In 1861–62, James Clerk Maxwell published “On Physical Lines of Force,” in which he laid out a detailed mechanical model of the ether and argued that it could account not only for electromagnetic phenomena but for light as well. In 1864, he followed with “A Dynamical Theory of the Electromagnetic Field,” in which he derived the electromagnetic equations from general dynamical considerations without invoking any mechanical model of the ether. Why the shift? Did Maxwell regard his mechanical model as mere scaffolding, to be cast aside once it had led him to the proper field equations? Or did he remain committed to the goal of a purely mechanical explanation, but find it useful to free his main results, particularly his electromagnetic theory of light, from dependence on the specifics of an admittedly speculative model? To understand the apparent shift Maxwell’s thinking underwent between 1862 and 1864, I propose that we look closely at what he was doing in 1863. He spent that year working hard for the British Association Committee on Electrical Standards, collaborating with telegraph engineers to establish the value of the ohm and laying the groundwork for measuring the ratio of electrostatic to electromagnetic units, a key quantity in his electromagnetic theory of light. This experience led Maxwell to adopt for a time an engineering approach that focused on establishing relationships between measureable quantities rather than devising hypothetical mechanisms. Maxwell’s electromagnetic work thus had closer ties to the technological context of the day than has generally been recognized.
Skip Nav Destination
Research Article| April 01 2015
Maxwell, Measurement, and the Modes of Electromagnetic Theory
Historical Studies in the Natural Sciences (2015) 45 (2): 303–339.
- Views Icon Views
- Share Icon Share
- Search Site
Bruce J. Hunt; Maxwell, Measurement, and the Modes of Electromagnetic Theory. Historical Studies in the Natural Sciences 1 April 2015; 45 (2): 303–339. doi: https://doi.org/10.1525/hsns.2015.45.2.303
Download citation file: