Ambiguity resolution and the evolution of homophones in English

Research output: Chapter in book / Conference proceedingChapter in an edited book (as author)Academic researchpeer-review


Based on a quantitative study of the evolution of homophones in English, we present an argument about why homophones occur. Zipf’s law, which states that word frequency decreases as a power law of its rank, can be seen as the outcome of form-meaning associations, adopted in order to comply with listener and speaker needs. This implies that one form can correspond
to many meanings (i.e., polysemy and homophony). We argue that
homophony is a desirable feature in communication systems, is stable, and
increases through time. When a large number of homophones emerge,
however, an impetus to avoid homophones comes into play. We suggest that
the evolution of diatones is a case of the avoidance of homophony. Related
to this, we examine the neural substrates of bisyllabic noun-verb homophones, using near-infrared spectroscopy. We show that noun and verb
categories are represented in different neural substrates in the left
hemisphere, and relate this to our historical data, explaining why the
actuation of diatone-formation was connected with production in frequent
homophones in the 16th century, but was connected with perception in
infrequent words after the 17th century.
Original languageEnglish
Title of host publicationEnglish Historical Linguistics: Change in Structure and Meaning.
Subtitle of host publicationPapers from the XXth ICEHL
EditorsBettelou Los, Claire Cowie, Patrick Honeybone, Graeme Trousdale
PublisherJohn Benjamins
ISBN (Electronic)9789027258205
ISBN (Print)9789027210647
Publication statusPublished - Feb 2022

Publication series

NameCurrent Issues in Linguistic Theory
PublisherJohn Benjamins


  • linguistic evolution
  • homophones
  • ambiguity
  • Zipf’s law
  • word frequency
  • diatones
  • neural substrates


Dive into the research topics of 'Ambiguity resolution and the evolution of homophones in English'. Together they form a unique fingerprint.

Cite this