Published online by Cambridge University Press: 01 January 2022
In this article I explore some statistical difficulties confronting going conceptions of ‘group’ as understood in accounts of group selection. Most such theories require real groups but define the reality of groups in ways that make it impossible to test for their reality. There are alternatives, but they either require or invite a nominalism about groups that many theorists abjure.
My thanks to Wes Anderson, John Basl, Clark Glymour, and Brian McLoone, who read early drafts of this article; to M. Maria Glymour and Michael Higgins, with whom I discussed statistics; and to four anonymous referees. They have together much improved this article.