The collective known by the pseudonym Nicolas Bourbaki has been a major representative and promoter of mathematical structuralism in France, above all through its Éléments de mathématique, a long-term publishing venture covering large areas of pure mathematics and now widely recognised internationally. This history of mathematics workshop, associated to the ANR-funded research projet “Bourbaki, new archives, new approaches” (BANANA), aims to study the rich archives of the collective, which, especially for the period 1953-1969, remain largely unpublished and unexplored. We intend to contribute substantially to the literature on Bourbaki (which has developed rapidly in recent years, in history as well as philosophy of mathematics) in a methodologically novel way, combining socio-historical methods with genetic criticism. The latter approach, well established in studies of literature or philosophy but still little used in the history of science, consists in studying the genesis of a text by examining its preparatory drafts; it is especially well suited to the study of the Bourbaki archives, which contain hundreds of « rédactions », that is, successive reports or drafts—intended for collective discussion by the group—relative to some book or chapter of the Éléments. We shall develop case studies at different scales (genesis of a book, of a chapter, or even of a single theorem or definition), examining how Bourbaki’s collective writing practices relate to the genesis of the text and of the concepts involved in it. This will also allow us to tackle epistemological questions, for instance relative to the definition and classification of mathematical structures and to the role of such epistemic values as generality, purity of method, rigour, and unification.