The persistence of numerous and recurring exceptions to free wage-labor in the contemporary world leads us to ask about the status of these “exceptions”: are they anachronistic vestiges of a feudal past or “traditional societies,” or are they a mode of the “normal” functioning of a capitalism that is otherwise firmly a part of modernity? ¹ Can we speak of modern slavery? If we are dealing with unfree forms of labor, how can they subsist in a system where “free labor” is dominant? Is primitive accumulation the prehistory of capitalism and thus not part of its proper history, or is it indeed an integral part of “historical capitalism”? It will be noted in passing that in the latter case, it is not sufficient to revert to the word “slavery” without interrogating its status: is it a complete description with explanatory aims, a rigorously constructed metaphor, or a much vaguer approximation?

An initial solution consists in tracing the ensemble (or a part) of these forms back to the history of pre-capitalism: a convenient solution, as it brings in exogenous, extra-economic factors ranging from sociological indicators to the “agential preferences” so dear to the neoclassical economists, but above all force. It has the advantage of acknowledging heterogeneity, but is less satisfactory in terms of logic, leaving more free space of analysis and explanation.

A second response also falls under the category of a value judgement, in being another argument for the idea that the most brutal kinds of exploitation persist in all settings and all periods. With regards to exploitation, there is “nothing new under the sun.” On the other hand, if this solution provides the anthropologist with a certain moral comfort – thus invested in taking a critical distance from theories of human progress through forms of labor and the economy – it possesses a number of intrinsic difficulties. In order to be credible, the argument for the persistence of slavery should avoid the metaphorical use of the term “slave,” which would quickly devalue its scientific worth: slavery needs to be demonstrated, and to do this slavery must be analyzed in its coexistence with free labor, as we will see. 

A theory of forms of unfree labor requires a theory of wage-labor. The exception must be understood through its distance from a norm, and as a result it proves necessary to produce a theory of how the two forms interact with each other, in both directions. 

We propose to show two things: on the one hand, it shall be argued that so-called “primitive” accumulation of capital takes place in a continuous, or ongoing [continuée] manner; on the other hand, in order to accurately identify the contemporary role of forms of unfree labor, it is necessary to take the same approach as found in my previous work on the constitution of historical wage-labor. ² That is, to think today, at one and the same time, the complementary coexistence of forms of free labor and unfree labor. This means revisiting the history of historical capitalism, including its most contemporary form, in light of the rise of its “classical” tendency towards authoritarian forms of extractive activity. But this also means seeking out the reason for this permanent reintroduction of the worst forms of exploitation, beyond the emphasis – ultimately tautological – on the amorality or amoralism of capitalism. We cannot understand the ferociousness of plantation-based mercantile capitalism in the periphery without taking into account the no less extreme difficulties of the proletarianization and putting to work of the “poor” [les pauvres] in the center. This double movement is indeed valuable for a clear understanding of the emergence, the trajectory, and disappearance of forms of labor that are compatible or incompatible on the same institutional level or the same market. 

Our point of departure will be the “scandal” of current forms of unfree labor, in order to move on to those forms of modern slavery and serfdom that are solidly anchored in the rise of capitalism. These “anomalies” raise both the question of “primitive accumulation” and that of the reciprocal and complementary character of systems of unfree and free wage-labor. 

Forms of Unfree Dependent Labor: The Contemporary Scandal

Contemporary globalization appears to fundamentally be a relative (sometimes absolute) reduction in the number of workers and wage-earners [salariés] in the fullest sense of the term (in not recognizing self-employed or autonomous laborers as wage-earners, whom legal scholar Alain Supiot calls “para-subordinate workers”). ³ At the same time, we are witnessing in the Global South the incorporation of a considerable part of the active peasant population into wage-labor; Chinese and Indian peasantries make up the primary, but not exclusive, part of this movement of the total growth of the working population. Such a movement seems to follow the description of the birth of European industrial capitalism in Manchester: an intense rural exodus, a clustering of the population in the metropoles, an increasing reduction of the population active in the primary unwaged sector, accompanied by a very high rate of growth. In short, a Rostow-like take-off, marking the irreversible transition from a “traditional” society to a modern society (with the emergence of the individual consumer and “maximizer” dear to microeconomics). ⁴ When viewed from the air, São Paulo, the north zone of Rio de Janeiro, Mexico, the Pearl River Delta in the hinterland of Hong Kong, all illustrate this global “Manchesterization.” If we include the informal sector, which provides services that the public sector does not, then the massive African metropolises of Cairo, Lagos, Johannesburg, and Dakar enter into this descriptive framework. 

This reassuring tableau of the evolution of human labor (reassuring since it ought to lead to development) needs to be tempered by several less uplifting facts, oft-neglected in polite indifference by international organizations and non-governmental organizations (NGOs). The latter stress that the vertiginous rise in social inequality is due not only to a lack of jobs [emplois] in the macroeconomic sense of the term, but to the lack of jobs with the guarantees provided by the International Labour Organization (ILO), working in close cooperation on these issues with the United Nations (UN).

In the general principles contained in its founding charter (the 1944 Declaration of Philadelphia), the ILO goes a long way, since it states in Point I: 

(a) labour is not a commodity; (b) freedom of expression and of association are essential to sustained progress; (c) poverty anywhere constitutes a danger to prosperity everywhere; (d) the war against want requires to be carried on with unrelenting vigour within each nation, and by continuous and concerted international effort.

The ILO’s program was very broad and articulated through a Keynesian and Beveridgean conception of full employment. ⁵ More than a half-century later, in its “Declaration on Fundamental Principles and Rights at Work and its Follow-up,” the ILO insisted more modestly on four points:

(a) freedom of association and the effective recognition of the right to collective bargaining; (b) the elimination of all forms of forced or compulsory labour; (c) the effective abolition of child labour; and (d) the elimination of discrimination in respect of employment and occupation. 

It is perhaps surprising that slavery in general or sex slavery (trafficking of women) in particular does not figure prominently among the ILO’s “recommendations.” In fact, after the abolition of slavery in Mauritania, slavery no longer had any legally sanctioned status, except in vestigial forms. On the other hand, the ILO considers the prohibition of slavery and human trafficking as falling under the 1948 Universal Declaration of Human Rights (thus the traffic in persons is mainly for forced sexual activity). The lapidary (and paradoxical, if we think about its implications with regards to contemporary globalization) first article – “labour is not a commodity” – aims to address this question in a preliminary fashion. This precaution had to be explicitly formulated as much by the French Constitution of 1791 as by the great classical liberal economist John Stuart Mill a half-century later, as well as in the debates around “rental contracts,” a transitional form toward free labor in slave economies. ⁶

This no longer appears to be an issue when the relation of waged employment finds itself qualified by jurists as a relationship of subordination between a clearly defined employer and an employee. This relationship of subordination defines, in turn, a service, its limits, its conditions, its specific period of time, and the social rights that are applicable. Labor bargaining is established as a crime. The globalization of the labor market, which tends to reduce work to a commodity, and the rendering of stable, protected work precarious, naturally poses problems for international labor standards that are careful to avoid situations of forced labor. We touch upon here an inherent difficulty to the question of constraint. There is a real difference between juridical constraint, such as forced labor (colonial corvée labor in Africa, for example), and economic constraint, which pushes the proletarianized peasant, without juridical obligation, to work as a dependent wage-laborer in the plantation economy, as an expatriate “cadet” in Europe – a rather hypocritical difference. The only way to resolve the question is ultimately found in Beveridge’s conception of “full employment.” ⁷ For the inventor of the welfare state, the criterion for full employment is not setting the population to work through economic constraint (which brings back a type of labor very close to legally forced labor) but the fact that the jobs held are “quality” jobs. The quality of these jobs is verified if an unemployed person is willing to move in order to obtain one. For Beveridge, the best guarantee this labor is not forced and is attractive geographically is that it can become the object of a collective contract. The freedom to create unions and the right of association for workers are thus the conditions for unforced forms of employment.

The question of human trafficking in connection with the sex trade might also appear to be a significant oversight in the ILO’s programs. Although it is not possible to precisely analyze its extent, prostitution represents one of the principal sources of revenue for organized crime. While persons who make a living in the sex trade selling their own body have begun to claim, throughout the world, to be recognized as “workers” in one of the most important services worldwide (the tourist industry and services) and to benefit from the same rights as workers in other sectors of economic activity, international organizations have been very reluctant to take this step. There have been some developments, but they depend in large part on the demands that are being raised. International organizations have approached sexual tourism under the auspices of the sexual abuse of minors and violence against persons.

The question of force in the case of adult prostitution is on the other hand quite ambivalent. Certain national laws target pimping, that is, the real employers of prostitutes, as well as the organized crime which supports the traffic of this particular line of work; but it is often difficult to prove physical force. The principle of having “control over one’s own body,” although important for women’s liberation in matters of contraception, limits efforts towards legislative action, and can prove that the road to hell is “paved with good intentions.” And while casual prostitution, tied to poor living conditions or poverty and often in the aftermath of civil war, is an activity necessitated by the need to survive, it does not fall under “forced labor.”

It would thus take a trial, too easy in our understanding, to denounce the blindspots of international organizations on the issue of slavery in the strict sense of the term or on the issue of sexual slavery. Especially as concerns labor and health issues, these organizations can only act with the full consent of members states and provide recommendations, not rules that would apply directly in a supranational fashion. An extremely powerful consensus of public opinion, under the auspices of crimes against humanity, was required in order to apply coercive measures to states, through the creation – and not without resistance from United States – of an international penal tribunal capable of indicting the leaders of sovereign states. The legal restrictions against sex tourism were accepted by states because of the pressure exerted by NGOs.

As an alternative, it seems more pertinent to bring out other limits of the ILO’s definition of free labor that are exhibiting a growing importance today. 

The first limit is conceptual and methodological. Although certain principles affirmed in the Declaration of Philadelphia (the right to a basic income, child welfare, and maternity protection: in short, “basic needs”) largely go beyond the sphere of dependent wage-labor, questions of domestic labor, sexual services, and childcare are not taken into account as constraining factors on activity. ⁸ However, this element is increasingly understood by feminist scholarship as being responsible for the persistence of employment discrimination, wage discrimination, and professional discrimination in the more developed countries. It is plausible to suppose that this also plays an important role in developing countries.

The second limit concerns the ILO’s almost exclusive focus on dependent wage-labor, which ends up leaving to the side forms of “autonomous, independent, or semi-dependent labor,” as much in the North as in the South. An important aspect of forced or confined labor in these forms involves – and in contradiction with the principles of the ILO – formally independent actors who exploit themselves, that is, they must give up the benefits which today come with wage-labor recognized as such. A significant proportion of live-in caregivers in agriculture, services, and micro-enterprises are managed by these “independents.” it should be borne in mind that the range of situations taken into account by defined ILO standards is far from complete. 

That being said, by focusing only on aspects of unfree labor mentioned in the 1998 ILO declaration, the emphasis is restricted to the widespread persistence of forms of domestic servitude, sexual abuse, and the recruitment of minors in countries particularly devastated by civil wars and ethnic killings (particularly Sudan). Supposedly free labor to repay debts, which is very extensive throughout the Indian subcontinent (a good 50 million people), should be included. In the latter case, the person remains formally free: they sign contracts of employment for the amount of time necessary to pay back the debt, but find themselves to be subjected to essentially a corvée, in the Medieval and not metaphorical sense of the term, since this obligation to work for the creditor is a substitute penalty for the debtor having defaulted on payment – the penal offense would entail imprisonment, and thus the deprivation of liberty. The cancellation of debt during Indian decolonization in 1947, and subsequent recurring experiments, have served no purpose, as debt mechanisms are reinstalled. Authors including Charles Gibson and A.J. Bauer have argued, apropos of the enganche system (a form of labor in which one was “hooked” to a debt contractor), that individuals chose to enter these relations of subjection, since they offered social protection and security, or a social and cultural environment, which free market relations could not do to the same degree. ⁹

Germane to this directly forced labor is the issue of child labor; which, according to some estimates, affects 246 million children around the globe. ¹⁰ As for coercive [contraignant] contract labor not involving forms of personal slavery, because the individual and/or their dependents (infants, spouses) remained confined to less valued social tasks under the effect of an institutional or social discrimination (the burakumin in Japan, the Dalits in India, the Copts in Egypt) or work and residency regulations (foreign migrants in countries with migrant labor, as opposed to population migrations), it comprises a substantial percentage of the global workforce: it suffices to think about how the current prosperity in coastal China depends on a system of internal passports, which deprives access to social rights (social security, schooling, housing, retirement) for those in the rural villages (even if they are in the process of being totally undermined) who migrate to the cities unauthorized. ¹¹ As in the USSR of old, China has its “undocumented” [sans-papiers]. They number over 100 million, a considerable order of magnitude (recall that the active population of France is 23 million). It’s also impossible to ignore the work of migrants deemed to be in “irregular situations,” according to the terminology of the states where they reside, while the International Labour Office (the permanent secretariat of the ILO) classes contract work of foreign laborers as irregular because it overrides the principle of the right to work. ¹² Of the 140 million international migrants in the world, a good 50 million find themselves in situations of legal subordination on the labor market. Foreign workers in the United States who do not possess green cards or J-1 permits, and who participate in programs of temporary seasonal work, obviously should also be placed under this rubric. 

Finally, if we account for the work of women who are compelled to do different types of work (independently of domestic labor) through traditional social structures (and of which the waged or paid portion only comprises a small part), we can estimate that around 500 million people in the world are working on behalf of others (dependent labor) but do not perform free wage-labor, abstracting from, or leaving to the side, the domestic labor of women. 

It does not appear that situation will improve any time soon – a matter for concern. The so-called countries of immigrants (specifically Australia and the United States) are severely restricting or tightening their legislation, which have started to resemble the policies of European countries; while in Africa, Asia, Central and Latin America, all the way to the archipelagos in Oceania where there is a significant degree of permeability across borders, states are in the process of strengthening manifestations of “national” sovereignty and transposing them onto the labor market under the effects of foreign and civil wars. We might recall the brutal expulsions of immigrant workers from Libya and Nigeria. Recently, in Côte d’Ivoire, populations that have settled and been active in the country for several decades have come to be brutally regarded as foreigners or undesirables, and prohibited from both paid and independent activities. This element has added to the “classical” migratory pressure feeding real, bare-handed attempts to cross the many electric barriers across fortress Europe, whether they are on the official border or displaced even further to neighboring countries, or to candidates for entry into the European Union or some privileged partner. ¹³

De facto slavery, peonage, contract work in order to pay off a debt incurred to cross borders which are more and more costly to broach, work whose geographic, social, and professional mobility is hampered, bridled by many, broadly institutional, mechanisms which form a large part of state interventions – such is the full tableau of the global labor market, which does not resemble a linear absorption of a reserve or a rural workforce into canonical wage-labor [le salariat canonique]. ¹⁴ At this level, there is no difference between the “undocumented” across the entire world: the policies of several legal systems are semi-interchangeable in their mechanisms and applications from one side of the planet to the other. 

In this sense, the dependent laborer’s struggle to win the basic liberty to freely sell their mere capacity to work to the highest bidder, as a wage-earner benefiting from the “normal” right to work without being subjected to the “special regime” reserved for foreigners, still has many days ahead of it. The freedom of residence is recognized in only a small number of countries that have generally been founded through European colonization, even if the Helsinki Accords (1977) paradoxically recognized the right to emigrate. Paradoxically, because the corresponding right to immigrate is the subject of one fraught international convention, the 1951 Geneva Convention Relating to the Status of Refugees, and only provides for an obligation to take in or admit those who can prove they have been persecuted. 

This dismal situation is not an accident of history, nor a recent incongruity. If we turn to the immediate past, or even further back, we are forced to make the same conclusion. The system of engagés and white indentured servants in the first European colonies in the New World, then the Asian coolies or “contract laborers in the Atlantic and Pacific economies, laborers in the ‘colonial corvée’ in Africa, these are the real, direct ancestors of the migrants under contract in Western Europe.” ¹⁵

But it is clearly the modern system of slavery and serfdom that constitutes the most complete example of unfree dependent labor. 

The Scandal of Slavery and Modern Forms of Serfdom

Following the works of Eric Williams, Andre Gunder Frank, Eugene Genovese, Manuel Moreno Fraginals, Immanuel Wallerstein, Sidney Mintz, and Jacob Gorender on the one hand, and those of Robert William Fogel and Stanley Engerman and Douglass C. North and Robert Paul Thomas, on the other, we now recognize the central role of the plantation economy in the rise of modern capitalism, which presupposes the birth of the absolutist modern state and mercantilist manufacturing in the 17th century just as much as the great English “factory” at the transition from the 18th to 19th centuries. ¹⁶ But what is less emphasized is the extraordinary persistence of slavery and serfdom within “liberal” capitalism all throughout the 19th century. ¹⁷ If we adopt the restrictive definition of capitalism, as only really emerging in the years between 1750–1780 (the Industrial Revolution and Parliamentary “enclosures”), then in the two and a half centuries of capitalism (1750–2000), we have to account for the 90 to 140 years of the persistence of slavery and serfdom as global and distinct legal systems in the central links of the world-economy. It was only abolished in 1836 for the United Kingdom and its empire, 1848 for France, but 1861 for Russia, 1865 for the United States, 1888 for Brazil and 1889 for Cuba, and 1907 for Zanzibar, one of the oldest sites of the Middle Eastern, then Portuguese, slave trade. For a survival of the pre-capitalist past (feudalism, or Oriental despotism, or traditional society according to the terminology), this is a very large remainder to explain. Whoever seeks to outline the dynamics of capitalism between 1917 and 1991 cannot abstract from communist socialism; mutatis mutandis, the enclave of slavery holds equal weight for liberal capitalism from 1780 to 1890.

The profits accumulated via the plantation economy prior to the Industrial Revolution and the formation of a wage-labor force [salariat] imprudently attached to it are no longer singular issues. Nor is the merchants’ and head planters’ training in the method of making “a large number of men work under one roof” – Marx’s definition of manufacture – through “brigades” enlisting women and children. 

Coffee, sugar, indigo, tea, oil, minerals, mahogany, and ebony were produced throughout the liberal industrial age by unfree labor. We can factor in coolie labor, defined as a quasi-slavery, and which from 1820 to 1924 was used to start Californian agriculture, the extraction of gold, and maritime transport. However, migrant populations that were transformed into foreigners by the apartheid system, and found in the compounds of South Africa, have extracted gold and diamonds for half the planet. 

To cap this demonstration, let us turn to the socialist world, presented as an alternative to liberal capitalism beginning in 1917. Its rapid industrialization features the exact same dual aspects: on the one hand, wage-earners benefiting from certain advantages (guaranteed employment, the indirect wage), and thus a form of wage-labor with its limitations or constraints partially ameliorated; and on the other hand, tens of millions of forced workers in labor camps, which were slow extermination camps but also real public works enterprises, mines. 

These two sinister examples of the persistence of forms of unfree labor in the recent present and the multi-faceted past of capitalist modernity requires us to reexamine the question of primitive accumulation. 

Primitive Accumulation Revisited

In the very short chapter devoted to the genesis of capitalist relations (“The Secret of Primitive Accumulation”) in the first volume of Capital, Marx writes: 

But the accumulation of capital presupposes surplus-value; surplus-value presupposes capitalist production; capitalist production presupposes the availability of considerable masses of capital and labour-power in the hands of commodity producers. The whole movement, therefore, seems to turn around in a never-ending circle, which we can only get out of by assuming a primitive accumulation (the “previous accumulation” of Adam Smith) which precedes capitalist accumulation; an accumulation which is not the result of the capitalist mode of production but its point of departure. ¹⁸

Marx resolutely discards the false explanation of initial wealth, Quesnay’s avance primitive, which made it possible to modestly cloak its origins, or indeed to impute it to a series of plunders of all types (from the pillage of Aztec and Incan riches, to the genocide of indigenous people) that are external to capitalism. Capitalism’s functioning would thus not be contaminated by this ghastly factor, as an economy that claims to be neutral or unconcerned with what it calls the “formation of agential preference.”

Under capitalism proper, “the silent compulsion of economic relations sets the seal on the domination of the capitalist over the worker.” Political violence provides, then, the initial drive without which the historical development of capitalism would never have been set in motion, but it becomes invisible (Marxists) or superfluous (neo-classical economists). Marx rejects the explanation that begins with money, because a pile of gold or cash only becomes capital when it confronts, as part of the means of production (that is, the condition of labor), labor stripped of the means of production, which becomes the condition of capital (with the potential for surplus-value). Otherwise, this money ends up as rent or income, but not capital. This is the class relation (of two antagonistic classes) that transforms money into capital.

Yet the problem of the origins of capitalism still remains, and on two levels. First there is the problem of origins or, if one prefers, the odor of the money that must be there in order to become capital. In the encounter between the proletarian and the money-owner, where does the latter’s money come from? Of course, money does not have an odor or smell. The profits from the slave trade and the interests paid to the traders from Nantes, Bordeaux, Le Havre, or Liverpool by the almost-always indebted planters, as Eric Williams has shown, would constitute the starting capital, without which it would have been impossible to transform the English poor into the proletariat in the Midlands factories. With this, the surplus extracted from the sweat and blood of slaves, like the sugar they produced, is indispensable to proletarianization. ¹⁹ The origin of capitalism is as troubled as the foundation of dynasties in Shakespeare’s tragedies – a history full of violence and fury: “In actual history, it is a notorious fact that conquest, enslavement, robbery, murder, in short, force, play the greatest part.” ²⁰ Capitalism is not Robinson Crusoe on his island, miraculously receiving all the tools necessary for his “civilized” survival. A considerable amount of wealth is required to put the poor to work, whether it is the merchant’s money or, more likely, the money accumulated by the modern state in protecting its fleets from pirates in the Caribbean, which finances the garrisons. Hence the question that torments the origins of capitalism with the Industrial Revolution: was there not a mechanism for explaining initial wealth, merchant capitalism for example? In this case, we are already in capitalism, before having entered it. The other solution is to presuppose a form of production that is no longer feudal but not yet capitalism: simple commodity production. What is inconvenient about this last solution is that it would never accumulate a sufficient amount of money [écus] prior to capital.

Let’s leave this difficulty to the side for a moment. Any founding of city or regime depends on an initial violence, which the discourse of legitimacy seeks to hide. But when Marx looks to understand primitive accumulation, he searches for a “historically determined” reason for the intrinsic logic of the system. He then advances the following explanation: for the capitalist relation to be established, there must be violence, because the valorization of money and its transformation into capital functions all the better if the separation is deepened, and thus the proletariat appears in all its dimensions. And this logic is not Hegelian in the slightest; nothing engenders itself without pain. 

Primitive accumulation is doubtless the accumulation of the proletariat. But we have only displaced the difficulty. For the primitive accumulation of the proletariat poses a further problem. In saying that this is not an easy process for capitalism, and that it must deploy the entire power of the state – for example, the English army chasing off the Irish tenants during the “enclosures,” leads us to a new difficulty: does proletarianization happen once and for all? In this case, after an initial dose of violence intended to proletarianize the landholding peasants and expel them, move them by force, capitalism would return to softer, more economic methods which hide this initial coup de force.

But let’s propose another hypothesis, at least as plausible as the first, presupposing that proletarianization must – just as capital must start from money which is not capital – start from reality. This reality covers the free poor in the center of the world-economy and the slaves or serfs in the periphery who are not part of the proletariat or working class; let’s put forth the hypothesis that this proletarianization, as a deprivation of liberty, must be reproduced on an expanded scale, then violence is no longer necessary once, but it is necessary to guarantee the maintain capitalism in a continuous fashion. And primitive accumulation becomes the continuous accumulation of the process of ongoing proletarianization. It is a matter of continuous creation, to use Descartes’s language. 

The text of chapter 26, “The Secret of Primitive Accumulation,” seems to point in the same direction we are indicating: 

The capital-relation presupposes a complete separation between the workers and the ownership of the conditions for the realization of their labour. As soon as capitalist production stands on its own feet, it not only maintains this separation, but reproduces it on a constantly extending scale. The process, therefore, which creates the capital-relation can be nothing other than the process which divorces the worker from the ownership of the conditions of his own labour; it is a process which operates two transformations, whereby the social means of subsistence and production are turned into capital, and the immediate producers are turned into wage-labourers. So-called primitive accumulation, therefore, is nothing else than the historical process of divorcing the producer from the means of production. It appears as “primitive” because it forms the pre-history of capital, and of the mode of production corresponding to capital. ²¹

Marx certainly provides us with a model of this pacification of the capitalist relation, or more exactly, this reduction in forms of exogenous violence to the sole functioning of economic mechanisms, when he analyzes the passage from absolute surplus-value to relative surplus-value. This pacification takes on a specific role under capitalism: it assures a regular supply of labor-power, whose savage and unlimited exploitation threatens reproduction. The general interest of capitalism, usually correctly interpreted by the state (including the liberal state), thus prevails over the interests of individual capitalists. But the consideration of this need to secure the reproduction of the working class comes at a cost: a slow but steady rise of the transfer costs in the wage-earner’s income. Can we say that the violence of primitive accumulation is blunted because of the mechanism of the globalization and socialization of exploitation? The problem or trouble for our purposes is that the passage from absolute surplus-value to relative surplus-value is produced within the fully deployed capitalist relation. Capitalism is already entirely present, fully equipped – the proletariat too – since the passage to relative surplus-value works or operates in step with the working class’s struggle over the limiting the length of the working day. 

If we stick to the tautological definition of the capitalist mode of production and surplus-value, which holds that only the presence of free wage-labor makes it possible to talk about the capitalist mode of production and surplus-value, we have a capitalist mode of production that is able to be recognized, isolated, and dated in space and time in the same way as Parmenides’s argument for the univocity of the One, but just as unusable and immobile as the latter. Historical primitive accumulation becomes the prehistory of capitalism. It is not part of its internal history. It is not part of the class struggle (this exclusion will have a significant impact on the way in which the workers’ movement in the core considered the struggles of the poor [pauvres] and modern slaves). 

There is nothing to see in modern slavery and forms of serfdom beyond non-capitalist modes of production (feudalism, Oriental despotism, primitive communism). Capitalism indeed has origins, but its beginnings hold an air of mystery, like the Hegelian passage from quantity to quality, or some kind of anti-Hegelian and Althusserian overdetermination. This explanation of capitalism appears weak when it is a question of discovering the real mechanisms of the historical genesis of the system of capitalist production. New difficulties are generated when it is a question of explaining the coexistence of forms of unfree dependent labor all throughout the expansion of capitalism across the globe. And yet we still impute the force and stamina of slavery in the sugarcane fields of Cuba, the coffee plantations of Brazil, and the cotton fields of the United States between 1790 and 1860 to the remnants of feudalism and transformations of the world market (a very high demand resulting from industrial and wage transformations in the core of the system); but do we resort to the same explanation for debt slavery in contemporary agriculture in Peru or India? ²²

Dale Tomich, on the basis of his study of the transformations in slavery after 1791 (the Haitian Revolution) in Martinique and then Cuba, accurately poses the underlying theoretical problem. ²³ Starting from the important controversy pitting Robert Brenner against Immanuel Wallerstein, he focuses on the impasse orthodox Marxism encounters in its response to the question of the nature of slavery within accumulation, which only repeats the vicious cycle Marx sought to escape. ²⁴ He clearly takes Wallerstein’s side in his global and systematic approach, whose central argument seems to hold to the principle that, through the circulation of money and commodity flows, and through the world market, the core of the capitalist system comes to extract surplus-value, including forms of unfree dependent labor, without the prior establishment of the canonical wage relation. This is what we have tried to show, on our part, in our work on slavery and the genesis of wage-labor. The creation of the incredibly complex institutional form of the fixed-length labor contract and wage-labor is not a formal precondition for the extraction of surplus-value, but the historical product of a struggle by dependent labor to win its freedom, and thus a social invention. ²⁵

Nevertheless, Tomich correctly emphasizes that Wallerstein’s initial impulse can lead to an “economism” and a “structuralism” detrimental to a careful empirical analysis of the terrain. The conflictual dimension and the institutional outcome – always specific – of the configurations of the primitive accumulation of the working class are at risk of dissolving into a reductive and homogeneous perspective. Isn’t it necessary to pose as a guiding methodological principle for historical and economic research that what took place in the “enclave of slavery,” or more broadly in systems dependent upon particular or “anomalous” forms of labor, as well as their evolutions, is not independent of: (a) internal forces (what we can call the governance of a system of slavery); (b) the internal dynamism of the system of free labor which is coexistent either at the global level or in the same territory; (c) the interactive effects produced between the two systems?

We have tried to show, for example, that the progressive enslavement of black people in Virginia and other North American colonies between 1620 and 1690 is inseparable from the extreme penury of the workforce in the plantations due to the escape from and rupture with the contract labor of whites (indentured servants) and blacks; that the system of contracting the workforce from the metropoles resulted in a policy of differential treatment between two types of labor; that the flight of slaves who sought to regain their freedom, specifically by demographic breach [brèche démographique] (sexual relationships and interracial marriages), led to the establishment of segregation. This segregation developed into actual apartheid (in Haiti, in North America, in South Africa at different moments) when the proportion of the black enslaved population arriving from Africa became overwhelming. ²⁶

In the case of contemporary international labor migration – like slavery, like coolie labor – the most productive methodological hypothesis is not that which considers the different forms of dependent labor (free/unfree, semi-free or bridled) as directly substitutable, but which locates how they are complementary. ²⁷ More broadly, it’s obviously necessary to think, at one and the same time, the failure to fix and put to work a proletariat at the core of the world-economy – this problem thus takes the form of the question of the poor – and the dazzling growth of the plantation economy. Colonial slavery and serfdom in Western Europe does not constitute the primitive accumulation of the proletariat (which already exists) but the working class. 

We shall try to illustrate this method of treating the juridical heterogeneity of dependent labor by returning now to Marx’s description of continuous and expanded accumulation of the proletariat. 

Proletarianization and Dependent Labor: A Complex, Nonlinear Process

On close examination, this relation which transforms dependent labor into a condition of capital presupposes a triple separation (Trenung): 1) the separation of the individual or productive unit from the means of production. This feature is generally held as a characteristic of “proletarianization,” and conforms to the definition of the pauvre given by Jean-Pierre Camus, the bishop of Besançon, in his Traité de la pauvreté evangelique (1634): “He alone is truly poor who has no other means of living other than his labour or industry, whether mind or body.” ²⁸ Since the 17th century, to be poor is to have no other means of living other than intellectual or physical faculties. The definitional precision of this passage is admirable, as it also includes the proletariat under intellectual labor, and does not fall back into the opposition between mental and manual labor. But there is another separation that plays a determinant role in the proletarianization: the separation of the individual or unit of production (family, community) from the product, which largely governs access to the market. This access breaks down into the right to sell the product of one’s labor on the market; or indeed into the simple tolerance and finally outright ban on selling the product. The very notion of the product of labor depends on this possibility. When the individual or group is denied the right and practical possibility to access the market where it can sell their activity or the product of this activity, they see the margins of freedom reduced. Regimes of slavery, serfdom, peasant production, and combined wage-labor (with another activity) are attenuated when this right is obtained, whether de jure or de facto. This is not a matter of property rights over a good or service, but the right to trade freely. One of the most historically common methods of limiting the individual to sell his or herself as a mere dependent laborer (free or not) is to forbid them from engaging in any other forms of commerce. This is why every diminished regime of dependent labor includes the right to land, money-holding, and the freedom to trade. Simple commodity production is highly dependent upon it.

We now come to the third form of separation observable in proletarianization: the separation of the mental or physical activity from the person of which it is the bearer or support (Träger, says Marx). This is what allows wage-labor to appear to be the hiring of services or capacities, as not the purchase of a person as in slavery or servitude (the latter two being described in the Code du travail as illegal subcontracting [délit de marchandage]).

Each of these conditions is open to variation, however. The separation of the worker from the means of production might concern the earth, housing, tools, or machines. A squatter who makes de facto use of a dwelling, whether it be the “allotment slave” [l’esclave mansé] (with relative autonomy and could own small plots of land), the English cottager, the maroon in the Antilles, the favelado in the middle of the city, is not in the same condition of proletarianization as Jean-Pierre Camus’s pauvre in the 17th century, the homeless [le sans domicile fixe], or the wage-earning tenant. ²⁹

The separation from the market might be formal (the slave under the harshest regime of slavery who neither has the right to hold money, nor goods of any kind, nor products cultivated or produced by him). The slaves or peasants in socialist regimes who can cultivate their own land and/or sell and trade their product on a market are less constrained than those who cannot. The wage-earner in a highly developed capitalist economy will be in a very different situation according to whether the co-product of his activity (co-product because it most often involves a joint production with an extremely complex machinery; if he were alone, he would not produce anything directly sellable on a market or requires a go-between which can be in his hands, at his disposal, or rather in the hands of the employer. This question is quite tricky in the case of the putting out system, where the merchant is a position of strength vis-a-vis the manufacturer, not because he directly controls the labor process, but because he is in a monopsony position. 

To add to the complexity, we can note that the third clause, that of the separation between labor-power and the person (only the first being the object of the purchase/sale transaction) is not the necessary and sufficient condition for the commodification of labor. Put otherwise: in order for there to be a labor market, it is not essential for the bearer of the activity to be free in both de jure and de facto terms. There exists a market of unfree, even semi-free labor (indentured servants, temporary slaves). The existence of such a market can be confirmed by empirical facts, on the one hand (commercial registers), and laws of functioning that even W. Stanley Jevons would not have disavowed, on the other: specifically, through the formation of the supply and demand of this particular commodity, in a separate fashion, through their confrontation, through the mechanism of price variation correlative to a variation of available quantities, and even of very sophisticated overall valuation mechanisms. We might even say that the labor market for slaves, indentured servants, in short, forms of unfree labor (workers under peonage, or workers tied to their employers through debt bondage) resembles more what economists and merchants have historically called a market than the very paradoxical market of free wage-labor. The existence of a free labor market does not necessarily imply that what is bought and sold there is free. ³⁰ This is a methodologically fundamental distinction when we examine the forms the labor market takes today. The fact that the latter present fluid and adjustive characteristics makes it difficult for us to infer from them a progressive tendency in workers’ freedom. 

The specificity of the labor-power commodity, which Marx so brilliantly analyzed, is the indeterminate character of its later use-value once the initial exchange is concluded, because it is within the power of the laborer to vary its implementation via machinery. But this also applies to the slave, as well as all the forms of unfree dependent labor, once the master implements a labor process in which the dependent laborer comprises a part of the conditions of work. 

Proletarianization, de-proletarianization, and re-proletarianization thus appear as the first, ongoing challenges in the class struggle. The juridical form of the money/labor transaction is not an empty form which would lay the framework for a wage-earner’s struggle over quantities and prices. The wage-earner’s struggle over prices and quantities is only one aspect of what is at stake. The market of rights and freedoms is doubtless more important in its consequences on the social costs of the functioning of capitalism. Robert Castel and Claudine Haroche have retraced the genesis of the movement of de-proletarianization and partial decommodification of the wage relation in the market order, when republican theorists of reform established the worker as possessing social rights in order to struggle against a proletarianization whose revolutionary and destabilizing effects they measured. ³¹

A fourth category should be added to the degrees of “radical” proletarianization: the dispossession of political, civil, and personal rights. We have set up the inventory of rights which are paths towards freedom in both slavery and other forms of historical serfdom. ³²

But does not the segmentation of the contemporary market display an equal complexity in the regulation of international migrants through different regimes of work cards, travel visas, and employment access? 

Through a general grid that can serve as an inventory for the degrees of freedom of dependent labor, it is easy to see that this fourfold order of conditions, often reduced too quickly to the privation of the means of production, opens onto a number of significant conditions. The table below provides a good illustration of this by restricting the column of rights to only the personal freedom of the dependent laborer. ³³

Table: The Four Dimensions of Proletarianization

Category of Dependent Labor	Separation from the Means of Production	Separation of the Product from the Market	Separation of Labor-Power from the Person	Freedom of the Bearer [Träger] of Labor-Power
Unfree Simple Commodity Producer	No	No	No	No
Free Simple Commodity Producer	No	No	No	Yes
Full Serfdom	No	Yes	No	No
Attenuated Serfdom	No	Yes	No	Yes
Modern Slavery	No	Yes	Yes	No
Enfranchised, Squatter	No	Yes	Yes	Yes
Attenuated Slavery	Yes	No	No	No
Squatter	Yes	No	No	Yes
Unfree Indebted Artisan	Yes	No	Yes	No
Free Indebted Artisan	Yes	No	Yes	Yes
Unfree Wage-Earner	Yes	Yes	Yes	No
Free Wage-Earner	Yes	Yes	Yes	Yes

The last section of this table might suggest that free wage-labor is the culmination of a process of proletarianization carried through to its end. Stripped of all the instruments of labor, deprived of all direct access to market of producers except for that of their own labor-power, split between their capacity to work and status as a person free to sell themselves to the highest offer, the wage-earner would be capitalism’s final word. This conclusion would be mistaken. Those who make the argument for “limited proletarianization” [prolétarisation restreinte], in line with our empirical observations on the hybrid and impure situations of proletarianization, have inferred that capitalism retreats before the consequences of a generalized expansion of wage-labor [salarisation]. Proletarianization can never come to an end, and this trait would be an indicator of underdevelopment. For our part, we tend rather to follow Sidney Mintz’s approach to the “peasant breach” by understanding the ensemble of forms of dependent labor. ³⁴ The path towards freedom constantly runs counter to the logic of capitalist proletarianization and often charts a trajectory that deviates from the expected trajectory of a “pure” proletarianization. ³⁵ This contradictory proletarianization is largely consolidated, at the center of the capitalist world-economy, through the construction of a wage-earner who becomes the “owner” of social rights, while access to certain goods (housing, assets, human capital) limits the classical modes of managing wage-labor and the middle classes. The precarization of the labor market (especially the attacks on the canonical model of the indefinite employment contract) and the involvement of a significant percentage of households in patrimonial governance [gouvernance patrimoine] of the financial assets of the economy reflect an attempt to combat this transformation. ³⁶

Along with dependent labor, forms of unfree labor are part of the overall economic picture of the ensemble of capitalist relations. 

– Translated by Patrick King

This text was originally published as “Formes de travail non libre: “Accumulation primitive. préhistoire ou histoire continuée du capitalisme?” Cahiers d’Études Africaines 45, no. 179/180 (2005): 1069–1092.