Networks, Communities, Systems and Sociotechnical Constituencies: Analysing Failure in a European Technological Initative

In late 1988, the news broke out that Europe was about to pursue the development of its own microprocessor capability in the form of a 'Europrocessor'. Soon after the Europrocessor' attempt foundered upon a complicated web of conflicting ideas as to which was the most appropriate path to follow. Why did it fail? In answering this question the paper raises other more fundamental issues, for instance, why is it that a large array of different, even competing, institutions tried to come together into a targetted European initiative? and what was it that prevented the conjunction of interests which would have made all the difference? Ultimately, what is the nature of the processes involved in the build up of policy initiatives like the 'Europrocessor'? An important contention of this paper is that the nature and mechanisms of the processes leading to the generation of a large-scale European initiative are not well understood. Conceptual approaches offering fruitful analytical avenues into the problem are badly needed. The paper contributes to such an endeavour through a combination of three aspects: 1) A selective search for some useful insights in the existing literature. Inputs come from political sciences, especially, the brand of 'bureaucratic politics' and 'policy networks/communities'; and from sociology and history, for instance, in the contributions of the 'social history' and 'the social construction of technology', as well as the 'actor-networks' and 'technological systems' approaches. 2) An exploration of the potential of the 'sociotechnical constituencies' approach. In particular, I explore its potential in relation to processes of 'programmatic alignment' (ie. the processes whereby a diverse range of interests succeed or fail to evolve into a coherent programmatic force). 3) The implementation of the 'constituency' approach to the particular case-study of the 'Europrocessor' initiative.