In the first part of this paper, issues concerning an “epistemology of time” are raised. The Derridean theme of the historial movement of a trace is connected to Prigogine's notion of an operator-time. It is suggested that both conceptions can be used to characterize the dynamics of experimental systems in contemporary science. It is argued that such systems have, to speak with Hacking, “a life of their own” and that this is precisely the reason for their inherent unpredictability.
In the second part, a case from the history of virology and cytomorphology is presented. The movement of a particular experimental system is followed in its peculiar tension between deferring and becoming different—i.e., between the persistence of a problem and the vagaries of its successive transformations. They are rendered possible by novel technical implementations, but they are not determined in any meaningful sense on a mere technological level.
In conclusion, a point is made for a deconstructive approach to the history of science.