An EU-funded project was established to enhance understanding of the nature of language variation, so-called logodiversity. The outcomes shed light on particular approaches to the problem of language acquisition.
The project 'Logodiversity: On the nature of linguistic unity and diversity' (LOGODIVERSITY) took as its starting point the standard Principles-and-Parameters model in generative grammar. It examined the challenges that arise in the use of this minimalist programme when accounting for variation in grammar.
The project examined the feasibility of constructing the range of options that a child could use to acquire language (parametric space) without postulating an a priori switchboard (the classical Parametric model).
Several architectural conclusions were drawn from work covering two major sub-tasks: construction of microparametric options and construction of macroparametric options.
With regard to the first, research showed that genuinely lexical parameters have rarely been posited in literature, even among microparameters. In fact, the vast majority of so-called lexical microparameters take the form of feature-bundling parameters, where 'bundle' is used to describe a single lexical item. However, treating bundling parameters as lexical results in a syntactic micro-structure (a treelet) that arises from a syntactic process.
LOGODIVERSITY explored the possibility that macroparametric effects are the results of learning biases. For instance, biases could occur when learners generalise the effect of a microparameter to similar categories to optimise the learning process for language acquisition.
One of the project's conclusions was that constructing 'a' parametric space does not mean reconstructing 'the' (standard) parametric space assumed in the (mature) Principles-and-Parameters model. LOGODIVERSITY emphasised that the learning path that emerges is far more complex than that entertained in the current literature.
The project work and its findings contribute to the body of knowledge on language acquisition and challenge existing models used to study sociolinguistic variation.
Provided by Cordis