Illiberal SCRIPTS Dataset: Corpus-Based Analysis of Norm Diffusion across Post-Soviet Regional Organisations (CIS, CSTO, SCO)

Abstract

This dataset accompanies the book chapter "Scripts in Circulation: Cross-Pollination of Anti-Liberal Norms in Eurasian Regionalism" (Gawrich & Nasibov, In: Panke, Diana; Libman, Alexander; Börzel, Tanja: Contestations of the Liberal Script in Regional Organizations in the Global South and the Global North. Oxford University Press (Forthcoming)).

It provides all reproducible analytical artefacts derived from declaratory documents of three intergovernmental organisations – the Commonwealth of Independent States (CIS), the Collective Security Treaty Organisation (CSTO), and the Shanghai Cooperation Organisation (SCO) – covering the period 2002–2024.

The dataset excludes verbatim text for copyright reasons and instead offers fully processed linguistic and statistical representations that enable complete replication of the study’s computational findings.

The analysis conceptualises “script cross-pollination” as the diffusion of illiberal normative vocabularies – such as sovereignty, stability, multipolarity, and non-interference – across regional organisations in post-Soviet Eurasia.

Using a transparent R-based pipeline, the study lemmatises and tokenises Russian-language summit declarations, constructs document–feature matrices and TF–IDF representations, and computes time-series trends, inter-organisational cosine similarity, and lead–follow diffusion patterns.

The dataset demonstrates how illiberal normative repertoires migrate, stabilise, and co-evolve across Eurasian organisations, showing that these institutions function as discursive relay stations of a shared counter-liberal script rather than as isolated challengers of liberal order.

All processing and analytical steps are documented in the included file illiberal_scripts_pipeline_public.R and further explained in HOW_TO_USE_outputs_README.md.

Contents File - Description illiberal_scripts_pipeline_public.R - Full R code for reproducing the analysis pipeline (CC-BY-NC). HOW_TO_USE_outputs_README.md - Documentation of data structure and usage. step02_dfm_lemmas.rds - Lemma-level document–feature matrix. step02_dfm_lemmas_aligned.rds - Lemma DFM aligned with metadata. step02_dfm_tfidf_lemmas.rds - TF–IDF weighted lemma matrix. step02_dfm_bigrams.rds - Bigram-level DFM (2-word combinations). step03_freq_illiberal_terms.csv - Yearly frequency of illiberal dictionary terms (CIS/CSTO/SCO). step03_crosspollination_similarity.csv - Cosine similarity between organisations by year. step03_first_use_by_org.csv - First observed use of each lemma by organisation. step03_adoptions_with_origin.csv - Term diffusion trajectories (origin → adopter + lag). step03_influence_edges_lag3.csv - Aggregated influence edges within a 3-year window.