手法を比較

	Phylogenetic Linguistics	Comparative Method (Historical Linguistics)
分野	言語学	言語学
系統	Process / pipeline	Process / pipeline
提唱年	2003	1861
提唱者	Russell Gray & Quentin Atkinson (modern Bayesian application); rooted in computational phylogenetics	Neogrammarians (Karl Brugmann, August Schleicher; building on Rasmus Rask, Jacob Grimm, Franz Bopp)
種類	Computational inference of language family trees and divergence dates from coded linguistic data	Systematic comparison of cognates to reconstruct a proto-language and establish genetic relationship
原典	Gray, R. D., & Atkinson, Q. D. (2003). Language-tree divergence times support the Anatolian theory of Indo-European origin. Nature, 426(6965), 435–439.	Campbell, L. (2013). Historical Linguistics: An Introduction (3rd ed.). Edinburgh University Press. ISBN: 9780748675593
別名	Linguistic Phylogenetics, Computational Language Phylogenetics, Phylogenetic Language Classification	Comparative Reconstruction, Comparative Linguistic Reconstruction, Method of Comparative Reconstruction
関連	4	4
概要	Computational phylogenetic linguistics borrows the statistical machinery developed in evolutionary biology — Bayesian inference, maximum likelihood, and distance-based network methods — and applies it to coded linguistic data, chiefly cognate-judged basic vocabulary, to infer language family trees and estimate when branches diverged. By treating linguistic characters like the molecular characters in a gene alignment and modelling their change probabilistically along a tree, the approach produces classifications with explicit measures of uncertainty and, when calibrated, dated phylogenies. Its best-known applications are the Gray and Atkinson and Bouckaert et al. analyses of Indo-European origins.	The comparative method is the foundational technique of historical linguistics for demonstrating that languages are genetically related and for reconstructing their unattested common ancestor. By systematically comparing cognate words across related languages and uncovering the regular, recurring sound correspondences between them — exemplified by Grimm's Law in Germanic — analysts reconstruct the forms of the proto-language and the sound changes that produced each daughter, and on that basis build the family tree. It is a qualitative, evidence-driven method distinct from the generic logic of cross-case comparison: here the 'comparison' is of linguistic forms governed by the regularity of sound change.