Article contents
Mortality on Long-Distance Voyages in the Eighteenth Century
Published online by Cambridge University Press: 03 March 2009
Abstract
![Image of the first page of this content. For PDF version, please use the ‘Save PDF’ preceeding this image.'](https://static.cambridge.org/content/id/urn%3Acambridge.org%3Aid%3Aarticle%3AS0022050700044375/resource/name/firstPage-S0022050700044375a.jpg)
- Type
- Discussion
- Information
- Copyright
- Copyright © The Economic History Association 1981
References
Notes
1 Struyck, Nicolaas, Les oeuvres de Nicolas Struyck (1687–1769), Vollgraff, J. A., ed. and trans. (Amsterdam, 1912), pp. 361–62.Google ScholarFurther interpretation of mortality among Dutch East India voyagers may be found in Bruijn, J. R., “De personeelsbehoefte van de VOC overzee en aan board, bezien in Aziatisch en Nederlands perspectief,” Bijdragen en mededelingen betreffende de geschiedenis der Nederlanden, 91 (1976), especially pp. 221–24. Bruijn's data indicate an average voyage-specific mortality on the outbound passage in the period 1730–1740 of 12 ½ per cent, which was somewhat higher than the usual rate on all outbound voyages.Google ScholarPer voyage experience may be followed in Bruijn, J.R., Gaastra, F.S., and Schöffer, I., eds., Dutch Asiatic Shipping in the 17th and 18th Centuries, 3 vols. (The Hague, 1979). Mortality rates have been computed by dividing the number of deaths on the voyage by the population at the midpoint of the voyage (here assumed to be the beginning population less half the deaths on the voyage). Russell Menard kindly pointed out to me the need to adopt this method rather than the more commonly seen computation that divides all deaths by the beginning population. (For the Breslau and model life table populations dealt with in Table 3, this adjustment is not needed because the data given are midyear populations.)CrossRefGoogle Scholar
2 Miller, Joseph C., “Mortality in the Atlantic Slave Trade: Statistical Evidence on Casuality,” Journal of Interdisciplinary History, 11 (Winter 1981), 388 and passim 385–423, argues that in ordinary voyages slaves died in largest numbers early in the voyage, after which the death rate declined. On unusually long voyages, however, the death rate again increased as food and water were exhausted. As Miller acknowledges, this interpretation of the available evidence awaits verification. In any event, the pattern of deaths within a voyage may indeed have been as Miller suggests without calling into question Struyck's observation that slower voyages had higher mortality levels. Struyck, and most historians of mortality on these voyages up to Miller, thought about death rates in voyage-specific terms. Their statements that the death rate increased when the average voyage lengthened meant that the longer a cargo of slaves remained at sea the larger would be the portion of slaves that would die. Sometimes these historians implied that the trend of the rate of loss turned upward with longer voyages. Miller's contribution is to show that higher rates for long voyages and lower rates for short voyages do not by themselves tell us anything about the trend of the rate of loss during the voyage.CrossRefGoogle Scholar
3 Stein, Robert L., The French Slave Trade in the Eighteenth Century: An Old Regime Business (Madison, 1979), p. 100; andGoogle Scholaridem., “Mortality in the Eighteenth-Century French Slave Trade,” Journal of African History, 21, 1 (1980), 35–41. See also Klein, Herbert S., The Middle Passage: Comparative Studies in the Atlantic Slave Trade (Princeton, 1978), passim, especially pp. 64–71, 86–93, 160–63, 193–203, and 229–41;CrossRefGoogle ScholarPostma, Johannes, “Mortality in the Dutch Slave Trade, 1675–1795, ” in Gemery, Henry A. and Hogendorn, Jan S., eds., The Uncommon Market: Essays in the Economic History of the Atlantic Slave Trade (New York, 1979), pp. 239–60;Google ScholarKlein, Herbert S. and Engerman, Stanley L., “A Note on Mortality in the French Slave Trade in the Eighteenth Century,” in Gemery and Hogendorn, The Uncommon Market, pp. 261–72;Google ScholarKlein, Herbert S. and Engerman, Stanley L., “Shipping Patterns and Mortality in the African Slave Trade to Rio de Janeiro, 1825–1830,” Cahiers d'études africaines, 15 (1975), 381–98; andCrossRefGoogle ScholarKlein, Herbert S. and Engerman, Stanley L., “Facteurs de mortalité dans le trafic français d'esclaves au XVIIIe siècle,” Annales économies sociétés civilisations 31 (11–12 1976), 1213–1224.CrossRefGoogle Scholar Most students of this question report that mortality declined over time, but the available data are sporadic in time and place. Information collected by Klein, Middle Passage, pp. 64–68, suggests the following pattern of voyage-specific mortality among slaves:.Google Scholar
4 Postma, “Mortality,” pp. 244–46; and Klein and Engerman, “Shipping Patterns,” p. 386.Google Scholar
5 Miller, “Mortality,” pp. 400 and 403–04.Google Scholar
6 In the literature on slave mortality considerable ambiguity arises because the time at issue in death rates is sometimes left vague; moreover, since the length of slave voyages tended to be reduced over time, the period of time implied in voyage-specific statements of death rates changes. This ambiguity might be ended by stating mortality rates at sea in daily, monthly, or annual rates, but of those a statement of monthly rates seems preferable. As Miller, “Mortality,” has shown, daily rates are quite useful for certain purposes, and they produce more precise rates. But daily rates are very small numbers. Annual rates might be taken to imply that mortality at the extraordinary levels of sea voyages was sustained for such a long period when in fact few voyages lasted a year and most were completed in a much shorter period. Monthly rates have the advantage of providing a common measure that can readily be used for comparative purposes while preserving sufficiently large figures. Either a linear or an exponential method of calculation might, in most cases, be used:Google Scholar
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20151017043611291-0710:S0022050700044375_eqn1.gif?pub-status=live)
exponential: Pt = P0 (1−k)t, where Pt, survivors at end of voyage, P0 = population at beginning of voyage, t = length of voyage in months, and k monthly probability of death. In most cases the difference between linear and exponential rates will be quite small (e.g., minimum per month slave mortality on French vessels: linear = 23.70, and exponential 23.74). The exponential method may be considered more appropriate because, if the probability of dying remains the same each month, the number of deaths will decline exponentially. But because the difference in result between the methods usually will be very small, in practice the linear method may be preferred. I wish to thank George Alter for explaining these distinctions to me. Monthly probabilities of death are converted to monthly death rates, which are central rates (i.e., deaths divided by the mid-month population), by the formula: k / (1 – 1/2k).Other methods of stating mortality rates remain useful. For example, in many instances voyage- specific rates will be preferred because they yield a merchant's rather than a demographer's measure of loss.
7 Coale, Ansley J. and Demeny, Paul, Regional Model Life Tables and Stable Populations (Princeton, 1966), pp. 6 and 8 of the tables;Google ScholarShaw, A.G.L., Convicts and the Colonies (London, 1966), pp. 116 and 363–64, for male and female (but mostly male) convicts transported to Australia andGoogle ScholarDavies, K. G., “The Living and the Dead: White Mortality in West Africa, 1684–1732,” in Engerman, Stanley L. and Genovese, Eugene D., eds., Race and Slavery in the Western Hemisphere: Quantitative Studies (Princeton, 1975), pp. 88–89, for Royal African Company employee deaths. Male model tables have been used because shipboard populations are believed to have been predominantly male in most cases. These calculations follow a stationary population approach.Google Scholar For the Nantes crew mortality rates I infer an average voyage length of 81 days (2.7 months) from Stein, “Mortality,” p. 37. From ibid., p. 41, table 8, I infer a weighted voyage-specific probability of death of 132.75 per 1000 and estimate passage times. For the coastal and middle passage taken together, note that the figure of 6 months represents an attempt to estimate average time spent on board by slaves.
8 It is difficult to give examples of slave mortality because, on the one hand, all the relevant information is seldom given in sources where this issue is considered and, on the other hand, available data are often drawn from different periods and relate to different trade routes (and thus voyage lengths and mixtures between coastal and middle passage portions) to the point at which averages drawn from them may have very little, perhaps no, value. Klein, Middle Passage, p. 55, reports an average mortality of 93 per thousand among 162,225 slaves arriving at Rio de Janeiro between 1795 and 1811. That (apparent) percentage may be converted to a mortality rate of 97.54 per thousand for the average voyage. But how long was the average voyage? In Klein and Engerman, “Shipping Patterns,” p. 386, an average saili.ng time of 39.1 days is reported for slaving ships arriving at Rio de Janeiro from several African regions between 1825 and 1830. If it is reasonable to use this average sailing time to interpret data on mortality among slaves making the passage to Rio between 1795 and 1811, the monthly mortality rate would then be 74.84 per thousand.Google Scholar
9 Stein, “Mortality,” pp. 36–37 and pp. 40–41.Google Scholar
10 Klein, Middle Passage, pp. 69–71 and p. 236.Google Scholar
11 Halley, Edmund, Degrees of Mortality of Mankind, Reed, Lowell J., ed., (Baltimore, 1942), p.6. Breslau's age structure is not known, so it is not possible to correct Halley's estimates. Struyck discusses Halley's work in Les oeuvres.Google Scholar
- 8
- Cited by