In recent years the research program labelled selection theory has seen a number of extensions that would certainly have pleased Donald T. Campbell. My opinion is that this expansion of the framework is a necessary evolutionary step in the long-term process of evaluation of the progressiveness of the program. However, even though such an expansion is necessary, it has not been generally accompanied by a more philosophical assessment and reconsideration of the fundamentals of the program. I wish to avert this trend. Campbell (1974) spoke of the thesis of the universality of selection as the “dogma” of his approach. I wish to reconsider, by exploring Campbell’s work and the recent work on selection theory, whether the universality thesis can be genuinely treated as an empirical hypothesis, and, above all, if it is a sound one. After Campbell proposed it, the thesis in all its different formulations was either snubbed or happily endorsed, but never seriously criticized. More recently a number of well-articulated criticisms have been raised against the universalism of the thesis, generally pointing to its more circumscribed range of application. With my project I first wish to analyze the nature and status of Campbell’s blind-variation-and- selective-retention model. In particular, I wish to focus on the metaphysical issues concerning the nature of selection processes (i.e. their logic and range of application, and the nature of the various formulations of the universality thesis) and on their characterization (i.e., the nature of the slippery notion of “blindness” of variation involved, their populational requirements, their limits). Being primarily an evolutionary epistemologist, I would then like to move to reconsider the epistemological implications of the universality thesis. The trend in recent years has been to give up the search for a logic of science in favor of a sociology of scientific validity. For instance, Hull and Campbell (1997) put stronger emphasis on the sociological analysis of the scientific process, seemingly in line with the naturalism of their epistemological approaches. Even though I consider the sociological approach as an indispensible part of a complete evolutionary epistemology, I believe that something has been left out by endorsing a hard-core sociological perspective. This is why I would like to investigate whether there are any good reasons to revive our interest in the “logic” of selection. My suspicion is that selection theory can provide a “logic” of discovery and justification (or, more properly, a sound account of the scientific processes of hypotheses generation and selection) somehow along the lines — aptly revised, of course — of what Popper sought (the logic of trial and error). In a sense, I would like to revive the deep implications of Campbell’s and Popper’s work by proposing that selection theory per se has normative value. Finally, I wish to consider what implications would a revived interest in the logic rather than sociology of scientific validity have on evolutionary epistemology as a naturalized epistemology.