Rhythm-based authorship recognition in syllabic and accentual-syllabic verse

Abstract

This contribution explores the extent to which rhythm-based features of poetic texts can contribute meaningfully to authorship recognition. We show that, although a binary categorization of languages as syllabic vs. accentual-syllabic fails to fully explain the differences. However, once we formalize accentual regularity as a continuum, our analysis shows that authorship attribution results improve as we move from the most to the least accentually regular languages. This result supports our hypothesis that accentual regularity is an inhibiting factor in authorship attribution as long as accentual regularity is understood as a continuous property.