Which comes first, realization or multiple realization? Hilary Putnam (1960) invoked the term ‘realization’ to refer to the relation that holds between physical devices and abstract computing machines, such as Turing machines or probabilistic automata. Putnam (1967) hypothesized that the relation between brain and mind is also realization. He contrasted his hypothesis—which he dubbed “functionalism”—with the competing hypotheses that mental states are to be identified with syndromes of behavior and behavioral dispositions, or that mental states are to be identified with brain processes. Instead, functionalism proposes that mental states are to be identified with functional states of whole organisms. Importantly, Putnam regarded functionalism as an empirical hypothesis, and one whose explication appeals to some technical notions, particularly to the idea of a probabilistic automaton. From this familiar story of the origins of functionalism, we can extract two lessons. First, ‘realization’ is a technical term. Second, realization can be the relation between physical systems and computing systems. The paradigm case of the mechanical realization of a probabilistic automaton, and the subsequent development of the functionalist hypothesis by extension of the computational example, fix the reference of the technical term.1 Crudely put, hardware realizes...