Three interrelated technical terms are used here. The "case under
consideration" is called a paksha, the "subject-locus." The "similar
case" is called a sapaksha, the "homologue." The "dissimilar case" is
called a vipaksha, the "heterologous." These three concepts are also
defined by the theory. The context is that of inferring a property A
(The signified in our new vocabulary) from the property B (the sign) in
a location S. Here the S is the paksha, the subject-locus. The sapaksha is
one which already possesses A, and is known to do so. And the
vipaksha is one which does not possess A. The "similarity" between the
paksha and the sapaksha is variously explained. One explanation is that
they would share tentatively the signified A by sharing the sign B. An
example would make it clear. Smoke is a sign of fire on a hill, because
it is present on that hill, and it is also present in a kitchen which is a
locus of fire, and it is absent from any non-locus of fire.
The third condition is easily explained. The sign must not be
present where the signified is not present. For otherwise, as we have
already noted, the sign will be deviating, and would be a "pseudo sign."
Why the second condition? Did Dinnaga (wherever in the earlier text you saw
as “diinamaga” please read as “DINNAGA” A BUDDIST BRAHMIN) overshoot his
mark?
Is not the second condition redundant (for the first and the third seem
to be sufficient to guarantee adequacy)? These questions were raised
in the tradition by both the Naiyayikas like Uddyotakara (circa
550-625 AD), and the Buddhists like Dharmakrrti (circa 600-660
AD).
KR: { AND MANY VAISHNAVAITES AND BRITISH BROUGHT THE AGE OF ADI SHANKARA
AT 750 AD, SINCE THE DHARMAKIRTHI IS SUPPOSED TO HAVE DEBATED WITH ADI
SHANKARA, A MISTAKEN OR WANTONLY TWISTED IDENTITY; EVEN THEN IF DHARMA
KIRTHI WAS IN 660 AD 60 YEARS OLD BY THEN AND DEAD, SHANKARA WAS IN 720 AD
ONLY ; SO A DEAD MAN COULD NOT HAVE ARGUED WITH ADI SHANKARA; AND SIKKIM
MONOSTRY RECORDS SHOW THAT DHARMA KIRTHI ARGUED WITH FEW COMMON NAME
SANKARAS WHO DIVED INTO GANGES AFTER THE DEFEAT IN THE HANDS OF DHARMA
KIRTI, AND ADI SHANKARA NEVER DIVED IN GANGES AT ALL}
Some, such as Dharmakrrti, maintained that it was slightly
repetitious but not exactly redundant. The second condition states
positively what the third, for the sake of emphasis, states negatively.
The second is here rephrased as: the sign should be present in all
sapaksha. The contraposed version can then be formulated with a
little ingenuity as: the sign should be absent from all vipakshas. For
sapaksha and vipaksha, along with the paksha, exhaust the universe of
discourse.
Other interpreters try to find additional justification for the second
condition to argue against the "redundancy" charge. Logically
speaking, it seems that the second condition is redundant, but
epistemologically speaking, a case of the co-presence of A and B may
be needed to suggest the possibility, at least, that one may be the sign
for the other. Perhaps Dinnaga's concern here was epistemological.
Dinnaga's Wheel of Reason/Sign
When a sign is identified, there are three possibilities. The sign may be
present in all, some, or none of the sapakshas. Likewise, it may be
present in all, some or none of the vipakshas. To identify a sign, we
have to assume that it is present in the paksha however; that is, the first
condition is already satisfied. Combining these, Dinnaga constructed
his "wheel of reason" with nine distinct possibilities, which may be
tabulated in Figure 1.1.
Of these nine possibilities, Dinnaga asserted that only two are
illustrative of sound inference for only they meet all the three
conditions. They are Numbers 2 and 8. Notice that either (- vipaksha
and + sapaksha), or (- vipaksha and sapaksha) would fulfil the
required conditions. Dinnaga is insistent that at least one sapak?ash
must have the positive sing. Number 5 is not a case of sound
inference; this sign is a pseudo-sign. For although it satisfies the two
conditions 1 and 3 above, it does not satisfy condition 2. So one can
argue that as far as Dinnaga was concerned all three were necessary
conditions. The second row does not satisfy condition 2 and hence
none of Numbers 4, 5, and 6 are logical signs; they are pseudo-signs.
Numbers 4 and 6 are called "contradictory" pseudo-signs - an
improvement upon the old Nyayasutra definition of contradictory.
The middle one, Number 5, is called "uniquely deviating" (asadharana),
perhaps for the reason that this sign becomes a unique sign of the
paksha itself, and is not found anywhere else. In Dinnaga's system, this
sign cannot be a sign for anything else, it can only point to itself
reflexively or to its own locus. Numbers 1, 3, 7, and 9 are also
pseudo-signs. They are called the "deviating" signs, for in each case
the sign occurs in some vipaksha or other, although each fulfils the
second condition. This shows that at least in Dinnaga's own view, the
second condition (when it is combined with the first) gives only a
necessary condition for being an adequate sign, not a sufficient one. In
other words, Dinnaga intended all three conditions jointly to formulate a
sufficient condition.
Development of the Wheel by Uddyotakara Dinnaga's system of nine
reason-types or sign-types was criticized by Uddyotakara, the Naiyayikas,
who argued that it was incomplete.
Dinnaga did not consider at least two further alternatives: (a) a
situation-type where there is no sapaksha, and (b) a situation-type
where there is no vipaksha. The sign's absence from all sapakshas (or all
vipakshas) should be distinguished from these two situations. Let us
use "0" for the situation-type which lacks any sapaksha, or vipaksha,
and "-" for the situation-type where the sign is present in no sapaksha
or vipaksha (as before). Hence combining the four possibilities +
sapaksha, ± sapaksha, - sapaksha, 0 sapaksha (no sapaksha) with the other
four (+, ±, -, 0) vipaksha, we get sixteen portions in our wheel of
reason, and the new wheel contains more sound inferences, that is,
adequate signs. For example, this is nameable, because this is knowable.
Here "knowability" is the sign, which is adequate and logical for
showing the name ability of an entity, for (in the Nyaya system)
whatever is knowable is also nameable (that is, expressible in
language). Now we cannot have a heterologous or vipaksha here, for
(Again, according to the Nyaya system) there is nothing that cannot be
named (or expressed in language). Within the Buddhist system,
another example of the same argument-type would be:
This is impermanent because it is a product. For Buddhists everything is
impermanent and a product. Later Naiyayikas called this type of sign
kevalanvayin, the universal positive- sign; that is, it is a characteristic
of every entity.
Uddyotakara captured another type of adequate reason or logical
sign, but he formulated the example of this reasoning (or inference)
negatively, that is, in terms of a counterfactual. This was done
probably to avoid a doctrinal quandary of the Nyaya school (to which
he belonged) in which the explanation of analytic judgements or a
prior knowledge always presents a problem. His typical example
was: The living body cannot be without a soul, for if it where it would
have been without life. This is the generalized inference called "universal
negative" -
kevalavyatirekin - in the tradition. The subject 5 which has a unique
property B cannot be without A, for then it would have been without
B. Since B is a unique property of 5, and since the presence of A and B
mutually imply each other, there is no sapaksha. But it is a correct
inference. Bhasaravajna (circa 950 AD) did not like the rather
roundabout way of formulating the inference-type. He said: The living body
has a soul, for it has life.
But this would verge on unorthodoxy in Nyaya, for (a) the statement
of the thesis includes the sign already, and (b) there seems to be a
necessary connection between having life and having a soul. The later
Nyaya went back to the negative formulation but got rid of the reflex
of the counterfactual that Uddyotakara had. If A and B are two properties
mutually implying each other such that B can be the definiens (lakshana)
and the class of those possessing A can be the definiendum, then the
following inference is correct:
The subject 5 differs from those that are without A, for it has B (and A is
defined in terms of B.) This seems to be equivalent to: S has A, for it
has B. The verbal statement "5 has A because it has B," however, does not
fully expose the structure of this type of inference. For one thing, in
this version it becomes indistinguishable from any other type of correct
inference discussed before. In fact, the special feature of this type of
inference is that the inferable property A is uniquely present in
S alone, and nowhere else, and hence our knowledge of the concomitance or
pervasion between A and B cannot be derived from an example (where their
co-presence will be instantiated) which will be a different case from the
S, the case under consideration. In fact, S here is a generic term and it
will be proper to say: all S have A, for they have B, and a supporting
example will have to be an S, that is, an
instance of S. To avoid this anomaly, a negative example is cited to cover
these cases. Thus we can say, a non-S is a case where neither A nor B are
present.T his will allow one to infer, for example, absence of B from
absence of A and also (since A and B are co-present in all cases) absence
of A from absence of B. But the evidence here is B.
Hence by seeing absence of B in all S we can infer absence of A. Such a
roundabout formulation was dictated by the peculiar nature of the
Dinnaga-Uddyotakara theory of inference.
Let us try to explain. In this theory, what legitimizes the inference of A
from the sign B is the knowledge that B is a logical sign of A. To have
that knowledge, we must have another item of knowledge, that B has
concomitance, an invariable connection, with A. The second item of
knowledge must be derived empirically, from an example
where it is certain that A as well as B is present. Without such an
example, we would not recognize B to be a logical sign of A. This
limitation precluded the possibility of inferring A from B, where the case
is such that all that have A are included in paksha, the subject locus of
the inference. The convention is that the said example cannot be chosen
from the members of the paksha, that is, of the set of S. Hence the
difficulty.
Uddyotakara saw this problem and extended the scope of the theory by saying
that in these cases, a negative example, a non-S having neither A nor B,
and absence of any counter-example (the sign's absence from all vipakshas),
will be enough to legitimize the inference. Udayana (circa 975-1050 AD)
later on defended this type
of inference as legitimate. For, he said, if we do not admit such
inferences as valid, our search for a defining property of some concepts
could not be justified. Suppose we wish to define cowl-hood: what is the
unique property of a cow? Now, suppose having a dewlap is a unique property
of a cow; it exists in all and only cows. What is the purpose of such a
"definition," if we can call it a definition (lakshana)? It is that we can
differentiate all cows from non-cows. How? We do it by means of the
following inference: cows are distinct from non-cows, for cows have
dewlaps. Of course, the statement "cows are distinct from non-cows" is
equivalent to the statement "cows are cows," but when it is put negatively,
the purpose of such inference becomes clearer.
KR Part 4 18 2 23
--
You received this message because you are subscribed to the Google Groups
"Thatha_Patty" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/thatha_patty/CAL5XZooStLTG1Ap6XQo2stb0Bo-C33JrWLT4TcKKL_3N5-hyPw%40mail.gmail.com.
