Augustin Jean Fresnel (1788-1827) was a French engineer and physicist who was a key figure in the move from an “emission” theory of light to a “wave” theory of light in the optical physics of the early-nineteenth century. Where a “ray” of light was generally taken to be a physical, if imperceptible, thing, which could (in theory) be counted, the new wave theory took a ray to be only a geometrical construct connecting a luminous source with a point on a wave front as it traveled through an ethereal medium (ether wave propagation!).
Fresnel was the son of an architect who, having a penchant for mathematics, began training at the new Ecole Polytechnique in Paris at the age of seventeen, where he received extensive instruction in methods of mathematical analysis, chemistry, and physics—an education that gave him both a background in natural philosophical conceptualizations as well as in practical technique.
Eager to make a “discovery” of any sort, he bounced between fields early on. After he left the Ecole in 1806, he worked as an engineer with the elite Corps des Ponts et Chausées (Bridges and Roadworks Corps) for three years, and in 1810 he invented two new processes for manufacturing soda, one of which was already in use, the other deemed too expensive by the experts. In 1814, intrigued by reports on something called the “polarization” of light, he turned to optics.
The pressing problems of optics when Fresnel entered the scene had largely to do with certain refraction phenomena, interference effects, and, indeed, polarization through crystals (the discussions on Wikipedia are pretty good: see, interference and polarization). Attempts to explain these phenomena revolved around the physical interaction between rays, but Fresnel posited explanations in terms of wave effects (for the optics nerds: this was, specifically, in terms of the generation of secondary wave fronts—an idea with links to Christiaan Huygens).
It is not entirely clear where Fresnel got the idea of using wave mechanics—he did not read English, and was not then aware of the content of English experimenter and wave proponent Thomas Young’s work—but his application was thorough, and tightly tied to his own experiments, which were specifically designed to test the efficacy of his theory work. Fresnel does not seem to have viewed the move as especially radical at the time. It only slowly became apparent, and not to everyone, that mathematical wave interpretations implicitly excluded emission interpretations.
Fresnel’s account of diffraction effects would win him a prize awarded by the French Academy of Sciences in 1819, even though the members of the prize board (Laplace, Biot, Poisson, Gay-Lussac, and Fresnel’s ally Arago) had a stake in maintaining the emission theory, and worded the prize announcement presupposing the answer would use an emission theory. As Jed Buchwald argues, because Fresnel’s formulation of the diffraction problem was not lethal to the emission interpretation, the prize board members would have felt they were “conceding comparatively little” in awarding the prize to him (there had been only one other applicant). The problem of polarization was the main stage.
As Buchwald points out, then, it was Fresnel’s application of his methods to polarization that cast the real blow, because it was here that theoretical explanations of specific experimental set-ups would ultimately make wave theories inescapable (if one subscribed at all to the need for algebraic theories).
Emission (or “selectionist”) theories described polarization in terms of the selection of differently oriented rays of light as they passed through a polarizing medium. These theories, for reasons I will not go into, were incompatible for explaining the results of certain experimental setups, which could distinguish between different kinds and degrees of polarization. The point was not immediately obvious, and, while Fresnel’s work was respected, wave theories were still not widely accepted when Fresnel died of tuberculosis in 1827.
Fresnel’s physical methodology was, when he started using it, quite novel. Precision experimentation was readily available and widely used, as was mathematical analysis. However, the careful integration of mathematical theory with precision experimentation so as to place strict demands on the theory was not. Notably, Thomas Young had used mathematical arguments and made precise experiments, but he had not presented one as feeding into the other in his published work.
Fresnel’s mathematical arguments for specific experimental results, particularly on the problem of polarization, made it impossible to work in mathematical optics by any means other than by subscribing to the wave theory, which denied the individuality of rays. When wave optics was taken up by the Cambridge mixed mathematicians and certain members of the new British Association for the Advancement of Science in the 1830s, they created a robust and self-sustaining theory community that was able to make progress on new kinds of problems. Those who did not subscribe to this program, or who (like David Brewster) supposed that it was possible to dispense with theoretical assumptions, were not able to participate in work at the vanguard. By the end of the 1830s, the emission theory had been thoroughly displaced.
Today’s post is based on Buchwald’s 1989 book The Rise of the Wave Theory of Light: Optical Theory and Experiment in the Early Nineteenth Century, which I want to write up for the hibernating “Canonical” series soon. He has some outstanding things to say about unarticulated conceptual disagreement and its impact on the possibility of rational agreement, as well as an unusual and admirable emphasis on the analysis and interpretation of historical scientific work beyond what was fully understood, much less articulated, by the actors. Buchwald needs to be understood, I think, as one of the most methodologically important authors to come out of the 1980s historiographical tradition. In grad school, I always thought of his work as overly fussy—it gets very technical, and is much more detailed and satisfying than the quick rundown here—but his commitment to explaining why technical details can matter in the big picture should be better appreciated.