INDIAN LOGIC PART 8   23 2 23   KR IRS

5 A SKELETAL THEORY OF INFERENCE



The last point may appear a bit enigmatic unless we give an account of a
skeletal theory of inference in the context of Indian logic. This skeletal
theory seems to be presupposed, consciously or unconsciously, in all the
representations of inference-patterns in India, although it became more
explicitly formulated somewhat later in the history. I shall present it as
a theory of substitution, where one property, by virtue of its logical
relation with another property, forces the substitution of the latter in
its place. That is (taking "p" to stand for the locus or paksha of the
inference," h" for the reason-property or hetu, and "s" for the
to-be-inferred property or sadhya:

(1) There is h-pervaded-by-sin pleads to:

(2) There is s in p.  Alternatively,

(3) P has h pervaded-by-s leads to:

(4) P has s.

In an historically earlier version, found in the Nyaya-sutra and other
contemporaneous texts, this was formulated as:

(5) There is h-connected-with-sin p leads to:

(6) There is s in p.

The spelling out of "connected-with. -.." in terms of "pervaded-by " was
how progress in the history of Indian logic was achieved, among other
things. We will have occasion to come back to the various ways in which the
phrase "connected-with-.. . " as well as "pervaded-by. " Were expanded. To
add flesh to this skeleton, I give an example:

(7) Sound has product-hood-connected-with-impermanence leads to:

(8) Sound has impermanence. This is an elaboration, presumably with
minimized distortion, of the following:

(9) Sound has impermanence because of its product-hood.

As we have seen in º3, (9) has generally been transformed, by almost

all modern interpreters, into a proto-Barbara: All products are
impermanent. Sound is a product. Therefore, Sound is impermanent. Or,
sometimes, it is rendered as:

Whatever is a product is impermanent.

Sound is a product. Therefore, Sound is impermanent.

This is equivalent in structure to the schema A in º3. Our "substitution"
model, however, follows more closely the actual analysis offered by the
Indian logicians. With this skeletal model before us, we can now look more
closely at the qualificand-qualifier distinction and its relation to the
subject-predicated istinction.



6 MASS TERMS



The Sanskrit logicians tried to explain the structure of the "atomic"
qualificative knowledge with a model that I have earlier called the
"property-location" model. This, in some respect, resembles what Strawson
(1959) has described as a "featuring-placing" language. In a
"feature-placing" language, Strawson notes, the subject-predicate
distinction has no place. The model sentence would be something

like" <p is here" or "there is <p here now." One advantage here is that

this language gives place-and-time-identifying expressions the status of
what are called logical subject-expressions and spatial and temporal
regions take the place of ordinary particulars. There are serious
limitations of such a language, as have been discussed by Strawson,
although he has pointed out that, in a feature-placing

language, "we can find the ultimate propositional level we are seeking
(Strawson 1959: 209)." In the above, we have seen that the Sanskrit
logicians concentrated upon a structure of knowledge episodes that is akin
to this form, for the locus, p, can be (in fact, has been) interpreted as a
patio-temporal location, where the to-be inferred property, s, is to be
located. In one formulation (see Dinnaga’s texts) the word "atra" is
explicitly used. This means "here" or even "here/now," if the understood
verbal element ("asti") is in the present tense. W. V. Quine, while he was
discussing the category of "mass terms" (a phrase coined by Otto
Jespersen),   which resemble the "feature universals" of Strawson, remarked
that these mass terms represent a primitive, archaic survival of a level of
thought, the one developmentally where the baby has not apparently learned
to identify particulars. Of course, the assumption involving
baby-psychology is open to question. However, the point is that our adult
language retains a considerable number of mass terms. Moreover, the
category of mass terms has been the "problem child" of quantification
theory, for the referents of these terms do not easily yield to
individuation and

hence we cannot quantify over them.



  The problem of fitting mass terms to quantification, or "feature words"
to sortals, is a genuine one. Quine's proposal has particularly been under
attack, for example by T. Parsons (1970), R. Sharvey (1978,1979), and Helen
Cartwright (1970). J. van Heijinoort (1974) has argued that the grammar of
the mass-term is "far from being a negligible side-show" (p. 264), for
"stuff-talk is an important part of our language, parallel to object talk"
(p. 265). It has been noted that in modern physics there has been "the true
systematization of stuff ontology"  (p. 266). It has further been noted
that abstract terms are also "much-terms," that is, the grammar of abstract
terms, such as prettiness and courage, is similar to the grammar of mass
terms. Sometimes it has been facetiously remarked that English may not have
real "count names" (Sharvey). A. N. Prior once suggested (1976: 183) that
"possibly all things are, or can be said to be made of stuff. "



    Our stuff-talk can be connected with property-talk, for there seems to
be an obvious connection between stuff-ontology and property ontology.
Suppose by "property" we mean non-universal, abstract features, or even
tropes, for example, the property of being a swimmer or the ability to
swim. This will be a non-universal, if we believe, as we probably should,
that this ability to swim varies from person to

person, for there may not be a single objective property that we can talk
about here. This will then be a perfect example of what the Nyaya call an
"imposed" property or upadhi. The use of the same expression "ability to
swim" would then be like the use of the term "water" for water found in
different spatio-temporal locations as the

river-water now is different from the water in this glass. Consider a
thought experiment. We may mentally integrate the individually located
water stuff in this world into a spatially integrated whole. "Water" then
becomes a singular term referring to this whole, which has a
spatio-temporal  spread. Then to talk about

the water in this glass we can delimit the stuff by its spatio-temporal
location. We can likewise conceptually integrate all the different
abilities to swim that are found in various agents into a "conceptual
spread," and to talk about John's ability to swim, we can delimit this
abstract feature, the ability to swim, by its spatio-temporal location, in
this case, John.





The purpose of this exercise has been to show that the problem of
individuation of a stuff like water is similar to that of an abstract
feature, or a non-universal property. Thus, consider: (1) The water in this
glass is cold, and

(2) John's ability to swim is poor (from: John is a poor swimmer).

The Sanskrit logicians would see them as equivalent to the following

analyses:

(3) Water, which is characterized by being a locate where such locatee-hood
is conditioned by a location-hood resident in the glass, has coldness (or
is cold-ness possessing).

(4) The ability to swim is characterized by being a locatee, where such
locatee-hood is conditioned by a location-hood

resident in John, has the quality of being poor. In both cases, we have to
add also that the locatee-hoodis delimited by the present time. This can be
further sharpened to take care of other well-known indexical.

 KR IRS 23 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/CAL5XZopQd93NBemKgme8YCM5OoLV-wNBSDGiC%3D8xkvOUTFxiTA%40mail.gmail.com.

Reply via email to