Science and Public Policy: National Security 99 



ROVER or ORION, thrust in for extra interest.) 



Now the issue is not whether this was a good or efficient 

 administrative structure, but rather to see the mode that was 

 actually exercised for transformation of the best scientific 

 specialities into developmental and systems engineering needs. 

 Some elements of this pluralism are so spectacular that they 

 stand out without deeper analysis. For example, in the field of 

 inertial guidance, the academic engineering studies, especially 

 of Professor Stark Draper at MIT, were so much closer coupled 

 to the necessary computing and other applied sciences in the 

 field than anything which an established industry or a new con- 

 tractor set up for such a purpose could possibly provide in the 

 time allotted, that it was directly agreed that this academic 

 laboratory should become the principal inertial guidance de- 

 velopmental and systems source. The really subtle factor, how- 

 ever, was that the MIT Laboratory for Instrumentation was 

 literally encircled by scientific talent in physics, metallurgy, 

 solid state, astrophysics, etc. In the ONR pattern, this latter 

 capability was not mission-directed but self-directed. Hence, 

 individual elements of the guidance mission could receive 

 exquisitely particular attention, either directly or indirectly 

 through the intellectual environment. The necessary confidence 

 that these delicate and incredibly precise guidance mechanisms 

 could be fitted into all the other elements of a complex system 

 was already dramatic tribute to the basic notion. This was 

 that enough elements of the total missile system would be 

 understood in fundamental physical terms so that inevitable 

 adjustments in guidance, propulsion, materials and warhead 

 would not be blocked by blind dependence on a single rigid 

 empirical engineering plan. Indeed, such resiliency can be 

 introduced only through some minimum level of basic under- 

 standing among the principal elements of a system, and also 

 corresponding availability of a common language for thought 

 and action in the integration and evolution of the system. At 

 least in retrospect, the value of the strategy of invoking non- 

 mission oriented basic research in the early conception of 

 complex new systems is unsurpassed. 



