ample, in the Leningrad Optical-Mechanical Associa- 

 tion the general management of the PO fully replaces 

 the plant managements. At the Svetlana Association 

 each enterprise retains a measure of autonomy, and 

 only the basic management functions are centralized. 

 The management of the PO is the same as the manage- 

 ment of the largest and most modern of the plants in 

 the association, i.e., the "head" organization of the 

 PO. The other plants are organized as branches of 

 this leading plant. At the Elektrosila Association 

 some of the plants are fully merged with the PO, 

 whereas others have retained some autonomy. A sim- 

 ilar pattern of structural diversity also character- 

 izes the science-production associations. According 

 to a model organizational statute on the NPO, issued 

 by the central leadership at the end of 1975, how- 

 ever, all units joining the association lose their 

 independence. Nonetheless, practice continues to 

 diverge from this uniform pattern. Figure 9-13 il- 

 lustrates the model management structure for a sci- 

 ence-production association. 



THE ORGANIZATIONAL SYSTEM: 

 WHOLE AND PARTS 



The previous discussion of the organization of R&D 

 in the USSR suggests certain features and themes that 

 deserve emphasis. Most basic, of course, is the for- 

 mal design of the whole edifice for science and tech- 

 nology as an "organizational system" of multiple and 

 well integrated parts, with elaborate but generally 

 internally consistent assignments and responsibil- 

 ities. This image of a highly centralized and coor- 

 dinated Soviet system that is able to pursue compre- 

 hensive and coherent S&T policies often prevails 

 abroad. 



The image, however, conceals as much as it re- 

 veals. Though highly centralized, the organizational 

 structure of Soviet science and technology is far 

 from monolithic. On the contrary, it is highly frag- 

 mented. An official at the top feels sometimes, in 



76 



