• “All the forms of assimilation, of Monte Carlos to experimentation, that I have presented so far (stability, error tracking, variance reduction, replicability, and so on) have been fundamentally epistemic, That is, they are all means and practices by which the researchers can argue toward the validity and robustness of their conclusions. Now I want to turn in a different direction, towardwhat amounted to a metaphysical case for the validity of Monte Carlos as a form of natural philosophical inquiry. The argument, as it was presented by a variety of people (including occasionally Ulam himself), was based on a purportedly fundamental affinity between the Monte Carlo and the statistical underpinnings of the world itself. In other words, because both Monte Carlo and nature were stochastic, the method could offer a glimpse of reality previously hidden to the analytically minded mathematician. As Ulam himself once put it, his and von Neumann’s hunt for the Monte Carlo had been a quest for a homomorphic image of a physical problem-where the particles would be represented by fictitious ‘particles’ in computation” (143–44)
    • Compare with the Meillassouxian strand of speculative realism, wherein mathematics (and by extension computation) are sufficiently “pure” so as to “touch the real.” The simulation or model becomes Borgesian: it tracks perfectly to reality, or at least provides us a mechanism by which we can ascertain a capital-T Truth inherent to reality as much as the simulation.
    • Galison even later calls the the simulacrum interpretation of Monte Carlo later on the same page.
    • “King’s view-that the Monte Chlo method corresponded to nature (got ‘back of the physics of the problem’) as no deterministic differential equation ever could—I will call stochasticism” (146).
  • Even in 1951, the idea of computer-as-actor or computer-as-autonomous-organism proliferated: “Two years later [King] amplified on these comments, arguing that the computer should ‘not be considered as a glorified slide rule’ but as an ‘organism’ that could treat a problem in an entirely new way” (144).
  • “To the Platonist, the stochasticist has merely developed another approximative method. To the stochasticist, the Platonist has interposed an unnecessary conceptual entity (the equation) between our understanding and nature-stochasticism, he or she claims, offers a direct gaze into the face of nature” (148).
  • The seductiveness of cyberspace as an alternative to coping with the harsh edges of the everyday is, and was, apparent to those who work with simulations. Years after working on simulations of theirst H-bomb, a physicist (then a young postdoc) told me: ‘I had a strange attitude toward the reality of hardware and the reality of explosions, which it’s hard for me to explain now. But [it] was intense and real at the time. I didn’t want to see the actual hardware of an atomic bomb in the laboratory . . . in the machine shop, in the metallurgical facility. And I didn’t want to see a nuclear explosion.’ T~he alternative world of simulation, even in its earliest days, held enough structure on its own to captivate its practitioners. And in their fascination they learned a new way of work and a new set of skills that marked them for a long time to come” (153–54).
  • A quote from a contemporary scientist, with fascinating parallels to the work of divination as speculative practice: “It seems to me that while it is true that cavemen have already used divination and the Roman priests have tried to prophesy the future from the interiors of birds, there was not anything in literature about solving differential and integral equations by means of suitable stochastic processes” (154). The Monte Carlo, as technique, is then “the culmination of a profound shift in theoretical culture, from the empyrean European mathematicism that treasured the differential equation above all, to a pragmatic empiricized mathematics in which sampling had the final word.”
  • “In this strong sense, the early Monte Carlo applications to the diffusion,of gases, the scattering of neutrons, the production of cosmic-ray showers simulated nature. On this view (the view of many early practitioners of the Monte Carlo), the Monte Carlo offered a resemblance relation between sign and signifier in a pre-sixteenth-century semiotic sense. As the Monte Carlo became a standard tool for the resolution of problems with no stochastic elements, King’s position became harder to defend, and the vision of simulations as offering a uniquely privileged vantage point began to dissolve. Nonetheless, the sense of direct access to a problem ‘as nature poses it,’’ ‘behind’ the equations never quite left the players of Monte Carlo” (155).
  • “If any theoretical representations stood in Plato’s heaven, they were the delicate hypersurfaces of differential equations-not batch-generated random numbers and the endless shuffle of magnetic tape. For experimenters, the Monte Carlo never came to occupy a position of ‘true’ experimentation, as exemplified in debates that continued decades later over the legitimacy of according doctorates to students who had ‘only’ simulated experiments. This left the engineers and their successors, the computer programmers, in a peculiar position. They spoke an intermediate language, a kind of formalized creole: the language of computer simulations understood both by theorists and by experimenters. (It was no accident that conferences flourished with names such as ‘Computing as a Language of Physics.)Accordingly, the simulators became indispensable as links between high theory and the gritty details of beam physics and particle collisions. But just as they occupied an essential role in this delocalized trading zone, they also found themselves marginalized at both the experimental and the theoretical end of particle physics” (155).
  • Looked at from a microscopic point of view, an atomic bomb does proceed in this way: disconnected fissions, scatterings; a Markovian universe plummeting into detonation. But from outside, we see a mission-oriented laboratory, a group of scientists allied with a military infrastructure struggling to create particular weapons and the intellectual superstructure to facilitate their design and implementation. Have we witnessed the cascade of the participants’ narrative of history into the narrative of the physicists’ simulations?” (156).