A basic assumption within truth-conditional semantics is that we use language to talk about the way the world is (or should be). Establishing reference is consequently seen as one of the basic functions of language. And articles - the basic tools to establish reference - have acquired a fundamental position in the semantics literature. Research in article semantics has however restricted itself to definite and indefinite articles and has largely ignored possessives like my, your, & In this project, I (i) argue that possessives should be included in the article paradigm, (ii) show how this insight adds a whole new dimension to ten years of article acquisition research and (iii) bring together semanticists, L2 acquisition researchers and language teachers as well as their research methodologies to come up with a new state of the art view on article semantics. In part 1 I work out an article semantics of possessives, study their cross-linguistic variation and end up with a typology of possessives. Part 2 is devoted to the predictions this new typology makes for the L2 acquisition of articles. These predictions are tested both experimentally and on learner corpora (electronic collections of texts produced by L2 learners). Part 3 builds towards the new state of the art view on articles by bringing together the theory-driven research of part 1 and 2 with error-driven analyses of learner corpora inspired by those of language teachers. Next to publications in international linguistics journals and workshops aimed at semanticists and L2 acquisition researchers, the project will make its results as well as the theoretical insights of three decades of article research available through a tri-lingual (Dutch-French-English) website aimed at language teachers that will be launched during its closing symposium.

The project unites an American researcher with wide experience in use of computational methods for child language corpus analysis as well as the study of early childhood stuttering with two Dutch researchers (Franken; Wijnen). The main objective is to extend a previous study led by Dr. Franken on stuttering interventions for children (the RESTART trial, funded by ZonMW project 94507417). The focus is now on factors in language development that may predict recovery versus persistence of childhood stuttering and/or response to early intervention. We are currently in the process of preparing data for detailed linguistic analysis. Dr Bernstein Ratner has proficient computational skills in the use of Dutch parsers and collaborates with Wijnen and Franken (and their respective labs) in coding and analyzing the spoken language samples from the large cohort of children that were studied in the RESTART trial (De Sonneville-Koedoot, Stolk, Rietveld, & Franken, 2015); this is the largest corpus of preschool speech-language samples of stuttering children collected to date. Professor Ratner’s visit would allow us to expedite the conversion of the data set to the international TalkBank format and to discuss the analysis of linguistic factors at onset of stuttering that may predict response to therapy or may promote spontaneous (untreated) recovery. The team would also prepare an NIH NIDCD R01 grant proposal to analyse linguistic factors related to development of speech fluency in early childhood, as well as examine potential shared mechanisms of effectiveness in the two stuttering therapies that can be exploited to improve therapies for childhood stuttering.

To make sense of a written or spoken text, it is crucial that people construct discourse relations such as CAUSE-CONSEQUENCE and CONTRAST between sentences. Connectives like “because” and “but” help people construct these relations, but many relations do not contain a connective. Little is known about what other cues can signal discourse relations, and whether people readily use such signals. By investigating how discourse relations are expressed in natural language, and how people process such expressions, I will gain deeper insight into how humans can effectively communicate via writing or in face-to-face situations.

This study will investigate the effectiveness of speech and language therapy in preschool children with severe developmental language disorders and special education needs. A developmental language disorder severely affects communication skills and has adverse effects on all aspects of a young childs development. There is a lack of information on the effectiveness of care and education provided to young children with severe language disorders. Results of effect studies are contradictory and in most cases aimed at older children. Furthermore there is a variety in organisation of care and education to children with developmental language disorders in countries over the world, so results of international studies cannot simply be generalized to the Dutch situation. In the proposed study the effects of intervention on growth in language skills will be investigated by following the progress of a large cohort systematically. Child factors, therapy factors and educational factors will be recorded as possible predictors of effect. Subsequently evidence based speech and language therapy according to a protocol will be compared to care as usual on schools for special education. The results of the study will lead to improvement of care and education that is offered to children with developmental language disorders.

Infants face the challenge of segmenting continuous speech into words and likely use statistical learning (SL) to aid this process. Individual variation in SL ability at an early age is associated with later vocabulary development and language disorders. An outstanding question is how such individual differences in SL arise. Rhythmic abilities are hypothesized to drive these differences, as well as general cognitive abilities (e.g. pattern recognition, working memory). This project will investigate if rhythmic and/or cognitive abilities predict individual performance on SL tasks. This will provide new insights on language acquisition and possibly early detection of language disorders.