Ecology is usually described as the study of organisms interacting with one another and their environments. From this view of ecology, viruses – not usually considered to be organisms – would merely be part of the environment. Since the late 1980s, however, a growing stream of micrographic, experimental, molecular, and model-based (theoretical) research has been investigating how and why viruses should be understood as ecological actors of the most important sort. Viruses, especially phage, have been revealed as participants in the (...) planet's most crucial food webs, even though viruses technically consume nothing (they do not metabolize by themselves). Even more impressively, viruses have been identified as regulators of planetary biogeochemistry, in which they control cycles such as carbon, nitrogen and phosphorus – cycles on which all life depends. Although much biogeochemical research black-boxes the entities filling functional roles, it is useful to focus a little more closely to understand how viruses can be held responsible for the global processes of life. This paper will give a brief overview of the history of virus ecology and tease out the implications of large-scale ecological modelling with viruses. This analysis suggests that viruses should be conceptualized as ecological actors that are at least comparable and possibly equal to organismal actors. Ecological agency can therefore be distinguished from standard interpretations of biological agency. (shrink)

Theoretical biology and economics are remarkably similar in their reliance on mathematical models, which attempt to represent real world systems using many idealized assumptions. They are also similar in placing a great emphasis on derivational robustness of modeling results. Recently philosophers of biology and economics have argued that robustness analysis can be a method for confirmation of claims about causal mechanisms, despite the significant reliance of these models on patently false assumptions. We argue that the power of robustness analysis has (...) been greatly exaggerated. It is best regarded as a method of discovery rather than confirmation. (shrink)

Some ecologists have argued that theoretical model building in population and community ecology has gone evidentially unconstrained. In the essay, I argue that "bottle experiments" offer ecological model building evidential constraints and illustrate this by considering work on chaotic models tested by the dynamics of flour beetles. Critics reply that these experiments are importantly unlike nonmanipulated natural systems and thus do not constitute genuine tests of the models. I conclude by considering two responses to this worry and a suggestion on (...) how to move forward on this debate. (shrink)