The conviction that capitalist production requires laborers who are not only dispossessed of autonomous means of reproduction, but are also legally free to offer their capacity to labor on the market, has been central to Marxist analyses of capitalism. Any endeavor to confront this thesis with the actual history of capitalism not only runs counter to the dominant content of traditional Marxist analysis, but also to the fundamentally optimistic Marxist philosophy of history. Notwithstanding exploitation, pauperization, injustice, and all the other evils of this historical epoch, Marx, Engels, and Marxists have also conceived of capitalism as one step further in the development of humanity.

Much like the proponents of capitalism, Marx and Marxists have explained that while direct violence against producers was constantly threatened and often applied in pre-capitalist forms of production, violence is no longer required for capitalist forms of exploitation. It is, indeed, even contrary to the successful production of surplus value, and hence profit. Capitalism, then, is conceived of as a marker of human progress in comparison to pre-capitalist forms of production, not only because of its surpassing of direct violence, but also because it brings about the social, technological, and ideological conditions which make socialism possible.

Nonetheless, while Marx, Engels, and Marxists concur with proponents of capitalism in their conception of this political-economic system as a progressive stage in the history of mankind, they are neither in agreement with the latter’s conception of capitalism as the final stage of history, nor with their explanation of its historical origin. This difference is spelled out at the end of the first volume of Capital in Marx’s explication of “primitive accumulation,” conceived of as historical prerequisite to the accumulation that becomes dominant in capitalism.

After having made fun of Adam Smith’s anecdotal narrative of the origin of capitalism, Marx explains that primitive accumulation of capital was “dripping from head to foot, from every pore, with blood and dirt.” ¹ The seven paragraphs of the chapter on “so-called primitive accumulation” contain extensive historical descriptions. However, they do not represent historical analysis as such. Instead, Marx highlights strategies and processes which resulted in the dispossession of laborers from autonomous means of reproduction, and in the accumulation of capital. While the latter was achieved through “colonialism, national debt, taxation, protectionism, trade wars, etc.,” ² the former was brought about through forceful appropriation and expropriation. All of these strategies were ripe with violence, and all of them made use of state power.

That these practices are considered to be conducive of the establishment of capitalist production, but not of its continuous reproduction, is made clear by the famous statement about violence being the midwife of any old society which is pregnant with a new one. ³ We might smile about Marx’s choice of word, since instead of using the word “Hebamme” (midwife), he entrusts historical progress to a male, designating him as “Geburtshelfer” (obstetrician). We will leave that aside. I will also leave aside the extensive debate on revolution in the course of which this sentence has, again and again, been cited. ⁴ But this statement is not only relevant for debates on revolution; it also concerns the analysis of capitalism. Having helped a woman in labor to give birth to new life, the mission of a midwife is fulfilled. Assuming that Marx chose his terms conscientiously, this also applies to the role of violence in history. And, indeed, it is in the same chapter that we are told that extra-economic violence, though exceptionally still made use of in capitalism, has been replaced by the “natural laws of production.” ⁵ In other words: the open violence of former times has been replaced by the silent force of market conditions.

Of course, neither Marx nor Marxists have conceived of capitalism as being devoid of violence. Instead, they have remarked, the site of violence has changed. It is no longer in the open, executed through the state, or – in places where the monopoly of legitimate physical violence has not yet been appropriated by the state – by open private violence. Once material conditions have forced men, women, and even children to offer their capacity to labor on the market, direct violence is no longer necessary to establish exploitation, because this result is achieved by the impersonal functioning of competition on the labor market. And it is this impersonal power which ensures the acceptance of dominance that is inherent in every capitalist labor relation. This capitalist form of violence, no longer in the open and no longer sporadic, has become the central element of the everyday life of capitalism. Its analysis forms the center of the Marxist analysis of capitalist labor, and rightly so. Nevertheless, by focusing exclusively on the violence inherent in the everyday command over the application of one’s capacity to labor, and hence over the bodies and the intellectual capacities of wage-laborers themselves, one only grasps specific historical developments of capitalist labor, albeit important ones.

I am rather convinced that Marx himself conceived of primitive accumulation as a historical phase which more or less ended when and where capitalist forms of production became dominant. Why else would he have stated that at the time of his writing, primitive accumulation had been more or less accomplished in Western Europe? ⁶ I am not opposed to re-interpretations which endeavor to make use of the analytical concept of primitive accumulation in order to grasp continual processes of expropriation as well as the extension of market structures into spheres of life heretofore outside the realm of competition. ⁷ Instead, my critique focuses on the assumption that the constitution of labor relations through direct violence and constraint is only prevalent at the fringes of capitalism, and that this will be overcome when the rationality of capitalist economic relations becomes dominant.

Just like proponents of capitalism, Marx, Engels, and Marxists have maintained that the forceful constraint of laborers is a hindrance to any strategy which aims at furthering the productivity of labor. If there was, indeed, slavery to be found in the epoch of capitalism, then this had to be conceived of as an alternative to capitalist production, albeit an alternative that was historically outdated and hence doomed to disappear. Marcel van der Linden and Karl Heinz Roth have recently labeled this a Eurocentric conception of the history of capitalism. ⁸ I would rather point to the fact that – with very few exceptions – it is not economic rationality that effects the end of forced labor. Instead, historical analysis of the concrete functioning of capitalism reveals that the end of the many different forms of direct violence in labor relations has usually been achieved by the application of state power, hence by political struggle. Capitalism has correctly been termed a “political-economic” system. In what follows I will try to explain why this should be taken much more seriously than has hitherto been the case.

Let us start out with the results of recent historical research on the history of wage labor in England. In contrast to developments on the continent, English laborers were legally free to sell their capacity to labor at a very early stage. However, until the mid-1870s – definitely long after the beginning of industrial capitalism in England – they were not legally free to end their contract at will. If they did so and their employer went to court, they had to expect a sentence in prison. Upon their arrival they were officially welcomed with a flogging, and during their stay many experienced forced labor in a treadmill. In his important work on the history of voluntary wage labor Robert J. Steinfeld is very firm in his rejection of any functional explanation of the persistent use of state violence to leave the duration of contracts to the will of employers. ⁹ He concludes that “what we call ‘free labor’ is defined essentially by the moral and political judgment that penal sanctions… should not be permitted to enforce ‘voluntary’ labor agreements.” ¹⁰ While coercion had long been a foregone conclusion in the political management of labor relations, its justification came into public doubt when suffrage was extended to more laborers in 1868. More and more members of parliament started to think that it might not be politically wise to uphold the legal inequality between employers and laborers, as far as labor contracts were concerned, by making use of penal law and penal institutions. Seven years after the extension of suffrage, the penal sanctions of so-called breaches of contract were repealed.

While corresponding developments were taking place in other states of the metropolitan core of capitalism, developments in France were somewhat different. This difference should be taken into account when debating the theoretical analysis of the political economy of capitalist labor. Before the Revolution, laborers in France, just like almost everywhere in Europe, had been under the control of the police. This subjection was symbolized in the fact that laborers had to present their livret (worker’s booklet) to the police when entering a new occupation, thereby proving that they had left their former occupation with the consent of their employer. Again just like almost everywhere in Europe, breach of contract was considered a penal offense in Ancien Régime France. The livret was hated, and it was abolished in the course of the Revolution. Although it was reinstituted at the beginning of the 19th century, it no longer functioned as a means to constrain employment-at-will as far as journaliers, meaning industrial wage-laborers, were concerned. (Instead, it was used for documenting advances of wages and the repayment of these credits. ¹¹ ) In spite of the fact that post-revolutionary development was a struggle about the removal or the preservation of revolutionary changes in society, and in spite of the numerous attempts to control and discipline workers, labor relations as such remained private in France, i.e. not under the direct supervision of the state. The political necessity to decree that the two parties of a contract were to be equal before the law was brought about by the political necessity of constituting state power after the revolutionary demise of nobility. It was a direct result of the Revolution, and had nothing to do with any economic rationality. Indeed, in France industrialized capitalist production was developed much later than in the United Kingdom. Some maintain that it started in the middle decades of the 19th century, others that it was negligible until the 1880s. In any case: there is no convincing functional economic explanation for the development of modern free wage labor at the end of the 18th century in France. ¹²

In spite of the restriction of the examples to only two national developments, it should now be clear that the actual history of capitalism does not support the assumption that the full legal and political autonomy of laborers is a fundamental requirement for capitalist forms of exploitation. Once this is established, we can start to tackle the problem of slavery.

The notion of the incompatibility of slavery with capitalist production is not only dear to proponents of capitalism, but also, at least until very recently, to most of its critics. More often than not the adherence to this conviction has been taken to be a touchstone by which to decide if an analysis can be accepted as Marxist or not. ¹³ Rather early on, however, Wilhelm Backhaus contended that Marx had to postulate this incompatibility because he conceived of the division between free wage-laborers and slaves as a hindrance to the development of a revolutionary class in the United States. ¹⁴ Of course, not only Marx and Engels but also later critics have suggested other explanations for their assertion. They have been successfully refuted by historical research.

Marx would have been correct in postulating that the enslaved workforce was exploited to an early death and therefore had to be continually replaced, if he had restricted himself to the West Indian sugar colonies. But the enslaved workforce in the United States was not only biologically reproduced, but growing. The mobility of labor, which indeed is a requisite for a capitalist labor market, was managed through the practice of hiring out slaves, as well as through the marketing of slaves. Where technology was introduced, as for example in Caribbean sugar mills, slavery was no obstacle to its competent operation. Every one of these arguments, and all the others that have been advanced in the course of the extensive debate on slavery, has been and will further be objected to. But as soon as we replace conviction by historical analysis, it is simply not possible to refute Robin Blackburn’s conclusion that “slavery was not overthrown for economic reasons but where it became politically untenable.” ¹⁵ And if some still insist on explaining the durability of slavery by pointing out that, in spite of its economic irrationality, it was sustained by the self-image of slave owners and their ladies, this explanation falters in the view of the many strategies aiming at the reproduction of a subservient labor force which differed from outright slavery only by name and legal definition. In the United States most former slaves were not only disappointed in their hopes to become landowners, somehow receiving “forty acres and a mule,” but in the decades after emancipation very many of them had to endure some form or another of “Ersatz” slavery. ¹⁶ The Black Codes, enacted by one Southern state after another in 1865 in order to effectively prevent the mobility of all black people, ¹⁷ and to legislate severe punishment for breach of labor contract, had to be repealed in the course of time. But white landowners continued to reject blacks as tenants, thereby forcing many of them to become laborers. More often than not they were denied any form of autonomous production. ¹⁸ The fact that the enticement of laborers by another employer was made a legal offence in every one of the former slave states shows that former slave owners still saw their workforce as property. The most important substitute for the legal status of slavery, however, was peonage. In spite of the fact that Congress outlawed any form of forced labor in 1867, laws sanctioning debt bondage existed in all of the Southern states. Cheap and bonded labor could also be acquired by renting prisoners. The sheer number of convictions sentenced to forced labor in the South proves the economic relevance of this practice. ¹⁹ Employers tried to prevent flight by using chains, dogs, whips, and weapons. ²⁰ Most of these forced laborers were black. However, by the beginning of the 20th century a growing number of whites also came to suffer under peonage and other forms of bonded labor.

Not only in the United States did abolition fail to do away with the reality of forced and bonded labor. So-called domestic slavery remained lawful in many parts of the world; the slave trade as well as slavery were secretly being continued; Black Codes, vagrancy laws, and especially peonage forced many a freed slave and many poor men, women, and children into jobs they were constrained to uphold, even if they would have preferred to exist in unemployed misery. In many colonies, forced apprenticeship, as well as laws against vagrancy, were used as substitutes for slavery. ²¹

When a global labor market came to increasingly replace the former slave trade in the middle decades of the 19th century, its dominant institutional form was a trade in labor contracts. ²² Even if recruiters made use of “coolie-catchers” who were known not to shrink from outright kidnapping, ²³ and even if the future laborers were not able to read or to write their name and had been deceived about the conditions of their servitude and the continent where they were going to be employed, the fiction of a labor contract to which both partners had freely agreed was upheld.

Coolies were made use of in the United States, on the Sugar Islands, in Peru, in Hawaii, in European Colonies in Southeast Asia, and in Australia. Most of them were recruited in China, India, and Malaya, but some of them also in Africa. Conditions varied in the different states and colonies, but some recurred in most coolie contracts. Officially, coolies were free to return to their home country at the end of their contract, but even if they would have been able to find a ship and pay their fare, many were tricked and forced into the renewal of their contract. In the early phases of the trade in coolies, these laborers often found themselves on plantations or in sugar mills whose owners and overseers were accustomed to slavery and saw no reason to change their manner of commanding labor. The hypothetical ability to leave their job at the end of a contract did not improve the working and living conditions of coolies while the contract lasted. From the 1860s onward, governments in the home countries tried to forbid or at least to control the trade, and some governments in receiving countries, especially in the West Indies, set up commissions to look into the realities of contract labor. The most important incentive to do so seems to have come from free wage-laborers in these countries, among them those former coolies who – for whatever reason – had not returned home after the end of their contract. They protested against the fact that the very low pay of coolies endangered their own demand for better wages. At the end of the 19th and in the first decades of the 20th century there were many places in world capitalism where unbiased observers were at a loss to detect differences between slavery and contract labor. Often coolies were forced to live at their place of work, and not only the managers of the tea-gardens in Assam but also those of tobacco plantations at the Eastern Coast of Sumatra tried to prevent the flight of their labor force by guarding them with fences, dogs, and armed watch personnel. Of the 85,000 coolies who had arrived in the modern hell of tobacco plantations in Sumatra in 1866, 35,000 had already died three years later. In 1873 planters in this colony were granted penal jurisdiction over their labor force. ²⁴

It was around that time that free wage labor had, indeed, become prevalent in the metropolitan core states of capitalism. This notwithstanding, employers could and still did make use of the state in order to prevent collective bargaining. More often than not armed forces were employed in order to prevent or break up a strike, and activists were regularly sentenced to stints in prison. It was only in the first decades of the 20th century that capitalism in the core metropolitan capitalist states started to become domesticated. ²⁵ While most authors take the development of social policy as having been vital for the transformations of capitalism, “domestication” focuses on the institutionalization of class conflict. This is most evident in the neutrality of state penal institutions in the face of strikes.

One might be tempted, and many Marxists have given in to this temptation, to describe these developments as the workings of tendencies inherent in capitalism, the most important aspect of this tendency being the fact that direct violence was no longer used as threat against the health and the life of workers, in order to make them subservient to exploitation. That this argument – as well as the philosophy of history from which it has been deduced – cannot be upheld in the face of the concrete historical functioning of capitalism has been brutally made obvious by the labor regime of National Socialism. Not only were strikes outlawed and trade unions dissolved, with their activists among the first to be imprisoned in concentration camps, but from the late 1930s onwards, millions of men, women and even children were forced to labor under severe conditions in the “Reich.” Their capacity to labor was used in private households, on farms and on ships, in institutions of the church and in factories.

And it was especially in factories that German employers disproved the conviction that forced labor cannot be used in modern industry. In spite of the fact that in some firms in the armament industry more than thirty and sometimes even more than sixty percent of the work force consisted of forced labor, ²⁶ the output of these firms grew threefold between 1942 and 1944. ²⁷ This growth was not halted when the German government decided to make available for private exploitation not only so-called “civilian” laborers from occupied countries, as well as prisoners of war, but also prisoners of concentration camps. All of these were delivered upon the request of private firms. Since forced labor had already been widely used since 1939, the application for prisoners of concentration camps seems to have been considered as falling into the range of normal management strategies. Some processes of production were adapted to these changes of the work force, for example by reducing the level of mechanization or by introducing production lines. ²⁸

The offer of concentration camp prisoners for private exploitation was motivated by the intention of the SS to further its political influence; ²⁹ it was not the result of political intentions to integrate economic rationality into the machinery of terrorism and destruction. Racism remained the guiding principle of strategy. It was, therefore, only in mid-1944 that the physical destruction of Jews was “postponed,” in order to reduce the scarcity of labor in the armament industry. ³⁰ While the percentage of forced labor, including prisoners of war, was very high in Germany during the Second World War, the percentage of concentration camp prisoners in the whole German workforce never amounted to more than 3%. Every one of their number was denied existence as an individual human being by being reduced to a number and a creature whose physical existence was permanently endangered.

When the International Court, which was set up in Nuremburg after the war, worked to investigate and punish the crimes against humanity committed by personnel of the Nazi regime, the use of prisoners for production was labeled “slavery.” Victims have also often used this term to sum up their experiences. Since then some historians have tried to compare German practices during World War II to other forms of slavery, most often slavery in the South of the United States. ³¹ While these endeavors offer insights into the specificities of the systems under consideration, they are nonetheless futile, because the abolition movement changed, once and for all, the content of the word “slavery.” It no longer refers to a status which, in the course of the 17th, 18th, and early 19th centuries, was defined by the laws of a colony or a sovereign state. Instead, it has become a political category, used to denote conditions of life and work considered to be irreconcilable with globally accepted norms of morality. The historical experience of Nazi Germany has proved that utter brutality is not irreconcilable with the functioning of advanced industrialized production.

There are two theoretical conclusions to be drawn from this degeneration of advanced capitalism to an economic system in which the extensive use of direct violence was practiced on a regular basis. Firstly: the domestication of capitalism hinges on the state safeguarding private and collective rights of laborers, and secondly: no level of economic development safeguards against the brutal use of direct violence against laborers.

It was only after the end of Second World that political struggle and political regulation resulted in what Robert Castel has called the civilization of labor, ³² and what I call the further domestication of capitalism. This, alas, was rather short-lived. Because not only has rigid competition on the world market led governments to sharpen the constraint to offer one’s capacity to labor on a very unfavorable market, but societies and governments all over the world also, once again, more or less tolerate forced labor. My definition of forced labor centers on the restriction of movement, i.e. the prevention of laborers from leaving the workplace and/or the job. Taking away the passports of foreign wage-earners implies that they are threatened with criminal procedure for lack of an identification card, and preventing them from leaving the workplace by barring doors and windows implies the threat of illness and even death. We all know numerous examples.

Political struggles against such practices of constraint have not become easier since the globalization of the labor market. They are, nevertheless, unavoidable. If it has always been analytically problematic to conceive of the violence inherent in primitive accumulation as being the characteristic of a certain historical phase of capitalism or of its fringes, globalization forces us to finally recognize that open violence and forceful constraint continue to present dangers in all phases and all places of capitalist labor regimes.

Indeed, I insist on the hypothesis that open constraint and repression are constant possibilities of capitalist labor regimes. Far from marking a certain epoch of capitalist development, violence is constantly hanging about in the wings of capitalist labor relations. It comes into the open when governments and societies refrain from decisive objection.