To reconstruct morphosyntactic evolution by ancient data: an overview of advantages and obstacles

Activity: Talk or presentationInvited talk


Title To reconstruct morphosyntactic evolution by ancient data: an overview of advantages and obstacles
Person and role
Description In spite of decades of careful and systematic analysis of morphosyntactic and typological evolution, manifested in datasets such as WALS, AUTOTYP, SAILS, or Grambank, many details of the mechanisms of prehistoric morphosyntactic change remain mysterious. At the beginning of the emergence of the typological approach in the 1960s, the aim of the research was to define universal patterns of linguistic typology (Greenberg 1963), by matching typological patterns cross-linguistically over family- and area boundaries. Even though typological research had an impact on historical linguistics, giving rise to the approach of ‘diachronic typology’, diachrony was not implemented in synchronically oriented typology.
The lecture will present a results of a project of typological reconstruction, using evolutionary models of character reconstruction (Maurits and Griffiths 2014), where we have developed a model for reconstruction that systematically allows the inclusion of ancient language data (Carling 2017; Cathcart et al. 2018). In the results, there are several interesting observations: first and most important, that we may increase the granularity in detail of the reconstruction. Further, by using ancient data, we are able to perform more careful observations on the change behavior of various traits. Meanwhile, we should raise skepticisms about our models of reconstruction: is it an artefact of the model that if the granularity increases, so does the uncertainty? What happens if we expand the datasets in one or the other direction? Using our results as a pilot, the lecture will discuss advantages and obstacles in evolutionary reconstruction, using ancient language data.

2017 Oct 19

Related organisations
2017 Oct 19

External organisation (University)

NameUniversity of Zurich