Annotated corpus sample for "Order across domains: a domain-neutral account of two Swedish prepositions for sequential order"

This dataset contains annotated corpus samples for two Swedish prepositions, före ('before') and efter ('after'), used in a study of their semantic structure. The data were extracted from four corpus types — social media, fiction, news, and academic texts — via Korp at Språkbanken, yielding a random sample of 4000 attestations per word (8000 total). Each token was manually annotated for semantic and grammatical categories. Methodological details are provided in the README file and in the article cited below.

The dataset supports the findings reported in Widoff (2026), "Order across domains: a domain-neutral account of two Swedish prepositions for sequential order," in Folia Linguistica. DOI: 10.1515/flin-2025-0160