War and Science: The Century-Long Transformation of Caltech's Physics Discipline (1891-2026)

Abstract

Caltech's physics discipline offers an unusually complete record of what happens to a science when its patron changes. Between 1891 and 2026 it passed through four funding regimes: private endowment, wartime contract, federal sponsorship, and the mixed order loosening since 1991. Each transition left traces in one archive. Reading that record requires two literatures that seldom meet. Big Science theory accounts for the scale of research; the Triple Helix accounts for who pays for that scale, and so who names the problems. Their model describes a peacetime arrangement. The Caltech record shows the wartime version that preceded it. Once the NDRC and then the OSRD began buying research in 1940-41, the industry pole did not disappear. It was absorbed into military demand. Caltech took more than eighty million dollars in federal contracts, second only to MIT, and about ninety-five percent of its war work went to rockets and guided missiles. What grew at the junctions—NOTS at China Lake, JPL, RAND—belonged to no single sphere, and their permanence made the substitution stick. Two features of that arrangement matter later. The arrangement was elastic, since patrons could be exchanged while the university stayed put: JPL passed from the Army to NASA in 1958, Owens Valley from the Navy to the NSF in 1970. It was also unequal, since costs fell first on the exposed: Tsien Hsue-shen, stripped of his clearance in 1955, and Margaret Burbidge, barred from the dome under her own name. After 1991 the security premise expired, and these institutions had to argue for themselves before publics, before Native Hawaiian communities at Mauna Kea, and before a Congress that refused the 47 percent cut proposed for NASA science. The closing section asks what the record implies for data access and siting, where the case is for defaults that can be reopened rather than settled once.