lines identify Empedocles’ own concerns here as those of an
ephemeral creature. Why do we find these divergent representations
of Empedocles both as a long-living exiled divinity (a daimôn), who
is superior to ‘much-destroyed mortal humans’, and, conversely, as
an ephemeral mortal?51
We should embrace, rather than try to explain away, the paradox
that Empedocles is indeed both a long-living, divine daimôn and
an ephemeral creature. Empedocles is a daimôn insofar as he
persists across successive incarnations as the same agent

49
Palmer (2013) 322–7 advances a similar view of the structure of the relation between
ordinary human experience, reasoning and revelation in Empedocles. Less convin-
cingly, Palmer finds the closest model for Empedocles and his Muse in Parmenides
and his goddess, and suggests that we should see the Muse, and not Empedocles, as the
speaker of most of the fragments of On Nature (at 313–14, he makes a good case for
B111). Empedocles’ Muse is not herself initiatory or eschatological and there is nothing
in Parmenides corresponding to Empedocles’ addressee, Pausanias (B1; cf. B4.2–3).
The triangular model Muse-Empedocles-Pausanias is better compared to the model
Muse-Hesiod-Perses in the Works and Days (where, indeed, poet and Muse are not
generally demarcated as different speakers: Ch. 2.4).
50
For the sense of fleeting transience in ephêmerios (‘for a day’), see e.g. Od. 21.85; Thgn.
656; Aesch. Pr. 547.
51
The texts discussed in this paragraph suffice to show this duality. I take no position on
whether in B112.4 – ‘I among you a deathless god, no longer mortal (θεὸς ἄμβροτος
οὐκέτι θνητός)’ – Empedocles expresses conviction in his essential divine nature as
a daimôn (Panagiotou (1983)), refers proleptically to his imminent-but-not-yet-realised
return to full godhood (Guthrie (1965) 246; cf. Trépanier (2014) 204), or merely
comments on how he is perceived among the people of Acragas (Palmer (2013) 311).
When Empedocles states ‘know these things clearly, having heard an account issuing
from a god’ (τορῶς ταῦτ’ ἴσθι, θεοῦ πάρα μῦθον ἀκούσας, B23), he may be referring
proleptically to his own status as a god among gods or, more likely, to the Muse from
whom his account issues (cf. παρ’ ἡμετέρης . . . Μούσης, B4.2).

335
 Retrospect and Prospect
(possessed of a continuous personal history which, from his
current vantage point, he can express with first-person pronouns
and past tense verbs). His identity as a daimôn pre-existed
his current incarnation in a human body and will survive its
dissolution. At the same time, however, Empedocles is also an
ephemeral human creature insofar as he is currently incarnated in
an ephemeral human form. The fact of his current incarnation is
hardly insignificant to the question of Empedocles’ identity.
Empedocles’ current imprisonment in a human form imposes all
sorts of constraints on him, physical as well as moral and cogni-
tive. As we noted, Empedocles recalls the terrible deeds which he
committed with his claws when he was incarnated as a wild beast
(Strasb. d 5–6 = B139). That Strife-dominated form of life
impelled Empedocles to commit those violent crimes. Incarnated
as a wild beast, then, Empedocles was morally and cognitively
constrained. Incarnated as a human rather than a wild beast,
Empedocles is presumably less constrained, both morally and
cognitively. He is less a slave to unthinking impulses. But this
does not mean that his current imprisonment in a human form
imposes no such limitations and constraints. Again, Empedocles is
still now trusting in mad Strife, presumably because he cannot
help doing so (B115.13–14).
Empedocles makes the issue of self-identity a central philoso-
phical concern.52 At first sight, it might have seemed that his
answer to the question ‘who am I really?’ was ‘I am really
a daimôn, who has a long-lasting life and who persists across
successive incarnations.’ We can now see, however, that
Empedocles will have to give a somewhat more complex and
qualified answer: ‘I am really a daimôn, who has a long-lasting
life and persists across successive incarnations, and who is cur-
rently incarnated in a human body.’ The added clause is significant
to who we are. Empedocles’ current imprisonment in a human
body is constitutive of his current conduct, disposition, impulses,
capacities and limitations. It makes a difference to what he is and is
not capable and worthy of, physically, morally and cognitively.
In this sense, ‘boy’, ‘girl’, ‘man’, ‘beast’, ‘bush’, etc. pick out

52
This centrality is emphasised in Osborne (2005) 295–7; Inwood (2006).

336
 6.2 The daimôn and Muse of Empedocles
importantly different phases in the existence of the same thing (the
transmigrating daimôn).53
After he lambasts the reckless fools, Empedocles promises
Pausanias – and us – that he will take us much further, but, inter-
estingly, only as far as mortal intelligence can reach: ‘But you, then,
since you have stepped aside here, you will learn (πεύσεαι); mortal
intelligence has certainly risen no further’ (οὐ πλεῖόν γε βροτείη
μῆτις ὄρωρεν, B2.8b-9). Empedocles recognises the ultimate limita-
tions imposed on Pausanias’ intelligence as well as on his own by
their current human incarnations. In the final analysis, the seemingly
incompatible designations ‘daimôn’ and ‘ephemeral creature’ are,
in fact, both appropriate. They pick out different aspects of
Empedocles’ complex current situation. In different contexts,
Empedocles will emphasise different aspects of his complex iden-
tity. When, for example, he argues against animal sacrifice, he will
allude to his awareness of transmigrations. When, however, he
wishes to invoke the Muse, he will acknowledge his current incar-
nation in a wretched and ephemeral human body and piously ask to
be told only whatever is appropriate for a creature in that circum-
stance to hear.
The Empedocles who emerged here operates within the same
broad tradition of theological epistemology as Hesiod,
Xenophanes and Parmenides. He too pursues questions of episte-
mology and contemplates the nature of speculative inquiry by

53
The answer to the question ‘who am I really?’ discussed here does not preclude the
additional answer ‘I am really a compound of elements.’ The material constitution of the
daimones is unclear. There is no reason to think that Empedocles specified it in some lost
lines. Still, we can – minimally – make limited inferences from what he does say. First,
Empedocles signals no further exceptions to the rule that all things are elemental
compounds apart from Love, Strife and the pure elements themselves. We would thus
expect the daimones to be an elemental compound of some sort. This is guaranteed if
they are capable of causing and undergoing bloodshed (although this particular point
depends on reading φόνῳ in B115.3, which is an emendation (as Gagné (2013) 462
reminds us), even if an almost universally accepted one). Empedocles describes incar-
nation as being dressed ‘in a tunic of foreign fleshes’ (B126). The elemental makeup of
daimones, then, seems to diverge from that of the bodies in which they are incarnated.
Finally, the elemental compound which is a daimôn persists across successive incarna-
tions, but we also know that it cannot be everlasting (cf. the long-lived (but not ever-
lasting) gods in B21.12): if not before, the daimôn must ultimately be homogenised into
the universal Sphere by Love or disintegrated into completely discrete elements under
total Strife. For a discussion, which goes beyond such basic points, of the daimôn as
a material and transmigrating substance, see Trépanier (2014).

337
 Retrospect and Prospect
integrating these issues into a broader set of reflections on the
nature of mortal and divine. Like Hesiod and Parmenides, but
unlike Xenophanes, Empedocles is the privileged recipient of an
extra-human, divine perspective on the cosmos, which issues from
a female deity. Unlike Parmenides, Empedocles does not take this
disclosure to encompass ‘all things’, but only what is right for
creatures trapped in human form to hear. In his own distinctive
way, however, Empedocles too thinks that the deliverances of
divine disclosure can be supported at least in part through con-
templation: universal cosmological and ethical principles dis-
closed by the Muse illuminate and are in turn illuminated by
critical reflection on (an extraordinarily broad range of) first-
hand experiences. Again, for Empedocles, as for Parmenides, the
mortal agent54 is not simply and strictly mortal, but possesses also
a higher-than-mortal, divine aspect and potential. And, with
Empedocles as with Parmenides, this idea again has important
epistemological consequences. Like all three thinkers,
Empedocles operates with his own version of an epistemically
significant interaction between mortal and divine agents, only
through which can the mortal attain knowledge or come by certain
insights and views.
Our discussion of Empedocles in this section was perforce
focused and limited. A comprehensive consideration of all the
ways in which we might relate these insights concerning his
epistemology to his intricate theological and cosmological sys-
tem – and of how these conclusions might bear on our under-
standing of different aspects of that system – would require a much
longer study. And yet, we were able to see how Empedocles offers
one particularly clear and significant, later instantiation of the
tradition of theological epistemology which was our guiding
theme in this book. This is not to say, of course, that Empedocles
is the only other viable candidate for a treatment along the lines
pursued in this book with respect to Hesiod, Xenophanes and
Parmenides. One could also examine similarly the interrelations
between epistemology and conceptions of (the interactions
between) mortal and divine in other early philosophers like

54
At least some mortal agents, if we are not all daimones.

338
 6.3 Final Remarks
Heraclitus and Anaxagoras, not to mention in such lyric poets as
Pindar, such dramatists as Aeschylus or such historians as
Herodotus – thinkers with whom scholars of Greek philosophy
converse less frequently. Indeed, it would also be interesting to ask
how Socrates and his divine sign might relate to this earlier
tradition, exemplified by Hesiod, Xenophanes, Parmenides and
Empedocles, in which the (typically personal and privileged)
interactions between mortal and divine played an important role
in one’s understanding of oneself as an agent and as a thinker. sed
haec hactenus.

6.3 Final Remarks: Reason and Revelation, Philosophy and

Religion Again
In Chapter 1, we examined ideas of ‘rationality’, which align
rather distinct intellectual phenomena and expect them to be
naturally concomitant: on the one hand, coherent, critical, argu-
mentative, inferential, questioning and explicative thinking; on the
other hand, transitions away from god-centred patterns of thought
and explanation and, in particular, transitions towards human
inquiries that forgo any appeal to divine interference or aid.
We may now summarise, in the broadest outline, conceptions of
reasoning and divine disclosure, which we went on to find and
analyse in the subsequent chapters.
With poetic inspiration, one does encounter claims for the truth
of divinely disclosed accounts, which are received and conveyed
apparently without the intervention of reasoning. Inspired poets
often stake such truth-claims for narratives the descriptive truth of
which neither they nor their audiences can assess.55 The epic
tradition, however, also reflects destabilising queries concerning
the status of inspired but unverifiable narrations. In particular,
Hesiod forcefully flags up the second-order question of whether
inspiration in fact warrants such truth-claims. Furthermore,
Hesiod identifies the factors which will inform and frame reflec-
tion on this question: above all, our understanding of the relation
between mortal and divine, and between male and female, and of

55
Ch. 2.2.

339
 Retrospect and Prospect
how this nexus of relations structures the world and our place
within it. Hesiod and his audience, moreover, can reflect on this
question critically and constructively by considering the prescrip-
tive import even of inspired verses whose descriptive truth-value
they are powerless to evaluate.56 Hesiod is no passive and unre-
flective recipient of revelation.
In Chapter 3.1, we explored the centrality of conjectural and
interpretative reasoning in Greek divination. We saw that, stan-
dardly, diviners or interpreters actively form conjectural beliefs by
fallibly decoding communications encoded in omens or oracles.
It remains the case, however, that the diviners must proceed with
the working assumption that what the gods communicate to them
are truths, even if mortals are liable to misinterpret divine com-
munications and so to fail to register those truths. So too, when
Socrates sets out to determine what Apollo could possibly have
meant by declaring that no mortal is wiser than him, he assumes in
advance that what the god says must be, in some sense, true (Plato,
Apol. 21b6–7). Xenophanes supplants such traditional notions
with a new model of divine disclosure and of reasoning. For
Xenophanes, the divine discloses not truth-evaluable messages
but evidence. Mortals reason out conjectures about the world
beyond their experience on the basis of a necessarily limited but,
through inquiries over time, ever-growing body of experiences,
which have been made possible for them to consider by the divine.
With Parmenides we can delineate a far more sophisticated,
specific and self-consciously developed form of rational thought,
which he himself associates with the term logos (DK28 B7.5).57
This type of thought is structured by the contrast (krisis) between
‘is’ and ‘is not’, as demarcated from the competing contrast

56
Esp. Ch. 2.3. We also saw there that Hesiod had certain reasons to motivate his very
postulation that the Muses inspire both truths and falsehoods.
57
This last point does not rely on the (viable) translation ‘reasoning’ for the word logos
itself. We could also render, say, ‘discourse’, ‘thought’, ‘reckoning’, vel sim. (cf.
Granger (2010) 31–3; Mourelatos (2013) 165). The significant point is that, with her
injunction for the mortal to judge by ‘logos’, the goddess looks forward to the subse-
quent arguments in Alêtheia; cf. Ch. 4.3; Ch. 5.4. Long (2009) analyses the emergence in
Heraclitus of an idea and ideal of rationality, which incorporates a complex nexus of
notions – meaningful discourse, account, measure, proportion, balance, structure, order,
plan, law, moderation, commonality and coherence – and which is centred on his term
logos.

340
 6.3 Final Remarks
between sensory oppositions (Light/Night, Hot/Cold, etc.). It is
manifested in the extended and sustained system of argumentation
developed in Alêtheia (DK28 B8). Furthermore, this form of
rational thought is unsustainable by the human mind. When
a human’s mind does sustain this thought, it is then a divinised
and no longer a human mind. When a human reasons about and in
terms of the contrast between ‘is’ and ‘is not’, follows Alêtheia’s
argumentation and ascertains for himself the true nature of
what-is, he is thinking with, or as, his divine soul. Finally, it is
only through his interaction with the goddess, and through her
disclosure and guidance, that Parmenides was first able to acquire
the ability to sustain such thought with his divine soul. It is only
through his initiatory and transformative text, which reprises this
interaction and this disclosure and guidance, that we can do the
same. We thus find in Parmenides a major, early exponent of
the subsequent philosophical tradition of assimilation to the divine
(homoiôsis theôi), which associated our capacity for sustaining
rational thought and contemplation with a capacity for identifying
ourselves with what is divine in us.58
Does the goddess also act for Parmenides as the guarantor of
truths? With regard to the first part of the poem, this question will
turn on whether we think that the initial and foundational princi-
ples established in DK28 B2 – however precisely we interpret
those principles – are meant to be justifiable self-sufficiently and
on the basis of their intrinsic logical self-evidence alone or, rather,
rely for their acceptability on the goddess’s authority. It is unclear
that this dilemma can be determined conclusively. Nonetheless,
the way in which the goddess rests her case on the intrinsic
impossibility of knowing or indicating what-is-not (‘for neither
could you know what-is-not (for that cannot be accomplished) nor
could you indicate it’, DK28 B2.7–8) perhaps suggests the former
alternative, as does her emphatic exhortation to the kouros to judge
and evaluate her examination for himself (DK28 B7.5–6; cf.
Chapter 6.1). There is also, however, the second part of the
58
Esp. Ch. 5.4. By identifying this emerging, self-conscious notion of rationality in
Parmenides, we do not, of course, exclude that Doxa too may display what we would
recognise as (systematic, explicative, coherent, abductive) ‘rational’ thought, as, for exam-
ple, when it explains the moon’s light as a reflection of the sun’s (DK28 B14–15; A42).

341
 Retrospect and Prospect
poem (Doxa), which is no less part of the goddess’s revelation, and
the correctness of whose accounts of multiple, heterogeneous and
changing things and processes cannot be established through the
same deductive means. When the goddess affirms Doxa’s status as
the best possible cosmology, she emphasises that this account will
be her own (σοι ἐγώ . . . φατίζω, DK28 B8.60–1). It is in her voice
that we, along with the kouros, are assured that the cosmology will
endow us with knowledge and understanding of, for example,
cosmic celestial bodies and phenomena (DK28 B10). In sum, it
is certainly possible that Parmenides, not unlike Empedocles, rests
the authority of his cosmological accounts not only on their
abductive, explanatory and observational appeal but also and
concurrently on the guidance and guarantees of his kind and
benevolent goddess. Nor could we exclude definitively that the
foundational principles in Alêtheia are similarly accepted concur-
rently on the basis of both logical appeal and divine authority.
Be that as it may, the essential function of this initiatory goddess
and of her revelation is to effect the cognitive transformation of the
mortal kouros, as we discussed above.
Empedocles, finally, makes a direct appeal to the ‘assurances’ of
his Muse (DK31 B4). He combines this appeal to divine revelation
with an emphasis on the significance of perceptual experiences
and the importance of properly analysing and contemplating those
experiences. Within the framework of divine disclosure,
Empedocles’ reflections on his extraordinarily rich repertoire of
experiences, and our own reflections on our own ordinary range of
experiences, illuminate and corroborate – and are in turn illumi-
nated by – universal and ethical principles, which Empedocles
could only have received from his goddess.
We began this book by discussing certain deep-seated tenden-
cies to dissociate rationality as secularising reasoning, from irra-
tionality, as non-reasoning, god-centred religiosity; to recognise
divine interference in human inquiry only as the transmission of
truths to unreasoning and passive recipients; and to isolate reli-
gious belief as special in kind, and as insulated from, and con-
trasted with, the activity of reasoning. We ended up finding that, in
the emergence of systematic epistemology in Hesiod,
Xenophanes, Parmenides and Empedocles, different forms of
342
 6.3 Final Remarks
critical reasoning and reflection, and different models of the rela-
tion and interactions between mortal and divine, play harmonious
and equally integral roles. In their different ways, each of these
thinkers operates with a model of divine disclosure which accom-
modates, facilitates or even necessitates critical human reflection
and inquiry.
With Hesiod, Xenophanes, Parmenides and Empedocles, we
explored in greater depth a phenomenon which in Chapter 1.2
we identified more briefly also in the cases of the Milesian cos-
mologists, the Hippocratic author of On the Sacred Disease and
such myth-critics as Hecataeus, Herodotus and Palaephatus.
We find in these thinkers pioneering strides in rational (coherent,
critical, argumentative, inferential, questioning and explicative)
thinking. We do not, however, find in them concomitant secularis-
ing moves away from god-centred patterns of thought and expla-
nation as such. The ancient evidence here resists our deep-seated
but historically contingent and even peculiar expectation that such
rational thinking should run counter in a general way to religious
and god-centred attitudes, or that religious belief must be a special
and qualitatively distinct kind of belief, self-consciously insulated
from procedures of criticism and evaluation (‘faith’).59
That said, we certainly encountered in the course of this book
reflections about the divine which were distinctive, original and not
infrequently subversive or critical of common and received reli-
gious and theological attitudes. Xenophanes offers the most striking
example. He questions what he identifies and portrays as
a pervasive (DK21 B10–12, B14) system of interrelated beliefs,
which was promulgated above all by Homer and Hesiod and which
conceived of the gods as fundamentally like us in how they come to
be, appear, think and conduct themselves. In rejecting divination,
moreover, Xenophanes rejects at least one ubiquitous religious
practice.60 Hesiod himself (whether we prefer to think of him as
59
On ‘religious belief’, see further Ch. 1.4.
60
Did Xenophanes reject other practices? Our evidence does not permit us to say. His
instructions for the ideal symposium (DK21 B1) do prescribe changes in certain
religious activities (even as they retain some traditional ritual patterns) by advocating
pointedly non-conventional prayers (for the capacity to act justly) and hymns (which are
to use only ‘pure speech’ and exclude stories about strife among gods). This is hardly
evidence for a wholesale overthrow of traditional forms of worship, however, and it has

343
 Retrospect and Prospect
a philosophical pioneer in his own right or as a key initiator of later
philosophical thought61) destabilises blithe acquiescence on the part
of poets in the truth of their inspired accounts. Parmenides’ cosmol-
ogy included a theological hierarchy that ran counter to the pre-
dominant Homeric and Hesiodic patriarchal scheme, placing centre
stage a governing goddess, who also appears to be the generator of
all other gods (DK28 B13). Even this unusual theological scheme,
moreover, is excluded from Parmenides’ account of the nature of
ultimate reality. Parmenides does not ascribe to what-is traditional
theological epithets. If what-is nonetheless possesses a divine sta-
tus, then it does so only in a highly abstract and non-traditional
sense.62 Empedocles’ belief in metempsychosis led him to reject
a religious institution no less central than animal sacrifice (DK31
B128, B136–7).
And yet, the engagements of these thinkers with established
religious and theological attitudes do not present us with traceless
rejections or simple dismissals, but with complex combinations of
divergences, criticisms and creative appropriations, even of those
very aspects of religion which are being criticised. Above all, their
engagements with received or current religious ideas are involved
and substantial. We are not dealing here simply with borrowings of
imagery and patterns of speech, which have been divested of their
religious significance. The categories ‘pre-Socratic philosophy’
and ‘early Greek philosophy’ have conditioned us to think that the
key (if not the only) intellectual interlocutors of a pre-Socratic
philosopher were other and earlier pre-Socratic philosophers.63
been well observed that Plato adopted many of Xenophanes’ theological criticisms
without proposing to abolish any civic cults (Cornford (1952) 146). What we can say is
that our evidence does suggest the expectation for a harmonious correspondence
between religious belief and practice. ‘Purity’ characterises not only the hymns pre-
scribed for the symposium (καθαροῖσι λόγοις, 14) but, equally, the floor (καθαρόν, 1) and
the water (καθαρόν, 8). Conversely, when Xenophanes criticises Homer and Hesiod for
‘attributing’ to the gods immoral actions, he uses a verb which standardly signifies the
consecration of a votive gift or dedication (ἀνέθηκαν, DK21 B11.1; cf. Hes. Op. 658;
Hdt. 2.159; for the observation, cf. Lesher (1992) 84). Homeric and Hesiodic myths
about thieving and cheating gods are thus recast as perverse and impious sorts of
offerings to the gods. Finally, Xenophanes reportedly advised the Eleans to fit their
practices in connection with Leucothea (they might either worship or lament her, but not
both) to their beliefs about her (do they take her to be a goddess or a deceased mortal?),
DK21 A13. The anecdote’s historicity is questionable, but it again associates
Xenophanes with the expectation for correspondence between belief and practice.
61
On this issue, see Ch. 1.5. 62 Ch. 5.5.2, Ch. 6.1. 63 Cf. Ch. 1.5.

344
 6.3 Final Remarks
But Xenophanes was thinking with and against poets and diviners
no less than Ionian cosmologists. He devoted serious reflection to
ideas of divine disclosure adumbrated in the traditions of poetry
and divination and, while he rejected those ideas, they exerted
a substantial influence on his own, alternative model of
disclosure.64 Again, Parmenides converses with the epic tradition
and with ideas of metempsychosis and initiation current in South
Italy no less than with Xenophanes or Ionian cosmology. To be
sure, Parmenides develops his own distinctive version of the
notion that the height of mortal excellence involves the promotion
of what is divine in the mortal and the attainment of divine
qualities, and he puts to novel philosophical use the models of
interactions between humans and personal gods found in epic
poetry and in the traditions of mystery initiations. But his debt to
these religious traditions remains profound.65
Xenophanes’ philosophical verses evince a very different
theology – different beliefs about and notions of the divine –
than, say, the representations of divination in Homer or the
extant inquiries at the oracle of Zeus at Dodona. They also
evince a very different mode of theologising – a different way
of constructing and conveying beliefs about and notions of the
divine. And yet, these different materials can be seen to parti-
cipate in a shared theological conversation about how
a knowledgeable divinity might comport itself in relation to
mortals who seek the truth. Again, Parmenides’ philosophical
verses evince a different theology and a different kind of
theologising than the representations of godlike mortals in
Homer or the promises of imminent deification in the initiatory
Gold Tablets. And yet, these different materials are relatable as
participants in an ongoing conversation concerning the nature
of the boundary between mortal and divine and the prospect of
traversing this boundary.66 By contrast with the idea of ‘natural
theology’, early Greek philosophical theology did not operate
on a qualitatively different plane to common and received
religious attitudes. It was not divorced from preoccupations
64
Ch. 3. 65 Esp. Ch. 5.2–3.
66
On the ways in which very different theological materials can interact and interrelate,
see similarly and more generally Eidinow, Kindt, Osborne and Tor (2016) 4–6.

345
 Retrospect and Prospect
with the gods elsewhere in Greek culture.67 We must remember
that Greek religion itself was never anything like a theological
unity. It encompassed a mass of often competing beliefs, values,
questions and reservations. The ultimate unknowability of the
gods was often emphasised and, by and large, philosophical and
other criticisms of particular religious attitudes, representations
and even practices were accommodated. I argued in this book
that the philosophical epistemology which emerges in Hesiod,
Xenophanes, Parmenides and Empedocles was an essentially theo-
logical enterprise. As such, this enterprise was not only influenced
by Greek religion but was itself one part and aspect of Greek
religion.68
For all these reasons, this study of early philosophical episte-
mology was – and had to be – a study also of poetic inspiration,
divination, mystery initiation, metempsychosis and other aspects
of early Greek conceptions of the relation and interactions
between gods and mortals. I strove in this book to examine
philosophical texts in the light of their cultural, religious and
literary context, without thereby reducing or assimilating them to
that context. I tried to show that, by combining the analysis of
systematic and critical reflection in those texts with this kind of
contextualisation, we can develop full-blooded interpretative
accounts which enable us to integrate the different aspects of our
evidence and free us from having to decide whether a Xenophanes
or a Parmenides was ‘a systematic thinker’ or ‘a religious sage’.
They were both. In the end, we did not have to choose between
taking seriously Parmenides’ goddess or his novel forms of argu-
mentative reasoning. Indeed, we must not choose.

67
On ‘natural theology’, see Ch. 1.3.
68
We emphasised these points in Ch. 1.3, as well as the fact that our evidence for Greek
philosophy outside of and before Classical Athens gives no hint of backlash against
theological criticisms (indeed, Xenophanes betrays no hesitation about taking his
teachings from city to city: DK21 B8, B45) nor supports the notion that philosophical
theology was typically thought to clash in a general way with traditional attitudes and
practices.

346
 APPENDIX

T H E T R A JE C TO RY OF T H E KO URO S’ JOU R NEY A ND

E S C H ATOL O GI C A L TOP OGR A P H Y I N PAR M E NI DE S :
S O M E I N C O N C L U S I V E R E MA R K S

In Chapter 5.3, we analysed the proem to Parmenides’ poem both

internally and in relation to a number of religious and literary
models that it appropriates. We saw that Parmenides represents
the journey of the kouros as that of the post-mortem (Hot, Light,
aethereal) soul and evokes the imagery and terminology of mystery
initiation as well as the scenery of the Hesiodic Underworld.
Furthermore, we saw that his journey is made possible by divine
agents of Light, Heat and the higher, celestial and aethereal regions,
with whom he himself is closely connected. We concluded that
these points should be recognised independently of a prior recon-
struction of the trajectory of the kouros’ journey, and, indeed, that
any such reconstruction will have to take them into account.
A consideration of the topographical markers internal to DK28
B1 and of cultural models which it evokes should lead us to the
conclusion that, the confident assertions of feuding scholars not-
withstanding, there are at least four (and probably more) viable
interpretations of the trajectory of the kouros’ journey, none of
which we can reasonably insist is demonstrably preferable to the
others. The journey could be (i) a katabasis; (ii) an anabasis; (iii)
a journey which follows the daily course of the Sun and, therefore,
combines in different stages both anabasis and katabasis; (iv)
intrinsically blurred.1 The proem does indeed display various

1
E.g. (i): Morrison (1955) 59–60; Burkert (1969) 1–15 (but Burkert ultimately suggests
(15) that it is best to think of the journey simply as one to the Beyond; cf. Primavesi
(2013) 41); Furley (1973) 1–5; Kingsley (1999) esp. 50–3; Göbel (2002) 161; Palmer
(2009) 52–3; Gemelli Marciano (2013); Ranzato (2015) 39–42; (ii): Diels (2003) 7–8;
Fränkel (1975a) 5; Jaeger (1947) 93; Verdenius (1949) 119; Cornford (1952) 118; Bowra
(1953c) 43–4; Roloff (1970) 171ff; Kahn (2009) 210–15. Interpreting the journey’s
trajectory as an anabasis does not imply interpreting it allegorically (or vice versa).
For Jaeger (1947) 96 and Verdenius (1949) 119–20, for example, the journey is an

347
 Appendix: The Trajectory of the kouros’ Journey
topographical markers and, in that sense, itself raises the question
of the journey’s trajectory.
Before turning to evaluate these alternatives, we will first con-
sider the possible topographical implications of Simplicius’
account of the goddess’s cyclical conveyance of souls (in Ph. 9.39.
19–21 = Coxon (2009) Testim. 207, quoted and discussed in
Chapter 5.2). Then, when we assess the different candidate inter-
pretations of B1, we can ask how they might relate to what
Simplicius could be taken to suggest. Simplicius certainly refers
to some sort of topography, and we may outline some interpreta-
tive options.
(a) ‘The invisible’, τὸ ἀειδές, especially given the eschatological con-
text, is an obvious reference to Hades (for invisibility, see Il. 5.845).
One possibility, then, is that ‘the visible’ simply signifies the region
which is visible or manifest to us, i.e. the habitat of living mortals,
while ‘the invisible’, the region we cannot see, signifies the post-
mortem destination of our souls.2 On this view, death would be
mentioned first (ποτέ) and birth second (ποτὲ δὲ ἀνάπαλιν).
(b) Now, it seems possible – although not obvious – that we should map
the antithesis ‘the visible’/‘the invisible’ unto Parmenides’ funda-
mental and pervasive Doxastic antithesis Light/Night (φάος . . . νύξ,
B9; cf. B8.56–9). One might support the identification of ‘the invi-
sible’ with Night by noting that Parmenides elsewhere describes
Night as ‘unseen’ with a similar alpha-privative adjective (νυκτὸς
ἀφάντου, B9.3). Aligning the two terms ‘the visible’ and ‘the invi-
sible’ with the Doxastic antithesis is compatible with the view that τὸ
ἀειδές signifies the soul’s post-mortem destination.3 This destination,
or ‘Hades’, would be a darker and colder region than that of living
humans.4
(c) We could, however, not only associate the two adjectives with the
two elements, but also identify τὸ ἐμφανές as the region inhabited by

anabasis and a genuine encounter with the divine; Palmer (loc. cit.) does not clearly
distinguish the two positions. (iv): Mourelatos (2008a) 14–16; Curd (1998) 19; Morgan
(2000) 78; Miller (2006); Cosgrove (2011) 38–9. See Kraus (2013) 453 for scholars who
have somehow related the kouros’ journey to that of the Sun. Options (i)-(iv) are not
exhaustive. Tarán (1965) 22ff offers a now dated but nonetheless instructive overview of
a range of diverse interpretations advanced in the literature.
2
See Rohde (1925) 373; Zeller (1919), 722, n.2; Diels (2003) 109–10; Burkert (1969) 28;
(1972) 284, n.33.
3
This is how Ranzato (2015) 224–5 reads Simplicius’ testimonium; cf. also Bollack
(2006) 327–8.
4
It is possible – although this point is itself contentious – that the proem describes the
destination of the souls of the deceased as ‘the House of Night’ (B1.9).

348
 Appendix: The Trajectory of the kouros’ Journey
prenatal and post-mortem souls and τὸ ἀειδές as the region inhabited
by living mortals. On this view, the goddess conveys (Hot) souls
from the celestial, fiery and aethereal region predominated by Hot to
our own darker, colder, invisible region of the universe, and vice
versa.5 On this view, then, the adjectives do not at all signify ‘that
which is visible / invisible to us’ but carry only the objective sense,
‘that which is in itself visible / invisible’. Birth, on this line, would be
the process mentioned first (ποτέ) and death second (ποτὲ δὲ
ἀνάπαλιν). What of the point that τὸ ἀειδές evidently recalls Hades?
We could counter that Parmenides is employing τὸ ἀειδές precisely to
refer subversively to our own cold and dark region of the universe as
the true glum Hades. Indeed, we noted in Chapter 5.2 that
Parmenides strikingly describes birth as ‘hateful’ (στυγεροῖο,
B12.4), using a term that standardly describes death and Hades.
Furthermore, our lower, sublunary region of the universe does indeed
have a larger proportion of cold, dark, ‘unseen Night’ (B9.3) than the
higher region of aether, and Parmenides did associate Cold with
death (see Chapter 5.2).6

There are, in sum, three possible construals of Simplicius’ topo-

graphical markers. (a) The goddess conveys the post-mortem, Hot
soul from our region of the world (the visible) to the afterlife (the
invisible). These two topographical markers are not further aligned
with the two elements of Doxa. (b) On the assumption that the two
adjectives are aligned with the two elements, the goddess conveys
the soul from our own relatively Hotter and Brighter region of the
universe (the visible) to a Colder and Darker afterlife (the invisible).
(c) The goddess conveys the soul from its own Hotter and Brighter
habitat (the visible) to our Colder and Darker region of the universe
(the invisible). (c) strongly suggests, and (a) allows, the widespread
conception of the aether as the prenatal source and post-mortem
destination of the aethereal soul. (b), according to which the post-
mortem, Hot and aethereal soul inhabits rather a region of Cold
Night, would constitute an idiosyncratic and diametrical divergence
from this conception, but should not for that reason be excluded in
5
See Karsten (1835) 272–4; Deichgräber (1959) 716.
6
Coxon (2009) 276 advances this same interpretation of τὸ ἀειδές as our darker vicinity on
the controversial premise that, in B1.9, ‘the House of Night’ refers to the ordinary,
mundane region which the kouros has left behind. A representation of our vicinity as
Hades is plausibly but controversially ascribed to Empedocles, see Rohde (1925) 403–4,
n.75; Dodds (1951) 153, with 174, n.114; Jaeger (1947) 148–9; Wright (1995), ad B118;
B120–1; cf. Schibli (1990) 121–2, n.38.

349
 Appendix: The Trajectory of the kouros’ Journey
a thinker as idiosyncratic and creative as Parmenides. Let us now
turn to the proem.

(i) katabasis
We must first address a point of syntax and punctuation
(B1.8b-10a): . . . ὅτε σπερχοίατο πέμπειν Ἡλιάδες κοῦραι,
προλιποῦσαι δώματα νυκτός εἰς φάος.7 Proponents of the katabasis
reading persuasively argue that it is far more plausible to read εἰς
φάος with προλιποῦσαι than with πέμπειν and, therefore, that we
should delete DK’s comma between νυκτός and εἰς. Parmenides
does not, then, say that the Daughters of the Sun conveyed him to
the Light, but rather: ‘when the maidens, the Daughters of the Sun,
made haste to convey me, after they had left behind the House of
Night for the Light’.8 The feminine participle προλιποῦσαι has
only the Heliades as its grammatical subject. Burkert infers that it
is they, and not the kouros, who initially leave behind the House of
Night. That the kouros should be initially placed in the House of
Night is, he argues, independently unlikely if Parmenides locates it
in the Underworld (a viable possibility).9 We may add that ‘light’
(φάος) is an extremely common marker for the ordinary world of
the living, precisely as contrasted with Underworldly, eschatolo-
gical locations, and often signifies (with εἰς, vel sim.) the destina-
tion of ascents (back) from Hades.10 Plausibly, then, only the
7
DK’s widely followed capitalisations, Νυκτός / εἰς φάος, manufacture an asymmetry
which would not have been present in any ancient text of B1. The term φάος (like ἤματος
in B1.11), I take it, at least strongly evokes the Doxastic element Light.
8
This is strongly suggested both by Parmenides’ word order (referring εἰς φάος back to
πέμπειν would generate a jarring hyperbaton, in particular since προλιποῦσαι . . . εἰς φάος
is immediately picked up by the participle ὠσάμεναι as the progression of a single train
of thought) and by lexical parallels for the use of προλιπεῖν (esp. in the aorist participle)
to express the trajectory from departure to arrival, cf. e.g. Mimn. fr. 12.3–4 W (ὠκεανὸν
προλιποῦσ᾽ οὐρανὸν εἰσαναβῇ); see Burkert (1969) 7; cf. Morrison (1955) 60; Mansfeld
(1964) 238; Furley (1973) 1–2 (listing further parallels); Kingsley (1999) 50–1; Palmer
(2009) 53.
9
Burkert, loc. cit.
10
World of living: Il. 18.61; Il. 19.103, 118 (φόωσδε for birth); Od. 4.540; Hes. Op. 155;
Aesch. Pers. 222; Aristoph. Av. 699; [Hippoc.] Vict. 1.4 (ἐξ Ἅιδου ἐς φάος . . . ἐκ τοῦ φάεος
ἐς Ἅιδην); see further LSJ sv. φάος, b. Ascents: Od. 11.223 (φόωσδε, of Odysseus, with
Od. 23.252); h.Hom. 2.335–8; Thgn. 711–12 (ἤλυθε . . . ἐς φάος ἠελίου, of Sisyphus; note
703: ἐξ Ἀίδεω . . . ἀνῆλθεν); Soph. Phil. 624–5 (κἀξ Ἅιδου θανὼν | πρὸς φῶς ἀνελθεῖν,
cf. 417); Eur. Alc. 457 (φάος ἐξ Ἀίδα), 1073–6 (ἐς φῶς . . . νερτέρων ἐκ δωμάτων), 1139

350
 Appendix: The Trajectory of the kouros’ Journey
Heliades are said to have left behind the House of Night for the
Light prior to conveying the kouros, whom it would be natural to
find initially positioned in the Light. Since, furthermore, the
dichotomic duality of Light and Night is prominent in the imagery
of the journey,11 it would seem plausible that, having left the
House of Night (δώματα νυκτός, B1.9) for the Light, the
Heliades convey the kouros back to that same House (ἱκάνων
ἡμέτερον δῶ, B1.25).
The journey’s topography clearly appropriates the scenery of
the Hesiodic Underworld (Chapter 5.3).12 Should it seem counter-
intuitive that the Daughters of the Sun are themselves initially
placed in the Underworld or in the House of Night, Greek tradition
affords clear parallels. Stesichorus writes that, in the evening,
Helios returns to the depth of night (ποτὶ βένθεα νυκτός) where
his wife, mother and children live (PMG 185).13 A similar picture
seems to be implied in Hesiod, not only since he has Day reside in
the House of Night whenever the latter ascends, but also given his
statement that, specifically on Sleep and Death, Helios does not
shine in the Underworld even when he descends from heaven (οὐδ᾽
οὐρανόθεν καταβαίνων, 759–62). In the Odyssey, Hermes leads
souls to Hades by ‘Sun’s gates’ (Ἠελίοιο πύλας, Od. 24.12).
Advocates of this trajectory will also rightly point out that the
Underworldly katabasis (especially that katabasis at the conclu-
sion of which the traveller meets and converses with a(n unnamed)
goddess) was a familiar model in contemporary South Italian
initiatory and eschatological journeys.14

(νέρθεν ἐς φάος τόδε); Aristoph. Ran. 1528–9 (of Aeschylus); cf. also Hes. Th. 157–8 (of
the offspring of Ouranos, kept deep within Gaia and not allowed ἐς φάος; cf. the Giants’
release from murky Erebus ἐς φάος: 626, 652–3, 669); Aesch. Pers. 630 (of necromancy:
πέμψατ’ ἔνερθεν ψυχὴν ἐς φῶς). For this topographical point, cf. Gemelli Marciano
(2013) 74.
11
πύλαι νυκτός τε καὶ ἤματός . . . κελεύθων (B1.11); Ἡλιάδες . . . νυκτός | εἰς φάος (B1.9–10);
cf. the images of alternating duality in δοιοῖς . . . δινωτοῖσιν | κύκλοις ἀμφοτέρωθεν (B1.
7–8); ἀμοιβούς (B1.14) and ἀμοιβαδόν (B1.19).
12
For the descent into Hades, note κατέβην δόμον Ἄϊδος εἴσω, Od. 23.252; cf. Il. 7.330; Il.
14.457; also Il. 8.13–16; Hes. Th. 720–5, 740–3; 750 (ἔσω καταβήσεται). Already in
Homer, then, one formulaically journeys downwards into Hades; contrast Burkert
(1969) 15; Primavesi (2013) 41–2; also Ganschinietz (1919) 2409.
13
Cf. Morrison (1955) 60; Burkert (1969) 9; Furley (1973) 4.
14
Burkert (1969) 17, 21–6; Kingsley (1999) passim; cf. Palmer (2009) 57–8; Morrison
(1955) 60.

351
 Appendix: The Trajectory of the kouros’ Journey
The katabasis interpretation is compatible with the reconstruction
of Parmenides’ eschatological ideas about the soul as established in
Chapter 5.2. We initially find the Daughters of the Sun, who empha-
tically connote Light, Heat and the higher, celestial regions, emerging
from the House of Night to the Light. Furthermore, Sun or Day
traditionally end each of their cyclical journeys in the House of
Night before ascending again. So, why should the Hot, fiery, aether-
eal soul not end each of its own life cycles in that same location of
Night, before ascending back to the Light of life (εἰς φάος)?
The katabasis interpretation of the trajectory of the journey in
B1 – interpretation (i) – will cohere with interpretations (a) and (b)
of Simplicius’ report.

(ii) anabasis
Although Parmenides undeniably draws on Hesiod’s Underworld
in his account of the journey’s scenery, he does not do so slavishly.
Most strikingly, Hesiod’s account offers no counterpart to
Parmenides’ description of the gates as aethereal (αἰθέριαι,
B1.13). Elsewhere in Parmenides, as we saw (Chapter 5.2), this
terminology designates a higher celestial region in which the
heavenly bodies are located and which is closely associated with
the element Hot. Burkert argues that, although by describing his
gates as ‘aethereal’ Parmenides locates them in the sky, his remark
that the threshold is made of stone (λάινος οὐδός, B1.12) – also
a departure from the Hesiodic bronze threshold (χάλκεος οὐδός,
Th. 811, cf. 750; Il. 8.15) – locates it down on the earth. The gates,
therefore, loom into the high aether but are based down below,
encompassing the domains of both earth and sky. Burkert finds
a parallel for this interpretation in the Hesiodic notion that Sky,
Earth, Sea and Tartarus all have their ‘sources and limits’ in the
same Underworldly location.15 Aeschylus offers a parallel use of

15
πηγαὶ καὶ πείρατ, Th. 736–8, 807–9; cf. ῥίζαι, 727–8; Burkert (1969) 11–12. Pellikaan-
Engel (1978) 57 adds that, in Th. 746 (cf. 518), Atlas is positioned in the Underworld
and holds the broad heavens. Burkert is followed by Pellikaan-Engel (1978) 57–8;
Owens (1979) 18; Furley (1973) 4; Gemelli Marciano (2013) 74–5 n.97. Morrison
(1955) 59 tenuously renders αἰθέριαι as ‘lofty’. The problem is not even mentioned in the
katabasis interpretations of Kingsley (1999) and Palmer (2009).

352
 Appendix: The Trajectory of the kouros’ Journey
the adjective to designate something (a cloud of dust), which
begins on the ground and reaches up to the heavens (αἰθερία
κόνις, Th. 81).16 Parmenides, furthermore, does closely juxtapose
his reference to the stone threshold and his (contrasting?) descrip-
tion of the gates as aethereal (καί σφας ὑπέρθυρον ἀμφὶς ἔχει καὶ
λάινος οὐδός· | αὐταὶ δ’ αἰθέριαι κτλ, B1.12–13).
Burkert’s ingenious hypothesis seems a possible solution, then,
but it could hardly be a compelling one. It is not clear why the fact
that Parmenides’ threshold comprises stone (rather than bronze)
should settle the question of its cosmic whereabouts. The gates
themselves possess a ‘bolted bar’ (βαλανωτὸν ὀχῆα, B1.16) as
well as hinges or pivots which are rich with bronze and fitted in
sockets with pegs and nails (πολυχάλκους | ἄξονας, B1.18–20), but,
nonetheless, the gates, minimally, reach up to the aether. Indeed, it
seems as likely as not that ‘stone’ is to be understood not only with
the threshold but also with the lintel (ὑπέρθυρον, B1.12), which
holds the gates from above.17 Furthermore, the stress in the juxta-
position of B1.12 with B1.13 could equally inhere in the contrast,
not between the stony composition of the threshold and the aether-
eal location of the doors, but simply between the statements (i) that
the doors are encompassed by a lintel and threshold (σφας
ὑπέρθυρον ἀμφὶς ἔχει, B1.12) and (ii) that they themselves fit into
a great architrave (αὐταὶ δ’ αἰθέριαι πλῆνται μεγάλοισι θυρέτροις,
B1.13). In this case, the adjective ‘aethereal’ itself would be
irrelevant for this contrast and constitute the only topographical
marker in B1.12–13. In sum, we cannot exclude the view that, in
B1.13, Parmenides locates in the aether both the gates through
which he travels, along with (presumably) the architrave into
which they fit, and his journey itself.18 This interpretation of the
description of the gates as ‘aethereal’ would correspond more
easily with Parmenides’ own use of this vocabulary elsewhere
than Burkert’s admittedly viable construal ‘reaching from below
up to the aether’.
The formula that expresses the passage of the kouros and kourai
beyond the gates (τῇ ῥα δι’ αὐτέων . . . ἔχον . . . ἵππους, B1.20–1)

16
Pellikaan-Engel (1978) 57. 17 As suggested by Coxon (2009) 276.
18
Cf. e.g. Bowra (1953c) 43; Dolin (1962) 96; Kahn (2009) 212–13.

353
 Appendix: The Trajectory of the kouros’ Journey
occurs only twice in epic. In both cases, it expresses the passage of
Hera and Athena through the gates of heaven.19 In both of these
Homeric passages, moreover, the Horae, among whom Hesiod
numbers Dikê (Th. 902–3), are charged with opening and closing
the gates of heaven. In Parmenides, ‘much-punishing Dikê’ holds
the keys of the gates and exercises a similar function
(B1.14–17).20 The anabasis interpretation, furthermore, is per-
fectly consistent with the syntax of B1.8b-10a as analysed above
(‘. . . made haste to convey me, after they had left behind the House
of Night for the Light’). Kahn reasonably insists that ‘for the
Light’ remains the only marker of a trajectory and that there is
no indication (I would qualify: no explicit indication) of a return
journey downwards or into Darkness as opposed to a continuous
progression into the Light.21 In B1.10, we learn that the Daughters
of the Sun journeyed into the Light (ἐς φάος). In the immediately
following line, in the same initial metrical position, it is said that
there (ἔνθα) are the gates of the paths of Night and Day through
which Parmenides travels. It seems eminently possible, therefore,
that the gates are placed in the destination of the Daughters’ initial
trajectory, i.e. in the Light.22 This is again supported by their later
description as ‘aethereal’.
Although the participle προλιποῦσαι (B1.9) indeed has only the
Daughters of the Sun as its grammatical subject, it certainly
implicitly includes also the horses and the chariot, which the
Daughters of the Sun drive as the charioteers (B1.5, 24).23
The party that leaves the House of Night to the Light, then,
could conceivably include implicitly also the kouros. If it does,

19
πύλαι . . . οὐρανοῦ, Il. 5.748–52 = Il. 8.392–6; cf. Coxon (2009) 10.
20
Coxon (2009) 277; Mourelatos (2008a) 14; cf. Burkert (1969) 11, n.2.
21
Kahn (2009) 211, 213, n.8.
22
The following words (νυκτός . . . κελεύθων), over and above evoking Hesiod’s
Underworld, constitute a word for word quotation from Homer’s apparent account of
the (near?) absence of darkness in Laestrygonia, Od. 10.82–6; cf. Coxon (2009) 275.
23
Contrast Burkert (1969) 7. Primavesi (2013) takes it that B1.1–5a describes a journey
which the kouros undertakes independently, while the guidance of the Heliades comes
into play only for a second leg of the journey, from line 5b onwards. But the text gives no
indication of such an abrupt shift. The words κοῦραι δ’ ὁδὸν ἡγεμόνευον (B1.5b)
complement the immediately preceding remarks about the agency of the mares and,
looking backwards to the previous lines as well as forwards, climactically identify the
maidens as the kouros’ guides and as (throughout the journey) the charioteers (n.b.
ἡνιόχοισιν, B1.24).

354
 Appendix: The Trajectory of the kouros’ Journey
then Parmenides subversively represents our vicinity (from which,
on the current interpretation under consideration, he was conveyed
to the Light) as the true Hades (a possibility which we indepen-
dently raised above). If, conversely, the House of Night retains its
traditional place in the Underworld, then it indeed could not be the
kouros’ starting-point (as Burkert argues), but then it would also
not have to be his destination. The main internal argument for the
katabasis reading rests on identifying the House of Night from
which the Daughters of the Sun initially depart (δώματα νυκτός,
B1.9) with the House to which the goddess welcomes the kouros
(ἡμέτερον δῶ, B1.25). That it must forgo this identification is
a weakness of the anabasis interpretation, but it is not a fatal one
in the light of the careful symmetrical tension in B1 between Night
and Day.
Parmenides speaks symmetrically of the ‘gates of the paths of
Night and Day’ (νυκτός τε καὶ ἤματος, B1.11). The ‘House of
Night’ (δώματα νυκτός, B1.9) signifies Night’s residence, which
the Daughters of the Sun leave behind when they depart to the
Light. We should not, then, hastily assume that the unspecified
House to which the kouros arrives through aethereal gates is the
same House of Night and not rather the House of its
counterpart, Day.24 Such a general progression in the kouros’
trajectory from the House of Night towards the House of Day
would cohere with the later characterisation of Night as ‘ignorant’
(νύκτ’ ἀδαῆ, B8.59) and with the suggestion that the knowing
man, who journeys to the goddess with or as his fiery and aethereal
soul, is a knowing Light (B1.3, with φώς punning on φῶς: see
further Chapter 5.3). It should be emphasised that none of the
possibilities outlined here require us to identify the narrating god-
dess herself as either Night or Day.25
24
Roloff (1970) 172, n.96 identifies it as the house of Day; cf. Diels (2003) 51.
25
The goddess cannot be Night, pace Morrison (1955) 60; Palmer (2009) 589; Primavesi
(2013) 64. The axiological nature of the contrast Light/Night (see Ch. 4.2, Ch. 5.2)
precludes this. Indeed, in B8.59 the goddess would be calling herself ‘ignorant’.
The Hesiodic House of Night in fact accommodates both Night and Day, although
never simultaneously (Th. 748–54). So, even if the kouros arrives at the House of Night,
this fact alone does not imply that he is welcomed by Night rather than by Day. Diels
(2003) 51, conversely, argues that αἰθέριαι (B1.13) identifies the location as ‘die
Wohnung der lichten Tagesgöttin’. But nothing in our direct or indirect evidence for
the goddess, or for the two elements, suggests an enantiomorphic duality of two

355
 Appendix: The Trajectory of the kouros’ Journey
We saw in Chapter 5.3 that the Daughters of the Sun evoke the
myth of Phaethon’s celestial journey, and that Parmenides, cast as
a new and successful Phaethon, travels in the chariot and in the
manner of the Sun. These associations favour, although they do
not force, the anabasis interpretation.
The anabasis interpretation is often accused of anachronism.26
But the idea that the post-mortem soul ascends to the aether was
current. This trajectory, moreover, is explicable within the frame-
work of Parmenides’ psychology and astronomy, according to
which the divine soul comprises aethereal fire (= Light) and this
same element predominates in the higher, celestial region of aether
(Chapter 5.2). On the anabasis trajectory, the path of the post-
mortem Hot soul, which the kouros is privileged to pursue while
still alive (B1.26–8), is that of an ascent to the aether.
The anabasis interpretation of the kouros’ journey – interpreta-
tion (ii) – will fit with interpretations (a) and (c) of Simplicius’
report.

(iii) anabasis – katabasis

We saw in Chapter 5.3 that the ‘way of the daimôn’ (B1.3), which
the kouros travels, is most likely the way of the Sun. At any rate,
the kouros certainly travels in the manner of the Sun, guided by the
Daughters of the Sun in what closely recalls the chariot and horses
of the Sun. The kouros, moreover, is journeying on the path of
Night and Day (B1.11). The way of the Sun, as indeed that of
Night and Day, might be most appositely represented, neither as an
anabasis nor as a katabasis, but as an anabasis followed by
a katabasis.27
goddesses, Night and Day. The Houses of Day (if that is the reference of δῶ at B1.25)
and Night (on any reading), then, more probably gesture towards the higher and lower
regions which are associated with the elements Light and Night respectively, rather than
the abodes of two goddesses. Nor is the goddess Justice. Justice is the gate-keeper, and
admits the kouros into his interview with the goddess (B1.14–25); cf. Burkert (1969) 13;
Furley (1973) 3, n.10. Parmenides’ goddess is left unnamed.
26
E.g. Burkert (1969) 3, 15; Palmer (2009) 52–3; cf. Owens (1979) 17, n.8.
27
Representing the journey in terms of the traditional concepts of the daily ‘ascent’ of the
Sun towards the heavens and its ‘descent’ from it is intelligible and legitimate from the
perspective of the mortal traveller himself, even if such terms would not be ultimately
adequate in a proper astronomical account of the sun within a spherical cosmological

356
 Appendix: The Trajectory of the kouros’ Journey
We may, therefore, entertain the following trajectory. (α) First,
the Daughters of the Sun leave the House of Night towards the
Light in order to find the kouros and guide him on the same course
which their father travels each day, conveying him in that same
upwards direction until they reach the gates of Night and Day
which are located in the aether (B1.6–13). (β) Thereupon, they
persuade Justice to open the gates of heaven (B1.14–20). (γ) They
now proceed through these gates and follow the Sun’s daily path
back down towards the House of Night (B1.20–1) where (δ) the
kouros finally meets the goddess (B1.22–3). The aforementioned
formula which occurs in the description of the journey beyond the
gates, ‘through them they held the horses’ (B1.20–1), conveys in
both of its other two occurrences in epic the descent of goddesses
through and from the gates of heaven.28 The words ἰθύς . . . κατ’
ἀμαξιτόν (B1.21) could mean no more than ‘straight . . . along
the road’, but could alternatively amplify, in keeping with the
Homeric resonance, a downward motion: ‘straight . . . down
the road’ (cf. κατά, Od. 14.253–4). On this interpretation, the
Daughters of the Sun go full circle, as the chariot of the Sun
does when it follows the path of each day.
This anabasis-katabasis model explains why the proem so
clearly evokes the images of both katabasis (through, for example,
Hesiod’s Underworld and mystery terminology) and anabasis
(through, for example, the Homeric gates of heaven and the
Phaethon myth). Parmenides’ journey, in its different stages,
involves both an ascent and a descent. Like the anabasis reading,
it employs the less forced interpretation of αἰθέριαι (B1.13) as ‘in
the aether’. Like the katabasis reading, it construes the movement
‘into the Light’ (εἰς φάος, B1.10) in line with a very common
usage, as an ascent from an Underworldly location towards our

system. In the same vein, Aëtius does not contradict himself when he reports that, for
Parmenides, the aether encircles everything and is ‘highest of all’ (περιστάντος δ’
ἀνωτάτω πάντων τοῦ αἰθέρος, A37). For cogent arguments for the view that
Parmenides, who recognised that the moon gets its light from the sun (B14–15), thought
of the heavenly bodies as spherical shapes which orbit around the earth, see Graham
(2006) 179–82; (2013a) 87–96; Mourelatos (2011) 170–80.
28
In Il. 8.392–6, Hera and Athena head for the battleground below before Iris stays them
(at 409ff). In Il. 5.748ff, they descend to Troy after pausing to confer with Zeus on the
topmost peak of Olympus. Note Il. 5.769: their chariot now flies ‘in between earth and
starry heaven’.

357
 Appendix: The Trajectory of the kouros’ Journey
brighter world of the living. Furthermore, it retains the attractive
and economical identification of the House at which the kouros
arrives (δῶ, B1.25) with the House of Night which was mentioned
earlier (δώματα, B1.9). Note that Hesiod mentions only a House of
Night, which accommodates alternately both Night and Day (Th.
744–54). I suggest that, on balance, the anabasis-katabasis inter-
pretation is the most satisfactory reading.29
What construal of the soul’s post-mortem path does this trajec-
tory imply? This will depend on whether, when she refers to the
way of the post-mortem soul which the privileged kouros travels
while still alive (B1.26–8), the goddess has in mind only his final
descent (katabasis) into her House of Night or both the anabasis
and the katabasis legs of his journey. In the former case, the
anabasis-katabasis interpretation of the kouros’ journey – inter-
pretation (iii) – will (just like the katabasis interpretation) cohere
with interpretations (a) and (b) of Simplicius’ report. In the latter
case, the eschatological path of the soul will involve both an ascent
and a descent, mirroring the daily path of the Sun. If so, then the
anabasis-katabasis interpretation of the kouros’ journey will
cohere with interpretation (a) of Simplicius’ report. The visible
will be the area of the cosmos which we inhabit, and the invisible
will be the path which the soul traverses (the daily path of the Sun)
before it returns again to the visible. But the anabasis-katabasis
interpretation could also cohere with a combination of interpreta-
tions (b) and (c) of Simplicius’ report: the goddess cyclically
conveys souls from their habitat in the aether (ποτὲ µὲν ἐκ τοῦ
ἐµφανοῦς) into the dark Underworld (εἰς τὸ ἀειδές) and back again
(ποτὲ δὲ ἀνάπαλιν), while in between – after the souls emerged
from the Underworld but before they reach the aether – they
inhabit our world of the living as incarnated humans.
29
We cannot confidently emend the corruption †κατὰ πάντ’ ἄτη† (B1.3). For discussions
and conjectures, see Lesher (1994a) 6 with n.9; Palmer (2009) 376–8; cf. Mansfeld
(2008). κατὰ πάντ’ ἄ<σ>τη (Burkert (1969) 6, n.14; Lesher (1994a); cf. Palmer (2009)
56–7, n.19) remains attractive, especially if one identifies the daimôn of B1.3 as the Sun.
To avoid conflict with B1.27, it must mean ‘over’ (as in Il. 4.276, 278; cf. Guthrie (1965)
7, n.) not ‘through’ all cities. This would again suggest that, at least at some stage of his
journey, the kouros travels (like the Sun) high in the sky. κατὰ πάντ’ ἄ<ν>τη<ν> would
have to mean something like ‘through everything straight forth’ (as Coxon had it in his
first edition). It is difficult to motivate the translation ‘to meet her face to face’ (Coxon
(2009) 48, 271–2) from the immediate context.

358
 Appendix: The Trajectory of the kouros’ Journey
(iv) Not Kansas
Alternatively, the contrary pull of all the different topographical
markers in B1, and the different cultural models it evokes, may
lead us to agree with Mourelatos’ assertion that ‘the topography of
the journey is blurred beyond recognition’.30 It would seem mis-
placed to consider (iv) a less ‘charitable’ interpretation. We explored
several exegetical alternatives and it would be rash to presuppose that
these semantic and imagistic ambiguities are there to be settled.
The proem may rather be gesturing simultaneously towards divergent
trajectories.31 By mixing imagery and topographical markers that
suggest a katabasis with others that point towards an anabasis,
Parmenides may be signalling, for example, that eschatological topo-
graphy cannot be coherently reduced to the categories of higher and
lower and that his journey to the revelation of what-is is a journey
beyond all such frameworks. As Patricia Curd put it to me, the proem
may indicate only that ‘we’re not in Kansas anymore’.
Since interpretation (iv) is intrinsically incoherent, it could not
‘cohere’ with any interpretation of the implications of Simplicius’
report concerning the topography of the afterlife in Parmenides.
But it would perhaps coexist most easily with interpretation (a),
since this construal commits Parmenides to the least specific and
concrete eschatological topography.
The important conclusion to draw from these considerations is
a methodological one. It would be misguided to exclude from our
interpretation of the proem certain representations or associations
on the grounds that they do not align with whatever view of the
trajectory of the kouros’ journey we happen to favour. Any ana-
lysis of the significance and function of the proem which is not to
be reductively lopsided would have to recognise, for example, its
appropriation of the Hesiodic Underworld and of the language and
imagery of mystery initiations, as well as the close association of
the kouros with a divine milieu that is emblematic of light and the
celestial aether.

30
Mourelatos (2008a) 15–16. Johnson (1999) questions whether Hesiod’s account of the
Underworld in Th. 721–819 aims for topographical coherence.
31
Compare Heraclitus’ celebrated use of what Kahn (1979) 87–95 calls ‘meaningful
ambiguity’.

359
 BIBLIOGRAPHY

Adluri, V. (2011) Parmenides, Plato and Mortal Philosophy. London.

Adomenas, M. (1999) ‘Heraclitus on religion’, Phronesis 44: 87–113.
Aikin, S. (2014) ‘Xenophanes the high rationalist: the case of F1: 17–18’, Epoché
19: 1–14.
Aikin, S. (2016) ‘So what if horses would draw horse gods?’, Sophia 55:
163–177.
Algra, K. (1999) ‘The beginnings of cosmology’, in Long (ed.), 45–65.
Alt, K. (1973) ‘Zum Satz des Anaximenes Über die Seele, Untersuchung von
Aetius ΠΕΡΙ ΑΡΧΩΝ’, Hermes 101: 129–164.
Andriopoulos, D. Z. (1975) ‘Parmenides’ fragment B16 and his theory of per-
ceiving and knowing’, in Actes de la XIIe Conférence Internationale d’Études
Classiques Eirene. Amsterdam, 553–561.
Annas, J. (1999), ‘Becoming like god: ethics, human nature, and the divine’, in
her Platonic Ethics, Old and New. Ithaca, NY, 52–71.
Anton, J. P. and Kustas, G. L. (eds) (1971), Essays in Ancient Greek Philosophy,
vol. I. Albany, NY.
Anton, J. P. and Preus, A. (eds) (1983), Essays in Ancient Greek Philosophy,
vol. II. Albany, NY.
Armstrong, J. M. (2004) ‘After the ascent: Plato on becoming like god’, OSAP
26: 171–183.
Assmann, J. (2001) The Search for God in Ancient Egypt. Ithaca, NY.
Austin, J. L. (1962) Sense and Sensibilia. G. J. Warnock (ed.). Oxford.
Austin, S. (2011) ‘Existence and essence in Parmenides’, in Cordero (ed.), 1–8.
Barker, E. (2006) ‘Paging the oracle: interpretations, identity and performance in
Herodotus’ History’, Greece & Rome 53: 1–28.
Barnes, J. (1982) The Presocratic Philosophers. London.
Barney, R. (2012) ‘History and dialectic (Metaphysics Α 3, 983a24–984b8)’, in
Steel (ed.), 69–104.
Beall, E. F. (1991) ‘Hesiod’s Prometheus and development in myth’, JHI 52:
355–371.
Beard, M. (1986) ‘Cicero and divination: the formation of a Latin discourse’, JRS
76: 33–46.
Belfiore, E. (1985) ‘“Lies unlike the truth”: Plato on Hesiod, Theogony 27’, TAPA
115: 47–57.
Bell, C. M. (2002) ‘“The Chinese believe in spirits”: belief and believing in the
study of religion’, in N. Frankenberry (ed.), Radical Interpretation in Religion.
Cambridge, 100–117.

360
 Bibliography
Benitez, R. and Tarrant, H. (2015) ‘Philosophy’, in Eidinow and Kindt (eds),
211–222.
Benzi, N. (2016), ‘Noos and mortal enquiry in the poetry of Xenophanes and
Parmenides’, Methodos 16.
Bergren, A. L. T. (1983) ‘Language and the female in early Greek thought’,
Arethusa 16: 69–95.
Bernabé, A. (2013) ‘The commentary of the Derveni papyrus: the last of
Presocratic cosmogonies’, Littera Antiqua 7: 4–31.
Bernabé, A. and San Cristóbal, A. I. J (2008), Instructions for the Netherworld:
The Orphic Gold Tablets. Leiden.
Bernabé, A. and San Cristóbal, A. I. J. (2011), ‘Are the “Orphic” gold leaves
Orphic?’, in Edmonds (ed.), 68–101.
Betegh, G. (2004) The Derveni Papyrus: Cosmology, Theology and
Interpretation. Cambridge.
Betegh, G. (2006a) ‘Greek philosophy and religion’, in M. L. Gill and P. Pellegrin
(eds), A Companion to Ancient Philosophy. Malden, MA, 625–639.
Betegh, G. (2006b) ‘Eschatology and cosmology: models and problems’, in Sassi
(ed.), 27–50.
Betegh, G. (2007) ‘On the physical aspect of Heraclitus’ psychology’, Phronesis
52: 3–32.
Betegh, G. (2009) ‘The limits of the soul: Heraclitus B45 DK. Its text and
interpretation’, in E. Hülsz (ed.), Nuevos Ensayos Sobre Heráclito. Mexico
City, 391–414.
Betegh, G. (2010) ‘The transmission of ancient wisdom’, in L. P. Gerson,
The Cambridge History of Philosophy in Late Antiquity, vol. I. Cambridge,
25–38.
Betegh, G. (2012) ‘The next principle (Metaphysics Α 3–4, 984b8–985b22)’, in
Steel (ed.), 105–140.
Betegh, G. (2014) ‘Pythagoreans, Orphism and Greek religion’, in Huffman
(ed.), 149–166.
Betegh, G. (2016) ‘Colocation’, in T. Buchheim, D. Meissner and N. Wachsmann
(eds), ΣΩΜΑ. Körperkonzepte und körperliche Existenz in der antiken
Philosophie und Literatur. Hamburg, 393–421.
Blank, D. L. (1982) ‘Faith and persuasion in Parmenides’, ClAnt. 1: 167–177.
Bodnár, I. M. (1989) ‘Reason and revelation in Parmenides’, Doxa 17: 61–67.
den Boer, W. (1973) ‘Aspects of religion in Classical Greece’, HSPh 77: 1–21.
Bollack, J. (2006) Parménide: de l’étant au monde. Paris.
Bollack, J. (2011) ‘From being to the world and vice versa’, in Cordero
(ed.), 9–20.
Bolton, J. D. P. (1962) Aristeas of Proconnesus. Oxford.
Bonnechere, P. (2003) ‘Trophonius of Lebadea: mystery aspects of an oracular
cult in Boeotia’, in Cosmopoulos (ed.), 169–192.
Bormann, K. (1979) ‘The interpretation of Parmenides by the Neoplatonist
Simplicius’, The Monist 62: 30–43.

361
 Bibliography
Bouché-Leclercq, A. (1879–1882) Histoire de la divination dans l’antiquité, 4
vols. Paris.
Boudouris, K. J. (ed.) (1989), Ionian Philosophy. Athens.
Bowden, H. (2010) Mystery Cults in the Ancient World. London.
Bowden, H. (2015) ‘Nameless gods: mystery cults, myth, ritual and Greek polis
religion’, Kodai 16: 31–42.
Bowie, E. L. (1993) ‘Fiction and slander in early Greek poetry’, in C. Gill and
T. P. Wiseman (eds), Lies and Fiction in the Ancient World. Exeter, 1–37.
Bowra, C. M. (1953) Problems in Greek Poetry. Oxford.
Bowra, C. M. (1953a) ‘Xenophanes on songs at feasts’, in Bowra (1953), 1–14.
Bowra, C. M. (1953b) ‘Xenophanes and the Olympian Games’, in Bowra (1953),
15–37.
Bowra, C. M. (1953c) ‘The proem of Parmenides’, in Bowra (1953), 38–53.
Boys-Stones, G. R. (2009) ‘Ancient philosophy of religion: an introduction’, in
Oppy and Trakakis (eds), 1–22.
Boys-Stones, G. R. (2010) ‘Hesiod and Plato’s history of philosophy’, in Boys-
Stones and Haubold (eds), 31–51.
Boys-Stones, G. R. and Haubold, J. H. (eds) (2010), Plato and Hesiod. Oxford.
Bredlow, L. A. (2011) ‘Aristotle, Theophrastus, and Parmenides’ theory of
cognition (B 16)’, Apeiron 44: 219–263.
Bremmer, J. N. (1983) The Early Greek Concept of the Soul. Princeton, NJ.
Bremmer, J. N. (1994) Greek Religion. Oxford.
Bremmer, J. N. (1999) ‘Rationalization and disenchantment in ancient
Greece: Max Weber among the Pythagoreans and Orphics?’, in Buxton
(ed.), 71–83.
Bremmer, J. N. (2007) ‘Atheism in antiquity’, in M. Martin (ed.), The Cambridge
Companion to Atheism. Cambridge, 11–27.
Bremmer, J. N. (2008) ‘Balaam, Mopsus and Melampus: tales of travelling
seers’, in his Greek Religion and Culture, the Bible and the Ancient Near
East. Leiden, 133–151.
Bremmer, J. N. (2014) Initiations into the Mysteries of the Ancient World. Berlin.
Bremmer, J. N. and Erskine, A. (eds) (2010), The Gods of Ancient Greece:
Identities and Transformations. Edinburgh.
Broadie, S. (1999) ‘Rational theology’, in Long (ed.), 205–224.
Brown, L. (1986) ‘Being in the Sophist: a syntactical enquiry’, OSAP 4: 49–70.
Brown, L. (1994) ‘The verb “to be” in Greek philosophy: some remarks’, in
S. Everson (ed.), Companions to Ancient Thought, vol. III: Language.
Cambridge, 212–236.
Brown, P. (1971) ‘The rise and function of the holy man in late antiquity’, JRS 61:
80–101.
Bryan, J. (2012) Likeness and Likelihood in the Presocratics and Plato.
Cambridge.
Burkert, W. (1969) ‘Das Proömium des Parmenides und die Katabasis des
Pythagoras’, Phronesis 14: 1–30.

362
 Bibliography
Burkert, W. (1972) Lore and Science in Ancient Pythagoreanism. E. L. Minar Jr.
(trans.). Cambridge, MA.
Burkert, W. (1983) Homo Necans: The Anthropology of Ancient Greek Sacrificial
Ritual and Myth. P. Bing (trans.). Berkeley, CA.
Burkert, W. (1985) Greek Religion. J. Raffan (trans.). Oxford.
Burkert, W. (1987) Ancient Mystery Cults. Cambridge, MA.
Burkert, W. (1992) The Orientalizing Revolution: Near Eastern Influence on
Greek Culture in the Early Archaic Age. Cambridge, MA.
Burkert, W. (1999) ‘The logic of cosmogony’, in Buxton (ed.), 87–106.
Burkert, W. (2000) ‘Revealing nature amidst multiple cultures. A discourse with
ancient Greeks’, The Tanner Lectures on Human Values 21: 125–151.
Burkert, W. (2004) Babylon, Memphis, Persepolis: Eastern Contexts of Greek
Culture. Cambridge, MA.
Burkert, W. (2005) ‘Signs, commands and knowledge: ancient divination
between enigma and epiphany’, in Johnston and Struck (eds), 29–49.
Burkert, W. (2008) ‘Prehistory of Presocratic philosophy in an Orientalizing
context’, in Curd and Graham (eds), 55–85.
Burnet, J. (1930) Early Greek Philosophy, 4th edn, London.
Buxton, R. (ed.) (1999) From Myth to Reason?: Studies in the Development of
Greek Thought. Oxford.
Buxton, R. (1999) ‘Introduction’, in Buxton (ed.), 1–21.
Buxton, R. (ed.) (2000), Oxford Readings in Greek Religion. Oxford.
Canevaro, L. G. (2015) Hesiod’s Works and Days: How to Teach Self-Sufficiency.
Oxford.
Casertano, G. (2011) ‘Parmenides: scholar of nature’, in Cordero (ed.), 21–58.
Caston, V. and Graham, D. W. (eds) (2002), Presocratic Philosophy: Essays in
Honour of Alexander Mourelatos. Aldershot.
Cerri, G. (2011) ‘The astronomical section in Parmenides’ poem’, in Cordero
(ed.), 81–94.
Chadwick, N. K. (1942) Poetry and Prophecy. Cambridge.
Chalmers, W. R. (1960) ‘Parmenides and the beliefs of mortals’, Phronesis
5: 5–22.
Cherubin, R. (2005) ‘Light, Night, and the opinions of mortals: Parmenides B8.
51–61 and B9’, Ancient Philosophy 25: 1–23.
Christopoulos, M., Karakantza, E. D., and Levaniouk, O. (eds) (2010), Light and
Darkness in Ancient Greek Myth and Religion. Lanham, MD.
Clark, R. J. (1969) ‘Parmenides and sense-perception’, Revue des Études
Grecques 82: 14–32.
Classen, C. J. (1989) ‘Xenophanes and the tradition of epic poetry’, in Boudouris
(ed.), 91–103.
Classen, C. J. (1996) ‘ΑΡΧΗ – its earliest use’, Scripta Classica Israelica 15:
20–24.
Clay, D. (2004) Archilochos Heros: The Cult of Poets in the Greek Polis.
Washington, DC.

363
 Bibliography
Clay, J. S. (1974) ‘Demas and aude: the nature of divine transformation in
Homer’, Hermes 102: 129–136.
Clay, J. S. (1984) ‘The Hecate of the Theogony’, GRBS 25: 27–38.
Clay, J. S. (1988) ‘What the Muses sang: Theogony 1–115’, GRBS 29: 323–333.
Clay, J. S. (2003) Hesiod’s Cosmos. Cambridge.
Clay, J. S. (2015) ‘Commencing cosmogony and the rhetoric of poetic authority’,
in P. Derron (ed.), Cosmologies et cosmogonies dans la littérature antique:
huit exposés suivis de discussions et d’un épilogue. Entretiens sur l’Antiquité
classique, 61. Vandœuvres, 105–147.
Cleve, F. M. (1965) The Giants of Pre-Sophistic Greek Philosophy: An Attempt to
Reconstruct their Thoughts, 2 vols. The Hague.
Clinton, K. (2003) ‘Stages of initiation in the Eleusinian and Samothracian
mysteries’, in Cosmopoulos (ed.), 50–78.
Cohen, B. (ed.) (1995), The Distaff Side. New York, NY.
Cole, T. (1983) ‘Archaic truth’, QUCC 13: 7–28.
Collard, C., Cropp, M. J. and Gibert, J. (2004) Euripides: Selected Fragmentary
Plays, vol. II. Oxford.
Compton, T. M. (2006) Victim of the Muses: Poet as Scapegoat, Warrior and Hero
in Greco-Roman and Indo-European Myth and History. Washington, DC.
Connor, W. R. (1988) ‘Seized by the nymphs: nympholepsy and symbolic
expression in Classical Greece’, ClAnt. 7: 155–189.
Coo, L. (2016) ‘Sight and blindness’, in M. Squire (ed.), Sight and the Ancient
Senses. London, 237–248.
Cook, A. B. (1914–1940) Zeus: A Study in Ancient Religion, 3 vols. Cambridge.
Cordero, N. -L. (ed.) (2011a), Parmenides Venerable and Awesome. Las Vegas, NV.
Cordero, N. -L. (2011b) ‘Parmenidean “Physics” is not part of what Parmenides
calls δόξα’, in Cordero (ed.), 95–113.
Cornford, F. M. (1933) ‘Parmenides’ two ways’, CQ 27: 97–111.
Cornford, F. M. (1935) ‘A new fragment of Parmenides’, CR 49: 122–123.
Cornford, F. M. (1939) Plato and Parmenides. London.
Cornford, F. M. (1950) The Unwritten Philosophy and Other Essays. Cambridge.
Cornford, F. M. (1952) Principium Sapientiae: The Origins of Greek
Philosophical Thought. Cambridge.
Cornford, F. M. (1974) ‘Mysticism and science in the Pythagorean tradition’, in
Mourelatos (ed.), 135–160.
Cosgrove, M. R. (1974) ‘The KOUROS motif in Parmenides: B1.24’, Phronesis
19: 81–94.
Cosgrove, M. R. (2011) ‘The unknown “knowing man”: Parmenides, B1.3’, CQ
61: 28–47.
Cosgrove, M. R. (2014) ‘What are “true” doxai worth to Parmenides? Essaying
a fresh look at his cosmology’, OSAP 46: 1–31.
Cosmopoulos, M. B. (ed.) (2003) Greek Mysteries: The Archaeology and Ritual
of Ancient Greek Secret Cults. London.
Coxon, A. H. (1968) ‘The text of Parmenides fr. 1. 3’, CQ 18: 69.

364
 Bibliography
Coxon, A. H. (2003) ‘Parmenides on thinking and being’, Mnemosyne ser. 4,
no. 56: 210–212.
Coxon, A. H. (2009) The Fragments of Parmenides: Revised and Expanded
Edition. R. McKirahan (ed. with new translations), M. Schofield (preface).
Las Vegas, NV.
Crystal, I. (2002) ‘The scope of thought in Parmenides’, CQ 52: 207–219.
Curd, P. K. (1998) The Legacy of Parmenides: Eleatic Monism and Later
Presocratic Thought. Princeton, NJ.
Curd, P. K. (2002) ‘The Presocratics as Philosophers’, in Laks and Louguet (eds),
115–138.
Curd, P. K. (2007) Anaxagoras of Clazomenae: Fragments and Testimonia,
a Text and Translation with Notes and Essays. Toronto.
Curd, P. K. (2010) ‘Thought and body in Heraclitus and Anaxagoras’, BACAP
25: 1–41.
Curd, P. K. (2011) ‘Thought and body in Parmenides’, in Cordero (ed.), 115–134.
Curd, P. K. (2013a) ‘The divine and the thinkable: toward an account of the
intelligible cosmos’, Rhizomata 1: 217–247.
Curd, P. K. (2013b) ‘Where are Love and Strife? Incorporeality in Empedocles’,
in McCoy (ed.), 113–138.
Curd, P. K. (2015) ‘Thinking, supposing, and physis in Parmenides’, Études
platoniciennes 12.
Curd, P. K. and Graham, D. W. (eds) (2008), The Oxford Handbook of
Presocratic Philosophy. Oxford.
Curd, P. K. and McKirahan, R. D. (eds) (2011), A Presocratics Reader, 2nd edn.
Indianapolis, IN.
Currie, B. (2002) ‘Euthymos of Locri: a case study in heroization in the Classical
period’, JHS 122: 24–44.
Daix, D. -A. (2006) ‘Réalités et vérités dans la Théogonie et Les Travaux et les
Jours d’Hésiode’, Mètis 4: 139–164.
Darcus, S. M. (1977) ‘Daimon parallels the holy phren in Empedocles’,
Phronesis 22: 175–190.
Darcus, S. M. (1978) ‘The phren of the noos in Xenophanes’ god’, Symbolae
Osloenses 53: 25–39.
Darcus-Sullivan, S. M. (1994) ‘The mind and heart of Zeus in Homer and the
Homeric Hymns’, ABG 37: 101–126.
Darcus-Sullivan, S. M. (1995) ‘The mind and heart of Zeus in the poetry of
Hesiod’, ABG 38: 34–47.
D’Avray, D. L. (2010) Rationalities in History: A Weberian Essay in
Comparison. Cambridge.
Dawkins, R. (2007) The God Delusion. London.
Deichgräber, K. (1959) ‘Parmenides’ Auffahrt zur Göttin des Rechts.
Untersuchungen zum Prooimion seines Lehrgedichts’, Akademie der
Wissenschaften und der Literatur in Mainz, Jahrg. 1958, no. 11. Wiesbaden,
629–724.

365
 Bibliography
Denyer, N. (1985) ‘The case against divination: an examination of Cicero’s De
Divinatione’, PCPS 31: 1–10.
Detienne, M. (1981) ‘Between beasts and gods’, in R. Gordon (ed.), Myth,
Religion and Society. Cambridge, 215–229.
Detienne, M. (1996) The Masters of Truth in Archaic Greece. J. Lloyd (trans.).
Cambridge, MA.
Detienne, M. and Vernant, J. -P. (1978) Cunning Intelligence in Greek Culture
and Society. J. Lloyd (trans.). Hassocks.
Diels, H. (1887) ‘Über die ältesten Philosophenschulen der Griechen’, in
Philosophische Aufsätze, E. Zeller gewidmet. Leipzig, 239–260.
Diels, H. (2003) Parmenides Lehrgedicht. Sankt Augustin.
Diels, H. and Kranz, W. (eds) (1951–1952), Die Fragmente der Vorsokratiker,
6th edn, 3 vols. Berlin.
Dilcher, R. (2006), ‘Parmenides on the place of mind’, in King (ed.), 31–48.
Diller, H. (1966) ‘Hesiod und die Anfänge der griechischen Philosophie’, in
E. Heitsch (ed.), Hesiod. Darmstadt, 688–707.
Dillery, J. (2005) ‘Chresmologues and manteis: independent diviners and the
problem of authority’, in Johnston and Struck (eds), 167–231.
Dillon, J. and Hershbell, J. P. (eds and trans) (1991), Iamblichus: On the
Pythagorean Way of Life. Atlanta, GA.
Dodds, E. R. (1951) The Greeks and the Irrational. Berkeley, CA.
Dodds, E. R. (1973) The Ancient Concept of Progress. Oxford.
Doherty, L. E. (1995) ‘Sirens, Muses, and female narrators in the Odyssey’, in
Cohen (ed.), 81–89.
Dolin, E. F. (1962) ‘Parmenides and Hesiod’, HSCP 66: 93–98.
Dover, K. J. (1988) ‘The freedom of the intellectual in Greek society’, in his
The Greeks and Their Legacy. Oxford, 135–158.
Drozdek, A. (2004) ‘Xenophanes’ Theology’, SIFC 4a ser. 2: 141–157.
Drozdek, A. (2007) Greek Philosophers as Theologians: The Divine Arche.
Aldershot.
Dueso, J. S. (2011) ‘Parmenides: logic and ontology’, in Cordero (ed.),
271–288.
Edmonds, J. (1931) Greek Elegy and Iambus, 2 vols. London.
Edmonds III, R. G. (ed.) (2011a) The ‘Orphic’ Gold Tablets and Greek Religion:
Further Along the Path. Cambridge.
Edmonds III, R. G. (2011b) ‘Sacred scripture or oracles for the dead? The semiotic
situation of the “Orphic” gold tablets’, in Edmonds (ed.), 257–270.
Edmonds III, R. G. (2013) Redefining Ancient Orphism: A Study in Greek
Religion. Cambridge.
Eidinow, E. and Kindt, J. (eds) (2015), The Oxford Handbook of Ancient Greek
Religion. Oxford.
Eidinow, E., Kindt, J. and Osborne, R. (eds) (2016), Theologies of Ancient Greek
Religion. Cambridge.

366
 Bibliography
Eidinow, E., Kindt, J., Osborne, R. and Tor, S. (2016) ‘Introduction: what might
we mean by the theologies of ancient Greek religion?’, in Eidinow, Kindt and
Osborne (eds), 1–11.
van der Eijk, P. J. (2005) Medicine and Philosophy in Classical Antiquity.
Cambridge.
Eisenstadt, M. (1974) ‘Xenophanes’ proposed reform of Greek religion’ Hermes
102: 142–150.
Emlyn-Jones, C. J. (1976) ‘Heraclitus and the identity of opposites’, Phronesis
21: 89–114.
Erler, M. (2002) ‘Epicurus as deus mortalis, homoiosis theoi and Epicurean self-
cultivation’, in D. Frede and A. Laks (eds), Traditions of Theology: Studies in
Hellenistic Theology, Its Background and Aftermath, Leiden, 159–182.
Evans, C. S. (2005) ‘Faith and revelation’, in W. J. Wainwright (ed.), The Oxford
Handbook of Philosophy of Religion. Oxford, 323–343.
Everson, S. (ed.) (1990), Companions to Ancient Thought, vol. I: Epistemology.
Cambridge.
Feeney, D. (1998) Literature and Religion at Rome. Cambridge.
Ferella, C. (2013) ‘Empedocle, il giorno spietato e l’esilio dagli dei: a proposito
di alcuni frammenti “demonologici”’, in R. di Donato (ed.), Origini
E Svolgimento Del Pensiero Greco: Studi per Jean-Pierre Vernant. Pisa,
29–47.
Ferrari, G. (1988) ‘Hesiod’s mimetic Muses and the strategies of deconstruction’,
in A. Benjamin (ed.), Post-Structuralist Classics. London, 45–78.
Feyerabend, B. (1984) ‘Zur Wegmetaphorik beim Goldblättchen aus Hipponion
und dem Proömium des Parmenides’, Rhein. Mus. 127: 1–22.
Finkelberg, A. (1986) ‘“Like by like” and two reflections of reality in
Parmenides’, Hermes 114: 405–412.
Finkelberg, A. (1997) ‘Xenophanes’ physics, Parmenides’ Doxa and
Empedocles’ theory of cosmogonical mixture’, Hermes 125: 1–16.
Flacelière, R. (1976) Greek Oracles. D. Garman (trans.). London.
Flower, M. A. (2008) The Seer in Ancient Greece. Berkeley, CA.
Floyd, E. D. (1992) ‘Why Parmenides wrote in verse’, Ancient Philosophy 12:
251–265.
Fontenrose, J. (1978) The Delphic Oracle: Its Responses and Operations with
a Catalogue of Responses. Berkeley, CA.
Fowler, R. L. (2010) ‘Gods in early Greek historiography’, in Bremmer and
Erskine (eds), 318–334.
Fowler, R. L. (2011) ‘Mythos and logos’, JHS 131: 45–66.
Fränkel, H. (1925) ‘Xenophanesstudien’, Hermes 60: 174–192.
Fränkel, H. (1974) ‘Xenophanes’ empiricism and his critique of knowledge
(B34)’, M. R. Cosgrove (trans.), in Mourelatos (ed.), 118–131.
Fränkel, H. (1975a) ‘Studies in Parmenides’, D. J. Furley and R. E. Allen (trans),
in Furley and Allen (eds), 1–47.

367
 Bibliography
Fränkel, H. (1975b) Early Greek Poetry and Philosophy. M. Hadas and J. Willis
(trans). Oxford.
Frede, D. and Reis, B. (eds) (2009), Body and Soul in Ancient Philosophy. Berlin.
Freeman, K. (1946) The Pre-Socratic Philosophers. Oxford.
Frère, J. (2011) ‘Mortals (βροτοί) according to Parmenides’, in Cordero (ed.),
135–146.
von Fritz, K. (1974) ‘Nous, noein, and their derivatives in Pre-Socratic philoso-
phy (excluding Anaxagoras)’, in Mourelatos (ed.), 23–85.
Furley, D. J. (1973) ‘Notes on Parmenides’, in E. N. Lee, A. P. D. Mourelatos and
R. M. Rorty (eds), Exegesis and Argument: Studies in Greek Philosophy
Presented to Gregory Vlastos. Assen, 1–15.
Furley, D. J. and Allen, R. E. (eds) (1970), Studies in Presocratic Philosophy, vol.
I: The Beginnings of Philosophy. London.
Furley, D. J. and Allen, R. E. (eds) (1975), Studies in Presocratic Philosophy, vol.
II: The Eleatics and the Pluralists. London.
Furth, M. (1974) ‘Elements of Eleatic ontology’, in Mourelatos (ed.), 241–270.
Gagarin, M. (1990) ‘The ambiguity of Eris in the Works and Days’, in M. Griffith
and D. J. Mastronarde (eds), Cabinet of the Muses: Essays on Classical and
Comparative Literature in Honor of Thomas G. Rosenmeyer. Atlanta, GA,
173–185.
Gagarin, M. (2002) Antiphon the Athenian: Oratory, Law, and Justice in the Age
of the Sophists. Austin, TX.
Gagné, R. (2007) ‘Winds and ancestors: the Physika of Orpheus’, HSPh 103: 1–23.
Gagné, R. (2009) ‘Mystery inquisitors: performance, authority and sacrilege at
Eleusis’, ClAnt 28: 211–247.
Gagné, R. (2013) Ancestral Fault in Ancient Greece. Cambridge.
Gallop, D. (1979) ‘“Is” or “is not”’, The Monist 62: 60–80.
Gallop, D. (1984) Parmenides of Elea. Toronto.
Ganschinietz, R. (1919) sv. ‘Katabasis’, in Pauly-Wissowa, Realencyclopädie
der classischen Altertumswissenschaft 10.2: 2359–2449.
Geertz, C. (1973) ‘Religion as a cultural system’, in his The Interpretation of
Cultures. New York, NY, 87–125.
Gemelli Marciano, M. L. (2002) ‘Le contexte culturel des Présocratiques: adver-
saires et destinataires’, in Laks and Louguet (eds), 83–114.
Gemelli Marciano, M. L. (2008) ‘Images and experience: at the roots of
Parmenides’ Aletheia’, Ancient Philosophy 28: 21–48.
Gemelli Marciano, M. L. (2014) ‘The Pythagorean way of life and Pythagorean
ethics’, in Huffman (ed.), 131–148.
Gemelli Marciano, M. L., Rossetti, L. and Pulpito, M. (eds) (2013), Parmenide:
suoni, immagini, esperienza. Sankt Augustin.
Gernet, L. (1981) ‘The origins of Greek philosophy’, in his The Anthropology of
Ancient Greece. J. Hamilton and B. Nagy (trans). Baltimore, MD, 352–364.
Gerson, L. P. (1994) God and Greek Philosophy: Studies in the Early History of
Natural Theology. London.

368
 Bibliography
Gigon, O. (1945) Der Ursprung der Griechischen Philosophie, von Hesiod bis
Parmenides. Basel.
Giordano-Zecharya, M. (2005) ‘As Socrates shows, the Athenians did not believe
in gods’, Numen 52: 325–355.
Göbel, C. (2002) ‘Mysticism and knowledge in Parmenides’, Downside Review
120: 157–174.
Goldhill, S. (1986) ‘Blindness and insight’, in his Reading Greek Tragedy.
Cambridge, 199–221.
Goldhill, S. (1991) The Poet’s Voice. Cambridge.
Goldhill, S. and Osborne, R. (eds) (2006), Rethinking Revolutions through
Ancient Greece. Cambridge.
Gomperz, T. (1906) Greek Thinkers: A History of Ancient Philosophy, vol. I.
L. Magnus (trans.). London.
Gomperz, T. (1929) ‘Zur Theogonie des Pherekydes von Syros’, Wiener Studien
47: 14–26.
Goodwin, W. W. (1889) Syntax of the Moods and Tenses of the Greek Verb.
London.
Gould, J. (1985) ‘On making sense of Greek religion’, in P. E. Easterling and
J. V. Muir (eds), Greek Religion and Society. Cambridge, 1–33.
Goulet-Cazé, M. -O. (1996) ‘Religion and the early Cynics’, in B. Branham and
M. -O. Goulet-Cazé (eds), The Cynics: The Cynic Movement in Antiquity and
its Legacy. Berkeley, CA, 47–80.
Graf, F. (2005) ‘Rolling the dice for an answer’, in Johnston and Struck (eds), 51–97.
Graf, F. (2013a) ‘A history of scholarship on the Tablets’ in Graf and Johnston
(eds), 50–65.
Graf, F. (2013b) ‘Dionysiac mystery cults and the Gold Tablets’, in Graf and
Johnston (eds), 137–166.
Graf, F. and Johnston, S. I. (eds) (2013a), Ritual Texts for the Afterlife: Orpheus
and the Bacchic Gold Tablets, 2nd edn. London.
Graf, F. and Johnston, S. I. (2013b) ‘Orpheus, his poetry, and sacred texts’, in
Graf and Johnston (eds), 167–186.
Graham, D. W. (1999) ‘Empedocles and Anaxagoras: responses to Parmenides’,
in Long (ed.), 159–180.
Graham, D. W. (2006) Explaining the Cosmos: The Ionian Tradition of Scientific
Philosophy. Princeton, NJ.
Graham, D. W. (2010) The Texts of Early Greek Philosophy: The Complete
Fragments and Selected Testimonies of the Major Presocratics, vol. I.
Cambridge.
Graham, D. W. (2013a) Science Before Socrates: Parmenides, Anaxagoras and
the New Astronomy. Oxford.
Graham, D. W. (2013b) ‘The theology of nature in the Ionian tradition’,
Rhizomata 1: 194–216.
Granger, H. (2007) ‘Poetry and prose: Xenophanes of Colophon’, TAPhA 137:
403–433.

369
 Bibliography
Granger, H. (2008) ‘The proem of Parmenides’ poem’, Ancient Philosophy
28: 1–20.
Granger, H. (2010) ‘Parmenides of Elea: rationalist or dogmatist?’, Ancient
Philosophy 30: 15–38.
Granger, H. (2013a) ‘Xenophanes’ positive theology and his criticism of Greek
popular religion’, Ancient Philosophy 33: 235–271.
Granger, H. (2013b) ‘Early natural theology: the purification of the divine
nature’, in D. Sider and D. Obbink (eds), Doctrine and Doxography: Studies
on Heraclitus and Pythagoras. Berlin, 163–200.
Grau, S. (2013) ‘Diogenes Laertius between tradition and innovation: philoso-
phers and θεῖοι ἄνδρες’, in R. García-Gasco, S. Gonzáles Sánchez and
D. Hernández de la Fuente (eds), The Theodosian Age (A.D. 379–455):
Power, Place, Belief and Learning at the End of the Western Empire.
Oxford, 183–189.
Graver, M. (2009) ‘Cicero’, in Oppy and Trakakis (eds), 119–131.
Gregory, A. (2013) The Presocratics and the Supernatural: Magic, Philosophy
and Science in Early Greece. London.
Gregory, A. (2014) ‘Parmenides, cosmology and sufficient reason’, Apeiron 47:
16–47.
Grice, H. P. (1957) ‘Meaning’, Philosophical Review 66: 377–388.
Griffin, J. (1980) Homer on Life and Death. Oxford.
Griffith, M. (1983) ‘Personality in Hesiod’, ClAnt. 2: 37–65.
Griffith, R. D. (1993) ‘Oedipus pharmakos? Alleged scapegoating in Sophocles’
Oedipus the King’, Phoenix 47: 95–114.
Griffith, R. D. (2009) ‘Putting your mouth where your money is: Eumolpus’ will,
pasta e fagioli, and the fate of the soul in South Italian thought from Pythagoras
to Ennius’, in G. Casadio and P. A. Johnston (eds), Mystic Cults in Magna
Graecia. Austin, TX, 131–136.
Griffiths, A. (1999) ‘Euenius the negligent nightwatchman (Herodotus 9.
82–96)’, in Buxton (ed.), 169–182.
Grottanelli, C. (2005) ‘Sorte unica pro casibus pluribus enotata: literary texts and
lot inscriptions as sources for ancient kleromancy’, in Johnston and Struck
(eds), 129–146.
Guthrie, W. K. C. (1935) Orpheus and Greek Religion. London.
Guthrie, W. K. C. (1952) ‘The Presocratic world-picture’, HThR 45: 87–104.
Guthrie, W. K. C. (1962) A History of Greek Philosophy, vol. I: The Earlier
Presocratics and the Pythagoreans. Cambridge.
Guthrie, W. K. C. (1965) A History of Greek Philosophy, vol. II: The Presocratic
Tradition from Parmenides to Democritus. Cambridge.
Guthrie, W. K. C. (1970) ‘Aristotle as historian’, in Furley and Allen (eds),
239–254.
Hack, R. K. (1931) God in Greek Philosophy to the Time of Socrates.
Princeton, NJ.
Hadot, P. (1995) Philosophy as a Way of Life. M. Chase (trans.). Oxford.

370
 Bibliography
Hainsworth, B. (1993) The Iliad: A Commentary, vol. III: Books 9–12. Cambridge.
Halliday, W. R. (1913) Greek Divination: A Study of Its Methods and Principles.
London.
Halliwell, S. (2011) Between Ecstasy and Truth: Interpretations of Greek Poetics
from Homer to Longinus. Oxford.
Harrison, T. (2000) Divinity and History: The Religion of Herodotus. Oxford.
Harrison, T. (2006) ‘Religion and the rationality of the Greek city’, in Goldhill
and Osborne (eds), 124–140.
Harrison, T. (2007) ‘Greek religion and literature’, in Ogden (ed.), 373–384.
Harrison, T. (2015a) ‘Beyond the polis? New approaches to Greek religion’, JHS
135: 165–180.
Harrison, T. (2015b) ‘Belief vs. practice’, in Eidinow and Kindt (eds), 21–29.
Harte, V. and Lane, M. (eds) (2013), Politeia in Greek and Roman Philosophy.
Cambridge.
Haubold, J. H. (2010) ‘Shepherd, farmer, poet, sophist: Hesiod on his own
reception’ in Boys-Stones and Haubold (eds), 11–30.
Havelock, E. A. (1958) ‘Parmenides and Odysseus’, HSCP 53: 133–143.
Hawes, G. (2014) Rationalizing Myth in Antiquity. Oxford.
Hedrick Jr., C. W. (2007) ‘Religion and society in Classical Greece’, in Ogden
(ed.), 283–296.
Heidegger, M. (1992) Parmenides. A. Schuwer and R. Rojcewicz (trans).
Indianapolis, IN.
Heidegger, M. (1999) ‘The last undelivered lecture (XII) from Summer Semester
1952’, W. McNeill (trans.), in Jacobs (ed.), 171–184.
Heidel, W. A. (1910) ‘Περὶ Φύσεως. A study of the conception of nature among the
Pre-Socratics’, Proceedings of the American Academy of Arts and Sciences 45:
79–133.
Heidel, W. A. (1943) ‘Hecataeus and Xenophanes’, AJPh 64: 257–277.
Heiden, B. (2007) ‘The Muses’ uncanny lies: Hesiod, Theogony 27 and its
translators’, AJPh 128: 153–175.
Heitsch, E. (1966) ‘Das Wissen des Xenophanes’, Rhein. Mus. 109: 193–235.
Heitsch, E. (1974) Parmenides: die Anfänge der Ontologie, Logik und
Naturwissenschaft. Munich.
Heitsch, E. (1983) Xenophanes: Die Fragmente. Munich.
Hermann, A. (2004) To Think Like God: Pythagoras and Parmenides, the
Origins of Philosophy. Las Vegas, NV.
Herrmann, F. -G. (2007) ‘Greek religion and philosophy: the god of the philoso-
pher’, in Ogden (ed.), 384–397.
Hershbell, J. P. (1972) ‘Plutarch and Parmenides’, GRBS 13: 193–207.
Hershbell, J. P. (1982) ‘Plutarch and Anaxagoras’, ICS 7: 141–158.
Hershbell, J. P. (1983) ‘Parmenides’ Way of Truth and B16’, in Anton and Preus
(eds), 41–58.
Hollmann, A. (2011) The Master of Signs: Signs and the Interpretation of Signs in
Herodotus’ Histories. Cambridge, MA.

371
 Bibliography
Hölscher, U. (1969) Parmenides: vom Wesen des Seienden. Frankfurt.
Horky, P. S. (2013) Plato and Pythagoreanism. Oxford.
Huffman, C. A. (2009) ‘The Pythagorean conception of the soul from Pythagoras
to Philolaus’, in Frede and Reis (eds), 21–43.
Huffman, C. A. (ed.) (2014), A History of Pythagoreanism. Cambridge.
Hülsz, E. (2013) ‘Heraclitus on Φύσις’, Epoché 17: 179–194.
Humphreys, S. C. (2004) ‘Dynamics of the Greek “breakthrough”: the dialogue
between philosophy and religion’, in her The Strangeness of Gods: Historical
Perspectives on the Interpretation of Athenian Religion. Oxford, 51–76.
Hussey, E. (1972) The Presocratics. London.
Hussey, E. (1990) ‘The beginnings of epistemology: from Homer to Philolaus’, in
Everson (ed.), 11–38.
Hussey, E. (2006) ‘Parmenides on thinking’, in King (ed.), 13–30.
Inwood, B. (2001) The Poem of Empedocles: A Text and Translation with an
Introduction, 2nd edn. Toronto.
Inwood, B. (2006) ‘Who do we think we are?’, in B. Reis (ed.), The Virtuous Life
in Greek Ethics. Cambridge, 230–243.
Iribarren, L. (forthcoming) ‘Vérité et référence: l’adresse des Muses à Hésiode
(Théog. 26–28) et son interprétation par Xénophane et Parménide’, in J. -
B. Gourinat (ed.), L’interprétation philosophique des mythes religieux. Paris.
Jacobs, D. C. (ed.) (1999), The Presocratics after Heidegger. Albany, NY.
Jacobs, D. C. (1999) ‘The ontological education of Parmenides’, in Jacobs (ed.),
185–202.
Jaeger, W. (1947) The Theology of the Early Greek Philosophers. E. S. Robinson
(trans.). Oxford.
Jameson, G. (1958) ‘“Well rounded truth” and circular thought in Parmenides’,
Phronesis 3: 15–30.
Janko, R. (1992) The Iliad: A Commentary, vol. IV: books 13–16. Cambridge.
Johansen, T. K. (1999) ‘Myth and logos in Aristotle’, in Buxton (ed.), 279–291.
Johansen, T. K. (2016) ‘Parmenides’ likely story’, OSAP 50: 1–29.
Johnson, D. M. (1999) ‘Hesiod’s description of Tartarus (Theogony 721–819)’,
Phoenix 53: 8–28.
Johnston, S. I. (2005a) ‘Introduction: divining divination’, in Johnston and
Struck (eds), 1–28.
Johnston, S. I. (2005b) ‘Delphi and the dead’, in Johnston and Struck (eds),
283–306.
Johnston, S. I. (2008) Ancient Greek Divination. Oxford.
Johnston, S. I. (2013a) ‘The myth of Dionysus’, in Graf and Johnston (eds), 66–93.
Johnston, S. I. (2013b) ‘The eschatology behind the Tablets’, in Graf and
Johnston (eds), 94–136.
Johnston, S. I. and Struck, P. T. (eds) (2005), Mantikê: Studies in Ancient
Divination. Leiden.
Journée, G. (2012) ‘Lumière et Nuit, Féminin et Masculin chez Parménide
d’Elée: quelques remarques’, Phronesis 57: 289–318.

372
 Bibliography
Jürrs, F. (1976) Zum Erkenntnisproblem bei den frühgriechischen Denkern.
Berlin.
Kahn, C. H. (1960) Anaximander and the Origins of Greek Cosmology.
New York, NY.
Kahn, C. H. (1969) ‘The thesis of Parmenides’, Review of Metaphysics 22:
700–724.
Kahn, C. H. (1971) ‘Religion and natural philosophy in Empedocles’ doctrine of
the soul’, in Anton and Kustas (eds), 3–38.
Kahn, C. H. (1973) The Verb “Be” in Ancient Greek. Dordrecht.
Kahn, C. H. (1976) ‘Why existence does not emerge as a distinct concept in
Greek Philosophy’, Archiv für Geschichte der Philosophie 58: 323–334.
Kahn, C. H. (1979) The Art and Thought of Heraclitus. Cambridge.
Kahn, C. H. (1981) ‘Some philosophical uses of “to be” in Plato’, Phronesis 26:
105–134.
Kahn, C. H. (1988) ‘Being in Parmenides and Plato’, PdP 43: 237–261.
Kahn, C. H. (2001) Pythagoras and the Pythagoreans. Indianapolis, IN.
Kahn, C. H. (2002) ‘Parmenides and Plato once more’, in Caston and Graham
(eds), 81–93.
Kahn, C. H. (2009) Essays on Being. Oxford.
Kahn, C. H. (2013) ‘The achievement of early Greek philosophy: a drama in five
acts: from Thales to the Timaeus’, in McCoy (ed.), 1–17.
Kamtekar, R. (2009) ‘Knowing by likeness in Empedocles’, Phronesis 54:
215–238.
Karsten, S. (1835) Parmenidis Eleatae Carminis Reliquiae, Philosophorum
Graecorum Veterum, i/1. Amsterdam.
Katz, J. T. and Volk, K. (2000) ‘Mere bellies? A new look at Theogony 26–28’,
JHS 120: 122–131.
Kearns, E. (2004) ‘The gods in the Homeric epics’, in R. Fowler (ed.),
The Cambridge Companion to Homer. Cambridge, 59–73.
Kearns, E. (2006) ‘Religious practice and belief’, in K. H. Kinzl, A Companion to
the Classical Greek World. Oxford, 311–326.
Kenney, E. J. (1971) Lucretius: De Rerum Natura Book III. Cambridge.
Kindt, J. (2006) ‘Delphic oracle stories and the beginning of historiography:
Herodotus’ Croesus Logos’, Classical Philology 101: 34–51.
Kindt, J. (2012) Rethinking Greek Religion. Cambridge.
Kindt, J. (2015) ‘Personal religion: a productive category for the study of ancient
Greek religion?’, JHS 135: 35–50.
King, R. A. H. (ed.) (2006), Common to Body and Soul. Berlin.
Kingsley, P. (1995) Ancient Philosophy, Mystery, and Magic: Empedocles and
Pythagorean Tradition. Oxford.
Kingsley, P. (1999) In the Dark Places of Wisdom. Inverness, CA.
Kingsley, P. (2003) Reality. Inverness, CA.
Kirk, G. S. (1970) Myth, its Meaning and Functions in Ancient and Other
Cultures. Cambridge.

373
 Bibliography
Kirk, G. S. and Stokes, M. C. (1960) ‘Parmenides’ refutation of motion’,
Phronesis 5: 1–4.
Kirk, G. S., Raven, J. E. and Schofield, M. (1983) The Presocratic Philosophers
2nd edn. Cambridge.
Kleingünther, A. (1933) ΠΡΩΤΟΣ ΕΥΡΕΤΗΣ, Philologus, Supplementband 26:
1–155.
Koning, H. H. (2010) Hesiod: The Other Poet: Ancient Reception of a Cultural
Icon. Leiden.
Kourouniotes, K. (1936) Eleusis. Athens.
Kowalzig, B. (2007) Singing for the Gods: Performances of Myth and Ritual in
Archaic and Classical Greece. Oxford.
Kraus, M. (2013) ‘Parmenides’, in H. Flashar, D. Bremer and G. Rechenauer
(eds), Die Philosophie der Antike Band 1: Frühgriechische Philosophie.
Basel, 441–530.
Kreisel, H. (1994) ‘Imitatio dei in Maimonides’ Guide of the Perplexed’, AJS
Review 19: 169–211.
Kurfess, C. (2016) ‘The truth about Parmenides’ Doxa’, Ancient Philosophy 36:
13–45.
Laks, A. (1990) ‘“The more” and “the full”: on the reconstruction of Parmenides’
theory of sensation in Theophrastus, De Sensibus, 3–4’, OSAP 8: 1–18.
Laks, A. (1999) ‘Soul, sensation and thought’, in Long (ed.), 250–270.
Laks, A. (2002) ‘“Philosophes présocratiques” Remarques sur la construction
d’une catégorie de l’historiographie philosophique’, in Laks and Louguet
(eds), 17–38.
Laks, A. (2003) ‘Phénomènes et références: éléments pour une réflexion sur la
rationalisation de l’irrationnel’, Methodos 3: 9–33.
Laks, A. (2013) ‘Phenomenon and reference: revisiting Parmenides, Empedocles, and
the problem of rationalization’, S. C. Humphreys (trans.), in R. G. Wagner and
S. C. Humphreys (eds), Modernity’s Classics. Berlin, 165–186.
Laks, A. and Most, G. W. (eds) (1997), Studies on the Derveni Papyrus. Oxford.
Laks, A. and Louguet, C. (eds) (2002), Qu’est-ce que la philosophie
présocratique? Villeneuve d’Ascq.
Lamberton, R. (1992) ‘The Neoplatonists and the spiritualization of Homer’, in
Lamberton and Keaney (eds), 115–133.
Lamberton, R. and Keaney, J. J. (eds) (1992), Homer’s Ancient Readers:
The Hermeneutics of Greek Epic’s Earliest Exegetes. Princeton, NJ.
Ledbetter, G. M. (2003) Poetics before Plato: Interpretation and Authority in
Early Greek Theories of Poetry. Princeton, NJ.
Lee, M. -K. (2005) Epistemology after Protagoras: Responses to Relativism in
Plato, Aristotle, and Democritus. Oxford.
Leigh, F. (2008) ‘The copula and semantic continuity in Plato’s Sophist’, OSAP
34: 105–121.
Lenz, J. R. (2007) ‘Deification of the philosopher in Classical Greece’, in M. J.
Christensen and J. A. Wittung (eds), Partakers of the Divine Nature:

374
 Bibliography
The History and Development of Deification in the Christian Traditions.
Madison, NJ, 47–67.
Lesher, J. H. (1983) ‘Xenophanes’ scepticism’, in Anton and Preus (eds), 20–40.
Lesher, J. H. (1984) ‘Parmenides’ critique of thinking: the poluderis elenchos of
Fragment 7’, OSAP 2: 1–31.
Lesher, J. H. (1991) ‘Xenophanes on inquiry and discovery: an alternative to
the “Hymn to Progress” reading of Fr. 18’, Ancient Philosophy 11:
229–248.
Lesher, J. H. (1992) Xenophanes of Colophon: Fragments, a Text and Translation
with a Commentary. Toronto.
Lesher, J. H. (1994a) ‘The significance of κατὰ πάντ᾽ ἄ<σ>τη in Parmenides Fr.
1.3’, Ancient Philosophy 14: 1–20.
Lesher, J. H. (1994b) ‘The emergence of philosophical interest in cognition’,
OSAP 12: 1–34.
Lesher, J. H. (2008) ‘The humanizing of knowledge in Presocratic thought’, in
Curd and Graham (eds), 458–484.
Lesher, J. H. (2013) ‘A systematic Xenophanes?’, in McCoy (ed.), 77–90.
Linforth, I. M. (1941) The Arts of Orpheus. Berkeley, CA.
Lloyd, G. E. R. (1966) Polarity and Analogy. Cambridge.
Lloyd, G. E. R. (1979) Magic, Reason and Experience. Cambridge.
Lloyd, G. E. R. (1987) The Revolutions of Wisdom: Studies in the Claims and
Practice of Ancient Greek Science. Berkeley, CA.
Lloyd, G. E. R. (1999) ‘Mythology: reflections from a Chinese perspective’, in
Buxton (ed.), 145–165.
Lloyd, G. E. R. (2002) ‘Le pluralisme de la vie intellectuelle avant Platon’,
A. Laks (trans.), in Laks and Louguet (eds), 39–53.
Lloyd, G. E. R. (2014) ‘Pythagoras’, in Huffman (ed.), 24–45.
Loenen, J. H. M. M. (1956) ‘In defence of the traditional interpretation of
Xenophanes’ Fr. 18’, Mnemosyne ser. 4, no. 9: 135–136.
Loenen, J. H. M. M. (1959) Parmenides, Melissus, Gorgias: A Reinterpretation
of Eleatic Philosophy. Assen.
Loney, A. C. (2014) ‘Hesiod’s incorporative poetics in the Theogony and the
contradictions of Prometheus’, AJPh 135: 503–531.
Long, A. A. (1966) ‘Thinking and sense-perception in Empedocles: mysticism or
materialism?’, CQ 16: 256–276.
Long, A. A. (1975) ‘The principles of Parmenides’ cosmogony’, in Furley and
Allen (eds), 82–101.
Long, A. A. (1996) ‘Parmenides on thinking being’, BACAP 12: 125–162.
Long, A. A. (ed.) (1999) The Cambridge Companion to Early Greek Philosophy.
Cambridge.
Long, A. A. (1999) ‘The scope of early Greek philosophy’, in Long (ed.), 1–21.
Long, A. A. (2002) Epictetus: A Stoic and Socratic Guide to Life. Oxford.
Long, A. A. (2009) ‘Heraclitus on measure and the explicit emergence of
rationality’, in Frede and Reis (eds), 87–109.

375
 Bibliography
Long, G. (ed.) and Paley, F. A. (commentary) (1861) The Epics of Hesiod.
London.
Lumpe, A. (1952) Die Philosophie des Xenophanes von Kolophon. Munich.
Luther, M. (1995) Table Talk. W. Hazlitt (trans.). London.
Mackenzie, M. M. (1982) ‘Parmenides’ dilemma’, Phronesis 27: 1–12.
Macris, C. (2009) ‘Pythagoras’, in G. Oppy and N. Trakakis (eds), The History of
Western Philosophy of Religion, vol. I: Ancient Philosophy of Religion.
Durham, 23–39.
Macris, C. and Skarsouli, P. (2012) ‘La sagesse et les pouvoirs du mystérieux τις
du fragment 129 d’Empédocle’, Revue de métaphysique et de morale 75:
357–377.
Manetti, G. (1993) Theories of Sign in Classical Antiquity. C. Richardson (trans.).
Bloomington, IN.
Mansfeld, J. (1964) Die Offenbarung des Parmenides und die Menschliche Welt.
Assen.
Mansfeld, J. (1995) ‘Insight by hindsight: intentional unclarity in presocratic
proems’, BICS 40: 225–232.
Mansfeld, J. (1996) ‘Aristote et la structure du De Sensibus de Théophraste’,
Phronesis 41: 158–188.
Mansfeld, J. (1999a) ‘Parménide et Héraclite avaient-ils une théorie de la percep-
tion?’, Phronesis 44: 326–346.
Mansfeld, J. (1999b) ‘Sources’ in Long (ed.), 22–44.
Mansfeld, J. (2005) ‘Minima Parmenidea’, Mnemosyne ser. 4, no. 58: 554–560.
Mansfeld, J. (2008) ‘A crux in Parmenides Fr. B 1.3 DK’, in S. Heilen, R. Kirstein,
R. Scott Smith, S. M. Trzaskoma, R. L. van der Wal and M. Vorwerk (eds),
In Pursuit of Wissenschaft. Festschrift für William M. Calder III zum 75.
Geburtstag. Hildesheim, 299–301.
Mansfeld, J. (2013) ‘Detheologization’, Rhizomata 1: 330–362.
Mansfeld, J. (2015), ‘Parmenides from right to left’, Études platoniciennes 12.
Marquardt, P. A. (1982) ‘The two faces of Hesiod’s Muse’, ICS 7: 1–12.
Marx, K. and Engels, F. (2008) On Religion. Mineola, NY.
Martin, A. and Primavesi, O. (1999) L’Empédocle de Strasbourg. Berlin.
Martin, G. (2009) Divine Talk: Religious Argumentation in Demosthenes. Oxford.
Mastronarde, D. J. (1994) Euripides: Phoenissae: Edited with Introduction and
Commentary. Cambridge.
Maurizio, L. (1997) ‘Delphic oracles as oral performances: authenticity and
historical evidence’, ClAnt. 16: 308–334.
Maurizio, L. (2013) ‘Technopaegnia in Heraclitus and the Delphic oracles:
shared compositional techniques’, in J. Kwapisz, D. Petrain and
M. Szymanski (eds), The Muse at Play: Riddles and Wordplay in Greek and
Latin Poetry. Berlin, 100–120.
Mayhew, R. (2008) Plato: Laws 10. Oxford.
McCoy, J. (ed.) (2013) Early Greek Philosophy: The Presocratics and the
Emergence of Reason. Washington, DC.

376
 Bibliography
McCoy, M. K. (1989) ‘Xenophanes’ epistemology: empiricism leading to skepti-
cism’, in Boudouris (ed.), 235–240.
McKirahan, R. D. (1994) Philosophy before Socrates: An Introduction with Texts
and Commentary. Indianapolis, IN.
McKirahan, R. D. (2005) ‘Assertion and argument in Empedocles’ cosmology or,
what did Empedocles learn from Parmenides?’, in Pierris (ed.), 163–188.
McKirahan, R. D. (2008) ‘Signs and arguments in Parmenides B8’, in Curd and
Graham (eds), 189–229.
Meijer, P. A. (1997) Parmenides beyond the Gates: The Divine Revelation on
Being, Thinking and the Doxa. Amsterdam.
Menn, S. (2012) ‘Critique of earlier philosophers on the good and the causes
(Metaphysics Α7-Α8 989a18)’, in Steel (ed.), 201–224.
Mihai, A. (2010) ‘Soul’s aitherial abode according to the Poteidaia epitaph and
the presocratic philosophers’, Numen 57: 553–582.
Mikalson, J. D. (2010) Greek Popular Religion in Greek Philosophy. Oxford.
Miller, E. L. (1968) ‘Parmenides the prophet?’, JHP 6: 67–69.
Miller, M. (2006) ‘Ambiguity and transport: reflections on the proem to
Parmenides’ poem’, OSAP 30: 1–47.
Miller, P. L. (2011) Becoming God: Pure Reason in Early Greek Philosophy.
London.
Minar, E. L. (1949) ‘Parmenides and the world of seeming’, AJPh 70: 41–55.
Mogyoródi, E. (2002) ‘Xenophanes as a philosopher: theology and theodicy’, in
Laks and Louguet (eds), 253–286.
Mogyoródi, E. (2006) ‘Xenophanes’ epistemology and Parmenides’ quest for
knowledge’, in Sassi (ed.), 123–160.
Montanari, F., Rengakos, A. and Tsagalis, C. C. (ed.) (2009), Brill’s Companion
to Hesiod. Leiden.
Montiglio, S. (2000) ‘Wandering philosophers in Classical Greece’, JHS 120:
86–105.
Montiglio, S. (2005) Wandering in Ancient Greek Culture. Chicago, IL.
Morgan, K. (2000) Myth and Philosophy from the Presocratics to Plato.
Cambridge.
Morrison, J. S. (1955) ‘Parmenides and Er’, JHS 75: 59–68.
Most, G. W. (1999a) ‘The poetics of early Greek philosophy’, in Long (ed.),
332–362.
Most, G. W. (1999b) ‘From logos to mythos’, in Buxton (ed.), 22–47.
Most, G. W. (2003) ‘Philosophy and religion’, in Sedley (ed.), 300–322.
Most, G. W. (ed. and trans. ) (2006) Hesiod, 2 vols. London.
Most, G. W. (2007) ‘ἄλλος δ’ ἐξ ἄλλου δέχεται: Presocratic philosophy and
traditional Greek epic’, in A. Bierl, R. Lämmle and K. Wesselmann (eds),
Literatur und Religion I: Wege zu einer mythisch-rituellen Poetik bei den
Griechen. Berlin, 271–302.
Most, G. W. (2011) ‘What ancient quarrel between philosophy and poetry?’, in
P. Destrée and F. G. Herrmann (eds), Plato and the Poets. Leiden, 1–20.

377
 Bibliography
Most, G. W. (2013) ‘Heraclitus on religion’, Rhizomata 1: 153–167.
Moulton, C. (1977) Similes in the Homeric Poems. Göttingen.
Mourelatos, A. P. D. (1965) ‘Phrazo and its derivatives in Parmenides’, Classical
Philology 60: 261–262.
Mourelatos, A. P. D. (ed.) (1974) The Pre-Socratics: A Collection of Critical
Essays. Princeton, NJ.
Mourelatos, A. P. D. (1976) ‘Determinacy and indeterminacy, being and
non-being in the fragments of Parmenides’, Canadian Journal of
Philosophy, suppl. vol. II: 45–60.
Mourelatos, A. P. D. (1979) ‘Some alternatives in interpreting Parmenides’,
The Monist 62: 3–14.
Mourelatos, A. P. D. (1989) ‘“X is really Y”: Ionian origins of a thought pattern’,
in Boudouris (ed.), 280–290.
Mourelatos, A. P. D. (2008a) The Route of Parmenides: Revised and Expanded
Edition. Las Vegas, NV.
Mourelatos, A. P. D. (2008b) ‘Heraclitus, Parmenides, and the naïve metaphysics
of things’, in Mourelatos (2008a), 299–332.
Mourelatos, A. P. D. (2008c) ‘Editing note’ to G. Vlastos, ‘“Names” of being in
Parmenides’, in Mourelatos (2008a), 386–388.
Mourelatos, A. P. D. (2008d) ‘The cloud-astrophysics of Xenophanes and Ionian
material monism’, in Curd and Graham (eds), 134–168.
Mourelatos, A. P. D. (2011) ‘Parmenides, early Greek astronomy, and modern
Scientific Realism’, in Cordero (ed.), 167–189.
Mourelatos, A. P. D. (2013) ‘Sounds, images, mysticism, and logic in
Parmenides’, in L. Rossetti and M. Pulpito (eds), Parmenide: suoni, immagini,
esperienza. Sankt Augustin, 159–177.
Mourelatos, A. P. D. (2014) ‘The conception of eoikōs/eikōs as epistemic stan-
dard in Xenophanes, Parmenides, and in Plato’s Timaeus’, Ancient Philosophy
34: 169–191.
Moyer, I. S. (2002) ‘Herodotus and an Egyptian mirage: the genealogies of the
Theban priests’, JHS 122: 70–90.
Murnaghan, S. (1987) Disguise and Recognition in the Odyssey. Princeton, NJ.
Murnaghan, S. (1995) ‘The plan of Athena’, in Cohen (ed.), 61–78.
Murray, P. (1981) ‘Poetic inspiration in early Greece’, JHS 101: 87–100.
Myers, F. W. H. (1897) ‘Greek oracles’, in his Essays-Classical. London, 1–106.
Mylonas, G. E. (1961) Eleusis and the Eleusinian Mysteries. Princeton, NJ.
Naddaf, G. (2005) The Greek Concept of Nature. Albany, NY.
Naddaf, G. (2006) ‘What is Presocratic philosophy?’, Ancient Philosophy 26:
161–179.
Nagler, M. N. (1996) ‘Dread goddess revisited’, in Schein (ed.), 141–161.
Nagy, G. (1990) Greek Mythology and Poetics. Ithaca, NY.
Needham, R. (1972) Belief, Language and Experience. Oxford.
Nehamas, A. (1981) ‘On Parmenides’ three ways of inquiry’, Deucalion 33/4:
97–111.

378
 Bibliography
Nehamas, A. (1999) ‘Plato on the imperfection of the sensible world’, in G. Fine
(ed.), Plato, vol. I: Metaphysics and Epistemology. Oxford, 171–191.
Nehamas, A. (2002) ‘Parmenidean being/Heraclitean fire’, in Caston and
Graham (eds), 45–64.
Nestle, W. (1942) Vom Mythos zum Logos. Stuttgart.
Nietzsche, F. W. (1922) Gesammelte Werke, vol. V. R. Oehler and M. Oehler
(eds). Munich.
Nietzsche, F. W. (1974) The Complete Works, vol. II: Early Greek Philosophy.
O. Levy (ed.), M.A. Mügge (trans.). Brooklyn, NY.
Nightingale, A. W. (2007) ‘The philosophers in archaic Greek culture’, in H. A.
Shapiro (ed.), The Cambridge Companion to Archaic Greece. Cambridge,
169–198.
Nutton, V. (2013) Ancient Medicine, 2nd edn. London.
Ober, J. (2010) ‘Socrates and democratic Athens’, in D. R. Morrison,
The Cambridge Companion to Socrates. Cambridge, 138–178.
Ogden, D. (ed.) (2007), A Companion to Greek Religion. Oxford.
Oppy, G. and Trakakis, N. (eds) (2009), The History of Western Philosophy of
Religion, vol. I: Ancient Philosophy of Religion. Durham.
Osborne, C. (1987a) Rethinking Early Greek Philosophy: Hippolytus of Rome
and the Presocratics. London.
Osborne, C. (1987b) ‘Empedocles recycled’ CQ 37: 24–50.
Osborne, C. (1997) ‘Heraclitus and the rites of established religion’, in A. B.
Lloyd (ed.), What is a God? London, 35–42.
Osborne, C. (2005) ‘Sin and moral responsibility in Empedocles’s cosmic cycle’,
in Pierris (ed.), 283–308.
Osborne, C. (2006) ‘Was there an Eleatic revolution in philosophy?’, in Goldhill
and Osborne (eds), 218–245.
Osborne, R. (1997) ‘The polis and its culture’, in Taylor (ed.), 9–46.
Osborne, R. (2016) ‘Sacrificial theologies’, in Eidinow, Kindt and Osborne (eds),
233–248.
Owen, G. E. L. (1960) ‘Eleatic questions’, CQ 10: 84–102.
Owen, G. E. L. (1974) ‘Plato and Parmenides on the timeless present’, in
Mourelatos (ed.), 271–292.
Owens, J. (1979) ‘Knowledge and katabasis in Parmenides’, The Monist 62:
15–29.
Paleothodorus, D. (2010) ‘Light and darkness in Dionysiac rituals as illustrated
on Attic vase paintings of the 5th century BCE’, in Christopoulos, Karakantza
and Levaniouk (eds), 237–260.
Palmer, J. A. (1998) ‘Xenophanes’ Ouranian god in the fourth century’, OSAP
16: 1–34.
Palmer, J. A. (1999) Plato’s Reception of Parmenides. Oxford.
Palmer, J. A. (2000) ‘Aristotle on the ancient theologians’, Apeiron 33:
181–206.
Palmer, J. A. (2009) Parmenides and Presocratic Philosophy. Oxford.

379
 Bibliography
Palmer, J. A. (2013) ‘Revelation and reasoning in Kalliopeia’s address to
Empedocles’, Rhizomata 1: 308–329.
Panagiotou, S. (1983) ‘Empedocles on his own divinity’, Mnemosyne ser. 4, no.
36: 276–285.
Parke, H. W. (1967) The Oracles of Zeus: Dodona, Olympia, Ammon. Oxford.
Parke, H. W. and Wormell, D. E. W. (1956) The Delphic Oracle, vol. II:
The Oracular Responses. Oxford.
Parker, R. (1997) ‘Gods cruel and kind: tragic and civic theology’, in C. Pelling
(ed.), Greek Tragedy and the Historian. Oxford, 143–160.
Parker, R. (2000) ‘Greek states and Greek oracles’, in Buxton (ed.), 76–108.
Parker, R. (2011) On Greek Religion. Ithaca, NY.
Pascal, B. (2008) Human Happiness. A. J. Krailsheimer (trans.). London.
Patera, I. (2010) ‘Light and lighting equipment in the Eleusinian mysteries:
symbolism and ritual use’, in Christopoulos, Karakantza and Levaniouk
(eds), 261–275.
Patin, A. (1899) ‘Parmenides im Kampfe gegen Heraklit’, Jahrbuch für klas-
sische Philologie, Supplementband 25: 489–660.
Pellikaan-Engel, M. E. (1978) Hesiod and Parmenides: A New View on Their
Cosmologies and on Parmenides’ Proem. Amsterdam.
Pendrick, G. J. (2002) Antiphon the Sophist: The Fragments. Cambridge.
Penelhum, T. (1983) ‘Skepticism and fideism’, in M. F. Burnyeat (ed.),
The Skeptical Tradition. Berkeley, CA, 287–318.
Petridou, G. (2015) Divine Epiphany in Greek Literature and Culture. Oxford.
Pfeiffer, H. (1975) Die Stellung des parmenideischen Lehrgedichtes in der
epischen Tradition. Bonn.
Phillips, E.D. (1955) ‘Parmenides on thought and being’, Philosophical Review
64: 546–560.
Picot, J. -C. and Berg, W. (2013) ‘Empedocles vs. Xenophanes: differing notions
of the divine’, Organon 45: 5–19.
Pierris, A. L. (ed.) (2005) The Empedoclean Κόσμος: Structure, Process and the
Question of Cyclicity. Patras.
Pierris, A. L. (2007) The Emergence of Reason from the Spirit of Mystery:
An Inquiry into the Origin and Nature of Ancient Greek Rationality, vol. II:
Mystery and Philosophy. Patras.
Pinto, R. (2016) ‘“All things are full of gods”: souls and gods in Thales’, Ancient
Philosophy 36: 243–261.
Plantinga, A. (2000) Warranted Christian Belief. Oxford.
Popper, K. R. (1970) ‘Back to the Presocratics’, in Furley and Allen (eds),
130–153.
Popper, K. R. (1992) ‘How the moon might shed some of her light upon the two
ways of Parmenides’, CQ 42: 12–19.
Popper, K. R. (1998) The World of Parmenides: Essays on the Presocratic
Enlightenment. A. F. Peterson (ed.). London, 93–104.
Price, S. (1999) Religions of the Ancient Greeks. Cambridge.

380
 Bibliography
Primavesi, O. (2008) ‘Empedocles: physical and mythical divinity’, in Curd and
Graham (eds), 250–283.
Primavesi, O. (2013) ‘Le chemin vers la révélation: lumière et nuit dans le
proème de Parménide’, Philosophie Antique 13: 37–81.
Pucci, P. (1977) Hesiod and the Language of Poetry. London.
Pucci, P. (2009) ‘The poetry of the Theogony’, in Montanari, Rengakos and
Tsagalis (eds), 37–70.
Pulpito, M. (2011) ‘Parmenides and the forms’, in Cordero (ed.), 191–212.
Rangos, S. (2012) ‘Empedocles on divine nature’, Revue de métaphysique et de
morale 75: 315–38.
Ranzato, S. (2015) Il Kouros e la Verità: Polivalenza delle immagini nel poema di
Parmenide. Pisa.
Reiche, H. A. T. (1971) ‘Empirical aspects of Xenophanes’ theology’, in Anton
and Kustas (eds), 88–110.
Reinhardt, K. (1974) ‘The relation between the two parts of Parmenides’ poem’,
in Mourelatos (ed.), 293–311.
Richardson, N. J. (1974) The Homeric Hymn to Demeter. Oxford.
de Rijk, L. M. (1983) ‘Did Parmenides reject the sensible world?’, in L. P. Gerson
(ed.), Graceful Reason. Toronto, 29–53.
Robbiano, C. (2006) Becoming Being: On Parmenides’ Transformative
Philosophy. Sankt Augustin.
Robertson, N. (2003) ‘Orphic mysteries and Dionysiac ritual’ in Cosmopoulos
(ed.), 218–240.
Robinson, T. M. (1975) ‘Parmenides on ascertainment of the real’, Canadian
Journal of Philosophy 4: 623–633.
Robinson, T. M. (2008) ‘Presocratic theology’, in Curd and Graham (eds),
485–498.
Rohde, E. (1925) Psyche: The Cult of Souls and Belief in Immortality among the
Greeks. W. B. Hillis (trans.). London.
Roloff, D. (1970) Gottähnlichkeit, Vergöttlichung und Erhöhung zu seligem
Leben: Untersuchungen zur Herkunft der platonischen Angleichung an Gott.
Berlin.
Rosen, R. M. (1990) ‘Poetry and sailing in Hesiod’s Works and Days’, ClAnt. 9:
99–113.
Ross, W. D. (1966) Aristotle’s Metaphysics, 2 vols. Oxford.
Rowe, C. J. (1983) ‘“Archaic thought” in Hesiod’, JHS 103: 124–135.
Rowett, C. (2013) ‘On calling gods by the right names’, Rhizomata 1: 168–193.
Rowett, C. (2014) ‘The Pythagorean society and politics’, in Huffman (ed.), 112–130.
Rowett, C. (2016) ‘Why the philosopher kings will believe the noble lie’, OSAP
50: 67–100.
Russell, N. (2006) The Doctrine of Deification in the Greek Patristic Tradition.
Oxford.
Santaniello, C. (2012) ‘Θεός, δαίμων, φρὴν ἱερή: Empedocles and the divine’,
Revue de métaphysique et de morale 75: 301–313.

381
 Bibliography
Santos, J. T. (2011) ‘The role of “thought” in the argument of Parmenides’ poem’,
in Cordero (ed.), 251–270.
Sassi, M. M. (2002) ‘La naissance de la philosophie de l’esprit de la tradition’, in
Laks and Louguet (eds), 55–81.
Sassi, M. M. (ed.) (2006), The Construction of Philosophical Discourse in the
Age of the Presocratics. Pisa.
Sassi, M. M. (2009) Gli inizi della filosofia: in Grecia. Turin.
Sassi, M. M. (2013) ‘Where epistemology and religion meet: what do(es) the
god(s) look like’, Rhizomata 1: 283–307.
Sassi, M. M. (2016) ‘Parmenides and Empedocles on krasis and knowledge’,
Apeiron 49: 451–469.
Schäfer, C. (1996) Xenophanes von Kolophon: Ein Vorsokratiker zwischen
Mythos und Philosophie. Stuttgart.
Schein, S. L. (ed.) (1996) Reading the Odyssey. Princeton, NJ.
Schein, S. L. (1996) ‘Introduction’, in Schein (ed.), 3–31.
Schibli, H. S. (1990) Pherekydes of Syros. Oxford.
Schofield, M. (1970) ‘Did Parmenides discover eternity?’, Archiv für Geschichte
der Philosophie 52: 113–135.
Schofield, M. (1987) ‘Coxon’s Parmenides’, Phronesis 32: 349–359.
Schofield, M. (1991) ‘Heraclitus’ theory of soul and its antecedents’, in
S. Everson (ed.), Companions to Ancient Thought, vol. II: Psychology.
Cambridge, 13–35.
Schofield, M. (1997) ‘The Ionians’, in Taylor (ed.), 47–87.
Schofield, M. (2003) ‘The Presocratics’, in Sedley (ed.), 42–72.
Schofield, M. (2007) ‘Heroic Journeys’, Presidential Address to the Classical
Association. Sherborne.
Schrödinger, E. (1954) Nature and the Greeks. Cambridge.
Seaford, R. (1986) ‘Immortality, salvation and the elements’, HSPh 90: 1–26.
Seaford, R. (2005) ‘Mystic light in Aeschylus’ Bassarai’, CQ 55: 602–6.
Seaford, R. (2010) ‘Mystic light and near-death experience’, in Christopoulos,
Karakantza and Levaniouk (eds), 201–206.
Sedley, D. N. (1999a) ‘Parmenides and Melissus’, in Long (ed.), 113–133.
Sedley, D. N. (1999b) ‘The ideal of godlikeness’, in G. Fine (ed.), Oxford Readings in
Plato: Ethics, Politics, Religion and the Soul, vol. II. Oxford, 309–328.
Sedley, D. N. (ed.) (2003) The Cambridge Companion to Greek and Roman
Philosophy. Cambridge.
Sedley, D. N. (2005) ‘Empedocles’ life cycles’, in Pierris (ed.), 331–371.
Sedley, D. N. (2007) Creationism and Its Critics in Antiquity. Berkeley, CA.
Sedley, D. N. (2010) ‘Hesiod’s Theogony and Plato’s Timaeus’, in Boys-Stones
and Haubold (eds), 246–258.
Sedley, D. N. (2013a) ‘The atheist underground’, in Harte and Lane (eds), 329–348.
Sedley, D. N. (2013b) ‘From the pre-Socratics to the Hellenistic age’, in
S. Bullivant and M. Ruse (eds), The Oxford Handbook of Atheism. Oxford,
139–151.

382
 Bibliography
Segal, C. (1992) ‘Bard and audience in Homer’, in Lamberton and Keaney
(eds), 3–29.
Shaw, G. (2004) Review of Kingsley (2003), BMCR 2004.07.43.
Shorey, P. (1911) ‘Note on Xenophanes Fr. 18 (Diels) and Isocrates Panegyricus
32’, Classical Philology 6: 88–89.
Sider, D. and Johnstone Jr., H.W. (1986) The Fragments of Parmenides. Bryn
Mawr, PA.
Sisko, J. E. (2003) ‘Anaxagoras’ Parmenidean cosmology: worlds within worlds
within the one’, Apeiron 36: 87–114.
Sisko, J. E. and Weiss, Y. (2015) ‘A fourth alternative in interpreting
Parmenides’, Phronesis 60: 40–59.
Slatkin, L. M. (1996) ‘Composition by theme and the metis of the Odyssey’, in
Schein (ed.), 223–237.
Smith, W. C. (1980) ‘Belief: a reply to a response’, Numen 27: 247–255.
Smyth, H. W. (1972) Greek Grammar, rev. by G. M. Messing. Cambridge, MA.
Snell, B. (1953) The Discovery of Mind. T. G. Rosenmeyer (trans.). Oxford.
Solmsen, F. (1957) ‘The vital heat, the inborn pneuma and the aether’, JHS 77:
119–123.
Sourvinou-Inwood, C. (2000) ‘What is polis religion?’, in Buxton (ed.), 13–37.
Steel, S. (ed.) (2012), Aristotle’s Metaphysics Alpha. Oxford.
Steele, L. D. (2002) ‘Mesopotamian elements in the proem of Parmenides?’, CQ
52: 583–588.
Stern, J. (1996) Palaephatus: On Unbelievable Tales: Translation, Introduction
and Commentary. Wauconda, IL.
Stern, J. (1999) ‘Rationalizing myth: methods and motives in Palaephatus’, in
Buxton (ed.), 215–222.
Stoddard, K. (2004) The Narrative Voice in the Theogony of Hesiod. Leiden.
Stokes, M. C. (1962) ‘Hesiodic and Milesian cosmogonies – I’, Phronesis
7: 1–37.
Stokes, M. C. (1963) ‘Hesiodic and Milesian cosmogonies – II’, Phronesis
8: 1–34.
Stokes, M. C. (1971) One and Many in Presocratic Philosophy.
Washington, DC.
Stratton, G. M. (1917) Theophrastus and the Greek Physiological Psychology
before Aristotle. London.
Stroh, W. (1976) ‘Hesiods lügende Musen’, in H. Görgemanns und E.A. Schmidt
(eds), Studien zum antiken Epos. Meisenheim, 85–112.
Struck, P. T. (2005) ‘Divination and literary criticism?’, in Johnston and Struck
(eds), 147–165.
Tarán, L. (1965) Parmenides: A Text with Translation, Commentary, and Critical
Essays. Princeton, NJ.
Tarrant, D. (1951) ‘Plato’s use of quotations and other illustrative material’, CQ
45: 59–67.
Taub, L. (2003) Ancient Meteorology. London.

383
 Bibliography
Taylor, C. C. W. (ed.) (1997) Routledge History of Philosophy, vol. I: From the
Beginning to Plato. London.
Tejera, V. (1989) ‘The expressive medium of the Ionian Presocratics’, in
Boudouris (ed.), 384–393.
Thalmann, W. G. (1984) Conventions of Form and Thought in Early Greek
Poetry. Baltimore, MD.
Thanassas, P. (2006) ‘How many Doxai are there in Parmenides?’, Rhizai 3:
199–218.
Thanassas, P. (2011) ‘Parmenidean dualisms’, in Cordero (ed.), 289–308.
Tor, S. (2009) review of Robbiano (2006), BMCR 2009.03.57.
Tor, S. (2010) review of Ustinova (2009), CR 60: 493–495.
Tor, S. (2012) review of Mikalson (2010), JHS 132: 281–282.
Tor, S. (2013a) ‘Mortal and divine in Xenophanes’ epistemology’, Rhizomata 1:
248–282.
Tor, S. (2013b) ‘Sextus Empiricus on Xenophanes’ scepticism’, IJSS 3: 1–23.
Tor, S. (2015) ‘Parmenides’ epistemology and the two parts of his poem’,
Phronesis 60: 3–39.
Tor, S. (2016) ‘Heraclitus on Apollo’s signs and his own: contemplating oracles
and philosophical inquiry’, in Eidinow, Kindt and Osborne (eds), 89–116.
Torgerson, T. B. (2006) ‘The εἰδὼς φώς and the traditional dichotomy of divine
and mortal epistemology’, Revue de Philosophie Ancienne 24: 25–43.
Trampedach, K. (2008) ‘Authority disputed: the seer in Homeric epic’, in
G. Dignas and K. Trampedach (eds), Practitioners of the Divine. Cambridge,
MA, 207–230.
Trépanier, S. (2003) ‘Empedocles on the ultimate symmetry of the world’, OSAP
24: 1–57.
Trépanier, S. (2004) Empedocles: An Interpretation. London.
Trépanier, S. (2010) ‘Early Greek theology: god as nature and natural gods’, in
Bremmer and Erskine (eds), 273–317.
Trépanier, S. (2014) ‘From wandering limbs to limbless gods: δαίμων as sub-
stance in Empedocles’, Apeiron 47: 172–210.
Tsagalis, C. C. (2009) ‘Poetry and poetics in the Hesiodic corpus’, in Montanari,
Rengakos and Tsagalis (eds), 131–177.
Tulin, A. (1993) ‘Xenophanes Fr. 18 D.-K. and the origins of the idea of
progress’, Hermes 121: 129–138.
Ustinova, Y. (2004) ‘Truth lies at the bottom of a cave: Apollo Pholeuterios, the
Pholarchs of the Eleats, and subterranean oracles’, PdP 59: 25–44.
Ustinova, Y. (2009) Caves and the Ancient Greek Mind: Descending
Underground in the Search for Ultimate Truth. Oxford.
Van Noorden, H. (2015) Playing Hesiod. Cambridge.
Verdenius, W. J. (1947) ‘Notes on the Presocratics’, Mnemosyne ser. 3, no. 13:
271–289.
Verdenius, W. J. (1949) ‘Parmenides’ conception of light’, Mnemosyne ser. 4, no.
2: 116–131.

384
 Bibliography
Verdenius, W. J. (1953) ‘Xenophanes 34, 3’, Mnemosyne ser. 4, no. 6: 197.
Verdenius, W. J. (1955) ‘Xenophanes Fr. 18’, Mnemosyne ser. 4, no. 8: 221.
Verdenius, W. J. (1956) ‘Xenophanes Fr. 1, 24’, Mnemosyne ser. 4, no. 9: 136.
Verdenius, W. J. (1964) Parmenides: Some Comments on His Poem. Amsterdam.
Vernant, J. -P. (1982) ‘The myth of Prometheus in Hesiod’ in his Myth and
Society in Ancient Greece. J. Lloyd (trans.). London, 183–201.
Vernant, J. -P. (1991) Mortals and Immortals. F. I. Zeitlin (ed.). Princeton, NJ.
Vernant, J. -P. (1991a) ‘Mortals and immortals: the body of the divine’, F. I.
Zeitlin (trans.), in Vernant (1991), 27–49.
Vernant, J. -P. (1991b) ‘Speech and mute signs’, F. I. Zeitlin (trans.), in Vernant
(1991), 303–317.
Vernant, J. -P. (1996) ‘The refusal of Odysseus’, in Schein (ed.), 185–189.
Versnel, H. S. (1987) ‘What did ancient man see when he saw a god? Some
reflections on Greco-Roman epiphany’, in D. van der Plas (ed.), Effigies Dei:
Essays on the History of Religions. Leiden, 42–55.
Versnel, H. S. (2011) Coping with the Gods: Wayward Readings in Greek
Theology. Leiden.
Veyne, P. (1988) Did the Greeks Believe in Their Myths? P. Wissing (trans.).
Chicago, IL.
Vlastos, G. (1946) ‘Parmenides’ theory of knowledge’, TAPA 77: 66–77.
Vlastos, G. (1970a) ‘Review of F. M. Cornford: Principium Sapientiae’, in
Furley and Allen (eds), 42–55.
Vlastos, G. (1970b) ‘Theology and philosophy in early Greek thought’, in Furley
and Allen (eds), 92–129.
Vlastos, G. (1973) ‘Degrees of reality in Plato’, in his Platonic Studies.
Princeton, NJ, 58–75.
Vlastos, G. (2008) ‘“Names” of being in Parmenides’, A. P. D. Mourelatos (ed.),
in Mourelatos (2008a), 367–388.
Wade-Gery, H. T. (1949) ‘Hesiod’, Phoenix 3: 81–93.
Wainwright, W. J. (1974) ‘Belief and conceptual relativity’, The Journal of
Religion 54: 305–312.
Walcot, P. (1963) ‘Hesiod and the law’, Symbolae Osloenses 38: 5–21.
Wallace, R. W. (1994) ‘Private lives and public enemies: freedom of thought in
Classical Athens’, in A. L. Boegehold and A. C. Scafuro (eds), Athenian
Identity and Civic Ideology. Baltimore, MD, 127–155.
Walsh, G. B. (1984) The Varieties of Enchantment. Chapel Hill, NC.
Warden, J. R. (1971) ‘The mind of Zeus’, JHI 32: 3–14.
Wardle, D. (2006) Cicero: On Divination Book 1: Translated with Introduction
and Commentary. Oxford.
Warren, J. (2000) ‘Epicurean immortality’, OSAP 18: 231–261.
Warren, J. (2007) Presocratics. Stockfield.
Warren, J. (2013) ‘Gods and men in Xenophanes’, in Harte and Lane (eds),
294–312.
Wedin, M. V. (2014) Parmenides’ Grand Deduction. Oxford.

385
 Bibliography
West, M. L. (1966) Hesiod: Theogony: Edited with Prolegomena and
Commentary. Oxford.
West, M. L. (1971) Early Greek Philosophy and the Orient. Oxford.
West, M. L. (1978) Hesiod: Works and Days: Edited with Prolegomena and
Commentary. Oxford.
West, M. L. (1983) The Orphic Poems. Oxford.
West, M. L. (1997) The East Face of Helikon: West Asiatic Elements in Greek
Poetry and Myth. Oxford.
West, S. (1991) ‘Herodotus’ portrait of Hecataeus’, JHS 111: 144–160.
Whitmarsh, T. (2015) Battling the Gods: Atheism in the Ancient World.
New York, NY.
Wiebe, D. (1979) ‘The role of “belief” in the study of religion: a response to W.C.
Smith’, Numen 26: 234–249.
Wiesner, J. (1997) ‘Wissen und Skepsis bei Xenophanes’, Hermes 125:
17–33.
von Wilamowitz-Moellendorff, U. (1916) Die Ilias und Homer. Berlin.
von Wilamowitz-Moellendorff, U. (1959) Der Glaube der Hellenen, 3rd edn,
2 vols. Basel.
Wildberg, C. (2011) ‘Dionysos in the mirror of philosophy: Heraclitus, Plato, and
Plotinus’, in R. Schlesier (ed.), A Different God? Dionysos and Ancient
Polytheism. Berlin, 205–232.
Willcock, M. M. (1978) The Iliad of Homer Books I-XII: Edited with Introduction
and Commentary. Basingstoke.
Wise, K. P. (1999) ‘Geology’, in J. F. Ashton (ed.), In Six Days: Why 50 Scientists
Choose to Believe in Creation. Sydney, 328–332.
Wittgenstein, L. (1966) Lectures and Conversations on Aesthetics, Psychology
and Religious Belief, Compiled from Notes taken by Yorick Smythies, Rush
Rhees and James Taylor. C. Barrett (ed.). Oxford.
Woodbury, L. (1971) ‘Parmenides on names’, in Anton and Kustas (eds),
145–162.
Woodbury, L. (1986) ‘Parmenides on naming by mortal men’, A. C. Bowen (ed.),
Ancient Philosophy 6: 1–13.
Woodruff, P. (2004) ‘Antiphon, Sophist and Athenian’, OSAP 26: 323–336.
Wright, M. R. (1995) Empedocles: The Extant Fragments. London.
Wyatt, Jr., W. F. (1992) ‘The root of Parmenides’, HSPh 94: 113–120.
Yonezawa, S. (1989) ‘Xenophanes: his self-consciousness as a wise man and Fr.
34’, in Boudouris (ed.), 432–438.
Yunis, H. (2011) Plato: Phaedrus. Cambridge.
Zaidman, L. B. and Pantel, P. S. (1992) Religion in the Ancient Greek City.
P. Cartledge (trans.). Cambridge.
Zeller, E. (1919) Die Philosophie der Griechen in ihrer geschichtlichen
Entwicklung, I. Teil, I. Leipzig.
Zhmud, L. (2012) Pythagoras and the Early Pythagoreans. K. Windle and
R. Ireland (trans). Oxford.

386
 INDEX LOCORUM

This index does not list exhaustively all of the passages that are merely referred to or
adduced as parallels in the book.

Aelius Aristides A17, 25

Or. 22.2–3, 268 A23, 20
Aeschylus A27, 148
Pers. 547, 211 B1, 20–21
Pr. 447–8, 195 Anaximenes
Pr. 499, 195 A7, 25
Pr. 975, 51 A7.1, 24–25
Supp. 100–3, 138 A7.4, 201
Th. 81, 352–353 A10, 24, 25
TrGF 3.68–73, 256 A15, 21
Aëtius B2, 21, 244
1.3.4, 21 Andocides
1.6.9, 36, 55 Myst. 30.10, 255
1.6.14–15, 55 Antiphon
1.7.2, 44 A9, 111
1.7.11, 20 Apollonius Rhodius
1.7.12, 25 scholia in Apollonium Rhodium
2.20.6, 30 Σ 77.15–16, 270
5.1.1, 104–106 Apuleius
5.1.4, 105 Met. 11.21.26–7, 269
Alcaeus Met. 11.23.21–31, 269
fr. 450 Campbell, 30 Met. 11.24.1–18, 269
Alcmaeon Aratus
A12, 244 Phain. 1.5–6, 110
B1, 115, 130 Archilochus
B2, 244 E1 II.50–1, 275
Alexander fr. 131 W, 182
in Metaph. 31.9–14, 220 Aristophanes
Anaxagoras Nub. 365–7, 44
A112, 229 Nub. 423–6, 44
B12, 22, 177 Nub. 1241, 255
B14, 290 Ran. 454–6, 269
Anaximander Aristotle
A11.5, 201 Cael. 294a21-8, 319
A11.6, 232 de An. 404a1-6, 233
A15, 20, 22, 25 de An. 404a27-31, 240

387
 Index Locorum
Aristotle (cont.) Div. 1.5.9, 115
de An. 405a18-b1, 38 Div. 1.6.11, 112
de An. 405a19-21, 23 Div. 1.13.23, 131
de An. 407b20-3, 244 Div. 1.14.24, 113
de An. 411a7-8, 20, 23 Div. 1.18.34, 112, 115
de An. 427a17-b14, 192, 240, 330–331 ND 1.25, 20, 25
EN 1177b31-1178a3, 251 ND 1.28, 231
fr. 15 Rose, 284 ND 2.168.3–4, 50
fr. 61 Rose, 277 ND 3.5–10, 50
fr. 192 Rose, 275 ND 3.29–52, 46
GA 736b29-38, 245 Clement
GC 318b6-7, 186 Protr. 2.21.2, 267
GC 325a13ff, 193 Strom. 3.3.14, 238
Metaph. 982b18-19, 55–56
Metaph. 984b3-4, 220 Democritus
Metaph. 984b11-19, 38 A128, 175
Metaph. 984b15-22, 277 B68, 148
Metaph. 984b22-985a10, 38, 55, B299.6–8, 148
319 Derveni papyrus
Metaph. 985a10-21, 38 col.14–17, 17
Metaph. 986b21-2, 314 col.17.12, 140
Metaph. 986b21-7, 143 col.19, 17
Metaph. 986b26-34, 193, 220 col.19.10, 140
Metaph. 986b33-987a2, 186 col.20, 284
Metaph. 988a14-17, 38 col.23, 17
Metaph. 988a33-34, 38 Diogenes Laertius
Metaph. 1000a5-19, 55 1.43–4, 148
Metaph. 1009b12-1010a3, 190–192, 240, 1.111, 151
330–331 2.40, 43
Metaph. 1009b13-15, 183–184 2.117, 46
Phys. 203b10-15, 20, 25 8.21, 276
Poet. 1460b10-11, 149 8.36, 152–153
Poet. 1460b35-1461a1, 149 8.41, 276
Rhet. 1377a19, 141 8.54, 323
Rhet. 1399b6–9, 141 9.18–20, 151, 152
Athenaeus 9.18.5–6, 314
12.526c, 154 9.18.11, 148
Augustine 9.20, 154
Civ. Dei 6.5, 36 9.20.6–7, 314
Civ. Dei 7.17, 115 9.21, 274–275
Civ. Dei 8.2, 24 9.21.1–3, 314
9.21.3–8, 314
Cebetis Tabula 9.22, 164
2.2.2–5, 275 9.22.11–12, 55
Cicero 9.52, 42
Div. 1.1.1, 105, 115 10.2, 56
Div. 1.2.3, 109 Diogenes of Apollonia
Div. 1.3.5, 104–106 A19.42, 244
Div. 1.3.6, 105 B5, 22

388
 Index Locorum
Empedocles B137, 322, 344
A85, 231 B146, 322
B1, 335 B147, 322
B2, 323–325, 330, 334 Strasb. a (ii) 21–30, 329–330
B2.3, 325 Strasb. d, 320
B2.8b-9, 337 Strasb. d 5–6 = B139, 321–322,
B3.3–5, 333–334 332–333, 336
B3.9–13, 328 Epicharmus
B4, 334, 335, 342 B9, 228, 229
B4.2–3, 335 B12, 194–195
B6, 334 B20, 251
B11, 326–327 B22, 228
B15, 327 Epicurus
B17.14, 325 Ep. ad Menoec. 123–4, 45
B17.20–6, 328–329 Sent. 1, 45
B17.26, 319 Epimenides
B21.12, 321, 325, 337 B1, 75
B23, 335 Epiphanius
B23.8, 325 Advers. Haeres. 2.2.9, 128, 130
B39, 319, 324, 330 Etymologicum Magnum
B62.1–3, 327 574.69–75, 127–128
B62.6, 175 Euripides
B82, 329–330 Bacch. 72–4, 255
B84, 330 Hel. 1013–16, 229, 230, 236–237, 241
B106, 191, 330–332 Ion 532ff, 113
B108, 191, 330–332, 333 Med. 145–7, 237
B109, 175, 178, 333 Med. 994, 237
B110, 325 Supp. 531–4, 229
B110.10, 174, 333 TrGF 5.286, 45
B111, 335 TrGF 5.370.71–2, 228–229, 230
B112.4, 273, 335 TrGF 5.506, 45
B113, 335 TrGF 5.771–86, 256
B115, 321, 326 TrGF 5.795, 129, 130
B115.3, 337 TrGF 5.839.1–2, 9–12, 229
B115.5, 325 TrGF 5.913, 44
B115.13–14, 320, 321, 333, 336 Euripides (?)
B117, 321, 326 TrGF 2.624, 45
B118, 238 Eusebius
B126, 337 PE 3.9.2, 140
B127, 321 PE 13.14.1.1–3.1, 50
B128, 320, 322, 344
B129, 151, 161, 275, 323, 324, 326, 332 Georgius Choeroboscus
B129.2–4, 325 in Theod. 4.1 200.3–5 Hilgard, 127–128
B131, 319, 333–334 in Theod. 4.2 88.27–33 Hilgard, 124
B131.1–2, 335 Gold Tablets
B132, 323 GJ 1, 272
B133, 331 GJ 1.15–16, 272
B134, 319 GJ 2.6, 245, 273
B136, 322, 344 GJ 2.11, 275

389
 Index Locorum
Gold Tablets (cont.) 1.53, 113–114
GJ 3.4, 273 1.60, 31
GJ 5–8, 272 1.182, 31
GJ 5.3, 245, 273 2.3, 47
GJ 5.5, 237–238, 243, 273 2.53, 53
GJ 5.9, 273 2.55–7, 30
GJ 6.3, 245, 273 2.123, 244
GJ 6.7, 272 2.143, 30
GJ 7.3, 245, 273 4.76, 148
GJ 7.7, 272 4.94–6, 276
GJ 8.12–15, 245, 273 6.75, 31
GJ 29.3–4, 245, 273 6.84, 31
Gregory of Nyssa 7.6, 145
ad Ablab. 3.1.38.19–39.7, 50 7.12–19, 31
7.137, 31
Hecataeus, 30 7.142, 266
F1, 29 7.142.3–143.1, 111
F26, 30 8.122, 109
F27, 30 9.33, 114
F35a, 33 9.92–4, 112, 195
F35b, 30 9.100, 31
F300, 30 9.120, 125
F302d, 30 Hesiod
T4, 30 fr. 278 M-W, 109
T12a, 148 fr. 278.5 M-W, 69
Heraclitus fr. 311 M-W, 256
B32, 204 Op. 1–10, 95–97
B34, 195 Op. 10, 69
B36, 236 Op. 11, 93
B40, 55, 322–323 Op. 11–26, 84
B41, 22 Op. 26, 318
B48, 204 Op. 42–105, 88–90
B57, 53 Op. 78, 66, 71
B61, 326 Op. 85–9, 92
B64, 22 Op. 106, 57
B67, 204 Op. 106–201, 317–318
B93, 48, 114, 266 Op. 108, 317
B101, 266 Op. 127–55, 70
B117, 233 Op. 178–9, 86
B118, 161, 233 Op. 182–3, 70
Herodian Op. 208–9, 83, 318
de Pros. Cathol. 3.1 535.28–35 Op. 218, 122
Lentz, 124 Op. 267–73, 34–35, 85–86, 100
περὶ Διχρόνων 3.2 16.17–29 Lentz, 124 Op. 282–5, 66
Herodotus Op. 293–5, 96
1.1–5, 30 Op. 323, 183
1.6–91, 31 Op. 373–5, 93, 183
1.30, 148 Op. 483–4, 100
1.32, 117–118, 122 Op. 635–8, 86

390
 Index Locorum
Op. 646–93, 97–101, 310 Th. 337–40, 30
Op. 660–2, 131 Th. 385–8, 58
Op. 685, 183 Th. 411–28, 84
Op. 708–9, 66 Th. 429–47, 83
Op. 788–9, 66 Th. 448–52, 84
scholia in Hesiodum Th. 468–96, 87
Σ Op. 130–1, 130a, 55 Th. 494, 90
Σ Op. 270–3, 35 Th. 498–500, 87–88
Σ Op. 661a, 100 Th. 510–14, 89
Σ Th. 23, 74 Th. 511b-14, 92
Σ Th. 535.1–2, 88, 90 Th. 535–616, 88–90
Th. 1–4, 84 Th. 537, 183
Th. 1–21, 61 Th. 653–7, 87
Th. 11–21, 74 Th. 721–819, 254–255, 359
Th. 22, 64, 73, 98 Th. 730–1, 87
Th. 22–3, 61 Th. 736–8, 352
Th. 24, 74 Th. 744–54, 355, 358
Th. 24–8, 61 Th. 746, 352
Th. 26–8, 61–103, 309–310, Th. 759–62, 351
311–314 Th. 760–1, 257
Th. 29, 75, 91 Th. 780–1, 119
Th. 29–34, 75 Th. 782–806, 66
Th. 30–1, 75–76 Th. 783–805, 87
Th. 31-2a, 79–80, 90, 318 Th. 807–9, 352
Th. 32, 76–78 Th. 811, 352
Th. 33, 77 Th. 837–8, 87
Th. 36–115, 73 Th. 869–80, 84
Th. 38, 76 Th. 881–5, 84
Th. 49, 58 Th. 886–900, 87
Th. 60, 84 Th. 902–3, 354
Th. 71–4, 84 Th. 904, 57
Th. 79, 319 Th. 940–2, 261, 263
Th. 80–92, 76 Th. 949, 263
Th. 93–5, 86 Th. 950–5, 263
Th. 95–100, 318 Th. 965–1022, 261
Th. 99–101, 78 Hippocratic corpus
Th. 105–6, 273 Aer. 22, 29
Th. 112–13, 84 Carn. 2, 245
Th. 114–15, 64 Hebd. 52, 236
Th. 115–16, 81 Morb. 4.2–3, 175
Th. 116, 56, 74 Morb. Sacr. 1, 27
Th. 174–7, 87 Morb. Sacr. 1–2, 27
Th. 175, 90 Morb. Sacr. 2, 27
Th. 207–10, 204 Morb. Sacr. 3–4, 28
Th. 217, 57 Morb. Sacr. 4, 28
Th. 225–32, 66, 71, 93 Morb. Sacr. 5, 27
Th. 231–2, 66 Morb. Sacr. 5–16, 27
Th. 233–6, 66, 68, 71, 90 Morb. Sacr. 10, 27
Th. 262, 90 Morb. Sacr. 14, 27

391
 Index Locorum
Hippocratic corpus (cont.) Il. 11.27–8, 119
Morb. Sacr. 16, 27 Il. 12.23, 262
Morb. Sacr. 21, 27 Il. 12.211–29, 111
Nat. Hom. 3, 236 Il. 12.315, 218
Nat. Hom. 6, 175 Il. 14.313–25, 261
Vict. 4.87, 28–29 Il. 15.158–9, 119
Vict. 4.89–90, 28–29 Il. 15.461, 138
VM 1, 130 Il. 16.53, 69
Hippolytus Il. 17.547–8, 119
Haer. 1.1.2, 20 Il. 18.115–21, 140, 263
Haer. 1.7.1, 24–25 Il. 18.128–9, 69
Haer. 1.26.1–2, 55 Il. 18.329, 69
Homer Il. 21.463–6, 232
Il. 1.62–3, 109 Il. 22.281, 75
Il. 1.69–70, 76–77, 109 Il. 23.64–107, 161
Il. 1.92–100, 109 Il. 23.304–50, 94
Il. 1.365, 109 Il. 23.440, 69
Il. 1.384–5, 109 Il. 24.222, 66
Il. 1.528–30, 138 Il. 24.527–33, 84
Il. 2.46, 75 Il. 24.592–5, 161
Il. 2.100–8, 75 Od. 1.1–4, 148, 264–265
Il. 2.299–300, 76 Od. 2.299–330, 110
Il. 2.303–30, 109 Od. 3.247–316, 68
Il. 2.335, 262 Od. 4.157, 69
Il. 2.353, 110 Od. 4.238–79, 94
Il. 2.484–93, 82 Od. 4.379, 109
Il. 2.485–6, 63, 78, 131 Od. 4.389–90, 109
Il. 3.65–6, 86 Od. 5.35, 261
Il. 3.108–10, 322 Od. 5.135–6, 265
Il. 3.277, 175 Od. 5.196–7, 133–134
Il. 3.396, 183 Od. 5.334–5, 263
Il. 4.200, 183 Od. 7.255–8, 265
Il. 4.235, 66 Od. 8.44–5, 260
Il. 4.370–405, 65–66 Od. 8.63–4, 194
Il. 5.1–3, 263 Od. 8.479–81, 265
Il. 5.121–32, 263 Od. 8.487–91, 81–82
Il. 5.438–42, 263 Od. 10.82–6, 354
Il. 5.748–52, 354, 357 Od. 10.138, 264
Il. 5.769, 357 Od. 10.302–3, 178
Il. 5.778, 69–70 Od. 10.490–5, 109, 194
Il. 5.815–61, 263 Od. 10.537–40, 109, 264
Il. 5.845, 233, 242, 348 Od. 11.16, 175
Il. 6.383–9, 67 Od. 11.50, 264
Il. 7.44–53, 112 Od. 11.89, 264
Il. 8.392–6, 354, 357 Od. 11.99–100, 266
Il. 9.115, 66 Od. 11.109, 175
Il. 9.312, 259 Od. 11.363–9, 94
Il. 9.508–9, 122 Od. 11.543–64, 161
Il. 10.482, 80 Od. 11.601–4, 263

392
 Index Locorum
Od. 12.186–91, 93–94 Homeric Hymn to Helios
Od. 12.267, 194 7, 261
Od. 12.323, 175 19, 122–123
Od. 13.109–12, 247, 257 Homeric Hymn to Heracles
Od. 14.296, 66 7–8, 263
Od. 14.379–87, 65 Homeric Hymn to Hermes
Od. 15.160–78, 111 54–9, 82
Od. 15.172–3, 113 368–9, 102
Od. 15.251–3, 115 427–33, 82
Od. 15.525–34, 112 560–4, 67, 77, 83
Od. 15.536, 129
Od. 16.136, 183 Iamblichus
Od. 17.153–4, 129 Protr. 47.25, 238
Od. 17.163, 129 VP 82–6, 238
Od. 17.218, 175 VP 143, 276
Od. 17.382–5, 113, 153 Inscriptiones Graecae
Od. 17.514–21, 94 I2 945, 228, 229
Od. 18.136–7, 182 II/III2 3661.5–6, 270
Od. 19.203, 65, 68, 70, 71, 94 II/III2 11466, 229–230
Od. 19.309, 129 II/III2 12599, 229–230
Od. 19.409, 206 Ion of Chios
Od. 19.547, 129 B4, 244
Od. 19.562–7, 69 Isocrates
Od. 20.195, 197 4.28.6–7, 267
Od. 20.351–62, 112 4.28.10–12, 270
Od. 21.406–9, 94
Od. 22.347–8, 80–81 Leucippus
Od. 22.420–5, 68 A1, 175
Od. 23.334–7, 265 Lucian
Od. 24.12, 257, 351 DDeor. 24.2, 256
Od. 24.258–9, 68 Lucretius
Hymni Homerici 3.121–9, 236
Homeric Hymn to Aphrodite 6.68–79, 46
82, 71, 94 Lycophron
255–79, 261 scholia in Lycophronem
Homeric Hymn to Apollo Σ Lycoph. 344, 118
131–2, 109, 129 Lysias
156–64, 94 6.51, 267
539a, 129
Homeric Hymn to Demeter Melissus
147–8, 86 B6, 287
216–17, 86
226, 272 Nonnus
235–42, 270–271 Dion. 38.192, 256
260–3, 270–271
480–2, 70, 267, 268, 270–271 Olympiodorus
Homeric Hymn to Dionysus in Alc. 69.21–70.4, 112
1–9, 34, 65, 93 in Phd. 1.3, 246

393
 Index Locorum
Orphic fragments B1.9, 254, 256, 258, 348, 349, 354,
fr. 32.iv.1–3 Kern = 568.1–3 355, 358
Bernabé, 273 B1.10, 357
fr. 168.15 Kern = 243.15 Bernabé, 257 B1.10–11, 354
fr. 226 Kern = 437 Bernabé, 236, 246 B1.11, 255, 256, 298, 350, 351, 355, 356
fr. 228a Kern = 436 Bernabé, 236, 246 B1.11–13, 248, 256, 352–353, 355, 357
fr. 229 Kern = 348 Bernabé, 238, B1.12, 255
243, 273 B1.14, 255, 351
B1.14–17, 354
Palaephatus B1.15–17, 260, 267
de Incred. pref. 9–12, 32 B1.16–20, 353
de Incred. pref. 22–4, 32 B1.17–18, 255
de Incred. 1, 31–32 B1.19, 255, 351
de Incred. 6, 32 B1.20–1, 353–354, 357
de Incred. 15, 32 B1.21, 265
Parke B1.22, 268, 272, 308, 315
oracle consultations at Dodona B1.22–3, 259
266 no.11, 109 B1.22–8, 313
272 no.27, 109 B1.24, 10, 255, 256, 258, 260, 267,
Parmenides 315, 354
A1.9–10, 231–232 B1.24-8a, 313
A1.11–12, 232, 240 B1.25, 351, 355, 356, 358
A16-17, 266 B1.26, 238, 274
A32, 239 B1.26–8, 255–256, 258, 259, 267,
A37, 164, 231, 239, 248, 357 356, 358
A40a, 164 B1.27, 272, 282, 298, 317, 358
A42, 165, 341 B1.28b, 164, 210, 218, 266, 289
A44a, 164 B1.28b-30, 155, 159, 167, 168, 208, 215,
A45, 232, 235, 241 217, 220–221, 259, 278, 279, 312
A46, 164, 169–175 B1.29, 10, 156, 296, 300–301, 308,
A46a-b, 164, 174, 231, 235, 249 312, 315
A47, 178 B1.29–30, 200
A51, 232 B1.30, 156, 164, 166, 167, 198, 210, 212,
A52, 164 215, 218, 219, 220, 282
A52-3, 249 B1.31–2, 209–215, 217, 299–300,
A53-4, 164, 165, 237 301, 302
B1, 10, 236, 252–261, 265, 277, B2, 278, 287, 294–295, 341
347–348, 350–359 B2.1, 266, 283
B1.1, 260, 265 B2.2, 266, 305
B1.1–5, 259–260, 354 B2.3, 298
B1.2–3, 272, 277–278 B2.6, 289
B1.3, 252, 255, 256–257, 258–259, 267, B2.7–8, 289
315, 355, 356, 358 B3, 304, 305, 306–307
B1.4, 256 B4, 221, 282, 292, 293
B1.5, 272, 354 B6, 198, 219–220
B1.7, 256 B6.1-2a, 304
B1.7–8, 256, 351 B6.2, 266, 282, 296
B1.8b, 238, 260 B6.3, 266
B1.8b-10a, 350–351, 354 B6.4–6, 197, 198, 216, 265, 307

394
 Index Locorum
B6.7, 190, 194 B8.51b-2, 167, 319
B6.9, 267, 298 B8.52, 166, 199–203, 215
B7, 221 B8.53–5, 188–190, 193, 203, 207, 216,
B7.2, 266, 282 218–219
B7.2–6, 266 B8.54, 197, 204, 206, 207,
B7.3–6, 194, 218–219 265, 298
B7.5–6, 184–185, 266, 281–282, 313, B8.56, 231
340, 341 B8.56–9, 175, 187, 188, 297, 348
B8, 12, 212, 281, 287, 288, 293, B8.57–9, 186, 187, 225
294–295, 341 B8.59, 185–186, 231, 249, 291, 355
B8.1–2, 294 B8.60–1, 164, 168, 201–202, 217–218,
B8.2, 266, 297 219, 220, 342
B8.3–7, 198 B9, 164, 207–208, 348
B8.4, 156, 315 B9.3, 298, 349
B8.6, 266 B9.3–4, 173, 177, 186
B8.6–21, 287 B10, 166, 168, 200–201, 208, 216, 217,
B8.7–9, 289, 304 220, 230, 286, 307, 342
B8.8, 266 B10.1, 178
B8.15–16, 159, 184–185, 281, 294, B10.4–5, 178, 289
298, 300 B10-11, 164, 258
B8.17, 205, 208 B11, 230–231
B8.17–18, 156 B12, 22, 164
B8.21, 315 B12.2, 186
B8.21–2, 198, 287 B12.3, 248, 315
B8.22–5, 287, 291 B12.3–6, 237–239
B8.22–31, 290 B12.4, 349
B8.25, 205 B13, 164, 238–239, 319, 344
B8.26, 186 B14, 258–259
B8.26–31, 198, 287, 291, 315 B14-15, 164, 165, 201, 341, 357
B8.27–8, 287 B15a, 164
B8.28, 156 B16, 164, 175–184, 196, 198, 199, 203,
B8.29, 186, 225 216, 217, 222–224, 239, 240, 241,
B8.34, 304, 305, 306–307 258, 282, 286, 307
B8.35-6a, 266, 304, 305–307 B16.1, 186, 265
B8.36b-41, 204–207, 297, 300 B16.1–2, 197, 198, 221
B8.37, 206 B16.2–3, 198
B8.38, 287, 291 B16.3, 298
B8.39, 198, 200, 315 B16.4, 298
B8.39–41, 218–219 B17-18, 237, 286
B8.40, 187, 208, 298 B18, 164, 239
B8.41, 186, 198, 287, 291 B19, 164, 203–204, 298
B8.42–9, 290 B19.3, 198, 206, 207, 218–219, 282
B8.43, 315 Pausanias
B8.44b-8, 291 1.34.4, 115
B8.49, 186, 225 4.10.6, 111
B8.50–1, 156, 282, 296, 300, 301 6.9.8, 275
B8.50–2, 199–200, 266, 312 6.17.6, 115
B8.50–9, 185, 198, 207 Pherecydes
B8.51, 164 F86b Schibli = A5, 244

395
 Index Locorum
Philodemus Phd. 80e-81a, 269, 270
Piet. 737–51, 46 Phd. 96a5-99d2, 37
Philoponus Phdr. 229d-e, 32
in de An. 86.30–1, 20 Phdr. 244b6-d5, 112
in Ph. 16.22.6–7, 201 Phdr. 244c, 127
in Ph. 21.30–22.15, 301 Phdr. 250b6-8, 268
Philostratus Phlb. 28c-d, 22
VA 8.19.40–4, 109 Phlb. 30d7-8, 22
Pindar Rep. 2.364e, 153
fr. 131 M, 245 Rep. 2.382d2-3, 72, 311
fr. 133 M, 243, 246 Rep. 5.478d5-7, 302
fr. 137 M, 269, 270 Rep. 6.500c8-d1, 251
N 6.1–2, 245 Rep. 10.607b1-6, 60
O 2.70, 257 Rep. 10.614c2, 247
O 2.85, 255 Rep. 10.615d5, 247
O 6.22–7, 265 scholia in Platonem
Plato Σ Lg. 715e, 140
Apol. 18b-c, 43 Soph. 242d4-6, 314
Apol. 19c, 44 Symp. 178b, 239
Apol. 21a, 109 Symp. 199a1-2, 255
Apol. 21b6-7, 340 Tht. 176a8-b2, 251
Apol. 22a, 127 Tht. 179e3-4, 314
Apol. 23b, 127 Tht. 180e, 208
Apol. 23d, 43 Tim. 40d6-e4, 50–51
Apol. 24b, 43 Tim. 72a1ff, 113
Apol. 26d, 44 Tim. 72b9-c1, 195
Apol. 29c9-d4, 127 Tim. 90b1-6, 250–251
Apol. 40a4-7, 127 Plutarch
Apol. 40b3-c3, 127 Ages. 3.3–5, 111
Cratyl. 398b5-7, 327 fr. 178.5–13 Sandbach, 268–269
Euth. 3c, 131 Lys. 22.5–6, 111
Euth. 6a-c, 51 Mor. 385c, 127
Gorg. 493a1-c3, 276 Mor. 415b-c, 318
Gorg. 523a, 35 Mor. 746b, 146
Lg. 4.715e8–716a1, 140 Mor. 756e10-11, 239
Lg. 10.887c-e, 41 Mor. 975b11, 195
Lg. 10.889a–890a, 44 Mor. 1114b-c, 164
Lg. 10.899b, 23 Mor. 1114d-e, 301–302
Lg. 10.899d4-900b6, 79 Peric. 6, 26
Lg. 10.909b, 153 Peric. 32.2, 42–43
Men. 81a10-c4, 243, 246 Porphyry
Men. 81b3-6, 243 Abst. 3.26.1–4, 244
Men. 91d-e, 42 Antr. 22.6–23.7, 247–248
Men. 92c4-7, 130 VP 30, 151, 323
Men. 99c7-d9, 262 Proclus
Parm. 130e5-134e8, 285 in R. 2.113.6–9, 277
Phd. 66d, 127 Protagoras
Phd. 69c, 269, 270 B4, 45, 47

396
 Index Locorum
ps.-Aristotle OT 298–9, 110, 115
Mu. 401a25ff, 140 OT 484, 110
Pepl. fr. 641.7–8 Rose, 229–230 OT 496–501, 110
ps.-Euripides TrGF 4.837, 267, 268, 270, 272
Rhes. 971–3, 255 Stesichorus
ps.-Plato PMG 185, 257, 351
Ax. 371d5-e4, 273 SVF
PW 1.264, 46
202, 266 1.538, 270
216, 154 2.42, 284

Satyrus Tertullian
vit. Eur. col.10, 42 de Carn. Christ. 5.1–4, 49
Sextus Empiricus de Praesc. Haeret. 7.9–13, 51
M 7.51, 128, 130 Thales
M 9.17–18, 44 A15, 20
M 9.20–1, 245 A22, 20, 23
M 9.49, 46 A23, 20
M 9.54, 44 Theognis
M 9.56, 42 139–41, 198
M 9.57, 42 373–82, 86
M 9.137–90, 46 713–14, 94
M 10.18–19, 56 731–52, 86
PH 3.2, 46 Theophrastus
PH 3.2–12, 46 Sens. 1–2, 169
PH 3.123.1–2, 55 Sens. 2.5–4.8, 169–175
Simplicius Sens. 3.2, 176, 181
in Cael. 7.556.12–14, 220 Sens. 3.2–4, 177
in Cael. 7.557.20–558.17, 301 Sens. 3.3, 183
in Cael. 7.558.8, 203 Sens. 3.3–4, 180
in Cael. 7.559.14–560.1, 303 Sens. 3.4, 249
in Ph. 9.22.26–31, 314 Sens. 4.1, 190, 195, 241
in Ph. 9.24.13–21, 21 Sens. 4.4–6, 180, 241, 249
in Ph. 9.29.18, 208 Sens. 4.5–6, 177, 231
in Ph. 9.30.20–31.7, 187–188 Sens. 4.7, 183
in Ph. 9.38.18–39.21, 235
in Ph. 9.38.30–9.39.18, 233 Xenophanes
in Ph. 9.39.14–19, 238–239 A1, 148, 154, 315
in Ph. 9.39.19–21, 233–236, 238, 239, A6, 152
243, 348–350, 352, 356, 358, 359 A7, 152
in Ph. 9.39.27–9.40.6, 234, 238 A12, 41–42, 79, 141, 315
in Ph. 9.143.10, 208 A13, 150–151, 314, 344
in Ph. 9.144.28, 234 A14, 141
Solon A24, 115
fr. 13 W, 35 A26, 150
Sophocles A28.7, 315
El. 25–6, 122 A30, 143
El. 62–4, 276–277 A31.9, 116

397
 Index Locorum
Xenophanes (cont.) B18, 105, 116–128, 131, 132–134, 316
A32, 141 B18.1, 143–144
A32-3, 118, 145, 148 B18.2, 141, 148, 149, 322
A33, 143 B20, 151
A35, 152 B21, 147, 148
A38, 119, 143 B22, 154
A39, 119 B23, 42, 115, 125, 139, 141, 150
A41, 143 B23-6, 125, 319
A41a, 135 B24, 116, 137, 150, 315
A43, 119, 143 B25, 21, 116, 122, 137–139, 143, 145,
A44, 119 150, 315
A45, 119 B26, 138, 141, 142, 143, 315
A46, 119 B27, 120
A52, 26, 104–106, 120, 144 B28, 134, 319
B1, 343–344 B29, 120, 139
B1.13–16, 141 B30, 119
B1.17–18, 152 B32, 119, 134, 143
B1.21–3, 147, 149 B33, 139
B1.24, 125, 141 B34, 60, 115, 128–133, 141, 146,
B2, 45, 147–149 150, 316
B2.6, 134 B34.2, 125, 130, 146
B2.10, 120 B34.3, 138, 145, 147
B2.11–14, 148, 152 B34.4, 41, 140, 146
B2.19–22, 153, 154 B35, 42, 146–147, 311–312
B3, 142, 147, 148, 149, 153, 154 B36, 124–126, 133–134, 136, 137, 139,
B5, 153 140–141, 145
B7, 147, 148, 151, 152–153, 161, B38, 121–122, 134–136, 139, 141, 150
244, 316 B45, 45, 104, 153, 346
B8, 45, 104, 148, 151–152, 153, 346 Xenophon
B9, 152 Anab. 5.6.29, 110–111
B10, 147 Anab. 6.1.31, 110–111
B10-16, 42, 119, 142, 343 Cyr. 1.6.2, 110
B11, 344 Cyr. 1.6.7, 211
B11.1, 120 Cyr. 1.6.46, 108, 110
B11-12, 125, 141, 143, 148, 312 Hell. 3.3.3, 111
B12, 149 Hipp. 9.9, 110
B14, 41, 143, 147, 148, 312, 343 Mem. 1.1.1, 43
B14.1, 315 Mem. 1.1.9, 110
B14-16, 125 Mem. 1.4.2–18, 45, 141
B15, 326 Mem. 1.4.15–18, 108
B15-16, 142–143 Mem. 4.3.11–13, 141
B16, 143, 148 Mem. 4.3.13, 117–118
B17, 148 Symp. 4.47–9, 108

398
 GENERAL INDEX

Abaris, 277 on Parmenides, 183–184, 186, 190–193,

Aelius Aristides, 268 220, 249–250
Aeschylus, 51, 138, 195, 256, 339 on vital heat, 245
Aithalides, 277 Athena, 28, 80, 89, 228–229, 259,
Alcmaeon, 38, 115, 130, 170, 244 262–263, 354
alêthea, and cognates
in epic poetry, 67–68 becoming like god (in philosophy),
in Parmenides, 156, 296 250–251, See also Parmenides:
Ameinias, 274–275 and becoming like god
Anaxagoras, 38, 339 Betegh, G., 17, 236
and perception, 169, 170, 178, 191 blindness and deafness, 194–195
and religion, 26, 42–44 Bryan, J., 147, 202
Mind in, 22, 37–38, 277, 290 Burkert, W., 8–9, 153, 243, 270, 352–353
Anaximander. See Milesian, the Burnet, J., 10, 11, 16
philosophers
Anaximenes. See Milesian, the Calchas, 76–77, 109, 110, 113
philosophers Cicero
Antiphon, 111 his character Cotta on traditional Roman
Aphrodite, 33, 71, 89, 94, 239, 261, 328 religion, 50
Apollo, 26, 28, 110, 115, 263, 276, 277, See on divination, 104–105, 109, 112, 113,
also Heraclitus: on Apollo 115, 131
Delphic Apollo. See divination: the Cole, T., 67
Delphic oracle Colotes. See Parmenides: Colotes on
approximation to the divine (in Homeric Cornford, F.M., 16
and Hesiodic poetry), 251, Coxon, A.H., 189
261–264, 270–271, Croesus, 113–114
317–318, 345 Curd, P.K., 18, 359
Apuleius, 269
Aristeas, 277 Delian maidens, 94
Aristophanes, 42, 44, 45, 46, 269 Delphic oracle. See divination: Delphic
Aristotle oracle
on becoming like god, 251, 281, 283 Demeter, 276, See also Homeric Hymn to
on early Greek philosophy, 37–38 Demeter
on Parmenides and others on Democritus, 7, 170, 178, 233, 240, 241, 277
perception and cognition, Denyer, N., 108
183–184, 190–193, 240, Derveni papyrus, 17, 140, 236, 284
330–331 Dillery, J., 153
physiologoi and theologoi, 38–39 Diogenes of Apollonia, 21, 22, 38,
on Hesiod, 55–56 170, 244
on initiation, 284 Diomedes, 262–263

399
 General Index
Dionysus, 238, 246, 261, 268, See also Empedocles
Homeric Hymn to Dionysus and Hesiod, 59, 319, 335, 337–338
Diopeithes, decree of, 42–43 and Parmenides, 319, 335, 337–338
divination, 14, 15, 26, 31, 45, 108–116, and Xenophanes, 319, 337–338
311, 345 daimonology and metempsychosis in,
and approximation to the divine, 238, 320–327, 330–333,
109–110, 114–115 335–337, 338
and knowledge-claims, 109–110, the daimôn as material, 337
114, 128 his attitude to divine disclosure,
and rationality. See rationality: and 334–335, 337–338, 342
divine disclosure: in divination his Muse, 319, 333–335, 337, 342
and zêtêsis, 127–128 ideas of consciousness in, 332–333
as conjectural, 110–114, 327, 340 Love and Strife
itinerant diviners, 153 exerting moral as well as physical
mantic families, 112, 114–115 influence, 320, 332–333, 336
practised by amateurs, 110–111, 113 Love as epistemic object,
technical vs inspired, 112 328–330, 331
the Delphic oracle, 26, 31, 48, 87–88, the question of corporeality, 290
111, 113–114, 127, 154, on experience and wisdom, 151, 320,
265–266, 340 322–333
Xenophanes on. See under Xenophanes on perception and understanding, 169,
divine disclosure, 6–7, See also Hesiod: the 170–171, 191, 241, 325,
Muses’ address, and entries 328–332, 333
under Empedocles, rejecting animal sacrifice, 322, 337, 344
Parmenides, Xenophanes self-identity in, 335–337
and rationality. See rationality: and theology and epistemology in, 6,
divine disclosure 318–339
divine revelation. See divine disclosure views on religion and philosophy in his
Dodds, E.R., 14, 16, 104, 263–264 thought, 17, 320
double motivation, 80, 262–263 Epicharmus, 194–195, 228, 230, 251
Doxa. See Parmenides: Doxa Epicurus and Epicureanism, 45–46, 56,
Doxastic thing, 156 104, 105, 251
epikleseis, 33
early Greek philosophy Epimenides, 75, 148, 151, 277
and religion. See religion, Greek: and his longevity, 151, 322
philosophy epistemically significant interactions. See
as a historiographical category, 54, 158, divine disclosure
344–345 epistemology, early Greek. See early Greek
as mechanistic and non-teleological. See philosophy: early Greek
Plato: on early Greek epistemology
philosophy eschatology. See mystery initiations and
early Greek epistemology entries under Empedocles,
relation to theology, 1, 53, 309, Euripides, Homer,
310–311, 337–339, 346 Parmenides, Pindar,
requiring an interdisciplinary Pythagoras and the
approach, 7–9, 346 Pythagoreans
scholarly schisms about, 7, 11–12, aethereal, 228–230, 242, 244,
17–18, 346 245, 356
Eleusinian mysteries. See mystery initiations etyma, and cognates, 68–69

400
 General Index
Euripides, 42, 44, 45, 46, 113, 256 echoes of divinatory language in, 76–77
eschatological ideas in, 228–229, 230, excursus on seafaring, 3, 97–101,
236–237, 241 131, 310
expressing an epistemological
faith. See religious belief framework, 53, 101–103,
309–310, 339–340
Gemelli Marciano, L., 18 his narrative of human Races, 57, 58,
Gernet, L., 162 317–318
gifts, divine, 86 his poetic persona, 64, 312
godlikeness. See approximation to the his staff, 75–76
divine, becoming like god on female and male, 84, 86–90, 91,
Gold Tablets, 237–238, 243, 244–245, 270, 93, 313
271–274, 345 on gods and natural, psychological and
Gould, J., 41 social phenomena, 56, 57–58
Gregory of Nyssa, 50 on Hecate, 83, 84, 85, 93
on Prometheus and Pandora, 57, 58, 66,
Harrison, T., 41 88–90
Hecataeus, 29–30, 33, 55, 148, 323, 343 on Zeus, 34–35, 57–58, 61, 74, 84–90,
Hecate. See Hesiod: on Hecate 91, 95–98, 99–101
Helios and Heliades, 256–257, 258–259, the Muses’ address, 2–3, 61–103,
264, 269, 315, 347, 351, 352, 309–310, 311–314, 317,
356–358 339–340
Heracles, 30, 140, 259, 263, 271, 274, 276 interpretations of, 62–63
Heraclitus, 6, 21, 30, 55, 169, 265, 322, its constitutive terms, 65–72
339, 340, 359 motivation for, 92–93
and traditional religion, 48 shepherds . . . mere bellies, 74–75, 313
on Apollo, 114 whenever we wish, 83–84, 85–86,
on names, 204 93, 100
on the soul, 38, 161, 233, 236 the prescriptive force of his narratives,
Hermes, 28, 66, 71, 89, 257, 351, See also 57, 72, 92, 339–340
Homeric Hymn to Hermes the proem to the Works and Days, 95–97
Hermotimus, 277 Hippocratic corpus, 17, 236, 245
hero cult, 275 On the Sacred Disease, 27–29, 343
Herodotus, 30–31, 47, 117–118, 125, 145, historiê. See wandering: as an Ionian ideal
339, 343 Homer
Hesiod on the soul after death, 161
ambivalence in, 3, 34–35, 59, 85–86, the cave of the Nymphs, 247–248, 257
99–100, 101–102, 314, 318 theological attitudes, 23, 30, 57–58
and Parmenides, 59, 254–255, 310–311, approximation to the divine and
312–314, 317–318, 352 godlikeness. See
and philosophy, 3, 52–60, 310–314, approximation to the divine
317–318 on divination, 108–115, 128–129, 345
and theodicy, 34–35, 84–86, 100–101 on Muses and poetic inspiration, 63,
and Xenophanes, 59, 310, 311–312, 314, 80–82, 93–94, 194
317–318 Homeric Hymn to Demeter, 79, 86, 268,
approximation to the divine and 270–271
godlikeness. See approximation Homeric Hymn to Dionysus, 34, 79
to the divine Homeric Hymn to Hermes, 77, 82, 83,
Aristotle on. See Aristotle: on Hesiod 94, 102

401
 General Index
homoia, 69–71 Muses. See under Empedocles, Hesiod,
homoiôsis theôi. See becoming like god, Homer
Parmenides: and becoming mystery initiations, 255, 267–273, 279, 351
like god and altered states of consciousness, 284
myth, 8, 14, 16, 17, 19–20, 25–26, 44, 56,
irrationality. See rationality: and 147, 320, See also theology:
irrationality Varro’s tripartite theology
and Greek religion. See religion, Greek: myth-critics, 29–33, 343
and irrationality
Isocrates, 267, 270 natural theology, 36–37, 39, 41, 345–346
Nehamas, A., 200
Jaeger, W., 16 Nestle, W., 16
Nestor, 68, 94, 152, 322
katabasis, 254, 257, 276–277, 347, Nietzsche, F., 62, 166–167
350–353, 357–358 Numenius, 247–248
Kingsley, P., 12, 18, 274
Odysseus, 15, 65, 66, 71, 81–82, 129,
Laks, A., 13 262, 263
Leonymus, 277 and Parmenides. See Parmenides: and
Lesher, J.H., 12, 17–18, 104, 106, 119, the Odyssey
131–132, 138, 149 as a poet-like figure, 94
Lloyd, G.E.R., 16–17, 48, 59, 152 On the Sacred Disease. See Hippocratic
Long, A.A., 158, 223 corpus: On the Sacred Disease
luxurious multiplicity. See Versnel, H.S. Orphica, 140, 257, 276, See also Derveni
papyrus, Gold Tablets
Mansfeld, J., 223–224 Orphic anthropogony, 246
mantikê. See divination Owen, G.E.L., 163
Martin Luther, 49
Marx, K., 51 Palaephatus, 31–33, 343
Melissus, 208, 220, 287 Palmer, J.A., 294–295
metempsychosis, 252, 272–273, See also Pandora. See Hesiod: on Prometheus and
entries under Empedocles, Pandora
Parmenides, Pindar, Plato, Parker, R., 41, 46, 114
Pythagoras and the Parmenides
Pythagoreans alêtheiê in. See under alêthea
mêtis, 87–90, 91, 94, 238, 330–332 and becoming like god, 5, 252–284,
Milesian, the philosophers, 16, 19–26, 343 313, 341
Anaximander, 20–21, 22, 25, 148, and Hesiod, 59, 254–255, 310–311,
201, 232 312–314, 317–318, 352
Anaximenes, 21, 23–26, 47, 201, 244, and Ionian cosmology, 292
290–291 and mystery initiations. See mystery
and traditional religion, 19–20, 22, initiations
25–26, 39 and oracles, 265–267
Thales, 20, 25–26, 38, 148 and sense-perception, 169–184,
all things are full of gods, 20, 23 190–195, 196, 211, 216–217
Mogyoródi, E., 139 and the Gold Tablets. See Gold Tablets
Mopsus, 114, 154 and the Odyssey, 264–265
Mourelatos, A.P.D., 190, 197, 198, 210, and the Pythagoreans. See Pythagoras
254, 264, 307, 359 and the Pythagoreans

402
 General Index
and vital heat, 231–232 what-is
and Xenophanes, 308, 314–318 as intelligent and divine, 290,
verbal echoes in Parmenides, 314–315 300–301, 307–308,
as transformative, 157–158, 315–316, 344
283–284, 341 as what really is, 295–301
Colotes on, 157, 158, 164, 301–302 in later ancient readers, 301–303
Doxa cannot be identified or aligned with
as deceptive, 168, 199–203, 206–207, Light (or Night), 186–187,
209, 214–215, 286 225–226, 280
expressing Parmenides’ own not a cosmogonic or cosmological
views, 167, 199–201, principle, 292, 300–301
215–216, 220 the unshaken heart of well-rounded
framed pejoratively, 167, 168 reality, 300–301, 308
its governing goddess, 233–239, 248, his theory of human cognition, 4,
315, 344 169–184, 195–196, 216–217,
its scope and nature, 164–165 239–240, 280
the reasons for it, 4, 160–161, its doxographic context, 169–175,
163–168, 196–197, 190–193, 195, 240–242
215–221, 278 responses to its paradoxical
Doxa as dialectical, 163–165 implications, 222–227
Doxa as pragmatic, 165–166 inscriptions about him discovered in
eschatological topography in, 242, Elea, 277
254–255, 347–359 intelligence and intelligibility in
his attitude to divine disclosure, 5, Alêtheia, 304–308
223–224, 266–267, 279–284, krisis (of Alêtheia vs of Doxa), 158–159,
340–342 184–190, 195–196, 226–227,
his embryology, 237–239 278–279, 280–282, 340–341
his goddess notion of rationality in, 340–342
anonymous, 267–268, 355–356 on aether, 230–231, 258, 349–350,
as aether, 248 352–353, 356, 358
main interpretations of, 11 on female and male, 248–249, 313
his Homeric language, 264 on naming, 188–190, 203–208, 300
his ontology, 4, 157, 211–214, 215, on the soul, 5, 161, 227–228, 232–237,
285–303 242–243, 247–248,
consubstantial interpretations, 249–250
290–293 and death, 234–237
Doxastic things as non-existent, and metempsychosis, 235–236,
illusions, 287–289 237–238, 242–243, 245, 256
Doxastic things as things-that-are, its divinity, 242–244, 247–248,
293–294 249–250, See also Parmenides:
Doxastic things available for and becoming like god
discussion, belief, (in)correct soul and mind in Parmenides,
accounts, 286, 303 239–242, 282
technical vs non-technical uses of ‘to the proem, 250–278, 279, 280, 347–348,
be’, 298–299, 300 350–359
the modal interpretation, 294–295 echoing Doxa’s imagery,
the problem of colocation, 290 159–160, 248
the two-worlds interpretation, the knowing man, 255, 258–259,
289–290 277–278, 355

403
 General Index
Parmenides (cont.) Prodicus, 44
the kouros’ journey Prometheus. See Hesiod: on Prometheus
as a journey of the soul, 238, and Pandora
255–260, 274, 276–277, 355 Protagoras, 42, 45, 47
its status, 13, 162, 280, 284, pseudea, and cognates, 65–67
347–348 Pythagoras and the Pythagoreans, 55, 148,
its trajectory, 254, 257, 347–359 274–277
the kouros as dependent, a social and political force in South
259–260, 280 Italy, 274
the two parts of his poem, 155–156 and metempsychosis, 151, 161, 238, 244,
distinguished by ancient 275, 323
commentators, 220 Pythagoras as (semi-)divine, 275–276
interpretative questions about the Pythagorean quietude, 274–275
relation between them, the katabasis of Pythagoras, 276–277
160–161
Pascal, 49 rationality
Persephone, 243, 246, 269, 272 and divine disclosure, 5–6, 10–12, 15,
Phaethon, 256, 258, 356, 357 51, 282–284, 334–335, 338,
Pherecydes, 59, 244, 246 339–343
Philo, 150 in divination, 106, 110–114,
Philoponus, 20, 201 145–146, 340
Phormio, 277 and irrationality, 10–19
Pindar, 246, 339 and religious belief. See religious
on death and initiation, 245, 269, 270 belief
on metempsychosis, 243 as a type of thinking, 2, 15, 343
Plantinga, A., 49–50 as secularising, 2, 14, 36, 342–343
Plato requiring disambiguation, 19, 35,
his myth of Er, 247–248 342–343
his ontological preoccupations and religion, Greek
Parmenides’, 285, 302 and irrationality, 33–35
on becoming like god, 250–251, 252, and philosophy, 2, 14, 36–48, 343–346
281, 283 evidence of social backlash, 42–45,
on divination, 112, 113, 130, 131, 195 46, 346
on early Greek philosophy, 21, 37–39 philosophical criticisms and
on metempsychosis, 243, 245 appropriations, 2, 5, 41–42,
on mystery initiations, 269, 270 45–46, 47–48, 274, 343–345
on poetic inspiration, 81 philosophy as part of religion, 2,
on Socrates’ trial. See Socrates: his trial 48, 346
Plutarch views of the relations between them,
on divination, 26, 127 16–18, 45
on mystery initiations, 268–269 beliefs in, 40–41, 48, 50, 51
on Parmenides, 157, 164, 239, cognitive approaches to, 41
301–302, 303 tolerance and flexibility of, 40,
poetic inspiration, 14, 80–82, 265, 45–47, 346
266–267, 339 unknowability of the divine, 34,
prayer, 22, 25, 26, 28–29, 41, 45, 46, 96, 47, 346
141, 343 religious belief, 48–51, 342–343
presocratic philosophy. See early Greek Rohde, E., 239
philosophy Rowe, C.J., 56–58

404
 General Index
sacrifice, 25, 26, 28, 41, 44, 45, 74, 88, Thriai, 67, 77, 83, 84
150–151, See also transmigration. See metempsychosis
Empedocles: rejecting animal Trophonius, 284
sacrifice
Schofield, M., 278, 279 Varro, 36
Sedley, D.N., 168 Versnel, H.S., 33–35, 41
Simplicius Vlastos, G., 184, 190, 224–227, 233
his evidence for Parmenides’ conception
of soul. See Parmenides: on the wandering
soul and divination, 153
his evidence for Parmenides’ as an Ionian ideal, 148–149, 322–323
eschatological topography. See in Empedocles, 321, 326
Parmenides: eschatological in Parmenides, 197–198, 216, 264–265
topography in in the Odyssey, 148, 264–265, 322
on Parmenides’ ontology, 301, 303 Wittgenstein, L., 49
on the two parts of Parmenides’
poem, 220 Xenophanes
Sirens, 84, 93–94 and Colophon, 142, 149, 154
Socrates and Eastern influences, 154
his divine sign, 127, 339 and Hesiod, 59, 310, 311–312, 314,
his trial, 43–44, 45, 46 317–318
Sophocles, 110, 113, 114, 195, 270, 272, and monotheism, 125
276–277 and Parmenides, 308, 314–318
soul. See entries on soul or metempsychosis and the idea of progress, 148–150
under Empedocles, Heraclitus, and theodicy, 141–142
Homer, Parmenides, Pindar, criticisms of traditional religious
Plato, Pythagoras and the attitudes, 41–42, 142–143,
Pythagoreans 149, 150–151, 343
as aethereal. See eschatology: aethereal on divination, 41, 104–107, 119–121,
as divine, 243–246, See also Parmenides: 123, 130–132, 139,
on the soul 144–154, 343
Stoicism, 20, 46, 105, 245, 251, 269 divinatory language in, 128–129,
152–153
Tabula of Cebes, 275 expressing a unified world-view,
Teiresias, 109, 110, 114–115, 150, 154, 148–150
194, 264 his attitude to divine disclosure, 3–4,
Tertullian, 49, 51 104–154, 310, 311–312,
Thales, See Milesian, the philosophers 316, 340
theodicy. See under Hesiod, intentionality in his notion of
Xenophanes disclosure, 106–107, 133–134,
theology, 8, 345, See also natural theology 136–142, 144–145, 310
Varro’s tripartite theology, 36 its role in human inquiry, 133–143,
theology of approximation, 251 144–145
Theophrastus rejecting one notion of disclosure and
his account of Parmenides in de promoting another, 106,
Sensibus. See Parmenides: his 116–128, 143–144
theory of human cognition insisting on a strict boundary between
on the two parts of Parmenides’ mortal and divine, 150–151,
poem, 220 252, 279, 316, 318

405
 General Index
Xenophanes (cont.) and personal experience, 129–130
on belief and practice in religion, theological methodology in, 142–143
343–344 Xenophon, 45, 117–118, 141
on his own wisdom, 107, 146–154 on divination, 108, 110–111
on his longevity, 151–152,
322–323 Zalmoxis, 276
on intelligent divine governance, Zeus, 17, 28, 32, 33, 44, 66, 110, 119, 140,
137–139 229, 246, 257, 261, 269
on knowledge, 106, 115, 128–133, 146, Hesiod on. See Hesiod: on Zeus
150, 316 his mind, 98, 100, 137–138, 183

406