Slovenian Social Assistance Rights text data collection (SSAR 1.0)

english

2025

volume 13, issue 2

legal language   readability measures   syntactic complexity   text complexity measures

Univerza v Ljubljani, Filozofska fakulteta
(Obvezni izvod spletne publikacije)

This paper presents the Slovenian Social Assistance Rights text corpus (SSAR 1.0), a newly compiled and openly available dataset intended to support linguistic and computational analysis of language complexity in the domain of social assistance law. The corpus includes both legally binding and interpretative texts and is analysed using a range of established readability formulas and syntactic complexity metrics. The results show that social assistance texts tend to exhibit significantly higher sentence-level complexity than selected reference texts from news portals and government websites, particularly in terms of sentence length, syntactic depth, and dependency structure. Importantly, some explanatory texts intended for beneficiaries were found to be among the most complex, pointing to a notable disconnect between communicative purpose and linguistic form. In contrast, texts from news portals displayed consistently lower structural complexity, offering valid empirical points of comparison. The study confirms that quantitative measures can reliably flag particularly complex texts and identify more accessible examples, providing a practical foundation for developing clearer legal drafting practices. By establishing a replicable methodology and empirical baseline, this work contributes to ongoing efforts to improve linguistic accessibility in Slovenian public communication.

06.03.2026