Towards a Semi-Automatic Detection of Reflexive and Reciprocal Constructions and Their Representation in a Valency Lexicon

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 3136–3144 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 3136 Towards a Semi-Automatic Detection of Reflexive and Reciprocal Constructions and Their Representation in a Valency Lexicon Václava Kettnerová, Markéta Lopatková, Anna Vernerová, Petra Baranˇcíková Charles University, Faculty of Mathematics and Physics, Institute of Formal and Applied Linguistics Malostranské námˇestí 25, 118 00 Prague, Czech Republic {kettnerova, lopatkova, vernerova, barancikova}@ufal.mff.cuni.cz Abstract Valency lexicons usually describe valency behavior of verbs in non-reflexive and non-reciprocal constructions. However, reflexive and reciprocal constructions are common morphosyntactic forms of verbs. Both of these constructions are characterized by regular changes in morphosyntactic properties of verbs, thus they can be described by grammatical rules. On the other hand, the possibility to create reflexive and/or reciprocal constructions cannot be trivially derived from the morphosyntactic structure of verbs as it is conditioned by their semantic properties as well. A large-coverage valency lexicon allowing for rule based generation of all well formed verb constructions should thus integrate the information on reflexivity and reciprocity. In this paper, we propose a semi-automatic procedure, based on grammatical constraints on reflexivity and reciprocity, detecting those verbs that form reflexive and reciprocal constructions in corpus data. However, exploitation of corpus data for this purpose is complicated due to the diverse functions of reflexive markers crossing the domain of reflexivity and reciprocity. The list of verbs identified by the previous procedure is thus further used in an automatic experiment, applying word embeddings for detecting semantically similar verbs. These candidate verbs have been manually verified and annotation of their reflexive and reciprocal constructions has been integrated into the valency lexicon of Czech verbs VALLEX. Keywords: valency lexicon, reflexivity, reciprocity 1. Introduction Reflexivity and reciprocity attract much attention in current theoretical linguistics, particularly from a typological per- spective, see e.g. (Nedjalkov, 2007; König and Gast, 2008; Evans et al., 2007; Frajzyngier and Walker, 2000a; Frajzyn- gier and Walker, 2000b). These language phenomena are typically encoded by reflexive markers (henceforth RMs). However, cross-linguistic studies evidence that RMs typi- cally serve various functions in a language, crossing bound- aries of reflexivity and reciprocity, see esp. (Geniušien˙e, 1987) and (Kemmer, 1993). In Czech, which is the focus of this work, RMs are involved in various constructions. In addition to reflexive (1) and re- ciprocal constructions (2), RMs encode impersonal passive constructions (3), middle constructions (4), anticausative constructions (5) or constructions with inherently reflexive verbs (6).1 (1) Petr Peter myslel thought jen only na of sebe. RM ‘Peter thought only of himself.’ (2) Rodiˇce parents si RM (vzájemnˇe) (mutually) pomáhali. helped ‘Parents helped each other.’ (3) Z from ˇCeské republiky Czech republic se RM elektˇrina electricity vyváží. exports ‘Electricity is exported from the Czech republic.’ (4) Špatnˇe badly se RM mi me spalo. slept ‘I did not sleep well.’ 1Similar functions are exhibited by RMs in other Slavic and Romance languages as well (Medová, 2009). (5) Okno window se RM rozbilo. broke ‘The window broke.’ …

Content truncated to 500 words