Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, AFRICAN HISTORY (africanhistory.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 15 October 2018

Numerical Data and Statistical Sources

Summary and Keywords

The use of numerical data and statistical sources in African history has expanded in recent decades, facilitated by technological advances and the digitization of primary sources. This expansion has included new analysis of traditional measures (population, government, and trade) as well as new sources of individual-level data such as census returns, marriage registers, and military and police records. Overall, this work has allowed for a more comprehensive quantitative picture of Africa’s history, and in particular facilitated comparisons within Africa and between African countries and other parts of the world. However, there remain misunderstandings about the collection, use, and interpretation of these data. Increasingly sophisticated methods of quantitative analysis can alienate scholars who have an intimate knowledge of the data and how they are produced, but lack specialist methodological training. At the same time, limited understanding of the origins and reliability of quantitative data can lead to misinterpretation.

Keywords: quantification, statistics, data, population, GDP, econometrics

Reconstructing Quantitative History

In 1982, Jan Vansina published an article lamenting the neglect of what he described as “lost corners of the world” in economic and social history and stressed the need for methodological innovation to fill the gap left by conventional written sources.1 While his focus was on the use of comparative ethnography, he might have made the same argument for the use of quantitative evidence in African history.

For many years, African history was notable for the limited coverage of available statistical sources. Even today, some observers describe what they call a “statistical tragedy” in Africa.2 In European history, for example, Gross Domestic Product (GDP) per capita can be calculated back to the Middle Ages.3 For most African countries, official estimates exist only from around 1950. Reliable population estimates based on direct data are only available from around the same time for most parts of the continent. Efforts to look further back have relied on proxies, conjectures, and backwards projections.

As Vansina was writing, African economic and social history was in a period of relative decline, which further reduced the demand for quantitative research.4 However, in recent decades, the field has undergone what one recent paper has described as a “data revolution.”5 While this is particularly true of economic history, it is also relevant to other subfields of African history, including social, medical, and political history. In this work, scholars have used old sources in new and more inventive ways, while at the same time locating and digitizing previously neglected ones. One particular feature has been the use of individual-level data, or “microdata,” which has allowed for the recovery of African people otherwise lost to conventional archival records.6 Another has been the use of spatial analysis, drawing attention particularly to the considerable geographical variation and subnational inequalities in economic and political structures across the continent. This work has revealed a much more comprehensive quantitative picture of Africa’s history.

Some lost corners still remain, and coverage has been uneven. The 19th and 20th centuries receive the most detailed analysis owing to the availability of data.7 Some efforts to extrapolate backwards remain controversial. Quantitative research relies heavily on records collected by colonial institutions and has been dominated by the study of British and French records. Other parts of Africa, including those countries colonized by Germany, Belgium, Portugal, and Spain, as well as Liberia and Ethiopia, which remained independent through most of the colonial period, have received much less attention.8 Further, the quality of data sources is variable, and debates continue about the appropriate interpretation of statistical sources.

This article reviews some of the key sources and methods used in quantitative research on African history. It begins by considering the ways in which the use of quantitative data has changed over time (see “Quantitative Methods in African History.”) Subsequent sections are divided according to types of data and includes sections on “Population and Demography,” “Economic Performance,” “Living Standards and Social Trends,” “Government Institutions,” and “Geography, Environment, and Space.” It concludes by addressing some of the wider debates on the opportunities and risks of quantitative research in African history (see “To Quantify or Not to Quantify.”)

Quantitative Methods in African History

Quantitative research is a broad church, and recent work in African history has encompassed methods ranging from simple descriptive statistics to more complex econometric analysis. They offer both opportunities and hazards to the historian using them. A general review of the methods used is provided, which will be elaborated on in discussions of specific data sources.

Descriptive statistics are nothing new in African history, and current work often builds on early datasets begun in the early days of African history in the 1920s.9 Research in this period was often data-rich, capturing available information on a variety of social and economic variables.10 Organizing these data into tables and graphs is the first step in understanding and analyzing them; beyond this, there are a number of simple statistical methods used to organize and present quantitative data.11

A further step is to use these data to build standardized indicators of, for example, economic or social outcomes. These include Gross Domestic Product (GDP) as a measure of economic output, measures of inequality such as the Gini coefficient, or composite measures of welfare such as the Human Development Index (HDI). Because these are calculated according to prescribed methods, they allow for comparisons of development over time. For most African countries, these calculations have been restricted to the postindependence period for which official estimates are available. However, there have been attempts to use historical data to look further back in time.12 Such calculations have the virtue of facilitating international comparisons. We can ask, for example, how Africa compared to Asia or Latin America through the 19th or 20th centuries or compare the patterns of growing and shrinking in African economies to those of preindustrial economies in other time periods.13

The use of econometric methods has potentially greater analytical power, particularly in identifying the relationships between different variables, but comes at the risk of further abstraction from the hard data. Econometric analysis has become increasingly important in African political and economic history in recent decades owing to the combination of advances in computing power as well as the greater availability of digitized data. In African history, it has particularly been used to examine the causal links between historical events and differences in recent development outcomes.14 This work is often neglected by historians who find the methods difficult to grasp while at the same time questioning many of the assumptions of formal modelling. Further, it has been criticized for “compressing” history, linking an event in the past to current outcomes without detailed study of the events that happened in between.15

Sources of Data

Population and Demography

Demographic data are key to most quantitative work, whether on economic, social, or political history. Population estimates provide the denominator in any comparisons of per capita income, trade, revenue, or scale of activity. Further, the theory that Africa’s low population density had important effects on market size and trade costs as well as political institutions has been influential in shaping the study of African economies over time.16

Measuring the population of African countries over the long run has been a persistent challenge. Little direct evidence has survived for the measurement of precolonial populations. An exception is Thornton’s study of the Kingdom of the Kongo from the 16th to the 18th centuries, which relies on baptismal registers kept by missionaries in the region.17 Similarly, Heywood and Thornton argue that the estimates of Hungarian traveller Lazlo Magyar made use of fiscal records of the kingdoms of central Angola.18

For later periods, colonial governments did make some effort to calculate the population of their territories. However, colonial censuses are widely considered to be unreliable. Limits in the financial resources and capacity of the colonial state meant colonial administrations had to make do with uncertain estimates by district officers for most regions. Further, resistance to colonial taxation and forced labor exactions made many Africans reluctant to be counted.19 Kuczynski’s 1949 survey of demographic statistics in the British Empire provides the following example for Kenya. Before 1914, the African population was accounted for using the following method, described by a District Commissioner: “In 1913–14 the only counting was of huts and was done by tribal retainers . . . Population figures were estimated by assuming an average of three persons per hut and entering a round figure which approximated to the results obtained.” In 1926, a proposal was made to extend the Kenyan census to include the African population, as opposed to only the expatriate communities. The Acting Chief Native Commissioner responded by noting that “there are certain areas in the Colony—the Northern Frontier District, for instance—in which it would probably not be possible to carry it out without armed intervention.” The proposal was not taken forward.20 Instead, it was not until 1948 that the first complete census of Kenya was taken, and even at that there were some areas of incomplete enumeration.21 Similar stories can be told for other countries. There were exceptions to this; for example, Doyle has used a series of more reliable colonial censuses for Uganda, supplemented by other colonial records as well as oral histories and qualitative accounts, to investigate the interaction of demographic and economic change in Bunyoro.22

In the absence of good direct evidence before World War II, most historical estimates of the population of precolonial Africa are made using backwards projection by assuming a reasonable growth rate and subtracting from the first reliable count, making adjustments for historical events that may have affected population numbers or fertility. Identifying the appropriate growth rate has been the subject of some controversy. For one set of estimates, Patrick Manning used the growth rate of better-documented India to develop new estimates for African countries.23 A critique of this method argued that India was far more densely populated than much of sub-Saharan Africa and therefore adopting growth rates of land-abundant countries in southeast Asia would be more accurate.24Figure 1 compares estimates from these two works for the population of West Africa from 1850 to 1950. Over such a long period, the differences generated by assumptions about growth rates become apparent. In the Manning estimates, the early population was higher but growth less rapid. In contrast, the higher growth rate assumed by Frankema and Jerven results in a lower estimate in 1850.

Numerical Data and Statistical SourcesClick to view larger

Figure 1. African population by two methods (millions).

Source: Manning (2010); Frankema and Jerven (2014).

Such methods are inexact and may have considerable errors, particularly for earlier periods. Other scholars have used a combination of qualitative and quantitative records to reconstruct data on indigenous populations from earlier periods (e.g., the Khoesan of South Africa).25

One area of African demographic history which has been the subject of considerable work is regarding the impact of the Atlantic slave trade on the African population. Many early estimates of the number of slaves taken from the region from the beginning of the trade in the 16th century through the 19th were based on propaganda produced during debates on abolition and were often wildly inaccurate. From the 1960s, a group of historians have compiled shipping records of slave voyages, cataloguing close to 35,000 voyages from 1514 to 1866. From these data, they estimated that around 12.5 million slaves departed Africa across this period (see The Trans-Atlantic Slave Trade Database).26 The database also includes information on a variety of other variables, including the place of departure, the duration of the voyages, and mortality rates during the voyage. When placed in the context of advances in estimating the population of Africa, these data allow historians to better understand the human cost of the trade and its impact on demographic patterns and the labor force.

A second area of demographic history which has been the subject of considerable literature is the study of mortality, both African and European. A series of works by Philip Curtin attempted to assess the extent of and reasons for high levels of European mortality on the West African coast, and how these changed over time.27 Unique sources from Liberia have also allowed for studies of the mortality of African American migrants.28

Economic Performance

Measures of economic performance are another area of quantitative research. Given the relatively limited data available, these are often based on proxies which need to be interpreted with care. However, such data allow for calculation of the economic impact of historical events like the slave trade and for African economies to be assessed in comparison to one another and to those of other regions.

Trade data are one of the key indicators historians use to understand economic patterns of the past. In the absence of other measures, the level of exports has been interpreted by economic historians as a broad indicator of economic expansion, one which is often better documented than agricultural production for domestic consumption.

Data on imports and exports survive, if not always comprehensively, back to the early modern period. At best, they can offer insights into both the quantity and variety of exports as well as their value. Figure 2 shows the quantity and value of cocoa exports from colonial Ghana as an example. Trade data can also give a sense of the degree of diversity in African economies. For most African countries, exports are dominated by a relatively small number of primary goods, including agricultural produce or minerals. As shown in the cocoa statistics, the values of these exports often tend to fluctuate, making producers and governments vulnerable to the volatility of their prices. Care must be taken to understand how trade statistics were constructed in order to understand how to interpret them. For example, during the 18th century, British trade statistics used official rather than market prices to assess the value of imports. Where these two prices diverge, the values reported in customs records can be misleading.29

Numerical Data and Statistical SourcesClick to view larger

Figure 2. Cocoa exports from Ghana, 1890–1938. Tons on left axis, values on right axis.

Source: Gold Coast, Blue Book (Accra: Government Printer).

While trade data have the advantage of being readily available, export production is just one sector of the economy, and understanding how changes in one sector relate to changes in the other sectors is difficult. This situation is particularly true for what is often referred to as the subsistence sector, or domestic agricultural production.30 In the past, it was common to assume that subsistence production increased with population; in other words, people produced just enough to feed additional people. However, this assumption has been challenged in the history of other regions, where research has shown that expansions in export production can change patterns of food consumption, with people consuming more calories as they become more affluent.31

Exports also do not capture the whole of the production of the manufacturing sector, as much of this is likely to be consumed domestically. Only a small number of countries, most notably South Africa, began to export manufactured goods in the second half of the 20th century. Data on the production of manufactures or the shares of the labor force employed in different sectors provide better measures of structural change. These data show two phases of structural change in the postindependence era. The first, during the “golden age” of growth from the end of World War II until the early 1970s, saw many Africans moving out of agriculture and into manufacturing. The higher productivity of these occupations contributed to the increase in per capita Gross Domestic Product (GDP) experienced by most countries across these years. From the 1990s, structural change resumed following the difficult years of the 1980s, except that workers leaving agriculture moved mainly into the service sector, into occupations with lower productivity growth.32

For much of the postindependence period, African economic performance has been measured primarily using measures of GDP. Methods of measuring the national economy date back at least to the interwar period when they were associated with the work of Simon Kuznets in particular. These were expanded and systematized during and after World War II when the System of National Accounts (SNA) was introduced to facilitate the measurement of growth and international comparisons.33 Today, GDP is calculated by both African governments as well as international agencies using official statistics on production and consumption.

Africa in particular has been the subject of some controversy. Critics have argued that the poor quality of underlying data can limit the usefulness of such measures, as they may distort pictures of African economic performance.34 However, assessments using alternative measures have shown no consistent bias, particularly if they are used over periods longer than a year or two.35 Further, it remains the easiest mechanism for comparing African economies to each other and to those of countries in other regions.

National averages mask sometimes considerable local inequalities in economic performance. Rural and urban divides play an important role in shaping both economic and political dynamics in many African countries. To understand these phenomena, it is therefore important to be able to measure economic performance at local levels across different periods. Unfortunately, such data are more difficult to find than national aggregates. In 2014, the African Economic Outlook report by the Organisation for Economic Co-operation and Development (OECD) referred to the lack of subnational data as an important obstacle in regional economic policy. The same holds true for historical research on changing regional dynamics over time. One source of data currently used as a proxy for subnational development is a measure of night light density from satellite photographs, which has been shown to correlate well with income levels.36 Though only available for very recent periods, it can be used to document change over time since 1992.37 It is also often used as an outcome measure in studies hoping to link historical patterns of, for example, political organization or economic production to current levels of subnational inequality.

Living Standards and Social Trends

To understand the implications of economic change, it is necessary to consider equity and poverty as well as growth.38 Several methods used to address this problem are reviewed. Understanding how the economic and political changes of the 19th and 20th centuries affected the way Africans lived, worked, married, and interacted with each other has been a more difficult frontier in terms of quantitative data, but there has been progress in recent years using new methods and newly digitized microdata from a variety of sources.

One way historians have approached this area is by using wages and prices. Methods developed in the study of Europe and Asia combine nominal wages with the prices for a basket of goods that might be consumed by a hypothetical worker and his family to calculate a measure called “welfare ratios,” a form of real wages. A welfare ratio of one, meaning that the wage of the breadwinner is sufficient to purchase one “basket” of goods, implies a subsistence income. A welfare ratio of more than one suggests an income above subsistence.39 Welfare ratios can also be used to examine the gap between different types of wage earners. This can include using wages for skilled as well as unskilled workers and changing the composition of the basket of goods to include more or higher-quality food (a “respectability” basket). The first paper to do this for African countries calculated welfare ratios for South African workers of different occupations and races.40 Subsequent work has compared African countries in the colonial period or used company records to reconstruct welfare ratios in the 18th century.41

Where more comprehensive measures of income for different groups in society exist, composite measures of inequality are able to be calculated. The most popular is the Gini coefficient, which indicates the distribution of income. Widely used for contemporary Africa, it has also been calculated for earlier periods in order to understand the distributional consequences of, for example, economic change in the colonial period. Calculations of inequality in Botswana show that it was already accelerating before the discovery of diamonds because of the structure of the cattle industry.42 Bigsten also found that inequality increased in Kenya during parts of the colonial and postindependence periods, though there were also times in which it fell. Nor was rising inequality restricted to the 20th century. The earliest calculations of inequality for sub-Saharan Africa come from the Cape Colony, which use production data to calculate incomes and show rising inequality in the 17th and 18th centuries.43 There are challenges in using such data with the surviving data for much of Africa, particularly in understanding inequality among farmers for whom we often have little data on levels of production or incomes even in 2018.

Microdata and African History

Broad distributional measures such as the ones discussed in “Living Standards and Social Trends” often miss the stories of individual lives. A relatively recent frontier in the study of African history attempts to combine quantitative approaches with a focus on individuals through the use of microdata. Perhaps the oldest of such efforts is the study of anthropometrics, which exploits a relationship between average heights and the living standards of children. Where children are undernourished or suffering frequently from illnesses, average height is generally lower. Economic development eventually leads to an increase in average height, although in the early stages of growth there may be a slight decrease, known as the “antebellum puzzle” in the United States or the “early industrial growth puzzle” in Europe.44 Height data on Africans appear in the Demographic and Health Surveys (DHS) data from the 1950s onward as well as in various sources from the 19th and 20th centuries, particularly military or police records. These data have been used either to understand the impact of economic change, measuring the average heights of the same groups over time, or to compare different populations.45

Another source of microdata to be digitized more recently are missionary records, particularly Anglican marriage registers. Each register contains data on the names of both spouses, their age at marriage, their occupation and that of their fathers, and their place of residence. Literacy can be inferred through the presence of a signature (or lack thereof). These have now been used to investigate a number of questions in Africa’s economic and social history, including changing gender relations and social mobility. They provide a rare quantitative window into areas of African life not often captured at an individual level but are subject to the criticism that they only capture a subset of the African population, namely, those married in the Anglican church.46

Probate records from South Africa are another source which has provided a window into the lives and households of individuals. Fourie uses these records to measure household assets at death and argues, based on this evidence, that living standards were better than has been sometimes assumed.47 These sources provide valuable evidence for a previously neglected period in the economic history of southern Africa, but unfortunately are limited to the settler population. They also provide some information on slaves or other laborers working on settler farms.

A final source of microdata is postindependence censuses and surveys, the data for which have increasingly been released online. While these are often from more recent eras, they are able to illustrate the experiences of different eras through the comparison of people born in different decades (often referred to as “birth cohorts”). Through such methods, it is possible to examine how educational opportunities have changed across generations, for example, by comparing education levels or occupations of one cohort with another.48 These data are also used as an outcome variable in work linking historical events to current development outcomes.

Government Institutions

The development of the African state through the precolonial, colonial, and postindependence regions is a topic of wide interest in a number of subfields. How do African states differ from each other or from governments in other regions? How has the environment, historical change, or the structure of the economy influenced state building?

One approach to these questions has been simply to measure the size of African states or subsidiary institutions. Kirk-Greene uses this method to consider the nature of the British colonial state, focusing particularly on the ratio of colonial government officials to population in different colonies.49 Measurement of the incidence of specific government functions, such as court cases, the exercise of the death penalty, or the provision of education has also helped illustrate institutional change over time.50

Looking beyond scale to the quality of institutions is notoriously difficult. However, quantitative research in African history has relied on several different types of measures, both direct and indirect. One which has long been used but has become increasingly popular in recent years is fiscal capacity, including data on taxation, public spending, and public debt. Such measures are most commonly used in the colonial and postindependence periods, and to some extent for the precolonial period. Another more controversial measure is the quantitative rating of indigenous governments according to their structure as perceived by early anthropologists. A third measure is the incidence of known conflicts.

Fiscal data provide perhaps the most direct measure of institutional capacity over time, but their interpretation can be difficult and they are only available systematically from the beginning of the colonial period.51 The history of taxation is often entwined with the history of political development, with taxpayers demanding concessions from governments in exchange for increased revenue. Further, the state’s ability to “capture” revenue from a larger share of the population is also thought to imply greater administrative reach in a number of areas. However, there are numerous exceptions to this general history, and colonial taxation also involved considerable coercion.52 Further, different types of tax collection have different implications for state capacity.

The use of fiscal data therefore generally requires going beyond total revenue and expenditure figures (generally normalized by population in comparative studies) to examine the structure of that revenue or expenditure. For example, trade taxes, which comprised the largest revenue source for coastal countries (particularly in West Africa), are often collected in just a few locations from a relatively small number of taxpayers. In contrast, direct taxes require wider engagement with the population. These are often expressed in terms of shares of the total, or as per capita figures in and of themselves.53

Users of fiscal data need also to keep in mind the political context that produces them. In studies of colonial taxation, the most common source used in British Africa had been an annual volume of statistics sent from colonial governments to the imperial government, known as the Blue Books for the color of their binding. The Blue Books originated as a questionnaire circulated to colonial governments from 1822 on as part of a program to stamp out corruption in colonial administrations in the West Indies, in particular.54 Over the 19th century, they expanded to include data on public finances, trade, wages, prices, the size of the public administration, prisons, schools, and a number of other topics. Because they record data on an annual basis, at least for most colonies, and are at least very similar in format and content, they are an invaluable resource for comparative research on the British Empire. However, their origins as a monitoring device suggest caution in the interpretation of the data they present, which was at times manipulated to paint a better picture for the imperial government.

An example from Kenya illustrates this point. In the early 20th century, the colonial government relied heavily on the collection of flat-rate direct taxes on African dwellings and on adult males (referred to as “hut” and poll taxes). Collecting these taxes was particularly difficult in the northeast of the country, where much of the population was comprised of pastoralists who migrated across a wide area through the year. As an alternative, local administrators began collecting a percentage tax on cattle sales in certain towns, despite the fact that there was no basis in law for such a tax. They were warned by the central administration in Nairobi to refer to revenue from this tax as “hut tax” for the purposes of colonial accounting records reported in the Blue Books.55 Relying solely on the Blue Books therefore risks presenting a misleading picture of the experience of taxation if not carefully contextualized and paired with local archival research. The same is true of other measures of fiscal stability such as measures of sovereign risk. British authorities intervened proactively to maintain high prices for West African bonds, thus minimizing the appearance of risk to investors.56

Measures of institutional quality prior to the colonial period are more difficult to acquire. One source used frequently in recent work is the Ethnographic Atlas compiled by George Murdock, published in 1967.57 The Atlas, based on material previously published in the journal Ethnology, contains a variety of ethnographic information on 485 African societies, with variables ranging from those related to political structure to social stratification and modes of agricultural production. The sources underlying the Atlas are diverse and based on ethnographic readings from the Yale Library and Murdock’s personal collection of monographs and articles. The most widely used indicator from the Atlas is a measure of the degree of political centralization. This variable measures the jurisdictional hierarchy or political authority beyond the local community and ranges from stateless societies (no levels beyond the community) via petty chiefdoms and larger chiefdoms to states and large states. A number of recent papers have documented a correlation between jurisdictional hierarchy and current development outcomes.58

Societies were observed at varying times, depending on the earliest period for which Murdock could find satisfactory data. For most African societies, this period was generally the first quarter of the 20th century, after long histories of exchange with Europe and other regions, leading many to question whether the Murdock data provide a reasonable proxy for precolonial institutions.59 Instead, other work has interpreted the Murdock data as reflecting colonial observations of African societies in the early colonial period.60

A third political measure intended to look beyond the colonial period is the measure of conflict. Such data are compiled through the quantification of qualitative historical narratives, which can be accomplished simply by indicating which years saw conflict in particular regions and which did not. A more complex version can attempt to grade conflicts on their level of severity, for example, by measuring their duration or the number of fatalities. These data extend into the precolonial period and can thus be used to estimate the degree of conflict between African states.61

Geography, Environment, and Space

A final set of sources used particularly in an effort to get beyond colonial data are sources related to geography and the environment. Environmental data used in historical research can be divided into two categories. One is cross-sectional snapshots of items such as soil quality. Econometric research often uses the results of this work to proxy for the economic potential of a region and control for underlying economic differences, which might explain regional inequalities. Another example is environmental suitability for particular diseases, such as malaria or tsetse fly, in an effort to estimate their historical incidence.62

Environmental data can also be studied over time, either annually or seasonally. Rainfall data were collected by colonial governments in a number of countries and allow historians to investigate periods of potential hardship for agricultural producers.63 Data on precipitation can also be reconstructed from wind patterns recorded in ship logbooks for southern Africa.64

To Quantify or Not to Quantify

While the sources of data have multiplied in recent years, to what extent is using numerical data productive for historians? This question has been the subject of fierce debate since the controversies of the “cliometric revolution” of the 1960s.65 Critics of quantitative work argue that historical data are insufficiently precise for statistical analysis, that to classify things in a quantitative way often requires imposing anachronistic assumptions on historical actors, and that reducing historical change to numbers obscures the more human and compelling parts of history—what American Historical Association President Carl Bridenbaugh referred to in 1962 as the “dehumanizing methods of social sciences.”66 African history has seen similar criticisms and debates.67 Data quality (see, e.g., “Economic Performance”) may also be a concern, and poor data may mislead.

Despite the potential pitfalls of selectivity and low-quality data, which can also affect qualitative research, quantitative evidence has been of great value to African history in facilitating comparative research and overturning old orthodoxies. Comparisons of living standards and of institutions in Africa have helped to unlock some questions about economic and political development. Quantitative evidence has also reshaped the characterizations of particular periods of African history. For example, using prices in precolonial Dahomey, which showed that they rose in response to scarcity, Robin Law challenges the notion that precolonial African economies were not market driven.68

The discussion of the literature highlights that quantitative research on African history is often conducted not only by historians but also by scholars in other fields, including economists and political scientists. Historians skeptical of quantitative methods should also bear in mind, as Jarausch and Coclanis note, that “historians are better trained to handle and preserve data from the past than social scientists who often take figures at their face value.”69 Further, the most fruitful insights come from the use of quantitative and qualitative sources together, and collaboration between scholars in all of these areas has great potential for the future.

Discussion of the Literature

The “data revolution” in African history over the past two decades has been driven in part by a dialogue between two literatures. In one, current levels of poverty in African countries are linked to historical patterns. One criticism of papers in this area is that the mechanisms they propose for linking past and present are often very speculative, the empirical value of the papers being the econometric results establishing a causal role of historical events in shaping current outcomes. The second literature attempts to document change over time, thus avoiding “compression of history.” Over time, the unit of analysis has shifted from focusing largely on the national units created under colonial rule to examining subnational inequalities or even individual-level data. This work has documented the diversity of African experiences owing to the economic and political changes of the 19th and 20th centuries.

There remain several important gaps. One is that quantitative methods have mainly featured in research on economic history. However, there is considerable potential for work in other fields such as in the use of colonial, missionary, or medical records in social history. In part, these represent differences in training between the subfields, where economic historians are more likely to have had training in quantitative methods. As a result, the potential gains from coauthorship are considerable.

Further, much of this research is produced by scholars working outside Africa. Differences in both training and research incentives have limited the extent to which African scholars have participated in these discussions.70 This limitation is unfortunate both for restricting the perspective of quantitative research as well as for the resulting neglect of local sources of quantitative data with which scholars based in countries being studies are more likely to be familiar.

Finally, considerable spatial and temporal imbalances exist in the coverage of this work. Colonial and postindependence periods have received much closer attention than the precolonial period, and British and French Africa are the subject of more research than other parts of the continent, which can be attributed primarily to the greater availability of data, particularly in metropolitan archives. Greater collaboration with African scholars and researchers in other fields may help to resolve these omissions. However, the fact remains that not all aspects of African history can be quantified, and quantitative research must be paired with qualitative investigations to ensure that quantitative sources are appropriately contextualized.

Primary Sources

Previous sections cited a range of sources used in quantitative research on different topics. This section highlights the main ones, but also notes the diversity of primary sources used in quantitative research.

Precolonial Data

Few systematic statistics beyond European trade and shipping records survive from the precolonial period, and thus quantitative work in this area requires additional methodological flexibility. The Ethnographic Atlas codifies anthropological research to provide comparative measures of African societies in the early stages of European intervention along a number of dimensions, from agricultural production to social and political scholars. A number of scholars use this as a measure of precolonial status, but this source remains controversial as the actual date of observation was often after the beginning of colonial rule. Environmental data and genetic data can also provide quantitative glimpses into the distant past, albeit only indirectly. Linguistic analysis offers another potential quantitative source on the precolonial period.

Colonial Reports

Colonial reports, including budgets and trade reports, are perhaps the most common sources found in recent quantitative evidence. These fall broadly into two forms. The first and most common are annual statistical reports by British, French, Belgian, Portuguese, and Italian administrations, of which the British Blue Books are the most prominent in recent work. The second are one-off reports, censuses and surveys which give snapshots at specific periods.

Colonial government reports have the advantage of being readily accessible and often easily comparable across colonies but have the disadvantage of giving data only on what colonial officials thought worthy of measurement and reporting. These often favor economic and financial data linked to governance or external trade but neglect granular studies of the lives and livelihoods of the indigenous population. They should also be viewed as products of a political relationship between the colonial and imperial governments.

Microdata

Microdata have also come from a range of colonial sources, although these are often sources emerging from internal bureaucratic processes rather than aggregated reports. These include police and soldier records used in studies of heights, or census microdata used for a variety of purposes. Medical records from colonial government hospitals are another such source, which are used to study the engagement of Africans with colonial medical services as well as the effectiveness of the care they received. Some medical records come from missionary institutions, which also house marriage registers, another recent source of microdata. The collecting and processing of archive-based microdata is highly resource intensive, and there is little doubt that additional sources of this type will become available as the rewards of such research are illustrated.

Postindependence Data

The postindependence period presents a mixed picture. For most countries, the quantity of available data increases in terms of the range of topics about which data are collected. Sources of data have also expanded, encompassing governments and private sector actors as well as international organizations. However, there are variations in quality and availability. In countries affected by civil wars or other conflicts, frequent gaps exist. In others, political imperatives may incentivize manipulation of data and thus require cross-checking with other sources.71

Further Reading

Surveys and Debates

Broadberry, Stephen, and Leigh Gardner. “Economic Development in Africa and Europe: Reciprocal Comparisons.” Revista de Historica Economica 34 (2016): 11–37.Find this resource:

Fenske, James. “The Causal History of Africa: A Response to Hopkins.” Economic History of Developing Regions 25 (2010): 177–212.Find this resource:

Fourie, Johan. “The Data Revolution in African Economic History.” Journal of Interdisciplinary History 47 (2016): 193–212.Find this resource:

Hopkins, A. G. “The New Economic History of Africa.” Journal of African History 50 (2009): 155–177.Find this resource:

Jerven, Morten. Poor Numbers: How We Are Misled by African Development Statistics and What to Do About It. Ithaca, NY: Cornell University Press, 2013.Find this resource:

Examples of Quantitative Work on Africa

Population and Demography

Eltis, David, and David Richardson. Extending the Frontiers: Essays on the New Transatlantic Slave Trade Database. New Haven, CT: Yale University Press, 2008.Find this resource:

Frankema, Ewout, and Morten Jerven. “Writing History Backwards or Sideways: Towards a Consensus on African Population, 1850–2010.” Economic History Review 67 (2014): 907–931.Find this resource:

Manning, Patrick. “African Population: Projections 1850–1960.” In The Demography of Empire: The Colonial Order and the Creation of Knowledge. Edited by K. Ittman, D. D. Cordell, and G. H. Maddox. Athens, OH: Ohio University Press, 2010.Find this resource:

Thornton, John. “Demography and History in the Kingdom of Kongo, 1550–1750.” Journal of African History 18 (1977): 507–530.Find this resource:

Economic Performance

De Vries, Gaaitzen, Marcel Timmer, and Klaas de Vries. “Structural Transformation in Africa: Static Gains, Dynamic Losses.” Groningen Growth and Development Center Research Memorandum 136 (2013).Find this resource:

Magee, Gary B., Lorraine Greyling, and Grietjie Verhoef. “South Africa in the Australian Mirror: Per capita Real GDP in the Cape Colony, Natal, Victoria and New South Wales, 1861–1909.” Economic History Review 69 (2016): 893–914.Find this resource:

Marwah, Hanaan. “What Explains Slow Sub-Saharan African Growth? Revisiting Oil Boom-Era Investment and Productivity in Nigeria’s National Accounts, 1976–85.” Economic History Review 67 (2014): 993–1011.Find this resource:

Prados de la Escosura, Leandro. “Output per Head in Pre-independence Africa: Quantitative Conjectures.” Economic History of Developing Regions 27 (2012): 1–36.Find this resource:

Living Standards and Social Trends

Bolt, Jutta, and Ellen Hillbom. “Long Term Trends in Economic Inequality: Lessons from Colonial Botswana 1921–1974.” Economic History Review 69 (2016): 1255–1284.Find this resource:

Frankema, Ewout, and Marlous van Waijenburg. “Structural Impediments to African Growth? New Evidence from Real Wages in British Africa, 1880–1965.” Journal of Economic History 72 (2012): 895–926.Find this resource:

Microdata

Meier zu Selhausen, Felix, and Jacob Weisdorf. “A Colonial Legacy of African Gender Inequality? Evidence from Christian Kampala, 1895–2011.” Economic History Review 69 (2016): 229–257.Find this resource:

Moradi, Alexander. “Towards an Objective Account of Nutrition and Health in Colonial Kenya: A Study of Stature in African Army Recruits and Civilians, 1880–1980.” Journal of Economic History LXIX (2009): 719–754.Find this resource:

Government Institutions

Besley, Timothy, and Marta Rynal-Querol. “The Legacy of Historical Conflict: Evidence from Africa.” American Political Science Review 108 (2014): 319–336.Find this resource:

Frankema, Ewout, and Marlous van Waijenburg. “Metropolitan Blueprints of Colonial Taxation? Lessons from Fiscal Capacity Building in British and French Africa, c. 1880–1940.” Journal of African History 55 (2014): 371–400.Find this resource:

Gardner, Leigh. Taxing Colonial Africa: The Political Economy of British Imperialism. Oxford: Oxford University Press, 2012.Find this resource:

Gardner, Leigh. “Colonialism or Supersanctions: Sovereignty and Debt in West Africa, 1871–1914.” European Review of Economic History 21 (2017): 236–257.Find this resource:

Kirk-Greene, Anthony. “The Thin White Line: The Size of the British Colonial Service in Africa.” African Affairs 79 (1980): 25–44.Find this resource:

Geography and Environment

Alsan, Marcella. “The Effect of the TseTse Fly on African Development.” American Economic Review 105 (2015): 382–410.Find this resource:

Papaioannou, Kostadis J., and Michiel de Haas. “Weather Shocks and Agricultural Commercialization in Colonial Tropical Africa: Did Cash Crops Alleviate Social Distress?” World Development 94 (2017): 346–365.Find this resource:

Guides for Historians on Quantitative Research

Feinstein, Charles H., and Mark Thomas. Making History Count: A Primer in Quantitative Methods for Historians. Cambridge, UK: Cambridge University Press, 2002.Find this resource:

Jarausch, Konrad H., and Kenneth A. Hardy. Quantitative Methods for Historians: A Guide to Research, Data and Statistics. Chapel Hill: University of North Carolina Press, 1991.Find this resource:

Notes:

(1.) Jan Vansina, “Towards a History of Lost Corners in the World,” Economic History Review 35 (1982): 165–178.

(2.) Shantayanan Devarajan, “Africa’s Statistical Tragedy,” Review of Income and Wealth 59 (2013): S9–S15.

(3.) Stephen Broadberry, Bruce Campbell, Alexander Klein, Mark Overton, and Bas Van Leeuwen, British Economic Growth 1270–1870 (Cambridge, UK: Cambridge University Press, 2015).

(4.) Anthony G. Hopkins, “The New Economic History of Africa,” Journal of African History 50 (2009): 155–177.

(5.) Johan Fourie, “The Data Revolution in African Economic History,” Journal of Interdisciplinary History 47 (2016): 193–212.

(6.) The author thanks Deb Oxley for this interpretation.

(7.) One exception is the Cape Colony in what is today South Africa, for which considerable quantitative evidence is available for the 18th century. See Johan Fourie, “The Quantitative Cape: A Review of the New Historiography of the Dutch Cape Colony,” South African Historical Journal 66 (2014): 142–168.

(8.) Ethiopia was briefly colonized by Italy from 1936 to 1941.

(9.) Fourie, “The Data Revolution,” 194.

(10.) For a review of early quantitative work in African history, see Joseph P. K. Smaldone, “Quantitative Research in African History,” Historical Methods 10 (1976): 20–28.

(11.) For a review, see C. H. Feinstein and Mark Thomas, Making History Count: A Primer in Quantitative Methods for Historians (Cambridge, UK: Cambridge University Press, 2002), chap. 2.

(12.) Robert Szereszewski, “The Process of Growth in Ghana, 1891–1911,” Journal of Development Studies 1 (1965): 123–141; Jutta Bolt and Ellen Hillbom, “Long Term Trends in Economic Inequality: Lessons from Colonial Botswana 1921–1974,” Economic History Review 69 (2016): 1255–1284; Arne Bigsten, “Welfare and Economic Growth in Kenya, 1914–1976,” World Development 14 (1986): 1151–1160; Johan Fourie and Jan Luiten van Zanden, “GDP in the Dutch Cape Colony: The National Accounts of a Slave-Based Society,” South African Journal of Economics 81 (2013): 467–490; Gary B. Magee, Lorraine Greyling, and Grietjie Verhoef, “South Africa in the Australian Mirror: Per capita Real GDP in the Cape Colony, Natal, Victoria and New South Wales, 1861–1909,” Economic History Review 69 (2016): 893–914; Leandro Prados de la Escosura, “Output per Head in Pre-independence Africa: Quantitative Conjectures,” Economic History of Developing Regions 27 (2012): 1–36; Leandro Prados de la Escosura, “World Human Development 1870–2007,” Review of Income and Wealth 61 (2015): 220–247.

(13.) Stephen Broadberry and Leigh Gardner, “Economic Development in Africa and Europe: Reciprocal Comparisons,” Revista de Historica Economica 34 (2016): 11–37.

(14.) This is now a substantial literature. For a review, see Fourie, “The Data Revolution” or James Fenske, “The Causal History of Africa: A Response to Hopkins,” Economic History of Developing Regions 25 (2010): 177–212.

(15.) Hopkins, “New Economic History.” See also Gareth Austin, “The ‘Reversal of Fortune’ Thesis and the Compression of History: Perspectives from African and Comparative Economic History,” Journal of International Development 20 (2008): 996–1027.

(16.) Gareth Austin, “Resources, Techniques and Strategies South of the Sahara: Revising the Factor Endowments Perspective on African Economic Development, 1500–2000,” Economic History Review 61 (2008): 587–624; Anthony G. Hopkins, Economic History of West Africa (London: Longman, 1973); Jeremy I. Herbst, States and Power in Africa: Comparative Lessons in Authority and Control (Princeton, NJ: Princeton University Press, 2000), chap. 1.

(17.) J. Thornton, “Demography and History in the Kingdom of Kongo, 1550–1750,” Journal of African History 18 (1977): 507–530.

(18.) Linda Heywood and John Thornton, “African Fiscal Systems as Sources for Demographic History: The Case of Central Angola, 1799–1920,” Journal of African History 29 (1988): 213–228.

(19.) Dennis D. Cordell, Karl Ittman, and Gregory H. Maddox, “Counting Subjects: Demography and Empire,” in The Demography of Empire: The Colonial Order and the Creation of Knowledge (Athens, OH: Ohio University Press, 2010), 8.

(20.) Robert R. Kuczynski, Demographic Survey of the British Colonial Empire (Oxford: Oxford University Press, 1949).

(21.) Ewout Frankema and Morten Jerven, “Writing History Backwards or Sideways: Towards a Consensus on African Population, 1850–2010,” Economic History Review 67 (2014): 915–916; Keren Weitzberg, “The Unaccountable Census: Colonial Enumeration and Its Implications for the Somali People of Kenya,” Journal of African History 56: 409–428.

(22.) Shane Doyle, “Population Decline and Delayed Recovery in Bunyoro, 1860-1960,” Journal of African History 41 (2000): 429–458.

(23.) Patrick Manning, “African Population: Projections 1850–1960,” in The Demography of Empire: The Colonial Order and the Creation of Knowledge, ed. Karl Ittman, Dennis D. Cordell, and Gregory H. Maddox (Athens, OH: Ohio University Press, 2010).

(24.) Frankema and Jerven, “Writing History Backwards.”

(25.) Johan Fourie and Erik Green, “The Missing People: Accounting for the Productivity of Indigenous Populations in Cape Colonial History,” Journal of African History 56 (2015): 195–215; Sumner La Croix, “The Decline of the Khoikhoi Population, 1652–1780: A Review and a New Estimate,” University of Hawaii at Manoa Department of Economics Working Paper Series 16–22 (2016).

(26.) David Eltis and David Richardson, Extending the Frontiers: Essays on the New Transatlantic Slave Trade Database (New Haven, CT: Yale University Press, 2008).

(27.) Philip Curtin, “The White Man’s Grave: Image and Reality, 1780–1850,” Journal of British Studies I (1961): 94–110; “The End of the ‘White Man’s Grave’? Nineteenth-Century Mortality in West Africa,” Journal of Interdisciplinary History 21 (1990): 63–88.

(28.) Tom W. Shick, “A Quantitative Analysis of Liberian Colonization from 1820 to 1843 with Special Reference to Mortality,” Journal of African History XII (1971): 45–59; Antonio McDaniel, Swing Low, Sweet Chariot: The Mortality Cost of Colonizing Liberia in the Nineteenth Century (Chicago: University of Chicago Press, 1995).

(29.) Brian R. Mitchell and Phyllis Deane, Abstract of British Historical Statistics (Cambridge, UK: Cambridge University Press, 1962), 275.

(30.) These are sometimes also referred to as the “traditional” and “modern” sectors.

(31.) See, for example, Stephen Broadberry, Johann Custodis, and Bishnupriya Gupta, “India and the Great Divergence: An Anglo-Indian Comparison of GDP per Capita, 1600–1871,” Explorations in Economic History 55 (2015): 58–75.

(32.) Gaaitzen De Vries, Marcel Timmer, and Klaas de Vries, “Structural Transformation in Africa: Static Gains, Dynamic Losses,” Groningen Growth and Development Center Research Memorandum 136 (2013).

(33.) For more on the history of GDP, see Diane Coyle, GDP: A Brief but Affectionate History (Princeton, NJ: Princeton University Press, 2014); Moshe Syrquin, “A Review Essay on GDP: A Brief but Affectionate History by Diane Coyle,” Journal of Economic Literature 54 (2016): 573–588.

(34.) Morten Jerven, Poor Numbers: How We Are Misled by African Development Statistics and What to Do About It (Ithaca, NY: Cornell University Press, 2013).

(35.) Broadberry and Gardner, “Economic Development,” 19–20.

(36.) Nonso Obikili, “An Examination of Subnational Growth in Nigeria: 1999–2012,” South African Journal of Economics 83 (2015): 335–356.

(37.) Kai Hu et al., “A Scientometric Visualization Analysis for Night-Time Light Remote Sensing Research from 1991 to 2016,” Remote Sensing 9 (2017): 802–832; Nils B. Weidmann and Sebastian Schutte, “Using Night Light Emissions for the Prediction of Local Wealth,” Journal of Peace Research 54 (2017): 125–140.

(38.) Bigsten, “Welfare and Economic Growth,” 1151.

(39.) Ewout Frankema and Marlous van Waijenburg, “Structural Impediments to African Growth? New Evidence from Real Wages in British Africa, 1880–1965,” Journal of Economic History 72 (2012): 895–926; Robert C. Allen, Tommy E. Murphy, and Eric B. Schneider, “Una de cal y otra de arena: Building Comparable Real Wages in a Global Perspective,” Revista de Historia Economica 33 (2014): 61–75.

(40.) Pim de Zwart, “South African living standards in global perspective, 1835-1910,” Economic History of Developing Regions 26 (2011): 49–74.

(41.) Ewout Frankema and Marlous van Waijenburg, “Structural Impediments”; Sophia du Plessis and Stan du Plessis, “Happy in the Service of the Company: The Purchasing Power of VOC Salaries at the Cape in the 18th Century,” Economic History of Developing Regions 27 (2012): 125–149; Klas Rönnbäck, Labor and Living Standards in Precolonial West Africa: The Case of the Gold Coast (New York: Routledge, 2016).

(42.) Bolt and Hillbom, “Long Term Trends.”

(43.) Johan Fourie and Dieter von Fintel, “A History with Evidence: Income Inequality in the Dutch Cape Colony,” Economic History of Developing Regions 26 (2011); 16–48.

(44.) For a review and further reading, see Fourie, “The Data Revolution,” 200.

(45.) See, for example, Alexander Moradi, “Towards an Objective Account of Nutrition and Health in Colonial Kenya: A Study of Stature in African Army Recruits and Civilians, 1880–1980,” Journal of Economic History LXIX (2009): 719–754.

(46.) Felix Meier zu Selhausen, “Missionaries and Female Empowerment in Colonial Uganda: New Evidence from Protestant Marriage Registers 1880–1945,” Economic History of Developing Regions 29 (2014): 74–112; Felix Meier zu Selhausen and Jacob Weisdorf, “A Colonial Legacy of African Gender Inequality? Evidence from Christian Kampala, 1895–2011,” Economic History Review 69 (2016): 229–257; Felix Meier zu Selhausen, Marco H. D. van Leeuwen, and Jacob L Weisdorf, “Social Mobility among Christian Africans: Evidence from Anglican Marriage Registers in Uganda, 1895–2011,” Economic History Review (2017). For a critique, see Michiel de Haas and Ewout Frankema, “Gender, Ethnicity, and Unequal Opportunity in Colonial Uganda: European Influences, African Realities, and the Pitfalls of Parish Register Data,” Economic History Review (2018)

(47.) Johan Fourie, “The Remarkable Wealth of the Dutch Cape Colony: Measurements from Eighteenth-Century Probate Inventories,” Economic History Review 66 (2013): 419–448.

(48.) Thomas Bossuroy and Denis Cogneau, “Social Mobility in Five African Countries,” Review of Income and Wealth 59 (2013): S84–S110; Rebecca Simson, “(Under)privileged Bureaucrats? The Changing Fortunes of Public Servants in Kenya, Tanzania and Uganda, 1960–2010” (PhD diss., London School of Economics, 2017), chap. 4.

(49.) Anthony Kirk-Greene, “The Thin White Line: The Size of the British Colonial Service in Africa,” African Affairs 79 (1980): 25–44. A similar study looking beyond Anglophone Africa is Peter Richens, “The Economic Legacies of the ‘Thin White Line’: Indirect Rule and the Comparative Development of Sub-Saharan Africa,” African Economic History 37 (2009): 33–102. The size of the police force in colonial Malawi, including Africans, is measured in John McCracken, “Coercion and Control in Nyasaland: Aspects of the History of a Colonial Police Force,” Journal of African History 27 (1986): 127–147.

(50.) For different approaches to court records, see James Fenske, “Land Abundance and Economic Institutions: Egba Land and Slavery, 1830–1914,” Economic History Review 65 (2012): 527–555; Stacey Hynd, “Murder and Mercy: Capital Punishment in Colonial Kenya, ca. 1909–1956,” International Journal of African Historical Studies 45 (2012): 81–101. Another literature considers school enrollments. See, for example, Ewout Frankema, “The Origins of Formal Education in Sub-Saharan Africa: Was British Rule More Benign?,” European Review of Economic History 16 (2012): 335–355.

(51.) For a study of precolonial fiscal systems, see Linda Heywood and John Thornton, “African Fiscal Systems as Sources for Demographic History: The Case of Central Angola, 1799–1920,” Journal of African History 39 (1988): 213–228.

(52.) Odd-Helge Fjeldstad and Ole Therkildsen, “Mass Taxation and State-Society Relations in East Africa,” in Taxation and State-Building in Developing Countries: Capacity and Consent, ed. Deborah Bräutigam, Odd-Helge Fjeldstad, and Mick Moore (Cambridge, UK: Cambridge University Press, 2008), 114–134.

(53.) Ewout Frankema and Marlous van Waijenburg, “Metropolitan Blueprints of Colonial Taxation? Lessons from Fiscal Capacity Building in British and French Africa, c. 1880–1940,” Journal of African History 55 (2014): 371–400; Leigh Gardner, Taxing Colonial Africa: The Political Economy of British Imperialism (Oxford: Oxford University Press, 2012); Jens Andersson, “Long-Term Dynamics of the State in Francophone West Africa: Fiscal Capacity Pathways 1850–2010,” Economic History of Developing Regions 32 (2017): 37–70; Bas De Roo, “Taxation in the Congo Free State, An Exceptional Case?,” Economic History of Developing Regions 32 (2017): 97–126.

(54.) Zoe Laidlaw, Colonial Connections, 1815–1845: Patronage, the Information Revolution and Colonial Government (Manchester, UK: Manchester University Press, 2005).

(55.) Gardner, Taxing Colonial Africa, 57. See also “Primary Sources.”

(56.) Leigh Gardner, “Colonialism or Supersanctions: Sovereignty and Debt in West Africa, 1871–1914,” European Review of Economic History 21 (2017): 236–257.

(57.) George P. Murdock, “Ethnographic Atlas: A Summary,” Ethnology 6 (1967): 109–236.

(58.) See, for example, Sanghamitra Bandyopadhyay and Elliott Green, “Precolonial Political Centralization and Contemporary Development in Uganda,” Economic Development and Cultural Change 64 (2016): 471–508; Nicola Gennaioli and Ilia Rainer, “The Modern Impact of Precolonial Centralization in Africa,” Journal of Economic Growth 12 (2007): 185–234; Stelios Michalopoulos and Elias Papaioannou, “Precolonial Ethnic Institutions and Contemporary African Development,” Econometrica 81 (2013): 113–152.

(59.) Denis Cogneau and Yannick Dupras, “Questionable Inference on the Power of Precolonial Institutions in Africa,” PSE Working Papers n2014-25 (2014).

(60.) Morgan Henderson and Warren C. Whatley, “Pacification and Gender in Colonial Africa: Evidence from the Ethnographic Atlas,” MPRA Paper 6103 (2014).

(61.) Timothy Besley and Marta Rynal-Querol. “The Legacy of Historical Conflict: Evidence from Africa,” American Political Science Review 108 (2014): 319–336.

(62.) Marcella Alsan, “The Effect of the TseTse Fly on African Development,” American Economic Review 105 (2015): 382–410; David N. Weil, “The Impact of Malaria on African Development over the Longue Duree,” in Africa’s Development in Historical Perspective, ed. Emmanuel Akyeampong et al. (Cambridge, UK: Cambridge University Press, 2014), 89–130.

(63.) Kostadis J. Papaioannou and Michiel de Haas, “Weather Shocks and Agricultural Commercialization in Colonial Tropical Africa: Did Cash Crops Alleviate Social Distress?” World Development 94 (2017): 346–365.

(64.) Matthew J. Hannaford, Julie M. Jones, and Grant R. Bigg, “Early Nineteenth-Century Southern African Precipitation Reconstructions from Ships’ Logbooks,” The Holocene 25 (2015): 379–390.

(65.) Jan W. Drukker, The Revolution that Bit Its Own Tail: How Economic History Changed Our Ideas on Economic Growth (Amsterdam: Transaction Press, 2006).

(66.) Carl Bridenbaugh, “The Great Mutation,” American Historical Review 68 (1963): 326. A more recent contribution making a similar point is Nick Cullater, “The Foreign Policy of the Calorie,” American Historical Review 112 (2007): 337–364.

(67.) For example, see discussion between Tony Hopkins and James Fenske: Fenske, “The Casual History”; Hopkins, “New Economic History.” See also critique of the Slave Voyages dataset in Bwedolyn Midlo Hall, “Africa and Africans in the African Diaspora: The Uses of Relational Databases,” American Historical Review 115 (2010): 136–150.

(68.) Robin Law, “Posthumous Questions for Karl Polanyi: Price Inflation in Precolonial Dahomey,” Journal of African History 33 (1992): 387–420.

(69.) Konrad H. Jarausch and Peter A. Coclanis, “Quantification in History,” in International Encyclopedia of the Social & Behavioral Sciences, ed. Neil J. Smelser and Paul J. Baltes, 2nd ed., Vol. 19 (Oxford: Elsevier, 2015).

(70.) For discussion of these issues as they relate to economic history, see Johan Fourie and Leigh Gardner, “The Internationalization of Economic History: A Puzzle,” Economic History of Developing Regions 29 (2014): 1–14; Erik Green and Pius Nyambara, “The Internationalization of Economic History: Perspectives from the African Frontier,” Economic History of Developing Regions 30 (2015): 68–78; Gareth Austin, “African Economic History in Africa,” Economic History of Developing Regions 30 (2015): 79–94.

(71.) For example, Marwah (2014; see “Further Reading”) draws different conclusions about investment and growth in Nigeria after independence using private corporate records as compared with official national accounts.