David Eltis, Stephen D. Behrendt, David Richardson, and Herbert S. Klein, The Trans-Atlantic Slave Trade: A Database on CD-ROM. Cambridge and New York: Cambridge University Press, 1999. (User guide, xi + 89 pp.) $195, ISBN: 0-521-62910-1.
The Trans-Atlantic Slave Trade Database is a monumental achievement that brings together in a single multisource data set the results of over thirty years of international research undertaken by many individual scholars working in English, Portuguese, Danish, French, Spanish, and Dutch on the largest transoceanic migration of any people prior to the outpouring of Europeans to the New World in the nineteenth century. The authors entertain "grand hopes" for this extraordinary resource, sponsored by the W.E.B. Du Bois Institute at Harvard University with additional support from the National Endowment for the Humanities, the Mellon Foundation, and the Ford Foundation, and taking seven years to complete. It will, they contend, not only "enable historians to develop new insights into the history of peoples of African descent and the forces that determined their forced migration," but will also "greatly facilitate the study of cultural, demographic, and economic change in the Atlantic world from the late sixteenth to the mid-nineteenth centuries" (p. 2). Are the hopes of the authors, the sponsors, and of Cambridge University Press justified? Or is the publication of this database more likely simply to facilitate an expansion of the esoteric and sometimes contentious numbers games that have frequently characterized slave trade studies? My sense is that the project is indeed likely to stimulate new research not just on the slave trade narrowly conceived, but also on a range of broader economic, demographic, and cultural issues on both sides of the Atlantic.
The database includes records of 27,233 trans-Atlantic slave ship voyages made between 1595 and 1866, accounting for between two-thirds and three-quarters of all trans-Atlantic slave voyages sailing after 1600. (Independent estimates of the volume of the trans-Atlantic slave trade after 1600 yield a "scholarly consensus" figure of 11.4 million departures from Africa and 10 million arrivals in the Americas. This suggests a total of between 34,482 and 35,561 slaving voyages in this period.) The authors standardized existing data sets compiled by individual researchers, collated voyages that appeared in several different data sets, and added new information from previously unexplored sources. Extraordinary care was used in the initial compilation and in subsequent editing of the database; the portion that I have used (and rechecked against original primary sources) is exceptionally accurate. Coverage of the British trade is fullest (the authors estimate that 90 percent of all voyages are included) and the eighteenth-century French and Dutch trades are also largely complete. Bigger gaps exist for the Portuguese, seventeenth-century French, and nineteenth-century Spanish, Danish, and North American trades. However the authors contend that the set "provides samples large enough to present the major trends over time" (p. 5).
Each entry in the database consists of a single slaving voyage, for which up to 226 pieces of information may be available. These include 162 data variables incorporating information collected from the sources such as dates at which the ship left from or arrived at various destinations during the voyage; ports of origin, slave purchase, and delivery; number of slaves embarked and disembarked, their demographic composition and mortality levels; details of ship construction, registration, armament, and crew size; names of captains and owners; the outcome of the voyage; and archival sources. An additional 64 imputed variables are calculated or imputed from the data to compensate for missing information and to facilitate analysis by consolidating or regrouping variables that have unwieldy numbers of individual codes. These include consolidation of geographic locations into regional and continental categories, and grouping of voyages into different temporal categories (year the voyage originated, year in which slaves were embarked, and year of disembarkation, and for the last also into periods of 5, 25, and 100 years). Outcomes of voyages (successful completion, wreck, capture, or insurrection somewhere en route, etc.) are reclassified in three ways from the perspective of slaves, captors, and owners. Other inferred variables group locations into major trading regions, estimate the numbers embarked or disembarked where full information is not available, and regroup data on age, sex, and mortality. How the estimations were made is clearly documented (the SPSS program that creates the imputed variables is included on the CD-ROM), so users can easily substitute different groupings or estimations.
The usefulness of the database for refining conventional slave trade studies is obvious, but it is the broader applications that go well beyond core issues such as the volume and demographic structure of the trade, Middle Passage mortality, and shipping productivity -- indeed far beyond the slave trade itself -- that are the most exciting. On the African side, data on slave exports from specific coastal outlets afford insights into the slave and commodity trades of particular African subregions and even single ports, and into African agency through resistance to the trade as evidenced in ship insurrections, as well as broader economic and demographic results. The authors contend that "the large role of Africans in the Atlantic world" is "perhaps the single most important preliminary feature to emerge from these new data" (p. 35). As the largest data base on any transoceanic trade of the seventeenth and eighteenth centuries, it facilitates research into European long-distance shipping activities, including investigations of changing shipping technologies and, from records of ports of origin, ship owners, and captains, connections between the slave and other colonial trades. For the Americas, new information on the numbers and demographic composition of forced migrants brought to particular destinations suggests a need to reassess current understanding of at least some local population histories. Even more importantly, the data set makes available more precise information on which parts of Africa supplied the different parts of the slaveholding Americas. Expanded evidence on the African origins of forced migrants will allow scholars to explore the impact of African heritage on New World societies, and to better assess patterns of cultural retention and adaptation. Most of the papers and articles initially derived from the data set (listed on pp. 55-56) deal with traditional slave trade topics. However David Eltis's The Rise of African Slavery in the Americas (Cambridge University Press, 2000) -- which builds on the database -- provides a striking example of how such quantitative measures can be utilized in combination with qualitative materials to address broader comparative economic and cultural issues including topics such as gender, ethnicity, and value systems.
Unlike many digitized data sets available through sources such as ICPSR that require some level of technical expertise to manipulate, the data on individual voyages can be readily accessed, queried, and rearranged, some basic analyses obtained, and the results graphed, viewed on interactive maps, and printed out simply by clicking on pull-down menus. Selected subsets can be saved for subsequent reference or downloaded into SPSS data files for modification and more refined analysis according to individual needs or preferences. Thus the database has importance for several different audiences. Scholars interested in quantitative aspects of the trans-Atlantic slave trade are but the most obvious group. It is also a readily accessible reference work that ought to be available to non-specialist as well as specialist users of college and museum libraries. Finally, it is a marvelous teaching resource, both for supplementing other course materials and from which students with varying levels of technical expertise can develop a wide range of research projects.
A closing caveat is perhaps in order since my enthusiastic endorsement of The Trans-Atlantic Slave Trade Database is definitely not that of a disinterested reader or casual user. Invited to present a paper exploring a subset of the database pertaining to the Chesapeake region for a conference held at Williamsburg in 1998, I at first imagined that this would entail little more than a cursory review of some revised numbers. That supposedly limited foray has since expanded into a multi-year research project incorporating additional information on Chesapeake slaving voyages; tracing connections between the slave, indentured servant, and tobacco trades; exploring social, demographic, and cultural implications for the region; and developing museum interpretations underscoring trans-Atlantic interconnections. My experience is surely not unique. An example of a related project is Gwendolyn Midlo Hall, ed., Databases for the Study of Afro-Louisiana History and Genealogy, 1699-1860: Computerized Information from Original Manuscript Sources (Louisiana State University Press, 2000).
Until recently compilations of slave trade statistics have seemed to reduce one of the darkest episodes in world history into a set of abstract and bloodless figures. This commendable collaborative effort now offers scholars not only an authoritative source for developing better statistics, but also an extraordinary resource from which to begin translating those esoteric numbers back into a more humanized history.
(System requirements: The following configuration is recommended to run the CD-ROM: Windows 95, 98, or NT operating system; 166 MHz Pentium processor; 32 MD RAM; 800 x 600 monitor resolution x 65,536 colors (16 bits); 6x speed CD-ROM drive; 84 MD available hard disk space.)
Lorena S. Walsh is the author of From Calabar to Carter's Grove: The History of a Virginia Slave Community (University of Virginia Press, 1997), and of "The Chesapeake Slave Trade: Regional Patterns, African Origins, and Some Implications," William and Mary Quarterly, forthcoming, January 2001.
Copyright © 2000 by EH-Net, all rights reserved. This work may be copied for non-profit educational use if proper credit is given to the author and the list. For other permission, please contact the EH.Net Administrator (email@example.com; Telephone: 513-529-2850; Fax: 513-529-3308). Published by EH.Net (October 2000). All EH.Net reviews are archived at www.eh.net/BookReview.
Posted: 19 October 2000