2. The Concept of a Lineage Mode of Production

During the early sixties, primarily because of the efforts of Claude Meillassoux and Maurice Godelier, Marxist anthropology in France began to extricate itself from the Stalinist dogma of the unilinear "five stages" of history and develop a formidable challenge to Structuralism, British functionalism, and other approaches to cultural and economic anthropology.^[8] Particularly noteworthy in terms of our present discussion were two works by Meillassoux: "The Economy in Agricultural Self-Sustaining Societies: A Preliminary Analysis," a 1960 essay (English translation, 1978) that interpreted kinship relations and the power of elders in terms of relations of production (communal subsistence agriculture) and reproduction (the control of women by elders by means of special "elite" goods used solely for the exchange of women, not for subsistence); and Anthropologie économique des Gouro de Côte d'Ivoire , a detailed monograph on the Guro of the Ivory Coast published in 1964, which provided a thorough description of the means and social organization of production and reproduction of a lineage-based, subsistence agricultural society and its transformation first by trade and then by the colonial system.^[9] The independent but overlapping work of Althusser and Godelier in the early sixties created a rigorous theoretical structure for the more descriptive and empirical studies of Meillassoux. From 1965 onward, a number of young anthropologists influenced by Althusser, most notably Emmanuel Terray and Pierre-Philippe Rey, began to apply the Structural Marxist concept of a mode of production to Meillassoux's work in order to create a framework for their own field research, and Meillassoux's subsequent development has been significantly influenced by their efforts. Taken together, the efforts of Meillassoux, Terray, and Rey have made a distinctively Structural Marxist contribution to the study of subsistence societies.

There can be no question of a comprehensive discussion of Structural

Marxist anthropology here. In the following three sections, I will take up only a few key problems and concepts most pertinent to the applicability of the concept of a mode of production to subsistence social formations organized by kinship. For purposes of exposition, I begin with the early work of Terray, which emphasizes the primacy of labor process and the forces of production; next I discuss Rey's assertion of the primacy of class struggle and the relations of production; and finally I turn to Meillassoux's later work, which, in my opinion, restores a proper balance between the forces and relations of production while simultaneously moving from a restricted to an extended concept of primitive modes of production.

Terray: The Primacy of the Labor Process

In Marxism and "Primitive" Societies , published in 1969 (English translation, 1972), Emmanuel Terray attempts to clarify the structural relationships of Guro society as described by Meillassoux. While fully acknowledging the importance of Meillassoux's work, Terray contends that Meillassoux's "general description" of the economic system of kinship or lineage communities is incomplete and too general. Meillassoux describes "self-sustaining" societies in terms of "cultivation of the soil, self-subsistence, the use of very short-term production techniques, and of human energy as the main source of power" (Meillassoux, quoted in Terray 1972, 97). For Terray, such a general concept, while in some ways adequate, suffers because it offers no "principle of variation" or structural characteristic whose historical specification would render the broad range of different kinship societies both comparable and comprehensible. To accomplish this latter task, Terray argues, it is necessary to recognize the existence of multiple modes of production within primitive social formations and to conceptualize the nature of their articulation and the social relations that "realize" or "represent" them. The key to the discovery of the dominant factor or factors in primitive societies is the study of their various "instruments of labor" from which hypotheses regarding the nature of their mode or modes of production might be derived. "It would then be possible," Terray concludes, "to analyze the concrete social formations . . . and to begin to reconstruct the relations of production of which these structures are one realization" (Terray 1972, 104).

Terray proceeds to reinterpret Meillassoux's findings by means of a detailed "inventory" of various economic activities described by the

latter (hunting, crop cultivation, animal husbandry, food gathering, and handicrafts) in order to ascertain the various functional-technical relations or "modes of cooperation" operative within each activity. These activities are then classified into two general forms of cooperation, "complex cooperation" (defined by use of a collective work implement—for instance, the net) and "simple cooperation" (teams of individuals, including the "production community" itself, performing identical or analogous labor). There are also solitary activities, such as hunting with a bow. From these labor processes Terray deduces two modes of production: a "tribal-village system" and a "lineage system." The former "realizes" relations of complex cooperation in the collective hunt and is characterized by voluntary cooperation, the availability of the means of production (nets and territory) to all, an egalitarian power structure (rotation of leadership of the hunt), equal distribution of the product after the hunt, and the dependence of non-producers (children and elders) on the producers. The lineage system, "realized" primarily in extended cooperation in agriculture, is characterized by the control of the means of production (the exercise of use rights in cultivated land, control over livestock, appropriation of necessary knowledge) and distribution of the social product by elders of the lineage.

While he insists on the articulation of both modes of production, Terray does not specify any relation of dominance and subordination between them; rather, he speaks of their mutual interrelationship or "cross-dominance." In a later self-criticism, Terray acknowledges that the tribal-village system would have been better understood as subordinate to the dominant lineage system. The elders of the lineage, having recourse to ritual, are able to "isolate hunting from other sectors of activity and thus neutralize its effects" and "control hunting, to the extent of making the hunting net a sacred object, a symbol of lineage continuity and thus under their control and surveillance" (Terray 1979, 32).^[10] The tribal-village system, Rey admits, is therefore marginal, and the reproduction of the lineage mode of production determines the limited place of the tribal-village system in hunting and in war.

For Terray, as for Meillassoux, the kinship system of the Guro is economically determined by production insofar as the former adapts to the exigencies of the latter (biological families being modified into social families to keep production units intact). Terray, however, is concerned to avoid the idea that kinship is derived directly from the economy as well as the opposite notion that it is independent of the modes of production:

I am not saying . . . that kinship relations are class relations, . . . I am saying that both are the complex result of the interplay of the economic, juridico-political, and ideological phases of the mode of production. . . . [A]nalysis . . . should distinguish:

—An economic aspect or level in which all the following are "realized": the division of the labor force . . . into kinship groups . . . corresponding to . . . production units; the division of the means of production between these units; the organization of consumption units; the structures of direction and control of production.

—A juridico-political level in which the following are "realized": the determination of the personal status of individuals; the regulations governing property and inheritance; relations of authority and their effects on the formation of those organizations . . . which ensure the smooth running of social life in general.

—An ideological aspect or level in which the ideological conditions for the functioning of the system are "realized." (Terray 1972, 144-45)

Terray relies here on a vulgar pluralist interpretation of the concept of overdetermination to avoid both economism and politicism: "concrete kinship relations must be seen as the product of a three-fold causality [economic, juridico-political, ideological] operating upon a given substratum [Meillassoux's "genealogical base"], as the combined effect upon it of the action of the three phases of the mode of production" (Terray 1972, 143). While Terray is correct to defer determination in the present to the matrix effect of the previously existing determinations of the complex whole, his explanation is inadequate to the extent that he fails to acknowledge a hierarchy among determining effectivities in either the past or the present. Terray also rejects Godelier's claim that kinship may function as a relation of production seemingly unaware of the theoretical advantages of Godelier's distinction between the function of economic ownership and the apparently non-economic character of the institution that exercises it. Godelier's framework, as we have seen, allows us to formulate a hierarchy of determinations within the framework of overdetermination. Terray's use of the term overdetermination, by contrast, cannot move beyond the unhelpful truism that everything causes everything else. Terray's methodology, admirable both for its insistence on the specificity of concrete labor processes and for its demonstration of the relevance of Marxist concepts to lineage-based, agrarian societies, cannot fully escape the charge that it has analytically separated the forces and relations of production at the expense of a satisfactory account of their articulation as a mode of production.

Terray considers the control over the means of production and the distribution of the social product exercised by elders to be cooperative

rather than exploitive. The relations of production, or property ownership, are interpreted as egalitarian, with elders having little or no power beyond the necessary administrative task of coordinating production and distribution. According to Terray, power is "vested" in the elder as a "representative of the productive community." Similarly, Terray interprets the elder's control over elite goods—goods such as gold, loincloths, ivory, and guns, which serve as a medium of exchange for nubile women—in terms of functional-cooperative responsibility rather than exploitive power. Finally, Terray views the elder's control over the distribution of women as functionally subordinated to their control over production: because elders control production, it logically follows that they would be granted control over reproduction as well. These positions place Terray at considerable variance with Meillassoux, who grants elders real power and locates it in their determinant place in reproduction rather than production. According to Meillassoux, the low level of productivity and the overwhelming importance of human energy in self-sustaining agriculture mean that human beings and the physical reproduction of the production unit are the axis of the entire production process. Control of the economy by the elders is objectively real, Meillassoux insists, but it operates indirectly, through control of the producer rather than the means of production, and is achieved via control over eligible women rather than over the producers themselves. While Meillassoux refrains from labelling the elders' power over women exploitive, he does view this power as the fundamental contradiction within "self-sustaining" societies and the source from which class relations might eventually develop. Terray concedes that elders possess a limited amount of power—they can delay the moment when the junior can have a wife and enter the lineage system—but because they cannot refuse it altogether, and because juniors eventually achieve independence, Terray insists that the elders' control over women is ultimately cooperative.

Rey: The Primacy of Class Struggle

Pierre-Philippe Rey (1971; 1975) rejects what he calls the "technicism" of Marxism and "Primitive" Societies , that is, Terray's reflectionist interpretation of relations within the primitive economy and his emphasis on the reciprocal nature of life-cycle relations in both economic production and the circulation of women. According to Rey, Terray's methodology masks the existence of power within lineage-based societies;

power is rendered invisible because Terray privileges relations of cooperation (the technical division of labor) over relations of exploitation (the social division of labor). Because of his emphasis on relations of cooperation, Terray's concepts of tribal-village and lineage modes of production are static rather than dynamic in outlook and unable to specify the contradictions within primitive societies or the internal logic of their development. On the basis of his own fieldwork in the Republic of the Congo, Rey strongly asserts the primacy of the relations over the forces of production within a lineage mode of production and the exploitive, rather than reciprocal, nature of the relationship between elders and juniors.

Rey agrees with Meillassoux that elders control the means of production by controlling the means of reproduction (and in the case of Rey's particular social formation, slaves), but unlike Meillassoux, Rey insists that this power constitutes class power. For Rey, elders form a ruling class not simply because they control surplus production without providing any corresponding productive labor but also because they use that surplus, or at least a large proportion of it, to insure their appropriation of future surpluses. This condition, Rey argues, can only be understood as exploitation: "Exploitation exists when the use of the surplus product by a group (or an aggregate) which has not contributed the corresponding surplus of labor reproduces the conditions of a new extortion of surplus labor from the producers" (Dupré and Rey 1980, 149). Elders, as a class extending across the exchange networks of several lineages, control the destiny of all juniors, as a class, by controlling the timing of their access to wives and, among the peoples studied by Rey, by imposing a constant threat of slavery should juniors violate existing social rules (transgressions by juniors must be "redeemed" in elite goods controlled by elders, who can always refuse to advance them and thereby reduce the transgressing junior to slavery).

Terray (1979) has come to accept Rey's position on the class nature of lineage modes of production, contending, however, that women and juniors are classes "in themselves" not classes "for themselves." Terray argues that the "naturalness" of the exploitation of women and juniors (strengthened by its apparent derivation from life cycles and biological distinctions), the low level of exploitation in "self-sustaining" societies (which produce so little surplus), and the strength of the "vertical" solidarity of a lineage (vis-à-vis other lineages) over the "horizontal" solidarity of class positions (within a lineage) all combine to prevent either women or juniors from ever acting as a class. Meillassoux (1981), by

contrast, continues to reject Rey's designation of the powers of men over women and elders over juniors as class power because these powers are based on stages of the life cycle, not on distinctions of social status. Older women and widows, Meillassoux points out, acquire a status more like a male, while juniors will become elders themselves with the passage of time. According to Meillassoux, the elder-junior relationship is better understood as a patron-client relationship of domination between individuals rather than as a class relationship of exploitation between groups of individuals.

To Meillassoux's objections, Rey simply responds that the question of class exploitation turns on structures and not persons—on the institutional existence of economic exploitation and not on the basis of such exploitation in age, gender, or status. Rey also emphasizes that the forms of exploitation in question are based on factors that extend beyond the life cycle: few men and no women will become elders, while an initial delay in obtaining a wife constitutes a permanent, not a temporary, limitation of future status. The fact that he defines exploitation in terms of its structured existence and not in terms of particular agents, genders, or phases of the life cycle would seem to give Rey the best of the argument. If all men and women of a certain age became elders and if the existing body of elders was unable to control "succession" by controlling the future status of juniors, Meillassoux would perhaps have a stronger case, but then, of course, Rey's definition of exploitation would no longer apply.

Rey also strongly asserts the primacy of the relations of production over the forces of production or, as he puts it, the primacy of "relations of exploitation" over "relations of cooperation." He insists, against Terray's initial view of relations of cooperation as both dominant and neutral, that "relations of production are social relations of production. As such they include both relations between the direct producers and their exploiters and the relations between the direct producers themselves: on the one hand relations of exploitation and on the other relations of cooperation. . . . Within the relations of production, the relations of exploitation determine the relations of cooperation because . . . class struggle does not only involve the struggle of the exploited against exploitation [but] also involves the struggle against relations of cooperation which are themselves dominated by the relations of exploitation" (Rey 1979, 42). Rey goes on to distinguish the "labor process" from the "valorization process": "the process of production is . . . the unity of the labor process and the valorization process, in other words,

the unity of the labor process and the creation of surplus value. The creation of surplus value, or the valorization process, determines the labor process" (Rey 1979, 43).

While Rey is correct to assert the primacy of the relations over the forces of production, his defense of this position is unacceptable from the perspective of Balibar's mode of production analysis. As we have seen, Balibar maintains that the forces and relations of production are a unity—a double connection (a property relation and a labor process) of the same three elements (laborer, non-laborer, and the means of production)—that defines the class struggle (class interests and powers). Rey, however, isolates the relations of production (laborers and owners) from the forces (laborers and other laborers) and then identifies the relations with class struggle (relations of exploitation) and the forces with mere technical organization (relations of cooperation). By refusing to recognize class struggle as constituted by both the forces and the relations of production, Rey slips back into a false opposition of economism and class struggle and transforms Balibar's concept of the primacy of the relations of production into a voluntarist assertion of the primacy of class struggle. Rey empties the forces of production of all significance, reifies class struggle, and lapses into a teleological voluntarism wherein class struggle is always somehow in advance of the productive forces. "Since the labor process itself . . . is the bearer of relations of production . . . to which they are adapted, as are all the relations of cooperation," Rey explains, "the labor process . . . cannot . . . be in advance of the relations of production. . . . Within a given mode of production the labor process and the productive forces are always slower to evolve than relations of exploitation, in other words than the class struggle" (Rey 1979, 43-44).

For Rey, the dominance of relations of exploitation over relations of cooperation is purely external. Because the relations of production have no structural significance for the labor process, the class struggle is reduced to a voluntaristic struggle for liberty. By contrast, Balibar defines exploitation as the internal effect of the unity of the forces and relations of production and the subsumption of the labor process to the reproduction of the relations. For Balibar, the dominance of the property relation is embedded in the very structure of the forces of production and the class struggle is inseparable from the uneven development of both the forces and relations and their relative correspondence or non-correspondence (the extent to which developing productivity facilitates or undermines the reproduction of the existing property relation). To

imply, as Rey does, that class struggle is independent of the productive forces, or worse, that it is always in advance of it, makes no sense at all. The object of revolutionary class struggle is not to liberate existing relations of cooperation from the tyranny of relations of exploitation; its object is to transform them both, that is, to transform the mode of production itself.

Meillassoux: Toward an Extended Concept of Lineage Modes of Production

In addition to their disagreements over the presence or absence of exploitation in lineage-based modes of production, Terray and Rey differ as to the source of the elders' control over eligible women. Terray, at least in Marxism and "Primitive" Societies , argues that elders' control over eligible women stems from their role as coordinators of production and distribution: because they are representatives of the lineages within the economy, elders are entrusted with responsibility for the distribution of wives and allowed to monopolize elite goods. Rey, by contrast, maintains that it is because elders have control over the circulation of women that they are able to exercise control over direct producers and extract economic tribute from them in the form of elite goods. Terray appears to be defending a form of "economism" in which economic control determines political control (although such control is understood to be non-exploitive), while Rey appears to take a more "politicist" position asserting the primacy of political control over economic control (and that such control is, in fact, exploitive). The entire opposition, however, stems from a mistaken assumption, shared by both Terray and Rey, that the concept of a mode of production is restricted to its labor processes and property relations, thereby excluding those political and ideological relations that provide for their reproduction. Omitting the structures of reproduction from the concept of a mode of production leads only to confusion regarding the nature of economic determination and class struggle. The correct approach, which we have already defended at the theoretical level, is to develop an extended concept of a lineage mode of production.

Claude Meillassoux achieves something very close to such a concept in his 1975 book, Maidens, Meal, and Money (English translation, 1981). Where Terray and Rey separate the economic-productive and political-reproductive powers of the elders in order to initiate a misleading debate over the primacy of one or the other, Meillassoux steers

the discussion onto the more fruitful ground of their interrelationship. For Meillassoux, as for Engels, "the determining factor in history is, in the last analysis, the production and the reproduction of the immediate essentials of life" (Engels, quoted in Meillassoux 1981, xi; my italics). Although he does not use the term, Meillassoux develops what amounts to an extended concept of a "domestic" mode of production composed of (1) the forces and relations of production: a labor process of "self-sustaining" agricultural "productive cells" whose collective "ownership" of land and tools is organized by patrimonial relations of kinship; and (2) a "domestic community" of productive cells organized, under the political control of the elders of each cell, into an exchange network involving elite goods and nubile women, a system established for and maintained by "the ordered manipulation of the living means of reproduction, that is: women" (Meillassoux 1981, xiii).

The restricted concept of such lineage-based societies, the forces and relations of production proper to "self-sustaining" agriculture, defines the given historical level of economic development. But precisely because of the low level of productivity, almost exclusively dependent on the unaided strength of the human individual, the number of people in each productive cell is always lower than the minimum required to ensure endogamous reproduction. Therefore, the problem of providing for sufficient availability and exchange of nubile women between individual production cells assumes paramount economic significance, according to Meillassoux. The relationship between labor process and property relations in the domestic mode of production is thus determined by the low level of productivity but dominated by the importance of reproducing human labor power. In the context of low productivity, Meillassoux concludes, the "clustering of [production] cells and the alliances between them depend less on requirements of production and exchange than on the imperatives of reproduction. . . . If there is a mode of production it is here, in this gathering of productive units, organized for reproduction" (Meillassoux 1981, 14).

Every mode of production, Meillassoux maintains, has its own "laws of population" not to be considered apart from the forces and relations of production. The growth of population is governed by constraints other than the fertility of women. Meillassoux distinguishes between two "primitive" modes of production: hunter-gatherer bands and agricultural communities. For hunter-gatherer bands, land is the "subject" of labor, and productive activity consists of taking from the soil what is naturally there. Relations of production and distribution

are more or less continuous but require little permanent or integrated social activity. Society is constituted and reconstituted around the free movement of adults between bands, and the positions of individuals within the band are voluntary, unstable, and reversible, what Meillassoux calls relations of "adhesion." The low level of investment and short duration of productive activity mean that social relations tend to be defined in terms of present participation in the common activities of production and consumption. As a result, kinship relations are not very significant, and terms such as brothers, sisters , and fathers refer to men of the same age, all nubile women, and men who can no longer hunt—not to lineage or kinship. Mating is loosely organized, but there is little concern for "filiation," the elaboration of social "relations of dependence" following from marriage and children.

In agricultural communities, by contrast, land is the "object," not the "subject," of labor. Labor must be invested in the soil in a continuous chain of successive tasks that keeps the producers together throughout the agricultural cycle and requires, among other things, sufficient surplus to support producers until the harvest. In contrast to hunter-gatherer bands, agricultural communities are characterized by relations of "filiation" rather than adhesion, a difference Meillassoux explains in terms of their radically different ways of exploiting the land. Agriculture encourages the formation of permanent and indefinitely renewed social ties and the circulation of goods between generations, whereas the solidarity created by agriculture arouses concerns linked to the physical and structural reproduction of the group. Meillassoux contrasts two possibilities for satisfying the new reproductive requirements: "gynostatic" societies, in which women stay with their communities and men come to join and procreate, and "gyneco-mobile" societies, in which women are exchanged between allied communities. In the former, the reproduction of the group rests entirely on the reproductive capacities of the women born within the group; in the latter, reproduction depends on the political capacities of the communities to negotiate an adequate number of women at all times. Meillassoux points out that gynostatic and gyneco-mobile relations tend to be mutually exclusive, the former associated with matriliny (a woman's brother having authority over children) and the latter with patriliny (a woman's husband having this control).

Meillassoux advances the hypothesis that the distinction between matriliny and patriliny corresponds to the relation of dominance that obtains between hunter-gatherer and agricultural modes of production

within societies characterized by an articulation of both. Agricultural production requires the continuous presence of producers but relatively few people; thus with the development of agriculture, the household acquires a social and functional existence. However, where hunting-gathering dominates agriculture, households remain small and gather through the mediation of the hunters of each household. These societies are relatively unstable, but their instability is between households, not independent individuals. Matrimonial relations tend to be gynostatic but require occasional abduction and warfare. Abduction, Meillassoux adds, "encapsulates all the elements of the enterprise of inferiorisation of women and anticipates all the others" (Meillassoux 1981, 29). It involves, paradoxically, a dependence on men as fighters not simply because of their superiority as fighters but also because of their inferiority as reproducers: men are more expendable. War is also, of course, the means by which adult men affirm and reproduce their superiority over women. Political activity, however, continues to lack cohesion, and continuous coercive authority does not extend beyond the household.

When agriculture is dominant—that is, when a sufficient agricultural surplus exists to support the community during the entire year with only limited need for supplementary hunting and gathering—one tends to find a corresponding difference within the sphere of reproduction and political authority. Agriculture is dominant, Meillassoux contends, not only because it receives most of the producer's energy but also because it determines the general social organization to which other economic, social, and political activities are subordinate. Other activities (war, hunting, and so on) predominate only as long as they last and then only within their own field of action. Men are needed in the fields, and therefore there is pressure toward non-violent marriage regulation. Relations between communities are more conciliatory and less predatory. Marriage relations tend toward patriliny and gyneco-mobility because adjustments to demographic and productive needs are more difficult under matrilineal-gynostatic systems. Thanks to the mobility of women, the group's reproductive capacity no longer depends on the number of women born within the group but on the political capacity of the leaders to negotiate them into the group.

With the existence of agriculture and patriliny there develops a stable and continuous lineage-based system of political authority of male elders over male juniors and of men over women. Meillassoux argues that because constraints on productivity are constant over time, reproduction acquires a dominant position within the social formation.

Gradually, political authority comes to dominate economic authority as the agrarian tendency to gerontocracy—stemming from the indebtedness of younger generations to those who came before—merges with the expansion of patriliny and the elders' control over women. Women are doubly exploited by these developments: it is through women that elders maintain authority over juniors, but it is also through women that juniors emancipate themselves. Pubescent women are exploited in both their production (which is turned over to the husband) and their reproduction (progeny are always controlled by men). Meillassoux not only provides a provocative explanation of the origins of patriarchy but also successfully links social reproduction and political authority to the forces and relations of production on which they rest. From his analysis of a lineage mode of production, Meillassoux shows that "controls over social reproduction are variable, they depend upon built-in political capacities of the society, on the strengthening civil power [the state]"; he also shows how the control over social reproduction "is based on the relations of production it seeks to maintain" (Meillassoux 1981, 33).

Despite the formidable obstacles to economic development in lineage-based societies, it would be a mistake to see them as without internal contradictions. For both Meillassoux and Rey, the general contradiction of a lineage mode of production is expressed in the tension between economic development and population growth on the one hand and the accumulation of power in the hands of lineage elders on the other. Leaving aside their differences over the class nature of the control exercised by the elders, there is general agreement between Meiltassoux and Rey regarding the significance of the contradiction between the expansion of the elders' power and the development of the domestic community to the point that it is sufficiently large and prosperous to reproduce itself from within. New technologies, crops, relations of cooperation, and so on not only improve the economic and demographic situation of lineage societies but also introduce a threat to the elders' power if the functional utility of that power in production or reproduction is brought into question. Rey (1979) gives a firsthand account of how elders among the Gagam attempt to control innovation (such as the introduction of yam and pea cultivation) and subordinate it to the existing social division of labor (in the production of millet, which is controlled by elders) while juniors or women struggle to preserve and expand the cultivation of such crops and bypass the intervention of the elders.

Meillassoux advances the hypothesis that increased production and

population growth create the potential for an internal dissolution of a lineage mode of production and the emergence of "seigneural" lineages possessing "real" class power over "dependent" lineages: that is, "the dominance of entire, organically constituted communities which endow all their members , irrespective of age or sex, with prerogatives and privileges over all the members of the dominated communities" (Meillassoux 1981, 81). Meillassoux acknowledges that we have no historical example of the evolution from a subsistence-kinship society to a tributary, feudal, or slave mode of production. He recognizes that the possibility of internal development is limited by the small surplus produced in kinship societies, and he admits that the known cases of transformation of lineage societies have been the result of articulations established by conquest, migration, or trade between lineage modes of production and other modes. However, Meillassoux contends that it is at least plausible that a domestic community might develop to the point that the power of the elders over women and juniors is threatened, in which case it is also plausible that such authority might be successfully maintained by coercion and thus gradually transformed into hereditary authority of one distinct branch of an elder's lineage organized by primogeniture. Segmentation of the community, the traditional means of resolving the contradiction between demographic expansion and politically enforced endogamy, might produce a similar outcome, Meillassoux suggests, if the centralized control of the original elder's lineage is not segmented also.

Such a hypothesis regarding the transformation of a lineage mode of production presupposes considerable "primitive accumulation" of wealth and power into the hands of the elder and his clients. Without conceding the class character of this accumulation, Meillassoux speculates that it might originate in the circulation of elite goods (goods exchanged for women) and in the elder's control over the production and possession of such goods. Elite goods should not, of course, have any exchange value outside of the circulation of women, but the entry of material and durable goods into marriage transactions—objects that continue to exist after women are "consumed" and that may be accumulated independently of any "woman standard"—introduces another contradiction within the system and the potential for change. Bridewealth is at least physically capable of entering into other exchange circuits besides that of marriage. If elite goods become the means to free oneself from other obligations, that is, if they come to acquire some degree of general exchange value, the possibility for domination by pro-

ducers or "owners" of such goods over non-producers and non-owners exists. Class domination, Meillassoux concludes, might result from monopoly control over the production and circulation of bridewealth goods by one lineage, the accumulation of exchange value and power in the hands of this lineage, and its subsequent use of force to strengthen its control over marriages and women and to protect itself against the possibility that its particular form of bridewealth "currency" might be "bypassed" by a new medium of exchange.

Imperialism as an Articulation of Capitalist and Lineage Modes of Production

An extended concept of a lineage mode of production provides a viable, historical materialist explanation of "self-sustaining" agrarian social formations organized by kinship that are often considered beyond the purview of Marxist analysis. In this section I will demonstrate that the concept of an articulation of two modes of production, defined extensively and restrictively in terms of dominance and subordination, provides an equally useful explanation of the transformation of lineage societies by capitalism. I will confine my discussion to the analysis of imperialism in Africa advanced by Rey and Meillassoux. Rey is of particular interest in this regard because of his comparative analysis of the transition to capitalism in West-Central Africa (from the era of the slave trade to the post-World War II era) and in Western Europe (the original transition from feudalism to capitalism from the sixteenth to the eighteenth century). Rey's argument, put forward in the massive monograph Colonialisme, néo-colonialisme et transition au capitalisme (1971) and the more speculative essay Les alliances de classes (1973), turns first on the structural differences between exploitation and exchange in feudal and lineage modes of production and the possible articulations of each of these modes with capitalism and second on the relations of dominance and subordination that obtain at a given conjuncture—specifically, the relative degree of development and thus the relative power of capitalism.^[11]

In contrast to Rey's work, Meillassoux's reflections on the articulation of capitalism and pre-capitalist modes of production pertain more to the contemporary global economy and to debates over the nature of "underdevelopment" in the Third World. In contrast to "dependency theory," which conceptualizes capitalist imperialism in terms of impersonal market relations between unevenly developed but nevertheless

capitalist sectors (metropole and periphery) of a global economy, Meillassoux views the global economy as a structured whole composed of distinct yet integrated modes of production dominated by capitalism. According to Meillassoux, the domination of capitalism results not from unequal exchange mechanisms stemming from differences in regional productivity (although these become increasingly significant as pre-capitalist modes of production are eliminated from the periphery) but from specific modes of domination and "super-exploitation" that can be fully understood only when the indigenous pre-capitalist mode of production as well as the complex structure of its articulation with capitalism are clearly identified.^[12]

Rey's argument may be briefly summarized. He contends that lineage modes of production, in contrast to feudal societies, are based on indirect rather than direct exploitation. For Rey, the primacy of marriage or elite goods in lineage societies—the fact that the class power of the elders is maintained by exchange mechanisms one degree removed from the forces and relations of production—creates a fundamental obstacle for capitalism as the latter attempts to integrate lineage societies (and the labor power and materials they supply) into its global system of commodity exchange. Lineage elders have relatively little interest in expanding the production of subsistence goods, Rey points out, and relatively weak mechanisms for accomplishing such expansion should they desire to do so. However, Rey argues that when capitalism (1) desires a greater degree of productivity than the lineage mode of production can or will provide and (2) attains the technological capacity to impose its will on the lineage mode, then capitalist relations will be "implanted," usually by violence, and forcibly articulated with the indigenous pre-capitalist mode of production in such a way as to destroy the autonomy of the latter and subordinate its reproduction to the reproduction of capital.

In Colonialisme, néo-colonialisme et transition au capitalisme , Rey charts the course of the articulation of capitalist and lineage modes of production in the Congo-Brazzaville region (now the Republic of the Congo). According to Rey, the exchange of slaves for manufactured goods characteristic of the pre-colonial period actually reinforced the lineage mode of production since it operated through the elders' traditional control over the double circulation of juniors (as slaves) and women (as brides). Chains of exchange between chiefs produced a flow of slaves from the interior to the coast and a counterflow of European "elite goods" from the coast to the interior. At the coastal end of the

chain, population and wealth became concentrated, while at the interior end, social formations lacking sufficient lineage organization and military power to become part of the chain of exchange became its victims and suffered steady depopulation and immiseration. All along the chain, chiefs and their lineages accumulated women and European goods and passed along "superfluous" males toward the coastal kingdoms of Ngoyo, Kakongo, and Loango, which controlled the three major slave ports. This type of exchange between "mercantile capitalism" (which I call "feudal commerce" for reasons made clear later in this chapter) and a lineage mode of production is characteristic of what Rey calls the "trade era" of imperialism. The case at hand did not really involve an articulation at all, he argues, since both modes of production remained largely autonomous: exchange took place within the framework of normal circulation for each mode, and the relations of production and reproduction were not structurally modified in either case. The total volume of exchange, however, was limited by the fact that lineage societies exchange only to satisfy the elders' needs for elite goods.

As the needs of capitalism increase (along with its power to impose its will), relations of "reciprocal exchange" between autonomous modes of production become increasingly unacceptable from the capitalist point of view. According to Rey, in West Africa the lineage mode of production supplied slaves efficiently enough, but when the European capitalists began to desire other products, such as gum, palm oil, india rubber, groundnuts, and so on, and to desire them in ever greater quantities, an impasse quickly developed. From the point of view of the elders, demands for economic rather than elite goods meant a significant increase in productivity, which in turn not only threatened to provoke crises of lineage segmentation and territorial control but also constituted a potential threat to their erstwhile monopoly over European-made goods. If not rigorously controlled, such economic "development" threatened to undermine the class power of the elders and the integrity of the lineage system itself. In addition, Rey notes the significance of the European processing plants that began to proliferate along the West African coast during the 1890s and break up the great kingdoms that had monopolized the three slave trading ports. The extension of European power dissolved these kingdoms into competing lineages with different chiefs "protecting" different plants (and their supply routes to the interior) while raiding those plants (and supply routes) protected by rival chiefs. The tension between the multiplication of trading ports and the needs of each for security from raiding,

coupled with the disappointing quantities of raw materials finding their way to the coastal plants from the interior, produced a change in strategy on the part of the Europeans and inaugurated what Rey calls the "colonial era" of the early decades of the twentieth century.

Unlike the trade era that preceded it or the "neo-colonial" era that succeeded it, the colonial period was characterized by coercion rather than exchange. Colonialism, according to Rey, is a political-military despotism designed to "implant" capitalist relations of production by force—building the necessary infrastructure, establishing plantations as well as market production by the indigenous population, creating a labor market for capitalist enterprises, and destroying the autonomy of the lineage mode of production. The military power and economic resources of a developed capitalist state standing behind the local coercive apparatus is required, Rey argues, to create and maintain a hierarchy of chiefs who will control land and collect taxes under the orders of a European commandant, conscript the initial labor force required to build roads and railroads and work the plantations, reorganize land distribution and property rights to appropriate land for the plantations and weaken the "self-sustaining" power of communal agriculture, and introduce money taxes and monetize bridewealth in order to compel wage labor and market production. Rey demonstrates how these policies were pursued with relentless brutality in the Congo-Brazzaville region from 1912 into the early 1920s (when military conquest was achieved), culminating with the hecatomb associated with the construction of the Congo-Ocean railway from 1925 to 1934.

The colonial period was necessary for capitalism, Rey maintains, even though it was initially unprofitable. It was the only way to "civilize" lineage-based societies, that is, to "implant" capitalism on foreign soil and establish conditions necessary for its autonomous development and reproduction—conditions that the lineage mode of production, unlike European feudalism, did not provide. After 1934 and down to the present day, in a stage that Rey calls neo-colonialism, the "free" sale of labor power and the growing sale of commodities became self-generating. As a result of the massive application of force during the colonial period, the "unity" of producers and consumers in a relatively self-sufficient subsistence economy was finally broken. Workers and products no longer have to be obtained for capitalism by force; conscripted workers who initially had money forced on them became wage earners and commodity buyers, while the men and women who remained in the villages became sellers of provisions supplying the new

labor force. Colonial despotism was relaxed to the extent that capitalism began to function according to its own laws, and the separation of the economy and the state characteristic of capitalism began to take place. By the fifties, Rey concludes, the economy of Congo-Brazzaville had been "restructured." Lineage modes of production continued to exist, but their reproduction had been subordinated to the needs of foreign-owned industry. Whatever integrity remained in the lineage forces and relations of production was preserved in order to subsidize the cost of labor power for capital and to support the "surplus" population for which capitalism as yet had no use.

The persistence of "tribalist" politics and lineage modes of production remains a "technical" obstacle to the internal economic development (the production of more use values for the local population) of the Republic of the Congo, as Rey notes at the end of Colonialisme , but one completely explicable in terms of the logic of exchange values within the global capitalist system: global capitalism is concerned primarily with the extraction of surplus value from the Congo and not with its economic development. In Maidens, Meal, and Money , Meillassoux extends Rey's analysis and argues that contemporary capitalism actually attempts to preserve lineage and other domestic modes of production in the Third World (as well as their attenuated survival in the form of the nuclear family in the First World) in order to insure a cheap supply of labor. From the "tribal reserve" system in South Africa, to migrant labor forces in the United States and Western Europe, to the unpaid labor of wage-earning parents everywhere, capitalism always seeks to exclude the cost of "indirect" wages (the cost of reproducing the worker) from the wage contract, thereby restricting it to "direct" wages paid to the worker on the basis of hours worked. Workers, however, fiercely resist this tendency, and in the First World they have some ability to do so. Despite the fact that indirect wages in the First World are heavily biased in favor of capital (they are paid for by taxes—forced savings on the part of wage earners—as well as by indirect wages from capital and are payable to workers only through the labor-regulating state), they remain unpalatable to capitalists. In addition, Meillassoux maintains that indirect wages tend to increase as the long-term logic of capitalist development breaks down the family's economic structure and forces family members out of the home and onto the labor market in order to preserve their household's standard of living. This process raises the cost of reproducing labor power and thus wages, since goods and services associated with the reproduction of labor power, previ-

ously provided by the family cheaply or for free, must now be purchased.

For these reasons, Meillassoux contends that capital is impelled by its own logic to search for cheaper labor outside the capitalist metropole: capital does not simply "react" to conditions of "unequal exchange" but seeks to create and maintain such conditions; therefore, the existence of these conditions can be understood only in terms of a globalized class struggle. For Meillassoux, the articulation of capitalism with modes of production based on subsistence agriculture and lineage relations of production means that capital pays little or nothing for the reproduction of the Third World worker. The absence of indirect wages in the Third World constitutes "super-exploitation" of labor power, a windfall labor "rent" for the capitalist equal to the difference between the cost of wage labor in the First World and the cost of Third World labor subsidized by subsistence production (less costs of political compulsion, salaries to pre-capitalist elites, costs of feeding and housing labor transported over long distances for extended periods, and costs necessary to cover the losses to the subsistence mode where the loss of manpower threatens its very existence). According to Meillassoux, the capitalist benefits from super-exploitation as long as the worker remains connected to the domestic economy, and this connection is preserved by the creation of a "double labor market" consisting of an "integrated" working class, reproduced within the capitalist mode of production and receiving indirect wages, and a "migrant" working class, which only partially reproduces itself within the capitalist sector and which receives wages so low that no integrated worker could afford to work for them. Finally, political mechanisms providing for the rotating movement of migrant labor plus discriminatory legal codes and ideological systems (racism) serve to restrict migrant access to other economic opportunities and keep the migrant worker from establishing communal ties and acquiring basic human rights.

The problem for capital, Meillassoux points out, is that the reproduction of the lineage mode of production is continually undermined under such conditions of articulation. The introduction of money into the domestic economy and the unequal productivity of the two modes of production serve as incentives for workers to emigrate. The wages brought back to the domestic mode of production are insufficient to compensate for the losses sustained by the accelerated drain of manpower, losses that can be made up only by purchases from the capitalist sector or the "largess" of the capitalists themselves. The grim outcome

of this process of dissolution of the lineage mode of production is "absolute proletarianization," a situation depicted by Meillassoux as a form of barbarism comparable to the Nazi concentration camps, which provided virtually free labor for I. G. Farben, Krupp, Thyssen, and other large (and still respectable) capitalist enterprises. Arbeit macht frei .