The PREFAB project aims to identify and analyze the prefabricated patterns of French interactions (e.g. comment dirais-je ‘how shall I put it’), in spoken corpora, interactional written corpora, and dialogues in fiction (these resources are already available). The project, initiated by LIDILEM, integrates researchers from ICAR, ATILF and BCL, research units with complementary skills. The modeling is based on construction grammars and includes syntactic, semantic, pragmatic and interactional dimensions. The innovative aspects of the project are : (a) the study of a very wide range of prefabricated patterns, from expressive expressions to metadiscursive patterns (b) an integrated approach to levels of linguistic processing (in a model based on "constructicons") (c) an innovative inductive methodology of corpus exploration (including treebanks) (d) the study of variation between sub-genres and mediums, including a comparison between French and German c. The data compiled will be freely available. They will contribute to the linguistic heritage and will be useful for language teaching as well as computer applications.

This project proposes to carry out a systematic and comprehensive study of contact-induced grammatical change during the history of Chinese, through investigating the abundance of available Chinese linguistic data. The Chinese language has undergone prolonged contact between languages of very different grammatical types throughout its history with these contacts impacting on Chinese linguistic development both synchronically and diachronically. Consequently, any study of contact involving Chinese cannot fail to be significant and valuable. During the past few decades, a significant body of literature has been published on the subject of language contact and its influence on language change, but it has primarily focused on inflectional languages and agglutinative languages, as opposed to isolating or analytic languages such as Chinese. Three periods will be covered: (i) Altaic-Mongolian influence [1000-1400]; (ii) Altaic-Manchu influence [1700-1900]; (iii) Altaic influence on Northwestern (Chinese dialects and/or) Sinitic languages (Modern period). The contact situations during these three periods are very different in terms of contact intensity, language policies, and other social and political factors. The aim of this project is to revisit and contribute some relevant solutions to some of the questions and issues raised by a good number of scholars with respect to the differences between Northern and Southern Chinese, especially the geographical distribution of various typological features in both varieties. Likewise, we will contribute to a better understanding of the areal diffusion and the typological characteristics that distinguish Northern Chinese languages from the Sinitic languages of Southern, Central and South-eastern China. Furthermore, this study will examine the nature of the languages in Yuan baihua and Northwestern Sinitic languages since some scholars have considered them as good examples of ‘creoles’ or ‘pidgins’ and perhaps even ‘mixed languages’ or ‘intertwining languages’. Thus, one of the main objectives of this project is also to address this issue in depth, discussing whether or not this hypothesis is well grounded. By thoroughly investigating and analysing the enormous amount of Chinese documents available from these three periods, research that has never been undertaken before, we will be enable to revise and enrich the present theories and methodologies of language contact studies such as CI (contact-induced) transfers, CI grammaticalization and analogical replication. By comparing and contrasting these socially and linguistically different contacts, and by thoroughly investigating and analyzing the enormous amount of Chinese documents available for these three periods, research never before undertaken, we hope to propose hypotheses and refine general theories on language contact. This will be based on the cases observed in the Chinese language to complement or challenge the existing assumptions and theories on language contact.

Students’ relationship to writing is a key determinant of school and university achievement and social integration in France. For this reason, writing still remains to be explored from the perspective of the writers’ actual skills, the linguistic aspects of their production and the contextual factors that constrain the production. In addition, the way in which teachers evaluate writing remains unclear, particularly the relationship between their comments on copies and the students’ correction/rewriting of their texts. What language skills do school and university students demonstrate in their texts and drafts according to their age and socio-cultural milieu? What role do the teachers’ practices play in influencing these manifestations? How can teachers be equipped in terms of their interventions on copies of their students? These are the questions that this pluridisciplinary research project sets out to answer through a collaboration between experts in the sociology of education and language, corpus linguistics, digital textual analysis and the didactics of writing. Based on a corpus of school and university students’ writing made available during the project, researchers will focus on characterizing certain scriptural skills (spelling and textual coherence) and better understanding of how teachers, through their interventions on copies, guide the writing of their students, in order to support them in the process of revision at school and university. This multidisciplinary project brings together the expertise of four laboratories: • Three teams in Language Sciences that combine an expertise in the analysis of learners’ writing and in natural language processing: o Clesthia (EA 7345) has developed the Ecriscol (school students’ writing) corpus. Researchers involved in the E-CALM project are experts in the genetic analysis of school students’ writing, macro- and micro-discourse structures and digital textual analysis. o CLLE-ERSS (UMR 5263) has developed the Resolco corpus (school writing involving the discursive competence of the writers). The researchers involved in are specialists in discourse analysis and in the didactics of French as a mother tongue, using a tool-based linguistic approach. o Lidilem (EA 609) has developed the Littéracie avancée (university students’ writing) and the Scoledit (writing acquisition) corpora. Researchers involved in E-CALM are experts in linguistics, in the didactics of writing and in natural language processing. • A host team in Educational Sciences (CIRCEFT-ESCOL, EA 4384) consisting of sociologists of education and language who study students’ relationship to knowledge and, in particular, the language of students from contrasted social backgrounds. The current work has three objectives: - Structuring and making available to the scientific community a vast corpus of school and university students’ writings. - Characterizing these writings and the expectations of the teachers from the point of view of the acquisition of spelling and coherence though socially-contextualized analyses. - Studying the modalities of writing in the pre-texts (plans, notes, drafts) and texts, in particular through the reciprocal influence of the written texts and the teachers’ interventions on the copies. Achieving these common goals will require a large amount of linguistic and technological work, leading to the following deliverables: - The availability of the corpus and all its enrichments. - The creation of investigation modules adapted to the corpus and its related queries. - An annotation method adapted to the specificities of the students’ writings, together with good practice guides for the collection and processing of the data. - The description of scriptural skills (morphography and coherence / cohesion) by grade level in relation to contextual variables. - Recommendations concerning the professional practice of copy correction.

As well as protecting linguistic and cultural diversity, which is clearly a priority for Europe today, multilingualism meets the need to acculturate speakers of a language. According to the Common European Framework of Reference for Languages, "Plurilingual and pluricultural competence promotes the development of linguistic and communication awareness". This linguistic awareness, fostered by the metalinguistic activity associated with contact with several languages, is often lacking in young writers when they first learn to write. It is one of the factors that helps to promote distancing and reflexivity, which are essential dimensions of written production, the active side of literacy, which all international surveys show to be extremely heterogeneous from one European country to another. In addition to the sociological, pedagogical and teaching parameters that influence this heterogeneity, the way in which teachers take multilingualism into account and the image of the language transmitted to pupils can be questioned as factors in the acquisition of writing. This goes hand in hand with preserving the richness and diversity of Europe's linguistic capital at school. The ÉPRISE project (Multilingualism and Writing : Reflexivity as a Key for Inclusive Education in Europe) focuses on the link between the metalinguistic activities generated by multilingualism and those required for writing. In both situations, attention is paid to language: spelling, morphology, syntax. How is this worked on in the classroom? How does the design of the writing sessions encourage it? Is there a connection between metalinguistic work and multilingual contexts? What ways of managing multilingualism could have an impact on writing skills and, consequently, on the socialisation of young writers? To answer these questions, the ÉPRISE interdisciplinary project will bring together linguists, computer scientists, educationalists, psychologists and legal experts from five European countries: France, Italy, Estonia, Spain and Belgium. The educational authorities in the various countries (Rectorats and Inspections Académiques in France, equivalent structures elsewhere) will be involved and associated with the work. By focusing the study on the management of plurilingualism, the aim is to uncover the variables that explain the success of school and collège pupils in producing written work, in order to provide teachers and political and administrative authorities with working tools that are in close touch with the reality of performance. By building up a multimodal corpus of students' writings, collections of teaching practices, surveys of speakers' sense of language and data on countries' linguistic and cultural contexts, we will consider parameters such as: the extent to which schools take account of the multilingual context, the language consistency, linguistic ideology, the place given to linguistic reflexivity in the school context, all of which are likely to shed light on the diversity of teaching choices and students' performances. The results will enable recommendations to be made at all levels of decision-making, in particular regional and national language policies. Through ongoing work in the field, both in the classroom and in teacher training, and by involving pupils and teachers in the tools created, the project will contribute both to the protection of linguistic diversity in Europe and to the dissemination of teaching practices conducive to pupils' literary development, producing original recommendations backed up by the results of analyses and experiments in the various participating sites.