One Hundred Years of Migration Discourse in The Times: A Discourse-Historical Word Vector Space Approach to the Construction of Meaning

Abstract

This study proposes an experimental method to trace the historical evolution of media discourse as a means to investigate the construction of collective meaning. Based on distributional semantics theory (Harris, 1954; Firth, 1957) and critical discourse theory (Wodak and Fairclough, 1997), it explores the value of merging two techniques widely employed to investigate language and meaning in two separate fields: neural word embeddings (computational linguistics) and the discourse-historical approach (DHA; Reisigl and Wodak, 2001) (applied linguistics). As a use case, we investigate the historical changes in the semantic space of public discourse of migration in the United Kingdom, and we use the Times Digital Archive (TDA) from 1900 to 2000 as dataset. For the computational part, we use the publicly available TDA word2vec models 1 (Kenter et al., 2015; Martinez-Ortiz et al., 2016); these models have been trained according to sliding time windows with the specific intention to map conceptual change. We then use DHA to triangulate the results generated by the word vector models with social and historical data to identify plausible explanations for the changes in the public debate. By bringing the focus of the analysis to the level of discourse, with this method, we aim to go beyond mapping different senses expressed by single words and to add the currently missing sociohistorical and sociolinguistic depth to the computational results. The study rests on the foundation that social changes will be reflected in changes in public discourse (Couldry, 2008). Although correlation does not prove direct causation, we argue that historical events, language, and meaning should be considered as a mutually reinforcing cycle in which the language used to describe events shapes explicit meanings, which in turn trigger other events, which again will be reflected in the public discourse.

LL(O)D and NLP perspectives on semantic change for humanities research

“…Thus, based on the principle of "triangulation", DHA takes into account a variety of datasets, methods, theories and background information to analyse the historical dimension of discursive events and the ways in which specific discourse genres are subject to diachronic change. Recent studies on linguistic change using diachronic corpora and a combination of computational methods, such as word embeddings, and discourse-based approaches argue that a discourse-historical angle can provide a better understanding of the interrelations between language and social, cultural and historical factors, and their change over time [165,184].…”

“…Based on these overviews and state of the art, we can say that automatic lexical semantic change detection is not yet a solved task in NLP, but a good amount of progress has been achieved and a great variety of systems have been developed and tested, paving the way for further research and improvements. An important aspect to stress is that this research has rarely reached outside the remit of NLP, and relatively few applications have involved humanities research (e.g., [121,165,184]). This is not particularly surprising, as it usually takes time for foundational research to find its way into application areas.…”

"Regrets for leaving the 'zoo'?": Regret construction strategies in the online discourses of Nigerian migrants

“…Many scholarly works all over the world have interrogated migration issues from different perspectives (Fina 2003;Priesto 2004;Guillem 2015;Eberl et al 2018;Obi-Ani et al 2020;Viola and Verheul 2020;de Haas 2021; de Haas 2016; among others). Migration discourse has been researched from the angle of its historical evolution (Viola and Verheul 2020), media representation of migrants (Priesto 2004;Eberl et al 2018;Marques and Ramos 2020), politics and migration (Heidenreich and Eberl 2020) and migration and identity construction (Li et al 1995;Fina 2003).…”

Discursive construction of migrant otherness on Facebook: A distributional semantics approach

“…Moreover, the use of word embeddings for discourse analysis seems also to be a promising research area. However, this opportunity has not been efficiently addressed so far in the discourse studies (see [7,8] as exceptions). This study's ambition is to contribute to the existing research in several ways.…”