Marking Universals and the Structure and Evolution of Kinship Terminologies: Evidence from Salish
Hage, Per, Journal of the Royal Anthropological Institute
The linguistic couples of marked and unmarked terms belong to the dyadic forms with a preeminence of one of the two opposites. Such dyads are deeply enrooted in the cultural anthropology of the world. Their discussion, launched at the beginning of our century by Robert Hertz [1881-1915], has opened wide perspectives for the comparative analysis of ethnic structures.
Roman Jakobson and Linda Waugh, The sound shape of language.
The most notable terminological inconsistency [in Palauan] was the presence of bifurcate collateral avuncular terminology in a system which exhibited Hawaiian cousin terminology.
Roland Force and Maryanne Force, Just one house.
In a letter to Trubetzkoy (1975), Jakobson predicted that the former's discovery of marking in phonology would have interesting applications in anthropology. Jakobson was certain that in culture as in language, opposing terms are always reducible to the relation a/not a and that the relevant problem is to discover which is the marked term. The most important application was made by Greenberg (1966; 1987; 1990 ), who integrated a generalized concept of marking into a cognitive-linguistic theory of kinship universals. Although Greenberg's theory is technical in nature, it is important for three reasons: (1) it directs attention to the interaction of cognitive-linguistic and social determinants of kinship terminologies by revealing their complementary effects; (2) it provides a valuable intellectual tool for the historical analysis of kinship terminologies: (3) it develops a connexion between anthropology and cognitive science. Marking, as Lakoff (1987) has shown, is a special case of prototype effects in classification systems. Such effects are clearly discernible in kinship, the most basic of all human classification systems.
In Greenberg's theory there are two fundamental determinants of kinship terms: marking rules and the avoidance of disjunctive categories. While conjunctivity is a simple concept and an 'item of faith' that motivates the analysis of kinship systems (Hage 1997; Lounsbury 1968 ), marking is a complex idea that usually enters into kinship analysis in an incidental way. The only explicit, systematic evidence for Greenberg's theory of marking is a cross-cultural study of sibling terminology by Nerlove and Romney (1967), subsequently amended by Kronenfeld (1974), and a brief analysis of Chinese kinship terms by Lin (1983). This neglect is unfortunate because Greenberg's theory, unlike the restricted and relativistic position represented by Scheffler (1987), is comprehensive and deductive and leads to a general understanding of kinship terminologies as hierarchically ordered systems. Marking rules produce predictable asymmetries even in elementary, logically-simple systems such as Kariera, and they result in uneven but predictable patterns of terminological change.
My purpose is to provide some comparative linguistic evidence for Greenberg's theory by showing the effects of marking on the structure and evolution of Salish kinship terminologies. It is natural to include an evolutionary perspective because Greenberg's theory, although it is stated synchronically, is implicitly diachronic, concerned with processes of origin and loss. In the Salish case the process is one of loss starting from a proto-terminology which, as Morgan (1871) said of the Spokane system, is 'minute' in its discriminations and 'opulent' in its nomenclature and ending in a variety of simpler forms. Following the presentation of the Salish evidence I consider the relation between marking, salience and prototype effects and problems of indeterminacy in the application of marking rules. I then discuss the relative importance of cognitive-linguistic and social determinants of nonprescriptive terminologies such as Salish, and briefly relate my results to some recent developments in the evolutionary study of kinship systems (Godelier 1998; Kryukov 1998). …