Bytes as Test Tubes: Computational Predictions and a Methodological Chemical Revolution Available to Purchase

We trace the emergence of computational methods in quantum chemistry, from their introduction in the late 1950s until their acceptance and use as a primary exploratory tool in the 1980s. We argue for the following theses: First, to strengthen the independence of computational models from experiments, quantum chemists used rhetorical strategies that tended to overstate the relation of their models with theory and understate the way they were guided by experiment. Second, alliances with organic chemists were crucial for disseminating computations. Third, pictorial representations facilitated the understanding of abstract theoretical entities, such as orbitals. Fourth, in the 1970s, evaluation strategies for the performance of models became a fundamental part of computational modeling in quantum chemistry. Fifth, the epistemological and institutional status of the new methods—called computer simulations—were uncertain and became the objects of negotiations and sometimes controversies between computational chemists and traditional bench chemists. Finally, the new methods were accepted as complementary to conventional experimental methods by the early 1980s.