As Jean-Paul Sanderson rightly notes in the introduction to Demographie Coloniale Congolaise: Entre Speculation, Idéologie et Reconstruction Historique, the question of how the population of Congo was affected demographically by European colonization is one that is both politically charged and poorly understood. Demographic statistics have been circulated (and recirculated) to support historical interpretations of varied ideological orientations. On the one hand, there are those, like Adam Hochschild, who have used shocking statistics — estimating that Congo's population decreased from twenty million to ten million under Leopold II's rule — to call out the violent regime of the Congo Free State. On the other hand, there are colonial apologists who have used demographic statistics that indicated significant population gains in post-World War II era of Belgian colonial rule to illustrate the supposed benefits of colonization. The problem with all of them, Sanderson points out, is that they uncritically reproduce dubious numbers, the origins of which remain largely uninterrogated, which has resulted in a poor understanding of demographic change in Congolese history. In this study, Sanderson sets out to remedy this problem by taking a deep dive into the colonial demographic archives, there surveying and systematically analyzing a variety of colonial sources for Congolese population statistics. The result is a meticulously researched and scientifically grounded study that contains rich data and thoughtful analysis, but is, in its scope and style, likely to appeal only to highly specialized audiences.

The book is organized into three sections, each of which addresses a different research question. In Section One, Sanderson seeks to answer the question, ‘What do we know or did we know about the Congolese population’? This section surveys both widely held ideas about Congolese demography and the sources from which those ideas have emerged. Most existing narratives of Congolese demographic history, he explains, identify three major eras of demographic change. The first is a period of exceedingly high mortality (and low birth rates), which lasted from about 1875–1925. It was set into motion by the violence of coastal Swahili-Arab slave traders and then further exacerbated by the arrival of Europeans, who ushered in the infamously violent Leopoldian regime, the disruptions of which were the catalyst for multiple disease epidemics that ensued (sleeping sickness, syphilis, Spanish Flu, etc.). The second period, which lasted from about 1925–47, witnessed a kind of demographic plateau, with neither a marked increase or decrease in population, as some of the most egregious forms of state violence ceased, but wartime disruptions curbed any demographic growth. In the third period, which lasted until independence in 1960, both the colonial state and private companies increasingly invested in developing medical infrastructure in the colony, for the self-serving purposes of maintaining a healthy and profitable labor force, at the same time that larger global developments in medical science (antibiotics, for example) offered improved treatments.

With some variation in emphasis and explanation, this demographic timeline appears to have been widely accepted in the literature on Congolese demographics. Sanderson argues that even if this timeline makes sense logically, given the known social and political history of Congo in these eras, few scholars have actually sought statistical demographic evidence to back it up. Indeed, many have uncritically cited the numbers produced in highly problematic sources, such as Henry Morton Stanley's field notes (the original source of the widely circulated, and oft-repeated, estimate that there were twenty million people in what became the Congo in the late nineteenth century). It is these oversights that Sanderson sets out to correct in the second and third parts of the book.

In Part Two, Sanderson asks, ‘What are the documents that allow us to study Congolese demography and what are their limits’? To answer this question, he offers a detailed survey of the kinds of sources from which colonial population data might be mined, assessing each for their merits and potential biases. Two kinds of sources emerge as the most important: census and survey data collected by state agents for taxation and labor recruitment purposes and, after 1930, medical data collected by the Fonds Reine Elisabeth pour l'Assistance Medicale aux Indigenes (FORAMI). In what is perhaps the most original part of this section, Sanderson interviews a number of Belgian former colonial agents, who worked in Congo between 1947–59, in order to get a sense of what the census gathering process looked like on the ground and what possible biases and inaccuracies the data might have. These interviews offer important insights into how colonial state power functioned (and didn't function) on the ground. We learn, for example, that it was not uncommon for agents to ‘extrapolate’ data when they could not actually get an accurate population count (which was often). More importantly, we learn that the process was highly unstandardized, the agents were poorly trained, the population resented and often resisted being counted, and the resulting data had a high (the former agents themselves estimate 10 per cent) margin of error.

In Part Three — framed around ‘What understanding of Congolese demography can be constructed from these documents’? — Sanderson applies statistical demographic analysis to the data discussed in Part Two. He begins with a case study of the region of Kwango, applying multiple methods of data analysis in order identify the most plausible estimates of population change in the region over time. He then uses those findings to test several ‘rear-projection’ methods to come to his best estimate for Congo's population in the late nineteenth century. Ultimately, the two most convincing methods produce a hypothetical Congolese population figure in 1895 of somewhere between 10.5 million and 15 million (185–7). The latter number is five million less than the twenty million figure that has traditionally been accepted, but it would still result net loss of five million people between 1895 and 1930 — a very high, but not entirely implausible, number. It is, in any case, the upper limit of what is possible, he argues. The former number (10.5 million) — which signifies a net demographic loss of half a million people between 1895 and 1930 — is his most conservative estimate.

In the end, this study offers a thorough analysis of the demographic sources, but it does not greatly revolutionize our understanding of demographic change in Congo during the colonial era. Instead, it largely affirms the hypotheses with which historians of the region were already working. To be sure, the difference between a net loss of ten million people (the previously circulated number, which Sanderson convincingly dubs demographically ‘impossible’) and a net loss of half a million people (Sanderson's most conservative estimate) is quite substantial. But Sanderson's estimate, it must be emphasized, remains highly theoretical, even if scientifically grounded. He readily admits that there are no reliable sources from that era that can confirm his math. And, importantly, both numbers connote a catastrophic loss of human life and potential. The question becomes then, is the horror of the Congo Free State best understood through the lens of statistics? The numbers do not really change what we know already about how the Congo Free State functioned through violence. Similarly, the marked improvements in mortality and birth rates in the late Belgian colonial period do not change the fact the colonial violence continued to permeate the lives of Congolese people during that era in ways that are not necessarily signified in birth and death rates. Demographic studies, while potentially powerful, will always be limited in their scope unless tied in a meaningful way to the lived experiences of the people they represent. This book, while well-researched and important, suffers from that narrowed scope. That reality, combined with the highly technical and at times repetitive nature of the prose, suggest that it will likely find limited readership outside of the field of demography.