Marlyse Baptista at the University of Pennsylvania


By Andries W. Coetzee
Feb 03, 2013 Bookmark and Share

Marlyse Baptista

Marlyse Baptista gave a colloquium presentation at the University of Pennsylvania on December 6. Her presentation was based on joint research with Susan Gelman from the Psychology Department and graduate student Erica Beck. In their research, this team tackles the problem of finding real time evidence for how convergence happens in creole formation. Convergence between the different source languages that contribute to a creole is often held up as an explanation for the grammatical structures found the creole. A problem faced with this explanation is that it is always, necessarily, post hoc. Marlyse and her team address this problem by using the so-called "artificial grammar learning" paradigm. In their experiment, they place participants in an artificial language learning situation similar to that observed during natural creole genesis, and then investigate how the participants acquire this artificial language. This represents some of the first research that probes into the psychological processes that contribute to creole genesis in the language laboratory. The title and full abstract of the presentation is given below.

Testing the convergence hypothesis in creole genesis

Contact linguists generally agree that multiple factors, both linguistic and extralinguistic, ultimately condition the linguistic make-up of contact languages in general and of pidgins and creoles in particular. Schuchardt (1888, 1914), Weinreich (1953), Muysken (1981; 1986), Thomason & Kaufman (1988), Lefebvre (1998; 2004), among others, provide thorough documentation of the linguistic processes and outcomes involved in creole genesis. On this issue, Mufwene (2001, 2008) has put forth the notion of competition and selection as a useful conceptual framework to account for creole formation. This framework assumes that linguistic features present in multiple source languages compete for survival in the newly created language. Based on this assumption, two questions emerge: how do the features that participate in creole formation get selected and why do some of the features get selected (or disfavored) over others?

In this respect, the process of convergence is particularly revealing, as features from multiple sources are believed to converge in the phonological, morphological and semantic components of the newly emerged language. Weinreich (1964), Sankoff & Brown (1976), Sankoff (1984), Keesing (1988), Siegel (2000) and Aboh (2009) have documented this process in a variety of languages and scholars like Mufwene (2008) and Siegel (2008) have proposed that a cluster of variables in combination with the morpho-phonological and semantic mapping of features favour some of the competing features to win in the make-up of creoles. Though not opposed in principle to the convergence hypothesis, scholars like Kouwenberg (2001) warn us that convergence is still considered by many as an ad-hoc, ‘after the fact’ type of process without solid evidence.

This collaborative paper presents results from an experiment whose objective is to test the morpho-phonological and semantic mapping discussed above and demonstrate that convergence is not an ad-hoc process; it is an actually observable phenomenon. The experiment was conducted with 93 monolingual, native English speakers, to test the hypothesis that speakers most readily make use of morpho-phonological and semantic elements that converge across multiple sources of input. We independently varied the morphological similarity and the functional overlap between elements in a new, artificial “language” and elements in the speaker’s extant language, to assess the conditions under which novel linguistic elements are acquired. Using three conditions labeled the congruent condition, the reverse condition and the novel condition, we tested whether speakers make use of the morphological and functional overlap across languages to more readily accept or reject features of the language. We predicted that participants would learn most readily in the congruent condition and preliminary results (results of recoding pending) seem to support the hypothesis.