The project proposes an interdisciplinary approach to tackle the topic of the identification of features, debates, actors and roles in migration narratives, taking into consideration their textual patterns and their evolution in time (2013-2020) within the Italian context, characterised by different migration issues as well as common and/or divergent discourses and policies throughout the years. The project will make use of innovative state-of-the-art techniques in corpus linguistics, information extraction, knowledge representation to retrieve, weight up and ultimately present the rhetorics of hetero-representations and self-representations of migrants (and ethnic minorities) over time. Data will be extracted from two ad-hoc assembled big corpora (ca. a billion words, from a. the media/press and b. social networks) and coupled with a set of information provided from different fields, namely: historical and philosophical studies, linguistics, law, demography, sociology and media studies. Such integration of different levels of information (that can be summarized as a mapping of textual data and metadata) will foster qualitative and quantitative analyses, with the ultimate goal of providing a dynamic portrait of changes in migration narratives and their event triggers. The object is to give a full multidimensional portrayal of representation of migration in Italy.
The outputs of the project will include structured information on migration discourse in the forms of research protocols, country reports on the dynamics of representation and its interpretation, and guidelines available for both policy makers, media workers and the civil society in order to facilitate actions improving objective representations, integration and inclusion. A set of scientific publications, addressing the topic issue from the specific points of view of linguistics, philosophy, sociology, media studies, demography and legal aspects will also be produced.
The main innovative feature of the project is the ambition of integrating different cross-field competencies on the topic of migration and anchor them to a data-driven approach, grounded on contemporary empirical information. In fact, although multidisciplinarity is used in several studies to tackle the topic of migration phenomena within more comprehensive perspectives (Baggiani et al., 2011; Vannini et al., 2018), a deep integration of qualitative and quantitative analyses on the same data-set is an approach that has been rarely embraced so far.
From the linguistic perspective, the creation of a large new corpus on migration represents a precious occasion to build a resource that includes official and cultural information in Italian targeted on the topic, and will be possibly used also for secondary analyses in the future. The metadata scheme to be added to the corpus, taking into account extra-textual information regarding entities, historical and socially-relevant facts associated to the texts, will require new standards of information labelling and visualization, which will contribute to the existing literature of language annotation in computer-mediated communications. Discourse analysis, and its related line of studies on migration and minorities, will also benefit from an investigation in corpus linguistics closely guided through the lens of different social sciences with an established tradition on migration representation.
At the same time, the prospective analysis on the corpus that the project envisages, will provide innovative and up-to-date data for social sciences, which, within the field of Italian migration studies, are used to deal with scarce or aggregated information, surveys (often collected in `unnatural contexts') or limited case studies, posing methodological problems about impact, coverage, replicability and comparability of the results. Textual data (especially those coming from social networks), sampled with a reasoned methodology, constitute genuine and abundant information instead, with the potential of being used both in a collocational environment (for qualitative research) and in a structured way (for information extraction). In this sense, one of the expected results is to deepen and possibly overcome the classical dichotomy of psycho-social and media analyses on migration representation, usually depicting migrants as threats or victims in a stereotyped manner. In particular, the analysis of the narrative dynamics will specifically aim at understanding not only the perception and representation of migrants, but, specifically, the relations between events, single actors, groups (including their different resources in terms of symbolic-discursive power, i.e. in the media field), of the overall phenomenon of migration, focusing on the rise or change of certain societal responses resulting also in prosocial behaviors, tolerance, as well as hate speech. The integration between empirical textual data on narratives and social analyses is also fundamental to improve the understanding of the mechanisms behind the political exploitation of information, aimed at intercepting people¿s consensus, that, within our complex era characterized by a massive access and use of online media and social networks, remains still partly uncovered in its fast-changing dynamics.
Finally, the interaction between linguistics, psychology, sociology, law, history, philosophy and demography will also stand out as an innovative feature in most of the areas mentioned, by developing a template procedure to connect and map textual data and external metadata in order to offer and authentic and large base of data to be interpreted in a multidisciplinary approach in order to provide a multidimensional overall portrait of representation of migration in time and all its influencing factors.