Role

The margin of autonomy implied by the variance and ambiguity of roles is a generator of system effects, effects that organizational sociologists insist on with vigour. This notion is so important that it is worthwhile illustrating it with a detailed example. The observers of the crisis in the American university system in the 1960s were struck by a surprising finding: the protest against the university system was mainly by students belonging to the best universities. Why? Precisely and mainly because of a system effect resulting from the ‘variance’ in the role of the university professor. This role contains typically two sub-roles: the role of teacher and the role of researcher. The existence of these sub-roles is the result of the dual function of a university: to produce and to disseminate knowledge. This duality of role gives the individuals who perform it a measure of freedom: they are, within certain limits, free to decide how much of each sub-role it suits them to perform. Let us consider the disadvantages and advantages associated with each of these two sub-roles. By its nature, the system of social compensation of a teacher is local. The ‘good’ teacher is appreciated by his students. He is seen in a good light by the management of the organization to which he belongs. But it is unlikely that a teacher’s reputation will extend beyond the walls of his establishment. By its nature, the system of compensation for a researcher is on the contrary, to use Merton’s term, cosmopolitan. The results of a discovery are, in theory, destined to be made available to the whole of the international scientific community. Therefore, by the nature of their roles, the compensation of the teacher is assigned by a local agency, the establishment; that of the researcher is assigned by central agencies. It is necessary to expect therefore that a system that does not accept the separation of the roles of teacher and researcher will make the second sub-role seem much more attractive. Let us now consider the system effect resulting from the variance of roles. The American university system is notable, if we compare it with the French system, for a high degree of mobility. University establishments, being unequal in prestige, are unequally sought after. This has the effect that an individual whose fame is in the ascendant will ‘normally’ seek to move to a more prestigious establishment. For their part, the prestigious institutions will endeavour to maintain and if possible to enhance their prestige by joining the competition to attract candidates with established reputations. But, by the nature of the sub-roles, reputations are rather more readily established on the basis of the quality of research work than on the quality of the teaching. We must make an exception of the liberal arts colleges, where a certain type of tutoring is highly valued and which have the capacity to offer their teachers a certificate of reputation that is negotiable on the wider university market. But in general, it follows from the special attraction of the compensations that go with the

sub-role of researcher that the best universities are also those where the teachers, being more often than not reputable researchers, have the tendency to interpret their role as teacher in the most restricted way possible by seeking to keep to the minimum the time they devote to this sub-role and to spend their energies exclusively in teaching only what is connected with their field of research. We arrive thus at a contradiction which explains the inversion of the correlation between quality and protest: the ‘best’ universities are those with the ‘best’ professors and the ‘best’ students. But these are also the ones whose professors concern themselves the least with the most numerous sector of the student population, the students who are at the beginning of their courses. Numerous, and aware of their quality, since they were admitted only after the most rigorous selection procedures, these students also had the feeling more frequently than students in less prestigious institutions that they were being let down by the teaching staff. This example illustrates in some detail a fundamental case in which we see that role variance can produce system effects with considerable social importance. The analysis of these effects is one of the main objectives of the theory and of the sociology of organizations. The reader who is anxious to go deeper into this point can refer to the works of Deutsch, Crozier, March, and Simon which contain numerous examples of system effect provoked by organizational systems.