IJEE Soapbox: on the uses and abuses of comparative demography

In a foreign country, few pleasures can match being in the field with a local biologist, steeped in the reticulate details of local natural history and passionate about the task of understanding the rhythms of the natural world. My wife Lynne and I recently spent an enjoyable day with Professor Steven Piper exploring the splendor of the Drakensberg Mountains near Durban, South Africa, seeking out highly localized endemic species of birds with lovely names such as the Orange-breasted Rockjumper and Gurney’s Sugarbird. Professor Piper is well-known in the ornithological community for his long-term population studies in southern Africa, and in particular for his studies of two endangered species of vulture found in these mountains, the Cape Vulture and Bearded Vulture. To monitor these birds, he and his assistants must put bands on the legs of adult and juvenile birds caught at the nest. To do that, they must first risk life and limb by clambering up cliffs, and then put up with the stomach-churning behavioral responses (I will spare you, the reader, the details) of vultures indignant at being manhandled. By tracking a number of banded individuals over time, vital statistics such as per capita birth and death rates can eventually be estimated. When carried out over a sufficiently long time period, and at a number of sites, a picture of how demographic processes vary over space and through time is gradually built up, with each hard-won datum implicitly representing many hours of human sweat and effort. The literature of ecology, conservation, and wildlife biology is implicitly replete with stories like this. (I sometimes think that it would be very instructive for publications to have appendices, quantifying the magnitude of human effort that has gone into producing each figure and table.) Given how difficult it is to gather comparative demographic data, it is important to think very clearly about how such data can be used, and the conceptual traps that may arise when interpreting demographic patterns. One such trap goes as follows: It seems intuitive that if a hypothesized environmental factor impacting