________________
Kārikās 2.40 and 444 occur both in the context of passages in support of the view that the sentence meaning, and by implication also the sentence, is one and indivisible, although secondarily a division in words and word meanings may be adopted. The unity and indivisibility of the sentence is demonstrated by the fact that, theoretically, a whole sentence meaning could be expressed by a single phoneme. The unity of meaning of a sentence becomes this way clear in one small (prosodial) unit (mātrā). 3.1 Because kārikā 40 speaks of the adoption of the relation of identification, it seems that the author asks us to imagine a situation in which the meaning of a whole sentence is by convention equated with a single phoneme. Even if we are favourably disposed towards the view that the sentence is one and indivisible, this argument given in support of it seems a bit vague and far-fetched. Punyarāja's commentary and the later interpreters are not of much help, because they only explain the general implication of the kārikās in the wider context. It is only the Vștti, the most ancient commentary available, which seems to make the example used in the kärikäs more concrete. 3.2 The Vrtti refers to a problem in the technical understanding of Pāņini's grammar which received some attention in the Vārttika's and in Patañjali's Mahābhāsya, and which was also important in Bhartshari's Mahābhāşyadipikā. The problem is connected with sūtra P 1.1.45, which defines the technical term samprasārana as follows: ig yanaḥ samprasāranam. This means:
When i, u, and in all possible prosodic varieties) replace y, r, I, v, this is
called samprasarana or vocalization'. In the Mahābhāsya (MBh) the question is asked, whether saprasāraṇa should be considered the technical term for the phonemes ik (i, u, , ! and their prosodic variants22) when they replace yan (y, r, 1, v); or whether it is rather the technical term for the sentence: ig yanah ‘i, u, ?, !. replace y, r, 1, V'. In the discussion in the MBh it is pointed out that in some technical contexts the term is required as denotative of the phonemes which replace the other phonemes, in others as denotative of the sentence that some phonemes replace others. The sūtra that is mentioned as specifically requiring the sentence interpretation of
22 The phrase "and their prosodic variants' is to be understood wherever I refer to the phonemes indicated by ik.
parade by the
Jan E.M. Houben, Pune, March 1994 (pre-final version)
Page 12
