Abhava : negation in logic, real non-existent, and a distinctive pramana in

the Mimamsa
AUTHOR(S)

Purushottama Bilimoria

PUBLICATION DATE

01-01-2008

HANDLE

10536/DRO/DU:30017023

Downloaded from Deakin University’s Figshare repository

Deakin University CRICOS Provider Code: 00113B

 Abhāva:
Negation in logic, real non-existent, and a
distinctive pramāṅa in the Mı̄mām. sā
Purushottama Bilimoria∗
School of International and Political Studies, Deakin University, Geelong Waterfront
Campus, Geelong, Victoria, 3127, Australia
School of Philosophy, Anthropology and Social Inquiry, The University of Melbourne,
Melbourne, Victoria, 3010, Australia
E-mail: [email protected]

The focus of this essay is on the Mı̄mām . sās’ radical epistemological the-
ory of negation, otherwise known as abhāva or “absence”, and which may
be read as nāsti or non-existent. Before we go any further, a word on
“Mı̄mām. sā” is appropriate, particularly for those who may not be familiar
with this particular school of classical Indian thought —perhaps in some
ways even pre-classical with its roots in the Brāhman.as of the Vedas— and
to which also arguably belongs the genesis of the Nyāya school of thought.
The term “mı̄mām . sā” literally signifies “commeasurement” — from the
root “mān”, “to think” and (with the additional gan.-prefix mı̄) — “to
interpret and align”, and this implies, in Zilberman’s words,
“achievement of consistency in reasoning [naya, later nyāya], as a
necessary precondition of making the meanings of words or sentences
(for Sanskrit it is all the same) comprehensible [19]”.

According to the definition given by Śabara in his bhās.ya on Jaimini-

sūtra I.1.2,
“... interconnected words or sentences which instruct in methods of
cooperation or congruency in actions are known as pūrva-mı̄mām
. sā,
or karma-mı̄mām . sā [19]”.

It is in this semantic field where internal consistency is aligned with the

normative pragmatic of “what is to be done” (actual or potential actions) —
and, counterfactually, what is not to be done (i.e., prohibition)— that
the doctrine of negation first makes its appearance in Indian philosophical
thought. With Kumārila Bhat.t.a, c. 7th century CE, the doctrine achieves
a decisive logical formulation in the dialectical tussles with the Nyāya and
∗ I should like to extends my thanks to J. L. Shaw, K. T. Pandurangi (my scholastic

guru), Frits Staal, Dhirendra Sharma (in absentia), and two anonymous referees for
their inputs into this paper, which I wish to dedicate to the memory of the ever absently-
present, Bimal Krishna Matilal.

Mihir K. Chakraborty, Benedikt Löwe, Madhabendra Nath Mitra, Sundar Sarukkai (eds.).
Logic, Navya-Nyāya & Applications. Homage to Bimal Krishna Matilal. College Publications,
London, 2008. Studies in Logic 15. pp. 43–64.
Received by the editors: 7 May 2007; 30 May 2007.
Accepted for publication: 27 August 2008.
44 P. Bilimoria

Buddhist philosophers. In the background are the rule-based formulations

where varieties of negations and self-defeating paradigms within language-
uses are schematically dealt with by the —even more vintage— Grammar-
ians, Pān.ini and Patañjali. (By “logical”, then, one means not so much
formal reasoning as in mathematical logic, but rather ratiocination encom-
passing linguistics, epistemology and ontology — the Vedanta added “meta-
physics” to this developmental ratio.)
I have three aims. First I should like to present an outline of the logical
theory of negation in the Mı̄mām . sā. Second, I shall make a connection of
this with the Mı̄mām . sā hermeneutic of moral judgments (which for them are
inscribed in śruti, the Vedas, in the form of Vedic injunctions, “the ought to
do” type of positive propositions or vidhis. And finally, we make a further
link of these two moves to its epistemological radicalism in standing up for
an independent and distinctive pramāṅa, valid instrument of knowing, in
respect of “absence” — and that too, on pain of being true to a realist
ontology. The most important part of the discussion here is the Mı̄mām . sā
treatment of negative propositions.

1
As mentioned above, the logical thesis begins with deliberations on rules
followed in expressions pertaining to certain actions. The concern is not
at all with indicative or purely descriptive expressions (for these are re-
garded as arthavāda, of auxiliary aid to the deliberations). To elaborate
on this, I shall use Zilberman’s succinct descriptors for action-statements.
Actions are divided subjunctively into two classes: prescriptive and pro-
hibitive. Normative injunctions (vidhi s), derived from the Vedic sentences,
which stimulate or are an accomplice to actions can take one of these forms
(1) regular, imperative or non-exemptive (nitya), i.e., categorically non-
conditional; (2) imposed by circumstances (naimittika), hence conditional
to some greater good; (3) optative, prudentially conditional, i.e., conducive
to some desirable award or potential effect (kāmya), and is non-prescriptive.
These together constitute dharma, the normative performatives: this is the
Bhāt.t.a a (followers of Kumārila) view; the Prābhākara extend the kāryatā,
practicum, of dharma only to the first two.1 The inducement or conducive-
ness toward the requisite effort is communicatively generated, after atten-
tion has been roused, in the form of a verbal energy in the so-compelled,
excited or inspired initiate by the vidhi -proposition, generally of the form:
“Let him be induced to do this and that”, and it abides in the subjective
1 anyākāryatā, anyā ces..tasādhanatā ... phalam . prati upayātam . phalasādha-
natvam . , kr.tim. prati pradhānatvam . tadadhı̄nasattākatvam ca kāryatva; ... loke
kriyākāryatājñānātpravr.ttāvapi kāryatājñānameva pravr.ttinimittah.. [11, p. 57]. Cf. also
[9, p. 63] where this passage is also cited.
Abhāva 45

intention (bhāvana) much like the subjunctive force of modality. It is ren-

dered by the imperative or optative form of the verb (liṅ) and it conceals
the deontic modality of the statement which proposes something as hitherto
not existent (non-existent), but as due or awaiting to be done (kāryatā): the
potential or portending action [19, p. 289].
Now, and this is a crux of the argument, the prior non-existent (or the
facticity of non-existence) is the logical negation of what has not yet been
actualised. The relation between the non-existent (or non-existence of the
action/thing), abhāva, and its counterpositive (pratiyogı̄) is that of incom-
mensurability. In its simplest form this could be expressed as ¬(p ∧ ¬p).
All subsequent schools of Indian thought that recognised the ideality of
non-existence accepted this formulation of negation. But there are other
possible kinds of negations as well; and even more so, the Mı̄mām . sā was
pressed to argue, in the case of subjunctive propositions, i.e., of the pre-
scriptive, vidhi, type. The problem then that faces the philosopher is to
interpret all parts of the vidhis and the larger passages in which they are
embedded, and to sort through the kinds and degrees of permitted (i.e., in-
cluded), proscribed or prohibited, delimited and bounded qualifications (i.e.,
inclusive-exclusive), as well as the wholly excluded conditionals governing a
particular action (such as, for example, the special sacrifice on the night of
equinox). The way this is done is by focusing primarily on the verbal dis-
playment that abides in the impersonal Vedic words (being authorless, there
is no question of inquiring into the intentions of the supposed author[4]);
and that focus narrows down to the verbal intentionality (śābdı̄bhāvana)
that presents a specific transcendental function as the specific property of
the intension of the verbal meaning.2 It follows that among the injunctions
there may well be resemblances, dissimilarities, or oppositions and defeats;
and so generating an order among these becomes of paramount importance.
In the analysis that ensues, two categories become important: affirma-
tive and negative meanings attached to the verbal expressions. Balancing
the effects, the consequent, of these would mean first examining their an-
tecedent casual potency (śakti ) — how much good or damage they can do;
this calls for dialectical mediation, and it may mean a recourse to non-formal
logic (deontic, subjunctive, subjective, aesthetic, causal, etc.).
It may be noted that, in fact, the Mı̄mām . sā were the first to intro-
duce negative kinds of propositions in Indian philosophy. Jaimini-sūtras at
VI.5.15, VIII.7,10, specifically discusses pratis.edha and nis.edha, and it may
be noted that this is well before the Nyāya-Vaiśes.ika and the Jaina and
2 Here Zilberman cites Man.d.ana Miśra as his authority for this rendering of “śabda
bhāvana”, as he calls it, which otherwise would be śābdı̄bhāvana, as “verbal energy” [19,
p. 289]. Arindam Chakraborty disputes this is a correct rendering of śābdı̄bhāvana, for
the bhāvana presumably is not in the verbal formation but rather is a disposition within
the hearer which propels him forthwith into action.
46 P. Bilimoria

Buddhists took up these concepts and developed them in rather different

and indeed logically rigorous ways. Although it is the case also, as men-
tioned earlier, that the Grammarian Pānin.i had already laid out some of
the terms and rules for interpreting opposition and tensions between con-
trary expressions or linguistics predicates. The contradiction is recognised
to be of the form (ϕ ∧ ¬ϕ). But the grammarian’s rules did not embed the
same degree of logical articulation, particularly in respect of non-indicative
or non-doxastic sentences, which the Mı̄mām . sā would herald in. Again,
the Mı̄mām . sā needed this in order to sort out the different kinds of nega-
tive injunctions, which —as already noted— Jaimini set out to do, so that
all kinds of pratibādhya-pratibhandakas or handicaps, contradictions and
incommensurabilities (classed as vipratis.edha, “mutual prohibitions” — in
and between the injunctions of equal force, mantras, and their supplemen-
tations, i.e., arthavādas, and the moot distinctions between these as well)
could be put in place. So here is an application of early Indian thinking
on negative propositions and how one is to “read” these in respect of their
ramifications for the intended purport of the Vedic codanās or vidhis. For
those with an interest in finding other ways to talk about contradictions or
self-defeating relations —a divergence from Aristotelian logic— this earliest
of Indian classical treatment may well be instructive.
In what follows I shall adopt the notation that Staal used when first
discussing the matter [16]. The use of the term “negation” has largely been
ignored by modern logicians in favour of sentential negation. The intended
reading of ¬F (x), as opposed to ¬[F (x)], is “x is non-F ”. and is famil-
iar from the discussion of Aristotelian syllogistic logic. The negation of
singular terms, as as F (¬x), however, will strike modern logicians as some-
thing unheard of. Its intended reading is “not that x is F ”. The point
of introducing these grades of involvement is not just to stay closer to the
grammatical form of the original statements (something modern logic cares
preciously little about), but to keep track of the different inferential roles the
Mı̄mām. sakas attribute to them. All of ¬[F (x)], ¬F (x), and F (¬x) mean
that x is not F , but ¬F (x) additionally carries the information that x is
something other than x is F . Granted, these concepts can be formalized us-
ing sentential negations (given identity and second order quantification are
expressible in logic). However, the exact meaning of the different negations
is what is discussed by the Indian philosophers. For example, Staal gives
a passage that states that F (¬x) is to be interpreted as “only x doesn’t F
[16, p. 64]”, prompting a quite different translation into modern logic.
There are four-fold division of negation, as follows (which in much mod-
ified form also came to be accepted by the Nyāya-Vaiśes.ika, although they
differed on the means of apprehending the stated negations):3
3 SV, Abhāva 2–4, p. 336: ks.ı̄re dadhyādi yannāsti prāgabhāvah.sa ucyate//2// nāstitā
Abhāva 47

1. prāgabhāva: antecedent non-existent or prior negation; the “not-yet

so” or absence of an effect (e.g., the negation of curd in milk, by its
theory of asatkārya: an effect is not prior to its cause; the pot’s non-
existence (ghat.ābhāva) before the clay is thrown and its counterpos-
itive, ghat.a, is manifested); this absence is causelessly beginningless
but ceases once the effect is produced in temporal space.
2. dhvam . sābhāva: posterior non-existent or deconstructive negation, the
“no-more” type of absence due to destruction of some being in time:
(e.g., the negation of milk in the curd; the destruction of the pot once
it falls to the ground and is shattered into pieces; or there is no book
now on the table; when the Faculty of Arts will all but be destroyed,
the absence of the Faculty of Arts will be in or haunt the University;
when my savings dwindle to zero; and Saddam Hussein is hanged).

3. anyonyābhāva: mutual and relational negation; the “not-related” type

of absence (x is not in y, e.g., the negation of the horse in the cow,
due to the absence of dew-lap; and vice versa; Rāma is not Laxmana
(being more honest of the two); Sı̄tā is not Rāvan.a’s wife, as this
rūpı̄yā belongs to Rāma; Philosophy is not Cinema Studies; my dogs
are not related to the cat in the same house etc.)
4. atyantābhāva: absolute negation; the “never-never” or impossible type
payaso dadhni pradhvam . sābhāva is.yate/ gavi yo’śvādy-abhāvas tu so’nyonyābhāva
ucyate//3// śiraso’vayavā nimnā vr.ddhikāt.hinyavarjitāh./ śaśaśr.n
. ga.dirūpen
.a
so’tyantābhāva ucyate//4//
In [2, p. 243]; also discussed in [12, pp. 27sqq]; compare the Naiyāyika Jagadı̄śa:
“prāgabhāva-dhvam . sayorapi uttarapūrvakālāveva” pointed out by J. L. Shaw in [15, p.
153]. Almost identical classification of these four kinds also appear in Jaina texts, notably
Āptamı̄mām. sā (Kumārila seemed to have been aware of this text), reported in [18, p.
24–25] — interestingly this very thesis on the possibility of the non-existent is built into
an example of syādvāda in its saptabhaginaya, seven-step reasoning, thus:
1. A thing is existent — from a certain point of view.
2. It is non-existent — from another point of view.
3. It is both existent and non-existent in turn — from a third point of view.
4. It is indescribable (that is, both existent and non-existent simultaneously – from
a fourth point of view).
5. It is existent and indescribable — from a fifth point of view
6. It is non-existent and indescribable — from a sixth point of view.
7. It is both existent and non-existent and indescribable — from a seventh point of
view.
[18, p. 25]. “sytā” means conditional “yes’, or more philosophically, it means the same
as “kathamcit”, “in some respect”, or “from a certain point-of-view” and “kadācit”,
“somehow” or “sometime”; thus, a proposition, P , could be true or false; if P , then A
is B.
48 P. Bilimoria

of negation in all possible worlds (e.g., absence of hardness in the

lower portions of the hare’s head, or its likelihood of ever growing
horns; Lucy in the sky with diamonds). If absolute nothingness mili-
tates against all possible-world emergences, and there is no temporal
space for being or things to materialise, ever, then it would be a case of
nityābhāva, Absence of Presence, eternality of Nothingness, Void. This
seems to be one sense in which the Mı̄mām . sā, especially Kumārila,
understood the Buddhist doctrine of śānyatā, “non-dependent non-
origination”, the truth of which he would be persuaded by except that
he couldn’t condescend to the absence of the timeless śruti proposi-
tions; even if # 2 above comes to pass, or there is global pralaya
(Armageddon), mathematical logic has to live on somewhere.
From # 3, following the Grammarian commentator Patañjali, one is able
to derive pratis.edha, “mutual prohibition”, and vipratis.edha where “two
rules with different meaning apply to one (word)”; which could be read as
“opposition (between two propositions) of equal force”.4 The question is
how to resolve the tension. The Mı̄mām . sá for their part begin by grouping
the negations into three categories, which I shall call limiting negations.
These are now described: vipratis.edha, pratis.edha or nis.edha, paryudāsa:
permissible, prohibited, and excluded, respectively.
The first kind is a contingent opposition, applying to individual instances
but is not considered to be universal (for there is no jāti, genus, that per-
vades across the two expressions). When a jar is destroyed, not all jars
are destroyed. You may be seen with the umbrella or long-brim hat this
morning even though it is not raining, for it may rain in the afternoon
(as it always does in Melbourne). Nevertheless, it has more of a force in
prescriptive sentences than in indicative or nominal constructions. Strict
prohibitions belong to the pratis.edha in its nis.edha form.
Negation proper admits of two types, prasajya (mutual prohibition) and
paryudāsa (exclusion). Let us take prasajya first: “snow is not black” or
“there is not-any snow that is black”, and it will take the form: ¬F (x). On
the other hand, Paryudāsa, in its simple form, is exclusion: “the jar is not
(here)”, “it has not snowed here”, F (¬x); A little more complex form may
involve negative implication, “p implies ¬q”, example of this would be: “if
Rāma is a ks.atriya, he is not a brahmin”; “if it is grey, it is not white”, “if
y is an asura, he is not a god”. This is as far as the Grammarians went
and they did not bother with what Staal calls the injunctive operator (the
obligation conditional), N ; where negation is attached either to F (what is)
or to N (what is to be done). The latter is compounded and is often a cause
for ambiguity when the negation is attached to or pervades the obligatory
4 Cf. [17, pp. 51–71].
Abhāva 49

predicate; thus: not-ought, ought-not, ought not to, you-not ought, you not-
ought, etc., etc. So now we shall see how these formulations are transformed
and tightened up when the injunctive operator, N , is introduced by the
Mı̄mām. sā.
Let us take Prasajya-pratis.edha (which is also called nis.edha) first and
take the injunction “na bhaks.ayet”, “he shall not eat”. It is a negation
of the positive injunction “bhaks.ayet”, “he shall eat” that is denoted by
N [F (x)], where F (x) denotes “he eats” (and modally, “it is necessary that
he eats”). Its injunctive force is not in prescribing an action other than
eating; rather it simply prohibits eating and so should be rendered as “he
shall-not eat” (colloquially, eating is a “no-no”). If “he shall eat” is sym-
bolised by “N [F (x)]”, its logical negation is symbolised by “N ¬[F (x)]” .
In this context, as J. L. Shaw notes, the negative injunction has not been
symbolised by “¬N [F (x)]” for “in standard deontic logic ‘¬N [F (x)]’ would
be equivalent to ‘P ¬[F (x)]’, where ‘P ’ stands for the permissibility opera-
tor. So ‘¬N [F (x)]’ would not express a negative injunction [14].” In other
words, the expression “¬N [F (x)]”, even though it appears as a negation of
the positive injunction (takes on, as it were, the whole sentence), it eo ipso
does not have the force of an injunction, and could mean “it is permissible
that he does not eat”. In that regard, it is more consistent with the sec-
ond, the paryudāsa, or “exclusionary” rather than the strictly prohibitive
type of negation, which must strike at a very specific part of the injunction
(represented by the verbal ending only, ākhyāta/lakara-pratyaya). And this
takes us to the second type of negation.
The second type is paryudāsa, or “exclusion” negation. Here the nega-
tion is connected with either the verbal root or with the noun (the nominal
indexical or the predicate object); thus in the injunction, neks.eta 5 , “he shall
not look”, the “not” is attached to the verbal root, so it should be rendered
as “he shall not-look”; more positively it prescribes something other than
looking (looking away, for instance). Curiously, there is positive injunction
here, because there is a preceding phrase “his vows are...” Hence, techni-
cally speaking, nothing is prohibited; there is no iks.ana-virodhi (opposition
to looking) for he never thought of looking (in the direction of Dharmakı̄rti’s
distractingly dancing mistress)! He has not been given a desirous option:
the expression is bereft of an optative ending in the negative, which is dif-
ferent from saying you can be enjoined to do W (kāryatā), even where W
cashes out into ¬F (x) = “not-look”. You are still doing your work. But
the negative could as well strike at the nominal indexical or the predicate
object: “he-shall not look”; “he shall not look-at-the-mistress”; thus
someone is excluded or something is occluded from the gaze.6 Symbolised
5 Staal takes this from the Mı̄mām
. sā-nyāya-prākāśa in [6].
6 Whereas Staal takes it to be the opposite and swaps the two. Staal gives the sym-
50 P. Bilimoria

respectively (per bold qualifiers) by N [¬F (x)], and N [F (¬x)], N [F (x, ¬y)].
We could go on and find sentences where one of the nouns is negated
(subject, qualia, or object), drawing on #3, the anyonyābhāva negation.
Thus it could take any one of these forms: N ¬[F (x)], N [¬F (x)], N [F (¬x)].
If the positive injunction of the expression F (x, y) is N [F (x, y)], [y is contra
x, so other than x], its negation or paryudāsa is N [F (¬x, y)], but it still
remains positive in its injunctive force (kāryatā).
The difference between the two is that in paryudāsa, there is a residue
of the essentially positive still lingering on, while the negative is secondary
or of second order, a semantic negation that strikes at the last member, not
necessarily the verb itself. Whereas in the former, prasajya, the essential
assertion is of a negation and there need not be any positive residue; it
does not apply to the last member of the negative compound but strikes at
the verbal ending (krı̄yaya saha yatra nañ) (nañ signifies the negation). So
the “paryudāsa” type of negation may be called “exclusion type of nega-
tion”; whereas prasajya-pratis.edha is a “prohibition type” of negation. And
this latter is also called “nis.edha”, which can spell out more radical forms
of negation the harder the negative strikes at verbal intension via the N -
operator.
In other words, one form of negation may suggest that doing x is not all
right at this moment, but it may be permissible at another moment, or by
someone else. “He shall not-eat”, inscribes into its propositional structure
the permissibility of eating at other times.7 The Mı̄mām . sās were interested
in delineating the kinds of negation that are prohibitory without a residue
of permissibility in any part of the semantic field, for one can easily say,
“Do not indulge in sex”, but if P got married tomorrow it may become
permissible. Likewise, “a Śūdra should not even as much as be permitted to
hear the Vedas recited”, but it does not necessarily exclude the remaining
castes from being present at a sacrifice, and so on. These belong to the
exclusionary type. While the Mı̄mām . sās are in search of the equivalence of
the strictly obligatory of the negative kind.
Clearly the kind of negation in the foregoing examples, if it be ad-
mitted as a valid kind of negation (which I presume it would not be in
Aristotelian logic, unless one brings in some other operators, like Church’s
bolic form of nis.edha as “(¬N )[F (x)], but which does not in deontic modal logic express
a negative injunction, so while it admits of permissibility it may or may not reign in pro-
hibition, which prasajya-pratis.edha or nis.edha must do. Staal’s candidate “(¬N )[F (x)]”
would denote there is no injunction or mandate for him to eat — and Staal says as much
when he renders “(¬N )[F (x)]” as “there is no mandate for eating”; whereas there is
a clear injunction prohibiting any eating: “shall-not”; the negation must strike at the
verbal ending not just the N operator. It is not a simple withholding of the taxes but
forfeiting the taxes to the taxman.
7 E.g., crossing communities, it would apply to the Muslim observance of Ramzaan,

where eating in the day hours is prohibited, but permitted after sunset.
Abhāva 51

sūtra (event): F (x) the door is locked (i)

paryudāsa : F (¬x) you may unlock not this, the other door
(ii) (not well formed negation; not gov-
erned by principle of noncontradiction,
as in Quine also)
prasajya-pratis.edha: ¬F (x) you may not unlock this door (iii)
(negation of the predicate, as in Aris-
totelean logic; governed by principle of
noncontradiction)

Figure 1. Grammarian

λ, I suppose), the Mı̄mām . sā would classify under paryudāsa, “exclusion”.

Yatra-uttara-padena nañ, “where the negative is connected with the next
word” — denoting here, “other than the verbal ending”, krı̄yaya saha yatra
nañ. And the “next word”, uttarapada, as suggested earlier, denotes the
second member of a negative compound (tatpurus.a or bahuvrihi, different
kinds of nominal compounds in Sanskrit, but not ruling out verbs) can be
either a verbal root or a noun, but such a negation does not strike at the
core of the injunction (because it excludes the verbal ending), hence it is
not properly a negative injunction. In other words, it has a qualificative
limiting function rather than an absolute prohibitory function, and there-
fore it is not consistent with the fourth category of negation either, i.e.,
atyantābhāva.
Bringing this part of the discussion to a close, it should be said that the
innovative element in the Mı̄mām . sā approach to prohibition and exclusion
vis-à-vis the grammarians in respect of the three types of negation, is that
negation can be applied to either N or F or all parts of a sentence, which
is impossible in grammar. This may not seem very novel from our modern,
post-classical point of view in logic, but at a time when logic was totally in
the control of a tool used principally by grammarians to structure the deter-
minants of proper speech, this indeed is quite a remarkable break-through.
For it attempts to mirror the world outside, while playing with modal pos-
sibilities and instantiations — of real krı̄yas and kāryas, actions and things,
being and events, in speech-forms; and this onto-logic further tries to un-
derstand, without compromise, the meaning of certain a priori negations
in the Vedic corpus, which the Mı̄mām . sā took to be unquestionably valid
(because of its apaurus.eyatva, freedom from personal errors, including that
of a possible deity).
52 P. Bilimoria

(event) F (x) the door is locked (iv)

vidhi (positive in- N [F (x)] the door ought to be locked (v)

junction):
paryudāsa I (exclu- N [¬F (x)] the door need not-to-be-locked (vi)
sion):
paryudāsa II (inclu- N [F (¬x)] not this, another door needs to-be-
sive-exclusion): locked (vii)
nis.edha (strict prohi- N ¬[F (x)] the door shall-not be locked (viii)
bition):
The anomaly or ¬N [F (x)] you told me not to lock the door
“wild card” here is: (ix) (ambiguous)

Figure 2. Mı̄mām
. sā

So we finally give the difference in formal expression in Figures 1 and 2.8

J. L. Shaw points out that a sharper distinction needs to be drawn between
vi/vii and viii, making the former permissible : if we substitute F for eating,
then vi/vii renders it permissible that x does not eat (he might not be
observing a dietary vow); however, the latter, viii, is an injunction against
eating: ought-not to eat. But I believe, from what has been demonstrated
above, that the Mı̄mām . sa actually achieve this quite successfully.
The Naiyāyikas, from what I understand to be their thinking, reduce
negation to two —at most three— main kinds of negation. From the list
given earlier by Kumārila Bhat.t.a, they accept the “not-yet” (prāgabhāva),
9
“no-more” (dhvam . sa), and — if J. L. Shaw is right — “never” (atyantā-
bhāva). But each of these is based on the fundamental recognition of the
self-defeating relation of a thing and its simultaneous absence (pratibādhya-
pratibandhaka bhāva), which is represented by (p ∧ ¬p); but the pot could
8 I am grateful to Staal — when I was very confused by the Mı̄māmsā formulations, I
.
went to Staal back in 1981, and we discussed some Mı̄mām . sā texts together in Berkeley
many moons ago. But I have also modified his formulations with the corrective provided
by J. L. Shaw (cf. Footnote 6).
9 In his [15, pp. 144–145], he classes them under “relational absences”, with certain

caveats built into the “temporal relation as the limiting relation of the property of being
the counterpositive”. Although in his other papers on negation, Shaw limits Nyāya nega-
tion to two main kinds: relational absence and mutual absence, represented respectively
by (1): x is not in y, or x does not occur in y, or the absence of x occurring in y; and (2):
x is not y, or x is different from y; where “x” and “y” are non-empty terms, and their
counterpositive are: (1’) x is in y, or x occurs in y, and (2’) x is y. I am indeed grateful
to J. L. Shaw for sharing his papers on negation with me, and I have drawn liberally
with his permission and kind guidance for the present essay. In particular, his [14] (cf.
Footnote 4), [13], and [15, p. 144].
Abhāva 53

also lack blueness, or there is blueness but no potness, or neither; thus

(p ∧ ¬b), (¬p ∧ b), (¬p ∧ ¬b), which is really to say, ¬(p ∧ b) ↔ ¬p ∨ ¬b.10 It
was also recognised that ¬p rejects p in a way different from that in which p
rejects ¬p. Or take the simple identity A is A; so A is ¬A, and ¬A is A are
self-defeating but in quite, even radically, different senses, precisely because
the prioricity of absence is marked in a different way in each of the copulas,
and one excludes much more than the other.11 (God is His own Absence, or
Nothingness; Eternal Absence or Nothingness is analytically God. The first
absence —exclusion— is part of the meaning of God, as the base or ādhāra;
while Absence in the second conjunct cannot be the base, metaphysically,
of God — although a Buddhist might think so!). I leave it to the scholars
of Nyāya to elaborate on the Nyāya theory of negation and its difference
from the Mı̄mām . saka’s theory that I have sketched above, and also to show
what the Nyāya owes to the Mı̄mām . saka in developing its particular view
in all its symbolic and logical sophistication, particularly when it comes to
double negation and contradictions.
Having as it were worked around with the law of non-contradiction with-
out giving it up (but limiting its pragmatic scope), Indian logicians appear
to have no hesitation in rejecting the law of excluded middle or “bivalence”;
so it turns out on this view to be quite reasonable to say that it is neither
true nor false that man will land on Mars by 2020. But to the question of
whether man is capable of landing on Mars, the bivalent answer may take
this form:
Either it is not in man’s power to land on Mars or (given the current
state of space-technology) it is not in his power not to land on Mars. It
is neither true that man will by 2020 land on Mars (because they never
have, and the future owes nothing to the present, as Bradley reminded us),
nor is it false that men will land on Mars (given other conditionals). If we
replaced “nor” with “and”, it will be self-defeating and therefore has to be
rejected: ¬(M ∧ ¬M ).
But what about ¬(M ∧ ¬M ) ∨ ¬(M ∨ ¬M ), given that paryudāsa may
permit negation of the whole statement, bivalence included? This is where
the Buddhists come into the picture. And we may only touch on this inter-
vention before moving onto Section 2 of the paper.
Firstly, the Buddhists also recognise the distinction between paryudāsa
and prasajya-pratis.edha, but they characterise the distinction in terms of the
10 Cf.[8].
11 A point made by K. C. Bhattacharyya in his celebrated essay [3, p. 576]; cf. also, [3,
pp. 599–601]. Bhattacharyya asks, what is the denial of “A is either B or C”? Is it “A is
either not B or not C”? Bhattacharyya’s response is “no” ; he says it that “A is either
B or not-B” is the logical negative of “A is either B or C”; but this is the proverbial
excluded middle again; however, Bhattacharyya feels strongly that the “indeterminacy”
reeking through such negations that evade absolute truth is the “limiting mystery of all
philosophy”! Cf. [3, p.6̇01].
54 P. Bilimoria

different modes of negation involved. In the former, a certain affirmation

of a positive entity or event is said to be involved and the negation is more
usually than not by implication (arthāpatti ) rather than by direct reference:
so when one means ks.atriya he uses the single expression “non-brahmin”,
and it is implied that he is a ks.atriya. Hence the commitment is quite
marked, so that even where one is seen to deny “the flower is red”, at least
he believes the flower to have some colour or that its former red-colour has
now withered to a mellow-yellow. While in the case of the latter, prasajya, it
negates directly whatever the opponent asserts, and there is no implication,
nor does it affirm the counterfactual (for other castes are “non-brahmins”
too, such as śūdra, vaiśya); it simply means he is “not a Brahmin”. There
is no commitment to anything here, so that when one says “Man is not the
creator of the universe”, there is no a priori commitment to any creator or
for that matter creation. One could as well say, “Nothing exists” (for there
never was any-thing).
Using semantics and pragmatics Shaw is able to describe the moves at
stake in the view more clearly, and I quote:

Pragmatics is involved when it is claimed that in paryudāsa negation

affirmation is primarily intended, but in prasajya-pratis.edha negation
affirmation is not primarily intended if there is any affirmation [at all].
Semantics is involved when it is said that both the negative sentence
and its implicate describe the same fact in paryudāsa negation. The
negative sentence “he is not a Brahmin” and its intended implicate
“he is a ks.atriya” describe the same fact, and the expression “not a
Brahmin” and “ks.atriya” refer to the same thing. But in the case
of prasajya-pratis.edha negation “he is not a Brahmin” and “he is a
ks.atriya” do not have the same meaning, and the former does not
imply the latter. [13, pp. 62–63].12

Furthermore, it may be observed that when a prasajya-pratis.edha nega-

tion is leveled against the weaker paryudāsa negation to which the oppo-
nent is bent on committing his adversary (because of its affirmative im-
plicate), despite admitting to the prasajya-pratis.edha alternative (but re-
call it left the two sentences unconnected), and thereby extract a posi-
tion, both the alternatives are negated and all presuppositions therein also:
G¬([F (¬x) ∧ ¬F (x)]) — where F (¬x) is paryudāsa, and ¬F (x) is prasajya-
pratis.edha type of negation, and G is the intended residue qualified by the
parenthesis. This makes possible a prasajya-paryudāsa type of negation.
Space does not permit us to develop this line of thought to its logical culmi-

12 By the way, the Buddhist view represented in the text here is attributed to one

Avalokitavarta, which Shaw finds discussed by Kajiyama; but also cf. [12].
Abhāva 55

nation in Nāgārjuna’s famous catus.koti, four-cornered dialectic, especially

the last premise:there is neither x nor non-x; symbolized by “¬(p ∨ ¬p)”.13

2
Now to the last part of the paper. As we have seen, the Mı̄mām . sā were in-
exorably committed to the absoluteness of negation and its manifestation at
least in the injunctive mode, initially in Vedic propositions. The paribhās.ā
(“meta-language”) rules became instructive with the development of gram-
mar for its application to more secular speech. The Mı̄mām . sās, however,
under Kumārila Bhat.t.a, followed a trajectory laid out in Śabarasvāmin’s
bhās.ya,14 or even before that in one Upavars.a15 ; their intuitions on negation
is taken a step further and extendedly applied to the perceptual encounter
with the objective world as well. In other words, they set out to make a
connection between negation and things seen or un-seen, and even to the
soteriological end of all perception and knowing: namely, apavarga or eman-
cipation. There is a negative underbelly to that as well. Let me explore
these insights further, beginning with the foray into the epistemological
frontier.
Let us revert back to the fourfold division of negation we began with:
antecedent, posterior, mutual, and absolute. Kumārila’s argument turns on
causality: if one did not admit these four kinds of negation it would not be
possible to differentiate between cause and effect (i.e., we would be com-
mitting the fallacy of running cause into effect, and vice versa). But what
precisely is the ontological status of “negation”? Is it simply a characteris-
tic, viśes.a, of an object whose negation is being effected here, or is it about a
phenomenon in its own right? The word used here is “vastuta” (not a mere
semantic substantive but a “thing-signifying” substantive), substantive en-
tity or object, which I have advisedly called a phenomenon as it could be an
event, or an episode, or a substantive absence: an existentially non-existent
object, rather an “entitative item” in that sense nevertheless.16
In his adhikaran.a on “Abhāva”, Kumārila observes: there are “those
who hold that negation being a non-entity (avastu) is not an objective
13 In its expanded form, without collapsing double negation into its positive assertion:

(p ∨ ¬p) ∧ (¬p ∨ ¬¬p)).

The stifling modern debate of whether this commits the Buddhist to rejecting the
excluded middle or, more damagingly, law of noncontradiction is completely beside the
point when viewed from the foregoing logicus negative genealogy.
14 abhāvo’pi pramānābhavo nāstı̄tyasyārthayasannikrstasyeti, I.i.5; 5.47
. . ..
15 Cf. [12, p. 74] with the analysis [12, pp. 26–29] (I have drawn liberally from this

discussion). For a useful philological treatment and philosophical discussion of Kumārila’s

understanding and various possible translations of the key definition (by no means the
only telling verse) in Ślokavārttika (abhāvapariccheda 11), cf. [7].
16 I think we are entering a Meinongian-Sylvian jungle here.
56 P. Bilimoria

reality (tuccha) and is without a self-existent character (nih.svabhāva).”17

Adverting to the fore-going classification of negation, he avers that such
a classification would not be possible in respect of a non-entity (avastu);
therefore, he concludes “negation must be an entity”. For what is the
negation of an effect, other than the absence of the cause.18 In other words,
all effects are non-existent putatively prior to their production; otherwise,
under # 1 and # 2, the thing would be always present in time. If there
were no prior absence of curd in milk, then, we would have to say, curd is
present in the milk at all times, and we would not be able to cognise the
milk, that is, the cause of the effect (curd). And # 1 is negated in # 2, the
counter-entity (pratiyogin).
Likewise, with #3: a cow is decisively absent in a horse, and vice versa,
even as both are species of the same genus of vertebrates. Thus, “all things
are positive from their own standpoint, but negative from that of the other”.
As for # 4, if the absolute absence of colour were not in the air, or fire in
the water, or horns in the hare, smell in the waters, etc, we would be forced
to say otherwise.
These may be crude examples, but Kumārila wants to underscore a
metaphysical point; he needs a theory of natural kind negation in order to
ground the fledgling Mı̄mām . sā epistemology of “Abhāva”. What do I mean
by the latter? Among the accepted pramān.as or valid means of knowl-
edge are perception, inference, analogy, and testimony: but these are all
in respect of things that exist naturally, or perhaps supernaturally. What
if perception itself fails to deliver any object? Or any of the pramān.a for
that matter? However, where there is perception of absence of any or all
of the entities, padārthas (to use a Nyāya term), should we bring in ab-
sence simply as the failure of cognition of the same or should we say that
there is a veridical perception of the absence in the locus (ādhāra) where
the object or entity or event would otherwise have been present? Of course,
even this can be interpreted in at least two ways, as we shall see shortly
with the Nyāya insisting that there is a perception of something “not there”
in a substantive base where it would have otherwise been; but the percept
of “absence” is as it were conjunctive rather than disjunctive (though the
theory suggests that it is disjunctive but not in the radical sense in which
Kumārila would have it, consistent with the trope of antyantābhāva or ab-
solute negation). The other way to interpret this absence is to suggest that
there is an inference (via arthāparthi 19 or implication) made in respect of
17 This is an intersperse by Pārthasārathi Miśra preceding SV, Abhāva 7, p. 336; cf.

Footnotes 15 and 19.

18 Pārthasārathi Miśra in Nyāyaratnākara on SV 7–8 (p. 336). We find this also in

Śāstradı̄pikā.
19 Although for Udayana its is more a case of anumāna, straight inference than it is of

implication.
Abhāva 57

something “x” in lieu of the substantive base which is perceived as lacking

what would otherwise have been there. The “absented x” is epistemically
derivative, rather than sui generis (svabhāvic). Kumārila argues that the
failure of perception in this instance, that is “non-perception” itself, consti-
tutes a karan.a, instrument (reason = liṅga), that warrants the postulation
of a separate pramān.a. Kumārila’s reasoning for his radical interpretation
is partly captured in this passage (it is not the best argument one can make,
nevertheless, one can appreciate his motivation and boldness in urging for
this warrantability):
Negation is cognised (prameyatvācca gamyate) as an entity such as a
cow, etc. For it is the object of inclusive and exclusive conceptions
and is an object of cognition. It is not merely fortuitous, that it is
an (incorrect) imposition or an erroneous notion. Therefore, the fact
that (negation defined in terms of) the universal and the concrete
particular is not false. (SV, Abhāva 8–10, p. 337).
Kumārila’s worry is that in order for us to be able to differentiate a
particular from all other things and affirm it in its generality, we must first
cognise it with its class character and then proceed with the differentiation.
The cow is first cognised as the bearer of cow-ness (genus, ākr.ti = jāti ),
but for the precise cognition of the cow that belongs to Devadatta we must
differentiate (vyāvr.tti ) it from all other cows and from all other objects that
are not cows. He believes that such a cognition of affirmation and denial
is possible only because every entity (vastu) has two-fold reality, namely,
that of its class and that of the individual.20 Moreover, when we cognise the
antecedent non-existence of a thing, after having affirmed (anuvr.tti ) it as be-
longing to the class of “abhāva”, we differentiate (vyāvr.tti ) it from the other
three kinds of negation. We also cognise posterior-negation as distinct from
antecedent, mutual and the absolute negations. In the same way the last two
are cognised as distinct from each other and from the rest. So, Kumārila
contends that like any other (positive) form of reality negation forms the
object of cognition, and that it (negation) can be expressed by affirmative
and negative propositions (a move we already encountered in respect of pos-
itive and negative injunctions), with M -predicates, for example: “He does
nothing”, “He eats nothing” — the predicated absence is to be treated as
the object of a distinct cognition (vastu). This affirmative propositional use
of negation indicates a class character (anuvr.tta-dharma) Thus, all negative
cognitions are members of the class of abhāva or absence. And this abhāva
is a padārtha requiring its own pramān.a rather than being simply seen as
a prameya that can be subsumed under the regular pramān.as — as the
Nyāya would have it.
20 Supplemented with Dhirendra Sharma’s translation (with slight modification), [12,

pp. 29sqq].
58 P. Bilimoria

Of course, the Buddhist recognise the need to differentiate and affirm;

but to them this simply amounts to discrimination and elimination (apoha)
by virtue of there being positive entities, and it does not warrant bringing
this under a separate entitative category, much less a distinctive pramān.a
as Kumārila seems to need to ground this perception of the negative entity.
Precisely, Kumārila retorts that he should not be misunderstood as argu-
ing for the existence of abhāva in the absence of some positive reality (that
would be a sort of Nietzschean nihilism). The abhāva inexorably bears re-
lation to its counterpositive (pratiyogı̄), except that it is not known simply
via cognition of the counterpositive as his Nyāya counterparts argue for.
His point is that when he says “x is its own absence” he means just that,
x has both a presence (in regard to its own form, i.e., svarūpa) plus an ab-
sence (in regard to the form of another object, i.e., pararūpa) in all possible
worlds (nityam . sad-asadātmake vastuni ). The two are logically related; but
they are also independent, inasmuch as they form the objects respectively
of two different cognitions: people sometimes cognise one, sometimes the
other (jñayate kaiścid rūpam
. kiṅcit kadācana).(SV 12, p. 337).
The realist contention is that “absence of the cloth in the jar” simply
means that the cloth in its non-existent form inheres in another object, the
jar, and as such, produces the cognition of its “non-existent form” in the
jar. Furthermore, Kumārila argues:
A judgment (nirn.aya), concerning a (positive) existent (bhāva) —
such as “this is (the jar) and nothing else”— is not possible without
reference to the cognition of absence of everything else. Nor is the
knowledge (sam . vr.itti) that “it (the jar) does not exist” possible, with-
out reference to the negated thing itself; for there can be no cognition
without an objective substratum. (SV Abhāva, 15–16, pp. 338–339).

What Kumārila seems to be suggesting in this passage, which is brought

out more clearly in a later passage, is that the function of perception is over
once our sense-organ fails to make contact with an objective substratum out
there (arthasyendriyasannikr..s.tasyabādha) and there thus is non-generation
of perception, inference, etc.; instead, there is a stark absence of that object
(nāstı̄tyasyārthasya); now this “absence” is presented phenomenologically
or noetically in the mental grasping, as he comments:
After the object (the place where the jar is not present) has been
perceived, and the counter-entity (the jar) has been remembered, then
follows the notion that it (the jar) is not, which is purely mental (and
as such) independent of the sense-organs. (SV Abhāva, 27, p. 341).

So, the judgment “p” implies denial of “not-p”; hence, all meaningful
positive judgments embed negative cognitive essence — they are not mu-
tually contradictory as Śaṅkara later wanted to argue; when the judgment
Abhāva 59

tilts to the latter, in full view, literally, of the absence of “p”, we say it in re-
spect of the negative cognitive entity, “not-p”. There cannot be a cognition
without reference to some object or other. Kumārila is being consistently
realist, perhaps a naı̈ve negative realist!
In short, from Kumārila’s standpoint, negation is not plain “ignorance”
(ajñāna) but rather “the knowledge of absence” (abhāvajñāna), which oc-
curs through the absence of knowledge of the counterpositive; in other
words, pratiyoginah. anupalabdhyā abhāvasya upalabdhih.. Another way of
putting this is to say that “negation is a cognition of real absence in the same
way in which affirmation is (a) cognition of real presence”. In Stcherbatsky’s
words: “The Mı̄mām . sakas viewed non-existence as a reality sui generis (vas-
tuvantaram)”.21 The Buddhist objection to this standpoint is that the pro-
cess of grasping the absence connected with the perception of the bare locus
(kaivalya) could be explained inferentially as tending towards the exclusive
or paryudāsa kind of negation. Of course, the neo-Nyāya, railing against
the Buddhist reductionism, also came to accept abhāva as designating a
real which is absent in the locus (entity — hence a padārtha); however, the
Naiyāyikas denied that its perceptual grasping (albeit, the non-cognition of
the otherwise present in the locus of absence) needs to be attributed to an
independent pramān.a. The judgment for them is in respect of the cognition
of the locus devoid of the suggested relation with the object negated there,
on the epistemic consideration that the negative cognition must refer to pos-
itive entities only. The perception of kaivalya, bareness, is in respect of the
locus (ādhāra, adhikaran.a, āśraya) and not in respect of the absented object
as such to which of course it is related by resemblance (i.e., relationally to
its counterpositive). This alternative standpoint that inscribes a relational
exclusion was first championed by Prabhākara, and by Śālikanāth Miśra22 —
who set out to refute abhāvaprāman.a, which brings the Prābhākarans closer
to the Buddhists (who re-tool it as dr.śyānupalabdhi ). The Naiyāyikas seem
to have expropriated the qualified reading in Prabhākara in order to qualify
the more radical standpoint of Kumārila: on the basis of the non-perception
of a perceptible object (dr.śyādarśana or yogyānupalambha), whence the bare
locus, the substratum (bhūtala) is cognised, the absence of the object is ap-
prehended through a special inferential trope (yogyānulapabdhi ). The free-
standing reference to anulapabdhi is left-out as being otiose, in the Nyāya at
least, while for later Mı̄mām . sakas, particularly with Pārthasārathi Miśra,
anupalabdhi comes to replace abhāva in naming the distinctive pramān.a
that is implicated, rendered it simply as “non-perception of the otherwise
21 Cited in [12, pp. 35–36]. This means that for these philosophers the “non-existent” is
a reality sui generis (vastuvantaram) and this “knowledge of absence” (not just absence
of knowledge) is admitted via yogya-pratiyogya-anupalabdhi, though not as an inference
(anumāna), but a special means of knowing, pramān . a, which they called abhāva.
22 Cf. Footnote 23.
60 P. Bilimoria

perceptible object because of its absence in the locus”. Śālikanātha, whom

I just mentioned, invidiously misrepresents the theory of abhāva and gives
it the Naiyāyika’s twist; he writes
the evidence to prove abhāva is the cognition that there is no jar on
the ground (bhūtale ghat.o nāsti) [10, p. 265].23

As we have seen, Kumārila makes no mention of bhūtala (or ādhāra, adhi-

karan.a, āśraya); it is prolix for him to introduce the cognition of the sub-
stratum and what it is lacking —for the non-existence of the jar becomes a
mere attribute of the ground— and the ground would be there even the jar
is perceived but we do not necessarily relate the perception of the jar with
the perception of the ground, which again is “bare to the bone” but for the
presence of the jar! It would be more parsimonious to accept a direct cog-
nition of the non-existence of the jar, regardless, rather than make recourse
to its relational counterpositive — which is really a post-abhāva intellectual
exercise!
The Prābhākarans reject this view because perception would require that
there is contact of the senses with the non-existent, which is not possible;
the only contact there is is with the ground or locus. So absence has to
be viewed as the counterpositive (pratiyogin) of the locus which at another
moment flagged that presence. It is not the pratiyogin of absence that comes
to the cogniser’s mind, rather it is the unified cognition of the locus-minus-
the-object as kaivalya or bareness; the absence, if you will, of the relation
between the ground and the jar. It is a disjunction built upon a prior
conjunction. The disjunction is the non-perception of a positive thing, or
it is a perception that underscores a dis-affirmation (literally, an-upalabdhi )
of a thing, by exclusion (nivr.tti ), not an affirmation of no thing (tuccha).
It is the mere presence as bare-boned locus (tanmātra), which means that it
is devoid of the relation with the other object: anupalabdhir hi bhāvanam
abhāvah.. It is a denial of the perception of the object (adr.śyānupalabdhi ;
which is like ’inference to the absence of the best argument’).
However, I contend that in its Bhāt.t.a formulation this doctrine ought
not to be reduced to the pramān.a (means of knowing) of anupalabdhi (non-
cognition) as it only achieves this latter re-naming and articulation with
Pārthasārathi Miśra and the Prābhākaran, Śālikanath Miśra. Both these
commentators are influenced by and condescended to the criticisms of Bud-
dhist logicians, especially Dharmakı̄rti (a contemporary of Kumārila, c. 7th
century CE) and Prabhākara (the other doyen of the Mı̄mām . sā) and indeed
23 Neither is it a prameya nor is it separate pramān
. a what we call “non-cognition” is
really a perception of the positive entity (the counterpositive), namely that the locus is
bereft of a certain entity that otherwise was there — the latter is recalled by memory
and thus provides the counterpositive as the positive entity in the apprehension along
with the locus that is indeed given in perception.
Abhāva 61

to the Naiyāyikas. The latter themselves did not begin with the kind of rad-
ical doctrine of negation that they end up with, particularly, in Vallabha
and Raghunāth Śiroman.i.
The Buddhists again enter at this juncture and insist that without the
bare locus nothing is perceived: or that all cognitions are in relation of one
thing to another, and so it is only with reference to the empty ground that
the negative judgment, “the jar is not there” is made possible. Dharmottara
in Nyāya-bindu-t.ikā, puts it thus: “the perception of the bare locus with
reference to the perceptible jar and the apprehension of this fact are the
basis of the negative cognition”. “Non-perception is due to the paryudāsa
type of negation” (which is really Dharmakı̄rti’s position in Nyāya-bindu).24
It is the Nyāya-Vaiśes.ikas who heed most to this objection, recognising
that since there is no direct sense-contact with the non-existent object it
cannot properly be said to be a case of perception; however, if we shift the fo-
cus to the locus —or if the locus is made the focal point of perception— then
the onus too shifts from the non-existent to the absence-marking-the-locus,
as though the invisible ink- traces of the once-present-but-now-absent object
is simmering to the clear-light of the mind that picks up the efflorescence. A
negative entity of the kind the Bhāt.t.as of the realist world are hell-bent on
positing is not a substance and it therefore cannot come into contact with
the senses (indriyasannikar .sāyogya). Thus, on the anti-Bhāt.t.a view cir-
cling around, it is far better to hypotheticate a unique relation (viśe .san.atā)
between the so-called negative entity (nivr.tti, by default) and the locus or
substratum which is at least pravr.tti, substantive. Thus, they maintain
that the reality of non-existence (abhāva) is perceived not as ordinary per-
ception goes but as qualifying the locus which is a perceivable substance
(Śrı̄dhara, though it is also in Vallabha). This is tantamount to the return
of the bhūtalamātra in another guise. Even Raghunātha Śiroman.i, who bol-
stered the logical character of negation by redefining abhāvatva as upadhi,
and introducing double negation without regress, could not bring himself to
accept the traditional Bhāt.t.a view of “abhāva” as an ontological padārtha
that calls for its own distinctive pramān.a; he accepts abhāva as a prameya,
yes, but only derivatively so. And much has been made of this compromised
Navya-Nyāya doctrine in recent literature on Indian philosophy, even as it
took the sails out of the more radical Bhāt.t.a view. Why then this resistance
to condescending to “abhāva” as an object (padārtha) not in respect of the
absent-marked locus as the perceptible object (prameya), but of absence
qua the non-existent itself that is neither perceived nor inferred but is the
result of a certain “non-cognition”?
Two implications of the Bhāt.t.a view are worth noting: where there
is absolute absence of valid knowledge of something, one can be assured
24 Notice that paryudāsa negation still involves or implies an affirmation, so this is

“catch-22” situation from which even the Buddhist could not escape!
62 P. Bilimoria

that it is absolutely non-existent. The omniscience of the Buddha, and the

existence of God, are two such instances. But what if as the Prābhākarans
insist, the absence of perception (or non-perception) is a given for non-
perceptibles (adr.śyavastu), as for some perceptibles too (as in the standard
examples of anupalabdhi )? The Bhāt.t.a’s retort that their theory of abhāva
does not preclude non-perceptibles — not due to some yogic feat, but by
virtue of the possibility of cognition of non-existents in the empirical plane
(which does not rule out their existence and positive perception thereof in
other possible and real worlds: Martians, for example, seeing green cheese).
Adr.śya should not be conflated with anupalabdhi ; something non-existent
does not mean it is imperceptible (to the mind, even if not to the senses).
Sensory contact was not deemed essential to the perception of absence (and
by the time of Gaṅgeśa nor for presence either), why then for other non-
existents?
Second, and in concluding this disquisition, the theory of negation has
ramifications for a theory of emancipation as well. If the life of sam . sāra
is full of pain or duh.kha as the Buddha also recognised (he wasn’t the
first to do so, the Śraman.a and Jainas preceded him), then the eradica-
tion of pain or the “absolute non-existence of pain” (duh.khātyanta-vomiks.o
apavargah.), accordingly to asatkārya theory of causality, namely, “non-
pre-existent effect” entails that its causes, including its “antecedent non-
existence” (prāgabhāva), be destroyed (dhvam . sa-ed) by the production of
the effect: so this implies a double negation of contraries. Thus, it is clear
that apart from the problem of negative judgment, postulation of the nega-
tive reality was necessary (also) for their doctrines of Causality and Emanci-
pation. Although, the Mı̄mām . sakas’ apavarga was not as negative, ironically
for the bhūtalamātrikas, as that of the Naiyāyikas where the end-state of en-
lightenment was some zombie-like, tirelessly pain-free fiddling in a state of
near-boredom25 ; rather for the Mı̄mām . saka, there was at least the rich har-
vest of the apūrva, the transcendental credits or “never-before merits” from
the rituals performed and the negative injunctions too carried out equally
diligently or with the same inner propulsion of the śābdı̄bhāvana, from which
a blissful state of svarga, or heaven in all worlds could be gleefully drawn
upon. Here the empirical world becomes the object of the negative percep-
tion of absence, a real negation at that; and realism is not ever compromised:
one does not need to run off to or with the temptresses of metaphysical or
internal realism of any variety, Indian or Western.
Vive la negation and negative realism of the Mı̄māṁsā!

Primary Sources
SV. = Ślokavārttika, with Nyāyaratnākara of Pārthasārathi Miśra, in: [1].
25 Cf. [5].
Abhāva 63

References
[1] Kumārila Bhat.t.a. Ślokavārttika with Nyāyaratnākara of Pārthasārathi
Miśra. Prāchyabhārati Series 10. Tara Publications, Varanasi, 1978.
[2] Kumārila Bhat.t.a. Ślokavārttika. The Asiatic Society, Calcutta, 1985.
Translated by Ganganatha Jha.
[3] Krishna C. Bhattacharyya. Some Aspects of Negation. In Gopinath
Bhattacharyya, editor, Krishnachandra Bhattacharyya Studies in Phi-
losophy, volume I and II. Motilal Banarsidass, Delhi, 1983.
[4] Purushottama Bilimoria. Authorless Voice, Tradition and Authority in
the Mı̄mām. sā — Reflections of Cross-Cultural Hermeneutics. Nagoya
Studies in Indian Culture and Buddhism: Sambhās.ā, 16:137–160, 1995.
[5] Purushottama Bilimoria. Nyāya. In Helaine Selin, editor, Encyclopedia
of the History of Non-Western Science: Natural Sciences, Technology
and Medicine. Springer-, Dordrecht, 2007.
[6] Franklin Edgerton. The Mı̄māṅsā Nyāya Prakāśa or Āpadevı̄: A Trea-
tise on the Mı̄māṅsā System by Āpadeva. Translated into English, with
an Introduction, Transliterated Sanskrit Text, and Glossarial Index,
volume 36 of Sri Garib Dass Oriental Series. Sri Satguru Publications,
Delhi, 1986.
[7] Birgit Kellner. There are no Pots in the Ślokavārttika. Journal of the
Oriental Institute, 46(3–4):143–168, 1997.
[8] Bimal Krishna Matilal. The Navya-Nyāya Doctrine of Negation: The
Semantics and Ontology of Negative Statements in Navya-Nyāya Phi-
losophy, volume 46 of Harvard Oriental Series. Harvard University
Press, Cambridge, Massachusetts, 1968.
[9] Jitendranath N. Mohanty. Dharma, Imperatives, and Tradition. In
Purushottama Bilimoria, J. Prabhu, and Ramesh Sharma, editors, In-
dian Ethics, Classical and Contemporary Challenges, volume 1. Ash-
gate Publishers, Aldershot, 2007.
[10] Śālikenātha. Prakaran.apañcikā. Indian Council of Philosophical Re-
search, New Delhi, 2004. with an exposition in English, K. T. Pan-
durangi.
[11] K.S. Ramaswami Sāstrı̄, editor. Rāmānujācārya, Tantrarahasyam. Ori-
ental Institute, Baroda, 1956.
[12] Dhirendra Sharma. The Negative Dialectic. Sterling, Delhi, 1974.
64 P. Bilimoria

[13] Jayasankar L. Shaw. Negation and the Buddhist Theory of Meaning.

Journal of Indian Philosophy, 6:59–77, 1978.
[14] Jayasankar L. Shaw. Negation: Some Indian Theories. In Dalsukh
Malvania and Nagin J. Shah, editors, Studies in Indian Philosophy: A
Memorial Volume in Honour of Pandit Sukhlalji Sanghvi, pages 57–78.
Institute of 0Indology, Ahmedabad, 1981.
[15] Jayasankar L. Shaw. The Nyāya on Double Negation. Notre Dame
Journal of Formal Logic, 29(1):139–154, 1989.

[16] Frits Staal. Negation and the Law of Contradiction in Indian Thought:
A Comparative Study. In Frits Staal, editor, Universals : Studies in
Indian Logic and Linguistics, pages 109–128, Chicago, 1988. Chicago
University Press.

[17] Frits Staal. Universals: Studies in Indian Logic and Linguistics.

Chicago University Press, Chicago, 1988.
[18] Satis C. Vidyabhusana. History of the Medieval School of Indian Logic.
Oriental Books Reprint Corporation, New Delhi, 1977.
[19] David B. Zilberman. Birth of Meaning in Hindu Thought. Boston
Series in Philosophy of Science. D. Reidel Publishing Co., Dordrecht,
1988.