36 Research and National Purpose 



field and also provided the motivation for its generous sup- 

 port and exploitation. 



Our system of science support in the U. S. is what I like to 

 term a mixed decision system, and I think one of the major 

 sources of our scientific and technological strength. By a mixed 

 system I mean one in which scientific choices are governed by 

 a wide diversity of priorities, institutional environments, and 

 motivations. Not all pure research is too pure, and not all 

 applied research is too applied. In the apt terminology of 

 Alvin Weinberg, our choices are seldom dominated exclusively 

 by either scientific criteria or social criteria, and our variety of 

 research institutions and sponsoring agencies emphasize these 

 two types of criteria in varying mixes. We recognize the fact 

 that the sponsor of research and its performer may quite 

 legitimately have different sets of motivations and priorities, 

 and that the very tension between the two may be highly 

 creative. 



One of the great paradoxes of this age of specialization is 

 that it is harder and harder to delimit the boundaries between 

 scientific fields or the relation between the scientific fields and 

 federal missions. The "modern" technological agencies — De- 

 fense, Space, Atomic Energy — draw scientific sustenance from 

 almost all areas of basic science. Within our present diversity of 

 support systems, how is an agency to decide just what is relevant 

 to its mission? How is it to decide when it can safely depend 

 on the research sponsored by other agencies? When should it 

 depend on in-house capability, and when should it look to the 

 academic community for intellectual support? When is an ac- 

 tivity which it sponsors to be considered above critical size? 

 It would be wrong to expect that these questions could be 

 answered a priori, that is, in the absence of past history. What 

 an agency sponsors will depend not only on its own mix of 

 skills and its appraisal of the skills of other agencies, but also 

 on previous history. One cannot set down a system of rules. 



Why should a mission-oriented agency sponsor or conduct 

 basic research? It needs a fund of knowledge adequate to the 

 fulfillment of its mission at a satisfactory rate of progress, and to 



