Soru

Historically, historical linguists relied exclusively on the comparative method—reconstructing proto-languages by identifying systematic sound correspondences across cognates—to trace language family trees. In recent years, computational linguists introduced Bayesian phylogenetic inference, applying evolutionary modeling algorithms borrowed from biology to lexical data matrices. Early proponents of phylogenetic methods claimed these models could objectively resolve longstanding debates, such as the geographical origin of Proto-Indo-European. However, prominent traditional linguists have questioned the core premise of applying biological mutation models to linguistic evolution, arguing that lexical replacement rates vary unpredictably due to socio-cultural factors like language contact and prestige borrowing, unlike genetic mutation rates which obey relatively stable probabilistic clocks.

While these traditional critiques rightly highlight the limitations of naive glottochronological assumptions, they risk dismissing the genuine utility of modern Bayesian framework refinements. Recent phylogenetic iterations incorporate relaxed-clock models that explicitly accommodate rate heterogeneity across branches, thereby accounting for varying cultural pressures. Moreover, when integrated with archaeological and ancient DNA evidence, phylogenetic probability distributions offer valuable quantitative constraints that purely qualitative comparative methods cannot provide. Thus, while computational trees should never supersede meticulous historical reconstruction, viewing them as inherently incompatible with qualitative linguistics reflects an unwarranted methodological rigidity.

Based on the passage, which of the following best describes the author’s attitude toward the traditional linguists' critique of Bayesian phylogenetic methods?