This afternoon, thanks to the initiative of Jim Grozier, I am giving a talk at the weekly High Energy Physics seminar at UCL. The subject will be my work on experimentation in early particle physics. While my “Strategies of Detection” paper mainly concerns the problem of how to build “mesoscopic” histories of experimental practices, my talk will repurpose my argument to discuss how we can articulate and evaluate experimental ingenuity and skill. This jibes with other thoughts I’ve had about whether it could ever be considered legitimate for a professional historian to write a celebratory narrative of scientific progress. The very notion triggers the raising of well-disciplined eyebrows: isn’t it the job of professional historians to problematize celebratory narratives? But, really, I can’t think of a good reason why not, and it seems to me there is substantial opportunity to improve the genre.
The celebratory genre, I think we can argue, remains in an undeveloped state, constrained by the domineering influence of citation practices. The objective of citation is to identify where credit is to be allotted, or where more information can be obtained, about a particular claim in a scientific paper. Typically, a citation is concerned with the claim itself. How the claim was developed may be discussed, but mainly if the claim is considered dubious or in need of qualification. Prizes, likewise, are usually given for achieving a particular result, or making a particular discovery. Whether the result or discovery was largely the product of serendipity, or great experimental skill and ingenuity may be taken into account. However, discussion is likely to be limited to a description of an experimental setup, or use terms such as “elegant”, devoid of contextual discussion of what would constitute a typical, mediocre experimental setup. I have seen few inquiries into what is considered to constitute “skillfulness” or “ingenuity” or “elegance,” and how that changes with time in various experimental cultures.
By defining what these sorts of terms mean, it may be possible to recognize, and more easily teach, what makes an experimenter—or theorist, for that matter—a good experimenter, theorist, etc., over the course of their body of work, even if they never made a historic discovery.
To a certain extent, we can get away with talking about the precision of instrumentation, but a very good experimenter, it seems to me, will develop a good sense of understanding what sorts of things need to be measured to make useful contributions to a body of knowledge, what sorts of things can be measured to a useful degree of precision, and then being aware of and being able to marshal various available experimental resources to those ends. There are, no doubt, other facets of experimental skill, but this encapsulates the facet I try to capture in my paper.
As is well established, when Carl Anderson discovered the positron, he was actually attempting to measure the energy spectrum of the secondary cosmic radiation. I argue that his methods for doing so were imprecise, but nevertheless represented a pathbreaking use of inferential reasoning. Within this work, Anderson’s discovery of the positron was, effectively, serendipitous. He was not looking for new particles, as, indeed, the inferential methods he used in making his energy measurements precluded discovery. However, his use of those methods also made it possible to identify particles that simply could not be interpreted as a known particle. The identification of the particle as a positive equivalent of the electron was by no means established (and was, in fact, subjected to subsequent experiments), but it was, after that point, difficult to deny that what was sometimes generically called the “hitherto unknown” had been seen. But, as I also argue, it was Patrick Blackett and Giuseppe Occhialini’s ability to aggregate evidence that made it clear that positrons were not simply those particles that could not otherwise be explained away as protons or electrons, that they were, in fact, very prevalent.
I would also point to Cecil Powell’s experimental and interpretive technique as particularly skillful. Aside from his work in developing the technology of the nuclear emulsion (in cooperation with industrial manufacturers of photographic film), his understanding of nuclear physics and the nuclear interactions that could take place in his emulsions, allowed him to make a deeply elaborate argument that the “primary” and “secondary” mesons he identified in his emulsions differed in mass. This was an extraordinarily important accomplishment because it was able to bring confidence to an interpretive matter that was at that time fraught with specious claims to the detection of particles of irregular mass. Once Robert Marshak and Hans Bethe’s work on Marshak’s “two meson” theory established that tracks with no evident nuclear disintegration were, indeed, very likely spontaneous decays, the Powell group was well prepared to make new discoveries by observing particles’ “decay modes”. Meanwhile, the imprecise counter experiments of Alikhanian and Alikhanov might well have been accepted as evidence of the “hitherto unknown,” but that method of experimentation would prove inadequate in an age of more rapid discovery.
While I would hesitate to presume that historians and philosophers can make major contributions to present-day experimental practices by helping to articulate experimenters’ strategies, I do believe clarifying and articulating the strategies experimenters used in the past could encourage scientists to speak more precisely and frequently about how experimental technique has evolved. This would, I think, make the persuasive power of experimental results more widely understood, both within and beyond the scientific community. It might also encourage the awarding of prizes for scientific skill, rather than just the results those skills produce.