When a Bedouin kills another in Cyrenaica, one of a number of consequences ensues. According to the Bedouin, the particular consequence is determined by the genealogical positions of the persons or groups concerned. In the first part of this chapter an analysis of the disturbances in social relationships precipitated by homicide is made, using a lineage model as a framework. In this part of the discussion the arrangement of the information corresponds closely to the view the Bedouin give of their relationships. Additional information is then presented to show that this lineage model neither covers several important areas of social relationships nor enables an accurate prediction of events to be made. This raises the issue: can the comprehensive power of the model be increased by complicating it, or does complexity destroy its utility? The argument advanced is that the lineage model is not a sociological one, but that it is a frame of reference used by a particular people to give them a common-sense kind of understanding of their social relationships. For sociological purposes this means that the lineage model, with its supporting theoretical presuppositions, must perforce be abandoned. In the latter part of the chapter some of the implications of adopting this position are discussed. Particular stress is given to the feud, for this is one of the few societies in which it occurs. What is meant here by a state of feud is a set of relationships between two tribal groups which are characterised by hostility whenever two or more of their members meet.