Here’s an excerpt from Robert Perry’s 1967 paper, “Innovation and Military Requirements: A Comparative Study.” In it, he finds that one of the inventors of the jet engine, Frank Whittle, had to work through private funding and early success was achieved through austere demonstrations.
American projects like at Lockheed adhered to “efficiency concepts then fashionable in America, so it was a very complicated and difficult development project.” This added cost and complexity meant the company was less willing to take private risk and instead relied on government.
As Perry found, “Neither Lockheed nor Northrop proved willing to make any substantial research and development investment in their own designs without assurance that the War Department would underwrite virtually all expenditures.” He continued:
Both Whittle and Ohain carried their [jet] engines through the preliminary demonstration stages at costs that certainly did not exceed $75,000 each. Lockheed, with what in retrospect seems to have been a reasonably attractive concept, invested $25,000 in evaluative design and then spent two years trying to induce the government to fund a $1 million feasibility demonstration.
Here’s what Perry takes away from it all:
The importance of early and cheap demonstrations of the technical feasibility of novel devices is apparent in a number of other cases — Watson-Watt and the radio-wave reflection detector, the Heinkel-von Braun experiments with liquid-fuel rocket aircraft, California Institute of Technology’s experiments with JATO rockets, even Fermi’s fission demonstration at Stagg Field.
… These examples suggest that, given a concept or invention that is inherently sound, a demonstration of feasibility can be provided at relatively low cost if the necessity of demonstrating feasibility is not subordinated to some other (and probably improper) objective. Only then is it entirely feasible to evaluate the prospect that a valid requirement for the innovation exists — or can be induced — and to proceed accordingly.
I think that’s in line with what people think of today. The S&T account in RDT&E budget activities 6.1 to 6.3 are designed to be rapid and austere demonstrations. Anything after that should be lined up in an acquisition program, once the requirement emerges.
Somehow, the reality of the situation doesn’t seem to work out. Not enough has been done to convince the experts that a better way is possible:
Having particular familiarity with the technical area the invention will affect does not seem to improve prospects of successful prediction, as witness the general inaccuracy of “experts” in estimating the probability that a workable turbojet engine could be developed within a reasonable period of time.
I think this is one of the most important lessons from the history of technology, and should give caution to the overly consensus-based system of acquisition decision-making today that starts with military “requirements.”
Leave a Reply