Glottochronology proposes to date language splits arithmetically
Morris Swadesh set out a method for converting the proportion of shared cognates in a fixed list of basic meanings into elapsed time since two languages diverged, assuming a universal replacement rate calibrated on documented histories. The proposal was attacked immediately and comprehensively: rates are not constant across meanings, languages or historical circumstances, and the cognacy judgements the arithmetic consumes are themselves the contested output of comparative work. Mainstream historical linguistics had largely abandoned the method by the 1970s.
Historical — it happened, and the record is settled.
Subjects
- Glottochronology · process · not located
- The comparative method · process · not located
Evidence
Partly verified — Core facts are sourced; some optional detail is deliberately absent.
Historical Linguistics: An Introduction, third edition
Supports the general account in this pack of the comparative method, the regularity principle, contact and borrowing, glottochronology and its critics, and the definition of a language isolate. V55 verification basis: this session had no access to any source — WebFetch was egress-blocked for every host attempted and the WebSearch budget was exhausted before authoring — so the cited work was not retrieved, and no page, volume or figure quoted from it was re-checked. No chapter or page is cited, because the book was not opened in this session; the citation identifies the work a curator should check the general account against. This is the source named in the Journey declaration.
Edinburgh University Press · 2013 · Partly verified
Campbell, L., Historical Linguistics: An Introduction, 3rd edn (Edinburgh University Press, 2013)
Recorded on Glottochronology
- The Neogrammarian doctrine: sound laws admit no exception
- Glottochronology proposes to date language splits arithmetically
- Language in the Americas, and its rejection
Related
Related because they share a subject in the Atlas — never because the text looks similar.