On cause and correlation in language change

Abstract

Abstract Studies of language change frequently wrestle with the problem of cause and correlation. It is comparatively simple to observe a correlation between historical trends. However, it is much more difficult to demonstrate that the changes in the frequency of a construction A were indeed the cause of the changes in the frequency of a construction B. We present a statistical method that can help to assess whether two historical changes are not only correlated, but also causally related. In particular, we use Granger Causality to determine whether the gradual replacement of ex-situ wh-interrogatives with clefted wh-interrogatives in 18th to 20th century Brazilian Portuguese resulted from word order change, and particularly the rise in the frequency of SV word order. Our results indeed suggest that SV word order Granger-causes the use of clefted wh-interrogatives, as well as declarative ‘that’-clefts, but not the other way around.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.