Characterizing Idioms: Conventionality and Contingency

TL;DR

This paper investigates the properties of idioms, focusing on their non-canonical word meanings and context-dependent meanings, using language models to measure and analyze these features.

Contribution

It introduces two measures for idiom properties and applies them to BERT and XLNet, showing these properties are independent and questioning the need for special idiom processing mechanisms.

Findings

Idioms exhibit non-canonical word meanings and context-dependent meanings.

The two properties are uncorrelated across idioms.

Standard language models capture idiom properties without special mechanisms.

Abstract

Idioms are unlike most phrases in two important ways. First, the words in an idiom have non-canonical meanings. Second, the non-canonical meanings of words in an idiom are contingent on the presence of other words in the idiom. Linguistic theories differ on whether these properties depend on one another, as well as whether special theoretical machinery is needed to accommodate idioms. We define two measures that correspond to the properties above, and we implement them using BERT (Devlin et al., 2019) and XLNet(Yang et al., 2019). We show that idioms fall at the expected intersection of the two dimensions, but that the dimensions themselves are not correlated. Our results suggest that special machinery to handle idioms may not be warranted.