Generalization in ecology and evolutionary biology: From hypothesis to paradigm

We argue that broad, simple generalizations, not specifically linked to con tingencies, will rarely approach truth in ecology and evolutionary biology. This is because most interesting phenomena have multiple, interacting caus es. Instead of looking for single universal theories to explain the great d iversity of natural systems, we suggest that it would be profitable to deve lop general explanatory 'frameworks'. A framework should clearly specify fo cal levels. The process or pattern that we wish to study defines our level of focus. The set of potential and actual states at the focal level interac ts with conditions at the contiguous lower and upper levels of organization , through sets of many-to-one and one-to-many connections. The number of in itiating conditions and their permutations at the lower level define the po tential states at the focal level, whereas the actual state is constrained by the upper-level boundary conditions. The most useful generalizations are explanatory frameworks, which are road maps to solutions, rather than solu tions themselves. Such frameworks outline what is understood about boundary conditions and initiating conditions so that an investigator can pick and choose what is required to effectively understand a specific event or situa tion. We discuss these relationships in terms of examples involving sex rat io and mating behavior, competitive hierarchies, insect life-histories and the evolution of sex.