MD Srinivas: the indian tradition in science and technology (excerpt)

jul 19th

interesting analysis by prof srinivas. relevant to the discussion we had on this blog about indian science vs. euro-science vs. faith, as well as the importance of language.

this adds to my contention that panini's grammar was the greatest invention of a single human mind in the history of mankind. there is much more to language than just communication. it is science, and nowhere has language evolved more than in india. the very system of the generation of consonant phoneme -- based on adding 5 variants such sibilant, epiglottal etc. to the basic ka, cha, ta, tha, pa sounds -- is itself a remarkable invention. compare indian alphabets -- totally systematic -- to say, english -- totally random. (see also the remarkable conjecture that mendeleev's periodic table ideas were an insight based on his familiarity with the recurring patterns of the letters in the devanagari -- and most other indian -- scripts. no wonder he talked about eka-silicon, dva-germanium, etc, as he knew samskrtam).

the 'reductionist' approach of euro-science, and the vanity that everything can be understood by reducing it down to smaller and smaller components, is now under challenge by the scholars asserting the 'emergent' principle of the whole being greater than the sum of the parts.

the indian approach -- it may be characterized as categorization and deductive/inductive reasoning from first principles/observation -- does not fall into the trap of mechanistic thinking.

Any study of the Indian tradition of science has to start with
linguistics. This is true not only because linguistics is the earliest
of Indian sciences to have been rigorougly systematized but also
because this systematization became the paradigm example for all other
sciences.

Like all sciences and arts in India, linguistics finds its first
expression in the Vedas. For most of the Indian sciences, the elements
of study and the categories of analysis were established in the
Vaidika period, and the basic data was collected and preliminary
systematization achieved simultaneously. Thus for the science of
linguistics, we find, in the siksha and pratisakhya texts associated
with the various Vedas, a complete and settled list of phonemes
appropriately classified into vowels, semi-vowels, sibilants and the
five groups of five consonants, all arranged according to the place of
articulation that moves systematically from the throat to the lips.

Phonetics and phonology are, therefore, taken for granted by all
post-Vaidika authorities on etymology (nirukta) and grammar
(vyakarana), including Yaska and Panini. In the pratisakhya literature
we also find the morpho-phonemic (sandhi) rules and much of the
methodology basic to the later grammatical literature.

Indian linguistics finds its rigorous systematization in Panini's
Ashtadhyayi. The date of this text, like that of much of the early
Indian literature, is yet to be settled with certainty. But it is not
later than 500 BC. In Ashtadhyayi, Panini achieves a complete
characterization of the Sanskrit language as spoken at his time, and
also specifies the way it deviated from the Sanskrit of the Vedas.
Using the sutras of Panini and a list of the root words of the
Sanskrit language (dhatupatha), it is possible to generate all
possible valid utterances in Sanskrit. This is of course the main
thrust of the generative grammars of today that seek to achieve a
grammatical description of language through a formalized set of
derivational strings. In fact, till the western scholars began
studying generative grammars in the recent past, they failed to
understand the significance of Ashtadhyayi: till then Paninian sutras
for them were merely artificial and abstruse formulations with little
content.

Patanjali (prior to the first century BC) in his elaborate commentary
on Ashtadhyayi, Mahabhashya, explains the rationale for the Paninian
exercise. According to Mahabhashya, the purpose of grammar is to give
an exposition of all valid utterances. An obvious way to do this is to
enumerate all valid utterances individually. This is how the celestial
teacher Brihaspati would have taught the science of language to the
celestial student, Indra. However for ordinary mortals, not having
access to celestial intelligence and time, such complete enumeration
is of little use. Therefore, it is necessary to lay down widely
applicable general rules (Utsarga sutras) so that with a comparatively
small effort men can learn larger and larger collections of valid
utterances. What fails to fit in this set of general rules should,
according to the Mahabhashya, then be encompassed in exceptional rules
(apavada sutras), and so on. In thus characterizing grammar, Patanjali
expounds perhaps the most essential feature of the Indian scientific
effort.

Science in India starts with the assumption that truth resides in the
real world with all its diversity and complexity. For the linguist,
what is ultimately true is the language as spoken by the people in all
their diverse expressions. As Patanjali emphasizes, valid utterances
are not manufactured by the linguist but are already established in
the practice in the world. One does not go to a linguist asking for
valid utterances, the way one goes to a potter asking for pots.

Linguists make generalizations about the language spoken. These
generalizations are not the truth behind or above the reality of the
spoken language. These are not idealizations according to which
reality is to be tailored. On the other hand what is true is what is
actually spoken in the real workd, and some part of the truth always
escapes our idealization of it. There are always exceptions. It is the
business of the scientist to formulate these generalizations, but also
at the same time to be always attuned to the reality, to always be
conscious of the exceptional nature of each specific instance. This
attitude, as we shall have occasion to see, permeates all Indian
science and makes it an exercise quite different from the scientific
enterprise of the West.

In Lingusitcs, after the period of Mahabhashya, grammarians tried to
provide continuous refinements and simplifications of Panini. Several
Sanskrit grammar texts were written. One of them, Siddhanta Kaumudi
(c. 1600) became eminently successful, because of its simplicity.
These attempts continued till the 19th century. Another form of study
that became popular among the grammarians was what may be called
philosophical semantics, where grammarians tried to fix and
characterize the meaning of an utterance by analyzing it into its
basic grammatical components. This, of course, is the major
application for which grammar is intended in the first place.

Grammars for other Indian languages were written, using Paninian
framework as the basis. These grammars were not fuly formalized in the
sense of Panini. Instead, they started with the Paninian apparatus and
specified the transfer rules from Sanskrit and the specific
morpho-phonemic rules (sandhi rules) for the language under
consideration. Such grammars for various Prakrit languages of the
North and also the South Indian languages continued to be written
until the 18th century. In the 16th century, Krishnadasa even wrote a
grammar for the Persian language, Parasi Prakasha, styled on the
grammars of the Prakrit language.

[From MD Srinivas, 2005, The Indian tradition in science and
technology: an overview, in: P. Parameswaran, ed., National Resurgence
in India, Thiruvananthapuram, Bharatheeya Vichara Kendram, pp. 52-62.]