Migration and Forced Labor in the Social Imaginary of Southern Mozambique, 1920–1964

The Function of Work to the Colonizer: Civilizing with the Law of the Whip

There is a large historical literature that examines migrant workers, women, forced workers, domestic workers, and mine workers in South Africa.⁸ Some of these works have sought to reconstruct how residents of southern Mozambique interacted with colonial rule; others are ethnographic works that, although produced through contact with the colonial state, can nonetheless illuminate the social imaginary of the local population, outside of colonial discourse.⁹ These works provide the basis for my investigation; although their objectives are different from my own, they provide sufficient material through which to reinterpret how residents of southern Mozambique navigated everyday life within the State of Exception, and how colonial law and everyday practice actually functioned, for both colonial rulers and the subject population.

The pioneers of Portuguese colonial occupation in Mozambique, men like Mouzinho de Albequerque and António Enes, called for colonial action to be more “effective.”¹⁰ To achieve this goal, they constantly sought to link together capital accumulation, military action, and territorial occupation—which, in turn, meant imposing forced labor and controlling the movement of workers to South Africa. These initiatives were based upon legislation that called for the “protection” of the “native,” who was defined as someone who “was not domesticated by work, occupying the category of a person of a different nature, or of a great child, who must be protected and guided along the route of education.”¹¹

The difficulties of establishing this dominion immediately became evident. For one thing, the Portuguese government lacked the resources necessary to effectively occupy the terrain. As a result, they resorted to a series of chartered companies,¹² which were given vast concessions and vast administrative powers. Mozambique was divided into four territories: the area north of the Lurio River was given to the Niassa Company; from the Ligonha River to the southern banks of the Zambeze was governed by the prazos, most of which were later acquired by the Mozambique Company or the Zambezia Company; the territory between the prazos and the 22nd parallel was given to the Mozambique Company.¹³ Only a small area in Tete and the territory south of the 22nd parallel were directly administered by the Portuguese state.¹⁴ In practice, Portugal had relinquished its authority over the majority of the colony’s territory.

This weakness notwithstanding, the Portuguese took up the colonial enterprise with a new modernizing rhetoric. Portuguese officials recognized that the types of mercantilism created by the slave trade were outmoded, and demanded new economic methods, defined primarily by legislation that justified “the moral obligation to labor of Africans” as the apex of colonial civilization.¹⁵ In Enes’ words:

The state . . . should not be scrupulous in requiring and, if necessary, forcing them to work—that is to say, improving them through work . . . in civilizing themselves through work, these backwards blacks . . . a race that up to now . . . has never produced through its own spontaneous effort even a single example of civilization. . . .¹⁶

In this new context, forcing “native workers” to work signified the reorientation of slave labor toward more modern methods of accumulating wealth, thus substituting the image of the slave for one of a “free worker” or a “contract worker.” Nonetheless, this process was not linear, nor did it occur overnight, for the population did not respond in the ways that Portuguese administrators hoped. The transition from slave labor to wage labor has been extensively discussed in the scholarly literature; here I will only highlight a few comparative passages that are important for situating my critique of the Portuguese colonial enterprise. Claude Meillassoux, in the late 1970s, observed a paradox in the articulation of the capitalist mode of production within African colonial contexts. After slave labor relationships were officially abolished, most Africans returned to their previous forms of dependency, and returned as well to their previous modes of production.¹⁷ There was no way to bring people into the wage labor system, and for this reason “the domestic mode of production was paradoxically destroyed and preserved, because even though it still exists, its existence and its decisions are directly linked to the capitalist sector.” Frederick Cooper, analyzing labor relationships analogous to slavery, echoed Meillassoux’s analysis when he observed that “regular and constant work is exactly what the slave, or the freed slave, did not want.”¹⁸ Valdemir Zamparoni, for his part, in discussing the theory and practice of forced labor in Portuguese colonies, noted that “the various regulations concerning native labor were very detailed about the obligations of the natives, but extremely vague and ill-defined regarding their rights.”¹⁹ In the same vein, José Capela described the difficulty that the Portuguese had in obtaining labor from the region’s inhabitants: “In fact neither the objective nor the subjective conditions made it possible to create free wage labor. Workers could avoid providing their labor because, for better or for worse, they possessed other methods of subsistence and it was not possible to materially or psychologically attract them to the labor market.”²⁰

As all of these authors make clear, whatever the argumentative importance of debates over labor policies in the metropolis, the relations of production did not actually change very much in colonized areas. Contrary to colonial rhetoric about the local population’s vagrancy and laziness, in reality people did not see any benefits, rights, or assurances that would come from their incorporation into the new colonial system as “free workers.” This was in part because very little changed among the colonizers, who continued to treat workers as slaves. As Capela recounts, colonizers

continued to conceptualize and to treat Africans not as the citizens imagined by the Civil Code, but instead as slaves that they would always own. The first director of Public Works in Mozambique, a liberal sent from Lisbon, recounted his shock when, at the end of the month, owners came to collect the salaries of the [Public Works] employees.²¹

Ironically, or perhaps paradoxically, the “civilizing” policy that aimed to use wage labor to transform the “customs and habits” of the so-called “natives” saw no need to transform the “customs and habits” of the Portuguese colonizers themselves, who were accustomed to the rhythms of the slave trade, developed over centuries of unequal relationships. In fact, in reading many of the reports from this era, one is struck by how quickly the colonizers aimed to reinstate slavery as the only possible method of controlling the population, owing to the supposed vagrancy and laziness of the population.

These reports deliberately ignored African labor migration, and they (along with accounts, like Capela’s, which are based upon them) are incomplete. By not including migration to the Transvaal mines and the Natal sugar plantations, they ignore a phenomenon that had taken on significant proportions within the African social universe. They also ignore its importance for understanding the legal and economic rationality of the colonized population. This rationality was ignored by colonial ideology. But this rationality must be understood if we are to understand the many ways that the population responded to colonial takeover and its subsequent impositions. There were many such responses: hidden forms of protest, like evading tax payments, working slowly, deserting work with the collusion of local chiefs, or sabotage; there were also more direct responses, such as protests from rural workers, strikes in urban centers, workers joining independent African churches, or written protests from intellectuals.²² All of these forms constituted part of the imaginary among workers confronting the power wielded by colonial rulers.

My economic bias in describing both the implementation of colonial rule and the responses to it is not random. This is because I am using the term “economic” to suggest particular aspects of colonial rule, namely those which originate in the rationality and subjectivity of colonizers and administrators. The term “economic” also captures the responses of people who were affected by these measures—a set of responses that encompasses a diverse set of people, men and women, old and young, chiefs and commoners, all of whom were forced to navigate the economic measures imposed by colonial rule, and whose economic and social positioning within their own worlds was being altered by the onset of colonial rule. Looking at the economic contours of colonial rule makes clear the serious obstacles that administrators and colonizers faced when attempting to implement their “civilizing project.” These obstacles were not created by an absence of understanding among the colonized population. This was the explanation offered by the colonizers, because in their particular worldview, it was only possible to explain an absence of cooperation by citing the “uncivilized” condition of African workers, and by calling them lazy, ignorant, pagans, and so on. But the colonized population understood very well its role within the colonial project; they simply did not accept it. From this point of view, the implementation of the entire repertoire of laws and statutes that sought to control the “native” population through the guise of “civilization” represented, for the colonized population, an intolerable and illegible state of exception that was ultimately incompatible with their cognitive universe.

The Function of Work to the Colonized: Economic Dynamics of Domestic “Resistance”

As noted, migration toward the Transvaal was a longstanding phenomenon in southern Mozambique. The initial migrations were to the sugarcane plantations in Natal; with the discovery of gold in 1886, this migratory flux coalesced into a permanent migratory system.²³ The low concentration of gold on the Rand mines forced companies to focus extensively upon obtaining enough labor to mine sufficient quantities of ore that it would maintain the profitability of their extraction. To that end, they formed the Witwatersrand Native Labour Association (Wenela), which was tasked with obtaining labor from outside of South Africa’s borders in order to assuage fears among the mine owners that there were not enough workers in South Africa, or rather not enough who could be brought to the mines.²⁴ For its part, the Portuguese colonial administration viewed this migratory system as a way of bolstering its tax receipts and counterbalancing the shortcomings of its own economic model. These shortcomings should not be interpreted (as has often been the case) as the product of low mechanization and the absence of capital investments that would have spurred more efficient production.²⁵ This was part of the reason, but the economic incompetence of Portuguese colonial rule was also created by their racist convictions that the African population was lazy and unproductive, thus justifying the widespread use of forced labor to serve the economic and “civilizational” needs of the Portuguese.²⁶

Analyzing these two dynamics—on the one hand, the mobility of labor migrants going to and from South Africa; on the other hand, forced labor imposed upon migrant workers in Mozambique—alongside each other is necessary in order to better comprehend both the colonial administration and the way that people responded to it. In particular, it demonstrates the illegibility of the colonial state to the colonized population and their consequent refusal to accept the discourse it propagated.²⁷ While forced labor was legally consecrated within Portuguese colonial rule, it was never viewed in the same light by the African population, meaning that forced labor was never able to fulfill its stated function of integrating the African population within the developmental rhetoric propagated by the colonial state. Understanding the limitations of both civilizing discourse and its specific manifestations (such as forced labor) is crucial to understand the colonial state in its proper context; as noted at the beginning of the article, there is limited utility in effecting a more narrowly legalistic criticism of the measures used by Portuguese administrators to control colonial subjects. Instead, as Veena Das and Deborah Poole have argued, it is more fruitful to examine “the many different spaces, forms, and practices through which the state is continually both experienced and undone through the illegibility of its own practices, documents, and words.”²⁸

The 1899 regulations which defined native labor, for example, were understood by colonizers and officials as a body of laws that would systematize the use of native labor. In their stated purpose, these regulations are valuable for the historian, because they provide a comprehensible compendium of orders that are easily accessible, and which can consequently be interpreted and interrogated. However, shifting the point of view to examine how these laws were understood by the people affected by them shows that the impact of the laws does not neatly align with their text; instead, they reverberated in ways that were not restricted to labor itself. For example, the word chibalo, used by the population of southern Mozambique to refer to forced labor, does not only mean forced labor; it also means injustice and suffering at the hands of someone else.²⁹ Most books on forced labor in southern Africa, starting with van Onselen’s influential work on Rhodesia, simply use chibalo as a synonym for forced labor, but examining the way that people actually use these terms makes it possible to locate both chibalo and “forced labor” within the logic that generated them.³⁰ Thus, discussing “forced labor” means examining not only labor, but also the broader colonial aspiration to dominate African subjects, grounded in racist beliefs of white superiority. Examining this aspiration shows how “forced labor” could acquire a positive ideological connotation as an instrument of “civilization.” In contrast, chibalo serves as a negative expression, created through the experience of subjugation. It too does not refer solely to labor; instead, it performs a negative ideological role that indicates random abuse and unjustified suffering.

Interviews with elderly Mozambicans graphically demonstrate this conception of chibalo as suffering, beyond the formal categories of voluntary, compelled, or forced labor utilized by the administration. For example, Mahawani Khosa, a former resident of Chokwe, in Gaza province, was interviewed by Alpheus Manghezi in 1979. Asked what he knew about chibalo, he responded:

Chibalo is a very old question because it began at soon as the whites set foot in Mozambique. They arrested us and forced us to work for nothing. They would arrest someone, make them do difficult work, beat them, and pay them 100 escudos per month. We suffered for many, many years, and it’s only recently that chibalo ended. Chibalo and the palmatória³¹ arrived at the same time, and now they are finished.³²

There are three significant aspects of this interview. The first is to associate chibalo with the arrival of the Portuguese in the region; as an emic expression, the word may have existed to designate certain types of labor that preexisted colonial occupation. Nonetheless, I am not aware of any work that discusses in detail the origins of the term chibalo, nor does it appear in the dictionaries of indigenous languages of southern Mozambique, other than as a reference to forced labor. The second significant aspect of the interview is that the interviewee discusses two forms of labor, one paid and the other not. According to the official colonial labor regulations, all forms of compulsory labor were paid labor, although the various works discussed in this article all point to a much more complex reality on this issue. Finally, violence forms the central connection between these two forms of work, and was constitutional within the practice of forced labor.

Another interview taken from those conducted by Alpheus Manghezi and his team, and which clearly shows the perception of chibalo as linked to the idea of suffering and injustice, comes from Ms. Mindawu Bila, another resident of Chokwe, who lived through the Limpopo cotton colonization scheme:

I was part of the forced cotton cultivation scheme, and we suffered. We worked under the close supervision of heartless overseers, who whipped us for the smallest violation of their rules. For example, they ordered us to burn the cotton stalks after we harvested it, but some of these overseers wouldn’t give us enough time for the stalks to dry before they were burned. We worked hard but they never paid us the correct price for our cotton. As a result, we eventually felt like it was too much and so we stopped growing cotton for the colonialists. They said that we all had to grow cotton and that all the money we earned would be ours. We thought this was a good idea, and so we decided to grow cotton—to earn money [italics mine]. But we soon understood that, although we had volunteered to grow cotton, they were always watching us, wielding the whip, forcing us to work. . . . They didn’t count our work in terms of months, because this wasn’t that type of chibalo, with a specific contract. Everyone did roadwork—they were all told to work on the roads for no pay. Today, after independence, life is better because we aren’t subject to chibalo. When men came back from the mines, they had to pay 100 escudos to the régulos—“the régulo’s bread.” Yeah, all the magaizas [miners returning from abroad] had to pay 100 escudos to the régulo, and the régulo didn’t do anything for them.³³ Any miners who didn’t pay this amount were taken to chibalo while they were home.³⁴

In contrast to the previous interview, Ms. Bila points to her initial inclination to participate in the labor system created by the Portuguese, which indicates a type of economic rationale related to the process of monetarization underway in the region—“grow cotton to make money.” However, over the course of the interview, it is evident that cotton cultivation was implemented in an arbitrary and unjust manner by the colonial administration. What appeared to be a contractual relationship, of growing cotton for a certain amount of time, was instead laced with irregularities (payment of taxes and fees), including the use of violence, illustrated by the whip and the overseers.³⁵ This interview again shows the idea of chibalo as a type of injustice, first for the lack of a specific length of time that the labor required, and also for the absence of any agreement regarding wages—both of which were fundamental within the economic logic of those affected. Through this interview, we can see that the distinction between forced and voluntary work made little sense for Mozambicans, because deception was central in both, and because the arbitrary violence of the colonial authorities meant that both were understood as being part of the same structure.

One final interview taken from those conducted by Manghezi in Chokwe in 1979 is from Ms. Oselina Marindzi:

I have to tell you something about my suffering. I suffered a lot because my husband went to the mines and never returned because of the fear of chibalo. What used to happen during those times is that my husband returned from the mines for a break, and then after he arrived, they would try to take him into chibalo. This is what the colonialists did to make us suffer. After he went to Joni [Johannesburg], never to return, I stayed here alone and I had to grow cotton to pay the hut tax that they said I owed to the colonial government. We were beaten by people like Albino Mabunda under forced cotton cultivation. They beat me and ripped up my cotton documents, saying that I was a woman who only caused problems, but I was not a woman who caused problems—I was just tired of being forced to grow cotton, cotton that didn’t give me any benefits. I was being punished because my husband had fled because he feared going into chibalo. Because of this suffering, when I used to return home after working in the fields, I would always be thinking, desperately, what am I going to do? Where am I going to go? How am I going to get there?³⁶

The suffering of Ms. Oselina, aside from being associated with the absence of concrete economic benefits from growing cotton, is also associated with her husband’s absence. Nonetheless, it is important to note that, as she points out, the causes of her situation are in fact based in the system of forced labor imposed by the colonial administration. Here chibalo acts as a disintegrating force, something that is cited in many other interviews. Luís Covane, in his work on the migration of workers to the Transvaal mines, shows how this system worked to divide families from agricultural communities, which they had to flee owing to the threat of chibalo. In this sense, chibalo was not only terrible because it was violent and arbitrary, but also and especially because it brought severe long-term consequences to the social structure of previously productive agricultural communities.

Disaggregating the many valences of chibalo from its ostensible synonym of forced labor makes it possible to more clearly understand that thing which historians habitually label as the “colonial state.” The colonial state is often accepted as a fact. This is itself based upon the self-presentations of colonizers and colonial officials, who presented their various laws and regulations as part of a universal logic. I instead propose moving toward a different heuristic, one which distinguishes between the self-proclaimed universal norms and the constant specific exceptions of the colonial administration, and which analyzes these twin poles as indivisible, continuous, and foundational. Thus, the “emergency” character of colonial rule, established to deal with an “exception,” served to justify a normalized set of administrative practices; in the case of Mozambique, a colonial administration established to deal with the exceptional problems created by so-called “lazy natives” was able to justify forced labor as the vessel of civilization—and, not coincidentally, affirm its own aspiration to sovereign power. Shifting the analysis to consider forced labor from the perspective of the African population reveals the illegibility of Portuguese laws and consequently the deceit within their claim to rational universality. This shift, in turn, makes it possible to adopt a less condescending interpretation of colonial rule; rather than analyzing the colonial administration from within its own field of vision, it becomes possible to understand it from the perspective of the colonized population as an external, extractive mechanism.

Shifting the perspective away from colonial laws, and into the perspective of the colonized, is not just useful in better understanding the colonial state; it is also essential in better understanding the social imaginary of colonized population. To that end, it is worth exploring the imbrication of forced labor with preexisting processes of social mobility, and using that imbrication as a lens through which to understand how colonized populations understood their labor. Numerous analyses have argued against the possibility of such an analysis, presenting mobility is an essentially involuntary response to capitalist impositions; as Jean-Paul Guademar has argued, for example, once “men surrender their behavior to the demands of capitalist growth, the capitalist strategy of mobility is at the same time a strategy of forced mobility.”³⁷ I want to move beyond this argument; even if this mobility is in one sense forced, because it is inevitable, it also sets off other processes that readjust the migrant’s social universe—readjustments which are not solely reducible to the political, economic, and juridical subordination of migrants.

In that sense, I think it is worth discussing one of the most interesting strategies drawn up by a specific group of individuals to confront the arbitrary and unjust reality of the colonial state. It is interesting because it opens a series of interpretive possibilities, which show us the full extent to which administration strategies for rounding up inexpensive chibalo labor had to confront the varied responses of the people affected by the forced labor system. This is the case of the Muchopes studied by Jeanne Penvenne.³⁸ The example concerns a system of city cleaning and trash and waste removal from the city of Lourenço Marques, the capital of colonial Mozambique. This work was performed by chibalo workers, and although it was extremely insalubrious, the muchopes, according to Penvenne, “became the chibalo workers preferred for these jobs and, as a group, knew how to take advantage of the situation.”

According to Penvenne, after 1908, the work of cleaning the city was given to a Portuguese firm, which used chibalo to obtain its workers. Because it was important and delicate work, any problems subjected the firm to fines assessed by the city administration. The muchopes, even though they were chibalo workers, rapidly understood the problems they could cause their employers. Thus they were able to come up with a system that allowed them to quickly fulfill their assigned cleaning tasks, leaving them more time outside of work, which they could use to perform other jobs, which provided them with concrete benefits. Among these were gardening, artisanal work, and above all repairing goods that they found in the trash of better-stocked houses. Although these “extra” jobs were considered “illegal” under chibalo, they were frequently tolerated by the supervisors responsible for overseeing these workers, precisely because if they were punished, they could respond by not properly cleaning the city—which would have caused problems for their employers.

Another strategy, derived from the muchopes, was the creation of a system of tipping, which consisted in providing exceptional services for those residents who agreed to pay, and punishing those who did not cooperate, or otherwise insulted them, by knocking over their trash cans. This experience, among others collected by Penvenne, is important in demonstrating that although chibalo was an arbitrary and contingent system of compulsion, it was confronted not only by a narrative which highlighted its injustices, but also by a population that mobilized strategies which responded to its economic realities. The most common strategy was migration to the South African mines, precisely because this mobility responded to the economic rationality that I have sought to trace in these cases. Mobility therefore represented a dynamic process of adaptation and innovation among migrants who were faced with the encroachment of capitalism across southern Africa. Rather than representing an involuntary response to colonial rule, both the Portuguese and South African authorities quickly grew to understand that the mobility of migrant workers was the most versatile and most complex driver of capitalist production in southern Africa.³⁹

To that end, the Portuguese tried to use their strategic position to take advantage of this phenomenon, signing a series of well-known accords with the South African administration between 1896 and 1909, which regulated the system of labor migration between Mozambique and South Africa.⁴⁰ Crucially important within these agreements was a deferred compensation system, in which miners were not paid until they returned to Mozambique. Portuguese authorities generated significant tax revenue from this system; as William Norman has observed, it allowed the Portuguese government not only to collect taxes directly from miners’ wages, but also manipulate the exchange rate of miners’ wages to Portuguese benefit, and further ensure that miners spent the money from their wages within Portuguese territory. Beyond those more tangible benefits, the system of deferred compensation gave the administration a sense that it actually controlled the migrants nominally under their jurisdiction.⁴¹

But while the colonial administration garnered significant advantages from these laws and treaties, their rule remained narrow, limited, and largely dependent upon African subordinates. Without local “chiefs,” known in Mozambique as régulos, it never would have been possible to develop an efficient system of tax collection. Luís Covane, for example, has noted that:

Much of the money that migrant laborers earned was used to avoid being conscripted into forced labor by local authorities. Although Portuguese officials knew that migrant workers gave gifts to their chiefs to be exempted from chibalo and other types of undesirable labor, they were unable to actually control this phenomenon, dependent as they were upon the chiefs. This also explains why the Portuguese constantly said that Africans became arrogant with the money they earned in South Africa, because they used it to bribe chiefs and avoid being conscripted for agricultural labor.⁴²

Notwithstanding colonial mechanisms of control, the legal structures of colonial rule, and even the economic advantages of deferred tax payments, a series of extra-legal arrangements between migrant workers and chiefs allowed them to escape colonial obligations. Thus, while the accords between the colonial administration and the South African mining houses were formally defined in a way to benefit both sides, their impact within African society was quite different. Since colonial laws and obligations held no benefit for migrants’ day-to-day life, they reacted by adapting preexisting social and cultural practices (in this case, the practice of providing tribute to local chiefs) to meet the new aim of evading colonial impositions.

One can see similar dynamics in the phenomenon of “clandestine migration” to South Africa. “Clandestine emigration” was a legal offense, invented by colonial legislation, which made it illegal for migrants to leave Mozambique without official authorization. The category can only make sense when viewed through the excessively narrow lens of colonial legislation; in reality, there was no way for the Portuguese to control the mobility of workers, nor were there any penalties they could apply to South African employers who hired “clandestine” workers.⁴³ From the point of view of migrants, meanwhile, the conceptual division between “legal” and “illegal” migration was not based upon the prescriptions of colonial law or intergovernmental accords; it was instead based upon the much more concrete necessity of earning a living. Migrants who went abroad without official authorization were aware that they were committing an “illegal” act, but for the majority of migrants, these laws were economically illegible and socially illegitimate.⁴⁴

Colonial authorities defined migration as “clandestine” because they imagined that it was being undertaken as flight from taxes and forced labor, thus making such “clandestine” migration an act of resistance. But in reality, migrants made their decisions in a much wider social universe than the one imagined by colonial authorities. Of course, the fear of chibalo had an effect on migratory processes. Nonetheless, migrants also made their calculations based on a different type of rationality, one grounded in the culture of migration. The act of migrating also implied the search for social status, not just flight from the colonial state. Physical mobility was intrinsically linked with social mobility; migrants were not only seeking to escape the colonial state, but also to better fulfill their social obligations. Fleeing from chibalo was, in that sense, a pragmatic motivation, rather than an outright rebellion facing the colonial state. Indeed, it would be hard to justify going to the mines if the sole objective was to flee from chibalo, because the mines were in some respects worse than chibalo, at least as regards their working conditions; the daily tasks in the mines were extremely difficult, demanding tremendous physical exertion, and unsafe conditions brought with them the constant danger of serious injury or death.

For that reason, the choice between working for the South Africans or working for the Portuguese was not a choice that reflected some intrinsic difference between one form of work and the other. Instead, it reflected the choices that migrants made, choices that reflected their understanding of the different modes of production that operated in Mozambique and South Africa. These choices reflected the fact that migrants were confronted with two competing possibilities, two different modes of production. Migrants were able to exploit their ability as migrants to move between these two different possibilities in ways that best met their specific needs. The coexistence of these two very different modes of labor made mobility such an important tool—not only as an economic strategy, but also as a social strategy. Migrants needed to monetarize their labor in order to navigate the world of colonial rule; mobility was an essential asset that allowed them to optimize their labor and effectively respond to the senseless and unjust demands made by colonial officials. But migrants were also operating within their own social and ideological realm, one which guaranteed them access to higher wages as needed, and thus allowed migrants to formulate the most appropriate response to the difficulties of their situation.