KAMALESWAR BHATTACHARYA

A NOTE ON FORMALISM IN INDIAN LOGIC

The late Professor Ingalls was among the great scholars of the generations
preceding mine it was given me to know. Although I was never his pupil
in the strict sense of the term, I met him several times at Widener 273
and had long discussions with him. I sent him some of my publications
and, whenever he received one that appeared to him significant, he
acknowledged it with some nice words in his elegant handwriting. I
cannot resist the temptation of citing a part of what he wrote after

receiving my book LAtman-Brahman

dans le Bouddhisme ancien
(1973):
. . . a thought occurs to me which may make you smile. It occurred to me while
reading the Sam
. yutta passage that you quote where the Buddha likens the discovery
of the dharman. am
a to the finding of a ruined city in the forest. Does not
. dharmat
this passage have a curious and happy illustration in your own life? for here an
inscription from the jungles of Cambodia has led you to what might be called the
basic truth of early Indian religion.

Four years later I started publishing my French translations of Navyanyaya. I chose that language for two main reasons. First, apart from
Fouchers Tarkasam
. graha (1949), which, though elementary, is not free
from defects, nothing was available in French at that time on Navyanyaya. Second, it appeared to me far more difficult to put Navya-nyaya
into French than to put it into English, a challenge which I thought
someone should attempt to meet. And Professor Ingalls, on receiving

the first instalment of my translation of the Siddhantalaks

. an. a with

s, wrote to me:
Raghunathas Ddhiti and the Jagad
. . . It adds appreciably to the small store of western tools for the interpretation of
Navya-nyaya . . . Putting Navya-nyaya into French strikes me as several degrees more
difficult than putting it into English, as French clings more persistently than English
to its traditional rules of wordbuilding and syntax. But you manage . . .

Today, I would have been happy to be able to honor his memory in a

worthy manner, by a significant contribution to Navya-nyaya studies
a field in which he was a pioneer in the West. But, to my great regret, I
can only present what I consider a frivolous note having been engaged
for some time past in an unusual work.
It has long been recognized that there was a formal logic in India.
And, in 1956, I. M. Bochenski emphasized for the first time, so far
Journal of Indian Philosophy 29: 1723, 2001.
c 2001 Kluwer Academic Publishers. Printed in the Netherlands.

18

KAMALESWAR BHATTACHARYA

as I am aware that formal logic was born in two and only two
cultural spheres: the western and the Indian.1 Why it was so is yet to
be determined precisely. The structure of the Indo-European languages
readily comes to our mind; and one scholar, at least, has expressed this
idea.2 But this was, perhaps, not the only factor involved. At all events,
no one today believes that Indian logic was influenced by Aristotle, or
that Aristotle was influenced by the logic of the Brahmins.
To be sure, Indian formal logic was not the same as Western formal
logic; but it was formal, nevertheless. Bochenski, on one occasion,
spoke of eine echte und korrekte, obwohl noch in vielem elementare
formale Logik,3 and, on another, said:
. . . wir es hier mit einer zwar schlechten, aber doch formalen Logik zu tun haben
. . .4

On another still, he said:

Mogen auch diese Ergebnisse dem westlichen Logiker bescheiden erscheinen, so
lat sich doch nicht verkennen, da sich die indische Logik . . . auf die Ebene einer
echten formalen obwohl bei weitem noch nicht formalistischen Logik erhoben
hat.5

Bochenski explained in which sense he called Indian logic a formal

logic:
. . . Da es sich um eine formale Logik handelt, sieht man daraus, da die indischen
Denker Formeln aufstellen, welche sich auf die Grundfrage der Logik beziehen,
namlich auf die Frage, ob etwas aus etwas anderem folgt oder nicht und zwar so,
da diese Formeln als allgemeingu ltig gedacht werden.6

All these questions need further investigation. Over the past years,
Nyaya studies have made progress but not so much as expected.
Particularly, concerning Navya-nyaya (New logic), what Staal wrote
forty years ago still remains valid to a great extent:
The study of Navya-nyaya logic is still in its infancy. Of the huge mass of manuscript
material only a fragment has been published. Even of the considerable amount of
published material only a small part is read.7

I might add that translations and interpretations appear nowadays

which serve no purpose, and editions of texts appear which cannot be
read.
I would have liked to present here a new edition of a part of Yajnapati
Upadhyayas Prabha (the earliest commentary known so far on the
. i)8 or of the Dus
. an. oddhara
of Narahari (Yajn
apatis
Tattvacintaman
9
son). But it has not been possible, for lack of time.
The point I wanted to make is that there was a formalism, too, in
Nyaya, and especially in the later phase of Navya-nyaya starting with
Raghunatha.

A NOTE ON FORMALISM IN INDIAN LOGIC

19

Formal Logic and formalistic logic (writes ukasiewicz) are two different things.
The Aristotelian logic is formal without being formalistic, whereas the logic of the
Stoics is both formal and formalistic. Let us explain what in modern formal logic
is meant by formalism.
Modern formal logic strives to attain the greatest possible exactness. This aim can
be reached only by means of a precise language built up of stable, visually perceptible
signs. Such a language is indispensable for any science. Our own thoughts not formed
in words are for ourselves almost inapprehensible and the thoughts of other people,
when not bearing an external shape, could be accessible only to a clairvoyant. Every
scientific truth, in order to be perceived and verified, must be put into an external
form intelligible to everybody. All these statements seem incontestably true. Modern
formal logic gives therefore the utmost attention to precision of language. What is
called formalism is the consequence of this tendency. In order to understand what
it is, let us analyse the following example.
There exists in logic a rule of inference, called formerly modus ponens and now
the rule of detachment. According to this rule, if an implication of the form If
, then is asserted and the antecedent of this implication is asserted too, we
are allowed to assert its consequent . In order to be able to apply this rule we
must know that the proposition , asserted separately, expresses the same thought
as the antecedent of the implication, since only in this case are we allowed to
perform the inference. We can state this only in the case where these two s have
exactly the same external form. For we cannot directly grasp the thoughts expressed
by these s, and a necessary, although not sufficient, condition for identifying two
thoughts is the external equality of their expressions. When, for instance, asserting
the implication If all philosophers are men, then all philosophers are mortal you
would also assert as second premiss the sentence Every philosopher is a man, you
could not get from these premisses the conclusion All philosophers are mortal,
because you would have no guarantee that the sentence Every philosopher is a man
represents the same thought as the sentence All philosophers are men. It would be
necessary to confirm by means of a definition that Every A is B means the same
as All As are Bs; on the ground of this definition replace the sentence Every
philosopher is a man by the sentence All philosophers are men, and only then will
it be possible to get the conclusion. By this example you can easily comprehend the
meaning of formalism. Formalism requires that the same thought should always be
expressed by means of exactly the same series of words ordered in exactly the same
manner. When a proof is formed according to this principle, we are able to control
its validity on the basis of its external form only, without referring to the meaning
of the terms used in the proof. In order to get the conclusion from the premisses
If , then and , we need not know either what or what really means; it
suffices to notice that the two s contained in the premisses have the same external
form.
Aristotle and his followers, the Perpatetics, were not formalists. As we have
already seen, Aristotle is not scrupulously exact in formulating his theses. The most
striking case of this inexactitude is the structural discrepancy between the abstract
and concrete forms of the syllogisms. Take as an example the syllogism with opposite
premisses quoted above, in our section 4. Let B and C be science and A medicine.
Aristotle states:
In variables:

In concrete terms:

If B belongs to all A
and C belongs to no A,
then C does not belong to some B.

If all medicine is science

and no medicine is science,
then some science is not science.

20

KAMALESWAR BHATTACHARYA

The difference of corresponding premisses, of which the two syllogisms consist, is

evident. Take, for instance, the first premiss. To the formula B belongs to all A
would correspond the sentence Science belongs to all medicine, and to the sentence
All medicine is science would correspond the formula All A is B. The sentence
in concrete terms, given by Aristotle, cannot be regarded as a substitution of the
abstract formula accepted by him. What is the cause of this difference?
Alexander gives three explanations of this problem: the first may be omitted as
unimportant, the last is a philosophical one and is, in my opinion, wrong; only the
second deserves our attention. According to this explanation, in formulae with the
verb to be predicated of something and, we may add, with the verb to belong to
something, the subject and the predicate are better distinguishable (
o)
than, we may add again, in formulae with the verb to be. In fact, in formulae with
to be the subject as well as the predicate is used in the nominative; in formulae
preferred by Aristotle only the predicate is in the nominative, and the subject is either
in the genitive or in the dative and therefore can be more easily distinguished from
the predicate. Very instructive, too, is the final remark of Alexander, from which it
follows that to say Virtue is predicated of all justice instead of the customary All
justice is virtue was felt in Ancient Greek to be as artificial as in modern languages.
There are still more cases of inexactitude in Aristotelian logic. Aristotle constantly
uses different phrases for the same thoughts. I shall give only a few examples
of this kind. He begins his syllogistic with the words A is predicated of all B,
but shortly he changes these words into the phrase A belongs to all B, which
seems to be regular. The words is predicated and belongs are frequently omitted,
sometimes even the important sign of the quantity all is dropped. Besides the form
A belongs to some B there are forms which may be translated A belongs to some
of the Bs. The premisses of the syllogism are combined by means of different
conjunctions. Syllogistic necessity is expressed in different ways and is sometimes
entirely omitted. Although theses inexactitudes have no bad consequences for the
system, they contribute in no way to its clearness or simplicity.
This procedure of Aristotle is probably not accidental, but seems to derive from
some preconceptions. Aristotle says occasionally that we ought to exchange equivalent terms, words for words and phrases for phrases. Commenting on this passage,
Alexander declares that the essence of the syllogism depends not on words but on
their meanings. This statement, which is manifestly directed against the Stoics, can
be understood thus: the syllogism does not change its essence, i.e. it remains a
syllogism, if some of its expressions are replaced by other equivalent expressions,
e.g. if the expression to be predicated of all is replaced by the equivalent expression
to belong to all. The Stoics were of a directly opposite opinion. They would say
that the essence of the syllogism depends on words, but not on their meanings. If
therefore the words are changed, the syllogism ceases to exist. This is illustrated
by Alexander with an example from the logic of the Stoics. The rule of inference
called modus ponens:
If , then ;
but ;
therefore ,
is the first indemonstratable syllogism of the Stoics. Both the Stoics and the Peripatetics seem mistakenly to regard the phrases If , then and entails as
having the same meaning. But if, in the syllogism given above, you replace the
premiss If , then by entails , saying:

A NOTE ON FORMALISM IN INDIAN LOGIC

21

entails ;
but ;
therefore ,
you get according to the Stoics a valid rule of inference, but not a syllogism. The
logic of the Stoics is formalistic.10

I have quoted a long passage, which may be familiar to most, if not

all, readers of this Journal. But it seemed to me necessary to do so, in
order to make my point crystal clear with no intention of equating
Indian logic with Western logic.

Nyaya, as is well known, operates with knowledges (jnana)

not as
pure psychological phenomena but in the logical form in which they
appear when expressed in language. Now, the formalism in question is particularly attested by the way Nyaya deals with contradictory
knowledges. Intuitively, more than one knowledge may be seen as contradictory to a single knowledge; for instance, the knowledges The moun an),
Fire has absence
tain has absence of fire (parvato vahnyabhavav
avav
an),

of occurrent-ness on the mountain (vahnih. parvatavrttitvabh

the mountain
Fire is the counterpositive of an absence occurring on

), and so on, to the knowledge

(vahnih. parvatavrttyabhavapratiyog

for, irrespective of their

The mountain has fire (parvato vahniman);
external form, all the former point to the same thing, the absence of
fire on the mountain. And the older school did admit, in effect, that all
these knowledges were contradictory to the knowledge The mountain
has fire. The new school, however, does not, on the ground that for two
knowledges to be logically contradictory to each other they must have
ak
ara).

the same form (saman

Thus, contradictory to the knowledge
is only the knowledge The
The mountain has fire (parvato vahniman)
an),
and not the
mountain has absence of fire (parvato vahnyabhavav
others as well.
The late Mahamahopadhyaya Kalpada Tarkacarya showed another
instance.11 I shall mention here another still and, in that connection,
an interesting piece of information given by Jagadsa.

To a knowledge of invariable concomitance (vyapti-j

nana)
is contra

dictory a knowledge of deviation (vyabhicara-j

nana).
Now, invariable

concomitance (vyapti)
has been defined in various ways; and the knowledge of invariable concomitance, as well as the knowledge of devi
ation (vyabhicara),
takes various forms, according to the definition of
invariable concomitance adopted. Is any knowledge of deviation, then,
contradictory to any knowledge of invariable concomitance? The older
school says yes; but the new school, no: to be contradictory to each
other, the knowledge of invariable concomitance and the knowledge

22

KAMALESWAR BHATTACHARYA

of deviation must have the same form. Thus, when one adopts the
definition of invariable concomitance as the fact that the object to be

established (sadhya)
is not the counterpositive of an absence residing in
av
apratiyogitvam],

what is a locus of the reason (hetu) [hetumannis. .thabh

the knowledge of invariable concomitance takes the form The object
to be established is not the counterpositive of an absence residing in

av
apratiyogi),

what is a locus of the reason (sadhyam

. hetumannis. .thabh
and contradictory to it is the knowledge of deviation: The object to be
established is the counterpositive of an absence residing in what is a

avapratiyogi),

locus of the reason (sadhyam

and not
. hetumannis. .thabh
also, e.g., the knowledge: The reason resides in what is a locus of the

avavadv

absence of the object to be established (hetuh. sadhy

abh
rttih. ),

which is contradictory to the knowledge of invariable concomitance:

The reason does not reside in what is a locus of the absence of the

avavadav

object to be established (hetuh. sadhy

abh
rttih. ), corresponding
to the definition of invariable concomitance as the fact that the reason
does not reside in what is a locus of the absence of the object to be

avavadav

established (sadhy
abh
rttitvam).

In this connection, Jagadsa makes

an interesting observation, while
criticizing his predecessor Bhavananda. In his interpretation of a

seemingly12 difficult passage of the Ddhiti on the Siddhantalaks

. an. a,
a passage which gave rise to a great deal of speculation among the
old commentators, Bhavananda, to avoid the difficulty, proposed
to replace, in the knowledge of invariable concomitance, the word
gotva cow-ness by the equivalent govrtti what resides in the cow:

What resides in the cow is not the delimitor

of the counterpositiveness of an absence residing in the locus of dewlap, etc. (govrtti
adimannis

avapratiyogit

sasn
anavacchedakam).
But Jagadsa
. .thabh
objects, saying that, in that case, the knowledge of invariable concomitance would not be contradicted by the knowledge of deviation, where
the word cow-ness (gotva) must occur.13
In another context, however, the older school appears to be more

formalistic than the new. The consideration (paramar

sa) immediately
preceding the inferential knowledge (anumiti) may take either the form
The subject has the reason pervaded by the object to be established

paks. ah. ) or the form The reason pervaded by

(sadhyavy
apyahetum
an

the object to be established is in the subject (sadhyavy

apyo
hetuh.
paks. e). And the older school maintains that from the first form of
consideration arises the inferential knowledge The subject has the

and from the second, the

object to be established (paks. ah. sadhyav
an),
inferential knowledge The object to be established is in the subject

A NOTE ON FORMALISM IN INDIAN LOGIC

23

(paks. e sadhyam).
But the new school holds that the inferential knowledge
takes invariably the form The subject has the object to be established

14
(paks. ah. sadhyav
an).
NOTES
1

Bochen ski 1996: 13. See also Bochenski 1952: 217.

Staal 1988: 74.
3
Bochen ski 1996: 483.
4
Bochen ski 1996: 505.
5
Bochen ski 1996: 509.
6
Bochen ski 1996: 516.
7
Staal 1988; 79.
8
Cf. Bhattacharya 1988. It may be noted that a second manuscript of the
Pratyaks. akhan. d. a of the Prabha has recently been traced.
9
Recently published in Gaekwads Oriental Series 179 (Vadodara, 1999).
10
ukasiewicz 1957: 1519.
11
See Tarkacarya 1964: 6263.
12
See Bhattacharya 1995: 396, n 51.
13
Bhattacharya 1995: 384, 394395 and n 45, with the references.
14
Bhattacharya 1995: n 27.
2

REFERENCES
Bhattacharya, Kamaleswar (1988): Yajnapatyupadhyayaviracitayam
.
Prof. N.S.
Tattvacintaman. iprabhayam Isvaravadavyakhyanam in: Lokaprajna:
anuja

ac
arya

Ram
Tat
Felicitation Volume, Puri: 275294.

.i
Bhattacharya, Kamaleswar (1995): Le Siddhantalaks
. aprakaran
. a du Tattvacintaman
. an

sa avec la D
de Gange
dhiti de Raghunatha Siroman
. ika de Jagadsa
. i et la T
Tarkalam
ara (suite), Journal Asiatique t. 283: 373406.
. k
Bochenski, I. M. (1952): Review of Ingalls, Materials . . ., Journal of Symbolic
Logic 17: 117119.
Bochenski, I. M. (1996): Formale Logik, 5th edn. Freiburg/Munchen: Verlag Karl
Albert 1956.
ukasiewicz, Jan (1957): Aristotles Syllogistic from the Standpoint of Modern Formal
Logic, 2nd edn. Oxford: Clarendon Press.
Staal, Frits (1988): Formal Structures in Indian Logic, Synthese: An International
Quarterly for the Logical and Psychological Study of the Foundations of Science
12(1960): 279286; Reprinted in Universals. Studies in Indian Logic and Linguistics.
Chicago and London: The University of Chicago Press: 7380.

Tarkacarya, Kalpada (1964): Nyayadar

sanabinduh. . Varan. as: Varan. aseyasam
. skrtavisva
vidyalayah. sam
. v. 2021 [AD 1964].

32 Route de Brie
Brunoy 91800
France