Construction of the Trans-Atlantic Slave Trade Database: Sources and Methods
Resistance and Price of Slaves
Finally, we consider two variables that will attract more attention than most. First, we use the data variable “African resistance” (RESISTANCE) to compute a “Rate of Resistance” variable which is available only on the time line and in Custom Graphs. This is simply the number of vessels experiencing some recorded act of resistance divided by the total number of vessels in a given year and expressed as a percentage. The second is the “Sterling cash price (of slaves) in Jamaica” variable (JAMCASPR), which may be used to track the price paid for slaves in the Americas as they were sold from the vessel. The Voyages Database contains prices for those on board 959 voyages. The full derivation of these data is described elsewhere, but a summary description is appropriate here. (23) Prices for human beings in the Americas were subject to as many influences as were prices in any other market. Key factors included the characteristics of the person being sold, the distance between slave markets in the Americas and Africa and the price of the captive in Africa. This variable attempts to adjust for several of these factors so that the underlying price trends become apparent to the user of Voyages. In most cases the data are taken from the slave traders’ accounts and correspondence. Our first goal was to ensure that we recorded a single category of captive – what was frequently referred to at the time as “a prime male”.(24) Second, we adjusted that price for the price differential between the market in which the slave was actually sold and the price in Jamaica. Thus, if the captive was sold in one of the eastern Caribbean islands we would make a small adjustment upwards to reflect the ten extra days sailing time it would take to reach Jamaica. Third, we converted all prices into pounds sterling. What we did not do was to express the price in constant pounds (adjusted for inflation) – in other words, in real terms. This variable is thus based on archival data, but the adjustments we have made in the interests of making it intelligible for users have the effect of converting this into an imputed variable.
The above discussion is not exhaustive in the sense that we not have touched on and explained every single variable in either the Voyages interface or the two databases offered for downloading. Many of the variables need no more explanation than is available on the Variable List page and readers are referred to this page for any missing explanations in the preceding pages.