Predicting Declension Class from Form and Meaning

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 6682–6695 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 6682 Predicting Declension Class from Form and Meaning Adina Williams@ Tiago PimentelD Arya D. McCarthyZ Hagen BlixË Eleanor ChodroffY Ryan CotterellD,Q @Facebook AI Research DUniversity of Cambridge ZJohns Hopkins University ËNew York University YUniversity of York QETH Z¨urich adinawilliams@fb.com, tp472@cam.ac.uk, arya@jhu.edu, hagen.blix@nyu.edu, eleanor.chodroff@york.ac.uk, ryan.cotterell@inf.ethz.ch Abstract The noun lexica of many natural languages are divided into several declension classes with characteristic morphological properties. Class membership is far from deterministic, but the phonological form of a noun and its meaning can often provide imperfect clues. Here, we investigate the strength of those clues. More specifically, we operationalize “strength” as measuring how much informa- tion, in bits, we can glean about declension class from knowing the form and meaning of nouns. We know that form and mean- ing are often also indicative of grammatical gender—which, as we quantitatively verify, can itself share information with declension class—so we also control for gender. We find for two Indo-European languages (Czech and German) that form and meaning share a significant amount of information with class (and contribute additional information beyond gender). The three-way interaction between class, form, and meaning (given gender) is also significant. Our study is important for two reasons: First, we introduce a new method that provides additional quantitative support for a classic linguistic finding that form and meaning are relevant for the classification of nouns into declensions. Second, we show not only that individual declension classes vary in the strength of their clues within a language, but also that the variations between classes vary across languages. The code is pub- licly available at https://github.com/ rycolab/declension-mi. 1 Introduction To an English speaker learning German, it may come as a surprise that one cannot necessarily pre- dict the plural form of a noun from its singular. This is because pluralizing nouns in English is rel- atively simple: Usually we merely add an -s to the end (e.g., cat �→ cats). Of course, not all En- glish nouns follow such a simple rule (e.g., child �→ children, sheep �→ sheep, etc.), but those that do + Figure 1: The conditional entropies (H) and mutual in- formation quantities (MI) of form (W), meaning (V ), and declension class (C), given gender (G) in German and Czech. not are few in number. Compared to English, Ger- man has comparatively many common morpholog- ical rules for inflecting nouns. For example, some plurals are formed by adding a suffix to the sin- gular: Insekt ‘insect’ �→ Insekt-en, Hund ‘dog’ �→ Hund-e, Radio ‘radio’ �→ Radio-s. For others, the plural is formed by changing a stem vowel:1 Mutter ‘mother’ �→ M¨utter, or Nagel ‘nail’ �→ N¨agel. Some others form plurals with both suffixation and vowel change: Haus ‘house’ �→ H¨aus-er and Koch ‘chef’ �→ K¨och-e. Still others, like Esel ‘donkey’, have the same form in plural and singular. The problem only worsens when we consider other inflectional …

Content truncated to 500 words