Sentenced without trial : re iling and re amping the Resource Dispersion Hypothesis

The resource dispersion hypothesis (RDH) offers a predictive model for how, in heterogeneous environments, some animals may form groups even without any direct benefits from doing so. Revilla (2003) levels a number of criticisms at the RDH, all of which we feel can be easily clarified or resolved (we address them in the order they appear in his critique). We argue that the RDH, despite its long history and despite our own efforts, still lacks convincing evidence to reject it. On the contrary, we see it as having weathered its critics very well. While there is steadily accumulating evidence in support of the RDH, there are very few empirical tests that provide decent evidence against it. During his Master’s degree, one of us (DDPJ) wrote a highly sceptical essay hacking away at assumptions of the RDH and arguing that it did not, on its own, provide a satisfactory explanation of sociality. Five years later, after thinking about it and working on it most of the days in between, he still holds those same views. However, one thing has most certainly changed. He is now convinced that good tests of the RDH (either trying to demonstrate or to falsify it) have simply not been done. As a result, theoretical arguments against the RDH remain essentially ineffectual because, even if it can be postulated that certain aspects of the RDH model seem unlikely (or unimportant), it remains to be demonstrated that these shortcomings are the case empirically. That is the challenge we still face. The underlying logic of the RDH still holds as relevant now as it did when it was first conceived (Macdonald 1983a), or when it first appeared as a formal model (Carr and Macdonald 1986), or when it was corroborated by more rigorous mathematical modelling (Bacon et al. 1991). After a critique published soon after the original paper (von Schantz 1984), the RDH – in its more developed theoretical form – remains unattacked by any published theoretical criticisms. Empirically, there has only been one study we know of (until Revilla’s work) that claims a falsification of RDH predictions (Baker et al. 2000), and their conclusions have been questioned (Blackwell 2001). But it is important to realise that, even if there were studies successfully falsifying the RDH in this or that location, or in this or that species, they would not provide good reason to reject it as a model elsewhere. As Roger Peters was at pains to remind us (Peters 1991, p. 53):