Calendar

• Gustavo Guajardo (The Arctic University of Norway)

dans le cadre DILIS

Hopper and Thompson (1980) propose that transitivity should be modelled as a scale and not as a categorical property and propose ten (mostly binary) parameters to characterize the components of transitivity. A weakness of this approach is the lack of hierarchical structure among the parameters, thus assuming that all parameters are equally important across constructions (Givón 1995, Malchukov 2006). Using machine learning, I will present an innovative framework to derive a dynamic, structure-dependent hierarchy for the transitivity parameters. This method provides us with a single numerical value of transitivity for each clause, which I have called the Transitivity Index (Guajardo 2021a, 2021b, 2021c). Besides adding hierarchy to the transitivity parameters, another advantage of this approach is that no a priori subjective decision need be made on the part of the researcher about which parameters are the most relevant for a particular construction. The machine learning algorithm discovers the importance of each parameter from the data itself. Guajardo, G. (2021a) The Transitivity Index: Using Transitivity as a Continuous Measure to Account for Clitic Case Alternation in Spanish Causative Constructions. PLoS ONE. https://doi.org/10.1371/journal.pone.0246834 Guajardo, G. (2021b) Co-occurrence Strength and Transitivity Effects on Spanish Clitic Case Variation with Reverse-Psychological Predicates. Frontiers in Psychology https://doi.org/10.3389/fpsyg.2021.712959 Guajardo, G.(2021c) Transitivity on a Continuum: The Transitivity Index as a Predictor of Spanish Causatives. Corpus Linguistics and Linguistic Theory https://doi.org/10.1515/cllt-2021-0019

link for presentation: https://cnrs.zoom.us/j/95974153878?pwd=cXJ5UnZtOHZMM2ZPOGZhWDhIOWhadz09