Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Computational Analysis of the Dehumanisation of Migrants from Syria and Ukraine in Slovene News Media

Jaya Caporusso, Damar Hoogland, Mojca Brglez, Boshko Koloski, Matthew Purver, Senja Pollak

TL;DR

This paper tackles dehumanisation in migrant coverage within Slovene news by adapting a English-language framework to a less-resourced language. It introduces new resources (Migrant, Moral Disgust, and Vermin term lists) and a Slovene Valence–Arousal–Dominance lexicon, together with a cross-lingual VA detector and an anchoring Kolmogorov–Smirnov test to compare corpora without explicit vector-space alignment. Applying these tools to two migration-crisis periods (Syria 2015–16 and Ukraine 2022–23) across large Slovene-news corpora, the study finds that discourse becomes more negative and aroused over time, while dehumanising language is not uniform: Ukrainian migrants are dehumanised less than non-Ukrainian migrants, though vermin-metaphor use can rise in some subcontexts. Overall, the work demonstrates a transferable, multi-faceted approach for measuring dehumanisation in media, delivers valuable Slovene-language resources, and offers insights into media framing of different migrant groups with potential policy and public-discourse implications.

Abstract

Dehumanisation involves the perception and or treatment of a social group's members as less than human. This phenomenon is rarely addressed with computational linguistic techniques. We adapt a recently proposed approach for English, making it easier to transfer to other languages and to evaluate, introducing a new sentiment resource, the use of zero-shot cross-lingual valence and arousal detection, and a new method for statistical significance testing. We then apply it to study attitudes to migration expressed in Slovene newspapers, to examine changes in the Slovene discourse on migration between the 2015-16 migration crisis following the war in Syria and the 2022-23 period following the war in Ukraine. We find that while this discourse became more negative and more intense over time, it is less dehumanising when specifically addressing Ukrainian migrants compared to others.

A Computational Analysis of the Dehumanisation of Migrants from Syria and Ukraine in Slovene News Media

TL;DR

This paper tackles dehumanisation in migrant coverage within Slovene news by adapting a English-language framework to a less-resourced language. It introduces new resources (Migrant, Moral Disgust, and Vermin term lists) and a Slovene Valence–Arousal–Dominance lexicon, together with a cross-lingual VA detector and an anchoring Kolmogorov–Smirnov test to compare corpora without explicit vector-space alignment. Applying these tools to two migration-crisis periods (Syria 2015–16 and Ukraine 2022–23) across large Slovene-news corpora, the study finds that discourse becomes more negative and aroused over time, while dehumanising language is not uniform: Ukrainian migrants are dehumanised less than non-Ukrainian migrants, though vermin-metaphor use can rise in some subcontexts. Overall, the work demonstrates a transferable, multi-faceted approach for measuring dehumanisation in media, delivers valuable Slovene-language resources, and offers insights into media framing of different migrant groups with potential policy and public-discourse implications.

Abstract

Dehumanisation involves the perception and or treatment of a social group's members as less than human. This phenomenon is rarely addressed with computational linguistic techniques. We adapt a recently proposed approach for English, making it easier to transfer to other languages and to evaluate, introducing a new sentiment resource, the use of zero-shot cross-lingual valence and arousal detection, and a new method for statistical significance testing. We then apply it to study attitudes to migration expressed in Slovene newspapers, to examine changes in the Slovene discourse on migration between the 2015-16 migration crisis following the war in Syria and the 2022-23 period following the war in Ukraine. We find that while this discourse became more negative and more intense over time, it is less dehumanising when specifically addressing Ukrainian migrants compared to others.
Paper Structure (54 sections, 1 figure, 1 table)

This paper contains 54 sections, 1 figure, 1 table.

Table of Contents

  1. Introduction
  2. Related work
  3. Data and resources
  4. Corpora
  5. Analysis Tools
  6. Static Embeddings
  7. Concept vector construction
  8. Migrant terms
  9. Moral disgust terms
  10. Vermin terms
  11. Concept vector construction
  12. Slovene Valence, Arousal and Dominance Lexicon
  13. Zero-shot Cross-Lingual VA Detection
  14. Methods
  15. Vector-based similarity analysis
  16. ...and 39 more sections

Figures (1)

  • Figure 1: Distributions of valence scores for $C_{syr}$ and $C_{ukr}$ according to the lexicon approach (A) and the transformer model (B).

Theorems & Definitions (1)

  • Example 1