Social Organization in Aboriginal Australia. WARREN SHAPIRO

ed to manipulate the notion, to invoke a "principle of flexibility," or to distinguish between the "ideology" and the "composition" of the groups. Sillitoe has chosen another course. He claims not only that there are no descent groups in Wola society (Southern Highlands), but that only individuals, not groups, are relevant in it. This reduction of society to a conglomerate of individuals i s made possible by taking descent theory at i t s face value. Consequently, the semonda the named, localized, exogamous, propatrilateral, and propatrilocal groups that compose Wola society-are not considered as descent groups. But it seems that, deprived of the possibility of accounting for these groups in terms of descent, Sillitoe is incapable of accounting for them altogether, except in purely empirical terms. He goes so far as to claim that they owe their "continued ordered existence as groups" to their "smallness" (p. 34)! And although he at one time concedes that "permanent, if somewhat vague, groups do occur and play a prominent part in the organization of Wola life" (pp. 30-311, in fact his model takes into account mainly two opposite terms: individuals and society. The emphasis, however, i s on the individual, since "it i s individuals that make up a society, which exists because it benefits them. This i s Melanesian philosophy pure and simple" (p. 4). It seems that Melanesian philosophers are of the Hobbesian persuasion, except that they substitute "exchange" for "contract" (after all, they are not Englishmen). We are told that exchange has been devised by the Wola to overcome the original state of "Warre" (p. 4). This theory implies that self-interested and rational individuals preexist society. Self-interest and rationality prompt individuals to associate by means of exchanges. Exchange is the social principle that combines the maximum degree of "sovereignty of the individual" (p. 78) with the minimum of constraint. Its nature is such that it promotes sociability, while at the same time promoting competition and, therefore, self-interest (pp. 5-6). From this point of view, society exists only because of a contingent agreement or complementarity of the atomistic (p. 288). selfinterested decisions of the individuals who exchange "only when they think it i s favourable to them" (pp. 10, 22). not whenever there i s an obligation to do so. From another point of view, however, society has an ontological priority over the individual since it "uses the self-interested drive of individuals to support itself by harnessing it to values which guide behaviour so that it feeds back and reinforces the social order" (p. 8). Whenever Sillitoe is unable to account for the actions of the individual in terms of costs and benefits as conceived by the utilitarian, Hobbesian model, he attributes them to the influence of values created by a society which itself acts as a "rational individual." Thus, what cannot be explained as individual interest is explained as society's interest (i.e.. in terms of "social functions"); but society is conceived of as an individual. It seems quite clear that Sillitoe projects his own utilitarian philosophy onto Melanesian ideas; at any rate, he does not quote any Wola statement that would confirm his claim that "this i s Melanesian philosophy pure and simple." I suppose that he would answer that the existence of this "philosophy" i s implied by what the Wola do. But, quite apart from the fact that what the Wola do seems much more complicated than S i l litoe makes it out to be, it i s dangerous to infer thought from action, especially if the actual ideas of those who act are not considered at all. At any rate, the "individualistic" and the "functionalistic" models are used side by side in Sillitoe's analysis without any attempt to explain how they relate to each other. Moreover, neither really accounts for the existence of different types of exchanges, nor for the fact that they occur on certain occasions or take place between certain categories of people (e.g., rather than between undifferentiated "individuals"). To account for these configurations, one must consider Wola society as a system of rules and values. But this i s precisely what Sillitoe does not do; his reference to rules and values i s quite rhapsodic. Finally, one must point out that the fact that individuals, not groups, are the subject of exchange in Wola society does not imply, as Sillitoe believes, that the Wola have an individualistic ideology based on the notion of the "sovereignity of the individual," nor that they conceive of society in Hobbesian terms. Self-interest exists in all societies, but it i s a far cry from individualism, which implies a particular form of ideological legitimation whose existence among the Wola Si l l i toe has not, in my opinion, convincingly demonstrated. Give and Jake i s valuable, however, because of the great amount of statistical data and case studies it furnishes on exchange; from this point of view it should be taken as a model by anthropologists working with similar societies.