Ijtimoiy-gumanitar fanlar

DEVELOPING A LONGITUDINAL CORPUS OF UZBEK CHILD SPEECH: SCHOLARLY NEED AND A CHILDES/TalkBank-BASED APPROACH

child language, longitudinal corpus, CHILDES, TalkBank, CHAT, corpus linguistics, morphosyntactic annotation, Balausa 1.

Authors

  • Zuxro AKBAROVA DSc, professor, Farg‘ona davlat universiteti, Farg‘ona, O‘zbekiston, Uzbekistan
  • Zarnigor UBAYDULLAYEVA Tadqiqotchi, Farg‘ona davlat universiteti, Farg‘ona, O‘zbekiston, Uzbekistan

This article examines the scholarly need for a longitudinal corpus of Uzbek child speech and considers how the
CHILDES/TalkBank framework can inform its development. The absence of Uzbek as a dedicated child-language corpus in the
current CHILDES index, together with the lack of a unified longitudinal and reusable format for existing national data, is identified
as the central research gap. Using a conceptual-analytical approach, the study compares RusLan-M, Balausa1, Turkish childlanguage data, CHAT transcription practices, and Universal Dependencies resources for Uzbek. Five core requirements are
proposed: regular longitudinal sampling, naturalistic audio-video data with metadata, CHAT-based transcription, morphosyntactic
representation adapted to Uzbek agglutinative structure, and ethical and technical quality control. The framework is intended to
support corpus-based study of early lexical-grammatical development, the productive emergence of grammatical categories, and
individual developmental dynamics. Formal contribution to CHILDES/TalkBank is treated as a subsequent stage contingent on
meeting repository requirements.