THE AUTHORSHIP OF THE NYAYAPRAVESA
By A. Berriedale Keith
The Indian Historical Quarterly
Vol 4:1, 1928, pp. 14-22.
p. 14
It should be noted, in view of the interest
naturally taken in the authorship of the
Nyayapravesa, which has been dealt with in an
interesting paper by Mr. Vidhusekhara Bhattacharya,
(I.H.Q., III, pp. 152-60) that Mr. M. Tubianski in a
paper on the authorship of the Nyayapravesa in the
Bulletin de l' Academie des Sciences de l' URSS
(1927, pp. 975-82) contends that it is certain that
the Nyayapravesa was not the work of Dignaga. Writing
without knowledge of Mr. Vidhusekhara Bhattacharya's
arguments, his chief evidence lies in a comparison of
the Nyayadvara known from a Chinese version and the
Nyayapravesa. His arguments, however, are not wholly
convincing.
(1) He points out that the Nyayapravesa adds some
fallacies of the thesis which are not found in the
Nyayadvara; this, of course, merely suggests
difference of date of composition,
(2) The dusanabhasas, 14 in number of the
Nyayadvara and even of the Pramanasamuccaya, are
omitted, all that is valuable in them being subsumed
under the hetvabhasas as in the Nyayabindu of
Dharmakirti. Now it may readily be admitted that the
dusanabhasas are merely an illegitimate relic of the
old Nyaya jati, and
___________________________________
1. Cf. my article 'Jataka' in Hastings' Encyclopaedia
of Religion and Ethics, vol. vii, p. 492.
p. 15
that their disappearance is valuable, but there seems
no good reason for denying that Dignaga himself
advanced to the point of rejecting these types;
difference of date again explains the situation. (3)
The terminology is more lucid, which again merely
proves later date, as is indicated also by the
improved form of exposition. (4) Of more importance
is the fact that Dharmakirti in his criticism of
Dignaga in the Nyayabindu uses the term
istavighatakrt in his criticism in lieu of
dharmavisesaviparitasadhana, the more effective name
introduced in the Nyayapravesa. Would this be
possible if the Nyayapravesa belonged to Dignaga? It
seems to me that it, certainly would; there is no
reason obvious why Dharmakirti should not have used
the older and more common name. All the arguments,
therefore, from contents may be disposed of as
inconclusive, on the score that they are consistent
with the Nyayapravesa being the later work. This
view, it must be noted, involves the rejection of Mr.
Vidhusekhara Bhattacharya's argument (I. H. Q., III,
155) that Dignaga in his comment on his
Pramanasamuccaya refers to the Nyayapravesa and not
to the Nyayadvara, and of his view that
Jinendrabuddhi in his comment on that work actually
cites, from the Nyayapravesa the definition
pratyaksam kalpanapodham; there is no real doubt that
the Nyayadvara was used in the Pramanasamuccaya, and
the definition is taken thence. Even if this proved
not to be the case, it would be necessary to remember
that the quotation is not absolutely assigned by
Jinendrabuddhi to the Nyayadvara (understood by Mr.
Vidhusekhara Bhattacharya as Nyayapravesa) but to the
Nyayadvaradi. It must also be admitted that the
references to the Nyayapravesa seen in Kumarila and
Parthasarathi Misra by Mr. Vidhusekhara Bhattacharya
(pp. 156, 157) are not conclusive evidence, in the
absence of any definite mention of that text and of
any proof that the doctrines cited are not found in
other parts of the writings of Dignaga. It appears to
me,
p. 16
therefore, that from the evidence adduced by Mr.
Vidhusekhara Bhattacharya and Mr. Tubianski alike no
certain conclusion can be drawn.
The external evidence is declared by Mr.
Tubianski to support his denial of Dignaga's
authorship. (1) The Chinese tradition ascribes the
Nyayapravesa to Sankara Svamin, and it preserves both
texts, while the Tibetan has only the Nyayapravesa,
omitting the Nyayadvara. Presumably, therefore, the
Chinese had larger materials for ascription, and,
since they adopted the Nyayapravesa as the basis of
their logical studies, presumably knew the author.
This argument, however, omits to note that, of the
two Tibetan versions of the Nyayapravesa one is based
on the Chinese version of the original, and it
deliberately gives the name of the author as Dignaga.
This is certainly strong evidence that there was a
Chinese tradition which ascribed the text to Dignaga,
and, therefore, derogates fatally from the
conclusiveness of the argument from the Chinese
tradition. (2)I-Tsing's list of Dignaga's works
contains nothing that can be identified with the
Nyayapravesa. But that is hardly the case. No. 4 of
I-Tsing's list appears to be the Hetudvarasastra1 or
Hetuvidyadvara, and it is not enough to say that it
cannot answer to Nyayapravesa, " as the last does not
treat exclusively of hetu." There is decidedly more
validity in Mr. Vidhusekhara Bhattacharya's
declaration that Hetudvara and Nyayadvara (equivalent
to Nyayapravesa) mean the same thing. Apparently,
therefore, I-Tsing did know the Nyayapravesa as well
as the Nyayadvara, which is doubtless meant by No. 6
in his list. (3) Mr. Tubianski holds that it is
possible to explain the Tibetan blunder, as he calls
it. They knew of the Nyayadvara from Dignaga's own
reference: to it in his comments on the
Pramanasamuccaya, and, as
___________________________________
1. There is no reason to identify this with the Hetu-
cakrahamaru only, which, according to the History
of Indian Logic, p. 300, n.1, is only a part of
the Nyayapravesa.
p. 17
they had no version of it, the similarity of name
induced them to take Sankara Svamin's Nyayapravesa
for the Nyayadvara, as is indicated by the style
Nyayapravesadvara given to one of the Tibetan
versions, This argument seems to possess no cogency,
and as a matter of fact there is not the slightest
evidence of ignorance on the part of the Tibetans. On
the contrary, the Tibetan index to the Bstan-hgyur
the Dkarchag , expressly warns against confusing the
Nyayapravesa with the Nyayadvara, which shows that,
if the two works of the same general tenor by the
same author were sometimes confused, critical opinion
was perfectly aware of the distinction. Nothing,
therefore, can be made of this argument , nor is it
necessary to suggest any ground why the Chinese went
wrong; as mentioned above, their mistake was probably
not general.
Is there then anything to support the Tibetan
tradition of authorship? Note should be made of N. D.
Mironov's contribution, made accessible in the
Garbe-Festgabe (1927, pp. 37-46), where he suggests
two small points as supporting the tradition of
Tibet. The last verse of the Nyayapravesa runs:
padarthamatram akhyatam adau dinmatrasiddhaye
yatra yuktir ayuktir va sanyatra suvicarita.
Mr. Mironov suggests that in dinmatrasiddhaye we
have an allusion to Dinnaga's name, and he thinks
this may be supported by the fact that Haribhadra in
his comment on anyatra writes Pramanasamuccayadau.
Mr. Tubianski objects that Haribhadra's remark merely
proves that he assumed the author of the
Nyayapravesa to be of the same school as Dignaga, and
this is true. But on the other hand, the remark is
specially apposite if the author really were Dignaga,
in which case it would be perfectly true and very
much to the point. Nor is it quite legitimate to pass
over the posible play in dinmatrasiddhaye; it can
carry no great weight, but it certainly improves Mr.
Mironov's argument. The results arrived at may be
summed up as follows:
p. 18
though largely negative, they represent all that can
be said with any approach to certainty. The
Nyayadvara, preserved only in a Chinese rendering, is
a work of Dignaga's, and was written before, and used
by the author in, his Pramanasamuccaya and
commentary. There is nothing in the Nyayapravesa
which is not compatible with the authorship of
Dignaga, if we assume, as me are perfectly entitled
to do, that it was written later than either of the
two works above mentioned, and embodies improvements
in minor detail. In any case it is essentially of the
same general type as his works, and call he used with
a high degree of probability as setting forth his
views. It is clear that in Tibet it was ascribed to
Dignaga; this was probably the case also to some
extent in China, and the most natural explanation of
a remark of Haribhadra's is that he thought the text
on which he commented of Dignaga's written after the
Pramanasamuccaya, to which reference is made in the
term anyatra used in the last line of the work, There
is no difficulty in holding that the work is
equivalent to No. 4 in I-Tsing's list of Dignaga's
works. Against the ascription to Dignaga there is but
one matter of weight, the declaration of the Chinese
tradition which ascribes the Nyayapravesa as
translated to Sankara Svamin, and even that is not
conclusive, because, while, as we have it, the
Chinese version of the Nyayapravesa ascribed to
Hiuen-tsaug gives Sankara Svamin as the author, the
Tibetan version derived from this Chinese version1
gives Dignaga, suggesting that in the text known to
the Tibetan translator from Chinese this ascription
was found. Moreover some weight attaches to the fact
that Hiuen-tsang is silent else where regarding
Sankara Svamin, which is rather curious if he really
translated an important text of his.
______________________________
1. It was translated first by one and later by another
Chinese monk
p. 19
Even, however, the recovery of the Sanskrit
version of the Nyayapravesa leaves us hopelessly in
the dark on many problems of the development of
Indian logic. I remain of opinion(1) that the most
satisfactory way of reading the history of its
development is to ascribe to Dignaga priority over
Prasastapada, and to explain, as does Dr. Randle(2)
the references of Dignaga to doctrines not found in
the Vaisesika Sutra but set out in the
Prasastapadabhasya, as dealing with the doctrines of
early commentators on the Vaisesika. That there were
such has never been doubted by any one; without them
the Bhasya, would never have assumed its present
form. I differ, however, from Dr, Randle(3) as
regards the question of Dignaga's doctrine of
indissoluble connection; Vacaspati (p. 127)
unquestionably denies that on Dignaga's view there
can be any indissoluble connection of real things,
but it is equally clear that Dignaga himself denied
the connection of reals. What other meaning can be
ascribed to the famous passage: sarvo 'yam
anumananumeyavyavaharo buddhyarudhenaiva
dharmadharmidhavena na bahihsadasattvam apeksate?
Dignaga appears to me in this passage, which is
unquestionably his, to declare clearly that all the
relations of probans and probandum have nothing
whatever to do with external reality-which on his
idealistic system(4) was beyond knowledge if it had
any existence at all--but depend upon the intellect.
That means, in the absence of any reason to deny the
obvious sense, that all the relations with which we
have to do are matters imposed by the intellect, and
accords admirably with the doctrine which regards the
intellect as the essential reality, Dr., Randle holds
that there is no evidence that Dignaga bases his
doctrine of indissoluble connection on his idealism,
but this evidence seems to me
_________________________
1. Indian Logic and Atomism (1921),
2. Fragments from Dinnaga, p.65.
3. Ibid., pp. 53, 54.
4. For his thing-in-itself, see Stcherbatsky, Nirvana,
pp. 153, 154, 161; for his ultimate idealism,
Keith, Buddhist Philosophy, p. 308.
p. 20
to be explicitly contained in the word buddhyarudha.
How else could Dignaga have expressed his doctrine?
Indeed it would be very satisfactory if other
logicians had given us anything half so explicit.
It seems further to me still proper to argue that
it is probable that the doctrine of indissoluble
connection was derived by Prasastapada from a school
in which that doctrine had a natural right exist. It
appears to be impossible to deny that such a doctrine
has such a right to exist from the standpotnt of such
a system as that of Dignaga; the denial of the Nyaya
school has no cogency for us, nor has it any
relevance to the question whether it was from Dignaga
that Prasastapada derived the doctrine of
indissoluble connection, which inevitably assumed a
very different aspect in its relation to the
Vaisesika as a realistic system. We are unlikely ever
to have any conclusive evidence for ascribing to
Dignaga the origin of the doctrine of indissoluble
connection, since it may have been evolved by a
Buddhist predecessor, but there is sufficient
evidence to show that he developed it, and, as it
admirably suits an idealistic position, the
probability that it is a Bauddha doctrine is
extremely great. We cannot safely assume that it is
an accident that Uddyotakara in his attack on the
doctrine of avinabhava associates it with Dignaga,
and assign the fact to "his normal habit of ignoring
Vaisesika logic.'' It is equally legitimate to hold,
and indeed far more probable, that Dignaga was
attacked because he was, if not the inventor, the
protagonist in the exposition and defence of the
doctrine. It is well to remember, that philosophic
doctrines emerge often from more than one mind, and
the origination is in many cases wholly impossible of
determination. But an idealist school is a more
natural source of a, doctrine of indissoluble
connection than a realist. Dr. Randle indeed seems to
hold (p. 26) that the Buddhist logic classified
inference "according as they are based on the real
relations of causality and identity," thus agreeing
with one interpretation of Vaisesika Sutra, ix, 2, 1,
but the term
p. 21
"real" appears to me to be wholly inapplicable to the
Buddhist view. The relations are clearly dependent on
buddha and in no sense real, as they doubtless are on
the Vaisesika view. As I pointed out,(1) writing
before the classification of inferences was known to
be found in Dignaga's Pramanasamuccaya, (2) the
classification is in no wise in disagreement with the
essential doctrine laid down by Dignaga. Any other
interpretation reduces Dignaga's view to hopless
confusion.
While the question of the invention of the
trairupya, or three canons of syllogism, is not
essentially bound up with the issue of priority of
discovery of the doctrine of indissoluble connection,
it seems to me that the effort(3) to ascribe it to
the Vaisesika school is implausible. Prasastapada
cites memorial verses, in which the doctrine is
asserted to be held by Kasyapa, which means, of
course, Kanada,(4) the author to whom the Vaisesika
Sutra is ascribed. It is argued that the effrontery
of such a claim, if the doctrine was really a
Buddhist innovation, would be incredible. But this is
to ignore the mode of thought prevalent in the
schools. Happily an illustration is available;
Sugiura tells us that Dignaga ascribes the doctrine
of the nine reasons, which follows from the
trairupya, to Socmock, i.e., Aksapada, though it is
patently not to be found in the Nyaya Sutra. In truth
the most that can be said for the Sutras of the two
schools is that there is a hint of the trairupya in
the Nyaya, v, 1, 34, and in the classification of
fallacies in the Vaisesika. It was quite enough for
Prasastapada that the explicit doctrine could be
fitted into his system, and it would have been
impossible for him, to judge from the spirit shown by
both schools in their writings, to accept anything as
given by the Buddhist philosophy. Here again the
simple explanation is that Dig-
____________________________
1. Indian Logic and Atomism p.10.
2. History of Indian Logic, pp. 280, 281.
3. Fragments from Dinnaga, pp. 66, 67.
4. S.B.H., vi, p.1
p. 22
naga formulated clearly what was implicit in some
degree in both Sutras, and that Prasastapada took it
over, without the willingness to admit his
appropriation.
Nor am I convined(1) that the Vaisesika Sutra, i,
2, 3 does not teach the subjectivity of the
universal; certainly i, 2, 8 does not negate that
doctrine and the obvious meaning of samanyam visesa
iti buddhyapeksamcm requires much explain ing away,
which doubtless it receives later. It is quite
correctly pointed out elsewhere (p. 71) that the
doctrine of the real universal does not appear to be
organically related to the Vaisesika realism, and
that even Prasastapada does not connect the doctrine
of real predicables with the Vaisesika realism of the
universal. It appears to me, therefore, as probable
that this doctrine of the real universal was not held
by Kanada, and that, though adopted by his scnool, it
proved intraatable and was only in part assimilated
by his system. No work more than the Vaisesika Sutra
gives the impression that the tradition of the school
was very far from accurate.
________________________
1. Fragments from Dinnaga, p.67.