Skip to main content

Form-frequency correspondences in grammar

Periodic Reporting for period 3 - FormGram (Form-frequency correspondences in grammar)

Reporting period: 2018-11-01 to 2020-04-30

This project documents and explains a substantial number of grammatical universals by demonstrating a link between cross-linguistic patterns of language form and general trends of language use. The claim is that frequently expressed meanings tend to be expressed by short forms, not only at the level of words, but also throughout the grammars of languages around the world (form-frequency correspondences). A simple example is the asymmetry in the coding of present-tense forms and future-tense forms in the world’s languages, as one out of a multitude of analogous cases: Present-tense forms tend to be short or zero-coded, while future-tense forms tend to be longer or to have an overt marker. This corresponds to a usage asymmetry: Present-tense forms are generally more frequent than future-tense forms, in all languages. The proposed explanation is that higher-frequency items are more predictable than lower-frequency items, and predictable content need not be expressed overtly or can be expressed by shorter forms. Form-frequency correspondences thus make language structure more efficient, but it still needs to be shown that there exists a mechanism that creates and maintains these efficient structures: recurrent instances of language change driven by the speakers’ preference for user-friendly utterances.

Understanding linguistic diversity is important for society in a general way, especially at a time when Europe is become more linguistically diverse because of increasing migration.
During the first 18 months, we focused on the following research topics: (1) possessive constructions, (2) number and mass nouns, (3) valency coding and valency-changing operations, especially differential object marking and causatives, and (4) locational marking. We also worked on some closely related topics that did not appear directly in the project application but that are equally important for the overall research question: third-person marking and marking of independent possessors. We also worked on the diachronic dimensions of all these grammatical domains and on the more general question on the relationship between diachronic regularities and functional-adaptive constraints in grammar.

We published about 20 research papers, and an edited volume on the topic of diachronic and functional explanations in grammar is in the making. We also organized two workshops at larger conferences (a workshop at the Societas Linguistica Europaea in Naples in 2016, and a workshop at the Deutsche Gesellschaft für Sprachwissenschaft in Saarbrücken in 2017).
"Most of our research results have confirmed the original hypothesis of thoroughgoing form-frequency correspondences. There are so far only two domains where it does not seem to be confirmed, though we have not studied these in detail yet: imperatives and definite articles. One of the things we regularly do is to examine alternative explanatory proposals and engage with colleagues who have advanced these proposals, or who have voiced skepticism or critiques of the approach that we are pursuing. We are taking various possible objections very seriously and are engaging constructively with diverse strands of comparative linguistics, including generative linguistics.

We also looked closely at the psycholinguistic literature, and one of the most intriguing observations we made is that an expectation-based (or ""surprisal-based"") approach is also gaining traction in psycholinguistucs, based on a rather different set of facts. We are planning to pursue the relationships between psycholinguistic findings and our findings from world-wide comparisons further, and to intensify the dialogue with psycholinguists.

Apart from the general importance of understanding linguistic diversity in an age of increasing migration, there is no immediate socio-economic impact and no wider societal implications, as we are engaged in rather pure basic research that aims to understand the nature of grammar and language."