Socio-economic indicators in and about African countries, whether produced by state apparatuses, international organizations, or academics, regularly come under attack.Footnote 1 This special issue aims to take a step back from these debates, and instead to encourage reflection on the specific characteristics of data produced on the African continent through historical and sociological analysis.
From the very beginning of this project, we were guided by the intuition that evaluating the impact of colonization on local forms of numerical knowledge was essential for analyzing these statistics.Footnote 2 The fact that data and categories produced by colonial administrations were designed primarily to dominate territories and populations had to be taken into consideration as well as the extent to which decolonization impacted statistical production.
By bringing together case studies on the contemporary history of counting work and workers, this special issue attempts to open a field of research at the intersection between African labor history, colonial history, and the social sciences of quantification. We begin this introduction with a critical review of the state of the art in these fields, pointing out the epistemological obstacles that have hindered a consideration of African labor history through the prism of quantification. We then turn to what this special issue contributes. The articles gathered here show the diversity in the uses of numbers and the motivations for producing them, and the power asymmetries between bureaucracies and the populations they count. First, counts of workers were used to build an information capacity of colonial and independent states, which at the same time obfuscated labor exploitation. Second, the articles in this special issue highlight the distinction between quantification efforts that construct the labor force for the needs of the economy, and those that contribute to the implementation of labor rights and social protection. Third, figures on work and workers were produced, used, and adapted by a variety of actors, who thereby imprinted their visions and conflicts on the colonial and postcolonial worlds of work. We conclude the introduction with a plea for future work to engage more deeply and critically with African agency in the production, use, and repurposing of labor statistics.
Statistics and labor in Africa: the blind spots in historical accounts
The statistical observation of labor in Africa falls within a double blind spot in the historical literature. While the historical sociology of quantification has only recently begun to examine colonial statistics, the renewal of African labor history opens new opportunities for reflecting on the historical shifts in the categorization and quantification of work and workers.
Colonial and postcolonial statistics are just beginning to attract the attention of historical sociology of quantification.Footnote 3 In the 1980s and 1990s, scholars documenting the role of statistics in the construction of nation-states largely ignored the fact that imperialism was constitutive of this process.Footnote 4 Narratives that were developed for Western states focused on the metropolitan territory, especially in the nineteenth century, while also identifying three earlier episodes of interest in the production of knowledge: (1) the geographical description of states in the sixteenth and seventeenth centuries, (2) the measurement of the power of the states from the mid-seventeenth century onward, and (3) the mathematization of data interpretation in the late eighteenth century.Footnote 5 The lack of interest among historians of statistics in its colonial dimension is surprising,Footnote 6 especially when one considers how much this scholarship has been influenced by Foucauldian analyses of statistics as an instrument of population control by states.Footnote 7 Colonial power seems to offer an ideal field in which to study “how statistics facilitated the fabrication of society as an ‘object of rule.’”Footnote 8 Yet studies on quantification, in both its cultural and epistemological dimensions, often overlook or elide colonial legacies.Footnote 9 Above all, the question of how colonialism shaped the production of figures on populations and territories outside Europe remains largely unanswered.
Of course there are a few rare studies that escape this tendency toward Eurocentrism, but they tend to focus exclusively on statistics and colonial rule in India and other regions under British colonial domination.Footnote 10 Moreover, they study population data (e.g., censuses and registers), and in some cases economic statistics, while leaving other kinds of statistics unexamined.Footnote 11 A major contribution of this body of work, however, has been to show the connection between practices of data collection and symbols of authority in the colonial context. Arjun Appadurai, for instance, argues that “the role of numbers in complex information-gathering apparatuses such as the colonial one in India had two sides”: the “justificatory” and the “disciplinary.”Footnote 12 And Umamaheswaran Kalpagam states that statistics “became the most important language in the narrative legitimation of modernity, that is, for telling stories about progress, of accumulation of wealth, control of nature, the well-being of humanity, and equally to counter those stories as well.”Footnote 13 Throughout this body of research, however, work and workers appear only in the background, if they appear at all.Footnote 14
Historians have also recently started to examine the statistical institutions of independent African states and their enduring colonial legacies.Footnote 15 In particular, researchers have explored how development discourse since the 1950s has been underpinned by the glorification of science and technology, including statistics and economic indicators, as purported instruments of modernization.Footnote 16 While developmentalism became the dominant framework and raison d’être for newly independent states and international organizations alike, statistics, as a tool for measuring so-called progress, rendered the lived realities of ordinary people and class differences invisible behind abstract macroeconomic metrics.Footnote 17
In its attempt to take a critical look at the quantification of work in Africa, this special issue benefits from the renewal of African labor history. After World War II, intellectuals, especially on the African continent, began to challenge the narrative of colonial sciences,Footnote 18 which for decades had depicted African labor as essential to the “civilization” and “development” of colonial territories.Footnote 19 Another line of research, adopting a comparative and global perspective, has analyzed the organization and forms of forced labor under colonial rule.Footnote 20 It has contributed to a reconsideration of the concept of wage labor, which was promoted in an ethnocentric way by international organizations such as the International Labour Organization as a hallmark of modernity. This body of work has also made clear that free and unfree labor were two sides of the same coin.Footnote 21 In the wake of Frederick Cooper’s research on labor in African societies,Footnote 22 others have also explored labor migration,Footnote 23 wage work and white-collar workers,Footnote 24 and women’s work.Footnote 25 Much remains to be done, especially as regards agricultural, craft, and domestic and reproductive labor, the last in particular being too often elided in archival sources and statistics.Footnote 26 Nevertheless, recent syntheses have identified dynamics that are common to different contexts, including the destabilization of rural societies through land dispossession, migration to urban centers, and the persistence of strong ties to the countryside.Footnote 27 Since the 1970s, many African economies, both north and south of the Sahara, have been profoundly disrupted by liberalization policies. The drastic downsizing of public sector employment and the proliferation of unregulated small-scale activities drew attention to the informal sector. This concept in itself reveals the difficulty that Western scholars have in adequately grasping non-Western forms of work and livelihood.Footnote 28
Counting work and workers in Africa: toward a new research agenda
The motive for this special issue came from a research project on colonial statistics in Africa carried out by an interdisciplinary team led by Béatrice Touchelay.Footnote 29 Addressing the tendency in the scholarship to focus on the British Empire, Touchelay’s project analyzes for the first time the French colonial empire in Africa through the lens of various kinds of statistical production, including demographic, economic, social, labor, and financial statistics.Footnote 30 Among the first findings of the project are that “quantifying in colonial societies means not only conceiving, enumerating, classifying and counting, but also standardizing, shaping, and coercing in order to dominate,” and that populations “effectively resisted these attempts at standardization.”Footnote 31 This special issue arises from this project, and interrogates more specifically the practices of counting, categorizing, and observing work and workers in various colonial and post-independence contexts.
Drawing on sources such as occupational censuses and quantitative observations conducted by labor inspectors, civil servants, and colonial intermediaries in the field, this special issue considers the wide range of counting practices that were employed and examines the shifting representations of work and workers in the moyenne durée. How did political and organizational contexts, which were marked by limited resources and staff dedicated to statistics, influence the production of numbers? How did these figures contribute to the racist, repressive or managerial, government of populations? Which aspects of labor were measured under colonialism, and which were ignored?
This special issue is grounded in the idea that knowledge production is always situated and localized. It thus contributes to highlighting the diversity of configurations in which labor was observed and counted. Although it does not cover all regions or all colonial empires, it does sketch out comparisons across the African continent, focusing especially on French colonies (Côte d’Ivoire, see Cissé’s contribution, and Algeria, see Joyeux’s and Mollard’s), but also on independent Egypt under British influence (see Labib’s contribution) and Portuguese Guinea (see Cerdeira’s contribution). Finally, Demeule’s article addresses tensions between Belgian and British authorities in the framework of international projects in sub-Saharan Africa as well as in independent Nigeria.
In the same way that we reject any reduction of Africa to a mere “pictogram,”Footnote 32 the articles avoid portraying African workers as a monolithic group. In North Africa, the use of national categories (e.g., “Algerian” or “Egyptian”) by the actors themselves was common before independence. In sub-Saharan Africa, self-identification categories like Soninke, Wolof, Mossi, Malinke, or Bambara coexisted with colonial categorizations based, for instance, on religion or region. Our challenge was therefore to navigate emic categories, locally grounded, while also critically engaging with the labels used in colonial statistics, especially the label “native.”
The focus on the mid-twentieth century (1920s–1960s) in this special issue was not specified by the call for papers, but it may reveal wider patterns in African labor statistics. Whereas attempts to gather vital statistics or to survey land came at an early stage in colonization projects,Footnote 33 numbers on work and workers, as well as economic expertise, developed mostly after World War I.Footnote 34 The articles in this special issue thereby allow us to better understand the period of late colonialism and especially the role of quantification in the organization of forced labor, at a time when forced labor was already facing international condemnation (see Cissé and Cerdeira in this issue). Some contributions take a close look at decolonization as a moment in the emergence of measurement standards (see Demeule and Labib), or as the starting point in the development of national statistical systems (see Mollard and Labib).
Information capacity and concealment in colonial and postcolonial contexts
The cases studied here demonstrate that data production was more a tool of communication to assert control, primarily between levels of the colonial administration, rather than a means of gathering knowledge. Numbers are thus analyzed both as a performance of colonial rule and modern state capacity and as a magnifying mirror of their shortcomings. Indeed, quantification allows governments to give governed societies and economies an appearance of transparency and legibility. Counting standardizes heterogeneous social worlds, and “enables us to make disparate things hold together.”Footnote 35 Labor statistics disseminate abstract models to portray reality and act on it, while at the same time reducing uncertainty and facilitating decision-making by authorities.
But while the capacity to inform and make populations legible is central to state authority,Footnote 36 colonial administrations often refrained from developing and funding counting infrastructures. Moreover, in certain circumstances colonial officials resisted measuring. Statistical tables, used by central administrations as tools of bureaucratic reporting rather than as instruments for acquiring knowledge about populations and work, were often sent back empty by local officials.Footnote 37 Therefore, the articles gathered here also discuss the role of ignorance and the precariousness of bureaucratic apparatuses (see especially Joyeux and Cerdeira). At the same time, the performance of modernity intended by counting was equally significant for (newly) independent states and their strongly interventionist agendas, as illustrated by the articles in this issue on Algeria and Egypt.
Counting to mobilize a labor force and counting to regulate labor relations
The second result of this special issue is to reveal the tension in the colonial and postcolonial contexts between figures produced to meet the occasional needs of the administration or private enterprises, and surveys or aggregate data on working conditions (unemployment, standard of living etc.). The main motivations for keeping quantitative accounts were to mobilize the labor force and to control migration to cities, neighboring colonies, or Europe (see Mollard and Cissé).Footnote 38 The periodic counting of workers contributed to the objectification of the colonized population, which was viewed as an economic resource to be allocated. But during the interwar period, labor in colonial territories became increasingly regulated at the international level,Footnote 39 and colonial administrations had to adapt to this new framework. One strategy was to simply ignore it; in other cases, the central authority did mandate the collection of information on workers but the numbers produced by local officials were so vague that they concealed the actual recruitment procedures, as discussed by Cerdeira on late Portuguese Guinea.
After World War II, development programs allowed colonial authorities (and later, independent states) to continue exploiting the workforce at a time when forced labor was increasingly delegitimized at the international level.Footnote 40 The “statistical argument”Footnote 41 was a particularly helpful “companion”Footnote 42 to promote developmentalist policies and social protection systems. From the 1940s onward, the quantified assessment of newly adopted social laws (regulations on workplace accidents, working hours, minimum wages, etc.), in French colonies for example, was indeed intended to relegitimate colonial rule as a response to its growing contestation. African workers began to be included in this regulatory framework, as if to more firmly bind them through a—still incomplete—social citizenship.Footnote 43 In the second half of the twentieth century, the increasing regulation of employment relations and social insurance by international organizations and the newly independent states went hand in hand with new requirements and opportunities for quantification.Footnote 44
Actors and dynamics of knowledge making
Lastly, another outcome of this special issue is to highlight the many actors involved in the process of compiling figures and the varying materiality of surveys, ranging from written reports to more standardized forms. By analyzing data production as a dynamic and multiscalar process involving individuals, local political structures, colonial bureaucracies, and international organizations, this special issue contributes to the social history of science and knowledge.Footnote 45 Recent developments in the fields of both the history of quantification and colonial history emphasize the role of intermediaries and non-state actors in the production of knowledge, including labor unions and workers themselves.Footnote 46
The various contributions show that statistics on work and workers emerged out of different and sometimes contradictory needs and objectives. Labib, for example, demonstrates how newly designed statistical categories for measuring employment and unemployment were closely tied to debates in 1940s and 1950s Egypt between competing visions of economic development and social modernization. The contributions reveal that figures were also produced and compiled by non-state actors, be they missionaries in Algeria (Joyeux), unions of colonial and African planters (Cissé), or economists and international experts (Labib, Demeule). For example, Cissé shows in his article that quantification was not the exclusive prerogative of colonizers: in December 1944, Félix Houphouët-Boigny and African planters used their own figures to claim rights and pressure the colonial authorities.
By shedding light on the relationship between knowledge making, political rule, and the control of labor, this special issue thus contributes to a deeper understanding of colonial and postcolonial domination. The study of figures and categories reveals how states envisioned the workers and societies they governed. It also provides information about the broad range of instruments available to political authorities. The articles highlight the dynamics of co-construction, as well as the interactions and conflicts between state and non-state actors in the production of knowledge intended to organize, dominate, or reform societyFootnote 47—showing that statistics are more than just techniques of state power.
These findings show that future research on statistics in Africa should certainly resituate the colonial moment in a longer historical perspective in relation to preexisting counting practices and their historical transmission. This includes the long-term responses and forms of resistance that African societies developed to the forms of accounting and bookkeeping that accompanied raids for the transatlantic slave trade, about which there is still little known. Future studies could also further explore agency in its multiple meanings. Particular attention should be paid to the bottom-up dynamics of negotiation and intermediation inherent in the construction of this data, as well as the strategic uses and alternative forms of statistical production used to claim rights and challenge power.Footnote 48 As scholarship in subaltern studies has shown, researchers must take seriously the possibility of workers and peasants navigating within an “autonomous political space” distinct from that of colonial or national elites.Footnote 49 Like other tools of imperialism, there is no doubt that statistics were also domesticated or, as Partha Chatterjee puts it, “selective[ly] appropriat[ed].” Footnote 50
Acknowledgements
We would like to thank the ILWCH editors and team for their support. We would also like to thank Ian Drummond for his very accurate and valuable editing and proofreading work. We are deeply indebted to Roser Cussó, Akosua Darkwah, Alexander Keese, and Béatrice Touchelay for their comments on earlier versions of the articles during a writing workshop at Heidelberg University. This workshop received funding from ANR, GIS-Gestes, Université Paris Nanterre, and Heidelberg University (Expanding Internationality Program/Exzellenzuniversität). AI (ChatGPT and DeepL) was carefully used to improve style and expressions in English.