Competence in lexical semantics

Proceedings of the Fourth Joint Conference on Lexical and Computational Semantics (*SEM 2015), pages 165–175, Denver, Colorado, June 4–5, 2015. Competence in lexical semantics Andr´as Kornai Institute for Computer Science Hungarian Academy of Sciences Kende u. 13-17 1111 Budapest, Hungary andras@kornai.com Judit ´Acs Dept of Automation and Applied Informatics, BUTE Magyar Tud´osok krt. 2 1117 Budapest, Hungary judit@aut.bme.hu M´arton Makrai Institute for Linguistics Hungarian Academy of Sciences Bencz´ur u. 33 1068 Budapest, Hungary makrai@nytud.hu D´avid Nemeskey Faculty of Informatics E¨otv¨os Lor´and University P´azm´any P´eter s´et´any 1/C 1117 Budapest, Hungary nemeskeyd@gmail.com Katalin Pajkossy Department of Algebra BUTE Egry J. u. 1 1111 Budapest, Hungary pajkossy@math.bme.hu G´abor Recski Institute for Linguistics Hungarian Academy of Sciences Bencz´ur u. 33 1068 Budapest, Hungary recski@mokk.bme.hu Abstract We investigate from the competence stand- point two recent models of lexical semantics, algebraic conceptual representations and con- tinuous vector models. Characterizing what it means for a speaker to be competent in lexical semantics remains perhaps the most significant stumbling block in reconciling the two main threads of semantics, Chomsky’s cogni- tivism and Montague’s formalism. As Partee (1979) already notes (see also Partee 2013), linguists as- sume that people know their language and that their brain is finite, while Montague assumed that words are characterized by intensions, formal objects that require an infinite amount of information to specify. In this paper we investigate two recent models of lexical semantics that rely exclusively on finite in- formation objects: algebraic conceptual representa- tions (ACR) (Wierzbicka, 1985; Kornai, 2010; Gor- don et al., 2011), and continuous vector space (CVS) models which assign to each word a point in finite- dimensional Euclidean space (Bengio et al., 2003; Turian et al., 2010; Pennington et al., 2014). After a brief introduction to the philosophical background of these and similar models, we address the hard questions of competence, starting with learnability in Section 2; the ability of finite networks or vectors to replicate traditional notions of lexical relatedness such as synonymy, antonymy, ambiguity, polysemy, etc. in Section 3; the interface to compositional se- mantics in Section 4; and language-specificity and universality in Section 5. Our survey of the litera- ture is far from exhaustive: both ACR and CVS have deep roots, with significant precursors going back at least to Quillian (1968) and Osgood et al. (1975) re- spectively, but we put the emphasis on the compu- tational experiments we ran (source code and lexica available at github.com/kornai/4lang). 1 Background In the eyes of many, Quine (1951) has demolished the traditional analytic/synthetic distinction, relegat- ing nearly all pre-Fregean accounts of word mean- ing from Aristotle to Locke to the dustbin of his- tory. The opposing view, articulated clearly in Grice and Strawson (1956), is based on the empirical ob- servation that people make the call rather uniformly over novel examples, an argument whose import is evident from the (at the time, still nascent) cogni- tive perspective. Today, we may agree with Putnam (1976): ‘Bachelor’ may be synonymous with ‘un- married man’ but that cuts no philosophic ice. …

Content truncated to 500 words