PROFESSIONALIZATION AND THE

ORGANIZATION OF MIDDLE-CLASS

LABOUR IN POSTI NDUSTRIAL

SOCIETY':'

Eliot Freidson

Every generation considers itself to be new and modern. And thisis proper, for the contemporary is by definition what is onlyjust being experienced in the knife-edge of the present. The

problem for the participant and the analyst alike, however, is to deter­mine whether the present is 'really' new and modern, or whether itis, as the elderly are characteristically inclined to believe, just variationon settled and familiar themes. If it is merely the latter, then there isno serious point to changing our way of looking at the present or ofassuming that it is leading us toward some qualitatively differentfuture. But if it is the former, then we must ask what the genuinelynew means for the future-whether it is part of a trend which leads tofar-reaching shifts of emphasis in basic social institutions, or whetherit is an isolated event. And if it appears to be part of a distinct trend,we must, as scholars and scientists, examine the value of our conceptualequipment for analyzing it. Unfortunately, our conceptual equipmentat any particular time is far more likely to be useful for ordering thepast events it was created to deal with than it is for understandinggenuinely new events of the present or the future.

Our own time seems to be one in which something genuinely newis emerging. Many analysts believe that major changes are taking placein the most advanced industrial societies. Those changes are seen tobe as basic and far-reaching as those which occurred two hundredyears ago and which are now summed up as the Industrial Revolution.Various writers have struggled to find an appropriate name for theemergent society, but it seems to me that until future historians, fromtheir more advantageous, backward-looking perspective, create andagree on their own epithet, Bell's term, 'postindustrial', is the most

*1 wish to thank Arlene K. Daniels for comments on an earlier version ofthis paper.

47

Eliot Freidson

useful, avoiding neologism while at the same time connoting departurefrom the society we know. (Bell, 1967.) I shall use it throughout thispaper.

But while many analysts seem agreed that what is new and modemin our time is something with profound consequences for the natureof society, their discussions remain fixed on substance. They are notconcerned about the abstract analytical implications of those changesfor the basic sociological concepts we have used to analyse themechanisms ordering industrial society. These analysts discuss tech­nology, including data-processing, automation and communications,they discuss its foundation in 'technique' and special knowledge, andthey discuss the economy, but they do not consider the social formsby which all will be organized and dispensed. They pay little attentionto social organization, apparently assuming that traditional formswill persist. But if change is to be so radical, should we expect thatour present received concepts of social organization are of suchuniversal pertinence that they can be applied without question? Is itnot quite plausible that habits of thought, strategies, foci and conceptsdeveloped for the analysis of industrial society may not be equallyuseful and well-fitted to the analysis of postindustrial society?

I shall argue in this paper that if what is being said about post­industrial society is true, then very basic sociological concepts for theanalysis of the social organization of work must be reconsidered. Oneof the most basic of such concepts is that of the division of labourand, particularly, the principle of authority which establishes, coordin­ates and controls specialized labour. Since the Industrial Revolution,administrative authority has been emphasized. However, in this paperI wish to suggest that a key to concepts better fitted to the emergentnew society lies in the logic embedded in the concept of professional­ization, which stresses a different principle of authority over labour.To show this, I shall briefly review what most writers agree are theimportant characteristics of the emergent society. Next, I shall pointout what this view implies about the usefulness of what is by nowour almost instinctive reliance on the concepts of administrativeauthority forged for the analysis of industrial society. Then I shalldiscuss how the concept of the authority of institutionalized expertiseimplicit in the idea of professionalization may prove far more usefuland fitting than the traditional concept of rational-legal administrativeauthority.

Professionalization and the Organization of Middle-Class Labour

The Prototypical Postindustrial Worker

Some of the elements of the postindustrial society are predicatedon the assumption that the trends of the past few decades will continueand even accelerate until a basic point of balance is passed. In thissituation, formerly minor segments of the work force will becomecritical. The decline of agricultural labour and the rise of industriallabour created such a change of balance between types of workers inthe development of industrial society. In the present-day, prophets notethe decline of industrial or manufacturing labour as a trend indicatingthat future workers will be engaged primarily in clerical, sales andservice work. (Fuchs, 1966.) Massive as the manufacturing or indus­trial base of the society will remain, its labour requirements will bothdecline and change, as machines tend themselves and other machines,and as the demand for highly trained labour expands while that forunskilled and semiskilled labour contracts. Just as men are still engagedin agricultural work now in advanced industrial society, so will menstill be engaged in manufacturing in the future, but the bulk of menwill be engaged in other kinds of work, and factory labour will ceaseto be archetypal in the emerging age.

If this redistribution occurs, who will be the archetypal worker inpostindustrial society? Most writers emphasize the significance ofworkers practising complex skills for which higher education is thoughtnecessary. (Cf. Bell, 1968; Lane, 1966; Etzioni, 1968.) Virtually allsee the new society as being 'knowledge-based', but some emphasizeone kind of knowledge-based worker while others emphasize another.Those like John Kenneth Galbraith (1968), who are interested instate and corporate manufacturing and commercial enterprise,emphasize the role of the expert who plans and makes decisions, andare prone in that context to use the word 'technocrat'. (Cf. Touraine,1971.) The workers they refer to are engineers, economists, systemsanalysts, and specially trained managers. Others, like Bennis (Bennisand Slater, 1969), Bell (1968) and Lane (1966) are somewhat moregeneral, emphasizing a broad class of professional-technical workers,and the educational and scientific estate (Price, 1965) which trains themand creates their knowledge. Still others, and most especially Halmos(1970), emphasize the role of the professional and semiprofessionalpersonal service worker-the teacher, physician, social worker, nurse,counsellor, and the like who are connected with the social, psycho­logical, medical and other 'helping' services of the welfare state.

49

Eliot Freidson

Each writer on the postindustrial society has a particular issue inmind, and so properly emphasizes one kind of worker over another.While all workers may be alike in possessing higher education, theyare educated in quite different subjects and by quite different methods,they perform quite different kinds of work, and have quite differentkinds of responsibility. Higher education, as such, no matter what thecurriculum and the job prepared for, does not discriminate analyticdifferences in function as basic as that between manager and worker.That difference is the key to understanding how work gets organized.And it is also the key to discerning how postindustrial society maydiffer from industrial society in ways not recognized by those whostress the role of knowledge in the abstract without asking how know­ledge gets organized as work.

What is the traditional difference between manager and worker?In industrial society, managers, administrators, supervisors, or otherofficials of the formal productive organization exercise authority overworkers. They establish the organization, determine the set of tasksnecessary to attain their production goals, employ, train, assign andsupervise men to perform those tasks, and coordinate the interrelationsof the various tasks so as to gain their ends. The worker, on the otherhand, is the one who performs those directly productive tasks whichhave been organized, supervised and coordinated by the manager. Itis this by-now self-evident distinction which becomes problematicin the postindustrial society.

What is of great importance about the forecasts of postindustriallabour is that they imply that large numbers of the prototypical'knowledge-based' tasks are productive rather than managerial.Research scientists, teachers and physicians, for example, character­istically perform a kind of productive labour, even if the labour is notnecessarily manual and oriented to the production of services ratherthan goods. In contrast, other knowledge-based workers performmanagerial functions, setting up frameworks of authority and com­munication to facilitate the performance of labour. Each has agenerically different function, the task of one being to organize thelabour of the other. What is different in the forecasts of the post­industrial society is that they imply that the capacity of managers tocontrol the productive workers is open to serious question in waysthat have not really existed in industrial society.

In industrial society, the prototypical productive worker-the lower

5°

Professionalization and the Organization of Middle-Class Labour

class factory hand-is ordered and controlled ever more systematicallyby management. But the prototypical productive worker of the post­industrial society-the middle class knowledge-based worker-maybe in a position to resist much managerial authority and control. Evennow, the ambiguous position of knowledge-based productive labourin the present conventional industrial scheme of administration ismirrored by its anomalous classification as 'staff' rather than as 'line'.The jobs or organizational positions are dependent on managementfor capital, supportive services and at least some lines of communica­tion, but the tasks of these workers are not. Their tasks are not createdby or dependent on management, nor are the qualifications to performthem so dependent. Finally, evaluation of the performance of thosetasks does not rest solely with management.

In essence, the prototypical worker of industrial society operatesclearly and unambiguously under the authority of management, butthe prototypical workers of postindustrial society may work in aradically different way which limits seriously the traditional authorityof management. Whereas 30 or 40 years ago one could talk of the'revolution' whereby management took authority from the owner,(Burnham, 1960) writers like Galbraith can now write that the'technostructure . . . not management, is the guiding intelligence-thebrain-of the [business] enterprise'. (Galbraith, 1968, p. 82.) Beingthe brain is not the same thing as setting goals and being in charge,but it does make the authority of management problematic. In earlierindustrial enterprises, management was the brain as well as theauthority.

The Limitation of Managerial Rationalization

Apparently, neither the tasks nor the status of these postindustrialworkers seem amenable to the kind of rationalization that was appliedby management to the factory line worker in industrial society. Indeed,it is precisely crisis in managerial rationalization and control (Cf.Berkley, 1971) which prophets of a new society imply when theypoint to the emergence of a growing collection of increasingly strategicproductive workers whose comparatively abstract skills require longtraining and are surrounded by a mystique of the esoteric and thecomplex. Those strategic skills are of such a nature that they resistmanagerial rationalization as manual and clerical skills have not beenable to do. They do so in part because of their intrinsically complex

51

Eliot Preidson

character, but even more because of the occupational organizationwhich grows up around them. Their occupational organization is afunction of professional socialization rather than on-the-job training.Some of the motivation to identification with institutionalized skillsand to solidarity with colleagues stems from the requirement of along period of formal education. Long training is a socially, economic­ally and psychologically costly investment which virtually presupposesexpectation of a stable life-long career, and fairly extensive bondsand common interests shared with others going through the sameprocess. Higher vocational education does not merely insert 'knowledge'into people's heads, but also builds expectations and commitments noteasily overcome by managerial or policy rationalization. Organizedspecialized occupational identities get constructed. Knowledge getsinstitutionalized as expertise. The structure of meanings and commit­ments can override organizational goals or commitments.

The sociological rather than merely technical or economic significanceof a long period of training in putatively complex and abstract skills-that is, of 'knowledge-based' skill-lies in its tendency to developinstitutionalized commitments on the part of those trained. Suchtrained workers are inclined to identify with their skill and with theirfellows with the same training and skill. They are prone to developnot merely general skill-class, or mass solidarity, as is sometimes thecase with industrial workers in trade unions, but disciplinary or occupa­tional solidarity. Their skill is not merely abstractly there as a potential,but it is institutionalized as a stable discipline or occupation. Suchtrained workers do not constitute a class of labour which can betreated as mere hands, to perform whatever tasks management mayinvent for them and then train them for. Rather, they are a kind oflabour with preexistent skills for which management may have a need,but which management must take more or less as given. Their tasksare institutionalized occupationally, and thus resist simplification,fragmentation, mechanization or some other mode of managerialrationalization of labour.

The Limitation of Managerial Authority

If it is true that a new kind of highly educated labour with special­ized skills is becoming more important to the emergent society, andthat the skills involved resist managerial rationalization, what are theconsequences? Even now we can see some of them. The character of

52

Professionalization and the Organization of Middle-Class Labour

managerial or administrative authority which developed in the Indus­trial Revolution is at present experiencing such radical changes thatthe traditional concepts of managerial authority and formal corporatestructure are being questioned. Indeed, an increasing number of writersargue that reality has deviated so far from the concepts of monocratic,'rational-legal' administration, formal organization, hierarchical orderand bureaucracy that new concepts are needed which pick out theanalytically significant elements of the emergent forms of organization.Such new concepts must, of course, perform the same analyticalfunction as traditional ones, the most important of which is to specifythe nature and source of the control and coordination of various kindsof specialized labour so that some productive goal can be reached. Iftraditional managerial Or administrative authority has lost much ofits strength, but specialization and division of labour continue un­abated, what does organize labour and how is it coordinated? Thisquestion has barely been considered by writers on the postindustrialsociety other than on the very broad ahJ vague level of societal orsocial policy goals. An answer to the question on a more concrete levelis hinted at, however, by passing references to professionalization.

As I have already noted, most writers agree on the strategicimportance of workers with skills requiring long formal education,skills rather more esoteric and abstract than the manual and clericalskills organized and even imparted by management in the past. Theproblem is how, if not by management, such special skills get organizedand coordinated. A number of writers have implied the answer. Ellulwrote, 'technique always creates a kind of secret society, a closedfraternity of its practitioners'. (Ellul, 1964, p. 162.) Skill can formthe focus, in short, of an occupational group which then claims theauthority of its institutionalized expertise over the performance of itswork. This tendency is recognized by Galbraith in his observation thatmen of the technostructure are inclined to identify themselves withtheir department or function rather than with the corporation as awhole (Galbraith, 1968, pp. 162-168) and by Bennis in his delineationof 'pseudo-species' as 'bands of specialists held together by the illusionof a unique identity and with a tendency to view other pseudo-specieswith suspicion and mistrust'. (Bennis and Slater, 1968, p. 66.) Thesuggestions all point to the development of solidarity among workerspractising the same specialized skill, an organized solidarity strongenough to resist the pressure toward integration and rationalization

53

Eliot Freidson

exerted by management. The solidarity is of specialty, as discussed byDurkheim (1964), not of skill-class. Such solidarity suggests thatprofessionalization will be a critical element in the organization ofproduction in postindustrial society.

'Profession', 'professionalization', and 'professional' are all extremelyambiguous words, much of their stubborn imprecision hinging on theirconfusing and sometimes incompatible multiple connotations. Butthere is no other word in the English language which can be used torepresent an occupation so well organized that its members canrealistically envisage a career over most of their working years,a career during which they retain a particular occupational identityand continue to practise the same skills no matter what the institutionthey work in. A similar form of organization is to be found in theskilled trade or craft, though the craft cannot claim the same kind ofknowledge-based skill as can the profession.

Such organized occupations can be contrasted with traditionalindustrial labour. As opposed to organized occupations with concreteand particular identity, much industrial labour revolves around formsof work which are rather more appropriately called jobs or positionsthan occupations, since they are merely specially constituted tasks in adivision of labour created, coordinated and controlled by management.Those jobs are created, dissolved and reconstituted by management onthe basis of changing production goals or needs, changing technology,or further rationalization for the end of greater productivity or lowercost. They have no social or economic foundation for their persistencebeyond the plants, agencies or firms in which they exist. The personsperforming concrete tasks constituted as jobs may not perform themfor long, since the job can get eliminated or reconstituted. The workersare identified primarily by the general substance and level of theirskill-e.g., manual or clerical, unskilled or semiskilled or skilled­and if they are organized into such corporate associations as unions, itis more on the basis of industry or skill-class than on the basis of sub­stantive skill and training. Such labour organizations are far moreconcerned with wages, working conditions, job security and advance­ment than with control over tasks constituting jobs.

For industrially organized workers the substance of the work andwho can do it are established and controlled by the manager oradministrator. Tasks are planned around some managerial goal andmen hired and trained to perform them. This is the norm for the way

54

Professionalization and the Organization of Middle-Class Labour

the Industrial Revolution has, historically, rationalized labour by firstdissolving preindustrial functional differentiation by craft and land,and then reorganizing the labour of massed, undifferentiated workersaround managerial authority. But workers organized by professionalor craft associations both now and in the past establish and control thesubstance of their work as well as who can do it. Specialized fragmentsof the labour market split off and become organized and stabilizedaround tasks which the workers have institutionalized. (Cf. Kerr,1954.) Organized occupations like professions and crafts whichinstitutionalize specialization are at once a survival and a revival ofpreindustrial modes of organizing work. They are alternatives toadministrative rationalization.

The key to assessing whether or not the organized occupation willgain even more strength in the future society is the degree to whichall work can be rationalized and therefore reconstituted and controlledby management. It seems to be implicit in discussions of the proto­typical worker of the postindustrial society that knowledge-basedwork, the work of middle-class experts, professionals and technicians,is by its very nature not amenable to the mechanization and rational­ization which industrial production and commerce have undergoneover the past century. If it is true that management cannot rationalizesuch work-whether for technological, economic, political-legal oreven ideological or class reasons-then it can only maintain anadministrative framework around it. Management remains dependentfor attaining its goals on the worker but is unable to really control whoperforms the work and how it is done. Instead of an industrial bureau­cratic structure, with authority organized vertically, management canonly organize supportive services vertically, leaving authority over thework to the workers themselves. Management may set goals, but itmay not set means or connect means to goals.

Coordination of the Division of Labour

But if management cannot exercise much authority over task, howare varied tasks coordinated? How is the division of labour organized?Traditionally, the division of labour has been seen as a purelyfunctional array of interdependent tasks performed by an aggregateof individuals possessing the necessary skills. For Adam Smith, thedivision of labour meant individual men competing with each otherfar employment to perform more or less abstractly conceived special-

55

Eliot Freidson

ized and interdependent tasks. Coordination of all tasks was supposedto take place in part by the natural operation of the forces of the freemarket and in part, albeit implicitly, by entrepreneurs who inventcheaper ways of producing goods. Any explicit social combination ofthe participants in the division of labour was considered unnatural.But in reality, if only out of economic self-interest, both labour andcapital were prone to organize into combinations designed to influencethe labour market in ways that aggregates of competing individualscould not. While Smith's concept of the division of labour ruled outsocial organization, in the historical reality of the Industrial Revolutionthe division of labour has been continuously subject to sociallyorganized forces and has never been a merely technical arrangementof specialized, interdependent tasks. Empirically, we must treat thedivision of labour as a social organization.

Historically, the social organization of the division of labour hasbeen constituted by interaction between two radically different ways oforganizing the human labour necessary to perform interdependenttasks, and of defining the tasks themselves. The common terms'bureaucratization' and 'professionalization' denote, albeit crudely,those two different modes of organization. In the former case, thecharacter of the task, who will perform it, the way it is performed andevaluated, and the way it will be related to others is created bymanagement. The worker is recruited and organized to perform it.The worker and his labour are mere plastic materials for manage­ment, materials organized into jobs by managerial conceptions of thetasks necessary for the production of some good or service for themarket. This mode of organizing the division of labour is implicit inSmith, and it is the mode characteristic of industrial society. Coordina­tion of the division of labour in an industrial organization can easilybe seen as a function of managerial or bureaucratic authority. In thecase of professionalization, however, the task, who is to perform it,and the way it is performed and evaluated is controlled by the menwho actually perform the productive labour. Labour is organizedinto specialized occupations which control their own tasks, and thedivision of labour is constituted as a congeries of such organizedoccupations. In this case, however, the mechanism by which the workof the various occupations are coordinated appears problematic. Howcan any regulation exist between occupations without the monocraticauthority of management?

Professionalization and the Organization of Middle-Class Labour

If, as is characteristically the case for industrial labour, the authorityfor the definition of tasks and their organization into jobs lies inmanagement, then the division of labour can be seen to be constitutedby the jobs which management creates, defines and maintains.Managerial authority coordinates the relationships between the pro­ductive tasks accomplished by every job. Bureaucratization refers to amonocratic ordering of functions into jobs governed by particularrational-legal rules. If, on the other hand, as may characteristicallybe the case for postindustrial society, the authority for definition andorganization of task comes from the prototypical workers themselves,then the division of labour must be seen to be constituted by theoccupations into which work is organized and the relationship betweenoccupations defined and ordered by the jurisdictions established amongoccupations. Those jurisdictions can be established in a variety ofways, the most formal being by exclusive licensing and by contract.

Occupational jurisdictions must be seen as establishing the boun­daries of institutionalized tasks and also, what is often overlooked,the occupational authority to coordinate interrelated tasks. Theyestablish, in short, their own species of hierarchical authority in thedivision of labour, authority predicated on institutionalized expertiserather than on that of bureaucratic office. Such authority gives someoccupations the legitimate right to command the work of otheroccupations. Even now, as a class, the professions provide examplesof how a structure of occupations can be ordered and coordinatedhierarchically by the authority of institutionalized expertise. Medicine,for example, gives orders to a wide variety of other workers in aninterdependent technical enterprise, and does so even when thoseworkers are in the employ of others. In medicine, the division of labouris ordered and coordinated by a dominant profession rather than bymanagement. (Freidson, 1970.) The division of labour thus does notneed management for its coordination; an at least logically possiblealternative to management exists in the form of the occupationalprinciple of authority over work (Freidson, 1973). Prophecies ofpostindustrial society suggest that there is a very real empiricalpossibility that the new division of labour may in fact require a shiftfrom managerial to occupational authority.

The Social Role of Knowledge

In this paper I have argued that if the 'knowledge-based' worker is

57

Eliot Freidson

to be prototypical in the postindustrial society, then concepts of themechanisms by which productive labour is organized, controlled andcoordinated must be examined closely. I suggested that the primemechanism of the Industrial Revolution was administrative or bureau­cratic authority, the strength of which was predicated on its capacityto rationalize tasks into jobs for which it could itself mobilize and trainlabour. 'Knowledge-based' labour, however, may be resistant torationalization both by the very nature of the skill and knowledge itpossesses, and by its tendency to organize itself into stable occupationssimilar to those of present-day professions. Indeed, I suggested that themechanism for organizing, controlling and even coordinating special­ized labour to be found among professions today-the authority ofinstitutionalized expertise-may be far more useful for visualizingthe substance of the Postindustrial Revolution than reliance on now­traditional notions of rational-legal authority and bureaucracy.

Underlying the analysis in this paper is the assumption that know­ledge is a problematic concept with implications which cannot be seenclearly until translated into human activity-into men who arerecruited, trained and then led to engage in the work of producing,communicating or practising knowledge, technique or skill. It isassumed, furthermore, that work is analyzed inadequately if it is seenas an individual activity, that when more than one person does thesame work, the possibility exists that the aggregate of workers willform a group. Should this occur, the purely rational, functional,technical quality thought to inhere in the work itself becomes incor­porated into and transformed by a social, political and economicenterprise. It loses its abstract purity and maybe even its virtue. Thus,whether or not one sees hope for a future in which 'knowledge'becomes the critical guiding force of postindustrial society depends onhow one can visualize the ways in which the use of that knowledge islikely to be organized and controlled.

New York University.

REFERENCES

Bell, Daniel: 'Notes on the Post-Industrial Society', The Public Interest,6 (Winter), pp. 24-45 and 7 (Spring), pp. 102-II8, 1967.

Bell, Daniel: 'The Measurement of Knowledge and Technology', pp. 145-246in E. B. Sheldon and W. E. Moore, Eds., Indicators of Social Change.Russell Sage Foundation, New York, 1968.

58

Professionalization and the Organization of Middle-Class Labour

Bennis, W. G. and Slater, P. E., The Temporary Society, Harper and Row,New York, 1969.

Berkley, G. E.: The Administrative Revolution, Prentice-Hall, EnglewoodCliffs, 1971.

Burnham, James: The Managerial Revolution, Indiana University Press,Bloomington, 1960.

Durkheim, Emile: The Division of Labor in Society, The Free Press, NewYork, 1964.

Ellul, Jacques: The Technological Society, Vintage Books, New York, 1964.

Etzioni, Amitai: The Active Society, The Free Press, New York, 1968.

Freidson, Eliot: Professional Dominance, Aldine-Atherton, Chicago, 1970.

Friedson, Eliot: 'Professions and the Occupational Principle', in E. Friedson(ed.): The Professions and Their Prospects, Sage Publications, BeverlyHills, California, 1973.

Fuchs, V. R.: 'The First Service Economy', The Public Intcrest 2 (Winter),7-17, 1966.

Galbraith, J. K.: The New Industrial State, New American Library, NewYork, 1968.

Halmos, Paul: The Personal Service Society, Schocken Books, New York,1970.

Kerr, Clark: 'The Balkanization of Labor Markets', pp. 92-IIO in E. W.Bakke (ed.): Labor Markcts and Economic Opportunity, John Wiley &Sons, New York, 1954.

Lane, R. E.: 'The Decline of Politics and Ideology in a KnowledgeableSociety', American Sociological Rcview, 31, pp. 649-662,1966.

Price, D. K.: The Scientific Estate, Harvard University Press, Cambridge,1965·

Touraine, A.: The Post-Industrial Society, Random House, New York, 1971.

59