Non-canonical Inflection: Data, Formalisation and Complexity Measures

Abstract

Abstract. Non-canonical inflection (suppletion, deponency, heteroclisis. . . ) is extensively studied in theoretical approaches to morphology. However, these studies often lack practical implementations associated with large-scale lexica. Yet these are precisely the requirements for objective comparative studies on the complexity of morphological descriptions. We show how a model of inflectional morphology which can represent many non-canonical phenomena [67], as well as a formalisation and an implementation thereof can be used to evaluate the complexity of competing morphological descriptions. After illustrating the properties of the model with data about French, Latin, Italian, Persian and Sorani Kurdish verbs and about noun classes from Croatian and Slovak we expose experiments conducted on the complexity of four competing descriptions of French verbal inflection. The complexity is evaluated using the information-theoretic concept of description length. We show that the new concepts introduced in the model by [67] enable reducing the complexity of morphological descriptions w.r.t. both traditional or more recent models.

Implementing a Formal Model of Inflectional Morphology

“…We introduce a means to easily implement formal analyses in a typologically sound framework that benefits from the data processing power available through computational approaches alone. 4 On an experiment carried out on modelling Maltese verbal inflection, we show the benefit for formal approaches to rely on computational approaches.…”

“…Moreover, the extent of the non-canonical phenomenon can be quantified with specific non-canonicity measures developed within the framework [3]. A preliminary version of the framework had been introduced in [2] and used in experiments described in [4]. The version described here is the one presented in [3].…”

“…As already mentioned, a partial Alexina implementation of a preliminary version of parsli was used in [4]. However, the work presented in this section goes far beyond, for at least two reasons.…”

“…1 We briefly describe the last version of parsli, on which Alexina parsli relies, in Section 2. In particular, we point out the major innovations with respect to earlier versions of parsli [2,4]; an extension of the Alexina lexical framework [5] used in the field of Natural Language Processing (NLP) for modeling lexical information and developing lexical resources. The morphological layer of the (original) Alexina formalism is sketched in Section 3.…”

Gender typology

“…Only then do we investigate how this space is populated with real instances. There has been interesting work in other areas of linguistics too, from phonology (Hyman 2009(Hyman , 2011 to formal semantics (Fortin 2011) and computational modeling (Sagot and Walther 2011). Given that they have to match up to a logically determined standard, they are unlikely to be frequent.…”