T.P.

Wiseman’s books, all with University of Exeter Press

except where indicated, include:

Clio's Cosmetics: Three Studies in Greco-Roman Literature

(1979. reprinted 2004 by Bristol Phoenix Press, UEP’s sister imprint)
Death of an Emperor: Flavins Josephus (1991)
Talking to Virgil: A Miscellany (1992)
Historiography and Imagination: Eight essays on Roman culture (1994)
Remus: A Roman myth (1995, CUP)
Roman Drama and Roman History (1998)
The Myths of Rome (2004, paperback, 2008)
Remembering the Romnn People (forthcoming, OUP)
 32

Unwritten Rome
T.P. WISEMAN

CLASSICAL
LIBRARY
CAMBRIDGE

UNIVERSITY

EXETER
PRESS
 J

/ c* p
Lio

Cover illustration: Silenus-head terracotta antefix, early fifth century BC,

from below the Basilica Julia in the Roman Forum. Reproduced by
permission ot the Soprintendenza per i Beni Archeologici di Roma.

First published in 2008 by

University of Exeter Press
Reed Hall, Streatham Drive
Exeter EX4 4QR
UK
witw. exeterpress xo.uk

© T.P. Wiseman 2008

The right of T.P.Wiseman to be identified as author of this

work has been asserted by him in accordance with
the Copyright, Designs and Patents Acts 1988.

British Library Cataloguing in Publication Data

A catalogue record for this book is available
from the British Library.

Hardback ISBN: 978 0 85989 822 5

Paperback ISBN: 978 0 85989 823 2

Typeset in Bembo and Gill Sans

by XL Publishing Sendees, Tiverton

Printed in Great Britain by Cromwell Press, Trowbridge

 For
Michael Crawford
William Harris
David Ridgway
(British School at Rome, 1962—3)

I
!

|
 Contents

List of illustrations viii

Acknowledgements x
1 Unwritten Rome 1
2 What Can Livy Tell Us? 24
3 Fauns, Prophets and Ennius’ Annales 39
4 The God of the Lupercal 52
5 Liber: Myth, Drama and Ideology in Republican Rome 84
6 The Kalends of April 140
7 Summoning Jupiter: Magic in the Roman Republic 155
8 Origines Indortnn 167
9 The Games of Flora 175
10 The Games of Hercules 187
11 Praetextae, Togatae and Other Unhelpful Categories 194
12 Octavia and the Phantom Genre 200
13 Ovid and the Stage 210
14 The Prehistory of Roman Historiography 231
15 History, Poetry and Annales 243
16 The House of Tarquin 271
17 The Legend of Lucius Brutus 293
18 Roman Republic, Year One 306
Bibliography 321
Illustration credits 344
Index of passages 346
General index 350
 Illustrations

1 'Perhaps the earliest inscription from Rome’ 2

2 Fragment of terracotta frieze 3
3 Archaic inscription from the Volcanal 3
4 Remains of the fourth-century BC city wall near Roma Termini
station 5
Signature plate attached to the lid of the ‘Ficoroni cista 6
6 Time chart of Roman history 9
7 Anna Perenna cult site at Piazza Euclide, Rome 19
8 Inscription from Anna Perenna’s cult site 20
9 Marble relief of Priapus 21
10 Praenestine mirror: Paniskos and Marsyas 59
11 Praenestine mirror: Pan Lykaios? 60
12 Fourth-century BC coins of the Arcadian federation and
Panticapaeum 61
13 Map of the site of Rome 64
14 Praenestine mirror: drinking scene 67
15 Praenestine mirror: bedroom scene 68
16 Detail of sarcophagus from the S. Praetexta catacomb, Rome 80
17 Cista no. 5, in the Staatliche Museen, Berlin 87
18 Cista no. 4, in the Staatliche Museen, Berlin 88
19 Cista no. 22, in the Badisches Landesmuseum, Karlsruhe 89
20 Cista no. 72, in the Villa Giulia Museum, Rome 91
21 Cista no. 66, in Vassar College, New York 93
22 Cista no. 45, in the Pierpont Morgan Library, New York 94
23 Lid of cista no. 45 (22) 95
24 Mirror in the British Museum 96
25 Cista no. 1, in the Walters Art Gallery, Baltimore 97
26 Cista no. 14, in the Field Museum, Chicago 98
27 Mirror in the Villa Giulia Museum, Rome 100
28 One side of cista no. 100, in the Villa Giulia Museum, Rome 101
29 Marsyas statue at Paestum 102
30 Cista no. 70, in the Villa Giulia Museum, Rome 103
31 Detail of cista no. 68, in the Villa Giulia Museum, Rome 104
32 Lucanian oinochoe 105
 ILLUSTRATIONS ix

33 Antefix from Satricum in the Villa Giulia Museum, Rome 106

34 Faliscan cup in the Villa Giulia Museum, Rome 107
35 Apulian dish in the British Museum 108
36 Bell crater with Dionysiac scene in the Cleveland Museum of Art 110
37 Etruscan cup (Clusium group) in the Vatican,
Museo Gregoriano 112
38 Etruscan cup (Clusium group) in the Musee d’art et d’histoire,
Geneva 113
39 Cista no. 82, in the Villa Giulia Museum, Rome 114
40 Cista no. 51, in the Museo Archeologico, Palestrina 115
41 Cista no. 31, in the British Museum 118
42 Cista no. 9, in the Staatliche Museen, Berlin 120
43 Cista no. 83, in the Villa Giulia Museum, Rome 121
44 Campanian jug 123
45 Republican Rome in its geographical context 177
46 The coins of M. Volteius, 78 BC 187
47 The ‘Sacra Via’ from the arch of Titus to the temple of Vesta 272
48 ‘Domus 3’, first reconstruction 274
49 ‘Domus 3’, second reconstruction 275
50 Grove of Vesta and Porta Mugonia, first reconstruction 277
51 Grove of Vesta and Porta Mugonia, second reconstruction 277
52 ‘House of the Tarquins’, surviving remains 278
53 ‘House of the Tarquins’, surviving remains and ‘probable
structures’ 278
54 ‘House of the Tarquins’, hypothetical reconstruction 279
 Acknowledgements

I am ver)’ grateful for permission to reprint the following chapters:

3: Jm/nis.139.3 (2006). 513-29

4: Journal of Roman Studies 85 (1995), 1-22
5: Christer Bruun (ed.), Hie Roman Middle Republic: Politics, Religion and Historiog­
raphy c.400-133 BC (Acta Instituti Romani Finlandiae 23, Rome 2000),
265-99
6: W.V. Harris and E. Lo Cascio (eds), Nodes Cainpanae: Studi di storia antica ed
archeologia dell'Italia preromana e roinana in memoria di Martin W. Frederiksen (Itala
Tellus 1, Luciano, Naples 2005), 227-43
8: Kathryn Welch and T.W. Hillard (eds), Roman Crossings: Theory and Practice in
the Roman Republic (Classical Press of Wales, Swansea 2005), 49-58
9: Bettina Bergmann and Christine Kondoleon (eds), The Art of Ancient Spectacle
(Studies in the History of Art 56, Washington D.C. 1999), 194—203: reprinted
by permission of the National Gallery of Art, Washington; all rights reserved
10: Edward Bispham and Christopher Smith (eds), Religion in Archaic and Republican
Rome and Italy: Evidence and Experience (Edinburgh 2000), 108— 14
IT.Symbolae Osloenses 77 (2002), 82-8: by permission of Taylor and Francis AS
(www.tandf.no/symbolae)
12: Ancient History Bulletin 19.1-2 (2005), 59-69
13. Geraldine Herbert-Brown (ed.), Ovid’s Fasti: Historical Readings at its Bimillen­
nium (Oxford 2002), 275-99: by permission of Oxford University Press
14:John Marincola (ed.), The Blackwell Companion to Greek and Roman Historiog­
raphy (Blackwell, Oxford 2007), 67-75
15:D.S. Levene and D.P. Nelis (eds), Clio and the Poets: Augustan Poets and the
Traditions of Ancient Historiography (Mnemosyne Supplement 224, Brill, Leiden
2002), 331-62
17: Mario Citroni (ed.), Memoria e identita (Studi e testi 21, Florence 2003), 21-38
18: Greece and Rome 45 (1998), 19-26: copyright the Classical Association, with
permission

In all these chapters, additional material is included in square brackets. Chapters

1,2,7 and 16, and the Afterword to chapter 18, are previously unpublished.
 CHAPTER ONE

Unwritten Rome

Brave men lived before Agamemnon,

Plenty of them—but they’re all unmournable,
Buried unknown in the long night
Because they lack the sacred bard.
Horace Odes 4.9.25-8

I
Brave men lived before Romulus, too. Rome dates from at least the thirteenth
century BC,1 a thousand years before the sacred bard first wrote heroic verse in
Latin. When it was first called ‘Rome’, nobody knows; but the archaeological
record reveals a substantial community already in the ninth century BC.2
Its size and expansion can be inferred from the first major stage in its devel­
opment, when the burial-ground in the low-lying area north of the Palatine
was abandoned, and a new one created nearly a mile away on the plateau of
the Esquiline.3 The archaeological context of that is Ciiltura laziale II Al, about
875 BC on the conventional dating.4 The Esquiline cemetery remained in use
for more than eight centuries, until Augustus’ friend Maecenas redeveloped it
for his suburban park,5 and it was already more than two hundred years old
when the people who used it began to scratch names on the vases that were
buried with the deceased. These graffiti are the earliest known writing in
Rome—an Etruscan s/ni[...] on a fragment of an ‘Italo-geometric’ plate [fig.

1 The terracing on the Capitol in the late Bronze Age (say 1300-1200 bc) may have removed
the trace of Middle Bronze-Age occupation, for which only sporadic and unstratified mate­
rial survives: see Lugli and Rosa 2001 (esp. 285), Baroni 2001.
2 Or the tenth, if the proposed recalibration of Ciiltura laziale II A-B is correct: see Bettelli
1997.191-8, Nijboer et al. 1999/2000. See Cornell 1995.48-57 for an accessible summary
of the archaeological data.
3 Summary in Holloway 1994.20-36.
4 Bettelli 1997.215-16; on the revised dating (n. 2 above) it would be about a century earlier.
5 Horace Satires 1.8.8-16; Bodel 1994.38-54.
 2 UNWRITTEN ROME

Fig. I 'Perhaps the earliest inscription from Rome’ (Colonna 1987.58): graffito
on an Italo-geoinetric plate from the Esquiline cemetery at Rome, first half of
the seventh century BC.

1], and a Greek ktektos or kleiklos on a Corinthian oil-flask, both in the seventh
century BC?
That was about the time when the Romans (as we can now safely call
them) undertook a huge public-works project. By depositing up to 20,000
cubic metres of fill, they raised the ground level of the basin between the Pala­
tine and the Capitol, to minimise flooding and provide the foundation for a
pebble-paved public area, the Roman Forum.7 Rectangular public buildings in
stone soon began to replace thatch-and-timber huts, and by the middle of the
sixth century BC they were being decorated with terracotta panels and friezes.
One fragment that happens to survive, probably from the third phase of the
'king’s house’ (regia) at the east end of the Forum,8 gives us our first glimpse
into the imaginary world of archaic Rome [fig. 2]. What we see is quite myste­
rious: panthers and a minotaur. The image must have a meaning, but it was
probably never written down.
Writing was, however, in public use by this time. The most spectacular
evidence for that comes from a sixth-century cult site now identified as the
Volcanal, close to the Comitium.9 It consists of the lower part of a square-cut
pillar, covered on all four sides with a long inscription [fig. 3]. About half the
text survives, enough to allow the conjecture that it consisted of regulations
and prohibitions connected with the sacred site. However, since it runs both
vertically and ‘boiistropliedoii’ (i.e. reading alternately left to right and right to

6 Colonna 1987.57-8, Poucet 1989.287-8. See respectively Bagnasco Gianni 1996.305 and
Solin 1983 (illustration at Holloway 1994.25).
7 Ammerman 1990; cf. Ainmennan 1998 and 2006.303-7 for the ‘Velabrum’ valley.
8 Cristofani 1990.61 (item 3.2.13), Holloway 1994.60—4.
9 Holloway 1994.81-7, Cornell 1995.94-5; full details in Cristofani 1990.58-9 (item 3.1.39).
Volcanal: Coarelli 1983.161-78.
 UNWRITTEN ROME 3

‘■y; :

Fig. 2 Fragment of terracotta frieze, found in the 1950s and attributed to the
reqia: mid-sixth century bc.

'i
-Z

^£*1 *

Fig. 3 Archaic inscription from the Volcanal, discovered by Giacomo Boni in

1901. Starting at the top right, it reads: [. . .]m kalato I rem /ic[. . ,]od ioiixmen I ta
kapia dotau[.. ,]in i te ri[. . .]»i quoi /id uelod ne<jn[. . .iouestod. ..
 4 UNWRITTEN ROME

left, or in this case alternately up and down), it is hard to imagine that it was
meant for direct communication with the citizens.10 The rules were cut in
stone to make them pennanent, but I doubt if one Roman in a hundred could
have read them for himself. It was the priest, through his crier, the kalator, who
told the People what to do and what not to do.11
What mattered in practice was not written documents but the winged
word:12

... iicrliiiin saepe ttiinin perciet attris

omnibus in papula, niisstini praeconis ab ore.
A single word often makes attentive the ears of everyone in the assembly, when
it is shot from the crier’s mouth.

Even the Twelve Tables, the Roman law code, supposedly set up in bronze or
ivory on die Rostra in 449 BC, were communicated through oral means,
taught as a ‘compulsory recitation’ to young Romans in school.13 Public
inscriptions were symbolic monuments, not information media.
Another use of writing in early Rome came to light in the 1870s, when the
city was being redeveloped as the capital of united Italy. Up on the high
ground where the new railway station was built, substantial lengths of the city
wall of republican Rome were discovered [fig. 4]. The wall was built in the
370s BC. Seven miles long and forty feet high, it was a huge undertaking.
Hundreds of thousands of tufa blocks were cut from the quarries of newly-
conquered Veii, marked by the masons with code-letters on the side that
didn’t show.14
Bitter experience had made it necessary. In 387/6 BC (perhaps the first reli­
ably dated event in Roman history),15 Rome was captured and sacked by a
marauding army of Gauls. New defences were necessary to make sure that
would not happen again—and it didn’t, for nearly 800 years. As well as the
great walls, Rome also created a new and more open political system, with
power shared between ‘patricians’ and ‘plebeians’ (as later Roman idiom
described them). It is likely that these new conditions made Rome attractive to
ambitious immigrants from other parts of Italy, including areas like Campania

10 So Cristofani 1990.58, rightly. Cf. Harris 1989.50-1 on public inscriptions in archaic

Greece.
11 E.g. Varro De lingua Latina 6.27 (pontifices), Ovid Fasti 2.527-8 (curio iiiaxiiniis). Kalatores:
C1L 6.36840.C.1 (= fig. 3, bottom right); cf. 2184-90, 36932, 37167.
12 Lucretius 4.563-4. Winged words: Homer Iliad 1.201 etc.
13 Cicero De legibtts 2.59 (lit carmen necessariutn), cf. 2.9; Harris 1989.151-3, not refuted by
Cornell 1991.31-2. Bronze tablets: Diodorus 12.26.1, Livy 3.57.10, Dionysius of Halicar­
nassus 10.57.7. Ivory tablets: Pomponius in Digest 1.2.2.4.
14 Saflund 1932.109-11.
15 Polybius 1.6.1-2 (from Timaeus?), cf. 2.18-20; Diodorus Siculus 14.110.1, 14.115, Justin
6.6.5.
 UNWRITTEN ROME 5

Fig. 4 Remains of the fourth-century bc city wall near Roma Termini station.
In this photograph, taken soon after excavation, the masons’ marks are clearly
visible.

and Etruria where the art of writing may have been more widely practised.16
One such, no doubt, was the fourth-century master craftsman Novius Plautius,
who made the ‘Ficoroni cista' [fig. 31, p. 104 below]. His praenonien suggests a
Campanian origin, and the inscription he attached to his work reveals a confi­
dent mastery of the medium [fig. 5].17
What else was writing used for in Rome at this time? Unlike the Greek­
speaking world (Novius Plautius was a contemporary of Aristotle), there is no
sign here that the poet’s song, the actors’ performance or the story-teller’s
narrative was yet thought worthy to be ‘committed to letters’ and preserved
for ever like laws and treaties.18 Another fine fourth-century cista from Latium
[fig. 22, p. 94 below] shows Agamemnon in a scene unknown to Greek liter­
ature, and identifies him as Aanenieiio;'9 evidently the artist, and no doubt his
patron too, knew the name not from written texts but from hearing it spoken.
The earliest certain evidence for narrative writing in Rome comes from the

16 For Etruria, see Colonna 1976, Torelli 1996.

17 CIL 1-.561 = ILLRP 1197.
18 Litteris mandare: e.g. Sisenna fr. 127P (Aulus Gellius 12.15.2), Caesar De belli} Gallico 6.14.3.
Early treaty inscriptions: Dionysius of Halicarnassus 4.58.4 (Rome and Gabii), Polybius
3.22.1-3 and 26.1 (Rome and Carthage), Cicero Pro Balbo 53 (Rome and Latins), Livy
4.7.10 (Rome and Ardea).
19 Battaglia and Emiliozzi 1976.146—50 (no. 45), tav. cxcii.
 6 UNWRITTEN ROME

’41 Ql(l 4 (b
//

k Wovt Lz
•^b^VTiOS.Mg.o RE CID
__

Fig. 5 Signature plate attached to the lid of the ‘Ficoroni cista’, in the Villa
Giulia Museum, Rome. The footprints are of the figures of Liber Pater flanked
by two satyrs which fonned the handle. The front inscription reads ‘Novios
Phutios made me in Rome’, the rear one ‘Dindia Macolnia gave [me] to her
daughter'. About 340 bc.

tomb of the Scipios on the Via Latina. The two earliest sarcophagi are those of
L. Scipio 'Barbatus’, consul in 298 BC, and his son L. Scipio, consul in 259 BC.
They cam’ the following inscriptions in ‘Saturnian’ verse:20

1. Cornelius Lucius Scipio Barbatus

Gnaiuod patre progiiatus fortis uir sapieusque
quoinsforma uirtutci parisiuiia fuit
consol censor aidilis queifuit apud nos
Taurasia Cisauna Saiuuio ccpit
subigit omne Loucanam opsidesque abdoticit
Cornelius Lucius Scipio Barbatus
father Gnaeus’ offspring, a brave and educated man,
whose beauty’ matched his character.
He was consul, censor, aedile among you,
Taurasia, Cisauna, Samnium he conquered,
he subjugated all Lucania and brought back captives.
2. hone oino ploinime consentient R[o»i<i/ic]
duonoro optunio fuise uiro
Luciom Scipione filios Barbati
consol censor aidilis hiefuel afpud nos]
heccepit Corsica Aleriaque urbe
dedet Tempcstatebus aide /»en7o|d]
This man most Romans agree
was the very’ best of good men,

20 ILLRP 309-10, illustrated in Coarelli 1996.219-20. Translation: Goldberg 1995. 62-3, very'
slightly adapted.
 UNWRITTEN ROME 7

Lucius Scipio, the son of Barbatus.

He was consul, censor, aedile among you,
he captured Corsica and the city of Aleria,
he fittingly dedicated a temple to the Weather.

The second-person-plural address (apud nos) can hardly have been meant in the
first instance for a group reading by lamplight in an underground burial
chamber. In the context, ‘you’ ought to be the Roman People, and the infer­
ence is surely right that the texts were first composed for delivery at the two
men’s funerals, about 270 and 230 BC.21 The spelling and the letter-forms show
that the inscription on Barbatus’ coffin was cut much later than that on his
son’s (one might guess about 190 Be);22 so there was evidently a written
version of the funeral poem about his achievements which could be quoted
two generations later, when his descendants wanted to make his sarcophagus as
impressive as his son’s.
Those were the two generations that saw the creation and first flowering of
Latin literature.23 We can know of that development only anachronistically,
through the works of authors two centuries later who took literature as much
for granted as the air they breathed. Because they had no notion of it them­
selves, they cannot convey to us what was involved for the pioneering cultural
entrepreneurs who first wrote plays and epic poems in Latin—Andronicus
from Greek Tarentum, known by his Roman-citizen name as Livius;24 Gnaeus
Naevius, from Campania and proud of it;2’ and Quintus Ennius from Rudiae
in rhe heel of Italy, trilingual in Oscan, Greek and Latin.26
At the same time, Romans were now writing history in prose. But these
authors were senators—Quintus Fabius Pictor, from an ancient patrician family,
and Lucius Cincius Alimentus—and they wrote in Greek.27 Fabius’ history was
evidently translated into Latin, but we do not know when; Latin historiography
proper began in the mid-second century, with Cato and Cassius Hemina.28 By

21 Zevi 1968. See Polybius 6.53-4 for Roman aristocratic funerals.

22 Coarelli 1996.217-23.
23 Brilliantly analysed by Feeney 2005, esp. 232-3 on ‘the differences between artistic and
textual production in Rome in, say, 260 BCE and in, say, 200 bce’, and 236—7 on the seini-
jjnicri (Suetonius De grammaticis 1.2) who created it.
24 Cicero Brutus 72 for his origin (from Accius, who evidently misdated his arrival in Rome);
Gruen 1990.80-3.
25 Aulus Gellius 1.24.2 (plenum superbiae Campanae); Gruen 1990.92-3.
26 Aulus Gellius 17.17.1 on his ‘three hearts’; see ch. 3 below.
27 Dionysius of Halicarnassus 1.6.2 (date), 1.74.1 (Cincius a senator); Appian Hannibalic War
27.116 (Fabius a senator).
28 Pictor in Latin: Varro in Quintilian 1.6.12, Aulus Gellius 5.4.3, Nonius 835L, Servius auctus
on Aeneid 8.630. Cato: Cicero De senectute 38 (at work on the final book in 150 bc). Cassius
Hemina: Censorinus De die natali 17.11 (146 BC); Rawson 1991.246.
 8 UNWRITTEN ROME

that time, Ennius’ great epic Aunales had already given the Romans the whole
story of their city, from Aeneas and Romulus right down to the wars they had
fought in themselves.
It is very easy, when following a summary narrative like the one we have
just constructed, to lose track of the passing of time. It is only when we set it
out in casual form [fig. 6] that the problem of ‘unwritten Rome’ becomes
apparent. Ever}' century represents a period well beyond the range of living
mentor}', at the limit even of what can be learned from the memories of grand­
parents. So when the Roman past was first written down by the poets,
playwrights and historians of the late third century BC, how much of it was
accurately remembered?

II
Historians writing in Greek must have been aware of the principles of Greek
historiography, as set out by the great masters of the genre. For Herodotus, it
was not enough to accept or reject existing stories; the historian must take
responsibility for his own narrative.29 Thucydides insisted on first-hand knowl­
edge, or the careful interviewing of eye-witnesses; for periods where that was
not possible, the historian must make rational inferences from the evidence
(TEk'pqpia, ‘indications’).30 It seems that Fabius Pictor and Cincius Alimentus
acted on these principles. ‘Each of them,’ wrote Dionysius,31 ‘related the
events at which he himself had been present with great exactness, as being well
acquainted crith them, but touched only in a summary way upon the early
events that followed the foundation of the city.’
In fact, ‘indications’ did exist which could have allowed Thucydidean
inferences about early Rome; but there were not many of them, and they were
not easy to use.
In the Treasury of the Aediles, next to the temple of Jupiter Optimus
Maximus on the Capitol, was a bronze tablet preserving the first treaty
between Rome and Carthage. Polybius believed it was made in the first year of
the Republic by the consuls L. lunius Brutus and M. Horatius. Fie provided a
translation for his Greek readers, but warned them that ‘the modern language
has developed so many differences from the ancient Roman tongue that the

29 Herodotus 1.5.3: ‘So this is what the Persians and Phoenicians say. I am not going to come
down in favour of this or that account of events, but 1 will talk about the man who, to my
certain knowledge, first undertook criminal acts against the Greeks. 1 cvill shocv who it cvas
who did this, and then proceed svith the rest of my account.’ Translation by Robin Water-
field (World’s Classics).
30 Thucydides 1.1.1 (tEiquilpdpEVOg), 1.20.1, 1.21.1 (TEKpqpta); 1.22.2-3 (eye-witnesses).
31 Dionysius of Halicarnassus 1.6.2 (trans. E. Cary, Loeb); ‘exactness’ is the Thucydidean
concept of di:pi|kta.
 UNWRITTEN ROME 9

1400
Middle Bronze Age occupation?
1300
Late Bronze Age: terracing on Capitol
1200

I 100

1000

900
Creation of Esquiline cemetery
800

700
First graffiti on pots [fig. 1];
Forum valley developed 600
Terracotta decoration [fig. 2]
Volcanal inscription [fig. 3] ‘House ofTarquin’, ch. 16
500 Tarquins expelled, ch. 17

‘Twelve Tables’ law code

400
city wall [fig. 4]
‘Ficoroni cista’ [fig. 5] Marsyas statue, ch. 5
300 Q. Fabius and Lupercalia, ch. 4
Scipio tombs
Livius Andronicus; Naevius Indi Florales, ch. 9
Fabius Pictor; Cincius Alimentus 200
Ennius; Cato
Cassius Hemina
100
Sulla, ch. 10; Varro, ch. 15
Livy, Dionysius
BC/AD Ovid
Valerius Maximus, ch. 15
Octavia, ch. 12
100
Fig. 6 Time chart of Roman history, Plutarch
from the first evidence of occupation 200
to the end of the Roman empire in
the West. Latin literature began in the 300
twelfth of these nineteen centuries.
Left: developments mentioned in this 400 Augustine
chapter. Right: a selection of dates
relevant to later chapters. 500
 10 UNWRITTEN ROME

best scholars among the Romans themselves have great difficulty in inter­
preting certain points, even after much study’.32
At the temple of Diana on the Aventine there was a bronze pillar with an
ancient inscription ‘in the letters used long ago in Greece’, naming the cities of
the Latin league. In the temple of Semo Sancus on the Quirinal there was a
wooden shield covered in ox-hide with an inscription ‘in ancient characters’
recording Rome’s treaty with Gabii.33 Dionysius dates both inscriptions to the
sixth century BC, but it is unlikely that either he or his source could read them.
Dionysius also reports that Romulus, after his second triumph, put up a
statue to himself at the Volcanal, with an inscription ‘in Greek letters’
recording his achievements. Elsewhere, following a different source, he inter­
prets the same ancient inscription as reporting the deeds of Romulus’
right-hand man Hostus Hostilius.34 This is, in fact, the very inscription found
in 1901 [fig. 3 above].35 Broken and covered over in the repaving of the
Forum in the late Republic,36 it was still complete and visible in the second
century BC when the authors Dionysius used were writing. ‘What is
disturbing,’ observes Tim Cornell,37 ‘is that Dionysius’ sources should have
been so grossly mistaken in their interpretation of the text and its monumental
setting.’ I think it is just what we should expect.
Livy reports an outbreak of plague in 363 BC, as a result of which the Senate
revived the ancient custom of ‘hammering a nail’. He explains in a digression:38

lex uetusta est, priscis litteris tierbisque scripta, nt qui praetor niaxinius sit idtbus Scptein-
bribiis daiiutn paiigat; ftxa fuit dextro lateri aedis louis optimi ntaxiini, ex qua parte
Mineniae templmn est. eiim clamim, quia rarae per ea tempora litterae erant, notam
nmneri aimoniiii fuisse femnt eoque Mineruae templo dicatain legem quia nuinents
Mineniae inuentmn sit. Volsiniis quoque claims indices numeri annornm jixos in templo
Nortiae, Etmscae deae, coinparere diligens taliuni nionunientoruni auctor Cincius
adfimiat.
There is an ancient law, written in archaic letters and phraseology, that he who
is praetor iiiaxiiniis should hammer a nail on the Ides of September. The text was
fastened up on the right side of the temple ofjupiter Optimus Maximus, where
the sanctuary of Minerva is. It is said that this nail marked the number of the

32 Polybius 3.22.1 (date), 22.3 (trans. Ian Scott-Kilvert, Penguin Classics), 26.1 (aediles’
treasury); see now Serrati 2006.114-18. For the consuls in Year One, see ch. 18 below.
33 Dionysius of Halicarnassus 4.26.5, 4.58.4.
34 Dionysius of Halicarnassus 2.54.2, 3.1.2 (cf. 1.87.2, tomb of Faustulus); Festus 184L shows
that it was the same site.
35 Coarelli 1983.166-9, Cornell 1991.27-9.
36 Festus 416L (postqiiani id Cotta strand), with Coarelli 1985.196—7.
37 Cornell 1991.28-9.
38 Livy 7.3.5—7; detailed discussion in Oakley 1998.73-81.
 UNWRITTEN ROME II

years, since letters were little known at that time, and that the law was assigned
to Minerva’s shrine because numbers are her invention. Cincius, a scrupulous
authority on records of this kind, asserts that at Volsinii too nails as markers of
the number of the years may be seen in the temple of the Etruscan goddess
Nortia.

Livy’s source L. Cincius (not to be confused with Cincius Alimentus the histo­
rian) was an antiquarian, like those learned Romans Polybius had consulted
about the Carthage treaty. Livy cites him not as a fellow-historian but as a
specialist in esoteric lore, and it seems that Cincius himself thought of his
scholarship in those terms. Besides the merely informative titles—On the
Calendar, On Archaic Words, etc.—he also wrote a substantial work called Myst­
agogica, as if by interpreting such ancient documents he was leading his readers
to a level of arcane knowledge denied to the uninitiated.39
Similar learned commentaries were applied to two texts in particular that
were proverbial for archaic phraseology: the Twelve Tables law code and the
hymn of the Salian priests.4" But even the experts were often baffled. The Salii
themselves hardly understood their own hymn,41 and when Cicero reports
conunentators admitting that they didn’t understand a particular point in the
Twelve Tables, we need not suppose his example was unique.42 Of one thing
we can be certain: when the commentators met a difficulty, they tried to solve
it from what they already knew. Whenever their guess was wrong, another
difference between archaic Rome and their own times was lost for ever.
To be fair to the Roman antiquarians, they did know that the distant past
might be very unfamiliar, and sometimes they succeeded in preserving
evidence of a lost world. Consider for instance this passage from the elder Pliny
on the wearing of crowns (coronae):4i

semper tainen auctoritas uel ludicro quaesitarum fuit. namque ad certamina in circntn per
ludos et ipsi descendebant et seruos sues equosque mittebant. hide ilia xii tabiilarum lex:

39 Festus 498L, citing book 2 of the Mystagogica on an inscription dated to 380 DC. Cf. Macro-
bius Saturnalia 1.12.12 (DeJastis), Festus 236L etc (De uerbis priscis).
40 Fortunatianus Hrs rhetorica 3.6 (124 Helm) on antiqua uerba ... abolita, nt sunt in xii tabulis et
Saliari canuitie. Cf. Horace Epistles 2.1.86-7 on the unintelligibility of the Salian hymn. For
jurists as antiquarians, see Harries 2006.176-82.
41 Quintilian 1.6.40. Commentaries on carmen Saliare: Aelius Stilo (Varro De lingua Latina 7.2,
Festus 124L, 132L), Cincius (Macrobius Saturnalia 1.12.12), Sabidius (Schol. Veronensis on
Aeneid 10.241).
42 Cicero De legibus 2.59 on lessuni (Crawford 1996.706-7): hoc ueteres interpretes Sex. Aelius L.
Acilius non satis se intellegere dixenmt. Sex. Aelius Paetus was consul in 198 BC (see Pomponius
in Digest 1.2.2.38 for his text and interpretation of the Twelve Tables); L. Acilius was
writing in the mid-second century BC (cf. Cicero De amicitia 6); Cicero also cites the opinion
of L. Aelius Stilo (late second century). For the Twelve Tables in the late Republic, see
Watson 1974.111—22.
43 Pliny Nat. Hist. 21.7 (Crawford 1996.708-10).
 12 UNWRITTEN ROME

'qni coronam pant ipsepeainiaue tins nirtutisque suae ergo duitiir ei. ’ qttain semi equine
meniisscnt petunia partam lege dici nemo dubitauit. quis ergo hoiios? tit ipsi mortuo
parentibusque eitts, diiin intus positus essct forisucferretur, sine fraude esset irnposita.
There was always distinction in the winning [of coronae], even in sporting
competitions. For they used to go down to the circus themselves to compete at
the games, as well as entering their slaves and horses. Whence that law in the
Twelve Tables: ‘Whoever wins a crown, either himself or his property, or it be
given him for his bravery'...’ Nobody has ever doubted that by ‘won by his
property the law means that [corona] which his slaves or horses had earned. So
what was the honour? That at his death and that of his parents [the corona]
might legally be placed on the body while it lay in the house or was carried out
[for burial].

Rightly or wrongly, Pliny infers from the law that in the early Republic char­
ioteers at the games might be Roman citizens competing in person, rather than
professionals employed by the four teams who raced in the Circus in the late
Republic and under the emperors.44
Such insights are not common. Perhaps the best example is in Macrobius,
where Furius Albinus demonstrates that dancing and singing were acceptable
among the Roman elite in the time of the elder Cato, and not in the time of
Cicero.4’ But there the contrasting social mores were only a century apart, and
the evidence for the earlier situation came from literary texts that offered no
interpretative problems. Evidence for the distant past was much less easy to
handle.
One method of enquiry much favoured by Roman antiquarians was via the
etymology of place-names. ‘Velabrum’, for instance, was a place in the low-
lying valley between the Palatine and the Capitol, reached by' going from the
Forum down the Vicus Tuscus to what is now the church ot S. Giorgio in
Velabro* Varro believed the whole valley had been a lake or marsh (pains)
crossed by ferry-boats, with Velabrum as the landing-stage.47 Although his
derivation of the name from ueliere, ‘to convey’, is far-fetched—other authors
derived it from ttela, ‘sails’48—the idea of a waterlogged valley' seemed plausible

44 See Rawson 1991.389-407, esp. 392-3 on this passage. For the fourfactiones, and their unat­
tested but early origin, see Cameron 1976.56-60.
45 Macrobius Saturnalia 3.13.4-15, citing speeches by Cato and Scipio Aemilianus (respectively
fir. 114-15 and fr. 30 Malcovati).
46 Porphyrio on Horace Satires 2.3.228 (Tuscus dicitur uicus quo itur ad Velabnun); cf. Livy
27.37.14-15.
47 Varro De lingua Latina 5.43-4, 5.156; Solinus 1.14 (cf. Varro Antiquitates huinanae fr. 10.4
Mirsch); Dionysius of Halicarnassus 2.50.2 (Xi|tvr)); Tibullus 2.5.33-4 (per uadd), Propertius
4.9.5—6 (Mibra ... stagnabant), Ovid Fasti 6.405-8 (per Hildas').
48 Ps.Acro on Horace Arspoetica 67 (Felabnini dictum quod uelis transiretur); cf. Propertius 4.9.6.
 UNWRITTEN ROME 13

enough as a genuine memory of the conditions of very early Rome.4’ But it

turns out that Varro’s inference was false: deep-core analysis now reveals that
the valley had not been a marsh at all.50
Like Cincius the ‘mystagogue’, Varro too thought of his studies as a
mystery.51 That was appropriate, since the ancient texts they sought out were
quasi-religious relics, carefully preserved in temples and cult sites.52 Even the
documents of the magistrates of the Republic were thought of in those terms,
as Dionysius shows in his discussion of chronology:53

&T]X.of)xai be e^ aXkiov xe itokkcov koi xwv Kakovpsvwv xipi]xiicd)v

uitopvnpaTwv, a SiabE/Exai jratg jxapa naxpog icai jrepl nokkou
noieixai wig pe0’ eavxdv eaopevotg wonep lepa naxptpa napabibovaf
jtokXoi 5’ eioiv bird xwv ti|.it)tikwv oikiov avbpeg enicpaveig ol
biacpukaxxovxEg avxa.
This is demonstrated by (among other sources) the so-called ‘censorial
commentaries’, which a son receives from his father and takes great care to
hand on to those who come after him as if they were ancestral holy relics.
Many are the distinguished men from censorial houses who preserve them.

Early Rome had no public records office. The outgoing magistrate evidently
took his writing-tablets with him, and kept them as private property.5-1 One
wonders how many can have survived from the fourth century or before, and
whether archaic Latin was any more intelligible when written on wax in
cursive script than when cut into bronze or stone in a monumental inscription.
Roman aristocratic houses also contained family trees in the atrium, the
ancestors’ named portraits linked by lines that might go right back to kings or
gods.” Not surprisingly, the portrait captions (imagimim tituli) were notoriously

49 E.g. Coarelli 1983.229: ‘Questa situazione e quella originaria, prima dei lavori di bonifica
dell’area.’
50 Ammerman 2006.306-7: ‘While the valley bottom was from time to time seasonally wet in
the winter months, especially when the Tiber was in flood for a few days, it was essentially a
dry zone in the other months of the year.’
51 Varro De lingua Latina 5.8 (adytum et initia); cf. n. 39 above. A book by Varro evidently enti­
tled A/ysnuwiji is cited by Fulgentius in the fifth century ad (Expositio sermomim aiitiquornm
11)-
52 Dionysius of Halicarnassus 1.73.1: the Roman historians took their material EK Jtakaitbv
Xoywv ev Itpuiq dEkxoig ow^opeviov.
53 Dionysius of Halicarnassus 1.74.5, seeking to establish the date of the Gallic sack as 120 years
after the expulsion of the Tarquins.
54 See Culham 1991.124-8, inferring from documents of 73 bc (Sherk 1969.135, lines 58-9)
and 39 BC (Reynolds 1982.57, lines 1-3) that the same applied in the late Republic; the
records are described as bfikxoi and Kqpibgaxa, waxed wooden tablets. Cf. Pliny Nat. Hist.
35.7 on codices et monimenta rerum in magistral gestamm.
55 Pliny Nat. Hist. 35.6 (stemmata uero lineis discurrebant ad imagines pictas), Statius Siluae 2.6.11
(stemma innctnin). Kings: Plutarch Numa 1.1. Gods: Suetonius Galba 2.
 14 UNWRITTEN ROME

unreliable as historical evidence;56 no doubt they were regularly repainted,

with the chance every time of new material being added to flatter the family.
However,, a more interesting objection is offered by a Roman author cited by
Plutarch:57

ton be Kai jrepl twv Noga too PaotkEcog /povcov, icafl’ oi)g yeyove,
veavik-fj btatpopa, Kairtep dpxfjg sig toutov KarayEoflat iwv
OTEfigaTiov aKpiPthg SokoOvtiov. dXXa KXcbbtog Tig ev dXeyxq) xpovwv
(oiiiw yap nwg EJuyeypajTrai to |3l/3Xlov) laxopl^ETat Tag gsv dtpyaiag
EKeivag avaypatpag ev rotg KsKrucotg jtdflEot Tfjg noXeoag fygpavio0ai,
Tag 6e vvv tpatvogsvag otjic dXTjflcbg ovyicEioQai 6t’ dvbpcbv
Xapt^ogEvtov Ttalv etg xa itp&xa ysvT] Kai roug tljritpavEOTaTOug otKoug
e£ ov jrpoat]K6vTwv EioPta^ogEvotg.
There is also a vigorous dispute about the time at which king Numa lived, even
though the family trees appear to have been traced down accurately to him
from the starting-point. But a certain Clodius in his Critical Enquiry into
Chronology (which is roughly how the book is titled) forcefully maintains that
those ancient records were lost in the sack of the city by the Gauls, and that the
ones presented nowadays are put together untruthfully by men who wish to
gratify certain individuals by thrusting them into the leading families and the
most distinguished houses when they have no right to be there.

The impheation is that the owners of the great houses of Clodius’ time (when­
ever that was) claimed that their family trees were authentic right back to
Numa; and that Clodius denied the validity of their pretensions by insisting
that the walls that carried the family trees could not be older than 387 BC.
The Romans believed (wrongly) that the Gauls had destroyed the actual
fabric of the city, which therefore had to be rebuilt from the ground up.58 The
reason for that belief was the rarity of authentic early texts. We may be sure that
Clodius’ critical enquiry did not restrict itself to family trees, for if the very walls
were destroyed, what chance was there for wax tablets? As Livy succinctly put
it, ‘any written records there were in public and private documents mostly
perished in the burning city’.59 Although there is no sign in the archaeological
record of any general conflagration in the early fourth century, the reason for
the Romans’ finn belief in it only makes sense if there really were very few
early texts surviving. It may be that only very few had ever existed.

56 Livy- 4.16.4, 8.40.4.

57 Plutarch Numa 1.1, cf. 21.2 on OVK dXqGfj OTEggOTa. See Crawford 1998, who points out
that stenimata began at the bottom of the wall.
58 Livy 5.55.3-5, 6.1.3 (uelut ab stirpibus ... renatae urbis); Tacitus Annals 15.43.1.
59 Livy 6.1.2: si quae [litterae] in coniiiieiitariis pontijiam aliisqne pnblicis prinatisque erant nwiin-
mmtis, imeiisa urbe pleracque intericre (see pp. 263-70 below for the chronicle of the pontifices);
Hutarch Camillus 22.1 (accurate chronology therefore impossible).
 UNWRITTEN ROME 15

How then was it possible for Livy, Dionysius and Plutarch to provide their
vivid and detailed political history of sixth- and fifth-century Rome? The answer
must be that by their time the gap in knowledge had been filled by conjecture.
If the author of that ‘critical enquiry into chronology’ had been able to see
Augustus’ triumphal arch decorated with fists of every yearly magistracy from
Lucius Brutus, and every triumph from Romulus, all dated in years from the
foundation of the city, he might well have wondered where the information
had been found. Some of the names and dates will have come from genuine
research by men like Cincius, others perhaps from a combination of aristo­
cratic pride (those portrait captions) and a patriotic determination to achieve a
continuous history of the Republic, year by year. Once there was a list of
names, a natural predisposition to suppose that the past was like the present
would allow historians (the second- and first-century writers whom Livy and
Dionysius used as sources) to populate the early Republic with optimates and
popnlares, nobiles and noui homines, operating in a wholly recognisable political
and military world.60
Livy knew from his reading in the antiquarians that there were no public
records surviving from the early Republic, and that one rare piece of evidence
attested praetor inaximns as the title of Rome’s chief magistrate in (presumably)
the fifth century.61 But he was not going to allow that knowledge to impugn
the narrative structure provided by his predecessors, which he could turn into
a great moral story of triumph and tragedy.

III
If we are right to infer, firstly that very little documentary evidence survived
from the pre-fourth-century world, secondly that what there was could be
read and understood only with the greatest difficulty, and thirdly that the
narrative historians of the late Republic felt free to ignore it anyway, the
problem of‘unwritten Rome’ remains unsolved.
Indeed it becomes ever more acute, with even Romulus and Remus now
claimed as historical figures.62 The eighth-century wall discovered below the
northern slope of the Palatine by Andrea Carandini in 1988 is roughly on the
line of the ‘poinerinm of Romulus’ described by Tacitus in a famous digression.63

60 See Badian 1966.18-23; Wiseman 1987a.257-8; Gabba 1991.80-5.

61 Livy 6.1.2, 7.3.5 (on which see pp. 306 and 310 below).
62 On the ‘historicity’ of the foundation legend, see for instance Grandazzi 1991, Carandini
and Cappelli 2000, Fraschetti 2002, Carandini 2003, Carandini 2006.
63 Tacitus Annals 12.24.1-2. Cf. Carandini 1997.497 and 580 on ‘la ricostruzione di Tacito,
basata sul ricordo recente delle porte e di tratti delle mura del Palatino, conservatisi fino
all’eta di Augusto [... | la descrizione di Tacito, redatta poco dopo 1’eta augustea, durante la
quale porte e tratti delle mura erano ancora in vista.’
 16 UNWRITTEN ROME

Carandini assumes that the Varronian date for the foundation of the city is
historical, and that our literary sources preserve the ‘living memory’ of oral
tradition.64 Others disagree, of course, but he defies his critics with a fine
disdain:65

Chi e che interprets le mura in termini di fondazione di un primo re chiamato

o meno Romolo: Andrea Carandini o oltre venti generazioni di Romani senza
soluzione di continuita nella loro memoria?

Those twenty generations represent the five and a half centuries back from
Fabius Pictor’s narrative of Romulus to the supposed date of the foundation of
Rome.66
Much the same length of time separated Geoffrey of Monmouth’s detailed
history of King Arthur from the date of his supposed reign in the sixth century
AD.6' Many scholars have wanted to believe in Arthur as at least a powerful
warlord,68 but of course no one supposes that Geoffrey’s conqueror of
Norway, Demnark, Gaul and Rome was ever anything but a fantasy. The
latest defender of the historicity of the reign of Arthur goes as far as he dares in
appealing to oral tradition:69

Working on the supposition of a historian writing down the words of the oldest
person available, recording what they had been told by, for example, a grand­
parent when they were a child, [. ..] 200 years is the maximum rime one can
reasonably expect oral tradition to survive without serious distortion.

(That would bring us within range of a hypothetical eighth-century written

source of the extant ninth-century Historic! Brittonum, which calls Arthur a war­
leader, dux bellonim. QED...) But even that, deeply implausible as it is, pales
in comparison with Carandini’s vision of twenty generations ‘with no break in
the continuity of their memory’.
Five or six centuries are also what separate the composition of Homer’s
Hind from the destruction of Troy VI (conventionally dated to about 1280 bc).
Notoriously, Heinrich Schliemann believed that the Homeric epics were liter­
ally true, and even after more than a century of debate on the historicity of the
Trojan War it is still possible to argue, in all seriousness, that ‘Homer’s Iliad

64 E.g. Carandini 1997.37: ‘La tradizione orale, raccolta e trammandata nei testi, e spesso
I'unica vivente memoria che si giunga dall’intemo di quel lontano passato.’
65 Carandini 2003.11-12. See Wiseman 2006a for a detailed discussion of Carandini’s method.
66 Fabius dated the foundation to 747 bc (Dionysius of Halicarnassus 1.74.1). Narrative: Fabius
Pictor FGrH 809 F4 = Dionysius of Halicarnassus 1.79.4-83.3, Plutarch Romulus 3.1-6.7.
67 Geoffrey dated the death of Arthur to AD 542. Narrative: Historic! regain Britanniae 8.19-11.2
(completed about 1136).
68 Excellent summary and discussion in Hutton 2003.39-58. For the written sources on fifth-
and sixth-century Britain, see Dark 2000.32-48, esp. 43-5 on ‘inadmissible evidence*.
69 Gidlow 2004.91, going well beyond his stated source (Dark 2000.43).
 UNWRITTEN ROME 17

[has] the status of a source text’.7" However, not even Joachim Latacz wants to
make Achilles and Agamemnon historical figures like Carandini’s Romulus.
Everything depends on the reliability of ‘oral tradition’—or ‘cultural
memory’, in the more fashionable recent term. We might appeal to modem
anthropological evidence,71 but perhaps it is more appropriate to listen to a
great historian writing in Rome in the second century bc. Polybius notes that
in beginning his main narrative in 220 BC he is dealing with events ‘in our own
time and that of our fathers’; either he himself had taken part in them or he had
heard about them from eye-witnesses. Then he goes on:72

to yap avtoTEpw jrpooka|.ipdvEiv Toig ypovotg, <bg cncof|v cttcofj^

ypacpetv, ouk e<paivE0’ fjptv aacpakEtg exelv ovte Tag 6iakq\|’eig ovte
xag diTocpdoEig.
For to include material further back in time, so as to write hearsay on hearsay,
did not seem to me to be safe for either judgements or assertions.

Cultural memory, or hearsay on hearsay? Twenty generations, or two at the

most?
As Moses Finley pointed out many years ago, wherever independent
evidence allows the comparison, traditional oral narrative about events in the
past ends up by distorting those events in a quite fundamental way.73 Every
time the tale was told, it was told to a particular audience on a particular occa­
sion, and their expectations and preconceptions changed over time with the
changing circumstances of the community. Oral tradition does not transmit
what has become obsolete or irrelevant. It is also hospitable to creative fiction,
as the Greeks and Romans understood very well.74 The will to believe is
always strong, and a good story-teller can easily activate it.7’

70 Latacz 2004.91, 138—on the basis of Ili/iisa (llios) and AMtiyaiva (Achaioi) in the Hittite
texts, and Daikipi (D.itiaoi) in an Egyptian inscription of the fourteenth century BC.
71 As do the two best discussions of oral tradition known to me, Davies 1984 and Hutton
2003.18-27.
72 Polybius 4.2.3.
73 Finley 1964.2—3, citing the Sent; oj Roland, the Nibelniigenlied, and the South Slav tradition
on the battle of Kosovo.
74 See for instance Hesiod Theogony 27 and Polybius 3.33.17 on ‘lies like the truth’; Thucy­
dides 1.21.1 and Tacitus Annals 3.19.2 on the achievement of instant myth; Seneca
Apocolocyntosis 1.2, Quintilian 1.8.21 and Lucian On Writing History 32 on saying whatever
comes into your head.
75 See Finley 1964.2 on classicists’ ‘will to believe’ in the Trojan War; Hutton 2003.12 on
fiction’s ‘capacity to overpower virtually any facts given sufficient time . Remember Mopsa
and Autolycus in The Winter’s Tale (act 4 scene 3): ‘Is it true, think you?’ ‘Very true, and but
a month old. Here’s the midwife’s name on’t, one Mistress Taleporter, and five or six honest
wives who were present. Why should I carry lies abroad?’
 18 UNWRITTEN ROME

Livy, like Homer and Geoffrey of Monmouth, was a good story-teller. But
the fact remains that he knew hardly anything about the real conditions of
archaic Rome, and what little he did know he ignored.

IV
In their attempts to evade the consequences of this unwelcome conclusion,
some modem scholars appeal to one particular category of data as impervious
to change. They take it as axiomatic that cults and rituals remain the same over
long periods of time, and therefore that reports of them in late authors are valid
as evidence for early Rome.76 Moreover, it is only a step from that to accepting
as necessarily ancient the aetiological stories offered by our sources to explain
those cults and rituals.77
But is this escape clause really valid? Even for the most obviously ancient of
Roman cults, as we shall see in chapter 4, the sources seem to imply a sequence
of developing rituals and interpretations reflecting the changing needs of
Roman society.78 And if the Lupercalia are too controversial an example, we
can take the new evidence about Anna Perenna,79 whose cult underwent
substantial changes in the first century AD.
Ovid gives a detailed account of Anna Perenna’s festival on the Ides of
March. It took place, he says, in an open field by the river bank, which a
contemporary calendar places at the first milestone on the Via Flaininia.8"
Eighty years later, Martial refers to Anna’s cult site as a sacred grove, evidently
not far from the Milvian Bridge, which was at the third milestone on the
Flaininia.81 Not only that, but he describes the grove as ‘delighting in the blood
of virgins’, which seems to imply some kind of pre-wedding initiation tor
young girls. Scholars infer an archaic rite unchanged for centuries,82 but Ovid
knows nothing about that; his celebrants dance and make love with girls who
are clearly sexually experienced. A further anomaly is that a calendar of AD 354

76 See for instance Coarelli 2005.32 on ‘una delle costanti pili note di ogni rito religioso, il suo
carattere stabile e conservatore’; Carandini 2006.412 on ‘le istituzioni sacre, come i culti,
quanto di piu durevole e congelato possa esistere nel mondo uniano e pertanto una delle
chiavi perintendere la storia dei primordi’; and Schultz 2006.12 on the Romans as ‘a people
who prided themselves on their scrupulous maintenance of ancient forms, [and] are very
likely to have tried to preserve the details of ritual observance for centuries’.
77 Carandini 2006.412 (my emphasis): ‘1 Romani, in particolare, cancellavano quasi nulla del
loro passato religioso e lejtfcnifario, conservandolo piamente attraverso secoli.’
78 See pp. 52-83 below on the Lupercalia; pace Coarelli 2005.30 on 'le ragioni intrinseche,
strutturali, che collegano in un insieme coerente tin rito di purificazione della citta con tin
rito di fertilita’.
79 Piranomonte 2002, on which see now Wiseman 2006b.
80 Ovid Fasti 3.523-30; Fasti Paticaui on 15 March (Degrassi 1963.172—3).
81 Martial 4.64.16-17 (cf. 23-4 on the proximity of Muluius).
82 E.g. Torelli 1984.57-66; Boels-Janssen 1993.23-39.
 UNWRITTEN ROME 19

dates Anna’s festival not to the Ides of March but to 18 June. Ovid knows
nothing of that either.83
We now know that during the first century AD the cult site was moved to a
sacred spring complex on the north side of the Monti Parioli [fig. 7], no longer
by the river and no longer on the Via Flaminia.84 We also know from the
inscriptions that at least by the second century AD it was the scene of mime
competitions [fig. 8].85 It seems beyond doubt that the place, date and nature of
Anna Perenna’s festival had all been changed.86

n
(

....
Fig. 7 Anna Perenna cult site at Piazza Euclide, Rome: the remains of a rectan­
gular basin with second-century AD inscriptions inserted into the wall.

The virgins’ rite of passage inferred from Martial recalls the custom that so
horrified Christian polemicists, of Roman brides surrendering their virginity
on the phallic image of the god Priapus.87 The only direct evidence for this
ritual defloration seems to be a marble relief once in Naples [fig. 9]; however,

83 Fasti Filocaliani (Degrassi 1963.248-9); not in Ovid Fasti 6.

84 Piranomonte 2002.34—7 for the date: coins in the basin from Augustus to the late fourth
century AD.
85 Piranomonte 2002.26-33; Slater 2005.319-20; Wiseman 2006b.55-9.
86 Anna’s myth was changed too: see Ovid Fasti 3.647—55, Silius Italicus 8.178—83, and
Wiseman 2006b.58-9 on the differences between them.
87 Lactantius Divine Institutes 1.20.36, Augustine City of Cod 6.9 and 7.24 (CSEL 40.292, 337).
 UNWRITTEN ROME

w
1

Fig. 8 Inscription from Anna Perenna’s cult site (on the left in fig. 7): ' The vow
[which] once I had made to the consecrated nymphs, who deserved it because
of the victory of my good patron C. Acilius Eutyches, we pay; and we attest in
verses that they are sacred, and we dedicated an altar to the welcome springs.
Eutychides, freedman.’ The metre is iambic senarii. An archinitintis L. Acilius
Eutyches, clearly a relative, is attested in AD 169 (C/L 14.2408 = ILS 5196).
 UNWRITTEN ROME 21

II
Fig. 9 Nineteenth-century drawing of a marble relief showing a girl about to be
prepared for marriage by Priapus. The stone was once in the gabinetto seareto of
Naples Museum; present whereabouts evidently unknown.

the best-omened time for Roman weddings was after the Ides of June,88 and
according to Martial it was in June that Priapus was ‘received by proud Venus’
(whatever that means) in a ceremony unknown to Ovid.89 Since Anna Perenna
shared her new cult site with the nymphs,90 and since from the first century AD
on' , ; ds Priapus is described as the nymphs’ companion,91 the conclusion is
inevitabl ■ that Anna’s grove, ‘delighting in the blood of virgins’, and Anna’s
1\ no'v on IS June at the auspicious time for weddings, were the setting
!’<•• initiation of young girls into matrimony by droit de seigneur.
■ important to understand what this new information implies. If one puts
.M m comparative anthropology and the history of religions ‘under-
,’ii autonomous discipline’,92 with data from various times and various
, .. vi.chronically juxtaposed, it is natural to see Martial’s description of
Anna Petenna’s sacred grove as evidence for an archaic initiation rite, and
thence to conclude, in a circular argument, that ancient rites survived
unchanged because the Romans were particularly conservative in religious
matters. 1 he discovery of the Anna Perenna inscriptions reveals how precar­
ious that method is. The outcome is just the opposite of the ‘history of
religions’ model: since the Romans were evidently quite happy to reorganise
an ancient cult in the first century AD, we have no reason to doubt that they

88 Ovid Fasti 6.219-34, cf. 713-14.

89 Martial 3.68.7-10; not in Ovid Fasti 6.
90 Piranomonte 2002.26-30 for the nyniphae sacratac (see fig. 8); Wiseman 2006b.55-9.
91 Petronius Satyricon 133.3, Corpus Priapeoruiii 33.1-4, CIL 14.3565.13-32 (Courtney
1995.148-9).
92 I borrow the phrase from Coarelli 2005.30.
 22 UNWRITTEN ROME

did so periodically throughout their history. Cult and ritual change, like every­
thing else in society.
What matters is not just what the new texts themselves say, but how they
have led to the explanation of otherwise baffling references in the literature. Of
course I may be prejudiced, but this example seems to me to vindicate the clas­
sicists traditionally empirical method—close reading of the sources, careful
consideration of what they may or may not presuppose, and the avoidance of
unexamined preconceptions.

There are no short cuts, there are no magic wands, there is no time machine to
take us back to unwritten Rome. All we have is evidence and argument—
what survives from the ancient world, and what we can make it tell us about
what we want to know.
‘What survives from the ancient world’ includes above all a huge and varied
corpus of literary texts, the writings of poets, playwrights, statesmen, scholars,
historians, satirists, novelists, theologians. The narratives such authors provide
of what they believed was early Roman history are of very little direct use to
us, since they could have had no knowledge of the reality of Roman experi­
ence before (say) the late fourth century BC. But the texts also tell us a great
many things about which their authors were uniquely well informed, and even
the most unhistorical narrative may sometimes reveal information the author
was not aware he was giving.
What the modern historian has to do was well described by R.G. Colling-
wood sixty years ago:93

As natural science finds its proper method when the scientist, in Bacon’s
metaphor, puts nature to the question, tortures her by experiment in order to
wring from her answers to his own questions, so history finds its proper method
when the historian puts his authorities in the witness-box, and by cross-ques­
tioning extorts from them information which in their original statements they
have withheld, either because they did not wish to give it or because they did
not possess it. [...] Where the scissors-and-paste historian said quite confi­
dently ‘There is nothing in such-and-such an author about such-and-such a
subject’, the scientific or Baconian historian will reply, ‘Oh, isn’t there? Do you
not see that in this passage about a totally different matter it is implied that the
author took such-and-such a view of the subject about which he says nothing?’

93 Collingwood 1946.237, 270. Francis Bacon’s empiricism is rightly perceived as inimical to

the idea of‘cultural memory’: see Carandini 2006.15 (‘Per i Baconiani la culture, al fondo, e
solo uno svantaggio nell’osservazione, perche moltiplicherebbe i preconcetti’), cf. 413 on
my own Bacoiiisiiio.
 UNWRITTEN ROME 23

This book is a sequence of such enquiries, to see whether it is possible to

extract usable evidence about pre-literary Rome.
We begin with the cross-questioning of two witnesses, Livy and Ennius;
then follow four studies (chapters 4-7) on Roman cults and rituals, and three
more (chapters 8-10) on a particular type of religious festival, the ‘games’ (Jtidi)
put on in honour of the gods; that in tum leads to three chapters (11-13) on
the nature of Roman theatre performances; two studies on the origin and
nature of Roman historiography (chapters 14-15) are followed by a discussion
of the interpretation of archaeological evidence (chapter 16); and the final two
chapters test the evidence for a defining moment in early Roman history, the
expulsion of the Tarquins and the establishment of a republic.
I

CHAPTER TWO

What Can Livy Tell Us?

i
When Titus Livius began his great history of Rome, some time around 30 BC,
Roman literature was more than two centuries old. Its history was already a
subject of learned controversy, and the earliest text that the scholars of his time
had been able to identify was a play by his namesake Livius Andronicus,
evidendy datable to 240 BC.1 The first historians of Rome, Fabius Pictor and
Cincius Alimentus, were writing at least a generation after that,2 so even
through his earbest sources Livy could hardly have had any sense of what
Roman culture was like in its pre-literary phase.
However, we need not assume automatically that Livy can tell us nothing
about ‘unwritten Rome’. If we read him carefully enough, it may be possible
to use the material he transmits, and even the phraseology he uses, to infer
something about that lost world. For instance, how did he deal with the idea of
an oral culture and the origins of literature? His treatment of 2-10 BC is lost, but
we do have his reaction to the work of Livius Andronicus in a later context.
In 207 the aged poet was asked to compose a hymn to J uno, for a ceremony
in expiation of alarming prodigies:3

decreuere item pontijices ut iiirgines ter nouenae per urbein eimtes carmen canerent. id cum
in louis Statoris aede discerent coiiditum ab Liuio poeta carmen, tacta de caelo aedis in
Aiientino Imionis Reginae ... tiini septem et uiginti iiirgines, indutae tiestem,
canneii in Iiinoiieiii Reginam canentes ibant, ilia tempestate forsitan laudabile mdibus
ingeniis, limit abhorrens et incoiiditiim si referatur.
The pontiffs also decreed that three bands of maidens, each consisting of nine,
should go through the city singing a hymn. This hymn was composed by the
poet Livius, and while they were practising it in the temple of Jupiter Stator,

1 Cicero Brutus 72-3 (from Atticus), Tusculan Disputations 1.3; Aullis Gellius 17.21.42 (from
Vano?). The coincidence of the names Titus Livius/Livius Andronicus may be of interest to
Shakespearean scholars.
2 Livy 1.44.2 (sm'ptorinii antiqiiissimus Fabius Pictor), 21.38.3 and 22.7.4 (Cincius and Fabius as
contemporary witnesses, 218-217 UC); cf. Dionysius of Halicarnassus 1.6.2 on Roman histo­
rians, tbv tloi irpeoflvTUTOt Kdivrdg te 4>d|3tog Kai Aeukioc KiyKtog.
3 Livy 27.37.7 and 13 (trans. Roberts, Everyman Library, slightly adapted).
 WHAT CAN LIVY TELL US? 25

the shrine of Queen Juno on the Aventine was struck by lightning. [. . .] Then
twenty-seven maidens, vested in long robes, walked in procession singing a
hymn in honour of Queen Juno, which was perhaps admired in those rude days
but would be considered very uncouth and unpleasing if it were recited now.

The hymn was ‘composed’ (condituni). That means it was written down, for
conciere—‘to store, put away, preserve’—is a regular word for literary composi­
tion, especially of poetry.4 The act of writing stores the poet’s words.
However, Livy comments that if it were repeated in his day it would seem
incondituin, evidently meaning something like ‘ardess’.
There are 29 other places in the extant text where Livy uses the adjective
inconditus. In 21 of them it means ‘disorderly, unstructured’, to describe a
crowd, an undisciplined army, or a city under mob rule.5 The other eight
passages, however, offer better parallels for what Livy says about Livius
Andronicus’ hymn. For instance, he reports that the priests at the Apollo
temple at Hieracome in Caria deliver the oracular responses uersibus hand incoii-
ditis.6 English translations render the phrase ‘in smooth and graceful verses’
(Everyman), ‘in verses not without polish’ (Loeb), ‘in verses of some literary
quality’ (Penguin), all indicating by contrast what tiersiis inconditi would have
been.
A more direct indication comes in the excursus on the origin of drama in
book 7, where Livy reports that Etruscan dancers were brought to Rome to
perform in the year 364 BC:7

imitari deinde cos iiiiientus, sinml inconditis inter se iocidaria ftiiidentes uersibus,
cocpcre.
Afterwards the young men [of Rome] began to imitate them, exercising their
wit on each other in burlesque verses.

That is the Everyman translation; the Loeb has ‘in uncouth verses’, the
Penguin 'in crude improvised verse’. The essential point is evidently that the
young men’s banter was not only unsophisticated but unscripted as well.
That sense also fits the remaining six examples of inconditus in Livy, all of
which describe the songs and jokes of Roman soldiers in the triumphal proces-

4 E.g. Cicero rl</ Atticiini 1.16.5, Lucretius 5.2, Virgil Eclogues 6.7, Horace Epistles 1.3.24,
Propertius 2.1.14; full references at TLL 4.153.74-154.29. The usage may go back to the
Twelve Tables: Crawford 1996.677-9.
5 Turbo-. 21.57.12, 22.45.3, 25.1.4, 25.13.10, 25.15.13, 29.1.22, 36.33.4. Multitudo: 24.29.1,
26.40.16, 27.32.8, 32.13.14, 43.10.5, 44.45.6. Army: 29.34.11, 30.11.5, 30.28.3, 35.28.3,
42.66.8, 44.39.1. Revolutionary Syracuse, 214-12 bc: 24.24.2, 26.40.1.
6 38.13.1.
7 7.2.5. Cf. 7.2.7, non, sicut ante, Fescennino tiers" similem incoiiiposituiii temere ac ntdeni attends
iaciebant; Servius auctus on Eclogues 6.7 defines inconditus as incoinpositus.
 26 UNWRITTEN ROME

sion? As Stephen Oakley rightly observes, ‘these incondita or inconditi uersus

exercised a curious fascination upon LJivy], who often alludes to them’.9 In
tact, there are eight such episodes, all in the first ten books, and since they
purport to be just what we are looking for, namely evidence for the oral
culture ot the early Republic, a close comparison of them may be useful for
our theme.

II
1: The batde of Algidus, 458 BC. Rome makes war on the Aequi after their
leader Cloelius Gracchus treats her ambassadors with contempt; the consul L.
Minucius and his army are surrounded and besieged; Cincinnatus is called from
tlie plough to be dictator; the Aequi are defeated and sent beneath the yoke. At
the dictator’s triumph, tables are spread with food for the soldiers before all the
houses:"1

epidantesqiie cuni carmine triuniphali et solleninibus iocis coniisantiinn rnodo ctimtnt

secuti sunt.
They regaled themselves and followed the chariot in the manner of revellers,
with triumph-songs and the joking appropriate to the occasion.

2: A. Cornelius Cossus, 437 BC. Rome makes war on the Veientes after their
king Lars Tolumnius has had Roman ambassadors murdered; Cossus, a hand­
some cavalry officer, kills Tolumnius in single combat. The dictator,
Mamercus Aemilius, holds a triumph:11

longe maximum triumphi spcctaculmn fuit Cossus, spolia opima regis interfecti gerens; in
cum milites cannina incondita acquantes eiun Romtilo canerc.
By far the most prominent sight in the triumphal procession was Cossus,
carrying the ‘spoils of honour’ of the king he had slain, while rhe troops sang
impromptu songs comparing him to Romulus.
In the procession all the citizens look not at the dictator’s chariot but at Cossus,
who hangs his spoils in a solletnnis dedicatio next to those of Romulus in the
Jupiter Feretrius temple. By order of the people, the dictator offers a gold
crown to Jupiter.12

8 4.20.2 (cannina), 4.53.11 (uerstis), 5.49.7 (ioci), 7.38.3 (iocus); 7.10.13, 10.30.9 (incondita as a
substantive, as at Varro Menippean Satires fr. 363B and Virgil Ecognes 2.4).
9 Oakley 1998.361.
10 3.25.7-8 (contempt), 3.26.7-10 (plough scene); quotation from 3.29.5.
11 4.17.1-6 (murder); quotation from 4.20.2.
12 4.20.3 (procession and dedication), 4.20.4 (crown).
 WHAT CAN LIVY TELL US? 27

3: M. Menenius, 410 BC. The plebeians complain about the illegal occupation
of ager ptiblicus; Menenius as tribune of the plebs proposes an agrarian law and
blocks the consul’s levy of troops, but a dangerous incursion by the Aequi
allows the consul to prevail. Though the Aequi are defeated, the plebs and the
soldiers are still resentful when the consul holds his ouatio:'5

alternis inconditi uersus militari licentia iactati qnibus consul increpitus, Meneni celebre
nomen fait, cunt ad onmein mentionem tribnni fauor circumstantis populi atm uocibus
militant certaret.
Impromptu songs were shouted in turns with soldiers’ freedom of speech, in
which the consul was abused but Menenius’ name was glorified, while at every
mention of the tribune the applause of the people standing by competed with
the soldiers’ voices.

The patricians are concerned less about the people than about the outspoken­
ness of the soldiers, which Livy describes as prope solletnnis.u

4: Camillus and the Gauls, 390 BC. During the siege of the Capitol, both sides
suffer from famine and a truce is arranged; terms are agreed for the Gauls to
withdraw, but Brennus contemptuously throws his sword on to the scale.
Then Camillus arrives, anti with the help of the gods defeats the Gauls in two
battles:13

triumphans in nrbeni redit, interqne iocos militares qnos inconditos iaciunt, Romulus ac
parens patriae conditorqne alter nrbis hand uanis landibus appellabatnr.
He returned to the city in triumph, and amid the impromptu jokes shouted by
the soldiers he was called ‘Romulus’, 'father of his country’, ‘the second
founder’—no empty praise.

5: Manlius and the Gaul, 361 BC. The Romans face a Gallic war-band at the
Amo bridge on the Via Salaria; a huge Gaul issues a challenge to single combat;
young Titus Manlius fights and kills him with the help of the gods, and takes
his torque; the Gauls are rooted to the spot by fear and astonishment. The
Romans joyfully escort Manlius to the dictator:16

inter canninuni prope <in> modinn incondita quaedam inilitariter iocnlantes Torqnati
cognomen audituni; celebration deinde posteris etiani Jamiliae honorifuit.

13 4.53.2 (lex agraria), 4.53.6 (ager publicus); quotation from 4.53.11—12.

14 4.53.13: Menenius was prevented from obtaining the post of trib. mil. cos. pot.
15 5.48.4 (truce), 5.48.8-9 (Brennus), 5.49.1 and 5 (gods); quotation from 5.49.7.
16 7.10.4 (gods), 7.10.12 (fear and astonishment); quotation from 7.10.13 (Oakley 1998.147
for the text).
 28 UNWRITTEN ROME

Amid the soldiers’ impromptu joking, almost like songs, the name ‘Torquatus’
was heard, which from then on was a mark of honour to the family’s descen­
dants too.

The dictator gives a speech in his praise at an assembly (contio), and rewards
him with a gold crown.17

6: Faliscans and Tarquinians, 356 BC. Rome makes war on Falerii and
Tarquinii; the Roman soldiers are at first terror-struck by the enemy priests
advancing like Furies with blazing torches, brandishing snakes. But then they
rout them and capture their camp:18

praeda ingenti parta uictores reuerterunt, militaribiis iocis cum apparatuiH hostium turn
siiiim mcrepantes panorem.
They captured a huge amount of booty and returned victorious, mocking in
their soldiers’jokes both the enemy’s paraphernalia and their own fear.

7: The elder Decius Mus, 343 BC. A Roman army advancing into Sammum is
trapped in a narrow valley; Decius, a military tribune, sees a high point unoc­
cupied by the enemy and takes a group of picked men to hold it; the Samnites
are dumb-struck, the main army escapes and Decius and his men, with the
help of Fortuna, fight their way back to it; the consul holds an assembly (conirn)
in his honour, but Decius urges an attack while the enemy are paralysed with
fear (pauore attoniti); the Samnites are routed.19 Decius is awarded a gold crown
and other decorations:2"
consoles ambo de Samnitibus triuniplianint sequente Decio insigui cum Imide donisque,
cum incondite militari ioco hand minus tribimi celcbre nomen qtiam consul esset.
Both consuls held a triumph over the Samnites, with Decius following,
conspicuous by praise and decorations, since in the soldiers impromptu jokes
the tribune’s name was no less celebrated than a consul.

8: The battle of Sentinum, 295 BC. On the Roman left wing the consul Decius
Mus (son of the hero of the last item) attacks the Gauls, but his cavalry are
panic-stricken; facing defeat, he utters the deuotio prayer and formally sacrifices
himself; the Gauls in turn are paralysed by dread; his colleague Fabius

17 7.10.14 (niirisipic pro comione earn piignam laudibtis tnlit).

18 7.17.3 (iielut lymphati), 7.17.4 (liana miracnla); quotation from 7.17.5.
19 7.34.1-37.3, elaborate narrative: 7.34.6 (Fortuna), 7.34.8 (admiratione pauentibus), 7.33.5, b
and 12 (Fortuna), 7.36.10 (pairore attonitos); pauor also at 7.36.3 and 13.
20 7.37.1-2 (including corona obsidialis); quotation from 7.38.3.
 WHAT CAN LIVY TELL US? 29

Rullianus defeats the Samnites on the other wing and takes the Gauls in the
rear. Fabius triumphs alone:21

milites triumphantem secuti sunt, celebrata inconditis militaribus non magis uictoria Q.
Fabi quatn mors praeclara P. Deci est excitataque meinoria parentis, aequata eucntu
publico priuatoque,filii laudibus.
The soldiers followed his triumph. In their impromptus the glorious death of
Publius Decius was celebrated no less than the victory of Quintus Fabius, and
the memory of the elder Decius, equaled by the outcome both privately and
publicly, was recalled by the praises of the younger.

The first point to make about these eight passages is an elementary one: Livy’s
sources were not reporting the soldiers’ songs at first hand. This is not evidence
like that of Suetonius and Velleius Paterculus on the soldiers’ triumph songs in
46 and 43 BC.22 It is true that Fabius Pictor or Cincius Alimentus could have
talked to old men whose grandfathers had told them what the soldiers sang at
the triumph in 295 BC; but item 8 is only the last in a coherent sequence which
has to be interpreted as a whole. The phenomenon must be, in some sense, a
literary one.
One of the group (item 3) stands out for its unusual context, political rather
than military. The combination of tribunician resistance to the levy and
agrarian legislation against illegal occupiers of ager pnbliais is powerfully remi­
niscent of the years 151-133 BC,23 and it is not unlikely that the episode was
elaborated, or even created, in the second century BC. But it is far from clear
why the elaboration should have taken the form of soldiers’ songs.
By definition, incondita ainnina are not written down. But at some point in
the development of the historiographical tradition on the early Republic—
between, say, 210 BC and 20 BC (from Fabius Pictor to Livy’s first
decade)—the content of the soldiers’ extempore songs on at least eight occa­
sions became "known’, and thus available as data for Roman history: The
question is: how?

21 10.28.10 (uelut lyinphaticus pauor), 10.28.13-18 (deuotio), 10.29.1 (nix humanae opis), 10.29.2
(uelut alienata mente), 10.29.4 (Furies and fomiidd), 10.29.7 (attoniti); quotation from 10.30.9,
referring to the elder Decius Mus’ deuotio in 340 bc.
22 Suetonius Diuus lulius 49.4, 51, 82; Velleius 2.67.3; Courtney 1993.483-5.
23 Levy: Livy Periochae 48 and 55 (consuls imprisoned by tribunes, 151 and 138 BC); Taylor
1962.19—27. Ager pnbliais: Plutarch Tiberius Gracchus 8.1—4 (C. Laelius, 140 bc?), Appian
Civil Wars 1.7—11 (lex Sempronia, 133 BC); on Laelius, see Astin 1967.307-10.
 30 UNWRITTEN ROME

Ill
Let us look more closely at what these stories have in common. All but two of
them explicitly celebrate the name of a particular hero, whether that of the
triumphant commander (item 4), his fallen colleague (item 8), a particularly
heroic soldier (items 2, 5, 7), or a champion of the plebs (item 3). One of the
two exceptions (item 1) is probably not an exception at all, since praise of
Cincinnatus is implicit in the story. (The other is item 6, where the soldiers
mock the enemy’s flummery and their own previous fear.) The phraseology is
consistent: praise, of course (Jans and lattdare),24 but also celebration (celeber and
celebrate), especially of the hero’s name.25
Where the hero is a junior officer, there may be an assembly (contid) in
which the commander praises and decorates him;26 direct speech may charac­
terise his fearlessness, initiative or good discipline.27 Direct speech is in any case
a very conspicuous characteristic of these episodes, from ‘satin salite?’ and ‘nae
uictisl’ (Cincinnatus at the plough, Brennus at the scales) to Decius’ nocturnal
lecture to the soldiers on Fortune and necessity.28 Speeches by the protagonists
frequendy establish the moral parameters of the story, as in the Romans’
response to Cloelius Gracchus’ insult (item 1), Cossus’ reaction at the sight of
the impious Tolumnius (item 2), the Gaul’s challenge at the Anio bridge (item
5), and Decius’ recognition of his inherited fate (item 8).29
Particularly interesting are the speeches Livy does not give us, but which
mark a dramatic turn of events. In item 3, Menenius’ defiant harangue to the
people is interrupted by a messenger with news of a military disaster; in item 4,
the action is precipitated by conloqnia between Gauls and Romans; in item 6,
the contemptuous speech of the consul and officers turns the soldiers from
panic to blind rage; in item 7, Decius interrupts the consul’s speech in his
praise by pointing out the opportunity for a decisive attack; and in item 8, the
speech of the pontifex M. Livius makes the consul’s self-sought death a Roman
victory.3" These speeches are an integral part of the action.

24 4.53.11; 5.49.7; 7.10.12 and 14; 7.36.7 and 9, 7.37.1, 7.38.3; 10.29.20, 10.30.9.
25 4.20.3 (celebritas); 4.53.11 (celebre nomen); 7.10.13 (cognomen celebration); 7.38.3 (celebre nomen);
10.30.9 (celebrata). Honorific comparisons: 4.20.2, 5.49.7 (Romulus); 10.30.9 (father).
26 Conlio: 7.10.14 (corona anrea); 7.36.9, 7.37.1 (corona anrea, etc); cf. 3.29.3 (corona anrea voted
to Cincinnatus by Minucius’ army), and 4.20.4 (corona anrea dedicated to Jupiter), where
populi iussu may imply a conlio.
27 4.19.3; 7.10.2-4; 7.34.4-7 and 13-14, 7.35.2-12, 7.36.5-6. (‘Marte nirtnte’ at 7.10.4, 7.36.5.)
28 3.26.9, 5.48.9; 7.35.2-12. Two other two-word speeches at 3.27.7 (‘adcelera, signijer!’,
'seipicre, miles!’).
29 3.25.8 ('lieonnn honiimnnqne sinuil niolata iura’); 4.19.1 (‘hicine est mptor foederis linmani uiola-
torque gentimn juris?’); 7.9.8 (‘qnem nunc Roma ninnn fortissimnm habet. . . ’); 10.28.12-13 (’quid
ultra niororfainiliarefatiim?’).
30 4.53.2-3; 5.48.4 and 8; 7.17.4; 7.36.9-10; 10.29.3-4. Another example may be 3.25.9 (item
1), if the news of the Sabine raid interrupted the tribunes.
 WHAT CAN LIVY TELL US? 31

Equally pervasive is the role of the gods. The action of item 1 begins when
the Aequian leader tells the Roman ambassadors to address their complaints to
a nearby oak-tree, and they take him at his word:

‘et haec sacrata querns et quidquid deonnn est audiant... ’

‘May this holy oak and whatever gods there are hear . .. ’

Similarly, what Cossus avenges in item 2 is Tolumnius’ sacrilegious murder of

ambassadors:

‘iam ego hanc mactatam ilictimam, si modo sancti quicquam in tern's di uolunt, legatonun
mambas dabo.’
‘If it is the gods’ will that there should be anything sacred in the world, 1 shall
give him to the shades of the ambassadors as a sacrificial victim.’

The Romans’ piety will be rewarded.31 In item 4, Camillus’ intervention is

attributed explicitly to divine help, in a very conspicuous ring construction:32

sed dique et homines prohibuere redemptos uiuere Romanos ... iam uerterat Jbrtuna,
iam deonim opes luimanaqiie consilia rem Romanam adiuuabant.
But both gods and men forbade the lives of Romans to be paid for . . . Now
fortune had turned, and the gods’ help and human strategy aided the Roman
cause.

In item 5, Manlius’ commanding officer commends his pietas as well as his

bravery:33

‘perge et nomen Romanian innictum iiinantibus dis praesta.’

‘Proceed—and with the gods’ help prove the Roman name to be invincible.’

For Decius in item 7 it is a particular divinity, Fortuna Populi Romani, who

favours his exploit.34 For his son, the deuotus of item 8, the gods who matter are
Tellus, the Di Manes, and the addressees of the pontiffs archaic prayer to acti­
vate the anger of caelestes and iiijeri.53
The sollennies precntiones of the self-sacrifice recall the sollemnis dedicatio of
Cossus’ ‘spoils of honour’ in item 2. Even the soldiers’ shouting, paradoxically

31 3.25.8,4.19.1.
32 5.49.1 and 5. Ogilvie 1965.737: ‘At that juncture divine intervention brings Camillus on to
the stage (49.1, cf. 49.5), and L. stresses that the Romans have earned their reprieve by their
piety.’
33 7.16.4.
34 7.34.6 (cf. 1.46.5)—also known as Fortuna Publica (ILLRP 112, Ovid Fasti 4.375-6, Fasti
Praenestini on 5 April), Fortuna Publica Populi Romani (Fasti Esquilini on 25 May), or
Fortuna Publica Populi Romani Quiritium (Fasti Caeretani and inagistronmi uici on 25 May).
35 10.28.13—17, with a reference back from 10.28.15 to 8.9.6 (‘lane, luppiter, Mars pater,
Quirine, Bellona, Lares, Dini Nouensiles, Di Indigetes...’); 10.29.1 (nix Immanae opis).
 32 UNWRITTEN ROME

at first sight, is twice described as solleninis (items 1 and 3); the point is that the
triumphal procession is a religious ritual, and the soldiers’ freedom of speech is
a part of it.36
After the consul’s self-immolation in item 8, the pontifex does what a
pontifex has to do, namely teach the people.37 Marcus Livius explains to the
soldiers that the Gauls and Samnites are in the power of the chthonic deities,
summoned and dragged down after the consul himself:38

‘Furianim acfonnidinis plena omnia ad hastes esse.’

‘Everything on the enemy side is filled with the Furies’ terror.’

The Furies appear also in item 6, where the situation is reversed; there it is
Etruscan priests who destroy Roman morale as they come on ‘like Furies’
(incessn jitriali) with their snakes and torches,39 but the soldiers are told not to
fear their empty show.
You know the gods are on your side when the enemy is dumbfounded,
thunderstruck, afflicted with irrational terror. Here too the vocabulary is
consistent. The Gauls in item 5 and the Samnites in item 7 are transfixed by
pinorand admiratio; the Samnites in item 7 and the Gauls in item 8 are attoiiiti,
the latter having been sent almost out of their minds (ueltit alienata tneiite) after
Decius’ self-immolation.4" A similar paralysis afflicts the Romans in the early
stages of items 6 (tiehit lymphati et attoiiiti) and 8 (ueltit lyniphatiais pauor)—but
the repeated ‘as if (ueltit) shows that it wasn’t the real thing.41 These are not
stories in which the gods are hostile to Rome.
How did Camillus arrive at the Capitol in item 4? Livy refers to the gods
involvement, and then blandly goes on: ‘For by some chance the Dictator
arrived’. But the truce had only been to allow negotiations; it would hardly
extend to allowing Camillus and his relief force to pass through the besiegers
lines. One can only assume that the Gauls were attoiiiti, as they had been when
Fabius Dorsuo, trusting in his gods, walked through their lines to make his
sacrifice on the Quirinal, and walked back again.42 It may be worth noting that

36 10.28.16, 4.20.3; 3.29.5, 4.53.13.

37 1.20.6-7 (nt... edocerct), Plut. Numa 9.4 (8l6doKO)V), Cicero De oratore 2.52.
38 10.29.4; cf. 8.9.7 for the deuotio prayer (‘hostesque popnli Romani Qiiiritinm terrore forinidine
niorteque adjidalis'), 8.9.10 for the effect of the dciiotus on the enemy (qiiacuinqne equo innectus
est, ibi hand seats quam pestifero sideri icti pauebant).
39 7.17.3; for Furies with snakes and torches, cf. (e.g.) Virgil Acncid 4.472-3, 7.447 and 456-7,
Ovid Mctainoqdioscs 10.349-50.
40 7.10.12; 7.34.8, 7.36.10; 10.29.7, 10.29.2. Compare 2.10.5 on Horatius Codes at the
bridge: ipso niiraado andaa'ae obslapejedt hastes. The Horatius story also celebrates one partic­
ular hero (2.10.2, turns ttir), features vivid dialogue (2.10.3-4, 7-8) and a pious prayer in
direct speech (2.10.11), and implies an internal audience (2.10.10, clamor Romanoruin).
41 7.17.3,10.28.10.
42 5.49.1 (tiaiii forte qtiadam ... dictator intenienit); cf. 5.48.4 (truce), 5.48.5 (relief forces);
5.46.2-3.
 WHAT CAN LIVY TELL US? 33

Fabius’ walk and the arrival of Camillus are two of four stories in Livy’s fifth
book where siege conditions seem to be arbitrarily suspended. The others are
the Roman soldier’s conversation with the soothsayer at the siege of Veii, and
the Faliscan schoolmaster’s walk to the besiegers’ camp with the children of the
local aristocracy.43 The question is not, I think, how strictly sieges were actu­
ally conducted in Italy in the early fourth century bc, but what type of
narrative could present such stories with any appearance of credibility.
Another recurring feature of these episodes is the presence of spectators
within the story. This is a device to produce enargeia, the vividness that presents
a scene as if it were before our eyes.44 The classic example is Thucydides’
account of the battle in the harbour at Syracuse, with its focus on the emotions
of the two armies watching from the shore. As Plutarch observed, the historian
seems to make his hearers into spectators; listening to the text is like watching
the action yourself.4’ It was a famous passage, imitated in characteristically
long-winded fashion by Dionysius of Halicarnassus to describe the 'battle of
the triplets’ that decided the fate of Alba Longa.46 Livy does it much more
economically in his description of Titus Manlius’ duel in item 5:47

ubi constitcre inter duos acies tot circa morraliinn animis spe nietiique pendentibiis. . .
When they took up their stand between the two armies, the hearts of the many
men standing round them were on tenterhooks of hope and fear.

There are similar ‘internal spectators’ in the other episodes: the crowd
receiving Cincinnatus in item I;48 the women and children watching Camillus’
battle against rhe Gauls in item 4;49 the frightened Roman soldiers in item 6;50
the Samnite army, and then the main Roman army, in item 7.51 And in all
eight cases the soldiers’ shouting at the triumph implies a crowd of spectators,
necessarily present even when Livy does not refer to them.52

43 5.15.6-7:5.27.1-2.
44 Est enim haec pars orationis >piiie rein constituat paene ante oeulos (Cicero Partitioiies oratoriac 20);
rerum, quasi gerantur, sub aspectum paene subiectio (Cicero De oratore 3.202); cf. Quint. Inst.
6.2.32 (enargeia), 9.2.40 (sub oeulos subiectio). For enargeia in Livy, see Feldherr 1998.4-12.
45 Thucydides 7.71.1-4; Plutarch Moralia 347a (otov Oeatf|V noiflocn TOV «KpoaTf]V).
46 Dionysius of Halicarnassus 3.18-20, esp. 3.18.2 (Oearai), 3.19.1-3 (Thucydides pastiche);
see Walker 1993.
47 7.10.9 (trans. Betty Radice, Penguin Classics); cf. 7.10.6 (uisti ac specie aestimantibus).
48 3.26.11-12 (ea f'reqitentia ... plebis concursus ingens).
49 5.48.3 (in conspectu habeittes ... coniitges et liberos).
50 7.17.3 (insueta ... specie)-, cf. 7.10.6 (n. 47 above) for species.
51 7.34.8 (cum omnium in se uertisset oeulos), 7.36.7-8 (coniectis in earn omnium oculis).
52 As he does at 4.20.3 (ciuium ora), 4.53.12 (fauor circunistantis populi).
 34 UNWRITTEN ROME

IV
rutting these common
Putting tnese elements togethi
common elements together—celebration of a particular protago­
nist, dramatic speech and dialogue, participation by the gods, the presence of
an audience—we may now come back to our question (p. 29 above) about the
transmission of the iiicoudita cannitta.
When Livy' describes Manlius’ single combat as ‘more in the manner of a
spectacle than by the rules of war’, he is describing a scene within his story, not
commenting, as he does elsewhere, on the nature of the story itself.53 Never­
theless, it may be significant that the theatre is in his mind.54 Similarly, when in
item 6 he refers to the paraphernalia of the Etruscan priests, he twice uses a
word (apparatus) which frequently refers to theatrical productions,” and he
applies it to a spectacle often seen on the Roman stage, that of Furies with
blazing torches.56
Little is known about the fabulae praetextae, Roman plays on Roman
subjects?7 The classic titles were Naevius’ Clastidium (on M. Marcellus in 2
BC), Naevius’ Romulus and Lupus (perhaps alternative titles for the same play),
Ennius’ Ambrada and Sabinae, Pacuvius’ Paulus (probably on Aemilius 1 au us
in 168 bc), Accius’ Aettcadae or Deans, and Accius’ BrutusA8 Of these, Aettea ae
or Decius dealt with our item 8, while the action of Clastidium was, a sing e
combat resulting in ‘spoils of honour’ (spolia opima), as in our item 2, 1 lutarc s
account makes much of the triumphal procession, with due mention o t e
soldiers’ songs.59
Ancient theorists of drama likened the praetexta to tragedy, but it is c ear
that the known plots were not ‘tragic’ in the modern sense of the w' i d.
Decius’ self-sacrifice, and Brutus’ execution of his sons and heroic de a th, were
exemplary acts in plays that ended well for the Romans; the other plots do not
have even that much ‘tragedy’ in them. The reason for the theorists cate gon

53 7.10.6 (spectaculi magis more qnani lege belli). For narrative itself as theatrica , set X-
ostentalioneni scaenae gaudentis niiraculis aptiora); Dionysius of Halicarnassus
Plutarch Romulus 8.7. „ .. ....... jn a
54 Cf. Feldherr 1998.101: ‘he explicitly draws attention to the “theatricality o t u sc ,
manner that gives the question of spectacle a thematic importance within t e e_Plst 1
55 7.17.5. Stage: 27.6.19, 27.31.1, 31.49.4, 31.50.2, 32.7.14, 40.45.6, 45.32.8; cf. also .
Ad Atticiim 15.2.3, 15.12.1, Adfamiliares 7.1.2, De officiis 2.56, Tusculan Disputatior • •
Valerius Maximus 2.4.2; Ulpian in Digest 7.1.15.5; Tertullian De spectacidis 4. .
56 Cicero Pro Roscio Anierino 66-7 (qiiem ad niodinn in fabnlis saepenninero uidetis), u soiien
(lit in scaena uidetis), De legibus 1.40 (sient in fabnlis). omi S
57 The surviving evidence is presented with very thorough discussion in Manuwa
below, pp. 194-9, 206-7.
58 Fragments in Manuwald 2001.134, 141-3, 162, 172, 180-1, 196-8, 220-1.
59 Plutarch Marcellus 6.4—8.5 (soldiets’ songs at 8.2); Livy’s treatment in book 20 may ave
been similar, to judge by Eutropius 3.6.1-2.
60 Texts and discussion in Manuwald 2001.29-51.
 WHAT CAN LIVY TELL US? 35

sation was that the praetexta, Eke tragedy, dealt with the deeds of quasi-histor-
ical kings and heroes; the criterion was merely ‘the dignity of the characters’
(personartnn dignitas) .6i
Polybius, who belonged to the generation between Pacuvius and Accius,
used ‘tragedy’ as a disparaging metaphor in discussing the Romans’ use of the
fear of the gods as an instrument of social control. The context is Polybius’
observation of how ‘the constant renewal of the good report of brave men’
inspired young Romans to sacrifice even life for the safety of the Republic; the
example he gives is Horatius Codes at the bridge, in a version of the story
where Codes does not survive.62 Polybius’ choice of the word tragoidia here
was pointed out nearly forty years ago by Santo Mazzarino, who also noted
that even the wretchedly few surviving fragments of praetextae include some
conspicuous prayers to the gods.63
He might also have mentioned the comment of ‘Mercury’ to the audience
in the prologue to Plautus' Anipliitriio:M

. . .ut alios in tragoediis

nidi, Neptununi V'irtutem Victoriani
iMarteni Bellonain, commemorate quae bona
nobis fecissent...
...as I've seen other gods in tragedies—Neptune, Virtus, Victoria, Mars,
Bellona—remind you of what good things they’ve done for you.

The ‘tragedies’ where the gods of Rome reminded the Roman people of their
benefits to them were probably plays on Roman subjects; and no doubt the
good things they had done included the divine assistance implicit in most of
our Livian episodes.
The sudden appearance of Camillus in item 4—the peripateia, as Ogilvie
aptly calls it—is easier to imagine on the stage than in real siege conditions; and
the same may be said of the soothsayer at Veii, the Faliscan schoolmaster, and

61 Diomedes in Cranunatici Latini I.489K (Manuwald 21)01.31): in quilms imperatorum iiegotia

agcbantur et publica et reges Romani uel ibices inducimtur, personaruin dignitate et sublimitate
tragoediis similes. Euanthius De fitbula 4.1 (Manuwald 2001.37): praetextatas [sic] a dignitate
personarum tragicanun ...
62 Polybius 6.54.2-6 (inspiration), 6.55 (Codes), 6.56.6-7 (dElOlbaipovta), 6.56.8
(EKTETpuyti)dT]Tai), 6.56.11 (xf) TOiauTfl Tpayoidia). Translation of 6.54.2 by lan Scott-
Kilvert (Penguin Classics).
63 Pacuvius Paulus fr. 1 Manuwald (‘pater supreme nostrae progenii patris...’); Accius Aeneadae or
Deans fr. 4 Manuwald fte sancte uenerans precibus, inuicte, iuuoco, | portenta ut populo patriae
uemmcent bene'). See Mazzarino 1966.63-4, in a chapter entitled ‘Society and historical
thought in the age of the praetexta’.
64 Plautus Amphitnio 41-5; Wiseman 1998.19-20. Cf. also Phaedrus 2.7.23-4 (di sunt locuti more
translaticio).
 36 UNWRITTEN ROME

Fabius Dorsuo.65 But any idea of stage performances influencing Livy’s narra­
tive method must be abandoned if Andrew Feldherr is to be believed:

[F]ar from claiming the drama as a model for the way he presents Roman
history', Livy consistently depicts the theatre as antithetical to his narrative in its
aims and effects ... Livy exploits Roman cultural constructions of the drama as
a socially pernicious and fundamentally alien form of spectacle to highlight by
contrast the salutary potential of his own history and its direct link to the
centers of Roman power.
According to this view, the theatre was ‘an institution isolated from the normal
conduct of civic life’, foreign and anomalous, quite inconsistent with Livy’s
patriotic purpose.66
It is a paradoxical view, as Feldherr admits: dramatic productions, as he
points out, took place ‘within the context of official civic festivals’, and ‘consti­
tuted part of the cult practices of the state religion’.67 Much depends on two
assumptions, that the praetexta was not an important dramatic form, and that
the action of plays seen in Rome, ‘in the vast majority of cases, took place
outside Rome, usually in a markedly Greek milieu’.68 That seems to infer
much too much from the surviving texts of Plautus and Terence. It is true that
Livy disapproved of the luxury and violence of the Indi scaenici in his own time,
as did Valerius Maximus a generation later, but he would surely have agreed
with Valerius that the theatre was one of‘the ancient and memorable institu­
tions of our city’.69 For the Romans’ attitude to their public games—‘especially
holy, solemn and sacred in their usages and institutions’—the language of
Cicero’s speeches is surely decisive.70
So there need be no a priori objection to the hypothesis that some or all of
our Livian passages originated as performance scenarios. And if that hypothesis
is available, it may account for the incondita cariniiia.

See above, nn. 42-3. Ogilvie 1965.737; cf. Dionysius of Halicarnassus 3.18.1 for theatrical
penpeteia, in Roman history.
66 Feldherr 1998.165-87; quotations from 165-6 and 170, cf. 181 (‘anomalous’). The argu­
ment about alienness (178-87) relies heavily on 7.2.4, cetemni pania qnoqne, nt ferine principal
omniti. etea ipsa peregrine resfuit. But when Livy says 'then it was new, small and alien’, he
rr S ’"T ',S "m‘‘ “ 'S familiar’ huBc- and Part °f Roman experience’.
67 Feldherr 1998.169, 174,
ro Tarin 1998.176, with 172 on the praetexta as ‘a not entirely successful experiment’.
' (hancmx opulentis regnis tolerabileni insaniam): Valerius Maximus 2.4.1 (violence), 2.4.6
. UXUr?'’ ,IS Ptt’face to book 2 (iniciam stihnn ... nostrae nrbis ... priscis et inetnorabilibns
tnstiMis). Rioting fans: Dio Cassius 54.17.4-5 (stasis, before 18 Be), Tacitus Annals 1.54.2
(discord,a, AD 14), 1.77.1-2 (seditio, AD 15).
./er0 l‘‘,nlsP‘aiin response 24 (on the Indi Megalenses): more institntisqne maxhne casti,
and amidssb^'' '''M' <resPectively

I
 WHAT CAN LIVY TELL US? 37

Horace, writing about 12 BC, complains that the theatre of his time is
obsessed with mere spectacle, and the example he gives is of a play on the
capture of Corinth in 146 BC, featuring an interminable triumphal procession
with wagon-loads of captured booty.71 In five of our eight Livy passages—and
in Plutarch’s account of Marcellus’ triumph, which may owe something indi­
rectly to Naevius’ play—specific mention is made of the amount of booty
carried at the triumph.'2 Perhaps the procession was a traditional joyful conclu­
sion to historical plays about Roman victories, rather like the closing revel
(komos) in Aristophanes;73 at Cincinnatus’ triumph, we are told that the soldiers
acted like revellers (coniisantes).14
That does at least offer a solution to the problem of extemporised verses
surviving into the much later literary tradition of historiography. Livy and his
predecessors may have heard those incondita cannina, and seen the exemplary
episodes they celebrated, in patriotic performances at the theatre games (ludi
scaenici). If so, it seems that Livy has given us some good evidence after all, not
so much in the stories he tells as in what can be inferred from his manner of
telling them.

V
It mav even be possible to support this hypothesis from a non-Livian source. In
the section ’On miracles’ of his collection of exemplary tales from history,
Valerius Maximus tells the story of C. Fabricius’ defence of Thurii against the
Lucanians and Bruttians in 282 BC. When the Romans were nervous about
attacking the besiegers, a young man of huge physique exhorted them, then
seized a scaling ladder and made his way through the enemy lines to their
camp, he climbed the rampart and shouted back to the Romans that the way
to victory was achieved. The Romans attacked the camp and thus drew the
Lucanians and Bruttians away from the siege; in the desperate battle that
followed, the heroic young man defeated the enemy single-handed; twenty
thousand were killed, five thousand taken prisoner, including the commander,
and twenty-three standards captured.
The following day Fabricius held an assembly, and announced that the
young man had won the ‘siege crown’ decoration {corona Italians)-, but he was
nowhere to be found. The Romans recognised that Mars himself had been

71 Horace Epistles 2.1.187-92; Brink 1982.432 on captiua Coriuthos (cf. Cicero De officiis 2.28,
Philippics 8.18; Livy 5.30.2, 38.43.10).
72 3.28.4, 4.19.6, 4.53.10, 7.17.5, 10.30.10; Plutarch Marcellus 8.1, cf. Eutropius 3.6.2 (n. 59
above).
73 Aristophanes Achamiaus 1190-1233 (triumphal); Peace 1316—57, Birds 1720-65 (hymenaeal).
Was there a wedding HUllOC at the end of Ennius’ Sabiuae?
74 3.29.5, coiiiisaiitium modo.
 38 UNWRITTEN ROME

among them, and Fabricius proclaimed a festival of thanksgiving (supplicate) to

the god:'’

<t lanreatis militibits iinigiin emu miimonmi laetitia oblati auxilii testimonium ei est
redditmn.
Wearing laurel wreaths and in great happiness of mind, the soldiers bore
witness to Mars of the help he had provided.

Although this concluding scene is not a triumph, the structure of the narrative
is very Eke that of the Livian passages.
It was probably not taken from Livy, who distances himself from miracle
stories and conspicuously omits the very similar epiphany of the Dioscuri at the
battle ot Lake Regillus.76 Other historians, however, were more hospitable to
such material,77 and 1 think we are entitled to take Valerius’ unknown source
as confinnation that our eight episodes reflect not just one historian’s way of
working but a real aspect of Roman fife—a way of presenting exemplary
actions to the public—that had a more general influence on Roman historiog­
raphy.
I think we can also infer that these dramatised episodes from ancient history
were put on regularly at the theatrical festivals. That seems to be implied by
Horace’s experience, since the fall of Corinth was hardly a topical subject in his
day. It may also be implied by two apparently puzzling comments by Cicero—
that the stage was a source of infonnation, and that ordinary working people
were fascinated by history.711 That was surely one of the main purposes of the
Roman games, to educate and inspire the citizen body by ‘the constant
renewal [as Polybius put it] of the good report of brave men’.79
What Livy can tell us is not (of course) how Roman heroism was celebrated
in the early Republic, but how the heroism of the early Republic was cele­
brated in his own time. It is easy to imagine how the experience of seeing such
exemplar)' deeds of bravery and piety recycled in stage performances year after
year may have conditioned the way historians constructed their narratives. Not
only that, but Polybius’ evidence takes us back to the middle of the second
century BC; even then, the Roman custom of representing the great deeds of
the past was a traditional one. We cannot be sure that it went right back to the
early Republic, but it is more likely than not that it predates Roman literature.
And if that is the case, then perhaps even Livy’s sophisticated literary history'
may have preserved some flavour of the culture of pre-literary Rome.

75 Valerius Maximus 1.8.6; the reference to the jjrarfns uietoriae was no doubt an .illusion to
Gradiuus, one of the names of Mars.
76 2.19-20; cf. Levene 1993.18-20, 152-3.
77 See below, pp. 247-62. , .
78 Cicero De legibus 1.47, Dejmibus 5.51-2 (on the latter passage cf. Wiseman 1987a._o--6).
79 Polybius 6.54.2; the Roman phrase would be res bene gestae.

1
 CHAPTER THREE

Fauns, Prophets, and Ennius’ Annales

i
The invention of Latin epic is attributed to Ennius by the grammarian
Diomedes in the fourth century AD:1

epos dicitur Graece carmine hexametro diuinarum rerum et heroicantni luimanariimqiie

cotnprehensio... Latine panto comimmius carmen anditnr. epos Latinnm primus digne
scripsit is qui res Romanonim decent et octo complextts est libris, qui et annales <in>scri-
bitnntr, quod singulonim fere annornm actus contineant, sicut publici annales qnos
pontifices scribaeque conficiunt, ttel Romais, quod Romanorum resgestas declarant.
Epes in Greek means the composition of the deeds of gods, heroes and mortals
in a hexameter poem. . . In Latin it is more often called carmen. The first Latin
epos worthy of the name was written by the poet who encompassed the deeds
of the Romans in eighteen books, which are entitled Annales because they
contain ihe events of practically each year, like the public annals composed by
the priests and scribes, or ‘the Romaid" because they make manifest the achieve­
ments of the Romans.

Ennius himself expressed his primacy in a famous passage, one of the most
frequently cited of all the surviving fragments of the Annals:2

scripserf alii rent

uorsilms qnos olim Faunei uatesque canebant.
Others have written of the matter [the first Punic War] in the verses which of
old the Fauns and prophets chanted.

The reference is to Gnaeus Naevius? Although Ennius thinks of him as a

fellow writer, it seems that Naevius doesn’t count as a proper ‘literary author’
because he used a metre appropriate to the pre-literary world.4

i Diomedes Ars grammatica book 3, in Crammatici Latini 1.483—4 Keil; I follow Skutsch
1985.46 in reading Romais for the MSS’ rotnanis.
2 Ennius Annales 206-7 Sk.
3 Stated explicitly by Cicero (Brutus 75-6), and obvious anyway.
4 Cicero Bnitus 71: ‘nee dicti studiosus quisquam erat ante hunt' ait [Ennius] ipse de se nec mentimr
in gloriando.
 40 UNWRITTEN ROME

That at least is my reading of the lines. But it is important to remember

how little we know about the culture and society of third-century Rome—
despite the confident assumptions of some recent scholarship. For instance,
Enrica Sciarrino argues that the audience for early Roman epic was not the
same as for drama, and that Livius Andronicus’ Odyssey translation ‘opened the
way to the encroachment of poets ... on more exclusive sites of social inter­
action .■ Perhaps so: but Homeric bards like Demodocus performed for the
games in the agora as well as for the ‘elite’ at the palace;6 and Oliver Taplin has
very plausibly argued that only a festival context can explain the length and
complexity of the Homeric epics themselves.7 Can we really be sure that
conditions were radically different in Rome? Why not imagine the Indi
Romani—or the Indi plebeii, for that matter—as the primary performance
context of oral narrative poetry, and then of Naevius’ and Ennius’ epics?
As for the ‘elite’ itself, the use of sociological models may be misleading if it
implies that Roman society was necessarily as hierarchical in the third century
BC as it was in the first. There was no differential seating at the Indi until 194
BC, when privileged seats were provided for the senators, and the sources for
that innovation (contradictory as they are) show clearly how controversial and
unpopular it was.8 Equally clearly, Pliny’s account of the equestrian insignia
implies a proliferation of hierarchical indicators from the Gracchan period
onwards; but not before.’
I am much more in sympathy with Sander Goldberg’s sense of how little is
really known about the world of Naevius and Ennius. But here too I have a
reservation. Is it really true, as he asserts,10 that ‘whatever archaic tradition
preceded [Ennius’ poem] remains beyond recovery’? If it is, then there is no
point trying to understand what Ennius meant when he called Naevius metre
the verses which of old the Fauns and prophets chanted’. I hope to show that
it may be possible, after all, to say something useful about the ‘pre-literary
world of third-century Rome.
Glimpses into that world, from people close enough to it to be well
infonned, are hard to interpret but disproportionately precious. 1 he most
famous of them is the elder Cato’s evocation of the custom, long obsolete in
his own time, of guests at a banquet rising in turn to sing the praises of the men

5 Sciarrino 2006.458.
6 1'
Homer Odyssey 8.12,16,109, 254-65 (irijora); 40-45, 62-70, 104-7 (palace).
T«wi:~ OAnn no
7 Taplin 2000.23-32.
8 Asconius 69-70C (citing Cicero’s Pro Come/io and Valerius Antias fr. 37P), Cicero De luinu-
picitin response 24, Livy 34.44.5 and 54.4-8.
9 Pliny Nat. Hist. 33.29-36.
10 Goldberg 2006.446.
 FAUNS, PROPHETS, AND ENNIUS’ ANNALES 41

of old.11 Cato also knew of professional praise singers at banquets, and claimed
that the inaiores despised them.12 Then there is Fabius Pictor’s account of the
procession at the Indi Romani, with its dancing choruses of Sileni in hairy tunics
and satyrs in goatskin loincloths, mocking the serious participants in a manner
that was literally satirical.13 Perhaps Ennius’ line can give us a similar insight
into this unfamiliar world?

II
Vorsibus quos olim Faitnei uatesque canebant is quoted six times in our extant texts.
Cicero (twice) uses it as evidence for progress in the literary arts, and Quin­
tilian to prove that poetry existed before the laws of metre.1'1 But it is the other
three citations that are more interesting for our purposes.
First, Varro, in his discussion of‘words which have been put down by the
poets’.15 He quotes the line, and then offers his exposition:16

Fauni dei Latiiiormn, ita ut Faunus et Fauna sit; hos uersibus quos uocant Saturnios in
siluestribus locis tradition est solitos Jari <jiitura, a> quo fando Faunos dittos, antiqui
poetas nates appellabant a uersibus tiiendis, ut <de> poematis cum scribam ostendani.
Fauni are gods of the Latins, in this sense, that there are both Faunus and Fauna.
It has been handed down that they are accustomed to speak the future in
wooded places, using the verses known as ‘Saturnian’, and called Fauni from

11 Cie, ■■.> Disp. Tusc. 4.3: graidssinuis auctor in originibus dixit Cato, inorein apud inaiores hunc
epuiani’t. Juisse, nt deinceps qui accuinberent canerent ad tibiam claroruin iiiroruni laudes atque uirtutes.
Cicc ro bnitus 75: atque utinain exstarent ilia cannina, quae inultis saeclis ante suain actatem in epulis
esse ^r"it.:t.i a singulis commas de claroruin uironun laudibus in originibus scriptuni reliquit Cato. Cf.
also Cicero Disp. Tusc. 1.3, Horace Odes 4.15.29-32, Valerius Maximus 2.1.10, Quintilian
1 16.20. Varro has a different version, in which praises are sung by pueri modesti (De uita
populi Hemani fr. 84 Riposati = Nonius 107-8L). For the background, see Rosier 1990 and
Zorzetti 1990; but also Horsfall 1994.70-73, a reminder that the Cato passage may be no
more than a caique of Creek scholarship on the symposia/! (e.g. Dicaearchus fr. 88), and
Goldberg 2006.431.
12 Aulus Gellius 11.2.5: praeterea ex eodem libro Catonis have etiam sparsim et intercise coininemi-
niinus: ... ‘poeticae artis honos non erat. si quis in ea re studebat aut sese ad conuiuia adplicabat
crassator uocabatiir'. For the background, see Peruzzi 1998.157-64, and Goldberg 2006.431-
4, with n. 9.
13 Dionysius of Halicarnassus 7.72.10 (FGrH 809 F13, p. 867), cf. 7.71.1 on Q. Fabius:
jtakuioTOTo; yup avi)p rwv ret 'PtupaiKd ouvia'^apEvaw, rccxl itiorrv ovk e§ d>v
quouoE (idvov, tikka Kai e§ wv auto? eyva) jrapExopEvog. For the background, see
Szilagyi 1981.
14 Cicero Brutus 71 (cf. 75), Orator 171; Quintilian 9.4.115.
15 Varro De lingua Gitina 7.5: dicani in hoc libro de uerbis quae a poetis sunt posita.
16 Ibid. 7.36 (repeated in Servius audits on Ceorgics 1,11, who alone has futnra after Jari); I
follow R.G. Kent’s Loeb text. See Aronen 1999 on ’Saturnian verses’, and Pasco-Pranger
2002.306-10 for an adventurous association of this passage with the story of Numa, Picus
and Faunus at Ovid Fasti 3.285-328.
 42 UNWRITTEN ROME

that ‘speaking’. The ancients called poets nates from ‘weaving verses’, as I shall
show when I write about poems.

Fauna was one of the names of the Bona Dea in the records of the pontifices;
another was Fatua, also derived from fari, to speak.17 She was thought of as a
prophetess, just as Faunus was thought of as a prophet.18 Some said she was a
Dryad, just as some said Faunus was Pan.19 Naturally, as divinities of the wild,
they ‘spoke’ in woods and groves.20
Prophecy is also the context of Cicero’s quotation of the Ennius line in De
diuinatione. Quintus is discussing the phenomenon of directly inspired divina­
tion, when a kind offuror in the soul enables certain individuals to prophesy
the future:21

codem enim modo midta a uaticinantibus saepe praedicta sunt, neqtte solum tterbis sed
etiam ‘uersibiis quos olim Failin' uatesque canebant’. similiter Marcius et Piiblicius nates
cecinisse dicuntur; quo de genere Apollinis operta prolata sunt.
For in that way many things have been predicted by those who chant prophe­
cies, not only in prose but also ‘in the verses which of old the Fauns and
prophets chanted’. The prophets Marcius and Publicius are said to have chanted
in the same way, and the secrets of Apollo were brought forth in that style too.

What matters here is the equivalence of uerba and uerstts, speaking in prose and
chanting in verse. Though Faunus is ‘the speaker’, his prophecies can equally
be described as chanted cannina.22 So Ennius’ Faiini ... canebant is in no way
paradoxical.
But who, or what, were the Fauni? It’s clear that Cicero did not share
Varro s interpretation of the plural as referring to two individual deities.

Macrobius Saturnalia 1.12.21-22: hanc [sc. Maiam] eandem Bonam Faunamqne, Opcin et
Fatuam pontifmm libris indigitari: Bonam quod omnium nobis ad uictuin bonoruin causa cst, Faunain
"Sl" """nantiiiniJanet, Opem quod ipsius auxilio uita constat, Fatuam a J'ando. . .
Justin 43.1.8. Fauna uxorfnit nomine Fatua, quae adsidue diuino spirit" inpleta uciuti per /iirorem
futurapraemonebat.Plutarch Mora/ia 268D (Quaestiones Romanae 20, on the Bona 1 >ca): <bg ol
put)oAoyowTeg loropovot, <t>auvou pev fjv yuvij tou pavTEto;...
1J Plutarch Caesar 9.3 (Dryad); Horace Odes 1.17.1-2, Ovid Fasti 2.267-80, 2.424, etc (Pan
Lykaios).
20 For instance in the first year of the Republic: Dionysius of Halicarnassus 5.16.2-3,
exp .lining that the Romans attribute (ptuval 6ai)l6vi0l to Faunus; cf. Servius on Aeneid
7.81 for Faunus named and Toil tpwvijg, quod uoce non signis ostendit futura. Calpumius
icu us (-r <>i;ncj 1.8-32) offers a literary version of the idea, with Faunus’ prophecy
insenbed on a beech tree in the grove.
21 Cicero De diuinatione 1.114-15 (cf. 1.4 on/nror).
j. /CS'U,S 0,1 Simians, quibus Faunus fata cecinisse hominibus uidetur, Calpumius
icu us -r. ._ an 34 (canere), 1.32, 35, 92 (canneii); cf. Ovid Fasti 3.323 on Faunus and
P ' II ^ra m "■> 'C"i" equivalence offari and canere in prophecy' is shown by
mtore "parcte ~ °n ^ateS' Fraefa"tes V'ondamfelicia Pelei cannina diuino cecinerunt
 FAUNS, PROPHETS, AND ENNIUS’ ANNALES 43

Faunus and Fauna. A little earlier in the De diuinatione he makes Quintus point
out that Fauni have often been heard in battles, clearly implying a particular
category of divinity.23 These Fauni were ‘half-gods’, dwellers in the country­
side, the companions of nymphs and satyrs;24 but they were still prophetic,
both in their own right and as the inspirers of prophecy in mortals.25
It is also clear that Cicero did not share Varro’s interpretation of nates as
meaning poets. Nothing more is known of Publicius, dismissed as nescio qtiis by
the sceptical Marcus in book 2 of the dialogue;26 but Gnaeus Marcius was a
well-known prophet, famous in Ennius’ time for having foretold the disaster at
Cannae.27 Such prophets were active throughout the history of the republic,
and we know from Livy and Horace that their oracular predictions were
collected in volume form.28
The final quotation of the Ennius line is in the anonymous Origo gentis
Ronianae, a treatise probably of the fourth century Al) but based on much
earlier material. (None of the thirty or so authorities cited is demonstrably
post-Augustan.) The author reports the successive reigns in Italy of Janus,
Satumus and Picus, and then goes on:29

post Picum regnauit in Italia Faunus, quern afando dictum uolunt, quod is soletfutura
praecincrc uersibus quos Saturnios dicimus; quod genus inetri in uaticatione Saturnine
primum prodittim est. eius rei Ennius testis est, cum ait ‘uersibus quos olim Fauni
uatesqiie canebant’.
After Picus Faunus reigned in Italy. They derive his name from ‘speaking’,
because he is accustomed to prophesy the future in the verses we call Saturnian,
a type of metre first used in the prophecy of Saturnia [or in a prophecy at

23 Cicero De diuinatione 1.101: saepe etiain et in proeliis Fauni auditi... esse dicuntur. So too in De
natura dcomm 2.6 (saepe Faunorum uoces auditae), though the sceptic’s answer at 3.15 uses the
singular.
24 Semidei: Ovid Metamorphoses 1.192-3. See for instance Lucretius 4.580-1; Virgil Eclogues
6.27, Georgies 1.10-11, Aeneid 8.314; Horace Epistles 1.19.4, Ars poetica 244; Ovid Metamor­
phoses 6.392-3.
25 Nemesianus Eclogues 2.73 (Fauni nates); Fronto De eloquentia 2.12, Teubner p. 141 (Fauni
iiaticinantium iiicitatores). Ct". Martianus Capella 2.167 on the longaeiiormn chori qui habitant
siluas, nemora, linos, laais, fontes ac fhuiios, appellanturque Panes, Fauni, Fones, Satyri, Siluani,
Nymphae, Fatui Fatuaeque uel Fantuae uel etiain Fanae, a quibus Jana dicta, quod soleant diuinare.
26 Cicero De diuinatione 2.113. It may be relevant that C. Marcius and T. ‘Publius' (Publicius?)
were among the first plebeian augurs, elected in 300 BC: Livy 10.9.2, with Wiseman
1998.103-4.
27 Livy 25.12.1-8 (prophecy discovered in 212 BC); cf. Cicero De diuinatione 1.89 (Marcias
quondamJ'ratres, nobile loco ortos), Festus 162L (in carmine Cn. Marci uatis).
28 Livy 25.1.12, 25.12.3 (libri uaticini); Horace Epistles 2.1.26 (annosa uolumina iiatimi). Prophets
active: nn. 47-8 below, discussion in Wiseman 1994.49-67.
29 Origo gentis Ronianae 4.4-5 (omitting a sentence condemned as a gloss since Gruner’s edition
in 1757). I think this passage disproves e silentio the attempt by Aronen (1999.63—9) to make
Satumus, as well as Faunus, an oracular deity.
44 UNWRITTEN ROME

Satumia], Ennius is our witness for that, when he says ‘in the verses which of
old the Fauns and prophets chanted’.

Satumia was the stronghold on the Capitol founded by Satumus,30 and it is

possible that there existed a supposed prophecy of its foundation. In this Euhe-
nierised telling of the myth, exploited by Virgil in Aeneid 7, Faunus is part of a
prehistoric dynasty of Latin monarchs; it was he who received Evander and his
Arcadian colonists.31 But he is a prophet in this guise too, providing dream­
oracles in his sacred grove to his son and successor Latinus.32

Ill
Prophecy was primarily an oral mode; the words used of it were ‘song’ and
‘singing’, airmen and canere.33 The same words defined the Camenae, those
‘cultural signifies of pre-poetic song’ who inspired the oral poet’s perform­
ance.34 But listening to a prophet was probably not like listening to a poet.
What did a prophet sound like? Ovid sets the scene, where he describes the
inspiration of the prophetess whose name was formed from carmen:3

panta tnora est, caelum nates ac nnmilia sninit

Jilqite stti tola pectore plena dei;
nix illam subito posses cognoscere, tanto
sanctior et tanto quant ntodo tnaior erat.
There is a moment’s pause. Then the prophetess assumes the powers of
heaven, and is filled with her god to the depths of her heart. Suddenly you
could hardly recognise her, so much holier she was, and so much taller than
before.

That was for a message of good news (‘laeta ainant"), but often the prophet s
demeanour was fierce and threatening;,36 a far cry from the song of the poet.

"
30 Satumia: - ■
j Origogentis ■ Romanae
- 3.1 and 3.7, citing Virgil Aeneid 8.357-8; see ajso Qvid Fasti
6.31, Pliny Nat. Hist. 3.68, Festus 430L. Placed in 1300-1200 bc (‘Bronzo recente, fuse
prima’) by Carandini 1997.120.
31 Virgil Aeneid 7.47-9; Dionysius of Halicarnassus 1.31.2, Justin 43.1.6, Origo gentis Romanae
5.1-3.
Virgil Aeneid 7.81-103, esp. 82 (fatidici genitoris).
33 Canere: e.g. Livy, 1.45.5, 1.55.6,, 5.15.4;J; see n. 22 above. Cannina: e.g. Livy 25.12.2-8,
.... -----
Festus 162L (Cn. Marcius). Primarily oral: Cicero De diuinatione 2.149 (sine tn uatein ...
andieris). For 'song and memory’ in general see Horsfall 2003.11—17, 36—17.
34 Festus (Paulus) 38L: Cantenac musae a canm'nibus sunt dictae, net quad canunt antiquonim
laudes...; cf. Varro De lingua Latina 7.26—7 (Casmenae>Cannenae>Cantenae), Servius on Ed.
3.59 (a cantn), Macrobius Commentary on Somnium Sdpionis 2.3.4 (a eanendd). "Cultural signi-
fier’: Sciarrino 2006.454.
35 Ovid Fasti 6.537-40 (cf. 1.503-6, also on Carmentis).
36 Lucretius 1.102 3 (iiatinii terriloquis... dictis), 1.109 (minis ... uatuni); Ovid Fasti 1.504 (torua).
 FAUNS, PROPHETS, AND ENNIUS’ ANNALES 45

(My use of the word ‘chant’ to translate Ennius’ canebatit is a crude attempt to
register the difference.)
Different sorts of cannina sounded different because they had different things
to say. In the pre-literary world (to put it in a modem idiom), the manner and
the message were inseparable. But once they were written down and collected
in books, cannina from widely different sources lost their distinctiveness.
Such collections of cannina antiqua are attested by Varro, Festus and Macro-
bius, who use them to cite (respectively) a description of a shepherds’ festival,37
a quasi-Homeric narrative of dawn,38 and a piece of father-to-son advice on
agriculture.39 A religious precept cited by Nigidius Figulus may be a fourth
example.41’ The variety and unpredictability of these haphazardly surviving
items are a salutary warning of how little we understand pre-literary culture.
For instance, we have no idea where, when or why the cannina of the Salii
were written down.41 Nor do we know how Dionysius of Halicarnassus found
his information about ‘songs’ in honour of Faunus, the twin founders and
Marcius ‘Coriolanus’,42 or whether the narrative carmen Priami and carmen Nelei
existed independently or could be cited only from a general collection.43 But
the fact that no authors are named may allow one inference, at least—that
these cannina were first composed in an oral culture and only later preserved in
writing.44 There is a clear contrast with Naevius’ poem: though often cited as
the carmen belli Piniici, it always has the author’s name attached.43

37 Varro De uita populi Rom,mi fr. 23 Riposati (Nonius 3IL), Horsfall 2003.46: etiam pcllis
bi.bn/us oleo perf'usas percurrebant ibique cemuabant. a quo tile uersus netus cst in carmiiiibus: 'ibi
piste,c> ludos Jaciunt coriis Consualia'.
38 Festus 214L: obstinet dicebant amiqui, quod nunc ostendit; ut in ueteribus carmiiiibus: ‘sed iam sc
nu.'a rri/tns .4un>ni obstinet suum pattern’.
39 Macrobius Saturnalia 5.20.18 (on Ceoigics 1.101), Horsfall 2003.45: in libro ciiim uetustissi-
morurrt canninum, qui ante omnia quae a Latinis scripta sunt compositns ferebatur, inuenitiir hoc
msticum uetus canticiim: ‘hiberno puluere, nemo into, grandia farra, camille, metes’. Also in Festus
(Paulus) 82L: in antique carmine, rum pater filio de agricultura praeciperet.
40 Aulus Gellius 4.9.1: Nigidius Figulus ... in undeciino coninientarionuii grainniaticoniin uersum ex
antiqiio cannine refert memoria hercle dignum: ‘religentem esse oportet, religiosus nefuas’.
41 Plural in Varro De lingua Latina 9.61, Festus (Paulus) 3L, Festus 124L, Macrobius Saturnalia
1.9.14 (Saliorum quoque antiqiiissiinis canninibus ... canitur), 1.15.14 (ut Salii in carmiiiibus
canuiit). Singular (carmen Salionmi or carmen Saliare) in Varro De lingua Gitina 5.110, 7.26-7,
Festus (or Paulus) 109L, 222L, 224L, 230L, 231L, Terentius Scaurus in Crammatici Latini
7.28 Keil.
42 Dionysius of Halicarnassus 1.31.2 on Faunus (Kai autdv (1)5 T0)V eTtr/wpiaiv Tiva
Ptnpaim daipdvwv Bvoiaig Kai tiidai^ yepaipovoiv), 1.79.10 on Romulus and Remus
(tbc tv rote iratpioic vpvotg bird 'PaipaiiDV eti Kai vuv aberat), 8.62.3 on Coriolanus
(aStiai Kai v|xveiTai npog jtavruiv ate Evoe[3f|g Kai diKatog avf|p).
43 Varro De lingua Latina 7.28 (in cannine Priami); Festus 418L, 482L, Charisius in Crammatici
Latini 1.84 Keil (in Nelei cannine).
44 As an analogy, cf. Varro De lingua Latina 6.18 on the togata praetexta data feist Apollinarilms
ludis, identified by the occasion, not the playwright.
45 Festus 306L, Nonius 290L, Priscian in Crammatici Latini 2.198, 234, 242, 351 Keil; cf. Aulus
Gellius 17.21.45.

CLASSICA
LIBRARY
CAMBR1DC
 46 UNWRITTEN ROME

These ‘old songs’ may have been hymns or narratives sung to the lyre, or
precepts to be given in a speaking voice. What the Fauns and prophets chanted
(in a metre that Ennius thought an epic poet shouldn’t use) was something
quite different, a phenomenon that persisted long into the fully literate society
of the late Republic and early Empire. The next section explores its later mani­
festations, which turn out to be unexpectedly relevant to Ennius’ own poem.

IV
Among the omens and portents associated with the outbreak of civil war in 49
BC, Dio reports that ‘certain oracles were chanted, purporting to be those of
the Sibyl, and some people became inspired and prophesied many things’.46
The prophecies Dio describes were no doubt like those delivered in earlier
crises: in 87 BC, for instance, the Senate took note of the ‘inspired predictions’
(fnribundae praedictioncs) of the prophet Cornelius Culleolus,47 and we happen
to know of similar warnings in 78 and 63.48
Attributing a particular oracle to the Sibyl was probably easier after 83 BC,
when the original cannina Sibyllac were destroyed with the temple of Capito-
line Jupiter. In 76 an embassy was sent to Erythrae to replace them; it came
back with about 1000 verses collected from individuals, and others were found
from other sources. But it must have been easy for professional prophets to
claim knowledge of items the embassy had missed.49 The official collection had
only limited authority, as Augustus showed when as pontifex niaxtiniis he
purged it of unsuitable items.50
What exactly the prophets foretold in 49 BC is not recorded, but it is
unlikely that it was encouraging. For the Romans had defied the Sibyl s
warning by restoring king Ptolemy six years earlier, and Dio records that the
citizens were afraid of the anger of the gods.51
In ad 19, at a time when the fear of civil war had returned,52 a quasi-

46 Dio Cassius 41.14.4: Xdyict nva dig Kat ifjg SiflvXXrig ovra pdero, raio/ol re Ttveg
yiyvopEVOl OU/va £0El(Jtt,OV. The Loeb translation takes fjbETO as from Elbto/oibct ( some
oracles ... were made known’), but Dio’s usage in similar passages elsewhere (mi. 53-5
below) suggests that it was from qbd), perhaps reproducing canebantur in a Latin source.
47 Cicero De diuinatione 1.4; cf. Plutarch Marius 42.4.
48 Sallust Histories 1.67.3M (uatum cannina), Cicero De consulatu suo 10.28—9 Courtney (mites
orada furenti pcctore fundebant). [Cf. also Appian Civil IRirs 1.121.563 on the 0f6Xl].TTOl
TtVEg who brought about the reconciliation of Pompey and Crassus in 71 BC.]
49 Dionysius of Halicarnassus 4.62.5-6 (from Vano), Lactantius Diuinae institutiones 1.6.11 and
14 (citing Vano and Fenestella).
50 Suetonius Dilins Augustus 31.1: solos rctiuuit Sibylliuos [libros], Itos quoque dilectu habito.
51 Dio Cassius 39.15.1-16.2 (Sibyl’s warning); 39.55.3, 56.4, 59.3, 60.4 (defiance); 39.61.1-4
(citizens’ fear, 54 Be).
52 Tacitus Annals 1.4.2, 1.16.1, 1.31.1, 1.35.3-4 (ad 14), 2.39.1 (ad 16), 2.59.2-3, 2.78.1 (ad
19); Suetonius Tiberius 25.1-2 (ad 14); SC de Pisone patre 45-9 (ad 19).
 FAUNS, PROPHETS, AND ENNIUS’ ANNALES 47

Sibylline oracle was chanted again, and this time Dio quotes it verbatim:
‘When thrice three hundred years have come and gone,
Then civil conflict shall destroy the Romans..

Tiberius tried to calm the People’s fears by insisting that the oracle was false,
and he repeated Augustus’ purging of all books of prophecy.53 But that could
not stop oral circulation, and the same oracle was remembered in ad 64, when
the fire of Rome was taken as its fulfilment.54 Nero assured the populace that
these lines could not be found in any collection of oracles, but in vain: they
simply chanted another, more specific, ‘Sibylline’ prophecy instead.55
What is interesting about these passages is not only the continued impor­
tance of oral prophecy for the Roman People as late as the first century AD, but
also the content of the Sibyl’s alarming forecast. Why should the citizens have
been afraid of the nine hundredth year of Rome? Surely they knew that AD 19
was ab urbe condita 771? All they had to do was go to Augustus’ arch in the
Forum and look at the list of consuls and triumphs; the AUC date was given
for every triumph and for every tenth set of consuls.56 Why did the Roman
People not believe so authoritative a source of information?
I think the point is this, that all human knowledge is fallible. Only the gods
know the truth, and only the nates—prophet or poet—is divinely inspired to
reveal it. That’s not how we see it, or how Cicero and Tacitus saw it, but to
most ordinary Romans it was probably self-evident.
Lines from the great poets could attain quasi-oracular status,57 and we can
be sure that after 27 BC everyone knew at least one passage from Ennius,
without necessarily knowing its context:58

53 Dio Cassius 57.18.4-5 (Xiphilinus): kdytbv te ti to; Kai StflvkkEiov, akkatg gEV ovbfv
no Ti")g nokfw; Zpdvip npooi]Kov, rtpo; de to irapovra adogEvov, ouy rjou/g otpag
t'Kivri* ekeye yup bit- rplg de rpii)Koaitov JtrpiTEkkogEvtDV evtavtinv 'Ptugaloug
Eglpukoc; oktt OTOUlg . . . The Loeb translator loses the sense of lldogEVOV (‘applied to the
situation then existing’); cf. n. 46 above.
54 Dio Cassius 62.18.3 (Xiphilinus): Kai gdktofl’ 8xi aurone; f| gvqgT] TOO koyioi) TOO
Kata TOV Tl|h!pldv ItOTE (toflevrog ef)opv|lei. Still unwilling to countenance chanting,
the Loeb translator renders CI0OEVTO5 as ‘the oracle which . .. had been on everybody’s
lips’.
55 Dio Cassius 62.18.4 (Xiphilinus): gETO|lak0VTE5 ETEpov koyiov dig Kai SlflukkElOV
bvTtog bv gdov eon de TOVTO* Eoyaro; Aiveadoiv grgpoKrovog f|yEgovEUoei. For
Ijdov the Loeb translator has ‘proceeded to repeat’.
56 C1L F.l, pp. 1-50; Degrassi 1947.1-87. Nineteen AUC dates survive on the fasti consulates,
more than ninety on the triumphales.
57 See for instance Cicero De republica 5.1: ‘moribus antiquis res stat Romana uirisque’ [Ennius
Annates 156 Sk|, quern quidem ille uersuni net breuitate net ueritate tamquam ex oraculo mihi quodam
esse ejfatus uidetur.
58 Ennius Annates 154-5 Sk, in Varro De re rustica 3.1.2: nam in hoe mine denique est ut dici possit,
non cum Ennius scripsit: ‘septingenti sunt panto plus ant minus anni augusto augurio postquam inclita
condita Roma est’. Applied to Augustus: Suetonius Diuus Augustus 7.2.
 50 UNWRITTEN ROME

747 BC (Fabius Pictor), or 728 (Cincius Alimentus), or 751 (Cato)74- md a

dynasty of Alban kings had been created to reach back to Aeneas.75
New patterns had to be found for the passing of time. In Virgil, Jupiter
prophesies three years for Aeneas, thirty for Ascanius, three hundred for the
Alban kings.76 In Livy, Camillus dates the sack of Rome to ab urbe condita 365,
the magical number representing a ‘year of years’.77
It is tantalising not to know on what grounds the learned senator Manilius
in 97 BC calculated the ‘great year’ marked by the life-cycle of the phoenix:
evidently the current era had begun in 312 BC (Ap. Claudius’ revolutionary
censorship), and the previous one in 852 BC.78 The latter date looks like a
century before the foundation; is it completely a coincidence that Jerome, who
had access to otherwise unknown traditions on early Rome, put the first
‘centenary games’ in 452 BC?79 Whatever arcane systems were devised, the
simple hundred-year saecuhtiH always retained its power.
The annalist Piso Frugi marked the opening of Rome’s seventh saecuhim
in 158 BC;80 other historians noted significant events in ab urbe condita 400 (a
patrician ‘backlash’) and 700 (a disastrous fire);81 and Claudius Caesar held
Secular Games, and displayed a fake phoenix, in the eight hundredth year of
the city.82 But as we have seen already, not everyone believed the chronology
of historians and emperors. If the prophets were chanting that thrice three
hundred years had come and gone, that was enough to cause panic among the
People.83
A late-republican expert on prophecy told Varro that provided the histo­
rians’ narrative about Romulus’ augury was true, the gods had promised that

74 Dionysius of Halicarnassus 1.74.1-2.

75 Diodorus 7.5.6-12 (from the Armenian version of Eusebius’ Chronicle), Livy 1.3.6-10,
Dionysius of Halicarnassus Ant. Roni. 1.71, Ovid Metamorphoses 14.609-21, Fasti 4.39-54,
Origo gratis Romanae 17.4—19.1. The new system was already in being in the third century
BC: see Forsythe 1994.113-23.
76 Virgil Aeneid 1.265-74. Servius (on Aeneid 1.272) was puzzled: qnoinodo treccntos annos (licit,
cum earn [i.e. gentem Hectoream] qnadringentis regnasse constet sub Albanis regibtts?
77 Livy 5.54.5; see Hubaux 1958.13-88. Magical: Papyri Graecae magicae XII. 139; cf. (e.g.)
Pliny Nat. Hist. 19.12, 34.33, Quintus Curtius 3.3.10, Heliodorus Aethiopica 9.226, Augus­
tine City of Cod 18.53.
78 Pliny Nat. Hist. 10.4-5: the 215th year of a 540-year era (if the numerals in Pliny s text are
reliable).
79 Jerome Citron. 112 Helm (Roinae ... agon centenarins pritnnm actus); cf. 88 Helm for the
unique version of Remus’ death (a Fabio Ronntli diice occistts). On the rival sequences ot ludi
saeculares, see Wiseman 1998.165-7.
80 Piso fr. 46 Forsythe (Censorinus De die natali 17.13).
81 Livy 7.18.1 (perhaps from Licinius Macer); Orosius 6.14.5, 7.2.11 (from Livy').
82 Tacitus Annals 11.11.1; Pliny Nat. Hist. 10.5.
83 Above, nn. 53-4.
 FAUNS, PROPHETS, AND ENNIUS’ ANNALES SI

the Roman People would last 1200 years, one century for each vulture.84
What is interesting is the reservation. Historians were notoriously capable of
lying,85 whereas prophets could claim that their knowledge came straight from
the gods. Poets were liars too, and yet privileged to see the gods;86 the Muses
themselves taught Hesiod that they could tell both lies and truth alike.87
Ennius was inspired by the Muses of Olympus, who were also the Latin
Camenae.88 He used the metre of Homer rather than that of the Fauns and
prophets, and no doubt his sweet-voiced Egeria sounded quite different from
the terrifying threats of the Lucretian miles.8’ Nevertheless, the title of his great
work suggests that he saw the world in much the same way that they did. It is
appropriate that later readers could treat his lines as the oracles of a prophet.90
Of course Ennius is rightly celebrated as a pioneering innovator, the
founder of a great tradition, a poet who deserves his place in Raphael’s
Parnassus next to Homer, Dante and Virgil. But he is no less interesting as a
witness of his own time, when the conditions of pre-literary Rome were still
familiar.

84 Varro Antiqnitates book 18 (fr. 4 Mirsch), cited in Censorious De die natali 17.15: ait
fuisse I ettinm Ronnie in angnrio non ignobilem ingenio magno, cninis doeto in diseeptando parent; cunt se
andisse dicentem, si ita esset ut traderent historic! de Romidi nrbis condendae attgnriis ac XU mdtiiris,
qnoniain C.\’.\’ annos incolnmis preterisset popular Romanns, ad ntille et ducentos pementumin.
85 Evidence and discussion in Wiseman 1993a.
86 Ovid Fasti 6.23 (‘ins tibi fecisti tinmen caeleste nidendi', says Juno), but also 6.253 (non equideni
nidi—naleant tnendacia natmn!).
87 Hesiod Theogony 22-8.
88 Ennius Annales 1 and 487 Sk.; n. 34 above.
89 Ennius Annales 113 Sk: olli respondit suatiis sonus Egeriai. Threats: n. 36 above.
90 Cicero De repnblica 5.1 (n. 57 above): cf. Cole 2006.532—3.
 CHAPTER FOUR

The God of the Lupercal

I
On 15 February, two days after the Ides, there took place at Rome the myste­
rious ritual called Lupercalia, which began when the Luperci sacrificed a goat at
the Lupercal. There was evidently a close conceptual and etymological connec­
tion between the name of the festival, the tide of the celebrants, and the name
of the sacred place: as our best-informed literary source on Roman religion,
M. Terentius Varro, succinctly put it, ‘the Luperci [are so called] because at the
Lupercalia they sacrifice at the Lupercal ... the Lupercalia are so called because
[that is when] the Luperci sacrifice at the Lupercal’.1
What is missing in that elegantly circular definition is the name of the
divinity to whom the sacrifice was made. Even the sex of the goat is unclear—
Ovid and Plutarch refer to a she-goat, other sources make it male"—which
might perhaps imply a similar ambiguity in the gender of the recipient.3 Varro
does indeed refer to a goddess Luperca, whom he identifies with the she-wolf
of the foundation legend; he explains the name as Ittpa pepercit, ‘the she-wolf
spared them’ (referring to the infant twins), so I think we can take this as an
elaboration
---------- 1 on the myth, and not much help for the ritual.4

Roman Society Presidential Address, November 1993. I am very grateful to those who
r°j'7kntT i°n tar''Cr vcrs'ons °f die argument, given at the British School at Rome, at Stan-
°n ■1 Webster Lecture) and at Berkeley. But my greatest debt is to the marvellous
collecnon of material in Dr Elisabeth Smits’ Utrecht thesis Faunas (Smits 1946).

1 Varro De lingua Latina 5.85, 6.13: Luperci, quod Lupercalibus in Lupercali sacra faciuut. ■ ■ Luper-
calia dicta, quod in Lupercali Luperci sacrafaciuut.
2 Ovid Fasti 2.361, Plutarch Romulus 21.4 (afyeg), Quintilian Institutio 1.5.66 (lucre per
capnini) .Servius on Virgil Aeneid 8.343 (de capro luebatur). The Luperci skinned the sacrificial
goat an used its hide for wearing and for striking those whom they met: Dionysius of Hali-
1:®0-1 (Tubero fr. 3P), Nicolaus of Damascus FGrH 90 Fl30.71, Festus (Paulus)
*n j ’ -VI ^St' 2-445-6, Plutarch Romulus 21.4-5, Valerius Maximus 2.2.9. They were
called or/M. evidently a fonn of capri: Festus (Paulus) 49L, cf 42L.
3 Amobius Aduersus gentes 7.19: dis feminis femimis, mares maribus hostias immolare. (Female
o enngs to aunus at Ovid Fasti 4.652 and Horace Odes 1.4.11—12 are regarded by the
commentators as ‘un-Roman’.)
4 Amobius Aduersus gentes 4.3 = Varro Antiquitates rerum diuinanim fr. 221 Cardauns.
 THE GOD OF THE LUPERCAL 53

‘Lupercalia’ is just one of the festival days (dies feriati) that are named in large
letters on the pre-Julian calendar. (Whether that list goes back to the early regal
period, as Mommsen thought, or no further than the fifth century BC, as is
argued by Agnes Kirsopp Michels in her book on the Roman calendar,5 it is
the earliest evidence we have for the Lupercalia.) There are forty-two such
names, of which thirty end in -alia-, and at least twenty of those thirty are
formed from the name of the divinity concerned—Liberalia, Floralia, Neptu-
nalia, Saturnalia, and so on. But there are others that are not (e.g. Agoualia,
Feralia, Vinalia), and I think it likely that ‘Lupercalia’ belongs in that category.6
Like the names of the ‘large-letter’ festivals, so too the names of the famines
(priests of individual divinities, of which thirteen out of fifteen are known)
must reflect the pantheon of early Rome.7 As one might expect, the two lists
overlap substantially: corresponding to the Jlainen Qiiirinalis and six of the
minor Jlamines—Cannentalis, Cerialis, Furrinalis, Portunalis, Volcanalis and Voltur-
nalis—are the calendar items Quirinalia, Cannentalia, Cerialia, Furrinalia,
Portunalia, Volcanalia and Volturnalia. But no name corresponds to the Luper­
calia, and in fact we know that there cannot have been a specialist Jlainen to
look after that cult; on 15 February the Jlainen Dialis, Jupiter’s priest, was in
charge.8 That fact, together with Varro’s circular explanation of Lupercalia-
Luperci-Lupercal, seems to imply that for the Romans no one god (or goddess)
was particularly associated with the ritual.
In modem accounts it is normally taken for granted that the divinity
honoured at the Lupercalia was Faunus, and that is indeed what Ovid says.9
Other authors, however, give other names, as we shall see; and for Faunus in
particular there is a strong pritnafacie argument against. His festival was the Ides,
13 February, very close to the Lupercalia but not the same day. It was in fact the
dedication day of Faunus’ temple on the Tiber island, founded in 193 BC,
which eiy’/r to imply that Faunus was thought of as closely associated with the
Lupercalia cult, but not himself the recipient of it.10

5 Michels 1967.207-20 for a history of the controversy. Cf. North 1989.574: ‘We do not
know when this form of calendar was introduced, though it may well have been in the
course of the republican period; its introduction might or might not have coincided with the
fixing of the list of festivals in capitals.’
6 The list is in Degrassi 1963.364-5; I have omitted the Ides of each month from the totals.
Degrassi gives ‘Lupercalia Luperco sive Fauno’, but that begs the question.
7 Varro De lingua Latina 5.84: horuni singnli cognomina liabent ab eo deo cui sacra faciunt (similarly
7.45, citing Ennius Annales 116-18 Sk); Vanggard 1988, esp. 24-8 [who however omits the
Jlainen Firbialis (ILS 6457)].
8 Ovid Fasti 2.282. (Conversely, minor famines without corresponding ‘large-letter’ festivals
are Falacer, Floralis, Palatualis, Pomonalis [and Pirbialis].)
9 Ovid Fasti 2.267-8, 303-4, 423-4; 5.99-102. Like Horace (Odes 1.17.1-4), Ovid assumes
the identity of Faunus and Pan (Fasti 2.423-4, 2.84, 4.650-3); see now Parker 1993.
10 Faunus temple: Ovid Fasti 2.193—4, Degrassi 1963.4, 223, Livy 33.42.10, 34.53.4. But see n.
51 below.
 S4 UNWRITTEN ROME

The Lttpercalia ritual and its associated myth, of the suckling of Remus and
Romulus at the Lupercal," have been a subject of inexhaustible fascination for
scholars both ancient and modem. In recent years the prevailing mode of
enquiry has been that of the comparativists, beginning in 1964 with Gerhard
Binder’s very influential monograph Die Atissetznng des Konigskindes,'2 and
continuing with Andreas Alfoldi’s formidably learned argument on ‘the struc­
ture of the pre-Etruscan Roman state’, almost entirely based on his comparative
interpretation of the Lttpercalia ritual and the myth of the twins.13 Since then, we
have had Christoph Ulfs book Das roinische Lupercalienfest, arguing from the
supposed parallel of African initiation rites, and Jan Bremmer’s anthropological
interpretation of the foundation story in the Bremmer and Horsfall collection
Roman Myth and Mythography.'4 (One might expect the name of Georges
Dumezil to appear in this doxology; but Dumezil’s only extended treatment of
the Lnpercalia was in an early work, and in his later years he disowned it.15)
The trouble with comparativist analysis is that it argues synchronically, and
makes no adequate allowance for change over time.16 Alfbldi, for instance,
claims to extrapolate, from the details given in our historical sources, the
model of a ritual which had served a society thousands of years earlier and
thousands of miles away (unimaginable to the Romans themselves) and which
remained essentially unchanged into the historical period, effectively dictating
to the Romans the performance of ritual acts that had little or no significance
for their own thought-world. Surely that is absurd. No doubt religious behav­
iour is inherently conservative; perhaps Roman religion was more
conservative than most. But any community’s dealings with its gods must
reflect, at some level, its own needs and preoccupations, and adjust, with what­
ever time-lag, as those needs and preoccupations change.
There have, of course, been some protests at some of the comparativists
excesses,17 but perhaps the time has come for a new look at the Lttpercalia from

11 Ovid Fasti 2.381-422, Origo gratis Roinanae 22.1, Servius on Aeneid 8.343; cf. Plutarch
Romulus 21.4, Dionysius of Halicarnassus 1.32.3—4, 1.79.7-8. Detailed analysis in Wiseman
1995.77-88.
12 Binder 1964, esp. 96-115 (ch. 10, ‘Der Romuhismythos und das Lupercalienfest ).
13 Alfbldi 1974, esp. 69-180 (chs 3-6, ‘Der Mythos von der Wblfin-Urahnin , Das
Luperkalienfesf, ‘Hirtenkriegertum und Mannerbund’, ‘Zweiteilung und Doppehnonar-
chie’).
14 Ulf 1982; Bremmer 1987a. For earlier literature see Ulfs bibliography, and add Pbtscher 1984.
15 Dumezil 1929.197-222; cf. Dumezil 1970.349 n. 33. The Lttpercalia offered no support for
Dumezil s ‘tripartite’ theory, on which see Momigliano 1984 — 1987.135—59, and Belier
1991.
16 Bremmer, however (1987a.38—43), does argue for a historically specific context.
17 E.g. Weiwei 1967, responding to Binder 1964; Versnel 1976, reviewing Alfbldi 1974. As an
analogy, cf. Momigliano 1984 on Dumezil, and 1989.55: ‘What Dumezil cannot do,
because it is contradictory in tenns, is to postulate an invariable Indo-European pattern as
the explanation of the continuously changing relations between the social groups of Rome.'
 THE GOD OF THE LUPERCAL 55

an empirical historian’s viewpoint, plotting the evidence over time and

looking for the ways the ritual developed in a constantly changing world. It
may help to keep the material under control if we focus in particular on that
puzzling question: to whom did the Lnperci sacrifice at the Ltipercal on the
morning of the Ltipercalid?

II
Our earliest evidence for the god of the Ltipercal dates back to the middle of the
third century BC. It is a fragment of Eratosthenes, and I draw particular atten­
tion to it because for some unaccountable reason it is not included in Jacoby’s
Fragments dsr griechischen Historiksr."1 It comes in the scholia to Plato’s Phaedms,
at the point where Socrates mentions the Sibyl. The scholiast gives a list of
Sibyls, including this item at no. 4:19

TETopTT) ’iTakiKij. n ev Epr)[.iiq Tfj; ’Irakia; tt)v dicnptpi’iv kaxoboa, fj;

vioG eyeveto Eiiavdpog, d to ev Pwpn tou navdi; tepov, rd
KUkOVgEVOV AOVJtEpKOV, KTlOag. JTEp'l EYPCHpEV ’EpaTOO0EVT|C;.
The fourth is the Italian [Sibyl], It was her lot to spend her life in the wilderness
of Italy; her son was Evander, who founded the cult-place of Pan in Koine,
which is called Luperkon. Eratosthenes wrote about her.

Clement of Alexandria, who was very well read in the Greek philosophers,
evidently had the same passage in mind when he wrote his Stromateis about AD
200. Arguing that the Sibyls, like Moses, pre-date Orpheus and the other sages
of Greece, he discusses the Phrygian and Erythraean Sibyls, and then goes on:.20
211

pspvT)Tat tovtwv 'IIpankEibi]; o flovTuco; ev tio jtEpl xpr]OTT|piiov. Etb

&e ti']v Aiyi’n’Tictv nat ti']v TtciXijv, ij to ev 'Pcopfl Kappukov wkijoev, rjg
uid; Euavdpoq d t6 ev 'Ptbpfl top flavor lEpdv to Aoujtepklov
KakoupEvov KTioag.
Heraclides Ponticus refers to them in his On Oracles. 1 say nothing of the Egyptian
[Sibyl], or the Italian one who dwelt at the Kannalon in Rome; her son was
Evander, who founded the cult-place of Pan in Rome, which is called Luperkion.

The last phrases, agreeing almost word for word with the Plato scholiast, show
that Clement too knew of the Italian Sibyl, and the cult of Pan at the Ltipercal,
from Eratosthenes.21

18 The passage is referred to, but not quoted, at FGrH 241 F26; no comment in Jacoby 1962.713.
19 Schol. Plato Phaedms 244b, Ruhnk p. 61. [Greene 1938 includes the passage only in the
apparatus critiais; no explanation is offered.]
20 Clemens Alexandrinus Stromateis 1.108.3; cf. Heraclides Ponticus fr. 130 Wehrli.
21 Eratosthenes was clearly expanding Heraclides’ list; see Parke 1988.23-36, who however
does not refer to these two passages.
 56 UNWRITTEN ROME

This fourth Sibyl must be the ‘Cimmerian Sibyl in Italy’ mentioned by

Naevius and Piso.22 She clearly owes her existence to the early identification of
Lake Avernus as the scene of Odysseus’ visit to the dead:23

The vessel came to the bounds of eddying Ocean, where lie the land and city of
the Cimmerians, covered with mist and cloud. Never does the resplendent sun
look on this people with his beams, neither when he climbs towards the stars of
heaven nor when once more he comes earthwards from the sky; dismal night
overhangs these wretches always.

That doesn’t sound much like the Bay of Naples. Nevertheless, Pliny and
Festus are explicit that the city of the Cimmerians had been next to Avernus,
between Baiae and Cumae; and though there is a textual corruption, the Origo
gentis Ronianae evidendy made it the home of the Sibyl.24 That idea probably
goes back at least as far as the fifth century: Aeschylus’ Psychagogoi seems to
have been set at Lake Avernus, with a chorus descended from Hermes; and
Hennes was the father of Evander.25
At some stage the story of Evander and his prophetic mother was moved to
Rome. In Eratosthenes she is still a Sibyl, living at the Cermalus, the site of the
Lnpercal;-1’ later, when the story of the Cumaean Sibyl had developed, with a
canonical date in the time of King Tarquin, Evander’s mother became
Camientis, ‘a prophetess before the Sibyl came to Italy’, as Livy puts it.-7 So
Eratosthenes item on the Italian Sibyl and the Roman cult of Pan belongs to a
quite early stratum of Roman legend. But how early?
Evander was an Arcadian from Pallantion, from which was named
Palatium, the Palatine hill.28 Pallantion was mentioned in Stesichorus Gery-
oneis, the sixth-century poem that told the story of Herakles’ tenth labour, the
cattle of Geryon. The Roman legend of Herakles and the cattle, which
explained the Forum Bovarium and the Ara Maxima, was evidently current by

loss «US '"St ~ Varro Aiitiquitates rcrum diuiiianmi fr. 56a Cardauns. Parke
lMS.il wrongly assumes she was invented by Naevius.
F74Rrr OtySSq’ "14~19 (trans- W' Shewring), Strabo 5.4.5 (244); cf. Sophocles 7 rGF 4
Thesproti'] "ekr""l‘""ei°" at ‘Aomos’ lbllt Ogden 2001.47-51 puts it at the Acheron in

Pliny Nah Hist. 3.61, Festus (Paulus) 37L; Grip) qcntis Ronianae 10.1, Sibylla in oppido quod
, nsf-i.e. Cinimmiim, as in Pliny?
’3 abnv K F273, 273a <Arist°Phancs Frogs 1266, Maximus of Tyre 8.2b); cf. n. 23
c (Sophocles). Hennes as father of Evander: Dionysius of Halicarnassus 1.31.1, 1.40.2
k, " yan"enta). Virgil Aeneid 8.138, 336 (bv Camientis), Pausanias 8.43.2 (by the
’6 Vano n r Lad°/1; °Vld Fasli 1472 ("ot a g°d)- Servius °>> 8-130 (Echemos).
27 Live 1 f '"■?™5"'™ 3-34- Hutarch Romulus 3.5 = Fabius Pictor FCrH 809 F4.3).
8 336 .i • C i” ° ° - 3 ^22l,l’ Dionysius of Halicarnassus 1.31.1, 1.40.2, Virgil Aeneid
dmio'uhu ’SLC ’I0 ”■ 35 bcl°W' Cllmacan Sibyl; Lactantius Div. inst. 1.6.10-1 f = Varro
V ‘1"‘^IC.rcn"" 56a Cardauns; Parke 1988.76-9.
28
no c tngua Latina 5.53, Dionysius of Halicarnassus 1.31.4, Livy 1.5.1, etc.
 THE GOD OF THE LUPERCAL 57

about 530 BC, when the hero’s deification was celebrated on a temple in the
Forum Bovarium itself.29 In our later literary sources Evander greets Herakles,
and his mother prophesies the apotheosis; but whether that too goes back to
the sixth century BC it is impossible to say.
Certainly the sixth century is too early for Pan. Even in Athens, the Arca­
dian god found a home only in the fifth century, after his help at the battle of
Marathon.30 In fact, the Athenian cult may well be helpful for our purposes. It
was set up in thanks for victory, in a cave below the Acropolis, close to the
point where the temple of Victory guarded the entrance to the citadel.31 The
Roman topography corresponds exactly.
The temple of Victory at the western comer of the Palatine was begun in or
about 307 BC, but not finished till 294.32 Archaeological evidence now reveals
why it took so long. First, it was a very large and imposing building, even
bigger than its later neighbour, the temple of Magna Mater. Second, the
building programme evidently involved more than just the temple itself; the
side of the Palatine overlooking the Forum Bovarium was built up with great
terracing walls in opus qnadratinn, and it is probable that the programme
included a new monumental approach, the cliuiis Victoriae. The effect must
have been like the entrance to an acropolis, with a Victory temple at the gate.33
We know from Dionysius that there was a clear conceptual and topograph­
ical relationship between the Victory temple and the Lnpercal. Both cults were
supposedly founded by Evander, and both came from Arcadia. The victory
goddess, according to Arcadian legend, was the daughter of Pallas son of
Lykaon, eponyms respectively of the Palatine (Pallantion) and the Lnpercal
(Lykaion). As Dionysius tells us, her temple was at the top of the hill, and the
Lnpercal cave at the bottom—just as in Athens.34 In view of this close associa­
tion of the two cults, it is not surprising to find that one of the names our
sources give us for Evander’s mother is Nikostrate, ‘victorious army’.35
There is another parallel with the Pan cave in Athens, this time involving
the foundation myth. In one version of the story of the twins, Mars ravishes

29 PA-/GF Stesichorus 85 (1’ausani.is 8.3.2); Enea 1981.121-2, Cristofani 1990.119-20 and tav.
IX.
30 Herodotus 6.105; Borgeaud 1988.133-62, Garland 1992.47-63.
31 For the topography see Travlos 1971.70-1, 148-57, 417-21; the juxtaposition ofPan’s cave
and the Nike temple is illustrated in Garland 1992, pl. 11.
32 Livy 10.33.9; begun by L. Postumius Megellus as aed. cur., dedicated by him as cos. II.
33 Pensabene [1998, esp. 26-34, 115-19); Wiseman 1981 = 1987a. 187-204.
34 Dionysius of Halicarnassus 1.32.3—33.1, with Wiseman 1981.35-6 = 1987a.187-8.
35 Strabo 5.3.3 (230), Servius on Aeneid 8.336. Otherwise Themis (Dionysius of Halicarnassus
1.31.1, 1.40.2) or Tiburs (Servius audits on Aeneid 8.336), the latter implying an identifica­
tion with the Tiburtine Sibyl. Cf. n. 27 above, and for the Tiburtine Sibyl (Albunea) see
Lactantius Div. inst. 1.6.12 = Varro Antiquitates renini diuiuaruui fr. 56a Cardauns, with
Coarelli 1987.103-10.
 58 UNWRITTEN ROME

their mother inside the Lnpercal cave itself, just as Apollo rapes Creousa in Pan’s
cave in Athens, and fathers on her the founder-hero Ion.36 Pan himself uses
caves for raping nymphs, and a comic version of that idea is applied by Ovid to
Faunus, his Roman Pan, in one of the aetiological stories for the Lttpercalia.57
So the Eratosthenes fragment enables us to see Pan in Rome in two
different areas of life: on the one hand, sex and conception; on the other, war
and victory. We shall be pursuing both those aspects—the former in sections
III and IV, the latter in section V—but first it is worth noting that the very idea
of Pan in Rome is not as paradoxical as one might think. Herakles had his cult
at the Ara Maxima by the sixth century BC; the Dioscuri had theirs in the
Forum in 484, and Apollo his in the Flaminian fields in 431; Asklepios was
brought from Epidauros to the Tiber island in 291.38 Why should Pan not have
been introduced at the Lnpercal some time in (say) the late fifth or early fourth
century?

III
An interesting contemporary sidelight on Eratosthenes’ Roman Pan is
provided by a series of engraved bronze mirrors from Praeneste. A third-
century example [fig. 10] portrays a sprightly little ithyphallic goat-legged Pan,
labelled PAINSSCOS for Paniskos, dancing with ‘Mannas’ the satyr;39 Marsyas
was a figure of some importance in Rome, the legendary ancestor of the gens
Marcia and the symbol of libertas, with his statue in the Forum.
A much more elaborate scene, dated to the third quarter of the fourth
century [fig, 11], evidently shows the Lnpercal myth itself, with a she-wolf in
the centre suckling human twins.41 To the left stands a male figure, naked but
for boots and a goatskin loosely knotted round his neck by the forelegs;4- he is
wild and unkempt, and carries a lagobolon, the shepherd’s throwing-stick. I
think he is Pan. The contemporary iconography of Pan in his native Arcadia

36 Servius audits on Aencid 1.273: repentino occiirsu lupi turbata refiigit in speluncani, m qua a Marte
compressa cst. Euripides Ion 491—506, 936—41; for Ion as a founder, ibid. 74, cf. 1571—94, I
am grateful to Christina Kraus for pointing this out to me. Cf. Borgeaud 1988.151-2 for the
cave as a ‘wild spot in the heart of town’.
37 Euripides Helen 188-90; Ovid Fasti 2.315, 332 (Faunus and Omphale in a very well-
appointed cave), cf. n. 9 above. A hint of rape from Silenus in his cave: Virgil Eclogues 6.13,
26 (cf. 6.27 for dancing Failin').
38 Herakles: n. 29 above. Castor and Pollux: Livy 2.42.5. Apollo: Livy 4.29.7. Asklepios: Livy
10.47.7, Epit. 11.
39 Gerhard et al. 1897.54, Taf 45; ILLRP 1201 for the inscription.
40 Torelli 1982.99—106; Coarelli 1985.91—119. For satyrs at Rome, see Wiseman 1988 =
1994.68-85, esp. 1988.4-5 = 1994.73-4 on Marsyas.
41 Adam and Briquel 1982; cf. Lexicon Iconographicnni Mythologiae Classicae 4.1 (1988) 131. For
an interpretation of the scene as a whole,"see Wiseman 1993b, 1995.65-71.
2 As in Philostratus Life ofApollonius 6.27: a libidinous satyr on Lemnos.
 THE GOD OF THE LUPERCAL 59

Fig. 10 Praenestine mirror, third century BC (ILLR.P 1201): Paniskos and

Marsyas.
60 UNWRITTEN ROME

Fig. 11 Praenestine mirror, fourth century BC (Adam and Briquel 1 ’<S ). on the
left, Pan Lykaios?

S /L4VS ™ 'n human f°nn> as a young man with a lagobolon, while Pan as a
man’s attested on the coins of the Black Sea colony of Panticapaeum,
ic was named after the god [fig. 12],43 The only sign of bestiality is very
o , trusive horns, effectively undetectable in the dishevelled hair of the ‘wild
man version, which is what I think we have here.

43 Arcadia: Head 1911.445, fig. 241; Hiibinger 1992.208, 210. Panticapaeum: Kraay and
Hirmer 1966.335, nos. 440-2. See in general Brommer 1949-50.
 THE GOD OF THE LUPERCAL 61

Fig. 12 Fourth-century BO coins ot (left) the Arcadian federation and (right)

Panticapaeum on the Black Sea: Pan in human form with hiqobo/ou, and as a
wild man.

The closest analogue to the figure on the mirror is a passage in Justin’s

abridgement of the first-century' BC historian Pompeius Trogus, on the Arca­
dian Evander at Rome:44

in Imins ilieibiis teinplmn Lycaeo, quern Craeci Puna, Rouuini Liipmmn appellant,
conslitnii; i/isnni dei siniulacnnn niidinn uiprimi pelle ainictinn cst, quo liabitu nunc
Rome imp< n alibtts diTiimtur.
At the >ot of this hill [the Palatine] he established a shrine to Lycaeus, whom
the GreeO . all Pan and the Romans Lupercus. The actual image of the god is
naked v. i .1 a goatskin cape, the costume in which the running is done nowa­
day' .a tin Laipeiralia in Koine.

We know that the Luperci of the first century BC wore their goatskins as loin­
cloths. and brandished goatskin thongs, not throwing-sticks.4’ However,
Trogus was evidently describing a statue which may have represented an
earlier state ot affairs, as illustrated on the mirror.
Trogus calls the god of the Lnpeiral ‘Lycaeus’—that is, Pan Lykaios, named
after the Arcadian Mount Lykaion.46 The Roman name he offers is merely a

-14 Justin 43.1.7.

45 Dionysius of ILdicarnassus 1.80.1 (Tubero fr. 3P), Nicolaus of Damascus FGrH 90 F130.71;
cf. Ovid Fasti 2.445-6 (thongs), 5.101 (dnetuti), Plutarch Romulus 21.4-5, Quacstioucs
Rommme 68 (A/oru/ki 280b). Justin's nunc is inexplicable.
46 Livy 1.5.1-2, Virgil Aeneid 8.343-4 and Servius ad lot., Dionysius of Halicarnassus 1.32.3,
1.80.1. Ovid Fasti 2.423—1, Plutarch Romulus 21.3, Caesar 61.1, Amony 12.1, Quaestiones
Ronianae 68 (Moralia 280c); see also Augustine City oj God 18.16 = Varro Detente populi
Romani fr. 29 Fraccaro. For the mysterious cults of Mt Lykaion, see Burkert 1983.84-93,
Borgeaud 1988.34-42. Hiibinger 1992.189-212.
 62
UNWRITTEN ROME

god Februfrius’ btok a

o
Arcadi ",0"te Lttpercal hoc fuisse ludicrnm ferunt, et a Pallanteo, urbe
A "a ^e"‘ Ptl^l'l,"t montem appellation; ibi Ettandnun, qni ex eo
^ M,'n ante ^pestatibus tennerit loca, solletnne allatum ex Arcadia
"" ' '""cnes Lycaeuni Pana uenerantes per liisinn atque lasciuiam
eunerent, quent Romani deinde uocarunt Inttttm.
hillT SJy^DneVen that time there existed this L"Pcrcal festival at the Palatine
and a“antlUm’ and then Ratine, after the Arcadian city of Pallantion),
ation r n I.61’ 3 man °^^rcacllan descent who held that region many gener-
s earner, had brought the rite from Arcadia and instituted it there; naked
g n ran about in shameless sport in honour of Lycaean Pan, whom the
Romans afterwards called Inuus.

titv for p3S °^Sexua^ Penetration (inire, to enter),’19 an appropriate iden-

the othe 1 ’ j1 jSe ramPant sexuality was one of his defining features. Ovid on
a HwaII lant 1 enta^es lycaean Pan with Faunus, a god of prophecy but also
eod of th10/ 6 strong sexual appetites.50 The idea of Faunus as the
—:j . e, "Percal g°es back at least as far as the historian C. Acilius in the
mid-second
----- J century bc.51

47 Livy fr. 63 Weissenbom = ‘Gelasius’ /irfii. Andromachum 11 12’ ^^^^'urie eXorta pesti-
inihi, cum saepemmicro in Romanis historiis legator Liuio oratore saepissui ((bellicosis
lentia infmita homimim milia deperisse atque eo frequenter iientiim nt in , li,ttrnr an etiam cultus
temporibus exercitus potuisset adscribi, ilia tempore deo tuo Febniano minim * hl,ec sacra illo
hie omnino nil proderat? illo tempore Lupercalia non celebrabantur. /n< """ Lupercalia autem
tempore non coepissc, quae ante Romulum ab Euandro in Italian! pi r n " " jjujus in secunda
propter quid instituta sunt, quantum ad ipsius superstitionis comimnta r‘'^^ ,r
decade loquitur nee propter morbos iiihibendos instituta coinnicinora st ritual of the
uidetur, mulienmi quae tunc acciderat exigendam. For febnian as t u P1 pjutarc sterihtatcni, tit ei
j1 Quaestiones
Luperci, see Varro De lingua Latina 6.13, 6.34, Ovid Fasti — ' ’ 75-6L Censorinus
Romanae 68 (Moralia 280b), Romulus 21.3, Numa 19.5, Festus ( au us
22.15.
48 Livy 1.5.1-2; cf. also Macrobius Saturnalia 1.22.2 (Pan ipse quern uouint 4ll(inSrns 69.2,
49 Festus (Paulus) 98L: init ponitur interdum pro coiicitbitu. E.g. Suetonius 13 2, etc); see
Seneca Epistles 95.21. Usually of animals (Varro De re rustica -.7.9, Livy
Arnobius Aduersus rentes 3.23 for Inuus as guardian of flocks. , 6 (with
50 Ovid Fasti 2.423-4. Faunus afando (or from cpcovr)): Cicero t c-r,.;us audits on
Pease’s commentary), Varro De lingua Latina 7.36, Origo gentis Romanae . , etc
Virgil Georgia 1.10-11, Servius on Acneid 7.47, 7.81. Sexuality. OraCpri wjth dear
Nonnus, no doubt from a Hellenistic source, makes Faunus the son o ■ >
reference to the Hesiodic ‘wild man’ Agrios, brother of Latinos: Hesiod The g y
Nonnus Dionysiaca 13.328-32, 37.56-60.
51 Plutarch Romulus 21.7 = Acilius FGrH 813 F2.
 THE GOD OF THE LUPERCAL 63

The three gods Pan, Faunus and Inuus are often identified,52 and also a
fourth, Silvanus, whose characteristics notoriously overlap with those of
Faunus.53 Silvanus is never mentioned in relation to the Lupercalia, but I think
one of his cult-places in Rome may well be associated with the ritual.
The naked Luperci ‘ran about’ this way and that: discurrere and diatheein are
the words most often used to describe them. According to Varro, they ran up
and down the Sacra Via.54 But Varro also calls their run a lustratio of the ancient
Palatine settlement, which should mean an encircling route round the hill; and
that is what Dionysius and Plutarch both imply.55 There is no real contradic­
tion, however. It was not a race, and the Luperci evidently spent much of the
day running about performing their antics; on the other hand, they began from
the Lupercal and they evidently ended in the Comitium, as is clear from the
Lupercalia of 44 BC, when a large crowd in the Forum, and Caesar on the
Rostra, were watching the climax of the show?6 That makes a very credible
lustratio of the Palatine [fig. 13], if we imagine a date for its institution when
the Velabrum was still a marsh, or a backwater of the Tiber.57
The Lupercal and the Comitium were both, paradoxically, the site of the
ficus Ruiuiualis, the fig-tree under which the she-wolf suckled the twins. The
duplication was explained by a miraculous relocation of the tree from Lupercal

52 Senius on Aeneid 6.775, Probus on Georgies 1.10, ps.Acro on Horace Odes 1.17.1, Rutilius
Namasianus De reditu 31-6 (multa licet priscum nomen deleuerit aetas, | hoc Inui castrum Jama
fuisse put.it, l <e:< Pan Tyrrhenis mutauit Maenala siluis | sine sinus patrios incola Faunus init; |
dun: rencuat largo niortalia seniinafetu, | Jingitur in uenerem pronior esse deus). Cf. Plutarch Numa
15.3 (Fauni like Panes).
53 Origo genii: Romanae 4.6. Silvanus as Faunus: Dionysius of Halicarnassus 5.16.3, Livy 2.7.2,
Valerius Maximus 1.8.5 (the voice in the silua Arsia). Silvanus as dangerous rapist: Augustine
City of God 6.9 (Varro Antiquitates return diiiinanmi fr. 111 Cardauns), 15.23. Dorcey
1992.33-40 vainly tries to argue away the similarities.
54 Discurrere: Festus (Paulus) 49L, Origo gentis Romanae 22.1; cf. Ovid Fasti 2.285 (of the god).
Diatheontes etc: Plutarch Quaestiones Romanae 68 (Moralia 280b-c), Romulus 21.5, Caesar
61.2, clunmy 12.1. Lupercorum persacram uiam ascension atqne descensum: Augustine City of God
18.12 = Varro De gente populi Romani fr. 21 Fraccaro.
55 Varro De lingua Lirina 6.34: turn februatiir populus ]n. 47 above], id est Lupercis midis lustratur
antiquum oppulum Pulutium gregibus humanis cinctum. Lustrare also at Ovid Fasti 2.32, 5.102,
Festus (Paulus) 75L, Censorinus 22.15; cf. Dionysius of Halicarnassus 1.80.1 (perielthein),
Plutarch Romulus 21.4 (peridrome), 21.8 (peritheein).
56 Cicero Philippics 2.85, Plutarch Caesar 61.3, Antony 12.1, Appian Civil Wars 2.109, Cassius
Dio 44.11.2.
57 Varro De lingua Latina 5.43-4, 5.156. Cicero (Pro Caelio 26) was probably right to date the
origin of the Luperci ’before civilisation and the rule of law’. [Deep-core analysis has now
disproved the Varronian picture of the early Velabrum: it is clear that the valley between the
Capitol and the Palatine had been neither a marsh nor a backwater, but an area of clay-beds
that was flooded only when the Tiber overflowed its banks (Ammerman 1998 and
2006.305—7, pp. 12—13 above).]
 UNWRITTEN ROME

a d °^^le s’te The dots represent the suggested beginning

en o the course of the Lttperci, each the site of a fig-tree: lower, the
"pern-, upper, ‘in front of the temple of Saturn’ (Pliny Nat. Hist. 15.77).
tem t e^ected by the wonder-working augur Attus Navius.58 One is
the £,0 t°.?UeSS t^at c^ere were necessarily two fig-trees, one at each end of
Dron ,rt nin’ Fig trees were associated with fertility and with goats, both
epithet and ^te authors give ‘Ficarius’ as an
aunus. In the passage where he describes Attus Navius’ miracle,

58 hi comitio: Tacitus Annals 13.58.1, Festus 168L; Conon FCrH -0 F48.8, . fI(M 5.54,
camassus 3.71.5; Torelli 1982.98-9. In Cemialo (i.e. Lupercal)-. Varro De 1 g Both
Livy 1.4.5, Ovid Fasti 2.411-12, Plutarch Romulus 4.1, Ongo genus Rom
(miracle of Attus Navius): Pliny Nat. Hist. 15.77. 9 above for
rus. See
59 Isidores Origines 17.7.17 (ficus a fecimditate)-, the wild fig-tree is capnjtats. See n.
n. -

60 Jerome Ad Isaiam 13.21 (PL 24.159, n. 66 below), Isidoros Origines 8.11.1N

mdgo hicubonem uocant, hunc Romani Faimum ficarium dicunt). Cf. 1 e agonius _.
for Fatuus Jicarius; Fatuus was another name for Faunus (e.g. Servius on . etiei
 THE GOD OF THE LUPERCAL 65

Pliny refers to a fig-tree, possibly the same one, that grew in front of the
temple of Saturn but had to be removed, with a sacrifice by the Vestals,
because it was undermining the statue of Silvanus.61
Silvanus and a fig-tree, in front of the temple of Saturn; and Propertius, in
his poem on Tarpeia, offers a grove of Silvanus, complete with a cave and a
spring, below the ctrx of the Capitol on the Forum side.62 The imagined land­
scape is very like that of the Lttpercal.b> The area in front of the Saturn temple
was supposed to be where the bones of Orestes were placed after he had
brought the image and cult of Artemis Tauropolos to Aricia. In the more usual
version of the myth, Orestes died in Arcadia.64 The details escape us, but it
looks as if the Comitium below the Capitol, like the Lnpercal below the Pala­
tine, was the site of an early cult imported from Arcadia, and that the two were
linked by the course of the Lttperci, from one fig-tree to the other.

IV
In dealing with Pan, Inuus, Faunus and Silvanus as gods of sexual energy and
desire, we must not omit two minor characters who are frequently associ­
ated—or indeed identified—with them in this respect. They are Incubus, or
Incubo. and Ephialtes: he who lies on you (in Latin) and he who jumps on you
(in Greek).'’-' They in turn are associated, or identified, with the pilosi, ‘the
hairy ones’.'"'

61 Pliny Nat. Hist 15.77 Ifnit et ante Satnmi aedem...); he gave the date of its removal, but the
numeral', have been lost from the text.
62 Propertius 4.5.3--6, cl. 13-14 for a spring at the site of the Curia; the ‘springs ofJanus' (Varro De
lingua Latina 5.156, Ovid Fasti 1.257-76, Metamorphoses 14.778-804) must have been thereabouts.
For ‘the wood below the Capitol' in what was later the Forum, see Dionysius of Halicarnassus
2.50.2; for the possibility that the Comitium was once a Incus, see Vaahtera 1993.103-7.
63 Dionysius of Halicarnassus 1.32.4, 1.79.8. 'Satyr country’: Wiseman 1988.12-13 = 1994.84.
64 Servius on Aeneid 2.1 16; cf. Herodotus 1.67-8 (bones of Orestes in Tegea), Pausanias 8.5.5,
Strabo 13.1.3 (582).
65 Servius on Aeneid 6.775, ps.Acro on Florace Odes 1.17.1, Augustine City of God 15.23,
Artemidorus Oneirocritica 2.37 (6 dE ’Et(lciTl); 6 UOTOg ELVUt TO) flavi VEVOglOTUl...),
Caelius Aurelianus De morbis chroma's 1.54—5 (Inaibonem aliqni ab hominis forma nel similitndine
nomen dicere dixerunt. . . item quiil.iin ueteres Ephialten nocauerunt, alii Epophelen, quod ntilis pati-
entibus perhibeatnr), Jerome I 'ita Panlli 8 (PL 23.23: nuns ex accolis eremi qnos uario delnsa errore
gentilitas Fannos satyrosqne et Incnbos nocans colit), ps.Augustine De spiritn et anima 25 (PL
40.789). For the etymology of Ephialtes, cf. Scholia Graeca in Hoineri Iliadem (ed. Dindorf)
3.248, Eustathius on the Iliad 560.10-11.
66 Jerome Ad Isaiam 13.21 (PL 24.159: et pilosi sahabnnt ibi, nel Incnbones nel satyros nel silnestres
qnosdam homines, qnos nonnnlli Fannos Jicarios nocant ant daemonnm genera intellegnnt), Mythogra-
plnts lAiticanns 2.24 Bode (Fanni antem sunt qni nnlgo Incnbae nel pilosi appellati sunt, et a qnibns,
dnm a paganis consnlerentnr, responsa nocibns dabantnr), Gregorius Magnus Moralia 7.36 (PL
75.786: qni namqne alii pilosi appellatione Jingantnr nisi hi qnos Graeci Panos, Latini Incnbos
nocant?), Isidorus Origines 8.11.103 (Pilosi, qni Graece Panitae, Latine Incnbi appellantnr, sine
Inni ab inenndo passim aim animalibns, nude et Incnbi dicnntnr ab incnmbendo, hoc est stnprando).
66 UNWRITTEN ROME

The pilosi happen to be attested first in the Vulgate, as the wild creatures
who Isaiah predicts will dance in the wilderness that once was Babylon, but St
Jerome took them from an earlier tradition in authors unknown to> us.67 Strabo
has Ephialtes along with Lamia, Gorgo and Mormolyke: as a bogey to frighten
treasure—snatch his cap
children;68 Petronius has Incubo as a goblin sitting on t*.
. it’s yours;69 coins of Bithynian Nicaea in the second amid third centuries
off, and
show Ephialtes,
AD show goat-legged and
Ephialtes, goat-legged wearing aa cap,
and wearing cap, in
in ms guise as Epopheles,
his guise
e pfi.il one . But above all Ephialtes and Incubus (or Incubo) were the
names of the god of nightmare, who sits on ;your chest
' ’ ” you>’—
while re -asleep and
stops you breathing.71 Peonies will keep __r him away; , Pliny, who ' tells”> us this,
calls the nightmare demons Fauni,72 and‘ since their nocturnal' assaults
----- i were
often sexual,73 it is clear that we are dealing with the libidinous and many-
faceted god of the Lttpercal.
At this point we return to the Praenestine mirrors, and in particular to a
pair, ated to the late fourth or early third century BC and clearly from the
ame workshop, which are engraved with related scenes.74 The first, now in
Baltimore [fig. 14], shows a Dionysiac scene surrounded by a vine with grapes.
e e a To the right, with an amphora ready to hand, a young
n, apparently naked, blows into a conch-shell(?); he is not garlanded, and
vo itt e horns appear from his hair. In the centre, a man with a garland on his
sprawls on the cushions, clearly dead drunk, while behind him, evidently
ry ng to revive him by pouring wine from a kylix, is a garlanded and bearded
gt e wit a thyrsus in the crook of his left arm. He seems at first to be hairy,
e eft forearm and right wrist show that he is wearing a tight-fitting hairy

67 Vulgite kolas 13.21,34.13; Jerome Ad iMiani 34.13 (PL 24.372: oiweiiiaun■ 1I1
68 -

scholia. . , tjlt. nlot juste,or

69 Petronius Satyrica 38.8, cf. Porphyrio on Horace Satires 2.6. — 1 s t]iat did it:
guarding treasure (references in Otto 1890.173), but it was more o c Hiibinger
e.g. Phaedrus 4.21, Martial 12.53.3, Festus (Paulus) 59 . or , inctuary on Mt
1992.198, 204: worn by dedicants (hunters and shepherds) at la

70 Lexicon Iconographiciiiii Mytholoipae Classicae 3.1 (1986) 802. Epopheles. '. . cf a]so
De morbis duonicis 1.3.54 (n. 65 above), Hesychius s.w. Opheles and Epopheles,
ArteraidorusOlle1rl,rn71rrt2.37(pEYdk«? Sdpionis 1.3.7,
71 Caehus Aurehanus De morbis chronics 1.3.54-/, M.urooius inT7—8 etc
Eustathius on the Iliad 561.8, on the Odyssey 1687.52; Aristophanes 11 asps R ’ ‘
72 Pliny Nat. Hist. 25.29, 30.84, Dioscorides Materia medica 3.140; cf. Aetius
1.84 (CMC 8.1.50) on the peony as ephialtia. 9 1 158-9)-
73 Caelius Aurelianus De morbis chronicis 1.3.56, Paulus Aegineta • 5 ( •
Incub(it)are in sexual sense: Plautus Persa 284, Pomponius Mela 3.83.
74 Gerhard et al. 1897.51-3, Taf. 42-3.
 THE GOD OF THE LUPERCAL 67

Fig. 14 Praenestine mirror, late fourth or early third century BC: drinking scene.

costume.7’ He looks rather sad; and the young man and the flute-girl seem to
have somewhat disapproving expressions.

75 Compare the ‘curious papposilen, obviously wearing a costume’ on a volute-crater of the

Arpi painter, also of late fourth- or early third-century date: Trendall and Cambitoglou
1982.924, pl. 362, better illustration in Trendall 1989 fig. 266. See also the actor playing
Silenos on the ‘Pronomos vase’ (Attic red-figure, c.400 BC): Seaford 1984.3-4, pl. III. Such
hairy-costumed performers evidently pre-date the genre of satyr-play, and appear on black­
figure vases from the first half of the sixth century onward: see Hedreen 1992.113-14,
125-6, 128, 163-4, pls 4, 31, 37, 44; also Hedreen 1994 pl. IV(a), where the hairy skin of
the ithyphallic silen/satyr stops at the elbows, and is therefore presumably a costume.
I
68 UNWRITTEN ROME

uc; bedroom
Fig. 15 Praenestine mirror, late fourth or early third century
scene.

For the second scene [fig. 15] (the mirror is in the Villa Giulia
Giuha a ’
we have moved from the triclinium to the aibicnlum. A lady, garlanded ana
an elaborate coiffure, lies on the bed naked, holding back the mantle round n
shoulders in a gesture of invitation. She is being pawed by what oo
same bearded figure in the hairy costume (his thyrsus is proppec uj.
him), while to the right a colleague in the same gear leaps up high, w it a a tore
in one hand and an amphora in the other. Between them the young man, us
 THE GOD OF THE LUPE.RCAL 69

horns now much more prominent, gallops across the bed playing the syrinx
with his left hand and pointing with his right at the lady’s private parts. They
still don’t look very happy, and it is worth noticing that they are not ithy-
phallic.753
Given that mirrors were often wedding presents, it is the marriage-torch
that tells the story for us. The bride awaits; the bridegroom is incapably drunk;
and the gods of sexual desire try' to wake him up and show him what to do.
The iconography even tells us who they are. The two figures in costume are
‘hairy ones’, pilosi. The one on the left is lying on the lady: incunibere, whence
Incubus or Incubo. The one on the right is leaping up: cphallesthai, whence
Ephialtes. The young man plays Pan’s pipes, and has Pan’s horns; but the
direction of his pointing finger is a clear instruction to inire, whence Inuus.
Inuus is named by Livy as the god of the Lnpercal. He had a cult-place,
Castrum Inui, on the coast of Latiurn between Antium and Ardea;76 it was
either identical with, or very' close to, the place called Aphrodision by Strabo,
Pliny and Pomponius Mela. Cicero refers to a birth-goddess, Natio, whose
shrine was one of a group in the territory of Ardea at which sacrifice was still
regularly offered in the first century BC. It looks as if Inuus was part of a
complex of ancient cults concerned with human reproduction and fertility.
Moreover. Antium and Ardea were the nearest ports to Praeneste, and linked
to the inland city by the cult of Fortuna. ” Fortuna, the guardian goddess of
Servius Tull ins. was a neighbour of the Lnpercal at Rome.79
Before . leave the mirror scene, let us remember that the pilosi Incubus
and Ephiaf - a ■ evidently in costume. Are we to think of them as supernat­
ural being-- • ■ ’ email performers? If the latter, are they performing in a stage
drama or a -.1 ritual? Perhaps these categories are too schematic: are they
human p ' ■ . impersonating supernatural beings, in a drama which is in
itself a ritua’ I he questions cannot be answered, but are relevant equally to
the LttpcHu! For according to Varro the Lttperci were ludii, players or

75a |The argument ignored by Gun 1998.1013-15, who believes that the woman on the bed
is ’sans doutr tine Menade', and that ‘la mise en scene suggere 1’idee d’une epiphanie
panique nocturne' (reference only to Borgeaud 1988.76).|
76 Livy 1.5.2, \ irgil . L in id (>.775, Martial 4.60.1, Silius Italicus 8.359; Rutilius Namasianus De
reditu 227-3(> (n. 52 above) contuses it with Castruin Novum in Etruria. Ct. Tomasetti
1910.460-1 on a villa 1‘iiapi in atiro Ardeatino, tenth century AD.
77 Strabo 5.3.5 (232), Pliny Nat. Hist. 3.57, Pomponius Mela 2.71; Cicero De iiatnra dcornin
3.47. For the archaic context, see Torelli 1993. Note that Horace (Odes 3.18.6) calls Faunus
I eneris sodalis, and that the Kutuli of Castrum Inui are Faiuiigenae in Silius Italicus 8.356. Ct.
also Vitruvius 8.3.2 for springs smelling of sulphur in Ardeatino—like Faunus’ oracle at
Albunea (Virgil Aeneid 7.84).
78 Coarelli 1987.74-9; cf. Torelli 1993.98: ‘il rapporto tra Preneste e Anzio e strettissimo.’
79 Ovid Fasti 6.476-9, 569-80; Coarelli 1988.305-28.
80 Cf. Seaford 1994.266-9 for men dressed as satyrs as part of wedding ritual; ibid. 308 for
weddings and Dionysiac mysteries, 270 n. 154 tor the mysteries as a spectacle.
 70 UNWRITTEN ROME

performers; the first stone theatre in Rome, begun in 154 BC but destroyed
soon after, was to have overlooked the Lttpercabf' and in Lactantius’ time
(though by then the circumstances were somewhat different) the Luperci even
wore masks.82

V
I think it is clear from all this that the Roman Pan attested by Eratosthenes is
perfectly explicable as the Hellenized form of an archaic Latin god of fertility.
But there is another aspect of his personality to be explored.
The longer version of Servius’ commentary on Virgil’s phrase gelida sub rape
Lupercal adds this learned comment to the identification of Pan Lycaeus:83

suut qui dicant /nine Il&va EvuctXlOV, detun bellicosuin; alii Liberum patrein, co
quod capro cifit dittina res, qui est hostia Liberi propria.
There are those who say that this Pan is Enyalios, th«le warlike god; others call
him Liber Pater, because a he-goat is sacrificed to him, which is the offering
appropriate to Liber.
Liber Pater, otherwise identified as Dionysus, is intelligible enough; but why
man-slaying Enyalios’, ‘the warrior with the flashing helmet’?8’1 That doesn’t
sound like Pan. However, Virgil at one point calls the Lupercal the cave of
Mars, and in one version of the foundation story Mars fathers the twins there.8’
There was also an aetiology of the running of the Luperci which derived it from
the victory of Romulus and Remus over Amulius, and their triumphant run
homewards waving their swords.86
That is appropriate to the cult supposedly founded below the temple of
Victory by Evander the son of Nikostrate, a cult analogous to that of Pan the
bringer of victory at Athens.87 The military associations are made to extend
also to the god’s Latin analogues: Faunus is a son of Ares in Dionysius and
Appian, and by a wonderful bilingual pun Inuus (’EvuoO^) is made the son of

81 Lndii: Varro Antiquitates rcnini diiiiiiamm fr. 80 Cardauns (Tertullian Di spicMUi/i. 5 ),

Schmidt 1989, esp. 88-9. Theatre: Velleius Paterculus 1.15.3; for the circums at
North 1992. ..
82 Lactantius Din. Inst. 1.21.45: niidi, uncti, coronati, ant personati ant Imo obhtt aimint. But see pp.
79-81 below.
83 Servius audits on Acneid 7.343.
84 Homer Iliad 2.651,22.132.
85
" Virgil Acneid 8.630, with Servius audits commentary (Fabius Annates Latim fr. 4P); Servius
auctns on Acneid 1.273 (n. 36 above).
86 Plutarch Romulus 21.6, from Butas’ elegiac Aitia.
87 See above, p. 57.
 THE GOD OF THE LUPERCAL 71

Enyo, the war-goddess whose Roman name was Bellona.88 The obvious
context for these versions is the Roman conquest of Italy; the real temple of
Victory was dedicated in 294, and the temple of Bellona was vowed in 296 and
dedicated a few years later.89
Now, an Inuus who is really Enuoits is no longer derived from ittire, and has
presumably lost his penetrative function. The source that gives us this eccentric
etymology carefully explains that Bellona’s son was goat-footed, and very
quick at running up and down hills. What defines him now is not sex but
speed.9" And the reason for that, I think, is that he has become the god of the
eqiiites.
The original Roman cavalry were the celeres, ‘the swift ones’—supposedly
the flying squad of three hundred horsemen who served as Romulus’ body­
guard.91 Their real origin was probably the late fourth century, when the
Romans, who had had no significant cavalry up to then, borrowed the idea
from the Sanmites.92 The defining ritual of the eqnester ordo, the parade (trans-
tiectio) to the temple of Castor ever}' 15 July, was introduced by Q. Fabius
Rullianus in his censorship in 304 BC,93 evidently an innovation in recognition
of a new elite corps. The first commander of Romulus’ celeres was called Fabius
in one surviving version of the story.94
In view of the later reputation of the Fabii, based on the patient caution of
the great Ciitiitator, it may seem paradoxical to associate them with cclcritas, of
all characteristics. But that is what their enemies evidently said of them. The
two most notorious Fabian exploits in the tradition of the early Republic are,
first, their offer to fight the Veientes on their own, which led to the disastrous
defeat at rite Cremera, and second, their rash engagement with the Gauls at
Clusium, which led to the even more disastrous sack of Rome. Before the
Ciinctiitor redefined their image, the Fabii could be represented as men who

88 Dionysius of Halicarnassus 1.31.2, Appian Basilike fr. 1 (identifying Faunus and Latinus).
Diomedes in Grammatici Latiui 1.475 Keil: <...> el Bellonae, ill est ’EVU0O5, filio, quern
capriuo pede humin poetae Jiuguut, quod suinnia moiitium et difficilia collium coucitato cursu caprae
more superaret, quotieus praedatoria uice grassaretur, citipedem lumc cursttm sibi repperisse testijicantmr.
89 Livy 10.33.9; Livy 10.19.17-21, Ovid Fasti 6.201-4.
90 Diomedes at n. 88 above. So too Faunus: Horace Odes 1.17.1—2 (uc/o.v), Ovid Fasti 2.285—
6.
91 Dionysius of Halicarnassus 2.13.2, Festus (Paulus) 48L, Pliny Nat. Hist. 33.35, Servius on
Aeneid 1 1.603 (a celeritate). The tribuuus celerum had the same relationship to the king as the
magister equimm to the dictator (and the Praetorian Prefect to the emperor): Pomponius in
Digest 1.2.2.15—19, Lydus De magistratibus 1.14, cf. 37.
92 lueditum Faticauum, FGrH 839 F1.3 (lines 19-22).
93 Valerius Maximus 2.2.9, De uiris illustribus 32.3.
94 Jerome Chronica on Olympiad 6.3 (Fotheringham p. 152): Remus rutro pastorali a Fabio
Roniuli duce occisus. The implement is significant, given the ancient etymology of Fabius from
words meaning ‘to dig’: Festus (Paulus) 77L, Plutarch Fabius Maximus 1.2.
 72 UNWRITTEN ROME

acted first and thought afterwards—gallantly courageous or dangerously head­

strong, according to your prejudice.95 The celeres themselves were similarly
controversial: according to Plutarch’s version, they were one of the symptoms
of Romulus’ increasingly tyrannical rule, and the first thing Numa did was to
disband them.96 Perhaps what was at issue, in the social and political context of
the late fourth century, was the behaviour of young aristocrats liberated from
the discipline of the phalanx by the innovation of swift and mobile cavalry.
Was it flair, promptness and glamorous high spirits, or rashness, arrogance and
mindless exhibitionism?
In his account of Roman military institutions, Polybius makes a very inter­
esting observation about the Roman cavalry. Nowadays, he says, they are
armed like Greek cavalry; but originally they used no body-armour and fought
naked except for loincloths (perizoutcitd), in order to maximise their speed in
getting on and off their horses.97 There is no reason to doubt the accuracy of
Polybius’ information, and it gives valuable confirmation of the military role of
the celeres in the late fourth century. It also directs us back to the Lupercalict.
‘Naked except for loincloths’ is also a description of the Luperci, as we meet
them in Dionysius and Plutarch. Ovid calls them ciuctuti, a very rare word
meaning not just ‘girt’ (that would be cincti), but ‘wearing the cuictus’—that is,
the minimal covering used by young men at exercise, sometimes called
campestre after the Campus where the young men trained.98 Poiripeius Trogus,
however (as excerpted by Justin), describes an ancient statue of the god of the
Lttpercal naked but for a goatskin used as a cape—exactly the costume of Pan
on the Praenestine mirror in about 330 BC." His assertion that that is also what
the Luperci wore directly contradicts the descriptions in Dionysius, Ovid and
Plutarch.
I suggest that the statue represented an archaic form of the Lupercus
costume, which was then superseded, perhaps for reasons of modesty, by the
goatskin loincloth. And since the ‘reformed’ Luperci were thus dressed (or
undressed) like the celeres, it is an economical hypothesis to assume that the
redefinition of Inuus as a quasi-ce/er dates from the same time. Faunus as the
son of Mars, Inuus as the son of Bellona, Pan as Enyalios, the run of the Luperci
as the triumphant waving of swords, Evander’s Lttpercal shrine associated with

p‘l I-^1 c” ^1vourahk‘ and hostile, are discussed by Montanan 1976.83—187,

a o • ,°n F‘ahh as celeres. Cf. Ovid Fasti 2.205 (the Fabii reach the Cretnera celeri

« EStSttJ54 ■'»- *
111 IS A °Ve campestre: Varro De lingua Latina 5.114, ps.Acro on Horace Epistles
99 » “ '1J7- l9 -5-
 THE GOD OF THE LUPERCAL 73

the temple of Victory above—all these things fit most comfortably into the
years between the vowing and the dedication of the Victory temple (c. 307-
294 BC). The key moment was no doubt the censorship of Q. Fabius Rullianus
in 304.11)0
Valerius Maximus provides a confirmation of the hypothesis, in the second
section of his chapter on traditional customs (de institutis antiquis)."1' The exam­
ples he offers are in hierarchical order, from the Senate and magistrates down
to the poptilus, interpreted in military terms as the infantry. His one item on the
equester ordo—of which the celeres were the original nucleus—refers to the two
occasions each year on which the young cavalrymen were allowed to ‘show
themselves off to the city: equestris item ordiuis iuueutus omnibus auuis bis tirbeni
spectaculo stti sub tnagnis auctoribus celebrabant. Those two occasions were the
parade (transuectio) on 15 July, and the Lupercalia on 15 February,
The uiagtii aiictores to whom Valerius refers are respectively Fabius
Rullianus, who instituted the equestrian parade as censor in 304, and Romulus
and Remus, who were supposed to have instituted the Lupercalia after their
grandfather Numitor, now restored as king of Alba, had given them permission
to found a city at the place where they had been brought up. The foundation
of the Lupercal by the twins is an aetiological explanation for the division of the
Luperci into two groups, the Fabiani and the Quinctiales;"'2 Valerius refers to this
element of the ritual with the phrase diuisa pastorali turba.
Many items in Valerius’ collection of anecdotes, including some in this
chapter,1,0 come from the family history of the Fabii Maximi. The fact that he
associates the Lupercalia with the cavalry, and that one of the two groups of
Luperci was named after the Fabii, makes one suspect that perhaps both the
events he mentions were innovations by Fabius Rullianus in 304. The parade
was a wholly novel institution, for which Fabius himself properly took the
responsibility; but if he also introduced changes in the traditional ritual of the
Lupercalia, they would need to be disguised as a return to ancient practice, with
an appropriate aetiology from the distant past.
Ovid too tells the story of the origin of the Lupercalia, and his narrative
includes a very revealing detail. While the meat was being prepared after the
sacrifice of a goat to Faunus, the twins and their followers were exercising
naked. It was reported that robbers were stealing their flocks. They ran oft in
pursuit in different directions; Remus and the Fabii caught the robbers,
brought back the booty, and helped themselves to the meat, which was now

100 See pp. 57, 70-1 above.

101 Valerius Maximus 2.2.9.
102 Ovid Fasti 2.375-8, Festus (Paulus) 78L, Festus 308L; CIL 6.1933, 6.33421, 11.3205.
103 Valerius Maximus 2.2.1 and 4. Fabian items take up nearly a page in the index to Kempf s
Teubner edition.
 74 UNWRITTEN ROME

ready; Romulus and the Quincti(li)i returned empty-handed, and had to go

hungry.104 The story is very similar to that of the Pinarii at the Ara Maxima:
they came late and had to go without.103 In each case, there must have been a
ritual reason why certain persons involved in the sacrifice were not allowed to
share the feast.
We know that the cult of Hercules at the Ara Maxima was reorganised in
312.1 suggest that the Lupercalia were reorganised in 304, that the two teams of
Luperci were introduced at that time, and that this doubling of the number of
the potential participants made necessary an aetiological explanation why only
one of the two groups could share the sacrificial meat. If we are right to
attribute the innovation to Fabius Rullianus in 304, it is no surprise that the
Fabian group was privileged.
It was Mommsen who pointed out that the Fabii and the Quinctii, after
whom the two groups of Luperci were named, were the only patrician geutes
who ever used the praenonien Kaeso. The name, he suggested, was derived
from caedere, to beat, with reference to the ritual at the Lupercalia in which the
young men beat whomever they ran into.106 Now, the best known Kaeso
Quinctius was the young patrician who was supposedly put on trial in 461 BC
for beating up plebeians. His combination of glamour and arrogance reflects, I
think, the ambivalent image of the celeres.107
The Fabii who bore the name Kaeso are also significant. There were at least
two of them, and possibly four. The two certain ones were K. Fabius Vibu-
lanus, who as consular tribune in 479 led the ill-fated Fabian army to the
Cremera, and K. Fabius Ambustus, who in 391 led the embassy to Clusium
which provoked the Gauls to march on Rome.108 That is, both the episodes in
which Fabian haste and thoughtlessness led to disaster were attributed to Fabii
called Kaeso. However, there also traces of a tradition which gave the name to
Fabian heroes. In one version of the story of the priest who during the siege of
the Capitol miraculously walked through the Gallic lines to perforin sacrifice,
his name is given as K. Fabius Dorsuo (otherwise Gaius).J09 And in one version

10S P)V!d /?K'L2‘359~80’ esP- 374 'hacc certe non nisi victor edcf.
on a C1 . 0,0 3 01,0(3 Romantic 60 (Moralia 278e-f), Servius on Aeneid 8.269, Servius auctus
neivaTanotherbl"'^ R0"“"“W 8’3, LyduS Dr ""'A03'0'"7’"* 1-23: Pinarii and TOO

cos a*S° Pk’beian K. Duilii (Xnir 450, cos. 336) and K. Acilii (grandfather of
HutarZZZ^ZXS °f Da,naSCUS FGrH HM71’
108 Livy 9 48 5-50 n* ^ionyS’US ^Halicarnassus 10.5-8.
of M~ F k a 7 D10?yslus of Halicarnassus 9.14.1, 15.3, 16.3, 22.5. Of the ‘three sons
Kaeso .V115) a' i*StUS SCnt tO Clusium in 391 (Livy 5.35.5), the senior was evidently
5.24 1) ° "* a rea^y hold the consular tribunate three times (Livy 4.61.4, 5.10.1,
sius Dio 7 fr. 25.5; Gaius in Livy 5.46.2 and 52.3, Valerius Maximus 1.1.11.
 THE GOD OF THE LUPERCAL 75

of the story of the exploration of the Ciminian forest during the war against the
Etruscans in 310, the heroic spy is named as K. Fabius (otherwise Marcus), the
brother of Fabius Rullianus himself.110
It looks as if the Kaesones, both Fabii and Quinctii, were exploited by both
sides in the political controversies of the late fourth century, as patrician
charisma strove with plebeian constitutionalism for ideological supremacy.
That in turn implies that the Luperci were now controversial, one of the
symbols of a patrician ideology which claimed credit for Rome’s military
success but was resisted by its opponents as arrogant and tyrannical.

VI

The half-century from the Caudine Forks to the defeat of Pyrrhus was a period
of intense conflict, crisis, and innovation. It is unrealistic to try to separate out
religious, political and military elements. Just as Appius Claudius’ censorship in
312 revolutionised both the cult of Hercules and the political influence of the
urban plebs, so that of his enemy Fabius Rullianus in 304 honoured the new
military role of the young aristocrats and also, 1 suggest, radically reorganised
the Lupercalia.
Two teams of Luperci were created, named after aristocratic gentes and
consisting of young cavalrymen, dismounted but stripped for action. The loin­
cloths that were the uniform of the celeres enabled them to show off their
physique without the undignified nudity of the traditional Luperci. The explic­
itly sexual associations of the Pan cult were minimised, Inuus, Faunus and Pan
himself being reinterpreted as divinities of war. The ritual run, if we may trust
the aetiology preserved in Plutarch, now took place with the waving of
swords. The Lupercal cave itself must have been involved in the ambitious
restructuring of the hillside below the new temple of Victory, which was dedi­
cated the year after Rullianus’ great victory at Sentinum in 205.
Such innovations would not please everyone, and when the victory in 295
was followed by three years of continuous pestilence,111 we can reasonably
guess that Rullianus’ enemies attributed it to the anger of the gods. This was
the epidemic that caused the Romans to summon Asklepios (Aesculapius)
from Epidauros in 292.112 The word used of it is hies, plague, a contagion of the
sort that lustration rituals like the Lupercalia were designed to prevent,113 and it
seems to have been sent by the god of the Lupercal himself.

110 Frontinus 1.2.2; cf. Livy 9.36.2, AL Fabium, Caesonem alii... tradnnt.
111 Valerius Maximus 1.8.2 (triennio before 293), Zonaras 8.1 (prophesied in 297).
112 Ovid Metamorphoses 15.622-744, Valerius Maximus 1.8.2, Livy 10.47.6-7, Plutarch Qnaes-
tiones Romanae 94 (Moralia 286d), Lactantius Inst. din. 2.7.13, De uiris illnstribus 22.
113 Lues: Ovid Metamorphoses 15.626, Lactantius Inst. din. 2.7.13. Purification: see above, nn.
47 and 55.
 76 UNWRITTEN ROME

The first-century AD medical writer Soranus, discussing the choking symp­

toms of nightmare, insisted that Incubus/Ephialtes was not a god but a disease,
of which the symptoms were pallor and emaciation.114 He quotes a Greek
authority ‘Silimachus the Hippocratic’ (possibly Callimachus, and if so third
century bc), as saying that many people in Rome had died of this disease,
incurring it through contagion as if in plague (uehit Itte). The reference must
surely be to the plague that caused the embassy to Epidauros, of which Ovid
mentions pallentia corpora among the symptoms.115 When Asklepios came to
Rome, he stopped on the way at Antium, in Inuus’ territory; he was finally
installed on the Tiber island, where a century later a temple was also set up to
Faunus. Not only that, but Ephialtes is named by the late medical writer Orib-
asius as ‘the sacred interpreter and minister’ of Asklepios.116 It seems that the
Romans, and Asklepios on their behalf, were going out of their way to mollify
an offended god.
Fifteen years later, Rome was affected by another epidemic. This time the
problem was miscarriages and still-births, both animal and human. The date,
-76 BC, is given by Orosius, and a fragment of Livy reveals that this was the
occasion when the Lttpercaha flagellation ritual was introduced. According to
ugustine,] Asklepios was no help; he announced that gynaecology was not his

AVhat happened can be read in Ovid, in a narrative transposed to the time

omulus. Juno, goddess of childbirth, was consulted; she announced, with
allusion to the name Inuus, that ‘the sacred he-goat must enter the
women of Rome . Pan is the goat-god, and we know that the Luperci, minis-
o an Lykaios, were called crepi, a corruption of capri, he-goats.118 One
o vious way to fulfil the command would be for the women of Rome to
su mit to sexual penetration by the Luperci. However, an alternative interpre-
jc p ]W1S ound and adopted. When the Luperci made their goat-sacrifice on
with V*Ct*ni was to be skinned and the pelt cut up into thongs,
tie women of Rome must allow themselves to be whipped. In

115 Caelius Aurelh’ ln,Cacllus Aureliamis De morbis chronicis 1.3.55-6.

(Polybius 19 I ^""rbis chronicis 1.3.57, Ovid Metamorphoses 15.627. For Callimachus
live emendation is T ^'Sl' etc)’ see von Staden 1989.480-3. The alterna-
116 Antium- Ovirl M 7siniac*lus > date uncertain but pre-Neronian (von Staden 1989.564).
Faunus:’ n. 10 ab 7 15'719~28- Valerius Maximus 1.2.8, De uiris illustribns 22.3.
’EcptaXinr . Plualtes: Oribasius Synopsis 8.2 (OVK EOTIV 6 KaXouptVO;
Oepdjtwv AoKknirtoO)'rh’. a)Xa ” TlS V°°°5 6 wtotpiftnc; Ifpb; Kai
renamed ferentarii (\r n cc eres were evidently portrayed in Asklepios’ temple, but now
117 Orosius 4.2 9 Livv fr M 7’57’ unannoured cavalry).
Holleman 1974 9(7/- “ Weissenbom (n. 47 above), Augustine City of Cod 3.17. See
118 Ovid Fasti 9 49? e, ;."'n°Vatlon already emphasised by Otto 1913.183-5.
“ (441 for ‘^cerhircns inito’). Crepi: n. 2 above.
 THE GOD OF THE LUPERCAL 77

that way, if the skin was broken, the he-goat would have ‘entered’ them.
According to Festus, the goatskin thongs were called ainiculum Itmoiiis; no
doubt ‘wearing Juno’s little cloak’ was a euphemistic fonnula for offering
yourself half-naked to the lash. Plutarch and Juvenal tell us that by the second
century ad ladies used merely to hold out their hands to the Luperci ‘like chil­
dren in school’; but Ovid’s phrase terga percutienda dabant suggests that originally
it was more serious than that.119
All the literary evidence makes it clear that the Lttpercalia ritual was an occa­
sion for laughter and enjoyment; the words used are paidia, sdos, hilaritas, lusus
and lasciuia. Naked young men, their bodies oiled or smeared with mud, ran
about striking anyone who got in their way. The fertility ritual introduced in
276 made the fun more brutal, and no doubt more exciting for the onlookers:
the young women were no longer to run away, but to offer themselves to a
flagellation that was a metaphor for sexual union. It was a female divinity who
demanded a carnal remedy, and a male interpreter of the divine will who had
it commuted to flagellation. No doubt the husbands of the women of Rome
were not eager to have their wives literally impregnated by the Luperci.
It is important to remember the sheer sexiness of the Lttpercalia. The young
men themselves were objects of desire, which is why Augustus would not
allow beardless boys to take part in the run. And from 276 onwards, the ritual
encouraged young married women to bare their bodies in public.12" One can
see why it was such a popular spectacle.
But the reason for its introduction was specific to the third century BC. The
secession of the plebeians in 287 shows that social and political tensions were
still acute; a m' and dangerous war with Pyrrhus began with costly defeats in
280 and 279 When pestilence returned, despite Asklepios, in 276, Juno
evidently demanded a return to traditional ways. A compromise was reached,
and once more the ritual of the Lupercalia was reformed. When the flagellation
ritual was introduced, the necessary aetiological explanation was again attrib­
uted to Romulus, attached this time to the story of the Sabine women.121 As in
304, innovation was disguised as a return to ancient custom.
This reconstruction is, of course, in the highest degree speculative. But
enough evidence survives to make a reconstruction possible, to explain the
phenomena in a coherent way, provided that we entertain the possibility of
development, controversy and change, and reject the premise that ritual must

119 Festus (Paulus) 75—6L, Plutarch Caesar 61.2 (‘shaggy thongs’, as at Antony 12.1), Juvenal
2.142; Ovid Fasti 2.445-6. There may have been an Arcadian precedent; see Pausanias
8.23.1 for the flagellation of women at the Dionysos festival at Aiea, on the instructions of
Delphi.
120 Suetonius Dinns Augustus 31.4. Matronae nudato corpore napnlabant: ‘Gelasius’ Adncrsns Ainlro-
niaclnnn 16 (CSEL 35.1.458).
121 Ovid Fasti 2.431-4.
 78
UNWRITTEN ROME

remain essentially the■ same despite fundamental changes in the community

whose prosperity it is ;
meant to guarantee. In times of very rapid change, like
the late fourth and early third centuries BC, far-reaching social and political
developments must have affected myth and ritual along with everything else.

VII
, 1 time Rome was convulsed by revolutionary change on that scale was
in the first century BC, and it is no surprise to find that the next development
in the ritual of the Lttpercalia takes place precisely then.
th nionS t’le sPecial honours granted to Caesar the dictator late in 45 BC was
e creation of a third group of Luperci, the Iidiani.'22 Their leader at the next
p rca a was the consul himself, M. Antonius, who used the occasion to
aesar the crown. Cicero was disgusted. Running as naked Luperci was all
ry we or young men, but quite incompatible with the dignity of a consul;
h h rl 6 I March, Cicero never tired of taunting Antony with the time
run into the Comitium audits, unctus, ebritts, and mounted the Rostra in
an attempt to make Caesar king.123
The Senate in 43 withdrew Caesar’s funding from the Luperci, and may
a o shed the Iiiliani; but the historian Aelius Tubero, writing in the thir—
, BC.’ c^ear^y implies three groups in his account of the origin of the ritual,
an it is ikely enough that the Triumvirs restored them.1-1 Suetonius tells us
f C ]<i were one of the ancient ceremonies that Augustus restored
° ivion, and Augustus himself in the Res gestae includes rhe Lupercal
among buddings he constructed. That probably implies a reorganisation analo-
& 1/°. one m 304, though all we know for certain about it is his
Pr°Th K10n °f b°yS he^0re a8e of puberty from taking part in the run.125
the collCOnCern SeXUal moratity is very characteristic. In the late Republic
Cice ° f^ ^u^erc' had evidently had a somewhat equivocal reputation;
cero, or instance, did not approve of his nephew becoming a member.'126 ’'
gustus re-emphasised the connection with the equester ordo,127 but seems to

122 Cassius Dio 44.6.2, 45.30.2, Suetonius Diuus Julius 76.1.

123 Cicero Philippics 2.84-7, 3.12, 13.17, Cassius Dio 45.30.1-5. missu, 1.80.2 =
124 Cicero Philippics 13.31 and fr. 19 (Nonius 418L); Dionysius of Halicam
Tubero fr. 3P.
125 Suetonius Diuus Augustus 31.4, Augustus Res gestae 19.1. r'Uc;nniis (1LLRP
126 Cicero Ad Atticu.n 12.5A. Late-republican Luperci included Ge^nius C^PP p,
696, Pliny Nat. Hist. 34.11), A. Castricius Myriotalentus (C1L 14..1US), I -Balbus
Surus (AE 1968.33, Panciera 1986); also M. Caelius Rufus and L H
(Cicero Pro Caelio 26), though the latter claimed to be a stnct moralist (ibid. -= - E
127 C/L 6.31200.b.2.5-9: equestrian honours to Drusus atf the Lupercal
trausuectio (15 July). Luperci as essentially equestrian. CIL 6—lou, .
1924.41 (second to third century ad).
 THE GOD OF THE LUPERCAL 79

have introduced attendants to minimise any danger to the young equites’ moral
well-being, and no longer required them either to wear goatskin or to brandish
goatskin thongs.
The gravestone of Ti. Claudius Liberalis in the Vatican Museum shows us
how a young eques who died at sixteen years of age was remembered in the
early Empire: on one side, Liberalis riding in the transuectio parade, attended by
a man with a flag; on the other, Liberalis as a Lupercus, wearing a substantial
loincloth certainly not of goatskin, carrying a whip, not a goatskin thong, and
escorted by two attendants.128 The dignity and moral probity of the equestrian
order are conspicuously on display, but what has happened to the hilaritas and
lasciuia of the republican ritual? The evidence of Plutarch and Juvenal, that in
the second century AD all the ladies had to do was to put out their hands to be
struck by the Luperci, suggests that the ritual’s traditional erotic charge had been
deliberately neutralised.
The next great period of crisis and innovation in Roman history—one
which involved the ultimate revolution in the Romans’ religious outlook—
was the late third and early fourth centuries AD, from (let us say) the building
of Aurelian’s wall to the founding of Constantinople. We have two images of
the Lupercalia from this period, and very astonishing they are.
The first is on a mosaic floor from Thysdrus in North Africa; there is a
scene for each month, and February is represented by the Lupercalia. Here the
Lupercus is not running but standing; he is wearing a substantial apron, and
raising a whip to bring down on the body of a woman who is being held in
place by the two attendants. She looks back over her left shoulder at where her
dress is raised to bare her body for the blow.12'' The same scene in greater detail
appears on a late third-century sarcophagus from Rome [fig. 16],130 Here the
hieratic pose of the Lupercus and the humiliating exposure of the woman are
even more explicit. The lady wears bracelets and an elegant coiffure, and the
Lupercus carries out the rite with a very conscious dignity. His equestrian rank
is symbolised by the man on the right carrying the uexillum. Behind him,
shouldering the tree-branch which is his regular attribute, appears Silvanus, the
god whose ancient grove by the temple of Saturn probably marked the ritual
conclusion of the Luperci's run.131 The Lupercalia are still just recognisable, but
fundamentally changed. This young equestrian is static and solemn, not naked

128 Illustrated and discussed by Veyne 1961), cf. Wiseman 1995.83, fig. 10. For the whip
(appropriate to a horseman), cf. Daremberg and Saglio 1896.1153-4. [See now Tortorella
2000.249; Tortorella’s article is the standard work on the iconography of the Lupercalia.]
129 Stem 1968, esp. 181—2 and pl. Ill, fig. 2; also in Holleman 1974.138 and Foucher 1976.278
[see now Tortorella 2000.252, cf. 245-6 for an analogous scene on a mirror in New York].
130 Schumacher 1968—9, Solin and Brandenburg 1980 [see now Tortorella 2000.253-4].
131 See p. 65 above; for the iconography ofSilvanus, see Dorcey 1992.17-19, with illustrations
2, 3 and 6.
80 UNWRITTEN ROME

1
s
6

§
s
5
2
z
i
£

£
g

~ ■-

V1

llI
g“
z <
■J g'

.«> <-»5
U-
 THE GOD OF THE LUPERCAL 81

but wrapped up tightly as far as the chest. The high spirits have disappeared,
replaced by a cold-blooded formality.
To us, this scene is repulsive. But it was evidently not offensive at the time;
when the Lupercus’ sarcophagus was re-used for the burial of a Christian lady
about AD 340, his portrait and inscription were removed but the flagellation
scene was not touched. Moreover, the Luperci in the late Empire seem to be
rising in status, with even senators appearing among them from the late second
century ad onwards.1'2 One such was Crepereius Rogatus, who early in the
fourth century created an elegant triple-naved larariuin, its design very like a
Christian basilica, on his property on the Viminal slope.133 The apse was deco­
rated with mosaics, showing the she-wolf and twins behind the altar, a male
figure with a spear at the top of the arch, and on the pilaster at each side a
Lupercus carrying a whip.134
If this was the headquarters of a sodalitas of Luperci, as Lanciani thought,
then they met in formal and luxurious surroundings, surroui and the god they
worshipped was evidently Mars. Inuus, Faunus, Pan Lycaeus, Incubo and
Ephialtes were \\ ell known to the learned (much of our evidence about them
comes from this period), but perhaps rituals involving the elite demanded
something mr.T dignified. Faunus, however, was still respectable in distant
Britain, where ie name is prominently featured on the silverware of the Thet­
ford treasure: and Macrobius shows how Pan and Inuus could be reinterpreted
as allegory ic ... .p. the sensibilities of serious persons. 135
The god or r . Lupercdl had to recede into the background, but the Luper-
caliu remains ... important element in the civic life of Koine. When a late
fifth-centm v , , probably Felix 111 rather than Gelasius—tried to abolish it,
a senator c iii ■1 . i omachus complained that the city was being deprived of
its protecti . i ■. pestilence and famine.13" The pope returned to the attack,
in that spirit > milky to the ’neutral ground’ of secular observance which
Robert M.n1 m l.as recently documented. 13 The episcopal polemic reveals

132 E.g. (.7/ >>. I3‘>“. (,.147-E (>.31716, 11.2106 (a Fabius from Clusiuml). If Lactantius was
right tli.it tlics sometimes wore masks (n. 82 above), that may have been to protect their
dignity.
133 Lanciani 189|, ami /•bmi.i urbis Romae (1893-1901) sheet 23: on the line of Via Cavour
below S. Maria Maggiore; Duliere 1979.255—9 and fig. 128 [see now Tortorella 2000.254—
5].
134 Vatican MSS Lat. 2733.llf.285: ‘in utroque ipsarum [i.e. parastatarum] latere dicto opi>ere
[i.e. musivo] duo midi luperci efficti erant, gestantes ferulas intortas.’
135 Johns 1986; Macrobius Sutuniuliu 1.22.2—7.
136 ’Gelasius . Aducrsus AudroiHM/mui 3. 13, 23 (CSEL 35.1.454, 457, 460—1); Duval 1977, esp.
246—50 for the date (suggesting c.488). [See now Cameron 2004b.512-13: ‘What Gelasius
saw was not a genuine survival but a picturesque revival.’]
137 Markus 1990.131-5. esp. 133: 'The attack on the Lupercalia is not so much an attack on
“remnants of paganism" as on traditions of Roman urban living.’
 82 UNWRITTEN ROME

that the flagellations still took place, but as a performance by ‘vile and common
persons of the lowest class’, to the accompaniment of obscene songs.138 The
pope challenges the senator and his fellow traditionalists: if the ritual is so
important, why don’t you do it the old way and run around naked yourselves
with your little whip?139 It is likely that the rite was suppressed not long after
that.
In Constantinople, on the other hand, the Lupercalia were still celebrated in
the tenth century, in a curious performance at the circus-races, where the
charioteers dismounted and ran on foot, using the reins on each other.14" A
springtime hymn was sung, which shows how the Lupercalia (15 February) had
been assimilated to the date for the start of spring and the coming of the
genitabilis aura Fauoni (7-8 February).141 An even more harmless association was
with the beginning of Lent; the Byzantine Lupercalia ritual was defined as
tuakellarikoii, to do with the eating of meat, no doubt to mark the last day
(before Easter) when it was allowed.142
That sort of tolerantly creative reinterpretation contrasts strongly with the
hostility of the Roman Church. One of the things that made the difference
was clearly the flagellation of women, which was giving trouble again a thou­
sand years later. In 1481 the Carmelite friar Baptista Spagnuoli of
Mantua—‘good old Mantuan’ to Holofernes in Love's Labour’s Lost
published his long poem De sacris diebus, which did for the Christian calendar
what Ovid’s Fasti had done for the pagan one. His fourth item under February
(after St Agatha on the 5th) concerns ‘the evil custom of Shrove T uesday’143
nothing less than a revived Lupercalia, with youths running about the town in
masks, taking advantage of their anonymity to manhandle the young married

138 Aduersus Andromachum 16, 19-20 (CSEL 35.1.458, 459): nites triuialesque personas, abiectos et
infmios. The cantilenac were evidently a charade, confessions of sexual misconduct to justify
the whipping.
139 Ibid. 17 (CSEL 35.1.458): ipsi celebrate more maiorum, ipsi cum resticulo nudi discurrite. Resticulo
(cf. n. 128 above) is Guether’s emendation for the MSS ridiculo; for discurrere, see n. 54
above.
140 Constantinus Porphyrogenitus De caerimoniis 1.79 (70), 82 (73); Duval 1977.223—13. Does
pvioyoOvTEt; CtkXt]kovg describe a ‘fossilised’ derivative of the thong-wielding Lnperci of
classical Rome (n. 2 above)?
141 Ibid. 1.82 (73); Ovid Fasti 2.148, Pliny Nat. Hist. 2.122, Columella 11.2.15; Duval
1976.264-7. Favonius: Lucretius 1.11, Horace Odes 1.4, etc; for the connection with
Faunus and the Lupercalia even in Horace’s time, see Barr 1962.
142 Duval 1977.226-7.
143 Mantuanus 1481, de carmsprhdi mala consuetiidine: Nam iuveties istis facit insanire diebus | Pan
Deus Arcadiae quondam: totasque per urbes | currere et acceptis facies abscondere lands. | Est pudor
in facie: facies t’e/ata pudorem | ora tegit: Serins est paiddum metuitque idderi. | Cuncta sub ignotis
pctulantia multibus audet | quae ablegat gravitas et quae proscribit bonestas. | Per fora per vicos it
personata libido: | et censore caretts subit omnia tecta voluptas: | nee minium palmas sed membra
recondita pulsat: | perque domos remanentJoedi vestigia capri.
 THE GOD OF THE LUPERCAL 83

women, bare their bodies and whip them. The same licensed sadism is attested
in Saxony in the seventeenth century, again as part of the ‘carnival’ permissive­
ness of Shrove Tuesday.144
Northern Europe may also provide a somewhat more spectacular manifes­
tation of the god of the Ltipercal. Behrend Pick’s derivation of Mephistopheles
from Opheles, one of the names of Ephialtes,145 may not be right—in the Faust-
buch of 1587 the name is spelt ‘Mephostophiles’146—but even so, the horns and
goat’s feet of Pan and his equivalents have made a contribution to the iconog­
raphy of Christian Europe which is literally diabolical.147

144 Paul Heinrich Tilemann, Couuueritatio liistorico-uioralis et juridica de eo quod justiuu est circa
nuditateni (1692), as cited in Mannhardt 1904.225-6: ‘Tempore quadragesimal! im Facht-
nacht mulieres sibi obviani factas inhonesto ioco interdum denudatis posterioribus virgis vel
etiam herba aliqua pungente feriunt’ (no mention of masks). Mannhardt (1904.252-6)
otFers many examples of striking on the hands (cf. n. 119 above).
145 Pick 1917 = 1931.105-12, citing the coins showing Ephialtes Epopheles at Nicaea and
elsewhere (n. 70 above).
146 So too in Marlowe: Bevington and Rasmussen 1993.127, 211 (the apparatus criticus reports
‘Mephostophiles’ for both texts).
147 Bemheimer 1952.93—101; cf. Merivale 1969 [and now also Boardman 1997].
 CHAPTER FIVE

Liber: Myth, Drama and Ideology

in Republican Rome

I
By name and by nature, Liber is the god of freedom. His gift of wine frees men
from cares, his mysteries free the soul from mortality, his power frees the seed
in sexual union;1 one of his many Greek names is Lyacus, ‘the releaser’
(Analog from XllElV).2 Though many explanations were offered by ancient
sources to account for his name, the simplest and most obvious was an ideo­
logical one: Liber a libertate3 Political freedom, libertas, was the defining quality
of the Roman Republic, achieved by the expulsion of Tarquin and under
threat ever after.4 It can hardly be accidental that Liber’s festival (the Liberalia)
was held on 17 March, just two days after the magistrates of the Republic
entered office and twenty days after ‘the flight of the king’ (7^<;'f'-
But the freedom of the Roman People was always potentially in conflict
with the dignitas of the principes tiiri.6 Roman politics, like Athenian, could be
conceived as a struggle between the many and the few,7 and what the many called
libertas the few redefined as licentia, anarchic and dangerously ‘un-Roman .8

1 Respectively: Seneca De tranquillitate animi 17.8, Servius on Virgil Georges 1166, Varro
"tiquitatcs rerum diuinarum fr. 93 Cardauns; Maltby 1991 337
2 E'ymologicum maj-num 571.18; Bnickmann 1893.87-8.
3 \y/-V*USl°|1.^ene^ 4-638, cf. 3.20 (causa hbcrtatis), 4.57 (Lyacus ... ante urbis libcrtatis est deus).
4 Wirszubski 1950; Hellegouarc’h 1963.542-59.
5 Mastrocinque 1988.46-7. Consuls enter office on Ides of March: Livy 26.26.5,
, >■ ’ etC on|y festivals marked in the ancient calendar between Regifugiiuii
M t, er "'k t',e t'V° horse-races in honour of Mars on 27 February and 14
th.- e ‘ound,1tion'story of the Republic, the expulsion of Tarquin was followed by
cL^n, SHt,On °f,hlS ProPert>' and its dedication to Mars (the Campus Martius).
6
Tadms^S'r LiVy 4’61’’ C£ 7’33’3’ diSSOdabileS ' ■ ■ M
7 felhm, 8-2- Multi, plurimi: Cicero De republica 2.39, 6.1. Pauci: Sallust
1.55.23^48 ?8~° 7’ 39J’ Bel1'"" I“s“r,hi"""‘ 31’ 31-9> 3L2°- 41-7- Histories 1.55.12,

8 E g. Cicero Pro Flatmo 16, De republica 1.68, 3.23; Livy 3.37.8, 3.53.6, 3.59.4, 23.2.1; Phae-
dnis 1.2.2-3; Tacitus Dialogus 40.2. Cf. Cicero De domo 110—11 and De legibus 2.42 on
Clodius’ shriniie of Libertas (teniplum Liccntiae).
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 85

Not surprisingly, therefore, it was when the Senate’s authority was strong, in
the early second century BC, that the ‘licentious’ aspects of Liber—the so-
called Bacchanalia—were stamped out throughout Italy in a ruthless police
action.9
The fluctuations of ideological attitude, of which the ‘Bacchanalia affair’ was
a particularly violent example, make it pointless to attempt a synchronic
account of Liber in the Roman Republic. Our literary sources almost all post­
date 186 bc, and naturally reflect the Senate’s hostile view. A tiny glimpse of
earlier attitudes is provided by a line of Naevius quoted in Festus’ dictionary:10

Liberalia Liberifesta, quae apitd Graecos dicuntur Aiovvoia. Naeuius: Libera lingua
loquiiuur ludis Liberalibus.
Liberalia: the festival of Liber, called Dionysia by the Greeks. See Naevius: ‘At
the Liberalia games we enjoy free speech.’

We know from Ovid that the Liberalia had once been hidi." Naevius’ refer­
ence may suggest that they were theatre games (Indi scaenici), but a line of
Plautus implies that the women celebrating the festival were also a part of the
performance, and from Livy’s hostile description, which may well go back to a
second-century' BC source, we may infer that ritual and stage spectacle over­
lapped.12
Before Naevius, there is no contemporary literary evidence—but that does
not mean that there is no contemporary evidence at all. A crucially important,
and much neglected, source of information is the corpus of engraved bronze
cistae and mirrors dating from the late fourth and early third centuries BC.
These artefacts are usually called ‘Praenestine’, because of the accident that
Praeneste’s cemeteries have been more thoroughly explored than those of any
other Latin city; but since one of the finest of the cistae was certainly made in
Rome, this material should be understood as illustrating the story-world of
Latium in general, just as contemporary vase-painting illustrates that of Etruria,
Campania and southern Italy.13

9 Livy 39.8-19 (licentia at 39.13.10), 1LLRP 511; Cicero De legibus 2.37, Valerius Maximus
6.3.7 (exercise of seueritas): Pailler 1988.
10 Festus (Paulus) 103L = Naevius fr. 113R; Wiseman 1998.35-43. [According to Csapo and
Slater 1994.209, ‘it is notable that there are no games [at Rome) for Dionysus or his Latin
equivalent.’!
11 Ovid Fasti 3.784-6 (later transferred to the Cerialia on 19 April).
12 Plautus Casina 980 (nunc Baccae nullae ludunt); Livy 39.10.7, 39.13.12-13, cf. 39.15.7
(concession Indian ac Ltsciniain); Cazanove 1983, esp. 103-13. Livy’s source: Wiseman
1998.47-8.
13 Battaglia and Einiliozzi 1979 and 1990: no. 68, the ‘Ficoroni cista' carries Novios Plautios’
‘made in Rome’ inscription (ILLR.P 1197). Vase-painting: Beazley 1947; Martelli 1987;
Trendall 1967 and 1987; Trendall and Cambitoglou 1982. A detailed comparative study of
the iconography of the Latin engravers and the Etruscan, Campanian and South-Italian vase­
painters is urgently needed.
 86 UNWRITTEN ROME

As on the vases, so on the engraved bronzes Dionysiac imagery is ubiqui­

tous. It is clear that all over Italy, Latium included, the iconography of the
thiasos was perfectly familiar in the fourth and third centuries BC, as it had been
in the sixth and fifth.14 When the consul in Livy’s narrative of 186 BC warns the
Roman People against outlandish foreign rituals, we should not take his words
as reliable evidence. If something like that really was said at the time, it can
only have been a disingenuous pretext for interference with a cult that had
become politically undesirable.15 Liber and libertas had to be controlled.

II
A cista in Berlin, no. 5 in the corpus [fig. 17], names LEIBER in a group of
gods witnessing what seems to be the initiation of the young Mars at the Liber­
als.lb The combination of the familiar iconography of the gods and the
mysterious scene of young Mars kneeling on an amphora under the sign of
Cerberus may serve as a reminder of how little we know; we need not expect
everything portrayed on these artefacts to correspond to what our literary
sources tell us about two or three centuries later. (It would be good to know,
for instance, what Liber’s right-hand gesture means. Is he warning Apollo off,
or is just the artist’s way of filling the space?) Another Berlin cista, no. 4 [fig.
18], portrays Liber in a younger guise, with Venus and Adonis(?), one of the
Dioscuri, Jupiter and Diana.17
A cista in Karlsruhe [fig. 19, p. 89] shows Liber and the satyrs with a female
figure. Since she and Liber balance each other in a very symmetrical composi­
tion, it is tempting to identify her as Libera.18 Or she may be Semele, mother
of Dionysus, brought back from the Underworld by her son to become the
goddess Thyone, whose Latin equivalent is Stimula; the Incus Stiumlae, centre

14 Early fifth-century antefixes etc: Cazenove 1986, esp. 185—90 for bibliography and discus­
sion. Note that the satyrs’ female partners were more often nymphs than maenads: Hedreen
1994. The Phase 1IB vase with the graffito enoin from tomb 482 at Osteria dell Osa (Gabii)
attests Latin familiarity with Dionysiac ritual as early as the eighth century BC: see Peruzzi
1998.81—90 [however, the reading may be cutin: Ridgway 1996J.
15 Livy 39.15.1-3, 39.16.8-9; Beard, North and Price 1998.1.93-6.
16 Suggested by Pairault Massa 1992.163—4; other suggestions in Battaglia and Emiliozzi
1979.52-4 and 61 (‘nota aggiunta’).
17 Battaglia and Emiliozzi 1979.49.
18 Cf Pliny Nat. Hist. 36.29 (a statue group of satyrs with Liber and Libera). Libera was some­
times identified with Persephone (Cicero De iiatura deortini 2.62, terrines 5.187, Dionysius of
Halicarnassus 6.17.2, 6.94.3, Arnobius 5.21), sometimes with Ariadne (Ovid Fasti 3.459-
516, Hyginus Fabtdae 224.2), sometimes with Venus (Varro Antiquitates renim diiiitiaruni fr.
93 Cardauns, CIL 8.15578). See now Jurgeit 1999.524, who identifies the scene as the
preparation of a bride by her girl friends, with Dionysus bringing his own bride Ariadne.
(But why are the satyrs so surprised? And which of the five young women is the bride?)
 5
A
m

K 3
-<

WgsWlMwSMpgSSsft X
0
>3
A
d_ I >Z
>
□
5
o
"13 5
3
te77.TKTI gs
CD

n
Fig. 17 Cista no. 5, in the Staatliche Museen, Berlin (Battaglia and Emiliozzi z
1979.50-54): the gods witness Minerva’s anointing of the young Mars (third
g
figure from the right, beneath Cerberus in the upper border). i
00
 00
00

c
z
5
77
H
m

7J
0
£

Fig. 18 Cistci no. 4, in the Staatliche Museen, Berlin (Battaglia and Emiliozzi
1979.48-50): the young Liber (second figure from the right).
 m

3
H

O
>
3

>
0
5
m

5
3
Fig. 19 Cista no. 22, in the Badisches Landesinuseum, Karlsruhe (Battaglia and
Emiliozzi 1979.95-7), with nineteenth-century additions omitted (Jurgeit s
1992): Liber and Libera(?) escorted by satyrs. CD

n
z

i
00
xO
 90 UNWRITTEN ROME

of the Bacchic rites in Rome, was also known as ‘Semele’s Grove’.19

This cista is also notable as the most elaborate example of a frequently recur­
ring iconographic motif, women naked at the washing-basin.20 Mauro
Menichetti, in his important book on the cistae, sees this as an allusion to the
‘feminine paideia’ of an aristocratic society, celebrating the beauty and seduc­
tiveness that lead to marriage,21 and it is certainly true that this scene seems to
have something in common with the women’s festival on 1 April, as described
by Ovid.22 We shall return to this motif; for the moment, however, I want to
concentrate on Liber and his followers, particularly the satyrs.
Cista no. 72, in the Villa Giulia [fig. 20], shows a wide range of satyr­
types.23 Note in particular the old fat satyr with the long beard, riding on a
goat; he reminds us of old Silenus in the Augustan poets, the father of the
satyrs, on his hollow-backed donkey.24 In the centre is a sober, mature satyr,
respectably clothed, playing a pipe; he too may remind us of Silenus in the
literary texts, this time as the source of wisdom and sage advice. 25
But neither characterisation of Silenus is necessarily applicable in the fourth
century bc. It is true that Silenus as father of the satyrs is already a familiar
figure in Athenian satiric drama, but Herodotus continues to use OlXqvd^ in
its original generic sense and not as a proper name. name. 26 Alcibiades in Plato s

19 Semele: Diodorus Siculus 4.25.4, Apollodorus 3.53, Hyginus 1 py/hjjns 3.177,

1997.62-4. Thyone Ttapa rd Oueiv, o EOtiv oppav: Scholiast'diuinun'ni fr. 13<» Cardauns.
Suda s.v. Thyone. Stimula from stiinulare: Varro Aiitiquitatcs renitn i.
Lucus: Livy 39.12.4, Ovid Fasti 6.503, CIL 6.9897; Cazenove 1986.56-66.
20 See also cistae nos. 24 (Leningrad), 67 (Providence RI, KI, authenticity ii)“
authenticity disputed),• 74 and 75
(both Villa Giulia). See below, pp. 111
Ill and 115 on nos. 50 50-51
—51 (and n b rolctta delh
21 Menichetti 1995.57: ‘Tutte queste raffigurazioni che pongono in prim
donna alludono in tin linguaggio grecizzante alia paideia femminik della c a,-cede alia
trainite cui si acquisisce quel potere di seduzione dell uomo inedianti i qn. 6 ■ njc aj
sfera di matrimonio.’ Ibid. 19: ‘La donna visibile sui grandi fregi muse no . di
una banalizzante prospettiva deH’oiL'OJ, ma e invece investita da una sa onz ,,snJissione di
una societ.i aristocratica che individua nella donna un veicolo pnvi egiato e

22 MM 4.13^62^^ matres numsque); Menichetti 1995.62-3 (and see now pp. 140-

23 Also an ithyphallic henn (Priapus?) on the altar; the young Pan, in human form'
horns in his hair, identified by the throwing-stick (pedum); and a goose u it a fourth
as on the base of one of the phallic pillars of the choregos monument at e os a . por
century). For the Dionysiac context of the Delos monument, see Co e , e‘sp
a goose as deliciae Priapi, see Petronius Satyricon 137.1.]
24 The locus classims is Ovid Ars aniatoria 1.541-8; also Fasti 1.399, 3.749, ... h it
25 Cicero Tusculanae 1.114, Virgil Eclogues 6.13-30; also Theopompus FGr o .
is not clear whether he used Silenos as a proper name or a description (cf. Hero otus
Olkqvog, Xenophon Anabasis 1.2.13 Tdv OCtTVpOV). A serious, clothed, elderly satyr ppc
on cista no. 96 (Villa Giulia), among respectable women in the gyitacceum; see also t y
playing Silenus and the Silenus with the mirror in the Villa of the Mysteries frieze at on p
26 Aeschylus Diktyoulhoi 805, Euripides Cyclops 82-4, 272, 431 etc; Herodoms 7._6.3, o
MapcfUEO) aOKOg. Seaford 1984.6-7.
 £
m

3
-<
I
0

>Z
o
□
o
5
5
s
TO

n
Fig. 20 Cista no. 72, in the Villa Giulia Museum, Koine (Battaglia and Emil- z
iozzi 1990.239—14): various satyrs.

SO
 92 UNWRITTEN ROME

Symposion likens Socrates to ‘those silenoi in the sculptors’ workshops which

the craftsmen portray as holding pan-pipes or auloi, and when you take them
apart you see they have statues of gods inside’; but since he then goes on to say
that Socrates is like ‘the satyr Marsyas’ even though he doesn’t play the atilos, it
is clear enough that he means not statues of Silenus but statues of silenoi in
general, of whom Marsyas was one.27 As for fat Silenus’ donkey, in Ptolemy’s
great Dionysiac procession in 270 BC there were five troops of donkeys ridden
by satyroi and silenoi alike.28
So it would be rash to identify either the old fat satyr or the sober one as
Silenus. On the contrary, the young dancing satyr at the far left of fig. 20, and
the one at the far right with the short beard, carrying a wineskin, may have a
better claim to the name. On cista no. 66, at Vassar College [fig. 21], a frag­
mentary scene showing Jupiter or Dis Pater, Diana, the Dioscuri and Victoria
includes at the far right a short-bearded satyr carrying a wineskin; he is named
as SILANOS.29 And on a cista in New York, no. 45 [fig. 22], the name
SILANVS is attached to the young satyr dancing at the far left with a naked
young woman, and playing kottabos at the same time.
On the lid of no. 45 a grossly fat old satyr, whom first-century BC Romans
might have identified as Silenus, is given the appropriate name EBRIOS [fig-
23, p. 95], He wears a crown of a veiy distinctive type, which appears again on
a mirror in the British Museum [fig. 24, p. 96], showing a mythological scene
involving Telis (Thetis?), Ajax and Alcumena.3" At the right, the little fat satyr
wearing the crown, and making good use of a drinking-bowl, may be Ebrios
again.
The same sort of crown is worn by a different sort of ‘satyr’-figure at a
drinking-party on a mirror in Baltimore [fig. 14, p. 67 above). He has a
straggly triangular beard and is evidently playing kottabos—or perhaps dropping
wine on the drunken figure sprawled in front of him. His companions are a
young woman, naked to the waist, playing the pipes, and a naked young man
(perhaps Pan, to judge by the horns in his hair), blowing on a conch-shell.
Unlike them, he is wearing a tight-fitting hairy suit, the sleeve visible at the left

27 Plato Sy/nposion 215a, 216d, 221d (silenoi); 215b-e, 216c (Marsyas); cf. Herodotus 7.26.3
(previous note).
28 Athenaeus 5.200e = Callixenus of Rhodes FGrH 627 F2 (p. 173.7-8); ibid. 197e, 197f-
198a, 199b (pp. 168.10—11, 168.21—5, 170.17—18) for generic satyroi and silenoi, 199a (p.
170.12) for a singular Silenus.
29 For the inscriptions, including [?DIES]PATER and [PJORLOUfCES], see Battaglia and
Emiliozzi 1990.202-4. They draw attention to the fact that the ‘thunderbolt’ in ‘Jupiter’s’
hand is ‘stilizzato in maniera insolito’. I suggest that it is not a thunderbolt at all but the edge
of the god s cloak, that his hand is resting on the ana of his throne, and that the inscription
could more naturally be restored as (D/SJ PATER.
30 I airault Massa 1992.168-70, Menichetti 1995.120: Ajax in the Isles of the Blessed, receiving
the amis of Achilles?
 'A 7
Ui
UJ
m

3
-<
I

Q 0
>3
>
>Z
0
0

fL
o
5
3

CD

n
z
Fig. 21 Cista no. 66, in Vassar College, New York (Battaglia and Emiliozzi
1990.200-6): far right, a satyr called Silanos.
I
vO
UJ
Fig. 22 Cista no. 45, in the Pierpont Morgan Library, New York (Battaglia and
Emiliozzi 1979.146-50): second figure from the left, a satyr called Silanus.
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 95

Fig. 23 Lid of cista no. 45 (fig. 22): a satyr called Ebrios.

forearm, which must mean that he is an actor or performer of some sort.31

On a cista also in Baltimore, no. 1 [fig. 25, p. 97], the hairy-suited figure
appears again, in the company now of two girl musicians and a young satyr
with an amphora. What is striking about this scene is the way the Dionysiac
thiasos merges into the mythological scene of the death of Medusa, the birth of
Pegasus, and the pursuit of Perseus by one of the Gorgons.32

31 For an interpretation of this mirror and its companion-piece (in the Villa Giulia), see pp. 66-
9 above.
32 Battaglia and Emiliozzi 1979.42: ‘Si tratta di due scene ben distinte: 1) nascita di Pegaso dal
corpo di Medusa; 2) thiasos.’ But the scenes are not distinct: the upright thyrsus divides the
composition into two unequal halves, but the tliiusos extends beyond it (veiled woman
dancing). Perseus with the head of Medusa appears also on cista no. 76 (Villa Giulia), along
with Peleus and Thetis.
 96 UNWRITTEN ROME

v.
Fig. 24 Mirror in the British Museum: ‘Telis’, Aiax, Alcuinena and satyr.

Before trying to pull together some of these data, let us comp e

typology of Latin satyrs. On cista no. 14, in Chicago [fig. 26, p. JbJ, we
thiasos including at the left the young Liber with a thyrsus, and at t e nr>
short, plump, short-bearded satyr with a cup. He reappears in 3^nI' SteJ\
scene on a mirror of doubtful provenance in the Villa Giulia [tig. -7, p.
carrying a caduceus and offering a garland to a tall young woman who seems^
be feeling inside her dress with her left hand. He appears again
cista in the Villa Giulia, no. 100 [fig. 28, p. 101], sitting beneath a basin and
 s
m

3
-<
I
□
>
3
>
>
Z
□
□
m

5
2
s
Fig. 25 Cista no. 1, in the Walters Art Gallery, Baltimore (Battaglia and CD

Emiliozzi 1979.41—3): Bacchic procession with Medusa and Perseus. n

z
g
*

vO
 \O
00

c
z
$
-j
m

73
0
3

Fig. 26 Cista no. 14, in the Field Museum, Chicago (Battaglia and Emiliozzi
1979.76-7): Bacchic scene with plump satyr at far right.
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 99

feeling the sexual parts of a winged naked female.33 What stories these scenes
refer to, we do not know.
A satyr with the same physique, short and plump, but with a fuller beard
and moustache, appears in an early-third century context of some importance
for the ideology of the Roman Republic. This is Marsyas, as portrayed in the
forum of the Latin colony of Paestum, founded in 273 bc [fig. 29, p. 102] 34
Since part of a similar statue has turned up at Alba Fucens, founded in 303 BC,35
it is a natural inference that both colonies had put up replicas of the Marsyas
statue that stood in the Comitium at Rome, and that the Rome statue, for
which our earliest direct evidence is in Horace, therefore dated from the
f°l TheX-smm Marsyas has shackles on his feet. Filippo Coarelli brilliantly

suniiestedthe abolition of slavery for debt (hc.vhih), m 326 or 313 BC, as the
likely context for the erection of the Marsyas statue m Rome. In mytholog-
• ..1 however, the reference must surely be to Marsyas musical contest
with Apollo, illustrated on cisM no. 70 in the Villa Giulia [fig. 30, p 103]
According to what Herodotus calls ‘the Phrygian story , victorious Apollo had
the presumptuous satyr hung up and flayed alive, but some mythologists were
evidently uneasy about this gruesome tale; Diodorus, for instance says that
Apollo did it but later regretted it.- Sihus tahcus and the elder Pliny attest
whit we mav call rhe Italian story', that Marsyas escaped and fled to Italy,
where he became the ancestor of the Marsi.39
How did he escape? Surely he must have been freed by his patron Liber,
whom no chains could resist.- And that, as Servius explains in his commentary
on the Aeneid is the meaning of those statues in the./ora of Italian cities?1

33 Cf. Menichetti 1995.r>2, who ignores the wings and takes this as another example of the sjera
I'eiiiniinile and seduction leading to marriage. Pairault Massa 1992.165-6 more plausibly
describes the winged figures as ’due "geni” femminili’, and rightly insists on their unique­
ness.
34 See Coarelli 1985.98-9, figs. 13-17.
35 See Liberatore 1995.
36 Colonial copies: Charax of Pergannim FGrH 103 F31. Rome: Horace Satires 1.6.115-17
with Porphyrio and ps.Acro ud /<>c.: illustrated on denarii of L. Marcius Censorinus in 82 BC
(Crawford 1974.377-8, no. 363) and on the anaglypha Traiani (Torelli 1982.99-106,
Coarelli 1985.91-119).
37 Coarelli 1985.97-100 and 102-11. The Twelve Tables specified shackles (coinpedes') for
debtors (Aulus Gellius 20.1.45). Date of abolition: Livy 8.28 (326 bc), Varro De lingua Latina
7.105 (313 bc); Oakley 1998.688-91.
38 Herodotus 7.26.3, Diodorus Siculus 3.59.2-5, 5.75.3.
39 Silius Italicus 8.502—4, Pliny Nat. Hist. 3.108; Weis 1992.367, 377.
40 Euripides Bacchae 613-21, Ovid Metamorphoses 3.696-700; in a different sense, cf. Tibullus
1.7.41—2 (Bacchus et adflictis requiem mortalibus adfert, | crura licet dura compede pulsa sonenf).
41 Servius on Aeneid 3.20, 4.58; Veyne 1961.
100 UNWRITTEN ROME

in liberis ciuitatiblis simulacrum Marsyae erat, qui in tutela Liberi patris est. ...
‘Patrique Lyaeo’, qui, lit supra diximus, apte urbibus libertatis est deus; unde etiam
Marsyas, eius minister, est in ciuitatiblis in Jbro positus libertatis indicium, qui erecta
maim testatur nihil urbi deesse.
In free cities there was a statue of Marsyas, who is under the protection of Liber
Pater. ... ‘To Father Lyaeus’, who as we said above is appropriately the god of
freedom for cities; whence Marsyas too, his servant, is placed in the forum in
cities as a symbol of freedom, with his hand raised to show the city lacks
nothing.

I.

V//

Fig. 27 Mirror in the Villa Giulia Museum, Rome: woman and satyr.
 LIBER: MYTH, DRAMA AND IDEOLOGY IN REPUBLICAN ROME 101

....__ > i_.

1
&'

Fig. 28 One side of cistu no. 100, in the Villa Giulia Museum, Rome (Battaglia
and Emiliozzi 1990.312-6): satyr and winged girls.
102 UNWRITTEN ROME

Fig. 29 Marsyas statue at Paestum (Coarelli 1985.98, fig. 13).

 5m
3
H

'A
>3

Z
□
6
O
5
?
5
CD

n
z

Fig. 30 Cista no. 70, in the Villa Giulia Museum, Rome (Battaglia and
Emiliozzi 1990.232-6): Apollo and Marsyas.
i

s
104 UNWRITTEN ROME

III
On a mirror in the Villa Giulia, signed by the artist Vibis (=Vibius) Pilipus,
Marsyas is portrayed dancing cheerfully with an ithyphallic goatlegged
Paniskos [fig. 10, p. 59 above]. Little Pan is clearly imitating him, as the fat old
satyr on the Ficoroni cista in the Villa Giulia, no. 68 [fig. 31], imitates the
Argonaut working out at the punch-bag. Marsyas ‘stamps his foot and waves
the tail of an ass’: thus Anne Weis in the Lexicon Iconographicnm, but the
stamping is clearly a dance, as illustrated on a near-contemporary Lucanian
oinochoe [fig. 32].42 The hand-gesture too is evidently part of the satyrs’ dance­
repertoire, as is clear from many scenes on Athenian and Italian vases," and it
brings us back to young Silanus on the New York cista [fig. 22, p. 94 above].

>

Fig. 31 Detail of the ‘Ficoroni cista’ (no. 68) in the Villa Giulia ! ■ .■tun.
(Battaglia and Emiliozzi 1990.211—25); Argonauts ami old sat' i.

ccording to the description of that scene in Le ciste pivnestine. the young

woman solleva ambedue le mani in un gesto che sembra th protesta’.'" But
men a misreadlnf the young satyr is not assaulting her, and her hand-move-
familin °’lesP°n to h's- ionographically, what we have here is a scene
r om the Satricum antefixes of the early fifth century hc |fig. 33], the

42 Weis 1992.368; Trendall 1967.141, no. 784 (plate 66.6). Is Marsyas on the mirror waving
his own tail? If so, we must think of him as a performer. 40-3
43 Athenian evidence conveniently collected in Brommer 1944: see figs 2, 4. 9. 10, 38. ' - •
67. An early Italian example, c. 470-460 Be:, is a cup from Vulci in the Rodin Museum in
Paris (Beazley 1947.25-7, Plate 1V.1, Martelli 1987 fig. 160); it carries the name ot Aules
Vipinas.
44 Battaglia and Emiliozzi 1979.147.
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 105

Fig. 32 Lucanian oiiioJu’e (Trendall 1967, no. 784): dancing satyr.

106 UNWRITTEN ROME

J'J '

Fig. 33 Antefix from Satricum in the Villa Giulia Museum, Rome (Giglioli
1935, tav. cclxxxv.2): satyr and nymph dancing.
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 107

Fig. 34 1 ill', .up m the Villa Giulia Museum, Kome (Giglioli 1935, tav.
cdxxii.di: s n ■performer ami dancing girl.

dance of a satyr \\ itli a nvinph.*’' But the ‘nymph’ on the New York cista is no
creature of the wild; she wears very elaborate jewellery. I think she is a dancing
girl playing the part of a nymph, analogous perhaps to the hairy-suited
performer on the Baltimore mirror [fig. 14, p. 67 above]. Indeed, we see the
two together on a fourth-century Faliscan cup in the Villa Giulia [fig. 34], A
closer parallel for Silanus and his partner is a scene on an Apulian dish in the
British Museum |fig. 35), showing a young satyr with the kottabos-cup on his
finger and a dancing girl using the turned-up hand gesture.46 The context

45 Not a maenad, as wrongly stated in Wiseman 1994.71 and 77; see Hedreen 1994. Also
mistaken is the description of the satyr’s gesture in Giglioli 1935.35: ‘con la mano destra fa
un gesto come di nluttanza'.
46 At Trendall and Cambitoglou 1982.281. the satyr is wrongly described as ‘a nude youth’.
 108 UNWRITTEN ROME

Fig. 35 Apulian dish in the British Museum (F J33); young • il'-.■ Dicing girl.

seems to be a symposion, but it is worth noticing that th< i >f the scene
shows two bacchantes attacking the disguised Pentheus.
A mid-fourth-century comment from an Athenian wi ..new southern
Italy may help us to see what is going on. In the seventh : . . of the Laws,
Plato discusses acceptable and unacceptable types of dance;'

boi] pev flaK/Eta t’ eotlv Kai nov Tavratg EjropEvcnv, tig Ntipt] a; te koi
flavag Kai S£iXi]voug Kai SaTupoog EjrovopdgovTig, <■’>; quoiv,
pipoOvTat KaTtpvwgEvovg, jtspl KaOappovg te Kai TtktTdg Ttvag
ajtoTEkouvTiov, ovpirav toOto Tfjg op/ijciEtog to yEVog oi’O’ tog
EtpT]viKdv oi>0’ tog rroXspiKOV oi>0’ on jtote liouXtrat pabiov
dcpoplaaoOat.
Any [dancing] that is Bacchic in nature, or that resembles those in which, while
celebrating certain ceremonies of purification and initiation, the dancers
‘imitate’, as the phrase is, drunken persons, calling them nymphs. Pans, Sileni
and satyrs—this kind of dancing is, as a whole, neither peaceful nor warlike,
and it is hard to determine what its purpose is.

47 Plato Laws 7.815c. For text and meaning, see England 1921.302—3 and Morrow 1960.362—
3 and n. 221; 1 have borrowed part of Morrow’s translation.
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 109

As a modern commentator observes, ‘despite the questionable character of

these dances, Plato hesitates to legislate against them; and this hesitation may be
due to a reluctance to lay hands upon long-established religious practices.’48
The Senate showed no such reluctance in 186 BC; presumably ceremonies of
purification and initiation were what the ‘ignoble’ Greco-Etruscan sacrifiaihis ct
nates had been offering.49
One detail of the Plato passage is particularly important for our purposes;
wq cpctotv introduces the technical term (.ItgoOvTai, which means ‘represent
in the form of a |Xl|J.O5’.’" We have already noticed examples of visually-repre­
sented imitation: Marsyas and Paniskos, the Argonaut and the old fat satyr. The
hairy body-suit, worn on the Baltimore mirror by someone at a symposion,31
was also one of the regular costumes for actors in satyr-play.52 The common
ground of drinking parties, Dionysiac ritual and dramatic mimesis, clearly
implied by Plato but not easy for moderns to visualise, has been explored from
various angles in recent years,’3 and it seems to me that interpretation of the
iconography of the Latin bronzes must also take account of it.
The ubiquity’ of the satyrs makes it natural to think first of satyric drama and
quasi-drama. I he hairy-suited papposilenos is a very familiar character in south-
Italian, ami particularly Paestan, vase-painting, frequently portrayed in boots,
which make i: ci . : that he is a performer even though the scenes are not obvi­
ously dramatic. \n Apulian bell-crater in Cleveland [fig. 36| shows him in a
mask, but a mrl. quite different from that of Attic satyr-play, and his
companion it. Ji - '.■tie is a slave from comedy.” Mime too is relevant here,
for even in , • -aty r-play was beginning to be associated with mime by the
fourth centm■ > In Italy the dramatic genres were even less watertight, as is
shown by .\p .- t and Paestan vases showing naked girl acrobats performing
with comic a i is.

48 Morrow !*>oo
49 Livy .39.8.3; , f. 3‘>.1 > 1 I tor the consul’s attempt to reassure the populus that ancient rites
were not being \ lolatcrl.
50 England 1921.302.
51 Evidently associated with a wedding ceremony (p. 69 above); cf. Seaford 1994.308 for the
analogies between marriage ami Dionysiac initiation. Hairy-suited satyrs on Delos, perhaps
to be associated with the phallephoria procession: Cole 1993.32-3.
52 See the T’ronomos Vase' (late fifth century), with the actor in the hairy suit at upper right:
conveniently illustrated m Seaford 1984 plate 111, and in Green 1994.44, fig. 2.19.
53 E.g. Seaford 1984.5-10, Seaford 1994.266-9, Green 1994.38-46 and 89-93. See in general
Green 1995.
54 E.g. Trendall 1987. plates 28e, 71c, 92, 93b. 99c, lOle, 105a-b, 105e, 107a, 107c, 120b,
162a. 162c. 241a, 241c: Green 1995.92-3, figs 4.3-4.
55 Green 1995.102-3 and plate 8; contrast the ’pronomos Vase’ (n. 52 above).
56 See Luce 1930.339, Simon 1982.19-20 and plate 8: Dionysus entertained by, respectively, a
niinia playing a satyr, and a satyr called Mimos.
110 UNWRITTEN ROME

*_______
Fig. 36 Bell crater with Dionysiac scene in the Cleveland Museum of Art:
South Italy, Apulia, early fourth century BC, 400-390 BC. Earthenware with
slip decoration, 37.8 x 40.3 cm.
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME Ill

As C.W. Dearden puts it (though I think the evidence justifies a less tenta­
tive formulation),37

the argument to be made is that the tight lines drawn around the individual
dramatic genre in fifth-century Athens may not have been so watertight in
Magna Graecia in the fourth century and later... The evidence is by no means
conclusive but it could be taken to suggest that vase painters in Southern Italy
and, more importantly, their customers, saw drama of whatever genre in a
comprehensive light.

So—to return to the New York cista [fig. 22, p. 94 above]—a young woman,
naked but adorned with jewellery, dancing with a satyr, might be seen either as
a nymph, or as a Bacchic worshipper imitating a nymph, or as a mima playing a
nymph on stage.
Another recurring motif may help us. A fourth-century Etruscan cup in the
Vatican [fig. 37] shows a dancing satyr, his left hand in the now-familiar
turned-up position, with two naked and bejewelled young women, who are
not dancing but using a mirror and a perfume jar; their clothes hang on a tree
behind. Another cup from the same ‘Clusium’ group [fig. 38] shows three such
young women, adorned in just the same way, playing the roles of Leda and her
maidservants—and the eagle shows that the scene is taken from Euripides’
Helen.3"
On cista no. S2 in the Villa Giulia [fig. 39, p. 114], at the right hand side of
the composition a young dancing satyr (same gesture of the hand) looks across
at a naked young woman wearing jewellery and using a mirror. Between
them, playing the atdai, sits an old fat satyr wearing the same type of crown as
Ebrios [fig. 23, p. 95 above]. The piping satyr reappears on two cistae at Palest­
rina, nos. 50 and 51 (fig. 40, p. 115], both illustrating the same scene; but this
time he sits and plays beneath a basin at which two naked and bejewelled
young women are washing.
For Menichetti, these are examples of the feminine paideia that leads to
marriage,31' but a different and more specific explanation is available if we keep
mimesis in mind. In the iconographic conventions of Athenian vase-painting,
the presence of a piper (auletes) indicates that the scene is not ‘real life’ but a
performance.Let us suppose for the sake of argument that the same conven­
tion applies to the piping satyr in cistae nos. 50-51 and 82.

57 Dearden 1995 plate 1 (quotations from 85 and 86). See also Hughes 1997, identifying a
naked female comic dancer (‘Konnakisj on a fourth-century Tarentine krater.
58 Euripides Helen 17-21; Martelli 1987.331, fig. 180.
59 See above, p. 90.
60 Green 1994.24 (‘a conventional symbol whose function is to indicate that what we have
here is a performance in the theatre’); Carpenter 1997.78 fa signal that the context tor the
scene is a performance’).
 112 UNWRITTEN ROME

Fig. 37 Etruscan cup (Clusium group) in the Vatican. Museo cii- .

14962): showgirls and dancing satyr.

The first of the two scenes /fig. 40, p. 1 15| is described in l.e ciste preHcstine
■ scena di toletta ed immagini di repertorio . . . sei personaggi aggruppati due
*L. Ue’ senza *1 niinimo legame narrative o, connmque, logico’.1,1 It is true that
this is not a, story that we can recognise, but it doesn’t follow that it is not a
story at all.' The naked, kneeling captive is not just a stock figure; he looks up
at the tall man with the staff, and the lady wrapped up in a cloak clearly belongs
in the anie scene, as does the young man to the right (not shown on no. 50).

61 Battaglia and Emiliozzi 1979.158-9 and 160.

62 There are plenty of unfamiliar stories on the cistac: e.g. the man. the dwarf and the gtr ssit i
the knife (handle of no. 47), the chariot-race of Venus and Auccna (lid ot no. 9), the attac
on the palace (no. 54), the Dioscuri and the pater poiniilionimi (no. 27). etc. For Aucena see
Peruzzi 1998.161-4.
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 113

Fig. 38 la:. . (Clushnn group) in the Musee d’art et d’histoire, Geneva:

showgirls . .. 1 eda and the swan.

It is not clear \ hetlier the bearded warrior with the horse, and the tall lady
accompanied In a w inged J,union carrying a parasol, are part of the same story,
or another. But the\ are surely in a story of some sort.
With no. 82 |fig. 39| we are on firmer ground. The identification of the
warrior to the left, w ith horse and dog, is uncertain, as is that of the naked
young man w ith spear and sword just to the left of the dancing satyr. But the
figures between them are recognisable: grieving Achilles, Iphigeneia,
Clytemnestra (or Diana?) at the window, Agamemnon.63 It is a scene familiar

63 Battaglia and Emiliozzi 1990.274—6: Menichetti 1995.66. [See Graham 1998 tor 'the
woman at the window’ as normally a prostitute or an adulteress; the latter would be appro­
priate for Clytemnestra.|
 c
z
§
H

z
7J
0
3

Fig. 39 Cista no. 82, in the Villa Giulia Museum, Rome (Battagli;ia and
Emiliozzi 1990.273—7): sacrifice of Iphigeneia, with piping satyr.
 5
m

3
-<
I
□
Z

>
z
o
5
m

5
5
S
g
TO

n
z
g
Fig. 40 Cista no. 51, in the Museo Archeologico, Palestrina (Battaglia and
Eniiliozzi 1979.160-2): unidentified scene, with piping satyr.
3
 116 UNWRITTEN ROME

fiom tragedy, and Bordenache Battaglia and Emiliozzi are right to point out its
indubbi riflessi di una rappresentazione teatrale’, including the window: ‘la
finestrella cioe, potrebbe indicate, in modo compendiario, lo sfondo architet-
tonico della scena stessa.’64
So our hypothesis about the piping satyr may be justified: one of the two
scenes is certainly theatrical, and the other could be. In one case the satyr is
accompanied by two women at the basin, in the other by a woman with a
mirror and a young satyr dancing—stock figures, according to the editors of
the corpus, but it is hardly possible to believe that they have their place in the
composition just because the artist had some space to fill.65 If we are right that
the piping satyr represents theatricality, then so, in some sense, must the other
figures.
Let us look again at Iphigeneia [fig. 39]:“

La fanciulla e rappresentata nell’atto di denudarsi, sollevando con ambedue le

mani un ampio mantello per lasciarlo ricadere dietro le spalle, quasi a sottolin-
eare, in maniera un po’ teatrale, 1’inumana condanna a metter fine alia sua
splendida giovinezza.
What sort of theatre would expose her ‘splendida giovinezza’ in this way? Why
should a tragic heroine have nothing on under her cloak? Perhaps we may
approach an answer if we notice that what little she is wearing is the same
jewellery as the girl with the mirror: necklace, bracelets, earrings. So too on
the New York cista, no. 45 [fig. 22, p. 94 above], the young woman dancing
with Silanus wears practically the same jewellery as the woman called I9O.K.4
in the main scene. There is no piping satyr there, but otherwise the juxtaposi­
tion of the dancing satyr and a young woman next to a narrative scene is very
close to what we have in no. 82.67
Apart from young Silanus and his partner, what is happening on the New
York cista? The scene is flanked by two female figures named as DOX/1 and
LAVIS (i.e. Ians?). Between them, from left to right: LADVMEDA with a
stag, in front of a bearded herm; /17/IX ILIOS, holding the reins of two horses;
a column with a notice attached (LECES, i.e. leges?); a naked young man with
a sword and two spears, named as SORESIOS; ACMEMONO, i.e.
Agamemnon, in a pose similar to that on no. 82; two horses looking out of a
double window; and a young man in a short cloak, named as ISTOR. It is a

64 Battaglia and Emiliozzi 1990.275. Windows often appear in theatrical scenes on Paestan
vase-painting: see Green 1995.109-10 and plate 10.
65 Battaglia and Emiliozzi 1990.276: Tincisore colma lo spazio rimasto libero, senza soluzione
di continuity, con tre figure di repertorio...*
66 Battaglia and Emiliozzi 1990.274-5 (cf. 1979.148).
67 For the similarities between the two scenes, see Battaglia and Emiliozzi 1979.148—9.
 LIBER: MYTH, DRAMA AND IDEOLOGY IN REPUBLICAN ROME 117

very symmetrical scene, and one which implies a contest: ‘Glory’ and ‘Praise’
at the sides, ‘the rules’ in the centre, two sets of two horses, and a man whose
name means ‘judge’ (lOTWp). How Laodamia, Ajax and Agamemnon fit in we
do not know, but their presence makes it clear that this is not just the celebra­
tion of a chariot-race victory (by Soresios?). The contest or race must belong
in the age of heroes, or in the afterlife. Here too, as in nos. 50-51, we have an
unknown story, but one which could have been a stage perfomiance.
My suggestion is that the young woman dancing with Silanus on no. 45,
the young woman with the mirror on no. 82, and the two young women at
the basin on nos. 50 and 51, are performers whom we might best describe as
niiinae, portrayed next to but outside the dramas in which they appeared. The
matching jewellery suggests that Silanus’ partner could have played Doxa, and
the girl on no. 82 could have played Iphigeneia; but what about the girls on
nos. 50-51? Here, very tentatively, I should like to make a further suggestion.
The editors of the corpus rightly note the ‘ampio mantello’ of which Iphi­
geneia is divesting herself on no. 82 [fig. 39 above]. Is it possible that before she
took it off, she was wrapped up in it like the woman behind the kneeling
captive on nos. 50 and 51 [fig. 40 above]? Young women wrapped up like that
are seen dancing in the Bacchic thiasos on no. 72 [fig. 20, p. 91 above] and
no. 1 [fig. 25, p. 97 above], and also on a cista we have not considered yet, no.
31 in the British Museum [fig. 41]. The scene on no. 72 is not obviously
dramatic, but it does include a piping satyr. Nos. 1 and 31 show ‘composite’
scenes, with, die thiasos merging into a particular story.“ On no. 31, Dis Pater
and Proserpina receive a messenger announcing the death of a youth, who
dismounts from his horse just as winged Victoria holds out a diadem to him;69
the thiasos consists of Liber, bowed and supported by a young boy, a grimacing
satyr holding a torch reversed, the wrapped-up young woman dancing,
another satyr, and a bearded male figure turning to observe the scene in the
Underworld.
I suggest that these ‘composite’ scenes are similar in intention to the piping­
satyr motif observed above—that is, they play on the viewer’s understanding
that Liber is among other things a god of drama,7" and thus indicate that his
followers are not only singing and dancing but also imitating (Plato’s word) the
legendary story. If that is so, then the young women so conspicuously wrapped
up may indeed be part of the performance.

68 See n. 32 above for the ‘merging’ of the thiasos and the Perseus and Medusa story on no. 1;
on no. 31 the scenes are linked by the turned head ot the man on the right.
69 Battaglia and Emiliozzi 1979.117; cf. n. 29 above for Dis Pater.
70 In Athens, the god of drama was Dionysos Eleuthereus: for the identification with Liber see
Alexander ‘Polyhistor’ FGrH 273 F109 (Plutarch Moralia 289a = Quaestioiies Romanae 104).
 00

c
z

H
m

73
O
3

Fig. 41 Cislii no. 31, in the British Museum (Battaglia and Emiliozzi
1979.116-18): scene in Hades, with satyrs.
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 119

IV
To sum up the argument so far, I think these cista scenes may imply a kind of
Dionysiac dramatic performance, perhaps not separable from Dionysiac ritual,
in which one feature was the participation of mimae. Their role may have
evolved from the imitation of nymphs dancing with satyrs, which would at
least help to explain its startlingly erotic quality; for what we seem to infer is a
dramatic convention by which young women performed wrapped up
completely in voluminous cloaks, and then removed them ('in maniera un po’
teatrale’) to appear spectacularly naked. Not only that, but the stories in which
they did so were sometimes plots from the repertoire of Attic tragedy like Ip/ii-
geneia at Aulis.
Two further cistae may illuminate the phenomenon. No. 9, in Berlin [fig.
42], shows CRISIDA (Chryseis?) and Ajax between two mounted Amazons
called CASENTER (Cassandra?) and O1NVMAMA, watched by a young man
in petasos and chlamys called ALSES. Behind Alses, ALIXENTR is carrying out
a version of the Judgement of Paris in which the contestants are ATELETA
(Atalanta?), ALSIR and HELENA. All three of the contestants are naked
except for jewellery (the disgruntled Helen is either dropping her cloak or
about to put it on again), and Crisida’s dress has fallen open at the front. On
no. 83, in the Villa Giulia [fig. 43], CREISITA and ELENA are at the
washing-basin by a fountain, both naked and bejewelled, with Helen holding
her cloak out behind her in a pose identical to that of Iphigeneia on no. 82. To
the left stand I ONDRl 'S (Tyndareus?), a figure whose name has been lost,
and a warrior called SECJI.LI 'Cl'S holding a horse; to the right are ACILES
with a horse, a slave called SLMOS, and ORESTE[s).
These scenes could be burlesque mythology, with characters and stories
deliberately mixed up. It they are from drama, it is probably paratragic. It may
not be accidental that Ajax and Atalanta, who feature on cista no. 9, both had
satyr-plays named after them. 1 But the scenes on the cistae may be closer to a
less formal dramatic genre, the paignioti, recently discussed in an important
article by James Davidson.72 Mimic, erotic, with plots from tragedy, paigiiia
could be appropriate entertainment for a symposion.73 The form dates back at
least to the fifth century' at Athens, with the work of a certain Gnesippus,
mentioned by Cratinus as a tragic poet, whom Athenaeus describes as ‘a
/wigHiUM-author of the cheerful muse’.74

71 Aristias’ Atalaute (Poll,lux 8.31), Polemaeus of Ephesus’ /lias (SIG 3'.1079): Steffen 1952.117,
257.
72 See Davidson 2000.
73 Plutarch Moralia 712e-f (Quaest. conv. 8); Athenaeus 14.638d-639a. TdlLElV and minius:
Suetonius Diuus Augustus 99.1; cf. Plutarch Caesar 10.2 for Ttlt'CeiV at the Bona Dea rites,
on which see Wiseman 1974.130-4. Laevius’ Eratapaiguia: Courtney 1993.118-43.
74 TGF 27 T1 = Athenaeus 14.638d (Tut'/viaypugou xfjg ikap&q |K>lior|;); Cratinus fr. 256.
 o

c
z
77
-j
z
73
o
3

Fig. 42 Cista no. 9, in the Staadiche Museen, Berlin (Battaglia and Emiliozzi
1979.64-5): Crisida and Ajax.
 m
5

3
H
:Z: >i'^.>-
□
fl
xr,n <A M K-fi-k. >

5®wBaWWWi|
3

i,^' U£g|f V-f”X A J /?

/> i 0
□
O
i 5
,\ > 3
ww ■/-'

g5
CD

n
z
g
Fig. 43 Cista no. 83, in the Villa Giulia Museum, Rome (Battaglia and
Emiliozzi 1990.277-80): Creisita and Elena.
3
 122 UNWRITTEN ROME

That description is reminiscent of the ‘cheerful tragedy’ pioneered by

Rhinthon of Tarentum in the third century BC.75 Rhinthon’s work has often
been associated with the scenes of comedy and mythological burlesque on
south-Italian vase-painting (the so-called ‘phlyax vases’), but they belong a
generation or more earlier.76 The natural inference is that the vases illustrate
various forms of fourth-century dramatic performance in various parts of
southern Italy, from sophisticated Tarentum to Oscan-speaking Campania,
influenced in varying degrees by the Athenian theatre (and we may add the
Syracusan too) but not bound by its generic rules.77 The scenes on the cistae
extend that range into Latium, and the spelling of the names implies a culture
in which the names of the characters of Greek mythology were known orally,
and not (or not primarily) by texts. So if in Latium we seem to find the un­
Attic phenomenon of iitiniae displaying themselves naked in roles otherwise
appropriate to tragedy, there is no a priori reason to doubt the validity of the
inference.
A late fourth-century jug in Naples [fig. 44] offers a close iconographic
parallel to Iphigeneia and Helen on cistae nos. 82 and 83 [tigs 39 and 43, pp. 114
and 121 above]: the jewellery, the cloak and the raised arms are the same, but
here the young woman’s arms and cloak are separately attached to two
uprights which look like oars. She is Andromeda, heroine of Euripides most
popular play, ‘an openly romantic tale, complete with magic, monsters, and a
happy ending’.78 In all earlier representations of Andromeda’s exposure to the
monster, she is richly dressed as a barbarian princess. The first time she is repre­
sented naked, as was the norm in later Roman art, is precisely here, in
Campania in the late fourth century, close in both time and space to the world
of the Latin cistae.19 Andromeda was above all a subject for the stage—not only
Euripidean tragedy but probably satyr-play too, and later in Latin plays by

75 Suda s.v. Rhinthon: 'PivOtuv, Tapavrtvog KtopiKog, dp/ip/b; Ti]> KaXoi>|ifVi]C

iXapoTpaycpdiag, o eon tpkoaKoypctcpia. Stephanus of Byzantium s.v. laras: PtvOtov,
Tapavrtvog cpXualj, ret rpayueft (lETappoOpt^tov ec id yekoiov. Most of his titles are
Euripidean (Herakles, Iphigeneia in Anlis, Iphigeneia in Tanris, Medea, Orestes, Telephns).
76 Trendall 1959.9; Taplin 1993.48—54.
77 See above, pp. 109—11. See Taplin 1993.89—99 on the ‘transplantation of Attic comedy to
Magna Graecia; ibid. 40—1 (plate 16.15) and 94 on Oscan-speaking Campania. Syracusan
theatre: comedies by Epichannus, Phonnis and Deinolochus, mimes by Sophron and
Xenarchus (Kaibel 1899.87—182); Epichannus wrote an Aniykos, a favourite subject on the
Latin bronzes (e.g. cistae nos. 8, 38, 49, 68), and an Atalantai (cf. cista no. 9) was attributed to
both Epichannus (frr. 14—18 Kaibel) and to Phonnis (Athenaeus 14.652a).
78 Euripides frr. 114—56 Nauck; for its popularity, Aristophanes Frogs 52—I (and the parody at
Thesniophoriaznsae 1009—1134), Athenaeus 12.537d, Lucian De historia conscribenda 1; quota­
tion from Michelini 1987.77 [and sec now Wright 2005].
79 Phillips 1968.9; Schauenberg 1981.788. Indeed, there is a precise parallel if (as Phillips
thinks, 1968.13) the Andromeda scene on cista no. *60, in the Louvre, is genuine—but alas,
it seems unlikely to be so: see R. Leprevots Trogan, in Battaglia and Emiliozzi 1990.187—8.

i
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 123

Livius Andronicus, Ennius and Accius.80 The comparable iconography makes

it likely, I think, that Helen and Iphigeneia on the cistae are on stage too.
The complex monographic evidence presented here indicates, I think, a
common fourth-century culture ot mimetic representation extending far
beyond the Greek cities of southern Italy into Latium and Etruria. In Apulia
[fig. 35, p. 108 above], we see a kottafeos-playing satyr, a piper, a naked
dancing-girl, and a scene from Euripides’ Baahae. At Etruscan Clusium [figs 37
and 38, pp. 112-13 above], we have a dancing satyr and a scene described in
the prologue of Euripides’ Helen, with the inimae losing their cloaks and
displaying themselves naked as they protect the delusive swan from the eagle.
In Naples [fig. 441, a naked niima plays Euripides’ Andromeda. In Latium [fig.

Fig. 44 Campanian jug (Trendall 1967, Campanian 2/138): Andromeda.

80 Satyr-play: a plot involving Perseus is interred from vase-painting (Brommer 1944.25-9, figs
21-3). Livius fr. 18R (Nonius 86L), Ennius fir. 97-106R (=112-19 Jocelyn). Accius frr.
100-118R (=380-97 Dangel).
 124 UNWRITTEN ROME

39, p. 114 above], the cista scene shows satyrs piping and dancing, and a naked
mima dropping her cloak as Euripides’ Iphigeneia at Aulis. There is no reason to
suppose that Rome was in any way immune to this culture of Dionysiac
mimesis, or that it had been somehow lost before literary drama was introduced
there by Livius Andronicus three or four generations later.

V
Livius Andronicus came from Tarentum.81 If the Varronian chronology is
correct, and he produced his first play in Rome in 240 BC,82 he was a near­
contemporary of Rhinthon and must have been very familiar with the
‘cheerful tragedy’ mode; perhaps his Andromeda was an example of it.83
Just at the time of his Roman debut, at a date variously transmitted as 241
and 238 BC, the plebeian aediles, who had their headquarters at the temple of
Liber, struck a blow for the Roman People by prosecuting wealthy
landowners who were grazing their beasts on public land.84 They used the fines
to build a new temple to Flora, immediately next door to the Liber temple,
and set up games in honour of the goddess. According to Ovid (who reports
Flora’s own account), the Senate refused to fund the games on an annual basis
until forced to do so by the goddess’ anger in 173 BC.85 The reason was
presumably the Senate’s disapproval of the licentious nature of the games,
which in later times at least involved mimae undressing on stage.86
Three and a half centuries of ill-attested cultural change lie between the
engraving of the cistae and our literary evidence for the Indi Florales. Inevitably,
the argument connecting them is a tenuous one. But inadequate though the
evidence is, I think there is enough of it to justify the tentative hypothesis that
the scenes on the cistae reflect one type at least of mimetic performance in fourth­
century Latium. If Greeks called it paignia, Latins might translate that as Indi.

81 Accius fr. 18 Funaioli (Cicero Brutus 72).

82 Varro Depoetis fr. 55 Funaioli (Aldus Gellius 17.21.42); Cicero Brutus 72, De senectute 50; cf.
Cassiodorus Chronica on 239 bc. For the chronological problem (Accius put it much later),
see most recently Oakley 1998.61-3.
83 Rhinthonica known at Rome: Cicero Ad Atticinn 1.20.3, Donatus De contoedia (>. 1, Evanthius
De contoedia 4.1, Lydus De niagistratibus 1.40.
84 Ovid Fasti 5.279-90; Velleius Paterculus 1.14.8 (241 bc), Pliny Nat. Hist. 18.286 (238 BC).
Cf Wiseman 1979.93—4 and n. 124, arguing for 240 BC (the consulship of C. Claudius
Cento) as a possible date for the origin of the Floralia; note that cento means 'patchwork , and
that at Flora’s games it was the custom to wear multi-coloured clothes (Ovid Fasti 5.356,
ctdtti uersicolore).
85 Ovid Fasti 5.291-330, Tacitus Annals 2.49.1; cf. Varro De lingua Latina 5.158. Festus 276L.
86 Valerius Maximus 2.10.8 (priscus nios ioconnn), Seneca Epistles 97.8; cf Ovid Fasti 4.946 and
5.331-56, Martial 1 pref, and 1.35.8-9, Tertullian De spcctaculis 17, Amobius 3.23 and 7.33,
Lactantius Institntiones diuinae 1.20.6—10, Augustine City of God 2.27, Historia Augusta Heli-
ogabahts 6.5, Ausonius 7.24.25—6. [See pp. 175—86 below for the games of Flora.]
87 For ludere and Jtat^Etv, see above, nn. 12 and 73.

I
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 125

This suggestion is of course hardly compatible with the late-republican

history of the origin and development of Roman drama, as reconstructed by
Varro and transmitted to us by Livy, Horace and Valerius Maximus.88 Varro’s
view depended on the assumption that drama derived from rustic festivals, and
that in the early Republic the Romans were peasants innocent of Greek
literary culture. As Horace puts it in a too-famous passage,89
agricolite priscifortes paruoque beati,
condita post frumenta leuantes temporefesto
corpus et ipsum animum spe finis dura ferentem,
cum sociis operum pueris et coniugefida
Tellurem porco, Siluanum lade piabant,
floribus et uino Genium memorem breuis aeui ...
Graccia capta fentm uictorem cepit et artes
intulit agresti Latio. sic horridus ille
defluxit mimerus Satnmius etgraue aims
munditiae pepulere; sed in longum tamen aeuum
maiisenmt hodieque manent uestigia ruris.
sems enim Graecis admouit acumina chartis,
et post /’imii j bella quietus quaerere coepit
quid Sophocles et Thespis et Aeschylus utile Jerrent.
The men of old were tough tanners content with little. After the harvest was
brought m. they relaxed body and mind at holiday time, for the mind endures
hard wotk by the prospect of an end to it. With the boys and faithful wives
who shared their labours, they appeased Tellus with a pig, Silvanus with milk,
and Genius, who knows how short life is, with flowers and wine.. .
When Greece was captured, it took the fierce victor captive and brought
the arts into rustic Latium. That's how the emde Saturnian metre drained away,
and elegance drove out the offensive smell; but traces of the farmyard remained
for a long time, and still remain today. For it was only late that (the Roman]
applied his intelligence to Greek pages, and in the peace after the Punic Wars
began to ask what use Sophocles and Thespis and Aeschylus could be.

This passage represents just one of the paradigms of early Rome in the anti­
quarian tradition; precisely the opposite picture is oftered by Cicero in De
republica, who argues that since Romulus post-dated Homer, the Romans at
the time of the foundation lived in an age when literary culture was long
familiar and primitive conditions had long been dispelled.9" Both notions are

88 Livy 7.2, Horace Epistles 2.1.139-76, Valerius Maximus 2.4.4; cf. also Virgil Georgies 2.380-
96; Schmidt 1989, Oakley 1998.40-58.
89 Horace Epistles 2.1.139-44, 156-63; detailed commentary in Brink 1982.175-215.
90 Cicero De republica 2.18—19: iam inueteratis litteris atque doctrinis onmique illo antiquo ex inciiha
homiuum uita errore sublato.
 126 UNWRITTEN ROME

learned reconstructions deployed as part of an argument, and neither has any

historical validity. How could Horace or Cicero have known what early
Rome was like?
Unfortunately, however, Horace’s epigram about Graecia capta has predis­
posed some modern scholars to reject the possibility of significant Greek
influence on Roman culture before the second century BC. So for instance
Richard Beacham, in an otherwise valuable book on Roman stagecraft, can
follow his account of the theatre in Magna Graecia with this assertion:91

It is most unlikely that Rome of the mid-fourth century was in any position
either geographically or culturally to benefit directly from such a sophisticated
theatrical culture. Roman visitors may possibly have been bemused observers
from time to time of performances in Greek theatres, but they would have
brought back only travellers’ tales of practices, entertaining perhaps to recount
to their families and friends, but aesthetically too advanced to be understood or
to influence the growth of whatever crude drama was germinating at Rome.
The influence of Horace’s patronising paradigm is clearly detectable here; but
the iconography of the cistae, and the artistic sophistication they represent, are
enough to refute it completely.
Varro’s ‘theoretical construction’ of the prehistory of the Roman theatre is
much indebted to Eratosthenes’ researches into Attic drama.92 That is a familiar
technique: in the absence of good evidence, analogy can provide the mate­
rial.93 But Varro also had an ideological reason to attribute a native and rustic
origin to Roman drama. He had strong views about the theatre in his own
time, particularly its portrayal of the gods,94 and it is very likely that he blamed
its deplorable aspects on Greek influence corrupting an honest native tradition.
So those tough old farmers are imagined as sacrificing not to Ceres (Demeter)
or Liber (Dionysus), but to Latin gods untainted by Greek myth Tellus,
Silvanus and Genius.
As the power of fertility both human and agricultural,9’ Genius receives
flowers and wine, offerings that Horace’s readers would associate with Flora

91 Beacham 1991.7-8.
92 See Horsfall 1994, esp.66-70; quotation from Rawson ! ^274
93 For examples in historiography, see Soltau 1909.73-71. 1 hire eviun y but it is
evidence for the Roman theatre (the coinnieutarii cited at Cicero Brutus an
not likely to have gone back to the fourth century BC. |7_9 n. note
94 Varro Autiquitates rcrum diuiuanuu frr. 7 and 10 Cardauns (cf. Wiseman • . ?
that one of the ‘tales unworthy of the gods’ to which he objects is the birr o g
cxguttis sauquinis uatus), as illustrated on cista no. 1. He was
95 Varro Autiquitates rerum diuiuanuu fr. 248 Cardauns: uim habet otmtium miimffffteitamim.
thought of as the father of Tages, who was bom from a ploughed field ( ’ .
De diuinatiouc 2.50); and he was identified as the Dirfamiliaris (Censonnus 3._), '
fested himself as the phallus in the hearth to beget Servius Tullius (1 liny 1 at-
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 127

Mater and Liber Pater; and it is Genius who is mindful of the shortness of
human life, which according to Ovid was Flora’s justification for the licen­
tiousness of her games.96 Liber and Flora were the gods of fertility honoured in
Roman public ritual, and their conspicuous omission here must be by delib­
erate choice.97 So the Varronian model for the origin of Roman drama is
indeed of great historical importance—but not as evidence for the real condi­
tions of the fourth century BC. Rather it is an example of the ongoing
ideological sensitivity of Liber and Flora.
Liber had his ancient cult and temple, shared with Ceres and Libera, at the
foot of the Aventine; it was the headquarters of the plebeian aediles, and their
use of it as an archive for senatorial decrees was represented as an important
aspect of the lihertas restored after the tyranny of the Decemvirs.98 The cistae
reveal total familiarity with Dionysiac imagery and ritual in the fourth century
BC, and seem also to attest a type of drama, or mimetic performance, which
combined mythic narrative with erotic burlesque. About 240 BC the People’s
representatives built a temple to Flora next to that of Liber, Ceres and Libera,
and introduced games of which the Senate refused to approve. At the end of
the third century BC we find Naevius, a politically controversial playwright,
referring to the Indi of the Liberalia as a forum for freedom of speech. In 186 BC
the specifically Dionysiac aspects of the cult of Liber were brutally suppressed.
Flora’s games, however, were recognised at last in 173 BC, perhaps as a neces­
sary' concession to popular feeling.99
In the light of all that, it is not surprising that a culturally conservative anti­
quarian like Vart o would create a prehistory' of Roman drama which left out
Liber completely. His reconstruction, as transmitted by Livy, Horace and
Valerius Maximus, cannot stand against the visual evidence of the cistae. The
scenes they show are hard to interpret, but they are at least a contemporary
source.

96 Horace Epistles 2.1.144, Ovid Fasti 5.353—4. Flora Mater: e.g. Cicero Venines 5.36,
Lucretius 5.739. Liber Pater: e.g. Varro De re rustica 1.2.19, Horace Epistles 2.1.5. Associa­
tion of sex, flowers and wine: Ovid Fasti 5.331-54 on Flora and Bacchus.
97 Liber as phallic god of fertility: Varro Antiquitates rcnim diiiiiianaii fr. 262 Cardauns (Augus­
tine City of Cod 7.21). Flora presides over all fertility, including human: Ovid Fasti
5.261-74. ’
98 Temple: Dionysius of Halicarnassus 6.17.2-4, 6.94.3 (‘vowed in 496 BC, dedicated in 493’),
with Wiseman 1998.35—6. Aediles: Livy 3.55.7 and 13 (3.55.1—56.1 on libertas restored in
449), 10.23.13, 26.6.19, 26.36.9, 33.25.2-3.
99 See above, pp. 84-5, 124. I have suggested elsewhere that Ampelius’ reference to Liber the
son of Flora (9.11) may reflect the post—186 situation, when Semele as Liber’s mother was
too closely associated with the Bacchanals: Wiseman 1998.41-2.
J
128 UNWRITTEN ROME

VI
The cult of Ceres, Liber and Libera and the cult of Flora both involved games
which were set up as a result of famine, after consultation of the Sibylline
books;100 the goddess of agriculture, the god of viticulture and the goddess of
the flowering spring had to be placated or the Roman People would starve.
But both sets of Indi were politically controversial: as we have seen, the Senate
refused for sixty-five years to make Flora’s games annual, and at some point in
the second century BC Liber’s games were taken away from the Liberalia and
merged with those of Ceres in April.101
Marsyas too, the servant of Liber,102 may have featured in these ill-attested
ideological tensions. His old enemy Apollo was honoured with Indi in 212 BC,
and it is clear from Livy’s account that the innovation was the Senate’s reaction
to the demands of agitators at a time of popular dissatisfaction. But Apollo
failed to bring about the end of the war, which had been the purpose of the
exercise; that was done by the Phrygian Magna Mater, who was brought to
Rome in 204 and granted her own temple and Indi.'03 Marsyas was part of the
Magna Mater’s story:104
ouvavaoTpEtpEoOai 6’ aifrfj ical tptkiav e/elv ejii jcXeov <i?aol Mapauav
rdv Opuya, Oaupa^dpEvov ejtI ouveoel ral otocppoofiviy Tfjg fttv
ouvEOEtng TEKpiptov XappdvoucH to pipTjoaoOcti Toi>g | 0oYYOvg Tfjg
jtoXuKaXapou oupcYY°5 [ieteveykeIv ejtI Tout; afiXoi’S xf|v oXqv
appovtav, Tfjg be oojcppoabvrig or||iEiov eIvcxl tpaat rd pt/pi Tfjg
TEXEuif]? djtEipaTov yevEoOai xa>v acppobtutcnv.
The man who associated with her and loved her more than anyone else, they
say, was Marsyas the Phrygian, who was admired for his intelligence and
chastity. A proof of his intelligence they find in the fact that he imitated the
sounds made by the pipe of many reeds and carried all its notes over into the
flute, and as an indication of his chastity they cite his abstinence from sexual
pleasures until the day of his death.
Here is a very different Marsyas from the presumptuous satyr who was taught
his place by Apollo. That surprising quality of sexual abstinence may have been
emphasised at the time when the rites of his master Liber—the Bncclnnialni held

100 Ceres, Liber, Libera: Dionysius of Halicarnassus 6.17.2-4, [Cyprian] De spectaadis 4.1.
Flora: Pliny Nat. Hist. 18.286, Degrassi 1963.132-3 (Fasti Pracnestini on f28~ April);
‘ Ovid'
cf. ~
Fasti 5.311-30 on the games made annual in 173.
101 Ovid Fasti 5.311-30 (Flora), 3.785-6 (Liber).
102 See above, pp. 99-100.
103 Livy 25.1.6-12, 25.12.2-15 (hidi Apollinares); 29.14.14, 34.54.3, 36.3.4 (Indi iMegalenses).
104 Diodorus Siculus 3.58.3. Cf. Solinus 1.8—9 (Cn. Gellius fr. 7P) on the Phrygian Megales,
ambassador of king Marsyas, who taught the Sabines the art of augury; his name evidently
alludes to the Magna Mater.

I
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 129

at Stimula’s grove close to the temple of Ceres, Liber and Libera—were alleged
to be orgies of sexual abuse.10’
It is tempting to see the suppression of the Bacchanalia in 186 BC as the
Senate’s reimposition of Apolline authority. Perhaps something of the sort may
be inferred from the censors’ arrangements in 179 and 174:106

theatruin et proscaeniitni ad Apollinis ... locauit ... scaenain aedilibus practoribus

praebeiidain . . . curaueriint.
[M. Lepidus the censor| contracted for a theatre and proscaenium at the temple
of Apollo . . . [The censors] saw to it that a stage be placed at the disposal of the
aediles and praetors.
Does this mean that the temple of Apollo in circa Flaininia was being organised
as the centre for Indi in general? (Only the Indi Apollinares were under the
praetor’s control; the aediles looked after the other games.) We don’t know
when Liber lost his Indi at the Liberalia,'"1 but this would certainly be an appro­
priate time. However, Ovid is explicit that a famine in 173 forced the Senate
to make Flora’s games annual; since Flora could be thought of as the mother of
Liber, it seems that the ideological issue was not quite settled after all.108
Forty years later, in the political crisis after the murder of Tiberius Grac­
chus, the Sibylline books ordered the Romans to placate Ceres antiqnissima',
that was interpreted to mean the Demeter of Enna in Sicily, whose daughter
Persephone was now identified as Libera. Cicero notes that the decenniiri sacris
facittndis deliberately did not involve the goddess at the Roman temple of
Ceres, Liber and Libera.109 Yet again, though the details escape us, there seems
to be an ideological question at stake. Though it was clear who would be
offended at the death of the People’s champion, the senatorial interpreters of
the will of the gods were evidently eager to move the confrontation out of
Rome.
At Philippi in 41 BC, ’the last struggle of the Free State’,11" Brutus’ password
was Libertas, and that of the Caesarians was Apollo. Part of the mythology of the
defeat was that Brutus had applied to himself the words of the dying Patroclus:
‘deadly fate and the son of Leto have killed me.’111

105 Livy 39.8—19; tor the location of the Incus Stinudae (Livy 39.12.5, Ovid Fasti 6.503, CIL
6.9897) see Cazanove 1986.56-66.
106 Livy 40.51.3, 41.27.5.
107 The tenninns post quern is Naevius fr. 113R. (p. 85 above).
108 Ovid Fasti 5.311—30; Ampelius 9.11 (n. 99 above).
109 Cicero Verrines 4.108, cf. 106, 109 and 5.187 for Libera.
110 Syme 1939.205.
111 Dio Cassius 47.43.1, Valerius Maximus 1.5.7, quoting Iliad 16.849. Plutarch s version
(Brutus 24.7), that ‘Apollo’ was Brutus’ own password, is probably an error; it is preferred
by Gurval 1995.98—100, who however ignores the Dio reference.
 130 UNWRITTEN ROME

The young Caesar claimed to be Apollo’s son, or Apollo himself,112 but

when he masqueraded as the god at a time of famine (probably in 40 or 39),
popular reaction was very hostile:113

ccua quoque tins secretior in fabulis fuit, quae uulgo 6a)bEKd0EOg uocabatur; in qua
deonim dearumque habitu discubuisse conuiuas et ipsum pro Apolline ornatum non
Antoni tnodo epistulae singulorum nomina amarissime enumerantis exprobrant scd ct
sine auctore notissimi uersus:
cum primuni istorum conduxit mensa choregum
sexque deos uidit Mallia sexque deas,
impia dum Phoebi Caesar mendacia ludit,
dum nona diuorum cenat adulteria,
omnia se a terris tunc nunrina declinamnt
fugit et auratos luppiter ipse thronos.
auxit cenae nimorem suinma tunc in ciuitate penuria ac fames, adclamatumque est
postridie omne frumentum deos comedisse et Caesarem esse plane Apolliitetn sed
Tortorem, quo cognomina is deus quadam in parte urbis colebatur.
There was besides a private dinner of his, commonly called that of the ‘twelve
gods’, which was the subject of gossip. At this the guests appealed in the guise
of gods and goddesses, while he himself was made up to represent Apollo, as
was charged not merely in letters of Antony, who spitefully gives the names of
all the guests, but also in these anonymous lines, which everyone• knows: ‘As
soon as that table of rascals had secured a choragus and Maliia saw six gods and
six goddesses, while Caesar impiously played the false role of Apollo and feasted
amid novel debaucheries of the gods, then all the deities turned their laces from
the earth and Jupiter himself fled from his golden throne.’ 1 he scandal of this
banquet was the greater because of dearth and famine in the land at that time,
and on the following day there was an outcry that the gods had eaten all the
grain and that Caesar was in truth Apollo, but Apollo the Tormentor, a
surname under which the god was worshipped in one part of the city.
There is an interesting parallel between this passage and the description of

Antony in Asia after the Philippi campaign:114

sig yohv EtpEoov elchovto^ auroO, yuvatKEc; p&v Etc; BaK/aq, avbps? 6s
teat Jtai&Eg Eig oatupovg Kai flavaq ifyoOvio SiEaKevaapEvoi, klttov 6e

11- Domitius Marsus fr. 8 (Courtney 1993.304-5), Dio Cassius 45.1.2, Suetonius Diuus
1995.87—136,
Augustus 94.4 (- Asclepiades of Mendes FGrH 617 F2). See now Gurval 1995.87-136,
, 'Vho puts Octavian’s association with Apollo very late.
113 Suetonius Diuus Augustus 70.1-2 (trans. J.C. Rolfe, Loeb Classical Library 1913). Famine:
Appian Civil Wars 5.67-8, Dio Cassius 48.31.1.
1Jutarch Antony 24.3—4 (trans. Ian Scott-Kilvert, Penguin Classics 1965); cf. 26.3 on
Antony m Cilicia.
 LIBER: MYTH, DRAMA AND IDEOLOGY IN REPUBLICAN ROME 131

Kai QupaiDV Kai ipakrnpicov Kai ovpiyyiov Kai avkibv f] rtokig fjv nkea,
Aiovuaov avxov dvaKakovgEvcov ^apiboTi]v Kai MEiki'/iov. fjv yap
dpekEt toioOtoi; Evioig, xoig 6e nokkoig ‘QgnoTij; Kai Aypicbviog.
acpnpeiTO yap EuyEVEig dvOpcbjtovg id dvra, gaoiiyiaig Kai Kokaiji
Xapi^d|.iEvo5.
At any rate, when Antony made his entry into Ephesus, women arrayed like
Bacchantes and men and boys like satyrs and Pans led the way before him, and
the city was full of ivy and thyrsus-wands and harps and pipes and flutes, the
people hailing him as Dionysus Giver of Joy and Beneficent. For he was so,
undoubtedly, to some; but to the greater part he was Dionysus Carnivorous
and Savage. For he took their property from well-born men and bestowed it on
flatterers and scoundrels.

In each case the divine identification has to be accepted. Opponents may insist
on the harmful aspects of the respective gods, but Octavian is Apollo, Antony
is Dionysus (i.e. Liber).
Moreover, it seems that this mythology’ reflects the same ideological
polarity that can be detected in the third and second centuries BC. For ‘Apollo
the torturer’ must surely be an allusion to the Marsyas myth, where the god’s
victory’ consists of Marsyas hanging from the tree, flayed alive. But Marsyas is
the servant of Liber, his statue is the guarantee of the People’s freedom, his
gesture announces that the city' lacks nothing.”’ If he is tortured, of course
there’s a famine’.

VII
It was not until the com supply was under control again, after the defeat of
Sextus Pompeius, that Octavian vowed his temple to Apollo and began
creating the great complex on the Palatine that would eventually be dedicated
on 9 October 28 uc.ll“ The site was on the brow of the hill, overlooking the
Forum Bovarium and the lower part of the Circus Maximus valley. Directly
opposite was the temple of Ceres, Liber and Libera, on the lower slope of the
Aventine ‘just above the starting-gates of the Circus’; described by Cicero as
pulcherriinHiii et nui^niftcentissinmni,"1 it was a splendid monument to Antony’s
patron god—and therefore, perhaps, offensive to Apollo and Octavian. In 31
BC, however, with the Apollo temple still under construction and Octavian

115 See above, pp. 99—100. The connection is made by Feeney 1991.220-1.
116 Vow: Velleius Paterculus 2.81.3, Dio Cassius 49.15.5. Dedication: Dio Cassius 53.1.3,
Degrassi 1963.209 (Fasti Antiates on 9 October).
117 Dionysius of Halicarnassus 6.94.3 (site), Cicero lertines 4.108. Had the old fifth-century
temple been rebuilt? If so, it is not known when or by whom.
 132 UNWRITTEN ROME

away fighting the Actium campaign, the eyesore was suddenly and fortuitously
removed:118
Kat jtvp aXXa te ovk dXtya Kai mhoO tov btjtobpdgov rcoXv to te
AT)gi]Tpiov ical ETEpov vadv ’EXitibo^ Ecp0Eipsv. sbo^av pev yap ol
eJjeXeuGepoi ovto 7iEJtotT)KEvai* Ttaot yap T015 ev te Tfj ’Irakis avTwv
overt Kat jtevte pvptabarv ovaiav fj Kai jeXelo) KEKTripsvotg to oy&oov
avTfj^ owteXeool ^keXevo6t), k<1k tovtov Kai Tapayal Kai tpdvot Kai
E|i7tpr|OEig wt’ ovtcov jtoXXal dysvovTo, Kai ov jrpoTEpov ys
KaTEOTTjoav jtplv fj Toig SjtXotg KaTa6anao0fjvat.
Fire also consumed a considerable portion of the Circus itself, along with the
temple of Ceres, another shrine dedicated to Spes, and a large number of other
structures. The freedmen were thought to have caused this; for all of them who
were in Italy and possessed property worth two hundred thousand sesterces or
more had been ordered to contribute an eighth of it. This resulted in numerous
riots, murders, and the burning of many buildings on their part, and they were
not brought to order until they were subdued by armed force.
Flora’s temple too must have been destroyed—and it is a remarkable fact that
in the Georgies, published the following year, the goddess ot bloom and growth
is not so much as mentioned.119
In 28 bc, the year of the dedication of the Apollo temple, Octavian on the
Senate’s instructions rebuilt 82 temples, ‘omitting none which deserved to be
rebuilt at that time’.12" What exactly does quod co tempore re/iei debebat mean? It
is clear that neither the Ceres-Liber-Libera temple nor the temple ot Flora
qualified for inclusion, since the date of their rededication is given by Tacitus
as AD 17, forty-seven years after the fire that destroyed them.121 Evidently
Augustus was content to do without those two temples for the greater part ot
his principate. For the citizens in the bustling Circus Maximus quarter there
was to be no rival centre of loyalty. They had only to look up, and there
against the blue sky was the great marble pediment of Apollo’s temple, with his
bronze quadriga flashing in the sun. It was from above (desuper) that Apollo
exerted his authority.122

118 Dio Cassius 50.10.3—4 (included in a list of portents, 50.10.6).

119 See Fantham 1992, esp. 49-52; contrast Lucretius 5.739, Cicero terrines 5.36, Varro De re
mstica 1.1.5-6.
120 Augustus Res gestae 20.4, cf. Livy 4.20.7 (templomm omnium conditorem ant restitutorem),
Suetonius Diuus Augustus 30.2, Dio Cassius 53.2.4—5.
121 Tacitus Annals 2.49.1: isdem temporibus denm aedis uetustate ant igni abolitas coeptasque ab
Augusto dedicaidt, Libero Liberaeque et Cereri iuxta circum maximum, qiiam A. Postiimins dictator
uouerat, eodemqne in loco Florae ab Lucio et Marco Publiciis aedilibus constitutam...
122 Propertius 2.31.9-11; Virgil Aeneid 8.705, 9.639.

I
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 133

It is a measure of Augustus’ effective power that he could do such a thing.

In 23 BC, when his life was threatened by serious illness, the Odes of Horace
were published, including one which must have touched some anxieties in
Rome:123

delicta maiorum iiinieritiis lues,

Romane, donee templa refeceris
aedesque labeiitis deonim et
foeda nigro simulacra fumo.
dis te minorem quod geris, imperas:
hinc onuie principium, hue refer exitiim.
di miilta negleeti dederunt
Hesperiae mala luctuosae.
Roman, though innocent you will pay for your ancestors’ misdeeds until you
have restored the gods’ temples and tailing shrines and statues foul with black
smoke. You only rule because you recognise the gods’ superiority: put down
every beginning and end to that. Neglected gods have inflicted many evils on
sorrowing Hesperia.

The following year there was famine. Surely Ceres, Liber and Flora must be
propitiated, as they had been in the past? But no: Augustus took the cura
aiiuonae on himself, and the crisis passed.124
Meanwhile, the talismanic statue of Marsyas still stood in the Forum,
crowned with flowers by the citizens he protected. The elder Pliny reports a
curious story: '

P. Miniatin'- aim demptam Marsuae coronam ejloribus eapiti suo inposiiissct atque ob id
duci aim in muada rriumuiri iussissent, appellauit tribiinis plebis, uec intercessere illi.
Publius Munatius took a chaplet of flowers from the statue of Marsyas and put
it on his o\\ n head. Ordered by the triumriri [nipir<iles| to be put in chains for his
offence, he appealed to the tribunes of the plebs, who refused to intervene.

But Pliny goes on to quote a letter of Augustus complaining about his daughter
Julia for doing just the opposite—not taking a garland from Marsyas but

123 Horace Odes 3.6.1-8 (the last of the ‘Roman Odes’); Dio Cassius 53.30.1-4 for Augustus’
illness.
124 Dio Cassius 54.1.1—I, Augustus Res gestae 5.1—2. It was to Terra Mater, not Ceres, that
sacrifice was made at the Secular Games in 17 BC (CIL 6.32323.136, but Ceres in Horace
Carmen saeculare 30); and it may have been an Augustan innovation to have the lectisteniium
to Ceres on the festival day of Tellus, at her temple on the Carinae (Amobius 7.32; Degrassi
1963.136-7, Fasti Praenestini on 13 December).
125 Pliny Nat. Hist. 21.8 (on the Uremia of wearing garlands during the day).
 134 UNWRITTEN ROME

putting one on him. The natural inference is that her gesti:ure was a statement
of libertas, which he interpreted as licentia and luxuria.'26
The letter was no doubt his justification to the Senate for the banishment of
Julia in 2 bc. Seneca’s description of that event also implies the significance of
Marsyas as a symbol of licentia:'27

Diuus Augustus Jiham ultra impudicitiae maledictum impudicam relcgauit et Jlagitia

principalis damns in publicum emisit: admissos gregatim adulteros, pererratam noctumis
coniissationibiis ciuitatem, forum ipsum ac rostra, ex quibtts pater legem de adtilteriis
tulerat,Jiliae in stupra placuisse, cotidiammi ad Marsyam concurstint, cum ex adultera in
quaestuariam uersa ins onmis licentiae sub ignoto adultera peteret.
The deified Augustus banished his daughter, whose immorality went beyond
her bad reputation for it, and he brought the scandals of the imperial house into
the open—the adulterers admitted in droves, the nocturnal parties roaming
round the city, the Forum itself (and the Rostra from which her father had
proposed the law against adultery) as the favourite place for her sexual
debauchery, the daily gathering at the statue of Marsyas, where from being an
adulteress she became a paid harlot, and claimed the right to every licence
under a lover she didn’t even know.

Marsyas was the servant of Antony’s patron god; the most conspicuous of
Julia’s lovers, now duly executed for treason, was Antony’s son.128

VIII
Apolline authority was still being exercised, but progressively from now on
matters slipped out of Augustus’ control. By ad 4 atrox fortuna had robbed him
of his sons Gaius and Lucius; by AD 7 Agrippa Posttnnus had been disinherited
and Pannonia was in revolt; what was worse, throughout that three-year
period there was disastrous famine in Rome.129
This was probably the time when Ovid was writing his Fasti. In that poem
Augustus Apollo is hardly mentioned;12" Ceres, Liber and Flora, on the other
hand, are given extensive treatment, as is required by their respective festivals.
For the Liberalia in book 3, Ovid elaborately declines to narrate the birth and
res gestae of the god, preferring to concentrate on the honey-cakes (Jiba from
Liber) provided at his festival:131

126 Pliny Nat. Hist. 21.9: apud nos exeinplum licentiae huitis non est aliud quant fiha diui AtigU-ti-
cuius luxuria noctibus coronation Marsuam litterae illius dei geinimt.
127 Seneca De benejlciis 6.32.1; cf. Velleius Paterculus 2.100.3 on Julia’s licentia (.qutdquid Mere
pro licito uindicans).
128 Velleius Paterculus 2.100.4, Tacitus Annals 4.44.3, Dio Cassius 55.10.15.
129 Dio Cassius 55.22.3, 55.26.1-3, 55.27.1, 55.31.3-4, 55.33.4.
130 Only at 4.951 and 6.91.
131 Ovid Fasti 3.715, 724-6.

I
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 135

nec refcram Semelen . ..

. . . non est carminis hums opus;
carminis huius opus causas exponere quare
uitisator populos ad sua liba uocet.
I shall not tell of Semele [or the birth from Jupiter’s thigh, or the conquest of
Thrace, Scythia and India, or Pentheus or Lycurgus or the Tyrrhenian pirates];
that is not the business of this song. The business of this song is to explain the
reasons why the Vine-planter invites the people to eat his honey-cakes.
Feeding the People is what matters. For the Cerialia in book 4, Ovid tells the
old story of Demeter and Persephone, ending with the goddess’ restoration of
the grain harvest:132
turn demum uoltiunque Ceres animumque recepit
imposuitque suae spicea serta comae;
larqaque prouenit cessatis messis in amis,
er uix conqestas area cepit opes.
Then at last Ceres recovered her looks and her spirit and put wreaths of com-
ears in her hair; a generous harvest came forth in the fallow fields, and the
threshing-floor could hardly hold the heaped-up riches.

And for the Floralia in book 5, Ovid has the goddess herself tell how she
punished the Senate’s neglect of her:133

'nos ./uo.ji'c tanqit honor; festis qaudemus et aris,

turbaque caelestes ambitiosa sumus...
at si neqleaimur maquis iniuria poenis
soluitiir et iustuni praeterit ira modum...
lonqa relerre tnora est correcta obliuia damnis.
me quoqtte Romani praeteriere patres.'
‘We too are touched by honour; enjoying festivals and sacrifices, we dwellers in
heaven are an ambitious lot... But if we are neglected the crime is paid for
with heavy punishments, and our anger goes beyond the just limit. .. It would
take a long time to list examples of forgetfulness corrected by disasters. Even I
was once overlooked by the Roman Senate.’

And the punishment she inflicted, in 173 BC, was famine. We should
remember that it was the Senate who had authorised the programme of temple
restoration in 28 BC, and thus left Ceres, Liber and Flora for all this time
without their temples in the Circus Maximus. Not even the subtlest of poets

132 Ovid Fasti 4.615—18; for the famine, see Metamporphoses 5.474-86 (from Homeric Hymn to
Ceres 305-13).
133 Ovid Fasti 5.297-8, 303-4, 311-12.
 136 UNWRITTEN ROME

could blame the princeps; but whoever was thought to be responsible, it is hard
to believe that in a time of famine Ovid’s readers failed to notice the topicality
of these passages.
This time, it seems, Augustus gave in. The temples were rebuilt, and reded­
icated in due course by Tiberius. But it would be wrong to see that as a victory
in the long ideological struggle. Palatine Apollo at the house of Augustus
retained his supremacy, and libertas was now defended by the Praetorian
Guard.134 Marsyas in the Forum still presided over the welfare of the city, but
now it took the form of imperial munificence, celebrated in grand marble
reliefs for the citizens to admire.135 The mythology of Liber and Marsyas had
been a republican phenomenon, an example of what Tacitus called plebis et
optimatiiim certaniina. It still had real life in it for the first generation or two of
the Augustan principate, but gradually the reality of power turned all such
phenomena into mere simulacra libertatis (another Tacitean phrase).136

IX
Inevitably, given the nature of the evidence, this story' of Liber. Indi and libertas
has been uneven and incomplete. And only now, at the end of it, are we in a
position to ask how it began.
According to Dionysius of Halicarnassus, the temple and games of Ceres,
Liber and Libera were vowed in 496 bc by the dictator A Postumius Albus,
victor of the battle of Lake Regillus, and inaugurated three years later. That
tradition, unknown to Livy, can be traced to the history or A. Postumius
Albinus in the mid-second century BC; but since the historian was a cousin of
the consul Sp. Postumius Albinus who in 186 BC was responsible for the
suppression of the Bacchanalia, his testimony on the origin of the cult is not to
be accepted without question.137 There is no contemporary evidence, but
events in Athens may suggest another possibility.
According to the Athenaion Politeia, the Pisistratids were expelled from
Athens in the archonship of Harpaktides, i.e. 511/10 BC.138 The dramatic
contests at the City Dionysia’, which took place towards the end of March at
the precinct of Dionysus Eleuthereus on the slope of the Acropolis, probably
began in 502/1 BC.137 The chronological sequence strongly suggests an ideo-
1

134 Josephus Antiquities 19.42 (a claim denied by an old-fashioned lover of liberty, Cassius
Chaerea).
135 Anaglypha Traiani: Torelli 1982.89-109, with figs IV. 1-16 (Marsyas at IV 8-9, 14).
136 Tacitus Annals 4.32.1, 1.77.3.
137 Dionysius of Halicarnassus 6.10.1, 6.17.2—4, 6.94.3; Wiseman 1998.35-6.
138 [Aristotle] Athenaion politcia 19.6; Rhodes 1981.191—9.
139 IC 2-.2318; Pickard-Cambridge 1968.101-7. Alleged earlier dates for victories by lhespis
and Choerilus are not reliable: West 1989.

J
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 137

logical connection between Eleuthereus and the freedom of Athens from the
tyranny of Hippias.140
Compare Rome. According to Polybius, the Tarquins were expelled
twenty-eight years before Xerxes’ invasion of Greece, i.e. 508/7 BC.141 The
Liberalia took place on 17 March at the temple of Ceres, Liber and Libera on
the slope of the Aventine overlooking the open space of the Circus Maximus.
At least by the first century BC, Liber could be identified as Dionysus
Eleuthereus.142 It is tempting to infer that the association of Liber and libertas
had the same origin as that of Eleuthereus and EXEV0epia, celebrating the
expulsion of Tarquin three years after the Athenians expelled Hippias.
Unfortunately, such synchronisms are suspiciously common in early
Roman history. The expulsion of the tyrant Tarquin/Hippias is paralleled by
the ambush and heroic death of the three hundred Fabii/Spartans, or the exile
by an ungrateful people of Coriolanus/Themistocles.145 The details of partic­
ular episodes are frequently reminiscent of Greek history—the war provoked
by the kidnap of prostitutes, the speech given at the congress of the enemy
alliance, the invaders who spare the defending commander’s estates, and so
on.144 The Tarquin story itself contains two classic examples, with Sextus at
Gabii playing first Zopyrus at Babylon and then Periander consulting Thrasy­
bulus.145
In particular, as Attilio Mastrocinque has pointed out, the events immedi­
ately following the expulsion seem to be largely borrowed from Herodotus.14’’
Porsena comes to bring back Tarquin, but abandons the siege because he is
impressed by the Romans’ courage and love of freedom; so Cleomenes came
to bring back Hippias, but abandoned the invasion because he was impressed
by the evils of tyranny.147 The Veientes attack Rome to bring back Tarquin,
there is a closely-fought battle, and Silvanus (or Faunus) speaks in favour of the
Romans; so the Persians attacked Athens to bring back Hippias, there was a
closely-fought battle, and Pan announced that he would fight on the Athe-

140 See Connor 1089, Seaford 1994.243—1.

141 Polybius 3.22.1; Walbank 1957.665-9. The date probably comes from Eratosthenes, who
used the dlil|iuoi5 of Xerxes as one of the fixed points of his chronology (FGrH 241 Fl (a)
= Clemens Alexandrinus Stramateis 1.138.1—3) and evidently included material on Rome
(F45 = Servius anctus on Aeneid 1.273); cf. Dionysius of Halicarnassus 1.74.2 for the impor­
tance of Eratosthenes for synchronising Roman and Greek chronology.
142 Alexander Polyhistor FGrH 273 F109 (Plutarch Morab’d 289a = Quaestiones Roinanae 104).
143 See respectively Pliny Nat. Hist. 34.17, Aulus Gellius 17.21.13, Cicero Brutus 41-3. On the
Fabii parallel see Soltau 1909.88-90, Ogilvie 1965.359-60.
144 See respectively Livy 2.18.2 (Aristophanes Achaniians 523-9), Dionysius of Halicarnassus
5.50.3 (Thucydides 1.72.2), Livy 2.39.5-6 (Thucydides 2.13.1).
145 Dionysius of Halicarnassus 4.55-6 (Herodotus 3.154, 5.92-3).
146 Mastrocinque 1988.33—1.
147 Herodotus 5.90—3 (including the story of Thrasybulus and the poppies, n. 145 above).
 138 UNWRITTEN ROME

nians’ side.148 Tarquinius ofCollatia is consul in the first year of the Republic,
but because his relationship with the exiled tyrant arouses suspicion, he goes
into voluntary exile; so Hipparchus of Collutos was archon in Cleisthenic
Athens, but because his relationship with the exiled Hippias aroused suspicion,
he was the first to go into ten-year exile under the new system of ostracism.149
As the last example shows, the template was not Herodotus alone;
Hipparchus’ ostracism probably comes from Androtion or one of the other
Atthidographers.15" It was evidently the events themselves, not just one narra­
tive of them, which provided the inspiration. Who did it? Possibly Fabius
Pictor, possibly Timaeus or some other Greek historian who dealt with Rome;
the involvement of the exiled Tarquins with Aristodemus of Cumae makes it
likely that the outline of their story was remembered — and elaborated?—in
Greek historiography.151
What these examples show is not just a means of creating plausible history
out of little or nothing (though it is that), but also a predisposition to think of
the Roman experience in Athenian terms. Rather than some Greek historian,
it may have been the Romans themselves who constructed their freedom
narrative on the lines of Athenian democracy. Late in the fourth century, at the
time of the Samnite wars, they set up a statue of Alcibiades at the Comitium.
A few years later, they placed the new temple of Victor, on the Palatine above
the cave of Pan (the Lupercal), and incorporated into the complex a newly-
fortified access to the hill (the clitttts Victoriae), thus reproducing the topography
of the Nike temple in the Propylaea of the Athenian acropolis. Nike, Pan and
Alcibiades had made the Athenians victorious; perhaps they would do the same
for Rome.152
By that time at least, we may reasonably imagine that Liber, hbcrtas and the
Libernlia were firmly associated in the Romans’ minds with Dionysus
Eleuthereus, eX£l)0Epta and the City Dionysia. But did the association go back
to the actual events—whatever they were—of the end of the tyranny at
Rome?
Given the Cumae connection, which brought the story' into Greek histori­
ography, Polybius’ date for the expulsion of the Tarquins may be roughly

148 Herodotus 6.105.2—3, Pausanias 1.28.4 on Pan; cf. Livy 2.7.2 (Silvanus), Dionysius of Hali­
carnassus 5.16.2-3 (Faunus).
149 [Aristotle] Athenaion politeia 22.4, cf. Livy 2.2.3-10; the version in Dionysius and Plutarch
makes Collatinus deserve his exile (Wiseman 1998.81—2 [and pp. 299—301 below]). The
particular parallel of Tarquinius Collatinus and "IJtJtapXOg KoXXuTEUg was pointed out by
Griffiths 1998.
150 For the sources of the Athenaion politeia, see Rhodes 1981.15-30.
151 Mastrocinque 1988.34—5. Detailed Cumaean history (possibly Hyperochus, FGrH 576)
inferred from Dionysius of Halicarnassus 7.3—11 on Aristodemus; Altbldi 1965.56—72.
152 Alcibiades: Pliny Nat. Hist. 34.26, Plutarch Numa 8.10. Victory temple complex: pp. 57-8
above. Pan and Marathon: n. 148 above.

I
 liber: myth, drama and IDEOLOGY IN REPUBLICAN ROME 139

right. As for Liber, his temple was decorated by the famous artists Damophilus
and Gorgasus, whose work was attested by a Greek inscription; no remains
survive, but a likely parallel is provided by the early fifth-century temple of
Mater Matuta at Satricum, evidently the work of itinerant Greek master­
craftsmen.153 So the expulsion and the temple may indeed have been roughly
contemporary, and Liber could have signified liberals even as early as about 500
BC.
Because of the absence of good contemporary evidence, and the conse­
quent impossibility of deciding the historical value, if any, of the
foundation-legends of the Republic,1’4 we cannot be sure that the ideological
significance of Liber and his Indi goes back to the very beginning of the
Republic. But it is certainly possible.

153 Pliny Nat. Hist. 35.154; Lulof 1996.204-8. Mater Matuta was identified as Ino, the sister of
Semele, and was therefore part of Liber’s story (Ovid Fasti 6.484, si damns ilia tua est).
154 I wish I could accept Mastrocinque’s estimate of the historicity of the literary tradition
(1988.235), but I think it is not enough to ‘rassegnarsi, J'aute de mieux, ad accettare senza
entusiasmo i dati della tradizione’ (182) just because modem reconstructions are no more
plausible than the ancient ones (213). Nor can 1 follow him in using Macrobius Saturnalia
1.7.34-5, on Brutus’ supposed reform of the supposedly Dionysiac Lares Compitales ritual,
as evidence for a historical connection between Liber and the new Republic (Mastrocinque
1988.37-41, 145—50; appendix on Liber at 245—75).
 CHAPTER SIX

The Kalends of April

i
The Kalends of April were one of the three days in the year when a man was
expected to give a woman a present (the other two were her birthday and the
Sigillarid). As Ovid points out in his Ars atitatoria, the economical seducer
should avoid those dates:1

sine dies suberit natalis sine Kalendae

qnas Venerem Marti continnasse iunat,
sine erit ornatus non, ntflit ante, sigillis,
sed regain positas Circus habebit opes,
differ opus. ..
If her birthday is coming, or the Kalends whose pleasure it is to have joined
Venus to Mars, or if the Circus, not as of old adorned with statuettes, is going
to have royal treasures on display—postpone your campaign.
Note the phrase Ovid uses to describe the ‘joining’ of Mat ch and April (the
months of Mars and Venus respectively). Appropriately for the context, it is a
sexual allusion; the first day of April is, as it were, the moment when Mars
enters Venus. The same expression is used in the Fasti, where the poet intro­
duces the 1 April ritual:2

et fonnosa Venns fonnoso tempore digna est,

Htque solet, Marti continuata suo est.
uere monet ettrnas materna per aeqtiora pttppes
ire nec Inberitas iain tininisse ininas.
Lovely Venus both deserves the lovely season and is joined, as usual, to her own
Mars. In spring she tells the curved ships to sail through the seas that bore her,
and no longer to fear the threats of winter.

i Ovid Ars ainatoria 1.405—9. For the Sigillaria (20 December, part of the Saturnalia), see
Macrobius Saturnalia 1.11.46—50, Suetonius Diuus Claudius 5, SHA Hadrian 17.3, Caracalla
1.8.
2 Ovid Fasti 4.129-32.

t
 THE KALENDS OF APRIL 141

What follows in Fasti 4, the description of the ritual of Venus Verticordia

and Fortuna Virilis, is notoriously controversial. Many important contributions
have been made in the last twenty years, since Jacqueline Champeaux’ seminal
work on Fortuna in 1982,3 but I think there is more to be said. This chapter
begins as a close reading of the Ovid passage, but then develops into a hypo­
thetical reconstruction of one aspect of the culture of archaic Italy. I offer it in
memory of Martin Frederiksen as an example of what literary scholarship and
archaeology can do for each other.4

II
Ovid Fasti 4.133—62:
rite deam colitis Latiae matresqne niimsque
et nos, qnis iiittae longaqne nestis abest.
135 anrea marmoreo redimicnla demite collo,
demite dinitias: tota lananda dea est.
anrea siccato redimicnla reddite collo:
mine alii flares, mine nona danda rasa est.
nos qnoqne sub niridi myrto inbet ipsa lanari,
140 cansaqne enr inbeat (discite) certa snbest.
litoie siccabat reroutes nnda capillos;
•I'dernnt satyri, tnrba protenia, deam.
sensit el epposita texit sna corpora [temporaj myrto;
tnta fnit facto iiosqne referre inbet.
145 dtsritc nunc qnare Fortimae tnra I 'irili
detis eo calida Igelida] qni locus umet aqua.
aicipit itle locus posito nelamine ennetas
et nitinm nndi corporis omne nidet;
nt regat hoc celetqne uiros, Fortuna Firilis
150 praestat et hoc panto Hire rogata facit.
nee pigeat tritnm nineo cum laete papaner
snmere et expressis mella liqnata Janis.
anti primnm ettpido I imns est dedneta manto,
hoc bibit; ex illo tempore mipta Juit.

3 Champeaux 1982.375—109. See also Torelli 1984.77-89; Greco 1985; Porte 1985.388—93,
458-60, 469; Coarelli 1988.293-301; Boels-Janssen 1993.321-35; Barchiesi 1994.207-16 =
1997.219-28; Menichetti 1995.61-3; Fantham 1998.115-23; Torelli 1999.65-8, 74-8,
177-8. [See now Schultz 2006.148, 202-3.]
4 Martin supervised my Oxford doctoral thesis on noui homines (1961-7). When 1 first met
him, I was a traditional text-based English classicist. I still am; but what little I know about
the archaeology of ancient Italy I owe to him, and to the British School at Rome, where I
was Rome Scholar in Classical Studies in 1962-3.
 142 UNWRITTEN ROME

155 stipplicibits uerbis illani placate: sub ilia

et forma et mores et bona fama manet.
Roma pndicitia proauontm tempore lapsa est;
Cumaeam, iieteres, consuluistis anum.
templa inbetfieri Veneri, qtiibus ordinefactis
160 inde Venus uerso nomina corde tenet.
semper ad Aeneadas placido, ptilcherrima, uoltn
respice totque tuas, diua, tuere minis.
You pay the goddess proper respect, matrons and young wives of Latium, and
you who don’t wear ribbons and long dresses.
Take off from her marble neck the golden necklaces, take off her wealth; the
goddess must be washed all over. Put back on her new-dried neck the golden
necklaces; fresh flowers, new roses, must be given her now.
Wash yourselves too, under green myrtle. She gives the order herself, and
there is a sure reason for it. Listen! She was naked, drying her dripping hair on
the shore, when a lecherous gang of satyrs caught sight of the goddess. She
realised, and put myrtle in the way to screen her body [orbrow]. That made her
safe, and she bids you remember it.
Now learn why you give incense to Fortuna Virilis in the place which is
wet with warm [or cold] water. That place receives all women without their
clothes, and sees each blemish of the naked body. Fortuna Vtnlis guarantees to
conceal it and hide it from men, and she does so when asked with a little
incense.
And don’t be reluctant to take ground poppy, in white milk and liquid
honey squeezed from the honeycomb. When Venus was first brought to her
eager bridegroom, this is what she drank, and from that moment she was a wife.
Appease her with words of supplication; in her control beauty and character
and good reputation stay fixed. Rome slipped from chastity in our ancestors
time, and the men of old consulted the old woman of Cumae. She ordered a
temple to be built for Venus, and it was duly done; since then Venus has a
name from the turning of the heart.
Goddess most fair, look always with a kindly face on the descendants of
Aeneas, and protect all your young wives.

The conspicuous ‘ring-composition’ of lines 133 (cieani, nurusqiie) and 162

(diua, mints) defines the passage as a single coherent unit. It is organised around
a sequence of imperatives (lines 135, 136, 137, 140, 145, 155), as the poet takes
on the role of master of ceremonies’.5 But it is also divided into seven recog­
nisable paragraphs:

5 A favourite technique: see also Fasti 1.663-96, 2.532-64, 2.623-38, 4.407-16, 4.731-82,
6.775-80.
 THE KALENDS OF APRIL 143

1 [133-4] definition of worshippers

2 [135-8] washing of Venus statue
3 [139-44] worshippers bathe, myrtle crown aition
4 [145—50] incense to Fortuna Virilis
5 [151-4] poppy drink, Venus as bride
6 [155-60] Venus Verticordia aition
7 [161—22] closing prayer to Venus

The first and last of these are a single couplet each; the second is a conspicu­
ously antithetical pair of couplets; the third is marked out by an internal ‘ring’,
with nos (qno)qne and inbet at 139 and 144, myrto at 139 and 143; so too is the
fourth, with turn/lure at 145 and 150 and the goddess’s name at 145 and 149.
The fourth paragraph is thus the central ‘omphalos’ of a symmetrical structure.
The construction of this complex passage makes it clear that the offering to
Fortuna Virilis is an integral part of the Venus ritual, and also that the same
group of women—Roman citizens, rich and poor—is being addressed
throughout. Ovid’s careful and deliberate presentation rules out any recon­
struction that presupposes a distinction between the women who worship
Venus and those who worship Fortuna.6 It was a single ritual with two
goddesses involved, no doubt the result of an ancient Fortuna cult being
combined with Venus Verticordia in the third or second century BC.7
The distinction that Ovid does make, in the first couplet, is sometimes
taken as being between respectable women and prostitutes.11 But that too is
surely wrong. Those who don’t wear the stola are just women definable as
hnmiliores, with no connotations of immorality; if Ovid had wanted to specify
prostitutes, he could have referred to women who wore the toga.9 As the

6 According to Fantham (1998.116-18), ‘these rituals honouring more than one deity may
well have been observed in different places’; it ‘seems unlikely that the same group is
addressed throughout’; ‘elite wives in the main ritual' are to be distinguished from ‘the
working women’s offering to Fortuna Virilis’. Contra Barchiesi 1994.209-10 = 1997.222,
surely rightly. Mommsen’s supplement to the Fasti Praenestini (wrongly attributed to
Degrassi by Fantham 1998.116) would distinguish honestiores worshipping Venus Verticordia
from hnmiliores worshipping Fortuna Virilis; but it is gratuitous, and has been rightly rejected
by all scholars since Champeaux (1982.379-80). [However, Schultz (n. 3 above) tacitly
accepts it; ‘the poet then conflates the two rituals’.]
7 Thus Champeaux 1982.383 and 390-4, Torelli 1984.78-81, Coarelli 1988.293.
8 E.g. Champeaux 1982.382 (Champeaux’ argument slides tacitly from ‘hutniliores’ to ‘liuniil-
iores et courtisanes’: 1982.384, 387, 390, 392, 393, 394, 404, 405, 407); Boels-Janssen
1993.328; Barchiesi 1994.209-10 - 1997.222; Pasco-Pranger 2006.149-52. ‘Patrician and
plebeian’ (Torelli 1984.83, 1999.67) is closer to Ovid’s formulation.
9 Cicero Philippics 2.44, Horace Satires 1.2.63 and scholiasts, Sulpicia [Tibullus] 3.16.3, Martial
2.39.2, 10.52, Juvenal 2.68—70. For ‘those who don't wear the stola’, cf. Ovid Ars ainatoria
1.31—2, 2.600; presumably the poet did not want to characterise all his female readers as
prostitutes.
 144 UNWRITTEN ROME

descendants of Aeneas are the Romans, so the ‘daughters-in-law of Venus’

(162) are the women of Rome, and the poet addresses all of them.10
The context of the third paragraph is the myth of Venus anadyomene,
already alluded to in the introductory passage (materna per aequora, 131). For
Ovid, as for Horace,11 April’s goddess is Rem/s marina. There were myrtles
growing where she emerged naked from the sea, and according to Servius she
hid herself behind the bush.12 That seems to be what Ovid asks us to imagine
(143—4); but such a banal scenario does not fit the aition at all well, since the
women who imitate her do so by wearing only myrtle garlands. Moreover, the
eleventh-century Vatican manuscript has tempora, not corpora, at 143. If that is
the true reading, and corpora a rationalising ‘correction’, then the protection
was not physical but magical.13 Crowned with myrtle, Venus was safe even
though the satyrs were watching.
Whatever the correct solution of the textual crux, that idea of magical
protection must be valid for the ritual, where it is certain that the women s
only protection was myrtle round the head. The parallel with the myth must
mean that there are men watching, like the satyrs.13’ Under normal circum­
stances, of course, the women would cover their bodies, but on the Kalends of
April it seems that wearing Venus’ myrtle garland makes ;t al! right.
Certainly the watching males are a necessary feature of the fourth para­
graph: the whole purpose of the offering to Fortuna Virilis is to make the less
beautiful women’s bodies attractive to the men (ttiros, 149). This privileged
exhibition a startling idea, but one clearly implied by Ovid’s text may
account for the goddess’s name, if ‘Fortuna Virilis’ expresses the men’s good
fortune. Where it took place depends on the solution of another textual uncer­
tainty. If cahda is the correct reading at 146, it was in the public baths; i£gclida,
it was outside, no doubt at the piscina publica. In any case, the piscina publica
must have been the site of the ritual from the late fourth century, when it was
built to exploit the water of the Aqua Appia,H until the introduction of public
baths (and warm water) in the second century BC.15

10 Cf. cimctas at 147, the middle point of the whole passage.

11 Horace Odes 4.11.15-16, cf. 3.26.5.
12 Servius on Eclogues 7.62: myrtuiii idea dicatum V'eneri uolunt quia cum a mari exisset latuit in myrto
ne nuda conspiceretur. . ,
13 Torelli 1984.83. I think Porte (1985.469) and Boels-Janssen (1993.327) are right to pre
tempora, though Porte wrongly supposes that the story is an Ovidian invention.
13a [Pare Pasco-Pranger 2006.154, who assumes without argument that men were exc u e .]
14 Since the Paestum colony had its own piscina publica (Greco 1985), 273 BC is a firm terminus
ante queni.
15 So Greco 1985.229-31, Coarelli 1988.299-300, Torelli 1999.65-7. Boels-Janssen 1993.329
less convincingly suggests the Tiber.

1
 THE KALENDS OF APRIL 145

A Praenestine cista, now in the Badisches Landesmuseum at Karlsruhe [fig.

19, p. 89 above], seems to offer a mythologised illustration of the ritual as it
may have existed in the fourth century BC.1(,The columns mark oft' the bath­
house or nymphaeum where the women are washing themselves at the
fountain; they are observed by two satyrs, one followed by Liber Pater, the
other by an unidentified goddess. The symmetrical composition makes one
think of Libera,17 but Fortuna Virilis is also a possibility.
Are we to think of men dressed as satyrs? That was certainly a familiar
phenomenon in Dionysiac cult;18 and we have visual evidence for it in two
contemporary bronze mirrors from Latiurn, one in the Villa Giulia in Rome
[fig. 10, p. 59 above], the other in Baltimore [fig. 14, p. 67 above].1'1 Marsyas in
fig. 10 is waving his tail, which must therefore be detachable, and the central
figure in fig. 14 is wearing the outfit of an actor playing a hairy satyr (visible at
the left forearm). It is not inconceivable that in the fourth century BC the
Kalends of April ritual involved a similar type of masquerade.
Since satyrs symbolise male sexual desire, the purpose of the ritual is clear
enough, and confirmed by Ovid’s fifth paragraph. The mythic aition (153-4) is
of Venus on her wedding night, taking a sedative before submitting to the
enthusiastic embrace of Vulcan.2" So the women ofRome must allow the men
to have their w ay. The nearest parallel is the Spartan custom, allegedly intro­
duced by Lycurgus, of having citizen girls dance naked at certain festivals with
the young men looking on, as an incentive to marriage by seizure (CtpJtayq).21
The community must encourage sex, but also control it. Marriage,
however defined, is the seed-bed of the Republic, producing warriors who
will defend its territory’.22 It is surely not accidental that the two festivals of

16 Jurgeit 1999.528-33, no. 885. It is no. 22 in the standard catalogue (Battaglia and Emiliozzi
1979.95-7, tav. cxiii-cxvii), but the illustrations there include some nineteenth-century
additions; see Jurgeit 1992.85-94. See Menichetti 1995.61-2 for the relevance of Fortuna
Virilis.
17 See p. 86 above.
18 See Plato Luu's 7.815c (nearly contemporary with the cista): p. 108 above.
19 Etmskische Spiegel 5 (1883) Taf. 43 and 45.
20 Sedative: Champeaux 1982.381 and 387-8, Porte 1985.459, Boels-Janssen 1993.326,
Barchiesi 1994.213 = 1997.225; Torelli 1984.83 assumes it is an aphrodisiac. The ‘eager
husband’ cannot be Mars, as assumed without argument by Borner 1958.218, tentatively
followed by Fantham 1998.121.
21 Plutarch Lycurgus 14.2 (E101OE ... TCtg KOpag yilgvug TE jtOgltELIElV Kai Ttpo; ispoig
riotv op/EioSai Kai adEtv rtbv veojv jtapovrwv Kai Oempevow); cf. 15.3 for apnayi).
The Spartan custom clearly influenced Plato (Lin’s 771e-772a).
22 Cicero De qfficiis 1.54: mim cum sir hoc natura commune animantium, ut habeant lubidinem procre-
andi, prima societas in ipso conittgio est, proximo in liberis, deinde ttna donuts, communia omnia; id
autem est principimn urbis et quasi seminarium rei publicae. Cf. Catullus 61.71-5, on Hymenaeus:
quae tuis careat sacris | non queat dare praesides | terra fmibus. For the marriage formula liberonun
quaerendomm causa, see Festus 312L, Valerius Maximus 7.7.4, Suetonius Diuus lulius 52.3,
Aulus Gellius 4.3.2.
 146 UNWRITTEN ROME

Carmentis the goddess of childbirth (11 and 15 January) come just over nine
months after the Kalends of April.23 Various stories of Roman history or
pseudo-history betray anxieties about the birth-rate;24 no less conspicuous are
anxieties about unregulated sexual activity, as in the Dionysiac rites banned in
186 bc.25 The combined festival of Fortuna Virilis, who gives men what they
want, and Venus Verticordia, who turns women’s hearts to chastity (Ovid’s
sixth paragraph),26 may be seen as an expression of those conflicting priorities.

Ill
So far, I have deliberately restricted the discussion to the Ovid passage and
what can be inferred from it. But other authors also refer to the Kalends of
April, if only briefly:
1. Verrius Flaccus Fasti Praenestini (ftiscriptiones Italiae 13.1.126—7):
frequenter initheres supplicant Fortunae Firili, liutniliores etiam in balincis, quod tn iis ea
parte corpora's] utique niri nndantur qua feniinarutn gratia desideratur.
The women en masse pray to Fortuna Virilis, those of lower rank even in the
baths, because there men reveal that particular part of the body by which the
favour of women is sought [or the beauty of women is desired |.

2. Macrobius Saturnalia 1.12,15:

non tauten negat Ferrins Flaccus hoc die postea constitution nt inatronae Feiien sacrum
facerent, cuius rei causain, quia Imic loco non coniienit, praetcreundimi est.
However, Verrius Flaccus agrees that on this day it was later laid down that the
married women should sacrifice to Venus. I must pass over the reason for it,
because in this place it is not appropriate.

3. Plutarch Numa 19.2:

.. .bsuiEpov 6e tov AjrpiXXtov, Ejrcbvupov ovto Tfjq AqpobiTi]?, ev to
Guotxn te Tfj 0eo) icai iatg KaXavbatg al yuvaikEg saTECpavtoHEvai
pvpaivp Xovoviat.

23 Carmentis and childbirth: Ovid Fasti 1.627-8, Plutarch Romulus 21.2 and Moralia 278b-c
(Quaestioues Romauae 56), Augustine City of Cod 4.11.
24 E.g. Ovid Fasti 2.429-48 (Romulus); Festus 478L (Tarquinius Superbus); Orosius 4.2.2,
Augustine City of God 3.18 (276 Be); Ovid Fasti 1.619-28 (195 Be).
25 Livy 39.8.6-7 (stupra promiscua)-, cf. 13.10, 13.12, 15.9, 15.12, 18.5-6.
26 Valerius Maximus 8.15.12: quo facilius uirginum midienimque mens a libidine ad pudidtiain
comierteretur. Ovid may be referring to the temple built in 119 BC (Obsequens 37); but since
the consultation of the Sibylline books in that year is recorded only in the context ot human
sacrifice (Plutarch Moralia 284a-b — Quaestioues Romauae 83), it may be better to take it as a
reference to the earlier dedication of a statue (Valerius Maximus loc. cit., Pliny Nat. Hist.
7.128). The date of the latter is evidently before 204 BC, since Pliny puts it before the story
ofQ. Claudia; however, Torelli (1984.79-81) argues for 178 BC.
 THE KALENDS OF APRIL 147

[Romulus named] the second month April after Aphrodite. In that month they
sacrifice to the goddess and on the Kalends the women bathe garlanded with
myrtle.

4. John Lydus De mensibus 4.65:

xalc; Toivov Kakavbat? Aixpikiai; al OEpval yvvatKWV (iirep opovoia?

Kai [liov oo'xppovog Exiptov xijv Atppo6ixr)w al 6e xoO nXi]0ovg
yvvatKEc; ev xotg xtbv avdpcbv PaXaveion; eXovovxo jxpoq 0EpajTEiav
auxfjc; pvpoivt] Eoxsppsvai.
On the Kalends of April the women of rank used to honour Aphrodite for the
sake of harmony and a respectable character; but the women of the people used
to bathe in the men’s baths, garlanded with myrtle in honour of the goddess.

All four passages clearly derive from the antiquarian tradition. Only Verrius
Flaccus (item I) refers to Fortuna Virilis; for the others, it is solely a festival of
Venus, and since the day is named as ‘Veneralia’ in the Calendar of Furius Filo-
calus, it is normally (and plausibly) assumed that after the Augustan age
Fortuna’s share in it was gradually forgotten.27 Lydus in item 4 was probably
using the Verrius Flaccus tradition (item 1), but assuming that the goddess
concerned was Venus. Since Verrius Flaccus and Plutarch use the present
tense, Macrobius and Lydus the past, the whole ritual was probably obsolete by
the fourth century ad.
What we can infer from these passages is the way sophisticated Augustan
Rome handled a ritual that had become embarrassing. Its erotic nature is
admitted in Verrius Flaccus’ prim little note in the Fasti Praenestini (item I),28
and implied by Macrobius’ reluctance to discuss the subject (item 2). How did
a lady of the Roman elite carry out her ritual obligations on the Kalends of
April? If she took the necessary bath at home, the proper place for the next
stage would surely be the leans genialis in the atrium.-' Named after Genius, the
god of human reproduction, it was the matroua’s particular place in the house?"
That may seem public enough, but of course the doors could be kept closed
and witnesses excluded. For the humiliores who had neither an atrium nor their
own water supply, there was no alternative to the traditional public display,
now held in the baths.

27 So Champeaux 1982.391-2, Boels-Janssen 1993.323-4; Fantham 1998.115 assumes it was

already the Veneralia in Ovid’s time.
28 It ‘seems absurd’ (Fantham 1998.120), but that is all the more reason to take it seriously as
evidence.
29 Horace Epistles 1.1.87 (lectus genialis in aula est), Cicero Pro Cluentio 14, Lucan 2.356-7—all
in the context of marriage.
30 Genius: Festus (Paulus) 83L; cf. Varro in Augustine City of God 7.13, Festus (Paulus) 84L,
Censorinus 3.1. Natrona's place: Laberius fr. 30R (Aulus Gellius 16.9.4), Propertius 4.11.85—
6. Asconius 43C.
 148 UNWRITTEN ROME

There may be an analogy in the flagellation ritual of the Lupercalia on 15

February. In the time of Plutarch and Juvenal, the women just held out their
hands to be struck, ‘like children in school’.31 If that was already the custom in
the Rome of Augustus, Ovid’s aetiology ignores the fact; he refers only to
women ‘offering their backs’.32 So too on the Kalends of April, he gives the
respectable ladies no option of avoiding the public exposure of their bodies.
Alessandro Barchiesi is surely right to see that as mischievous.33

IV
The temple of Venus Verticordia, referred to by Ovid in his sixth paragraph,
was situated in the valley of the Circus Maximus (ttallis Murcia), presumably
close to the piscina pnblica. The evidence is in a passage of Servius, commenting
on Virgil’s reference to the rape of the Sabine women:34

uallis autein ipsa tibi circenses editi sunt idea Murcia dicta est, quia quidam uicinum
inontein Murctun appellation iiolunt; alii quod famuli Veneris Verticordiae ibi fuerit,
circa quod neiniis e niurtetisfuissent, iiuuutata littera Murciam appeliatain; alii Murciam
a Murcido, quod est marcidiiin, dictam uolunt; pars a dea Murcia, quae cum Bacchanalia
essentfurorein sacri ipsius iniirciduni faceret.
The valley itself where the circus games were held is called Murcia for this
reason, because some think the nearby hill is called Mons Murcus; others say
that because this is where the temple of Venus Verticordia was, around which
there had been a myrtle grove, [the valley] was called Murcia by the change of
a letter; others think it is called Murcia from Murcidus, i.e. marcidus (rotten);
some think it is from the goddess Murcia, who, when the Bacchanalia were
taking place here, made slow the madness of the god’s ritual.
The myrtle grove is located by Varro at the temple of Venus Murtea, which
must therefore be another name for the same goddess.35 It can hardly be an
accident that this was where the story of Rome’s primeval ‘marriage by
seizure’ was set.
If our inference from the Ovid passage is valid, the Fortuna Virilis cult must

31 llutarch Caesar 61.2, Juvenal 2.142; contrast Gelasius Adm Amir. 16 (CSEL 35.1.458).
inatronae nudato corpore publice uapidabant. See now Stefano Tortorella in Carandini and
,, n -J r‘ 2000-2‘,4~55> especially the illustrations at 246, 252 and 253.
,7 , V1 . as,‘ 2-445-6, iussae sua terga puellae | pellibns exsectis percutienda dabant.
33 Barchiesi 1994.209-10 = 1997.222; cf. 1994.212 for his malizia (1997.224, ‘mischief).
34 Servius audits on Aeneid 8.636; Coarelli 1988.95-7.
35 Varro De lingua Latina 5.154 on ad Murciae: alii dicunt a nuirteto declination, quod ibi id fuerit;
C'l‘"^"es"x'"'" "la"et’ 9"od ibi M medium etiam nunc Murteae Veneris. Cf. Pliny Nat. Hist.
' ~. *n context of myrtles and the Sabine women story, 119—20): quin et ara uetus fiiit
I- enen Murteae quam nunc Murciam uocant.

I
 THE KALENDS OF APRIL 149

have been in the same place. Plutarch mentions it in his list of the Roman
aspects of the goddess in De fortuna Romanoruin:}b

eoxt be Kat napa xdv xfjc; Acppobixt^ entxakapiov ptogov appevog

Tv/qg etbog. [Ejxixakaoiov Bemadakis]
... and by the altar of Aphrodite Epitalarios there is a shrine of male Tyche
[Fortuna VirilisJ.
That evidently gives us yet another name for the goddess of the myrtle grove;
but why should Venus be associated with a basket (xdkapog)? Mario Torelli
sees an allusion to the Bacchic cista inystica containing the phallus, and given the
cult’s Dionysiac associations (the watching satyrs, and Servius’ reference to the
Bacchanalia), that is certainly a possibility. Filippo Coarelli, however, suggests
a better solution, one that brings in the site’s association with the story of the
rape of the Sabines.37
The shout of good wishes at Roman weddings was 'Talas(s)io'—or
'Thalas(s)io', as it was sometimes spelt.38 (All the spellings presumably render an
archaic Talasio from a time when Latin did not aspirate or duplicate conso­
nants.) Many authors explain it as the dative form of the name of an influential
Roman whose agents seized the most beautiful of the Sabine girls and carried
her off; challenged by rival suitors, they shouted that she was ‘for T(h)alas-
sius’.3’ It seems an improbable name for a Roman, even one of Romulus’
time, and it is not surprising that the antiquarian tradition offered a different
explanation:
1. Festus 478—SOL:

[t<i/rts]sioriein m nuptiis I arm ait [siypimn esse /anipiii, xdkapov, id est quassilluni;
i[nde eiiiin Si>]/irmn appellari talassionem. [Sed ... | historianim scriptor Talassium
[<"'•.. |
Varro says that ‘talassio' at weddings is talaros, the sign of wool-working, i.e. a
basket; that is the reason for the custom of calling ‘talassio'. But . .. the histo­
rian says [‘girl for Talassius’ story’ follows].

2. Plutarch Moralia 27 If (Quaestiones Romanae 31):

bid xi 6 jrokvOpvkqxog cibexai Takdoiog ev xotg ydgotg; jtdxEpov tkjtd

Ti'ig Tuuotag; Kai yap xdv xakapov xdkaoov bvopd^ovof Kai xfjv

36 Plutarch Moralia 323a (De fortuna Roinaiionun 10).

37 Torelli 1984.78; Coarelli 1988.298-9.
38 Unaspirated: Catullus 61.127, Festus 478-80L, Sidonius Apollinaris Epistles 1.5; Plutarch
Quaesriones Romanae 31, Romulus 15.1-4, Pompey 4.3. Aspirated: Livy 1.9.12, Festus 492L,
Servius on Aeneid 1.651; Martial 1.35.6—7, 3.93.25, 12.42.4, 12.95.5. MSS vary: [Virg.|
Catalepton 12.9, 13.6.
39 Livy 1.9.12, etc; Servius on Aeneid 1.651 makes Tlialassio the nominative form of the name.
 150 UNWRITTEN ROME

vvpipriv elaayovTsg vaKog frrtocrrparvvuaaiv*' ami) 6’ siacpEpEi pev

TjAaicaTT)v icai Tijv axpaKTOv, £ptq) 5e rf|v 0vpav jtEpunecpei top
avbpog. fj to XsyopEvov bird twv iuTopiKojv (xXt)0e£, 5ti...
Why is the famous ‘talasios’ sung at weddings? Is it from spinning? For they call
the wool-basket talasos. And when they bring the bride in, they spread a fleece
beneath her; she herself brings a distaff and the spindle, and she uses wool to
garland the bridegroom’s door. Or is what the historians say true [‘girl for
Talassius’ story follows]?

3. Plutarch Romulus 15.2-4:

Ss^Tiog 6s SuXXag 6 Kapytiboviog, oute pouomv ovte yapiToiv EjnSfiiig

dvijp, eXeyev fjpitv otl Tfjg apjtayfjg ouv0r|pa ti]v cf covf]v eSidke ravrriv 6
'PwpnXog. cmavTEq ofiv tov Takaaiov E0da>v ot Tag itap0Evoug
Kopi^ovTsg* Kai bia touto xotg yapoig JtapapsvEi to s'0og. oi 6e
jtXeiotoi vopi^ouotv, <nv Kai ’Id[3ag eoti, itapaKkrpiv slvat Kai
jrapaKskETiaiv Elg tpiXspyiav Kai xakaoiav, o&ta) tote xotg 'EXXr]viKoig
dvopaoi tojv ’iTaXtKwv ^TtiKE/upsvcov. Et 6e touto lu) Xeyetoi KaKwg,
aXX’ E/pcovTo 'Pwpaioi tote tq> dvopaxi Tijg TaXauiag, KaOartEp
ETEpav av Tig aixiav eikcioeie Jti0avcoTEpav. ejtei yao ol Saflivoi Jtpog
Toiig 'Pcopaioug TtoXspijaavTsg 6u]XXayr]uav, eyevovTo uvv0f]Kai JtEpi
twv yuvaikajv, biting p]6ev aXXo spyov xoig dvdpaaiv ij til JtEpi Tijv
TaXaoiav uitoupycbai.
[‘Girl for Talassius’ story.] But Sextius Sulla the Carthaginian, no stranger to
either the Muses or the Graces, told me that Romulus gave this word as the
signal for the abduction; so those who were carrying oil the girls kept shouting
‘talasios’, and that is why the custom survives at weddings. Most authorities,
including Juba, think that it is an exhortation and encouragement to diligence
and spinning, since Greek words had not yet been submerged by Latin ones.
And if this is not mistaken, and the Romans did at that time use the word talasia
as we do, then one might conjecture a different and more plausible explanation:
when the Sabines made peace after their war with the Romans, there were arti­
cles in the treaty about the women, to the effect that they should do no work
for their husbands except for tasks involved with spinning.
It is possible, as Coarelli suggests, that Plutarch’s ‘Aphrodite Epitalarios
belongs in this aetiological context. In that case Venus the Turner of Hearts
(Verticordia), Venus of the Myrtle Grove (Murtea), would be also Venus ot
the Spinning-Basket.
However, item 3 above incorporates a third explanation, otherwise unat­
tested. The learned Sextius Sulla thought the cry was Romulus’ signal for the
‘marriage by seizure’. Unfortunately Plutarch doesn’t explain why the word
would have been appropriate in that context, but I think a reason can be

i
 THE KALENDS OF APRIL 151

found. Let us suppose that baskets have nothing to do with it, that the
rhotacism of items 1 and 2 (TCtkapoq = TaXaaoi;) is an unnecessary expedient,
and that Talasio originally meant Thalassio—that is, ©akaooicp, ‘for
Poseidon’.4"
The games to which Romulus had invited the Sabines and their daughters
were in honour of Poseidon Hippios, Latinised as Neptunus Equester.41 The
god was identified with Census, whose underground altar was in the valley of
the Circus Maximus at the point where the chariots turned. The turning-post
was called nietae Mttrciae, just opposite the myrtle grove of Murcia who was
also Venus Murtea.42
The reason for Poseidon Hippios was brilliantly revealed by Fausto Zevi in
an article demonstrating the historicity of Rome’s 'Corinthian kings’, the
Tarquinii.43 If the first Indi circenses were set up by the son of Demaratus, as the
tradition insists,44 then naturally they would be on the model of the Isthmian
Games, and in honour of the same god. The valley flooded every winter, when
you needed a boat to reach the Aventine,4’ so the idea that it was somehow
Poseidon’s realm was not as strange as it seems today. It may explain why the
scalae Cad, down from the Palatine to the Circus Maximus, were called ‘the
steps of the Fair Shore’.46 At the bottom of those steps, at or near the Lupercal,
was a building decorated with sea-shells (perhaps a iiymphaeiini), which its
sixteenth-century discoverers naturally identified as a temple of Neptune.47
It Consus was Poseidon, and his altar was next to the myrtle grove, and the
goddess of the myrtle grove was not only Venus Murtea but also Aphrodite
Epitalarios, then perhaps we can explain that mysterious epithet as another
unnecessary rhotacism. Was she originally Epitalasios, an early Latin transliter-

40 For 0akucioio£ as Poseidon, see Aristophanes Hasps 1519 (= adesp. fr. 69 Nauck),
Plutus 396.
41 Dionysius of Halicarnassus 1.33.2, Livy 1.9.6, Plutarch Romulus 14.3, Moralia 276C (Quaes-
tiones Romanae 48), Servius aiutus on Aeneid 8.635, Servius on Aeneid 8.636, Lydus De
niagistratibus 1.30.
42 Dionysius of Halicarnassus 2.31.2, Plutarch Romulus 14.3, Tertullian De spectaadis 5.7, 8.6
(apud nietas Murcias, cf. Apuleius Golden Ass 6.8.2); Humphrey 1986.60-2, 95-7, 258-9,
290-1. For the place-name ad Murciae, see Varro De lingua Lilina 5.154, Livy 1.33.5, Inscrip-
tiones Italiae 13.3.60 and 78 (elogimn of M’. Valerius Maximus).
43 Zevi 1995.307-8. [See below, p. 233 n. 18.]
44 Cicero De republica 2.36, Livy 1.35.7-9, Dionysius of Halicarnassus 3.68.1, De iritis illustrilms
6.8.
45 Cicero Ad Quintuni J'ratreni 3.5.8 (54 BC: inagna tris aquae usque ad piseiiiam publicum), Varro De
lingua Latina 5.43; cf. Varro De lingua Latina 5.44, 5.156, Propertius 4.9.5-6, Tibullus
2.5.33—1, Ovid Fasti 6.405-8, Plutarch Romulus 5.5 (Velabrum).
46 Plutarch Romulus 20.4: itctpct TOOL kEyopEVOUC [loOpou; Ktliffc UKTljc.
47 Lanciani 1990.230: discovered by Ulisse Aldovrandi in 1549 between S. Teodoro and S.
Anastasia. [It was rediscovered in 2007, and identified—prematurely, 1 think—as the
Lupercal itself.]
 152 UNWRITTEN ROME

ation of EJllQaXaoOLO^? That would at least account for the myrtles: if the
flooded valley was thought of as the sea, then this is where ‘marine Venus
emerged naked before the satyrs’ lustful eyes. And here, as Plutarch says, was
the shrine of Fortuna Virilis.

V
According to the earliest surviving version of the myth, Aphrodite came ashore
on the coast of Cyprus.48 The goddess was honoured there by a ritual of‘sacred
prostitution’, as we know from a passage in Justin:49

mos emt Cypriis uirgiues ante nuptias statutis diebus dotalem pcauiiam qnaesitnras in
qtiaestuin ad Urns man's mittere, pro reliqua pudicitia libameiita I'etieri solntnras. hanim
igitur ex miuiero LXXX admodum ttirgiues raptas minibus inponi Elissa ittbet, tit et
innentus matrimonia et ttrbs siibolem habere posset.
It was the custom among the Cypriots to send their virgins before marriage to
the sea-shore on fixed days, for employment to earn money for their dowries,
and to pay a first-fruit offering to Venus in return for chastity thereafter. So
Elissa ordered as many as eighty of these girls to be taken and put on the ships,
so that her young men could have marriages and her city could have posterity.
Elissa is Dido, in flight from Tyre. Her motive in seizing the Cypriot girls was
the same as that of Romulus with the Sabines, to provide women for a new
city- The story may well come from Timaeus, just at rhe time when the legend
of the Sabines was probably taking shape.5"
Parallels to the Cypriot custom are reported from Lydia, Syria, Armenia
and Babylonia,’1 but also, more relevantly, at the Creek city of Locri
Epizephyrii in south Italy. Hard pressed in a war with their neighbours of
Rhegium in 477/6 BC, the Locrians vowed to Aphrodite that if she gave them
the victory their daughters would serve the goddess as prostitutes each year on
her festal day; she did, and the vow was carried out. Later it lapsed, during a
difficult war with the Lucanians, but Dionysius II of Sicily demanded that it be
reinstated, this time with the service lasting a month.52

48 Hesiod Theogony 193, Homeric Hymns 6.1-5; cf. Homer Odyssey 8.363. Homeric Hymns 5.58,
Herodotus 1.105.3 (Paphos).
49 w<n!-i■5_4-5, Presumably from Pompeius Trogus in the first century BC; MacLachlan

Tinlaeus FGrH 566 F82, cf. 60. That the Sabines story originated with the incorporation of
", nstonca a ines into the Roman state in 290 (cities sine suflragio) and 266 (cities optima
/>■ *S st4onI’ 7 suggested by the tradition that T. Tatius’ Sabines were given citiitas sine
sitl/ragm (Serous on Aeneid 7.709); see Wiseman 1995.127.
51 earchus fr. 43a Wehrli (Athenaeus 12.516), Herodotus 1.199.2, Strabo 11.14.6, Lucian De
CO yna 6; cf also Valerius Maximus 2.6.15 (past tense) on Sicca in north Africa.
52 Justin 21.3.1-4: uonerant, si uictoresJorent, nt die festo Veneris uirgines suas prostitnerent.

I
 THE KALENDS OF APRIL 153

At Locri and in Cyprus, the duty imposed by the goddess was carried out
by citizen women at specific times, as a kind of public contribution. Else­
where, at wealthy cult centres that could afford to feed and clothe them, the
goddess’s precinct might have a permanent staff of slave girls (hierodottloi). The
best known examples of the latter custom are Eryx, at the western tip of Sicily,
and Corinth?3
When Strabo discusses ‘wealthy Corinth’, and tries to account for that
ancient epithet (no longer valid in his own time, of course), the sequence of his
material is as follows?4 first the city’s geographical situation, enabling it to
exact tolls both from north-south traffic by land and from east-west traffic by
sea; then the Isthmian Games, bringing in spectators from all over Greece; then
the city’s Bacchiad rulers, enriched by exploiting the market but later expelled
in Cypselus’ coup; then Demaratus, the wealthy Bacchiad who took refuge in
Etruria and whose son moved to Rome to become king Tarquinius ‘Priscus’;
and finally the famous cult of Aphrodite, with more than a thousand hierodoitloi
attracting the custom of sea-captains from all over the world. It gave rise to the
proverb ‘Not for every man is the voyage to Corinth’, implying ‘only for the
fortunate’?3
Strabo’s train of thought entitles us, I think, to extend Zevi’s inference
about Demaratus’ son and the Isthmian Games: if he brought one of Corinth’s
profitable culls to his new city, why not the other one as well? If Poseidon
Hippios (Neptunus Equester) could be superimposed on Consus, couldn’t
Aphrodite (Venus) be superimposed on Fortuna?
I suggest, therefore, that at some time in the seventh or early sixth century
BC, two Corinthian cults, each designed to bring in visitors from far afield,
were established in the valley between the Palatine and Aventine hills. In
August there were horse-races in honour of Poseidon/Consus; and all the year
round there were /litwdoii/oi serving the needs of men who could afford an
offering to Aphrodite/Fortuna. The latter, at least, could only be maintained
by the munificence of a wealthy ruler; but wealth, as Zevi rightly insists,36 was
the recurring theme of the Tarquins’ story from first to last.
What would happen to such a “tyrannical’ cult when the tyrant was
expelled and the funding cut off? The goddess must still be honoured. I suggest

53 Strabo 6.2.6, Diodorus Siculus 4.83 (Eryx); Pindar fr. 122 Snell, Strabo 8.6.20, 12.3.36
(Corinth); MacLachlan 1992.157-60.
54 Strabo 8.6.20; for Corinth after the destruction of 146 BC, see Cicero Adfaniiliares 4.5.4 (Ser.
Sulpicius on oppida qnodam tempore Jlorentissinia ... nunc prostrata et dimta). By Strabo s time
there was a Roman colony.
55 oil naVTOg avdpdg Eg KopivOov EO0’ 0 irkoOg, Latinised by Horace (Epistles 1.17.36) as
non minis homini contingit adire Corint/mm.
56 Zevi 1995.294-8.
 154 UNWRI r f l N KOMI

that the community adopted the Cypriot system, as the Locrians did a genera­
tion later: citizen women would serve, but only on the goddess’s festal day. Just
as August was the driest month of the year at Koine, and therefore the most
suitable for horse-races in the valley, so in April the river was at its height,
swollen with melted Apennine snow, and the valley was flooded like the sea.’7
The connection with Liber, attested on the Karlsruhe cista and by Servius’
definition of Murcia,58 may well belong to this stage. Certainly the proximity
of the two festivals in the calendar—the Liberalia on 17 March and Fortuna
Virilis on 1 April—is consistent with a conceptual link, while Liber's role as
the guarantor of public freedom makes his cult necessarily post reges exactos,
whether or not the date of 493 BC for his temple is accurate.’9
The next stage, I suggest, was a long-drawn out ideological dispute, prob­
ably in the fourth and third centuries BC, between orgiastic traditionalists and
puritanical modemisers.60 The cult of Venus Obsequens, introduced in 295,
may have been an attempt to control the goddess (‘Venus who complies); if
so, die effect was surely cancelled out by the introduction ot Venus Erycina in
215.61 The cult of Venus Verticordia, whenever it began, was evidently
designed to shift the emphasis of Fortuna Virilis. It celebrated chastity and
marital concord,62 and offered the ladies ot Koine a respectable reason (in
imitation of the goddess’) to do what they traditionally had to do on that day
of the year.
All of that, I think, is necessary to make sense ot the Ovid passage in Fasti 4.
To return to the very beginning, I take it as confirmation ot the hypothesis
offered here that on the Kalends of April a man was expected to give a woman
a present. It was the last vestige of the long-forgotten origin of the ritual in
sacred prostitution.

57 For the modem statistics, see the table in Le Gall 1953.12.

58 See above, at pp. 145, 148.
59 Liber and iibcrtas: Servius on Aeneid 3.20, 4.57 and 4.638. See pp. 84 . hove.
60 Such a conflict can be inferred about the Lupercalia, roughly betu een ■ • - 6 He: pp.
70-8 above.
61 For Erycina and prostitutes, see Ovid Fasti 4.865—76. I cannot follow Torelli on
Venus Obsequens: aliquot matroiias ad popidum stupri dautuatas petunia intiltauit (Lit t i'
can hardly mean that the adulterous women were ‘obliged to prostitute themselves near
new sanctuary’.
62 Valerius Maximus 8.15.12, Pliny Nat. Hist. 7.120, Lydus De inensibus 4.65

J
 CHAPTER SEVEN

Summoning Jupiter:
Magic in the Roman Republic

I
There is a long passage in Varro’s De lingua Latina where the great antiquarian
seeks to explain the archaic word inliciunt. In the course of his argument he
cites early attestations not only of the noun itself but also of the verb inlicere (‘to
entice, bring in’), and its cognates pellicere and elicere.' All but one of the cita­
tions are from early texts; the exception is his example of elicere (‘to elicit, bring
out’), which is not a text but the name of a Roman cult site:2

sir Elicit louis ara in Auentino, ab eliciendo.

Thus the altar ofJupiter ’Elicius’ on the Aventine, from ‘eliciting’.

The name of the altar was explained by a famous story of Numa ‘eliciting’
Jupiter from heaven, bringing him down to the Aventine to bargain about
what sacrifice he would accept to expiate thunderbolts.3
Livy has a much less sensational explanation of the name, in keeping with
his policy of distancing himself from miracle stories and divine epiphanies.4 In
his account of Numa’s institution of the pontifices, he reports that their duties
included teaching the citizens what portents sent by lightning or other visible
means should be recognised and properly expiated:’

ad ea elicienda ex nientibus diuinis lotti Elicio arain in Auentino dicauit detiiuqite

consiiluit augttriis quae suscipienda essent.

1 De lingua Latina 6.86—95, citing censoriae tabulae (86), coiiuitcutarii consulates (87), the coiniiien-
tarius of a quaestor’s indictment (90), Chorus Proserpinae (a play?) and Pacuvius’ Henniona
(94), and the Conunentarii of Marcus lunius (95); lunius is probably the early jurist M. lunius
Brutus, praetor c. 140 BC (Cicero Brutus 175, Pomponius in Digest 1.2.2.39).
2 Varro De lingua Latina 6.94 (R..G. Kent’s Loeb text; the MSS read iobis uisa uni).
3 Valerius Antias fr. 6P (Amobius Aduersus nationes 5.1): .. .quibus ad terras Hindis luppiter posset
et sacriftdis elid. Ovid Fasti 3.327-8: eliciiutt caelo te, luppiter, unde minores | nunc quoquc te cele­
brant Elidmnqne uocant. Cf. Plutarch Numa 15.3-6, who naturally does not have the Latin
etymology.
4 Levene 1993.16—30 on Livy’s attitude to the supernatural; see also pp. 245-6 below.
5 Livy 1.20.7.
I
156 UNWRITTEN ROME

To elicit that information from the divine mind, Numa consecrated an altar to
Jupiter Elicius on the Aventine, and enquired of the god by auguries which
phenomena should be fonnally recognised.

Since augury was a traditionally acceptable means of ascertaining the will of the
gods, this version preserved the dignity of sober history; Livy will not tell the
story of Numa bargaining with Jupiter in person, though we can be sure he
knew it.
There is a sequel to this episode, reported by Livy a few paragraphs later in
his account ofNuma’s warlike successor Tullus Hostilius. Rome was struck by
a plague, and the citizens were anxious to regain the goodwill of the gods, as in
Numa’s time:6

ipsum regent tradunt uoluentem commentarios Numae, ann ibi quaedam occulta
sollemnia sacrijicia loui Elicio facta inuenisset, operation his sacris sc abdidisse; sed non
rite iuitum out ciiratum id sacrum esse, nec solum nullatn ei oblatam caelestiuin spcciem
sed ira louis sollicitati praua rcligiouefidmine ictum cum domo conflagrasse.
They say that the king himself was turning over the commentaries of Numa,
and when he found certain secret and solemn sacrifices that had been made to
Jupiter Elicius, he hid himself away in order to carry out these rites. But the
ritual was not properly entered on or carried out. Not only was no divine
appearance manifested to him, but Jupiter was angry at being solicited by
perverted religion. Struck by a thunderbolt, the king and his palace were
destroyed by fire.

If Tullus was expecting a ‘divine appearance’ (caelestiuin species), what he found

in Numa’s commentaries must have been instructions for more than just
augury.
We happen to know what it was from one of Livy’s sources, whose account
he chose not to reproduce:7

L. Piso primo atmalium auctor est Tnllnni Hostilium regent ex Numae libris rodent quo
ilium sacrificio loiiein carlo dcuocare conation.. .
According to Lucius Piso in the first book of his Annals, king Tullus Hostilius
used the same sacrificial ritual as Numa, from Numa’s books, to try to call
Jupiter down from heaven.

But even what Livy does tell us about the death of Tullus is enough to disprove
the sanitised version of Numa and Jupiter Elicius that he has just offered; it

6 Livy 1.31.8. Cf. Dionysius of Halicarnassus 3.35.1-2 (with a rationalised version at 35.3—1),
Plutarch Numa 22.7.
7 Pliny Nat. Hist. 28.14 (Piso fr. 17 Forsythe); cf. Nat. Hist. 2.140 (fr. 20 Forsythe). Piso too
had the ritual performed panini rite.
 SUMMONING JUPITER: MAGIC IN THE ROMAN REPUBLIC 157

presupposes the story which at that point he declined to tell, of a face-to-face

encounter between the king and the god.
Why did Tullus fail where Numa had evidently succeeded? ‘Incorrect
ritual’ is an odd explanation: what could count as the proper ritual for such an
act? It must have been self-evident to Livy’s readers that someone who tried to
compel a god to appear before him was asking for a thunderbolt. However,
they did know that there were people who claimed to be able to do that, and
elicere was the verb used for what they did; it was the imposition of a binding
spell by what we would call magic, and what Livy here calls ‘perverted reli­
gion’ {prana religio).*
In 56 BC, for instance, Cicero was attacking the credibility of a hostile
witness in a court case:9
quae te tanta prauitas mentis tenuerit, qui tantus furor, ut cum maudita ac nefaria
sacra susccperis, cum inferorum animas elicere, cum puerontm extis deos manes mactare
soleas, . . .
What great perversion of mind, what great madness possessed you, that
although you have undertaken unheard-of and impious rites, although it is your
habit to bring out the spirits of the underworld and sacrifice the entrails of chil­
dren to the gods of the dead, [yet you ignore the auspices of traditional
religion).
The witness, Publius Vatinius, was a Pythagorean, and that philosophical sect
had always been associated with magical practices.’" Of course we need not
believe forensic invective, but the allegation must have been at least an intelli­
gible one to the senators and equires on the jury. In the early years of Tiberius,
a young aristocrat with an interest in ‘the rites of magicians’ who could ‘bring
out the shades of the underworld by incantations’ ended up facing a charge of
treason, and took his own life.”
These examples from the world of public life provide a context for the
more lurid descriptions of magic in the poets—Horace on the witches Canidia
and Sagana in the graveyard,12 Tibullus on the witch who will help adulterous

8 Attested already in Plato: Republic 364c (fiyvprai 6t Kat |tavtEl5 ... Eltaytiiyaii; TlOtv
Kai KatabEopotc rov; Otou;, a)? tpaoiv, rtEiOovrEg atpioi witjpeteIv); Laws 909b.
9 Cicero In latiiiiiim 14; Ogden 2001.149-50.
10 Sources in Guthrie 1962.253—1; cf. also MacMullen 1966.95-127, Ogden 2001.116-23.
Vatinius was an associate of the learned P. Nigidius Figulus (Scholia Bobiensia 146St), who
was described as Pythagoricus et magus (Jerome Chron. on 45-44 BC); for Pythagoreanism in
late-republican Koine see Griffin 1994.707-10.
11 Tacitus Annals 2.27.2 (magonun sacra), 2.28.2 (temptatus ut infernos umbras canniuibus eliceret).
12 Horace Satires 1.8.23—50, esp. 28—9: cruor in fossam confusus, ut inde | oiaiits elicerent, animas
responsa daturas. In Epode 5, Canidia is about to sacrifice a child, as Cicero claimed Vatinius
did. See Ogden 2001.199-200.
 158 UNWRITTEN ROME

lovers,13 and above all Lucan on the terrifying Erictho, who can summon not
only spirits of the dead but the Furies themselves.14 Lucan states explicitly that
Thessalian witches like Erictho had power to compel even the gods by their
incantations.15 What he describes as foul and barbaric sorcery seems to be
exactly what Tullus Hostilius was punished for trying to do; yet Tullus was
following the instructions of good king Numa, and Numa himself—as Livy
was reluctant to reveal—had evidently been able to compel Jupiter, and call
him down from heaven.
We do not know when the story about Numa ‘eliciting’ Jupiter was first
created; but it is very clear that by the time Livy was writing attitudes to such
activities had shifted fundamentally. That is not surprising. ‘In archaic times
magic was “embedded”, that is, an integrated part of everyday practice.
Nobody thought about it as “magic”.’16 Circe’s powers are taken for granted in
the Odyssey, including her instructions for calling up the shades of the dead.17
But the disciplines of the city-state defined more rigorously what was accept­
able in the community’s relationship with the gods, and what was not. The
creation of magic as a category came about when certain practices were iden­
tified as both objectionable and alien—the malign inventions of the Persian
magoi.18 When did that happen in Rome?
The tradition of Numa as a pupil of Pythagoras probably goes back at least
to the fourth century BC.19 At that time it was accepted doctrine that
Pythagoras had travelled to Persia to learn the wisdom of ‘Zaratas the Chal-
daean’—i.e. Zoroaster.21’ Three centuries later both Cicero and Livy rejected
the whole idea of Numa’s connection with Pythagoras, partly on chronolog­
ical grounds but also because Rome’s wise king could surely not have derived
his wisdom from foreign sources.21 Their attitude was the result of a major shift

13 Tibullus 1.2.43-54, esp. 47-8: haec cantujinditqne solum manesqne sepnlcris | elicit.
14 Lucan 6.507-830, esp. 732—3: ‘iam nos [die Furies] ego nomine uero | eliciam. . . ’ CL Pliny
Mi/. Hist. 30.14 for Nero’s wish to use magic ‘to command the gods’ (imperare dis eoncupinit),
Ogden 2001.152-3.
15 Lucan 6.440-51, 492-9, 527-8 (esp. 443-4, 497-8, 528 for the farm ina); cf. Statius Thebaid
3.144—5 (plurima ... imperet ad superos), Apuleius Metamorphoses 1.8.4 (deos inJimare), 3.15.7
(coguntur minima), 3.18.3 (minimum coactonim uiolentia).
16 Ankarloo and Clark 1999.xv. Cf. Ogden 1999.85: ‘there is no easy way to separate binding
spells from what can uncontroversially be termed “religion” in an ancient context.’
17 As noted by Pliny Nat. Hist. 30.6; Homer Odyssey 10.504—40, carried out at 11.23—50.
18 See Gordon 1999, esp. 162-5, 229-31, 244-52.
19 Numa and Pythagoras: Diodorus Siculus 8.14, Cicero De republiea 2.28 (falsiim), Livy 1.18.2
(falso), Dionysius of Halicarnassus 2.59.1, Ovid Metamorphoses 15.1—484, Plutarch Numa 1.2—
4, 8.4—10, 22.4; Storchi Marino 1999, esp. 82—3 on the date of the tradition.
20 Aristoxenus fr. 13 Wehrli.
21 Cicero De republiea 2.28-9 (non esse nos transmarinis nee inportatis artibiis emditos, sed genuinis
domesticisque uirtutibus), Livy 1.18.2-4 (non tarn peregrinis artibiis quam disciplina tetrica ac tristi
ueteruni Sabinorum).
 SUMMONING JUPITER: MAGIC IN THE ROMAN REPUBLIC 159

in Roman attitudes in the late third and second centuries BC, marked by the
Senate’s successive purges of ‘alien’ religious experts—‘prophets and sacrifices’
in 212, Bacchanals in 186, Chaldaean astrologers in 139.”
It was in 181 BC that the tomb of Numa was ‘discovered’ on the Janiculutn,
containing no body but several volumes of pontifical law and Pythagorean
philosophy. When the content of the books became known, the urban praetor
demanded to read them, declared that they were dangerous to religion, and
with the backing of the Senate had them publicly burned in the coiiiitiiim.23 We
may, I think, interpret this mysterious episode as someone’s attempt to validate
the traditional belief in a Pythagorean Numa by providing ‘authentic’
evidence. If so, the authorities defeated it; but the book-burning did not stamp
out the belief itself, which was still being defended by Plutarch three centuries
later.24
It is against that background that we can now consider the story that Livy
wouldn’t tell, but which Varro alluded to in the passage with which we began.
We have three authorities for it: the historian Valerius Antias, writing probably
in the mid-first century Al),2’ the poet Ovid, writing at the very beginning of
the first century AD,26 and the biographer and philosopher Plutarch, writing
early in the second century AD.27

II
Numa was worried about thunderbolts. They were portents, signs of the gods’
anger; but what was the proper sacrifice of expiation?28 He consulted his divine
consort Egeria for advice.29
Egeria was the nymph of a spring near the ancient grove of Diana at

22 213 Be: Livy 25.1 8-12 (alieno errore . .. extemo rim). 186 BC: Livy 39.8-19, esp. 15.2-3 for
the consul's speech on prauae et extemae reliqiones. 139 BC: Valerius Maximus 1.3.3, Livy
Oxyrliynclius epitome 54.
23 Livy 40.29 (mm animum odiiertisset pleraqipie dissoluendanmi religionnm esse, 29.11), Valerius
Maximus 1.1.12, Pliny Nat. Hist. 13.84-7, Plutarch Numa 22.2-5; detailed analysis of the
discrepant sources m Forsythe 1994.207-15. Pythagorean content: Cassius Hemina fr. 37P
(Pliny Nat. Hist. 13.86), Piso fr. 19 Forsythe (Pliny Nat. Hist. 13.87), Valerius Antias fr. 9P
(Livy 40.29.8).
24 See Gruen 1990.158-70, esp. 167-8.
25 Valerius Antias fr. 6P (Amobius Aduersus nationes 5.1). For the date of Antias, see Wiseman
1979.113-35 and 1998.75-89, Forsythe 2002.99-103.
26 Ovid Fasti 3.259-392.
27 Plutarch Numa 15.
28 Valerius Antias fr. 6P (cum procuraiidi Jiilminis sdentiam non haberet essetque HU cupido noscendi);
Ovid Fasti 3.285—8.
29 Valerius Antias fr. 6P (Egeriae mom'tii), Ovid Fasti 3.289—90, Plutarch Numa 15.5 (cf. 13.1).
Both Ovid (Fasti 3.261—76) and Plutarch (Numa 15.2) introduce the story with a preface
on Egeria.
 160 UNWRITTEN ROME

Aricia.311 But she was also a goddess at Rome, one of the Ccinieiiae, whose
sacred grove and spring were just outside the Porta Capena.31 She was Numa’s
wife (or lover), and his adviser on relations with the gods.32 (Sceptical histo­
rians preserved the story by reporting it as what Numa himself told the Roman
People, an invention to confirm his authority.33) She was thought of as having
magical powers, for instance in miraculously furnishing Numa’s frugal table
with costly food and drink—a feat otherwise attributed in ancient authors to
sorcerers and witches.34
Thanks to St Augustine’s interest in the subject, we know that Varro dealt
at length with Numa in his dialogue De ailtn cieoriini.35 The long peace of
Numa’s reign was due to the benevolence of the gods, which was achieved,
according to Varro, by Numa’s practice of hydromancy.36 This technique,
which originated in Persia and was later employed also by Pythagoras (thus
Varro avoided the chronological impossibility of Numa as a Pythagorean
disciple), involved seeing the gods’ reflections in water; Numa was thus able to
consult them on what form of ritual they required him to use.37 Because he
‘brought out’ (egerere) water for this purpose (no doubt from the spring of the
Camenae), the story arose that Numa had a divine counsellor called ’Egeria’.
By this means Varro was able to avoid attributing to the pious king a sexual
relationship which would have been unworthy of the dignity of the gods.38
Varro did, however, attribute to Numa the desire to keep secret his method
of communicating with the immortals. He had the books that explained the

30 Virgil Aeneid 7.762—4; Ovid Fasti 3.263—75, Metamorphoses 15.487—92; Statius Siluae
5.3.290-1, Martial 6.47.3.
31 Camenae: Livy 1.21.3; Dionysius of Halicarnassus 2.60.5 (‘Muses’); Ovid Fasti 3.275, Meta­
morphoses 15.482; Martial 6.47.3-4, Juvenal 3.16; Plutarch Numa 8.6, 13.2 ( Muses). Porta
Capena: Juvenal 3.10-20, cf. Martial 2.6.15-16.
32 Cicero De legibus 1.4 (fabula), Dionysius of Halicarnassus 2.60.4-5 (pi>0oX.oyoOatv); Ovid
Fasti 3.262, 276, Metamorphoses 15.482-4; Martial 10.35.13-14; Plutarch Numa 4.1-2
(Xdyoc), Moralia 321c (puO(i)dEOTepov).
33 Livy 1.19.5; Plutarch Numa 4.8, 8.6, 15.1.
34 Dionysius of Halicarnassus 2.60.5-7, Plutarch Numa 15.2; cf. Origen Contra Celsum 1.68 (TCt
epya Tiiiv yoijrwv), Philostratus Fita Apollonii3.27 (Indian sage), 4.25 (empousa at Corinth).
35 Augustine City of God 3.9, 7.34-5 (Varro Logistorici frr. 42-4 Chappuis).
36 Varro Logistorici fr. 42 Chappuis: quid ille [Numa] molitus sit et quibus artibus deos tales sibi uel
illi duitati consoriare potuerit, Farro prodit. For compulsion of gods by hydromancy (also called
lecanomancy), see Papyri magici Graeci IV.222-60 (Ogden 2002.205-6), with discussion in
Ogden 2001.192-3.
37 Varro Logistorici fr. 44 Chappuis: lit in aqua irideret imagines deorum [lie/ potius hidificatioiies
daemonum, comments Augustine], a quibus audiret quid in sacris constituere atque obsemare debet.
Cf. also Strabo 16.2.39 (C762) on Persian ubpopavTElg.
38 Augustine City of God 7.35: quod eigo aquam egesserit... idea nympham Egeriam coniugem didtur
habuisse, qiiemadmodum in supra scripto libro Farronis exponitur. Cf. Augustine City of God
18.10: Marcus Farro non miltfabulosis aduersus deosfdem adhibere figmentis, ne de maiestatis conun
dignitate indignmu aliquid seutiat.
 SUMMONING JUPITER: MAGIC IN THE ROMAN REPUBLIC 161

technique buried with him, and when in 181 bc they accidentally came to
light, ‘the senators agreed with the dead king’ and told the praetor to bum
them.39 Varro had no problem with the use of hydromancy as such—he had
witnessed hydromantic divination at Tralles during the Mithridatic war4"—but
he understood why Numa and the senators of 181 bc did not want the tech­
nique divulged. He took the view that the understanding of the wise was not
something to be shared with the population at large.41
His interpretation of Numa’s piety precisely exemplifies the tripartite
scheme he set out at the beginning of the Dit’ine Antiquities:'12 ‘mythical
theology’, with its tales unworthy of the divine majesty, offers the story of
Numa and Egeria as lovers; ‘physical [i.e. natural] theology’, which philoso­
phers understand, provides what Varro regards as the true explanation, that
Numa used hydromancy to consult the gods; ‘civic theology’ is represented by
the cult ofjupiter Elicius, and the very particular offerings made at the altar on
the Aventine.
How do these Varronian categories relate to each other? Varro himself took
it for granted that ‘mythical theology’ was necessarily secondary' to ‘natural
theology’’; the truth was turned into myth by the poets.43 Augustine accepted
that, though naturally without endorsing the truth of the philosophers’
theology. Modern scholars are more likely to take the opposite view, assuming
a primeval myth subsequently rationalised by etymology. But not all myths are
‘primeval’ (whatever that unhelpful term signifies), and since the poets Varro
had in mind were clearly inventing their stories for the stage,44 the origin of
‘mythical theology’ is bound up with the question of the origin of the Roman
Indi scaenici.
Where Augustine disagreed with Varro was about the separation of the first
and third categories Since the theatre games were an integral part of pagan
religion, ‘civic theology’ too depended on myths unworthy of the gods.4’
Over a century earlier, Arnobius had argued that this very narrative of Numa
and Jupiter was both grossly offensive to divine dignity and at the same time

39 Augustus City Co*/ 7.34, citing Varro’s own words (Logistorici fr. 43 Chappuis): ubi cum
primores quasdain causas legissent, cur quidqiie in sacris fieri! iiistinuuni, Ntiniae mortuo senatns
adsensus est, eosque libros tainquani religiosi patres conscripti praetor ut combureret censuerunt.
40 Apuleius Apologia 42.
41 Varro Antiquirates diuinae fr. 8 Cardauns (Augustine City of God 6.5): sic alia, quaefacilius intra
parietes in schola quant extra inforo Jerre possum aures.
42 Varro Antiquitates diuinae trr. 6—10 Cardauns (Augustine City of God 6.5, 6.12).
43 Augustine City of God 7.35: ita eiriin solent res gestae aspersione mendacionuii injabulas uerti. See
pp. 246—9 below for the same assumption in Varro’s Degente populi Romani (Augustine City
of God 18.10-13).
44 Augustine City of God 6.6 (di poetici theatrici ludicri scaenici), 6.7 (theologia Jabulosa theatrica
scaenica); Wiseman 1998.17—24.
45 Augustine City of God 6.6-7 passim.
 162 UNWRITTEN ROME

essential to the Romans’ traditional civic cult.46 On that point, at least, the
critics had a good case. The whole comic tale of Egeria’s advice was indeed
embedded in the cult practice of the Roman Republic.

III
Our sources call Egeria a nymph or a goddess interchangeably. But I think it is
clear that when her story was first created she was only a nymph—a semidea,
one of the humblest members of the immortal world.47 She did not know the
answer to Numa’s problem, so she advised him to consult powers one step up
in the divine hierarchy, the deities of the woods and hills of Rome, Picus and
Faunus.48 Ovid describes them as horned and hoofed; elsewhere, he identifies
Faunus with Pan. According to Plutarch, ‘one might liken them to satyrs and
Pans’—but with an added characteristic to which we shall come in a
moment.49
Below the Aventine was a holm-oak grove, with a cave and a spring where
Picus and Faunus used to come to drink.56 Numa left bowls of wine there, and
hid in the cave with a picked group of twelve young men. Picus and Faunus
came, drank deeply, and fell asleep; the twelve leapt out and bound them fast,
only releasing them when they promised to tell Numa what he needed to
know?' The obvious analogue for this story is Midas’ capture ot Silenus, to
learn from him the workings of the universe;52 but there are also echoes of the
consultation of Proteus, since Plutarch refers to Picus and Faunus shape­
shifting in their attempt to avoid capture.53
Picus and Faunus are known from other sources as oracular deities, and in
the Euhemerised version of their myth—where they feature as kings of ancient

46 Amobius Aduersus nationcs 5.1—4 (exploiting Valerius Antias fr. 61’).

47 Seinideae: e.g. Ovid Heroides 4.49, Metamorphoses 1.192.
48 Ovid Fasti 3.292 (Romani numen utnmique soli), 3.303 (silucstria mmiina), 3.315—16 ('di SIllllUS
agrestes et qui dominenmr in altis | montibns’).
49 Ovid Fasti 3.312, cf. 2.361, 4.663, 5.99-101 (horns and hoofs); 2.423-4, 3.84, 4.650-3
(Faunus as Pan); Plutarch Numa 15.3.
50 Ovid Fasti 3.295-9, cf 302 and 305 for the antrum; Calpurnius Siculus 1.8—12 describes a
generic grove-cave-spring habitat for Faunus.
51 Only Antias (fr. 6P) mentions the twelve casti iuuenes, in what was probably an aetiology of
the leaping Salii (Wiseman 1998.21-2).
52 Herodotus 8.138.2—3, Xenophon Anabasis 1.2.13, Theopompus FGrH 115 F75; Gantz
1993.138 for the monographic evidence. Virgil (Eclogues 6.13-30) has Silenus captured by
two boys and a nymph.
53 Plutarch Numa 15.4; cf. Homer Odyssey 4.454—61 (Proteus and Menelaus), Virgil Georgia
4.437-47 (Proteus and Aristaeus). Proteus could be thought of as a magician: Petronius
Satyricon 134.12, Pliny Nat. Hist. 30.6.
 SUMMONING JUPITER: MAGIC IN THE ROMAN REPUBLIC 163

Latium—they are described as skilled prophets or augurs.’4 But Plutarch knew

another and more sensational interpretation of their powers:55

duvapst 6e cpappciKtov Kai bEtvoTpit xfig nspl ta 0Eia yoi]TEiag kEyoviai

tafiia Tolg oep’ 'Ekktjvcov 7tpooayopEV0Eioiv 'Idaioig AaKTvkotg
oocpi^opsvoi jTEptiEvat if|v ’Irakiav.
It is said that they go around Italy practising the same craft, by the power of
their potions and the skill of their sorcery regarding divine matters, as those
called by the Greeks the Idaean Daktyloi.

According to Ephorus the Idaean Daktyloi were wizards, who moved from their
native Phrygia to the island of Samothrace and astonished the people there by
their skill in ‘incantations, initiations and mystery rituals’.51’ The Samothracian
mystery cult was well known to the Romans; Varro claimed that Aeneas had
brought the Penates from there, that Tarquinius Priscus had been an initiate,
and that some Roman religious terminology was derived from Samothrace.57
The main purpose of the Numa story was to explain why the offerings
made at the altar ofJupiter Elicius consisted of fish, hair and onions: the wise
king had discovered that the god wanted live human heads, but would be satis­
fied with live fish, human hair and onion-heads.Our sources report three
different versions of how Numa discovered this.
The first is the simplest: Picus and Faunus told him.59 But how did they
know? As Faunus points out in Ovid’s version, rustic divinities are hardly
experts on thunderbolts:60

‘magna petis ne< quae inonitn tibi discere nostro

Jas sit: Itabent jinis nmnina nostra snos.
di snintis agrestes ct qui dotnineninr in altis
inontibns: atl’itrinin est in sna tela loin'.’

54 Oracles: Plutarch Moralia 268f (Picus); Virgil Aeneid 7.81-103, Ovid Fasti 4.649-68
(Faunus). Prophet-kings: Augustine City oj'God 18.15 (Picus); Plutarch Moralia 268d, Origo
gentis Romanae 4.4 (Faunus). See Ogden 2001.91-2 on Virgil’s story of Latinus consulting
the oracle ot his father Faunus, ‘an amalgamation of a tiekuomaiiteioit and a hero-oracle’;
Ovid, however, thinks ot the oracular Faunus as a hoofed divinity (n. 49 above).
55 Plutarch Numa 15.3.
56 Ephorus FGrH 70 F104 (Diodorus 5.64.4): vitap^avtag de ydrjTag dmTt)beuoai Tag TE
Eitwdag Kai TekEtag Kai pvoTijpia.
57 Varro Amiquitares humanae fr. 2.8 Mirsch (Aeneas), Amiquitares diuinae fr. 205 Cardauns
(Tarquinius); De lingua Latina 5.58 (a phrase in rhe augunuu libri), 7.34 (camillus); tor Varro on
Samothrace, see also Anriquitates diuinae fr. 206 Cardauns (Augustine City oj God 7.28),
Logistorid fr. 40 Chappuis.
58 Amobius Aduersus nationes 5.1 (Antias fr. 6P), Ovid Fasti 3.337—44, Plutarch Numa 15.5.
59 Plutarch Numa 15.4: TOVTOVg epaoi . . . TOV ETti TOi: KEpaVVOig EKbldClgai KaOappdv,
6; itotEtiai HE'/.pi vvv did Kpopjwaiv Kai rpr/wv Kai paividaiv.
60 Ovid Fasti 3.313—16.
 164 UNWRITTEN ROME

‘You’re asking a lot, and for something you are forbidden to know by our
instruction. Our powers have their limits. We are gods of the countryside,
masters of the high hills; Jupiter’s weapons are under his control.’
Somehow, Numa had to talk to Jupiter himself. Hence the second and third
versions of the story, which differ slightly but significantly.
According to Valerius Antias, Picus and Faunus told Numa ‘by what means
and by what sacrifices’Jupiter could be brought out of heaven.61 That cautious
formulation allows the assumption that the pious king did nothing unlawful,
and that Jupiter came of his own free will. ‘It is rational enough,’ observed
Plutarch,62 ‘to suppose that the deity would not place his affection upon horses
or birds, but rather upon human beings eminently distinguished by virtue, and
that he neither dislikes nor disdains to hold conversation with a man of
wisdom and piety.’
The third version, however, stated without concealment that Picus and
Faunus brought Jupiter out by the arts of magic, and that Jupiter was angry.63
Ovid is explicit that he was compelled to come:64
‘deme tamen nobis uincula, ’ Picus ait:
‘luppiter hue neniet, ualida perductus ab arte.
itubila promissi Styx mihi testis erit. ’
emissi laqueis quid agant, quae carmina dicant,
quaque traliaut superis sedibus arte louem,
scire ttefas homiiti: nobis coitcessa caneiitiir
quaeque pio dici uatis ab ore licet.
‘Just take the shackles off us,’ said Picus, ‘and Jupiter will come here, drawn by
powerful art. Cloudy Styx will be the witness of my promise.’ What they did
when they were set free, what incantations they said, by what art they dragged
Jupiter from his high seat, it is forbidden for man to know. I shall sing what I ni
allowed, what may be spoken from the mouth of a pious bard.
Although the story has a happy ending—-Jupiter is amusi;ed by Numa’s inge-
nuity and goes back to heaven in a good mood6'1—the fact remains that he has
been forced to obey a magic spell, just as if wicked Erictho had been practising
her polluta ars.,'b

61 Amobius Aduersus nationes 5.1 (Antias fr. 6P): illos statim perdocuisse regent quibus ad terras modis
luppiter posset et saenjidis elid.
62 Plutarch Numa 4.3, trans. John and William Langhorne (1770).
63 Plutarch Numa 15.5: EVtot di ou xoug haipovaq tpaotv vjto0EO0at tov KaOappov,
dXX’ EKtivoug piv KaTayayEtv rdv Aia payEUoavrag, tov de Oeov opyi^dpEvov ...
64 Ovid Fasti 3.320-6.
65 Plutarch Numa 15.6 (idv pfv 0e6v <5oteX.0eiv iXeoi yEVopEVOV); Ovid Fasti 3.343 (risit).
66 Lucan 6.509; see above, p. 158.
 SUMMONING JUPITER: MAGIC IN THE ROMAN REPUBLIC 165

I think it is inconceivable that this story is a late invention. Why should

anyone want to impute the use of forbidden magic to the pious king, if the
tradition did not exist already? Besides, the problematic element in the story
seems to be inherent in the cult itself—not only the god’s title (from elicere) but
also the nature of the offerings, since hair and fish were much used in magical
rituals.67 Some scholars resort to the concept of‘folktale’,68 which does at least
imply antiquity, but is otherwise an unhelpful description of the aetiological
explanation of a public cult.
Certainly the story as we have it is a comedy, perhaps first created for the
stage.69 I think it is likely that ‘tales unworthy of the gods’ were characteristic
of the plays put on at the Roman dramatic festivals, and that such plays were
intended to be educational as well as entertaining.711 At any rate, what we seem
to have here is an old story designed to encourage the Roman People to
believe that their wise king could call on magical powers—and that the even­
tual outcome was the shield that fell from heaven, a divine talisman to
guarantee Rome’s prosperity.71

IV
Modern work on magic in the ancient world tends to move straight from
Hellenistic Greece to late-Republican Rome.72 The Roman Republic’s
knowledge of magic is assumed to be pan of its assimilation of Greek culture,
supposedly in the earls second century BC.73 The tacit premise, that Rome was
free ot outside cultural influences until ’after the Punic Wars’ (as Horace put
it),'4 is demonstrably talse: the Greek language was known in Latium by about
800 BC, and Greek mythology is attested at Rome as soon as recognisable
iconography becomes available in the sixth century.75 Given the evident early
interest in Pythagoras, and the prohibition of casting spells in the Twelve

67 Hair: e.g. Euripides Hippolytns 513, Apuleius Metamorphoses 3.16; Ogden 1999.14-15. Fish:
e g. Ovid Fasti 2.577-8, Apuleius Apologia 27.
68 Rose 1928.318; Grant 1971.145-6.
69 Valerius Antias fr. 6P (Arobius Adttersus nationes 5.1, excogitata et eoinparata derisid), Plutarch
A'uih.i 15.6 (JlvOdidi) octi yeXoict); see above, n. 44.
70 Augustine City of Cod 4.26-7; Varro De lingua Latina 6.18 (a play that docuit popidiini). For a
possible example, see Wiseman 2004a. 116-17 (on Ovid Fasti 2.583-616).
71 Ovid Fasti 3.259—62, 349—92; implicit also at Plutarch Numa 13.1. See Wiseman 1998.21—2
for the suggestion that the ’twelve chaste youths’ of Antias fr. 6P were an aetiology for the
Salii.
72 E.g. Luck 1999.120: Theocritus followed by Nigidius Figulus.
73 E.g. Dickie 2001.124—8, esp. 126 on ’new ways of thinking and new forms of behaviour’ in
the time of the elder Cato.
74 Horace Epistles 2.1.156-63, a disastrously influential passage.
75 See briefly Wiseman 1994.6-8, 26—8; on the Gabii graffito, see Ridgway 1996, Peruzzi
1998.19-22, Wiseman 2004a.l3-16.
 166 UNWRITTEN ROME

Tables,76 we can be confident that the Romans of the fifth and fourth centuries
BC were well aware of the magic arts.
They evidently disapproved of the malicious use of such arts against indi­
viduals, but we have no reason to suppose that the arts themselves were
regarded as un-Roman. On the contrary, the implication of the Numa and
Jupiter story is that in the archaic period (whenever it was that the Elicius cult
was set up) they were taken for granted as an acceptable way for a wise and
pious king to make contact with the gods. The trouble came in the second
century BC, when the Senate arrogated to itself the duty of deciding which
traditional practices were acceptable, and which not. The burning of the
Pythagorean ‘books of Numa’ in 181 BC was an early manifestation of Roman
authoritarianism, later so familiar under the emperors.77 But it seems that the
story of Numa and Jupiter was firmly fixed in the popular mind, no doubt
through the medium of stage performances, and so it survived to be preserved
in literature.
As a poet, Ovid could tell the story straight, so long as he virtuously disso­
ciated himself from any knowledge of magic; so too could a frivolous historian
like Valerius Antias, much to the indignation of Arnobius.78 But a philosoph­
ical antiquarian could allow only a rationalised reference to hydromancy, while
serious historians, protecting Numa’s reputation, referred to magic and the com­
pulsion of deities only as sacrilegious rites, as in the exemplars' case of Tullus
Hostilius, blasted by the angry god.79 Plutarch, the Greek philosopher who
wanted to insist on Numa’s Pythagoreanism despite the chronological impos­
sibility,80 narrated the story only as a mythos—not what really happened, but a
Platonic ‘noble lie’ told to the unruly Romans by Numa the philosopher-
king.81
For our purpose, the value of the story is as an example of how something
that belonged exclusively to the archaic world ofpre-literary Rome can never­
theless become visible to us, even through the medium of literature itself.

76 Tabulae VIII.1 and 4: texts and discussion at Crawford 1996.677-9, 682—I.

77 See Dio 52.36.1-3 (Maecenas’ advice to Augustus), and Paulus Sententiae 5.23.17-19
(Ogden 2002.279) on the punishment of magicians, including the public burning ot their
books.
78 Arnobius dduersus uatioiies 5.1: quid? ilia quae historiae continent graues seriae airiosae ... poet­
arum sunt excogitata lasciuia?
T? See above, pp. 160-1 (Varro), 155-7 (Livy and Piso).
80 Plutarch Numa 1.2-3, 8.3-10, 14.2-3, 22.3-4.
81 Plutarch Numa 15.1, cf. 4.8 (‘not ignoble’ story about Egeria), 8.1—3 (Plato).
 CHAPTER EIGHT

Origines ludorum

si Latinis ciuitatem dederitis, credo, existimatis nos ita nt constitistis in condone habituros
locum aut ludis etJestis interf'utiiros. nouue illos omnia occupaturos pntatis?
‘I suppose you imagine that if you give the Latins citizenship, you’ll have space
at public meetings and take part in games and festivals in just the same way as
now. Don’t you think they’ll take over everything?’
C. Fannins (cos. 122 BC)1

The consul’s argument tells us what mattered most about being a Roman
citizen: participation in politics, and in the games and festivals of the gods. The
audience at the games was the Roman People, and their reactions could be
taken as those of the citizen body as a whole.2 When Polybius comments on
the importance of religious belief in holding the Roman state together, he
explains it by the use of ’tragedy’ to influence the People;3 presumably that
refers to plays performed at the ludi saienici, which often featured the Furies
punishing evildoers and the gods announcing to the audience the benefits they
had conferred on diem. *
So the origins and development of the 'public games’, as the Romans called
them, are a significant part of the political history of the Republic, and the
appearance of a learned and detailed monograph on the subject is a reason for
gratitude and congratulation? Deservedly, Frank Bernstein’s book will be for
the foreseeable future the standard work on the history of the ludi, and for that

i Oratio de sociis et nomine Latino contra Gracchum fr. 3 Malcovati (Julius Victor 6.4 = Rhetores
Latini 402 Halm).
2 Cicero Pro Sestio 106, 116—18, De hanispicum response 22-5, In Pisonem 65, Ad Atticuni
2.19.3, 14.3.2, Philippics 1.36.
3 Polybius 6.56.8 and 11: Eiti roooOtov yap eKTeTpaytobnioL koi jrapaafjictai tovto to
(ttpoc [sc. deiotdaipoviaj nap’ aurot? tic te tou; icax’ idiav pioug teal xa Kotva
rf]C jioXeu)? uiote pf| KUTakiTtEiv vrtEppo/.rjv. [...] kEijiEiat toic abqXoic; tpdfloig
Kai Ttj TOtautr] Tpayipdia Ta n/.ljOq Ouve/Eiv. See Mazzarino 1966.61-2, Zorzetti
1980.64-5.
4 Cicero Pro Rosdo Amerino 67, In Pisonem 46, De hanispicum response 39, De legibus 1.40, Acad-
einica 2.89 (Furies); Plautus Amphitmo 41-5 (gods).
5 Bernstein 1998.
 168 UNWRITTEN ROME

very reason it is important to draw attention to any arguments in it which may

seem to be less than compelling.

II
A convenient summary of the dates of origin of the different Indi is offered in
Bernstein’s Appendix 2:6
Indi maximi/Romani about 509 or 507
Indi plebeii probably 220
Indi Ceriales probably 220 or 219
Indi Apollinares 208
Indi Megalenses 191
Indi Florales 173
Indi Victoriae (Sultanae) 81
Ln di Victoriae Caesaris 45
In fact, only the last two of these dates are secure. Cicero and Livy are explicit
that the Indi Romani were instituted by Tarquinius Priscus;7 Livy gives 212 for
the origin of the Indi Apollinares,8 and 204 and 194 (as well as 191) for that of
the Megalenses;1 the Indi Florales dated from 241 or 238, according to Velleius
Paterculus and Pliny respectively.111
In the case of the Apollinares, which were the responsibility of the urban
praetor, we know that at first they were vowed by the praetor each year and
held on a day of his choosing, and then in 208 they were made a nnual with a
fixed date.11 The Florales similarly were not at first a regular part of the
calendar, and we do not know how often (or even whether) the plebeian
aediles of each year held them before the consuls of 173, acting on a senatns

6 Bernstein 1998.358.
7 Cicero De republica 2.36 (eimdem priminn ludos maximos, qui Romani dicti stmt, Jecisse accepitims),
Livy 1.35.9 (sollenmes deinde ammi mansere Indi, Romani magnique ttarie appellati); Eutropius
1.6.1 and De ttiris ilhistribus 6.8 are probably dependent on Livy. Cf. also ps.Asconius 217
Stangl: Romani Indi sub regibus itistituti sunt magnique appellati, quod magnis impensis dati.
8 Livy 25.12.1-15 (concluding hacc est origo ludonnn Apollinarium), 27.23.5.
9 Livy 29.10.4-11.8 and 14.5-14 (concluding Indi fuere, Megalesia appellati, 204); 34.54.3
(Megalesia ludos scaem'cos A. Atilius Scrramts L. Scribonius aediles entitles primi fecenmt, 194);
36.36.4 (ludique ob dedicationem eius facti, qttos primos scaenicos fuisse Antias Valerius est auctor,
Megalesia appellatos, 191).
10 Velleius Paterculus 1.14.8 on colonial foundations (.. .proximoqtte anno Torquato Senipro-
nioque consulibits Bnindisium et post triennimn Spoletium, quo anno Floralitim liidonun factum est
initium); Pliny Nat. Hist. 18.236 (itaqne iidem Floralia IV kal. easdem instituenmt ttrbis anno
DXVJ ex oraculis Sibyllae, ut omnia bene dcjlorescerent). Cf. Ovid Fasti 5.277—94 on the Publicii,
plebeian aediles no doubt in one or other of those years; also the Flora coins of C. Clodius
Vestalis in 41 BC (Crawford 1974.521, no. 512), which may have been for the two-
hundredth anniversary.
11 Livy 27.23.5-7, by a law of the urban praetor P. Licinius Varus.
 ORIGINES LUDORUM 169

consultmii, made them annual.12 But it is worth noticing that Livy doesn’t
mention the reform of 173;13 evidently it was a less important stage than the
founding of the games 65 or 68 years earlier.
Bernstein argues that the Megalenses only became annual with the dedica­
tion of the temple of Magna Mater in 191, and applies the same logic to the
Indi Romani: in honour ofJupiter Optimus Maximus, they can only have been
regular after the Capitoline temple was dedicated.14 That is of course a possible
solution (in both cases), but it is certainly not a necessary one. No doubt Rome
already had a cult of Jupiter before the Capitoline temple was built, and the
famously wealthy Tarquins could certainly afford to put on games every year.15
There is a particular problem with the Indi Romani, arising from what ought
to be the best of evidence, a statement by an aedile-elect concerning the games
for which he was going to be responsible:16

habeo rationem quid a populo Romano accepenm; mihi ludos sanctissimos maxima cum
cura et caerimoma Cereri Libero Liberaeque faciundos, mihi Florent matrem populo
plebique Romanae ludorum celebritate placandam, mihi ludos antiquissimos, qui primi
Romani appellati sunt, cum digmtate maxima et religione loui lunoni Mineruaeque esse
faciundos.
‘1 think carefully about the task the Roman People has laid upon me. I shall
have to put on, with ali possible care and ceremony, the most sacred games of
Ceres, Liber and Libera' 1 shall have to secure the goodwill of Mother Flora for
the people and plebs of Rome by the popularity of her games; and for Jupiter,
Juno and Minerva, widi the utmost dignity and religious piety, 1 shall have to
put on the most ancient .mes, the first that were called "Roman".’

This puzzling passage wa. brilliantly explained by Lily Ross Taylor long ago.17
The ludi Ceriales and Fiorales were certainly the plebeian aediles’ responsibility,
as the Romani and Megalenses were that of the curule aediles;1" it is therefore
inevitable that Cicero was a plebeian aedile, and that the third games he refers
to were the ludi plebeii.
According to one strand of the historiographical tradition, the Plebeian
Gaines went right back to the beginning of the Republic.19 What Cicero

12 Ovid Fasti 5.295—330; rhe famine that caused the Senate to act may have been the result of
the plague of 175—I (Obsequens 10, Livy 41.21.5—8, 42.2.7).
13 There was an opportunity at 42.2.3—7, on prodigia and the pax deorum; cf. also 42.10.7—8 on
the locust cloud in Apulia.
14 Bernstein 1998.193—5 on Magna Mater, 50—1 on luppiter Optimus Maximus.
15 For the historicity of the tradition about the Tarquins’ wealth, see Zevi 1995.
16 Cicero 1'errines 5.36; cf. Pro Murena 40 (ego qui trinos ludos aedilis feceram).
17 Taylor 1939, esp. 194—7 on the games.
18 Dio Cassius 37.8.1 (curule), 47.40.6 and Ovid Fasti 5.287—92 (plebeian).
19 Valerius Maximus 1.7.4 (491 BC); ps.Asconius 217 Stangl (plebeii ludi, quos exactis regibus pro
libertate plebis fecenmt, an pro recondliatione plebis post secessionem in Auentinuni).
 170 UNWRITTEN ROME

implies is that they were originally the only ‘Roman Games’, and that only
after 367, when the curule aediles took over, did the Indi plebeii and Romani
become two separate events.20 I suspect that this narrative was the creation of
Licinius Macer, himself an ex-tribune of the plebs, whose history was probably
coming out at just the time Cicero was speaking.21

III
Bernstein, however, rejects Taylor’s argument outright, and thinks the Cicero
passage must be emended. That is a desperate expedient, and not a convincing
one—for what would be the point of calling them ‘the ancient games which
were first called Plebeian’? It is clear that Bernstein is driven to it by the ill-
founded belief that the ludi plebeii were a creation of the late third century, and
that the Circus Flaminius was created in 220 BC to provide a site for them.22 That
idea comes with the authority of Mommsen,23 but even so it must be wrong.
In both title and structure, the Indi plebeii are analogous to the Romani.
Neither games are named after a god; both are centred on Jupiter’s day,24 in the
only two months of the Roman calendar which feature no ‘large-letter’ festi­
vals except the Kalends, Nones and Ides. That in itself suggests that the Indi
plebeii and Romani coexisted at the time the calendar was put together (no later
than the fourth century BC), and in any case the very existence of a specifically
plebeian festival suggests a fifth- or fourth-century origin.
By contrast, the Indi that were introduced in the third century all have very
specific foundation stories arising out of the particular issues of the day. The
Florales resulted from the plebeian aediles fining landowners who encroached
on the public land, the Apollinares from the praetor’s alleged discovery of the
prophecies of On. Marcius after a crack-down on ‘seers and sacrificers , the
Megalenses from the Sibylline Books’ advice on how to drive the foreign
enemy out of Italy.2’ The only foundation story that survives for the Indi plebeii.

20 Livy 6.42.12—14 (reaisantibus id iiiuniis aedilibus plcbis, 367). Cf. Dionysius ot Halicarnassus
5.57.5 for ‘the games named after the city’ in 500 BC (probably from Aelius 1 ubero:
Wiseman 1994.54—5); it is not clear who was thought to be organising them.
21 See Wiseman 1995.107—8 and 135—6. Macer was tribune in 73; he died in 66, when his
history had reached at least as far as 299 BC (Livy 10.9.10, fr. 19P).
22 Bernstein 1998.79—80 and 158—63: ‘problematisch ist aber, dass die Angabe des Relativsatz
iiber die Roinaiii als Indi antiquissiini zu plebeii emendiert werden muss' (79).
23 Mommsen 1887.519—20: ‘Wann diese Spiele eingerichtet worden sind, ist zweifelhaft,
wahrscheinlich erst im J. d. St. 534.’ The doubt has disappeared in Mommsen 1893.335:
‘Ludi plebeii post Romanos magnos antiquissiini et honoratissimi instituti sunt a.u.c 534.’
24 Ides sacred to Jupiter: Ovid Fasti 1.56, Macrobius Saturnalia 1.15.14-15, Lydus De inensibus
3.10. The idea that the calendar must pre-date the dedication of the Capitoline temple,
because ‘of the feast-days which it marks with large letters none is connected with that cult’
(Rose and Price 1996.274), is simply invalid.
25 See notes 8-10 above.
 ORIGINES LUDORUM 171

that they celebrated the liberty of the plebs after the expulsion of the kings or
the secession to the Aventine, is in a late Ciceronian scholiast. Too late to be
reliable, according to Bernstein; but if there had been a genuine historical origin
for the games in 220 bc, one might have expected a scholiast to know it.26
Finally, the building of the Circus Flaminius is certainly irrelevant to the
question. It is true that Valerius Maximus anachronistically puts the htdi plebcii
in the Circus Flaminius in 491 BC;27 but there is no other evidence that C.
Flaminius’ creation was ever used for major htdi, and ample evidence that it
was a piazza surrounded by very substantial public buildings, quite unsuitable
for chariot racing.28 It is also true that theatrical perfonnances were held in the
piazza at the htdi Apollinares;2'* but that was only because the temple of Apollo
was there, and no connection with the htdi plebeii in honour of Jupiter.
Mommsen’s theory has nothing to recommend it.

IV
In Bernstein’s list, the htdi plebcii and Ceriales belong together. Since the first
reliable evidence for them comes in Livy', respectively in 216 and 202 BC, he
assumes that they were introduced soon before that time."1 The reason for the
assumption is his belief in the ’hellenizing’ of the Roman religious experience
in the third century' BC.J1 But it is an illusion to suppose that Greek ideas influ-

26 See n. 19 above; Bernstein 1998.79-80.

27 Valerius Maximus 1 7.4; he may have been imagining the open space of the prata Jlaminia,
the site enclosed by C. Flaminius in 220 (Livy 3.54.15). For the gloss in circo [„ui.vi»i<> et
Flantinio] spectaculi Ax o at Valerius Maximus 4.4.8, see Wiseman 1976.45.
28 See Humphrey 1986.540-5, Viscogliosi 1993. The only Indi Vano knew about in the
Circus Flaminius were the horse-races of the obscure Indi Tattrei (De lingua Lutina 5.154).
29 Plutarch Cicero 13 (the temples ot Bellona and Apollo were adjacent); cf. Livy 3.63.7 (in prata
Jlaminia, ttbi nunc aedes Apollims est), 40.51.3 (censor of 179 lets contracts for theatnmt et
proscaenium ad Apollinis).
30 Bernstein 1998.157—8; Livy 23.30.17 (plebei Indi aedilium Al. Aurelii Coltae et M. Clandii
Marcelli ter instaurati), 30.39.8 (Cerialia ludos dictator et magister equitmn e.v senalus consulto
fecenuit). [So too Csapo and Slater 1994.208: ludi plebeii ‘instituted in 220 Be’, Ceriales ‘cele­
brated from 202 BC’ (no evidence or argument offered).|
31 Bernstein 1998.227: ‘Veranderungen und Neuerungen wurden in der Regel aufein gewan-
deltes Religionsverst’andnis zuriickgefiihrt. Neue, vor allem griechisch-hellenistische
Gottesvorstellungen ftihrten zur Erweiterung des Ritus und veranlassten die Einfiihrung
neuer ludi publici. Uiese Ausweitung des ofFentlichen Spielbetriebs in Rom vollzog sich
allerdings auf mehreren Ebenen. Nur vereinzelt konnte aufgrund der Uberlieferungssitua-
tion beobachtet werden, wann es zu einer Vermehrung der Spieltage kam. Gleichwol diirfte
von einer sukzessiven Zunahme ausgehen sein. Andererseits wurden in einem ziemlich
kurzen Zeitraum funf weitere statarische ludi publici, die ludi plebeii, Ceriales, Apollinares,
Megalenses und Florales, eingerichtet. Diese Tatsache ist um so bemerkenswerter, als es von
spatarchaischer Zeit bis in das spate 3. Jahrhundert nur die ludi maximi bzw. Romani fur
luppiter Optimus Maximus gegeben hat.’ I think there is a circular argument hiding in that
final sentence.
 172 UNWRITTEN ROME

enced the Romans only then; the iconographic evidence illustrates the phenom­
enon at least as far back as the sixth century.32 So there is no a priori reason to
attribute the origin of the undatable games to the latest period possible.
In fact, each of the Livy passages offers no more than a terminus ante quern.
Since they refer to exceptional circumstances which happened to occur in
those years, the games concerned may have been regularly held for many
generations before then, and in the case of the Indi plebeii we have seen that
they probably were. I think the same applies to the Ceriales.
If the games of Ceres had really been introduced in 220 or 219 BC, one
would expect some account of their introduction to have survived, as it has for
the games of Flora, Apollo and the Great Mother. Certainly one would not
expect a foundation story attributing it to the battle of Lake Regillus in the
early fifth century BC; but that is what we have, not in a late scholiast but in the
narrative of Dionysius of Halicarnassus, very probably taken from a second-
century BC source.33 We needn’t suppose that the account is necessarily
historical. The point is rather that if the Romans of the second and first
centuries BC believed that Ceres’ games were introduced in the early
Republic, the existence of that tradition puts the onus of proof on those who
claim, like Bernstein, that they were really introduced nearly three centuries
later, in circumstances that happen not to have been recorded.
The moneyer C. Memmius claimed on his coins in 56 BC that one of his
ancestors was the first to hold the Indi Ceriales: MEMMIVS AED. CER1ALIA
PREIMVS FECIT.34 At first glance, that seems a good argument for a late
foundation date; the first securely dated Memmius was praetor in 172 BC, so an
aedile a generation or two earlier would bring us to the period suggested by
Bernstein. Indeed, all the standard works confidently place the moneyer s
ancestor as ‘aedile ?211’ and father of the praetor.35 But despite the question
mark, the date 211 is not uncertain; it is worthless. It should be only ’before
210’, when Livy started recording the names of plebeian aediles.36 Here too,
we have no more than a terminus ante quern.
It is important to remember that the late-republican Memmii claimed to be
a very ancient family indeed, descended from Assaracus and the kings of Troy.37

32 For a brief statement, see North 2000a. 15-16.

33 Dionysius of Halicarnassus 6.10.1, 6.17.2-4, 6.94.3. Probably from the history of A. Postu-
mius Albinus, consul in 151 and descendant of the victor of Lake Regillus: see Wiseman
1998.35-6 and 86-7.
34 Crawford 1974. 451-2, no. 427.
35 Miinzer 1932; Broughton 1951.276; Sumner 1973.87.
36 See Miinzer 1932.303, Broughton 1951.277 n. 4.
37 Virgil Aeneid 12.127 (genus Assarad Mnestheus'), 5.117 (inox Italus Mnestheus, genus a quo
nomine Mennni); the name is a caique on gt|ivi]o0ai = nieniinissc. Mnestheus was evidently
named on the late-republican monument of C. Memmius C.f. at Ephesus: see Torelli
1997.152-74, esp. 164-7.
 OR/G/NES LUDORUM 173

What were they supposed to have been doing between the arrival of Aeneas in
Italy and the praetorship of 172 bc? Compare the Sicinii, who claimed descent
from the aristocracy of Alba Longa; they too are first securely attested in the
early second century BC.3S But someone in the historiographical tradition gave
the Sicinii a heroic role in the early history of the plebs, from 493 to 387 BC; it
was almost certainly Licinius Macer, honouring a fellow-popuhtris of the post-
Sullan years. ’9 The Memmii too had popularis credentials (the radical tribune of
111 BC was the moneyer’s grandfather),4" and since the cult of Ceres, Liber and
Libera was the centre of the early plebeian movement, it is quite possible that
Macer—and the Memmii themselves—attributed the origin of Ceres’ games to
a plebeian aedile of the early fifth century BC.
The following fifth-century plebeian aediles are named in the historical
tradition: L. lunius Brutus and L. Sicinius, 492; T. lunius Brutus and C. Visel-
lius(?) Ruga, 491; L. Alienus, 454.41 The lunii Bruti clearly owe their existence
to the story’ of the Liberator, though in a version which evidently denied the
execution of his sons.42 The Sicinii we have considered already; Visellii and
Al(l)ieni are otherwise known only as late-republican tribunes of the plebs.43
Memmius the aedile would fit as easily into that pseudo-historical context as
into the real history of the late third century.
So the coin legend does not disprove the presumption that the ludi Ceriales,
like the plebeii, are more probably an early fifth-century than a late third-
century' phenomenon. As noted above, that doesn’t mean that the account of
their origin in Dionysius, as the result of a vow for victory at the battle of Lake
Regillus, is necessarily historical. On the contrary’, one can see that it incorpo­
rates an allusion to a quite different version, that the games were vowed at a
time of famine.44 As with the plebeii,45 different accounts of their origin were

38 Dionysius ot Halicarnassus 3.13.4 (Alban); Cn. Sicinius, praetor in 183 and 172 Be. One of
the supposed consuls ot 487 BC is variously named as ‘T. Sicinius’ (Livy 2.40.14, Festus
180L), ‘T. Siccius’ (Dionysius of Halicarnassus 8.64.1), and ‘Sabinus’ (Fasti Hydatiaui,
Chronicle ot 354, C/irunnwi Pasdi.i/e),
39 Sallust Histories 3.48.8 (speech of Macer) on L. Sicinius tr. pl. 76. For Macer and the alleged
early Sicinii, see Ogilvie 1965.337, 382 (on Livy 2.40.14 and 58.2), and Briscoe 1971.10;
Ogilvie’s briet comments are justified in detail by Hodgkinson 1997.
40 Sallust Juqurtha 27.2, 30.3—34.1; for the stemnia, see Sumner 1973.85-90.
41 Dionysius ot Halicarnassus 7.14.2, cf. 6.89.1 for the full names (492 bc); 7.26.3, cf. 6.89.1
for ‘Visellius’, name corrupted in the MSS (491); Livy 3.11.5, cf. Dionysius of Halicarnassus
10.48.3 (454). There is also M". Marcius (440 bc), named by Pliny Nut. Hist. 18.15.
42 For the dispute, see Plutarch Brutus 1.4-5 (Posidonius FGrH 87 F 40).
43 CIL I2 744 (lex l/isellia), cf. Cicero Brutus 264 on C. Visellius Varro; Campbell 2000.216,
321-2 (lex .. . Allieiui); cf. Crawford 1974.471, no. 457, on A. Allienus.
44 Dionysius of Halicarnassus 6.17.3—1; ps.Cyprian De spectaculis 4.1 (cum urbem fames oceupasset,
ad auocationem populi adquisiti sunt ludi scaeuici et Cereri et Libero dieati postmodum). For the ludi
of Liber, see Festus (Paulus) 103L, Ovid Fasti 3.784-6.
45 Ps.Asconius 217 Stangl (n. 19 above).
 174 UNWRITTEN ROME

evidently circulating as the historiographical tradition took shape. Which, if

any, was the right one is now beyond our knowledge.

V
The main importance of the Memmius coin-legend is that it reveals how
significant the games were in late-republican society. It mattered to be able to
claim credit for having introduced them. A moneyer of the previous year, C.
Servilius C.f, had made what may be a similar boast about the Florales
(FLORAL. PRIMVS),46 and traces of similar ‘primus’ claims are easily
detectable in the historical sources on the origins of the various Indi.*7
There was a very lively literature on the subject; not only Varro’s De
scaenicis originibns, De qnaestionibns scaenicis, and the ninth and tenth books of his
Antiqmtates dininae (on Indi circenses and scaenici respectively), but also Sinnius
Capito’s Spectacnla, and whatever other late-republican sources Suetonius had
for his two books of Historia lndicra.*s Like the De fantiliis Troianis of Varro and
Hyginus, these works no doubt incorporated the competitive historical claims
of ambitious Romans engaged in the contentio nobililatis of late-republican poli­
tics.49
As it turns out, Bernstein’s innocent list of the origins of the Roman games
conceals a fennent of aetiology, pseudo-history and self-promoting publicity.
And that is just what we should expect. How the games beg in could be a
subject for plays put on at the games themselves,50 and for most people what
they saw on the stage was a large part of what they knew about the past, an
uncheckable amalgam of fact and legend.51 For us, what matters most is what
the consul pointed out to the populace in 122 bc: the games were part of the
definition of Roman citizenship.

46 Crawford 1974.447, no. 423, interpreting the legend as [/lumen] Floral[ts] pnnms; the tradi-
tional interpretation F/ora/[M| primus [/cn7| is defended by Badian 1984.56 .
47 E.g. Livy 27.23.5 and 7, 34.54.3, 36.36.3 (Valerius Antias fr. 40P); Censonnus De <
17.10 (probably also from Antias). ...
48 Varro: frr. 70-7 and 307-18 Funaioli; Augustine City of God 6.3 on the Antiqmtates.
Capito: Lactantius Diuinae institutiones 6.20.35; cf. Festus 186L, 438L, .<>< • ut °
Aulus Gellius 9.7.3, Suda s.v. ’Trankullos’; cf. Tertullian Dr spectaadis 5 (he found Ins ma -
rial apud Suetonium Traiiquilluni uel a quibus Tranquilhis accepit).
49 Lucretius 2.11 (eertorc ingenio, contendere nobilitate), Horace Odes 3.1.10—13 (hie gi nerostor .
contendof). .
50 Megalcnsia was the title of comedies (togatae) by Afranius and Atta; Ovid Fasti 4.2 -6 or a p ay
about the arrival of the Great Mother. Afranius and the mime-writer Laberius also
plays called Conipitalia (for the ludi conipitalicii see Cicero In Pisonent 8 and Ascomus -)•
some idea of the content may be inferred from Pliny Nat. Hist. 36.204 and Macrobius Satur­
nalia 1.7.34-5.
51 CCicero De..................
legibus 1.47- (scaena as source of opiniones), Pausanias 1.3.3 (learning from choruses
and tragedies’); cf. nn. 3-4 above, and Wiseman 1994.16-20.
 CHAPTER NINE

The Games of Flora

i
Spectacle can take many forms, but in the Roman Republic it came to have a
particular significance, in the festivals or ‘games’ (Indi) that were put on by the
aediles at public expense, at regular dates throughout the year, in honour of the
gods of Rome. Each of these festivals consisted of two different types of enter­
tainment: ‘stage games’ (Indi scaenici), dramatic performances in temporary
theatres erected for the occasion, and ‘circus games’ (Indi circenses), chariot races
and wild-beast hunts in the Circus Maximus.
These were the occasions when the Roman community met en masse to
honour its gods and celebrate its identity. They were a powerful force for
social cohesion, and the right to ‘view’ them (speetaadnm means ‘a viewing’)
was one of the chief privileges of citizenship. By the mid-first century BC,
which is the period much of our evidence comes from, there were seven such
festivals: the Indi iVkyalenses, for the Great Mother of the Gods (4-9 April); the
Indi Ceriales, for Ceres (12—18 April); the Indi Florales, for Flora (28 April—2
May); the Indi Apollinares. for Apollo (6-12 July); the Indi Romani, for Jupiter
(4-12 September); the Indi I ictoriae, for Victory (26—31 October); and the Indi
plebeii, for Jupiter again (4-12 November). The dates refer to the ‘stage games’
only; the ‘circus games’ came afterwards in each case.
That makes forty-nine days of public stage performance every year. What
went on? We have texts only tor one type of comedy (Plautus and Terence),
played by an all-male cast in masks. It is obvious that this cannot have been the
only dramatic genre in republican Rome. Entertaining the Roman People and
its gods for five, six, seven or nine consecutive days necessarily required plenty
of variety, but we are very ill informed about the different types of perform­
ance that must have been offered. What follows is an investigation of one
particular dramatic festival, the Indi Florales in April and May.

II
Flora was a goddess of flowers and blossoms, and a deity of some significance
in Rome. Fabius Pictor gave her name (in Greek) to a character in the foun­
dation story; according to Varro, her cult was introduced by Titus Tatius and
 176 UNWRITTEN ROME

his Sabines in the reign of Romulus; John Lydus, from what source we do not
know, even reports that ‘Flora’ was the sacred name of Rome itself’
The original site of her cult was just at the point where the ancient Via
Salaria, coming over the high ground from Antemnae, descended into the
floodplain of the Tiber before entering the city at the foot of the Capitol; her
grove and altar, and later her temple, evidently sheltered below the north-west
slope of the Quirinal, facing west across the meadows of the Campus Martius
[fig. 45].2 She was one of the select divinities that had their own designated
priests, the Jlamiiies allegedly created by Numa.3 She appears on coin-types
minted in 57 and 41 BC; the former, by C. Servilius C.f, carries the legend
FLORAL. PRIMVS, interpreted by Michael Crawford as Floralis primus, refer­
ring to an ancestor of the moneyer who was the goddess’ first Jlamen.4 If that is
right, no doubt the ancestor was a legendary one, like Marcius the first
pontifcx.5
The other coin-type, by C. Clodius Vestalis in 41 BC, probably refers to the
moment in the third century BC when Flora received a new temple and the
htdi Florales were introduced.6 In or about the year 240 BC, two plebeian

1 Fabius Pictor FGrH 809 F 4a (Plutarch Romulus 3.3), on Antho the daughter of Amulius,
who saves the life of her cousin, the mother of the twins; Varro De lingua Latina 5.74 (citing
aimales); Lydus De mensibus 4.73: iepailKOV de Okdipa oiovel tivOovUtt (compare 4.75.
'Pwpq <W.«pa).
2 Varro De lingua Latina 5.158, Vitruvius 7.9.4, Martial 5.22.3—I, 6.27.1. Coarelli 1995.254.
‘sulle pendici del colle che scendono verso Via della Panetteria,’ just below the Quirinal
Palace. (The descent of the Via Salaria is represented by Via di Porra l-’inciana and Via
Francesco Crispi.)
3 Vano De lingua Latina 7.45, quoting Ennius Annalcs fr. 116—18 Skutsch; cf. Vanggaard 1988.
4 Crawford 1974.447, no. 423, pl. LI. The traditional interpretation ‘FLORAL|iaj PRIMVS
[fecit]’, referring to an aedile of 173 BC (cf. Ovid Fasti 5.329-30). is restated by Badian
1984.56-8. But Badian’s appeal (1984.56) to the legend MEMMIVS AE1L CERIAL1A
PRE1MVS FECIT on C. Memmius’ denarii (Crawford 1974.451-2, no. 427 [see p. 172
above]) as ‘inspired by that of the “first Floralia” in the previous year’ is unconvincing. As
explicit—perhaps
Badian himself observes (1984.57), Memmius’ legend ‘could not be more explicitperhaps
by contrast with Servilius’ obscure conciseness’; surely the only reason for the explicitness
would be to distinguish it from a different message?
5 Livy 1.20.5; cf. also Plutarch Moralia 264c-d (Juba FGrH 275 F 91) on the pontifex
Cornelius; Plutarch Numa 10.1 on the Vestal Verania. Badian (1984.57-8) argues that since
the Servilii were among the Alban families brought to Rome by Tullus Hostilius, there
could not have been ajlanien Servilius under Numa. But that is to impute too much consis­
tency to ‘accepted belief. Numa was supposed to have created the priesthood of the Vestal
Virgins, but its previous existence in Alba is a necessary premise of the foundation story
(indeed, in one version the Vestal mother of the twins was called Servilia: Anthologia Palatina
3.19.pref). Perhaps the (laminate, mentioned by Livy 1.20.2—3 in just this context, was also
thought of in Servilian tradition as an Alban institution.
6 Crawford 1974.521, no. 512, pl. LX1L See also Wiseman 1979.92-4 and n. 124, arguing
that Vestalis was a descendant of C. Claudius Cento cos. 240 and that he attributed the
Floralia to Cento's consulship. Pliny Nat. Hist. 18.286 dates the games to 238 BC, Velleius
Paterculus 1.14.8 to 241. Note that Cento means ‘patchwork’, and that at Flora’s games it
was the custom to wear multicoloured clothes (Ovid Fasti 5.356, cultu uersicolore).
 THE GAMES OF FLORA 177

rz/'vx -^a
z/r
/// >(
M (L? .//I
r Vi
?r
M
/—

J
Fig. 45 Republican Rome in its geographical context.
 178 UNWRITTEN ROME

aediles with the appropriate name of Publicius punished rich landowners who
were illegally occupying public land; they brought the culprits to trial before
the People, fined them heavily, and with the proceeds built the Clivus Publi­
cius up from the Circus Maximus to the Aventine.7 At the bottom of the
Clivus, next to the ancient temple of Ceres and Liber, they built a new temple
of Flora and instituted her games.8 The Senate, evidently resenting this mani­
festation of popular sovereignty, refused to honour Flora, but she made them
change their minds (as she explained in person to the poet Ovid), and the
games were duly established as regular and annual in 173 BC.9
According to Pliny, Flora’s games were set up at the behest of the Sibylline
books. So too was the cult of Venus of Eryx, with her tradition of ritual
prostitution.10 This same period saw the introduction of the erotically charged
flagellation spectacle at the Lupercalia, and of the Saturnalia with its carnival-
esque role-reversals.11 Members of the senatorial elite may have disapproved,
but licensed popular revelry seems to have been not only permitted but
even prescribed as a means of honouring the gods in the Rome of the third
century BC.
Whatever the reason, Flora’s games were notoriously licentious.12 Her day
was 28 April, and by the Augustan period at least, the Indi scaenici extended till
2 May—five days of theatrical performances, followed by one day of Indi
circenses in the Circus Maximus. In fact, the whole festival probably took place
in the Circus, immediately in front of the new temple, where there was plenty
of room for the erection of ad hoc stages and auditorial3 Tin Circus was
famous for showgirls and prostitutes,14 and it was precisely showgirls and pros-

7 Ovid Fasti 5.279-94, Varro De lingua Latina 5.158. Festus (276L) has a variant attributing the
Clivus to the curule aediles L. and M. Publicii Malleoli.
8 Tacitus Annals 2.49.1; according to Ampelius 9.11, Liber was the son of f lora. Games: Ovid
Fasti 5.277, 291-2, Velleius Paterculus 1.14.8, Pliny Nat. Hist. 18.286.
9 Ovid Fasti 5.295-330. The ideological dispute evidently gave rise to rival myths. Flora tells
Ovid (Fasti 5.195-228) that she was the nymph Chloris—from /Xcopbc. the fresh green of
spring growth—and that Zephyrus abducted and married her, and made her the goddess of
flowers. A much less flattering version, alluded to by Ovid (Fasti 5.191. hoininuni seittentia
fallax), made her out to be a wealthy prostitute who made the Roman People her heir, and
left them a fund to pay for games in her honour (Lactanctius Dininae instttniiones 1.20.5—10,
Minucius Felix Octavius 25.8, Cyprian De nanitatc idoluni 4).
10 Pliny Nat. Hist. 18.286, Livy 22.9.8-10 (215 BC). For the Sibyl’s innovations in the third
century BC, see North 1989.616-18.
11 Lupercalia: Livy fr. 63 Weissenbom; cf. Orosius 4.2.2 (for the date 276 Be), Ovid Fasti
2.425-52 (by order ofJuno); for the context see pp. 75—8 above. Saturnalia: Livy 22.1.19—
20 (217 Be), Accius Annates fr. 4 Courtney (Macrobius Saturnalia 1.7.36), Pliny Letters
2.17.24, Ausonius 7.24.15; Macrobius Saturnalia 1.24.23.
12 Ovid Fasti 4.946, 5.331—56, Martial l.pref., Historia Augusta Elagabalus 6.5, Ausonius
7.24.25, Amobius 3.23, Augustine City of God 2.27.
13 Hanson 1959.16-17; see also ps.-Acro on Horace Satires 2.3.182.
14 Priapea 27 (deliciae populi, magno notissima Circo, | Quinctia, nibratas doeta tnonere nates), Sueto­
nius Nero 27.2, Juvenal 3.65.
 THE GAMES OF FLORA 179

titutes who performed at the ludi Florales.'5 At least by the late Republic, it was
a traditional part of the entertainment that they should undress at the audi­
ence’s demand. On a famous occasion in 55 BC, the presence of Marcus Cato
in the audience inhibited the usual calls for the striptease; when he realized, he
left the theatre and the show went on.16
Nanirally, Christian polemicists seized on this indecency as typical of
paganism.17 Cicero, as plebeian aedile-elect, had included the ludi Florales
among the solemn responsibilities of his office:18
habeo ratioiiem quid a popido Romano acceperiin; ... mihi Floram matron popido
plcbiqite Romanae ludorum celebritate placandain.
I think carefully about the task the Roman People has laid upon me; ... I shall
have to secure the goodwill of Mother Flora for the People and plebs of Rome
by the celebration of games.
St Augustine was appalled: Cicero was a serious man, and claimed to be a
philosopher!19 ‘Secure the goodwill’, indeed! This placatio of the goddess was
petidantissitna, iinpudentissinia, nequissinia, inunundissima. .. However over the
top Augustine’s rhetoric may have been, one can see his point. How could a
striptease show be a traditional part of the public religion of Rome, formally
entrusted to the dignified magistrates of the Republic? Why didn’t Cato stay
where he was, and let rhe force of his moral authority prevent the disgraceful
exhibition?
1 think an answer to the puzzle may be found if we ask another question:
what were the plays in which Flora’s girls performed? They were not, of
course, anything like the formal genres of comedy and tragedy, with their male
actors and masks. These were ‘mimes’, traditionally sexy and farcical,2" but not
restricted to sex and farce. A whole collection of ethically improving one-liners
was put together from the mimes of Publilius Syrus.21 The epitaph of a mime­
actress ot the late Republic emphasizes her skill in dance, and in all the Muses’
arts.” Mimes had plots and dialogue; there was more involved than just ‘bumps
and grinds’. But what sort of plots offered parts to numerous young women
and required them to undress?

15 Lactantius Diuinae institutiones 1.20.10: meretrices, quae turn mimannn fungantur officio. Minute:
Valerius Maximus 2.10.8. Meretrices: Ovid Fasti 5.349, Seneca Epistulae 97.8, Martial 1.35.8-
9, Tertullian De spectaadis 17.3, Amobius 7.33.
16 Valerius Maximus 2.10.8, Seneca Epistulae 97.8, Martial l.pref.
17 Lactantius Diuinae institutiones 1.20.6—10, Tertullian De spectaadis 17.3—I, Amobius 7.33.
18 Cicero Fem’nes 5.36 (70 Be).
19 Augustine City of God 2.27 (uirqrauis et philosophaster Tullius)-, see also n. 58 below.
20 Osad Tristia 2.497—520; see Rawson 1993.
21 Collected in Duff and Duff 1934.14—111; see Giancotti 1967.275-462.
22 ILLRPW3 = CIL F 1214 = ILS 5213, Eucharis Liciniae 1.; see Wiseman 1985.30-5.
 180 UNWRITTEN ROME

I have argued elsewhere that much of what went on at the Roman

dramatic festivals represented the Romans’ own mythology and history—aeti-
ological stories to explain gods and rituals, exemplary stories to celebrate the
deeds of men.23 I believe that there were performances of this sort at Rome
long before Livius Andronicus introduced the formal genres of Greek drama. If
that hypothesis is valid, then we may be able to trace some possible scenarios
for Flora’s games in the mythological and historical narratives offered by our
literary sources.

III
In the second or third year of the Republic,24 Rome was besieged by Lars
Porsena ofClusium, in alliance with the exiled Tarquins. The literary tradition
narrates the siege as a succession of heroic exploits by ‘the three prodigies and
wonders of the Roman name’: Horatius Codes, Gaius Mucius, and Cloelia.2’
All three stories were probably aetiological in origin: to explain a statue,
supposedly of Horatius, at the Volcanal,26 the ‘Mucian Meadows’ across the
Tiber,27 and another statue, of a girl on a horse, at the top of the Sacra Via.28
All three stories were told in various forms,29 and among the variants of one,
the Cloelia narrative, we find our first example.
Cloelia was one of a group of boys and young women sent by the Romans
as hostages to Porsena’s camp across the Tiber. What is common to all the
versions of the story is that she escaped back to Rome, the Romans
honourably gave her up, and Porsena equally honourably freed her in admira-

23 Wiseman 1994.10-19; Wiseman 1995.133-8.

24 Second: Livy 2.8.9-9.1, Plutarch Publicola 16.1-2. Third: Dionysius of 1 lalicarnassus 5.21.1.
25 Floras 1.10.3 (ilia tria Romani nominis prodigia atque iniracida, Horatius, Mucins, Clodia); thus
e.g. Livy 2.10-13, Plutarch Publicola 16-19, De uiris illnstribns 11-13. Georges Dumezil
interpreted Horatius and Mucius as reflections of the Indo-European 'one-eyed god and
‘one-handed god’ (Dutnezil 1988.139-59, Dutnezil 1973.274-6); for the context in
Dumezil’s theory, see Belier 1991.146—56. The concept was extended to < loelia by Luciano
Arcella (Arcella 1985.30-1 on ‘i tre eroi-mostri’). Quite apart from the general problems
raised by Dumezil’s theories, on which see Belier 1991, there are good reasons for consid­
ering at least two of the stories to be comparatively late inventions: Horatius exploit at the
bridge seems to presuppose a wall circuit that included the Janiculum and can therefore
hardly predate the third century BC (Gjerstad 1969, Wiseman 1979.44), while Mucius sacri­
fice of his right hand explained the use of the cognomen Scaevola by the Mucii (Livy 2.13.1,
Plutarch Publicola 17.3), which is first attested in a praetor of215 BC (Livy 25.3.6).
26 Aulus Gellius 4.5.1—4, Plutarch Publicola 16.7, De uiris illnstribns 11.2; cf Livy 2.10.12 (in
comitio), Dionysius of Halicarnassus 5.25.2 (in the Forum). See Coarelli 1983.161-2, 168-9.
27 Festus 131L, Livy 2.13.5, Dionysius of Halicarnassus 5.35.1, De uiris illnstribns 12.6.
28 Livy 2.13.11, Dionysius of Halicarnassus 5.35.2, Seneca Consolatio ad Marciam 16.2, Plutarch
Publicola 19.5, Annius Fetialis in Pliny Nat. Hist. 34.29 [see pp. 289—90 below]; cf. De uiris
illnstribns 13.4 (in jbro).
29 Horatius: in Polybius 6.55.3, unlike the later writers, he sacrifices his life. Mucius: flpflTCll
pfv Bird JtoXXdjv Kai btatpopebg (Plutarch Publicola 17.1).
 THE GAMES OF FLORA 181

tion of her bravery. But how did she cross the river? (The bridge, of course,
had been broken at the time of Horatius’ exploit.) According to one version,
she got hold of a horse and rode it across.30 Since the statue that gave rise to the
story was of a girl on a horse,31 this version is likely to be the original one. But
it did not go unchallenged: some said she swam across, and explained the statue
by having Porsena honour her with the gift of a war-horse, to show that she
had the courage of a man.32 It is a variant of the ‘swimming’ version that
concerns us here.
Dionysius of Halicarnassus relates that the girl hostages asked their guards
for pennission to bathe in the river. Permission was granted. The girls then
asked the men to withdraw a little from the bank until they had bathed and
dressed again, so as not to see them naked. The guards behaved like
gentlemen, and the girls, now unsupervised, followed Cloelia’s suggestion and
swain across en masse. Plutarch has the same version, but he does not just
follow Dionysius; it is clear that the two authors were independently using the
same source.33
Also found only in Dionysius and Plutarch is the episode of an ambush by
the Tarquins as the girls are being returned to Porsena. The heroine of this
story is Valeria, daughter of Publius Valerius Publicola; the author from whom
Dionysius and Plutarch took it was probably Valerius Antias.34
The mass escape offers no starring role for Cloelia. She is no longer the
uniquely courageous heroine, but at the most prima inter pares, with no partic­
ular claim to an honorific statue. The story’s aetiological significance is lost, but
what has been gained—carefully signaled by the request for the guards to with­
draw—is the scene where the girls get undressed. It is, however, a scene which
the non-visual medium of Historiography is ill equipped to convey, and I
imagine it was originally conceived for the stage: the guards withdraw, but we
the audience are privileged to watch.
There is nothing paradoxical in this hypothesis. The historians of Rome

30 Valerius Maximus 3.2.2, Plutarch Publicola 19.4, Fionas 1.10.7, De uiris ilhistribus 13.1.
31 Seen. 28 above. As a curiosity, cf. Arcella 1985.39: ’statua equestre come enfatizzazione e
cristalizzazione del ri fmto di un’unione esogamica.’
32 Swimming: Livy 2.13.6, Virgil Aeueid 8.651, Dionysius of Halicarnassus 5.33.1, Plutarch
Publicola 19.1, Polyaenus 8.31. Gift from Porsena: Dionysius of Halicarnassus 5.34.3,
Plutarch Publicola 19.4, Cassius Dio 4.14. Gift from Romans at Porsena’s suggestion: Servius
on Aeueid 8.646.
33 Dionysius of Halicarnassus 5.33.1, Plutarch Publicola 19.1; also Polyaenus 8.31, who adds the
detail that they wrapped their dresses round their heads as turbans, presumably to have
something to wear on the other side.
34 Dionysius of Halicarnassus 5.33.2-4, Plutarch Publicola 19.2-3. Plutarch Publicola 19.5
reports a variant that attributed the ’girl on a horse’ statue to Valeria. I think it can be shown
(see Wiseman 1998.75—89) that Antias systematically retold the famous stories of the early
Republic with a ’Valerian dimension’ in each case.
 182 UNWRITTEN ROME

were well aware that ‘historical’ episodes were presented on the stage, and that
some of the material in their own sources might be suspected of having origi­
nated there.35 Most historical drama was serious, and literary historians equated
the genre offttbula praetexta with Greek tragedy.36 But the praetexta was not the
only dramatic form that exploited the history of Rome. What sort of play was
it that starred the glamorous Quinta Claudia in the story of the Great Mother’s
arrival in 204 bc?37 The Indi scaenici of the Roman Republic no doubt
presented performances in many more various forms than later antiquarians
were ever able to categorise;38 for some of them, no doubt (to borrow modern
phraseology), the contribution of the producer and director will have been
more important than that of the scriptwriter. The mimes at the Floralia could
be an example.
I suggest, therefore, that the version of the Cloelia story reported by
Dionysius and Plutarch started life as a lightheartedly patriotic showpiece for
the girls at the games of Flora, and that it was then incorporated, probably by
Valerius Antias,39 into a historical narrative from which our two surviving
authors took it in good faith. If we accept that as a working hypothesis (and the
nature of the evidence means that it can hardly be anything more), we may
look for analogous examples in some other unexpected places.

IV
The Indi Flornlcs ran from 28 April to 2 May.4" The first day of May, which fell
in the middle of the games, was the festival of Bona Dea, the womens
goddess, at her shrine below the so-called ‘little Aventine’.41

35 Livy 5.21.9 (fall ofVeii), Dionysius of Halicarnassus 3.18.1 (triplets’ fight in war with Alba),
9.22.1-3 (Fabii at the Cremera), Plutarch Romulus 8.7 (foundation legend).
36 Diomedes in Grammatici Latiiti 1.489K: in quibus imperatonim negotia agebanlur et publica el reges
Romani uel duces inducmitur, personanmi dignitate et [personanini] sublimitate tragoediis similes.
Praetexta fragments in Manuwald 2001.131-248; see Zorzetti 1980, Flower 1993 [and pp.
194—9 below].
37 Ovid Fasti 4.291—348: inira, sed et scacna testificata, loquar (326). Cf. also Varro De lingua
Latina 6.18, on the Nonae Caprotinae.
38 See Zorzetti 1980.83: ‘La pretesta arcaica .. . non e che la punta di un iceberg, la parte cioe
emergente nel campo del teatro letterario di un ben piu nutrito gruppo di manifestazioni
cultural!, in cui il popolo romano era sollecitato ad esprimere, per mantenerla o adattarla, la
coscienza della propria tradizione.’
39 See n. 34 above. Antias was interested in the stage: frr. 18P, 22P, 55P (Censorious 17.8-11)
on the ludi saeailares, fr. 37P (Asconius 69C) on the Indi Romani, fr. 40P (Livy 36.36.4) on
the ludi Megalenses, fr. 46P (Livy 39.22.9) on votive ludi. Two items in his history have been
thought to derive from dramatic sources: see Wiseman 1998.21—3 on fr. 6P, Mazzarino
1966.451 and 542 on fr. 48P.
40 Ovid Fosti 4.943-8, 5.183-6; Degrassi 1963.449-51.
41 Ovid Fasti 5.148-58, cf. Cicero De domo 136 (sub Saxo); Degrassi 1963.453.
 THE GAMES OF FLORA 183

That is the scene of Propertius 4.9, one of the legendary narratives in a

book of Callimachean aetiology. What has to be explained is the Forum
Bovarium, and its cult of Hercules at the Ara Maxima. In twenty lines, Prop­
ertius tells the story made famous in the eighth book of the Aetteid, how
Hercules overcame the monstrous catde-thief Cacus. And then:42

.. . sicco torquet sitis ora palato,

terraque non nilas feta nu'nistrat aquas.
sed procul inclusas audit ridere puellas,
Incus iibi umbroso fecerat orbe nenius,
femineae loca clausa deaefontesque piandos,
impiine et nullis sacra retecta uiris.
His palate is dry, thirst tortures his mouth, and the earth, teeming with water,
offers him none. But far off he hears the laughter of girls, enclosed where the
wood had made a grove with a circle of shade—the secret place of the women’s
goddess, and her sacred waters, and rituals never with impunity revealed to
men.

The parched hero hurries to the precinct and begs for admission: ‘O you who
sport [or dance] in the grotto of the grove. . ,’43 But the priestess refuses him
(‘May the gods give you other springs; this water flows for girls’),44 at which
Hercules forces his way in and drinks the stream dry.4’
What girls, what sport, what dance? The grotto (antriun) mentioned by
Propertius features also in Juvenal’s wonderfully overheated description of the
rites of Bona Dea, where the women of Rome, in a drunken frenzy, challenge
the brothel-keepers’ slav e girls to a contest of erotic dancing.46 It is far from
clearwhetherjuven.il was thinking about the Kalends of May festival or about
the Bona Dea ritual in December at the house of a senior magistrate (the occa­
sion notoriously desecrated by Publius Clodius in 62 BC).47 However, since the
May festival took place at the time of the htdi Florales, when we know the
mimes were performed by mere trices, Juvenal’s reference to the girls from the
brothels may allude to them.
Propertius’ poem has been much studied of late,48 but none of its recent

42 Propertius 4.9.21-6 (the translation of line 22 is adapted from G.P. Goold’s Loeb Classical
Library version).
43 4.9.33: 'o luci quae luditis antro'. For ludere as ‘dance’ (Thesaunis linguae Latinae7.1772.45-74,
1789.48-63), see Cairns 1992.74.
44 4.9.59-60: 'di tibi dent alios fontes; haec lynipha puellis ... Jluit'.
45 4.9.65, exhausto iamjlumine.
46 Juvenal 6.314—34 (328 for the antrum); cf. Wiseman 1974.130-4.
47 Cicero Ad Atticuni 1.12.3, 1.13.3, Plutarch Cicero 28, Caesar 9-10, Scholia Bobiensia 85-91
Stangl; Juvenal refers to the Clodius scandal at 6.335—15.
48 Cairns 1992, Anderson 1992, Fox 1996.169—75. See also Hardie 1996.222-3. 225: the girls
‘are celebrating the festival of the Bona Dea with their dance’ (223).
 184 UNWRITTEN ROME

interpreters has picked up J.C. McKeown’s suggestion that in this poem, as in

others, Propertius may have been ‘mythologising mime’.49 Although it is not
quite what McKeown meant (he sees Hercules as a mimic komast), I take this
poem as another scenario for Flora’s girls, but conceptually reversed. Proper­
tius focuses on Hercules; for the stage, the scene would be inside the precinct,
with the audience again in the role of voyeurs.
Bona Dea shared the first day of May with the guardian gods of Rome, the
Lares Praestites. They were the twin sons of Mercury by a nymph called Lara,
whose story is told by Ovid as an aetiology of‘the silent goddess’.5(1 Jupiter is
pursuing the nymph Juturna, but she persistently eludes him. Frustrated
beyond endurance, he summons all the nymphs of Latium and issues a stem
warning: next time Juturna runs away, stop her before she gets in the water!
The nymphs agree, but Lara can’t hold her tongue. Not only does she warn
Juturna, she even goes to Juno and tells her what Jupiter is up to, sympathizing
with married women.51 Furious, Jupiter tears out her tongue and banishes her
to the underworld, giving Mercury, as psychoponipos, the job of escorting her
there. On the way, in a grove (no doubt the grove of his mother Maia, whom
some identified as Bona Dea),52 Mercury takes advantage of her condition and
fathers on her the twin Lares.
Since the action is almost wholly articulated in speeches—Jupiter to the
nymphs, Lara to Juturna, Lara to Juno, Jupiter to Mercury—1 infer that this too
is probably the elegiac reworking of a dramatic original. If so, the assembled
nymphs, like the girls in Propertius’ precinct,53 may suggest what sort of
performance was involved. In the spring, as Horace tells us, the nymphs dance
naked in the warm west wind; the bride of the west wind, a nymph who
became a goddess, is Flora herself.54
As we have noted already, these two scenarios in Propertius and Ovid both
involve divinities whose festival day fell in the middle of Flora’s games. But
they have something else in common, which is particularly relevant to our
enquiry. Here is Hercules, dying for a drink, pleading at the gate in ‘words
unworthy of a god fnerba minora deo]’; here is Jupiter, dying for something else,
his treatment by the elusive Juturna ‘unworthy of so great a god [tanto non
patienda deo]’. Both phrases echo, no doubt deliberately, the polemical intro-

49. McKeown 1979.77-8. I

50 Ovid Fasti 2.583-616, on the Feralia (21 February) and dea Muta.
51 Ovid Fasti 2.605, miserataquc ttuptas.
52 Maia as Bona Dea: Macrobills Saturnalia 1.12.20—1, citing Cornelius Labeo. Maia as the
goddess of the month of May: Ovid Fasti 5.79-106. The Bona Dea precinct was close to
Mercury’s temple: see Wiseman 1995.113, 139, 204, and also 67-71, on the myth of the
silent goddess.
53 Cf Hardie 1996.223-5, on the analogy of water-nymphs and human girl dancers.
54 Horace Odes 4.7.5-12, Ovid Fasti 5.195-220.

I
 THE GAMES OF FLORA 185

duction to Varro’s Antiquitates reruiii ditiinartttn, in which the great antiquarian

attacked traditional mythology as ‘contrary to the dignity and nature of the
immortal gods’.” And what Varro says about this deplorable type of theology
is that it belongs in the theatre.56
Much of our knowledge of Varro’s work comes indirectly from Christian
writers, who exploited it for their own polemical purposes. Their arguments
confirm that the disgraceful stories, unworthy of the gods, to which Varro
objected (but which they seized on as characteristic of paganism), were
performed on the stage at the public games.’7 Not only that, but Amobius and
Augustine, in passages clearly dependent on Varro, associate them in particular
with the games of Flora.58
V
So I think we have found three stories that have some claim to be the sort of
thing the tnitnae performed at the Indi Florales—one preserved in historical
narrative, the other two in aetiological elegy. And perhaps we can even see
how such performances may have been structured—for we need not imagine
any great complexity of plot. (No doubt choreography was more important.)
First, on come the girls, with enough dialogue to establish who they are
and what they are doing: Porsena’s hostages; worshippers of Bona Dea;
nymphs of the streams of Latium. They dance, and in due course take off their
dresses as required by the plot; it is striking that all three scenarios involve
water and streams or rivers.59 When the audience has duly enjoyed the spec­
tacle, enter the heavies: the Tarquins, attacking the girls on their way back to
the chivalrous Potsena; Hercules, causing a rumpus at a place where men are
forbidden; Jupiter (with Mercury in attendance), haranguing the nymphs,
tearing out Lara's tongue.

55 Propertius 4.9.32, Ovid Fasti 2.586; Varro Antiquitates diuinae fr. 7 Cardauns (Augustine City
of Cod (>.5): mult,i contra dignitatem et naturam imntortalimn ficta.
56 Varro Antiquitates diuinae fr. 10 Cardauns (Augustine City of God 6.5): prima theologia inaxitne
accommodata est ad t/teatrum. This was the theology quo inaxitne uttmtur poctae (fr. 7), and by
‘poets’ Varro primarily meant dramatists: cf. Augustine City of Cod 6.6 (using Varro) on di
poetici theatrici ludicri scaenici.
57 E.g. Tertullian Apologeticus 15.1—3, Amobius 4.35, Augustine City of God 6.6-8.
58 Amobius 7.33: existimatue tractari se honorifice Flora, si sttis in ludis flagitiosa conspexerit res agi et
migration ab hipanarilms in theatra? itane istud non est deornm imminuere dignitatem, dicare et
consecrate turpissimas res eis, quas censor animus respuat... ? Augustine City of God 2.27 (having
just quoted Cicero I 'errines 5.36 on the Floralid): Italic, inquant, piidendain ueraeque religioni
auersandam et detestandain talium ntmtimmt placationeni, fabulas in deos inlecebrosas atque criini-
nosas, Itaec ignominiosa deornm uel sceleritate turpiterque facta uel sceleratius turpiusqiie conficta
octtlis et atiribus publicis ciuitas rota discebat. Varro is cited by name at Amobius 7.1 and Augus­
tine City of God 3.4 (respectively frr. 22 and 20 Cardauns).
59 Cf. GritFin 1985.88-111 (on ‘the pleasures of water and nakedness’), and Traversari 1960.
Could the temporary theatres of the late Republic have been flooded: It is hard to imagine it.
 186 UNWRITTEN ROME

One imagines these scenes performed in very broad burlesque; the tongue­
tearing scene would be intolerable in any other mode. Cato, of course, is not
in the audience, and if his fellow philosopher Marcus Varro is there, he will
not approve of this treatment of the gods. But it is all done for the best of
motives. Heroic Valeria escapes from the ambush and brings help; Porsena
banishes the Tarquins and the Romans are saved.60 Hercules breaks in and
drinks his fill—and that, citizens, is why women may not worship at the Ara
Maxima. Exit Mercury, ogling the helpless Lara; she will conceive, and bear
the guardian gods of Rome. And at each satisfactory ending, no doubt, the
girls dance again before they make their bow.
To moralists like Cato and religious purists like Varro, it was all very
disgraceful. But to the Roman People, and the ambitious aediles whose job it
was to entertain them at the games, it was surely a proper way to honour the
goddess at her holiday time. For all their raunchy style, Flora’s games were like
the other dramatic festivals in providing an opportunity for the creation and
re-creation of the Romans’ concepts of their gods, their city, and their past.

60 Dionysius of Halicarnassus 5.34, Plutarch Publicola 19.5-6.

 CHAPTER TEN

The Games of Hercules

i
In 78 BC, the year of Sulla’s death, a nioneyer called M. Volteius issued the
following series of denarii [fig. 46]:
i. Obv.t Laureate head of Jupiter r. Rev.: Capitoline temple; below,
M.VOLTEI.M.F.
2. Obi’.: Head of Hercules r. Rev.'. Erymanthian boar r.; in exergue,
M.VOLTEI.M.F.
3. Obi’.: Head of Liber r., wearing ivy-wreath. Rev.: Ceres in Inga of
snakes r., holding torch in each hand; behind, control-symbol; in
exergue, M.VOLTEI.M.F.
4. Obi’.: Helmeted bust r., draped (helmet bound with laurel-wreath); behind,
control-symbol. Ret1.: Cybele, wearing turreted crown and veil, in Inga of

Fig. 46 The coins of M. Volteius, 78 bc.

 188 UNWRITTEN ROME

lions r., holding reins in 1. hand and patera in r. hand; above, control­
numeral; in exergue, M.VOLTEI.M.F.
5. Obi’.: Laureate head of Apollo r. Rev.: Tripod with snakes coiled round front
leg and rearing head above; on 1., S.C.; on r., D.T.; in exergue,
M.VOLTEI.M.F.

What these images have in common was recognised by Mommsen in 1860:

they represent, he thought, the Indi Romani (in honour of Jupiter), the ludi
plebeii, the ludi Ceriales, the ludi Megalenses and the ludi Apollinares.1
There are some problems, however. First, can the tetrastyle temple on the
reverse of no. 1 really be that of Jupiter Optimus Maximus on the Capitol?
Perhaps it can: as Michael Crawford points out, ‘since the temple had been
destroyed and not yet rebuilt at the time of this issue, the representation is not
necessarily accurate’. Next, whose is the bust on the obverse of no. 4? Duncan
Fishwick suggests ‘the Cappadocian Goddess’, whose cult was similar to that of
‘the Phrygian Goddess’ Cybele. In the civil war of 88 BC she encouraged Sulla
in a dream to smite his enemies; Plutarch’s account, which probably comes
from Sulla’s memoirs, identifies the goddess as ‘Selene, Athena, or Enyo’, and
the garlanded helmet is indeed sometimes found in the iconography of
Athena.2
What surely confirms Monunsen’s hypothesis is the combination of images
on no. 3. In the late third century BC (the evidence conies from Naevius),
Liber had his own ludi at the Liberalia; by Ovid’s time, he bad lost them and
now shared the games of Ceres.3 Evidently the same was true in 78 BC, since
he shares the coin issue with her as well. So four of the five issues represent
major annual ludi, identified by the divinities in whose honour they were held.
But what about Hercules?
Mommsen assumed that Hercules must represent the ludi plebeii, otherwise
absent from the list. But the ludi of Flora and of Victoria are absent too, so that
in itself is not a sufficient argument. More important is the fact that rhe ludi
plebeii, like the Romani, were centred on a ‘feast ofJupiter’ (epulum lotus) and
were therefore presumably thought of as offered to that god.4 Certainly there is
nothing in our surviving sources to connect the Plebeian Games with
Hercules. The logic of the Volteius coin issue seems to require that in 78 BC

1 Crawford 1974.399—402 (no. 385.1—5), whose descriptions I repeat (except 'border of dots
for each type). Ludi: Mommsen 1860.620—1 n. 451, followed by Crawford 1974.402.
2 Crawford 1974.400; Fishwick 1967.152-4; Plutarch Sulla 9.4; Lexicon Iconographiaim
Mythologiae Classicae s.v. Athena nos. 302, 305 (Thurii coins), Athena-Minerva no. 2
(Pompeii).
3 Naevius fr. 113K (Festus [Paulus] 103L), cf. Ausonius 7.24.29-30, Ovid Fasti 3.785-6. See
now Wiseman 1998.35-51 [and ch. 5 above].
4 Degrassi 1963.449-52, 525-6 (ludi Florales 28 April-3 May, ludi I'ictoriae Sullanae 26 Oct.—1
Nov.). Epula louis: Degrassi 1963.509, 530 (ludi Romani 13 Sept., ludi plebeii 13 Nov.).
 THE GAMES OF HERCULES 189

there were games of Hercules comparable with those of Ceres, Apollo and the
Great Mother.
Certainly there were games of Hercules at Rome in the first century bc.
Two late-republican inscriptions attest them. The first was found on the Via
Appia between the sixth and seventh milestones:
JR.MAG.LVDOS
herJCOLEI. MAGNO
JNEO.FECIT
The first letter is thought to be the remains of a cognomen. In the third line
Mommsen read [im theatro lig\neo; Whatmough in 1921 suggested [i/i circa
Flamijneo, which may be preferable in that the temple of Hercules Magnus
Custos stood in the Circus Flaminius. The god’s cult day was 4 June, on which
date the late-imperial calendars report ‘Indi’ {fasti Silnii) or ’Indi in Minicia’ {fasti
Filocali). The Severan marble plan identifies as A//N/[CL4] the rectangular
portico north of the theatre of Balbus, not far from the probable site of the
Hercules Custos temple (modem Via Arenula); Filippo Coarelli plausibly
suggests that that by the late empire the Porticus Minucia fmmentaria provided
a better site for the games than the Circus Flaminius piazza?
The second inset iption comes from the Caelian, near Quattro Coronati:
MAG.HE[rc
SVFFRAGlO.PAG.PRlM[i creati
LVDOS.FECERV[nt
Here too we have games given by tnagistri, and it seems that the men in charge
are not only uKioisfri He[rculani] but also tnagistri pagi, their responsibility defined
both topographically and by reference to the god. The nearest parallel may be
the Indi Tarpeii or Capirolini, set up in honour of Jupiter Feretrius on the
Capitol and organised by a collegium of‘those who lived on the Capitol and the
<ir.v’.6
Games held in a portico, or in a piazza like the Circus Flaminius (which
despite its name was not a race-track), were clearly not Indi circenses with
chariot-racing.7 Theatrical performances would be possible, and perhaps a
small-scale uenatio (Augustus once flooded the Circus Flaminius to show croc­
odiles), but these local games were clearly on a less ambitious scale than the

5 C/L 6.335 = P.985 = ILLR.P 703. Degrassi 1963.284-9 (fasti Filocali), 269 (fasti Silnii)-,
Coarelli 1997.296-345 (Porticus Minucia), 498-503 (Hercules Custos temple); cf. Zevi
1993 (esp. 679—92), whose siting of the Porticus Minucia fnniietitaria inside the Circus
Flaminius itself is refuted by Coarelli 1997.304-10.
6 CIL 6.30888 = 1-.984 = ILS 6081 = ILLR.P 701. Piso fr. 7P (Tertullian De spectacnlis 5),
Livy 5.50.4 (collegium . .. ex iis qni in Capitolio atque arce habitarent).
7 Pace Coarelli 1997.499, who restores the first inscription '.. .hides [rirceiires] .
 190 UNWRITTEN ROME

series of great annual Indi entrusted to the aediles. If the Indi Tarpeii are indeed
analogous, they may have consisted of amateur athletics and boxing.8
The magistri in the second inscription were the first to be elected by their
pagns. Mommsen thought that this might mean they were the first elected after
Clodius restored the collegia in 58 BC. However, the ban in 64 BC had
exempted religious associations, and we know that even before Clodius’ law
was passed Sex. Cloelius, as magister collegii, was holding Indi compitalicii in
honour of the Lares.9 But whether or not that was the occasion, the wording
of the inscription makes it clear that some sort of change had taken place in the
selection of those responsible for the games of Hercules.
That may offer a solution to the dilemma presented by our two near­
contemporary sources of information: the coins of Volteius imply games like
the Indi Romani and the other annual festivals put on by the aediles; the inscrip­
tions attest games put on by magistri, evidently on a much smaller scale. Is it
possible that the games of Hercules were at first what the coins imply, but were
then reorganised as merely local and entrusted to magistri elected by their
fellow-pagani?
The hypothesis is in two parts: first, that major Indi of Hercules had been
instituted before 78 BC, and second, that they were soon afterwards ‘demoted
to local games of a particular pagns. Neither event is attested, but it seems to me
that both are perfectly possible.93

II
Sulla took the gods seriously. In the dedication passage of his Memoirs he
advised Lucullus to regard dreams as reliable divine messages, and rhe Memoirs
themselves were full of prophecies, portents and communications from the
gods. Fortuna, Venus, Apollo and Bellona feature particularly, but they were
not the only deities he honoured. In 79 he consecrated a tithe of his whole

8 Circus Flaminius piazza: Humphrey 1986.540-5. Crocodiles: Cassius Dio 55.10.8 (2 Be).
Liidi Tarpeii: Ennius Aniiales 1 fr. li Sk (Scholia Bernensia on Virgil Georgies 2.384), sic Indus
edidit lit caestibus dimicarent et cinsn contenderent.
9 Cicero In Pisonent 8, Asconius 7C (cf. 75C for exemptions); Pliny Nut. Hist, 36.204 (Lares);
Lintott 1968.77-83.
9a [See Keaveney 2005 for a determinedly sceptical reaction to the second part of the hypoth­
esis. However, his own preferred solution, that the inscriptions are pre-Sullan and that
‘games that were lowly fell into the hands of Sulla, the great devotee of Hercules, and it was
he who gave them their enhanced standing’ (Keaveney 2005.223), takes no account ofpnni[i
creati] in the Caelian inscription, and fails to explain the absence of the ‘enhanced’ games of
Hercules from the calendar fasti and from Ovid Fasti 6.209—12.]
 THE GAMES OF HERCULES 191

property to Hercules, with huge banquets for the populace, and it is possible to
conjecture why the hero-god was important to him.1"
The campaigns in the War of the Allies which made Sulla’s reputation as a
general, and won him the consulship in 88 and the command against Mithri­
dates, were concentrated in Campania; also in Campania, at Mount Tifata, he
defeated the armies of Norbanus and the young Marius in 83; and it was to
Campanian Cumae that he retired, after laying down his power, to spend his
last years writing his memoirs.11 Campania, and the Bay of Naples in particular,
was a part of Italy particularly associated with Hercules. It was at the Campi
Phlegraei that the hero met and defeated the rebellious Giants, at Bauli that he
penned up the cattle of Geryon, at Pompeii that he held his triumphal proces­
sion; he built the causeway across the Lucrine bay, and he founded the town
called Herakleion, the Roman Herculaneum.12
Sulla derived his cognomen from ‘Sibylla’. The Sibyl’s home at Cumae,
which was Sulla’s home too in his retirement, was a part of this Herculean
neighbourhood. A pair of huge tusks was preserved there in the ancient temple
ot Apollo, and identified as those of the Erymanthian boar.13 Since the boar
was what M. Volteius chose to use on the reverse of his issue no. 2, it must
somehow have symbolised the games of Hercules at Rome.
Ovid is explicit that it was on the instructions of the Sibyl that Sulla set up
the temple of Hercules Magnus Custos in the Circus Flaminius:
altera pars circi Custode sub Hercule tuta est,
quod deus Luboico carmine inunus habet.
numeris est tempos, qui Nonas Lucifer ante est:
si tituluiu quaeris, Sulla probauit opus.
The other part of the Circus is under the protection of Guardian Hercules; the
god holds office through the Euboean oracle. The time of his taking office is
the day before the Nones; if you are enquiring about the inscription, it was
Sulla who approved the work.
Since the inscription from the Via Appia specifies Hercules Magnus as the
recipient of the games, it is reasonable to infer that the games were set up at the

10 To theion, to daimonion, etc: Plutarch Sulla 6.5-7 (Sulla fr. 8P), 7.2-6, 9.3-4, 14.7, 17.1-2,
27.3—1. Ibid. 6.5 (Fortuna); 7.6, 9.4, 27.6, 30.2 (Bellona); 12.4—5, 19.6, 29.6 (Apollo); 19.5,
34.2 (Venus); 35.1 (tithe to Hercules). Cf. Velleius Paterculus 2.25.4: Sulla thanks Diana for
his victory at Mt Tifata.
11 89 Be: Sulla trr. 9-10P, Appian Civil Wars 1.50-1, Orosius 5.18.22, Velleius Paterculus
2.16.2. 83 BC: Appian Civil Wars 1.84-6, Plutarch Sulla 27.4-5, Velleius Paterculus 2.25,
Cicero Philippics 2.27. Cumae villa: Appian Civil Wars 1.104, cf. Valerius Maximus 9.3.8, De
uiris illustribus 75.12 (Puteoli).
12 Diodorus Siculus 4.21.5—22.2, cf. Strabo 5.4.4 (Giants); Servius on Virgil Aeneid 7.662
(Bauli, Pompeii), Dionysius of Halicarnassus 1.44.1 (Herculaneum).
13 Macrobius Saturnalia 1.17.27, Charisius Ars granuuatica DOB (Sibylla); Pausanias 8.24.5.
 192 UNWRITTEN ROME

same time as the temple, as had happened a century earlier with the Magna
Mater and the Indi Megalenses. No doubt ‘Hercules the Great Guardian’, like
his neighbour Bellona, played some part in Sulla’s personal myth.14 What part
it was we do not know, but the events of 89 and 83 BC offered plenty of
opportunity for the protective hero to earn his temple and his games.
But what happened afterwards? Why should the games have been reduced
to the humbler sphere of the local magistri?
Here we must remember the ideological sensitivity of Indi. The Indi plebeti,
for instance, represented plebeian freedom—either from the Tarquins or from
the patricians, according to the alternative origin stories offered by a
Ciceronian scholiast. The Indi Romani were the subject of a dispute between
the plebeians and the patricians at a turning point in the ‘struggle of the orders’,
the election of the first plebeian consul; the story cannot be made sense of in
Livy s abbreviated account, but the games were evidently thought to be an
important issue. The Indi Florales were founded by plebeian aediles, and paid
for with fines exacted from landowners encroaching on public land; for over
sixty years the Senate refused to recognise the games, and only yielded when a
famine was attributed to the goddess’s anger.1’
A similarly contentious background may be inferred for the Indi Apollinares.
In 213 BC the urban praetor carried out a police action, on the Senate’s instruc­
tions, against ‘prophets and sacrificers’ in the Roman Forum. These people
were evidently considered a threat to the authority of the Senate and magis­
trates, and their prophetic books and sacrificial manuals were duly confiscated.
The following year, however, it was announced that two important prophecies
of Cn. Marcius had been discovered among this material. The Senate ordered
a consultation of the Sibylline Books; the Sibyl evidently agreed with Marcius,
and the games of Apollo were set up as a result.16 It is natural to infer that the
Senate was alarmed by the prophets’ influence, and made sure that any innova­
tions that had to be conceded would be carried out under senatorial control.
That is, the Indi Apollinares were set up in order to forestall something more
dangerous.16’
The vicissitudes of the ‘games of Liber’ may give us some idea of what that
something more dangerous’ was. As we noted in the context of Volteius coin
issue no. 3, the Liberalia had been Indi in the time of Naevius, but the games

14 Ovid Fasti 6.209-12 (4 June), cf. 199-208 for Bellona; Livy 36.36.3-4 (Magna Mater,
temple and games).
15 Ps.Asconius 217St (on Cicero Verrities act. pr. 31); Livy 6.42.12—14; Ovid Fasti 5.279—330.
16 Livy 25.1.6-12, 12.2-15. An ancestor of Sulla was involved in the consultation ot the Sibyl
(Macrobius Saturnalia 1.17.27).
16a [See now North 2000b. 100-102. Keaveney, however (2005.222 n. 45). finds this whole
argument ‘pointless and irrelevant’.]
 THE GAMES OF HERCULES 193

were later merged with the Indi Ceriales. The Naevius fragment—'libera lingua
loqiiiinur Itidis Liberalibus’—suggests that the games exploited the ideological
implication of the god’s name, and therefore that their suppression may have
had a political dimension. The most likely context for it is surely the consuls’
crack-down on the ‘Bacchanals’ in 186 BC, a particularly brutal example of
senatorial authority being exerted over what could be perceived as a rival focus
of loyalty. The leader of the Bacchic cult was a ‘prophet and sacrificer’, just like
those supposedly subversive characters in the Forum in 213 BC.17
If that could happen in the second century' BC, perhaps it could also happen
in the first, but in the opposite ideological direction. What we have inferred
from the combination of Volteius’ coins and the inagistri inscriptions is a
‘demotion’ of Sullan games. The likely context for that is the reform move­
ment of 70—67 BC, when the Sullan oligarchy was tainted with gross
corruption and abuse of power, and Sulla’s more contentious legislation was
reversed.18 There would have to be a good reason to take the god’s games out
ot the hands of the aediles, but what the reason was we can only guess. Since
Sulla’s Indi I 'ictoriae continued as a regular part of the aediles’ annual
programme, mere association with the dictator was clearly not enough to
justify it.
The Victoria games celebrated the battle of the Colline Gate, which could
quite reasonably be presented as a victory over an external enemy.19 If the
Hercules games were associated with specifically civil war campaigns, or with
victories over allies who accepted the Roman citizenship in 89 (and were now
at last enrolled by the censors of 70—69), one can imagine hostile tribunes not
wanting the spilling of citizen blood to be publicly commemorated. No
offence to the god. who would still be honoured on 4 June, but his games
would now be a local affair rather than a celebration by the people as a whole.
We cannot know whether it happened like that; but some such hypothesis
seems to be required to make sense of the conflicting evidence we have.
Perhaps rhe games of Hercules may count as another example of the inextri­
cable interrelation of‘religion’ and ‘politics’ in the Roman republic.

17 See n. 3 above; exhaustive treatment of the events ot 186 in Pailler 1988. Sacrificulus er nates;
Livy 39.8.3, cf. 25.1.8.
18 Crook, Lintott and Rawson 1994.210—15, 223—8 (Seager); 327-8 (Wiseman).
19 Velleius Paterculus 2.27.1—2 on Pontius Telesinus as a quasi-Hannibal [disputed by Keav-
eney 2005.221-2],
 CHAPTER ELEVEN

Praetextae, Togatae and Other

Unhelpful Categories

... ttel qtti praetextas uel qtti docttere togatas.

.. .whether those who have staged praetextae or those who have staged togatae.
Horace /Its poetica 288
cur hoc, togata praetexta data feitf Apollinaribus litdis dociiit popiihim.
The togata praetexta given at the Games of Apollo taught the People why this
was so.
Varro De lingua Latina 6.18
Horace clearly implies that praetextae and togatae were not the same thing;
equally clearly, Varro refers to a play that is both. That is because Varro had his
own categorisation of dramatic forms (reported for us by the fourth-century
grammarian Diomedes), which did not coincide with what Horace took for
granted.1 In exploring the reasons for that, I want to take further Patrick
Kragelund’s excellent discussion about the nature of the sources.2
It was characteristic of Varro to organise whatever subject he was addressing
into categories and sub-categories,3 and the two main genera into which he
divided drama were defined according to national costume. Greek plays were
palliatae, Roman togatae. Each genus was subdivided into four species: respec­
tively tragedy, comedy, satyr-play, mime, and their equivalents praetexta,
tabernaria, Atellana, planipes.'* Diomedes takes all this as gospel, and describes
Horace’s view as a communis error, but there is no need to accept his verdict. In
my view it is much more likely that Varro’s schema is an attempt to define the
indefinable, to impose artificial order on essentially disorderly data, and that
Horace more accurately represents the reality of his time.
It does not follow, however, that Horace was distinguishing between tragic

1 Varro fr. 306 Funaioli (Diomedes in Graniniatici Latini 1.489 Keil).

2 Kragelund et al. 2002.11-17.
3 E.g. Varro De lingua Latina 5.10—13, De re nistica 1.9.1—6, 2.1.11—12, De phiosopua ap.
Augustine City of God 19.1; see Rawson 1991.327-9.
4 Diomedes in Graniniatici Latini 1.489-90 Keil; see Kragelund et al. 2002.14 and n. _8.
 PRAETEXTAE, TOGATAE AND OTHER UNHELPFUL CATEGORIES 195

and comic Roman plays? The natural sense of the two temis he uses ought to
be that praetextae dealt with the doings of senators and magistrates, and togatae
with those of ordinary citizens. The example Varro refers to (not, in fact, a
good argument for his own categories) shows that stories of Roman history’
might be wholly un-tragic; the Nonae Caprotinae play not only has a happy
ending, for the Romans at least, but is clearly erotic in nature? Could the
scene where the girls are handed over to the lecherous Latins have been
perfonned by male actors in masks? It seems unlikely—but what place was
there for miinae in a praetexta?
Perhaps our own categories are too rigid. Any attempt to define a ‘dramatic
genre’ is bound to be artificial. Performance takes place in an infinite number
of ways, and its forms are constantly changing. Whatever the moral, political or
religious constraints in particular places at particular times, the only rule
perfonners consistently observe is the necessity of pleasing their audience.
Even the apparently fundamental distinction between tragedy and comedy
is too crude to be helpful. At least as far back as the fifth century BC, comedy
exploited tragedy, not only mocking it but seeking to borrow its authority,7
and in Periclean Athens there was recognisable common ground between
tragedy and the ‘cheerful muse’ of erotic mime? Plays entitled Konioiilotragoiilia
are known from fourth-century Athens, and probably also from early fifth­
century Syracuse.9
Evidence for drama in the west is provided by the red-figure vases of South
Italy and Sicily from the late fifth century to the end of the fourth; particularly
important are the mid-fourth-century theatrical scenes on the pottery of
Lucanian Paestum, which show a flourishing tradition of‘tragic’ mythological
themes played as btoad farce.1" At much the same time—the second half of the
fourth century—engraved bronze dstae from Latium appear to illustrate a type
of Dionysiac performance in which heroic myth is combined with erotic spec­
tacle by female performers whom one can only describe as inimae.'1 At
Tarentum in the early third century, Rhinthon specialised in a form called

5 As is normally assumed, e.g. by Kragelund (Kragelund et ill. 2002.16, Niall Rudd’s transla­
tion); judicious discussion in Brink 1971.319-21.
6 Wiseman 1998.8-1 1, 68.
7 See now Silk 2000, esp. 42-98, 415-17. For tragic parody in Old Comedy, see Bowie
2000.322-4.
8 On Gnesippus, described by Cratinus (fr. 256 K-A) as a tragic poet, see Athenaeus 14.638d
(Itai'/vtaYpatfOU if); ikapag |JOVOI];); Davidson 2000, esp. 48-9 on the Cratinus passage.
9 Alcaeus and Anaxandrides (Kassel-Austin PCG 2.9—10 and 249—50); Dinolochus (cf. Sudd
s.v. for his date), cited by the Antiatticista Bekkeri 112.29 (Kaibel CCF 149).
10 TrendaU 1989, esp. 12, 262-4; ibid. 196-232 for Paestum; Trendall 1987.433, index of
‘vases with theatrical subjects’. [See now Pontrandolfo 2000.)
11 Battaglia and Emiliozzi 1979 and 1990: nos. 9 (judgement of Paris), 22, 45, 50, 51, 82 (Iphi-
geneia at Aulis), 83; see pp. 109-27 above.
 196 UNWRITTEN ROME

‘cheerful tragedy’.12 One of his plays was Amphitryon, and Plautus’ play of that
name a century later is described in Mercury’s prologue speech as a tragico-
nioedia.13
Halfway between Rhinthon and Plautus, at just the time when the Taren-
tine Andronicus produced his first play at Rome,14 erotic mime performances
were featured at the plebeian Indi Florales.15 I think there is good reason to
believe that the repertoire of the mimae included scenarios that claimed to be
historical.16 It is certain that ‘mime’ was a polymorphous mode, ranging from
the outrageously licentious to the morally exemplary;17 we know that it could
deal with contemporary events and characters in a satirical way,18 and there is
no reason to suppose that it would regard any subject as beyond its scope.
The type of drama that was normally called fabttla togata (in Horace’s defini­
tion, not Varro’s) flourished in the second and first centuries BC: its classic
authors were Titinius, Quinctius Atta and Afranius. Like mime, it could be
obscene;19 like mime, it could be ethically improving.2" Known titles common
to both forms are Aquae caldae (Atta and Laberius), Augur (Afranius and
Pomponius), Compitalia (Afranius and Laberius), Ftillones (Titinius and three
different mime authors), Saturn (Atta and Pomponius), and I7irgo (Afranius and
Laberius). Satura is interesting, given the Dionysiac context of the perfomi-

12 Suifa s.v. (dp/tiydg Tfjg KaXovpevrjg IXaporpayoidiac); Stephanas of Byzantium s.v.

Tams ('PlvOwv ... rd rpayucd perc(ppuO|iit,<i)v eg to yeXoTov).
13 Rhinthon fr. 1 Kaibel (Athenaeus 3.111c), Plautus Auiphitruo 50—63; Christenson 2000.7—
12, 47-50.
14 240 BC: Varro De poetis fr. 55 Funaioli (Aulus Gellius 17.21.42), Cicero Biutus 72, De senee-
tute 50; Oakley 1998.61—3. Livius Andronicus from Tarentum: Accius fr. 18 Funaioli
(Cicero Brutus 72).
15 241 bc: Velleius 1.14.8. 238 BC: Pliny Nat. Hist. 18.286. Made annual in 173 bc: Ovid Fasti
5.295—330. Mimae, erotic: Valerius Maximus 2.10.8 (priscuni tuareut iotvnim); cf. Ovid Fasti
4.946 (tori luoretu libcrioris), 5.331—54, Seneca Epistles 97.8, Martial 1 pref, and 1.35.8—9.
Plebeian: Ovid Fasti 5.279-94, cf. 352.
16 See pp. 175—86 above, esp. 180—5 on Dionysius of Halicarnassus 5.33.1 and Plutarch Piibli-
cola 19.1 (Cloelia and the Tarquins), Propertius 4.9.21—70 (Hercules and Bona Dea), Ovid
Fasti 2.583-616 (conception of the Lares). Cf. also Wiseman 1998.64—74, esp. 72-4 on
Ovid Fasti 3.661-96 (Anna Perenna and the secession of the plebs).
17 Mime and ethologia: Diogenes Laertius 3.18 (Sophron and Plato), Cicero De oratore 2.242-4.
On Publilius’ moral maxims (cf. Seneca De tranqiiillitate animi 11.8), see Giancotti 1967.275—
462; Rawson 1991.570-1.
18 E.g. Cicero Ad familiares 7.2.3 (53 bc), Ad Atticnm 14.3.2 (44 BC); Suetonius Domitian 10.4
(ad 93), where exodinm probably refers to mime. For Laberius’ treatment of contemporary
issues, see Wiseman 1985.187-8.
19 Quintilian 10.1.100, Ausonius Epigrams 67.2—4 (both on Afranius). Priapus was appropriate
to both modes: Macrobius Saturnalia 6.5.6 (Afranius), Augustine City of God 6.7 (mime).
20 Seneca Epistles 8.8: non attingain tragicos nec togatas nostras, habent enim hae qnoqne aliquid setter-
itatis et sunt inter comoedias et tragoedias mediae. Afranius compared with Menander: Cicero De
Jinibus 1.7, Horace Epistles 2.1.57, Macrobius Saturnalia 6.1.4.
 PRAETEXTAE, TOCATAE AND OTHER UNHELPFUL CATEGORIES 197

ances illustrated on the fourth-century cistae, and the evidence of Vitruvius and
Horace for satyr-play in Rome.21 Compitalia is interesting too: if it was an aeti­
ology for the ludi coinpitales, it may have had a ‘historical’ plot about Servius
Tullius or Lucius Brutus.22
Two togata authors, Atta and Afranius, wrote plays called Megalensia, for
which the coming of the Great Mother in 204 BC is clearly a possible subject.
As Patrick Kragelund rightly points out,23 the play about Quinta Claudia to
which Ovid refers is analogous to the Nonae Caprotinae play mentioned by
Varro. He assumes they were praetextae, which may be right in one sense, since
Roman magistrates must have featured in both. But neither can have been
remotely like a tragedy; the main action of both was essentially comic.24 Wher­
ever we look, the Varronian categories seem to dissolve.
An added complication came in Varro’s own time, with the introduction
by Bathyllus and Pylades of the solo ballet sometimes called pantoniiinus. The
plot might be a tragedy, a comedy, an Atellana, a togata, or a theme from a poet
like Virgil or Ovid.2’ Since Lucian includes Polycrates of Samos and Seleucus
and Stratonice among the themes appropriate to tragic dance,26 there was
evidently no prohibition on historical subjects; nor on contemporary ones,
since fulsome flattery of the reigning emperor was always a safe and popular
theme.27
I am not wholly convinced by Patrick Kragelund’s idea of a ‘transition from
the republican system focused on the ludi to the more diversified imperial
literary scene’, in which ‘the ludi no longer were a viable setting for aristocratic
self-promotion' 26 A precious but neglected description of ludi in Augustan
Rome reveals an aristocrat in charge (erat Jactnrus ludos quidaiu nobilis),-' and just
the same creative mixture of performance genres as we have detected in the

21 Cistae: n. 1 1 above. Satyr-play: Vitruvius 5.6.9 (cf. 7.5.2), Horace /Irs poetiea 220-50;
Wiseman 199-1.68-85.
22 Pliny Nat. Hist. 36.204 (Ser. Tullius), Macrobius Saturnalia 1.7.34-5 (Brutus). Performances
at eompita: Propertius 2.22a.3-8.
23 Kragelund et a/. 2002.18-20.
24 Ovid Farti 4.293-326, with Fantham 1998.154: ’these lines sketch a kind of comedy.’ See in
general pp. 210-11 below.
25 E.g. [Seneca Cuiirrunersiae 3.pref. 10 (Pylades for comedy, Bathyllus for tragedy);] Juvenal
7.86—92 (tragedy plots), Plutarch Quaestiones conuittiales 7.8 (comedy), Tertullian De spectac-
ulis 17.2 (Hrei/ana), Pliny Nat. Hist. 7.159 (togatae), Suetonius Nero 54 (Virgil’s Tumus),
Ovid Tristia 2.519 and 2.7.25 (inea poeniata and cannina nostra). [Possibly also satyr play: See
Persius 5.123 on Bathyllus’ Satyr.)
26 Lucian Saltatio 54, 58; cf. 72 on dance as a source of knowledge about the past (blbaOKOUOCL
de nokka raiv jrakai).
27 E.g. Domitian: Pliny Panegyric 54.1, clearly referring to pantoiniini (cf. 46.4); for the eonnnis-
sioues, cf. Suetonius Diuus Augustus 89.3 (Augustus discouraging flattery).
28 Kragelund et al. 2002.34—5.
29 Phaedrus 5.7.16 (cf. 5.5.4—10, dines et nobilis).
 198 UNWRITTEN ROME

republican evidence. Phaedrus is telling the story of a tibicen called Princeps,

Bathyllus’ regular accompanist, who had fallen and broken his leg while
‘flying’ on the stage-crane in the course of his performance. Now conva­
lescing, he makes a guest appearance at the noble’s games:30

aulaeo misso, denolntis tonitribus,

di sunt locuti more translaticio.
time chorus ignotum modo reducto caiiticum
insomiit, cuius haecfuit sententia:
'laetare incolumis Roma saluo principe. ’
The curtain was removed, thunders were rolled, and the gods spoke in the
traditional way. Then the chorus started singing a song unknown to the man
just returned, which went like this: ‘Rome, rejoice, you are safe while the prin­
ceps is safe.’

The thunder rolls; the gods speak ‘in the traditional way’, perhaps in a
prologue (in Plautus’ time gods speaking from the stage were a feature of
tragoedid);31 the chorus sings a patriotic song about the safety of the emperor. It
could be the start of a praetexta about the safe return of Augustus in 24 or 19 or
13 BC—or it could be a mime, a togata, or any mixture of the available forms.
So multifarious is the theatrical world implied by the sources that to say ‘it .
must have been’ (or ‘it cannot have been’) this form or that would be merely
to announce one’s own prejudice. The evidence does not allow us to make
that sort ofjudgement.
The same applies to the Octavia, about which Patrick Kragelund’s discus­
sion seems to me exemplary.32 The text presents itself as a play for
perfonnance; what sort of evidence or argument would be needed to show
that it cannot be what it purports to be? The text makes repeated reference to
the traditional liberties of the Roman People;33 what sort of evidence or argu­
ment would be needed to show that it was written not for a popular audience
at the Indi scaenici, but for a literary coterie at a private recitation
One feature seems to me decisive, the absence from the text of the day of
the royal wedding. The ghost of Agrippina refers to it in advance (lines 593-7),
and Poppaea’s nurse describes it in retrospect (693—709); in the intervening
scene, the chorus of citizens tells us it has already happened (671—3, 683-5). So
more than twenty-four hours have passed, without explanation, between
Agrippina’s exit at 645 and the entrance of Octavia and the chorus at 646. At a

30 Phaedrus 5.7.23-7; Henderson 2001.95—118, cf. 220 n. 42, where Henderson is surprisingly
uncurious about what sort of show it was.
31 Plautus Amphitruo 41—4.
32 Kragelund ct al. 2002.7-11, 41-50.
33 Kragelund 1982.
 PRAETEXTAE, TOGATAE AND OTHER UNHELPFUL CATEGORIES 199

redtatio, where the text has to do all the work, that would be an artistic failure.
In a stage performance, music and action can carry the plot forward without
words; and as we have seen, the Roman theatrical tradition offered many
different styles in which the haughty harlot’s wedding-orgy could be played.'4
Was there a play about the fall of Messallina in AD 48? Tacitus’ description
of her marriage to Silius reads like a satyr-play scenario.3’ What Tacitus’ later
text owes to the Octavia no doubt comes indirectly (perhaps through Cluvius
Rufus, a historian much involved with Neronian theatre),31' but is nevertheless
important evidence for the way such plays might affect historical narrative. We
can see that clearly enough in Livy’s stories of Tullia and Verginia, and may
suspect it also in Plutarch’s treatment of Gaius Gracchus.37
So when Octavia attributes Messallina’s downfall to the anger of Venus,31*
and the chorus dwells on the crime of Tullia and the deaths of Verginia, the
Gracchi and Messallina herself,39 I think we are entitled to infer (at least provi­
sionally) an on-going tradition of historical drama from which, much more
than from historians, the Roman People derived its knowledge of the signifi­
cant past. Whether the literary elite called such plays praetextae, togatae, or
anything else, seems to me of peripheral importance. For understanding the
history of Roman drama, the Varronian categories are more of a hindrance
than a help.

34 I have sketched one possibility in Wiseman 2004a.265-72. Snpcrba paelex: Octavia 125.
35 Tacitus Annals 11.31.2, strepente circnni procaci choro; the doomed Silius wears tragic cothurni.
36 Tacitus HwiuZs 14.60.5 (Octavia 896—8), 63.2 (Octavia 933—46). Cluvius: Suetonius Nero
21.2, Dio Cassius 63.14.3; Wiseman 1991.111-18.
37 Tullia: Livy 1.46.3 (tragicnni scelus), 59.13 (Furies). Verginia: Livy 3.58.10-11 (ghost). Grac­
chus: see now Beness and Hillard 2001.
38 Octavia 257—9; Messallina had caused the deaths of two of Venus’ descendants, Julia Livilla
and Livia Julia (Seneca Apolocyntosis 10.4, Dio Cassius 60.18.4).
39 Octavia 304—8 (Tullia), 295—303 (Lucretia and Verginia), 882-6 (Gracchi), 947-51 (Messal­
lina). Lucretia was the tragic heroine of plays by Accius and Cassius: Kragelund et al. 2002.21
and nn. 46—7.
 CHAPTER TWELVE

Octavia and the Phantom Genre

Octavia: A Play Attributed to Seneca, edited with introduction and commentary

by Rolando Ferri (Cambridge Classical Texts and Commentaries 41),
Cambridge U.P., 2003. pp. x + 471.

i
In one sense, as an example of what traditional classical scholarship can do with
a Latin text, Rolando Ferri’s commentary on the pseudo-Senecan Octavia is
superb. On manuscript readings and textual questions, on Latin idiom and
semantics, on prosody and metrical technicalities, his detailed notes provide a
treasure-house of erudition and judicious comment, with particularly thor­
ough citation of parallels in the Senecan corpus and other ’Silver Latin’
authors. A worthy addition to the ‘orange cover’ Cambridge series, it will be
an indispensable aid to the study of this important and surprisingly neglected
text. In another sense, however, those who are interested in rhe history and
society of Rome as well as her language and literature may find it less than
completely helpful.
Ferri begins his Introduction with a short section on the genre of Octavia,
announcing somewhat dogmatically that reference to the play as Octavia prae-
texta is ‘a practice which should be abandoned’. He sees it as influenced much
more by Greek tragedy than by the republican fahtilae practextae, and concludes
with this firm statement (pp. 2-3):
Political caution, a propensity for themes increasingly irrelevant to popular
audiences at large, and a long-tenn process of ‘gentrification’ of literature at
Rome made practextae more suitable for recitation in the auditoria of a few aris­
tocratic patrons than for onstage performance before large theatre audiences.
The only evidence offered for this far-reaching generalisation is a reference
forward to section 5 of the Introduction (‘Structure and Dramatic Technique
in Octavia), where the same view is taken as already established (pp. 54—5):
The cultural setting in which Octavia must be placed is that of the flourishing
production of dramas for recitation in Roman literary circles. Abundant
evidence for the popularity of tragedy in these circles comes from Pliny,
 OCTAVIA AND THE PHANTOM GENRE 201

Juvenal and Martial. Whether any of these dramas were in fact composed for
regular theatre performance is doubtful, and a multiplicity of factors have been
thought to account for this: the lack of state patronage, the corruption of
popular taste, a snobbish reluctance, on the part of the elite, to produce elabo­
rate dramas in the style of fifth-century Athenian tragedy and expose them to
the whimsical reaction of the uneducated.
Note ‘must’ in the first sentence, insisting on the relevance of one genre alone,
‘the Senecan Rezitationsdrania’ (p. 56). Note too the confident assumption of
‘lack of state patronage’, which an author as well read as Fem must know to be
untrue: Tiberius didn’t put on shows, but all the other emperors did, and of
course the praetors continued to present the regular Indi scaenici every year; one
of the playwrights we happen to know about (because the theatre audience
barracked him) was P. Pomponius Secundus, consul in AD 44 and author of
tragedies, including a praetexta entitled /leneas.1
It is important to understand that there is no ancient evidence for dramatic
texts written solely for recitation. Otto Zwierlein coined the phrase Rezita-
tionsdrama to describe what he believed Seneca’s tragedies to be;2 but of course
there is a lively and still unresolved debate on whether (or how) Seneca’s
tragedies were staged, and it is by no means clear that Zwierlein’s assumption
is correct.3 The scene of Tacitus’ Dialogic is set the day after Maternus’ tragedy
Cato had been ptesented at a recitatio, and the interlocutors refer repeatedly to
that type of presentation; but when Marcus Aper says to him ‘1 summon you
from the lecture-room and the theatre to the law court’ it is clear that Maternus
is not thought or as v. nting for recitatio alone.4 The whole point is that his plays
were dangerously controversial; there would be no chance of having them
staged under the present circumstances, but that does not mean that Maternus
never wanted them staged at all.
What there is evidence for is the playwright presenting his drama to a small
audience of friends before risking it in front of the poptilns Romanns in the
theatre;5 but that is something very different. As Pliny observed, if you write
tragedy you need actors and a stage.6 It is surely counter-intuitive to imagine

1 Tacitus Annals 11.13.1: ,-lr Claudius ... theatralem populi lasciuiam seneris edictis iucrepuit, quod
in I’ubliiiiii Poniponiinn consularein (is cannina scaenae dabat) ... probra iecerat. Tragedies: Quin­
tilian 10.1.98, Pliny Letters 7.17.11. Aeneas: Charisius in Gramniatici Latini 1.132 Keil;
Manuwald 2001.243-8.
2 Zwierlein. 1966, esp. 156-65.
3 See for instance Kragelund 1999, Harrison 2000.
4 Tacitus Dialoqus 10.5: mine te ab auditoriis et theatris infomm et ad causas et ad uera proelia uoco.
Cf. 2.1, 3.3, 9.4, 10.2, 11.2 (recitatio); 9.3, 10.7 (auditoria).
5 Horace Ars poetica 386—9, 438-52 on honest criticism (cf. 419-37 on flatterers), 474 for the
recitator. Contrast 125, 179-92 (scaena); 113, 153-5, 248-50 (theatre audience).
6 Pliny Letters 7.17.3: ... air trqgoediam, quae non auditorium sed scaenaui et adores, cur lyrica, quae
non lectoreni sed clioruni et lyrani poscunt.
 202 UNWRITTEN ROME

an author choosing a dramatic genre with no expectation at all of dramatic

performance. When Ferri supposes that ‘a reciter could have supplemented his
words with expressive gesture or even prefaced his recitation with a short
description of what was to happen’ (p. 55 n. 137), the reader may wonder
about the value of a hypothesis that needs such special pleading to make it
plausible.
It is very striking that Ferri has Octavia ‘come out of the palace’ at the start
of the text, ‘withdraw into her apartments’ at line 33, and ‘return on stage’ at
line 57 (pp. 120, 122). We also have the Nurse’s ‘appearance on stage’ at line
34 (p. 134); Poppaea ‘appears on stage’ at line 690 (p. 321), and the chorus
comes on stage’ at line 762 (p. 342). Ferri does sometimes use scare-quotes
(e.g. p. 308, ‘as she leaves the “stage”. . .’), but it seems clear enough that
normally as he reads the text he imagines a dramatic scenario and not a recita­
tion. No doubt he would say that he is imagining an imaginary dramatic
scenario (p. 226: ‘conventionally portrayed as “outside” the royal palace,
Seneca nervously paces up and down. ..’), but I think most readers will find
that an unnecessary complication.
Why should the text not be written for performance on the stage? Ferri
seems unaware that the burden of proof is on him: the default position must be
that a text in dramatic form is a text for performance, unless it can be shown
that that is impossible. Ferri does look for non-dramatic characteristics at pp.
59-63, but the results are unimpressive, falling far short of the absolute
cogency his position requires. The fact is that he assumes the 'Reziiationsdrama
hypothesis’ a priori, and thus feels no obligation to prove it.

II
In the same way, he assumes a late date of composition. ‘Consideration of the
play’s structure strongly suggests that it was composed by someone who
worked from written sources’ (p. 9)—i.e. after the appearance of what Ferri
calls ‘the historical vnlgata', the histories of Cluvius Rufus, Fabius Rusticus and
the elder Pliny. His principal reason for this seems to me quite extraordinary.
‘We may imagine that in 68-69 much of what happened in the household of
the princeps was still shrouded in obscurity’ (p. 10); ‘unless the auctor Octaviae
was one of the political protagonists of the Neronian court, he would have
found it impossible to have first-hand, independent knowledge of the inside
story of imperial intrigues’ (p. 16); ‘this dwelling on the details and personalia
ofJulio-Claudian Rome presupposes an “audience” well acquainted with the
stories of court intrigue of the period, as perhaps made popular by Flavian
historians’ (p. 398).
I doubt if many readers will share Ferri’s belief that nobody in Rome knew
what happened in Nero’s court until the Flavian historians told them. This
 OCTAVIA AND THE PHANTOM GENRE 203

very text refers to the glamour of the court (fnlgor, line 34) and the constant
operation of Jama (line 67), and as the note on the latter passage rightly
observes, ‘the role of fania in creating public opinion at Rome, and often
bringing about important events, is constantly stressed in Tacitus’ historical
works’ (p. 314). Besides, Ferri himself points out that the character of Nero in
Octavia is not the same as in Tacitus, Suetonius and Dio. He explains it by
‘reasons of literary decorum’ leading the playwright ‘to sketch a figure more
suited to tragedy’ (p. 248)—but one might prefer to suppose that the play­
wright simply wrote before the ‘historical vnlgata’ had been created.
That seems particularly clear in the case of one key element in the plot, the
explicit linking of the popular rising in support of Octavia and the fire of
Rome two years later. We know from Tacitus that his sources were divided
over whether Nero was responsible for the fire;7 but those who said he was
(reflected in Suetonius and Dio) evidently did not attribute to him the motive
of vengeance on the popnhis Romanns which is so striking a feature of the play.
No ‘historical vnlgata' here, then (tacitly admitted at p. 363); and the emphasis
on the people’s hunger at line 833, evidently referring to the brief food
shortage after the tire (p. 365), might suggest that the playwright was a
contemporary, reflecting the immediate impact of events. But no: at p. 15
Ferri interprets the invention of the vengeance motive
as a clear sign that the author was composing from written sources, and
collecting in his portrayal of Nero everything that was best known and, as it
were, typical about Nero: Nero had to be recognizable as the legendary emperor
that everybody bad read about.
To describe that as a non seqnitnr would be putting it mildly.
Over twenty years ago, simultaneously but independently, Patrick
Kragelund and T.D. Barnes made the case for the reign of Galba (June 68 to
January 69) as the date of Octavia.3 Ferri discusses their arguments at pp. 5-9,
and is unconvinced. In particular, he dismisses Kragelund’s inferences from the
‘populism’ of the play (p. 7):

[T|he lines in which the chorus summons itself to rebel against the princeps
display little Republicanism: the rebels only want to restore the Claudian
princess to her legitimate share in the government ... The praetexta is remark­
ably vague and non-committal on all constitutional issues regarding the position
of the princeps. No traces of the so-called Senatorial opposition under Vespasian
can be detected. The language in which political issues are discussed applies to
situations which range through the whole of the first century.

7 Tacitus Annals 15.38.1: forte an dolo prindpis incertuni (nam utminque airfares prodidere).
8 Kragelund 1982, esp. 38—54; Barnes 1982.
 204 UNWRITTEN ROME

Ferri prefers to see the author’s motive as exploiting the Flavians’ supposed
sympathy for the memory of Claudius (pp. 16—17). In that case it is surprising
that there is nothing in the text that even hints at the coming dynasty, but that
is an argument Ferri legitimately makes about Galba too. Much more impor­
tant is the reference to the ‘Senatorial opposition’, and his later judgement that
‘the authority of the “Roman people” .. . need mean no more than the
Roman Senate’ (p. 315). But the Roman Senate did not regard the Gracchi as
heroes (lines 882-6, no comment in Ferri’s notes); far from being language
common throughout the first century, that passage explicitly contradicts the
views of Velleius Paterculus, Valerius Maximus, and Lucan.9
I think the problem here is the basic premise of Ferri’s political analysis, as
made clear at p. 316:

It is risky to draw any conclusions about the author’s political stance from the
confidence that the people may be won over by a just cause . . . Rousing the
people to an insurrection has no parallels in extant Roman imperial history .. .
This nostalgic celebration of Republican Rome does not lead to an open
condemnation of the Principate as such.

Of course there is no sign in the text that the populus Romanns wanted to over­
throw the principate. In 49 BC, the first Caesar crossed the Rubicon ‘in order
to free the Roman People from its oppression by an oligarchic faction , a
motive repeated almost word for word by his adopted son when he marched
on Rome five years later;111 in AD 41, when the consuls restored the Republic
after the death of Caligula, the People were delighted at the prospect of
Claudius being imposed as emperor, because they disliked senatorial authority
and needed the protection of a princeps ‘to keep the Senate’s rapacity in
check’.11 The Caesars were the People’s tribunes (tribunicia potestas meant what
it said),12 and the last thing the populus Romanns wanted was to go back to an
oligarchic Republic. In Octavia the people dislike Nero not because he is a
princeps but because he is an overbearing tyrant. They love the daughter of
their old emperor, just as they had loved the daughter of Augustus, and now,
as then,13 they react with violence when their favourite is wronged.
An emperor could lose the People’s allegiance by abandoning his responsi­
bility to protect them. Nero realised that in 64, when he decided not to go to

9 Velleius Paterculus 2.3.2, 2.6.2; Valerius Maximus 4.7.1, 7.2.6b, 9.4.1; Lucan 6.796.
10 Caesar Civil War 1.22.5 (his own speech at Corfinium): tit se ct poptihiin Romanian Jactione
pauconnn oppresstini in libcrtatein iiindicaret. Augustus Res gestae 1.1: rem pnblicant a dominatione
factionis oppressam in libcrtatein uindicatii.
11 Josephus Antiquities 19.228; cf.l 15, 158 (and 189, 272 for a different view, probably from a
different source); Suetonius Claudius 10.4, Dio 59.30.3.
12 Tacitus Annals 1.2.2 on Augustus: ... ad tuendain plebeni tribuuicia itire contention.
13 Dio 55.13.1 (demonstrations in AD 3).
 OCTAVIA AND THE PHANTOM GENRE 205

Greece;14 but the effect of the fire more than cancelled out the popularity he
earned by staying in Rome, and at the end of 66 he went to Greece anyway.
Popular reaction to his death was mixed, but it is clear that some, at least, of
the populace were delighted at the news.15 The circumstances of summer 68
were unprecedented—the last of the Caesars dead, having failed the People—
and nobody knew at the time what was going to happen next. The attitudes
presupposed in Octavia may be easier to understand in that context than under
the safely established successor dynasty of the Flavians.
Ferri, however, has convinced himself that the author of Octavia was writing
(for recitation) under Domitian. And he places him very specifically (p. 26):
The author of Octavia may have been an old pupil of Seneca, a survivor left
with little to rejoice at by the advent of the Flavian dynasty, who had found a
haven in the house of the last surviving grand ladies of that circle [e.g. Lucan’s
widow Polla Argentaria], and set about composing Octavia hoping to ingratiate
himself with them.
That would be surprising, given the anti-Senecan reaction of the Domitianic
period (pp. 26—7); nor is it clear how many of Seneca’s pupils were ‘on the
margins of the literary field’, with ‘a mind hill of remembered verse, but ... ill
at ease with the tools of those who had made them’, as the anonymous author
is described later (pp. 31, 39). However, Ferri himself describes this idea as a
fantasy (p. 72), and it is hard to disagree.

III
If the hypotheses put forward about the nature and date of Octavia by a Latinist
of Ferri’s quality are not acceptable, the critic has an obligation to offer some­
thing better. 1 think that can be done, but only if we rid ourselves of
preconceptions about ‘the corruption of popular taste’ or ‘the gentrification of
literature’.
It is very easy, especially for classicists, to forget that drama is perfonnance
first and literature only second (if that). The aediles of the Republic had about
fifty days of Indi scaenici per year to fill with spectacle for the Roman People—
and that only counts the regular festivals, not the extra Indi for triumphs,
funerals, temple dedications and so on. For the praetors under the principate
(they were given the responsibility in 23 Be), the total was substantially more.
How much of what they put on had written texts that would be preserved
afterwards? The texts of Plautus, Terence and Seneca predispose us to think
that comedy and tragedy on the Greek model were the norm, and that seems

14 Tacitus Annals 15.36.3—4.

15 Suetonius Nero 57.1, Dio 63.29.1; cf. Tacitus Histories 1.4.
 206 UNWRITTEN ROME

to be confirmed, at least as far as tragedy is concerned, by the surviving frag­

ments of lost plays. The index of Otto Ribbeck’s Tragicorum Roinanorum
fraginenta lists the tides of 141 tragedies on Greek mythological themes, but of
only sixteen fabulae praetextae on Roman subjects. However, there are good
reasons for supposing that those statistics may be misleading.
In the first place, fabulae praetextae were sometimes on topical subjects. The
classic examples are Naevius’ Clastidium (on M. Marcellus’ victory in 222 bc),
Ennius’ Atnbracia (on M. Nobilior’s victory in 189 BC) and Pacuvius’ Paullus
(probably on L. Paullus’ victory in 168 bc). But the very fact that they are
classic examples—plays so famous that their texts were preserved—may make
them untypical. More valuable for our purposes is the evidence in Pollio’s
letter to Cicero in June 43 BC, reporting the games organised by L. Balbus in
Spain: one of the plays put on was a praetexta about Balbus’ own part in the
civil war five years earlier. There is no reason to suppose that the text of that
was ever preserved, and we would never have known about it if Pollio’s letter
had not been incorporated into Cicero’s collected correspondence. There may
have been many such plays on topical subjects, significant at the time but soon
forgotten, which never found their way into libraries and were therefore never
quoted by scholars and grammarians. ‘C’etaient des pieces de circonstance,
wrote Gaston Boissier in 1893,16 ‘et le plupart n’ont pas du survivre aux
circonstances memes pour lesquelles on les avaient faites.’ In which case the
Ribbeck statistics tell us nothing about the theatrical reality of the time.
Secondly, the evidence of authors who happen to mention theatrical matters
seems to imply that Roman themes were important. Plautus, for instance, in the
prologue to Ainphitnio, refers to the way gods like Neptune, Virtus, Victoria,
Mars and Bellona appear in tragedies to tell the audience what benefits they
have conferred on Rome. Two hundred years later, Phaedrus reports a
performance that began with a roll of thunder and ‘the gods speaking in the
traditional way’, followed by a choral song of thanks for the safety of the priit-
ceps.'7 It is very likely that topical themes under the principate involved much
loyal flattery; Augustus had to tell the praetors to keep it to a minimum, and
Trajan too did his best to discourage it after gross excesses under Domitian.
Themes from Roman history are of course attested in the fragments
(Ennius’ Sabinae, Accius’ Brutus, and so on), and historical writers not infre­
quently comment that stories they have found in their sources seem
appropriate to plays. Livy calls the death of Servius Tullius ‘a crime from
tragedy’, and remarks that the transfer of Juno Regina from Veii to her new

16 Boissier 1893.107.
17 Plautus Ainphitnio 41—4; Phaedrus 5.7.16—27.
18 Suetonius Diuus Augustus 89.3 (cf. 43.5 and Cicero Ad Atticuiu 16.5.1 for commissio referring
to ludi scaeuici); Pliny Pancgyricus 54.1-2.
 OCTAVIA AND THE PHANTOM GENRE 207

home on the Avenrine involved miracle stories ‘more appropriate to the

stage’.” Plutarch makes a similar comment about the foundation story, and
Dionysius about the battle of the triplets and the Fabii at the Cremera.20 The
strong inference, that there were indeed influential plays on these subjects, is
confirmed in the case of the foundation story by Naevius’ known play (or
plays) on Romulus;21 but all we need to insist on is a weaker inference, that
these were the sort of historical stories that plays might be written about.
Livy’s treatment of the Juno Regina story suggests that drama could
provide aetiologies for particular Roman cults. When Ovid tells of the coming
of the Magna Mater to Rome and the miraculous vindication of Quinta
Claudia, he comments that the story is ‘amazing, but attested by the stage’.22
And then there is Varro’s account of the Nonae Caprotinae ritual on 7 July,
explained by a pseudo-historical story set in the immediate aftermath of the
sack of Rome by the Gauls: ‘the People learned the reason for this from the
togata praetexla given at the Games of Apollo.’23
It may be significant that Varro does not attribute that play to an author.
Does it date back to a time before the magistrates responsible for the games
dealt with authors at all? (We might imagine them hiring the actors to do a
performance, and telling them what was required.) If that is the case, then the
fact that Varro knows the play may imply that it was repeated every year as a
traditional item on the programme, part of the educational function of the
games. The same may be true of plays about historic Roman victories. When
Horace complains of the modern taste for mere spectacle in the theatre, his
example, featuring a four-hour triumphal procession, is of a play on the taking
of Corinth in 146 lie.24 That was hardly a topical subject at the time he was
writing; but it may be evidence for the way the citizens of Rome acquired
their knowledge of rheir own history.
Of course such conjectures cannot be proved. All I want to insist on is the
importance of Roman themes in the normal business of the Roman stage.
Two final witnesses may be enough to prove the point. Here is Horace again,
this time in the Hrs puetini:25

nil inteniptatuni uostri liquere poetae,

hit niiniinum ineruere decus uestigia Craeca

19 Livy 1.46.3 (sce/eris tragici exemption), 5.211.9 (ad ostentationeni scaenaegaudentis niiraailis aptiora).
20 Plutarch Romulus 8.7; Dionysius of Halicarnassus /bit. Rom. 3.18.1, 9.22.3.
21 Varro De lingua Latina 7.107, Festus 334L, Donatus on Terence Adelphoc 537; Manuwald
2001.141-61.
22 Ovid Fasti 4.326: mira sed et scaena testificata loquar.
23 Varro De lingua Latina 6.18: air hoc, togata praetexta data eis Apollinaribus ludis docuit populum;
Manuwald 2001.66—71.
24 Horace Epistles 2.1.187—93; Manuwald 2001.71—5.
25 Horace Ars poetica 285—8; for the sense of the fourth line see above, pp. 194-5.
 208 UNWRITTEN ROME

ansi descrere et celebrare domestica facta,

uel qui practcxtas uel qui docuere togatas.
Our poets have left nothing unattempted, earning not the least glory when they
have dared to leave the Greek tracks and celebrate our own deeds, whether
producing praetextae or togatae.
A generation later, Manilius tells us that one born under the constellation
Cepheus may be a talented actor, playing either Romans or Greek heroes:
scaettisque togatos | ant magiios heroas aget.26 There is no evidence here for ‘the
progressive disappearance of praetextae from the stage’ (Ferri p. 2).

IV
Having suggested above that ‘the Senecan Rezitationsdraina' is a phantom genre,
created out of preconceptions about Seneca and a misunderstanding of the recita-
tiones of Matemus in Tacitus’ Dialogus, I need to offer something in its place.
Here too there is evidence, if one is prepared to look for it, which shows authors
continuing to write for stage performance throughout the early principate.
Ovid counters Augustus’ charge of obscenity against his Ars ainatoria by
pointing at the plots of erotic mimes, where successful adulter)' always wins the
prize. Then he goes on:27
quoque minus prodcst, scaena est lucrosa poetae,
tantaque non porno crimilia proctor emit.
inspice ludorum stimptus, Auguste, tuoriim:
empta tibi magno talia miilta leges.
The stage is profitable for poets, the less edifying it is. The praetor pays high
prices buying such crimes. Take a look at the accounts of your own games,
Augustus: you’ll see that many such plots were expensively bought for you.
Writing for money isn’t the most glorious way of achieving literary' fame, as
the elder Seneca pointed out about Abronius Silo, who wrote scripts for the
pantoniimi (‘he didn’t abandon his gift, he polluted it’).2K But that is not the
point at issue here. Lucan himself left fourteen volumes of salticae fabiilae;
Statius sold his Agaue to Paris, the star actor-dancer of his day.29 Martial did not
think it at all demeaning; he lists the literary specialities of his friend Varro
Sophoclean tragedy, Horatian lyric, mime and elegy—without any sense that
the third genre was less respectable than the others.30

26 Manilius Astronoinica 5.282 (1 follow G.P. Goold’s ordering of the text in the Loeb edition).
27 Ovid Tristia 2.507-10.
28 Seneca Suasoriae 2A9 (pamoniimisfabulas scripsif). Cf. Anthologia Palatina 9.543 (Crinagoras 39
Gow-Page): Philonides writing for Bathyllus.
29 kite Lucani; Juvenal 7.86-7.
30 Martial 5.30.1—4 (faamdi scacna Catidli at line 3).
 OCTAVIA AND THE PHANTOM GENRE 209

Authors had to live, and there was a market to supply. The biggest demand
was for the paiitoiiiiini, who drew huge crowds of enthusiastic and sometimes
violent fans. The subject matter of their art was totally eclectic, including
Roman themes as well as Greek. Pliny tells us that Stephanie, who had the
distinction of performing in the Secular Games of both 17 bc and AD 47, was
the first to dance as a togatus; Nero himself tried to recover his popularity in 68
by offering to dance Virgil’s Tumus.31 Like modem dramatists writing for tele­
vision, Roman authors evidently adapted traditional genres to the new
medium, and since Lucian includes ‘teaching historical events’ among the lofty
aims of the actor-dancer,32 the praetexta must have been given the same treat­
ment. Even those who wrote in the traditional form were no doubt affected by
the new theatrical conditions. Ferri briefly notes the possible influence of
mime on the Ocrui'ki (pp. 294, 309, after D.F. Sutton); 1 think it can be
assumed, and exploited as a solution to the main problem in the text as we
have it, the way the marriage of Nero and Poppaea seems to have been ‘spliced
out of the sequence of events’.33
Lucian defined the possible range of the actor-dancer’s material as every­
thing ‘from the emergence of the world out of Chaos to the days of Cleopatra
of Egypt’.34 But that was for Greek readers, ending where Greek history
ended. Romans could still find new historical themes in the turbulent history
of the principate. As we noted above, under normal circumstances the
contemporary concerns of the Indi scaenici were no doubt confined to loyal
praise of the princcps and his family. But when there was a dynastic crisis, the
house of Caesar could provide plots as dramatic as the house of Atreus. It is
easy to imagine a changed atmosphere in the theatre after the disgrace ofjulia
in 2 BC, the death of Gennanicus in AD 19, the destruction ofSejanus in Ala 31,
the death of Tiberius m AD 37, the assassination of Gaius in AD 41, the fall of
Messallina in AD 48. the execution of Agrippina in AD 59, and so on. All the
playwright needed was a stage villain safely dead or disgraced, and the full
melodrama of the imperial court could be staged with impunity.
That, I think, is the background we need if we are to understand the
Octavia. We can, if we want, interpret it merely as a literary text, to be
explained by its relation to other literary texts; but it seems to me more fruitful
to read it as a document of the Roman stage, composed for performance at one
of the most dramatic moments of Roman history.

31 Pliny Nat. Hist. 7.159; Suetonius Nero 54.

32 Lucian Saltatia 72; see above, p. 197.
33 Problem: Smith 2003.403-4 (quotation from 404). Hypothetical reconstruction: Wiseman
2004a.265-72.
34 Lucian Saltatia 37.
 CHAPTER THIRTEEN

Ovid and the Stage

scaena sonat Indiqne uocant: spectate, Quirites.

Ovid Fasti 4.187

I
The fourth book of the Fasti contains Ovid’s account of the Indi Megalenses, the
games of the Great Mother. The din of her procession is deafening, but he has
lots of questions to ask. The Mother deputes to the Muses the job of explana­
tion, and Erato tells him what he needs to know. There are ten questions and
ten answers, of which much the longest (102 lines out of the total 194) is the
story of how the Phrygian goddess came to Rome. A major part of that story
(Fasti 4.297-328) concerns her reception at Ostia and the miracle of the
grounded ship.1
The tale of Claudia Quinta’s undeserved reputation, and her vindication by
the goddess as she pulled the ship from the sandbank single-handed, was
known to Propertius and probably to Livy (though he didn’t use it), but
evidently not to Cicero or Diodorus.2 Erato reveals that it was the plot of a
play: mira, sed et scaena testificata loqnar (326). What sort of play was it? The stan­
dard catalogues of Ribbeck and Klotz include it among the fabnlae praetextae;3
on the other hand, since it must have been particularly relevant to the Indi
Megalenses, it is worth noting that the title ‘Megalensia’ was used by two of the
known authors offabnlae togatae*
According to the grammarians of late antiquity, praetextae were tragedies—
or at least, serious plays—and togatae were comedies.5 However, Seneca

1 On which see Wiseman 1979.94—9 and Bremmer 1987.105-11.

2 Propertius 4.11.51-2, Livy 29.14.12; Cicero De hanispiaun response 27, Pro Girfto 34,
Diodorus Siculus 34.33.2.
3 Ribbeck 1897.335; Klotz 1953.371.
4 Nonius 829L (Afranius), Servius on Eclogues 7.33 (Atta).
5 Donatus De comoedia 6.1, Diomedes in Craminatici Latini 1.482 and 489 Keil, Lydus De
magistralibus 1.40.
 OVID AND THE STAGE 211

describes togatae as 'between comedy and tragedy’,6 and Varro seems to refer to
a play on a somewhat similar historical theme (an aitioti for the Nonae
Caprotinae festival) as a togata praetextci.7 Certainly the attractive young noble­
woman of Ovid’s scenario, with her ‘ready tongue for the strict old men’,8
sounds like a character in a comedy, however pious and patriotic its outcome;
and a crowd of men hauling on a rope has comic potential too, as in Aristo­
phanes’ Peace and Aeschylus’ satyr-play Diktyonlkoi.9
A detailed discussion of‘Ovid at the htdi Megalenses' came to this conclusion
about the Claudia passage:10

It would seem that dramatic entertainment at Roman festivals provided, as did

the medieval mystery plays, a representation of the cult story appropriate to the
festal day. [. . . ] At all events we may conclude that Ovid deliberately accentu­
ated the flavour of his descriptions of Roman holidays if not by the inclusion of
appropriate material from the festal stage at any rate by dramatic colour and
stage effects.

For this passage, at least, the conditional in the last sentence is unnecessarily
cautious: Ovid evidently is including ‘material from the festal stage’. To make
sense ot it, however, requires a hard look at two inter-related questions. First,
what evidence is there for the sort of performances the Romans watched at the
Indi scaenici in the Inst century BC? And second, what can we infer from Ovid’s
texts about the kind ot drama he used as a source for his stories?

II
For most aspects ot the social and cultural history of the late Republic, the
works ot Cicero otter an unparalleled wealth of information. But not for the
stage, and there may be a reason for that. The audience at the games was the
populiis Romanns-," Cicero’s relationship with the Roman People was funda­
mentally changed by the execution of the conspirators on 5 December 63 BC;

6 Seneca Epistles 8.8: iptam multi poetoe dicimt quae pliilosophis ant dicta sunt aut dicenda! non
adtingam tragicos, nee legates nostras: habent enini line quoque aliquid seneritatis et sunt inter comoe-
dias ac tragoedias mediae. Ct. Diomedes in Granimatici Latini 1.489—90 Keil, classing
praetextatae as a type of togatoe. See Brink 1971.319-20.
7 Varro De lingua latina 6.18, on which see Drossart 1974 and Wiseman 1998.8-11.
8 Fasti 4.309—10: cultus et ornaris mine prodisse capillis | obfuit, ad rigidos promptaqiie lingua senes.
9 Fasti 4.297—304, with Fantham 1998.154: 'these lines sketch a kind ot comedy.’ Emphasis
on funis: 4.297, 325, 331, 333 (also Propertius 4.11.51); cf. Aristophanes Peace 458-519,
Aeschylus fr. 46a. 16-21 Radt.
10 Littlewood 1981.387.
11 Cicero Pro Sestio 106, 116—18, De hamspiemn response 22—5, In Pisoneni 65, Ad Atticuni
2.19.3, 14.3.2, Philippics 1.36.
 212 UNWRITTEN ROME

and the greater part of his oeuvre, including almost all the surviving correspon­
dence, was written after that date.
As a young outsider challenging the entrenched nobilitas, Cicero could
speak on the People’s behalf—as in the prosecution of Verres, the debate on
Pompey’s eastern command and the defence of the popularis tribune C.
Cornelius.12 He was still able to exploit that popular goodwill in the first weeks
of his consulship, when he presented himself as a consul popularis in order to
defeat Rullus’ land bill.13 Up to that point, we may be sure that Cicero was
greeted with wami applause whenever he took his seat in the theatre at the ludi
scaenici. But the execution of citizens without trial must have changed all that.
Cicero himself evidently presented it as a conscious act of aristokratiafr and
from then on it seems he was regularly hissed at the games, except when he
could shelter behind the popularity of Pompey.15 Not surprisingly, though he
still went to the games when politics required it, Cicero in his late years was
not an enthusiastic theatre-goer. With that in mind, let us see what the
Ciceronian evidence has to offer on the subject of the Roman stage.
In 80 BC (he was twenty-six) Cicero defended Sex. Roscius on a charge of
parricide. In the course of a long purple passage on the heinousness of the
crime, he appealed to what the jury had often seen on the stage, Orestes or
Alcmaeon pursued by Furies for the murder of his mother:16

nolite enini putare, quern ad inodiini infabulis saepeiniincro uidetis, cos qui aliquid impie
scelerateqtic cominiserint agitari et pcrtcrreri Furiarunt taedis ardentibus. sna qitenique
fraus ct suns terror maxima uexat. ..
Do not imagine that those who have committed some impious or criminal act
are driven in terror by the blazing torches of the Furies, as you often see in
plays. What hounds them is above all their own crime and their own terror.
The phrase taedis ardentibus is used by Cicero on two other occasions where he
is making the same point, and otherwise only in a quotation from tragedy.

12 E.g. Cicero Perrines 134-7, Perrines II 5.174-6; Pro lege Manilia 63-4, 69-71; Asconius 71-
79C. The pro Comelio speeches are reconstructed by Kumaniecki 1970.
13 Cicero De lege agraria 2.6-10.
14 Plutarch Cicero 22.1, cf. Cicero Ad Attieum 2.3.4 (dpiOTOKpciriKUig). Plutarch’ss source at
this point was probably Cicero’s Jtepl VJtaretag, the Greek monograph on 1his -----consulship:
Pelling 1985.315.
15 Cicero Ad Attieum 1.16.11, q.v. for Cicero’s view of the Roman People at this time (also Ad
Attieum 2.1.8, 2.16.1).
16 Cicero Pro Roseio Amerino 67, cf 66 {uidetisne quos nobis poetae tradidemnt...) and 46 for
fabula as ‘play’ (a comedy by Caecilius).
17 Cicero In Pisonem 46 (nr in scaena uidetis). De legibns 1.40 {sieut in fabulis), Aeademiea 2.89
(tragedy); cf. De haruspicum response 39 {in tragoediis) on the Furies and madness. The topos
dates back at least as far as Aeschines {In Timarelmm 190): |.u) yap OlfO0E .. . TOlig
rjoeflrpcoTcig, KuOctntp sv ring rpaytpbiaig, Iloivdg eXavveiv Kai KoXa^Etv bqotv
i)|tgevaig.
 OVID AND THE STAGE 213

Ennius wrote an Alcineo and a Eiimenides, and Pacuvius evidently had the
Furies on stage lying in wait for Orestes at Delphi.18 Such plays were no doubt
what Polybius had in mind when he drew attention to the Romans’ use of
tragedy for controlling the populace;19 fear of punishment by the gods was
always a powerful deterrent from wrongdoing, and one medium of exemplary
moral education was evidently the tragic stage.2" Cicero’s rationalisation of it is
merely the recognition that educated people will not take the imagery literally.
It was probably some time in the late seventies that Cicero defended the
great comic actor Q. Roscius in a private suit brought by a business partner.
How absurd to imagine that a man of Roscius’ stature would stoop to fraud for
a mere fifty thousand sesterces!21

pro deiini liomiiiiimqiie Jidem! qui HS CCCI333 CCCI333 CCCI333 quaestus

facere noluit (mini eerie HS CCCI333 CCC1333 CCCI333 merere et potuit et
debuit, si potest Dionysia HS CCI33 CCI33 merere), is per sniiiiiuim fraudem et
malitiam et perfidiani HS 13 appetiit?
1 ask you! A man who declined to make a profit of 300,000 sesterces used utter
deceit, wickedness and treachery to make 50,000? For he certainly could and
should have earned 300,000 if Dionysia can earn 200,000.

Who was Dionysia? The jury evidently knew without being told; they must
have seen her often on the stage. Aulus Gellius, quoting a bon mot of Q. Hort­
ensius, calls her a saltatricula, which must mean a mime actress; and a fragment
ot Varro’s Menippean Satires, probably dateable to just about the time of the
Roscius case, reveals the sort of performance she may have starred in:22

18 Ennius frr. 16-31, 14-1-8 Jocelyn (= 19-32, 132-6 R’), cf. Accius zl/cwiro, frr. 608-20
Dangel (= 58-70 |<’). Pacuvius: Servius on Aeueiil 4.473 (fr. LUI R')
19 Pol. 6.56.8: tjTi rooovrov yup EKTtTptiYQi'iU’tm Kai JtapEtofjKTai toOto to pfpog [sc.
deioi6ai|ioviu| nap’ avrotc; Etc; te too; kut’ idiav ploug Kai ta Koiva tf]q ndXEwg
t'boTE pi] KaraXintiv VJtEp|3oXl]V. Fading a community of philosophers, the people have to
be controlled (5t>. 11): keijurat rot; ddqkoi; (fofiotg Kai xfj TOtaut]] I(H4YHlbi<# ta
ItkljOl] OWE/riV. See Mazzarino 1966.2.61—2 and Zorzetti 1980.64—5.
20 Anstotle A/etiip/iysics 12.8.20 (1074b4—5), Diodorus Siculus 1.2.2 (on f| T(i>V EV qbou
puOoXoyia), Cicero In Catilinam 4.7-8, De letiibus 2.15-16; cf. Cicero De legibus 1.47 on the
stage as a source of o/ibiioni’s. See Liebeschuetz 1979.39—54 for the ethical demands of
Roman religion, and Rawson 1991.570-81 for theatrical moralising (‘the theatre ...
provided much ot the mental furniture of the poor’, 581).
21 Cicero Pro Roiiio coinoedo 23 (Roscius was a rich man, and did not accept a fee for his
performances).
22 Varro Menippean Satires 513 Astbury (Nonius 563L); cf. Lucian Saltatio 41 for Actaeon as a
subject for tragic mime. [Atalanta was evidently another mythological mime subject: see CIL
6.37965.21 (the second- or third-century AD epitaph for Allia Potestas): quid crura? Atalantes
status illi councils ipse, translated by Horsfall 1985.256 as 'What about her legs? She had quite
the pose ot Atalanta on the comic stage’. The reference at that date is probably to mime, and
we know that already in Ovid’s time (Amores 3.2.29—30, Hrs ainatona 3.775) Atalanta s legs
were famous for more than just running (Horsfall 1985.263).| Hortensius: Aulus Gellius
1.5.3 (against L. Torquatus in 62 DC).
 214 UNWRITTEN ROME

crede mihi, plures dominos semi comedernnt qnam canes, quod si Actaeon occupasset et
ipse prius snos canes comedisset, non nngas saltatoribns in theatro fieret.
I tell you, more masters have been gobbled up by their slaves than by their
dogs. But if Actaeon had got in first and eaten his dogs himself, he wouldn’t be
rubbish for dancers in the theatre.

Presumably the saltatores (masculine) played the dogs who tore Actaeon to
pieces; naked Diana and her nymphs would be roles for saltatriailae.
In his defence of Q. Gallius in 64 BC,23 Cicero reminded the jury of what
he and they were seeing at the games—the huge success enjoyed by one
‘dominant’ playwright with his ‘Banquets of Poets and Philosophers’,
presenting Euripides arguing with Menander, Epicurus with Socrates.24 It is
not at all clear what dramatic genre was involved. On the one hand, the
subject matter might suit the more ethically-improving end of the mime­
writers’ repertoire, known to us mainly from the collection of sententiae
attributed to Publilius Syrus.25 On the other hand, the banquets of philoso­
phers had been satirised in Lycophron’s third-century satyr-play Menedenios,
and both Vitruvius and Horace clearly imply that satyr-play was a familiar part
of the Roman theatrical experience in the first century BC; indeed, Q. Cicero
evidently produced Sophocles’ satyr-play ‘The Banqueters’ to amuse Caesar s
officers in Gaul.26
Coming now to Cicero’s career after the consulship, we find him reporting
political allusions at the Indi Apollinares of 59 and 58 BC (the latter at second
hand, since he was in exile at the time); the plays concerned are tragedies, a
praetexta, and a comic togata.27 In the pro Caelio of April 56, delivered at the
very time of the Indi Megalenses, he quotes from tragedy and comedy, and
alludes to tragedy, praetexta and mime.28 At Pompey’s lavish games in 55, he

23 For the date, see Ramsey 1980.

24 Fr. 4 Puccioni = J 52.8: unus quidam poeta dominatur, homo perlitteratns,
Jerome Ad Ncpotianuiii 22.2:
cuius sunt ilia conuiuia poetanmi acphilosophariim.... See Giancotti 1967.119—28 and Wiseman
1994.80, both of whom wrongly date the speech to 66 BC.
25 Conveniently available in the Loeb Minor Latin Poets (ed. J.W. Dull and A.M. Duff. 1934)
3—111; cf. Rawson 1991.579f. Giancotti (see previous note) assumes that the play Cicero
refers to was a mime, and tentatively suggests D. Laberius as the author.
26 Athenaeus 2.55d, 10.419e—420c (Lycophron); Vitruvius 5.6.9, 7.5.2, Florace Ars poetica
220-50; Cicero ad Q. fratrein 2.16.3 (July 54 BC), SwdtTJtVOUc; SotpoDfroVC, quamquam a
te actam fabellam uidco esse festiue, mdlo modo probaui (for the vulgar content of the play, cf.
Athenaeus 1.17d). See Wiseman 1994.68-85.
27 Cicero Ad Atticum 2.19.3 (unknown tragedy); Pro Sestio 118 (Afranius’ Simulans), 120—2
(Accius’ Eurysaces), 123 (Accius’ Brutus).
28 Cicero Pro Caelio 1 (diebus festis ludisque publicis), 18 (Ennius’ Medea exsuf), 36 (unknown
comedy), 37 (comedy by Caecilius), 38 (Terence’s Adelphi), 64—5 (tabulae, mime), 67 (tragic
Eqmis Troiauus?): see Hollis 1998. The prosopopoeia of Ap. Claudius Caecus at Pro Caelio 34
is presented as a stage scenario (itagrauem personam induxi, 35), necessarily a praetexta.
 OVID AND THE STAGE 215

found the mimes soporific and the tragedies disappointing; also on the
programme, though he gives no details, were ‘Greek and Oscan shows’.29
In 54 BC, at a time when his political volte-face may have brought him
temporarily back into popular favour, he was at Apollo’s games again,
evidently in support of the praetor responsible for them:30

redii Romain Fontei causa a.d. vii Id. Quint, itcni spectatum, primum magno et aeqtia-
bi/i plausu—sed hoc ne curaris, ego ineptus qui scripserim. deinde Amiphonti operain. is
erat ante maim missus qitam productus. ne ditttius pendeas, palmam tidit. sed nihil tain
pusilhtm, nihil tarn sine tioce, nihil tarn. . . uerum haec tn tecum habcto. in Andromacha
tamen maiorfuit quam Astyanax, in ceteris parent habuit neminem. quaeris nunc de
Arbuscula. ualde placuit. Indi magnijici etgrati; uenatio in aliiid tempos dilata.
I returned to Rome for Fonteius’ benefit on 9 July, and went to the theatre. To
begin with, the applause was loud and steady as I entered—but never mind
that, I am a fool to mention it. To proceed, 1 saw Antipho, who had been given
his freedom before they put him on the stage. Not to keep you too long in
suspense, he won the prize; but never have I seen such a weedy little object,
not a scrap of voice, not a—but don’t say I said so! As Andromache at any rate
he stood head and shoulders above Astyanax. In the other roles he didn’t have
his equal. Now you'll want to know about Arbuscula. First-rate. The games
were fine and much appreciated. The hunt was put off to another time.

Antipho was evidently a tragic actor. Arbuscula must have been a inima,
presumably also a novice and an ex-slave. Twenty years later Horace refers to
her as saying ‘It’s enough if the knights applaud me’ when the rest of the audi­
ence booed her off the stage; in 54 she may have been like the young
freedwoman Epicharis, starring in ‘the games of the nobles’ at fourteen.31
A letter to Trebatius the following year reveals another aspect of the mimic
stage—topical satire, of which Trebatius may be a victim if he stays in Gaul too
long without getting rich.32 The mimes used to come first at the games, in the
morning programme; a letter to Paetus in 46 implies that a mime now follows
a tragedy, in the slot previously occupied by a fabula Atellaiia.iy Whenever they
came, Cicero found them as tedious at Caesar’s games that autumn as he had

29 Cicero .4<lJamiliares 7.2.1-3; for Greek shows, cf. Ail Atticmn 16.5.1, Plutarch Marins 2.1 and
ZtLRP 803.13 = CIL I2.1214.13.
30 Cicero .4<i Atticmn 4.15.6 (Shackleton Bailey’s translation); cf. 4.5.1 and AdJamiliares 1.9.4—
18 on the ’palinode’.
31 Horace Satires 1.10.76-7; ILLR.P 803 = CIL P.1214, with Wiseman 1985.30-35.
32 Cicero AdJtmiliares 7.11.2: si diutiusjrustra ajueris, non modo Laberium sed etiam sodatem nostnmi
Falerium pertimesco; mira enim persona induci potest Britannia iuris consult!. For Valerius Catullus
the mimographer, see n. 41 below.
33 Cicero AdJamiliares 9.16.7 (contrast 7.2.1).
 216 UNWRITTEN ROME

done at Pompey’s nine years earlier.34 What excited him were the serious
genres of tragedy and prcietextci that could be taken as commenting on the high
issues of contemporary politics, as at the ludi Apollittares in July 44 BC.3’
However, even the mimes’ ad-libbing had been worth hearing at that year’s
Indi Megctleiises, immediately following the Ides of March.36
The ubiquity of mime, in its many forms, is the main thing that emerges
from the Ciceronian evidence. Small wonder that in Lucretius the man who
has been watching the games all day sees dancers in his dreams, and not the
masks of comedy or tragedy.37 Despite recent doubts on the subject, it is more
likely than not that this popular and versatile dramatic form influenced, and
even overlapped with, the literary genres of‘high culture’.3B
The star mima of the forties BC, mistress of Antony and Cornelius Gallus,37
sang (and danced?) in a performance of Virgil’s sixth Eclogue. So at least Servius
tells us; Donatus too believed that the Eclogues were performed on the stage,
and it is not easy to see why these late sources should have invented the idea.4"
It has been argued that Catullus the poet, Catullus the mime-writer and
Catullus the theorist of mime were one and the same person—a startling
notion, but not at all inconsistent with what we can infer about the literary
culture of the mid-first century BC.41 The same applies to the Augustan age,
and particularly to the elegiac poets, since ‘contemporary mime is precisely the
sort of literary production which we should expect to find exploited in
elegy’.42 With that in mind, we may turn to Ovid, whose own works, he tells
us, were often ‘danced’ on the stage.43
4

34 Cicero AdJaniiliares 12.18.2 on Laberi et Ptiblilii poemata (cf. 7.2.1); Pro Rabirio Postitmo35 for
a dismissive comment about mimes in 54 BC.
35 Cicero Philippics 1.36, Ad Attiaim 16.2.3 (Accius’ Tereus); cf. Ad Attiaim 16.5.1 (Brutus had
expected Accius’ Brutus).
36 Cicero Ad Attiaim 14.3.2 (9 April 44): til si quid Jtpay|taTLK6v liabes rescribe; sin minus, populi
EJttorinaotav et mimonmi dicta perscribito.
37 Lucretius 4.973-83, cf. 788-93; the mask is exploited in a different context (4.296-9).
38 See McKeown 1979 and Fantham 1989; the latter calls mime ‘the missing link in Roman
literary history’. Contra Rawson 1993, who insists on ‘the vulgarity of the Roman mime ,
but with so many proper reservations that the essentials of the McKeown-Fantham view
remain valid. Evidently mime could be both vulgar and sophisticated, morally sententious
and obscene.
39 Volumnia Cytheris: Cicero Ad familiares 9.26.2; Ad Atticuni 10.10.5, 10.16.5, Philippics 2.20,
2.58, 2.62, 2.69, 2.77, Plutarch Antony 9.4-5, Pliny Nat. Hist. 8.55 (Antony); Servius on
Eclogues 10.1, De uiris illustribus 82.2 (Gallus).
40 Servius on Eclogues 6.11, Donatus Vita Vergilii 26 OCT; see Van Sickle 1986.17—23.
41 Wiseman 1985.183-98, 258-9, and 1994.92-4. For the theorist, see Scholia Bemensia on
Lucan 1.544: in libro Catulli qui inscribitur JTEpi pipoXoyubv (Muller’s reading for quis cribitnr
permimologiaruin).
42 McKeown 1979.71; among the mime-influenced passages he goes on to discuss are Horace
Satires 1.2.127-34, Propertius 2.29, 4.7, 4.8 and 4.9.
43 Tristia 2.519-20: et mea sunt populo saltata poemata saepe.
 OVID AND THE STAGE 217

III
For the following fifteen episodes in the Metamorphoses and Fasti, various
scholars in the past thirty years or so have suggested direct influence from the
Roman theatre:
1. Pentheus and Acoites (Met. 3.562-83, 692-700): from Pacuvius’
tragedy Pentheus?44
2. The Calydonian boar-hunt (Met. 8.273-413): from a mime by
Laberius?4’
3. Circe, Picus and Canens (Met. 14.320-434): from a satyr-play?46
4. Pomona and Vertumnus (Met. 14.622-94, 765-71): from a satyr­
play?47
5. Priapus and Loris (Fasti 1.391-440): from a mime?48
6. Faunus, Hercules and Omphale (Fasti 2.303-56): from a satyr-play?4'*
7. Jupiter, Lara and Mercury' (Fasti 2.583-616): from a mime?’"
8. Ariadne and Liber (Fasti 3.459—516): from a mime?’1
9. Anna, Aeneas and Lavinia (Fasti 3.543-656): from a mime?52
10. Anna Perenna and Mars (Fasti 3.675-96): from a mime by Laberius?53
11. Silenus and the bees (Fasti 3.738-60): from ‘burlesque drama’?’4
12. Priapus and Vesta (Fasti 6.319—18): from a mime or satyr-play?”
13. Ino, Hercules and Carmentis (Fasti 6.501-50): from a mime or satyr­
play?’"
14. Fortuna and Servius (Fasti 6.573—80): from a mime?57
15. Servius, Tarquin and Tullia (Fasti 6.585-624): from a praeiexta?’8

44 Otis 1966.371—2: Pacuvius ap. Servius on Aeneid 4.469.

45 Horsfall 1979b: Laberius fr. 176 Bonaria = 148 IL3.
46 Wiseman 1994.83: cf. Steffen 1952.134 on Aeschylus’ satyr-play Circe (frr. 113a—115 Radt).
See further p. 22(1 below.
47 Brief reference at Wiseman 1994.82, 84.
48 McKeown 1979.76; cf. also Littlewood 198(1.317 and Wiseman 1998.23-4.
49 Satyr-play: Littlewood 1975.1063-7: Steffen 1952.230-4, 241-2 on Omphale satyr-plays by
Ion and Achaios. ‘Popular drama’: Fantham 1983.196-8, 200-1.
50 Wiseman 1998.24.
51 Barchiesi 1994.230 = 1997.243: ‘una scena di teatro leggero’. The putative drama must
post-date Catullus 64.
52 Giancotti 1967.63—5; cf. also McKeown 1979.75-6 and Barchiesi 1994.232 = 1997.245.
The putative drama must post-date Virgil Aeneid 4.
53 Giancotti 1967.61-3: Laberius fr. 10 Bonaria = 2 R’; cf. also McKeown 1979.76, Little­
wood 1980.316-17, Wiseman 1998.72-4.
54 Brief reference at Littlewood 1980.317. See further pp. 223-4 below.
55 Mime: McKeown 1979.76; cf. Littlewood 1980.317. Satyr-play: Wiseman 1994.82.
56 Wiseman 1998.48-51.
57 Barchiesi 1994.217 = 1997.229; Wiseman 1998.27-30.
58 Wiseman 1998.30—4: Fasti 6.612, mira quidem, sed ramen acta.
 218 UNWRITTEN ROME

There is more to be said about one or two of these, and further examples may
be added to the list. But first, it may be helpful to expose certain prejudices that
have hampered enquiry hitherto.
First, the idea that ‘mythological burlesque . . . seems not to have been a
common feature in mime’.59 That goes against the explicit evidence of Varro,
who refers to mimes about Liber and the Nymphs, and of Augustine on
Priapus, the protagonist of nos. 5 and 12 in the list above:60
niiinquid Priapo niinii, non etiain sacerdotes enormia pudenda fecernnt? an aliter stat
adorandns in locis sacris, qttain procedit ridicnlns in theatris?
It is not only the mimes who give Priapus an enormous phallus; the priests do
the same. He stands there in his sacred places to claim men’s adoration in just
the same guise as he comes on the stage to provoke laughter.
The context is Augustine’s challenge to Varro’s distinction between the gods
of the poets and the gods of the city, theologia fabttlaris and theologia cittilis.6'
Since the whole lengthy passage is devoted to Varro’s argument, it is likely that
Priapus too featured in the mimes of Varro’s time, and not just Augustine’s.62
Second, it is said that ‘satyr play [was] a genre which despite Horace s
encouragement had probably lapsed for good in Augustan Rome’.63 Why
should we suppose so? Horace, like Vitruvius, provides the evidence that in
one form or another it was still a living genre, as it probably had been since at
least the fourth century BC.64
Coupled with this is a third unfounded prejudice, that ’the theory of
decorum’ prevented any common ground existing between the world of the
satyrs and that of contemporary Rome: ‘no common language or socially
acceptable container can exist for a mixture of themes and styles which
amounts to a breakdown of the conventional hierarchies.’6’ But it that were
the case one would hardly expect dignified senatorial families to claim descent
(as they did) from Pan, Silenus and Marsyas,66 or Suetonius to include in his

59 McKeown 1979.75; contra Horsfall 1979b.331, ‘both mimes and Atellan farces on mytho-
logical themes are attested’. [See n. 22 above.)
60 Varro Antiquitates diuinae fr. 3 Cardauns; Augustine City of God 6.7 (CSEL 40.284), trans. H.
Bettenson (Penguin Classics 1967).
61 Augustine City of God 6.5 = Varro Antiquitates diuinae frr. 7-10 Cardauns.
62 City of God 6.2—10 passim (Varro Antiquitates diuinae frr. 2a—12, 47, 62 Cardauns). Agahd (fr.
39d) included Priapus in his Varro fragment; Cardauns (fr. 35) is more cautious.
63 Fantham 1983.187; cf. Barchiesi 1994.231-2 = 1997.244, ‘un genere letterario di cui
sappiamo poco, o, sopratutto, non possiamo richiamare con certezza la presenza a Roma: il
dramma satiresco’. (7 he phrase ’genere letterario’ begs a big question.)
64 Horace and Vitruvius: n. 26 above. Fourth century BC: for satyrs and the Bacchic thiasos on
bronze cistae, sometimes apparently in the context of performance, see above, pp. 86—124.
65 Barchiesi 1994.234 = 1997.247.
66 See the coin-types of the Vibii Pansae (90 and 48 Be), the lunii Silani (91 BC) and the Marcii
(82 bc): Crawford 1974.336-7, 346, 377, 464-5, 467.

I
 OVID AND THE STAGE 219

narrative of Caesar’s crossing of the Rubicon (as he does) a supernatural figure

piping to shepherds like Pan.67
What we do not know is the extent to which satyr-play and mime could be
combined on the Roman stage. Mutual influence of the two genres can be
detected as early as the fourth century BC in Athens,68 and is likely to have
increased in the innovative world of Hellenistic theatre. At Rome, the jesters
in the processions at funerals, triumphs and festivals could be described as
either mimes or satyristai; the satyr-dance called sikinnis was one of the
elements of mime as performed by Bathyllus and Pylades; and Horace’s purist
polemic in the Ars poetica clearly implies that in his time the ancient rustic form
of satyr-play had given way to a more sophisticated urban type, with obscene
dialogue like that attributed to mime.69 The interdependence of the genres is
illustrated two generations after Ovid by Petronius’ Satyrica, a novel full of
references and allusions to mime, with a plot that depends on the anger of
Priapus.70
Satyr-play naturally had a chorus of satyrs; mime, or one type of it at least,
had a chorus of dancing girls71—and once we are free of the notion that iniinue
never played mythological burlesque, we can cast them as nymphs.'2 Nymphs
and satyrs had danced and made love together since the beginning of Greek
iconography, and their interplay of sexual desire was still celebrated in the an
of Rome.73 It would be very’ surprising if they were kept strictly separate in
Roman popular drama. In the Ovidian episodes listed above (nos. 3, 4, 5 and
12), we find them both symmetrically separate—naiads and dryads in love with
Picus: satyrs, Pans, Silenus and Priapus in love with Pomona—and also
together, pursued and pursuing in the two stories of the 'feast of the gods’.74
The natural inference is that the same corps de ballet of male and female dancers
could put on different scenarios in different styles, as the producers of each set
of games requited.

67 Suetonius Diuus lulius 32; Wiseman 1998.60-2.

68 See Simon 1982.19-20 and Plate 8, and Beazley 1963.1519 no. 13: Dionysus being enter­
tained by, respectively, a satyr called Mimos and a minia dressed as a satyr.
69 Dionysius ot Halicarnassus 7.72.10—12, Suetonius Diuus lespasiauus 19.2; Aristonicus ot
Alexandria ap. Athenaeus 1.20e; Horace ,4rs poetica 234-50, esp. 247 iiiuiiuuda crepeut igno-
miuiosaque dicta (cf. Ovid Tristia 2.497 on uiinios obscaena iocautcs).
70 Full details and bibliography in Panayotakis 1995.
71 Ct. Ovid Fasti 5.349—52 on the iiiimae at the ludi Florales. See Wiseman 1998.70-1.
72 See pp. 213—14 above on the Actaeon scenario attested by Varro.
73 See Hedreen 1994 on the sixth century BC, and Dierichs 1997.41-55 (who wrongly calls all
the female partners ‘maenads’). [See above, pp. 106—8.)
74 Ovid Metamorphoses 14.326-32, 637-41, Fasti 1.405-14, 6.323-4, 333-5. Just nymphs at
Fasti 2.589-98, just satyrs at 3.737-58 (nos. 7 and 11). Cf. also Fasti 2.307 (no. 6, Faunus’
farewell to the mountain nymphs), 3.467 (no. 8. Bacchus' girl captives) and 6.499—500 (no.
13, sea-nymphs bringing Ino and Melicertes): the dancers perform first, then leave the stage
as the play proper begins?
 220 UNWRITTEN ROME

IV
It is clear enough from the late-Republican evidence that performers like
Dionysia, Arbuscula and Cytheris were not just chorus girls but stars in their
own right,73 no doubt with enough pulling power to have shows created
specifically as vehicles for their particular talents. If that was the case, we may
have an example in Ovid’s treatment of Circe in Metamorphoses 14 (no. 3
above). The nymphs and Nereids are there, but in a subordinate capacity as
Circe’s maids;76 no one is allowed to upstage the ‘elemental sexuality’ of the
sorceress.77
Her first scene is with Glaucus, himself the subject of an Aeschylean satyr­
play.78 He wants Scylla, a girl who plays with the Nereids,79 but Circe wants
him. Spumed, she poisons Scylla’s bathing pool.80 Now here comes handsome
Picus, king of Latium, galloping after wild boar. All the nymphs pined for him,
but he loved just one, and married her—Canens, the singer, whose song
perhaps the audience have already heard.81 Circe lures him with her magic and
begs for his love. Spurned again, she turns him into a woodpecker. Enter his
companions, who threaten her with their spears; more sorcery', and they are
turned into monstrous beasts.82 ‘This magic and its results,’ writes a modern
critic,83 ‘project the disordered, irrational state of Circe’s whole being and
convey her passion’s blind, willful megalomania.’ What a part for an actress!
No doubt she watches from above in triumphant malice as the dancers mime
the search for their missing king, and the scene ends with the singing nymph s
despairing lament.84
One of the reasons for inferring a dramatic origin for this Ovidian story' is
the lavish use of dialogue.85 The same applies to the Lara, Ino and Tullia
episodes in the Fasti (nos. 7, 13 and 15 above), whereas the lecherous pursuits

75 See above, nn. 22, 31, 39.

76 Metamorphoses 14.264—7, cf. 311 (exploited by Ovid to make the Pious episode a story
within a story).
77 Segal 1968.442; Metamorphoses 14.25—7, cf. Virgil Aeneid 7.189 capta atpidine. For sorcery as
a mime subject, see Theocritus 2 (taken from a Sophron mime, according to the scholiast),
and Page 1941.328-31.
78 Steffen 1952.119-23: Aeschylus frr. 25c—35 Radt.
79 Metamorphoses 13.735-7, 898-903.
80 Metamorphoses 14.8—74. Scylla’s pool is like Thetis* (p. 221 below): 14.51—I, 11.229—37.
81 Metamorphoses 14.341, fcmiuea modulator canniiia note.
82 Metamorphoses 14.320-415. For variants of the Picus-Circe story, see Moorton 1988.
83 Segal 1968.438; Metamorphoses 14.320-415.
84 Metamorphoses 14.416—34. For a search in the final scene, but with a happier outcome, cf.
Fasti 3.649-56 (no. 9 in the list above).
85 Metamorphoses 14.12-39, 335-7, 372-85, 397-402. Another reason is Shakespeare’s
borrowing of the name Titania (14.438) and the transformation of man into beast (’Bottom,
thou art translated!’).
 OVID AND THE STAGE 221

of Priapus and Faunus (nos. 5, 6 and 12) are perfonned without words,86 in a
tiptoeing silence before the sudden uproar of the denouement. A combination
of both styles is suggested by an episode that I think has not yet been consid­
ered in this context, that of Peleus and Thetis.87 Here too we have an erotic
quest, but with a very different outcome.
First, a dialogue between the Nereid Thetis and the prophetic seagod
Proteus, another figure from Aeschylean satyr-play.88 He tells her to conceive:
her son will be mightier than his father, which is why Jupiter has stopped
pursuing her and granted permission to do so to the hero Peleus. The next
scene is the sea-cave where the naked nymph comes to sleep. Peleus finds her,
pleads with her to accept his love, and seizes her when she refuses. But Thetis
is a shape-shifter, and when she turns herself into a tigress even Peleus is
discouraged (like Circe’s transformations, this scene is a choreographer’s chal­
lenge and opportunity). Dialogue again, as Proteus tells Peleus to bring
bondage equipment next time. And so he does:89
promts erat Titan inclinatoque tenebat
Hesperium tentone /return, cum ptdchra relicto
Nereis inyro/ifiir consueta cubilia ponto.
iri.v bene uitpineos Peleus inuaserat artus;
ilia nouat formas, donee sua membra teneri
sensit et in partes diuersas braccltia tendi.
turn deinum itiqemuit ‘neque’ ait ‘sine numine iiincis'
exhibita esrque Thetis, confessam amplectitiir heros
et potitur noris inqentique implet Achille.
The sun was sinking, touching the western strait in his descending chariot,
when the beautiful Nereid left the sea and mounted her accustomed couch.
Unfairly, Peleus had taken control of her virgin limbs; she changed her shape,
until she realised that her legs were being held and her amis stretched out on
either side. Then at last she cried ‘You only win with divine help’—and Thetis
was displayed. She admitted defeat; the hero embraced her, took possession of
what he had prayed for, and filled her with huge Achilles.
Two details may allude to stage performance. First, the grotto itself, Thetis’
private bedroom: ‘hard to tell whether nature or art had made it—but no, it
was artificial.’9" And then the denouement, with the nymph on display to the

86 Priapus woos Loris with ‘nods and signs’, and she rejects him with a glance (Fasti 1.417-20).
87 Metamorphoses 11.221—65.
88 Steffen 1952.137-9: Aeschylus frr. 210-15 Radt.
89 Metamorphoses 11.257—65. For dancers performing metamorphosis, see Lucian Saltatio 19,
57.
90 Metamorphoses 11.235—6, natura foetus an arte | ambiguum, mqgis arte tauten.
 222 UNWRITTEN ROME

audience as well as to Peleus, in the position of total exposure used to portray

Andromeda and Hesione.91 She gives in, so Peleus can release her (as Perseus
and Hercules do in the other stories), no doubt for a closing pas de deux.92
As early as the fourth century BC, to judge by the visual evidence, mytho­
logical performances in which tniniae appeared naked were familiar in Latium;
the literary sources on the Indi Florales attest similar spectacles from the third
century BC to the first.93 The discovery of Ariadne could be an example. L.
Pomponius wrote a satyr-play Ariadne9'1 and Catullus’ dramatically ‘disobe­
dient ekphrasis’ in poem 64 has the distraught heroine lose her clothes before
delivering her great aria of passion and despair; meanwhile the god and his
satyrs are looking for her, and he is burning with desire.95
Ovid’s sequel to Catullus’ Ariadne (no. 8 above) is much more decorous,
ending with a confirmation of conjugal love. It is an aitioit not only for the
constellation of Ariadne’s crown but also for Ariadne as Libera, the mysterious
third occupant of the Roman temple of Ceres, Liber and Libera, allegedly
founded after the battle of Lake Regillus.96 Annual games were part of the cult,
the Liberalia on 17 March and the Cerealia on 19 April. Ovid's aition for the
Liberalia (no. 11 above) is also—like the Ariadne narrative—surprisingly
uncharacteristic of the licentious wine-god and his satyrs.
It begins with a long and deliberately anticlimactic priamel:97

tertia post Ictus lux est celeberrinta Baccho;

Bacche,fane nati, dttni ttia Jesta cano.
nee rejerani Setnelen, ad qnam nisifnlnnna secinn

91 Not, 1 think, ‘spreadeagled on the ground’, as suggested by Grinin 1997.140. Andromeda

[fig. 44, p. 123 above]: Schauenberg 1981.787-90; for illustrations see particularly 628 no.
20 (Campanian vase, fourth century BC), 640 no. 146a (temple relief at Falerii, second
century BC), 641 no. 152 (grave relief of performance at Roman games, second century All).
Hesione: Oakley 1997.628-9, illustrations at 386-9 nos. 18-41.
92 Hardly an imitation of the sex act itself, though in some contexts that may have been accept­
able on the mimic stage: Ovid Tristia 2.515, imitantes turpia iniinos; Valerius Maximus 2.6.7,
quorum argunienta inaiore ex parte stuproniin continent actus.
93 See above, pp. 109-24. Floralia: Valerius Maximus 2.10.8 (prisons nios ioconim), Seneca Epis­
tles 97.8; cf. Ovid Fasti 4.946 and 5.331—56, Martial 1 pref, and 1.35.8-9, I ertullian De
spectaculis 17, Arnobius Aduersus nationes. 3.23 and 7.33, Lactantius Institutiones. 1.20.6-10,
Augustine City of Cod 2.27, Historia Augusta Elagabalns 6.5, Ausonius 7.24.25f.
94 Porphyrio on Horace Ars poetica 221; Wiseman 1994.70.
95 Catullus 64.60-70, 251-3 (fc quaereus, Ariadna, tuoque incensus amore, 253); note the music-
and-dance scene at 64.254—65. See Laird 1993, esp. 20 (‘sound, movement and1
temporality’) and 29 (‘succession of time, movement, sound, and especially speech ) on the
qualities that make the ekphrasis ‘disobedient’: he relates them to rhetoric, but they seem
even more relevant to dramatic performance.
96 Ovid Fasti 3.459—516, esp. 469 (coniunx), 498 (coniugis), 512 (Libera). Temple: Dionysius of
Halicarnassus 6.10.1, 6.17.2-4, 6.94.3, on which see Wiseman 1998.35-6.
97 3.713-26: partus for panms at 716 is Heinsius’ emendation. For uitisator at 726 some MSS and
editors read uilis anus (cf. 765), an unnecessary simplification.
 OVID AND THE STAGE 223

luppiter adfcrret, partus biennis eras;

nee, puer nt posses mature tempore nasci,
expletum patrio corpore matris opus;
Sithonas et Scythicos longum narrare triumphos
et domitos gentes, turifer hide, tuas;
tu quoque Thebanae mala praeda tacebere matris,
inque tuum furiis acte, Lycurge, genus;
ecce libet subitos pistes Tyrrhenaque monstra
dicere, sed non est carminis huiiis opus.
canninis huius opus causes exponere quare
uitisator populos ad sua liba uocet.
The third dawn after the Ides is famed for Bacchus: Bacchus, favour a poet
while I sing of your festival. But I shall not tell of Semele, to whom Jupiter
came—if he had left his thunderbolts behind you would have been born
unarmed. Nor shall I tell how a mother’s function was fulfilled in your father’s
body, so that you could be bom a boy at the proper time. It would take too
long to narrare your Sithonian and Scythian triumphs and the conquest of the
peoples of incense-bearing India. You too [Pentheus], wrongful prey of your
Theban mother, will not be spoken of, nor you, Lycurgus, driven by the Furies
against your own kin. See, I would like to tell of sudden fish and Tyrrhenian
miracles, but th it is not the purpose of this song. The purpose of this song is to
explain why the Vine-planter calls the peoples to his own honey-cakes.

The Vine-planter is Liber, his own cakes are liba—and the song is of what
happened when the god discovered honey.
The stories Ovid is not going to tell here—though he tells them in the
Metamorphoses—were about the god’s miraculous origin and miraculous
powers, well known in Koine as subjects for tragedy. (Pacuvius wrote a
Pentheus, Naevius a Lycurgus, Accius a Tropaeuni Liberi which presumably dealt
with the god’s triumphant progress.) The story he does tell here could hardly
be further from tragedy, and it isn’t really satyr-play either, though the satyrs
feature in it:1'8

ibat harenoso saryris comitatus ab Hebro

(non habet ingratos fabula nostra iocos),
iamque erat ad Rhodopen Pangaeaque Jlorida nentum...
He was on his way from sandy Hebrus, attended by the satyrs (my story
contains no unwelcome frolics), and already they had come to Rhodope and
flowery' Pangaeus. . .

These are satyrs for a children’s party. They enjoy their honey, and they laugh

98 3.737-9.
 224 UNWRITTEN ROME

when old Silenus gets stung on his bald head, but their usual phallic antics—
hinted at in iocos"—are explicitly ruled out. Ovid at the Liberalia leaves out
sex, plays down the god’s awesome power, and draws attention to the fact that
he is doing so.
A reason why may be hinted at a few lines further on, in one of the expla­
nations for the granting of the toga uirilis at the Liberalia. In the old times when
the Romans were horny-handed fanners, when senators worked their fathers’
fields and the consul took up the fasces straight from the plough, the population
lived in the country and only came to town for the games; so perhaps that was
why the day was chosen for the ceremony, to get a crowd together to congrat­
ulate the new adult.100 A parenthesis interrupts the argument:101

sed dis, non studiis ille dabattir honor;

luce sna hidos uuae commcntor habebat,
quos cum taedifcra nunc habet ille dea.
But that honour [games] used to be given to the gods, not to popular enthu­
siasm; the discoverer of the gr ape used to have games on his own day, but now
he shares them with the torch-carrying goddess [Ceres].

The games of the Liberalia are attested for the time of Naevius, and it is likely
that their cancellation had something to do with the Roman government’s
violent suppression of the ‘Bacchanalia’ in 186 BC.102 That event was long
remembered; it is cited in Cicero’s treatise on the laws as a salutary example of
traditional moral standards.103 Augustus is unlikely to have disagreed, especially
in the light of Antony’s masquerade as Liber-Dionysus.104
The ancient temple of Ceres, Liber and Libera was burned do wn in 31 BC;
astonishingly, it was not included in the temple-restoration programme of 28,
but rededicated only in AD 17. Augustus evidently did without it for most of
his principate. That must surely have been a deliberate gesture to demote
Liber: since Ceres had also a joint cult with Tellus at the Carinae, the goddess
of the harvest could now be honoured without any unfortunate association

99 Barchiesi 1997.133, 240: ‘Ovid uses the word ioci and its cognates to refer to a whole comic
and sexual sphere in the Fasti. .. The use of terms like ioci, iocosns, obscenus, and alsofabula
seems to be reserved for burlesque tales, almost always of a sexual nature.’
100 Fasti 3.779-88.
101 3.784—6. For studio used of actors’ rowdy fans, cf. Tacitus Annals 1.16.3, Dialog"! 29.3; for
the phenomenon in general, see Valeius Maximus 2.4.1, Tacitus Annals 1.77, 4.14.3,
11.13.1, 13.24.1, Suetonius Tiberius 37.2, Nero 20.3, Pliny Epistles 7.24.7.
102 Naevius fr. 113 IV; Wiseman 1998.35-43.
103 Cicero De legibns 2.37, seueritas niaioruin in defence of miilieruin faina; cf. Varro Antiquitates
diuinae fr. 45 Cardauns. Livy’s great set-piece narrative at 39.8-19 must have been very
familiar to Ovid and his readers.
104 Seneca Suasoriae 1.6, Dio Cassius 48.39.2, Plutarch Antony 24.2-3, 26.3, 60.3, 75.3-4; see
above, pp. 130-1.
 OVID AND THE STAGE 225

with Antony’s patron god.105 In that context, it seems permissible to wonder

whether Ovid’s account of the Liberalia reflects the god’s reduced prestige.
Since the derivation of Liber from liberi is found already in Cicero,106 it may
not be absurd to suggest that the Liberalia festival was actually reorganised as a
show suitable for children. Whatever the explanation, Ovid seems to be illus­
trating a deliberately emasculated version of the Roman games in honour of
the god. The phallic comedies he does bring on (nos. 5, 6 and 12 above) are all
safely set in a Greek landscape.107

V
Discussion of Liber has brought us back to our starting point, Ovid’s use of
stage material about the gods and goddesses who were honoured at the Indi
scaem'ci themselves.108 My last example concerns the goddess Flora, whose Indi
Florales (28 April—2 May) were famous for just the sort of erotic entertainment
that Ovid’s narratives have allowed us to infer.109
Flora was closely associated with Liber (his mother, in one story), and her
temple was next to his at the Circus Maximus; it must have been burned down
at the same time, and like his was only rededicated in AD 17,110 Her games had
been controversial from the start, so here too we may guess that Augustus
chose to distance himself from so uninhibited a festival.111 Ovid, however,
gives himself a long and affectionate interview with Flora, enjoying her phys­
ical presence and emphasising the Liber-like liberty of her games.112
He begins by asking her to tell him in person who she is, ‘for the opinion
of men is fallacious’. That may refer to the story known from Christian
sources, that Flora was a successful prostitute who bequeathed her riches to the

105 Dio Cassius 50.10.3-4 (31 Be), Augutus Res gestae 20.4 (28 bc), Tacitus Annals 2.49.1 (ap
17); perhaps Augustus' hand was eventually forced by the great famine of AP 4-7 (Dio
Cassius 55.22.3. 26.1-3, 27.1, 31.3—1, 33.4). Ceres and Tellus: Arnobius Adnersns nationes.
7.32, Fasti Praenestini on 13 December (Inscriptiones Italiae 13.2.5371), Horace Cannen saecu-
lare 29-30; see I’alombi 1997.154-8 on frr. 577 and 672 of the Severan Forma Urbis,
showing twin temples in Tellnre. The goddess of plenty on the Ara Pads is indistinguishably
Ceres or Tellus: see most recently Spaeth 1996.125-51, esp. 133-5.
106 Cicero De natnra deonint 2.62, quod ex nobis nates liberos appellanins.
107 Fasti 1.393 (Gmeriu), 2.313 (Lydia), 6.327 (Ida).
108 Fasti 4.326, with Littlewood 1981.387.
109 See pp. 175—86 above.
110 Tacitus Annals 2.49.1, eodem in loco (n. 105 above); Ampelius 9.11, secundns Liber ex Mcrone
et Flora; Ovid Fasti 5.335—16 for Flora and Bacchus, in the context of her games.
111 Games not recognised by the Senate till 173 BC (Ovid Fasti 5.295—330); Augustus’ Palatine
Vesta installed on the Floralia (ibid. 4.943—54), on which see Barchiesi 1994.122 =
1997.133.
112 Fasti 4.946 (scaena ioci ntoreni liberioris habet), 5.331—2 (quare lasciuia tnaior | hisforet in Indis
liberiorqne iocns); 5.195, 199, 275, 376 (physical presence). Note that the narrator, uniquely
in the Fasti, is named as Naso the poet (5.377).
 226 UNWRITTEN ROME

Roman People, out of which games were set up in her honour."3 But she
knows better:114

sic ego. sic nostris respondit diua rogatis

(dinii loquitur nemos cjjlat ab ore rosas):
‘Chloris erain quae Flora uocor: corrupta Latino
nominis est nostri littera Graeco sono.
Chloris eram, nymphe tempiJelicis, ubi attdis
rem forttmatis ante fuisse niris.
quaefuerit mihifonna gratte est narrare modestae... ’
That was what I asked. This is what the goddess replied to my question (and
while she spoke, she breathed spring roses): ‘1 who am called Flora used to be
Chloris; a Greek letter in my name has been corrupted in Latin speech. I used
to be Chloris, a nymph of the happy field where once upon a time, as you have
heard, fortunate men had business. It’s hard to tell you modestly what my figure
was like...’

The nature of the business’ is one of Flora’s tactful reticences. But who were
the fortunate men who enjoyed it?
In Homer, the Elysian Field is a country club with a tight membership
policy, Menelaus only gets in as a son-in-law of Zeus. In Hesiod, you qualify if
you fought in the epic wars at Thebes or Troy. By Pindar’s time the criterion
is a moral one: whoever have thrice dared to keep their soul utterly free from
deeds of wrong. 115 Just as evil-doers are threatened with the punishments of
Tartarus, so this is the reward for a life of virtue.
Tartarus was a real threat for pious Romans; it is presupposed by Lucretius’
great poem, and references to it are found even in senatorial speeches."6 As we
saw with reference to the Ciceronian evidence about drama, tragedy put the
threat on the stage in the form of the Furies, to enforce moral behaviour in the
way Polybius had described."7 One might infer as a corollary that the rewards
of virtue could be advertised on the comic stage.
The paradise of Hesiod and Pindar was the Isles of the Blessed, but the
facilities there were the same as in Elysium—the river of Oceanus at the ends
of the earth, and the mild breath of Zephyrus, god of the west wind. Pindar

113 Fasti 5.191 (ipsa dace, quae sis: hominum sententia fallax); Lactantius Institutiones 1.20.5 10,
Minucius Felix Octavius 25.8, Cyprian De Militate idoluni 4. For a well-known prostitute
called Flora in the first century BC, see Plutarch Pompeius 2.2 1, Varro Menippean tins
136 Astbury, Philodemus 12 Gow-Page (Anthologia Palatine 5.132).
114 Fasti 5.193-9; for rem at 198, see Adams 1982.203.
115 Homer Odyssey 4.561-9, Hesiod IP.vks and Days 166-73, Pindar Olympians 2.68-77.
116 Lucretius 1.62-110, 3.37-56 etc; Cicero In Catilinam 4.7-8, Sallust Catiline 54.20, Histones
2.47.3M.
117 See above, pp. 212-13.

I
 OVID AND THE STAGE in

adds flowers and garlands; in Hesiod it is ‘sweet-smelling fruit’.118 In the

Aeneid, the abode of the blessed is a combination of grassy turf and shady
woods, including a grove scented with laurel.119 All these items appear in
Flora’s story. Chloris the nymph was ravished by Zephyrus, who gave her a
garden and made her the goddess of flowers; she was visited by Juno, on her
way to Oceanus to complain about Jupiter, and she gave her a herb to make
her pregnant. Outside Ovid, Martial refers to Flora fleeing into her laurel­
grove to escape the attentions of Priapus.120
At some time after Hesiod (whose text was interpolated to make it fit),
Kronos came to be thought of as the king of the blessed ones.121 The Latins
called Kronos Saturnus, and attributed to him the naming of their own land;
for it was there that he hid (latebat) when banished by Jupiter. He was
welcomed by primeval Janus, and his reign was the golden age.122 The Siitnniia
regna were either here but impossibly long ago or now but impossibly far away,
ideas symbolised respectively by "the castle of Saturnus’ at the foot of the
Capitol, supposed origin of the treasury of the Roman People, and by ‘the
tower of Kronos’, a feature of Pindar’s Isles of the Blessed.123 Here and now
was only possible at the god’s own festival, the Saturnalia.
The Roman cults of Saturnus and Flora had much in common. Both were
ancient,124
124 both were extended with new features in the third century BC, and
in both cases the innovation involved popular licentia, freedom from the
normal constraints of society'. In 241 or 238 the plebeian aediles built ‘rustic
Flora’ a new temple next to Liber’s, with luxurious Corinthian columns, and
honoured her with permissive games which the Senate refused to make annual
until forced to do so by the goddess’s anger.12’ In 217, with Hannibal already
in Italy and social cohesion an urgent necessity', the Senate not only allowed

118 Oceanus; Homer Odyssey 4.568, Hesiod Herts and Days 170, Pindar Olympians 2.71.
JlEtparu yaili;: Homer Odyssey 4.563, Hesiod ll'orfes and Days 168. Zephyrus: Homer
Odyssey 4.567, cf. Pindar Olympians 2.72 ciuptll neputveotoi. Flowers and fruit: Hesiod
ll’orks and Days 171, Pindar Olympians 2.72, 74.
119 Virgil Aeneid 6.6.38, 642, 656, 674 (grass); 639, 658, 673 (groves).
120 Ovid Fasti 5.200—12, 229—60 (for Flora and Zephyrus, cf. also Lucretius 5.738-40); Martial
10.92.10. [See Bremer 1975.268—74 for the erotic implication of flowery meadows, from
Archilochus to Petronius.)
121 West 1978.195, on Hesiod Works and Days 173a.
122 Virgil Aeneid 8.314-27, Festus 430L, Justin 43.1.3-5. Janus: Ovid Fasti 1.232-53, Origo
gentis Romanae 3.1—7, Macrobius Saturnalia 1.7.19-24. See Versnel 1993.89—135 (Kronos
and the Kronia’), 136—227 (’Saturnus and the Saturnalia’).
123 Satnrni castrnm: Festus 430L, Varro De lingua Latina 5.42, Origo gentis Romanae 3.6 (aerariuni).
Kpdvou TVpotC: Pindar Olympians 2.70.
124 Flora had her ow n Jlamen (Ennius Annales 2.117 Sk, ILS 5007); the Saturnalia were part of
the archaic calendar (Fasti Antiates, 17 Dec.).
125 Velleius 1.14.8, Pliny Nat. Hist. 18.286 (date); Tacitus Annals 2.49.1 (site), Vitruvius 1.2.5
(columns); Ovid Fasti 5.279-94 (aediles), 297-330 (Senate). Rnstica Flow. Martial 5.22.4
(her old temple at the foot of the Quirinal).
 228 UNWRITTEN ROME

but instructed the populace to celebrate Saturnus’ festival every year with
drunken revelry and gambling, wearing symbolic caps of liberty (pi7/e/).126
Ovid tactfully refrains from asking Flora herself about her games:127
qnaerere conabar qnare lascittia niaior
his foret in Indis liberiorqtte ioctts,
sed niihi sncairrit ninnen non esse setteruni
aptaque deliciis nmnera ferre deani.
tempora sntilibns cingnntnr pota coronis
et latet iniecta splendida ntensa rosa...
scaena lettis decet hanc: non est, niihi credite, non est
ilia coturnatas inter habenda deas.
turba quideni cur hos celebret meretricia Indos
non ex d[fficili causa petita subest:
non est de tetricis, non est de magna professis,
nolt sua plebeio sacra patere choro,
et inonet aetatis specie dinn Jloreat uti;
contenini spinani cum cecidere rosae.
I was trying to inquire why these games have more sexiness and freer fun; but
it occurred to me that the goddess’s nature is not strict, and the gifts she brings
are appropriate to pleasure. Drunken brows are wreathed with stitched
garlands, and thrown roses cover the polished table. [There follow ten lines on
what Flora and Liber have in common, including a cross-reference to . Iriadne's cn’iwi.J A
lightweight stage is proper for her; she isn’t, believe me she isn't, to be counted
among the goddesses in tragic boots. As for why a troupe of prostitutes
perfonns these games, the reason isn’t hard to find. She is not one ot the
frowners, not one of those who make big claims. She wants her festival open to
a plebeian chorus, and she warns us to use the beauty of youth while it s in
flower: the thorn is despised once the roses have fallen.
Note the deae cotnrnatae at line 348. Since goddesses wearing cotnrni are
goddesses on the tragic stage, we may infer that Flora is on the stage too, but in
the ‘barefoot’ genre of mime.128 Ovid’s emphatic ‘believe me. . . with his
earlier admission that he couldn’t keep his eyes off her,129 has a particular
impact if we imagine the theatre. The goddesses of tragedy are static, masked,

126 Livy 22.1.19-20; Seneca Epistles 18.3-4, Martial 11.6, 14.1; full details in Versnel
1993.146-50.
127 Ovid Fasti 5.331-6, 347—54. This part of the argument first appeared in JACT Review 24
(Autumn 1998) 6-8.
128 Mimes were excalccati (Seneca Epistles 8.8) or planipedes (Juvenal 8.191, Diomedes in Grain-
niatici Latini 1.490 Keil, etc).
129 Ovid Fasti 5.275-6: talia diccnteiii tacims inirabar; at ilia | 'ins tibi discendi, si qua requiris’ ait.

i
 OVID AND THE STAGE 229

and played by male actors; the goddesses of mime are showgirls who sing and
dance.130
So I suggest that here too Ovid was alluding to a stage scenario familiar to
his readers. It was a performance for the torchlit nights of the Indi Florales- the
scene was the Elysian Field, and the girls played the nymphs whose availability
was one of the rewards of the righteous; after the show, they would be avail­
able in real life.131 Chloris was the leading lady’s role (partes tuae, as Ovid says to
Flora at 5.184), in what was probably an episodic plot like that suggested above
for Circe.132 Enter Priapus, whose attentions she avoids; enter Juno, whose
pregnancy problem she solves (hoping Jupiter doesn’t find out); enter
Zephyrus, who successfully pursues her; and the play ends, as it should, with an
aition for the cult of the goddess in whose honour the games are held.133
Ovid calls Flora’s girls a chorus plebeius, and in telling the stoiy of how the
plebeian aediles set up her games he emphasises their significant name by
pointedly repeating the words populus, publicus and Piibliciits.13'* Flora was a
goddess of the People; it was the populus who in 55 BC were inhibited by
Cato’s presence from enjoying the ‘ancient custom’ of her games.135 That the
popular idea of Elysium was a garden of fleshly delights may be illustrated from
Ovid’s own text, in the passage on the Anna Perenna festival: those who have
spent the day drinking and making love in her riverside meadow and grove are
hailed on their return as ‘the blessed ones’.136 The purpose of such festivals was
to make the mythical real, for a day.
But there was no escape from the social hierarchy. Even the gods were
subject to it, as Ovid explains in his item on the birth of Maiestas; so too in the
Metamorphoses, Jupiter’s neighbourhood is carefully defined as ‘the Palatium of
the sky’, away from the dwellings of the plebeian gods.137 Nymphs, who dwell
on earth, are among the humblest of the immortals, but by pleasing the greater

130 Tragedy: Plautus Aniphitruo 41-3 (Virtus, Victoria, Bellona). Mime: Augustus City of God
7.26 = CSEL 40.340, Amobius 4.35 (Venus, Cybele); and cf. nn. 60-2 and 74 above.
131 Ovid Fasti 5.361—8 (deliais noctunta licentia nostris | couuenit, 3671), cf. Dio Cassius 58.19.1—
2; for iniuiae as uieretrices (Fasti 5.349), cf. Seneca Epistles 97.8, Lactantius Institutioiies
1.20.10, Amobius 7.33; for ‘after the show’, cf. Plautus Carina 1016-18, Truailciitus 965-6,
and Cicero De Jluibus 2.23 (Indus et quae sequuutur) with Wiseman 1985.44—5.
132 See p. 220 above.
133 Martial 10.92.10 (Priapus), Ovid Fasti 5.229-60 (Juno, 230 for Jupiter), 201-11
(Zephyrus).
134 Fasti 5.352, 283-94. At the time, the aediles probably spelled their name Populicius (cf.
ILLRP 35 = CIL P.28).
135 Seneca Epistles 97.8, Valerius Maximus 2.10.8 (quetti [Catuiieni] abeunteni ingenti plausu
populus proseattus priscutn uioreiu ioconun in scaeiiain reuocauit), cf. Martial 1 pref.
136 Ovid Fasti 3.523—40; Wiseman 1998.64-74. [See also Tibullus 1.3.57-64, and Mace
1996.239—10 on the erotic afterlife.]
137 Fasti 5.19-32, Metamorphoses 1.173-4. For ilia turba quasi plebeiorum dcorum, see Augustine
City of God 4.11, 6.1, 7.1, 7.2 (CSEL 40.178, 270, 302, 305), certainly from Varro.
 230 UNWRITTEN ROME

gods in their amorous moods they may earn promotion to goddess status—as a
slave-girl may become a libertitia.'3S Flora was just like the mima who played
her. And the grand and humbler gods were just like the audience in the
theatre, from the senators in the orchestra to the ordinary citizens in the cattea,
chewing their roasted nuts.139 The illusion of the theatre merged the two
worlds.
Ovid, I suggest, exploited that illusion in the Fasti. He could take advantage
of what all his readers had vividly in their minds, the doings of gods and men
made visible on the stage in the festivals of the Roman People.

VI
‘Peleus and Thetis’ and ‘Flora (The Musical)’ are offered as further items for
the list on p. 217. Naturally, all these suggestions about Ovid’s ‘sources’ (or
‘influences’, or ‘intertexts’) are hypothetical, and hypotheses are judged by
how well or badly they explain the phenomena.
What needs explanation is not Ovid but the stage. The history of Roman
drama cannot be understood just from the texts of Plautus, Terence and
Seneca. The dramatic festivals were so central a part of the community life of
Rome that allusions to what went on at them can be found throughout the
vast range of our literary sources; but such allusions are rarely explicit, since the
authors had no need to explain what their readers knew already. ()ur problem
is to imagine what they took for granted.
One such passage is Erato’s declaration at Fasti 4.326: ‘1 shall tell you
marvels, but marvels attested also by the stage.’ That at least is direct evidence
for Ovid’s use of theatrical material, and it encourages inference from other
Ovidian episodes as possible testimony for the history of Roman show business.

138 On earth: Ovid Metamorphoses 1.192—5 (under the protection ofjupiter—Augustus). Promo­
tion: e.g. Juturna (Virgil Aeneid 12.138-41), Pomona (Ovid Metamorphoses 14.635-42,
765-71), Carna (Ovid Fasti 6.101-28).
139 Lucretius 4.78-9, Horace Ars poetica 249.

i
 CHAPTER FOURTEEN

The Prehistory of Roman Historiography

i
The Greeks come before the Romans, in this book as in all accounts of the
ancient world. That priority is not a historical datum—on the contrary, the
city-states of Athens and Rome came into being at much the same time—but
in the discussion of ancient literature it is inevitable. Uniquely, and astonish­
ingly, in the sixth century BC Greeks used the medium of alphabetic writing
not just for lists or laws or epitaphs, but also to preserve the songs of epic bards
and lyric poets; and so they created a literature. Neither Latins nor (so far as we
know) Etruscans did that until three centuries later.
It is not that the Latins and Etruscans were backward, or peripheral.
Horace’s famous lines on the rude farmers of the Roman Republic, wholly
innocent of Greek culture until ’the peace that followed the Punic Wars’,1 are
demonstrable nonsense. Of course classicists want to believe what a classic
author tells them; but a few archaeological glimpses of the archaic world of
Latium and south Etruria may help us to overcome that mindset.
The earliest alphabetic Greek inscription known from anywhere was
scratched on a pot about 800 BC in an Iron-Age community in Latium, just 20
km east of Rome;2 Euboean potters can be detected working at Veii in the
eighth century BC;3 in the mid-seventh century, Kleiklos, ‘famed for fame’, is
attested on a Corinthian vase in the Esquiline cemetery at Rome,4 and
Aristonothos, ‘bastard noble’, painted the blinding of Polyphemos on a mixing
bowl made for an Etruscan magnate at Caere;’ other Greeks known as living
and working in Etruria at that time include Larth Telikles and Rutile
Hipukrates;*’ in the sixth century, lonians and Samians were frequenting the
trading post of Gravisca,7 at about the same time as a bronze plaque attests the

i Horace Epistles 2.1.156-63.

2 Ridgway 1996.
3 Ridgway 1992.131-7.
4 SEG 31.875.
5 Schweitzer 1955.
6 Ridgway 1988.664-5.
7 SEG 27.671, 32.940-1017.
 232 UNWRITTEN ROME

cult of the Dioskouroi at Lavinium,8 while at Rome an Attic black-figure

krater showing Hephaistos appears at the Volcanal,9 and a terracotta statue
group portrays the apotheosis of Herakles in the Forum Bovarium.10
These are just the most striking examples of an archaeological record which
led its most authoritative interpreter to the following conclusion:11 ‘The effect
of the refined Ionian civilisation on the cities of Tyrrhenian Italy is widespread
and deeply felt in the second half of the sixth century BC. We could even say
that there conies into being a genuine cultural and artistic koine consisting
equally of the Greek colonies and the Campanian, Latin and Etruscan centres.’
So it should be no surprise that Hesiod—or a sixth-century pseudo­
Hesiod makes Latinos, eponym of the Latins and ruler of ‘the famed
Etruscans , a son of Kirke and Odysseus.12 Kirke’s island was just oft the coast
ofLatium, directly across the water from the north coast of Sicily, where in the
early sixth century lived the greatest poet of the Greek West, Stesichoros of
Himera. Though he wrote in lyric metres, Stesichoros’ grandeur and ambitious
heroic narratives put him in almost the same category as Homer himself.13 A
late source, not necessarily untrustworthy, says that Stesichoros related Aeneas’
voyage to the West, and recent discoveries have revealed much of his Gery-
oneis, the narrative of Herakles’ tenth labour which ultimately lies behind the
Roman story of Hercules’ meeting with Evander.14
Pallantion, Evander’s Arkadian home, was mentioned in the Geryoneis, and
one of the versions of the poet’s own life says that he was born tn Pallantion
but went into exile—just like Evander.15 If that reflects the characteristic biog­
raphical method of borrowing episodes from an author's works as it they were
evidence for his life,16 perhaps we might even infer that the Roman story goes
back to Stesichoros himself. Of course we do not know enough to assert that,
but it is not inconceivable. Rome and early Greek poetry did not exist in sepa­
rate worlds.
The same is true of early Greek prose. Those Ionian dedications at Gravisca
were set up by people not unlike the pioneer logopoios Hekataios of Miletos,
whose ‘circuit of the earth’ naturally included western Italy—not only the
predictable islands (Elba, Capri etc) but also inland centres like Nola and
Capua in Campania, the latter of which he derived from ‘Capys the Trojan .'7

8 1LLR.P 1271a.
9 Coarelli 1983.176-7.
10 Cristofani 1990.115-18.
11 Pallottino 1981.44 (my translation).
12 Hesiod Thcogony 1011—16; West 1966.435-6.
13 Quintilian 10.1.62.
14 Horsfall 1979a; Page 1973.
15 Suda s.v. ‘Stesichoros’.
16 Harvey 2004.298-300.
17 Hekataios FGrH 1 F 59—63.
 THE PREHISTORY OF ROMAN HISTORIOGRAPHY 233

Rome would not have been beyond his scope. The early fifth century may be
when Promathion of Sanios wrote his Italika, which contained the earliest
version of the Romulus legend; and the unknown author of the ‘Kymaian
chronicle’, which probably dealt with the descendants of Demaratos of
Korinth who ruled in Rome, may also belong to that period.18 Towards the
end of the fifth century Hellanikos of Lesbos reported a view of Rome’s
origin—founded by Aeneas ‘with Odysseus’—which looks like a combination
of two separate traditions existing already. Hellanikos also seems to have
known some Latin, since his account of Herakles’ return with the cattle of
Geryon includes a derivation of‘Italia’ from uitulus.19
He may have regarded Latin as a Greek dialect.20 Certainly in the fourth
century BC Herakleides of Pontos called Rome ‘a Hellenic po/is’, and Aristotle
believed it had been founded by Achaians blown off course by storms in the
return from Troy.21 Theophrastos knew the Roman colony at Kirkaion,
where the inhabitants pointed out the tomb of Elpenor, and his circumstantial
report of a Roman attempt to found a city in Corsica shows that he was well
informed.22 At this point we have visual evidence again, with the engraved
bronze mirrors and cistae which attest the thorough familiarity of Latin and
Roman craftsmen with Greek artistic traditions; the inscribed names on some
of the scenes depicted provide vivid evidence of their creative exploitation of
the stories of Greek mythology.23
In the third century BC, Eratosthenes of Kyrene found it natural to include
Romulus, son ,'f Ascanius and grandson of Aeneas, in his chronological
researches, and it is not at all paradoxical that his contemporary Kallimachos
used the story of‘Gaius the Roman’ to illustrate the virtues of Punhellas.24

II
It is clear, then, that as tar back as our information extends—more than half a
millennium before Horace imagined the dawn of Hellenic consciousness in his
Roman peasants—Rome and her Latin and Etruscan neighbours were an inte-

18 Wiseman 1995.57-61 (Promathion); Alfbldi 1965.56-72 (chronicle); Zevi 1995

(Demaratos). [Gallia 2007 attacks the Alfoldi/Zevi hypothesis on a priori grounds; his argu­
ment depends on the very questionable assumptions that an early fifth-century Greek source
interested in the Tarquins is an impossibility (‘no such record could have existed’, p. 64), and
that the story ot Tarquin’s association with Aristodemus and death at his court was the
invention of a Roman annalist.)
19 Hellanikos FGrH 4 F 84, 111.
20 Gabba 2000.159-65.
21 Herakleides of Pontos fr. 102 Wehrli; Aristotle fr. 609 Rose.
22 Theophrastos History of Plants 5.8.1-3.
23 Wiseman 2004a.87-l 18. [See above, pp. 86-124.)
24 Eratosthenes FGrH 241 F 45; Kallimachos Aitia 4.106.
 234 UNWRITTEN ROME

gral part of the Greek world. What effect did that have on the Romans’
perception of themselves?
Whether or not the expulsion of the Tarquins was inspired by that of the
Peisistratids, the Romans must have been aware of the parallel. The early­
republican cult of Liber may well have been influenced by the Athenian
Dionysos Eleuthereus; much more certain is the influence of the ‘Solonian’
law-code on the Twelve Tables, visible even in the fragments that survive.2’
Moreover, there are structural elements in the ‘history’ of the early Republic,
as we have it in Livy and Dionysios, which strongly suggest the influence of
Athenian events: the attack ofPorsenna to restore Tarquin parallels that of the
Spartans to restore Hippias, the exile of Collatinus parallels the ostracism of
Hipparchos, the exile of Coriolanus parallels the ostracism of Themistokles,
and so on.26 The cults of the Spartan Dioskouroi in the Forum and of Arkadian
Pan at the Lupercal, and the ‘chapels of the Argives’ listed in a liturgical docu­
ment quoted by Varro,27 show that Athens was not the onl}' influence, but by
the end of the fourth century her imperial democracy may well have seemed a
particularly appropriate paradigm for the Roman republic’s domination of
central Italy. It was at the time of the Samnite wars that the statue of Alkibi-
ades, ‘the bravest of the Greeks’, was erected in the Coniitium.2’’
That choice implies familiarity with recent historiography—1 hucydides,
Xenophon, Ephoros of Kyme. And there is reason to suppose that some
Romans also read the ‘Atthidographers’, whose narratives of the long history
of their city’s political development conspicuously prefigured the later histori­
ography of Rome. A particularly revealing parallel is the description in
Kleidemos of Theseus’ battle with the Amazons in what was in historical times
the middle of Athens:29 the topographical details are used as ‘evidence in just
the same way as the Lacus Curtius and the luppiter Stator temple in the story
of Romulus’ battle with the Sabines in what was in historical times the middle
of Rome.
Kleidemos was probably an exegetes,30 using his cultic expertise to create the
material for his history. In Rome the annales of the pontijices, considered by
Cicero to be fundamental for the development of Roman historiography, may
have begun in 300 BC with the creation of the reformed college of pontijices,
now with plebeian parity of membership.31 It is usually thought that the aiinales
began much earlier than that; but the main evidence for that position has been

25 Crawford 1996.560-1.
26 Mastrocinque 1988.32-5.
27 Varro De lingua Latina 5.45-54.
28 Pliny Nat. Hist. 34.26.
29 Kleidemos FGrH 323 F 18.
30 Kleidemos FGrH 323 F 14.
31 Livy 10.9.2; cf. Cicero De oratore 2.51-3.
 THE PREHISTORY OF ROMAN HISTORIOGRAPHY 235

shown to be textually unreliable,32 and in any case it may be thought unlikely

that the unreformed patrician college had been interested in making its knowl­
edge public.
A similar point may be made about another potential source of historical
information, the list of annual magistrates. Such lists certainly existed after the
power-sharing compromise of 367 BC, which required the election of one
plebeian and one patrician consul each year; however, magistrate lists reaching
right back to the expulsion of the Tarquins were available to the historians of
the first century BC, and to whoever created the sequence of consuls and
triumphs inscribed on Augustus’ triumphal arch, and it is usually thought that
they are broadly reliable.33 But if the reform of 367 marked the final achieve­
ment of a polis constitution which required formal record-keeping, it may well
be doubted whether the patrician magistrates of the previous 140 years felt the
need for such an archive. We know that some later historians, faced with an
absence of authentic record for the years before the Gallic sack of the city in
387, were driven to the conclusion that the Gauls had burned the city, and all
the records had perished in the fire.34 In fact, as the archaeological evidence
shows, there was no fire. The suspicion must be that what passed for the ‘early­
republican’ magistrate list in Livy’s time was the result of antiquarian
scholarship of the same kind as that of the Atthidographers who reconstructed
the earliest Athenian archon-list.
It happens to be recorded that in 304 BC the aedile Cn. Flavius set up a
shrine of Concordia to mark the reconciliation of the ‘orders’, and that in the
inscription he dated it to 204 years after the dedication of the Capitoline
temple.3’ That may imply that the only way of counting the years before 367
BC was by the nails that were annually driven into the wall of the temple of
luppiter Optimus Maximus on the Capitol.36 That great monument to the
power and ambition of the archaic tyranny, conceived on the lines of the cult
of Zeus at Olympia, still dominated the Romans’ memory of their own past.37
Kleidemos was writing in the mid-fourth century BC. Whether or not his work
in particular was influential in Rome, it now seems clear that it was in the
generation after that—the pivotal period at the turn of the fourth and third
centuries—that the main outlines of "Roman history’ were formed.38

32 Humm 2000.106-9, on Cicero De republica 1.25; contra Cornell 1995.13-15.

33 Cornell 1995.218-21.
34 FGrH 840 F 3. [See above, pp. 13-14.]
35 Pliny Nat. Hist. 33.119-20.
36 Livy 7.3.5.
37. Purcell 2003.26-33.
38 Gabba 2000.16-19.
 236 UNWRITTEN ROME

III
How that came about is a question fraught with methodological difficulty.
What can we know about the communal memory of a pre-literary society,
when our evidence comes from much later literary texts whose authors
(remember Horace!) had little or no understanding of it? Any hypothesis can
only be tentative—but even so, there are hints in our literary sources that may
help us to imagine some aspects of the Rome of 300 BC.
The most famous such item, first exploited by the Dutch scholar Perizonius
in 1685, is Cicero’s reference to a passage in the elder Cato’s Origines about a
custom—obsolete in Cato’s own time—of guests at banquets singing songs to
the music of the pipe ‘in praise of famous men’.39 Modern scholarship is
divided on the value of this evidence, which on the one hand is certainly
consistent with what we know of the archaic syniposion, but on the other might
be just Cato’s exploitation of Greek antiquarian scholarship about the archaic
past.4" My own view is that it may well have been a genuine memory from the
pre-literary world, offering an insight into how the past was remembered in
one stratum of Roman society. One might imagine that banquets where the
friends and relatives of an Appius Claudius or a Quintus Fabius were gathered
were not unlike those implied by the poetry of the early Greek elegists, with
the values of an aristocratic elite being rehearsed and reinforced by the example
of admired ancestors.41
Aristocrats also had more public ways of making sure the deeds of their
ancestors were remembered. Cicero refers to the survival of early funeral
orations, Livy to the inscriptions attached to ancestral portraits, and the earliest
Scipionic epitaphs give us an idea of the sort of information that might be
transmitted.42 The fact that both Cicero and Livy thought it was unreliable
need not concern us: what matters is the creation of communal memory
through the pride of noble families.
Was that all there was? One influential modem view holds that the creation
of the Roman historical tradition was solely the work of the elite, as it the
Roman ruling class were identical with the Republic.43 But since the Latin for
‘Roman history’ is res (gestae) popttli Romani,44 that can hardly be true. The
deeds of noble leaders do indeed feature prominently in the later historical
tradition, but they do not monopolise it. Where should we look for the
communal memory of the People as a whole?

39 Cato fr. 118P; cf. Momigliano 1957.

40 Horsfall 1994.70-3; contra Zorzetti 1990.
41 Wiseman 1994.30-2.
42 Cicero Brutus 62; Livy 8.40.2; ILLR.P 309-10 [pp. 6-7 above].
43 Timpe 1988.283-5.
44 Sallust Cat. 4.2, Livy pref. 1.
 THE PREHISTORY OF ROMAN HISTORIOGRAPHY 237

It may be helpful to apply to the citizens of Rome something that Pausanias

wrote in the second century AD about the Athenians, who believed that their
democracy was set up by Theseus. As Pausanias rather snobbishly comments,
‘there are many false beliefs current among the mass of mankind, since they are
ignorant of historical science and consider trustworthy whatever they have
heard from childhood in choruses and tragedies.’45 The ‘choruses’ he refers to
were part of everyday experience, hymns in honour of the gods at the sacrifices
and rituals that were familiar to every adult and child in the poll's; and since
Plato says much the same thing about ‘stories repeated in prayers at sacrifices’,46
I think we are entitled to use Pausanias here as evidence for more than just his
own time and place.
There are three passages in Dionysios of Halikamassos’ history of early
Rome where the author alludes to hymns still sung in Rome in his own time.
At 1.31.2 he refers to Faunus, king of the Aborigines, ‘a man of prudence as
well as energy, whom the Romans in their sacrifices and songs honour as one
of the gods of their country’; at 1.79.11 he comments that the young Romulus
and Remus were looked on as offspring of the gods, ‘and as such they are still
celebrated by the Romans in the hymns of their country’; and at 8.62.3 he says
of Marcius Coriolanus ‘though nearly five hundred years have already elapsed
since his death down to the present time, his memory has not become extinct,
but he is still sung and hymned by all as a pious and just man’. A possible
context for the first of these may be the sacrifice to Faunus at the Tiber Island
on 13 February, for the second, either the Lupercalia two days later, scene of
the suckling of die twins, or the Parilia, anniversary of the foundation of the
city, on 21 April; and for the third, the sacrifice to Fortuna Muliebris at the
fourth milestone of the Via Latina on 6 July.47 It is not necessary to suppose
that the hymns Dionysios knew in the late first century BC were themselves
archaic compositions; the mere tact that such hymns were sung at Roman
sacrifices is enough to attest one more traditional source of communal
memory.
Tragedies, according to Pausanias, were the other main source of the ordi­
nary Athenian’s historical knowledge. Here too there is a Roman analogy, in
Plautus’ casual remark that gods in tragedies regularly reminded the audience
of the good things they have done for Rome. That suggests plays on Roman
historical themes, no doubt the fabulae praetextae discussed by historians of
Roman drama from Varro onwards.4!t Since the earliest praetexta our sources
refer to is Naevius’ Clastidiiini, on the single combat of the consul Marcellus

45 Pausanias 1.3.3.
46 Plato Laius 887d; cf. Buxton 1994.21-6.
47 Dionysios 8.55.3-5 gives the date and the reason.
48 Plautus Amphitnio 41—4; cf. Kragelund et al. 2002.
 238 UNWRITTEN ROME

with a Gallic king in 222 BC, modern scholars have usually inferred that the
genre was an invention of Naevius himself—and no doubt that is true as far as
literary drama is concerned. The question is, was there drama before there was
literature?
One revealing piece of evidence is Varro’s citation of a play on a quasi-
historical subject not by the author’s name, or even the title, but by the
context of its performance: ‘the People were taught the reason for this by the
togata praetexta presented at the Games of Apollo.’49 What matters here is the
assumption that performances at the theatre games ‘teach the People’. The
particular play Varro refers to can hardly predate 212 BC, when the Indi Apol-
linares were inaugurated, but the principle may well date back to before our
imagined 300 BC horizon, since the Indi Romani certainly, and the Indi plebeii,
Ceriales and Liberates probably, were already in existence at that time (see
above, pp. 168—74). Moreover, it is clear from the iconography of the bronze
mirrors and cistae mentioned above that Latins and Romans in the fourth
century BC were familiar with dramatic performance, in a Dionysiac context
which might even involve the plots of Euripidean tragedies (see above, pp.
109-24).
So it may well be that when the citizens of Rome gathered at the ‘games
in honour of their gods—luppiter Optimus Maximus at the Indi Romani and
plebeii, Ceres, Liber and Libera at the others—they were taught what they
needed to know about the gods, about their rights and duties, and about the
history of their city, by exemplary narratives presented in dramatic form. As
always, our inference can only be a provisional hypothesis. But it is at least
consistent with the little we know about plays presented in the later Rome of
our literary sources, which dealt with divine punishment of wrongdoers,’" the
power of the gods as revealed on earth,51 the miraculous ways in which the
Romans of the past overcame their enemies,’2 and the rewards of victory as
manifested in the triumphal celebration.53 It is, I think, reasonable to assume
that much of Rome’s communal memory consisted of what the citizen body
saw regularly performed before its eyes.
The very fact that later historians might suspect material in their sources of
having been invented for the stage is enough to confirm that dramatic
performance was one of the ways in which historical knowledge could be
created and perpetuated. There were of course other ways, which hardly need
argument: the instruction of the young by parents and teachers, for example,

49 Vano De lingua Latina 6.18.

50 Cicero In Pisonent 46.
51 Ovid Fasti 4.326.
52 Livy 5.21.9.
53 Horace Epistles 2.1.187-93.

J
 THE PREHISTORY OF ROMAN HISTORIOGRAPHY 239

or the commercial activity of professional story-tellers.54 But the contexts

suggested above, preserving the remembered past of the leading families or of
the citizen body as a whole, may be more important for our purposes, as
possible conduits for the historical tradition of the Roman People.

IV

In his preface to the Lays of Ancient Rome, Macaulay notes with regret that the
Romans—unlike Sir Walter Scott—never sought out and recorded examples
of oral poetry in order to prevent them trom being forgotten. But Macaulay
was wrong: it seems that at least one such collection was indeed made, ‘a book
of very ancient carniina, which was said to have been put together before
anything written in Latin’.” Only scraps from it survive, preserved in learned
authors from Varro to Macrobius, but some of them may be of interest for our
subject.
Here for instance is a quasi-Homeric moment quoted by Festus from the
uetera carniina: ’But now Aurora, withdrawing from the sky, reveals her
father.’56 That must be from a narrative poem, as perhaps was this line, quoted
by Varro in his De nira populi Romani: ’There the shepherds hold the Consualia
games with hides.” The reference is evidently to Consus’ altar in the valley
north of the Aventine; that was where the Sabine women were abducted at the
Consualia on 21 August in the first year of Rome, so the ‘shepherds’ may well
be Romulus’ men. Two other carniina were certainly narrative, entitled ‘Priam’
and ‘Nelens’.56 Priam’s sceptre was one of the divine talismans of Rome,39 and
Nelens was one of a pair of divinely begotten twins who were exposed,
rescued, and brought up in secret before freeing their mother from servitude;
the parallel with Romulus and Remus extends even to the cradle (skaphe)
which featured in the recognition scene.6'1
The most important feature of these tantalisingly enigmatic fragments is
their anonymity, which suggests that they did indeed originate in a world
before authors. The transition to literature proper is marked by the carmen belli
Pttnici, which is always attributed in our sources to its author Cn. Naevius. This
was a poem of which the written text was preserved—but even so, it was not
designed primarily for reading. Before C. Octavius Lampadio divided it into
seven books, it existed as ‘a continuous script’,61 which I think means that

54 Pliny Epistles 2.20.1, Dio Chrysostom 20.10.

55 Macrobius Saturnalia 5.20.18.
56 Festus 214L.
57 Nonius 3IL.
58 Varro De lingua Latina 7.28 (an invocation to the Casmenae); Festus 418L, 482L.
59 Servius auctus on Aeneid 7.188.
60 Aristotle Poetics 1454B, Dionysius of Halicarnassus 1.82-3.
61 Suetonius De grainin. 2.2.
 240 UNWRITTEN ROME

Naevius wrote it for recitation in the traditional way: the script was his
personal property, not copied until after his death.
Naevius carmen is an important milestone in the development of Roman
historical consciousness. No doubt bigger, better and more comprehensive
than any of its predecessors, it must have taken five or six hours to deliver. Its
audience of Roman citizens at the Indi Romani, perhaps—will have learned
about Aeneas flight from Troy, Jupiter’s prophecy of Roman greatness, the
origins of Carthage and its enmity with Rome, the foundation of Rome by
Aeneas grandson Romulus, and the course of the great Punic war in which
the poet himself had fought. Ennius was right about Naevius’ Saturnian metre,
in which of old the Fauns and prophets sang’.62 It was indeed a traditional
form but it was used at a high level of literary sophistication,63 and Naevius’
poem deserves to be thought of as the first true history of Rome.
There was another form ofliterary sophistication, however, which despised
the historical epic and its mass audience.64 Those who knew the Greek histo­
rians knew Thucydides’ pointed contrast between the performances of poets
and logographoi, competing for a particular audience’s favour, and his own
historical research, entrusted to a written text which would be a possession for
ever. The Hannibalic War was a time when men might well remember
Thucydides, and his insistence that the war he narrated was the greatest and
most terrible in history: no doubt that had been true two centuries earlier, but
the Romans knew it was true no longer. Even so, the first historiai ot Rome
were not (or not only) Thucydidean war narratives.
Fabius Pictor and Cincius Alimentus—and Postumius Albinus too, in the
next generation—were Roman senators who wrote their histories in Greek.
Despite authoritative opinion to the contrary,66 that must mean that at least
they hoped for a Greek readership. Since 1974 we have known that Fabius
Pictor s history began with ‘the coming of Herakles into Italy’;67 Plutarch tells
us that Fabius’ detailed and theatrical narrative of the overthrow of Amulius by
Romulus and Remus was taken from a Greek author, Diokles of Peparethos;68
Postumius Albinus is likely to be responsible for casting the victory of his
ancestor Aulus Postumius at Lake Regillus into the ‘purely Homeric’ narrative
followed by Livy and Dionysios.69 That looks like prima facie evidence that
these authors were writing for an international Greek-speaking audience.

62 Ennius Amiales fr. 207 Sk.

63 Goldberg 1995.73-82.
64 Kallimachos Anth. Pal. 12.43.1-4.
65 Thucydides 1.21.1, 22.4.
66 Gruen 1992.231.
67 SEG 26.1123.3a.
68 Plutarch Romulus 3.1, 8.7.
69 Wiseman 1998.86-7.
 THE PREHISTORY OF ROMAN HISTORIOGRAPHY 241

Part of their motive was no doubt to present Rome in a sympathetic light.

Fabius will have presented Herakles as the ancestor not only of himself but of
the great commander who defied Hannibal.'" One of the few surviving items
of Cincius’ history reassuringly presents the Roman ruling class as a strong
oligarchy trusted by the People: the Senate decrees the execution of Sp.
Maelius without trial, and entrusts the deed to Servilius Ahala, who takes a
sword and cuts Maelius down in the Forum; the citizens are indignant, but
when he shouts that he has killed a tyrant by order of the Senate, they accept
the sentence without demur.'1
One can hardly imagine Naevius, that scourge of the nobility,72 telling the
story to the Roman People in quite those tenns. And of course there were
other differences too between the Saturnian bard and the Hellenizing histo­
rians. ‘Naevius . . ., by choosing to write a poem rather than a history, did not
submit the Roman past to the process of rational elucidation in the interest of
truth which is characteristic of Greek historiography.’73 But there is no need to
privilege one style over the other. It was a combination of their methods (and
their prejudices) that defined the past of a city which in their day had already
seen a century and a half of military struggle and success, three centuries of
creative tension between the respective interests of the many and the few, and
more than half a millennium (pace Horace) of varied cultural evolution. It is
not surprising that the resulting tradition of Roman historiography was so rich
and complex.

Recommended reading
First, six chapters from three separate volumes of the second edition of The Cambridge
Ancient History. Dav id Ridgway, ’The Etruscans’, vol. 4 (1988), 634-75; R.M. Ogilvie
and A. Drummond, ‘The sources for early Roman history’, vol. 7.2 (1989), 1-29; M.
Torelli, ‘Archaic Rome between Latium and Etruria’, ibid. 30-51; A. Momigliano,
’The origins of Rome’, ibid. 52-112; A. Drummond, ‘Rome in the fifth century’,
ibid. 113—71 and 172—242; Elizabeth Rawson, ‘Roman tradition and the Greek
world’, vol. 8 (1989), 422-76.
Tim Cornell’s chapters in CAH2 7.2 are subsumed into his magisterial synthesis The
Beginnings of Rome (London, 1995), an essential work; much of interest will also be
found also in Stephen Oakley’s chapter ‘The Annalistic Tradition’, in his Commentary
on Liny Books 1-7-A', vol. 1 (Oxford, 1997), 21-109, and in the essays in Erich Green’s
two collections. Studies in Creek Culture and Roman Policy (Leiden, 1990) and Culture
and Identity in Republican Rome (Ithaca NY, 1992). Two valuable studies on Roman
religion are relevant to the theme: the chapter ‘Early Rome’ in Mary Beard, John

70 Plutarch Fabius 1.1.

71 Cincius Alimentus FCrH 810 F 4.
72 Aulus Gellius 3.3.15.
73 Momigliano 1990.91.
 242 UNWRITTEN ROME

North and Simon Price, Religions of Rome (Cambridge, 1998), vol. 1, 1—72, and Denis
Feeney’s short but stimulating Literature and Religion at Rome (Cambridge, 1998).
All those are predominantly background reading. For the central problem of how
to understand non-literary culture, three recent items may be recommended, not least
as examples of contrasting scholarly approaches: Nicholas Horsfall, The Culture of the
Roman Plebs (London, 2003); Nicholas Purcell, ‘Becoming Historical: the Roman
Case’, in David Braund and Christopher Gill (eds), Myth, History and Culture in Repub­
lican Rome (Exeter, 2003), 12—40; and chapters 2—7 of T.P. Wiseman, The Myths of
Rome (Exeter, 2004).

i
 CHAPTER. FIFTEEN

History, Poetry, and Annales

I Livy and Vano

Livy begins his preface with a conspicuous half-hexameter and an allusion to

Ennius, and ends it with the wish that he could exploit the prefatory conven­
tions of epic poetry.1 In the central passage, however, he emphasises the
differences between what he does and what the poets do:2

quae ante conditani condendaniue urbein poeticis ntagis decora fabulis quant incomiptis
rentni gestannn nionninentis tradita, ea nee adjinnare nec refellere in aninio est.
1 intend neither to endorse nor to refute those traditions of events before the
city was founded or planned which are more appropriate to the stories of poets
than to the uncorrupted records of history.

More appropriate in what way? Livy is evidently expressing a generic distinc­

tion between poetry and history,3 but on what criterion?
The temporal phrase ante ccniditani condendaniue urbein implies that
chronolog)' is what marks off poetic from historical material, and that is
perhaps what we should expect. Just a few years earlier, in 43 UC, Varro had
published his major work on history and chronology, the four books De gente
popttli Romani,'* in which he had carefully distinguished three periods of the
past (discrimina teniponun).3 The first, the (ibl]X.OV or obscure, ran from the
creation of mankind to the first flood (that of Ogyges, not Deucalion), and
could not be calculated in years. The second, about 1600 years from the flood
to the first Olympiad, he called HvSlKOV, because many of the supposed events
of that period were f'abitlosa—and it is clear from the fragments that he attrib-

1Livy pref. 1, 13. On §1, cf. Quintilian 9.4.74 (T. Liuius hexametri exordia coepit) and Ennius
/bimils. fr. 494 Skutsch; Moles 1993.141—2. On §13, see Moles 1993.156-8.
2 Livy pref. 6: this passage and pref. 13 are the only places in the whole of Livy's extant text
where the noun poeta or the adjective poeticns occurs.
3 Moles 1993.148: ‘decora is used of what is generically appropriate.'
4 Fragments in Peter 1906.10-24 and Fraccaro 1907.255-86; for the date, see Fraccaro
1907.77-8, and Horsfall 1972.124-5.
5 Fr. 3P (Censorinus 21.1—5); cf. Fraccaro 1907.255-7.
 244 UNWRITTEN ROME

uted the fibulae to poets.6 Only the third period, from the first Olympiad
onwards, could be called LOTOpiKOV, because the events that took place in it
were recorded in ‘true histories’.
Varro s choice of 776 bc as the dividing-line between the realms of poetic
fiction and historical truth corresponds quite closely with Livy’s dating, since it
just allows the beginning of the foundation story (condenda ttrbs) to fall within
the historical period. The dispute over the Alban kingship which resulted in
Amulius overthrow of Numitor and murder of his son, and the consecration
of Numitor’s daughter as a Vestal Virgin, took place three years before the
conception of Romulus and Remus, and thus, on Varro’s chronology, in 775
BC.7
However, it is not as simple as that. As Livy’s next sentence makes clear, the
particular poetic legend he has in mind is the conception story itself:8
datur haec ticma autiqiiitati ut tniscendo hiimana diirinis primordia ttrbiimi aitgitstiora
facial; et si cut popttlo licere oportet couseerare origities sttas et ad deos refine auctores, ca
belli gloria est popttlo Romano ut atm siiimi coitditorisque pareutem Mortem potissimimi
ferat, tarn et hoegeutes liumanae patiautur aequo ammo quam imperium patiuutur.
Antiquity is granted this licence, to make the origins of cities more august by
mingling the human with the divine; and if any people ought to be allowed to
sanctify its origins and claim gods as founders, such is the martial glory of the
Roman people that when it declares Mars in particular as its parent and the
father of its founder, the nations of the world may put up with that too, as
readily as they put up with our imperial rule.
So now it seems that what marks off the poet’s territory from rhe historian s is
the participation of gods in human affairs, the ‘mingling of human and divine .
And if the poets say that Mars really appeared on earth to lather the twins on
Rhea Silvia, they must mean that he did so in 772 BC, four years into Varro s
historical period.9
Of course it is absurd to be so pedantic. Both Varro and Livy were serious
authors dealing with a serious issue, and neither of them was content with a
merely mechanical solution.

6 Fr. 6P (Augustine City of God 18.8), on the birth of Minerva from the head ofjupiter: poetis
etfabidis, non historiae rebusque gestis cst adplicatidimi (the antithesis accepted as Varronian by
Fraccaro 1907.42). Fr. 13P = 23F (Augustine City of God 18.12), on Jupiter and Europa:
quod de Ione poetae cantant, as opposed to historica ueritas. Fr. 17P = 29F (Augustine City of
God 18.16), on the metamorphosis ofDiomedes’ companions, which Varro thought was not
fictitious: non Jabidoso poeticoque niendario scd historica adtestatione confimiant.
7 Conception on 24 June 772 (Varro in Plutarch Romulus 12.3—6), TETCtpiti) etel after Ilia s
consecration (Dionysius of Halicarnassus 1.77.1).
8 Livy pref. 7.
9 Above, n. 7: Plutarch citing the calculations of Varro’s friend Tarutius.

i
 HISTORY, POETRY, AND ANNALES 245

The way Livy dealt with the problem has been well explored by David
Levene.10 There were plenty of allegedly supernatural events in Roman
history, and Livy’s consistent technique in dealing with them is to distance
himself with an exculpatory formula (dicitur, utferunt, traditur memoriae, etc),
sometimes offering an alternative explanation on the merely human level;11 but
he will often report, as a fact, that contemporaries accepted the event as super­
natural, and by narrating the favourable outcome he allows his reader to infer
that they were right to do so.12 Sometimes he leaves out a famous divine-inter­
vention story altogether, but leaves a hint to remind his readers of it, as when
he refers to the caelestia anna of the Salii, or the dictator’s vow to the Dioscuri
at the battle of Lake Regillus, or the insigne nomen of Claudia Quinta at the
reception of the Magna Mater in 204.13
One particularly revealing instance of his technique is the narrative of the
disappearance of Romulus.14 After the sudden stonn that left the king’s throne
mysteriously empty, the Romans were prepared to believe the patres who had
been standing close to him when they said he had been snatched up by a whirl­
wind. On the initiative ofjust a few people (presumably the patres themselves),
the Romans now hailed Romulus as a god and the son of a god. But then Livy
goes on (1.16.4):

fttisse credo rum quo pie aliquot qui discerptuni regent patrinn manibiis taciti argnerent;
nianaiiit eiiini hate qttoque ted perobscura Jama; illant alteram admiratio niri et panor
praesens nobilitauit
1 believe that even then there were some who claimed in secret that the king
had been torn to pieces at the hands of the patres; for this tale too has spread,
very obscure though it is. Admiration of Romulus and the fear felt at the time
has made the former version famous (or noble].

What made the Romans believe in Romulus the god was the prudent judge­
ment (consilium) of one man, Julius Proculus, who reported to the people’s

10 Levene 1993, esp. 16—30. See also Feldherr 1998.64—78 and Forsythe 1999.87-98.
11 E.g. 1.4.7 (Acea Larentia as Ittpa), 1.31.4 (either divine voice or response of hantspices),
5.22.5 (either divine inspiration or young man’s joke). Note also the use of arte (e.g. 1.36.6,
2.7.2, 5.22.6) to mark the transition to what Livy is prepared to endorse.
12 E.g. the warning voice at 5.32.6-7: Livy commits himself only to the fact that M. Caedicius
reported to the tribunes that he had heard it, but acceptance of its genuineness is part of the
religio which enables Rome to survive the disaster (5.50.5, 5.52.11).
13 1.20.4, 2.20.12 (nihil nee diuinae nee hnmanae opis dictator praetennittens), 29.14.12; cf. respec­
tively Ovid Fasti 3.259—392, Dionysius of Halicamasus 6.13 and Ovid Fasti 4.291-344.
Cicero uses the same technique at De repnblica 2.13 (orantibus), 2.37.4 (scintilla ingenii), 2.45.3
(caede maetdatus): see Zetzel 1995.170, 192, 201. Otherwise unattested divine-intervention
stories may be inferred from Livy 1.12.7 on Romulus andjupiter Stator (tamquam caelesti ttoce
iussi) and 10.19.17 on Ap. Claudius and Bellona (uelut intrigante dea).
14 Livy 1.16.2-8; Levene 1993.132-3.
 246 UNWRITTEN ROME

assembly that Romulus had descended from heaven and told him to tell the
Romans of his deification.
Livy reports as fact Proculus’ report, and the belief it created,15 but his use
of the word consilium clearly implies that it was a piece of deliberate policy to
keep the people happy—a ‘noble lie’ (yewatOV IpevSog), as in Plato’s
Republic. His use of the ambiguous word nobilitauit may be an allusion to that
famous passage; so too Plutarch describes as ‘not ignoble’ (Oil tpavXov) the
tradition that Numa invented the story of his divine mistress and adviser
Egeria, for the same reason.16 Livy may not have been as much of a philoso­
pher as Plutarch, but he was interested in the subject, and wrote philosophical
dialogues.17 So an allusion to Plato is not out of the question.
As for Varro, who certainly was a philosopher (among many other things),
his complex argument in De gente populi Romani about myth, history and
divine intervention can be reconstructed from book 18 of Augustine’s De ciui-
tate Dei, which is largely based on it.18 Varro ended book 2 ot De gente with the
Trojan War, which divided his ‘mythical’ period—from Ogyges’ flood to the
first Olympiad—into two unequal parts, of which the latter was closer to the
‘historical’ category.19 But even in the earlier period, as Augustine remarks in a
passage which is agreed to be Varronian,20 there were true histories which
were turned into myth by the poets.
Augustine has just given a long list offabnlae Jictae, all of which involved
physical impossibilities: Triptolemus and his winged serpents: the Minotaur;
the Centaurs; three-headed Cerberus; Phrixus and Helle on the flying ram; the
Gorgon’s gaze; Bellerophon on Pegasus; Amphion and his miraculous lyre;
Daedalus and Icarus; Oedipus and the Sphinx; Antaeus the son ot of Earth. Then
he comments:21
hae fibulae helium ad usque Troianum, ubi secundum libritm .Maims I airo de populi
Romani gente fmiuit, ex occasione historiarum, quae res ueraciter gestas continent, ita
sunt ingeniis hominum Jictae lit non sunt opprobriis iiiimiiiuin adjtxae.
Down to the time of the Trojan War, which is where Marcus Varro ended the
second book of De gente populi Romani, these mythical stories were made up by
the ingenuity of men, taking the opportunity' offered by histories which contain
things that truthfully happened, in such a way as not to link them with slanders
on the divine powers.

15 Livy 1.16.5 (consilium, fides), 1.16.8 (tides, twice).

16 Livy 1.16.4; Plato Republic 3.41413-4151); Plutarch Numa 4.8 (also Lycurgus and his lawcode
‘from Apollo’).
17 Seneca Epistles 100.9.
18 See Fraccaro 1907.23-67 for an analysis of Augustine’s sources in book 18.
19 Fr. 3P (n. 5 above), fr. 14P = 28F (Augustine City of God 18.13).
20 Fr. 14P = 28F (Augustine City of God 18.13).
21 Augustine City of God 18.13.

i
 HISTORY, POETRY, AND ANNALES 247

That last phrase alludes to a Varronian obsession, the ‘tales unworthy of the
gods’ which made up the theologia fabttlaris of poets and playwrights,” and
Augustine goes on to some examples of such stories, such as Jupiter’s alleged
abduction of Ganymede (not true, it was king Tantalus who did it, just as it
was the Cretan king Xanthus who abducted Europa).23
But whether slanderous or not, these myths were made up by poets out of
true historical records. That is how Varro and Augustine saw it; we of course
see it the other way round, as the creation of ‘history’ by the rationalising of
myth. So Erichthonius was not conceived by the earth from Vulcan’s ejacula­
tion as he tried to rape Minerva: that was an opinio fabtdosa arising from the
‘fact’ that the infant was found in the temple of those two gods.24 And the
dispute between Minerva and Neptune over the city of Athens was settled not
by a vulgar display of miracle-working, as the myth has it, but by a ballot of the
Athenians themselves in which the women defeated the men by one vote; as
Augustine observes, for Varro this was non fabulosa sed historica ratio.25
So the chronological criterion for poetic myth or factual history was by no
means hard and fast. Nor, more surprisingly, was that of supernatural interven­
tion in human life.
It may well be that Varro had a strictly natural ‘historical’ version of all
those miraculous stories listed by Augustine. But he certainly did not rule out
the supernatural a priori. In book 3 of De gente he reported as historical fact the
metamorphosis of the companions of Dioinedes into birds, citing as support
the transformation of Ulysses’ men by Circe, and the werewolves of the Arca­
dian Lycaeus cull.2" Augustine insists on Varro’s judgement—historica adtestatio,
notfabulosuin poetirtiinqne ntendaeiinn—and it leads him into a long digression on
the powers of daeinoiies.21
That was a matter of some importance to Augustine, and he returns to it
more than once in De ciuitate Dei. In book 10, discussing the ‘miracles of the
gods of the gentiles’, which resulted from the power of demons, he cites as
examples the migration of the Penates from Alba to Lavinium, Attus Navius’

22 Varro Antiqnitates diuiii.ie trr. 7 and 10 Cardauns (Augustine City of God 6.5); see Wiseman
1998.17—24. Varro De petite popti/i Rotnoni fr. 8P = 18F (Augustine City of God 18.10):
Manns I um> non unit J.ibulosis adiierstis deosjidetn adhibcre figmenlis, tie de maicstatis eonim digni-
Mte indigniitn illiquid sentiat.
23 Augustine City of God 18.12 (Europa), 18.13 (Ganymede)—both passages with explicit
reference to the stage. Xanthus’ abduction of Europa counted as historica ueritas (cf. n. 6
above).
24 Fr. 13P = 24F (Augustine City of God 18.12).
25 Fr. 7P = 17-18F (Augustine City of God 18.9-10).
26 Fr. 17P = 29F (Augustine City of God 18.16-17). Contrast Pliny Nat. Hist. 10.127 (on the
Jabula of Diomedes’ companions), 8.81—2 (on the Arcadian story as Gniem credulitas).
27 Augustine City of God 18.18: he leaves open the possibility that Apuleius Golden .4ss was a
factual narrative. On daeinones, see Brenk 1986.
 248 UNWRITTEN ROME

cutting of the whetstone, the voyage of Aesculapius’ snake from Epidaurus to

Rome, Claudia Quinta pulling the ship from the mudbank, and the Vestal
who proved her innocence by carrying water in a sieve. When Tertullian had
dealt with the same question, he too cited Claudia Quinta and the Vestal with
the sieve, but also the epiphany of Castor and Pollux and the reddening of
Doniitius’ beard.28 Both writers knew Varro’s works well and used them
extensively, and I think it is likely that all these examples come from him.
One of them certainly did, as is proved by a neglected passage in De ciiiitate
Dei book 22:29

nani si ad eorum iniracula ueniamus, quae facta a dis suis obponunt martyribus nostris,
noime etiam ipsa pro nobisfacere et nobis reperientur omnino profcere? nani inter magna
iniracula deoruiu suorum profecto magnum Hind est quod Varro coinmemorat, V'estaleitt
uirginem, cum periclitareturfalsa suspicione de stupro, cribrum implcsse aqua de Tiberi et
ad sitos indices nulla eius parte stillante portasse. quis aquae pondus supra cribrum tenuit?
quis tot cauernis patentibus nihil inde in terrain cadere permisit? responsuri sunt: ‘aliquis
deus ant aliquis daemon.'
And now if we turn to examine the miracles of paganism, the achievements of
their gods which they oppose to those of our martyrs, we shall find that those
achievements support our side, and supply us with invaluable assistance. In fact,
among their greater miracles is the one recorded by Vano concerning a Vestal
Virgin who had come under suspicion of unchastity and was in danger of her
life. The story goes that she filled a sieve with water from the Tiber and carried
it to her judges without spilling a drop. Now who was it that kept mat weight
of water in the sieve? Who was it who prevented the water from pouring out
on to the ground from all those gaping holes? The reply will be, 'Some god, or
some demon.’
The Vestal’s name was Tuccia, and her trial before the pontifices took place
about 230 bc.3" The way Varro dealt with her story is not beyond detection.
At the beginning of the twenty-eighth book of the Natural History, for
which Varro was one of his named sources, the elder Pliny discusses the power
of words, in particular prayers and incantations. One of his prime examples is

28 Augustine City of God 10.16 (ilia ... iniracula deonnn gentiliuni, quae coinniendat historia);
Tertullian Apologetiais 22.12. For Doniitius, see Suetonius Nero 1.1 and Plutarch Aemilius
Paulins 25.3-4.
29 Augustine City of God 22.11, trans. H. Bettenson (Penguin Classics). The passage escaped
the notice of Cichorius 1922.20-1, Miinzer 1939 (see also n. 31 below), and Broughton
1951.227.
30 Livy Epitome 20, mentioned after the subjection of Sardinia and Corsica in 231. The MSS of
Pliny Nat. Hist. 28.12 give anno urbis DCV1III (145 bc), to be emended to DXX11II (230
BC): Miinzer 1897.177 n. 1 and 1937.206.
 HISTORY, POETRY, AND ANNALES 249

the deprecatio which enabled Tuccia to carry the water in the sieve.31 When
Valerius Maximus tells the Tuccia story, he quotes her rash prayer to Vesta as
the cause of the miracle.32 Dionysius of Halicarnassus, in a part of his Roman
Antiquities for which he certainly used Varro, inserts into his account of the
foundation of the Vesta cult a digression on the manifestations of the goddess:
two Vestals, Aemilia and Tuccia, are cited as having prayed to Vesta and been
empowered to perform miracles.33 Aemilia’s prayer is quoted verbatim;
Tuccia’s is only referred to, but is clearly an important part of the story.34
From this triangulation of the extant sources, the outline at least is visible of
the Varronian treatment of Tuccia’s story as evidence of the power of Vesta’s
Hinnen. Perhaps it was one of a series of examples of divine manifestation in
Roman history', as in Tertullian and Augustine.3'’ It was certainly a question as
important to Varro’s theology as it was to theirs.
Livy' had quite different priorities, either omitting miracle stories altogether
or distancing himself from them by the use of oratio obliqua.36 He wanted his
readers to draw appropriate moral conclusions from his narrative whether or
not they believed in supernatural manifestations. It was a controversial ques­
tion—and though he may have relegated such stories to the ‘appropriate to
poetic legends' category', there is no reason to suppose that all Romans would
agree with him.

II Cicero
The locus classics for the respective modes of poetry' and history is the conver­
sation that opens Cicero’s De Iegibus. It takes place by the river Lins, in the
grounds of the Cicero house at Arpinum. This ancient oak, says Atticus—it

31 Pliny N«if. Hist. 28.12. ,/u.i usu aqu.un in cribro tidit; cf. 28.10, polleautue illiquid uerlui et iiicauta-
meiita eminiiiiiiii. See Miinzer 1897.177—8 for Varro as the source, and 1937.205: ‘Es liegt
ohne Zweitel bier iiberall eine mehr antiquarisch als annalistische Uberlieferung zugrunde.’
See n. 29 above: Miinzer was evidently unaware of the Augustine passage which confirmed
his conjecture.
32 Valerius Maximus 8. Labs.5 [see now Mueller 2002.50-2]. For Valerius’ use of Varro, see
Bloomer 1992.113-26.
33 Dionysius of Halicarnassus 2.68-9 (EKtqdvEia at 68.1 and 69.3); Varro cited at 2.21.2,
2.47.4 and 2.48.4. For Aemilia (also Valerius Maximus 1.1.7, Prop. 4.11.53-4), see Miinzer
1920.173-7 and 1937.199-203 [and Mueller 2002.47-8]: the legendary version of a real
event in 178 UC (Livy Epitome 41, Obsequens 8).
34
' ’ Dionysius of Halicarnassus 2.68.4 (Aemilia), 2.69.2 (Tuccia, tf|V OeOV E3tlKClk<i)Oa|I£Vr|V
liytnova Tij; odov yEVEoOat).
35 See n. 28 above. The subject may have been treated in the last book of Varro’s Aiitiqiiitates
diuinae (on the di seleeti, including Vesta: frr. 229 and 281-3 Cardauns). or in Curio de ailtu
deonmi (cf. Augustine City of Cod 7.34—5), or just possibly in Gallus de admirandis (cf. Macro-
bius Saturnalia 3.15.8).
36 See p. 245 above. He evidendy reported Tuccia’s condemnation (Livy Epitome 20); if he
mentioned the miracle as an alternative tradition (as suggested by Miinzer 1937.207), it will
have been to reject it.
 250 UNWRITTEN ROME

must be the very one where Marius saw the eagle in Marcus’ epic poem. If you
like, says Quintus; but Marius’ oak was planted by a poet, and that’s the sort of
tree that never dies.37 So Atticus consults the author himself: was it his own
verses that planted the tree, or did Cicero have a factual source for what he
wrote?38
Marcus replies by asking Atticus two questions of his own. Did Romulus
really appear to Julius Proculus (on the Quirinal, close to Atticus’ house in
Rome), and did the north wind really carry off Oreithyia (at the Ilissus, close
to Atticus’ old house in Athens)?39 ‘Why do you ask?’, says Atticus.40
nihil sane, nisi tie nimis diligcnter inquiras in ea quae isto modo memoriae sint prodita.
‘No reason — except this, that you shouldn’t inquire too carefully into events
that are recorded in that manner.’
At first sight, it looks straightforward enough: if isto inodo means ‘like a poet,
then modus should mean something like ‘generic convention’. But this modus is
a means of memoriae prodere, which is the historian’s job. "
In any case, Atticus is still not satisfied:
atqui nmlta quacrimtur in Mario Jictane an nera sint, et a nommllis, quod et in recent!
mcmoria et in Arpinati homine uersere, ueritas a te postulatur.
‘But people ask about many things in the Marius, whether they are true or
invented; and since you’re dealing with recent events and a man ot Arpinum,
some people are demanding the truth from you.’
Evidently not everyone thought it was enough just to say ‘this is poetry, not
history’. But Marcus insists that it is:
et mehercule ego me cupio non mendacem putari; sed tamen nonnuili isli, hie noster,
faciimt imperite qiii in isto periculo non ut a poeta sed ut a teste ueritatem exigant; nee
dubito quin idem et cum Egeria conlocutum Niimam et ab aquila Tarqninio apicem
inpositum patent.
‘Well, I certainly don’t want to be thought a liar! But my dear Titus, those
“some people” of yours are naive in demanding the truth in such a case, not as
from a poet, but as from a witness. I’ve no doubt they think Numa really talked
with Egeria, and the eagle really put the cap on Tarquin’s head!’

37 Cicero De legibns 1.1—2 (1.14 for the Liris bank). Cicero’s Marins'. Courtney 1993.174 8.
38 Cicero De legibns 1.3: mine ucrsus hanc quercnm seuerint, an ita factum de Mario, ut scnbis,
accepcris.
39 Ibid.: certcn ... iieninine sit... sicenim est tradition. (Plato Phaedrus 229B for the Ilissus, surely
a deliberate allusion by Cicero.)
40 Cicero Dr legibns 1.4 (the following three quotations).
41 TLL 8.677.70-8: e.g. Cicero De diuinatione 1.55 (onmes historic!), Nepos Themistocles 10.5
(Thucydides). The phrase occurs sixteen times in Livy (memoriae tradere seven times, memoriae
mandate twice).

4
 1

HISTORY, POETRY, AND ANNALES 251

So, says Quintus, you think that different laws are to be observed in a history
and in a poem? Of course, replies Marcus, for in a history what matters is truth,
in a poem it’s mostly enjoyment.42
That ‘mostly’ (pleraqtte) is important, as is the concession Marcus immedi­
ately makes, that there are countless fabiilae in Herodotus and Theopompus.43
He knows that his schematic distinction between poetry and history’ is not
universally valid, and that people like Atticus, naive or not, are entitled to ask
whether a poem on a historical subject is to be believed. So if, as Denis Feeney
puts it, the question at issue in this conversation is ‘what you can narrate in a
particular genre’, ‘the appropriate treatment of the data of received tradition’,44
it is clear that more than one type of answer could legitimately be given.
That is spectacularly demonstrated in another dialogue of which Cicero and
his brother were the interlocutors (this time without Atticus), namely De
diuiuatioiie. Do the gods really communicate their will to men?45 In book 1
Quintus argues that they do: although its workings are mysterious, genuine
divination can and does exist, as is proved by the overwhelming testimony of
past authorities.46 In book 2 Marcus speaks for the sceptics: since divination is
inherently improbable and cannot be explained, the supposed evidence for it
must be either invented or explicable by other means.47
Although the negative argument comes second, Cicero as author is careful
to keep both possibilities open. The Academic philosopher’s purpose is ‘to
compare arguments, to set out what can be said for any proposition, and
without asserting his own authority to leave the audience’s judgement free and
unimpaired’.4b Who knows how many of the audience agreed with Quintus?
The distinction between history and fabtihi is central to the whole discus­
sion. Quintus insists on the authority of written records, which may be either
poetry or prose.4'1 One of his prime exhibits is the whetstone cut in half by
Attus Navins:5"

42 Cicero De legibus 1.5: Quippe, ruin in ill,i [historial omnia ad iieritatem, Quinte, referatur, in hoc
[poematel ad delectationem pleraqne.
43 Ibid.; cf. Strabo 1.2.35 (C43). For myths in history see Marincola 1997.117-27.
44 Feeney 1991.259, 260; the whole chapter on 'epic of history’ (250-312) is fundamental.
45 Cicero De diuinatione 1.1: magnified quaedam res el salutaris, si modo est idla, quaque proximo ad
deonun uiin natura mortalis possit accedere.
46 E.g. 1.12: nihil est iinreni quod non longinqiiitas temponim excipiente memoria prodendisque monil-
mentis efficere atque assequi possit.
47 E.g. 2.27: hoc ego philosophi non esse arbitror, testibus mi qui ant casu tieri am malitiafalsijictiue esse
possunt.
48 2.150: ... conjerre causas et quid in quamque senteiitiam dici possit expromere, nulla adhibita sua
auctoritate indicium audieiitiuni relinquere integrum ar liberum. See Beard 1986, Schofield 1986.
49 Scriptores: 1.31. Scribere, with named author: poetry 1.22, 87; prose 1.33, 36, 46, 48, 50, 53,
56, 121, 130. Scripruin est etc., anonymous: 1.31, 39, 52, 89, 101, 121. Historia: 1.37, 38, 49,
50, 55, 121. Annales: 1.43, 51, 100. Monumental 1.12, 33, 72, 87.
50 1.33; cf. Livy 1.36.4—5, who narrates it with utfenmt.. . fenmt... memorant....
 I

252 UNWRITTEN ROME

negenius omnia, comburamus annales, ficta haec dicamus, quiduis denique potins quoin
deos res Innnanas curarefateamur.
‘Let s deny it all, let’s bum the annals, let’s say it’s an invention—anything at all
rather than admit that the gods care about human affairs!’

Which is precisely the answer Marcus gives in book 2: philosophy shouldn’t

accept invented stories.51 Again, such stories may be in poetry or prose. After
all, says Marcus, why should I believe Herodotus any more than Ennius?’2
Literary genre seems not to be the main issue here.
However, there is a passage in book 1 which does address the status of
poetic stories. Quintus has been citing examples from Greek historians,
including Philistus’ report of what the mother of Dionysius of Syracuse
dreamed when she was pregnant.53 That reminds him of the dreams of Ilia and
Priam in Ennius, which he admits are invented fabulae; but he quotes them all
the same, because they are no different from historical examples like the dream
of Aeneas in Fabius Pictor.5'*
All three of those dreams (since Ennius’ Ilia was the daughter of Aeneas)
were far back in what Varro called the mythical period. But now Quintus takes
a historical example:55

sedpropiora uideamus. cninsnani modi est Superbi Tarqninii somniinn, de quo in Bruto
Acci loquitur ipse?
‘But let’s look at something more recent. Of what sort is Tarquimus Superbus
dream, which he talks about himself in Accius’ Brutus?'

What exactly is Quintus asking here? Cuitisiiain modi est? c.m hardly mean
‘What genre is it?’, because the genre is obvious: it’s a play, aJiihula praetexta. At
this point his concern is not primarily with poetry or prose,’" but with fact or
fiction in any genre. Accius’ play was on a historical subject, so the question
‘Did Tarquin really have that dream?’ is a perfectly reasonable one, even
though Marcus in book 2 doesn’t bother to answer it.’7

-a m,em A"i Niuii; nihil debet esse in philosophia commenticiis fabellis loci.
5- ..lb: Herodotmii cur ueraciorem ducam Eimio? num minus die potuit de Croeso quam de Pyrrho
f‘"X‘rc Ennius? Fmgere also at 2.27, 58, 136; fibula and/or coininenticiiis at 2.27, 8(1. 1 13.
3 n l '^' 0 | n W'ffdlaces; num etiam Graiorum historia mentila est? Chrysippus on the
e p ic oracle (1.37-8), Chrysippus and Philistus on dreams (1.39); the theme resumes after
t e igression with Heraclides Ponticus, Dinon, and one of the Alexander-historians (1.46-

o4 1.40-3 (fabulae at 40 and 43, ficta and commenticium at 42): Ennius Annales 1.34-50 Skutsch
and Scaemca aO-61 Jocelyn; Fabius Pictor FGrH 809 Fl
55 1.43: Accius Brutus 651-72 Dangel (= 17-38R).
56 He ends the digression at 1.46 with age mmc ad externa redeamus: Greek v. Latin is as impor­
tant as prose v. poetry.
57 Cf. De diuinatione 2.136 on niulta ... a te ex historiis prolata somnia, not citing the examples
rom Ennius
from nntus or Accius and dismissing those from the Greek historians (quis enim auctor
istonmi?).
 HISTORY, POETRY, AND ANNALES 253

What Marcus’ answer would have been, we can guess from the contemp­
tuous reference in De legibus to ‘events recorded isto modo’. For him, as no
doubt for Cicero the author, the only modus that counted as valid history’ was
one where the author guaranteed his report with evidence and argument; mere
narrative had no authority in itself.’8 But not everyone took that view. Cicero
is careful to let us see both sides of the argument, and to give them spokesmen
of equal status. Quintus speaks not only for the naive majority, the impend, but
also for thinking people, readers of philosophical dialogues, who chose not to
be bound by the scepticism of the Academy.

III Dionysius and Plutarch

Twenty years or so after Livy made his prefatory' comment about stories more
suitable to poetry, Dionysius of Halicarnassus faced the problem of narrating
the conception of Romulus.
He reported as fact the rape of Ilia in the grove sacred to Mars. Someone
did it5’—but who? Some say it was one of the girl’s suitors (a version hardly
consistent with Ilia being a Vestal); others that it was Amulius himself, dressed
in armour to terrify her (a version evidently designed to rationalise the story
that Mars was responsible).6" But those are the exceptions:
ol 6e nXetoTot itvOoXoyovot top baipovo; ei'bwXov, ov to xinpiov ijv,
jtoXXci kuI tiXXa Tip jtiiOei bctiudvia epya jipoacunovTEg lyXiou te
dtpaviopdv uiq vibiov icat t.dipov ev ovpavip KitTuo/ovTa.
Most authors relate the legend that it was an apparition of the divine power to
whom the place was sacred, adding to the incident many supernatural occur­
rences, including a sudden disappearance of the sun and darkness spread over
the sky.
The apparition was bigger and more beautiful than a man; they say that it
comforted the girl afterwards (’by which it became clear that it was a god’), and
then disappeared up to the sky wrapped in a cloud.61
Dionysius is aware that more is involved here than just a disagreement
among authorities. He suggests two possible approaches. One is to reject such
stories outright because they attribute to the gods the frailties of humanity; on

58 Cicero De legibus 1.4. For the rival modes of historic and uphegesis (inquiry and story-telling),
see Wiseman 1993a. 135—8.
59 Dionysius of Halicarnassus 1.77.1, with (hd^ETal TIC EV TO) TEgEVEl delayed for emphasis
to the end of the sentence.
60 Ibid.: TIVEC J1EV «.TO<( aiVOUOl ... oi dE ... (cf. Plutarch Romulus 4.2 for the second
version).
61 Dionysius of Halicarnassus 1.77.2; Origo gentis Rouuiuie 20.1 (from Fabius Pictor book 1).
 254 UNWRITTEN ROME

this view (which is that ofVarro), sexual desire is ‘unworthy of the incorrupt­
ible and blessed nature’ of the gods. The other approach is to suppose that the
substance of the universe is mixed, and that between human and divine a third
category exists, a race of dainiones who couple sometimes with mortals, some­
times with gods. It is not for a historian to decide:62
onto? pev oov xpf) nspl rwv tolwvSe 86^t)S e/etv, ... otiie Kaipo? ev Tip
napovn hiaoKOJtetv apKEt te doa cptXoadcpoig JtEpl auxcbv eXe/Sti.
What view should we take of such matters? This is not the appropriate place to
consider it; what the philosophers have said on the subject is sufficient.
The only safe thing to say is ‘The Romans believe that the twins were the sons
of Mars’.63
The next miracle story in the narrative is the suckling of the twins by the
she-wolf. Dionysius reports it from Fabius Pictor as an apparently supernatural
event,64 but then offers the rationalised version in which the lupa was really
Acca Larentia.65 He attributes the latter to ‘those who think that mythical
material does not belong in historical writing’, but he himself does not privi­
lege that version. Like Livy, no doubt he wanted to satisfy readers of bot
persuasions.66
So too on the disappearance of Romulus, Dionysius juxtaposes the more
legendary’ story, that he was swept up to heaven by his father Mars, and t e
‘more credible’ version, that he was killed by the senators.6 But this time e
adds a comment of his own:
EOIKE 6’ OU pucpctv dcpoppi’jv JtapEXELV TOtg 0EO3TOLOUOI T& 0V1]TU Kai El?
oupavdv dvaPi[3d^ouoi rag tpu^ag twv tbttcpavcbv rd ounpavia ek tov
0eou JtEp'i Tiyv ouyKptaiv toO avdpog Kai tt)v StaKpiatv.
But the things that happened by divine agency concerning both the formation
and the dissolution of that man seem to provide no little material for those who
make mortality divine and take the souls of famous men up to heaven.

62
62 1.77.3 (cf. n. 22 above for Varro); pace Gabba 1991.124-5, Dionysius does not endorse die
1.77.3 (cf.
first (or either) of these alternatives. [For the dainiones, see Plato Symposium 20. .J
63 2.2.3 (Dionysius’ summary of book 1).
64 1.79.7, daipdvtdv Tl xptjpa. 1.83.3 for Fabius FGrH 809 F4(b). ,
65 1.84.4; cf. Livy 1.4.7 (inde locum fabulae ac miracido datum), Plutarch Romulus 4.3 (EJtl TO
HtlOoibES EETpo;il]V Tfj tpiipp Jrapao/Etv), Onj>i> gentis Romanae 21.1 (from Va enus
Antias). .
66 1.84.1. Cf. 1.8.3 for die various categories of his readership, including those who like their
enjoyment undisturbed (ddxXrjTOg).
67 2.56.2-4; for the pvOwdEOTEpa/JtlQavwTEpa contrast, cf. 1.39.1 and 41.1 on Hercules (Ta
gEV puQtKaiTEpa rd 6’ dXi]0EOTEpa).

i
 HISTORY, POETRY, AND ANNALES 255

He refers to the two mysterious eclipses of the sun that took place at the
conception and the death of Romulus, even though the former had been part
of his ‘mythical’ narrative of that event.68
The impression that Dionysius’ sympathy did not lie with the sceptics is
strengthened by his treatment of Numa. He knows, of course, the standard
explanation of the Egeria story, that it was Numa’s ‘noble lie’ to impress the
people; but he attributes it to ‘those who remove all legends from history’, and
conspicuously fails to endorse it himself, saying that such matters relating to the
gods would require too long a discussion.69
When he comes to Numa’s establishment of the Vesta cult, his digression
on the historical evidence for the goddess’s miracle-working powers is intro­
duced with a very explicit declaration:70
0001 pev ovv xa; ciOeov; ciokouoi tpiXoooq lug, el 6f| Kai tpiXoaotpia;
avxa; Sei kciXetv, airaoa; btaovpovxs; xa; EJtitfavEia; xtbv 0eiov xa;
nap’ "EXXrioiv t) Pappctpot; yEvopEva; Kai xaiixa; eI; yEXwxa JtoXvv
a^ovot xa; loxopiag aXai^ovEiat; avOpwnivai; avia; avaxi0EVXE;, to;
ov&evl 0Eii)v ptXov avOptbjxwv oudEvo;. boot 6’ ouk dnoXvovoi xfj;
dvOpwntvi]; ejripEXeta; xou; 6eov;, aXXa Kai xoi; dyaOoi; evpevei;
Eivai vopitovoi Kat xoi; kokoi; SuopEVEi; 6ia jtoXXi]; eX1]Xv06xe;
loxopia;, ovbr xavxa; vnoXqipovxai xa; EJiupavEia; Eivai anioxov;.
Those who practise the atheist philosophies (if they deserve to be called
philosophies at all), who ridictde all the manifestations of the gods which have
taken place among either Greeks or foreigners, will reduce these reports as well
to mocking laughter and attribute them to human trickery, on the grounds that
none of the gods is concerned with anything human. But those who do not
absolve the gods from the care of humanity, but after going deeply into history
conclude that they are favourable to the good and hostile to the wicked, will
not suppose that even these manifestations are incredible.
‘Going deeply into history’ is just what Dionysius claimed to be doing, and this
sort of subject matter was well worth recording.71 Indeed, it was an essential
part of the usefulness of history for statesmen and politicians, as Dionysius
points out when narrating the part played by daiiiiones in revealing a conspiracy
in the ninth year of the Republic.72

68 2.56.6, cf. 1.77.2.

69 2.60.5-61.3 (quotation from 61.1); see above, n. 16.
70 2.68.1-2 (see above, nn. 33-4), citing the reports of oi Ovy'/paifElg. [Ct. Livy 10.40.10 on
doctriiui decs spemens.]
71 1.5.4, 3.18.1, 7.66.1-3, 11.1.4-5 (akribeia); 2.68.1 (Jtdvv b’ aElOV ... laxopnoai).
72 5.56.1, cf. 5.54.1—3; not in Livy. (Probably from Tubero: see Wiseman 1994.54—7.) Diony­
sius writing for politikoi: 1.8.3, 5.75.1, 11.1.1.
 256 UNWRITTEN ROME

That could hardly be further away from our Polybian preconceptions about
what is appropriate to history.73 But perhaps it would not have surprised Varro,
or the Q. Cicero of De diuinatione. I think Emilio Gabba was mistaken in
saying that Dionysius ‘sought, as far as it lay within his power, to leave aside
any myth involving divine intervention in human affairs’.74 On the contrary,
he not only related such myths but even drew his readers’ attention to the
philosophical issues involved in them.
The clearest statement of Dionysius’ position comes in book 8. Coriolanus
has yielded to his mother and withdrawn his army; the Senate has honoured
the women of Rome by the dedication of a temple to Fortuna Muliebris.
There follows an episode which Livy omits, but which Dionysius expressly
describes as ‘appropriate to the form of history’.75 The matroiiae had had a statue
of the goddess made at their own expense; on the day of the dedication, this
statue spoke, not once but twice (so there could be no mistake), confinning
the propriety of the women’s act.76 The event was recorded in the chronicle of
the poiitifices, and Dionysius reports it for two very specific reasons:77
... ibtavopGcfiaEcog eveko twv otopevtov pr]T’ ejt'i rate ripaic raig nap’
avGpwjtwv xaipeiv roiig 0Eoug pi)T’ ejt'i xatg dvooioig Kai abtKoig
jrpa^Eotv ayavaKtetv ... iva roig psv EuXafisaTEpoiq nspl to ovvexeiv
ag Jiapa twv npoydvaiv 6d§ag ujtep toO datpoviou napeXafiov
dHETapeXTiTog fj rotauTT) jipoalpsoig Kat (kfiaia diapcvri, rolg 6’
wrepopcboi tcov naiptiov E0iopd>v Kai pi]0ev6g noioOai to baipoviov
T<x)v avOpojjtivajv Xoyiaprfiv Kupiov paXiora psv dvaOcollai Tavri)v tt]v
fioljav, el 6’ dviaiaig exovolv, eti paXXov auToig dnExtiavEoGai Kai
KaKofiaipovEOTEpotg Eivai.
To correct those who think that the gods are neither pleased by honours
received from mortals nor offended at unholy and unjust acts. . . [and] that
those who are more scrupulous in preserving the beliefs about the divine power
which they have inherited from their forefathers may keep their conviction
firm and unwavering, while those who despise ancestral custom and claim that
the divine power has no control over human reason may either abandon that
view or, if they are incorrigible, become even more hateful and god-forsaken.

73 Cf. Polybius 3.47.6—9 on the participation of gods and heroes in real events as 7taoi]g
loropiag fiXXotpubraTov.
74 Gabba 1991.118; but cf. Dionysius of Halicarnassus 1.40.5 (epiphany of Hercules in 312 BC,
story promised), 2.68.3 (much more to say on Vesta’s epiphanies), 6.13 (Castor and Pollux
at Lake Regillus), etc. What Dionysius rejected at 2.18.3—19.1 was not myth as such (Gabba
1991.119-20) but ‘tales unworthy of the gods’ (n. 22 above).
75 8.56.1 (ioTopEtrai at 56.2, loropiav at 56.4). For the loiopiag OXHlia. cf. 1.41.1, with
Fox 1993.44-5.
76 8.56.2-4; cf. Valerius Maximus 1.8.4 and Plutarch Coriolanus 37.3.
77 8.56.1 (rbg al Ttbv IfpoqiavTwv ntpie/ovoi ypatpai) = Annales pontifimtun fr. 20 Chas-
signet; cf. Frier 1979.118.
 HISTORY, POETRY. AND ANNALES 257

Yet again we find that the question asked was not so much ‘What is appro­
priate to the genre?’ as ‘What ought we to believe?’.
That order of priorities is even more evident in Plutarch, whose biograph­
ical format allowed him more space than Dionysius had for philosophical
reflection. On the disappearance of Romulus: can we really believe that the
body of a great man, and not just his soul, might be borne up to heaven? On
Numa and Egeria: although the gods must love virtuous mortals, and may
instruct wise rulers, surely they cannot take sexual pleasure in a mortal body?
On the appearance of Castor and Pollux in the Forum at the time of the battle
of Lake Regillus: why should it not be credible, when even in modern times
the divine power can bring instant news of battles far away?78 These are impor­
tant questions, and Plutarch is not dogmatic about them. ‘If anyone says
otherwise, “Broad is the way”, as Bacchylides has it.’79
As for the speaking statue, of course it was impossible (statues have no vocal
cords), and if the historical evidence forces us to believe it, then we must
assume that some dream-like power of the imagination caused people to
believe they heard it speak.8" But as Plutarch knows, there is also quite another
way of looking at it:
ov pf|v dkXa rot; v;r’ Evvoia; Kat qtkia; npo; tov Oedv ayav spnaOto;
e'/ovoi, Kat pqhfv «0eteiv ppd’ avuivEoOat Ttbv Toiourtov buvapevoig,
pEya npo; rticTiv eoti to 0avpciotov sal pf| kuO’ qpu; Tij; too 0eoO
bvvupEW'.
Nevertheless, for those who are passionate in their devotion and love for the
divine, and incapable of rejecting or denying anything of this kind, the main
argument for belief is the wonderful and transcendant nature of the divine
power.

On this view, miracles are impossible only in human terms: divine power is
beyond our understanding. Plutarch himself urges a cautious middle way,
between empty superstition and arrogant contempt for the gods.81
It seems therefore that what we are disposed to think of as a matter of
literary decorum was in reality a profound philosophical and theological ques­
tion. Those who argued that myths had no place in history' were not just
defining a genre; they were putting into practice a controversial belief about

78 Plutarch Romulus 28, Numa 4, Aemilius Paulins 25; [Nock 1972.535-6,] Wardman
1974.163-6.
79 Plutarch Numa 4.8. Other gnomic quotations in the same context: Romulus 28.6 (Pindar),
Coriolanus 38.4 (Heraclitus), Camillus 6.4 (Delphi motto).
80 Plutarch Coriolanus 38.3, OJtOV d’ f)p&; fj tOTOpta TOD.ok djtoflldtETai Kai TtOavoi;
paprootv. . . For Romans who believed in animate statues, ct. Lucilius 484—9M = 524-
9W (on terriculas Lamias, Fauni quas Pompiliique instilum Numae).
81 Plutarch Coriolanus 38.3—1; Camillus 6.4 for another talking statue (Juno's, at Veii).
 258 UNWRITTEN ROME

the nature of the gods. Dionysius and Plutarch were both familiar with that
school of thought,82 but neither endorsed it himself, and there is no reason to
suppose that they were unusual in that.83

IV Valerius Maximus and Livy

My final witness is an author whose evidential value has been long neglected
because of a misconception about the purpose of his work. It is a widely held
view—but one based on no evidence that I know of—that Valerius Maximus’
collection of historical exempla was written as a handbook ‘for students and
practitioners of declamation’.84 The author’s own statement of his aim is set
out clearly enough in the preface, which deserves quoting in full:85

(1) urbis Romae exterarumque gentium facta simtd ac dicta memoratu digna, quae apud
alios lathis diffusa stmt qtiam ut breuiter cognosci possint, ab inlnstribus electa auctonbus
digerere constitui, ut documenta sumere uolentibus longae inqitisitionis labor absit. (2) nec
mihi cimcta complectendi cupido incessit: quis enim omnis aeui gesta modico uolumimmi
immero comprehenderit, ant quis compos mentis domesticae peregrinaeque historiae seriem
felici superiorum stilo conditam uel attentiore cura uel praestantiore facundia traditurum se
sperauerit? (3) te igitur huic coepto, penes quern hominum deorumqiie consensus maris ac
terrae regimen esse uoluit, certissima salus patriae, Caesar, inuoco, cuius caelesti
prouidentia uirtutes, de quibtis dictums sum, benignissime foiienttir, uitia seuerissime
iiiudicantur: (4) nam si prisci oratores ab Ione Optimo Maximo bene orsi sunt, si excel-
lentissimi nates a mmiine aliquo principia traxerunt, mca paniitas co instills adfauorem
trnmi decucurrerit, quo cetera diuinitas opinione colligitur, tua praesenti tide patemo
auitoque sideri par uidetur, quorum eximio fulgore mulmm caerimoniis nostris inclutae
claritatis accessit: rcliquos enim deos accepimus, Caesares dedimus. (5) et quoniam
initimn a cultu deomm petere in animo est, de condicione eius summatim disseram.

82 No myths in history: Dionysius of Halicarnassus 1.84.1, 2.61.1. No divine involvement in

human affairs: Dionysius of Halicarnassus 1.77.3, 2.56.6, 2.68.2, 8.56.1; Plutarch Coriolanus
38.3, Camillus 6.2—4. T ,
83 Cf. Dio Cassius 55.1.4 on a miraculous apparition in 9 bc: OavjtaOTOV pt'V OW . .. Oil
|1EVTOI Kai djtlOTEtv EXW. Dio became a historian when id datpoviov appeared in his
sleep and encouraged him. See n. 159 below, and Millar 1964.179—81: ’an unquestioning
adherent of traditional pagan belief and observance’ (179).
84 Bloomer 1992.1, cf. Fantham 1996.133: ‘a reference work for orators, or more likely
deciaimers, needing precedents for the subjects of their speeches.’ Contra Skidmore
1996.xvi-xvii, 53-4. Wardle 1998.12-15 hedges his bets: ‘I would argue that the serious
moral purpose envisaged by Skidmore should be combined with a primary' audience ot those
involved in declamation...’ (The reference to declamantes in the preface to Julius Paris
epitome is evidence only for the fourth or fifth century AP.)
85 Text as in J. Briscoe’s 1998 Teubner edition and D.R. Shackleton Bailey’s Loeb (2000); R.
Combas’ Bude (1995) reads alacritatis (with the best MSS), not claritatis, at §4. I have added
the section numbers. [See now Mueller 2002.17-19 for a good defence of alacritatis.]
 HISTORY, POETRY, AND ANNALES 259

(1) I have decided to set in order a selection, from distinguished authors, of the
memorable deeds and sayings of Rome and of foreign nations, which are too
widely scattered in other writers for knowledge of them to be acquired in a
short time, in order to save those who wish to receive the lessons the labour of
long research.
(2) I have not been seized by the desire to include everything. For who could
embrace the events of all time in a small number of volumes? Who in his right
mind could hope to hand on the whole sequence of Roman and foreign
history, recorded in the happy style of previous writers, either with more
careful accuracy or with more outstanding eloquence?
(3) You therefore, Caesar, the surest salvation of our country, 1 invoke for this
enterprise — you whom the consensus of men and gods has wished to be the
ruler of land and sea, you whose heavenly providence most generously favours
the virtues, and most strictly punishes the vices, of which 1 am about to speak.
(4) For if the orators of old rightly began with Jupiter Best and Greatest, if the
most excellent bards took their beginnings from some divine power, all the
more properly will my humble self have had recourse to your favour, in that
the rest of divinity is inferred from mere opinion, whereas yours is seen by
immediate frith as equal to the star of your father and grandfather, whose
exceptional brightness added much splendid brilliance to our ceremonies; for
the other gods we have received [from the past], the Caesars we have given [to
the future).
(5) And since it is my intention to seek my starting point in the worship of the
gods, I shall deal briefly with its nature.

Valerius uses the idiom of historiography —not just omnis acid gcsta and
historiae series (§2), but the title phrase itself, /inni tic dicta inemoratu digna (§1),
which is reminiscent of Cicero’s definition of the subject-matter of history in
De oratore.*"' His opening phrase, and cupido imessit in §2, seem to be borrowed
from Sallust.87 But the most striking feature of this passage is its repeated inter-
textual reference to a famous preface of about sixty years earlier, that of Livy
himself.

86 Cicero De oratore 2.63: unit enim, qiioiiiam in rebus nunjiiis memoria^ue dignis consilia prinniin,
deinde acta, posted euentus exspectentur, et de consiliis significari quid scriptor probet et in rebus bestis
declarari non solum quid actum aut dictum sit, sed etiam quo inodo... [Cf. Tacitus Annals 3.65.1
on exemplary dicta Jactaque as the subject matter of antia/es.]
For inemoratu dignum, and the more frequent nienioria diginiin, see Oakley 1998.138-9;
the idea goes back at least to Herodotus (1.16.2, 1.117 d^lOJinYnTOTara). Servius auctus on
Aeneid 1.373 (digua inemoratu) and Isidore Etyniologiae 1.44.3 (digna nienioria) preserve Verrius
Flaccus’ account of the amiales inaximi: Frier 1979.27-37, 40-1.
87 Sallust Catiline 6.1 (urbem Romain, imitated by Tacitus Annals 1.1.1), 7.3 (cupido incesserat); cf.
Guemni 1981.29-60, ‘Valerio Massimo e Sallustio’. Tony Woodman reminds me that oiiine
aeiium (§2) may be an allusion to Catullus 1.6, and thus to Nepos’ Chronica.
 260 UNWRITTEN ROME

The first allusion defines Valerius’ purpose (§1). Livy had carefully regis­
tered the ethical value of history:88

hoc dlud est praecipue in cognitione rerum salubre acfrugifenim, omnis te exempli docu­
menta in inlustri posita momunento intueri; hide tibi tuaeque rei pnblicae quod imiterc
capias, hidefoedum inceptu foedum exitii quod nites.
It is this in particular which is salutary and beneficial in the knowledge of
history, that you behold lessons of every sort of example set forth in a conspic­
uous monument. From there you may take, for yourself and your republic,
what to imitate and what, corrupt in origin and corrupt in outcome, to avoid.

Valerius repeats the key word documenta and applies to it the verb sumere, corre­
sponding to Livy’s capias. Failure to recognise this has resulted in some strange
translations of documenta sttmere uolentibns,89 but the allusion and the meaning
are both clear enough: documenta is from docere, and history teaches.
Livy’s conspicuous second-person address to the reader is transferred by
Valerius to the emperor (§3)—no longer the citizen of a republic choosing
what to imitate or avoid, but a god-like arbiter meting out reward and punish­
ment. Even so, Valerius’ uirtutes and uitia are recognisably rhe quod imitere and
quod nites of Livy’s preface. Similarly Livian are the passages where Valerius
identifies accuracy and style as the possible ways to surpass one’s predecessors
(§2),90 and appeals to the poets’ habit of beginning with the gods 4)?' Finally,
where Livy had referred to the magnitudo of his rivals, Valerias speaks of mea
paruitas (§4).92
In view of all this, and of the brute fact that in his time Livy’s huge narra­
tive was the classic statement of domestica historia, one might expect Valerius to
share Livy’s attitude to stories of miracles and divine intervention. But no: in
his first book alone, he presents gods and goddesses acting as protagonists in
one story after another,93 and his moral lesson on the ubiquity of divine

88 Livy pref. 10; Moles 1993.152-5.

89 E.g. ‘those who want concrete evidence* (Bloomer 1992.14); ‘those who wish to embrace
the examples’ (Wardle 1998.29); ‘chi volesse compulsare tali fond’ (Faranda 1971.65). The
key explanatory text is Varro De lingua Latina 6.62, defining documenta as texempla docendi
causa.
90 Livy pref. 2, noui semper scriptores ant in rebus certins aliquid allatnros se aut scribendi arte rudeni
uetustatem superaturos credimt. Parallel noted by Herkommer 1968.79 n. 1.
91 Livy pref. 13, precationibus deorum dearumque, si ut poetis nobis quoqne mos esset...; cf. n. 1
above.
92 Livy pref. 3, nobilitate ac magnitudine eomm qui nomini efficient meo consoler. Valerius perhaps
alludes to his own cognomen Maximus, itself evidently a badge of nobilitas (Skidmore
1996.113—17); but if so, the allusion is modest, not ‘ludic’ (Wardle 1998.70).
93 1.1.18 (Apollo), 1.1.19 (Aesculapius), l.l.ext.l (Juno), 1.6.12 (Jupiter). [Add 3.2.1 (di
immortales keep Codes safe), 3.4.4 and 4.5.3 (acts of Fortuna), 8.15.5 (gods send crow for
Valerius).]

4
 HISTORY, POETRY, AND ANNALES 261

concern for mortals, and divine punishment of wickedness, is spelt out again
and again.94
The fire bums on Servius Tullius’ head;9’ the voice ofSilvanus is heard in
the wood;96 Castor and Pollux fight for the Romans at Lake Regillus;97 T.
Latinius reports his vision to the Senate, and is cured;98 the statue of Fortuna
addresses the matrons;99 queen Juno ofVeii says she wants to go to Rome;1*1 a
divine vision warns Decius Mus and his colleague in a dream;11” the angry
Hercules blinds Appius Claudius and wipes out the Potitii.102 Miracle stories
which Livy either omits altogether or keeps in carefill oratio obliqua are narrated
by Valerius in a confident indicative mode which leaves no room for doubt.103
Particularly revealing is what he does with the material he takes from
Cicero. In book 2 of Cicero’s De natura deonim, the Stoic spokesman Lucilius
Balbus uses divine epiphanies as part of his argument for the existence of the
gods; for instance, on the day of the battle of Pydna in 168 BC, P, Vatienus of
Reate was met by two young men on white horses who told him that king
Perses had just been captured. In book 3, however, C. Cotta pours scorn on all
such stories as mere unsubstantiated rumour. Valerius ignores the sceptic’s
point, and reports the epiphany story as simple fact,1114 just as he does with two
dream stories adduced by Quintus Cicero in book 1 of De diuinatione and
refuted by Marcus in book 2.105
Valerius was writing moral protreptic, not philosophical argument. He sets
out his position with admirable clarity:106

94 1.1.ext.3, lento euini gradu ad uindictam sui diuina procedit ini; 1.6.1, sic Inniiaiia consilia casti-
gantur ubi se caeles/ibus praeferunr, 1.6.12, quibus apparel caeleslium iiiinieii ... uohisse; 1.7.1,
quid ergo aliud ptiianms quant diuino nurnine ejfectimi ....’; 1.8.2, iiiinieii ipsius dei ... compro-
bauit.
95 1.6.1 (Livy 1.39.1, Jerunt). Also L. Marcius in 211 BC: 1.6.2 (Livy 25.39.16, miracula addunl).
96 1.8.5 (Livy 2.7 2, adicittnt miracula).
97 1.8.1 (not in Livy: see n. 13 above).
98 1.7.4 (Livy 2.36.8, traditum nienioriiie esr).
99 1.8.4 (not in Livy); see above, pp. 256—7.
100 1.8.3 (Livy 5.22.6, tabulae adieetunt esf).
101 1.7.3 (Livy 8.6.9, dicilur).
102 1.1.17 (Livy 9.29.10, traditur).
103 Not only in book 1: see also 5.6.2 on M. Curtius and the chasm (Livy 7.6.1, diciturjemnt),
5.6.3 on Genucius Cipus (not in Livy: see n. 112 below), 8.Labs.5 on Tuccia (Livy Epitome
20, cf. pp. 248-9 above). [See also Mueller 2002.39 on Valerius Maximus 1.1.20 (contrast
Livy 42.28.12) on the anger ofjuno Lacinia, and 2002.40—1 on Valerius miracles in general.)
104 Cicero De natura deonim 2.6 (3.11—13); Valerius Maximus 1.8.1, with close verbal parallels.
For the name (not Vatinius) see Tavlor 1960.262-3.
105 Cicero De diuinatione 1.56 and 59 (2.136-41); Valerius Maximus 1.7.5-6, verbal parallels
again.
106 Vai. Max. 1.8.7. Wardle 1998.59 unfortunately translates Kempfs 1888 Teubner text,
which made nonsense of the passage by printing Novak's conjecture uera for iniiiii in the
final phrase.
 262 UNWRITTEN ROME

nee me praeterit de motu et noce deoruni immortaliuni hnmanis ocnlis anribnsqne percepto
quam in ancipiti opinione aestimatio uersetur, sed quia non nona dicuntur sed tradita
repetnntur, jidcm auctores uindicent: nostrum sit inclutis litteraruni inonunientis conse-
crata pcrindc ac liana non refugisse.
I am not unaware how controversial is the question of the perception by
human sight and hearing of the movement and voices of the immortal gods.
But since what is said is not new, but a repetition of what has been handed
down, my authorities may defend their own credibility. Let it be my part not to
have avoided, as if they were false, events that have been consecrated in famous
literary monuments.

The particular event that gave rise to this declaration was the miraculous
double migration of the Penates from Alba Longa back to their home in
Lavinium. We know some of the famous literary monuments in which that
story was consecrated: the fourth book of the pontiffs’ Annales inaxinii,107 the
second book of L. Cincius Alimentus’ history,108 the second book of L. Julius
Caesar’s Pontificalia,lm the first book of Q. Aelius Tubero’s Historiae,"0 and the
first book of Dionysius of Halicarnassus’ Roman Antiquities.'" Not one of them
was the work of a poet.

V Poetry and Annates

So Varro, Cicero, Dionysius, Plutarch and Valerius Maximus have brought us

to an unexpected conclusion. Livy’s comment on what is appropriate to poets’
stories, far from being a truism, turns out to be a partisan statement of philo­
sophical scepticism. Other historians took the opposite view, and did accept
miracle stories and divine epiphanies as a proper part of their subject matter.
The issue was one not of literary convention but of theological belief.
The common ground of prose and poetry is well illustrated by an item in
Valerius Maximus’ chapter on patriotism. The praetor Genucius Cipus, leaving
the city on campaign, finds that he has grown horns; the soothsayers tell him
that it signifies he will be king if he returns; he goes into voluntary exile. The

107 Origo gentis Ronianae 17.3 = Annales pontifieuni fr. 3 Chassignet; Frier 1979.50-2.
108 Origo gentis Ronianae 17.3 = FCrH 810 F9 (unnecessarily categorised as doubtful): Richard
1983.165-6. [See now Cameron 2004.329-34 on OCR's source citations, esp. 330 on this
passage.]
109 Origo gentis Ronianae 17.3, cf. 9.6, 15.5; Richard 1983.139.
110 Origo gentis Ronianae 17.3, not in Peter’s HRR fragments. Preliminary material on Tubero
in Wiseman 1979.135-9.
111 Dionysius of Halicarnassus 1.67.2 (OaOpct geyiOTOV XeyeTai yEveoOat). Possibly from
Timaeus (cf FCrH 566 F59); but Dionysius used authors who used the pontifical chronicle
(1.73.1, 1.74.3, 4.2.1, 7.1.6, 8.56.1), and Tubero was one ofhis Roman patrons (Dionysius
of Halicarnassus Thucydides 1).

I
 HISTORY, POETRY, AND ANNALES 263

same story is told in Ovid’s Metamorphoses."2 Valerius’ account of the Penates’

migration offers a similar parallel. He introduces it as something 'recognised in
its own time [which] has come down to later generations’—-just what Virgil
puts more succinctly when he has a miracle to report: priscaftdes facto, sed fama
perennis."3
It was a way of thinking very familiar to the Romans. To believe in rnira-
des and divine intervention is to believe that the gods are concerned with the
world of humans, that they take note of moral excellence and its opposite, and
reward or punish it accordingly. Cicero, in his guise as legislator, made it the
programme of his law-code to secure that belief in the minds of the citizens,
and we have seen both Dionysius and Valerius Maximus deliberately selecting
their material to do the same.114
So it should not be a surprise to find that the pontifices—whose duty it was
to instnict the people,115 and whose chronicle was consulted as a repertory of
precedent and example116—included in that chronicle such exemplary para­
digms of divine power as the migrating Penates and the speaking statue of
Fortuna Muliebris.117 They also included one of the weirdest of all such mani­
festations, the phantom phallus that appeared in the royal hearth and fathered a
miraculous child on queen Tanaquil’s maidservant. Dionysius, Pliny and
Plutarch all have that story, and so does Ovid, in a poem about ‘sacred matters
drawn from the old annales’.'"' We may as well burn the annales, says Quintus
Cicero, it we’re not going to believe in the gods’ concern for mortals. He was
talking about the story of Altus Navins and the whetstone, which no doubt
also featured in the pontiffs’ chronicle.119
Modem historians treasure the idea of the antiales pontijicnin as an ancient
archive, guaranteeing at least the framework of early Roman history.12" In my

112 Valerius Maximus 5.6.3, noui atque iuauditi generis prodigiuni; Ovid Metamorphoses 15.565-
621. Pliny (Nat. Hist. 1 1.123) calls Cipus fabulosus, like Actaeon.
113 Valerius Maximus 1.8.7, quod suo saeeulo cognition iiiananit <id posteros (trans. Wardle
1998.59); Virgil Aeneid 9.79. Cf. Ovid Metamorphoses 1.400 (quis hoc credat, nisi sit pro teste
netustas?), Fasti 4.203—4 (pro magno teste netustas | creditor: acceptaru parce tnouerefideni).
114 Cicero De legibus 2.15—16; see above, nn. 70, 77 (Dionysius), 94 (Valerius).
115 Livy 1.20.6-7, on Numa’s appointment of the first pontifex: ... ut esset quo consultant plebes
ueniret... ut idem pontifex edoeeret . . . Cicero De oratore 2.52, on the notice-board put up in
the house ot the pontifex maximus: potestas ut esset populo cognoscendi.
116 Livy 8.18.11-12, 27.8.8-9; Frier 1979.98-100.
117 See above, nn. 77, 107.
118 Dionysius of Halicarnassus 4.2.1—3 = Annales pontificum fr. 13 Chassignet; Pliny Nat. Hist.
36.204, Plutarch Moralia 323b—c (De fontina Rontanorutn 10); Ovid Fasti 6.625-36; also
Amobius Aduersus nationes 5.18. Sacra. . . annalibus eruta priscis: Ovid Fasti 1.7, cf. 4.11.
119 Cicero De diuinatione 1.33 (p. 252 above); cf. Frier 1979.222.
120 See most recently Cornell 1995.13—15 and Oakley 1997.24—7. Servius auctus (on Aeneid
1.373) says the pontifex maximus recorded digna memoratu ... doini militiaeque terra niarique
gesta; but that evidently reflects the eighty-book edition described by Verrius Flaccus (Frier
1979.27-37).
 264 UNWRITTEN ROME

view that is a somewhat sanguine idea. It was certainly an archive, but what the
pontiffs chose to record for the good of the citizens was not necessarily what
the historian of res gestae would like to know. It is ironic that the very item
which is supposed to show how far the chronicle went back—the report in the
annales maxinii of an eclipse on 21 June ‘about 350 years after the foundation of
the city’—is cited by Cicero not from a historian but from Ennius’ epic
poem.121 Indeed, Ennius evidently took the title, and perhaps the concept, of
his Annales from the pontiffs’ chronicle, long before any prose writer ever used
that title for a historical work.122
The word annales occurs in the first book of the Aeneid, where the ship­
wrecked hero introduces himself to his disguised mother:123

'o dea, si prima repetens ab origine pergam

et iiacet annalis nostrorum audire laboruni,
ante diem clausa coinponet Vesper Olynipo.
nos Troia antiqua, si uestras forte per auris
Troiae nomen iit, dinersa per aequora uectos
forte sua Libycis tempestas appulit oris.
sum pins Aeneas, raptos qui ex hoste penatis
classe ueho meciim,faina super aethera notus;
Italian! quaero patriam, et genus ab Ione sumnio.’
‘O goddess, if I were to start at the beginning and retrace cur whole story, and
if you had the time to listen to the annals of our suffering, before I finish the
doors of Olympus would close and the Evening Star would lay the day' to rest.
We come from the ancient city of Troy, if the name of Troy' has ever reached
your ears. We have sailed many seas and by the chance of the winds we have
been driven ashore here in Libya. I am Aeneas, known for my' devotion. I carry'
with me on my ships the gods of my house, the Penates, wrested from my
enemies, and my fame has reached beyond the skies. I am searching for my
fatherland in Italy. My descent is from highest Jupiter.’
By annales in line 373 Aeneas refers to the long story he will eventually tell in
books 2-3, a narrative full of miracles and divine interventions. But the ancient
commentators took the word to be a specific allusion to the pontiffs chronicle;
indeed, it is from their commentary on that line that much of our information

121 Cicero De republica 1.25 = Annales pontificmn fr. 8 Chassignet = Ennius Amides 153 Sk (r.v
hoc die quern apud Emiiiim et in amidibus maximis consignation iiidenms); Skutsch 1985.311—13.
The year date is very uncertain: see most recently Humin 2001). 106-9.
122 Skutsch 1985.6-7; Frier 1979.216-19. The first certain use of Amides as the title of a prose
history was evidently by L. Piso censorius after 120 BC: Wiseman 1979.12-16, Forsythe
1994.42 (‘Piso’s work is the earliest Roman history’ for which a strictly annalistic treatment
of the early and middle republic is attested’).
123 Virgil Aeneid 1.372—80, trans. D. West (Penguin Classics).

4
 HISTORY, POETRY, AND ANNALES 265

about the chronicle derives.124 They inferred that Aeneas is to be thought of as

a pontifex, which seems reasonable enough given his self-definition as pins
Aeneas and the carrier of the Penates.
The Aeneid was being composed at the time of Augustus’ restoration of the
82 temples, when the people wanted him to be pontifex niaxinius.125 Horace in
the last of his ‘Roman Odes’ was repeating the traditional lesson that the gods
control everything, and success depends on their goodwill.126 Aeneas the pious
pontifex, ancestor of Augustus, fits well into that context.
On 6 March 12 BC, Augustus finally did become pontifex niaxinnts, to the
huge enthusiasm of the Roman people;127 and at some point not far distant
from that date (between Cicero and Verrius Flaccus), a vast new eighty-book
edition of the annales pontifiann was produced.12,1 It seems to have been a char­
acteristically Augustan project. Just like the great lists of consuls and
trinniphatores that decorated his triumphal arch, or the annotated statues of the
heroes of the Roman past that thronged his new Forum, so too the accumu­
lated (and no doubt expanded) documentation of all past ponti/lces maxiini
would encompass the whole of Roman history in a teleological sequence
culminating in Augustus himself.129
It we are right to see miracle stories and reports of divine intervention as
part ot the pontiffs’ traditional material—and two of the four certain fragments
ot the eighty-book edition are clear examples of that13"—it is worth remem­
bering that Augustus’ own memoirs were full of such stories,131 and according

124 Macrobius Saturnalia 3.2.17; Servius auctiis on Aeneid 1.373.

125 Augustus Res tjesf.ic 10.2 ipopulo id saierdotium deferetile mihi), 20.4 (temples, begun in 28 BC).
126 Horace Odes 3.6, esp. 5-6: dis te niinorem quod geris, iniperas: hint oiniie principiiiiii, hue refer
exitinn. Compare Cicero De legibus 2.15: sir igitur hoc him a prineipio persuasion ciuibus, dominos
esse omnium at moderatores dens, eaque quae gerantur eonun geri indicia at iiuminc.
127 Augustus Kes gestae 10.2, cf. Ovid Fasti 3.415-28. For the importance of the pontificate in
Augustus’ sell-presentation, see Bowersock 1990.
128 Frier 1979.193-200; Chassignet 1996.xxxvi-xl refuses to accept Frier's conclusion, but
nowhere engages with his arguments.
129 For learning and scholarship as tools of Augustan authority, cf. Wallace-Hadrill 1997.11-
22; the eighty books were ‘a great collection of aetiological explanations of Roman rites
and institutions’ (Rawson 1991.15).
130 Amiiiles poiuifieum frr. 3 and 5 Chassignet (Origo gentis Romatiae 17.3, 18.3); the others are
frr. 4 and 7 (ibid. 17.5 and Aulus Gellius 4.5.1-6).
131 Augustus De uita sua frr. 4, 6, 7, 12 Malcovati (Tertullian De anima 46; Pliny Nat. Hist.
2.93; Servius auetus on Eclogues 9.46; Plutarch Brutus 41.5): prophetic visions seen by Cicero
and M. Artorius, the comet signifying Caesaris atiimam inter deomm inimortaliiini nuuiina
receptam, and the instant punishment of a haruspex who revealed the secrets of the gods. For
the autobiography and its context, see Ando 2000.139-43: 'to remind people of the reasons
for their gratitude and their loyalty’ (141).
Sulla’s memoirs were similarly full of divine interventions (frr. 6, 8, 9, 15, 16, 18. 21
Peter); and a later imperial author, Marcus Aurelius in his Meditations (2.11, 9.27, 12.28),
was quite sure that the gods exist, care about human affairs, and communicate their will
through dreams and oracles.
 266 UNWRITTEN ROME

to Suetonius one of his favourite entertainments was listening to aretalogi,

story-tellers who specialised in tales of the deeds and power of the gods.132 Yet
that was the type of story that Livy in his preface saw as appropriate to poetry.
Perhaps this is the place to bring in something Livy says in book 5, where
the phraseology, clearly reminiscent of the passage in the preface, may cast
some light on what sort of poetry he had in mind:133
sed in rebus tain antiquis si quae similia ueris sint pro ueris accipiantur, satis habeam;
haec ad osteiitatioiiein scenae gaudentis iniraculis aptiora quam adJidcm neque adfinnare
neque refellere est operae pretium.
But in matters so ancient let me be content if some things like the truth are
accepted as true. It is not worth the trouble either to endorse or to refute these
things: they are more appropriate to the ostentation of the stage, which delights
in miracles, than to credibility.
This refers to an episode which Livy carefully introduces as a fabula, the next
act of which was the speech ofjuno Regina, signifying her willingness to go to
Rome.134 Since fabula can also mean the plot of a play, perhaps, as in Varro,13’
the poetae Livy was thinking of were dramatists.
The stage that delighted in miracles was among other things a vehicle for
the moral and political education of the Roman people,,13fc and that included
instruction about the gods—how they helped Rome,137 L«>. how they
...-7 championed
------ ,
virtue,138 how they punished sinners.139 That is something not often mentioned
in histories of Roman drama, but the evidence for it is clear enough. Nor is it
often mentioned in accounts of the pontiffs’ chronicle that those responsible
for the eighty-book edition saw fit to include (at book 11) a learned explana­
tion for an iambic senarius line, malum consilium consultori pessimum est, allegedly
sung by children round the city when the wicked machinations of Etruscan
haruspices were foiled.140 Perhaps it was a catch-phrase from a popular play on

132 Suetonius Diuus Augustus 74. Examples of aretalogy are listed, with bibliography, in Engel­
mann 1975.37, 55-6.
133 Livy 5.21.9. The echo of pref. 6 is noted by (e.g.) Moles 1993.148—9, Kraus 1994.283-4,
Forsythe 1999.43.
134 Livy 5.21.8 (iuseritur hide loco fabula), 5.22.6 (indefabidae adicctuin est uocein quoque dicentis uelle
auditani). Fabula praetexta inferred by Ribbeck 1881.
135 Augustine City of God 6.5 (= Varro Autiquitates diuinae frr. 7—10 Cardauns), cf 6.6 (on di
poetici theatrici ludicri scaenici).
136 Moral: full documentation and discussion in Rawson 1991.570—81. Political: Polybius
6.56.8 and 11, with Zorzetti 1980.64—5; Cicero Pro Rabirio Postumo 29, with Leigh
1996.186-8.
137 Plautus Amphitruo 41—5 (gods in tragedies reporting their benefactions).
138 Ovid Fasti 4.326 (the miracle of Magna Mater and Q. Claudia).
139 Cicero Pro Roscio Aincrino 66—7, In Pisonem 46, De legibus 1.40, Academica 2.89, De hams-
picuni response 39 (the Furies and their blazing torches as a commonplace of tragedy).
140 Annales pontijicuin fr. 8 Chassignet (Aulus Gellius 4.5.5). Cf. Pliny Nat. Hist. 28.15 (citing
annales) for a similar story.

4
 HISTORY, POETRY, AND ANNALES 267

the subject; at any rate, it suggests common ground in the preoccupations of

the pontiffs and the stage.
What I think may have happened is this.
First, it was an ancient custom (though we don’t know how ancient) for
the pontifex ttiaxiitius to put up in the public part of his house a notice-board
listing events for the information of the Roman People. According to Cato,
crop failures and eclipses were characteristic items; no doubt the main purpose
was to make known items that might reveal the gods’ attitude towards Rome,
and thus to encourage the citizens in pious and god-fearing behaviour.141
This material was also accumulated into a chronicle, variously known as
amides pttblici,142 annales popitli Romani,w annales pontifiami,144 or amiales
ma.vimi.143 Presumably the chronicle consisted of a series of volumes,146 added
to by successive pontijices niaxinti, and kept in a place where they could be
consulted by the learned and literate.147
1 imagine the ordinary people had little use for the chronicle. A bulletin
board was one thing (Roman citizens were used to reading notices, or having
notices read to them144), but they needed no consolidated archive to discover
how the immortal gods had manifested their power and majesty in times past.
They had prophets (nates) to tell them that, and to keep the fear of divine
punishment firmly before their eyes.149 And those who were too sophisticated
to ‘listen to a prophet’ (one of the signs of superstition, according to Cicero)
would get the same message from the performances at the Intli saienici.151'

141 Cicero De oninw 2.52, Servius uiunis on Aeneid 1.373, Dionysius of Halicarnassus 1.74.3;
Cato Origines fr. 77P = Annales pontifiaini tr. 2 Chassignet (Aulus Gellius 2.28.6). Cf. Livy
1.20.6 on the pontiliees’ responsibility for oinnia publiea prinataqne sacra.
142 Cicero De repnblica 2.28; Seneca Consolatio ad Polybinnt 14.2; Diomedes in Granimalici Latini
1.484K.
143 Cicero De dcnno 86.
144 Cicero De legibns 1.2.6; Ongo mentis Romanae preface, 17.3, 17.5, 18.3.
145 Cicero De onttore 2.52, De repnblica 1.25; Aulus Gellius 4.5.6; Festus (Paulus) 113L; Macro-
bius Saturnalia 3.2.17; Servius mirnu on Aeneid 1.373.
146 Cf. Horace Epistles 2.1.26 (poiitifiann libras), which could be a reference to the chronicle. Its
original form may be inferred by comparison with the evidence for Etruscan ‘linen books’
in the fourth century uc: see most recently Torelli 1996.13—22, esp. 19-20, figs 2-3.
147 E.g. by Ennius, and probably by L. Cincius Alimentus about 200 uc: see above, nn. 121
and 108.
148 Public notices: e.g. Cicero De lege agraria 2.13 (text of law), Ad Attiann 2.20.4, 4.3.3 (edicts,
speeches); cf. Millar 1998.45, 103. Private notices (lost property, debts, etc): e.g. Plautus
Ruderts 1294; Cicero Pra Quinetia 50, Ad Attiann 1.18.8; Propertius 3.23.23; Seneca De
beneftdis 4.12.2.
149 Lucretius 1.102—9 and Dionysius of Halicarnassus 5.54.3, with Wiseman 1994.49-67. Cf.
Livy 25.1.7—10 (nates in the Forum, 213 bc); Isidore Onlines 6.8.12 (praecepta of the nates
Marcius).
150 See above, p. 266. Cicero De diuinatio 2.149 on superstitio: instat enim et urget et quo te antique
nerteris persequitur, sine tu nateni sine tn omen andieris...
 268 UNWRITTEN ROME

Part of the prophets’ task was taken over by the poets of early Roman liter­
ature (which is why in post-Varronian Latin nates can also mean a poet). In the
preface to book 7 of his Annales, a famous passage, Ennius carefully distanced
himself from the Saturnian metre of‘the Fauns and the nates’, but his subject­
matter, especially in the early books, must have substantially overlapped with
theirs.151 About seventy years later, L. Piso the ex-censor published a dry
historical chronicle in prose—a very different author and a very different work,
but he evidently called it Annales, and what it had in common with Ennius’
epic poem may be seen from how they dealt with the Trojan pre-history of
their subject. Ennius’ Anchises foretells the future, Venus having given her
lover the power of prophecy; Piso’s Aeneas passes through the victorious
Greek forces at Troy by a miracle, thanks to his piety.152 Either of those items
could have come from the pontiffs’ chronicle, which indeed both authors
evidently used.153
It is difficult for moderns to imagine a world in which prophecy, poetry,
history and moral exhortation were not always thought of as separate concep­
tual categories. The best evidence is provided not by the great thinkers but by
the commonplace minds of antiquity. Here is Diodorus Siculus in the preface
to his Universal History:'5''
el yap f| rtuv ev $bov gvOokoyia Tijv ujeoGeoiv jrE7iXaopEVT)V rtouaa
jtokka oupPakkEiai wig avOptrnwig jrpdg EucfrfiEtav rcI <fri:aioavvT)v,
nooq) paXXov wwXj]jtTeov rf|v jtpocpnTiv Ttjg aXrjOEtag iowpiav, Tijg
oXt| cpiXoootpiag oIove'l prjTpdjwXtv ovoav, EjriOKEvaoffat Ta ijOi]
ptaXkov npog KaXoicayaQiav;
If the [poets’] mythology of Hades, fictitious though its subject-matter is, makes
a great contribution to piety and justice among men, how much more must we
assume that the prophetess of truth, history, which is as it were the mother-city'
of philosophy in general, is even more able to equip men s characters for a
noble way of life?

151 Ennius Annales fr. 206-7 Skutsch (quoted by Varro De lingua latino 7.36; Cicero Brutus 71,
75-6, Orator 157, 171, De rliuinationc 1.114; Quintilian 9.4.115; Orixo .ventis Romanae 4.5);
cf. frr. 54-5 and 110-1 ISk (Romulus’ deification), 113Sk (Egeria), 139Sk (Lucumo and the
eagle), etc.
152 Ennius Annales fr. 15-16 Skutsch; Piso fr. 3 Forsythe (= 2P). Forsythe 1994.96-7 wrongly
translates miracnlo as ‘by way of inspiring awe and amazement’: awe and amazement are the
result, not the cause, of the miraculmn, as is clear from Livy (5.46.3) on the similar story' of
Fabius Dorsuo.
153 Ennius: n. 121 above. Piso fr. 33 Forsythe = Annales pontificum fr. 5 Chassignet (Onge mentis
Romanae 18.3), a story of divine punishment.
154 Diodorus Siculus 1.2.2, cf. Polybius 6.56.12; Polybius, not a conunonplace mind, treats it
as a means of social control.

J
 HISTORY, POETRY, AND ANNALES 269

Narrative set in the underworld was reserved to the poets. So too was narrative
set in heaven, as Servius remarks about the Aeneid:'55
.. .constat ex diuinis huinanisque personis, continens tiera aimjictis; mini Aeneam ad
Italian! uenisse nianifestnni est, Venerem uero locutam cum Ione missumue Merciirium
constat esse composition.
It is made up of divine and human characters, and contains both truth and
fiction; for Aeneas’ journey to Italy is plain fact, whereas Venus’ conversation
with Jupiter and the mission of Mercury are agreed to be invented.
Virgil himself marks the transition from the underworld back to the historical
mode by the passage of Aeneas through the ivory gate.151' Everything that
happens on earth, including divine manifestations and even Jupiter brought
down to earth to bargain with Numa, could appear equally in a historian, and
for the same hortatory’ purpose.1’7
P. Mucius Scaevola, who was pontijex inaximus from 130 to about 115 BC,
ended the practice of putting up the notice-board in his house. Presumably,
though we cannot be sure, that means that under the next seven pontifices
maxinii there were no further additions to the pontiffs’ chronicle.1’11 But during
those hundred years there was plenty of material to be recorded, as the disasters
of the dying Republic offered evidence of the gods’ anger. We see the late
Republic through the rational eyes of Cicero, but other sources can give us a
different angle; Dio, for instance, reveals the anxiety’ of the people in 55-54 bc
when Gabinius restored the king of Egypt against the express instructions of
the Sibyl.'5”
The prophets were busy throughout that time.16" So too were the histo­
rians, the poets and the playwrights, recording and dramatising the

155 Servius on .'linen/ 1 pref. The clearest statement is by an anonymous poet of the first
century AU (zlctmi 74—91), inveighing against the stage (plurima pars scaenae return cst Jdllacia,
75) and the mentiti nates (79): his examples of their mendosae mdgata licentia famae (74, cf.
libertas at 91) are the sinners and lawgivers in the underworld, the wars of the gods, and
Jupiter’s adulterous disguises.
156 Virgil Aeneid 6.893—9. Cf. Donatus uita I'eigilii 38 and ps.Probus uita Vergiia ad fin., quoting
a poem on Augustus’ refusal to allow the Aeneid to be burned: tn, maxiine Caesar, | non sinis
et Latiae consults historiae. According to Servius (on Aeneid 1.382), Lucan’s epic was a historia,
not a poema; it is noteworthy that the underworld narrative is not in the poet s voice but at
third hand, as the revived corpse tells what he says the ghosts told him (Lucan 6.776-9).
157 Valerius Antias fr. 6P (Amobius Aduersus nationes 5.1); cf. Wiseman 1998.21-3.
158 L. Metellus Delniaticus, ?115—103 Be; Cn. Domitius Ahenobarbus, 103—r89; Q. Mucius
Scaevola, ’89-82; Q. Metellus Pius, 82-63; C. Julius Caesar, 63—44; M. Aemilius Lepidus,
44-12.
159 Dio Cassius 39.15.1-4, 55.3, 56.4, 59.3, 60.4-61.4 (esp. 61.1 on the Tiber flood, 61.3 on
the anger of TO batpovtov); cf. n. 83 above.
160 E.g. Cicero De diuinatione 1.4 and Plutarch Marins 42.4 (87 bc); Sallust Histories 1.67.3M
(78 bc); Cicero De consulatu suo quoted in De diuinatione 1.18 (63 BC).
 270 UNWRITTEN ROME

catastrophes.161 When the crisis was eventually over, the charismatic son of
Divus lulius would find no shortage of pious examples and moral paradigms to
present to the grateful people, whether in his own memoirs or in a great new
edition of the pontiffs’ chronicle.162 What I think we have to infer is a complex
process of mutual borrowing and mutual influence among the poets, the
prose-writers and the pontifices, lasting from at least the third century BC to the
time of Augustus. The eighty books of the new pontiffs’ chronicle will have
contained a vast amount of stuff taken from literary sources, but those sources
themselves may well have been influenced by the chronicle in its earlier mani­
festations. And the whole enterprise would have been meaningless without the
fundamental motivation that we have found in one ancient source after
another—the need to understand the true relationship between gods and men.
If this is a credible reconstruction, then Livy’s position becomes a little
clearer. I take it that when he wrote his preface, at a time when guilt and
anxiety were still prevalent, the old pontiffs’ chronicle was largely forgotten
and the new one not yet created. What we see in him is not so much Augustan
piety as a tolerant and patriotic form of Ciceronian scepticism.
For Livy, divine intervention was not appropriate to ‘uncorrupted
history—but we know that other historians thought it was. For Cicero,
different rules applied when history was written in prose or in verse but
Quintus had had to ask him whether that was his view, which shows that not
everybody thought so.163 The very fact that purists tried to insist on rational
criteria entitles us to infer that the general attitude was different from theirs.
Even in the sophisticated Rome of the first century BC, for many readers the
distinction between the proper pursuits of poets and historians was far from
clear-cut, and certainly not a simple matter of literary genre.

161 Playwrights: no fragments or titles known, but lost plays—on the death of C. Gracchus, the
crossing of the Rubicon, the Ides of March—are inferred by Wiseman 1998.52-63 and
Weinstock 1971.353-4. [See Keaveney 2003 for a sceptical view.]
162 See above, p. 265.
163 Livy pref. 6; Cicero De legibus 1.5.

J
 CHAPTER SIXTEEN

The House of Tarquin

i
The archaeological evidence for archaic Rome has been spectacularly
increased in the last twenty years by Andrea Carandini’s ongoing excavations
on the north slope of the Palatine, between the Arch of Titus and the temple
of Vesta [fig. 47]. His first campaign, begun in 1985, revealed the remains of
four substantial houses, dateable to about 530 BC and evidently occupied
continuously for the next three centuries.1 Destroyed in the third century
(serious fires are recorded at Rome in 241, 213 and 210 BC),2 they were rebuilt
and survived in their new form until ‘Nero’s fire’ in AD 64.
Enough survives of one of the sixth-century houses (‘domus 3’) to show
that it was designed round an atrium [figs 48-9, pp. 274-5]? The remains of
the other three are much more fragmentary (indeed, hardly anything at all
survives of’domus 1’ and ‘domus 2’), but they are confidently reconstructed
by Carandini on the assumption that they too were atrium houses? That is not
what a cautious archaeologist would do, but Andrea Carandini is not a cautious
archaeologist. He takes the generous view that reconstruction is the highest
aim of excavation, ‘the one moral justification for our work’. He believes that
archaeologists have an obligation to make explicit, and visible, for the lay
public what they as experts can infer from the remains they have uncovered.
He has nothing but contempt for those of his colleagues who are too 'academ­
ically coniine il Jaut' to take risks. Tn life and in research, you get nowhere
without risk and error.’5

1First published in Carandini 1990, whence Holloway 1994.67 (plan of 'donnis 3'), Smith
1996.252 (general plan). Detailed excavation report in Carandini and Carafa 1995.215-59;
historical summary and reconstruction drawings in Carandini 1995.50-67.
2 Orosius 4.11.8-9, Augustine City of God 3.18 (241 BC); Livy 24.47.15-16 (213 bc),
26.27.1-5 (210 uc).
3 Plan in Carandini 1990.98 (= Holloway 1994.67) superseded in Carandini 1995.60-1, tav.
57-8.
4 Carandini 1995.52 and 54, tav. 49 and 51, for the surviving remains of‘domus 1 and 2;
contrast 53 and 55—9, tav. 50 and 52—6, for the reconstructions.
5 His position is stated with candour and eloquence in Carandini 2000.148-55; the phrases
quoted in the text are from pp. 149 and 155.
 272 UNWRITTEN ROME

* ~ Arch of
Li Titus

-.JU J
f , " T ’ ’ T ’ ’ [HfV',. «
•
’
a

■ °= __________________
___
Fig. 47 The ‘Sacra Via’ from the arch of Titus to the temple of Vesta. A: site of
‘doinus 3’ (figs 48—9). B: site of‘house of Tarquin’ and "Porta Mugonia’ (figs 52-4).

Error is indeed always possible, as Carandini knows perfectly well: Recon­

structions, by their nature, are nothing but combinations of hypotheses ...
Any scientific hypothesis may lead us into error, especially in the absence of a
critical sense, and thus be taken as proved fact.’6 That is the risk Carandini is
proud to take. The trouble is that the more plausible the reconstruction looks,
the less motivation there will be to apply one’s critical sense and ask whether
the data justify the hypotheses.
It is important to understand the motive for Carandini s method. As he says
himself, what drives him is ‘the magic of discovering the materials to construct
something . .. the pleasure of giving shape to the shapeless . He cites wit i
approval Italo Calvino’s idea of a ‘mental cinema’ for ever projecting images
inside our heads:11

6 Carandini 2000.153 (‘Le ricostrnzioni sono per loro nature nient’altro che combinazioni di
ipotesi’), 154 (‘Qualsiasi ipotesi scientifica pud indurre in errore, specie in assenza di senso
critico, ed essere quindi assunta come verita provata’).
7 Carandini 2000.73.
8 Carandini 2000.155 (citing Calvino 1988.83): ‘Cid accade quando pensiamo, quando
conversiamo, quando leggiamo ... e sopratutto quando sogniamo e in un secondo rius-
cianio ad ammobiliare un intero casteilo e a vestire tutti i cortigiani. Cosi anche
1’archeologo, partendo da indizi e framnienti, non pud impedirsi di autoproiettare ne a
propria mente il “cinema mentale” di quelle cose e di queU’edificio finalmente capiti. Se e
onesto, lavora su questa imagine, la perfeziona e la trasfonna in una fignra che sia visibile
anche ad altri, grazie a un disegno di cui e responsabile.’
 THE HOUSE OF TARQUIN 273

That’s what happens when we think, when we talk, when we read ... above
all when we dream, and in one second succeed in furnishing an entire castle and
putting all the courtiers in costume. So too the archaeologist, starting from
indications and fragments, cannot stop himself projecting in his mind the
‘mental cinema’ of those objects and that building finally understood. If he is
honest, he works on the image, perfects it, and transfonns it into a shape visible
to others as well, by means of an illustration for which he takes the responsi­
bility.

Note the phrase ‘cannot stop himself. Carandini is not just recommending
good professional practice; he is describing a psychological disposition.
So it is no surprise to find the same phrase at the climax of his presentation
of the sixth-century houses:’'

We cannot stop ourselves imagining the following scenes: Ocrisia in the

kitchen, possessed by the phallus of the Lar emerging from the hearth to beget
Servius Tullius, as was believed to have happened in the palace of Tarquinius
Priscus in the grove of Vesta; the mistress of the house weaving in her chamber;
the store-house under lock and key, full of furnishings and foodstuffs; the
throne, the marriage bed, the boxes of writing-tablets in the tablinuin; the
banquet-songs sung in the dining-room; the servants on the upper floor; the
clients in the iitiui ’i (about five hundred of them could get in for the Hilutiitio);
and the family on the terrace, nearly twice as wide as the Sacra Via, following
the course of the Luperci, the triumphal procession or other rituals in the
narrow street below, and three metres away the terrace of the family opposite.

Naturally, the hypothetical rooms in Carandini’s reconstruction drawings are

all labelled, as he sees them in his mental cinema.
None of this is meant as criticism. How else can hypotheses be created, if
not by the exercise of an informed imagination?1" The purpose of this intro­
ductory section has been merely to make clear Carandini’s method, before
considering his hypothesis about the house ofTarquin.

9 Carandini 1995.51: ‘Non possiamo impedirci di immaginare le seguenti scene...

10 As one of Carandini’s distinguished predecessors protested when criticised—'quasi che 1 in-
tuizione non fosse la somma di dati positivi, che sfuggono all’occhio delle moltitudini’
(Giacomo Boni, letter ofjune 1899 quoted in Wiseman 1987b.132).
 V

274 UNWRITTEN ROME

Fig. 48 ‘Domus 3’, first reconstruction (after Carandini 1990.98).

 the house OF TARQUIN
275

vicus
TV TV
]
penus
WiT

hortus tablinum chamber^

triclinium
of the |1
mater
Z familias
o ___
Puteal '|CUlinaQ
2 vestibulum |
\ focus
&
puteal

ala impluvium
ala

’irant cubiculum
.. burial cubiculum

cubiculum

taberna
taberna

& ‘D°mUS 3’’ —nd reconstruct (after C I i

ion (after Carandim 1995.60, fig. 57)
 Tlf> UNWRITTEN ROME

II
I think it is no accident that at the end of the first stage of his excavation
programme Andrea Carandini had in his mind a story set ‘in the palace of
Tarquinius Priscus in the grove of Vesta’." There are only two ancient sources
for the site of the house of Tarquinius Priscus. Solinus, probably using Varro,
says that it was ‘by the Porta Mugonia, above [or beyond] the top of the Nova
Via’,12 while according to Livy one of its windows looked out on the Nova
Via, ‘for the king lived by the temple of Jupiter Stator’.13 In 1990 Carandini
thought he knew where the Nova Via and the Porta Mugonia were, and that
they adjoined the grove of Vesta [fig. 50].14 All that area would be explored in
the second stage of his programme.
How the second stage altered Carandini’s conception of the archaic topog­
raphy can be seen in the sketch-plan he published in 2004 [fig. 51]. The Nova
Via and the Porta Mugonia have now been moved, the grove of Vesta is much
reduced from its former size, and a large area is newly marked off as ‘the house
of the Tarquins and of Servius Tullius’. Carandini’s paradoxical description in
1995 of the house of Tarquinius Priscus as ‘in the grove of Vesta’ (why should
there be a palace in a sacred grove?) marked a transitional phase in his reinter­
pretation, after the remains of an early house were discovered at the eastern
end of the area marked on the 1990 plan as Vesta’s grove.H
The fiill report of the new excavations is still to come, ' but Carandini has
now provided some preliminary information about ‘the house of the Tarquins.
Three construction periods are identified: ‘second haJ/iast quarter of the
seventh century’ (construction in scaglie di tufo rosso e atppellaccio); second quarter
of the sixth century’ (construction in scheggiotti di tufo rosso) ; last quarter of the
sixth century or first half of the fifth’ (construction in opera quadrata di cappel­
laccio).'1 The first two of these phases are interpreted as the house of Tarquinius
Priscus and the ‘official’ house ofServius Tullius, respectively.Ih Rebuilt with a
grandiose cryptoporticus after the fire of 210 BC, the house survived to the time
of Augustus, when it was destroyed and replaced with a warehouse (Jiorretiin).

11 Carandini 1995.51: ‘come si credeva fosse awenuto nella regia di Tarquinio I risco nc‘ 1 .
12 Solinus 1.24 : ad Mugoniam portain supra summam notiani uiani. Cf. Varro De him popnh ■
fr. 7 Riposati = 290 Salvadore (Nonius 852L) for the list of the kings kings ’ houses.
ousts.
13 Livy 1.41.4: perfenestras in iionani iiiam uersas, habitabat eniin rex ad louis Statons.
14 Carandini 1990.97, fig. 4.2; contrast Carandini 2004.89, fig. 5. ...
15 Carandini and Carafa 1995.264: ‘sulla base della scoperta in quest area durante indagim
ancora in corso di structure in cappellaccio. . . ’
16 Carandini, Carafa and Filippi, forthcoming.
17 Carandini 2004.46, cf. 68. 73. .
18 The concept of a king’s ’official’ residence, though evidently important to Caran mi ( Precl
sione necessaria’, Carandini 2004.53, cf. 56, 63 etc), is never explained, and seems to e an
anachronism. Where did the king actually live?
19 Carandini 2004.46, cf. 59, 63-4.
 THE HOUSE OF TARQUIN 277

Fig. 50 Grove of Vesta and Porta Mugonia, first reconstruction (after Carandini
1990.97).
Fig. 51 Grove of Vesta and Porta Mugonia, second reconstruction (after Caran­
dini 2004.89, fig. 5).

Carandini’s plan of the house very helpfully distinguishes preserved

remains, probable structures and hypothetical structures,2" though naturally the
degree of probability (and the criteria employed to establish it) will not
become clear until the full excavation report is published. When the three
categories are presented separately [figs 52-4], the extent of Carandini’s hypo­
thetical reconstruction becomes clear.

20 Carandini 2004.93, fig. 9c.

 THE HOUSE OF TARQUIN 277

Fig. 50 Grove ot \ ; nJ Porta Mugonia, first reconstruction (after Carandini

1990.97).

Fig. 51 Grove ot Vesta and Porta Mugonia, second reconstruction (after Caran­
dini 2004.89, tig. 5).

Carandini’s plan ot the house very helpfully distinguishes preserved

remains, probable structures and hypothetical structures,2" though naturally the
degree ot probability (and the criteria employed to establish it) will not
become clear until the full excavation report is published. When the three
categories are presented separately [figs 52-4], the extent of Carandini’s hypo­
thetical reconstruction becomes clear.

21) Carandini 2004.93, tig. 9c.

 278 UNWRITTEN ROME

A
X%
k r k. r

-4 4
I* . I‘
:S. 't :S. n

JOm

Fig. 52 ‘House of the Tarquins’, surviving remains (after Carandini 2004.93, fig. 9c).
Fig. 53 ‘House of the Tarquins’, surviving remains and p-.ibable structures'
(after Carandini 2004.93, fig. 9c).

It is based, once again, on the reasonably well attested layout of domus 3

[figs 48-9 above], Carandini at first believed that the two houses were contem­
porary, and he attributed the newly-discovered one to 1 aiqumius Superbus in
the late sixth century.21 Now that its origin has been taken back a century
earlier and attributed to the first Tarquinius, ‘domus 3 is a much less obvious
paradigm for its layout; so Carandini now suggests, without argument, that t le
atrium houses of his first excavations deliberately imitated the pic existing
house of the Tarquins.22
We must assume that they imitated it down to the last detail, for Carancnu
reproduces the layout of‘domus 3’ very precisely [fig. 54), even to the mystc
rious corridor across the back of the house. (The corridor did not appeal in his
first reconstruction of ‘domus 3’,23 and it is far from clear what pm post it
served.) However, the rooms on either side of the tablimmi are reversed in
comparison with ‘domus 3’, the kitchen, the chamber of the mater Jamilim
(gineceo)\ and the storeh
..louse exchanging places with the triclinium and its associ-

21 Carandini 1995.42; Carandini 1997.ill.xx.

22 Carandini 2004.63 (‘verosiniilniente’), 69 (‘e dune]ue possibile’), 72 (‘verosiniilniente’).
23 Carandini 1990.98, whence Holloway 1994.67.
 THE HOUSE OF TARQUIN 279

Fig. 54 ’House ot the I'arquins’, hypothetical reconstruction (after Carandini

2004.93, fig. 9c).
 280 UNWRITTEN ROME

ated anteroom.24 There must be a good reason for these choices, but it cannot
be archaeological; that part of the site is empty of ‘preserved remains’.2’ Note
too the identification of the front part of the tablinum as ‘thalamos Fortunae’.
Since a thalamos should be a closed chamber, there must be a compelling reason
to identify it as a space in a huge open atrium that could hold five hundred
people.26
Of course there are reasons, and Carandini makes them very clear. They are
to be found in Livy, Dionysius, Ovid and Plutarch—in the stories which stim­
ulate his imagination and direct his mental cinema.

III
There are seven episodes of the ‘myth-history’ of sixth-century BC Rome
which are set in different parts of the palace of the Tarquins. They are

(1) the conception of Servius Tullius,

(2) the fire portent,
(3) the assassination plot,
(4) Tanaquil at the window,
(5) the goddess and the king,
(6) the eagle and snake portents,
(7) the girl on a horse.

1. The conception of Seivius Tullius.

and rfjg EOTtag twv paoiXeitov, ecp’ fjg aXXag te Poipaiot ouvteXovolv
LEpoupyiag teal Tag curd tcov Seutvcov curap/ag ayi^oucnv, wrtp to
Jiupdg avao/Eiv Xeyovoiv aiboiov avbpdg. tovto 6e OtaoaaOai ti]v
’Ok'pioiav jrpdtTT]v cpepouoav wug eiwfioTag TtEXavoug Eiri to irvp. • ■
They say that from the hearth [/icsfin] in the palace, on which the Romans oiler
various other sacrifices and also consecrate the first portions of their meals, there
rose above the fire a man’s privy member, and that Ocrisia was the first to see it
as she was carrying the customary cakes to the fire. . ."7

24 Bafflingly labelled ‘I’cstibulmn’ (Carandini 1995.51, ‘il vestibolo del grande triclinium ), PP
283-4 below for the meaning of the Latin word. . .
25 Carandini 2004.69: ‘Abbiamo alcune tracce archeologiche fric], ma non suftici P
proporre una ricostruzione archeologica vera e propria. Possiamo pero sempre a
spasso—riempire quello spazio..... , t
26 Plutarch Moralia 273B (6 KaXoUpEVOg Tvxng OaXapog); for the atrium as a public space 6 <
commimem omnium usum) see Varro De lingua Latina 5.161.
27 Dionysius of Halicarnassus 4.2.1 (Loeb translation).
 THE HOUSE OF TARQUIN 281

.. .itctpOsvov tt|v ’OKppaiav ajrapypara Kai koipfjv ekciotote

kappctvonoav card Tfjq paaikiKng TpaitEtrig ejti Tfjv eotiov Kopi^Eiv* Kai
tote tuxeTv pev auTtyv, foOTTEp Eiw0Ei, Tip jTi’p'i Tac. ctnapxag
ETtpukkovoav, aicpvibiov 6e Tfjq epkoyoe; papav0Eiar|q popiov avbpo?
dvareivai yovqtov ek Tfj<; Eoxiag...
Ocrisia was a maiden who took the first-fruits and the libations on all occasions
from the royal table and brought them to the hearth [/iesri.1]; and once on a
time, when she chanced, as usual, to be casting the offerings on the tire,
suddenly, as the flames died down, the member of a man rose up out of the
hearth. . ,28
Ovid’s version is only slightly different:
Ikinc secttin Tanaqnil sacris de more peractis
inssit in ornatnm Jimdere iiina focnm;
hie inter cineres obscaeni forma uirilis
ant fnit ant nisa est. . .
After performing with her the sacred rites in due form, Tanaquil
ordered Ocrisia to pour wine on the hearth [/belts], which had been
adorned. There among the ashes there was, or seemed to be, the shape
of the male organ. . ,29
Carandini imagines this as taking place in the kitchen (ai/imi),3" but the sources’
emphasis on formal sacrifice, particularly Ovid’s ‘adorned hearth’
(omnium.. . focnm), seems to rule that out.
Where iiw the hearth in a palace in the sixth century BC—or rather, where
was it assumed to be when tins story was first created? One possible analogy
might be the ‘central hearth’ iu Agamemnon’s palace (as imagined by
Aeschylus), where the sheep stood ready for sacrifice to greet the king’s return;
or we might think of the hearth m the palace of Admetus, king of the Molos-
sians, where Themistocles took refuge in about 470 BC.31 One could imagine
either scene in the central courty ard of the sixth-century Etruscan palace at
Murlo.32

28 Plutarch Moralia 323b (Loeb translation).

29 Ovid Fasti 6.629—32 (Loeb translation).
30 Carandini 1995.51; Carandini 2004.63.
31 Aechvlus Agamemnon 1056 (EOTtac neoogqdkov), Thucydides 1.136.3.
32 Holloway 1994.56-9, fig. 4.7.
 282 UNWRITTEN ROME

2. The fire portent.

co tempore in regia prodigium uisit euentuque mirabile fuit. ptiero donnienti, ctti Serttio
Tullio fuit nomen, caput arsisse ferunt imiltorinn in conspectu; phtrinio igitur clamore
inde ad tantae rei miracuhtm orto excitos reges.. .
At this time there happened in the house of the king a portent which was
remarkable alike in its manifestation and in its outcome. The story is that while
a child named Servius Tullius lay sleeping, his head burst into flames in the sight
of many. The general outcry which so great a miracle called forth brought the
king and queen to the place. . ,33
Where were those numerous witnesses? Dionysius’ version is more specific.

KaflqpEvov yap hot’ avTofi iXEao'uoqi; paktoQ’ rgiEpaq ev xfi jraoxabt

tow paotXeiwv koi kotevex0evto5 ecp’ vjtvov, nvp dnEkapipEy and xfjg
KEtpaXfiq auxoD, fj te frqTT)p auxov Kai f) toO paoikscng yvyf]
nopEvogEvat <5ta xfjg naaxabog e0Ecioavxo Kai ndvxeg boot ovv xatg
yuvat^lv ETvy/avov tote napovxEg.
When he had fallen asleep one day while sitting in the portico [pasto] of the
palace about noon, a fire shone forth from his head. This was seen by his
mother and by the king’s wife, as they were walliking through the portico, as
1"’"" thrnurh
well as by all who happened to be present with them at that time.34

Where would ‘the portico of the palace’ be in an atrium house? The scene
seems to be imagined in a colonnaded courtyard or peristyle, but there is no
room for that in Carandini’s reconstruction.

3. The assassination plot.

The sons of Ancus Marcius plan to kill Tarquinius Priscus:
ex pastoribus duo ferocissimi delecti ad facinus, qnibiis consueti cram utcrque agrestibus
ferramuitis <annati,> in uestibulo regiae quant potucre tumiiltiiosissinie specie rixae in
se onines apparitores regios conuertunt; inde, cunt ambo regent appellarent clamorque
eoriiin penitns in regiatn pcruenisset, uocati ad regent peigunt.
Two very desperate shepherds were selected to do the deed. Armed with the
rustic implements to which they were both accustomed, they feigned a brawl in
the entrance-court fiiestibiiliini] of the palace and, making as much noise as
possible, attracted the attention of all the royal attendants; then they appealed to

33 Livy 1.39.1 (Loeb translation). Cf. Valerius Maximus 1.6.1 [domesticoniin oculi adnotauenmt).
34 Dionysius of Halicarnassus 4.2.4 (Loeb translation). For itaOTUf as the Greek for porticos, see
Plutarch Galba 25.5.
 THE HOUSE OF TARQUIN 283

the king, until their shouts were heard inside the palace and they were sent for
and came before him.3’

The forecourt and the king’s attendants are not mentioned in Dionysius’
version of the episode (he just says the brawl took place ‘near the palace’), but
he too makes it clear that the assassins were outside, and called in by the king
to have their case heard.36
There is no forecourt in front of Carandini’s atrium houses, just a row of
shops with the terrace above from which the family could look out across the
narrow street to their neighbours opposite. He puts the uestibulum inside the
house, identifying it as the first part of the atrium itself. On this matter 1 think,
with respect, that he is demonstrably wrong.
For the meaning of the term uestibulum we have something rather rare in
our sources, an explicit definition from an expert authority. It is provided by
Aulus Gellius, in his discussion of common errors:3'

aiiiinaduerti enini quosdam hautquaquam indactas uiros opinari uestibulum esse partem
damns priorem, quant uttlgus atrium uocat. C. He/iiis Gallus in libra de significatione
uerbaniin quae ad ins citiile pertinent uestibulum esse elicit non in ipsis aedilms iieqne
partem aedium, sed locum ante iamtam damns uaeuiim, per quem a ilia aditus acccs-
susque ad aedis est, cum dextra sinistraque iamtam tecta saepiimt iliac iiincta atque ipsa
ianua procul a uia est, area uacanti intersita.
For I have observed that some men who are by no means without learning
think that the vestibule is the front part of house, which is commonly known as
the atrium. Gaius Aelius Gallus, in the second book of his work On the Meaning
of Words Relating to the Cit’d Cut', says that the vestibule is not in the house itself,
nor is it part of the house, but is an open space before the door of the house,
through which there is an approach and access [adirus .taessiisqiie) to the house
from the street, while on the right and left the door is hemmed in by buildings
extended to the street and the door itself is at a distance from the street, sepa­
rated from it by this vacant space.

Aelius Gallus' definition is confirmed by the consistent usage of ancient authors

from Plautus onwards. The uestibulum was in front of the house,38 and one

35 Livy 1.40.5 (Loeb translation).

36 Dionysius of Halicarnassus 3.72.2 (7tkr]aiOV TWV flaOlkEltDV), 72.3 (fionuXfoaC avtoiig 6
flaoiXevg).
37 Aulus Gallius 16.5.2—3 (Loeb translation) = Aelius Gallus fr. 7 Funaioli; repeated almost
verbatim in Macrobius Saturnalia 6.8.15—16.
38 E.g. Plautus Mostellaria 817 (ante aedis), Varro de lingua Latittal.M (ante doinuiii), Varro De re
nistica 3>.9.7 (metaphor for yard in front of henhouse), Vitruvius 6.7.5 (.nite ianiios = Greek
ItpoSupa), Aulus Gellius 16.5.8 (ante ianuam), Servius on Aeneid 4.507 (tie quit introeat).
 284 UNWRITTEN ROME

went out of the house to reach it;39 it provided an approach to the door,40 and
might even be shared between two neighbouring houses.41
The ticstibnliini was where the clients congregated in the early morning,
waiting for the moment when they would be admitted to pay their respects.42
There was plenty of opportunity there for quarrels and fights,43 as in the story
of the brawling shepherds. The king’s attendants (apparitores) were probably
another authentic touch in the story, if, as seems likely, the lictors of a magis­
trate waited in his nestibnliiin.‘>4
The more important the house, the bigger the uestibnlnm. There might be a
colonnade, portrait statues of the ancestors, spoils of battle hung on the walls.45
Pompey the Great’s nestibtilniii even featured the rams of captured warships,
while that ofNero’s Golden House was designed to contain a colossal statue of
the emperor 120 feet high.46 When Catullus imagines the palace of king
Peleus, he has the nestibtilniii lined with tall trees—beeches, laurels, poplars,
cypresses.47 The forecourt of the king of Rome’s palace was no doubt thought
of as similarly grand.
How is it that Carandini’s reconstructed houses leave no room for so
conspicuous a feature? The answer, I think, is that having once decided that
the remains imply an atrium plan, he was over-influenced by the best-known
form of that layout, in the town-houses of Pompeii and Herculaneum.4’’ But
there was no need for grand forecourts in the small towns or Campania.41 For
the capital, that paradigm is fundamentally misleading.

PsetMits 953-5, </!»' exit in nestibnhnn)',

39 Varro De lingua Latina 7.81 (on Ballio at Plautus
Statius Siltiae 4.4.41-2 (iiestibnlo ... rogant exire clientes). \ ~> ( I’ this in Digest
40 Aditus: Cicero Pro Caecina 35, Pro Ahlone 75, Orator 50; Vitruvius
10.3.19.1.
41 Paulus in Digest 10.3.19.1; cf. Cicero Pro Milone 75 (Clodius building wall across his sister s
iiestibuluni).
42 E.g. Cicero Ail Atticinn 4.3.5, De oratore 1.200, Vitruvius 6.5.1-2, Seneca Epistles 84.12,
onso atio ad' Marciam 10.1 (cxchisonuii clicntimii tiirba referta liesfibula), Statius Sihiae 4.4.41-2,
uetonius Gains 22.2, Aulus Gellius 16.5.8-10 (priusqiiam admittcrcHtnr), Nonius 75L (dinn
mtroemidi darctur copia).
la r>neCa ^,st^es ^4.12: ilia tiimultuosa rixa salutaiitium limina.
44 De utns illustribus 20A: fasces lictorios foribns apposilos.
V 7SJU1,Sj//'''1‘' 817 Pliny Nat. Hist. 35.7 fans el circa limina . . . imagines),
irgi cnci 7.177—86 (statues and spolia). The ionic colonnade behind the shrine of
Sorrento base may represent the uestibuhim of Augustus’ house (Wiseman
1 >91.107—9).
Cicero Philippics 2.68; Suetonius Nero 31.1 (the Hadrianic temple of Venus and Rome prob-
47 r Y ^Pr“cnts the area of Nero’s planned uestilmlum).
Catullus 64.285-93 (wedding presents from a river god).
48 Marquadt 1886.213-50; McKay 1975.32-63.
Le idi'ToOs'' 5 ' °n th°se "',o don’t need uestibiila. For the few Pompeian exceptions, see
 THE HOUSE OF TARQUIN 285

4. Tanaquil at the window.

After the attack on Tarquinius, Tanaquil closes the palace doors to conceal the
king’s death and contrive the succession of Servius Tullius. The crowd outside
is noisy and anxious:
aim clamor impetusque iiiultitudinis nix sustineri posset, e.v superiore parte aediutu per
fenestras in noitain uiain uersas—hahitabat eniin rex ad louis Staton's—populiiin
Tanaquil adloquitur.
When the shouting and pushing of the crowd could hardly be withstood,
Tanaquil went up into the upper storey ot the house, and through a window
looking out upon the Nova Via—for the king lived near the temple of Jupiter
Stator—addressed the populace.'’"
The topographical reference was no doubt self-explanatory to Livy’s readers,
but for us the position of the Jupiter Stator temple and the course of the Nova
Via are very controversial questions.'’1
In 1990 Carandini’s line for the Nova Via was that argued by Filippo
Coarelli in 1983 [fig. 50, p. 277]; it corresponds to the orientation of the
archaic atrium lestae, behind which Giacomo Boni had found traces of early
paving.52 There are difficulties w ith Coarelli’s hypothesis (as there are with all
attempts to reconstruct this most elusive of ancient Roman streets), but it does
have the merit of fitting coherently into a comprehensive interpretation of all
the relevant literary and archaeological evidence. In Carandini’s 2004 version,
however [fig. 51, p. 2771, all coherence has been abandoned.’3
The paving found by Boni is now an ‘internal street’ inside the Vesta sanc­
tuary, leading nowhere and apparently closed at both ends.’4 Four different
pre-iniperial Novae Viae are now identified: the first two, for which there is
no archaeological evidence, are hypothesised as running outside the respective
eighth- and seventh-century Be: archaic walls;” the third, on a totally different
line and dated to the mid-sixth century, is inferred from paving and the

5l) Livy 1.41.4 (Loeb translation).

51 See Coarelli 1983.26—33, 227—55; contra Ziolkowski 1989.231—6. For a new examination of
the ancient evidence, see Wiseman 21104b on the Nova Via and Ziolkowski 2004 (esp. 8—
19, 32-8, 46—50, 58—64) on the Jupiter Stator temple. I remain unconvinced by
Ziolkowski’s argument that the temple was close to the Arch of Titus, tor three reasons:
according to Appian (Cit’d liars 2.11) it was close to the Forum; according to pseudo­
Cicero (Antequam in exsilium iret 24) it was at the foot of the Palatine; and according to
Plutarch (Cicero 16.3) it was ‘at the beginning of the Sacra Via as you go up to the Palatine’.
52 Carandini 1990.97, fig. 4.2; Carandini and Carafa 1995.262—3; Coarelli 1983.227—30 (contra
Ziolkowski 2004.43—6).
53 Carandini 2004.38-41, 87-92 (figs 3-9).
54 Carandini 2004.39, "percorso di servizio e chiuso’.
55 Carandini 2004.39 (‘Del percorso alto-arcaico della Nova Via non vi e traccia’), 88 (fig. 4);
cf. Carandini 1997.ill.xvii—xix and xxx.
 286 UNWRITTEN ROME
c
remains of a gate found in the second phase of the excavations;56 the fourth,
evidendy late-republican and on a different line again (coming out above the
Arch of Titus) is that identified by R. Santangeli Valenzani and R. Volpe in
1986.37 It is the third of these, misleadingly described as only a slight adjust­
ment of the Coarelli hypothesis,58 which concerns us here.
Coarelli’s Nova Via is outside the archaic city gate, as its name (via, not
tiicus) clearly requires; and it is below the grove of Vesta, as Cicero explicitly
states.59 Carandini’s Nova Via is inside the gate and above the grove.60 The
house of Tarquinius Priscus was ‘above [or beyond] the top of the Nova Via’,61
but Carandini has the ‘top of the Nova Via’ sloping down towards it.62 These
difficulties are not mentioned, and it is clear that for Carandini they simply
don’t count in comparison with the scenario his mental cinema projects. For
that, we must take the Tanaquil scene in conjunction with another story about
the window.

5. The goddess and the king.

In the thirty-sixth of his Roman Questions, Plutarch asks: ‘Why do they call one
of the gates the Window . . . and why is the so-called Chamber of Fortune
beside it?’ He gives two possible answers:65

noiEpov oxi SepPtog 6 Paoiksfig EUTvzEOTUTog yfivopEVog bo§av eo/e

ifi Tv/Il ovveivai cpornbofl btfi Ouitifiog itpog aurov: ij toOio uev pfiOog
eotiv, Eitfii 6s Tapruviov npioKov too fiaoiXEwg dnoOuvovTog f) yvyf'i
TavaKvXXlg gpcpparv ouaa tcotl paoikiKi5! bta 0vpidog jipoirutpaoa wig
noXiiaig evetv/e teal ovvejieioev outobEi^at paotkEa tov Xtpptov, eo/e
taurqv 6 Tonog ti1|v ^irwvvplav;
Is it because King Servius, the luckiest of mortals, was reputed to have inter
course with Fortuna, who visited him through a window? Or is this but a fable,
and is the true reason that when King Tarquinius Priscus died, his wife
Tanaquil, a sensible and a queenly woman, put her head out of a window and,
addressing the citizens, persuaded them to appoint Servius king, and thus the
place came to have this name?

56 Carandini 2004.46-7, 92-3 (fig. 8 and 9c); cf. Carandini 1997.ill.xx xxi anc x.
57 Carandini 2004.92, fig. 8; Santangeli Valenziani and Volpe 1996.348.
58 Carandini 2004.24, 38, 46 (‘la ricostruzione di Coarelli e nostra’); Coarelli s in
ricostniita invece leggennente pin a monte’ (38). • j,
59 Cicero De diiniiatione 1.101: .. ,<i luco l7cstae, qni <i Palati radice in nouain iiiani < <!«■•
60 Carandini 2004.36 (‘retrostante il lucus Vestae’), 89 (fig. 5). ‘bevond’
61 Solinus 1.24: supra suinmaiii nouain ilium. Though supra can sometimes mean y
(Wiseman 2004b.l70), it would be hard to understand it as ‘beyond and below. ,
62 Sumina nona via at Carandini 1997.ill.xx and xxxii; Carandini 2004.89 (hg. 5). 73 (tig. Q.
63 Plutarch Moralia 273b-c; I have slightly adapted the Loeb translation, w ic ren
(JVVElvai as ‘have converse with’. Plutarch uses the same verb elsewhere of emus a
Fortuna (see next note).
 THE HOUSE OF TARQUIN 287

Ovid and Plutarch (in another passage) tell us that the gate was called Porta
Fenestella, ‘Littlewindow Gate’, after the goddess’ amorous visits.64
Since we know from Solinus that Tarquinius Priscus lived by the Porta
Mugonia, it is an easy conjecture that that gate was also known as Porta
Fenestella because it was next to a little window in the upper floor of an
adjoining house.65 Now, the second phase of Carandini’s excavations has
revealed the remains of a sixth-century gate in opus quadratuin, which can be
reconstmcted as abutting on to the very house he identifies as that of
Tarquinius Priscus [figs 52—3, p. 278 above].66
The resulting scenario requires extended quotation (the translation and the
footnotes are mine):67

The piers of the archaic Porta Mugonia consist of two chambers (in one: of
which were found sacrificial remains ot the fifth century); that implies the
t
possibility’ of access to these rooms, ot ascent to the attic ot the arch, and of its
illumination by windows.68 In particular, the western pier abutted on to the
perimeter wall of the doiniis Regia, which probably predated it. In this abutment
an opening could have been arranged into the fabric of the house, through
which the upper storey of the gate, equipped with one or more windows (over
the passage or at the sides’), could have been put in direct communication with
the upper floor of the royal residence. We could then imagine Fortuna entering
at night by the fcnesuUa of the gate, getting into the house of Servius by an
opening and a passage or corridor, and ‘coming down’ to the ground floor.69
By taking an identical route in the opposite direction, Tanaquil could have
appeared at that same window of the house-gate, " to calm the crowd that filled
the Nova Via71 and put before the people Servius Tullius’ succession to the
throne of Tarquinius (who was dead, but supposedly only wounded).

64 Ovid Fasti 6.577-8 (node domain pania solita est intrare fenestra;] unde Fenestellae noniina porta
tenet); Plutarch Moralia 322f (cilCITE Kill OVVETVUI bOKEtv aurii) Ttjv Tu/.qv did TIVO?
Ot'pibog KiiTu|kii.vovoitv el; to diopattov, rjv vvv ipevEotekkuv itvXqv kuXovoiv).
65 Solinus 1.24 (n. 12 above); Wiseman 1998.28-9.
66 Carandini 2004.68 (attributed to Servius Tullius), 93 (fig. 9c). Argument must of course
remain provisional until the full excavation report is published.
67 Carandini 2004.68—9.
68 Carandini cites Trajan’s Column (see for instance Lepper and Frere 1988 plates vi, xxxv,
Lxxiii) for ‘numerosissime porte urbiche con finestrella o finestrelle soprastanti’. But if the
window ot a guardroom above was so regular a feature of city gates, why should it have
given its name to just this one?
69 See Plutarch Moralia 322f (n. 64 above) for Fortuna ‘coming down’ (from heaven) into the
house—not, of course, coming downstairs once she was inside.
70 An evasive phrase. Contrast Carandini 2004.60 on Tanaquil at the window of the house,
which is what Livy says.
71 Cf. Carandini 2004.60 and 61, insisting that the windows at the back ot the house fie finestre
del retro’) opened on to the Nova Via. Why was the crowd there, and not at the front doorr
 288 UNWRITTEN ROME

If Fortuna’s window of entry is the same as that at which Tanaquil

appeared, she must have been on the second floor, as Livy states. She is said to
‘come down’—we may imagine to the ground floor, where the royal bed was
in the tabliinim (so it is not a question of a simple bedroom on the second
floor).72 The king dies in his own bed; a new sovereign takes power, is
proclaimed king from the Regia /Porta Mugonia/Fenestella,73 and is thus legiti­
mated as sovereign by a ‘divine marriage’ which takes place in that same bed: as
it were, a death followed by fertilisation and resurrection—on the model of the
dying god.74

Now it becomes clear why Carandini’s hypothetical plan of the house of the
Tarquins was based so closely on that of the later ‘domus 3’, even down to the
mysterious corridor, and why the only change in the layout was to put the
chamber of the mater familias (what he calls the gineceo) on the other side.75 But
it doesn’t work: in the reconstruction of ‘domus 3’, the corridor is on the
ground floor only.76
Let’s remind ourselves of what the sources actually say. Tanaquil spoke to
the crowd from a window of the house, not from a gate; and far from
announcing a new king (for Servius’ succession was by no means guaranteed),
she was at pains to conceal the death of the old one.77 Fortuna came through a
window into the house, not a gate; and far from celebrating a sacred marriage
in the public part of the house, she was pursuing a shameful affair in a private
chamber.78

6. The eagle and snake portents.

Servius’ successor (and murderer) Tarquinius Superbus was wanted by various
portents that his reign was in danger. Dionysius reports that a pair o eag es
built their nest in a tall palm tree that stood in ‘the garden next to the pa ace ,
but were driven out by a flock of vultures.79 Livy tells us that a huge sna snake
'e

72 See n. 69 above. But Carandini insists (2004.67 n. 265): ‘Il aibiailnin non doveva trovarsi
11 cubiaduni
c1,^0 ’ 1"es,rdh al secondo piano della casa, visto die la Fortuna vi deve "scendere”.’
74 Cf r a- °V-i ^e-<!‘a^P0rla (‘palace-gate’) is another evasive phrase.
• aran ini -004.66-7 on the Athenian Icpdc; yupo? between Dionysus and the wife of
intro ™ a™' 'N°n VU dUnqUC “’’Pedimento a mettere in scena il mito erotico e
75 5 r°nizzante 1 ortuna a di Servio nd talamo di una dimora regale.. . ’
/5 See above, pp. 278-80.
,004n93nfig^9b)60 an<^ fig and 60’ so to° *n t*le ‘house of the Tarquins’ (Carandini
77 Ovirl JP' “8'’ ab°ve), 41.5 (inbet bona aniiiio esse ... regent ... iam ad se redisse, etc).
78 (ei? rd 6wpdTiov)'rt"">5 <""°rC^' 579 (""w P“del^ Plutarch Moralia 273b (OdXcqto;), 322f

79 belowSpIS’95 Hallcarnassus 4-631 (T^>v napaiceinEVov totg flaoiXtioiq tcflitov); see

 THE HOUSE OF TARQUIN 289

appeared out of a wooden column and drove those present ‘into the palace’ in
terror,80 A combination of the two accounts would imply a garden with a
portico, perhaps the same portico that featured in Dionysius’ version of the
miraculous fire round young Servius’ head. As we noted on item 2 above,
there seems to be no room for that in Carandini’s reconstniction.

7. The girl on a horse.

The elder Pliny begins his account of Roman equestrian statues with that of
Cloelia, the girl who led the escape of the Roman hostages from Porsenna’s
camp in the second year of the Republic.81 According to Livy, the statue of
Cloelia on a horse was ‘at the top of the Sacra Via’.82 Pliny’s account continues
as follows:

hanc priinam cum Coclitis publice dicatam crediderini ... nisi Cloeliae qnoqne Piso
traderet <i!> iis positam qui una opsides fuissent, redditis a Porsenna in honorem eius. e
diuerso Annins Fetialis eqiiestrem quae jiierit contra louis Statoris aedem in uestibido
Superbi dooms Faleriae Juisse Pnblicolae consulis Jiliae.
I should have held the view that her statue and that of [Horatius] Codes were
the first erected at public expense . . . were it not for the statement of Piso that
the statue of Cloelia also was erected by the persons who had been hostages
with her, when they were given back by Porsenna. as a mark of honour to her;
whereas on the other hand Annius Fetialis states that the equestrian [statue)
which stood opposite the temple of Jupiter Stator in the forecourt [iicsfifeiiluin]
was of Valeria, daughter of Publicola the consul.83

Since Plutarch reports that some said the statue on the Sacra Via was not
Cloelia but Valeria, it is clear that Annius Fetialis was not referring to a
different statue;84 there was only one ‘girl on a horse’, and authors dithered over
who she was.
What matters for our purposes is that Annius Fetialis said the statue was in
the uestibiiliim of the house of the Tarquins. For Carandini, that means it was

80 Livy 1.56.4: angitis ex colunma lignea elapsus cum terrorem (iiyaiiique in regiam fecissct.
81 Pliny Nur. Hist. 34.28; see above, pp. 180-1, on the Cloelia story.
82 Livy 2.13.11 (in sutiuna sacra uia Juit posita uirgo insidens equo); cl. Dionysius ot Halicarnassus
5.35.2 (EJU t!|C LEpug OdOV), Seneca Consolatio ad Marciam 16.2 (in SJ.r.1 uia celebemmo Lwk
Plutarch Publicola 19.5 (avciKELTai dE TT)V lEpav odov TOpevoilEVOi; tic naZutlOV .
Plutarch Moralia 250F (EJtl tqg 6do0 Tfjg lEpaj kEyogEvqc), Servius on Aettcid 8.646 .in
sacra uia).
83 Pliny Nat. Hist. 34.29. I have slightly adapted the Loeb translation, which renders rquesnrm
quaejiierit contra.. . as ‘an equestrian figure which once stood opposite... (there is no ’once
in the Latin).
84 Plutarch Publicola 19.5, Moralia 250f. Ziolkowski 2004.11-12 argues for two different
statues; rightly rejected by Carandini 2004.61 (’quasi che di statue equestri temminili potesse
pullulate la cirri agli esordi della Repubblica’).
 290 UNWRITTEN ROME

inside the house83—but in that case how could Annius identify it as ‘opposite
the temple of Jupiter Stator’? Ignoring that, Carandini makes it part of the
royal family’s domestic cult. He thinks it was really the statue of a goddess; at
first she is identified as Venus Equestris (mentioned very briefly by Servius),
and then, without argument, she becomes Fortuna Equestris, a wholly unat­
tested figure supposedly associated with the Fortuna who ‘ensnared’ Servius
Tullius.86 Naturally her statue stood in one part of the atrium, facing the
‘shrine-room’ of Fortuna in the other.87
All of that might be possible if the tiestibiilum were a room inside the house;
but as we saw on item 3 above, it is not. An alternative hypothesis is worth
considering, that the statue of the girl on a horse stood in a forecourt open to
the Sacra Via—which is what our evidence tells us.88

IV
Methodologically, it is important to distinguish three separate questions.

1. What was a sixth-century Roman king’s house really like?

2. What was it believed to be like when these stories were first created?
3. What was it believed to be like when our surviving sources were
writing?
Carandini normally assumes the identity of the first two questions and it is
indeed possible (though not of course the only possibility) that the stones
themselves dated back to the sixth century BC. And although he is a ways
prepared, where his argument requires it, to posit errors or misunderstan ings
in our literary evidence, more usually he assumes the essential identity o t e
second and third questions too.
His basic premise was made clear in La nascita di Roma. Oral tra ition,
gathered and handed down in the texts, is often the only living memory t rat
reaches us from within that distant past.’89 As with the first two questions, it is
indeed possible for later written texts to transmit authentic information a out a
much earlier world (this book could hardly have been written otherwise), ut

85 Carandini 2004.61 (‘nella stessa dimora regia’), 93 (fig. 9c).

Equestris'), 61 (‘una

tion of the cult of Fortuna Viscata on the Palatine (1 lutarch i < ra i<
Carandini 2004.63).
87 Carandini 2004.65; cf. 69 (‘tablinum/sacrarium’), 75 (‘sacrario/talamo ).
88 Open to the street: pp. 283-4 above. Sacra Via: n. 82 above. <ono
89 Carandini 1997.37 (in the context of‘residui delle piu antiche memone . .. qua
stati fissati per iscritto’): ‘La tradizione orale, raccolta e tramandata net testi, e spesso
vivente memoria che ci giunga dell’intemo di quell lontano passato.
 THE HOUSE OF TARQUIN 291

it is very far from automatic. In particular, when we are dealing not with anti­
quarian scholarship but with poets’ and historians’ narratives, there is little
likelihood of details being transmitted which were inconsistent with what their
readers took for granted. When Livy says that Tanaquil spoke to the crowd in
the Nova Via, ‘for the king lived next to the Jupiter Stator temple’, he is not
handing down a living memory, but appealing to his readers’ knowledge of the
city in their own time.
Of course neither Livy nor Dionysius nor Ovid nor Pliny nor Plutarch had
any idea of what ‘the house of the Tarquins’ was like in its sixth- or even filth­
century form. And it is not likely that their sources knew, either.90 According
to Carandini, the house was destroyed in the fire ot 210 BC, and splendidly
rebuilt some time after that.91 The new building in turn was evidently
destroyed under Augustus, but it may have helped to form the image of the
king’s palace in the minds of the authors on whom our literary tradition
depended.
The palace that they imagined had a colonnade (item 2 above), which perhaps
surrounded a garden (item 6); it had at least one window that opened on to the
Nova Via (item 4), evidently next to the ancient Palatine gate (item 5); and it
had a substantial forecourt (item 3), which opened on to the Sacra Via (item 7).
How many of those features belonged to the house as it was in the second
and first centuries bc? One obvious possibility is the forecourt (iicsnliiiliiiii), so
conspicuous a feature of the great houses of the late Republic.92 The king’s
forecourt was opposite the temple of Jupiter Stator,93 and in the story of the
assassins it was where his attendants were waiting;99 so it may be significant that
in the second century AD the college of lictors had its own cult of Jupiter
Stator, attested by dedication inscriptions evidently found near this very site.9’
However, we must be careful not to assume too easy a continuity. There
were several major fires in this area under Augustus, one of which destroyed
the statue of Cloelia ‘and the houses nearby’.90 Such fires may have resulted in
substantial changes to the urban landscape (Ovid, for instance, mentions a new

90 There may be one exception, if Dionysius of Halicarnassus 4.63.1 (on the eagles in Tarquin s
garden) came from an early Greek source: see below, pp. 295-6.
91 Carandini 2004.46 (cf. 59, 69, 73, ‘late-republican’): ‘splendidamente riedificata in opera
cementizia un certo tempo dopo 1’incendio del 210 a.C., dotata questa volta di un cripto-
portico (segno evidente di un edificio pubblico) sottostante un grande arrio con mosaico a
tessere nere.’
92 See p. 284 above.
93 Pliny Mir. Hist. 34.29: contra louis Statoris aedeui.
94 Livy 1.39.5 (unities apparitores regios), cf. 39.6 and 41.1 (lictors); see n. 44 above.
95 CIL 6.435, 31295a; CoareUi 1983.32-3.
96 Fires: Dio Cassius 54.24.2 (14 bc), 55.8.5-6 (7 bc), 55.26.4 (AD 6); Augustus Res gestae 20.3
(undated). Statue: Dionysius of Halicarnassus 5.35.2 (EAE'/ETO d’ £|tTplioEto; crept TOC
irkqoiov oiKtag yevoqEvq; fjtfaoioSai).
292 UNWRITTEN ROME

way through from the Nova Via to the Forum), and that no doubt accounts for
Carandini s discovery of a warehouse of Augustan date on the site of the house
of the Tarquins.97 But we know that Cloelia’s statue was replaced: Seneca
mentions it as shaming the effeminate young men of his time, in about AD 40.98
Then came the great fire of AD 64, which destroyed the temple of Jupiter
Stator, along with most of the centre of the city. The replacement statue must
have been lost as well, since Pliny in the 70s ad refers to it in the past tense.99
After Nero s fire the urban landscape was altered even more drastically,
with the Sacra Via straightened into a formal approach to the uestibiiluni of the
Golden House with its colossal statue of Nero.100 It’s hard to imagine any rival
uestibiiluni being allowed to remain. The temple ofjupiter Stator was rebuilt,101
but the lictors dedications to the god can no longer have been in the histori­
cally significant context of the king’s forecourt opposite the temple. And if the
girl on a horse statue was replaced again, as it probably was,102 its new setting
on the Sacra Via can hardly have suggested any connection with the house of
the Tarquins.
Andrea Carandini believes that the continuous reconstruction of archaic
features of the urban landscape represents a continuity of memory that could
last for seven centuries or more.103 In my opinion that is just wishful thinking.
Of course the statue meant that the Roman People still remembered Cloelia
(or was it Valeria?), but it was not ‘the living memory of oral tradition’. When
Annaeus Fionas in the second century ad told the story of Porsenna’s siege of
Rome, he referred to Horatius, Mucius and Cloelia as ‘those three prodigies
and marvels of the Roman name .. . who today would seem mythical if they
were not in the history books.’104
of new evidence for
The unique importance of excavation as the source
archaeologists. When the full
early Rome puts a particular responsibility on a
excavation report on ‘the house of the Tarquins’ is published in due course, it
will be interesting to see what the mute stones can be made to tell us.

97 Ovid Fasti 6.396: <]i<u nona Romano nunc ilia iunctaforo cst. Warehouse, n. 1 9 abo\
98 Seneca Consolatio ad Marciam 16.2 (for the date see Griftin 1976.396-7).
99 Pliny Nat. Hist. 34.29 (quaefaeril). Temple destroyed: Tacitus Annals 13.41. •
100 Coarelli 1999.227 (‘aspetto rettilineo e luonumentale’). Colossus: Mama ~ , •
/tacidis 2.1—4; Suetonius Nero 31.1 (ucstibidmn),
, Dio Cassius 66.15.1 (t'V rft lfp<?
101 It is listed in the fourth-century ‘Kegionary catalogues’ (Coarelli 1983.28).
102 . conspicintus (or was Servius
Servius on Acncid 8.646: statua cqncstris quant in sacra .iuia hodieque
just reproducing what his source said?).
103 Carandini 1997.519, 633 and 2000.89, on the ‘Romulean’ Palatine walls.
104 Fionas 1.10.3: qui nisi in annalibns foreut hodiefabnlae uiderentnr.
 CHAPTER SEVENTEEN

The Legend of Lucius Brutus

I Introduction

In the construction of Roman identity, the legend of L. lunius Brutus has a

significance rivalled only by that of Romulus himself. After the foundation of
the city, what could be more important in the collective memory than the
means by which freedom and constitutional government were attained? The
legends of Brutus and Romulus are similar in this respect also, that the narra­
tives of them in the sources that survive to our time are patently an amalgam of
many different versions, dating from many different chronological contexts.1
This chapter is an attempt to work out a possible ‘stratigraphy’ of the various
versions of the Brutus legend.
1 begin from the assumption, justified by a splendid recent article by Fausto
Zevi,2 that the story of L. Taiquinius Superhets’—his wealth, his Corinthian
ancestry, his rule in Rome, his expulsion from the city, his alliance in exile
with Aristodemus of Cumae, and his bequest to Aristodemus of the wealth
which the Romans had confiscated—is guaranteed as essentially historical by
the existence of an early and well-informed Greek historical source on the
career of Aristodemus,3 which will have narrated that material long before the
Roman historiographical tradition began.
However, there is no reason to suppose that the unknown Greek historian
took any interest in the internal history of Rome.4 For him, it would be
enough just to say that ’the Romans overthrew Tarquinius’; no doubt he iden­
tified the context as Tarquinius’ absence at the siege of Ardea, but we cannot
assume that he cared about the particular motives of the man or men who led
the revolution. The stories about Brutus, Lucretia, Collatinus, Publius Valerius
and the rest come from the Roman tradition, the modalities of which, before

1 I have tried to son out some of the different strata in the Romulus story in Wiseman 1983
and 1995.
2 Zevi 1995.
3 For the ‘Cronaca Cumana’, source of Dionysius of Halicarnassus’ narrative at 7.3—11, the
best treatment is still that of Alfoldi 1965.56—72.
4 On this fundamental point I differ from Mastrocinque 1988.35, 105, 235.
 294 UNWRITTEN ROME

the beginning of Roman historiography about 200 BC, are notoriously obscure
and controversial.5
A very recent treatment of Lucius Brutus calls him simply ‘a fictitious revo­
lutionary hero’.6 Essentially, I think, that judgement is undeniable: by the time
the Roman historiographical tradition took shape, the record of the real events
of the expulsion of the Tarquinii must have been distorted and romanticised
beyond recognition. If there was indeed a real Lucius Brutus in 507 bc (as we
know there was a real Poplios Valesios), the relationship between what he
really did and what Fabius Pictor attributed to him, after ten generations of oral
recycling and elaboration, is analogous to the relationship between the ‘real’
Agamemnon of the Bronze Age, or the ‘real’ King Arthur of the fifth century
AD, and the characters who bear those names in Homer and Geoffrey of
Monmouth.
But that is not to say that the traditions that have come down to us are
worthless. On the contrary, they are of very great interest, and it may even be
possible to extract some historical inferences from them. And the first thing to
notice is the ubiquity of portents, prophecies and oracles.

II Warnings for the King

The earliest surviving version of the story comes in a passage from Accius’ play
Brutus, quoted in Cicero’s De diuinatioue.7 Tarquinius dreams of a wonderful
flock of sheep; two brother rams are selected for him, and he sacrifices the
more splendid of the two. Then the other one attacks him, and he falls down
gravely wounded’. As he lies on the ground, he sees the sun travelling in a
new course towards the right (dextrorsum).
The flock must be the Roman people; since rams are the leaders of the
flock,8 the sacrifice must represent Tarquinius’ policy of getting rid of danger­
ously prominent citizens; the attack by the surviving brother of course refers to
Brutus? This is how the king’s prophets interpret the dream in Accius’ play:

prom uide ne quem tu esse hebeteui deputes aeque ac peens,

is sapieutia nmnituni pectus eqregie gerat
teque regno expellat; uaiu id quod de sole osteiituni est tibi,

5 See for instance Cornell 1986; Ungem-Stemberg 1986; Ungem-Stemberg 1 788, Tinip
1988; Wiseman 1994.1-22; Gabba 2000.11-23; Poucet 2000.27-75.
6 Weiwei 2000.48-57. , .
7 Cicero De diuinatioue 1.44—5 (Accius Brutus 651—72 Dangel = 17—38R). For detai e ana ysis
see Mastrocinque 1983.
8 So Artemidorus Oueirocritica 2.12, with a play on kpiOJ and KpeiWV.
9 For the murder of Brutus’ elder brother (and his lather), see Dionysius of Ha icanussus
4.68.1-2, Zonaras 7.11 (both of whom also report the murder of his father), Livy .3 . , '
uir. ill. 10.1.
r

THE LEGEND OF LUCIUS BRUTUS 295

populo commutationem rerum portenditfore

pcrpropinquam. haec bene uerruncent populo! nam quod ad dexteram
cepit cursum ab laeua signum praepotens, pidchemune
auguratum est rem Romanam publicum summum fore.
'Be careful, in case the man you think stupid like a sheep should have a heart
well fortified with wisdom, and drive you from your kingdom. For your
portent of the sun means that the people’s affairs will be changed, and very
soon. May these signs turn out well for the people! For because the mighty star
took its course to the right, the fairest augury has been given, that the Roman
state will be supreme.’

A warning for Tarquinius may be good news for Rome. And the details work
exactly, as they should: the ram makes Tarquinius fall (from power); he is not
killed, but badly hurt (exiled and impoverished).
In the historical tradition, the king receives different warnings. One of
them is reported by Dionysius:"’

... jTpoOt'GJtioavToc avTto tov daiiiovtov tqv gEkkovoav yEvi]OEO0ai

JTEpi tov oikov ovpqopav Ttokkoig pttv Kat aD.otq oitovoig, TEkEVTaicp
6f TiodE. aiETOi ovviovtec, Etc; tov irapaKEipEvov Tote PaoikEiot^ Kfprov
eapo; topa veotticiv ejtXu.ttov ejti Kopvqfj qoiviKo; tnpqkov. tovtidv be
TWV alETCOV ajTTf]V«5 ETt TOV? VEOTTOVg E/OVTCOV yVJTEIJ CtOpOOl
npooiTETao0EVTE5 Tqv te veotticiv btEqdpqoav Kai TOV? veottovc;
diTEKTEivav ical Tovg aiETOV^ irpooidvra; and Tqg voptqg apvTTOVTEg te
Kai naiovTEg T015 xapoot" and tov cpoivtKo; d.Tip.aoav.
The divine power had forewarned him of the future disaster to his house by
many signs, and particularly this final one. A pair of eagles came in the spring to
the garden by the palace and built their nest at the top of a tall palm tree. While
the young of these eagles were still unfledged, a flock of vultures flew to the
nest and destroyed it, killing the young birds. When the eagles returned from
hunting, the vultures drove them from the palm tree, tearing them and beating
them with their wings.

It we assume, as I think we must, that here too the details of the prophecy
accurately foretell what is to come, then the conclusion is inevitable that this
omen story was created tor a narrative quite different from the one all our
surviving sources take for granted. The eagles must be the king and his queen;
the vultures must be the Romans. But in this story the eagles’ young are still
helpless chicks, and the vultures kill them.

10 Dionysius of Halicarnassus 4.63.1—2; also in Zoharas 7.11.

 296 UNWRITTEN ROME

That is quite inconsistent with the developed story of the Tarquin clan, in
which Sextus, Arruns and Titus are adult warriors who share their father’s exile
and live to fight again.11 It must be a remnant from an older and simpler tradi­
tion, in which the tyrant was a comparatively young man, with young children
who were killed in the uprising. It may even come from the putative early
Greek source; certainly the total absence of any allusion to Brutus or to the
Romans’ reasons for expelling the king, coupled with the uncomplimentary
symbolism of the vultures (scavengers and carrion-eaters),12 strongly suggests a
non-Roman origin for the story.
Livy reports another portentum tembile: a snake appeared out of a wooden
column and caused terror and panic. Zonaras adds that the king was dining with
friends, and the huge snake ‘drove him and his guests out’.13 Since Livy refers to
fttga in regiam, I think we must imagine the scene as the same ‘garden by the
palace’ referred to by Dionysius in the story about the eagles, and the wooden
column as part of a peristyle. Out of the apparently lifeless wood there suddenly
appears a fierce and threatening force that drives the king out of his garden. The
allusion to Brutus is clear enough; but why is it the king and his friends who are
driven out, rather than the king and his family? Is this omen too a remnant from
an early version of the story, predating the Roman theme ot reges exacti?
Two further portents are alluded to by the elder Pliny, as the final item in
his discussion of dogs:14
Canetti locutum in prodigiis, quod equideni adnotauerini, accepiintis et serpcntein latrasse
cum pulsus cst regno Tarquinius.
So far as I have noted, a dog speaking and a snake barking have been reported
[only] among the prodigies when Tarquinius was expelled from his kingdom.
Evidently Pliny knew a narrative of the events which was independent of those
in our surviving sources. The snake he refers to must be the same one as in
Livy and Zonaras—but why did it bark like a dog? To find an answer, we must
move on to the next stage in the story: perturbed by these warning signs,
Tarquinius sends to Delphi to find out what the future holds.1’’

11 Arruns dies killing Brutus in the king’s first attempt to regain power by force (Livy 2.6.6—9,
Dionysius of Halicarnassus 5.15.1-2, Valerius Maximus 5.6.1, Plutarch Publicola 9.1-2);
Sextus and Titus fall at the battle of Lake Regillus (Dionysius of Halicarnessus 6.11.1-2,
12.5), though Livy has Sextus killed by his old subjects at Gabii (1.60.2).
12 Note that Plutarch (Romulus 9.6—7) feels it necessary to apologise for the vultures that inau­
gurated the foundation of Rome; cf. Jocelyn 1971.54-7.
13 Livy 1.56.4 (angitis ex columtta lignea elapsus cum terrorem Jugamqtte in regiamfecisset.. .), Zonaras
7.11 (otptg peyag CTicpavelg avrdv te teal rovg ovooirovg £§EpakE).
14 Pliny Nat. Hist. 8.153.
15 According to Dionysius of Halicarnassus (4.69.2), his reason for the consultation was a
plague afflicting children and pregnant women; Apollo gave instructions nt pro capitilms
capitilms supplicaretur, which Tarquinius interpreted as human sacrifice, commuted by Brutus
as consul (Macrobius Saturnalia 1.7.34-5, on which see Mastrocinque 1988.37-41).
 THE LEGEND OF LUCIUS BRUTUS 297

III The Oracles

In Zonaras’ narrative (but not in those of Livy or Dionysius), we come across
the talking dog:16

bta rot Taura Eg Aekcpoug Titov te koi Appovvra roue viovg EJtEjtipe.
toO be Airokkaivog xpijoavrog tote rfjg ap/qg EKnEOEto9ai avrov ote
kuidv avBpwjtivr) cpwvp xpTjoatTO, ayaSaig eXjuoiv pwpqro, pq olq0Eig
jiote YEVEO0ai rd piavTEupa.
Because of these portents he sent his sons Titus and Amins to Delphi; but when
Apollo declared that he would be driven from his realm only when a dog used
human speech, he was borne up with confidence, believing that the oracle
could never be fulfilled.
The explanation is clear enough. Brutus is pretending to be an idiot, unable to
speak,1' The king’s sons have taken him with them to Delphi as a source of
mockery and amusement; in short, they treat him like a dog.18 When this dog
speaks, Tarquinius’ reign will be over.
That, I assume, is why the snake in Pliny barks: it represents Bnitus himself.
It lurked within the wood of the column, just as the rod of gold was hidden
within the wooden staff that Brutus offered to Apollo.19 As for the talking dog,
either Pliny misinterpreted his source in making it a real portent (one of the
warning signs), rather than the unsuspected fulfilment of the oracle, or else the
theme was dealt v i.h differently by different authors. For it is clear that there
were many versions of the story of how the supposed idiot overthrew the king,
and that the narratives in our surviving texts are a palimpsest of various
elements.
Naturally, all our sources focus on the famous story of the king’s sons’ ques­
tion about the succession, and how Apollo’s answer is made to apply to Brutus
himself. But what exactly was Apollo’s answer? There is a revealing difference
between the Latin and the Greek authorities. In Livy, the god says that
whoever shall first kiss his mother will have imperium summimi at Rome; in
Valerius Maximus and in pseudo-Aurelius Victor, it is simimam potestatem.20
Those, of course, are phrases careftilly chosen to be applicable also to the
shared authority of a consul. Servius is less careful: he makes Apollo simply

16 Zonaras 7.11; cf. Livy 1.56.10 (perfectis paths maildan's), Dionysius of Halicarnassus 4.69.3 (0>5
6e Tapt'YEVi]Oquav eiri rd piavTEiov oi veuvlckoi Kai roue xpqopoug EXajtov iratp
aiv EirEpcfOpoav).
17 The adjective brutus is often used of dumb creatures: e.g. Pacuvius 176R. (Nonius 109L: el
obnoximn esse ant bniluni atir elmguem pines), Seneca Epistles 121.24 (tadtis qiioipie et bnitis).
18 Livy 1.56.9 (ludibrium), Dionysius of Halicarnassus 4.69.1 (tva '/El.WTO ItapE/J]); Dio
2.11.10, Zonaras 7.11 (tboitEp Tl a0uppa).
19 Livy 1.56.9, Dionysius of Halicarnassus 4.69.3, Valerius Maximus 7.3.2, De uir. ill. 10.2.
20 Livy 1.56.10, Valerius Maximus 7.3.2, De uir. ill. 10.3.
 298 UNWRITTEN ROME

promise power (imperium), and comments that Brutus did indeed ‘obtain the
power’, as if he were truly succeeding the king.21 That is what the Greek
sources say too (Apollo promises TT|V 'Pa>p.ai(ov dpXT]V or TO icpctTO?),22 and
since the oracle would have been in Greek it is not unreasonable to privilege
their version.
‘The rule’ and ‘the power’ are not the same as ‘the kingship’, which is how
the oracle can apply to Brutus. But equally they do not imply a shared
authority. Apollo tells Brutus that he can be the ruler, not the co-ruler, of
Rome. It seems to me that this crucial element in the story must predate the
institution of the consulship.
When supreme authority is shared between two men, one might expect a
foundation myth to reflect that duality. I have argued elsewhere that that is the
probable origin of the story of Remus and Romulus, and that what made the
myth necessary was the introduction of the consulship as a power-sharing
device in 367 bc.23 Not everyone accepts that suggestion for the origin of the
twins story, just as not everyone believes that the consulship was a fourth­
century innovation.24 But the normal assumption that the consulship dates
right back to the origins of the Republic, and that Brutus was always imagined
as sharing power with Collatinus, simply creates more trouble than it’s worth.
The canonical story of the first year of the Republic, as it appears in Livy,
Dionysius and Plutarch, is patently an artificial combination of rival claims to
have inaugurated the Republic, and can only be made to fit the assumption of
an original consulship by the most implausible succession of events—an exile,
two deaths in office, and three elections ofsuffect consuls.25 The identity of the
‘first two consuls’ as Brutus and Collatinus depends on the forced amalgama­
tion of two originally separate stories, those of Brutus the idiot and the rape of
Lucretia—for why should Brutus be present at Lucretia’s suicide.' In any case,
some authorities claimed that the first consuls were Brutus and M. Horatius, or
Brutus and P. Valerius.26 Three hundred years of oral story-telling, followed by
two hundred years of creative historiography, have produced a chaotically
overcrowded narrative.

21 Servius on Aeueid 3.96: uam Brutus etjilii Tarquiiiii cum oraculum Delphici Apollinis peterent,
respoiisum est eius imperium fore qui primus matrem reuersus oscularctur. quod solus Brutus agiioscens
de naui egressus simulaiis casum oscidatus est terrain; unde et potitus imperio est.
22 Respectively Dionysius of Halicarnassus 4.69.3 and Dio 2.11.12 (= Zonaras 7.11); in both
narratives the king’s sons assume the question is about fiaotkfia.
23 Wiseman 1995.103-10.
24 I am glad to see that the latter view is shared by Weiwei 2000.49—50.
25 I have tried briefly to account for the absurdities in ‘Roman Republic, Year One (pp. 306—
12 below).
26 Polybius 3.22.1 (M. Horatius), Pliny Nat. Hist. 36.112 (P. Valerius); the first chapter of
Plutarch’s Publicola (1.3-4) seems to presuppose a rejected tradition of Valerius as the first
consul.
 THE LEGEND OF LUCIUS BRUTUS 299

Much of the chaos can be cleared up by the simple expedient of forgetting

the consulship. Despite constant repetition to the contrary, there is no good
reason to believe in an authentic consular list going back even to the fifth
century, much less the sixth.27 On the contrary, Livy gives us good evidence—
which he himself was unable to exploit—that in the early Republic the chief
magistrate was called praetor niaximus, a title which surely rules out any notion
of equally shared power.21* If that office, or something like it, was what Apollo
promised to Brutus, the wording of the oracle would at least make sense.
So let us assume, for the sake of argument, that the earliest fonn of the
Romans’ story of their liberation presupposed a single supreme magistracy, to
which Brutus was elected—thus succeeding the king, as the oracle story
requires. Let us further assume that by the second half of the fourth century bc,
when the power-sharing consulship was taken for granted as the essential and
inevitable magistracy of the Republic, the story was elaborated to take account
of the new conditions, and a colleague was found to share office with the
Liberator. No doubt various names were proposed, but we know which one
eventually became canonical—that of tragic Lucretia’s husband, L. Tarquinius
Collatinus. I think the context of his incorporation into the story can be dated
quite securely.

IV The Man from Collatia

The rape of a virtuous lady by the king’s son is a perfectly intelligible motive
for the overthrow of the monarchy. But why must her husband be a
Tarquinius himself? It complicates the story, and makes necessary the inven­
tion ot an entire branch of the royal family which had remained miraculously
untouched by both wealth and arrogance.29 The reason is, of course, that
Collatinus’ name is needed to justify getting rid of him:3"

consults enim alterius, cum aliud nihil offenderet, nomen iiiuisiiin dnitati fait, niiniiini
Tarquinios regno adsuesse; initium <1 Prisco factum; regnasse dein Ser. Tiilliimi; tie

27 See for instance Wiseman 1996.313-15, in a review of Cornell 1995. The absence of fifth­
century records is attested by ‘Clodius’ (Claudius Quadrigarius?) in Plutarch Numa 1.1; he
gave the wrong explanation, but the fact that he had to explain it shows that the phenom­
enon itself was real enough.
28 Livy 7.3.5: Lex uetusta est, priscis litteris uerbisque scripta, ut qui praetor maximus sit idibus Septem-
bribus clauum pangat; Jixa fait dextro lateri aedis louis optinii maximi, ex qua parte Minemae
templum est. At 7.3.7 he cites diligens talium nionumentonim auctor Cindus.
29 For Demaratus’ supposed younger son ‘Egerius’ (from egens), see Livy 1.34.2-3, 1.57.6,
Dionysius of Halicarnassus 3.50.3, 4.64.2-3. The Latin name (cf. Festus 128L for Manilas
Egerius of Aricia) is enough to show that he is a secondary element in the story.
30 Livy 2.2.3; cf. also Cicero De oratore 3.40 {cum autem consilium hoc prindpes cepissent, cogna-
tionem Superbi nomenque Tarquiniontm et memoriam regni esse tollendam...).
 300 UNWRITTEN ROME

iiiteruallo quidem facto oblitum tamquam alieni regni, Superbum Tarquinium iielut
hereditatem gentis scelere ac tti repetisse; pttlso Superbo penes Collatinum imperium esse,
nescire Tarquiuios priuatos uiuere; non placere nomen, pericttlostim libertati esse.
The citizens disliked the name of one of the consuls, even though he had done
nothing else wrong. ‘The Tarquinii have got too used to royal power! Priscus
started it; then Servius Tullius reigned, but even in that interval of extraneous
rule it wasn’t forgotten, and Tarquinius Superbus claimed it by crime and
violence as if it was a family inheritance. Superbus has been expelled, but now
Collatinus has power! The Tarquinii don’t know how to live as private citizens.
The name is hateful, a danger to liberty.’

So Brutus’ consular colleague goes into voluntary exile, to make way for P.
Valerius. The parallel with Athenian ostracism has often been noted, and
Attilio Mastrocinque has convincingly included it as one of a whole sequence
of alleged events in the Romans’ story of their liberation which are evidently
based on equivalent motifs in Athenian history.31
One particular detail, recently pointed out by Alan Griffiths,32 reveals that
the parallel in this case is taken not from Herodotus but from a fourth-century
Atthidographer, possibly Androtion, used by the author of the Aristotelian
Constitution ofAthens:33
Kai irpokog (boTpaKioSr) twv ^keivov ovyyEvwv "Inirap/oc Xappou
KoXXuteij5, Si’ bv Kai pakiora t6v vbpov eOijkev 6 KXci00EVi]g,
E^sXaoat pouXogEvog aikov. ol yap A0t)vaiot roug twv TUpawwv
tpiXoug, boot pi] ouvE^apapidvoiEV ev rate; xapa/aig, eiiov oikeiv ti'jv
jioXiv, xpwpEvoi rfi ElcoOupx tou bqpou jrpaoTqTC d>v ijyfptbv Kai
jtpooTdTt]g rjv Tjinap/og.
The first of [the tyrant’s] relatives to be ostracized was Hipparchos son of
Charmos, of the Collutos deme. It was because of him in particular that Cleis-
thenes passed the law, wanting to drive him out of the city. For the Athenians,
showing the people’s customary forebearance, allowed the tyrant s friends to
live in the city, with the exception of those who had been involved in his
crimes during the disturbances; of these the leader and champion was Hippar­
chos.

Late in the fourth century BC, at the time of the Samnite wars, the Romans
put up a statue of Alcibiades in the Comitium.34 For the ambitious and ener­
getic city-state that dominated Latium and was now extending its control over

31 Mastrocinque 1988.32—5; p. 34 for Collatinus.

32 Griffiths 1998.
33 Athcnaiou politeia 22.4.
34 Pliny Nat. Hist. 34.26, Plutarch Numa 8.10: Pythian Apollo had told them to honour ‘the
bravest and the wisest of the Greeks’ (they chose Pythagoras as the wisest).
 THE LEGEND OF LUCIUS BRUTUS 301

all of central Italy, the history of fifth-century Athens must have been an
example and an inspiration. I think Emilio Gabba was right to identify this
particular period as the time when Rome’s traditions about her own past were
being somehow organised into the more or less coherent story' that was later
transmitted by Fabius Pictor and his successors;3’ and if that is the case, then the
phenomenon identified by Mastrocinque, of episodes evidently constmcted or
elaborated on the model of Athenian history, is easily explicable.
For the particular case of the foundation-myth of the Republic, 1 suggest
that the old story of heroic Brutus succeeding the king was reshaped to fit the
now unquestioned republican model of two equal ruling magistrates.
Lucretia’s husband would make a fitting colleague for Bnitus, but since he had
no traditional role to play in the liberation story, he would have to be got rid
of again (like Remus or T. Tatius). The Athenian model offered a solution.
Make Lucretia’s husband a relative of the tyrant; put their married home in
Collatia (at whatever cost to the coherence of her story);36 and the caique of
Collutos and Collatia provides an honourable exit for Bnitus’ necessary' but
inconvenient consular colleague.

V lunius and the Herons

During the last quauer of the fourth cenmry BC, one of the most prominent
Roman commanders was C. lunius Bubulcus. He is so named seven times in Livy:
9.20.7-9 Consul 317, campaigns in Apulia
9.28.2 Consul 11313
9.29.3 Dictator 312
9.30.1 Consul 111 311, campaigns in Samnium (9.31)
9.38.15 Magister equitum 310, campaigns in Samnium (9.40.8-9)
9.43.25 Censor 307
10.1.8 Dictator II 302

Diodonis names him three times just as C. lunius; but Valerius Maximus,
referring to his censorship, calls him ‘C. lunius Brutus Bubulcus’, and in the
Capitoline consular and triumphal fhsti he is consistently ‘C. lunius C.f.C.n.
Bubulcus Brutus’.37 Similarly, his elder relative lunius Scaeva—so named in

35 Gabba 2000.20—2; cf. Wiseman 1994.14—16, suggesting dramatic performance as one of the
means.
36 The morning after the rape, Lucretia must either travel to Rome herself (Dionysius of Hali­
carnassus 4.66.1), or else summon her menfolk from Rome and Ardea (Livy 1.58.5), thus
awkwardly putting the dramatic exposure of her body and the resulting popular indignation
at Collatia rather than Rome (Livy 1.59.3—5).
37 Diodorus Siculus 19.17.1, 19.77.1, 20.3.1; Valerius Maximus 2.9.2; Fasti coitsitlares for 317,
313, 312, 311, 309, 307 bc (also his son at 277 BC, cf. Livy 27.6.8 C. lunius Bubulcus’);
Fasti triumphales for 311 BC.
 302 UNWRITTEN ROME

Festus as one of the tresuiri responsible for the colony at Saticula—appears in

Livy as ‘lunius Brutus’ or ‘lunius Brutus Scaeva’.38
The natural inference is that these distinguished plebeians, the first of their
family to achieve the consulship, were originally nicknamed as ‘clumsy
(scaeuus) and ‘yokel’ (bnbiilats), insults which they or their descendants adapted
to their honour by adding the name of the hero who pretended to be stupid.
And if, as seems likely, the old story of Brutus the liberator was being elabo­
rated at just this time, it may be that a nomen gentilicium was only now for the
first time attributed to him.
L. lunius Brutus, as we may now call him, took over the army at Ardea as
soon as Tarquinius had left.39 The end of the siege, and the subsequent truce
between Ardea and Rome, are presumably real historical events. The impor­
tance of Ardea in the sixth and fifth centuries I3C, and her status as an equal of
Rome, can be inferred from the tradition reported by the Greek historian
Xenagoras, that Rhomos, Anteias and Ardeias were brothers, the sons of
Odysseus and Circe.40 It may also account for a puzzling quotation in Festus
from the fourth-century Sicilian historian Alcimus:41

Alcimus ait, Tyrrlienia Aeneae natuin fdium Romulum Jitisse, atque co ortam Albani
Acneac neptem, cuiusJilius Rhodius coudidcrit urban Romain.
Alcimus says that Romulus was the son of Aeneas’ wife Tyrrhenia, and from
Romulus was bom Aeneas’ granddaughter Alba, whose son, called Rhodius,
founded Rome.
Editors sometimes emend ‘Rhodius’ to ‘Rhomus , but the principle of the
dijjicilior lectio counts against that easy solution. An alternative explanation is
available if we remember that ardea means ‘heron , Eptp&lO^ oi pqi I 5 in
Greek.42 So Rhodius can be the eponym of Ardea, which evidently claimed at
one time to be the mother-city of Rome. ,
In the tradition exploited by Virgil, Ardea is the city of Tiirnus at le
Daunus.43 Daunus is more familiar as the eponymous king of Daunia in nort ern
Apulia, whose son-in-law Diomedes founded Arpi (Argyrippa) in that territory.

38 Festus 458L; Livy 8.12.13 (oirtg. eq. 339), 8.29.2 (cos. 325).
39 Livy 1.60.1-2, Dionysius of Halicarnassus 5.1.2.
40 Xenagoras FGrH 240 F29 (Dionysius of Halicarnassus 1.72.5).
41 Alcimus FGrH 560 F4 (Festus 326-8L).
42 The three-syllable spelling is attested by the grammarian Herodian (2.924 Lentz), who cites
an example from Hipponax (fr. 16 West). For ordco, see Servius on Aeneid 7.412: iiam Ardea
quasi ardua dicta est, id est iiiaqna et nobilis, licet Hygimis in Italicis urbibus [fr. 11P] ab augurio auis
dictum uelit. illud iiamque Ouidii in Metamorphoseos [14.573—80] fabidosum est, incensam ab
Hanuibale Ardeam in hanc auem esse comiersam. (Servius misinterprets Ovid, whose barbarus is
Aeneas, not Hannibal.)
43 Virgil Aeneid 10.688, 12.22; cf. 8.146, 10.616, 12.90 etc.
44 Daunos and Diomedes: Ovid Fasti 4.76, Antoninus Liberalis Metamorphoses 37. Arpi:
Lycophron Alexandra 592, Strabo 6.3.9, Justin 22.2.10, Solinus 2.10, Servius on Aeneid 8.9.
 THE LEGEND OF LUCIUS BRUTUS 303

But the idea of Daunii in Latium goes back at least as far as Lycophron;45 more­
over, there were traditions which brought Diomedes himself to Latium, and
even made him instrumental in the foundation of Rome.46
A Homeric scholiast’s report on the death of Diomedes includes a version
which must belong somehow in this context:47

.. .o0ev aviov rpvydvTa cpao'iv fjicEiv el; ’I|3i]piav Kam, w; pev tive;,
6okocpovr]0f]vai uito Aavvov tov paoikEtog, to; <5e evioi, airokEoSai
VJid Tovviov tov Aavvov iraibb; ev kvvtiyeoIoi;* o9ev avrov ptv
djtoOEtooEV AOqva, tov; 6e Etaipov; el; Eptpbiov; gETEpakEV.
They say he fled from there [i.e. from Argos, where his wife and her lover had
plotted to kill him] and came to Iberia, where according to some authorities he
was treacherously murdered by king Daunus, and according to others he was
killed by Daunus’ son lunius while out hunting. As a result Athena deified him
and turned his companions into herons.

‘Iberia’ is clearly a mistake: this story must be about Apulian Daunia.48 The first
version of Diomedes’ death, which may go back as far as Mimnennus,49 makes
Daunus a treacherous enemy. The second, however, seems to presuppose the
friendly Daunus who gave Diomedes his daughter. The death in the hunt was
presumably accidental, a tragedy requiring Athena’s intervention. It was also a
conveniently blameless way of removing Diomedes and allowing his brother-
in-law lunius to be Daunus’ heir—thus justifying Roman control over
Daunia, as expressed in the foundation of the colony at Luceria in 314 I)C.5U C.
lunius Bubulcus campaigned in Apulia as consul in 317, and held his second
consulship in 313; he could well have been one of the tresuiri coloniae
deducendae. In any case, his involvement in the subjection of Daunia must
surely be the origin of the story of lunius son of Daunus.
The herons, however, may be relevant to ‘Daunian’ Ardea. The name of
the new colony reflects that of Luceras, a legendary king of Ardea who assisted
Romulus in his war against T. Tatius’ Sabines; the Romulean tribe of'Luceres’

45 Lycophron .-fle.v.mdru 1253-5, on Aeneas: KtioEl be yiopav ev roitot; BopEiyovwv |

iinip Aarivov; Aavviov; T’ U)KlogEvr)V. Cf. Polybius 3.91.5 for Daunii in Campania,
and in general Briquel 1974. 8-19.
46 Diomedes in Latium: Solinus 2.14 (giving the Palladion to Aeneas), Appian Civil H'ars 2.20
(founding Lanuvium). Diomedes sends Rhomos son of Emathion to found Rome: Plutarch
Romulus 2.1.
47 Schol. Iliad 5.412 (b), Erbse 2.64—5; I owe my knowledge of this important passage to Curti
1993. For the stories of Diomedes in Italy, see Malkin 1998.234-57.
48 The hunt may be an allusion to the etymology of Canusium, another of Diomedes’ founda­
tions (Servius on Aeneid 11.246, cf. Horace Satires 1.5.91-2).
49 Cited by the scholiast on Lycophron Alexandra 610, though it is not clear how much of the
story was in Mimnemius.
50 Livy 9.26.1—5, Diodorus Siculus 19.72.8-9. Ancient dedications in the temple of Athena at
Luceria were taken as evidence for the dominion of Diomedes (Strabo 6.3.9, 284).
 304 UNWRITTEN ROME

was supposedly named after him.51 A likely explanation is that Ardea was
involved in some way in the Roman colonisation of Daunia, and that her
Daunian mythology was invented to reflect that.52
If so, then the presence at Ardea of lunius Bubulcus’ alleged ancestor
Brutus is likely to be another feature of the elaboration of the legend in the late
fourth century bc. Unlike Collatinus, however, it was an element which lost
its significance in later times. Not all the levels of the palimpsest remained
equally important.

VI Conclusion
Our sources on the revolution that brought freedom to Rome date from the
late second century BC (Accius) to the twelfth century AD (Zonaras), between
four and seventeen centuries after the event. Close analysis of the narratives
they offer enables us to identify (hypothetically, of course) some of the
elements out of which they were composed, and thus some of the stages by
which the story was built up over the centuries.
First, we can infer an early account (probably in a Greek historian) where
the tyrant’s young children were killed in the revolt. 1 he author s sympathies
were evidently with Tarquinius, the eagle driven out by vultures, and naturally
there is no sign of any heroic liberator.
Second, there may have been a version where the champion of freedom did
indeed appear, as a monstrous snake, but where only the king s companions
shared his exile. That looks like the earliest trace of the patriotic Roman story.
The version in which the snake barked like a dog perhaps belongs to a later
stage, featuring the embassy to Delphi, the king s sons and their i lot
companion, and Apollo’s oracle about the dog that speaks. Grown-up sons
imply a whole royal family to be driven from power, as implied by t e ater
Roman date-formula post reges exactos (not regent exactnni).
Whether that stage also included the sons’ enquiry about the succession is
not clear. But Apollo’s reply on that subject clearly' implies a post-expu sion
regime in which Brutus alone would have power, and it therefore pre-dates
what I believe to be the introduction of the consulship in 367 BC.
The creation of the consulship evidently made necessary the adjustment of
the liberation story to accommodate the idea of shared power. 1 hat seems to
have happened at a time when the Romans were elaborating their own history
on the model of that of Athens, and the tale of Tarquinius Collatinus as Brutus
temporary colleague is best explained as a caique on the ostracism of

51 Festus (Paulus) 106L. Note also ‘Leukaria’, daughter of Latinus and mother of Khomus (I iony-
sius of Halicarnassus 1.72.6, ef. Plutarch Romulus 2.1), though she may be a caique on AIM.
52 So Curti 1993: perhaps the colonists of Luceria were from Ardea?
 THE LEGEND OF LUCIUS BRUTUS 305

Hipparchus of Coliutos. This must be the point at which the story of the rape
of Lucretia—by a third son of Tarquin evidently not involved in the question
of the succession’3—is linked to the Brutus story by the expedient of having
the supposed idiot present at her suicide.
The consulship was a power-sharing device, with one patrician and one
plebeian in office each year. Since Tarquinius Collatinus must have been
thought of as a patrician, the expedient of retrojecting the new system to the
origins of the Republic required the Liberator himself to have been in some
sense a plebeian. It is clear from the nomenclature of the first consular lunii,
Scaeva and Bubulcus, that their plebeian family claimed him in the second half
ot the fourth century, and that ever afterwards Brutus was thought of as the
ancestor of the plebeian gens Itinia.
lunius Bubulcus’ involvement in the conquest and settlement of Daunia
evidently gave them an even earlier ancestor, lunius the brother-in-law of
Diomedes; it may also have given rise to an element in the liberation story, as
Bnitus takes over the Roman army at Ardea and makes peace with the tyrant’s
victims.
If the argument presented here is valid, it implies a fertile and creative
narrative tradition existing long before the introduction of literary historiog­
raphy at Rome. And I think that is what we should expect, in a pre-literary
culture making sense of changing conditions over a period of two centuries.
‘Oral tradition’ does not 'hand down the memory of events’; it elaborates,
recycles, omits, invents, creates a succession of stories for a succession of audi­
ences with ever-changing priorities.
Moreover, the process goes on even after the introduction of written
history. Historians always liked to differ from their predecessors, either with
new material or by presenting a different slant on traditional narratives?4 In the
case of the Bnitus story, the assassination of Caesar caused a serious reappraisal
on the part of some writers. Either Marcus Bnitus the tyrannicide was no true
descendant of the Liberator,35 or the reputation of Lucius Brutus himself must
be rethought in more critical tenns (as in Virgil’s Aeneid, where the arrogance
of Tarquinius himself is attributed to him).56 Like the story of Remus?7 so the
story of Bnitus took on a new aspect in the Rome ot Augustus.
Identity, the constniction of a people’s idea ot itself, is always changing; and
myth, the memory by which that identity is established, is always changing too.

53 Note that in Dionysius of Halicarnassus Sextus is the eldest son, and offers to make Lucretia
queen (4.55.1, 4.65.2); contrast Livy 1.53.5.
54 See for instance Wiseman 1998.75—89.
55 See Plutarch Brutus 1.2—3.
56 Virgil Aeneid 6.817—23 (animamqtte superbain, 817); cf. Dionysius of Halicarnassus 5.8.1,
Plutarch Publicola 6.4 (ambivalent judgements of Brutus character).
57 Cf. Wiseman 1995.144-50.
 CHAPTER EIGHTEEN

Roman Republic, Year One

I
The Romans knew that they had once been ruled by kings, and they believed,
perhaps rightly, that the fall of the monarchy had taken place at what we would
call the end of the sixth century BC. The texts that tell us this—Livy, Diony­
sius, Plutarch, etc—all depend on a historical tradition that can be traced back
as far as the second half of the third century BC, when the Roman literary
genres of historical drama, historical epic, and prose historiography began.
Before that, we do not know how the Romans conceived or recorded the
memory of their own past.
There is some exiguous contemporary evidence. The word ‘king’ (rex)
appears on two inscriptions found at deep levels in Giacomo Boni’s excavation
of the Forum in 1899-1903; both may well be of sixth-century date.1 Another
inscription—which does not survive, but was seen and copied by a Roman
antiquarian in the first century BC—evidently referred to the ’chief magistrate’
(praetor inaxitiins) of the post-monarchic regime.2
Early evidence of a different kind is provided by the Roman ritual calendar,
which probably dates back at least to the fifth century bc.2j It marks 24
February as ‘Flight of the King’ (Regifugitim), the last item before the beginning
of the new year on 1 March.3 In the third century BC, and perhaps before, the
newly-elected consuls entered office on 15 March. Does the sequence
commemorate a real event, the flight of the king followed by the election of
republican magistrates and their entry into office after a nineteen-day vacuum
of power? Or was the ‘event’ a story to explain the sequence?
Of one thing we can be reasonably sure. Whenever, and however, the
power of the king was ended, there will have been stories to account for it,
foundation legends for the infant Republic.

1 A fragment of a bucchero bowl from the Regia, illustrated in Momigliano 1989.76, fig. 25;
and the ‘Lapis Niger’ stele at the Volcanal (C7L I2 1 = ILLR.P3).
2 L. Cincius, quoted in Livy 7.3.5-7 [see pp. 10-11 above].
2a |So I believed in 1998; I now think the fourth century BC is no less likely.]
3 See Ovid Fasti 2.685, 851; Festus (Paulus) 347L; Plutarch Quaestiones Ronianae 63 (= .Moralia
279d); Ausonius 7.24.13; Polemius Silvius in C1L I2 p. 259 = Degrassi 1963.265.
 I
ROMAN REPUBLIC, YEAR ONE 307

/. The Story of Lucius Junius

Lucius Junius was the youngest son of a worthy citizen foully murdered by the
king’s agents in the days of Tarquin the Proud. The other sons were killed
too,Jj but Lucius pretended to be an idiot, and was contemptuously spared by
the assassins. He had to keep up the pretence as he grew up, and became
known to all as Stupid Lucius. The king’s sons, who were as arrogant as their
father, used Stupid Lucius as a butt for their cruel mockery.
One day, an ominous portent was reported at the palace. The king decided
to consult the oracle at Delphi, and sent his sons on the mission. They took
Stupid Lucius with them for amusement on the long journey. They brought
rich gifts for the god, and sneered at the simple staff of cornel-wood which was
Stupid Lucius’ offering—but Apollo could see, as they could not, that within
the hollow wood was a rod of pure gold.
After the Pythian priestess had given them the instructions they had come
for, the royal princes asked the question that was really on their minds. Which
of them would succeed their father as king? And this was the god’s reply:
’Whichever of you shall first kiss his mother, he shall have supreme authority
in Rome.’ They didn’t notice that Stupid Lucius was in the temple too.
As the king’s sons came out, each of them scheming how to get home first,
Stupid Lucius tripped and fell flat on his face. As usual, they mocked his clum­
siness. They didn’t see that he had kissed Mother Earth.
When they returned to Rome, they found the citizens furious with indig­
nation at Tarquin’s cruelty and injustice. All that was needed was a leader.
Confident of the gods’ support, Lucius Junius at last threw off his disguise and
led the Roman People in righteous revolt against their tyrant. Tarquin and his
sons were banished from Rome for ever, and the free citizens resolved to elect
a chief magistrate of their own. The choice fell naturally on Stupid Lucius
Junius, as he was ever afterwards known. For the nickname that had been an
insult was now a badge of pride, and his descendants use it still.

2. The Story of Lucretia

Lucretia was a beautiful and virtuous lady, daughter of Spurius Lucretius and
devoted wife of Lucius Collatinus, in the days of Tarquin the Proud. Her
husband was an excellent citizen; it is true he was related to the Tarquins, but
he had none of their power, none of their wealth, and none of their vices. He
and Lucretia loved each other and were faithful to each other; and when
Collatinus had to go away to the wars, Lucretia stayed at home and prayed tor
his safe return.

3a [My mistake: the sources give Lucius only one brother. So he is the younger son. not the
youngest.]
 308 UNWRITTEN ROME

But Lucretia was beautiful, and Sextus, the king’s son, lusted after her. Late
one evening he came to his kinsman’s house. Lucretia was puzzled (did he not
know her husband was away?), but she received him politely and offered him
the hospitality of the house, as she was in duty bound to do. That night he
came to her bedroom. He offered his love, but she rejected it with horror. He
demanded her obedience, but she proudly refused. He threatened her with his
sword, but she was not intimidated.
‘Very well,’ said Sextus. ‘I will take you by force and kill you afterwards.
Then I will kill a slave and put his body in your bed, and I shall tell your
husband and your father that I slew the adulteress and her paramour.’ Then,
and only then, Lucretia yielded.
Next morning Sextus rode away triumphant, and Lucretia sent urgently for
Collatinus and her father. When they heard what had happened, of course they
told her she was not to blame, but Lucretia would not be consoled. ‘I cannot
live with this shame,’ she said. ‘But give me vengeance!’ With that, she drew a
dagger from her dress, and drove it into her heart.
Horrified, Collatinus and Lucretius watched her die. In grief and anguish,
remembering her last words, they laid the bloodstained body on a bier and
took it in procession to the market-place. A crowd gathered, and the two men
told the tragic tale. It was enough. The citizens had long been indignant at
Tarquin’s cruelty and injustice. Now, stirred to righteous fury at Sextus
villainy, they rose against the tyrants. Tarquin and his sons were banished from
Rome for ever, and the free citizens resolved to elect a chief magistrate of their
own. Who better than the husband, or the father, of brave Lucretia?

3. The Story of Publius Valerius

Publius Valerius was the best of men. He was brave and handsome and
generous, and though he was rich, he used his wealth to benefit his fellow-citi­
zens. When bad king Tarquin was banished from Rome, and the free citizens
resolved to elect a chief magistrate of their own, the choice naturally fell on
him.
Valerius had a fine house on the Velia, overlooking the market-place,
conveniently close for consultation. But alas, there is malice even in a free
republic, and his enemies whispered that it was a tyrant’s castle, dominating the
citizens. Valerius never listened to gossip, but his close friends warned him of
the slander. One morning, as the citizens came to the market-place, they
looked up to the Velia and saw the house was gone. Valerius and his friends
had demolished it overnight, stone by stone and timber by timber.
When the chief magistrate, now homeless, came into the market-place that
day, he ordered his lictors to lower their fasces in the presence of the citizen
body. The Roman People was sovereign, not he. And so, in honour of his
modesty, Publius Valerius was called ‘The People’s Friend’.
 ROMAN REPUBLIC, YEAR ONE 309

4. The Story of Marcus Horatius

Marcus Horatius was a Roman of the old school, stem, unbending and
devoted to the public good. When bad king Tarquin was banished from
Rome, and the free citizens resolved to elect a chief magistrate of their own,
the choice naturally fell on him. And that is why you can see his name on the
architrave ofjupiter’s temple on the Capitol.
Everyone knows how the Capitol got its name. When they were digging
the foundations of the great temple, the workmen found a human head, mirac­
ulously preserved. Even the Etruscan soothsayers had to admit that the place
had been marked out by the gods to be the head of things, the centre of power.
That was in the days of the first Tarquin, the father of the tyrant.
The temple was huge and magnificent, and took many years to build. The
first Tarquin died, and his wicked son inherited the project. But naturally
Jupiter did not want his temple dedicated by a king so hateful to gods and men
alike. So he saw to it that Tarquin was banished when the temple was almost
finished, and the duty of dedicating it fell to Marcus Horatius.
Now, the dedicant of a temple must be free of all connection with grief or
mourning. Horatius had a son, a fine young man who was at that time on mili­
tary service. Just as rhe dedication was under way, a tearful messenger
interrupted the ceremony. "Bad news, Marcus Horatius! There is sickness in
the camp, and your son is dead.’ The people groaned in horror, but Horatius
thought only of his duty, "That is nothing to me,’ he said. ‘Cast out the body;
1 will not grieve or mourn.’ And so the temple was dedicated with no ill omen.
It was Horatius’ nephew, another devoted patriot, who saved Rome the
following year by holding the bridge against the Etniscan host ot Lars Porsenna
and his ally, the banished Tarquin.

II
Told like that, the stories are clearly independent, and mutually inconsistent.
That is what one expects in an oral tradition, when every telling ot a story is a
particular occasion, to a particular audience, for a particular purpose. One
imagines that part, at least, of the purpose was to honour the noble family
whose ancestor was commemorated in each of these exemplary tales.
But the stories are not told like that in our surviving literary sources. Livy
and Dionysius were dependent on a tradition of written history going back
nearly two hundred years; in the case of Plutarch, it was three hundred.4 Add

4 The main narrative sources for the first year of the Republic are Livy 2.1.7-8.9, Dionysius
ot Halicarnassus 5.1.2—19.5, and Plutarch Publicola 1.4-14.5. Up-to-date modem discussion
in Cornell 1995, ch. 9; cf. also Drummond 1989.172-90. For a fascinating exploration ot all
the aspects of the Brutus legend, assuming (wrongly, in my view) its essential historicity, see
Mastrocinque 1988.
 310 UNWRITTEN ROME

to that another three centuries—more than half the entire history of the
Republic—back from the origins of Roman historiography to the time of the
events themselves (whatever they were), and we have to take account of half a
millennium of creative story-telling.
One of the things that evidently happened during that time was that the
praetor niaxintns was forgotten. The tradition takes it totally for granted that the
king was replaced by consuls, two magistrates of equal authority, holding office
for a year. The earliest contemporary evidence for the consulship is the
sarcophagus inscription of Gnaeus Scipio Barbatus (about 260 BC);5 but it is
universally accepted that the institution dates back at least a century before
that. Many modem historians even believe that the tradition is right, and that
the consulship was introduced with the Republic; but they need pretty fancy
footwork to explain away the evidence for the praetor niaxinitts.
Once you accept the consulship as the original republican magistracy, two
of the stories can at once be amalgamated: L. Brutus (‘Stupid’) and L. Collat-
inus (Lucretia’s husband) can be the first two consuls. All you have to do is find
a reason for Brutus to be present at Lucretia’s suicide. As for Valerius, Horatius,
and Lucretia’s father, they can be brought in as ‘sufFects’, replacing consuls
who died in office or abdicated. And so, by the time our sources were writing,
it was accepted doctrine that there had been five consuls in the first year of the
Republic.
St Augustine, taking that in good faith as genuine history, naturally inferred
that the Republic had begun with a chapter of disasters:6
Indeed, the consuls did not complete their year of office. For Junius Brutus
deposed his colleague Lucius Tarquinius Collatinus from office and expelled
him from Rome; and soon afterwards Brutus himself fell in battle, after
inflicting wound for wound upon his enemy [one of Tarquin’s sons]. . . .
Besides this, Lucretius, the substitute for Brutus in the consulship, was carried
off by illness before the end of the year. And so Publius Valerius, successor to
Collatinus, and Marcus Horatius, brought in to fill the place of the departed
Lucretius, completed that year of mourning and misery, the year which saw
five consuls, and the year in which the new Roman commonwealth solemnly
inaugurated the consulship, the new office of authority.

The Roman historians, who were nothing if not patriotic, can hardly have
intended that reading. That it was possible at all is eloquent testimony to a
forced and factitious narrative.
What really happened cannot be known—though the evidence for burned
buildings, at an archaeological level that may well correspond to the later sixth

5 ILLRP 309 [p. 6 above]; Wachter 1987.301-42.

6 Augustine City of God 3.16 (Penguin translation).
 ROMAN REPUBLIC, YEAR ONE 311

century bc, suggests that ‘the flight of the king’ was a violent episode. Archae­
ology has also revealed Publius Valerius as a historical character: at about the
right date, ‘the companions of Poplios Valesios’ made a dedication to Mars at
Satricum, a Latin town some forty miles south of Rome.7 But that local
warlord seems to have little in common with the democratic constitutionalist
of the Roman story.
The trouble with archaeological discoveries is that they encourage the
Schliemann fallacy: find the site of Troy, and you’ve proved the Iliad is true.
Even the best modern historians sometimes succumb to this temptation.
The study of early Koine has been put on a wholly new footing by Tim
Cornell’s brilliant synthesis T/ic Bct’inunit’S of Rome (1995). Cornell has no illu­
sions about the tradition on the birth of the Republic: ‘it has the appearance of
a historical romance, and forms a self-contained saga of connected stories.’ But,
he goes on,

there is no reason in principle why the tradition should not be a romanticized

version of events that really happened. It is arbitrary to dismiss the rape of
Lucretia (for instance) as fiction, when we have no way of knowing whether it
is fiction or not.8

That is, it purports to be true; it could be true; why should it not be true?
Further,

we might be tempted to argue that the overthrow of Tarquin was followed by

a contused period of turmoil in which various members of his family and other
leading figures struggled for power, replacing each other in rapid succession...

—and the note makes it explicit that

they would include his relatives Brutus and Collatinus, but perhaps also
Valerius Publicola ['The People’s Friend’], who held the consulship three years
in succession, and in the traditional story was suspected of aiming at kingship.’

So perhaps there really were five ’consuls’ in Year One.

How do we know Valerius held the consulship three years running?
Cornell takes it as a fact because he believes that

the practice of recording the names of the men who held the chief magistracy
must go back to the very early years of the Republic, and it is certain that contin­
uous lists were kept in written form.1"

7 Illustrated in Momigliano 1989.97, fig. 33; see Stibbe et al. 1980.

8 Cornell 1995.217.
9 Cornell 1995.217-18, 439.
10 Cornell 1995.13 (my emphasis).
 312 UNWRITTEN ROME

If that were so, then documentary evidence would guarantee the five names as
authentic. But it is, to put it mildly, an adventurous hypothesis.11 It’s also
inconsistent with Cornell’s own suggested model for Year One: why should
the victor in that putative power struggle carefully record his rivals’ names as
equal to his own?
So forget any idea of archival evidence. What we have is the tradition, and
what matters is how we handle it. ‘In each case,’ says Cornell,

one must ask, first, whether there are grounds for regarding a story as ancient,
or as a relatively late invention; and second, whether there are reasons for
thinking that it might be based on fact.

Well and good, as far as it goes. But then: ‘The burden of proof lies as heavily
on those who wish to deny as on those who wish to affirm.’12 And that, with
the greatest respect, just will not do.
The burden of proof is on whoever challenges the prinui facie presumption.
And what is the prima facie presumption here? Not, I think, that authors writing
five hundred years later, in a tradition of written history no more than two
hundred years old, are likely to have reported the events accurately, or even
recognisably. In such circumstances, to treat ‘Why shouldn't it be true?’ as a no
less valid question than ‘Why should it?’ comes pretty close to abdicating the
historian’s responsibility.13
Better, in any case, to ask a different type of question. What sort of stories
are they, and how may they have come about?

Afterword (2008)

That is the sort of question the various studies in this book have tried to
address. The expulsion of the Tarquins and the establishment of the Republic
were a formative moment in Rome’s history, and of course a subject that

11 See Wiseman 1996.313-15 for arguments against. Stephen Oakley, in the introduction to
his magnificent new commentary on Livy 6-10, defends a position close to Cornell’s: 'it is
very hard to see whence Pictor and later annalists drew the basic framework of their narra­
tive, if there were no state records which were in some sense official’; the Romans belief
that the annales ina.viiiii went back to the beginning of the Republic ‘does not amount to
proof that the chronicle was already in existence in the fifth century or before, but it would
be surprising in a partly hellenized and partly literate society if the state did not keep records
of some kind’ (Oakley 1997.24, 25). 1 think that begs a big question about the nature of’the
state’; and Oakley’s general treatment of archival evidence seems to me more relevant to the
fourth century (the period covered by Books 6-10) than the fifth or late sixth. [See now
Oakley 2005.479-84 for further discussion of the problem.]
12 Cornell 1995.11.
13 Cf Oakley 1997.102: ‘accepting annalistic information unless it is proved to be wrong [is]
an absurd procedure given the inadequacies of our sources.’
 ROMAN REPUBLIC, YEAR ONE 313

mattered deeply to the Romans themselves. By the time Livy was writing his
history, a satisfactorily coherent narrative had been evolved from the princes’
trip to Delphi to the peace treaty with Porsenna.14 Some of the elements out of
which it was created can be identified without too much difficulty, but it is not
so easy to see which of them, if any, date back to the shadowy world of
unwritten Rome.
The story of Brutus and the story of Lucretia were both subjects for the
literary drama of the second and first centuries BC,1’ and no doubt also for the
pre-literary dramatic performances we have hypothesized in previous chapters.
Such an origin seems likely for the story Livy and Ovid tell about how Sextus
Tarquinius came to lust after Lucretia:16 as the siege of Ardea was dragging on,
the young royals amused themselves with drinking sessions; at one such, they
decided to ride off home that very night to see what their respective wives
were up to; in Rome, the ‘royal daughters-in-law’ were enjoying themselves
at a party'; at Collatia, chaste Lucretia was spinning wool with her maids.17
Since the wives who failed the test are conspicuously anonymous, one suspects
that their party' was originally a dance-interlude performed by miiiiae, light
relief of the kind we have inferred for other stories from drama.18 It is possible
too that the expulsion of the tyrants was presented on the stage by the Furies
pursuing Tarquin’s murderous queen.19
To the stories of the consuls of Year One we may add the three tales of
Roman heroism in Year Two which supposedly persuaded Porsenna to lift the
siege of Rome and abandon his alliance with the exiled Tarquins.2" These
were the exploits of Horatius Codes, Mucius and Cloelia, ‘those three prodi­
gies and marvels of the Roman name, who if they were not in the
history-books would seem mythical today’.21 The seven stories celebrate
heroes or heroines of six different aristocratic families, the lunii, Lucretii,
Valerii, Horatii, Mucii and Cloelii. We may guess at the date of their respec­
tive origins from the periods when each family was most prominent in Rome,

14 Livy 1.56.4—2.15.7, cf. Plutarch Publieola 2.3—19.6.

15 Brutus: Accius Brutus (fragments and discussion in Manuwald 2001.220-37). Lucretia: Vano
De lingua Latina b.7 and 7.72, citing a tragedian called Cassius (Manuwald 2001.237-43).
16 Livy 1.57.4—11, Ovid Fasti 2.723-80; also Dio Cassius 2.11.13-15, De uiris ilhislribus 9.1-2.
Contrast Dionysius of Halicarnassus 4.64.2—4, where Sextus conies to Collatia from Gabii,
not Ardea, having desired Lucretia for a long time.
17 Regii nineties: Livy 1.57.4, De uiris illustribus 9.1. Reside minis: Livy 1.57.9, Ovid Fasti 2.739,
De uiris illustribus 9.1. Cf. Ogilvie 1965.219, 222 (’pure New Comedy’).
18 See pp. 180—1 above on Porsenna's hostages, 198-9 (with Wisenun 201 >4a.269-70) on
Poppaea’s wedding.
19 Livy 1.59.13, with Michels 1951.19—20; for Furies on stage, see pp. 212-13 above.
20 Livy 2.13.8 (in adniirationem uersus), Plutarch Publicola 17.5 (dyuoDtic Kai 0Ul!|iaoaC TO
CfpOVT||J.a Kai tf|V apETTty TWV ’Ptupatwv). Dionysius of Halicarnassus (5.21.1) puts the
events in Year Three.
21 Fionas 1.10.4. For Cloelia see above, pp. 180-1, 289.
 314 UNWRITTEN ROME

as indicated by the consulships (or other chief magistracies) attributed to them

in the later Roman tradition.22
The plebeian lunii took effective advantage of the end of the ‘struggle of
the orders’, with six consulships and two dictatorships in the thirty-five years
from 325 to 291 BC; and they continued to be a distinguished family, with ten
more consulships (and another dictatorship) at regular intervals down to 62
BC.23 The plebeian Mucii have a similar record, though starting a century later,
with six consulships between 220 and 95 BC.24
No inference can be made from the patrician Valerii, who were hugely
influential throughout the republican period, and beyond.25 The other three
patrician families are more interesting, and important as a contrast with the
lunii and Mucii. Counting consulships and ‘consular tribunates’ together, the
Lucretii have nine chief magistracies between 508 and 381, the Horatii seven
between 507 and 378, and the Cloelii three between 498 and 378. No one
knows how reliable those records are, but the early high status of the Horatii, at
least, is confirmed by the use of their name for one of the original rural tribes.
It is natural to conclude that the stories of Lucretia and her father the
consul, of M. Horatius the consul and Horatius Codes, and of the heroine
Cloelia, all date from the fifth century or early fourth, while those of Brutus
and Mucius were created in the late fourth and late third centuries respectively.
We must be cautious, however. The old families may have continued: Lucretii
of senatorial rank are known from at least the late third century BC, Cloelii
from at least the early second, and even Horatii from at least the mid-first.-"
In this context it is worth considering also the names of the seven kings.
That of Romulus is certainly ancient, known already to a fourth-century
Sicilian historian (as the son of Aeneas and Tyrrhenia), anil no doubt associated
with the fifth-century Romulii, or Romilii, who gave their name to another
of the original rural tribes.27 Nor is there any reason to doubt the antiquity of

22 Magistrates after 366 Be probably come from authentic lists, and may be taken as historical;
those before that date are presumably the work of Roman antiquarians (p. 235 above), and
their status as historical data is very doubtful.
23 Consulships: 325, 317, 313, 311, 292, 291, 277, 266, 249, 230, 178, 167, 138, 109, 77, 62
BC. Dictatorships: 312, 302, 216 BC.
24 220, 175, 174, 133, 117, 95 BC—all with the cotp/onieti Scaevola, explained by their
legendary ancestor’s exploit (Plutarch Publicola 17.3).
25 On my count, 43 consulships between 505 and 31 BC; five dictatorships between 501 and
302 BC; 18 military tribunates with consular authority between 414 and 367 BC. For the
creation of Valerian pseudo-history in the late Republic, see Wiseman 1998.75-89.
26 Lucretii (now plebeian): e.g. praetors in 205, 172 and 171 BC. Cloelii: e.g. a rex sacrorum in
180 (Livy 40.42.11) and a senator in 39 BC (Sherk 1969.158, line 8). Horatii: a legatiis in 43
BC (Cicero Adfainiliares 12.30.7).
27 Alcimus FGrH 560 F 4; Festus (Paulus) 331L for Romulia tribus; T. Romilius Rocus Vati-
canus was supposedly consul in 455 BC and subsequently one of the Decemvirs (Degrassi
1947.24-5, Diodorus Siculus 12.5.1, Livy 3.33.3 etc).
 ROMAN REPUBLIC, YEAR ONE 315

the Tarquinii, a family known to Vel Saties of Vulci in the fourth century (one
of the scenes painted in his tomb was of Gnaeus Tarquinius of Rome being
killed by Marcus Camillus) and probably already to a Greek historian of
Cumae a century before that.28 Numa’s family name, Pompilius, is totally
mysterious and practically unparalleled in the Republic.29
That leaves Ancus Marcius, Tullus Hostilius and Servius Tullius, all of
whom bear the names of families attested in later times. A Tullius was suppos­
edly consul in 500 BC; the much later Tullii, including three first-century
consuls, evidently did not claim any connection.'" The March, on the other
hand, had eight consulships and a dictatorship in the 76 years from 357 to 281
BC (and then twelve consulships between 186 and 38 bc), a profile very like
that of the lunii;31 and the Hostilii, like the Mucii, first appear in the late third
century and continue at a high rank in the second.32 If it makes sense to think
of Horatius Codes as a creation of the fifth century, lunius Brutus of the late
fourth, and Mucins of the late third, the same inference may reasonably be
made about Servius Tullius (fifth century), Ancus Marcius (late fourth) and
Tullus Hostilius (late third).
So far as we can tell from Naevius and Ennius, both of whom knew
Romulus as the grandson of Aeneas,33 unwritten Rome did not have a contin­
uous narrative of the regal period, but a series of independent stories about
good king Numa, wicked king Tarquin, and so on. The idea of precisely seven
reigns, of which the years could be counted, was the achievement of literary
historiography, and it depended on the chronologies of Timaeus and Eratos­
thenes, who provided dates for the foundation of the city and the expulsion of
the kings.34 It is quite possible to imagine Fabius Pictor or Cincius Alimentus
exploiting the family pride of the Marcii and Hostilii (the latter their own
contemporaries) to link up the traditional stories into seamless ‘real history’.
Of course one cannot say for certain that that is what happened. But the
possibility (to put it no more strongly than that) should be borne in mind as an

28 Francois Tomb: e.g. Cornell 1995.135-9, Wiseman 2004a.41-3. Historian of Cumae: Zevi
1995; p. 293 above.
29 Ogilvie 1965.601: 'No Pompilius is known between Numa and Catiline's friend (Q.
Cicero, Cuinni. Pi t. 10).’ Claims of descent from Numa were already challenged by Roman
authors (Plutarch Nimiu 21.1, cf. 1.1).
30 Crawford 1974.297 for a moneyer in 120 BC; the consulships were in 81, 63 and 3IJ BC. No
connection: Cicero Brutus 62.
31 Consulships: 357, 352, 344, 342, 310, 306, 288, 281, 186, 169, 162, 156, 149, 118, 91, 68,
64, 56, 39, 38 bc. Dictatorship: 356 BC. Compare n. 23 above. Descent from king: Sueto­
nius Diuus lulius 6.1 (Marcii Reges).
32 Three praetorships in 209 and 207, then consulships in 170, 145 and 137 BC. Descent from
king implied by Tullus Hostilius, tribune designate in 43 (Cicero Philippics 13.26).
33 Servius audits on Virgil Aeueid 1.273. See pp. 49-50 above on Ennius' chronology.
34 Eratosthenes FCrH 241 Fl and F45 (see p. 49 above); Polybius 3.22.1-2 (from Timaeus, or
Fabius following Timaeus').
 316 UNWRITTEN ROME

antidote to the opposite assumption, as recently expressed by a very distin-

guished historian:35

As is well known, there is an ancient and uncontested tradition that makes the
reign of Ancus Marcius, fourth king ofRome (traditional dating 646—616 bc),
the moment when Roman dominion expanded as far as the Tyrrhenian coast
(usque ad mare imperium prolatum, Livy 1.33.9). . .

As Fausto Zevi rightly points out, ‘the story of Ancus Marcius had been grafted
on to a stock of Greek historiography’—but he assumes it was done very early,
with an authentic Ancus mentioned in the fifth-century Greek source that
evidently reported the alliance of Aristodemus of Cumae with the exiled
Tarquin.36 With the greatest respect, I find it hard to believe that the putative
Greek historian would have concerned himself with any details about Rome
before the son of Corinthian Demaratus arrived there. The grafting was much
more probably done centuries later, by a Roman historian and for a Roman
reason, to create a continuous narrative of the kings.
As we saw in the first chapter, real evidence for early Rome was rare and
hard to interpret. One potential source of information was provided by
honorific statues that had been set up in the past. There was a group of eight
on the Capitol, one with a drawn sword, which by the late Republic at least
were identified as the seven kings plus Lucius Brutus.37 But they were not the
canonical seven, since Titus Tatius was included,38 and in any case they were
probably no older than the fourth century BC;39 what they represented when
they were first made is anybody’s guess. The same may be said of the old
statues identified as Porsenna,40 Horatius Codes,41 Mucius,1- and Cloelia.43 We

35 Zevi 2005.30, cf. 31 on ‘the historical truth of the regal foundation ot Ostia’.
36 Zevi 2005.32: ‘Since [the tradition about Aristodennis| concerns a highly polemical contest
against the patricians responsible for driving out the Tarquins, it is more than probable that
it would have mentioned the name of the king of Koine by whom Tarquin was kindly
received and introduced into the highest social levels of the day. that is to say that same
Ancus Marcius, along with perhaps mention of his military endeavours and conquests made
by or with him.’ More than probable?
37 Cassius Dio 43.45.4 (seven plus Brutus), Plutarch Brutus 1.1 (drawn sword); Brutus statue
also mentioned by Pliny Nat. Hist. 34.23.
38 Asconius 29C, Pliny Nat. Hist. 34.23.
39 See Holscher 1978.328-31, Coarelli 1999b,369.
40 Plutarch Publicola 19.6, ‘near the Senate-house, of simple and archaic workmanship’.
41 Livy 2.10.12 (in comitio), Dionysius of Halicarnassus 5.25.2, Plutarch Publicola 16.7 (‘in the
temple of Vulcan’), Aulus Gellius 4.5.1-6 (in comitio 5.1, in area Volcani 5.4), De uiris illus-
tribus 11.2 (in Vokanali).
42 De uiris illustribus 12.7, site not given.
43 Livy 2.13.11 (in summa sacra uia), Dionysius of Halicarnassus 5.35.2 (on Sacra Via), Seneca
Cimsolatio ail Marciain 16.2 (in sacra uia), Pliny Nat. Hist. 34.29 (contra louis Staton's aedem in
uestibulo Supcrbi dooms), Plutarch Publicola 19.5 (on Sacra Via as you go to the Palatine),
Plutarch Moralia 250f (on Sacra Via), Servius on Aeneici 8.646 (in sacra uia).
 ROMAN REPUBLIC. YEAR ONE 317

know that the identity of the bronze girl on a horse was disputed (Cloelia or
Valeria?),44 and it seems that the ‘Horatius Codes’ statue at the Volcanal could
also be identified as Romulus, or as a play-actor struck by lightning, whose
remains were buried there.45 What seemed to the historians to be evidence for
their history of early Rome was probably nothing of the kind.
Two items on Porsenna preserved in the literary sources are interestingly
inconsistent with the historians’ bland portrait of the chivalrous king. The first
is the elder Pliny’s quotation from a treaty between Porsenna and the Roman
People; since it forbade the Romans the use of iron except for agricultural
purposes, it was evidently a treaty of surrender.'"’ It is impossible to know how
authentic the supposed document was, but Pliny goes on to quote ‘very
ancient writers’ (netiistissimi auctores), from whom he may have got the infor­
mation, and Tacitus too takes it for granted that the city was surrendered to
Porsenna.47
The other item was an old custom still in use in Livy’s time: whenever the
quaestors carried out an auction of public property, the first business was
always a symbolic ‘sale of the goods of king Porsenna’.48 That would seem to
commemorate the sale of booty from a defeated enemy,4'’ in this case an
unwelcome master who has been forced to withdraw. Livy, to his credit, is
aware of the inconsistency of this ritual with the tale of goodwill and mutual
admiration he has just narrated, but in the end he is happy to find a 'probable
explanation’, that the departing Porsenna generously left his headquarters fully
stocked for the Romans’ benefit; Dionysius and Plutarch report this version as
accepted fact?"
This episode offers a glimpse into what may lie behind <»iy item in the
smooth and plausible narrative of early Rome that is offered by Livy, Diony­
sius and Plutarch. It so happens that in this case we can detect a rival version,
possibly based on documentary evidence, in the inconspicuous allusions of
Pliny and Tacitus, and a contradictory oral tradition, embedded in the

44 See p. 289 above.

45 Romulus: Dionysius of Halicarnassus 2.54.2, with p. 10 above on rival interpretations of the
Volcanal complex. Ludius-. Festus 37t)L.
46 Pliny Nur. Hist. 34.139: infoedere quod expulsis regibus papula Romano Mil Porsina, noininalim
corupreliensuni inueninius ne fem nisi in agri cultu uteretur.
47 Tacitus Histories 3.72.1 on the Capitoline temple of Jupiter, quant non Porsenna dedita urbe
neque Galli capta tenierare potuissent.
48 Livy 2.14.1: trios traditus ab antiquis usque ad nostrum aetatem inter cetera sollentnia matter, bona
Porsennae regis uendendi. Cf. Dionysius of Halicarnassus 5.34.4 (quaestors), Plutarch Publicola
19.6 (auction).
49 So Ogilvie 1965.268 ("it will have been a semi-religious commemoration ot a Roman
success’), who notes the parallel with the audio Veientium (Festus 430L, ct. Plutarch Romulus
25.5, Aloralia 277c—d).
50 Livy 2.14.1 (liuic ram paeatae profeetioni ab urbe Regis Etnisei abhorrent mos...), 2.14.3 (pro.v-
titiurn uero est. . .); Dionysius of Halicarnassus 5.34.4, Plutarch Publirola 19.5-6.
318 UNWRITTEN ROME

quaestors’ symbolic ritual at auctions. Without the fortuitous survival of that

counter-evidence, we might have been tempted to say ‘As is well known,
there is an ancient and uncontested tradition’ that the Romans of the new
Republic successfully persuaded Porsenna to abandon his alliance with the
Tarquins.
In fact, what Livy offers is not ancient tradition at all, but a literary narrative
progressively created by his predecessors and elaborated by himself. If the terms
of ‘Porsenna’s treaty with the Roman People’ were available to Pliny, they
were surely available equally to Livy. But he ignores them, just as he ignored
the evidence for the praetor iitaximus.51 The reason is obvious: to try to take
account of the implications of such data would have made it impossible for
Livy to write his sort of history at all. He wanted to present a morally exem­
plary narrative which would be like a monument for his readers to
contemplate.52 The story was something he had inherited from his predeces­
sors; odd bits of evidence that were incompatible with it just had to be passed
over.
What would Livy have made of the ‘companions of Poplios Valesios’, if he
had known about the Satricum inscription? He might perhaps have identified
them as the friends who warned Publius Valerius that his house on the Velia
was making him unpopular with the citizens,’3 but he could not do what the
modem historian does, and take them as evidence for a quasi-Homeric archaic
society.54 That would have made nonsense of his whole narrative. Using such
‘indications’ to make inferences about an impenetrably distant past is what
Thucydides thought a historian ought to do,’5 but neither Livy nor Dionysius
nor Plutarch was interested in following his example.
Even so, there is material in their narratives that can be exploited. Livy did
mention the ‘sale of the goods of king Porsenna’, even if he had to explain it
away; Plutarch did report the summoning of Jupiter by magical means, even
though he regarded the story as ‘fabulous and ridiculous’;’6 and when we infer
a traditional performance genre that kept old tales of Roman heroism alive, it
is Livy himself who unwittingly provides the evidence,’7 with his repeated

51 Livy 7.3.5; pp. 1()— 11 above.

52 Livy preface 10: hoc Hind est praecipue in cognitione rerun! salubre ac frugiferuni, omnis te exempli
documenta in inlustri posita mominiento intueri.
53 Livy 2.7.6-7, cf. Plutarch Publicola 10.3 (TtilV cpiktov <5tt'E,ldvT(i>v).
54 E.g. Cornell 1995.144: ‘The otherwise rather unusual word sodales seems to have a particular
social significance in this context and to refer to the armed followers of an independent
warlord. The most obvious comparison is with the Homeric hetaimi, and it is no accident
that this is exactly how Dionysius of Halicarnassus translates it (9.15.3).’
55 Thucydides 1.20.1-21.1 (p. 8 above).
56 Plutarch Numa 15.6 (TUOTa pev OVV TCI pv0d>6r) Kai yEkota. . .); see pp. 162-6 above.
57 See chapter 2 above.
 ROMAN REPUBLIC, YEAR ONE 319

reports of soldiers singing at triumphs. More often it is what the poets say
Ennius on ‘Fauns and prophets’, Ovid on the Kalends of April—that enables us
to get an insight into unwritten Rome. Such insights are very uncertain, quite
different from the confident narrative of the historians, and they are often very
surprising, because the past is a foreign country where they do things differ­
ently. But I hope this book has shown that they are worth seeking out.

CLASSICAL
A
LIBRARY '
CAMBRIDGE
 Bibliography
Adam and Briquel 1982: Richard Adam and Dominique Briquel, ‘Le miroir prenestin
de 1’antiquario comunale de Rome et la legende des jumeaux divins en milieu latin
a la fin du IVe siecle av. J.C.’, Melanges de PEcole franfaise de Rome (Antiqiiite) 94:
33-65.
Adams 1982: J.N. Adams, The Latin Sexual Vocabulary. London: Duckworth.
Altbldi 1965: Andrew Altbldi, Early Rome and the Latins (Jerome Lectures series 7).
Ann Arbor: University of Michigan Press.
Altbldi 1974: Andreas Altbldi, Die Struktur des voretniskischen Romerstaates. Heidelberg:
Winter.
Ammennan 1990: A. Ammerman, ‘On the Origins of the Roman Forum’, American
Journal of Archaeology 94; 627—16.
Ammennan 1998: Albert J. Ammerman, ‘Environmental Archaeology in the
Velabrum, Rome: Interim Report’, Journal oj Roman Archaeology 11: 213-23.
Ammennan 2006: Albert Ammennan, ‘Adding Tune to Rome’s imq^o’, in Lothar
Haselberger and John Humphrey (eds). Imaging Ancient Rome: Documentation—
Visualization—Imagination (JRA Supplement 61, Portsmouth Rl): 297-308.
Anderson 1992: William S. Anderson, ‘The Limits of Genre: Response to Francis
Cairns’, in Karl Galinsky (ed.), The Interpretation of Roman Poetry: Empiricism or
Hermeneutics? (Studien zur klassischen Philologie 67, Frankfurt: Peter Lang): 96-103.
Ando 2000: Clifford Ando, Imperial Ideology and Provincial Loyalty in the Roman Empire.
Berkeley: University of California Press.
Ankarloo and Clark 1999: Bengt Ankarloo and Stuart Clark, ’Introduction’, in Valerie
Flint et al., Witchcraft and Magic in Europe: Ancient Greece and Rome (London:
Athlone Press): xi—xvi.
Arcella 1985: Luciano Arcella, ‘Il mito di Cloelia e i Valerii’, Studi e materiali di storia
delle religion! 9: 21—12.
Aronen 1999: Jaakko Aronen, ‘Perche il verso satumio fu chiamato “satumio ?, in
Nicole Blanc and Andre Buisson (eds), Imago antiquitatis: religions et iconographie du
monde romaine (Paris: de Boccard): 53—72.
Astin 1967: A.E. Astin, Scipio Aemiliantis. Oxford: Clarendon Press.
Badian 1966: E. Badian, ‘The Early Historians’, in T.A. Dorey (ed.), Latin Historians
(London: Roudedge and Kegan Paul): 1-38.
Badian 1984: E. Badian, ‘The House of the Servilii Gemini: A Study in the Misuse ot
Occam’s Razor’ , Papers oj the British School at Rome 52: 49-71.
Bagnasco Gianni 1996: Giovanna Bagnasco Gianni, Oggetti iscritti di epoca orientalizzante
in Etruria (Biblioteca di Studi Etruschi 30). Florence: Olschki.
 322 UNWRITTEN ROME

Barchiesi 1994: Alessandro Barchiesi, Il poeta e il principe: Ovidio e il discorso augusteo.

Bari: Laterza
Barchiesi 1997: Alessandro Barchiesi, The Poet and the Prince: Quid and Augustan
Discourse. Berkeley: University of California Press.
Barnes 1982: Timothy D. Barnes, ‘The date of the Octavia', Museum Helveticum 39:
215-17.
Baroni 2001: Irene Baroni, ‘Livelli di occupazione dell’eta del Bronzo nel Giardino
Romano: il Bronzo recente’, Bullettino della conunissione archeologica comtmale di
Roma 102: 291-8.
Barr 1962: W. Barr, ‘Horace Odes I, 4’, Classical Review 12: 5—11.
Barre 1840: L. Barre, Herculaneum et Pompei vol. 8 (Musee Secret). Paris.
Battaglia and Emiliozzi 1979: Gabriella Bordenache Battaglia and Adriana Emiliozzi,
Le ciste prenestine, I Corpus: 1.1. Rome: Consiglio nazionale delle ricerche.
Battaglia and Emiliozzi 1990: Gabriella Bordenache Battaglia and Adriana Emiliozzi,
Le ciste prenestine, I Corpus: 1.2. Rome: Consiglio nazionale delle ricerche.
Beacham 1991: Richard C. Beacham, The Roman Theatre and its Audience. London:
Routledge.
Beard 1986: Mary Beard, ‘Cicero and Divination’, Journal of Roman Studies 76: 33-46.
Beard, North and Price 1998: Mary Beard, John North and Simon Price, Religions oj
Rome, two volumes: 1 A History, 2 A Sourcebook. Cambridge University Press.
Beazley 1947: J.D. Beazley, Etruscan Vase-Painting. Oxford: Clarendon Press.
Beazley 1963: J.D. Beazley, Attic Red-Figure Vase-Painters, ed. 2. Oxford: Clarendon
Press.
Belier 1991: Wouter W. Belier, Decayed Gods: Origin and Development of Georges Dumezil’s
‘Ideologic Tripartie’ (Studies in Greek and Roman Religion 7). Leiden: Brill.
Beness and Hillard 2001: J.L. Beness and T.W. Hillard, ‘The I heatricality of the
Deaths of C. Gracchus and Friends’, Classical Quarterly 51: 135—40
Bemheimer 1952: Richard Bernheimer, Wild Men in the Middle Ages: a Study in Art,
Sentiment and Demonology. Cambridge MA: Harvard University Press.
Bernstein 1998: Frank Bernstein, Ludi publici: Untersnchungen zur Entstehung mid
Entivicklung der offentlichen Spiele im republikanischen Rom (Historia Einzelschrift 119).
Stuttgart: Franz Steiner.
Bettelli 1997: Marco Bettelli, Roma, la citta prinia della citta: i tempi di una nascita (Studia
archeologica 86). Rome: L’Erma di Bretschneider.
Bevington and Rasmussen 1993: David Bevington and Eric Rasmussen (eds). Doctor
Faustus A- and B-Texts (1604, 1616). Manchester University Press.
Binder 1964: Gerhard Binder, Die Aussetzung des Konigskindes (Beitrage zur klassischen
Philologie 10). Meisenheim am Gian: Hain.
Bloomer 1992: W. Martin Bloomer, Valerius Maximus and the Rhetoric of the New
Nobility. London: Duckworth.
Boardman 1997: John Boardman, The Great God Pan: The Survival of an Image (Walter
Neurath Memorial Lecture 29). London: Thames and Hudson.
Bodel 1994: John Bodel, ‘Graveyards and Groves: A Study of the Le.v Lucerina , Amer­
ican Journal of Ancient History 11: 1—133.
Boels-Janssen 1993: Nicole Boels-Janssen, La vie religieuse des matrones dans la Rome
 BIBLIOGRAPHY 323

archaique (Collection de 1’Ecole fran<?aise de Rome 176). Rome: Ecole franfaise de

Rome.
Boissier 1893: Gaston Boissier, ‘Les fabulae praetextae', Revue de philologie 17: 101-8.
Borner 1958: Franz Borner, P. Ovidius Naso: Die Fasten, vol. 2. Heidelberg: Carl
Winter.
Borgeaud 1988: Philippe Borgeaud, The Cult of Pan in Ancient Greece (trans. Kathleen
Atlass and James Redfield, orig. edition Rome 1979). University of Chicago Press.
Bowersock 1990: G.W. Bowersock, ‘The Pontificate of Augustus’, in Kurt A.
Raaflaub and Mark Toher (eds). Between Republic and Empire: Interpretations of
Augustus and his Principate (Berkeley: University of California Press): 380-94.
Bowie 2000: Angus Bowie, ‘Myth and Ritual in the Rivals of Aristophanes , in David
Harvey and John Wilkins (eds). The Rivals of Aristophanes: Studies in Atheman Old
Comedy (London: Duckworth and Classical Press of Wales). 317 3). . .. ,
Bremer 1975: J.M. Bremer, ‘The Meadow of Love and Two Passages in Euripides
Hippolytus’, Mnemosyne 28: 268—80.
Bremmer 1987a: J.N. Bremmer, ‘Romulus, Remus and the Foundation of Rome in
J.N. Bremmer and N.M. Horsfall, Roman Myth and Mythography (BIGS Supple­
ment 52, London: Institute of Classical Studies). 2o 18.
Bremmer 1987b: J.N. Bremmer, ‘Slow Cybele’s Arrival’, in J.N. Bremmer and N.M.
Horsfall. Roman Myth ami Mythography (BICS Supplement 52, London: Institute of
Classical Studies): 105-12. , . , . ,
Brenk 1986: F.E. Brenk, ‘In the Light of the Moon: Demonology in the Early Impe­
rial Period’, in H. Ternporini and W. Haase (eds),_ Aufstieg und N.edergang der
romischen H'eh 2.16.3 (Berlin: De Gruyter): 2068-2143.
Brink 1971: C.O. Brink, Horace on Poetry: The Ars Poetica. Cambndge University ress.
Bnnk 1982: C.O. Brink. Horace on Poetry: Epistles Book II. Cambridge University ress.
Briquel 1974: Dominique Briequel, Le probleme des Dauniens, de I Ecole
/nmprise de Rome: Aniiquite 86: 7—10. . .. ,
Briscoe 1971: J. Briscoe, ‘The First Decade’, in T.A. Dorey (ed.), Lmy (London.
Routledge and Kegan Paul): 1-20. .... ,, n.,i;„. iv
Brommer 1944: Frank Brommer, Satyrspiele: Bilder gnec use ler asen.

1949-5U: Frank Brammer. 'Pan im 5. and 4. Jtartataenv. Cta’.

^^7“^ B^Son. a. W—
-......
wJ'wX Barken. NW ™
rial KM and (t„„, Perer Barg. ong. edrnon Berta 1972). Berkeley.
University of California Press. Contexts of Mythology.
Buxton 1994: Richard Buxton, Imaginary Greece.
T^openra. 4.9; "Hereule, BxW rta Dtaenno.a
Genre’, in (el). 7*r E”“”' “
 324 UNWRITTEN ROME

Hermeneutics? (Studien zur klassischen Philologie 67, Frankfurt: Peter Lang): 65-95.
Calvino 1988: Italo Calvino, Lezioni americane: Sei proposte per il prossimo millennia.
Milan: Garzanti.
Cameron 1976: Alan Cameron, Circus Factions: Blues and Greens at Rome and Byzan­
tium. Oxford: Clarendon Press.
Cameron 2004a: Alan Cameron, Greek Mythology in the Romnan World (American Clas-
sical Studies 48). New York: Oxford University Press.
Cameron 2004b: Alan Cameron, ‘Vergil Illustrated between Pagans and Christians:
Reconsidering “the late-4th c. Classical Revival”, the dates of the manuscripts, and
the places of production of the Latin Classics’, Journal of Roman Archaeology 17.502-
25.
Campbell 2000: Brian Campbell, The Writings of the Roman Land Surveyors (JRS
Monographs 9). London: Society for the Promotion of Roman Studies.
Carandini 1990: A(ndrea) C(arandini), ‘Domus aristocratiche sopra le mura e il
pomerio del Palatino’, in Mauro Cristofani (ed.), La grande Roma dei Tarquini: cata-
logo della mostra (Rome: L’Erma di Bretschneider): 97—9.
Carandini 1995: Andrea Carandini, ‘Racconto breve dello scavo con disegni’, in
Palatimu e Sacra Via I: Tavole = Bollettino di archeologia 34: 1—75.
Carandini 1997: Andrea Carandini, La nascita di Roma: Dei, Lari, erot e uomini all’alba di
ima civiltd. Turin: Einaudi.
Carandini 2000: Andrea Carandini, Giornale di scavo: Pensicri spars* di un drc/ico/oijo.
Turin: Einaudi.
Carandini 2003: Andrea Carandini, ‘Il mito romuleo e le origini <1: Roma , in Mario
Citroni (ed.), Memoria e identitd: La cultura romana costruisce L sua imagine (Studi e
testi 21, Florence: Universita degli Studi di Firenze): 3-19.
Carandini 2004: Andrea Carandini, Palatino, Velia e Sacra Via: Paessqggi urbani attraverso
il tempo (Workshop di archeologia classica, Quademi 1). Rome: Edizioni
dell’Ateneo.
Carandini 2006: Andrea Carandini, Remo e Romolo: Dai rioni dei (Juiriti alia citta dei
Romani (775/750-700/675 a.C.). Turin: Einaudi.
Carandini et alii 1997: A. Carandini, G. Ricci, M.T. D’Alessio, C. De Davide, N.
Terrenato, ‘La villa dell’Auditorium dall’eta arcaica all’eta imperiale’, Romische
Mitteilungen 104: 117-148.
Carandini and Cappelli 2000: Andrea Carandini and Rosanna Capelli (eds), Roma:
Romolo, Remo e la fondazione della citta. Milan: Electa.
Carandini and Carafa 1995: Andrea Carandini and Paolo Carafa, Palatium e Sacra Via I:
Prima dclle mura, I’etd civile mura e I’etd case arcaiche = Bollettino di archeologia 31-3.
Carandini, Carafa and Filippi forthcoming: Andrea Carandini, Paolo Carafa, Dunia
Filippi, Palatium e Sacra Via III.
Carpenter 1997: Thomas H. Carpenter, Dionysian Imagery in Fifth-Century Athens.
Oxford: Clarendon Press.
Cazenove 1983: Olivier de Cazenove, ‘Lucus Stimulae: les aiguillons des Bacchanales ,
Melanges de I’Ecolefranfaise de Rome (Antiquite) 95: 55—113.
Cazenove 1986: Olivier de Cazanove, ‘Le thiase et son double: Images, statuts, func­
tions du cortege divin de Dionysos en Italie centrale’, in L'association dionysiaque
 I

BIBLIOGRAPHY 325

dans les sodetes anciennes (Collection de 1’EFR 89). Rome: Ecole franijaise de Rome.
Champeaux 1982: Jacqueline Champeaux, Fortuna: Le culte de la Fortune a Rome et dans
le monde remain des origins a la mart de Cesar, vol. 1. Rome: Ecole franyaise de Rome.
Champeaux 1987: Jacqueline Champeaux, Fortuna: Le culte de la Fortune d Rome et dans
le monde remain des origins d la inort de Cesar, vol. 2. Rome: Ecole franyaise de Rome.
Chassignet 1996: Martine Chassignet, L’annalistique romaine, vol. 1. Paris: Les Belles
Lettres.
Christenson 2000: David M. Christenson, Plautus: Amphitmo (Cambridge Greek and
Latin Classics). Cambridge University Press.
Cichorius 1922: Conrad Cichorius, Romische Studien: Historisches, Epigraphisches, Liter-
aturgeschichtliches mis t’ierJahrlmnderten Roms. Leipzig: Teubner.
Coarelli 1983: Filippo Coarelli, ll foro romano: periodo archaico. Rome: Quasar.
Coarelli 1985: Filippo Coarelli, Il foro romano: periodo repubblicano e augusteo. Rome:
Quasar.
Coarelli 1987: Filippo Coarelli, I santuari del Lazio in eta repubblicana (Studi N1S arche-
ologia 7). Rome: Nuova Italia Scientifica.
Coarelli 1988: Filippo Coarelli, Il foro boario dalle origini alle fine della repubblica. Rome:
Quasar.
Coarelli 1995: F. Coarelli. ‘Flora, templum (in Colle)’, in Eva Margareta Steinby (ed.),
Lexicon Topographicum [Jrbis Romae 2, D-G (Rome: Quasar): 254.
Coarelli 1996: Filippo Coarelli, Rei’ixit ars: arte e ideologia a Roma dai modelli ellenislici
alia tradizione repurb : aria. Rome: Quasar.
Coarelli 1997: Filippo Coarelli, Il Campo Marzio dalle origini alia fine della repubblica.
Rome: Quasar.
Coarelli 1999a: Filippo Coarelli, ‘Sacra Via’, in Eva Margareta Steinby (ed.), Lexicon
Topographicum ( ri-i: Romae vol. IV (Rome: Quasar): 223-8.
Coarelli 1999b: Filippo Coarelli, "Statuae regum Romanomm’, in Eva Margareta
Steinby (ed.), Lexicon Topographicum L'rbis Romae vol. IV (Rome: Quasar): 368-9.
Coarelli 2005: Filippo Coarelli, ’I percorsi cerimoniali a Roma in eta regia’, in
Emanuele Greco (ed.), Teseo e Romolo: Le origini di Atene e Roma a con/roiito
(Tripodes 1, Athens: Scuola archaeologica italiana): 29-42.
Cole 1993: Susan Guettel Cole, ‘Procession and Celebration at the Dionysia’, in R.
Scodel (ed.), Theatre and Society in the Classical ll'or/d (Ann Arbor: University of
Michigan Press): 25-38.
Cole 2006: Spencer Cole, ‘Cicero, Ennius, and the Concept of Apotheosis at Rome’,
Aretlmsa 39.3: 531—48.
Collingwood 1946: R.G. Collingwood, The Idea of History. Oxford: Clarendon Press.
Colonna 1976: Giovanni Colonna, ‘“Scriba cum rege sedens’”, in Melanges offerts d
Jacques Heurgon: L’Italie preromaine et la Rome republicaine (Collection de 1 Ecole
franyaise de Rome 27, Rome): 187—95.
Colonna 1987: Giovanni Colonna, ‘Etruria e Lazio nell’eta dei Tarquini’, in Mauro
Cristofani (ed.), Etruria e Lazio arcaico: Atti dell’incontro di studio (Quademi del
Centro di studio per 1’archeologia etrusco-italica 15, Rome: CNR): 55-66.
Connor 1989: W.R. Connor, ‘City Dionysia and Athenian Democracy’, Classica et
Mediaevalia 40: 7—32.
 326 UNWRITTEN ROME

Cornell 1986: Timothy J. Cornell, ‘The Value of the Literary Tradition Concerning
Archaic Rome’, in Kurt A. Raaflaub (ed.), Social Struggles in Archaic Rome: New
Perspectives on the Conflict of the Orders (Berkeley: University of California Press):
52-76.
Cornell 1991: Tim Cornell, ‘The Tyranny of the Evidence: A Discussion of the
Possible Uses of Literacy in Etruria and Latium in the Archaic Age’, in Literacy in
the Roman World (JRA Supplement 3, Ann Arbor): 7—33.
Cornell 1995: TJ. Cornell, The Beginnings of Rome: Italy and Romefrom the Bronze Age
to the Punic Wars (c. 1000—264 bc). London: Routledge.
Courtney 1993: Edward Courtney (ed.), The Fragmentary Latin Poets. Oxford:
Clarendon Press.
Courtney 1995: E. Cortney (ed.), Musa Lapidaria: A Selection of Latin Verse Inscriptions
(American Classical Studies 36). Atlanta GA: Scholars Press.
Crawford 1974: Michael H. Crawford, Roman Republican Coinage. Cambridge Univer­
sity Press.
Crawford 1996: M.H. Crawford (ed.), Roman Statutes (BICS Supplement 64), two
volumes. London: Institute of Classical Studies.
Crawford 1998: M.H. Crawford, ‘Numa and the Antiquarians’, Faventia 20/1: 37-8.
Cristofani 1990: Mauro Cristofani (ed.), La grande Roma dei Tarquinii: catalogo della
mostra. Rome: L’Erma di Bretschneider.
Crook, Lintott and Rawson 1994: J.A. Crook, Andrew Lintott and the late Elizabeth
Rawson (eds), Cambridge Ancient History, second edition, vol. 9. 7 he Last Age oj the
Roman Republic, 146-43 BC. Cambridge University Press.
Csapo and Slater 1994: Eric Csapo and William J. Slater, The Context oj Ancient Drama.
Ann Arbor: University of Michigan Press.
Culham 1991: Phyllis Culham, ‘Documents and Donuts in Republican Rome,
Libraries and Culture 26: 119-34.
Curti 1993: Emmanuele Curti, ‘The Use of Myth in Roman Propaganda , unpub­
lished paper given at a conference in London, 23 November 1993.
Daremberg and Saglio 1896: Ch. Daremberg and Edm. Saglio (eds), Dictionnaire des
antiquites grecques et romaines, vol. 2.2. Paris: Hachette.
Dark 2000: Ken Dark, Britain and the End of the Roman Empire. Stroud: Tempos.
Davidson 2000: James Davidson, 'Gnesippos paigniagraphos: the Comic Poets and the
Erotic Mime’, in David Harvey and John Wilkins (eds). The Rivals oj Aristophanes:
Studies in Athenian Old Comedy (London: Duckworth and Classical Press of Wales):
41—64.
Davies 1984: J.K. Davies, ‘The Reliability of the Oral Tradition’, in Lin Foxhall and
John K. Davies (eds), The Trojan War: Its Historicity and Context (Bristol Classical
Press): 87-110.
Dearden 1995: C.W. Dearden, ‘Pots, Tumblers and Phlyax Vases’, in Alan Griffiths
(ed.), Stage Directions: Essays in Ancient Drama in Honour oj E.W. Handley {BICS
Supplement 66, London: Institute of Classical Studies): 81—6.
Degrassi 1947: Atilius Degrassi (ed.), Inscriptiones Italiae, XIII Fasti et elogia, fasc. 1 Fasti
consulares et triumphales. Rome: Libreria dello stato.
Degrassi 1963: Atilius Degrassi (ed.), Inscriptiones Italiae, XIII Fasti et elogia, fasc. 2 Fasti
 BIBLIOGRAPHY 327

iiimi Nuniani et luliani. Rome: Istituto poligrafico dello stato.

De Puma 1987: Richard Daniel De Puma, Etruscan Tomb-Groups: Ancient Potter}' and
Bronzes in Chicago’s Field Museum of Natural History. Mainz: von Zabem.
De Simone 1981: Carlo De Simone, ‘Gli Etruschi a Roma: evidenza linguistica e
problemi metodologici’, in Gli Etruschi e Roma: Atti dell incontro di studio in onore di
Massimo Pallottino (Rome: Giorgio Bretschneider): 94-103.
Dickie 2001: Matthew W. Dickie, Magic and Magicians in the Greco-Roman World.
London: Routledge.
Dierichs 1997: Angelika Dierichs, Erotik in der Romischen Kunst. Mainz: von Zabem.
Dorcey 1992: Peter F. Dorcey, The Cult if Silvanus: a Study in Roman Folk Religion
(Columbia Studies in the Classical Tradition 20). Leiden: Brill.
Drossart 1974: P. Drossart, ‘Le theatre aux nones caprotines (a propos de Varron, De
lingua Latina 6,18)’, Revue de philologie 48: 54-64.
Drummond 1989: A. Drummond, ‘Rome in the Fifth Century 11: the Citizen
Community’, in F.W. Walbank et alii (eds), Vie Cambridge Ancient History (ed. 2)
V11.2: The Rise of Rome to 220 bc (Cambridge University Press): 172-242.
Duff and Duff 1934: John W. Duff and Arnold M. Duff (eds), Minor Latin Poets (Loeb
Classical Library 284). Cambridge MA: Harvard University Press.
Duliere 1979: Cecile Duliere Lima Romana: recherche; d’iconographie et essai d mterpreta-
(Etudes de philologie, d’archeologte et d’histoire 18). Brussels: Institut
historique Beige de R ome.
Dumezil 1929: Georges Dumezil, Le probleme des centaures: etude de mythologie comparer
mdo-europeenne. Pans: Geuthner. ... ... v ■
Dimiezil 1970: Georges Dumezil, Archaic Roman Religion (trans. Philip Krapp, ong.
Duntzil "97?r‘%6)‘ Universit>' of C^ic^° Hisloires romaines. Pans: Galli-
“mezu 1973: Georges Dumezil, Mythe et epopee 111. num
mard.
1988: Georges ISnn^U. Al.r-Orn-e At X
»“"«» ./SeeeX,, (t«m. Derek Colrman. «• «•"- P” N™
York: Zone Books.
le printemps’, Annates de
UVal 1976: Yves-Marie Duval, ‘Les Lupercales, Junon et
iCwn'v “ T * “■ 253“72' I , de Consranrhople
aux Lupercales de
a 1J77- Yves-Marie Duval, ’Des Lupercales de C01
°nie , Rt'ime des etudes latines 55: 222—70. Rome: Palombi.
W1; „rl ( r (exhibition a > 8“■ &
>9’5: Helmut Engelmann, V,e Del- A-W
notes aux religion, orienries dans I’ Empire rorna'" h Univeniry Press.
1921: E.B. England, 71,e «/«»»•Source,.ndMom-
1983. Elalne .Sexua, comedy m Oeids W

«'»Ron"n L“'”y
m 1989: R. Eiaine Fantham, ‘Mime, tne
Fin/*0'7’’ ClassiMl ^orld 82: 153-63. Georgic and And-Georgic
: and Flora: Georgic
f-Lnr 1992: Elai„e FanIham, 'Cere,. Liber °
Cambridge Philological Society 38: 39-56.
enients in Ovid’s Fasti’, Proceedings of the C. Culturefrom Cicero to Apuleins. Bald-
1996: Elaine Fantham, Romat1 Literary
 328 UNWRITTEN ROME

more: Johns Hopkins University Press.

Fantham 1998: Elaine Fantham, Ovid: Fasti Book IV (Cambridge Greek and Latin
Classics). Cambridge University Press.
Faranda 1971: R. Faranda (trans.), Detti e fatti memorabili di Valerio Massimo. Turin:
Utet.
Feeney 1991: D.C. Feeney, The Cods in Epic: Poets and Critics of the Classical Tradition.
Oxford: Clarendon Press.
Feeney 2005: Denis Feeney, ‘The Beginnings of a Literature in Latin’, Journal of Roman
Studies 95: 226-40.
Feldherr 1998: Andrew Feldherr, Spectacle and Society in Livy's History. Berkeley:
University of California Press.
Finley 1964: M.I. Finley, ‘The Trojan War’, Journal of Hellenic Studies 84: 1-9.
Fishwick 1967: Duncan Fishwick, ‘Hastiferi’, Journal of Roman Studies 57: 142-160.
Flower 1995: Harriet I. Flower, ‘Fabulae Praetextae in Context: When Were Plays on
Contemporary Subjects Performed in Republican Rome?’, Classical Quarterly 45:
170-190.
Forsythe 1994: Gary Forsythe, The Historian L. Calpurnius Piso Frugi and the Roman
Annalistic Tradition. Lanham MD: University Press of America.
Forsythe 1999: Gary Forsythe, Livy and Early Rome: A Study in Historical Method and
Judgment (Historia Einzelschrift 132). Stuttgart: Franz Steiner.
Forsythe 2002: Gary Forsythe, ‘Dating and Arranging the Roman History of Valerius
Antias’, in Vanessa R. Gordon and Eric W. Robinson (eds), G'lkistes: Studies in
Constitutions, Colonies, and Military Power in the Ancient World, Offered in Honor of
AJ. Graham (Leiden: Brill): 99-112.
Foucher 1976: L. Foucher, ‘Flagellation et rites de fecondite aux Lupercales , Annales
de Bretagne et des pays de I’Ouest 83: 273.
Fox 1993: Matthew Fox, ‘History and Rhetoric in Dionysius of Halicarnassus , Journal
of Roman Studies 83: 31-47.
Fox 1996: Matthew Fox, Roman Historical Myths: The Regal Period in riiiyiisMn Litera­
ture. Oxford: Clarendon Press.
Fraccaro 1907: Plinio Fraccaro, Studi Varroniani: De gente populi Romani libri IV. Padua:
Draghi.
Fraschetti 2002: Augusto Fraschetti, Romolo il fondatore (Quadrante Laterza 112).
Rome: Laterza.
Frier 1979 (= 1999): Bruce W. Frier, Libri annales pontijicum maximorum: The Origins of
the Annalistic Tradition (Papers and Monographs of the American Academy in
Rome 27). Rome: American Academy (ed. 2 1999, Ann Arbor: University of
Michigan Press).
Gabba 1991: Emilio Gabba, Dionysius and the History of Archaic Rome (Sather Classical
Lectures 56). Berkeley: University of California Press.
Gabba 2000: Emilio Gabba, Roma arcaica: storia e storiografia (Storia e letteratura 205).
Rome: Edizioni di storia e letteratura.
Gallia 2007: Andrew B. Gallia, ‘Reassessing the “Cumaean Chronicle”: Greek
Chronology and Roman History in Dionysius of Halicarnassus’, Journal of Roman
Studies 97: 50-67.
 BIBLIOGRAPHY 329

Gantz 1993: Timothy Gantz, Early Greek Myth: a Guide to Literary and Artistic Sources,
two volumes. Baltimore: Johns Hopkins University Press.
Garland 1992: Robert Garland, Introducing Neu’ Gods: The Politics oj Athenian Religion.
London: Duckworth.
Gerhard 1868: E. Gerhard, Gesammelte akademische Abhandlungen mid kleine Schriften.
Berlin.
Gerhard et al. 1897: E. Gerhard et al., Etmskische Spiegel 5. Berlin: Georg Reimer.
Giancotti 1967: Francesco Giancotti, Mimo e gnome: studio su Decimo Laberio e Publilio
Siro (Biblioteca di cultura contemporanea 98). Messina: G. D Anna.
Gidlow 2004: Christopher Gidlow, The Reign ofArthur: From History to Legend. Stroud:
Sutton.
Giglioli 1935: G.Q. Giglioli, L’arte etmsca. Milan: Fratelli Treves.
Gjerstad 1969: E. Gjerstad, T’orsenna and Rome’, Opusciila RomanaT. 149-61.
Goldberg 1995: Sander M. Goldberg, Epic in Republican Rome. New York: Oxford
University Press.
Goldberg 2006: Sander Goldberg, ‘Ennius after the Banquet, Arethusa 39.3. 427 17.
Gordon 1999: Richard Gordon,’ ‘Imagining Greek and Roman Magic’, in Valerie
Flint et al., H'itchcra/i and Magic in Europe: Ancient Greece mid Rome (London:
Athlone Press): 159-275
Graham 1998: AJ. Graham, ‘T he Woman at the Window: Observations on the “Stele
from the Harbour” oi l hasos’, Journal of Hellenic Studies 118: 22-40.
Grandazzi 1991: Alexandie Grandazzi, Lu fondation de Rome: Reflexion sm Phistoire.
Paris: Les Belles Letters.
Grant 1971; Michael Gram, Roman Myths. New York: Scribners.
Greco 1985: Emanuele Greco, ‘Un santuario di eta repubblicana presso 11 foro di
Paestum’, La parola del passato 40: 223-32.
Green 1994: J.R. Green, Theatre in Ancient Greek Society. London: Routledge.
Green 1995: J.R. Green Theatrical Motifs in Non-Theatrical Contexts on Vases of
the Later Fifth and Fourth Centuries’, in Alan Griffiths (ed.), Stage Directions: Essays
Ancient Drama in Honour of EAT. Handley (BICS Supplement 66, London: Insti-
tute of Classical Studies) : 93^121. r
Greene 1938: G.C. Greene, Scholia Platonica (Philological Monographs 8). Haverfor-
due: American Philological Association. ,
Gnffin 1976: Miriam T. Griffin, Seneca: a Philosopher in Polit.cs. Oxford: Clarendon
Press.
Grijn 1985: Jasper Gnffin, Latin Poets and Roman Life. London: Duckworth. ,
Gnffin 1994: Miriam Gnffin, ‘The Intellectual Developments of the Ciceronian Age
‘"M- Crook, Andrew Linrott and Elizabeth Rawson, eds
W-^ry, ix (ed. 2) The Last Age of the Roman Republic, 146-43 bc (Cambndge
Gnffin 1997?APHSR GnffimH Commentary on Ovid Metamorphoses Book XI (Henna-

Did Early Roman History Come From.’,

GmenP ^Erichs. Gme“ Smd^ and Roman Policy (Cincinnati
 330 UNWRITTEN ROME

Classical Studies n.s. 7). Leiden: EJ. Brill.

Gruen 1992: Erich S. Gruen, Culture and National Identity in Republican Rome (Cornell
Studies in Classical Philology 52). Ithaca, N.Y.: Cornell University Press.
Guerrini 1981: Roberto Guerrini, Studi su Valerio Massimo (Biblioteca di studi antichi
28). Pisa: Giardini.
Gurval 1995: Robert Alan Gurval, Actium and Augustus: The Politics and Emotions of
Civil War. Ann Arbor: University of Michigan Press.
Gury 1998: Franfoise Gury, ‘A propos de 1’image des incubes Latins’, Melanges de I’Ecole
franfaise de Rome (Antiquite) 110: 995-1021.
Guthrie 1962: W.K.C. Guthrie, A History of Greek Philosophy, I The Earlier Presocratics
and the Pythagoreans. Cambridge University Press.
Habinek 2005: Thomas Habinek, The World of Roman Song: From Ritualized Speech to
Social Order. Baltimore: Johns Hopkins University Press.
Hanson 1959: John A. Hanson, Roman Theater-Temples (Princeton Monographs in An
and Archaeology 33). Princeton University Press.
Hardie 1996: Alex Hardie, ‘Pindar, Castalia and the Muses of Delphi (the Sixth
Paean)’, in Francis Cairns and Malcolm Heath (eds.), Papers of the Leeds International
Seminar, Ninth Volume (Leeds: Francis Cairns): 219-57.
Harries 2006: Jill Harries, Cicero and the furists: From Citizens' Law to the Lauful State.
London: Duckworth.
Harris 1989: William V. Harris, Ancient Literacy. Cambridge MA: Harvard University
Press.
Harrison 2000: George W.M. Harrison (ed.), Seneca in Performance. Condon: Duck­
worth and Classical Press of Wales.
Harvey 2004: David Harvey, ‘Herodotus Mythistoricus: Arion and rhe liar? (1.23—4) ,
in Vassos Karageorghis and loannis Taifacos (eds), The World oj Hctodotus (Nicosia:
University of Cyprus): 287-305.
Head 1911: B.V. Head, Historia Numorum (second edition). Oxford: Clarendon Press.
Hedreen 1992: Guy Hedreen, Silens in Attic Black-Figure Vase-Painting: Myth and
Performance. Ann Arbor: University of Michigan Press.
Hedreen 1994: Guy Hedreen, ‘Silens, Nymphs and Maenads’, Journal oj Hellenic Studies
114:47-69.
Hellegouarc’h 1963: J. Hellegouarc’h, Le vocalmlaire lathi des relations et des partis poh-
tiques sous la republique (Publications de la Faculte des Lettres et Sciences Humaines
de 1’Universite de Lille, 11). Paris: Les Belles Lettres.
Henderson 1979: A.A.R. Henderson, P. Ovidi Nasonis Remedia Amoris. Edinburgh:
Scottish Academic Press.
Henderson 2001: John Henderson, Telling Tales on Caesar: Roman Stories from Phaedrus.
Oxford University Press.
Herkommer 1968: E. Herkommer, Die Topoi in den Proomien der rbmischen
Geschichtswerke (diss.). Tiibingen.
Hodgkinson 1997: MJ. Hodgkinson, C. Licinius Macer and the Historiography of the
Early Republic. University' of Exeter PhD thesis.
Holleman 1974: A.W.J. Holleman, Pope Gelasius I and the Lupercalia. Amsterdam:
Hakkert.
 BIBLIOGRAPHY 331

Hollis 1998: A.S. Hollis, ‘A Tragic Fragment in Cicero, Pro Caelio 67?’, Classical Quar­
terly 48: 561-4.
Holloway 1994: R. Ross Holloway, The Archaeology of Early Rome and Latium.
London: Routledge.
Holscher 1978: Tonio Holscher, ‘Die Anfange romischer Reprasentationskunst’,
Mitteilungen des Deutschen Archaeologischen Instituts, Roemische Abteil 85. 315-57.
Horsfall 1972: Nicholas Horsfall, ‘Varro and Caesar: Three Chronological Problems’,
Bulletin of the Institute of Classical Studies 19: 120—8.
Horsfall 1979a: Nicholas Horsf'ill, ‘Stesichorus at BoviUae?’, Journal of Hellenic Studies
99: 26-48.
Horsfall 1979b: Nicholas Horsfall, ‘Epic and Burlesque in Ovid, Met. VlI1.260fF, Clas­
sical Journal 74: 319-32.
Horsfall 1985: Nicholas Horsfall, ‘C1L VI 37965 = CLE 1988 (Epitaph of Allia
Potestas): a commentary’, Zeitschrift fur Papyrologie mid Epigrapluk 61: 251-72.
Horsfall 1994: Nicholas Horsfall, ‘The Prehistory of Latin Poetry: Some Problems of
Method’, Rii’ista di Jilologia 122: 50-75.
Horsfall 2003: Nicholas Horsfall The Culture of the Roman Plebs. London: Duckworth.
Hubaux 1958: Jean Hubaux, Rome et Veies: Recherches sur la chronologic legendaire du
uwyen d^e rornain. Paris: Les Belles Lettres.
Hubinger 1992: Ulrich Hubmger ’On Pan’s Iconography and the Cult in the Sanc-
nury of Pan on the Slopes of Mount Lykaion’, in Robin Hagg (ed.) The Iconography
of Greek Cult in the Hi.h.if and Classical Periods (Kemos Supplement 1, Liege. Centre
d Etude de la Religion Grecque Antique): 189-212.
Hughes 1997: Alan Hughes. ’RONNAKIS: A Scene from the Comic Theatre , Echos
Ju "tondeclassique 41: 237-46. ... .,
Hunim 2000: Michel Humm, ’Spazio e tempo civici: ritorma delle tnbu e nforma del
calendario alia fine del quarto secolo a.C.’, m Christer Bruun (ed) 1 he Roman
^public: Politics, Religion and Historiography '.400-133 bc (Acta Instituti
Koniani Finlandiae 23, Rome): 91-119. .
Humphrey 1986; John H. Humphrey, Roman Circuses: Arenas for Chartot Racing.
London: B.T. Batsford
Hutton 2003: Ronald Hutton, Witches, Druids and King Arthur. London: Hambledon
and London.
J^°by 1962: Felix Jacoby, Die Fragment' der griechischen Historiker: Zweiter Toil B,

17: 44-74

?'■ Hen.g and

Jurgeit 1999 f 8, Oxford Comm di Karlsruhe’, Bollettino
jurge 11992: Fntzi Jurgeit, ‘Intervene! ottocenteschi sulla cista
d arte 74-75- 85-uu

” SC1-X
"" F?,a(Terra Italia 5), two volumes. Rome: Gruppo
 332 UNWRITTEN ROME

Kaibel 1899: Georgius Kaibel (ed.), Comicorum Craecorum fragmenta, I fasc. 1. Berlin:
Weidmann.
Keaveney 2003: Arthur Keaveney, ‘The Tragedy of Gaius Gracchus: Ancient Melo­
drama or Modem Farce?’, Klio 85: 322-32.
Keaveney 2005: Arthur Keaveney, ‘Sulla and the Games of Hercules’, L’Antiquite
Classique 74: 217-23.
Keaveney 2006: Arthur Keaveney, ‘Livy and the Theatre: Reflections on the Theory
of Peter Wiseman’, Klio 88: 510-15.
Klotz 1953: Alfredus Klotz, Tragicorum fragmenta (Scaenicorum Rontaiiorum fragmenta I).
Munich: Oldenbourg.
Kraay and Hirmer 1966: C.M. Kraay and M. Hirmer, Greek Coins. London: Thames
and Hudson.
Kragelund 1982: Patrick Kragelund, Prophecy, Populism, and Propaganda in the 'Octavia'
(Opuscula Graecolatina 25). Copenhagen: Museum Tusculanum.
Kragelund 1999: Patrick Kragelund, ‘Senecan Tragedy: Back on Stage?’, Classica et
Mediaevalia 50: 235-47.
Kragelund et al. 2002: Patrick Kragelund et alii, ‘Historical Drama in Ancient Rome:
Republican Flourishing and Imperial Decline?’, Symbolae Osloenses 77: 5—105.
Kraus 1994: Christina S. Kraus, “‘No Second Troy”: Topoi and Refoundation in
Livy, Book V’, Transactions of the American Philological Association 124: 267-89.
Kumaniecki 1970: K. Kumaniecki, ‘Les discours egares de Ciceton Pro Comelio ,
Mededelingen van de Koninklijke Academic voor Wetenschappen, latteren en Schone
Kimstcn van Belgie, KI. Lett. 32.4: 3-36.
Laird 1993: Andrew Laird, ‘Sounding Out Ecphrasis: Art and Text in Catullus 64,
Journal of Roman Studies 83: 18-30.
Lanciani 1891: Rodolfo Lanciani, ‘Miscellanea topografica: insigne Larario del Vico
Patrizio’, Bidlettino della commissione archeologica comunale di Roma 19: 305-1 1.
Lanciani 1990: Rodolfo Lanciani, Storia degli scavi di Roma, vol. 2. Rome: Quasar.
Latacz 2004: Joachim Latacz, Troy and Homer: Towards a Solution oj an Old Mystery.
Oxford University Press.
Leach 1993: E.W. Leach, ‘The entrance room in the House of Julius Polybius and the
nature of the Roman vestibulum’, in Eric M. Moonnann (ed.), functional and
Spatial Analysis of Wall Painting (Annual Papers on Classical Archaeology, Supple­
ment 3, Leiden: Babesch): 23-8.
Le Gall 1953: Joel Le Gall, Le Tibreflenve de Rome dans I’antiquite. Paris: Presses univer-
sitaires de France.
Leigh 1996: Matthew Leigh, ‘Varius Rufus, Thyestes and the Appetites of Antony ,
Proceedings of the Cambridge Philological Association 42: 171—97.
Liebeschuetz 1979: J.H.W.G. Liebeschuetz, Continuity and Change in Roman Religion.
Oxford: Clarendon Press.
Lepper and Frere 1988: Frank Lepper and Sheppard Frere, Trajan’s Column: a New
Edition of the Cichorius Plates. Gloucester: Alan Sutton.
Levene 1993: D.S. Levene, Religion in Livy (Mnemosyne Supplement 127). Leiden:
Brill.
Liberatore 1995: Daniela Liberatore, ‘Un Marsia nel Foro di Alba Fucens? Una
 332 UNWRITTEN ROME

Kaibel 1899: Georgius Kaibel (ed.), Comicorum Graeconmi fragmenta, I fasc. 1. Berlin:
Weidmann.
Keaveney 2003: Arthur Keaveney, ‘The Tragedy of Gaius Gracchus: Ancient Melo-
drama or Modem Farce?’, Klio 85: 322-32.
Keaveney 2005: Arthur Keaveney, ‘Sulla and the Games of Hercules’, L'Antiquite
Classique 74: 217-23.
Keaveney 2006: Arthur Keaveney, ‘Livy and the Theatre: Reflections on the Theory
of Peter Wiseman’, Klio 88: 510-15.
Klotz 1953: Alfredus Klotz, Tragicorum fragmenta (Scaenicorum Romanorum fragmenta I).
Munich: Oldenbourg.
Kraay and Hirmer 1966: C.M. Kraay and M. Hirmer, Greek Coins. London: Thames
and Hudson.
Kragelund 1982: Patrick Kragelund, Prophecy, Populism, and Propaganda in the ‘Octavia’
(Opuscula Graecolatina 25). Copenhagen: Museum Tusculanum.
Kragelund 1999: Patrick Kragelund, ‘Senecan Tragedy: Back on Stage?’, Classica et
Mediaevalia 50: 235-47.
Kragelund et al. 2002: Patrick Kragelund et alii, ‘Historical Drama in Ancient Rome:
Republican Flourishing and Imperial Decline?’, Symbolac Osloenses IT. 5-105.
Kraus 1994: Christina S. Kraus, '“No Second Troy”: Topoi and Refoundation in
Livy, Book V’, Transactions of the American Philological Association 124: 267—89.
Kumaniecki 1970: K. Kumaniecki, ‘Les discours egares de Ciceron Pro Comelio’,
Mcdedelingen van de Koninklijke Academic voor ll'etcnsihappen, Letteren en Schone
Kunsten van Belgie, KI. Lett. 32.4: 3—36.
Laird 1993: Andrew Laird, ‘Sounding Out Ecphrasis: Art and Text in Catullus 64’,
Journal of Roman Studies 83: 18-30.
Lanciani 1891: Rodolfo Lanciani, ‘Miscellanea topogralica: insigne Larario del Vico
Patrizio’, Bullettino della commissione archeologica conmnale di Roma 19: 305-11.
Lanciani 1990: Rodolfo Lanciani, Storia degli scavi di Roma, vol. 2. Rome: Quasar.
Latacz 2004: Joachim Latacz, Troy and Homer: Towards a Solution of an Old Mystery.
Oxford University Press.
Leach 1993: E.W. Leach, ‘The entrance room in the House of lulius Polybius and the
nature of the Roman vestibulum’, in Eric M. Moormann (ed.), Functional and
Spatial Analysis of Wall Painting (Annual Papers on Classical Archaeology, Supple­
ment 3, Leiden: Babesch): 23-8.
Le Gall 1953: Joel Le Gall, Le Tibrefleuve de Rome dans 1’antiquite. Paris: Presses univer-
sitaires de France.
Leigh 1996: Matthew Leigh, ‘Varins Rufus, Thyestes and the Appetites of Antony,
Proceedings of the Cambridge Philological Association 42: 171—97.
Liebeschuetz 1979: J.H.W.G. Liebeschuetz, Continuity and Change in Roman Religion.
Oxford: Clarendon Press.
Lepper and Frere 1988: Frank Lepper and Sheppard Frere, Trajan’s Column: a i\'etv
Edition of the Cichorius Plates. Gloucester: Alan Sutton.
Levene 1993: D.S. Levene, Religion in Livy (Mnemosyne Supplement 127). Leiden:
Brill.
Liberatore 1995: Daniela Liberatore, ‘Un Marsia nel Foro di Alba Fucens? Una
 BIBLIOGRAPHY 333

proposta di identificazione’, Ostralra 4- 249 b5 Clarendon Press.

ST 'Ti,75WR (W «-*
Artistry of the Fasti’, Latonius 34: W60-7- of March 3.523-710); a
Littlewood 1980: R.J. Littlewood, ‘Ovid and Studies Latin
Further Study in the Artistry of the Fast., m Cart P
Literature 2 (Collection Latomus 168, Brusse )- poUtics; Ovid at the
Littlewood 1981: R.J. Littlewood, ‘Poetic Artistry .
Liidi Megolenses (Fasti 4.179-372), Ciassiea « eta steinby (ed.), Lexicon »P‘
Liverani 1996: P. Liverani, ‘laniculum’, in Eva Marg3
graphicum Urbis Ronnie vol. IV (Koine: Quasar). Vases and Fragments
Luce 1930: Stephen Bleecker Luce, Attic , v 1 rie
Corinth’, American Journal oj Archaeology 34. rhssical Literature, in
Luck 1999: Georg Luck, ‘Witches and Sorcerers m Rollll. (Lon o •
Flint et a!., IVitchcraft and Magic in Europe. . "
Athlone Press): 91-158. „ a ‘Prime evidenze > °P
Lugli and Rosa 2>iOl; Francesca Lugli and Carlo dello
terrazzamentc del Capitolium nell’eta deli Bronx lte’, Bullettiiio dello coininis-
sione urcheolooim roniiinule di Roma 102: 28 ( (|lf ^te /in ia<
Lulof 1996: Patricia S. Lulof, The Ridge-Pole S'*1 *‘?;sis publishers. -des (22
Satricuin (Scnnium 9, Satricum 5). Amsterdam. jqexV £yegy by 11
Mace 1996: Sarah Mace, ‘Utopian and Eronc Fusi° 233-47- , c(h(|ics in
West-)’, Z-itschnJt Jiir Papyrologie mid Epigmp '' dtution and Aphroc i
MacLachlan 1992: Bonnie MacLachlan, ‘Sacred 1 . Trf(Uon, Unrest,
Religion/Sciences Religieuses 21/2: 145-62. Ri>HbHI Orller-
MacMullen 1966: Ramsay MacMullen, Ene"'1^ University E||ltlicily.
and Alienation in the Empire. Cambridge, N • . . CoImi'"',,i<”
Malkin 1998: Irad Malkin, The Ret',nb‘ l-' ‘ . (Arca 25). Leeds:
Berkeley: University of California Press- £(y»u>og'w
Maltby 1991; Robert Maltby. A Lexicon oj • " 2nd ed. Berlin:
Francis Cairns. .1(|1J Feldkidte- vo -
Mannhardt 1904: Wilhelm Mannhardt, H*' ' rofundissimi opus abs°-
Bomtraeger. . r:s et theolog* P* . pouched.
Mantuanus 1481: Fratris Baptistae Nb"tvUn'p'ltiers| in edibus ’ (j(eWrischen Gaining
ludssinmm de sacris diebus.. - P^'^^praetext^' SP"re" C‘"
Manuwald 2001: Gesine Manuwald, Fa ’» Beck. ylacieiit
Historiography.
T““”"

Press. prii'Otlel’1’" (H—s ■>" „T->fico'6””-

Marquadt 1886: Joachim Marquadt, Novara: Istituto ge b
chen Alterthumer VII). Leip^ Jtgli E^ Novara:
Martelli 1987: Marina MarteUi, La ‘
De Agostini.
 334 UNWRITTEN ROME

Mastrocinque 1983: Attilio Mastrocinque, ‘La cacciata di Tarquinio il Superbo:

tradizione romana e letteratura greca’, Athenaeum 61: 457—80.
Mastrocinque 1988: Attilio Mastrocinque, Lucio Giunio Bruto: Ricerche di storia, religionc
e diritto suite origini della repubblica romana. Trento: edizioni La Reclame.
Mazzarino 1966: Santo Mazzarino, Ilpensiero storico classico, II. 1. Bari: Laterza.
McKay 1975: A.G. McKay, Houses, Villas and Palaces in the Roman World. London:
Thames and Hudson.
McKeown 1979: J.C. McKeown, ‘Augustan Elegy and Mime’, Proceedings of the
Cambridge Philological Society 25: 71-84.
Menichetti 1995: Mauro Menichetti, Quoins forma virtutei parisuma fuit. . . Ciste prenes-
tine e cultura di Roma medio-repubblicana (Archeologia Perusina 12), Rome: Giorgio
Bretschneider.
Merivale 1969: Patricia Merivale, Pan the Goat-God: His Myth in Modern Times.
Cambridge MA: Harvard University Press.
Michelini 1987: Ann Norris Michelini, Euripides and the Tragic Tradition (Wisconsin
Studies in Classics). Madison: University of Wisconsin Press.
Michels 1951: Agnes Kirsopp Michels, ‘The Drama of the Tarquins’, Latomus 10: 13-
24.
Michels 1967: Agnes Kirsopp Michels, The Calendar of the Roman Republic. Princeton
University Press.
Millar 1964: Fergus Millar, A Study of Cassius Dio. Oxford: Clarendon Press.
Millar 1998: Fergus Millar, The Crowd in Rome in the Late Republic (Jerome Lectures
22). Ann Arbor: University of Michigan Press.
Miller 1916: Konrad Miller, Itineraria Romana: romische Reisewcge an der Hand der tabula
Peutingeriana dargestellt. Stuttgart: Strecker.
Moles 1993: John Moles, ‘Livy’s Preface’, Proceedings of the Cambridge Philological Asso­
ciation 39: 141-68.
Momigliano 1957: Arnaldo Momigliano, ‘Perizonius, Niebuhr and the Character of
Early Roman Tradition’, Journal of Roman Studies 47: 104—14.
Momigliano 1984: Arnaldo Momigliano, ‘Georges Dumezil and the Trifunctional
Approach to Roman Civilization’, History and Theory 23: 312-30.
Momigliano 1987: Arnaldo Momigliano, Ottauo contributo alia stona degli studi classici e
del mondo antico (Raccolta di studi e testi 161). Rome: Edizioni di storia e letter­
atura.
Momigliano 1989: A. Momigliano, ‘The Origins of Rome’, in F.W. Walbank et alii
(eds), The Cambridge Ancient History (ed. 2) VII.2: The Rise of Rome to 220 BC
(Cambridge University Press): 52-112.
Momigliano 1990: Arnaldo Momigliano, The Classical Foundations of Modem Historiog­
raphy (Sather Classical Lectures 54). Berkeley: University of California Press.
Mommsen 1860: Theodor Mommsen, Geschichte des rbmischen Miinzu’esens. Berlin:
Weidmann.
Mommsen 1864: Theodor Mommsen, Romische Forschungen, vol. 1. Berlin: Weid­
mann.
Mommsen 1887: Theodor Mommsen, Romische Staatstrecht (ed. 3), vol. 2.1. Leipzig:
Hirzel.
 bibliography 335

Mommsen 1893: Theodor Mommsen (ed.), Corpus Inscriptionuni Litinormn vol. 1 part
1 (ed.2). Berlin: Reimer.
Montanari 1976: Enrico Montanari, Roma: moment! di una presa di cosaenza soaale (Ctu
siamo 3). Rome: Bulzoni.
Moonon 1988: R.F. Moorton, ’The Genealogy of Latinus in Virgil’s Aeneitf, Transac­
tions of the American Philological Association 118: 253-9.
Morrow I960: Glenn Raymond Morrow, Plato’s Cretan City: A Historical Interpretation
of the Laws. Princeton University Press.
Mueller 2002: Hans-Friedrich Mueller, Roman Religion in Valerius Ahixiinns. London.
Routledge.
Munzer 1897: Friedrich Munzer, Beitriige
Berlin: Weidmann. ~nr Quellenkritik der blaturgeschichte des Plinius.
Munzer 1920: Friedrich Munzer,
Metzler. Rbinische Adebparteien und Adelsfamilien. Stuttgart:
Munzer 1932: F. Mtin;-
Miinzer 1937:1FriedriIV1.hM 0) • Paulys Real-Encyclopadie 15: 603-4.
logos 92: 47-67 andi .* ”nzcr> Die romischen Vestalinnen bis zur Kaiserzeit’, Philo-
.............................u:d 199—2">">
Munzer 1939: F Minir •
Nijboer et al. . UCtia (12) • Panlys Real-EncyclopiidieIX 768-70.
Santis, 'A Hitth Cl J Nijboer, J. van der Plicht, A.M. Bietti Sestieri, A. De
41/42: 163-76 * °n° tor the Earl>’ lron ASe >n Central Italy’, Palaeohistoria
Nock 1972: Arthur
Harby Nock Essays on
Clarendon
~“*<vuuon Kress Pr< Religion and the Ancient World. Oxford:
Nonh 1989:J.a. North. -
Religion iin ~ Republican Rome’, in F.W. Walbank et al.
(Canrb77,|
(Cambridge University prt History, VII part 2 (ed. 2) The Rise of Rome to 220 BC
Nonh 1992: J.A. North,‘De '
,0 Dr. H/j.jz Cnn'chshi ^tOnstru'-t*ng Stone Theatres’, in Apodosis: Essays presented
75~83. ’ 10 "Birk his Eightieth Birthday (London, St Paul’s School):
Nonh 2000a: J.A. North R
J-A. North,
University Press. ’ 0,1111,1 RAigion (New Surveys in the Classics 30). Oxford
North 2000b: J.A?'r;u
l. ,t ^'’ 1 rophet and Text in the Third Century BC’, in Edward
Bispham and Christ
Italy: Evidence j pv^t,'er
llH (eds). Religion in Archaic and Republican Rome and
Oakley 1997: J.H. Oak/’ (Edlnburgh University Press): 92-107.
623—9 and 8.2- 386—i?’ ^es*one • Lexicon Iconographicnm Mythologiae Classicae 8.1:
Oakley 1997: S P 11
Clarendon Press 3 ey’
Commentary on Livy Books VI-X, vol. I. Oxford:
Oakley 1998: S.P. Oaklev A r
Clarendon Press. Y’ Co"""e"'“ry on Livy Books M-X, vol. II. Oxford:
Oakley 2005: S.P o H
Clarendon Press Commentary on Livy Books l'l-X, vol. IV. Oxford:

Greek and R Clgden, Binding Spells: Curse Tablets and Voodoo Dolls in the
Ancient oman Worlds , in Valerie Flint et al., JL'itchcraft and Magic in Europe:
Anctent Greece and Rome (London: Athlone Press): 1-90.
 336 UNWRITTEN ROME

Ogden 2001: Daniel Ogden, Greek and Roman Necromancy. Princeton University Press.
Ogden 2002: Daniel Ogden, Magic, Witchcraft and Ghosts in the Greek and Roman
Worlds. New York: Oxford University Press.
Ogilvie 1965: R.M. Ogilvie, A Commentary on Livy Books 1—5. Oxford: Clarendon
Press.
Otis 1966: Brooks Otis, Ovid as an Epic Poet. Cambridge University Press.
Otto 1890: August Otto, Die Sprichtvorter nnd sprichivortlichen Redensarten der Romer.
Leipzig.
Otto 1913: W.F. Otto, ‘Die Luperci und die Feier der Lupercalien’, Philologus
72:161-95
Page 1941: D.L. Page, Select Papyri 3 (Loeb Classical Library). London: Heinemann.
Page 1973: Denys Page, ‘Stesichorus: the Geryoneis’, Journal of Hellenic Studies 93: 138—
54.
Pailler 1988: Jean-Marie Pailler, Bacchanalia: La repression de 186 av.J.-C. d Rome et en
Italic (BEFAR 270). Rome: Ecole fran?aise de Rome.
Pairault Massa 1992: Fran^oise-Helene Pairault Massa, Iconologia c politico nell'Italia
antica: Roma, Lazio, Etruria dal VII al I secolo a.C. (Biblioteca di archeologia 18).
Milan: Longanesi.
Pallottino 1981: Massimo Pallottino, Genti e culture dell’Italia preromana (Guide allo
studio della civilta romana 1.2). Rome: Jouvence.
Palombi 1997: Domenico Palombi, Tra Palatino cd Esquilino: Velia, Carinae, Fagutal
(RINASA Supplement 1). Rome: Istituto nazionale d’archeologia e storia dell’arte.
Panayotakis 1995: Costas Panayotakis, Theatrnm Arbitri: Theatrical Elements in the
Satyrica ofPctronins (Mnemosyne Supplement 146). Leiden: Brill.
Panciera 1986: Silvio Panciera, ‘Ancora sull’iscrizione di Cornelius Sums magister scrib-
arum poetarum’, Bullettino della commissionc archeologica comunale di Roma 91: 35-44.
Parke 1988: H.W. Parke, Sibyls and Sibylline Prophecy in Classical Antiquity. London:
Routledge.
Parker 1993: Hugh C. Parker, 'Romani tinmen soli: Faunus in Ovid’s Fasti’, Transactions
of the American Philological Association 123: 199-217.
Pasco-Pranger 2002: Molly Pasco-Pranger, ‘A Varronian Vatic Numa? Ovid’s Fasti
and Plutarch’s Life of Numa’, in D.S. Levene and D.P. Nelis (eds), Clio and the Poets:
Augustan Poetry and the Traditions of Ancient Historiography (Mnemosyne Supplement
224, Leiden: Brill): 291-312.
Pasco-Pranger 2006: Molly Pasco-Pranger, Founding the Year: Ovid’s Fasti and the
Poetics of the Roman Calendar (Mnemosyne Supplement 276). Leiden: Brill.
Pelling 1985: C.B.R. Pelling, ‘Plutarch and Catiline’, Hermes 113: 311-29.
Pelling 2002: Christopher Pelling, Plutarch and History: Eighteen Studies. London: Clas­
sical Press of Wales and Duckworth.
Pensabene 1998: Patrizio Pensabene, ‘Vent’anni di studi e scavi dell’Universita di
Roma “La Sapienza” nell’area sud-ovest del palatino (1977—1997)’, in Carlo
Giavarini (ed.), Il Palatino: area sacra sud-ovest e Dointts Tiberiana (Studia archaeo-
logica 95, Rome: L’Erma di Bretschneider): 1-154.
Peruzzi 1998: Emilio Peruzzi, Civilta greca nel Lazio preroinano (Studi dell’Accademia
Toscana ‘La Columbaria’ 165). Florence: Leo S. Olschki.
 BIBLIOGRAPHY 337

Peter 1906: Hermann Peter, Historiconim Romanomm reliquiae vol. 2. Leipzig: Teubner.
Phillips 1968: Kyle M. Phillips, jr, ‘Perseus and Andromeda’,/litiericaiiJournal ofArchae­
ology 72: 1—23.
Pick 1917: Behrendt Pick, ‘Ein Vorlaufer des Mephistopheles auf antiken Miinzen’,
Jahrbnch tier Goethe-Gesellschaft 4: 153-64.
Pick 1931: Behrendt Pick, Aujsiitze zur Niimismatik mid Archiiologic. Jena: Fronunann.
Pickard-Cambridge 1968: A.W. Pickard-Cambridge, The Dramatic Festivals of Athens
(rev. ed.). Oxford: Clarendon Press.
Piranomonte 2001: M. Piranomonte, ‘Annae Perennae nemus’, in Adriano La Regina
(ed.). Lexicon Topographicnm Urbis Romae: Sitbiirbimii 1, A-B (Rome: Quasar): 59-
63.
Piranomonte 2002: Marina Piranomonte (ed.), II santuario della mnsica e il bosco sacra di
Anna Perenna. Milan: Electa.
Pontrandolfo 2000; Angela Pontrandolfo, ‘Dioniso e personaggi fliacici nelle imagini
pestane’, Ostraka 9.1: 117-34.
1 orte 1985: Danielle Porte, L’etiologie religieuse Juns les Rules d'Ovide (Collection
d etudes anciennes). Paris: Les Belles Lettres.
Potscher 1984: Walter Potscher, ‘Die Lupercalia—eine Strukturanalyse’, Grazer
Beitrdge 11: 221-49.
1 oucet 1989: Jacques Poucet, ‘Reflexions sur 1'ecrit et 1’ecriture dans le Rome des
premiers siecle.', Gitonins 48: 285-311.
Poucet 2000: Jacques Poucet, Les Rois de Rome: tradition et historic. Louvain: Academic
royale de Belgique 2000).
Purcell 2003: Nicholas Purcell, ‘Becoming Historical: The Roman Case’, in David
Braund and Christopher Gill (eds). Myth, History and Culture in Republican Rome
(University of Exeter Press): 12-40.
Ramsey 1980: J. T. Ramsey, ‘A Reconstruction of Q. Gallius’ Trial for Ambitus: One
Less Reason for Doubting the Authenticity of the Commeiitariohmi Pelitionis',
FHstoria 29; 4( )2-21.
Rawson 1771: Elizabeth Rawson, 'The Literary Sources for the Pre-Marian Anny ,
Papers oj the British School at Rome 39: 13-31.
Rawson 1985: Elizabeth Rawson: Intellectual L'fe in the Lite Roman Republic. London:
Duckworth.
Rawson 1991; Elizabeth Rawson, Roman Culture and Society: Collected Papers. Oxford:
Clarendon Press.
Rawson 1993: Elizabeth Rawson, ‘The Vulgarity of the Roman Mime , in H.D.
Jocelyn and H.V. Hurt (eds), Tria Gistra: Essays and Notes Presented to John Pinsent
(Liverpool: LCM): 255-60.
Reynolds 1982: Joyce Reynolds, Aphrodisias and Rome: Documents from the Excavation of
the Theatre at Aphrodisias (JRS Monograph 1). London: Society for the Promotion
of Roman Studies.
Rhodes 1981: PJ. Rhodes, A Commentary on the Aristotelian Athenaion Politeia.
the Aristotelian
Oxford: Clarendon Press.
Ribbeck 1881: Otto Ribbeck, ‘Ein historisches Drama’, Rheinisches Museum 36: 321-
2.
 338 UNWRITTEN ROME

Ribbeck 1887: Otto Ribbeck, Tragicorum Romanorum fragmenta, ed. 3. Leipzig:

Teubner.
Richard 1983: Jean-Claude Richard, Pseudo-Aurelius Victor, Les origines du people
romain. Paris: Les Belles Lettres.
Richter 1901: Otto Richter, Topographic der Stadt Rom (Handbuch der klassischen
Altertums-Wissenschaft 3.3.2). Munich: Beck.
Ridgway 1988: David Ridgway, ‘The Etruscans’, in The Cambridge Ancient History (ed.
2), vol. IV Persia, Crcece and the Western Mediterranean c. 525—479 B.C. (Cambridge
University Press): 634—75.
Ridgway 1992: David Ridgway, The First Western Creeks. Cambridge University
Press.
Ridgway 1996: David Ridgway, ‘Greek Letters at Osteria dell’Osa’, Opuscula Romana
20: 87-97.
Ritschl 1862: Fridericus Ritschelius (ed.), Priscae Latinitatis monumenta epigraphica ad
archetyporumJidem exemplis lithographis repraesentata. Berlin: Reimer.
Rose 1928: HJ. Rose, A Handbook of Creek Mythology, Including its Extension to Rome.
London: Methuen.
Rose and Price 1996: HJ. R(ose) and S.R.F. P(rice), ‘Calendar, Roman’, in Simon
Hornblower and Anthony Spawforth (eds), The Oxford Classical Dictionary (ed. 3,
Oxford University Press): 274.
Rosier 1990: Wolfgang Rosier, ‘Mnemosyne in the Symposion', in Oswyn Murray (ed.),
Sympotica: a Symposium on the Symposion (Oxford: Clarendon Press): 230-7.
Siiflund 1932: Gosta Saflund, Le mura di Roma reptibblicana: Saggio di archeologia romana
(Acta Instituti Romani Regni Sueciae 1). Lund: Gleerup.
Santangeli Valenziani and Volpe 1996: R. Santangeli Valenzani and R. Volpe, ‘Nova
Via’, in E.M. Steinby (ed.), Lexicon Topographicum Urbis Romae 3, H-O (Rome:
Quasar): 346-9.
Schauenberg 1981: Konrad Schauenberg, ‘Andromeda’, Lexicon Iconographicum
Mythologiae Classicae 1.1: 774—90 and 1.2: 622—42.
Schmidt 1989: Peter Lebrecht Schmidt, ‘Postquam Indus in artem paulatim verterat:
Varro und die Friigeschichte des romischen Theaters’, in Gregor Vogt-Spira (ed.),
Studien zur vorliterarien Periode im friihen Rom (ScriptOralia 12, Tubingen: Gunter
Narr): 77-133.
Schofield 1986: Malcolm Schofield, ‘Cicero For and Against Divination , Journal oj
Roman Studies 76: 47-65.
Schultz 2006: Celia E. Schultz, Women’s Religious Activity in the Roman Republic.
Chapel Hill: University of North Carolina Press.
Schumacher 1968-9: W.W. Schumacher, ‘Antikes und Christliches zur Auspeitschung
der Elia Afanacia’, Jahrbuch fiir Antike und Christentum 11-12: 65-75.
Schweitzer 1955: B. Schweitzer, ‘Zum Krater des Aristonothos’, Ronrische Mitteilungen
62: 78-106.
Sciarrino 2006: Enrica Sciarrino, ‘The Introduction of Epic in Rome: Cultural Thefts
and Social Contexts’, Aretlmsa 39.3: 449-69.
Seaford 1984: Richard Seaford, Euripides Cyclops, with Introduction and Commentary.
Oxford: Clarendon Press.
 BIBLIOGRAPHY 339

Seaford 1994: Richard Seaford, Reciprocity and Ritual: Homer and Tragedy in the Devel­
oping City-State. Oxford: Clarendon Press.
Segal 1968: Charles P. Segal, ‘Circean Temptations: Homer, Vergil, Odd’, Transac­
tions of the American Philological Association 99: 419-42.
Serrati 2006: John Serrati, ‘Neptune’s Altars: The Treaties between Rome and
Carthage (509-226 Be)’, Classical Quarterly 56: 113-34.
Sherk 1969: Robert K. Sherk, Roman Documents from the Greek East: Senatus consnlta
anil epistulae to the Age of Augustus. Baltimore: Johns Hopkins University Press.
Silk 2000: M.S. Silk, Aristophanes ami the Definition of Comedy. Oxford University Press.
Simon 1982: Erika Simon, The Ancient Theatre (trans. C.E. Vafopoulou-Richardson).
London: Methuen.
Simon 1997: Erika Simon, ‘Silenoi’, Lexicon Iconographicinn Mytliologiae Classicae 8.1:
1108-1133.
Skidmore 1996: Clive Skidmore, Practical Ethics for Roman Gentlemen: the Ubrfc of
Valerius Maximus. University of Exeter Press.
Skutsch 1968: Otto Skutsch, Stiulia Enniana. London: Athlone Press.
Skutsch 1985: Otto Skutsch, The Annals oj Quintus Ennius. Oxford: Clarendon Press.
Slater 2005: WJ. Slater, ‘Mimes and Mancipes', Phoenix 59: 316-23.
Smith 1996: Christopher John Smith, Early Rome ami Litiuni: Economy and Society
c. 1000 to 500 BC Oxtord: Clarendon Press.
Smith 2003: Joseph A. Smith, ‘Flavian Drama: Looking Back with Octavia’, in A.J.
Boyle and WJ. Dominik (eds), Flavian Rome: Culture, hmye, Text (Leiden: Brill):
391-430.
Smits 1946: Elisabeth Clementine Henriette Smits, Faumis. Leiden: Leidsche Uitgev-
ersmaatschappij
Solin 1983: Heikki Solin, ‘Varia onomastica: V. K/.tllf/.oJ, Zeitschrift fiir Papyrologie
mid Epigraphik 51: 180—2.
Solin and Brandenburg 1980: H. Solin and H. Brandenburg, ’Paganer Fruchtbarkeits-
ritus und Martyriumsdarstellung?: zum Grabrelief der Elia Afanacia im Museum
der Pratextat-Katakombe zu Rom’, Archaologische Anzeiger 1980: 271-84.
Soltau 1909: Wilhelm Soltau, Die Anfange der roemischen Geschichtsclireibimg. Leipzig.
Spaeth 1996: Barbette Stanley Spaeth, The Roman Goddess Ceres. Austin: University of
Texas Press.
Steffen 1952: Victor Steffen, Satyrographoimini Graeconmi.fr.ymeiita. Poznan: Poznahskie
Towarzystwo Przyjaciol Nauk.
Stem 1968: H. Stem, ‘Un calendrier remain illustre de Thysdrus’, in lardo antico e alto
niedioei’o: la Jorma artistica nel passaggio dall’antichita al medioevo (Rome: Accademia
del Lincei): 177-200.
Stibbe et al. 1980: C.M. Stibbe, G. Colonna, C. De Simone, H. Versnel, Lapis Satn-
canus: Archaeological, epigraphical, linguistic and historical aspects oj the neii’ inscription fron
from
Satricum (Archeologische Studien van het Nederlands Instituut te■ Rome, Scripta
Maatschappelijk
Minora 5). The Hague: Ministerie van Cultuur, Recreatie en 1
Werk.
Storchi Marino 1999: Alfredina Storchi Marino, Numa e Pitagora: sapientia constituendae
civitatis. Naples: Liguori Editore.
 340 UNWRITTEN ROME

Sumner 1973: G.V. Sumner, The Orators in Cicero’s Brutus: Prosopography and
Chronology (Phoenix Supplementary volume 11). University of Toronto Press.
Syme 1939: Ronald Syme, The Roman Revolution. Oxford: Clarendon Press.
Syme 1958: Ronald Syme, Tacitus. Oxford: Clarendon Press.
Szilagyi 1981: J.C. Szilagyi, ‘Impletae modis saturae’, Prospettiva 24: 2—23.
Taplin 1993: Oliver Taplin, Comic Angels, and Other Approaches to Greek Drama through
Vase-Paintings. Oxford: Clarendon Press.
Taplin 2000: Oliver Taplin, ‘The Spring of the Muses: Homer and Related Poetry’, in
Oliver Taplin (ed.), Literature in the Greek World (New York: Oxford University
Press): 4-39.
Taylor 1939: Lily Ross Taylor, ‘Cicero’s Aedileship’, American Journal of Philology 60:
194-202.
Taylor 1960: Lily Ross Taylor, The Voting Districts of the Roman Republic (Papers and
Monographs of the American Academy in Rome 20). Rome: American Academy.
Taylor 1962: Lily Ross Taylor, ‘Forerunners of the Gracchi’, Journal of Roman Studies
52: 19-27.
Timpe 1988: Dieter Timpe, ‘Miindlichkeit und Schriftlichkeit als Basis der friihromis-
chen iiberlieferung’, in Jiirgen von Ungem-Sternberg and Hansjbrg Reinau (eds),
Vergangenheit in muitdlichcr Uberlieferung (Colloquium Rauricum 1, Stuttgart:
Teubner): 266-86.
Tomasetti 1910: Giuseppe Tomasetti, La campagna di Roma antica, medioevale e modema,
vol. 2. Rome.
Torelli 1982: Mario Torelli, Typography and Structure of Roman Historical Reliefs (Jerome
Lectures 14). Ann Arbor: University of Michigan Press.
Torelli 1984: Mario Torelli, Lavittio e Roma: riti iniziatrici e matrimoi'io tra archeologia e
storia. Rome: Quasar.
Torelli 1993: Mario Torelli, ‘Gli aromi e il sale: Afrodite ed Encl neli emporia arcaica
dell’Italia’, in Attilio Mastrocinque (ed.), Ercole in occideute (Labirinti 2, Universita
degli studi di Trento): 91—117.
Torelli 1996: Mario Torelli, ‘Riflessioni sulle registrazioni storiche in Etruria , Eutopia
5: 13-22.
Torelli 1997: Il rango, il rito e I’immagine: alle origini della rappresentazione storica romana.
Milan: Electa.
Torelli 1999: Tota Italia: Essays in the Cultural Foundation of Roman Italy. Oxford:
Clarendon Press.
Tortorella 2000: Stefano Tortorella, ‘L’adolescenza dei gemelli, la festa dei Lupercalia e
1’uccisione di Amulio’, in Andrea Carandini and Rosanna Cappelli (eds), Roma:
Romolo, Remo e la fondazione della cittd (Milan: Electa): 244—55.
Traversari 1960: Gustavo Traversari, Gli spettacoli in acquit nel teatro tardo-antico. Rome:
L’Enna di Bretschneider.
Travlos 1971: A Pictorial Dictionary of Ancient Athens. London: Thames and Hudson.
Trendall 1959: A.D. Trendall, Phlyax Vases (BICS Supplement 8). London: Institute
of Classical Studies.
Trendall 1967: A.D. Trendall, The Red-Figured Vases of Lucania, Campania and Sicily.
Oxford: Clarendon Press.
 BIBLIOGRAPHY 341

Trendall 1987: A.D. Trendall, The Red-Figured Vases of Paestum. London: British
School at Rome.
Trendall 1989: A.D. Trendall, Red-Figure Vases of South Italy and Sicily: a Handbook.
London: Thames and Hudson.
Trendall and Cambitoglou 1982: A.D. Trendall and A. Cambitoglou, The Red-Figured
Poses of Apulia, vol. 2. Oxford: Clarendon Press.
Ulf 1982: Christoph Ulf, Das romische Lupercalienfest (Impulse der Forschung 38).
Darmstadt: Wissenschaftliche Buchgesellschaft.
Ungem-Stemberg 1986: Jurgen von Ungem-Stemberg, ‘The Formation of the
“Annalistic Tradition”; the Example of the Decemvirate’, in Kurt A. Ra.itl.iub
(ed.), Social Struggles in Archaic Rome: Netv Perspectives on the Conflict of the Orders
(Berkeley: University of California Press): 77-104.
Ungem-Stemberg 1988: Jurgen von Ungem-Stemberg, ‘Uberlegungen zur friihen
romischen iiberlieferung im Lichte der Oral-Tradition-Forschung’, in Jurgen von
Ungem-Stemberg and Hansjbrg Reinau (eds), Vergangenheit in miindlidier L'ber-
lieferung (Colloquium Rauricum 1, Stuttgart: Teubner): 237-65.
Vaahtera 1993: Jyri Vaahtera, ‘On the Religious Nature of the Place of Assembly’, in
Unto Paananen er al. (eds), Senatus Populusque Romanns: Studies in Roman Repub­
lican Legislation (Acta Instituti Romani Finlandiae 13, Helsinki University Press):
97-116.
Vanggaard 1988: Jens H. Vanggaard, The Flarnen: .4 Study in the History and Sociology of
Roman Religion. Copenhagen: Museum Tusculanum.
Van Sickle 1986: John Van Sickle, Poesia e potere: il niito Virgilio. Bari: Laterza.
Versnel 1976: H.S Versnel, review of Alfoldi 1974, Bibliotheca Orientalis 33: 391-401.
Versnel 1993: H.S. Versnel, 'Transition and Reversal in Myth and Ritual (Studies in Greek
and Roman Religion 6.2). Leiden: Brill.
Veyne I960: Paul Veyne, ‘Iconographie de la “transvectio equitum” et des Luper-
cales’. Revue des etudes anciennes 62: 100-10.
Veyne 1961: Paul Veyne, *Le Marsyas “colonial” et 1’independence des cites’, Rei'tie de
philologie 35: 87-98.
Viscogliosi 1993: A. Viscogliosi, ‘Circus Flaminius’, in Eva Margareta Steinby (ed.),
Lexicon Topographicum Urbis Romae 1 (Rome: Quasar): 269-72.
VonStaden 1989: Heinrich von Staden, Herophilus: Ute Hrt of Medicine in Early Alexan­
dria. Cambridge University Press.
Wachter 1987: Rudolf Wachter, Altlateinischer Inschriften: sprachliche mid epigraphische
Untersuchungen zu den Docuinenten bis etiva 150 v.Chr. Bern: Peter Lang.
Walbank 1957: F.W. Walbank, .4 Historical Conunentury on Polybius, vol. 1. Oxford.
Clarendon Press.
Walker 1993: Andrew D. Walker, ‘Enargeia and the Spectator in Greek Historiog­
raphy’, Transactions of the American Philological Association 123: 353-77.
Wallace-Hadrill 1997: Andrew Wallace-Hadrill, ‘Mutatio morum: The Idea of a
Cultural Revolution’, in T. Habinek and A. Schiesaro (eds), The Roman Cultura
Revolution (Cambridge University Press): 3-22.
Wardle 1998: D. Wardle, Valerius Maximus: Memorable Deeds and Sayings Boa 1.
Oxford: Clarendon Press.
 342 UNWRITTEN ROME

Wardman 1974: Alan Wardman, Plutarch’s Lives. London: Elek.

Watson 1974: Alan Watson, Law Making in the Later Roman Republic. Oxford:
Clarendon Press.
Weinstock 1971: Stefan Weinstock, Diems Julius. Oxford: Clarendon Press.
Weis 1992: Anne Weis, ‘Marsyas I’, Lexicon Iconographicum Mythologiae Classicae 6.1:
366-78.
Weiwei 1967: Karl-Wilhelm Weiwei, ‘Das Angebot des Diadems an Caesar und das
Luperkalienproblem’, Historia 16: 44—69.
Weiwei 2000: Karl-Wilhelm Weiwei, ‘Lucius lunius Brutus—ein fictiver Revolution-
sheld’, in Karl-Joachim Holkeskamp and Elke Stein-Holkeskamp (eds), Von
Romulus zu Augustus: grosse Cestalten der rbmischen Repnblik (Munich: 2000): 48-57.
West 1966: M.L. West, Hesiod: Theogony. Oxford: Clarendon Press.
West 1978: M.L. West, Hesiod: Works and Days. Oxford: Clarendon Press.
West 1989: M.L. West, ‘The Early Chronology of Attic Tragedy’, Classical Quarterly
39:251-4.
Wirszubski 1950: Ch. Wirszubski, Libertas as a Political Idea at Rome During the Late
Republic and Early Empire. Cambridge University Press.
Wiseman 1974: T.P. Wiseman, China the Poet and Other Roman Essays. Leicester
University Press.
Wiseman 1976: T.P. Wiseman, ‘Two Questions on the Circus Flaminius’, Papers of the
British School at Rome 44: 44—7.
Wiseman 1979: T.P. Wiseman, Clio’s Cosmetics: Three Studies in Greco-Roman Litera­
ture. Leicester University Press.
Wiseman 1981: T.P. Wiseman, ‘The Temple of Victors’ on the Palatine’, Antiquaries
Journal 61: 35-52.
Wiseman 1983: T.P. Wiseman, ‘The Wife and Children of Romulus’, Classical Quar­
terly 33: 445-52.
Wiseman 1985: T.P. Wiseman, Catullus and His World: A Reappraisal. Cambridge
University Press.
Wiseman 1987a: T.P. Wiseman, Roman Studies Literary and Historical. Liverpool:
Francis Cairns.
Wiseman 1987b: T.P. Wiseman, ‘Con Boni nel foro: I diari romani di W. St. Clair
Baddeley’, Rivista dell’Istituto Nazionale d'Archeologia e Storia dell'Arte serie 3, 8-9:
119-49.
Wiseman 1988: T.P. Wiseman, ‘Satyrs in Rome?: The Background to Horace’s rlrs
poetica’, Journal of Roman Studies 78: 1-13.
Wiseman 1991: T.P. Wiseman, Flavius Josephus: Death of an Emperor (Exeter Studies in
History 30). University of Exeter Press.
Wiseman 1993a: T.P. Wiseman, ‘Lying Historians: Seven Types of Mendacity’, in C.
Gill and T.P. Wiseman (eds), Lies and Fiction in the Ancient World (University of
Exeter Press): 122-46.
Wiseman 1993b: T.P. Wiseman, ‘The She-Wolf Mirror: an Interpretation’, Papers of
the British School at Rome 61: 1—6.
Wiseman 1994: T.P. Wiseman, Historiography and Imagination: Eight Essays on Roman
Culture. University of Exeter Press.
 BIBLIOGRAPHY 343

Wiseman 1995: T.P. Wiseman, Remus: a Roman Myth. Cambridge University Press.
Wiseman 1996: T.P. Wiseman, ‘What Do We Know About Early Rome?’, Journal of
Roman Archaeology 9: 310—15.
Wiseman 1998: T.P. Wiseman, Roman Drama and Roman History. University of Exeter
Press.
Wiseman 2004a: T.P. Wiseman, The Myths q) Rome. University of Exeter Press.
Wiseman 2004b: T.P. Wiseman, ‘Where was the Nova Via?’. Papers of the British School
at Rome 62: 167—83.
Wiseman 2006a: T.P. Wiseman, ‘Andrea Carandini and Roma Qnadrafa’, Accordia
Research Papers 10: 103—26.
Wiseman 2006b: T.P. Wiseman, ‘The Cult Site of Anna Perenna: Documentation,
Visualization, Imagination’, in Lothar Haselberger and John Humphrey (eds),
Imaging Ancient Rome: Documentation—Visualisation—Imagination (JR.4 Supplement
61, Portsmouth I<1): 51-61.
Wright 2005: Matthew Wright, Euripides’ Escape-Tragedies: H Study of Helen,
Andromeda, and Iphigeneia among the Taurians. Oxford University Press.
Zetzel 1995: James E.G. Zetzel, Cicero De re publica: Selections. Cambridge University
Press.
Zevi 1968. Fausto Zevi, Considerazioni sull’elogio di Scipione Barbato’, Studi iiiisccl-
lanei 15: 69-73.
Zevi 1993: Fausto Zevi, Per 1 identificazione della Porticus Minucia Fnimentaria ,
Melanges de I’hcole frangaise de Rome (Antiquite) 105: 661-708.
Zevi 1995: Fausto Zevi. ‘Demarato e i re “corinzi” di Roma’, in A. Storchi Marino
(ed.), L iniidiiiv.i dell antico: studi in memoria di Ettore Lepore, 1 (Naples: Luciano
editore): 291 -_> 14.
Zevi 2005: Fausto Zevi, ‘From Ancus Marcius to Apollodorus ofDamascus: Notes on
the Port Area of Ostia’, in Simon Keay, Martin Millett, Lidia Paroli and Kristian
Strutt (eds), Tortus (Archaeological Monographs 15, London: British School at
Rome): 30-42.
Ziolkowski 1989; Adam Ziolkowski, ‘The Sacra Via and the Temple ofJupiter Stator’,
Opuscula Romana 17: 225-39.
Ziolkowski 2004: Adam Ziolkowski, Sacra I ia Tuwity Years After (Journal of Junstic
Papyrology, Supplement 3). Warsaw: Raphael Taubenschlag Foundation.
Zorzetti 1980: Nevio Zorzetti, Lt pretesta e il teatro latino arcaico (Forme matenah e
ideologic del mondo antico 14). Naples: Liguori.
Zorzetti 1990: Nevio Zorzetti, ‘The Carmina Com'ivalia', in Oswyn Murray (ed.),
Sympotica: a Symposium on the Symposion (Oxford: Clarendon Press): 289-307.
Zwierlein 1966: Otto Zwierlein, Die Rezitationsdramen Senecas (Beitr'age zur klassis-
clien Philologie 20). Meisenheim am Gian: Hain.
 Illustration Credits
Fig. 1 De Simone 1981 tav. 111(a): Giorgio Bretschneider editore, Rome.
Fig. 2 Rome, Antiquarium forense, inv. 1918: Soprintendenza per i Beni Archeo-
logici di Roma.
Fig. 3 Richter 1901.364 (Abb. 32).
Fig. 4 Richter 1901, Taf. 5 (opposite p. 42).
Fig. 5 Ritschl 1862, pl. I (a).
Fig. 7 Piranomonte 2002.18: Soprintendenza per i Beni Archeologici di Roma.
Fig. 8 Piranomonte 2002.26: Soprintendenza per i Beni Archeologici di Roma.
Fig. 9 Barre 1840, fig. 27b: British Library.
Fig. 10 Gerhard et al. 1897.54.
Fig. 11 Gerhard et al. 1897.172.
Fig. 12 © The Trustees of the British Museum.
Fig. 14 Gerhard et al. 1897.52-3.
Fig. 15 Gerhard et al. 1897.51-2.
Fig. 16 German Archaeological Institute, Rome: Felbermeyer, Nc-g. D-DAI-Rom
1932.0120.
Fig. 17 Monunienti dell’Instituto 9, tav. LVIII.
Fig. 18 Moniunenti dell’Instituto, Supplement, tav. XIX.
Fig. 19 Drawing by Dr F. Jurgeit Blanck (Jurgeit 1999.1.528-33, 2.754-5).
Fig. 20 Monunienti dell’Instituto 10, tav. XLV1.
Fig. 21 Istituto di Studi sulle Civilta Italiche e del Mediterraneo Amico, Rome.
Fig. 22 Monunienti dell'Instituto 9, tav. XXIII.
Fig. 23 Monuiiieiiti dell’Instituto 9, tav. XXII.
Fig. 24 Gerhard et al. 1897, no. 120.
Fig. 25 The Walters Art Museum, Baltimore (54.136).
Fig. 26 Istituto di Studi sulle Civilta Italiche e del Mediterraneo Antico, Rome.
Fig. 27 Gerhard et al. 1879, no. 44.
Fig. 28 Soprintendenza per i Beni Archeologici dell’Etruria Meridionale (foto n.
224535).
Fig. 29 Soprintendenza archeologica per le province di Salerno, Avellino e
Benevento.
Figs 30-31 Istituto di Studi sulle Civilta Italiche e del Mediterraneo Antico, Rome.
Fig. 32 Soprintendenza per i Beni Archeologici di Puglia, inv. 8154.
Fig. 33 Soprintendenza per i Beni Archeologici dell’Etruria Meridionale (foto n.
4559).
Fig. 34 Soprintendenza per i Beni Archeologici dell’Etruria Meridionale (foto n.
117795).
 ILLUSTRATION CREDITS 345

Fig. 35 © The Trustees of the British Museum.

Fig. 36 © The Cleveland Museum of Art, 1999. John L. Severance Fund, 1989.73.
Fig. 37 © Monumenti musei e gallerie ponteficie (Museo Gregoriano, foto n.
XXXV.31.16).
Fig. 38 Musee d’art et d’histoire, Geneva (inv. 23471): photo Yves Siza.
Fig.s 39-42 Istituto di Studi sulle Civilta Italiche e del Mediterraneo Antico, Rome
(fig. 42 redrawn by Sean Goddard).
Fig. 43 Monumenti dell’Instituto, Supplement, tav. XV.
Fig. 44 Soprintendenza per i Beni Archeologici delle province di Napoli e Casetta.
Fig. 46 © The Trustees of the British Museum.
Fig.s 47-54 Drawn by Sean Goddard.
 Index of Passages

[ARISTOTLE] De republica 2.18—19 125

Ar/ienaioii politeia 22.4 300 2.36 168
In yatiiiimn 14 157
ARNOBIUS In Ferrelli 5.36 169, 179
Adtiersus tiationes 5.1 164, 166 Pro Q. Gallio fr. 4P 214
Pro Roscio Anierino 67 212
AUGUSTINE Pro Roscio comoedo 23 213
City of Cod 2.27 179
3.16 310 CLEMENT OF ALEXANDRIA
6.7 218 Stromaleis 1.108.3 55
10.16 247-8
18.13 246 DIO CASSIUS
22.11 248 41.14.4 46
50.10.3- 4 132
CALLIMACHUS 57.18.4- 5 47
Aitia 4.106 233
DIODORUS SICULUS
CATULLUS 1.2.2 268
64.60-70 222 3.58.3 128
64.251-3 222
64.285-93 284 DIOMEDES
Grainiiiatici Luiiii 1.483—1 Keil 39
CICERO
Ad Atticiim 4.15.6 215 DIONYSIUS OF HALICARNASSUS
Ad familiares 7.11.2 215 1.74.5 13
De diuinatione 1.12 251 1.77.1- 3 253- 4
1.33 252, 263 1.84.1 254
1.40-3 252 2.2.3 254
1.44-5 294-5 2.56.2-6 254- 5
1.101 286 2.61.1 255
1.114-15 42 2.68-9 249, 255
2.27 251 4.2.1 280
2.80 252 4.2.4 282, 289
2.115 252 4.63.1- 2 295
2.149 267 5.33.1- 4 181
2.150 251 5.56.1 255
De legibus 1.1-5 249-51 6.10.1 172
2.15-16 263 6.17.2- 4 172
De natura deonim 2.6 261 6.94.3 172
De oratorc 2.63 259 8.56.1-4 256
 INDEX OF PASSAGES 347

ERATOSTHENES LIVY
FGrH 241 F 45 49, 233 Pref. 6 243
Pref. 7 244
ENNIUS Pref. 10 260
Annales 15—16Sk 268 1.5.1-2 62
154—5Sk 47-8 1.16.4 245
206—7Sk 39, 240 1.20.7 155
1.31.8 156
C. FANN1US 1.35.9 168
Oratio de sociis fr. 3 Malcovati 167 1.39.1 282
1.40.5 282-3
FESTUS 1.41.4 276, 285, 291
75-6L 77 1.56.4 288-9, 296
103L (Paulus) 85 1.56.9 297
326-8L 302 1.57.4-11 313
478-80L 149 2.2.3 299-300
2.7.6-7 318
FLOROS 317
2.14.1
1.10.3 180, 292 3.25.8 31
3.29.5 26
‘GELAS1US’ 31
4.19.1
Aduersus Androiitachum 16—20 81-2 4.20.2 26
4.53.11-12 27
A. GELLIUS 266
5.21.9
16.5.2-3 283 5.49.1 31,32
5.49.5 31
HERODOTUS 27
1.5.3 5.49.7
8 7.2.5 25
7.3.5-7 10, 299,318
HESIOD 31
Theogony 101 1-16 , 7.10.4
232’ 7.10.9 33
7.10.13 27-8
HOMER 28
Odyssey 11.14-19 7.17.5
56 7.38.3 28
32
HORACE 10.29.4
29
Hrs poetica 234-50 10.30.9
219 171-2
285-8 23.30.17
194, 207-8 24
Epistles 2.1.139—44 27.37.7
125 24
2.1.155-63
27.37.13
125, 231 171-2
30.39.8
Odes 3.6.1—8 133, 265 40.51.3 129
4.9.25-8 1 129
41.27.5
JOHN LYDUS LUCRETIUS
De mensibus 4.65 147 De rerun: natura 4.563-4
4
4.73 176
MACROB1US
JUSTIN 146
Satiinw/iii 1.12.15
18.5.4-5 152
43.1.7 61,72
 348 UNWRITTEN ROME

MANILIUS PLATO
Astronomica 5.282 208 Lams 815c 108-9
887d 237
MARTIAL
10.92.10 227 PLAUTUS
Amphitruo 41—5 35, 237
ORIGO GENTIS ROMANAE Casina 980 85
4.4-5 43
PLINY
OVID Natural History 8.153 296, 297
A is amatoria 1.405-9 140 18.236 168
Fasti 1.7 263 21.7 11-12
2.297-312 135 21.8 133
2.359-80 73-4 28.10-12 248-9
2.425-52 76-7 28.14 156
2.583-616 184 34.29 289, 292
2.723-80 313 34.139 317
3.313-16 163
3.320-6 164 POLYBIUS
3.523-40 229 3.47.6-9 256
3.713-26 135, 222-3 4.2.3 17
3.737-9 223 6.25.3-4 72
3.779-88 224 6.54.2 38
3.784-6 85, 224 6.56.8-11 167
4.129-32 140
4.133-62 141-2
4.187 210 PLUTARCH
4.326 210, 230 Antony 24.3—4 130-1
4.615-18 135 Coriolanus 38.3-4 257
5.101 72 Moralia 'll 1 f 149-50
5.184 229 273b-c 286
5.193-9 226 322f 287
5.331-6 228 323a 149
5.347-54 228 323b 281
6.209-12 191 Numa 1.1 14
6.396 291-2 4.8 246, 257
6.537-40 44 15.3 162, 163
6.629-32 281 15.6 318
Metamorphoses 1.173-4 229 19.2 146
11.221-65 221 Pnblicola 19.1 181
14.8-415 220 Romulus 15.2—1 150
Tristia 2.507-10 208
PROPERTIUS
PAUSANIAS 4.9.21-6 183
1.3.3 237
SCHOLIASTS
PHAEDRUS on Homer Iliad 5.412 303
5.7.23-7 198 on Plato Phaedrus 244b 55
 INDEX OF PASSAGES 349

SCIPIONIC EPITAPHS VALERIUS MAXIMUS

ILLRP 309-10 6 preface 258-60
1.7.4 171
SENECA 1.8.1 261
Consolatio ad Marciani 16.2 292 1.8.6 38
De beireficiis 6.32.1 134 1.8.7 262
2.2.9 73
SERVIUS 5.6.3 262-3
on Aeneid preface 269 7.3.2 297
3.20 99-100 8. Labs.5 249
3.96 297-8
4.58 99-100 VARRO
7.343 70 De gente populi Romani fr. 3P 243-4
8.636 148 De lingua Latina 5.8 13
5.74 175-6
SOL1NUS 5.85 52
1.24 276, 287 6.13 52
6.18 194, 238
STRABO 6.94 155
1.2.8 66 7.5 41
6.2.6 153 7.36 41
Menippean Satires 513A 213-14
SUETONIUS
Di'hus An^rts'its 70.1—2 130 VELLEIUS PATERCULUS
1.14.8 168
TACITUS
Dialogns 10.5 201 VERRIUS FLACCUS
Histories 3.72.1 317 Fasti Praenestini on 1 April 146

TERTULL1AN VIRGIL
Apologeticns 22.12 24S Aeneid 1.372-80 264
6.817-23 305
THUCYDIDES 6.893-9 269
1.20.1 8,318 9.79 263

ZONARAS
7.11 296, 297
 General Index

Abronius Silo, scriptwriter 208 Alcibiades 90—2

Academy, scepticism of 251,253 statue of 138, 234, 300
Acca Larentia 254 Alcimus, historian 302
Accius, dramatist 34, 122-3, 206, 223, Alcmaeon, matricide 212
252, 294-5, 304 Alcumena, on mirror scene 92, 96
Achilles, on cista scenes 113-4, 119, Alfbldi, Andreas 54
121 Algidus, battle of 26
Acilius Eutyches, C. and L. 20 Alienus, L., supposed tribune 173
Acilius, C., historian 62 Alixentr (Paris), on cista scene 119-20
Actaeon, dance story 214 Alses, on cista scene 119—20
Admetus, king of Molossians 281 Alsir, on cista scene 1 19-20
Adonis, on cista scene 86, 88 Amulius, Alban king 70, 240, 244, 253
aediles, responsibility for Indi 169-70, ancestral portraits 13—14, 236
186, 205 Anchises, as prophet 268
Aelius Gallus, C., jurist 283 Ancus Marcius, king 282,315—16
Aelius Tubero, historian 78, 262 Andromachus, senator 81 -2
Aemilia, Vestal Virgin 249 Andromeda, exposure of 122—3, 222
Aemilius, Mam., diet. 437 BC 26 Androtion, historian 138, <00
Aemilius Paulus, L., cos. 168 BC 34, Anna Perenna 18—21,217,229
206 Anna, in Ovid 217
Aeneas 50, 144, 173, 201, 252, 264-5, annates 252, 263-8
268-9, 302 annalistic narrative 48-9
in Alcimus 302 Annins Fetialis, on Valeria 289-90
in Ennius 8, 49 Anteias, eponymous hero 302
in Naevius 240 Antemnae 176—7
in Stesichorus 232 Antipho, actor 215
Aequi, defeat of 26, 27 Antium, Latin city 69, 76
Aeschylus 56,211,221,281 Antonins, lullus 134
Aesculapius, cult of 58, 75-6, 77, 248 Antony (M. Antonins) 78,216
Afranius, togata author 196, 197 and Liber 130-1,224—5
Agamemnon 17,281,294 Aper, M., in Tacitus 201
on cista scenes 5, 94, 113-4, 116-17 Aphrodision, in Latium 69
agerpublicus, occupation of 27, 29 Aphrodite Epitalarios 149-51
Agrippa Postumus 134 Aphrodite, see Venus
Agrippina, empress 198, 209 Apollo 58, 129, 188, 190
Ajax, on cista and mirror scenes 92, 94, and Augustus 130-2, 134, 136
96,116-7,119-20 and Marsyas 99, 103, 128, 131
Alba Fucens, Marsyas statue at 99 Augustan temple of 131—2
Alba Longa 33, 50, 173 on cista scenes 86—7, 99, 103
Alba, granddaughter of Aeneas 302 republican temple of 129,171
 GENERAL INDEX 351

response to Tarquin’s sons 297-8, and aretalogi 265-6

299, 307 and Flora 132,225
Appian, historian 70 and Liber 132-3,136, 224-5
April, month of Venus 140 arch of 15, 47, 235, 265
Apulia, performance culture in 108, as ponttfex maxinitis 265, 270
123 memoirs of 265, 270
Ara Maxima, Hercules cult at 56, 58, purges prophecies 46, 47
74, 183 rebuilds temples 132-3, 265
Arbuscula, actress 215, 220 reforms Lupercalia 77, 78-9
Arcadia, cults imported from 57, 62, Aventine 12, 151, 162
65, 234 Avemus, lake 56
archaeological reconstruction 271-80
Ardea, Latin city 69, 293, 302, 303-4, Babylonia, sacred prostitution in 152
305, 313 Bacchanalia 148
Ardeias, eponymous hero 302 suppression of 85, 109, 127, 128-9,
Argives, chapels of 234 136, 146, 159, 193, 224
Ariadne 217,222 Bacchiads, Corinthian aristocrats 153
Aricia, Diana cult at 65 Bacchylides 257
Aristodemus of Cumae 138, 293, 316 Bacon, Francis 22
Aristonothos, vase-painter 231 banquet songs 40-1,273
Aristophanes 37, 211 Barchiesi, Alessandro 148
Aristotle, on Rome 233 barking snake, omen of 296, 297, 304
Annenia, sacred prostitution in 152 Barnes, T.D. 203
Arnobius, Christian polemicist 161-2, Bathyllus, dancer 197,198,219
166, 185 Bauli, and Hercules 191
Artemis T.nucpolos 65 Beacham, Richard 126
Ascanius, son of Aeneas 50 Bellona 71,190,192
Asinius Pollio, C. 206 as mother of Inuus 70-1,72
Asklepios, see Aesculapius on stage 35, 206
Assaracus, Trojan king 172-3 Bernstein, Frank 167-74
Atalanta, in mime? 213 Binder, Gerhard 54
on cista scene 119—20 Bona Dea, festival of 182-4
Athena (see also Minerva) 188, 303 Boni, Giacomo 285, 306
Athenaeus, on paignia 119 booty, as part of triumph 37
Athenian acropolis 57, 136, 138 Bremmer, Jan 54
Athenian drama, influence of 119, 122 Brennus, Gallic chieftain 27,30
Athenian history, as paradigm for Brutus, see lunius
Roman 136-8, 234, 300-1, 304
atrium, in Roman houses 13-14, 147, Cacus, cattle-thief 183
271, 273, 278-80, 282 Caesar, see Julius
Atthidographers 138, 234, 235, 300 Caligula, death of 204
Atticus, T. Pomponius 250 Callimachus, medical writer 76
Attus Navius, augur 64, 247-8, 251-2, Callimachus, poet 233
263 Calvino, halo 272-3
Augustine, St 160,218,310 Calydonian boar-hunt 217
on Flora 179, 185 Camenae, Latin Muses 44, 51, 160
on miracles 246-9 Camillus, M„ kills Cn. Tarquinius 315
Augustus 132—4, 189, 208 Camillus, see Furius
and Apollo 130—2, 134, 136 Campania 122, 191
 352 UNWRITTEN ROME

Canens, in Ovid 217,219 on poetry and history 249—52

Cannae, prophecy of 43 on pontifical chronicle 234
Cappadocian goddess 188 on prophets 267
Capys, founder of Capua 232 on sophisticated Romulus 125—6
Carandini, Andrea 15-16, 271-92 on stage as source of information 38
carmen Nclei 45, 239 quotes Ennius 41, 42
cannen Priaini 45, 239 Cicero, Q. Tullius 42, 43, 214, 250-3,
Cannentis, prophetess 44, 56, 145-6, 256, 261, 263, 270
217 Cincinnatus, L. Quinctius 26, 30, 33,
cannina, collection of 45-6, 239 37
Carthage 8-10, 240 Cincius Alimentus, L., historian 7, 8,
Casenter (Cassandra?), on cista scene 11, 50, 240-1, 262, 315
119-20 Cincius, L., antiquarian 11, 13, 15, 24,
Castor and Pollux 58, 71, 232, 234, 29
245, 248, 257, 261 cinctuti, in Ovid 72
on cista scenes 86, 88, 92-3 Circe, witch 158, 232, 247, 302
Castrum Inui 69 in Ovid 217,220,221
Cato, M. Porcius, historian 7, 41, 50, Circus Flaminius 129, 171, 189, 191
167 not a race-track 171, 189
on banquet songs 40-1, 236 Circus Maximus 12, 178
Cato, M. Porcius, at Indi Florales 179, overlooked by Apollo 131, 132
186, 229 cistae, engravings on 5-6, 85-104, 111—
Catullus, C. Valerius 206, 222, 284 21," 195, 197, 233, 238
cavalry, Roman 71-2, 75 City Dionysia, at Athens 136, 138
celeres, early cavalry 71—2, 73, 74, 75 Claudia, Quinta 182, 197,210—11,
as Romulus’bodyguard 71,72 245, 248
censors, records of 13 Claudius, Ap., ecus. 312 BC 50, 75,
Cerberus, on cista scene 86 261
Ceres 129,187,224-5 Claudius Caesar 50
shares temple with Liber and Libera Claudius Liberalis, Ti. 79
127, 129, 131-2, 135-7, 173, 222 Claudius Marcellus, M., res. 222 BC
Cermalus, home of Sibyl 55, 56 34, 206, 237-8
Chaldaean astrologers 159 Clement of Alexandria, on Sibyls 55
Champeaux, Jacqueline 141 Cleomenes, analogue for Porsenna 137
Chloris, nymph 226, 229 Cleopatra, as end of history 209
Cicero, M. Tullius 38, 69, 131, 157, cliuus Pnblicins 177
210, 212, 224, 236, 250, 263, 286, cliiitts Victoriae 57, 138
294 Clodius, on chronology 14
as aedile 169,179 Clodius, P„ tr. pl. 58 BC 183, 190
as sceptic 43, 47, 251-2, 261, 270 Clodius Vestalis, C., moneyer 176
executes conspirators 211, 212 Cloelia, legendary heroine 180-2, 292,
on Failin' 42-3 313 "
on history 259 statue of 180-1, 289-92, 316-17
on Liber 225 Cloelii, history of 313—4
on hidi 36,211-16,226 Cloelius, Sex. 190
on Indi Romani 168, 169-70 Cloelius Gracchus 26, 30
on Lnpcrcalia 78 Clusium, Etruscan city 71, 74
on miracle stories 261 performance culture at 112—3, 123
on Numa and Pythagoras 158 Cluvius Rufus, historian 199, 202
 GENERAL INDEX 353

Clytemnestra, on cista scene 113—1 Demaratus of Corinth 153,233,316

Coarelli, Filippo 99, 149, 150, 189, Demeter, see Ceres
285-6 Demodocus, Homeric bard 40
Colline Gate, battle of 193 demons (daitnones) 247-8, 254, 255
Collingwood, R.G. 22 deuotio 28, 31
Comitium, scene of Lupercalia 63—1 Diana 10,159-60,214
communal memory 17, 236—9, 292 on cista scenes 86-8, 92-3, 113-4
Constantinople, Lupercalia at 82 Dido, see Elissa
consulship, origin of 298-9, 304-5, Dindia Macolnia 6
310 Dio Cassius, historian 46-7, 203, 269
Consus, games of 151, 153, 239 Diodonis Siculus, historian 99, 210,
Corinth 151, 153 268, 301
Coriolanus, see Marcius Diokles of Peparethos 240
Cornelius, C., trial of 212 Diomedes, grammarian 39, 194
Cornelius Balbus, L. 206 Diomedes, in Italy 247, 302-3
Cornelius Cossus, A. 26, 30, 31 Dionysia, iiriiiM 213,220
Cornelius Culleolus, prophet 46 Dionysius 11, and Locri 152
Cornelius Gallus, and Cytheris 216 Dionysius of Halicarnassus, historian 8,
Cornelius Scipio ‘Barbatus’, L. 6-7 10, 13, 57, 63, 72, 255, 262, 263,
Cornelius Scipio, L., cos. 259 BC 6-7 317
Cornell, Tim 10,48,311-12 on Alban triplets 33, 207
Cratinus, comic poet 119 on Cloelia 181, 182
Crawford, Michael 176, 188 on early Rome 15, 306, 309, 317,
Cremera, Fabii defeated at 71, 74, 207 318
Creousa, mother of Ion 58 on Faunus 70, 237
Crepereius Kogatus, lararittm of 81 on history and drama 207
Crisida (Chryseis?), on cista scenes 119- on hymns 45, 237
21 on Lake Regillus 136, 240
Cumae 191 on Indi Ceriales 172, 173
home of Sibyl 56, 142 on miracle stories 253-6
Tarquin at 138,293,316 on purpose of history 255
Curiatius Maternus, dramatist 201, 208 on Romulus 254-5
Cybele, see Magna Mater on Servius Tullius 280, 282, 289
Cyprus, sacred prostitution in 152-3, on Tarquinius Superbus 288, 295-6
154 on Year One 298
Cytheris, ininui 216, 220 Dionysus Eleuthereus 136-7, 138,234
Dioscuri, see Castor and Pollux
Daktyloi, Idaeau 163 Dis Pater, on cista scene 92-3, 117-8
Damophilus, artist 139 divination, authenticity of 42-3, 251-2
Datums, Daunia 302-4 divine epiphanies 37-8, 155-7, 162-5,
Davidson, James 119 244-6, 248, 253, 260-1, 262, 265-
debt-slavery 99 6, 269, 270
Decemvirs, tyranny of 127 Domitius, beard of 248
Decius Mus, P„ cos. 295 BC 28-9, 30, Doxa, on cista scene 94, 116-7
31, 32 drama, and ritual 36, 69, 92, 95, 108-9,
Decius Mus, P., tr. mil. 343 BC 28, 29, 119, 127, 145, 161
30, 31, 261 drama, origin of at Rome 7, 25
Delphi, Tarquin’s sons at 296-8, 307, drama scenes, on cistae 113-7
313 drama subjects, hypothetical:
 354 UNWRITTEN ROME

battle of triplets 207 on Naevius 39, 40, 240

C. Gracchus 199 on prophets 39, 41—2, 268, 319
capture of Corinth 37, 38, 207 on Saturnian metre 240
Circe 220 plays of 34, 206, 213
Cloelia 180-2 title of his epic 48—9, 51
comic gods 165 Enyalios, as Pan 70, 72
deuolio of Decius 34 Enyo, war-goddess 70—1, 188
dynastic crises 209 Ephialtes 65-6, 69, 76, 81
Etruscan hanispices 266-7 as (Ep)opheles 66, 83
Fabii at Cremera 207 Ephorus, historian 163, 234
Flora 225-30 epic, origin of at Rome 7-8, 39
Hercules and Bona Dea 182-4 Epicurus, on stage 214
Juno Regina 207 Epidaurus 75-6, 248
L. Brutus 313 Epopheles 66
Lara and Mercury 184 equester ordo, rituals of 71—2, 73, 78-9
Lucretia 313 Erato, in Ovid 210, 230
Marcellus in Gaul 34 Eratosthenes 126, 315
Messallina 199 on Romulus 49, 233
Numa and Jupiter 166 on Sibyls and Pan 55, 58, 70
origins of Indi 174, 211, 229 Erictho, witch in Lucan 158, 164
Peleus and Thetis 221—2, 230 Erymanthian boar 187, 191
praise of emperor 197-8, 206, 209 Erythrae, seat of Sibyl 46
Q. Claudia 182,197,207 Eryx, sacred prostitution at 153, 178
Roman triumphs 37—8, 207, 318 Etruria, Greeks in 231-2
Romulus 34, 207 performance culture in 123
Tullia 199,206,313 Etruscan priests, apparatus of 28, 32, 34
Verginia 199 Etruscan saecula 48
drama subjects, Ovidian 217 Euhemerus, mythographer 44, 162
dramatic genres, permeability of 109- Euripides 111', 119, 122-4, 214, 238
11, 179, 182, 186, 194-8, 209, Eutychides, dutiful freedman 20
214, 219 Evander, Arcadian in Rome 44, 56-7,
Dumezil, Georges 54 61, 62
and Hercules 57, 232
Ebrios, satyr 92, 95, 111 founds Lnpercal 55.61,72
Egeria, nymph 51, 160, 162 mother of 55, 57, 70
consort of Numa 159-60, 161, 162,
250, 255 Fabii, as Lupcrci 73—1
elicere, meaning of 155, 157-8, 165 characteristics of 71-2
Elissa (Dido), abducts Cypriot girls 152 defeated at Cremera 71,74, 137,
Elysian Fields 226-7, 229 207
empirical method 22-3, 54-5, 230, Fabius, commander of celeres 71
272, 312 Fabius, Kaeso 75
enargeia 33 Fabius ‘Cunctator’, Q. 71, 241
Enna, cult centre of Demeter 129 Fabius Ambustus, K. 74
Ennius, Q., poet 7-8, 39-51, 243, 252, Fabius Dorsuo, C. or K. 32—3, 36, 74
264, 315 Fabius Pictor, Q., historian 7, 8, 24,
chronology of 48-50 29, 41, 138, 175, 240-1, 252, 301,
on Andromeda 122-3 315
on foundation of Rome 47—8, 49 chronology of 49, 50
 L

GENERAL INDEX 355

on Romulus 16, 254 Fishwick, Duncan 188

Fabius Rullianus, Q. 28-9, 71, 73, 75 Jlaiiiines 53, 176
Fabius Rusticus, historian 202 Flavian dynasty 205
Fabius Vibulanus, K. 74 Flavius, Cn., aedile 235
Fabricius, C. 37-8 Flora (Mater) 126-7, 132, 175-6, 225-
fabula Atellana, equivalent of satyr-play 30
194, 197, 215 and Priapus 227, 229
fabula praetexta, historical drama 34—5, bride of Zephyrus 184,227
36-7, 182, 194-9, 200, 201, 206- Jlanien of 176
10, 237-8, 252, 313 mother of Liber 129,225
fabula tabentaria, form of comedy 194 rival explanations of 225-6
fabula togata 194-5, 196-7, 208, 210— temple of 124, 127, 132, 176-8,
11, 214 225, 227
Falerii, defeat of 28 Florus, on history and myth 180, 292,
Faliscan schoolmaster 33, 35 313
fallible knowledge 47, 50-1 forecourt, see iiestibiiliini
family trees, in atrium 13-14 Fortuna 69, 190
famine in Rome 129, 130, 131, 133, chamber of 279-80, 286, 288
134, 135-6, 173, 192 helps Romans 28, 30, 31
Fannins, C„ cos. 122 BC 167, 174 lover of Servius Tullius 217, 286-8,
Jasti, of consuls and triumphs 15, 47, 290
235, 265,299,311-12 ‘Fortuna Equestris' 290
Fatua. as ‘speaker’ 42 Fortuna Muliebris 237, 256, 261, 263
Fauna, as ‘speaker’ 41-2 Fortuna Virilis 140-7, 148-9, 152,
Fauni 39, 40, 42-3, 46 153-4
Faunns 63,64,81 Forum Bovarium, derivation of 56-7
and Omphale 58, 217, 221 Frederiksen, Martin 141
an? Picus 162-4 Fulvius Nobilior, M. 49, 206
as ‘speaker’ 41-2, 137 Furies 28, 32, 158, 223
as king in Italy 43, 45, 162-3, 237 on stage 34, 167, 212-13, 226, 313
as Pan 42, 62-3, 162 Furius Camillus, M. 27, 31, 32-3, 35,
as prophet 41-2, 44, 62, 162-3 49, 50
as >on of Mars 70, 72, 75 Furius Filocalus, calendar of 147
god of Lupercal? 53, 62, 73
songs about 45, 237 Gabba, Emilio 256,301
temple of 53, 76, 237 Gabii treaty 10
Februarius, god of Lupercal? 62 Gabinius, A. 269
Feeney, Denis 251 Gallius, Q., trial of 214
Feldherr, Andrew 36 Gelasius, Pope 81
Felix 111, Pope 81 Gellius, Aulus 213,283
Ferri, Rolando 200-5, 209 Genius, god of reproduction 125, 126—
Festus, learned author 45, 56, 77, 149, 7, 147
239, 301-2 Genucius Cipus 262-3
Ficarius, epithet of Faunus 64 Geoffrey of Monmouth 16,18,294
Ficoroni cista 5-6, 85, 104 Germanicus, death of 209
fiction and history 17 Geryon, cattle of 56
ficus Ruininalis 63-5 Giants, defeat of 191
Finley, Moses 17 Gnesippus, dramatist 119
fires in Rome 47,203,271,276,291-2 goats, at Lupercalia 52, 61, 64, 70, 73, 76
 356 UNWRITTEN ROME

gods on stage 35, 167, 198, 206, 237 ‘history of religions’ 21

gods’ care for mankind 252, 255-8, Holofernes 82
260-1, 263, 266 Homer 16, 18, 40, 51, 158, 226
Goldberg, Sander 40 not historical source 16-17, 294
Gorgasus, artist 139 Horace (Q. Horatius Flaccus) 43, 99,
Gorgo, in Strabo 66 157,184,215
Gracchi, honoured in Octavia 204 on neglected gods 133, 265
Gracchus, Ti. Sempronius 129 on origin of drama 125-6, 127
Gravisca, lonians at 231, 232 on praetextae 194, 207-8
‘great year’ 50 on rustic Latium 125, 165, 231, 233,
Greek historiography, principles of 8, 236, 241
240, 241 on satyr-play 218,219
Griffiths, Alan 300 on spectacle 37, 38, 207
Horatii, history of 313-4,315
hairy suits 66—9, 92, 95, 107, 109—10, Horatius, M„ consul in Year One 8,
145 298, 309, 310
Harpaktides, archon 136 Horatius Codes 35, 180, 181, 292,
Hecataeus, on Italy 232 309, 313, 315
Helen, on cista scenes 119-21, 123 statue of 289,316-17
Hellanicus, on Rome 233 Hortensius, Q. 213
Hephaistos, at Volcanal 232 Hostilii (see also Tullus), history of 315
Heraclides Ponticus 55, 233 Hostilius, Hostus 10
Herakles, see Hercules hydromancy 160-1, 166
Herculaneum 191, 284 Hyginus, on Trojan families 174
Hercules (Herakles) 188, 222, 232 hymns, as source of knowledge 45, 237
and Ap. Claudius 261
and Bona Dea 183, 184, 185—6 Ilia (see also Silvia), mother of twins
and cattle of Geryon 56-7,183, 252, 253
191, 232-3, 240 inconditus, in Livy 24-9
and Sulla 190-1 Incubus (or Incubo) 65-6, 69, 81
games of 187-93 as disease 76
in Ovid 217 Ino, in Ovid 217, 220
Roman cult of (see also Ara Maxima) Inuus 62,71,75,76
58, 75, 183 as Pan 62—3, 65, 81
Hercules Custos, temple of 189, 191 as son of Bellona 70-1,72
Hercules Magnus 189,191 cult site of 69, 76
Hermes, father of Evander 56 Ion, conceived in cave 58
Herodotus 8, 137-8, 251, 252 Iphigeneia, on cista scene 113—I, 123
on Marsyas 90, 99 irrational terror 32
herons 247, 302—3 Isaiah, on pilosi 66
Hesiod 51,226-7,232 Istor, on cista scene 94, 116-7
Hesione, exposure of 222 lulius Caesar, L., author of Pontificalia
Hipparchus of Collutos 138, 300—1, 305 262
Hippias, analogue for Tarquin 137-8, lulius Proculus 245—6, 250
234 lunii, history of 313-4,315
history and drama 199, 206—7, 238, lunius, son of Daunus 303, 305
266 lunius Brutus, L., consul in Year One
history and poetry 243-53, 262, 266, 8, 34, 293-305, 307, 310, 311,
269-70 313, 315
 L
GENERAL INDEX 357

statue of 316 Lake Regillus, battle ot 136, 172, 173.

lunius Brutus, L., supposed tribune 222. 240, 245, 257, 261
173 Lamia, in Strabo 66
lunius Brutus. M., assassin of Caesar Lara, mother ot Lares 184, 185,21/,
129, 305 220
lunius Brutus, T., supposed tribune Lares I’raestites, festival ot 184
173 Lares, games of 190
lunius Bubulcus, C. 301-4, 305 Latacz, Joachim 17
lunius Scaeva 301—2, 305 Latin literature, origins of 7-8, 24, 40-
1
Janus, as king in Italy 43, 227 Latinius, T. 261
Jerome, on centenary games 50 Latinus 44, 232
on pilosi 66 Latium, performance culture in 123-4
Juba, on t.t/<is(sjio 150 Lavinia, in Ovid 217
Julia, daughter of Augustus 133-4, 204, lectus j>cniulis 147
209 Leda, played by mimat 111, 113
Julius Caesar 204, 305 Lent ritual 82
and Luperid/hi 63, 78 Levene, David 245
Juno 76-7,184,227,229 Liber (Pater) 6, 70, 127, 187, 223
Juno Regina 25, 206-7, 261, 266 and Antony 130—1,224—5
Jupiter, address of 229 and libertas 84-6, 127, 137, 138-9
and Manyas 99-100,128,131,136
and Lara 184,185,217
and Numa 155-8, 163-5, 269, 318 and plcbs 127, 173
as Dionysius Eleuthereus 137, 138,
on scenes 86-8,92
234 '
Jupiter Lhcius, altar of 155-6, 161,
163, 165, 166 as god of drama 117
Jupiter Feretrius 26, 189 from liberi 225
Jupiter Optimus Maximus, temple ot 8, in mime 218 ____
10, 46, 169, 187-8, 235, 309 on cista scenes 86-9, 96, 98, 117-8,
Jupiter Stator, temple of 234, 276, 285, 145, 154
289-91 son ot Flora 129
temple of 124, 127, 129, 131--,
Justin, historian 61,72, 152
135-7, 139, 222, 224
Juturna, pursued by Jupiter 184
Libera, as Ariadne 222
Juvenal, on Bona Dea cult 183 as Persephone 129
on flagellation ritual 77, 79, 148 on rist.i scene? 86,89,145
shares temple with Liber and Ceres
Kaeso, significant name 74-5 127,129, 131-2,135-7,173,222
kukitor 3—4
Libcmlin 84,86,137,222-3
King Arthur, supposed historicity of Imli at 85, 127-8, 129, 188, 19--3,
16, 294
222, 224, 238
Kleidemos, historian 234-5 libertas, and licentia 84,134
Kleiklos(?), buried on Esquiline 2, 231 Brutus’password 129
Kragelund, Patrick 194, 197, 198, 203 symbolised by Marsyas 58, 100, 131,
Kronos (see also Saturn) 227 133-4, 136
Licinius Macer, historian 170, 173
La(u)is, on cista scene 94, 116-7 Livius, M„ pontfex 30, 32
Laberius, mime writer 196, 217 Livius Andronicus 7, 24—5, 40, 122—4,
Lactanrius 70 180
Ladumeda (Laodamia) 94, 116-7
 358 UNWRITTEN ROME

Livy (T. Livius), historian 43, 50, 56, Lucretia 293, 298, 299, 301, 305, 307-
62, 172, 199, 210, 297 8, 311, 313, 314
as philosopher 249 Lucretii, history of 313—4
on ancestral portraits 236 Lucretius, Sp., consul in Year One
on ancient records 14, 15 308, 310, 314
on Bacchanalia 86 Lucretius, T., poet 51, 216, 226
on Cloelia 289 Lucullus, L., and Sulla 190
on deification of Romulus 245-6 ludi, under Augustus 197—8
on early Rome 15, 18, 306, 309, ludi Apollinarcs 168, 171, 175, 188, 207,
317- 18 214-5, 216, 238
on flagellation ritual 76 origin of 128, 168, 170, 172, 192
on history and drama 199, 206-7, 266 ludi Capitolini 189
on incondita carmina 24-9, 37-8, ludi Ceriales 129, 136, 169, 175, 188,
318- 19 193, 222
on Inuus 62, 69 origin of 168, 171-2, 173-4, 238
on Jupiter Elicius 155-6,158,159 ludi circcnses 175, 178
on Lake Regillus 136, 240 ludi compitalicii 190
on ludi Apollinarcs 128, 168 ludi Florales 169, 175-86, 188
on ludi Ceriales 171-2 licentious nature of 178—9, 196,
on Indi Megalenses 168 222, 225, 227-9
on ludi plebeii 171—2 origin of 124, 127, 168-9, 170, 172,
on ludi Romani 168, 192 174, 176-8, 192, 227, 229
on miracle stories 38, 155, 245, 249, ludi maximi 168
256, 261 ludi Megalenses 169, 175, 188, 210, 214,
on Numa and Pythagoras 158 216
on origin of drama 125, 127 origin of 128, 168-9, 170. 172, 192
on poetic stories 243-4, 253, 262, ludi plebeii 169-70, 175, 188, 238
266, 270 origin of 168, 170-1, 173-4, 192
on Porsenna 317,318 ludi Romani 168-70, 175, 188, 192,
on praetor maximus 10-11, 299 238, 240
on Servius Tullius 282 ludi scaenici 40, 85, 124, 175, 182, 205
on Tarquinius Superbus 288-9, 296 educational purpose of 36, 38, 165,
on theatre 36 167, 174, 180, 186, 199, 207, 213,
on triumphal songs 25-9, 33 226, 238, 266, 267
on Tullus Hostilius 156-7 effect on historians 38,181-2
on value of history 260,318 epic performances at 40
on Verginia 199 origins of 167—74
on Year One 298, 299-300 ludi Tarpeii 189-90
preface of 243-4, 259-60, 270 ludi Fictoriae Caesaris 168
time of writing 24 ludi lA'ctoriae, Sullan 168, 188, 193
Locri Epizephyrii, sacred prostitution at Luperca, goddess 52
152-3, 154 Lupercal 55, 57, 78, 234
Lotis, in Ovid 217 as cave of Mars 57-8,75
Lucan, 158, 204, 208 derivation of 52, 57
widow of 205 fig-tree at 63—1
Luceria, foundation of 303 Lupcrcalia 18, 52—83
Lucerus, king of Ardea 303-4 drama at? 69—70
Lucian, on dance 197, 209 flagellation at 74, 76-7, 79-81, 82,
Lucilius Balbus, in Cicero 261 148, 178
 GENERAL INDEX 359

nature of ritual 52—4, 61-5, 69-70, Markus, Robert 81

74, 77 marriage by seizure 145, 149-50
Ltiperci, armed 70, 72, 75 Mars 37-8, 140
as crepi 76 cave of 57-8, 70
as performers 69—70 fathers twins 57-8, 70, 244, 253
costume of 61,72,75,79 on cista scene 86
in late Empire 81 on stage 35, 206
masked 70, 82 Marsyas, ancestor of Marcii 58, 218
Ltiperci Fabiani 73-4 ancestor of Marsi 99
Luperci hiliani 78 and Apollo 99,103,128,131
Ltiperci Quinctialcs 73—4 and Liber 99-100, 128, 131, 134,
lustration ritual 63, 75 136
Lyaeus, name of Liber 84, 100 and Magna Mater 128
Lycaeus 61, 247 as silenos 92
Lycophron, on Daunii 303 on mirror and cista scenes 58-9, 99,
Lycurgus, lawgiver 145 103-4, 109
Lycurgus, enemy of Liber 223 statues of 58, 99-102, 131, 133-4,
Lydia, sacred prostitution in 152 136
Lydus.John 147, 176 Martial 208,227
Lykaion, Arcadian mountain 61 on Anna Perenna 18, 19-21
Lykaon, Arcadian hero 57 Mastrocinque, Attilio 137, 300-1
Mater Matuta, Satricum temple 139
Macaulay, Thomas Babington 239 Matemus, see Curiatius
Macrobius, learned author 12, 45, 81, Mazzarino, Santo 35
>46, 239 McKeown, J.C. 184
Maecenas, C, 1 Medusa, death of 95, 97
Maelius, Sp. 241 Memmii, history of 172-3
magic 50,165-6,318 Memmius, supposed aedile 172, 173
definition of 158 Memmius, C., moneyer 172, 174
magistri, in charge of Indi 189-90 Memmius, C., tribune 173
Magna Mater (see also Indi Megalenses) Menelaus 226
128, 182, 187-8, 197, 207,210, Menenius, M., tr. pl. 410 BC 27,30
245 Menichetti, Mauro 90, 111
temple of 57, 128, 192 mental cinema 272-3, 280, 286
Maia, grove of 184 Mephistopheles 83
Maiestas, in Ovid 229 Mercury, father of Lares 184,185-6,
Manilius, learned senator 50 217
Manilius, poet 208 Messallina, empress 199, 209
Manlius Torquatus, T. 27-8, 31, 34 Midas, captures Silenus 162
Mantuanus, poet 82 iniinae, at Indi Florales 178-9, 196
Marathon, battle of 57 niimae, dancing girls 107, 109, 111-19,
Marcellus, see Claudius 124, 185-6, 195,213-4,215,219,
March, month of Mars 140 222, 228-30,313
Marcii (see also Ancus), history of 315 mime 179, 208, 214-9
descended from Marsyas 58, 218 and elegiac poetry 184,216
Marcius, legendary pontifex 176 and satyr-play 109, 219
Marcius, Cn., prophet 42, 43, 192 and tragedy 195
Marcius Coriolanus, Cn. 137, 234, 256 in Anna Perenna ritual 19, 20
songs about 45, 237 satirical 196, 215-16
 360 UNWRITTEN ROME

mimesis, Dionysiac 108-9, 111, 117, commentaries of 156, 160—1, 166

119, 124, 127, 145 interview with Jupiter 155-6,158,
Mimnermus, on Diomedes 303 163-5, 269, 318
Minerva 87, 247 tomb of 159
Minicia (portions) 189 Numitor, grandfather of twins 73, 244
Minotaur, on Regia 2—3 nymphs 43, 58, 229-30
Mommsen, Theodor 74, 170, 171, and Anna Perenna 20-1
188, 189, 190 dancing 106-8, 119, 184, 219
Mormolyke, in Strabo 66 imitation of 108-9
Mucii, history of 313—4 in mime 214,218,219,229
Mucius, C., legendary hero 180, 292, ofLatium 184, 185
313
statue of 316 Oakley, Stephen 26, 48
Mucius Scaevola, P., cos. 133 BC 269 Oceanus 226-7
Munatius, P., and Marsyas 133 Ocrisia, maid of Tanaquil 263, 273,
Munucius, L., cos. 458 BC 26 280-1
Murcia, goddess 148, 151, 154 Octavia, daughter of Claudius 204
Murlo, palace at 281 Octavia, historical drama 198-9, 200-5
Muses, address Hesiod 51 date of 202—3
myrtle, as protection 142,144 mime in 209
Octavian, see Augustus
Naevius, Cn. 7,56,241,315 Octavius Lampadio, C. 239
epic poem of 39, 45, 239-40 Odysseus, and Circe 232, 247, 302
on Liberalia 85, 127, 188, 192-3, 224 Ogilvie, Ikoberr 35
plays of 34, 206-7, 223, 237-8 Ogyges’ flood 243, 246
nails, as year-counters 10—11, 235 Oinumama, on cista scene 11 9-20
Naples, performance culture at 123 Omphale, in Ovid 217
Natio, birth goddess 69 oral tradition 17-18, 239. 290, 292,
Neptune 35, 206, 247 294, 298, 305, 309
Neptunus Equester 151,153 Orestes 65, 212—13
Nereids 220, 221 on cista scene 119, i 2 1
Nero, emperor 47, 202-5, 209 Oribasius, medical writer 76
uestibiihim of 284, 292 Origo gentis Roniaiiae, date of 43
Nicaea, coins of 66 Origo gentis Roniaiiae, on Cimmerians?
nightmare demons 66 56
Nigidius Figulus, P. 45 Orosius, on miscarriages 76
Nike, Athenian temple of 57, 138 Ovid (P. Ovidius Naso):
Nikostrate, mother of Evander 57, 70 episodes from drama? 217, 225, 230
‘noble lie’ 166,246,255 on Anna Perenna 18
Nonae Capi'otiuae, play on 194, 195, on Apollo 134
197, 207 on Ariadne 222
Norbanus, defeated by Sulla 191 on Camientis 44
Nova Via 276-7, 279, 285-6, 287, on Ceres 134-5, 224
291-2 on Circe 217,220
Novius Plautius 5-6 on erotic mimes 208
Numa Pompilius, king 14, 72, 155-66, on Faunus and Pan 53, 58, 62, 217
176, 315 on flagellation ritual 76, 77
and Egeria 159-60, 250, 255 on Flora 124, 127, 134, 135, 178,
as Pythagorean 158-9,160,166 225-30
 GENERAL INDEX 361

on Genucius Cipus 263 Paris, dancer 208

on Hercules temple 191 Paris (Alexandras), see Alixentr
on Kalends of April 90, 140-6, 148, Patroclus, and Apollo 129
154,319 Pausanias, on false beliefs 237
on Lara and Jupiter 184 Pegasus, birth of 95, 97
on Liberalia 85, 134—5, 188, 222-5 Peleus and Thetis, in Ovid 221-2
on Indi Megalenses 210-11 Peleus, palace of 284
on Lnpercalia 52, 73—1, 148 Penates, migration of 247, 262, 263
on Lnperci 72 Pentheus 217,223
on neglected gods 135 Periander, analogue for Sex. Tarquinius
on Nova Via 291-2 137
on Numa and Jupiter 159, 166 Persephone, as Libera 129
on old annales 263 on rishi scene 117-18
on Persephone 135 Perseus, on rishi scene 95, 97
on Picus and Faunus 162-4 Petronius, novelist 66, 219
on plague 76 Phaedrus, on drama 198, 206
on Q. Claudia 197,207,210-11 phantom phallus 263, 273, 280-1
on seduction 140, 154 Philippi, battle of 129
on Servius Tullius 217, 281, 287 Philistus, historian 252
on silent goddess 184 phlyax vases 122
on wearers of stoLi 143 Phrygian goddess 188,210
provides dance plots 197, 216 Pick, Behrend 83
Picus and Canens, in Ovid 217, 219-
Pacuvius. plays of 34, 206, 213, 217, 20
Picus, as local god 162-4
Paest.im, drama at 109, 195 as king of Latium 43, 162-3, 220
Marsyas statue at 99, 102 pi/osi 65-6, 69
paignii ;i, form of drama 119, 124 Pinarii, at Ara Maxima 74
Palatium, derivation of 56, 57, 62 Pindar, on Elysian Fields 226-7
Pallantion, Arcadian town 56, 62 piper, sign of drama 111
Pallas, son of Lykaon 57 piscina publica 144, 148
Pan 55, 58, 76, 83, 218 Pisistratids, expulsion of 136-7, 234
as Enyalios 70, 72, 75 Piso Frugi, L., historian 49, 50, 56,
as Faunas 61, 62-3, 162 156, 268
as Inuus 62-3 plague at Koine 75-6, 77
as Lupercus 61 planipes, equivalent of mime 194, 228
as Picus 162 Plato 166,237,246
at Lnpercal 55, 57-8, 138, 234 on Bacchic mimesis 108-9, 117
in Athens 57-8, 70, 137-8 on silenoi 90-2
in human form 60-1, 66-9 Plautus, comic author 36, 85, 196, 205,
on Arcadian coins 58, 60-1 283
on mirror scene? 58, 60, 66-9, 92 on gods in tragedies 35, 198, 206,
Pan Lykaios 61-2, 70, 81 237
Paniskos, on mirror scene 58-9, 104, plebeian gods 229-30
109 Pliny (C. Plinius Secundus, the elder)
Pans, imitation of 108-9, 131 56, 69, 202
panthers, on Regia 2-3 on conception of Servius 263
Panticapaeum, coins of 60 on coronae 11-12
pantominius 197, 208-9 on dancing togatiis 209
 362 UNWRITTEN ROME

on equestrian insignia 40 Pomponius, mime writer 196, 222

on Fauiii 66 Pomponius Mela, geographer 69
on ficus Rtuniualis 64—5 Pomponius Secundus, P., playwright
on Indi Florales 168, 178 201
on Marsyas 99, 133 pontifex, duty of 32, 263
on Porsenna’s treaty 317—18 pontifices, chronicle of 48, 234—5, 262,
on portents 296-7 263-70
on power of words 248-9 Poppaea, in Octavia 198, 202, 209
on statues 289, 292 Porsen(n)a, Lars 137, 180-1, 185-6,
Pliny (C. Plinius Secundus, the younger) 234, 289, 309, 313, 317-8
201 statue of 316
Plutarch 33, 72, 188, 199, 317 Porta Fenestella 286-8
on Cloelia 181, 182 Porta Mugonia 276, 279, 287-8
on date of Numa 14 Poseidon Hippios 151, 153
on early Rome 15, 306, 309, 317, Postumius Albinus, A., historian 136,
318 240
on flagellation ritual 77, 79, 148 Postumius Albinus, Sp., cos. 186 BC
on Fortuna Virilis 149 136
on history and drama 199,207 Postumius Albus, A., dictator 136,
on Kalends of April 146-7 240
on Lupercalia 52, 63 Praenestine bronzes 58—60, 66-8, 85
on Luperci 72, 75 praetor niaxiinus 10, 15, 299, 306, 310,
on Marcellus’ triumph 34 318
on miracle stories 257-8 Praetorian Guard, as defenders of libertas
on Numa and Jupiter 159, 164, 166, 136
246, 318 praetors, in charge of Indi 205, 208
on Numa and Pythagoras 159, 166 Priam, king of Troy 239, 252
on Picus and Faunus 162-3 Priapus 19-21,219
on Servius Tullius 263, 281, 286-7 and Flora 227, 229
on talas(s)io 149-50 and nymphs 21,219
on Year One 298 in mime 218
poets as liars 51 in Ovid 217-18,221
Polla Argentaria 205 Princeps, piper 198
Pollux, see Castor principate, popularity of 204-5
Polybius, historian 17, 49, 72 Promathion of Samos 233
on ‘tragedy’ 35, 167, 213, 226 Propertius, Sex. 184—1,210
on Carthage treaty 8-10, 11 prophecy, as oral mode 44, 240
on expulsion ofTarquins 137, 138— continuing importance of 46-7,
9 269-70
on nature of history 256 prophets 39, 40, 42—3, 46, 159, 192-3,
on Roman moral education 35, 38 240, 267-8, 294
Polycrates of Samos, dance subject 197 books of 43, 192
Pomona, in Ovid 217,219 sound and appearance of 44, 51
- Pompeii 191, 284 Proserpina, see Persephone
Pompeius, Sextus 131 Proteus, shape-shifter 162, 221
Pompeius Trogus, on Luperci 61, 72 Ptolemy, Dionysiac procession of 92
Pompey (Cn. Pompeius Magnus) 212, public notices 267
284 Publicii, aediles 177, 229
Pompilii, family of Numa 315 Publicius, prophet 42, 43
 GENERAL INDEX 363

Publilius Syrus, mime writer 179 sack of Home by Gauls 4, 14, 49, 71,
Pylades, dancer 197, 219 207, 235
Pythagoras, Roman interest in 157, Sacra Via 272, 273, 277, 289-92
158-9, 165 sacred prostitution 152-4
Pythagoreans, and hydromancy 160 sacrificers 159, 192-3
and magic 157, 158-9 saeaila 48, 50
Salii, hymn of 11, 45
Quinctii, as Luperci 73—I Sallust (C. Sallustius Crispus) 259
Quinctius, Kaeso 74 Samothrace, mystery-cult at 163
Quinctius Atta, togata author 196, 197 Satricum, Latin city 104, 106, 139,
Quintilian, quotes Ennius 41 311,318
Saturn, temple of 64-5, 79
Raphael’s Parnassus 51 Saturnalia 178, 227-8
rcdtatio, of drama 198-9, 200-2 Satumia, prophecy of or at 43-4
Regia, decoration of 2-3 Saturnian verse 6, 39-40, 41-2, 43-4,
Regi/iigiuni 84, 306 46,51,240
Remus and Romulus 54, 57-8, 62, 70, Satumus, as king in Italy 43, 227
73-4, 77, 239, 244, 298, 301, 305 satyr-play 109, 119, 122, 194, 199,
songs about 45, 237 211,214,217-9, 223
supposed historicity of 15-16 and mime 109,219
Rhegium, war with Locri 152 satyrs 41,43,92
Rhinthon, dramatist 122, 124, 195-6 dancing 104-13, 219
Rhodrus, founder of Rome 302 imitation of 108-9,131,145
Rhomos, eponymous hero 302 in Ovid 223-4
Ribbeck, Otto 206, 210 on minor and dsta scenes 89-101,
ritual and performance 62, 92, 95, 111-18, 145
107—9, 117, 119, 127, 145 piping 111,114-6, 117
not static 18-22,54,77-8 spying on Venus 142, 144, 152
Roman calendar, date of 53, 170, 306 scalae Cad 151
Roman Forum, origin of 2 Schliemann, Heinrich 16, 311
Roman historiography, origin of 6-7, Sciarrino, Enrica 40
235—6, 240—1, 306 Scipio, see Cornelius
Roman People, as audience at luili 211 Scipios, epitaphs of 6-7, 236
history of 236-7 secession of p/efes 77,171
Rome, date of 1, 49-50 Seci. Lucus, on dsta scene 119, 121
sacred name of 176 Secular Games 50
Romulii, Romilii 314 Sejanus, fall of 209
Romulus (see also Remus) 8, 10, 15, Selene, and Sulla 188
26, 27,49-51,72, 293 Seleucus, dance subject 197
as grandson of Aeneas 49, 240, 315 Seinele, mother of Liber 86,223
as son of Aeneas 302, 314 grove of 90
deification of 245-6, 250 on dsta scene? 86, 89
statue of 317 Semo Sancus, temple of 10
Roscius, Q., trial of 213 Seneca, L. Annaeus (the elder) 208
Roscius, Sex., trial of 212 Seneca, L. Annaeus (the younger) 134,
Rubicon, crossing of 204, 219 202, 205,210-11,292
tragedies of 201, 205, 208
Sabine women, abduction of 148-51, Sentinum, battle of 28-9, 75
152, 239 Servilius, C., moneyer 174, 176
 364 UNWRITTEN ROME

Servilius Ahala 241 Stimula, grove of 86, 90, 129

Servius Tullius, king 217, 261, 276, Strabo, geographer and historian 66,
282, 285-9, 300, 315 69, 153
conception of 263, 273, 280-1 Stratonice, dance subject 197
Servius, Virgilian commentator 70, Suetonius, biographer 29, 174, 203,
144, 216, 290 218-19, 266
on fact and myth 269 Sulla, L. Cornelius 187, 190-2
on L. Brutus 297—8 Sutton, D.F. 209
on Marsyas 99-100 Syracuse, drama at 122, 195
on icallis Murcia 148, 149, 154 Syria, sacred prostitution in 152
seven kings, canonical list 49, 314—5
Sextius Sulla, friend of Plutarch 150 Tacitus, Cornelius 47, 132, 199, 203,
shape-shifting 162, 221 317
showgirls, see niiniae on loss of liberty 136
Shrove Tuesday, abuse of 82-3 on recitatio 201, 208
Sibylline prophecy 46-8, 128, 170, on Romulus’ ponteriuni 15
178, 191, 192, 269 talas(s)io 149—51
Sibyls 55-6, 142, 191 Tanaquil, queen 263, 280, 281, 285-8,
Sicinii, history of 173 291
Sicinius, L., supposed tribune 173 Taplin, Oliver 40
sileni, silenoi 41,92 Tarentum, drama at 122, 124, 195
imitation of 108-9 Tarquinia, defeat of 28
Silenus (Silanus), on cista scenes 90-4, Tarquinius Collatinus, consul in Year
104, 107, 116-7 One 138, 293, 298-301, 304-5,
ancestor of lunii Silani 218 307-8, 310-11
as source of wisdom 90—2, 162 Tarquinius Priscus, king 56, 163, 168,
in Ovid 217,219,224 250, 282-3, 285-7, 309
Silius Italicus, poet 99 Corinthian origin 151, 153,293,
Silvanus 63,79,125,126 316
speaks from grove 137,261 house of 273, 276-83, 285-91
statue and grove of 64—5, 79 Tarquinius Superbus, king 217, 278,
Silvia, Rhea (see also Ilia) 244 288, 293-7, 300
Simos, on cista scene 119, 121 dream of 252, 294-5
Sinnius Capito, on origins of Indi 174 expulsion of 84, 137, 171, 192,234,
Socrates, as silenos or Marsyas 90-2 235, 293-4, 306-9, 311, 312-13
onstage 214 Tarquinius, Arruns 296, 297, 307
Solinus, Julius 276,287 Tarquinius, Cn„ killed by M. Camillus
solleinnis, in Livy 26, 27, 31-2 315
Solon, law code of 234 Tarquinius, Sextus 137, 296, 305, 308,
Soranus, medical writer 76 313
Soresios, on cista scene 94, 116-7 Tarquinius, Titus 296, 297, 307
Sparta, marriage by seizure at 145 Tarquins, in exile 180, 185, 293, 304,
speaking dog, omen of 296-7, 304 316
Spes, temple of 132 wealth of 153, 169, 293, 307
spolia opima 26,31,34 Tatius, T., Sabine king 175-6, 301,
Statius, as scriptwriter 208 303
statues, evidence from 316-17 statue of 316
Stephanio, dancer 209 Taylor, Lily Ross 169, 170
Stesichorus, on Pallantion 56, 232 Telis, on mirror scene 92, 96
 GENERAL INDEX 365

Tellus, earth goddess 125, 126, 224 Valerii, history of 313-4

Terence (P. Terentius Afer) 36, 205 Valerius Antias, on Numa andjupiter
Tertullian, on miracles 248, 249 159, 164, 166
T (h)alassius 149—51 on Valeria 181,182
Themistocles 137, 234, 281 Valerius Maximus, historian 204, 258—
Theophrastus, on Rome 233 60
Theopompus, historian 251 on equites and Luperailia 73
Theseus, battle with Amazons 234 on Genucius Cipus 263
Thetis, sea-nymph 221-2 on lunius Bubulcus 301
on mirror scene? 92 on L. Brutus 297
Thucydides 8, 33, 234, 240 on Indi plebeii 171
Thurii, siege of 37 on miracle stories 260-3
Thyone, deified Semele 86 on origin of drama 125, 127
Thysdrus, mosaic from 79 on siege ofThurii 37
Tiberius Caesar 47, 209 on theatre 36
Tibullus, on witches 157-8 on Tuccia 249
Timaeus, historian 49, 138, 152, 315 Valerius Publicola, P„ consul in Year
Titinius, tojpita author 196 One 181,293,298,308,310-11,
Titus, arch of 272, 286 318
Tolumnius, Lars 26,30,31 Valesios, Poplios 294, 311, 318
Tondrus (Tyndareus?), on cista scene Vano, friend of Martial 208
119, 121 Vano, M. Terentius:
Torelli, Mario 149 as mystagogue 13
tragedy'. and mime 195 as philosopher 186
and praetexta 34-5 on cult of Flora 175-6
Trajan, controls flattery 206 on Aryei 224
transu xtio equitnm 71, 73, 79 on chronology 243-4, 252
Trebadus, C. 215 on dancers 213-14
triumphal procession, songs in 25-9, 33 on divine intervention 246-9
Trojan War, date of 49 on Faiuii 41
supposed historicity of 16-17 on foundation date 16
trumpet portents 48 on house of Tarquin 276
Tuccia, Vestal Virgin 248-9 on Jupiter Elicius 155,159
Tullia, murderous queen 199, 217, 220 on Lnperci and Lupercalia 52, 53, 63,
Tullius, see Servius 69-70
Tullus Hostilius, king 156-7, 158, 166, on mimes 218
315 on Numa 160, 166
Turnus, father of 302 on old cannina 45, 239
Twelve Tables 4,11-12,234 on origin of drama 125-6,127
Tyrrhenia, wife of Aeneas 302, 314 on origins of Indi 174
on Romulus’ augury 50-1
uullis Murciu 148, 151, 153 on saecula 48
uestibuliini, in Roman houses 282-4, on Samothrace mysteries 163
289-91 on stage stories 161, 185
Ulf, Christoph 54 on talas(s)io 149
underworld stories 268-9 on tales unworthy of gods 126, 160,
184-5, 247, 254
Valeria, supposititious heroine 181, on togata praetexta 194-5, 197, 207,
186, 289, 292 211,238
 366 UNWRITTEN ROME

on Trojan families 174 Virgil (P. Vergilius Maro) 50, 132, 263,
on Venus Murtea 148 302
subdivisions in 194 on annales 264
tripartite theology of 161, 218 on Faunus 44
Vatienus, P. 261 on Hercules and Cacus 183
Vatinius, P., Pythagorean 157 on L. Brutus 305
Veii, Veientes 26, 137 on Lupercal 70
siege of 33, 35 provides dance plots 197, 209, 216
Vel Saties of Vulci 315 Virtus, on stage 35, 206
Velabrum, etymology of 12 Visellius Ruga, C., supposed tribune
Velleius Paterculus, historian 29, 168, 173
204 Vitruvius, on satyr-play 218
Veneralia 147 Volcanal 180,232
Venus 140, 190, 199, 268-9 inscription at 2—4, 10
born from the sea 144, 152 Volteius, M., moneyer 187—8, 191,
on cista scene 86, 88 193
receives Priapus 21 Vulcan 145, 247
wife of Vulcan 145 Vulcatius, haruspex 48
Venus Equestris 290 vultures, omen of 295-6, 304
Venus Erycina 154,178
Venus Murtea 148,150,151 walls of Rome 4—5
Venus Obsequens 154 weddings, best time for 19-21
Venus Verticordia 140-3,146,148, Weiss, Anne 104
150, 154 will to believe 17
Verginia, tragic heroine 199 ‘winged words’ 4
Verres, trial of 212 witches 157-8, 160
Verrius Flaccus 146-7, 265 women washing, on cista scenes 89-90,
Vertumnus, in Ovid 217 110, 115-6
Vesta 217,249,255 writing, use of 1—5, 231
grove of 273,276-7,285-6
Via Appia 189,191 Xenagoras, historian 302
Via Flaminia 18, 19 Xenophon, historian 234
Via Latina 237
Via Salaria 27, 176 ‘year of years’ 50
Vibis Pilipus (Vibius Philippus) 59, 104 Zephyrus, in Ovid 184, 227, 229
Victoria (Victory), temple of 57, 71, Zevi, Fausto 151, 153, 293, 316
73, 75, 138 Zonaras, historian 296, 297, 304
on cista scenes 87, 92-3, 117-8 Zopyrus, analogue for Sex. 1 arquinius
on stage 35, 206 137
Vicus Tuscus 12 Zoroaster 158
Zwierlein, Otto 201