11. Carrying a direction (anuvŗtti) over a span of sutras: anuvrtti

The span of sutras 1.1.11 to 1.1.19 (see posts 7, 8, 9) is an illustration of the way a new concept or idea or proposition is introduced, and made implicit in a series of successive sutras, sometimes in a nested manner. Thus, 1.1.11 (see post 7) introduced a species called pragŗhyam (unaltered in sandhi, held in check), by presenting one category of it: dual words ending in long vowels īT ūT eT. The next sutra, 1.1.12 (see post 8), presents another variety of it, expressed concisely by the phrase adaso māt, but we have to supply the word  pragŗhyam to complete the idea. This is a carrying forward of a concept from the starting sutra of a chain, to the subsequent sutras which make sense only when the carried-forward word or phrase is supplied in our mind. This carry-forward is termed anuvŗtti.

I like to make some worldly sense out of these technical terms in Sanskrit grammar, because that may fix the meaning of these mysterious words better in our mind. After all, the grammarians of old must have chosen these technical names with some idea of the nature of the function or character of the species described. For denoting the function of carrying forward of a term from earlier to following sutras, why didn’t they choose some other, descriptive word like ‘cup-bearer’ or ‘chariot-driver’ or ‘repeating performance’ or something like that? The word anuvŗtti must, after all, mean something connected to the concept.

The word vŗtti  is explained in the vocabulary given at the back of Michael Coulson’s Teach Yourself Sanskrit in the following terms:

vŗtti f. behaviour, conduct; (grammar) synthetic expression (i.e. by cpds.). 

A host of words are related to the root

vŗt (I vartate) proceed, currently exist, abide, happen; (of promises, etc.) be entered upon. 

Guna (first augmentation or lengthening of the ŗ short vowel) gives vart, which leads to the verb forms of ‘to be’, as well as the sense of ‘turning’ (as in English revert). The famous text on horse-whispering in the Mitanni court used obviously Indo-Aryan or Sanskrit terms to denote the circuits of the training ground by the animals: eka-vartana, panca-vartana, and so on. The second augmentation or vŗddhi gives vārt, from which we have vārtta, news, a reflection of what exists, vartate.

The particle anu- denotes movement towards, or following, or a partiality to, something; it gives a benign and meliorative colour to the following word. So anu-vŗtti would denote something like ‘following (some preceding element) in character’, and in fact Coulson’s vocabulary does have a specific entry for the verb:

anuvŗt anu+vŗt (I anuvartate) go after, attend upon   

and anuvŗtti would be the character or nature of following upon (a preceding element). Indeed my supplying the last phrase (within brackets) itself is a faint echo and an illustration of anuvŗtti! Rama Nath Sharma gives it the label recurrence (chapter 4 of volume I of his treatise, see the Resources page on tabs above). A last thought on the meaning of the word anuvŗtti would be to think of it as an ‘enveloping’ statement (the echo of the sounds n and v may help relate the Sanskrit term and the meaning in English!).

The starting sutra or rule 1.1.11 is termed the adhikāra-sūtra (directing sutra). We carry forward the terms  īT ūT eT pragŗhyam as an anuvŗtti to the second sutra in this block or domain, 1.1.12 adaso māt, supplying the complement pragŗhyam, thus īT ūT eT  adaso māt pragŗhyam (but not the term dvivacanam or dual number from the head sutra, which we may sense from the fact that forms of adas with the stem ending –m- are not necessarily dual). This implies that we may have to choose the exact words to carry forward in anuvŗtti from the sense of the successive sutras and how the ideas are being built up.  This is also noticeable in sutra 1.1.13, śe (see post 8), where carry-forward (anuvŗtti) supplies only the term pragŗhyam. Similar is the case in 1.1.14 nipāta ekāc anāŋ, where we supply only the term pragŗhyam to complete the thought: there is no intent to carry forward īT ūT eT (long vowel ending) or dvivacanam (dual number) as conditions for a particle (nipāta) to be pragŗhyam (unaltered in sandhi), as long as the particle is not a meaningful āŋ (see Post 9!).

A related phenomenon is the nesting of rules by successive anuvŗtti in what computer programmers may call an indented arrangement. In sutra 1.1.15 is provided the single word ot, to which we have to add, by  anuvŗtti, the terms nipāta (particle) from 1.1.14, as well as of course pragŗhyam from the first in the series or domain of the head rule or adhikāra-sūtra (directing sutra). Sutra 1.1.16 carries ot from 1.1.15 and the original pragŗhyam, but not nipāta from 1.1.14. Sutra 1.1.17 and 1.1.19 carry the terms śakalyasya anārşe from 1.1.16, meaning ‘according to Sakalya in non-Vedic’ (interesting word, applying vrddhi to ŗşi, sage, with a negative prefix, an-), but not nipāta from 1.1.14 or ot from 1.1.15.   Sutra 1.1.19, the last in this domain, carries the term pragŗhyam from the head of the series, but not śakalyasya anārşe from 1.1.16, or  ot from 1.1.15, or nipāta from 1.1.14. There does not seem to be any clear instruction about this, and thus it appears that a certain amount of prior knowledge may have to brought into play to decide which terms are to be carried forward by anuvŗtti for each given sutra in a domain or sub-domain.