Song-Title Originality and Reuse in a Large Music Catalog, 1800–2026: An Aggregate Data Paper

This data paper presents a versioned aggregate dataset for studying measurable novelty and reuse in song titles across 7,096,331 eligible catalog records and 6,168,623 first-observed artist-normalized-title associations spanning 1800–2026. The release reports yearly title novelty, earlier-title reuse, cross-artist reuse, concentration, fixed-sample diversity, language-label aggregates, uncertainty intervals, and trend summaries. Title reuse increased across the 1953–2025 analysis window while fixed-sample distinct-title diversity also increased, demonstrating that full-catalog reuse ratios and standardized diversity measures answer different questions. The public package contains aggregate data and documentation only: no lyrics, audio, source code, personal data, or item-level catalog records. Version 2 is archived at Zenodo under DOI 10.5281/zenodo.23559663 and licensed under CC BY 4.0.