Areal Diffusion, Genetic Inheritance, and Problems of Subgrouping: A North Arawak Case Study

Abstract

Abstract To establish subgroupings within a language family one needs to recognize <a set of changes common to a particular subgroup which has occurred between the period of divergences of the family as a whole and that of the subgroup in question’ (Greenberg 1953: 49). The shared changes should be significant and fairly unusual, not the sorts of changes that recur aH over the world and could well have happened independently in each of the languages (e.g. palatalizing a velar consonant next to a front vowel). Note that the criteria! features must be shared innovations. If a number of languages within a given family share retentions from the proto-language this does not require a period of shared development and does not constitute evidence for subgrouping. Shared loss (e.g. loss of a system of possession markers, pronominal suffixes, or tense inflections) is also not a good criterion for subgrouping. What may be significant is the way the loss is replaced; if each of the languages loses tense inflection on verbs but then makes reference to time through an innovated system of auxiliaries (with the auxiliaries being cognate in form) then this shared innovation does provide evidence for their constituting a subgroup.