Explicit codification in Pāṇini and generative grammar

Generative grammar and Pāṇini’s grammar – so-called Aṣṭādhyāyī, dating from ca. 4th century B.C. – differ in scope and goals: while generative grammarians search for the limits of variation in natural languages to explain what makes language acquisition possible in the first place, Pāṇini provides a distributional and variationist account of old Indo-Aryan, with the twofold goal of faithfully recording that language and of regularizing its usage in the relevant sacred texts. Despite these important differences, the codification of linguistic phenomena operated by generative grammarians bears some resemblance to the one operated by Pāṇini. Thus, in this study I analyze the codification of long-distance agreement in generative grammar and the codification of compounding in the Aṣṭādhyāyī. I show that both instances of codification are explicit, in the technical sense that they specify (rigorously formulate) the rules – filters and operations – that license all well-formed tokens of long-distance agreement and compounding while simultaneously excluding all ill-formed ones. Finally, I argue that the explicit character of these instances of codification is a major part of the reason why they are still likewise considered as successful in the contemporary scientific community.