Sāṁkhya: Dualism without substances

Ferenc Ruzsa

Dualism may be understood as the view that there are two irreducible substances, soul and matter.
They are irreducible in that neither can produce the other, and as they are substances, they are
independent of each other, i.e. both could exist without the other.
Of course the theories of some of the staunchest materialists would meet the above criteria. The
ancient Greek atomists considered the psychê to be made up of atoms, but since these are a distinct
class of atoms, and atoms are uncreated and indestructible, we could say that we have two kinds of
substance here: soul atoms and all other atoms. Presumably we have to add that the spiritual substance
must be immaterial in some sense: it may be not in space or at least not occupying space, not resisting
the movement of matter. Unfortunately this restriction would exclude Jainism (where j va, soul is
spatially circumscribed and can be stained by karmic matter) and that is an undesirable consequence.
We may try to stipulate instead that a person’s soul must be an absolute unit, thereby excluding Colin
McGinn’s hyperdualism1 as well as those versions of Ved nta where a human soul is just a part of a
cosmic entity, Brahman. Or we may try to emphasise the number two – in atomism there are many
kinds of atoms, not only soul and matter. Unfortunately this would make the atomists pluralists instead
of the monists most historians take them to be. Further, all those theories that postulate e.g. time as a
distinct substance would cease to be dualist.
So dualism is far from a clear-cut category and we have to live with it. Old Vai eṣika had nine
substances (the five elements, space, time, manas and soul), and therefore it should be best categorised
as pluralism. Later on the non-materiality of the soul came to be more important and its liberated state
(free from any influence of matter) received more detailed treatment, so labelling it as dualism should
not be objected to.
Classical S ṁkhya is also a case in point. It is rightly considered the paradigmatic case of ancient
Indian dualism yet all the relevant criteria suggested so far are somewhat problematic. The name of
this school is derived from the word saṁkhy , number, and it is normally supposed that the idea
suggested is that here all features of the world are given in numbered lists. However, in its
foundational text, the S ṁkhya-K rik (SK)2 we are never told that there are two substances. When
the fundamental categories of the system are first mentioned, they are “the manifest, the unmanifest
and the knower” (vyaktâvyakta-jña, SK 2). The first two are material principles, the knower is, of
course, the soul. In the next verse a list of the well-known twenty-five entities (tattva) of the system
are given in four groups: the unmanifest, the seven productive and the sixteen unproductive manifest
tattvas, and finally the soul. So it appears as the system of three, or four, or twenty-five principles, not
of two!
Still, there can be no doubt that S ṁkhya is a dualism: puru a (the ‘person’, the standard term
here for the soul) and prakr i (nature) are regularly contrasted, their relation analysed, and the aim of
the system is the absolute and final separation of the two. Perhaps it is not a self-conscious dualism: it
Sāṁkhya: Dualism without substances F. Ruzsa 2

does not call itself ‘the system of two’, and it has no appropriate category for substance (or kind of
substance), like res for Descartes.
S ṁkhya’s conception of the soul (as it is normally interpreted) is again very far from the
standard European notion. It is definitely not the res cogitans, the thinking substance – it does none of
the functions of the Cartesian soul. To quote Larson,3 it is an ‘eccentric ghost’, ‘contentless
consciousness’. It does not seem to meet the criteria of substance, since it has no qualities and cannot
act or move – in fact it is absolutely unchangeable. Constantly changing matter and its more stable
formations again cannot be regarded as substances: as I will try to show in this paper, S ṁkhya in its
classical version is a strict form of substance reductionism, i.e. substances are only constantly
changing combinations of qualities and functions, nothing more.

The Sāṁkhya classic

The S ṁkhya here presented is varakrṣṇa’s version of the system, because it is both the most
interesting philosophically and the most influential formulation. S ṁkhya itself is immensely old,
clearly older than Buddhism, and had many significantly different forms.4 When, around the beginning
of the first millennium CE, many philosophical schools tried to standardize their teaching, there were
several attempts also in S ṁkhya. Perhaps the extremely short Tattva-Sam sa (Summary of the
principles) is the earliest survivor,5 but it is only a list of key terms without any recognisable
philosophical argument. When the classical Sū ras of the other schools were already written6 and
therefore philosophical style, argumentation and organization were significantly more developed, there
appeared the SK (ca. 4th cent. CE), a compendium in 72 verses written in the ry metre attributed to
varakrṣṇa. It seems to consist of a core text of some 50 verses with several additions and minor
reworking.7 I refer to the author of the core te t as ‘ varakrṣṇa’, although it is possible that it is in fact
the name of the (or a) person who added some extra verses.
varakrṣṇa clearly wanted to write the philosophical classic of the school. Therefore, first and
foremost, he construed a relevant and coherent system where all the tenets are argued for without
reference to any kind of traditional authority. At the same time he continued a long tradition and that is
most apparent in his choice of terminology; however, the at times astonishingly archaic words are
given modern definitions and the ancient agricultural/tantric imagery of the fertile, female nature
(Mother Earth) and the passive male spirit ruling over it is consistently understood as a metaphor only.
Thirdly, he tried to produce a text acceptable to all branches of the school and therefore avoided
controversial issues, either not mentioning them at all or at times picking an intentionally ambiguous
phrasing.8
varakrṣṇa’s success is really amazing. His te t became the absolute authority within the school
to which additions were possible (as in the medieval S ṁkhya-Sū ra) but whose formulations
remained standard ever since. As S ṁkhya concepts permeate other philosophies (especially Ved nta),
Hindu religious thought and literature (notably the Pur ṇas) and also traditional science, his
formulations were known and used everywhere.
As a philosopher, he can be compared to the greatest. There are at least three points where his
ideas may be considered fully relevant even in present day discussions. They are his conception of the
guṇas; his version of dualism, distinguishing the ontological and the epistemic sphere of
consciousness; and his substance-reductionism in general and in the specific cases. His denial of
emergence, his causally grounded epistemological theory and many of his single arguments are still
exciting and may often be felt essentially correct.
Sāṁkhya: Dualism without substances F. Ruzsa 3

Unfortunately he was not well understood in several respects. Already the person(s) extending the
SK to its present form added lots of philosophically uninteresting idiosyncratic material: long lists of
possible mental states, or of causes of non-perception (trivial ones, like “it is too far”), thereby also
disturbing the original composition and obscuring the tight internal logic of the whole. Partly because
of this, partly because of his philosophical depth combined with extreme conciseness, and partly
because of the general acceptance of some new ideas (notably the absolute unchangeability of the
soul), even his most faithful commentators often misrepresent his position and fail to grasp some of
his arguments. The voluminous Yukti-D pik , often called “the most important commentary of the SK”
is no betterś it could be described much more aptly as “the most important post-SK treatise in the
S ṁkhya tradition (formally a commentary on the SK)”. As has already been noticed by some
scholars, there was a definite break in the tradition after the SK, at several places the commentators
plainly just did not know what the author’s intention was.9

The twenty-five entities

Classical S ṁkhya is widely known as the system of the twenty-five tattvas. Tattva, literally ‘that-
ness’ or ‘being that’ means reality, principle, element, essence or substanceś here it will be translated
as ‘entity’. No definition of the term is given, and the concept itself is unimportant in the SK – it
occurs but once. Intuitively we could say that a tattva is a fundamental, fairly stable kind of reality;
everything else can be described as a (more or less temporary) combination of tattvas. It would be
very tempting to render tattva with ‘element’ś however, the unanimous practice of the translators
reserve this world for the five bhū as (earth, water, fire, air and ether).
varakrṣṇa inherited this list and as it was central to the tradition it was out of the question to
change it. However, several signs testify to his less than enthusiastic attitude to it: he never mentions
the number twenty-five, his single use of the term is not technical and he does not give all the
members of the list. Still, it is the fundamental map of the basic categories of S ṁkhya, so it is
reasonable to start with an overview. (It may be helpful time and again to look at the summary table at
the end of this section.)
Besides the two fundamental, imperceptible entities (soul and unmanifest materiality) there are
the twenty-three empirical entities of the manifest world: thirteen internal (psychic and biological) and
ten external (physical) entities. The latter are the five elements and the five sensibilia; together they
make up the inanimate world, but of course they form also the bodies of living beings. They constitute
the “field of activity”, k rya (lit. ‘what is to be done’, duty, work, effect)
The elements are called bhū as or viśe as, ‘beings’ and ‘kinds’, they are the gross material
substances of the world. It is a question for empirical science what exactly the elements are and how
many e ist – varakrṣṇa seems unconcerned, he never lists them, although he retains the traditional
number five. The category itself is relevant philosophically, but the members of the group are
irrelevant.
The five sensibilia or sense objects are those that the five senses can react to: colour, sound,
smell, taste and tangibility (including hot/cold). They are the tan-m ras or a-viśe as, meaning ‘only
that’ and ‘uniform’. Both names probably e press that each of them is an unmi ed simple (a colour
has no sound etc.), in contrast to the elements that have more than one sensible property (earth has
colour, smell etc.).10 The not exactly natural terminology of anm ras (instead of ‘qualities’ or
‘perceptible qualities’) is justified in S ṁkhya by its characteristic theory of the three guṇas,
Sāṁkhya: Dualism without substances F. Ruzsa 4

‘qualities’ – it is so fundamental for the school that it was imperative never to use the word ‘quality’
for anything else but rajas, tamas or sattva.
The psycho-physiological functions of a living being are together called the instrument, karaṇa
(lit. ‘doing’, action, means of action). It is conveniently split up into a centre and a periphery. The
latter, called the external instrument (b hya karaṇa), connects the centre to the physical world.
Inward-moving information comes through the five senses, outward-moving instructions go through
the five active powers.
The senses are not to be confused with the physical organs where they reside: hearing is not the
ear – it is the receptivity to specific entities of the physical world, i.e. sound and those things that
produce sound. The Sanskrit term, buddhi-indriya, clearly e presses thisŚ it means ‘power of
becoming aware’.
The five active powers (karma-indriya, lit. power of action) represent a surprisingly refined
biological theory veiled in a bafflingly archaic terminology – they are called speech, hand, foot, anus
and lap. No wonder varakrṣṇa felt the need to e plain themŚ “and the function of the five [active
powers] is speaking, taking, movement, e cretion and joy.”11 Nowadays we would perhaps express the
same idea with the concepts of consuming and excretion (including breathing in and out), movement,
reproduction and communication. A fairly comprehensive list, since perception is dealt with
separately, although growth and healing might perhaps be added.
The innermost core of a living being is the internal instrument, antaḥkaraṇa, consisting of the
manas, egoity and understanding. Although the concept matches quite well our concept of mind, great
care must be exercised in the use of this term on account of the unfortunate, but consistent practice of
Indologists who – mislead by the distant etymological connection – regularly translate manas with
‘mind’.12
Manas is, in fact, the lowest function of the mind, directly connected to the senses and the active
powersŚ “Manas is the coordinator,13 and it is a power reckoned among both kinds [i.e. the senses and
the active powers]. It is internal, for its objects belong to the three times [past, present and future];
therefore it works in both areas, [internal and e ternal].”14 As a sense, it takes the isolated pieces of
information of each sense and reproduces a coherent internal picture of the world and passes it on to
the conscious awareness; for we normally see an elephant, not a big patch of grey, and we hear the
same elephant, not some unrelated sound. As an active power, it controls the other active powers,
partly automatically, partly in response to a conscious decision “above”. But it is not only the eleventh
power, it is also part of the internal instrument, because the external instrument (i.e. the other powers)
can only operate in the present, while manas can produce images of the past or even imagine things
for the future.
Egoity is a very interesting psychological concept15 and it has a central function in the S ṁkhya
meditational practice leading to the liberation of soul from matter. At first sight it appears superfluous
in a system where each individual has its own soul, but in fact it has a very important role to play. Its
name, ahaṁ-k ra (‘I-maker’) and its synonyms mama-k ra (‘my-maker’) and asmi (‘I am-ness’, i.e.
the intuition that “I am [this or such]”)16 together with its succinct definition abhim na
(‘appropriation’, SK 24) are all that we are told, but they are e pressive enough and the idea is widely
known in all Hindu schools and in Buddhism as well. Perhaps the best analysis is aṅkara’s
introduction to his commentary on the Brahma-Sū ra, where it is called adhy sa, ‘superimposition’ or
rather ‘projection’. Nature is in itself single and continuous, but we all split it up into I and not-I
(ahaṁ-k ra), mine and not mine (mama-k ra). We feel certain facts of the world as belonging to our
being: I am old, I am fat, I am rich, I am a father, I am a scholar (asmi ). But in fact the real subject,
Sāṁkhya: Dualism without substances F. Ruzsa 5

the soul is not fat nor is it a scholar, it has no money nor offspring. So egoity is the “illegal” e tension
of the ontological subject, the soul, producing thereby our familiar phenomenal selves. Secondarily we
further e tend our authority over inanimate objects and even other persons, the similar use of “my”
obscuring the processŚ my hand, my suggestion, my car, my wife…
Understanding, buddhi comes closest to the Cartesian res cogitans, except for it being a material
entity. Its definition is simply “Understanding is judgment”, but it is added that “virtue, knowledge,
dispassion and sovereignty are its good (s vika, sattvic) forms, the dark ( masa, tamasic) forms are
the opposites of these.” 17 So judgment is here not only a value-neutral cognitive act, but includes also
moral judgments and decision-making as well. This is also the seat of the relevant motives (passion
and power); therefore it appears that buddhi corresponds to our notions of rationality, volition and
emotions. And it is here, and only here that soul and matter meet – therefore this is the point where
they can part ways at liberation. “Since understanding prepares all e periences for the soul, again only
it can distinguish the subtle difference between materiality and soul.”18
The last two principles – immaterial consciousness or soul and undetectable unmanifest
materiality – are never met with in experience, only inference can prove their existence. They will be
dealt with later in separate sections.
It is worthy of notice that the whole system seems to be basically an anthropology: it represents a
person (a human or any other living being). Although there is room for inanimate objects, provided by
the elements and sensibilia, these physical entities also constitute biological bodies.
Further, this anthropology focuses on information processing, it shows a refined, bidirectional
multi-layered system. To illustrate the point with the inward path: it starts with physical objects (the
elements); the sensibilia carry the information from them, then pass it on to the corresponding senses;
manas coordinates the information, arranging it into a coherent picture; egoity relates it to the subject
(“I see”, “he wears my shirt”, “I am afraid of dogs”19); understanding grasps the situation conceptually,
and presents it to the soul, that experiences it.20 It is clear that such a detailed analysis of cognition is
fuller and therefore perhaps superior to and potentially more promising than the more abstract,
widespread two-step model of immediate and conceptualised perception (nirvikalpaka–savikalpaka
pratyak a).
Here is a table of the twenty-five entities (in the bottom row), normally listed from left to right,
together with their groups and combinations. Mere groups are signified by the plural (like “11
powers”), whereas combinations, i.e. functional higher units are in the singular – like “3 internal
instrument”, read “the internal instrument that can further be analysed into three entities”. In brackets
appear those names and numbers that are not actually used in the SK.
Sāṁkhya: Dualism without substances F. Ruzsa 6

[25] entities (tattva)

conscious
[24] nature (prakr i)
(cetana)
[2] inferable only
[23] manifest (vyakta)
(liṅgin)
10 field of activity
13 instrument (karaṇa)
(k rya)
3 internal instrument 10 external instrument
(antaḥkaraṇa) (b hya [karaṇa])
11 powers (indriya)
5 active
5 senses 5 sensibilia 5 elements
powers
(buddhîndriya) ( anm ra) (bhū a)
(karmêndriya)

sight (cak us) speech (v c) colour (rūpa) [earth]

unmanifest under- egoity hearing(śro ra) hand (p ṇi) sound(śabda) [water]
soul (puru a) materiality standing (ahaṁ- manas smelling(ghr ṇa) foot (p da) [smell] [fire]
(avyakta) (buddhi) k ra) tasting (rasana) anus (p yu) [taste] [air]
skin (tvac) lap (upastha) [touch] [ether]

Cosmogony or psychology?
A particularly nasty problem in reading S ṁkhya texts is the unclear distinction between cosmology
and cosmogony on the one hand, and psychology and individual evolution on the other. This is but one
appearance of the microcosm–macrocosm homologization pervading practically the whole of Indian
culture, most apparent in the ubiquitous pantheism, where the cosmos is viewed as the body of God in
one sense or another. The classic e ample is Arjuna’s vision of Krṣṇa in the the 11th canto of the
Bhagavad-G (a remarkably S ṁkhyaistic text). This is a truly archaic feature of Hinduism and
S ṁkhya, going back to prehistoric ideas and appearing already in the gveda,21 although it is perhaps
not of Aryan origin.22 A somewhat similar situation obtains in Vasubandhu’s Abhidharma-Kośa,
heavily influencing Tibetan Buddhism in this respect; here mental (meditational) states and
cosmogonical events are identified.23
The problem, according to Michel Hulin is that “the S ṁkhya thinkers did not pay much attention
to dilemmas that are crucial to us, like ‘is there only one cosmic buddhi [intellect] or as many buddhis
as individual beings?’”24 Eli Franco points out the absurdity in no unclear termsŚ “Typical
psychological and individual terms like cognition, ego, mind, sense organs, and even hands, feet,
tongue, anus and penis, become trans-individual and obtain cosmological dimensions.”25
Hulin’s e planation is that since the followers of S ṁkhya strove to leave this world of individual
material existence, the confusion was not relevant for them:
in the wake of discrimination, there is no ground anymore to contrast the personal with the
universal perspective. As for the “temporary” continuation of individual, psychic experience,
the S ṁkhya thinkers, quite understandably, were prepared to admit a certain degree of
apparent contradiction within it, as a mark, so to say, of its ultimate lack of authenticity.26
Sāṁkhya: Dualism without substances F. Ruzsa 7

Johannes Bronkhorst found traces in several commentaries indicating that the ambiguity was resolved
in some sub-schoolsŚ the first principles were cosmic, the last ones individual. “If the thinkers of
classical S ṁkhya did indeed not confuse these two, they must have somewhere drawn a line, in the
middle of their evolutionary scheme, to distinguish between cosmological and psychological (or
rather: individual) essences (tattva).”27 His carefully worded conclusion is,
Does this [ambiguity] still hold true for the main thinkers of classical S ṁkhya? As we now
know, the answer must be a qualified no. It is true that cognition and ego – i.e. mahat/buddhi
and ahaṁk ra – appear to have been shared, and therefore cosmological, entities for some,
though not all S ṁkhyas. Other elements – in particular mind, sense organs, as well as hands,
feet, tongue, anus and penis – were looked upon as only individual, not trans-individual or
cosmological entities. The anm ras remain enigmatic… (p. 688.)
Bronkhorst did not analyse the SK itself, probably because he thought it self-evident that it is the very
source of the confusion. But that is not trueŚ in the relevant passages varakrṣṇa’s meaning is always
individual. He just selected a wording that could be interpreted cosmogonically, thereby
accommodating his text to the needs of some traditionally minded subschools.
If somebody really wanted to combine a cosmological account with the description of individuals,
that was easily done in a coherent way – as e.g. Vijñ nabhikṣu (16th century) did in his commentary
on S ṁkhya-Sū ra 3.10:
If there is one subtle body (liṅga, transmigrating entity), how could the experiences of each
person be different? On this the Sūtra saysŚ
“10. Differentiation to individuals is according to the particular karma.”
Even if at the beginning of the creation there is but a single subtle body, connected to the god
Hiraṇyagarbha (Golden Womb), still later it will be differentiated to individuals, its parts will
also be separate as individuals. As now the parts of the single subtle body of the father will be
separate as the subtle bodies of his sons and daughters. The Sūtra says the reasonŚ “according
to the particular karma”, and that means that according to the acts (karma) causing experiences
etc. of different beings.28
This model was widely used in Ved nta te ts, e.g. in Sad nanda’s Ved n a-S ra. But the idea itself is
ancient, already in the Puru a-hymn of the gveda (10.90) everything in the world originates from the
different parts of a primordial cosmic giant.
But clearly varakrṣṇa did not want anything like that – there is no sign of a two-step creation,
first cosmic, then individual. We have instead the problematic, ambiguous passage. Is he talking about
the origin of the world, or an individual, or both?
The contact of the soul and the unmanifest: that causes the creation. From nature the Great,
from that egoity, from that the group of sixteen; also from among those sixteen from five, the
five elements. … Egoity is appropriation. From it, two kinds of creation proceedŚ the eleven
powers and the five sensibilia. … The sensibilia are simpleś from them, the elementsŚ five
from five.29
In S ṁkhya technical terminology, the Great is just another name of the buddhi, understanding;
probably it is just short for the Great Self ( man Maha ). At the beginning of the series we have the
clearly single, cosmic unmanifest nature, at the end the plural, but again cosmic elements (earth, water
etc.) and sensibilia (smell, sound etc.). In between there are entities characteristic only of living
beings: understanding, egoity, the coordinator manas, the senses and the biological faculties.
Sāṁkhya: Dualism without substances F. Ruzsa 8

As an abstract, philosophical creation story it is very attractive. After the first contact of
unconscious matter with soul, there arose sentience (buddhi) in matter; and with it, differentiation
started (I and not-I, subject and object – that is, the principle of egoity); with this difference, the
subject could react to the object (perception), and then the subject could influence the object (the
karmêndriyas, the powers of action); the object, as perceived, distinguished the sensibilia sound,
colour etc.; and finally, the combinations of the sensibilia resulted in the five basic elements.
If this is the creation of a single Great Person, then it must have started with a single, very special
soul – and he would be God, śvara. Then why does everybody contrast godless (nir śvara) S ṁkhya
with theist (sêśvara) oga? Even worse, did varakrṣṇa simply forget about all the simpler
individuals like us, since there is not even a mention of the origin of other beings?
I think that this text is not a creation story at all,30 and its primary reading is just a list of the
components of any living being, indicating the interrelations of the components. varakrṣṇa may have
belonged to that group in the S ṁkhyan fold who taught that the connection of the soul and matter is
beginningless; or he may have been agnostic on the point. But, since he was trying to write the classic
for all subschools, he offered this description in a form that can be read as the beginning of such a
connection, either as a new soul entering an eternal world, or even as a creation or re-creation story.
Later tradition preferred the last version, because the idea of the cosmic cycles gained general
acceptance in Hinduism.
The key to the possibility of this multiple interpretation is the term sarga, so far carelessly
translated in its usual sense of ‘creation’. But it can have other meanings as wellŚ etymologically it is
pouring out, emission; and it can mean a troop and a herd as well. In the SK it is a frequent word, let
us consider its occurrences:
In verse 21, sarga is the result of the contact of soul and matter (in the next verses followed by the
description of the twenty-three manifest entities). In verse 24, we have the sarga of the eleven powers
and the sarga of the five sensibilia; in 46, the sarga of mental states (pratyaya); in 52, the sarga of the
transmigrating entity (liṅga, consisting of eighteen tattvas) and the sarga of the dispositions (bh va).
Verse 53 speaks about the divine sarga with eight divisions, the fivefold animal sarga and one kind of
human sarga; taken together, the sarga of living beings. In 54, sarga is clearly the same, i.e. of living
beings. Finally, in 66 we find a pun: here sarga is both ejaculation and the continuation of physical
existence.31
It seems that the dominant meaning of sarga in the SK is clearly ‘group’, especially of some
fundamental entities or categories of the system. And thus the beginning of the apparent “creation-
story” will be simplyŚ “The group (of the twenty-three manifest entities) is the result of the contact of
matter with soul.”32 The following series of Ablatives (‘from …’ – starting with ‘From nature the
Great’, SK 22 etc. quoted above at note 29) do not necessarily express origination, rather they suggest
direction of dependence.
To sum upŚ varakrṣṇa gave the description of the constituents of a human being in a traditional
form, so that those to whom it was important could read it as an emanation scheme, either of an
individual or of the whole cosmosś or of both. In varakrṣṇa’s philosophy such an emanation has no
role to play and no attempt is made to substantiate it. Still, it is a value-ordered presentation starting
from the innermost core of a rational human and moving outward, “down” to the physical constituents
of the body and of the external world.
Sāṁkhya: Dualism without substances F. Ruzsa 9

Soul as pure consciousness

In the complex system of the twenty-three entities of manifest nature we have all the functions of the
Cartesian res cogitans fulfilled by understanding, egoity and manas. It seems to be a closed
naturalistic description of the worldś but in S ṁkhya there is also an immaterial entity, the soul,
puru a. What could its function be? It is not needed even for such popular purposes as immortality,
since the transmigrating entity is immortal, but it is just a complex of the intangible material entities.33
Puru a is pure consciousness; today it is easier to grasp the concept than half a century ago. If we
feel, as most of us do, that our computers are fully unconscious; and if we think that it would not
change even if they were even more advanced and would be able to do whatever an average human
intellect can do; and such computer would control a nice human-looking artificial body – we got the
idea. Consciousness is what a robot does not have, while a dog has it.
This concept is extremely exciting philosophically as Paul Schweizer demonstrated in his
classical paper.34 Instead of the familiar body/mind dualism, we have here a ‘new’ type,
mind/consciousness dualism. Schweizer showed that “the ancient S nkhya-Yoga version of dualism
provides a more felicitous dividing line between substances than does the Cartesian parsing of mind
and matter”, because it “avoids one of the most serious pitfalls of Cartesian dualism, since on the
Indian account, mental causation does not violate physical conservation laws”. This is because the
mind (the internal instrument), in itself unconscious, is a subtle, but material organ, while
“consciousness plays no causal role in the transformation of mental structures, but rather is a passive
‘witness’ to some small portion of these”.35
Another, equally important feature of this analysis is that it neatly separates the two salient
problems of the magic of our internal experience, which are often considered two sides of the same
coinŚ subjectivity and intentionality. Intentionality, or ‘intrinsic aboutness’, or the ‘semantic content’
of our minds’ information-processing will largely appear as an illusion.
It is the conscious, subjective aspect of visual perception which serves to motivate the
introduction of a distinct metaphysical category, not the causally induced representational
structure of perception, since it is theoretically feasible that the latter can be explained in terms
of unconscious mechanisms, of generally the same sort that would be applied in the case of
robotic ‘vision’. … It is consciousness, rather than content, which provides the most
compelling impetus for dualism.36
It is worthy of noticing (as Schweizer does) that this concept of consciousness does not automatically
entail a substance dualism. One might be inclined rather to follow Searle, who in his Chinese room
thought-e periment suggested that consciousness is the result of the brain etc. having special “causal
powers”.37
Since consciousness in this interpretation is detached from intentionality, it seems logical to
construe it as completely without content. Indeed, that is what happens in later S ṁkhya; this tendency
is already dominant in the commentaries of the SK. Now this rather peculiar notion was perhaps best
described by Gerald Larson, so I will quote him at length:
[S ṁkhya is] a dualism between a closed, causal system of reductive materialism
(encompassing “awareness” or the “private” life of the mind), on the one hand, and a non-
intentional and contentless consciousness, on the other. … [C]onsciousness (puru a) cannot
think or act and is not ontologically involved or intellectually related in any sense to
primordial materiality other than being passively present. Consciousness, in other words, is
sheer contentless presence (s k itva).38
Sāṁkhya: Dualism without substances F. Ruzsa 10

It is outside the realm of causality, outside space and time, completely inactive, utterly simple,
unrelated apart from its sheer presence, uninvolved in emergence or transformation, without
parts, completely independent … neither an object nor a subject (in any conventional sense),
verbally uncharacterizable, a pure witness whose only relation to primordial materiality is
sheer presence, utterly isolated, completely indifferent, … a nonagent…39
Where can this disquieting, anomalous conception of the subject come from? It is difficult not to
recognise the Ved ntic idea of absolute pure consciousness, i.e. man=Brahman, and I think this
influence must have been overwhelming in the long run. In Advaita Ved nta, however, quite logically
there is only one of this pure abstraction of consciousness shared by all sentient beings, while in
S ṁkhya each person has their own unique soul.
In the S ṁkhya-Yoga tradition, however, there is a strong introspective evidence for a concept
like this, since here meditation is over-important. The fundamental form of meditation is emptying the
mind, which means completely relinquishing the internal stream of thought, abandoning the normally
never-stopping silent debate inside. In this state the meditator experiences nothing, thinks nothing,
feels no change at all – yet they still remain themselves. “ oga is the suppression of the functions of
the mind. Then the soul remains in its own form.”40 Therefore while for us a non-thinking self is a
very difficult and perhaps unnatural idea, something like this was almost self-evident in the Indian
tradition.
The timeless, causally unrelated, contentless consciousness characterised by Larson above,
unfortunately, can have no function in the system (or in a person), and therefore practically destroys
the S ṁkhya position. (Historically it did disappear.) As ntarakṣita, the great Buddhist scholar in the
8th century objected in his Tattva-saṁgrahaŚ “if consciousness endured always in the same form, how
could [it] be the enjoyer of many kinds of objects?”41 And there can be no doubt that in S ṁkhya the
soul is the “enjoyer”, the e periencer (bhok r), for it is one of the key arguments used by varakrṣṇa
to prove the soul’s e istence (in SK 17). In fact his concept of the soul was far less abstract than that
of the commentators.42
varakrṣṇa clearly maintained that the soul is inactive (a-kar r, SK 19–20), but only in the sense
that it does not move in space43 and it cannot move a physical object. It is also “qualityless” (not tri-
guṇa, SK 11), but only in the very special sense that it does not have the three guṇas of nature, which
are but the fundamental aspects of materiality. So it means only that the soul is immaterial; but it can
have individual features and changing states. “There are many souls; for birth, death and the
instrument are regulated individually, and we do not act at the same time.”44 This would be clearly
meaningless, if the souls were not responsible somehow for the individual acts.
The soul is the subject – an isolated, neutral spectator or witness.45 This aloofness means only that
the soul cannot be killed or objectively harmed, but it can really feel badŚ “In this world the conscious
soul suffers from the pain caused by aging and death.”46 Like someone on learning the loss of their
riches from the news can really be pained without any physical harm done. The soul identifies with the
body-mind complex and therefore feels the processes of life as its own. It is clearly stated to be the
knower and experiencer.47
Although it is repeatedly said that the puru a is not the agent, this can mean only that it lacks the
ability to physically move an object. The terminology of the two great systems of the material world –
the “instrument” (karaṇa), the thirteen psycho-physiological entities; and the “field of activity”
(k rya), the ten physical entities – seems to require an agent (kar r), but we are told that “no-one
operates the instrument”ś48 only “the passive one seems like the agent, although the guṇas are the
agent”.49 At another place, however, the soul seems to govern or superintend (adhi- ṭh , SK 17) the
Sāṁkhya: Dualism without substances F. Ruzsa 11

person. How is this possible, since it cannot move or change anything that is material? The only
answer that we get again and again is that nature follows the soul’s aim or purpose (SK 13, 17, 31, 42,
56, 58, 60, 63, 65, 68, 69).
To sum upŚ varakrṣṇa’s soul or consciousness, the source of subjectivity and inherently private
experiences is an immaterial entity not in space, but in time. It has temporal states, it is conscious of
our experiences: it reacts to materiality (more exactly, to understanding or perhaps the whole of the
mind). It “sees” and it suffers. It cannot move physical objects, but nature (more e actly,
understanding) reacts to its states, “follows its aim”.50

Materiality, the three qualities and the unmanifest

The S ṁkhya analysis of matter is another highlight of the system. Prakr i, ‘nature’ is not like the
everyday European mechanical concept of dead matter, modelled more or less on billiard balls. It is
not dead stuff, rather an alive creative process: it is essentially productive. It is not a loose heap of
individual substances like atoms but continuous or inseparable. Most importantly it is made up of the
three qualities.51
The theory of the three guṇas or qualities is the most important contribution of S ṁkhya to Indian
culture. In varakrṣṇa’s presentation it is a refined, grand vision of everything there is, e cept for the
soul. It is a relevant description of everyday physical processes as well as mental phenomena. Here
again we meet with the problem: are the guṇas psychological or cosmological? In this case, however,
the answer is clearly “both”. And it is not due to varakrṣṇa’s failure to grasp the difference, rather to
a deep insight into the fundamental, common structure.
The three qualities, tamas, rajas and sattva52 represent three aspects of any material phenomenon;
as some reflection shows, it is impossible to conceive of any material object where any of the three is
missing. Tamas is inertness, the tendency to remain the same; therefore mass, conservation and
stability, rigidness. Rajas is energy, mobility, the impulse to move and change, repulsion and
separation. Sattva is information, ordering, structure, coherence, cohesion and attraction. These are the
three threads (another meaning of the word guṇa) of which reality is spun. They are omnipresent, but
not distributed equally – lightning is more rajasic than a tree, a book is more sattvic than a stone. Their
inherent dynamism, cooperation and rivalry drives all processes from evolution to decay.
There is a facile, ubiquitous tendency to identify the guṇas (at least as psychic factors) with the
triad happiness, pain and bewilderment (sukha, duḥkha and moha). Although the commentaries
frequently do this, thereby divesting the theory from much of its explanatory power (but making it
easier to grasp), there is nothing in the SK that could justify or even allow this simplification.
There is another material entity beyond the twenty-three already discussed, namely the unmanifest
(avyakta; or principal, pradh na). This is a theoretical construct, it can never be experienced. Its
supposition is necessitated by the unity of the world: there must be something invisible that keeps the
whole thing together, that guarantees that everywhere the same things or events are possible, that
provides for the universal validity of natural laws. The idea is somewhat similar to empty space, but
more alive, perhaps it could be compared to a physical force field, especially a gravitational field in
the general theory of relativity.
Since the unmanifest is the underlying cause of all material entities, its properties can be deduced
from its effects. The S ṁkhya theory of causation is about the relation of two things (not events). It is
a mildly deterministic theory stating that nothing “une pected” appears in the process, the features of
Sāṁkhya: Dualism without substances F. Ruzsa 12

the effect are grounded in the features of the cause(s). Basically this is a rejection of the possibility of
emergence (in the strong sense).53 This theory necessitated the acceptance of the non-material soul:
since there is no consciousness in the guṇas, but we are conscious, without the soul it would be a case
of emergence.54
As all material entities are characterised by the three qualities, the same must be true of the
unmanifestŚ “the unmanifest is proven to be such [as described in SK 11, see note 51], since the effect
consists of the qualities of the cause”.55 Perhaps the difference is that since the unmanifest is
unmanifest (i.e. imperceptible), here the guṇas must be present in their potential form.

Substance reductionism
The three qualities are not properties or attributes of something else (a substance or substrate):
forgetting about the soul for the moment, there is nothing beyond them. They do not characterise an
object different from them – they themselves make up the whole word and all the things within it. This
well-known position of S ṁkhya is, in fact, not distinct from the basic concept of the three qualities. It
logically follows from the statement that everything is characterised by the three guṇas. For if they
were properties of a substance different from them, that substance in itself should be without the three
qualities, and that is ex hypothesi an impossibility.
So the three qualities do not qualify substances, they constitute them. There is nothing in the
substances but the three guṇas. This seems to be a particularly clear case of a strong substance
reductionist theory, yet S ṁkhya is not normally understood in this way in modern research. The
reason is not very clear for me; I suspect that perhaps the very idea is so alien to our regular European
ways of thinking that most students could not take this statement at face value. Since the position of
S ṁkhya (stating that everything is the three guṇas and nothing else) could not be called into question,
the only way open was to reinterpret the guṇas as not qualities at all. Perhaps few scholars would now
go so far as directly stating that they are substances (as Dasgupta or Hiriyanna did 56); rather they opt
for an innovative terminology, like Eliade’s ‘modes of being of prakr i’ or Larson’s ‘constituent
process’. The most widespread solution is to use the word ‘constituent’.57
One difficulty with this approach is that constituents can occur singly, and you can mix them, but
in standard S ṁkhya there is no such thing as “pure sattva” to which some “unmi ed rajas” could be
added. The guṇas always appear togetherś they could be called ‘constituent qualities’ (for qualities
cannot appear alone), but perhaps ‘aspects of materiality’ is more directly illuminating.
In any case, if the guṇas were something else and not qualities (in some sense), why would they
be called that? The usual answer is similar to Wezler’sŚ “The conception of the three guṇas … was
developed at a time when Indian thinkers ‘had not yet learned’ to distinguish between substance as
such and i t s qualities or properties.”58 I find this suggestion implausible, considering varakrṣṇa’s
extremely refined conception of the three guṇas and at the same time the plain impossibility of the
supposition that he was not perfectly aware of the Vai eṣika analysis of substance and quality (dravya
and guṇa).
This consensus of modern scholarship started to change with Wezler’s fundamental paper from
1985 quoted above, where he identified some S ṁkhya tradition where substance is considered
nothing but a combination of qualities – this time, however, of undisputable qualities, i.e. the sensibilia
colour, sound etc. He found this doctrine mentioned both after varakrṣṇa (in the Jaina Mallav din’s
Dv daś ra-Naya-Cakra and its commentary by Siṁhasūri) and also much earlier (2nd cent. BCE) in
Sāṁkhya: Dualism without substances F. Ruzsa 13

Patañjali’s Mah -Bh ya. Patañjali’s commentator, Kaiyaṭa even connects this theory with the three
standard qualities of S ṁkhyaŚ “Sattva, rajas and tamas are the qualities. The five qualities sound etc.
are developed from them and so consist of them. Pots and the like are compounds made of the [five],
and there is no substance as a whole beyond them.”59
Bronkhorst in two papers60 added further witnesses of the doctrine (Bhartrhari, his commentator
Puṇyar ja, Vasubandhu and Dharmap la), and connected it to an earlier phase of S ṁkhya where the
sensibilia were among the twenty-five entities. He suggested that possibly even varakrṣṇa belonged
to this phase, provided that he understood the anm ras as the sensibilia.61 Alex Watson in his 2006
book aptly summarized previous research, adding R makaṇṭha (10th century) to the known witnesses.
“Despite all these references in the te ts of other schools, neither Wezler nor Bronkhorst were able to
find a statement of this view in a surviving S ṅkhya source. It is a measure of how much that was
central to this tradition is lost… [I]t is striking that an author as late as R makaṇṭha refers to this as
S ṅkhya view centuries after it had been abandoned by S ṅkhyas.”62
Now it seems that this idea was not at all given up by S ṁkhya, at least not before Vijñ nabhikṣu
in the 16th century. In the Yukti-D pik itself there are at least four references to precisely this form of
the doctrine, i.e. where it is stated that substances are but combinations of the qualities and these are
not the usual S ṁkhya guṇas (tamas, rajas, sattva).63 So now we see that this sort of substance
reductionism is proper S ṁkhya, well attested in the classical period – still, it is atypical. It is like
saying that a house is made of clay. A house is built of bricks, and bricks consist of clay; and clay is a
kind of matter. So it is true, a house is made of clay, but S ṁkhya normally would say that it is made
of bricks. Or, of matter.
S ṁkhya, like most Indian traditions, thinks that everyday objects are made up of the five
elements (earth, water, fire, air, ether).64 What is peculiar to S ṁkhya is the position that the elements
consist of the five sensibilia.65 This is a surprising theory; however, it is perfectly convincing, as can
be seen through the following argument: All that I know of the world I learn through the senses. So I
have absolutely no ground to suppose that there is anything in a pot beyond its sensible qualities,
colour, shape, sound etc.
This theory is unambiguously present in the SK “emanation theory” where it states that the five
elements arise out of the five anm ras, which clearly stand for the sensibilia.66 Therefore it is not
particularly important to state separately that everyday objects (as consisting of the elements) consist
of the sensibilia.
Once we have convinced ourselves that substance reductionism is an important element in
S ṁkhya, we are prepared to realise that the guṇas are indeed what their name says – they are
qualities. S ṁkhya is substance reductionist at the deepest level: matter itself, prakr i, is nothing but
its constituent qualities, the three fundamental aspects of any material reality, inertia, energy and
information.
Not only nature itself, but all its manifestations, all material objects and phenomena are reducible
to the three qualities. “Therefore, recognising that bodies, even up to Brahm , cannot abide, for they
are but compounds of sattva etc., like pots…”67 Now this is the fundamental insight of S ṁkhya,
compared to which the particular cases (like the reduction of the objects to the elements, or directly to
the sensibilia; or the reduction of the sensibilia to the three guṇas68) are of little importance and are
therefore seldom mentioned in S ṁkhya texts proper. Perhaps some of these seemed more unusual to
other traditions and therefore they remarked on them pointedly, making these particular forms of
S ṁkhya reductionism more conspicuous in non-S ṁkhyan texts.
Sāṁkhya: Dualism without substances F. Ruzsa 14

varakrṣṇa himself was a conscious and consistent substance reductionist: he never uses a word
denoting substance, like dravya, vastu, dharmin or svabh va.69 The tattvas, ‘entities’ are again either
not substances or clearly reducible; even their names are often action nouns like buddhi,
‘understanding’ or adjectives like avyakta, ‘unmanifest’. We have seen that he derives the elements
from the sensibilia, and he defines all the psycho-physiological factors as functions, not things.
Understanding is judgment, egoity is appropriation etc. He even says that the three factors of mind can
be viewed either separately or as a unitŚ “The particularity of the three is their function; this is not
common. Their common function as an instrument is the five winds (breath etc.) [i.e. the life
functions].”70 And this is not an exception – in the table of the entities above it is clear that there are
several functional groups among them and most of these can be seen as a unit (while distinctness is
supposed to be a characteristic feature of substances). Only nature as a whole, the Universe, or its
abstraction, the unmanifest could be seen as proper substances: but they, too, are nothing but the
combination of the three qualities. Even agency, which is normally thought to be a privilege of
substances, is attributed to the qualities (SK 20, guṇa-kar r ve).
Substance reductionism is an idea that is difficult to grasp for most people in the European
tradition while it is a fundamental and stable feature of S ṁkhya, aiva Siddh nta and Buddhism.
Why could these people accept this concept so easily? Since these philosophies seem to be
independent of the Vedic tradition, it is tempting to speculate that a different linguistic background
may have been at work here. In most European languages and also in Sanskrit the opposition noun–
adjective is particularly strong. We can only speak in substance–quality language, and it makes it next
to impossible to think in another way. In the Dravidian languages (as in Tibetan and Chinese: that may
be relevant for Buddhism) there is no separate grammatical category for adjectives, so for people who
can think in these languages substance reductionism (to qualities) is no more alien than analysing a
substance in terms of its components, e.g. elements.71

The Buddha’s master

Insubstantialism is a very characteristic philosophical doctrine; and we see now that it is shared by
S ṁkhya and Buddhism.72 Is this a mere coincidence? Clearly not, for there are a lot more identical or
similar teachings in the two systems. The innumerable common points between early Buddhism and
classical Yoga – which is, after all, a version of S ṁkhya – have been repeatedly pointed out.73 Let us
mention a few central features shared by the SK and the Buddha.
The starting point of both philosophies is the universality and unavoidability of suffering, duḥkha.
This is the first noble truth ( rya-satya) of the Buddha, and the very first word of the SK; its ground is
the transience of all existence: aging and death.74 And, of course, both traditions promise to show the
way out.
They give an analysis of the world only in as much it is essential for liberation. They do not give a
description of the cosmos, the Earth, the Sun or the seas; they do not talk about the gods, although
their existence seems to be naturally accepted.75 In fact their description of reality is basically an
anthropology only. Clearly the picture is almost identical (with the obvious difference of the puru a,
the non-material soul in S ṁkhya), although the terminology differs. Both analyse the human being as
a compound of five entitiesś listing these, I first give the S ṁkhya concept, then the Buddha’s term
taken from the frequently mentioned group of the ‘five skandhas’. 1. Body (and all e ternal physical
realities): k rya–rūpa; 2. the connection between external and internal: b hyakaraṇa–vedan ;76 then
Sāṁkhya: Dualism without substances F. Ruzsa 15

the mind analysed into three factors, i.e. 3. coordinator: manas–saṁjñ ; 4. personal colouring:
ahaṁk ra–saṁsk r ḥ; 5. conceptual thinking: buddhi–vijñ na.
In the liberating praxis of both, a key feature is the disidentification of the subject with exactly
these factors.77 Even the wording is practically identical: “Practicing so with the entitiesŚ ‘I am not
[this], [this] is not mine, [this] is not the I’…” in the SKś “He considers [the five skandhas] so: ‘this is
not mine, I am not this, this is not my self’” in the e tremely important sermon of the Buddha, The
simile of the cobra.78
Lastly, probably as a logical consequence of their rejecting static substances, both traditions
emphasise causally determined processes of change, in S ṁkhya called pariṇ ma, in Buddhism
pra ya samu p da (typically translated ‘evolution’ or ‘change’, and ‘dependent origination’ or
‘conditioned arising’, respectively).
Now with so many points in common, it is natural to ask: who learned from whom? Although the
two schools must have been mutually influencing each other for centuries, still I think that the basic
direction can be identified. The Buddha was not shy to point out if something was his own discovery
or innovation, like the Middle Way (rejecting ascetic practices) or his gentle meditation technique
(‘letting go’ instead of oga’s ‘oppressing’). In case of his anthropology, the five skandhas, he clearly
uses a system he inherited; for in his very first sermon he merely mentions it without feeling the need
to explain the idea.79
After Gautama had left his home seeking the way to liberation from suffering, he had two
teachers. Ar ḍa K l ma, his first teacher, was – according to his earliest detailed biography80 – a
S ṁkhya philosopher. Of course, this late tradition could be mistaken; the older sources are less
e plicit. From the Buddha’s account of his own education, we learn only that K l ma had a school
(gaṇa) with two texts (ñ ṇav da and herav da, ‘knowledge-speech’ and ‘speech of the elders’,
probably a sū ra and its commentary), where they practised awareness and meditation (sati and
sam dhi). The deepest, last teaching, that Gautama himself directly realised (sacchi-kata) was
kiñcañña- ya ana, that seems to mean ‘the state of not having anything’.81 And this sounds
suspiciously similar to the disidentification-meditation mentioned above.
According to unanimous tradition, Gautama came from Kapila-vastu, ‘the ground of Kapila’ –
and Kapila is, of course, the legendary founder of the S ṁkhya school. So it seems perfectly
consistent with all that we know of him that the inherited part of the Buddha’s doctrine comes from an
early form of S ṁkhya.
What remains to be proven is that substance-reductionism is an old enough feature of S ṁkhya to
make it plausible that this is actually the source of the Buddha’s (and Buddhism’s) insubstantionalist
doctrine. Part of it is already clear, since the five-skandha anthropology (and especially the term rūpa,
‘form’ for the body) is in itself a reductionist approach, dissolving the unified concept of person as one
substance – and we saw that the Buddha inherited this concept.
Pañca ikha is one of the most famous ancient S ṁkhya masters. In the Mah -Bh ra a (12.211–
212) he leads king Janaka to liberation; this text is one of our most important sources for early
S ṁkhya.82 The text is somewhat unclear (as many philosophical texts in the Mah -Bh ra a), but the
key motifs of substance-reductionism and of disidentification are unmistakable:
Endless suffering will not cease for him who, through wrong views, sees this collection of
qualities as being himself. But where could the continuous flow of suffering, clinging to the
word, find a place, when the view is ‘not-self’, and also ‘I am not’ and ‘not mine’?83
Sāṁkhya: Dualism without substances F. Ruzsa 16

Janaka, the king of Videha (where the Buddha also spent some time) is more famous as the patron
of jñavalkya, the sage of the Brhad- raṇyaka Upani ad. Many of his questions and even words are
the same as those of Pañca ikha here. jñavalkya’s master was Udd laka ruṇi,84 the greatest
teacher of the Ch ndogya Upani ad. Now ruṇi’s teaching has many significant S ṁkhya features,85
most notably a substance-reductionism to three qualities! He calls them ‘forms’ or ‘colours’ (rūpa),
not guṇas, but the idea is exactly the same. They are present in everything, and there is nothing else in
the objects.86
And, as luck would have it, it can be proven that the Buddha knew this teaching. For he refers to a
very unusual view, shared only by ruṇi and jñavalkya,87 according to which a person at death
simply dissolves into the Great Being, also called the Self, man. The Buddha, of course, thinks that
this is a harmful theory:
There is this wrong viewŚ “The Self and the world are the same. After death I will be thatŚ I
will stay identical for eternal times, constant, stable, eternal, essentially unchanging.” He
considers it so: “this is mine, I am this, this is my self.”88
So the Buddha inherited his substance-reductionist ideas from the proto-S ṁkhya circles of ruṇi and
jñavalkya. What his own innovation was, he made perfectly clear: rejecting their man-substance,
he created a system entirely without substances, and so – to the dismay of many – a religion with no
soul.

Bibliography

Indic texts quoted

Gauḍap da’s commentary on the SKŚ
Mainkar, T.G. (2004), S ṁkhyak rik o śvarakr ṇa: With the Commentary of Gauḍap da.
(The Vrajajivan Indological Series 36.) Delhi: Chaukhamba Sanskrit Pratishthan. (Includes text
and translation.)
Mah -Bh ra a:
Dandekar, R.N. (ed) (1971–1976), The Mah bh ra a: Text as Constituted in the Critical
Edition. Poona: Bhandarkar Oriental Research Institute.
Mūla-Madhyamaka-K rik :
Kalupahana, D. J. (1986), Mūlamadhyamakak rik o N g rjuna: The Philosophy of the Middle
Way. New York: State University of New York. (Includes text and translation.)
Ny ya-Sū ra:
Nyaya-Tarkatirtha, Taranatha and Tarkatirtha, Amarendramohan (eds) (1936–1944),
Ny yadarśanam wi h V sy yana’s Bh ya, Uddyo akara’s V r ika, V caspa i Miśra’s
T paryaṭ k Viśvan ha’s Vr i. Calcutta: Metropolitan Printing & Publishing House. (Repr.
Delhi: Munshiram Manoharlal, 1985.)
Translation:
Vidy bhūṣana, Satisa Chandra (tr.) and Sinha, Nandalal (rev., ed) (1930): The Ny ya Sū ras o
Gotama. AllahabadŚ P ṇini Office. (Repr. Delhi: Motilal Banarsidass, 1990.)
Vijñ nabhikṣu’s commentary on the S ṁkhya-Sū ra:
Garbe, Richard (ed) (1895), The S ṁkhya-Pravacana-Bh ya or Commentary on the Exposition
Sāṁkhya: Dualism without substances F. Ruzsa 17

o he S nkhya Philosophy by Vijñ nabhik u. (Harvard Oriental Series 2) Cambridge, Mass.:

Harvard University.
SK, S ṁkhya-K rik :
Ruzsa, Ferenc (ed) (2014b), śvarakr ṇa’s S ṁkhya-K rik .
https://www.academia.edu/4757287/Sa_khya-Karika_-_reordered_text_and_words
Translation:
Larson, Gerald James (1979), Classical S ṁkhya. An Interpretation of its History and Meaning
(2nd edn). Santa Barbara: Ross/Erikson, pp. 255–277.
YD, Yukti-D pik :
Wezler, Albrecht and Motegi, Shujun (eds.) (1998), Yuk id pik : The Most Significant
Commen ary o he S ṁkhyak rik . Stuttgart: Franz Steiner Verlag.
Yoga-Sū ra:
Maas, Philipp André (2006), Sam dhip da: Das ers e Kapi el des P añjalayogaś s ra zum
ersten Mal kritisch ediert. Indologica Halensis – Geisteskultur Indiens. Texte und Studien, 9.
Shaker Verlag, Aachen.
Pali texts are quoted from the Vipassana Research Institute’s online edition, http://www.tipitaka.org.

Secondary sources
Bakker, Hans and Bisschop, Peter (1999), “Mok adharma 187 and 239-241 reconsidered”, Asiatische
S udien / É udes Asia iques, 53/3, 459–472.
Bronkhorst, Johannes (1983), “God in S ṁkhya”, Wiener Zei schri ür die Kunde Südasiens, 27,
149–164.
Bronkhorst, Johannes (1994), “The qualities of S ṁkhya”, Wiener Zei schri ür die Kunde Südasiens,
38 (Orbis Indicus, Festschrift G. Oberhammer), 309–322.
Bronkhorst, Johannes (1997), “S ṁkhya in the Abhidharmako a Bh ṣya”, Journal of Indian
Philosophy, 25, 393–400.
Bronkhorst, Johannes (1999), “The contradiction of S ṁkhya. On the number and the size of the
different tattvas”, Asia ische S udien / É udes Asia iques, 53/3, 679–691.
Bronkhorst, Johannes (2007), Greater Magadha. Studies in the Culture of Early India. (Handbook of
Oriental Studies, Section 2: South Asia, 19.) Leiden: Brill.
Bronkhorst, Johannes (2010), “Kundakunda versus S ṁkhya on the soul” in Nalini Balbir (ed), Svasti:
Essays in Honour o Pro . Hampa N gar jaiah or his 75 h Bir hday. Krishnapuradoddi: K. S.
Muddapa Smaraka Trust, pp. 215–226.
Dasgupta, Surendranath (1922–1955), A History of Indian Philosophy. Cambridge: Cambridge
University Press. (Repr. Delhi: Motilal Banarsidass, 1988.)
Dilthey, Wilhelm (1890), “Beiträge zur Lösung der Frage vom Ursprung unseres Glaubens an die
Realität der Außenwelt und seinem Recht” [The Origin of Our Belief in the Reality of the
External World and Its Justification], Sitzungsberichte der königlich preußischen Akademie der
Wissenschaften zu Berlin, 1890/2, 977–1024.
Franco, Eli (1991), “Whatever happened to the Yuktidīpik ?”, Wiener Zei schri ür die Kunde
Südasiens und Archiv ür Indische Philosophie, 35, 123–138.
Sāṁkhya: Dualism without substances F. Ruzsa 18

Hacker, Paul (1985), Grundlagen indischer Dichtung und indischen Denkens. (Ed. Klaus Rüping.)
Wien: Publications of the De Nobili Research Library 12.
Hiriyanna, Mysore (1932), Outlines of Indian Philosophy. London: G. Allen & Unwin Ltd. (Repr.
Delhi: Motilal Banarsidass, 1994.)
Hulin, Michel (2007), “The Ego-principle (Ahaṁk ra) as a Key Concept in the S ṁkhyak rik ”, in
Victoria Lysenko and Michel Hulin, Classical Indian Philosophy Reinterpreted. Delhi: Decent
Books (Centre for Sri Aurobindo Studies), pp. 47–60.
Jacobsen, Knut A. (1999), Prakr i in S ṁkhya-Yoga. (Asian Thought and Culture 30.) New York etc.:
Peter Lang.
Jakubczak, Marzenna (2012), “Why didn’t Siddh rtha Gautama become a S ṁkhya philosopher, after
all?” in Irina Kuznetsova, Jonardon Ganeri and Chakravarthi Ram-Prasad (eds), Hindu and
Buddhist Ideas in Dialogue: Self and No-Self. Farnham: Ashgate, pp. 29–45.
Körtvélyesi Tibor (2011), “Ötből áll az ember. Listákba szedett fogalmak a páli kánonban” [Man is
made up of five. Concepts arranged in lists in the Pali Canon], Vallás udományi Szemle, 7/1,
42–66.
Lang, Karen C. (2010), “Candrakīrti’s critique of the S ṁkhya concepts of puruṣa and prakrti”, in
Johannes Bronkhorst and Karin Preisendanz (eds), From Vasubandhu to Caitanya. Studies in
Indian Philosophy and Its Textual History. (Papers of the 12th World Sanskrit Conference.
Volume 10.1), Delhi: Motilal Banarsidass, pp. 53–59.
Larson, Gerald James (1987), “Introduction to the philosophy of S ṁkhya”, in Gerald James Larson
and Ram Shankar Bhattacharya (eds), S ṁkhya: A Dualist Tradition in Indian Philosophy.
(Encyclopedia of Indian Philosophies 4.) Princeton: Princeton U.P., pp. 1–103.
McGinn, Colin (1993), “Consciousness and CosmologyŚ Hyperdualism Ventilated”, in Martin Davies
and Glyn W. Humphreys (eds), Consciousness: Psychological and Philosophical Essays.
Oxford: Blackwell, pp. 155–177.
Motegi, Shujun (1999), “The teachings of Pañca ikha in Mok adharma”, Asiatische Studien / É udes
Asiatiques, 53/3, 513–535.
Ruzsa Ferenc (1997a), A klasszikus szánkhja ilozó iája. [The philosophy of classical S ṁkhya]
Budapest: Farkas Lőrinc Imre Könyvkiadó.
Ruzsa, Ferenc (1997b), “The triple suffering. A note on the S ṁkhya-k rik ”, paper read at the Xth
World Sanskrit Conference, Bangalore.
https://www.academia.edu/472497/The_triple_suffering._A_note_on_the_Sa_khya-karika
Ruzsa, Ferenc (2003a), “Inference, reasoning and causality in the S ṁkhya-k rik ”, Journal of Indian
Philosophy, 31/1–3, 285–301.
Ruzsa, Ferenc (2003b), “The types of suffering in the Mah vyutpatti and the P li Canon”, Acta
Orientalia Academiae Scientiarum Hungaricae, 56/1, 49–56.
Ruzsa, Ferenc (2004), “The Meaning of ruṇi’s Promise”, Indologica Taurinensia, 30, 229–235.
Ruzsa, Ferenc (2007), “The fertile clash. The rise of philosophy in India”, in Dezső, Csaba (ed),
Indian Languages and Texts through the Ages. Essays of Hungarian Indologists in Honour of
Pro . Csaba Tö össy. Delhi: Manohar, pp. 63–85.
Sāṁkhya: Dualism without substances F. Ruzsa 19

Ruzsa Ferenc (2009), “A só-hasonlat az upaniṣadokban” [The simile of salt in the Upaniṣads], in
Bangha Imre and Mészáros Tamás (eds), A Danubio ad usque Gangen. Klasszika- ilológiai,
indológiai és magyar kul úr ör éne i anulmányok a 80 éves Vekerdi Józse köszön ésére.
Budapest: Typotex, pp. 201–211.
Ruzsa, Ferenc (2010), “The authorlessness of the philosophical sūtras”, Acta Orientalia Academiae
Scientiarum Hungaricae, 63/4, 427–442.
Ruzsa, Ferenc (2013), “The influence of Dravidian on Indo-Aryan phonetics”, in Jared S. Klein and
Kazuhiko Yoshida (eds), Indic Across the Millennia: from the Rigveda to Modern Indo-Aryan.
Bremen: Hempen Verlag, pp. 145–152.
Ruzsa, Ferenc (2014a), Key issues in Indian philosophy. (Unpublished DSc thesis.) Budapest:
Hungarian Academy of Sciences.
https://www.academia.edu/4531513/Key_issues_in_Indian_philosophy
Ruzsa, Ferenc (forthcoming), “Is the Cosmic Giant an Indo-European myth?” in Joel Brereton and
Theodore Proferes (eds), Papers of the Veda section of the 13th World Sanskrit Conference.
http://www.academia.edu/472488/Is_the_Cosmic_Giant_an_Indo-European_myth
Schweizer, Paul (1993), “Mind/Consciousness Dualism in S ṅkhya- oga Philosophy”, Philosophy
and Phenomenological Research, 53/4, 845–859.
Searle, John R. (1980), “Minds, brains, and programs”, Behavioral and Brain Sciences, 3/3, 417–424.
Steinkellner, Ernst (1999), “The a ṭitantra on perception. A collection of fragments”, Asiatische
S udien /É udes Asia iques, 53/3, 667–677.
Szegedi Mónika (2009), “Szándéketika és pszicho-kozmikus összeomlás: apokaliptika a
buddhizmusban” [Intention based ethics and psycho-cosmic collapse: apocalypse in Buddhism],
Vallás udományi Szemle, 5/3, 32–53.
Tandon, S. N. (1995), A Re-appraisal of Pa añjali’s Yoga-Sutras in the Ligh o he Buddha’s
Teaching. Dhammagiri: Vipassana Research Institute.
Watson, Alex (2006), The Self's Awareness of Itself. Bhaṭṭa makaṇṭhas’s Argumen s agains he
Buddhist Doctrine of No-Self. Wien: Publications of the De Nobili Research Library 32.
Watson, Alex (2010), “R makaṇṭha’s concept of unchanging cognition (ni yajñ na). Influence from
Buddhism, S ṁkhya and Ved nta”, in Johannes Bronkhorst and Karin Preisendanz (eds), From
Vasubandhu to Caitanya. Studies in Indian Philosophy and Its Textual History. (Papers of the
12th World Sanskrit Conference. Volume 10.1), Delhi: Motilal Banarsidass, pp. 79–120.
Wezler, Albrecht (1985), “A note on Mah bh ṣya II 366.26: guṇasaṁdr vo dravyam. (Studies on
Mallav din’s Dv da ranayacakra II)”, in Buddhism and Its Relation to Other Religions: Essays
in Honour of Dr. Shozen Kumoi on His Seventieth Birthday, Kyoto: Heirakuji Shoten, pp. 1–33.

Notes

1
McGinn (1993).
2
The SK will be quoted based on my edition, where some text-critical information is also given.
When a variant reading specified in the edition seems relevant for the purposes of this paper, it will be
Sāṁkhya: Dualism without substances F. Ruzsa 20

marked here with ‘v.l.’ All Sanskrit texts quoted will be standardized, with hyphens, sandhi-markers
and punctuation added. All translations are mine except when noted explicitly.
3
Larson (1987: 77).
4
“[T]hough the Mok adharma as we have it offers already a bewildering diversity of often
contradicting views, the historical reality at the time of its first composition was still more complex –
each ashram, so to speak, having its own competing version of proto-S ṁkhya philosophy and being
keen on having it canonized.” Bakker et al. (1999: 468).
5
Most scholars consider this text very late (c. 14th century). I have tried to show elsewhere that
this opinion is not well-founded and that in all probability it is earlier than the second century CE.
Ruzsa (2014a: 101–107).
6
At least the Ny ya-Sū ra is clearly referred to.
SK 5: trividham anumānam khy am, “it has been e plained that inference has three kinds.”
SK 6: sāmānyatas tu drṣṭād atîndriy ṇ m prasiddhir anumānāt, “Things beyond the senses are
proved through inference by analogy”.
Ny ya-Sū ra 1.1.5: atha tat-pūrvakaṁ trividham anumānam: pūrvava , śe avat, sāmānyato
drṣṭaṁ ca, “Then inference is based on that [i.e. perception]; it has three kinds: like before, like the
rest, and by analogy [lit. seen by genus].
A quasi-quotation from N g rjuna’s Mūla-Madhyamaka-K rik (16.1&5) again seems
unmistakable:
1 saṁsk r ḥ saṁsaranti cen, na ni y ḥ saṁsaranti te;
saṁsaran i ca nâni y ḥ. sattve 'py e a samaḥ kramaḥ.
5 na badhyante, na mucyanta udaya-vyaya-dharmiṇaḥ
saṁsk r ḥ; pūrvava sa vo badhyate na, na mucyate.
“1. If composites transmigrate, they do not transmigrate as eternal; but temporary things do not
transmigrate. The logic is the same with a living being. 5. Composites, whose nature is to rise and pass
away, cannot be bound or freed. Similarly a living being is not bound and not freed.”
Compare SK 62: asm n na badhyate 'ddh na mucyate nâpi saṁsarati kaś-cit. “Therefore no-
one is bound, nor freed, and also does not transmigrate.”
Also the fact that the plurality of souls needed proof (SK 18) indicates that a system of Ved nta
was known to varakrṣṇa.
7
I have argued for this in Ruzsa (2010: 428–430), and in more detail in Ruzsa (1997a). A possible
reconstruction of the core text will be found in my edition of the SK.
8
A particularly nasty example from SK 41: vin viśe air na ti ṭhati nir- śrayaṁ liṅgam. “The
transmigrating entity cannot stay with no substrate, without the elements”, i.e. without a gross material
body. The message is perhaps that even gods must have a gross body and there is no Bardo, antara-
bh va, intermediate existence between dying and rebirth. But if we read it as vinâviśe air (i.e. sandhi
of vin + aviśe air), the meaning will be “without the subtle elements”, which is but a scholastic
insistence on the precise build-up of the liṅga, the transmigrating entity.
9
Larson (1987: 18–19); Bronkhorst (1994: 315); Ruzsa (1997b), in more detail in Ruzsa (2014a:
97–114).
10
In many traditions, including many commentaries of the SK, the elements are called mah - or
s hūla-bhū as (great or gross elements), while the anm ras are referred to as sūk ma-bhū as, ‘subtle
elements’. These subtle elements are theoretical entities (although advanced yogis supposedly can see
Sāṁkhya: Dualism without substances F. Ruzsa 21

them) with no clear function or justification; they are subtle earth, subtle water etc. They make up the
subtle body, which we carry on while transmigrating, while the gross elements (the gross body) is left
behind.
This interpretation is alien to the SK and therefore it is seriously misleading to translate – in this
text – these entities as ‘gross and subtle elements’. Although no list of the anm ras is given, it is
perfectly clear that they are the sensibilia:
28 rūpâdi u pañc n m locana-m ram i ya e vr iḥ
34 buddhîndriy ṇi te m pañca, viśe âviśe a-vi ay ṇi
“The function of the five [senses] is merely observing colour etc. … Among them [i.e. among the
“powers”] there are the five sensesś their objects are the elements and the aviśe as [= anm ras].”
This makes sense only if the end of the list of the twenty-five entities, the last ten, i.e. the anm ras
and the elements, starts with ‘colour’, and further, the list should contain the objects for the other
senses as well. It follows that the anm ras refer to the sensibilia.
But varakrṣṇa, as discussed above in note 8, with one of his characteristic ambiguous phrases
makes room for the approach where the anm ras are constituents of the transmigrating entity (liṅga),
and are therefore best understood as subtle elements. In SK 40 the ambiguity remains: mahad- di
sūk ma-paryantam saṁsara i … liṅgam, “the liṅga [consisting of the entities] from the Great (i.e.
buddhi, understanding) down to the subtle [elements] transmigrates”. However, the sentence could
also mean only that “the liṅga transmigrates from the great down to the tiny”.
11
SK 28: vr iḥ / vacanâd na-viharaṇô sargânand ś ca pañc n m.
12
I am not aware of any philosophical tradition where manas would be a regular term for ‘mind’.
In Yoga citta is used, in Ny ya and Vai eṣika man (working together with manas) is the mind.
Ved nta seems mostly to follow S ṁkhya usage.
13
Saṁkalpaka, ‘who fits together’, ‘who joins’, ‘connector’ is often misunderstood (already by
the Indian tradition), based on the everyday meaning of saṁkalpa, ‘intention’, leading to an
interpretation of manas as volition.
14
This is the less frequent, but probably more original version of SK 27:
saṁkalpakam atra manaḥ, ac cêndriyam ubhaya h sam khy am.
antas tri-k la-vi ayaṁ; asm d ubhaya-prac raṁ tat.
15
Quite independent of the Indian tradition, Wilhelm Dilthey’s (1890) unique analysis seems to
grasp the problem in some of its depth. He substitutes for the usual model (the I, the subject, as already
given cognises the external world) a dynamic approach where the I arises in the process of frustrating
interactions with external reality; the frustration defines it as alien, as not-I, as object.
16
In the SK we have only ahaṁk ra, but the expression in SK 64 clearly shows that at least these
three aspects, but most probably also the three terms were known: evaṁ a vâbhy s n:“nâsmi, na me,
nâham” i y, “Practicing so with the entitiesŚ ‘I am not [this], [this] is not mine, [this] is not the I’.”
17
SK 23: adhyavas yo buddhir. dharmo, jñ naṁ, vir ga, aiśvaryam
s vikam etad-rūpaṁ. masam asm d viparyas am.
It is interesting to note that the four sattvic forms of buddhi are the four basic religious aims (as
clearly spelled out in SK 44 and 45): dharma for Vedism, knowledge for S ṁkhya, dispassion for
Buddhism and the other śramaṇa religions, sovereignty (or magic powers) of Yoga.
18
SK 37: sarvam pratyupabhogaṁ yasm puru asya s dhaya i buddhiḥ
Sāṁkhya: Dualism without substances F. Ruzsa 22

sâiva ca viśina ṭi punaḥ pradh na-puru ân araṁ sūk mam.

19
The last example is somewhat hypothetical, suggesting that egoity adds the colouring of our
personalities (saṁsk ras: past impressions, memories, inclinations, unconscious traumas, instincts) to
the information processed. This idea was inspired by Tibor Körtvélyesi’s (2011: 49–50) analysis of the
fourth Buddhist skandha, i.e. saṁsk ras.
20
This information-processing series is not my idea, it is varakrṣṇa’s. He uses the verb ‘to see’
(drś) in connection with both the senses (SK 4, 5, 30) and the soul (19, 21, 65, 66, 59, 61). He calls the
external instrument (which includes the senses) the sense object (vi aya) of the mind (the internal
instrument; SK 33). The whole idea is nicely expressed in SK 35–36. See also the quotation in note
42.
21
Ruzsa (forthcoming).
22
Ruzsa (2007).
23
Szegedi (2009: 42–52).
24
Hulin (2007: 59).
25
Franco (1991: 123).
26
Hulin (2007: 60).
27
Bronkhorst (1999: 683).
28
nanu liṅgaṁ ced ekaṁ, tarhi kathaṁ puru a-bhedena vilak aṇ bhog ḥ syuḥ? a râha: ‘vyakti-
bhedaḥ karma-viśe . 10.’ yady api sargâdau Hiraṇyagarbhôp dhi-rūpam ekam eva liṅgam, a hâpi
asya paśc d vyak i-bhedo, vyakti-rūpeṇâṁśa o n n vam api bhava i; ya hêd n m ekasya pi r-
liṅgadehasya n n vam aṁśa o bhava i pu ra-kanyâdi-liṅgadeha-rūpeṇa. a ra k raṇam ha: karma-
viśe d i i; j vân ar ṇ ṁ bhoga-hetu-karmâder i y ar haḥ. (p. 90)
Bronkhorst (1999: 688) also refers to this example.
29
SK 21, 22, 24 and 38. – The highly suspect verse 25 would even more increase the cosmogonic
feel of the te tŚ “The sattvic eleven proceeds from the Vaikrita (‘Modified’) egoityś the group of the
sensibilia from the Bhūt di (‘Beginning of the Elements’)ś both from the Taijasa (‘Brilliant’)”. – The
Sanskrit original:
21 puru asya … pradh nasya … saṁyogas: a -kr aḥ sargaḥ.
22 prakr er mah ṁs, tato 'haṁk ras, asm d gaṇaś ca oḍaśakaḥ;
asm d api oḍaśak pañcabhyaḥ pañca bhū ni.
24 abhim no 'haṁk ras. asm d dvi-vidhaḥ pravartate sargaḥ:
aindriya ek daśakas, nm raḥ (vv.ll. ek daśakaś ca gaṇas, anm raḥ) pañcakaś
câiva.
25 s vika ek daśakaḥ pravar a e Vaikr d ahaṁ-k r ;
Bhū âdes nm raḥ (v.l. anm raḥ), sa masas; Taijas d ubhayaṁ.
38 tan-m r ṇy a-viśe s; ebhyo bhū ni, pañca pañcabhyaḥ.
30
Not surprisingly, the regular counterpart of creation, i.e. an apocalypse or cosmic dissolution,
pralaya is not even mentioned. SK 69 is clearly a later addition; in any case, it does not speak about
the dissolution of the world, but of the beings or elements (bhū a), and it says only that it occurred in
the system of the highest r i (Kapila): sthity-utpatti-pralay ś cin yan e ya ra bhū n m, “where the
abiding, arising and dissolution of the beings are thought of.”
Sāṁkhya: Dualism without substances F. Ruzsa 23

31
Verses 21 and 24 have been quoted above, note 29.
46 e a pratyaya-sargo: viparyayâśak i-tu ṭi-siddhy- khyaḥ.
guṇa-vai amya-vimard asya ca bhed s u pañc śa .
52 na vin bh vair liṅgaṁ; na vin liṅgena bh va-nirvr iḥ.
liṅgâkhyo bh vâkhyas asm d bhava i dvidh (v.l. dvividhaḥ pravartate) sargaḥ.
53 a ṭa-vikalpo daivas, tairyagyonaś ca pañcadh bhava i,
m nu yaś câikavidhaḥ – sam sa o bhau ikaḥ sargaḥ.
54 ūrdhvaṁ sattva-viś las, amo-viś laś ca mūla aḥ sargaḥ,
madhye rajo-viś lo: Brahmâdi-stamba-paryantaḥ.
66 dr ṭ mayê y upek aka eko. dr ṭâham i y upara âny .
sati saṁyoge 'pi tayoḥ prayojanaṁ ṇâs i sargasya.
32
This shows, incidentally, that it is somewhat superficial to call understanding or mind (buddhi
and antaḥkaraṇa) material entities. Without the soul there is no mindŚ “buddhi and ahaṁk ra …are
mixtures of matter and consciousness.” “The buddhi, the first material product, has two causes, the
prakr i and the puru a principle, and is like a knot made of two ropes.” Jacobsen (1999: 225–226).
33
At death only the gross body made up of the five elements dies. The subtler, invisible parts of a
living being (the subtle body, sūk ma-śar ra or liṅga) move on to build a new gross body. So this
transmigrating entity is immortal; and it consists of material entities (tattvas of prakr i) – the 13-fold
instrument (i.e. understanding, egoity, manas, the 5 senses and the 5 active powers) and perhaps the 5
anm ras.
34
Schweizer (1993).
35
Schweizer (1993: 858, 849, 850).
36
Schweizer (1993: 852).
37
Searle (1980: 422). Searle, however, focuses on intentionality, not on consciousness.
38
Larson (1987: 77).
39
Larson (1987: 81).
40
Yoga-Sū ra 1.2–3: yogaś ci a-vr i-nirodhaḥ. ad dra ṭuḥ svarūpe 'vas h nam.
41
Eka-rūpe ca cai anye sarva-k lam avas hi e / n n -vidhârtha-bhok r vaṁ kathaṁ
n môpapadyate? (Tattva-saṁgraha 288.) The translation and the sanskrit text is taken from Watson
(2010: 92). This paper is an excellent comparison of four positions about the eternal, immaterial self:
Ny ya (self without essential consciousness), S ṁkhya (self as pure consciousness), Ved nta (self as
unchanging knowledge) and aiva Siddh nta (active self).
42
Candrakīrti, another great Buddhist scholar in the 7th century, still knows this more natural
idea, for he writes about the S ṁkhya theory of cognition: “When the sound etc. has been grasped by
the hearing etc. superintended by the manas, understanding makes a judgment. Then the soul becomes
conscious of the object as judged by the understanding.” (śabdâdi u śro râdi-vr ibhir
manasâdhi ṭhi bhiḥ parigrh e u buddhir adhyavas yaṁ karoti. tato buddhy-avasitam arthaṁ
puru aś cetayate. Lang, 2010: 56).
43
So it does not transmigrate (na saṁsarati, SK 62).
44
SK 18: janma-maraṇa-karaṇ n ṁ pra i-niyam d, a-yugapa pravr eś ca
puru a-bahutvaṁ siddhaṁ; tri-guṇâdi (v.l. trai-guṇya)-viparyay c câiva.
Sāṁkhya: Dualism without substances F. Ruzsa 24

45
Vi ayin and vivekin, SK 11; kevalin, madhya-stha, sâk i, and dra ṭr, SK 19.
46
SK 55: a ra jar -maraṇa-kr aṁ duḥkham prâpnoti cetanaḥ puru aḥ
47
jña, SK 2, and bhok r, SK 17, 37, 40.
48
SK 31: na kena ci k rya e karaṇam.
49
SK 20: guṇa-kar r ve 'pi a h kar êva bhava y ud s naḥ.
50
The historical process is more complex than an unidirectional movement towards a more
abstract conception of the soul, for Kundakunda, an important Jain author before varakrṣṇa, had a
conception much like him, while already attributing to the S ṁkhyas the idea of absolutely
unchanging soulŚ “If the soul does not change by states like anger, then transmigration will be
impossible – or the S ṁkhya position will be true.”
apariṇamaṁte hi sayaṁ j ve kohâdiehi bh vehiṁ
saṁs rassa abh vo pasajjade, saṁkha-samao v .
Bronkhorst (2010: 219).
51
SK 11Ś “The manifest is made up of the three qualities, continuous, object, common,
unconscious, essentially productive. The unmanifest also. Soul is its opposite and also similar [to some
other properties of the unmanifest].”
tri-guṇam, aviveki, vi ayaḥ, s m nyam, ace anaṁ, prasava-dharmi
vyaktaṁ; a h pradh naṁ. tad-vipar as, a h ca pum n.
52
These traditional names are better left untranslated. Tamas means darkness, rajas the
atmosphere, sattva existence, essence or living being.
53
The name of the theory, sat-k rya (‘e istent-effect’) is wrongly construed as “the pree istence
of the effect in the cause”ś it could rather be understood as “effect of e istent(s)”, meaning that there
must be a feature in the cause(s) explaining a feature of the effect. Ruzsa (2003a: 286–287).
54
SK 17: tri-guṇ ‘di-viparyay d … puru o 'sti, “There is the soul, for there is the opposite of
‘made up of the three qualities’ [SK 11] etc.”, i.e. for we find in the world features opposite to those
characterising matter (like ‘unconscious’). – The argument is not begging the question, for it is not an
arbitrary postulate that the guṇas lack consciousness: the guṇas are (abstractions and generalisations)
based on experience, and in non-sentient objects we do not find the slightest trace of consciousness.
55
SK 14 k raṇa-guṇâ maka v k ryasyâvyak am api siddham.
The way this inference works has been more fully treated in Ruzsa (2003a: 293–295).
56
Dasgupta (1922–1955: 241–245); Hiriyanna (1932: 271–272).
57
Hacker’s (1985: 112) position may appear representativeŚ “Diese Lehre hat europäischen
Forschern zunächst einmal einige Schwierigkeiten bereitet, weil wir sonst das Wort guṇa überall mit
„Eigenschaft“ übersetzen können, und hier scheint nach unseren Denkmöglichkeiten nur ein Wort wie
etwa „Konstituente“ zu passen … Frauwallner aber, der hier tiefer sah…, hat die Bedeutung
„Eigenschaften“ beibehalten und hat ganz deutlich ausgesprochen, daß eben die Eigenschaften hier als
Substanzen aufgefaßt werden. Das ist uns Fremd, aber wir müssen uns an diese Fremdheit gewöhnen,
sonst können wir eben indische Philosophie nicht studieren.”
58
Wezler (1985: 26).
Sāṁkhya: Dualism without substances F. Ruzsa 25

59
sattva-rajas-tam ṁsi guṇ ḥ, tat-pariṇ ma-rūp ś ca ad- mak eva śabdâdayaḥ pañca guṇ ḥ.
tat-saṅgh a-rūpaṁ ca ghaṭâdi, na u ad-vyatiriktam avayavi-dravyam as î i s ṁkhy n ṁ
siddhân aḥ. Wezler (1985: 10).
60
Bronkhorst (1994) and Bronkhorst (1997).
In the second paper he brings to our notice Vasubandhu’s testimony in his Abhidharma-Kośa-
Bh ya (III.50a). Although Bronkhorst himself believes that the passage presents “S ṁkhya in its
classical form, and not in its pre-classical shape, in which no unchanging substrate of properties had
yet been introduced”, he clearly overlooked the unequivocal statement of the S ṁkhya position
denying a substrate beyond the qualities: kaś câivam ha, dharmebhyo 'nyo dharmîti? “Who said that
the substrate is different from the properties”? Bronkhorst (1997: 394).
61
Bronkhorst (1994: 311–312). I think that it is the only possible reading of the SK (see note 10).
This understanding of the anm tras as the sensibilia lived on in the commentaries at the side of the
new, ‘subtle elements’ interpretation. See e.g. Bronkhorst’s main source for “classical” S ṁkhya, the
Yukti-D pik on SK 24cd: anm r ṇ ṁ śabda-sparśâd n m ayaṁ nm raḥ sargaḥ, “This nm ra
creation is of the anm ras, sound, touch etc.” Wezler, Albrecht & Shujun Motegi (eds.): Yuk id pik .
The Mos Signi ican Commen ary o he S ṁkhyak rik , Franz Steiner Verlag, Stuttgart 1998, p. 195.
Bronkhorst’s key argument for the substantiality of the anm ras is that they have more than one
quality. This is, however, not standard “classical” S ṁkhya; it appears only in the Yukti-D pik .
Besides the passage that Bronkhorst and Wezler analysed, it is mentioned cursorily in the introduction
to SK 22 (p. 187): eka-rūp ṇi anm r ṇîty anye; ekô ar ṇî i v r agaṇyaḥ. “Others say that the
anm ras have one form each; V rṣagaṇya says that each has one more [than the previous anm ra].”
So this is clearly the unique position of a single master, not accepted by others.
62
Watson (2006: 187–188). In footnote 196 he also mentions (but does not comment on) the
references to the substance reductionist position in later S ṁkhya that I sent him.
63
Yukti-D pik (YD) on SK 16c (pp. 163–164): ha: … na hi vo dharmebhyo 'nyo dharm ! …
ucya e: … ya h senâṅgebhyo 'nanya vam sen y ḥ. “Opponent: For you, there is no substrate other
than the properties! … AnswerŚ [True,] as an army is not different from its parts.”
YD on SK 23a (p. 189), a Buddhist objects: dharma-dharmiṇor ananya v d … vânya vam i i
dosaḥ. “… the properties and the substrate are not different. Or … they would be different, and that is
unacceptable [for you S ṁkhyas].”
YD on SK 15b (p. 144): na câikaiko rūpâd n ṁ dravyâk raḥ, samud ya-dharma v . “Colour
etc. singly do not have the form of a substance, for that is a property of their combination.” – A
note in ms. D adds: rūpâdi-pañcaka-vyatiriktaṁ vad dravyaṁ nâsti, “there is no substance beyond
the fiveŚ colour etc.”
YD on SK 34d (p. 218): sabda-sparśa-rasa-rūpa-gandha-samud ya-rūp mūr r, “physical
bodies are combinations of sound, touch, taste, colour and smell.”
64
It is so obvious that it is seldom mentioned; one example is Gauḍap da on SK 17 (p. 91): idaṁ
śar raṁ pañc n ṁ mah bhū n ṁ saṁgh o var ate, “this human body is a compound of the five
great elements.”
65
How e actly they are related is not specified by varakrṣṇa. The commentators’ preferred
version is that from sound arises ether, if we add touch, we have air; adding colour, fire; adding taste,
water; and earth has all the five sensibilia.
Sāṁkhya: Dualism without substances F. Ruzsa 26

66
In most commentaries the ‘subtle element’ interpretation of the anm ras appears only in the
discussion of their role in the liṅga, the transmigrating entity.
67
asm saṅgh a-m ra v sa vâd n ṁ ghaṭâdiva / brahmaṇaḥ parijñ ya deh n m
anavasthitim. YD 50cd (p. 248).
68
śabda-sparśa-rūpa-rasa-gandh ḥ pañca ray ṇ ṁ sukha-duḥkha-moh n ṁ saṁniveśa-viśe ḥ,
“The five (sound, touch, colour, taste and smell) are particular arrangements of the three, happiness,
pain and bewilderment.” Steinkellner (1999: 670, fr. 4).
69
Actually dharmin and svabh va do occur in SK 11 and 55, but only in the sense of ‘essentially’.
śraya, ‘substrate’ or ‘support’ is freely used, but it is not necessarily a substanceŚ in SK 12 we are
told that the qualities are the substrates for each other (anyo'nyâbhibhavâśraya-janana-mi huna-
vr ayaś ca guṇ ḥ).
70
SK 29 sv lak aṇyaṁ vr is rayasya; sâi bhava y as m ny .
s m nya-karaṇa-vr iḥ pr ṇâdy v yavaḥ pañca.
71
It is not suggested that S ṁkhya has any particular connection with the Dravidian speaking
South (although many important philosophers writing in Sanskrit, e.g. aṅkara and R m nuja, belong
there). It seems that Dravidian languages were spoken in Northern India before the arrival of the Indo-
Aryans, and the continuous substrate influence of Dravidian is apparent in the development of all
Indo-Aryan languages (including Sanskrit) for much more than a millennium. So there is nothing
inherently improbable in the supposition that some thinkers, known to us only through Sanskrit te ts,
could speak and think in a Dravidian language. – I have argued for the influence of Dravidian on
Aryan languages starting already at the time of the composition of the Rgvedic hymns and continuing
at least to the period of late Prakrits (ca. 500 CE) in Ruzsa (2013).
72
That the Buddha himself already had a strong anti-substantialist attitude is clearly shown
beyond the well-known an- man, ‘no-self’ doctrine by his philosophical term for the human bodyŚ
rūpa, ‘form’ – that is, a quality, not a substance.
73
Tandon (1995).
Ruzsa (2003b); Ruzsa (2014a: 97–114), i.e. ch. IXŚ “Pain and its cure. The aim of philosophy in
74

S ṁkhya”.
75
SK 54 mentions Brahm as the highest being within creation. The Yoga-Sū ra and the
commentators of the SK speak about vara, the Lord, i.e. God, although he is not the creator. See
Bronkhorst (1983). Brahm Sahaṁ-pati, Sakka the king of the gods, and M ra the evil appear
frequently in stories about the Buddha.
76
Here, however, the Buddhist approach is more passive. The S ṁkhya concept includes the
powers of action as well, while vedan is only sensation.
77
As Marzenna Jakubczak (2012: 42) so clearly emphasized in her nicely perceptive paper, in
S ṁkhya “the aim is not to identify directly with puru a, but rather to keep disidentifying with the
present phenomenal self by means of constant realisationŚ ‘I am not, not mine, not I’ (n smi na me
n ham; cf. SK 64; M[ajjhima]N[ik ya] 109.15–16).”
78
evaṁ a vâbhy s n: “nâsmi, na me, nâham” i y (SK 64). – rūpaṁ (etc.) “n' etaṁ mama, n' eso
'ham asmi, na m' eso a ” i samanupassa i (Alagaddûpama-Sutta, Majjhima-Nik ya 22).
Sāṁkhya: Dualism without substances F. Ruzsa 27

79
Saṁkhittena, pañc' up d na-kkhandh dukkh . “Summarily, the five skandhas of appropriation
are painful.” Dhamma-Cakka-Ppavattana-Sutta, Saṁyutta-Nik ya 5.12.2.1 (= 56.11) = Vinaya-Piṭaka,
Mah -Vagga 1.6.
80
This is A vaghoṣa’s Buddha-carita, written half a millennium after the Buddha’s time.
81
Mah -Saccaka-Sutta, Majjhima-Nik ya 36.
82
Motegi (1999).
83
Mah -Bh ra a 12.212.14–15:
14 imaṁ guṇa-sam h ram ma-bh vena paśya aḥ
asamyag-darśanair duḥkham anantaṁ nôpaś mya i.
15 an mêti ca yad dr ṭaṁ, te nâhaṁ na mamêty api:
vartate kim-adhi ṭh n prasak duḥkha-saṁtatiḥ?
84
Bronkhorst (2007: 226–231). – On the curious inversion of motifs between the two great sages I
have written in Ruzsa (2009).
85
As it was perfectly clear in antiquity: the first important discussion in the Brahma-Sū ra (1.1.5–
11) tries to prove that this interpretation is false – thereby showing that it was widespread to quote the
Sad-Vidy , i.e. ruṇi’s teaching to his son vetaketu in the 6th chapter of the Ch ndogya Upani ad, as
a scriptural authority for S ṁkhya.
86
See Ruzsa (2004). It is interesting to note that ruṇi teaches sat-k rya (‘the effect of e istent’)
in the most literal sense, since his fundamental entity is not called prakr i but sat, ‘e istent’.
87
Ch ndogya Upani ad 6.8–10; Brhad- raṇyaka Upani ad 2.4 and 4.5
88
Yam p' idaṁ diṭṭhi-ṭṭh naṁ ‘so loko so a , so pecca bhaviss mi: nicco dhuvo sassa o a-
vipariṇ ma-dhammo, sassati-samaṁ tath'eva ṭhass mî’ i am pi ‘e aṁ mama, eso 'ham asmi, eso me
a ’ i samanupassa i. From The simile of the cobra (Alagaddûpama-Sutta, Majjhima-Nik ya 22).