Workshops > Workshop 1C - 11:45-1:30

Form-meaning co-variation in grammaticalization

Joan Bybee
University of New Mexico

Joan Bybee
Rena Torres-Cacoullos
Shana Poplack
Benedikt Szmrecsanyi

The variationist method is ideally suited to the examination of grammaticalization in progress because it enables us to track changes over time in the configuration of factors contributing to variant choice. By operationalizing hypotheses about grammaticalization and proposing multivariate quantitative models of speaker choices among different constructions, retention (persistence) of earlier meaning features and generalization along grammaticalization paths can be detected in changing co-occurrence patterns. In a complementary fashion, a diachronic perspective on grammaticalization can help elucidate the linguistic conditioning of morphosyntactic variables in synchrony.

Crosslinguistic study of grammaticalization has revealed that both form and meaning co-vary: greater reduction and fusion of grammaticalizing elements corresponds to greater semantic reduction or change. However, closer study of language-specific cases shows that form and meaning do not change in lockstep; rather, the relation between formal reduction and semantic generalization is probabilistic while change is in progress.

In this workshop, panelists explore both the formal and semantic diagnostics of grammaticalization and how their variation can be studied. Based on studies of English I don’t know and in spite of, genitive variability (the President's speech vs. the speech of the President), Portuguese future constructions, and Spanish progressives, participants will explore:

  • What are indicators of semantic change and how to operationalize hypotheses about changes in function (meaning) in order to find evidence for the progress of constructions along grammaticalization paths;

  • What kind of changes in form are predicted in grammaticalizing constructions and how to find evidence from changing distributions for the fusion of erstwhile independent elements into single units.

The formal diagnostics will include coding for phonetic reduction as well as syntactic diagnostics. These necessarily differ according to the construction under consideration, but include tests for constituency and category changes, such as separability of elements, conjunction, and presence or absence of priming.

20-minute presentations by each of the panelists will be followed by questions and discussion from workshop participants.

Joan Bybee, University of New Mexico
Shana Poplack, University of Ottawa
Benedikt Szmrecsanyi, Freiburg Institute for Advanced Studies
Rena Torres-Cacoullos, Penn State University