The open texture of functions: a framework for analyzing functional concepts in molecular biology

In recent times, the exponential growth of sequenced genomes and structural knowledge of proteins, as well as the development of computational tools and controlled vocabularies to deal with this growth, has fueled a demand for conceptual clarification regarding the concept of function in molecular biology. In this article, we will attempt to develop an account of function fit to deal with the conceptual/philosophical problems in that domain, but which can be extended to other areas of biology. To provide this account, we will argue for three theses: (1) some authors have confused metatheoretical issues (about the meaning and application criteria of terms) with metaphysical ones (about teleology); this led them to (2) look for explicit definitions of "function", in terms of necessary and sufficient criteria of application, in order to make the concept of function eliminable; however, (3) if one leaves metaphysical worries aside and focuses on functional attribution practices, it is more adequate to say that the concept of function has an open texture. That is, that a multiplicity of application criteria is available, none of which is sufficient nor necessary to attribute a function to a trait, and which only in concert form a clear picture. We distinguish this thesis from some usual forms of pluralism. Finally, we will illustrate this account with a historical reconstruction of the ascription of a water transport function to aquaporins.