Fishing at Gwaimasi: the interaction of social and ecological factors in influencing subsistence behaviourThis treatise describes fishing behaviour of people living at the village of Gwaimasi in the interior lowlands of Papua New Guinea. The concern is not with fishing as such but wdth the way in which social and ecological factors interact to influence behaviour. Iargue that an adequate explanation for behaviour must incorporate both dimensions, and suggest a conceptual basis for achieving this. The study then illustrates how at least one aspect of social organization, the size of resource-sharing groups, could be incorporated within a frame of analysis -evolutionary ecology -that has typically focussed on the explanatory significance of ecological factors.The primary thesis asserts that the probability of a particular pattern of behaviour being reproduced depends on both social and ecological factors. Variation in either domain can be expected to affect the kinds of behaviour observed, but the two domains do not affect behaviour in the same way. Ecological factors constrain production, the material outcome of action; social factors constrain consumption, the use that can be made of that outcome. The two domains are complementary, not contradictory, in their influence; in as much as production and consumption themselves are two facets of any action, ecology and society should be seen as mutually constitutive systems mediated by the actor.Evolutionary ecologists have tended to analyse constraints on production, particularly extrinsic constraints on production, in explaining behavioural variability.While constraints on consumption are not ignored, they are usually discussed only in terms of the intrinsic requirements of the actor. I argue that consumption, like production, is subject to extrinsic constraints; the use that an individual can make of a particular outcome of action may depend on the actions of others. In particular, where a resource has declining marginal value, the amount that is likely to be received from others will limit the amount that an individual can usefully procure; as a corollary, the iii expectation of having to distribute produce among others will increase the amount that can be usefully procured. In each case, the effect will depend on the size of the resourcesharijig group.These arguments are illustrated by analyses of the fishing behaviour of people atGwaimasi over a period of 57 weeks, Eariy chapters set the scene for those analyses, introduce the actors and position fishing within their broader subsistence arrangements.The community was small and isolated, the people hunter-horticulturalists using about 50km^ for subsistence. Skulls of neariy all fish caught within that area during the survey were purchased, and details recorded of fisher, technique, location and context of capture.In addition, records were kept of where individuals were based within the local area each day. Analyses of these data indicate that whether people chose to fish, as well as where and how they fish...