Spatial Representation - Experimental and cross-linguistic studiesVOLUME 1: CODING MANUAL

Résumé En

This coding manual was developed during a number of cross-linguistic studies regarding the representation and expression of space (the “Langacross” Project) in oral language productions. The four main languages studied initially were English, French, German, and Chinese. In the course of the project, other languages were added (Greek, Russian, Italian, Korean, Tamil), as were other populations. Originally native speakers, child and adult, as well as adult L2 learners were studied. Later bilingual and child L2 data as well as aphasic patient populations were studied using the same materials and theoretical framework.The work is embedded in a hybrid approach that encompasses formal, functional, and cognitive aspects of language and its acquisition. It takes as its starting point the extensive typological research developed by Talmy (1978, 1985, 2000, 2013) about static and dynamic space. Talmy proposed that the basic motion scene is composed of an entity (the Figure) that is moving in relation to another entity (the Ground), as well as motion per se and the Path of motion. Other types of information (less central to the motion event) such as Manner of motion and Cause of motion are also frequently expressed in the world’s languages. Talmy furthermore proposed that one can distinguish several types of languages, such as those that are verb-framed vs. satellite-framed, based on how they express the most crucial part of the motion event, i.e., whether they express the Path of motion in the verb root (for verb-framed languages) or in satellites (for satellite-framed languages).Multiple studies were conducted based on this typology, and it was soon found that the verb-framed / satellite-framed divide is valid particularly when the Path concerns a boundary crossing (cf. Aske 1989, Slobin 2004, and others). Other authors also questioned the possibility of fitting all languages into a bipartite dichotomy. For example, in languages such as Chinese, it is less obvious to define word classes than in other languages because of the absence of morphology, and because verb-like items can be serialised and take on satellite-like features. As a result, such languages might have to be placed in a separate class within Talmy’s typology, which is not clearly satelite-framed, nor clearly verb-framed. Other proposals concern placing languages along a Path- or Manner-salience continuum.Our studies aim to look in detail into the expression of motion (voluntary and caused motion) across different languages and the impact it may have on language acquisition, including first language acquisition, simultaneous bilinguals, and second language learners at different ages and levels of proficiency. Understanding the impact of the typological differences may bring us closer to the larger research question regarding the relationship between linguistic and cognitive development or, more generally, the relationship between language and thought.The methodology and coding system presented in this manual allows for such in-depth study of the expression of space (and related temporal information), using formal, functional and semantic categories of two types: 1) general categories that may apply to all languages and 2) language-specific codes that correspond to one particular system. Particular attention is placed on three aspects to be analyzed: what type of information is expressed (information focus), how much of it is expressed (semantic density), by what particular linguistic means (parts of speech) and/or structures it is expressed (information locus and architecture).The manual begins with a general introduction to the data sets that stem from the studies on which the coding system was developed. This system is then described in detail and is organized by coding « fields » (or « levels », as they are called in CHILDES). For each field, the reader will find a general introduction concerning what is coded in it, and why. This part also includes the general coding format for that Field which must be followed strictly if one wants to use tailor-made search programmes that were developed in conjunction with this manual. Following the general introduction, all the different individual codes are provided for this field, with ample examples in English, French and German1 (information on other languages is provided in the appendices when relevant). Finally, a detailed list of additional remarks, with further explanations on how to code, language- and population-specific codes and warnings about coding pitfalls are presented. It is therefore1 Examples in these three languages were not translated, based on the assumption that researchers interested in these languages are proficient in them.3important to always read the full description of a field’s coding before embarking on the coding itself.We wish the users of this manual a lot of coding fun and would appreciate it if they could give us feedback based on their experience with it and/or suggestions for improvements.---------------------