In chapter 17 of his book, Rewriting the Soul: Multiple Personality and the Sciences of Memory, Ian Hacking makes the disquieting claim that “perhaps we should best think of past human actions as being to a certain extent indeterminate.”<xref ref-type="fn" rid="FN1">1</xref> Against what may appear like the self-evident conception of the past as fixed and unalterable, Hacking suggests that when it comes to human conduct and experience, there are reasons to adopt a more flexible view. This suggestion has caused lively debate, in the journal History of the Human Sciences and elsewhere.<xref ref-type="fn" rid="FN2">2</xref> Central to this debate is the question of what it means to use a recently invented vocabulary to redescribe past human affairs. In particular, it is asked: How do the linguistic, cultural and social differences between past and present matter to the possibility of such a redescription’s being true? We who do research in the humanities and social sciences often make retroactive redescriptions of precisely this sort. Hence, the debate is clearly of some general importance for how to conceive the goals and methods of our inquiries. My overall aim in this paper is to clarify what we may learn from the clash between Hacking and his critics.