Competence, performance and language policy / Peti-Stantić, Anita. - 2016 - 3-4 str.
If we assume that language is one of complex cognitive faculties in many respects parallel to others such as vision or memory, we have every right to ask what the scientifically-informed language policy should look like today. By aiming to develop language performance of an individual within the specific language community (at a supra-national, national or sub-national level), I see language as a multidimensional capacity immersed in cognitive and conceptual world of an individual and, by extension of the “same except” principle, language community. When speaking of language policy, the main issue at stake is the status of lexicon and grammar and their mutual interrelation. Unlike generative grammar, in which the words and rules are set apart, in the theoretical framework of Parallel Architecture there is no strict division between the lexicon and the grammar. Rather, words and standard rules are at the opposite corners of a multidimensional continuum that includes all sorts of mixed items such as idioms and meaningful constructions. Important consequence of this view is that semantics is not necessarily in a one-to-one relation to syntax, but most often in a many-to-many relation, which needs to be stated explicitly. Using platform of Parallel Architecture as a foundation, I performed informed experimental psycholinguistic research. This is exactly the type of research that can and, I believe should, direct language policy, which should go beyond merely reducing itself to the compliance to the normative rules of grammar. Therefore I will speak of three rather diverse, yet interconnected phenomena that point to the importance of informed decision-making within the domain of language policy. Firstly, I solicit the analysis of so-called light verbs in Croatian, which are mostly used in a conversational style, from the normative and psycholinguistic treatment point of view. Secondly, I subject the specific problems of syntax-semantics mismatches to scrutiny. These problems are usually considered as unnatural collocations observed in translation services of EU institutions. Finally, I wrap up by advocating for many individual and social benefits of developing competent and critical speakers and readers by building specific types of world-webs at an early age. I use all of these different, but interrelated phenomena as an example of the importance of experimental psycholinguistic research with proper theoretical grounding in the sphere of language planning and language policy. As a consequence, such a position should inform language beliefs and explicit language policies in a much more efficient way than normative standard positions that were dominant in Croatia for decades.