Passing like flowers: the marriage regulations of the Tugen of Kenya and their implications for a theory of Crow-omaha
The dissertation presents an alliance model of Crow-Omaha marriage-system constraints based on the ethnographic description of the marriage and kinship relations of the Tugen of Kenya. Both Crow-Omaha and asymmetrical alliance constraints are shown to be logical derivatives of this model. Furthermore, the parameters on which the model is based seem to constrain all marriage alliance systems, regardless of the mode of lineation utilized. Implicit in the model is a theorem which stipulates that systems operating within Crow-Omaha and asymmetrical alliance constraints require specific minimum numbers of alliance units. These numbers are shown to be those reported to be associated with Crow-Omaha systems in a variety of symbolic contexts. Several implications of importance for anthropological theory follow from the model. First, different social classes in stratified regimes operating with Crow-Omaha constraints are likely to have different minimum numbers; the higher the stratum, the lower the number. This suggests an explanation for the origin of the common Indo-European symbolic oppositions, where three is sacred, thirteen denotes misfortune, and numbers in between like five and seven are associated with good luck. Since Crow-Omaha systems, in contrast to asymmetrical alliance systems, involve more prohibitions, and therefore, • larger minimum nUmbers of alliance units, they are less amenable to the manipulation for the advantage of a particular alliance group than are asymmetrical alliance systems proper. The model thus has implications for our understanding of the process of social stratification. The similarity of kin classification systems typically associated with both Crow-Omaha and asymmetrical alliance is shown to result from interaction between the processes of kin classification and marriage alliance. Crow-Omaha kin classification is meaningless in the absence of Crow-Omaha marriage constraints, but people with such constraints n~ed not have a Crow-Omaha classification. If they do, then they have the simplest kin classification which encodes the maximum amount of information about the relative social statuses of interacting persons vis a vis the alliance relationship linking their kin groups. variations within the class of Crow-Omaha systems are shown to be related to different value-assignments of the parameters of the model. The dissertation demonstrates that theoretical discussions about Crow-Omaha or asymmetrical alliance in recent years are mistaken or misleading. Most notably it falsifies Needham's claim that Omaha and, by implication, Crow kinship systems do not constitute a class of systems with any analytical value or validity. Furthermore, Radcliffe-Brown's notion of lineage solidarity is shown to be irrelevant. Finally, the model suggests an ecological-adaptive basis for Crow-Omaha marriage rules. The suggestion is speculative, but it follows from the theory, and provides an area of some interest for further research.