Abstract

Biology, in contrast to other historical disciplines such as cosmology or geology, is not explicitly articulated with physics. More specifically, its unifying principle, evolution by natural selection, is currently not formulated in physical terms. This hinders any attempt to explore whether this principle may apply to other physical systems, beyond life as we know it, or to understand the origin of life in a physico-chemical framework. To better understand whether an explicit articulation is achievable, we first aim to clarify, on the basis of examples, how principles are articulated within the physical sciences, or between the physical sciences and other scientific fields. This leads us to establish a typology where we emphasize that physical principles involve both “rules” in the form of mathematical relationships between concepts, and “premises”, defining the conditions and objects to which they apply; articulations may take place at these two levels. We then ask whether the principle of evolution by natural selection may fit in such a typology of articulations. We contend that addressing this question is made difficult by an apparent but ineffective distinction between rule and premises in current accounts of the principle of natural selection. These reduce evolution by natural selection to the iteration of a constant rule, thus failing to recognize that biological evolution is a process that recursively modifies its own modes of operation, e.g., through changes in inheritance systems or levels of individuality. While this may be ignored when focusing on paradigmatic cases of natural selection (as formalized by population genetics, where connections with physics are recognized), it becomes a patent problem in more general formulations of natural selection. We conclude by discussing whether this problem could be resolved, through a formal and general description of this principle, where rules and premises would be truly independent or, alternatively, whether its heuristic value, within biology or beyond, is just of a different nature than that of physical principles.