To quote this page: AAE Van der Geer 1998. The Bhasa Problem. A statistical research into its solution. PhD Rijksuniversiteit Leiden, the Netherlands

Back << >> Further

Back to my PhD

The Sanskrit Case

As to Sanskrit poetry, however, scholars seem to be much less interested in such studies, except for, for instance, Pollock (1977) and Klein (1991a-b, 1994). Much has been, and still is, written on several aspects of classical poetry, but there is a remarkable lack of knowledge and literature on the connection between metre on the one hand and syntax on the other. Such a relation can be expected to be present in Sanskrit lyrics as well.

A very general rule on the relationship between metre and syntax in Sanskrit is that the end of a verse regularly coincides with the end of a word, because each verse in a stanza is independent of the rest in structure (Macdonell 1958:437). Another rule, or rather observation is that in the Rgvedic anushtubhs 'External enjambment', or the continuing of a sentence beyond the end of a strophe, does not occur. If these poets needed to combine two stanzas, they evidently preferred not to mix up their particular syntactic structure (Gonda 1975f:324).

These kinds of rules are too general and non-specific to learn something about the real relationship between the two linguistic entities. The easiest way to get insight into the problem is gathering a large quantity of verses, and determining in each verse the relation between the two. In this way ad hoc rules are found, or more precisely general observations. When this method is applied to the Veda, a striking coincidence of the syntactic unit with the metrical unit is found. The caesurae of the odd pâdas often coincide with the sentence end, or function as a colon stop. Even in cases where the stanza has developed into a bipartite structure, they remain real caesuras, that is to say, they make a syntactic border visible (Gonda 1975e:290). The simplest stanzas are of course those in which each line consists of a whole, independent sentence. It could be expected that the larger number of these simple structures should occur in the oldest text. This, however, seems not to be the case. Gonda (1975f:324) finds in the Rgvedic anushtubh stanzas only about 14 instances of stanzas comprising four sentences which are equally distributed over the four pâdas. Only in the 10th mandala he finds 25 instances. Many more stanzas are found comprising two instead of four sentences, whereby the pâdas ab, as well as the pâdas cd, consist of a single, syntactically complete sentence. There are 166 such instances in total, comprising 46 cases in the 10th mandala (Gonda 1975f:335).

It is important to keep in mind that the syntactic units coinciding with metrical units should not be mistaken for independent syntactic units. For example, yena prânanti virudhah (AthV. 1.32.1d) "by which the plants breathe'' is a complete syntactic unit, but is not independent of the other part of the half-stanza to which it belongs (Gonda 1975d:262).