7. ‘Aboutness’ and semantic knowledge

This essay establishes the use of computational methods to study the semantics of a historical corpus compiled from the ancient Indian text of Yajnavalkya Smriti. Although the use of corpora has been extended to the study of computational semantics, in addition to grammar usage and patterns of language, they have mostly been limited to word-sense disambiguation, structural disambiguation or analysing a semantic space in terms of calculating semantic distances and determining relations between words within a corpus. This study adopts the use of computational semantics on ‘aboutness’ and ‘knowledge-free analysis’ within a limited aspect of ‘aboutness’ based on the methodology of Philips (1985). The text-only corpus of this study comprises 36,000 words of verses of the Yajnavalkya Smriti text translated by Vidyarnava (1918; 2010) from the original Sanskrit to English. In this study, ‘aboutness’ and ‘knowledge-free analysis’ aim to find the semantics of collocations and bring out bias-free information on the inheritance rights and the status of women in ancient India as described in the text. The application of the ubiquitous yet rarely applied concept of ‘aboutness’ is used to derive semantics in an unprecedented manner from an ancient historical text.This research on ‘aboutness’, which has seldom been used in computational semantics (Yablo, 2014), opens up avenues for further research on corpus analysis for extracting semantic knowledge from ancient texts to minimise knowledge bias.