Professions today: self-criticism and reflections for the future

This article revisits my 1977 work The Rise of Professionalism, outlining what I now believe was wrong in my book and what I think may still be useful in order to move on to some ideas about the present situation. One of the predominant questions is whether we should completely merge the study of professions with that of knowledge-based occupations. And if we accept that the age of experts has entirely supplanted the former public life of trustee professions, what consequences derive? What trends will future research need to consider? These tentative and incomplete ideas should be read as calls for further research.

search for a "new working class", which had moved from Serge Mallet's eminently political preoccupations in 1963 France to later analyses in the United States. Indeed, I started asking myself what was so special about being a professional, if ill-paid architects, who could be laid off from one day to the next, responded to an organizing attempt that it was "unprofessional" to unionize? University teachers faced with a student strike and daily police raids on their campus said the same to the colleagues who also went on strike. My initial plan was to study the architects' failed unionizing effort, and I followed their meetings for some months, before realizing that professions were, in themselves intriguing.
In approaching the abundant sociological literature on these special, and especially esteemed occupations, I found out that most of it seemed to repeat that professions were special: because they fulfilled important social needs, because they were communities, because they knew more than those who used their services, because they were trusted, because they pursued the clients' interest before their own. It was never clear how they got to be that way. We could say that sociology approached professions as objets trouvés, which may explain the overriding sociological concern with medicine, the professional status of which could not be doubted. So, first, I looked to history in trying to deconstruct what I found and tell a different story. I shall have to return later to my dependence on secondary historical works, which induced problems.
Certainly, in looking for a more satisfactory account of how professions came into being I did not think that later David Sciulli would brand me, together with Randy Collins and Andrew Abbott, a "revisionist" that had to be neutralized. Sciulli (2008) held that the revisionist work on the professions was characterized by a conceptual triad: they were expert occupations and nothing more, their monopolies were unnecessary, and the social consequences of their existence were of limited scope.
A detailed response to Sciulli's interpretations and misinterpretations would be out of place and it can be found elsewhere (see Larson, 2008). The narrative I proposed was neither a conceptual triad nor had pretensions to be a universal theory. It was simple, and only focused on England and the United States, two very different societies that nonetheless shared, as Eliot Freidson wrote, "a comparatively passive state apparatus with a strong but by no means unambivalent laissez-faire ideology and a small civil service" (Freidson, 1994: 17).
In these two countries, since the nineteenth century, the occupations that we still call professions had organized in waves of association, suggesting that the organizers responded to a similar context of challenges, opportunities and resources. "Society" did nothing nor did a generic class, like the bourgeoisie. To emphasize that the process by which professions organized was led and followed by real reformers and practitioners, I called it "a project".
For me, professions were situated historically and not invariant forms of organizing work. Looking at the history of reform movements, I saw that their project had the structural effect of translating one order of scarce resources into another. It aimed at establishing structural links between relatively high levels of formalized education and relatively desirable positions and rewards in the social division of labor.
What do I mean by saying that the effect of professionalization was structural and that the agents involved in the project aimed at establishing structural links? I want to underscore, first, that structures are not inevitable: agents construct them; even etymologically the first meaning of structure is construction. Perhaps more importantly, agents sustain these structures inadvertently, by participating in the institutions that express and undergird them. Second, structures, as they last, acquire an objective character that actors can read as obligatory, as "the way things are done" objectively, as a social passage that appears inevitable although it is entered by choice, because one needs or wants to engage in it. Third, one structure understood as "obligatory" passage has structural effects on other structures. Thus, here, advancing professionalization replaced apprenticeship as the main way of entering a practice and corralled the entrants into courses of training that had to become (and gradually became) the inevitable way of entering a protected market of services and labor. Apprenticeship, the privileged path for entering a guild, remained characteristic of important manual crafts; in formalized professional training, it still constituted the complementary clinical or practical dimension; experience always translated formal training into real practices.
Once established, the structural links were inevitable: on the one hand, what we now call credentials -formal, certifiable and certified education under professional control; on the other hand, market positions that could guarantee a respectable living and social status. Credentials and market shelters, for the excesses of unregulated competition were the main incentive for seeking reform. Shelters, of course, involved monopolistic tendencies, for it was necessary to offer inducements to new recruits; otherwise, they would have had no reason to enter the harder and more exclusionary path.
Certified knowledge was necessary in the professionals' self-presentation to the public, but it was not sufficient. Knowledge by itself, be it in Latin or vernacular, be it classic or "modern", abstract (as Abbott, 1988, thinks) or empirical, restricted or created in excess by over training, was never sufficient by itself to establish the superiority of the knowers vis-à-vis their rivals, even those less knowledgeable and less trained. The kind of knowledge mattered, of course, but it was affirmed and applied in markets of services that were structured in different ways and subjected the professional project to different conditions. In cultural and ideological terms, this process could be conceived as a prelude to meritocracy.
Guided by Karl Polanyi more than by Karl Marx, I saw professionals transacting a fictitious commodity in Polanyi's sense: the producers themselves had to be "produced" and their services made recognizable, that is, different from alternative forms of service, and hopefully better. Their acquired capacity was for sale; it was a fictitious commodity because educated labor is embodied in individuals that have not been produced for sale, even if their services were intended for exchange. In a competitive market the "production of producers" branded their services as superior. This branding cannot emerge from the market itself. Ultimately, professional movements must address their claims to the state as institutional guarantor, justifying them by principles that reside outside the market (Polanyi, 1957).
In sum, the path toward a sheltered market involved certifying knowledge and what I called the negotiation of cognitive exclusiveness within certain market conditions; importantly, even in laissez-faire societies this involved persuading the authorities and the public that the services offered were better than the competition. That path took varied ways and a long time. It could be emulated and so it was, introducing differential levels of privilege in the growing service class.
I do not believe that the professionalizing occupations all reached the same end-point, nor do I subscribe to Wilensky's (1964) interesting idea of "the professionalization of everyone" but I think that replicating the structural links between formalized and relatively systemic training and the social division of labor is a goal that was adopted by many different status-seeking occupations. Indeed, a second wave of professionalization was enacted at the end of the nineteenth century and in the twentieth by occupations born in the shade of the mostly large and mostly bureaucratic public and private organizations that employed them. I called them techno-bureaucratic professions: social workers, planners, managers and administrators of different kinds, librarians and archivists, specialties spawned by the large hospitals and by other state agencies. Closer to engineering than to medicine, they showed the way of the future for practically all expert occupations: not independent but employed, still enjoying (or claiming) technical control, still rooting their work identity in a field of knowledge, still seeking to advance this field or to keep up with its progress… but were they still professions? That is a question we have not quite resolved.

The faults and the omissions
The greatest flaw of The Rise of Professionalism is its abstraction and generality. They come in large part from my dependence on the secondary sources that were available, not primary materials. Certainly, the book did not focus on what professionals do in what Andrew Abbott was going to call, much later, the profession's work core, that which jurisdiction defends above all against intruders or usurpers. The goal was to tell a convincing story of what professions are and how they came about. At the time, there was not too much of exceptional quality in either history or sociology, except for the works on medicine by Robert Merton and his students, and the studies inspired by Everett Hughes in Chicago -Howard Becker and his co-authors with Boys in White and Eliot Freidson's Profession of Medicine (Freidson, 1974). The language that I use in parts often bears the Marxist mark of the time. For this, I ask indulgence, although I would still say that professionals tend to consider the use-value of their work as important as its exchange-value, and I would still rely on Louis Althusser's (1971) work on ideology.
Second, because of severe constraints of time and resources this work is exclusively focused on England and the United States. Since the collective effort initiated by Rolf Torstendhal at SCASSS (Swedish Collegium of Advanced Studies in the Social Sciences) in 1987, many students of professions accept the premise of two models (at least!): one "Anglo-American" led by reform movements started in the civil society, and one "Continental", closer to a civil service model, often prompted by positive initiatives of the state. A bimodal process obviously means that the search for a single, invariant ideal-type of profession has been archived. As Konrad Jarausch was to write later, "When comparativists tried to apply professionalization terminology, it soon became evident that categories derived from an Anglo-American model hardly fit the bureaucratic German pattern. Only recently have some sociologists come to realize that there might be a continental variant of professionalization with different dynamics from the British original" (Jarausch, 1990: 9). Matthew Ramsey's important comparative work on medical monopoly showed how the protection of a strong state could grant social privileges and power to a profession before it had demonstrated any cognitive or technical superiority. Monopoly, he argued, depended on essentially political conceptions of what role the state should play (Ramsey, 1984;Larson, 1990: 27-32).
The state was important for all European societies, and especially so in Germany. As Jarausch notes, the enlightened absolutist state wanted to upgrade its administrative effectiveness through training, especially in the law but also in secondary teaching. The state established and administered various levels of strict examinations, which were therefore not controlled by the professionals themselves. Nevertheless, gaining and proving expertise through academic achievement emancipated various professions from too powerful patrons. That expertise, protected by the professional corporation (the German Kammer), later became the anchor of their status within the large organizations for which Central European professionals typically worked.
Third, my work focused on the historical context within which modern professions were formed. The conditions for professional organizing were ripening toward the end of the eighteenth century (although they would only develop more fully in the second half of the nineteenth). I emphasized in this work what my teacher Arthur Stinchcombe called changes in "social technology". following his thesis that new organizational forms tend to appear at the time when it is precisely possible to found them. Basic societal variables have a positive effect both on the motivation to form new organizations and on the latter's stability; they are "literacy, urbanization, money economy, political evolution, and previously existing organizational density" (Stinchcombe, 1965: 155). To the economic and technical factors, I added cultural and ideological conditions, as a determinant of the motivation to organize. They seemed particularly relevant in the case of organizations which, like the professions, aim at increasing "the amount of trustworthiness among strangers" (Stinchcombe, 1965: 160).
The professional project was part of these incomplete but awesome transformations: various reform projects that led to the emergence of new types of professional societies appeared in roughly the same period (55 years in Britain, 47 in the United States). I undoubtedly overstated in this analysis the discontinuities between the pre-industrial past and the market society; a more attentive observation of history would often have dispelled them. However, even the profession of law, which was the first to disengage itself in fourteenth century Europe from the tutelage of the church, did not develop until the nineteenth the stable and intimate connection with training and examinations (or "objectively" verified competence) that came to be a stamp of profession. Also, the ideology of merit spread slowly after the French Revolution and was not sustained by the parallel opening of access to education. As Michael Schudson (2006) remarked, the idea of professionalism could not have been an effective ideological support for the social order before the 1950s, when higher education credentials became the hallmark of "getting ahead". Only then could the ideology of meritocracy become the widely-shared credo of elites, no matter how much the absence of equality of opportunity belied its premises.
Fourth, one of my most serious sociological mistakes echoed the same pattern of taking professions as objets trouvés that I was critiquing. I overstated a consistency in the professional project that could only be discerned ex post facto.Itw as wrong to assume that a profession existed as a unified actor at the onset and too often, it appears in my book that professional unity was found, rather than having to be produced. In her compelling study of English medicine on its road to the Medical Reform Act of 1858, Elizabeth Popp Berman (2006) shows that the modern profession itself had to be created for professionalization to advance. It was an arduous job. I emphasized like she did that the emergence of hospitals provided doctors with an identity-forming organizational base, and that this was one important advantage medicine had in its road to professional recognition; but I did not go into the real workings of any field "before the project". This induced many mistakes.
Fifth, in part because I did not deal with primary sources, I glaringly ignored that the classic professions were mostly male and mostly white. This is a grievous omission, especially when it was in many cases clear that the feminization of the personnel in a field stood in the way of professionalization. In Germany, for instance, when primary school teachers tried to follow their state-sponsored secondary level colleagues and upgrade their own status through training, gender was an unsurmountable obstacle. In the early twentieth century, German social workers, also predominantly women, faced a somewhat different problem: their source of recruitment made it easy to marginalize these mostly upper-middle and upper-class women and to thwart their professionalizing efforts within the male-dominated Weimar social welfare bureaucracy. Gender was undeniably a factor in the professional narrative of success or failure. 1 The demographic composition of knowledge-based occupations (more than just professions) has changed and continues changing. The changes raise important research questions not only about the continued forms of discrimination and exclusion -which are challenged because they are still present -but also about the forms of inclusion. In emphasizing the ways in which women and minorities are included, Celia Davies (1996) noted the affinities between bureaucracy and profession and their common relation to the ideal of masculinity, embodied and promoted by both. Capitalist rationalization, Davies argued, was male, and still is masculine. We still must study and clarify the effects that changes in personnel may have on the deployment of expert work, whether diversification will change professional practices, or the distribution of rewards, or whether it will mostly modify the collective social advantages that the profession enjoys. I did not do anything to bring up, much less advance, this preeminent research topic.
Finally, two omissions: I did not deal with the independent dynamics of state and university, the key institutions addressed by the professional project. Undoubtedly, their history, their internal politics and their external political effects are independent of the professions and particularly important to the latter, especially in what regards the state. Certainly, even in the laissez-faire ideology and comparatively less interventionist states of Britain and the United States, the political climate mattered very much in the reception that professional reform movements would find. And certainly also, the openness or, on the contrary, the inhospitality of the university -dependent on culture, ideology, politics and, of course, funding -made a difference in the professionalizing efforts of newer, not yet "learned" occupations. Any empirical study of a concrete professionalization movement would have to insert it, not only in what Andrew Abbott later called "a system of professions" but also, and perhaps more significantly, in a structured stage where state and university were the key players. We go back, therefore, to the fault of excessive abstraction and generality with which I started this section.
Second omission: it was implicit that I saw the professions as special communities of discourse, endowed with the authority of speaking on and for their field and, in so doing, constructing its meaning for the lay public. But it was too implicit. As I started working on architecture, I placed this discursive capacity at the center of my study.
Ordinary practitioners apply the discourse within which they have been formed and act within the unquestionable doxa of their specific field -the unstated agreement about what questions to ask and not to ask that Pierre Bourdieu (1981) posited for scientific disciplines. They know, as professionals, what they can challenge and how to debate, the procedures that rule the accumulation of symbolic capital in their field. The power-effect of these specific cultural capitals is pervasive, not difficult to discern but easy to resent and perhaps as lasting as the effect of economic capital.

What is still useful in The Rise of Professionalism
First, I would defend the emphasis on the structural link I emphasized between certified knowledge and relatively uniform or standardized training, and positions in the social division of labor. The forms in which this link is manifested vary, but knowledge-based occupations (however we name them) seek and affirm this link. Florent Champy has noted that, in most countries, it is the same "old" professions that are in some way protected by law from unqualified competitors; more questionably, he adds that market closure shelters new occupations that perform activities in the national interest. I do not subscribe to this notion, which may be more current in France than in other countries, but I agree, indeed, that status advantages are not distributed randomly (Champy, 2009). I would argue that one of the tasks of the sociology of professions today must be to ask what happens when the essential structural link is severed, or at least weakened by the overproduction of producers and the maladjustment in the supply and demand of high-level diplomas. The problem is undoubtedly affecting the academic profession and I shall return to it in what follows.
Second, I think that my comparative analysis of the resources available to medicine and engineering offered some important insights. It aimed at trimming down the paramount importance that sociologists accorded to medicine, so often treated as the archetypal profession, the queen among objets trouvés.
Medicine did not start out being favored in what I called the "negotiation of cognitive exclusiveness", on the contrary. Fierce battles for preeminence in the market of services were fought between allopathic and homeopathic physicians, and they were not resolved by demonstrable cognitive superiority of the eventual victors. However, medicine worked in a market so structured, that it would allow it to attain a well-protected shelter (if not a totally effective monopoly) when it had some effective technological superiority to show, after having finally accepted the discoveries of bacteriology.
Engineering, on the other hand, operated in a rapidly changing technological environment, where demand far exceeded supply and where the devisers of innovation would logically enjoy cognitive exclusiveness. But this market was inherently subordinated to large and powerful employers, hindering professional efforts.
I based on this comparison my structural approach to market control, identifying some key resources in the negotiation of cognitive exclusiveness and the establishment of market shelters. The argument is that variation in these resources inflects the path to success of particular occupations.
I cannot detail the conditions of variation here, but key resources include: the nature of the service that is marketed; the type of market; the type of clientele; the cognitive basis; the formalization and codification of training, or "the production of producers"; the power relations within which the professional market is placed; the profession's affinity with the dominant ideology (Larson, 2012: 47-49). From my perspective, medicine entered its professionalization project with structural resources that engineering did not replicate, despite being just as able to claim a scientific knowledge base (arguably, the claim was more justified in the nineteenth century for engineering than for medicine). Medicine's market was structured in ways that definitely eluded engineering: for a long time, medicine had unorganized individual clients, a unified organizational base in the hospital, a comprehensible object, which mattered to everybody regardless of social class and affected the governance of the state; perhaps above all, it had a potentially limitless market for its services.
That medicine did not reach its position of dominance because of its knowledge base becomes even clearer after World War II: not medical research, but new sciences like epidemiology, virology and molecular biology led the way, while the most notable medical advances, "like the artificial kidney, penicillin or the polio vaccine of were feats of research and engineering". They "raised such broad social issues that they moved the question of who controlled science outside of the control of the physician and the medical profession" (Wailoo, 2004: 650).
Of course, I did not intend to say that sociologists of professions should forget about medicine and take engineering for their only model. In the United States, the entrenchment of medicine in hospitals, which were large, technologically advanced and bureaucratically administered, may have eroded some aspects of dominance for beginning or lower level doctors, but it did not reduce it for the profession as a whole. Nurses have not become an alternative to doctors, despite the upgrading of their functions and increasing specialization. Doctors, especially in teaching hospitals, have steadily blurred the line between themselves as practicing professionals and the managers of health systems. In front of a surplus of physicians, they have been able to emphasize a non-existent shortage of specialists. In England, however, as in most countries with a national health system, the power situation of medicine has changed dramatically since 1948, as has the situation of allied health professions. They have all moved closer to techno-bureaucratic professions.
Third, I gave to the latter -the techno-bureaucratic professions -the importance they deserve. This led me to discard the hard distinction between profession and bureaucracy, as almost opposite forms of authority and of control of work, in the current situation of work of most professionals. In Professionalism. The Third Logic, Eliot Freidson (2001) tried to rebuild this opposition. But rather than emphasizing contrast or opposition, it may be more fruitful to consider the rise of different forms of professionalism within organizations that move toward a post-bureaucratic mode. Paul Adler and his associates have keenly traced the emergence of new cooperative communities among knowledge-based occupations, not only in "post-bureaucratic" settings, but also in less coercive, more enabling bureaucracies (Adler, Kwon and Heckscher, 2008).
Fourth, I think I was right to insist on the Weberian notion of "calling", or beruf. In my view, emphasis on the intrinsic value of work distinguishes professionalism and the professional's identity more than a hypothetical disinterest. The use-value of work is cardinal in my discussion of "anti-market principles" in the professional project. This value came in part from pre-industrial transfers incorporated in the ethical base of the modern conception of profession: disinterest, as an echo of noblesse oblige, averse to purely commercial pursuits and insistent on the duties as well as the rights of high rank. The market arguably upholds the notion that everyone has the right to be served (which is not incompatible with the idea of "universal service"), but the classic professions also affirmed, against the market and in preservation of pre-industrial notions of community responsibility, that all have needs, and that needs must be met. As T. H. Marshall (1965) emphasized, the advancement of the professions' functions stood for the expansion of social needs. Incidentally, the intrinsic interest and significance of one's work seems to be part of what makes professional work a target of populist anger, a symptom of "elitism" in the battles around expertise that we see brewing today.
In sum, the search for status, trust and autonomy through the certification of superior knowledge was a lasting strategy. Other occupational groups could follow it, even some which might have already achieved as much market closure as they could expect. However, in the new century, the strategy did not necessarily involve the pre-industrial transfers that went into the notion of calling: it could be a mere strategy. Privileges could be sought without postulating an ethical base. This argument, however, leads us directly to the dilution of profession into the broader and even less defined category of experts. Many years ago, Brint (1996) had announced the ethical problems posed by the dilution of what he called "trustee professionalism", the kind with a somewhat communal mooring, and the dominion of "unmoored" expertise, often anchored in governmental organizations.

Studying professions today
Professionalism today is contradictory. Three crucial sources of contradiction or erosion are demographic, cognitive and organizational, as I will try to illustrate in part with examples from architecture and teaching, which are in some respects almost opposite occupations. But I want to begin with some general effects of the dilution of profession into expertise, centered around questions of trust and secondarily of identity.
Trust in established professions is eroded but public esteem is still present; in the United States, the public trusts the military the most (78%), followed not too close by teachers (72%), medical doctors (66%), scientists (65%) and engineers (63%). The word "professional", the idea of something "professionally done", connote competence and dedication and perhaps a hint of immunity, for professionals should not be questioned by lay people. Lay people have no choice: they must trust the competence of experts, mechanics and electricians as much as that of lawyers and engineers.
Yet, we are undeniably in a situation of wide mistrust of experts. Politicians, mainly of the right, are quick to use it, like Michael Gove, the UK minister of justice, in the Brexit referendum: "Britain has had enough of experts" he declared, and incongruously compared the pro-EU economists to the Nazis damning Einstein's theories. After the first six months of the Trump administration, it is clear that the 45 th president takes for expertise what he wants, especially if it comes from friendly news sources like the Fox channel; but his stance only maximizes a pre-existent problem.
In part, the distrust of experts may be related to the expansion of government's reach and supervision. Thus, in 1992, Terence Johnson argued that "the extension of the capacity to govern depended on expertise in its professionalized form and the development of expert jurisdictions and systems of notation, documentation, evaluation, calculation and assessment". (op. cit. Evetts, 2006: 524). Experts may be implicated in procedures that citizens fear, or which they think burdensome, especially if they do not know exactly what they are, or which experts do what. I believe that trust is related to understanding what experts do and knowing what they represent in the social division of labor; the less is known about the scope and effects of their interventions, and the more suspect they are. The multiplication of expert specialties incomprehensible in name and function, relatively autonomous and free from public account, can only augment mistrust. The absence of possible dialogue with experts does the rest.
If one argument implicitly refers to the power of experts and to its lack of transparency, in America Tom Nichols argues that expertise is dead, or at least seriously threatened. The "omnicompetence of the common man" (which, for Richard Hofstadter, was fundamental in the "original populistic dream") clashes with the overwhelming complexity of the modern world. Feeling at the mercy of sophisticated knowledge elites, the common man responds with Internet and Google-fueled rejection of anything but purely technical advice. From many professionals, Nichols hears "stories not about patients or clients or students raising informed questions, but about them telling the professionals why their professional advice was actually misguided or even wrong. The idea that the expert was giving considered, experienced advice worth taking seriously was simply dismissed" (Nichols, 2017).
In the age of Trump, it is tempting to grant at least part of Nichols's thesis about "the death of expertise". The historian Beverly Gage explains that anti-elite sentiment in America can be seen as "a repeating cycle of… revolt", but the "up versus down" antagonism has been appropriated by the conservative movement some fifty years ago: "That meant redefining the term away from class and toward culture, where the 'elite' could be identified by its liberal ideas, coastal real estate and highbrow consumer preferences… Trump has also ventured beyond mere name-calling, turning the 2016 election into a competition between knowledge systems: the tell-it-like-it-is 'people' versus the know-it-all 'elites'. His campaign insisted for months that pollsters and technocrats and media would be proven wrong by his electoral success. The fact that he did win dealt a blow to an entire worldview, one in which empirical inquiry and truth-telling were supposed to triumph in the end. The question, now, is whether it's possible to run an executive branch based on hostility toward experts and professionals of all political stripes" (Gage, 2017).
A second point is in order here: the common man's feelings of helplessness may be what sustains mistrust or ignorant dismissal, but they are in turn sustained by widespread antagonism toward the cosmopolitan and multicultural cities and by deep resentment of real and growing inequality (Cramer, 2016). Authority-denying mistrust is fueled by uneven access to knowledge capital (despite the Internet illusion) and the uneven distribution of the rewards of expertise. Symbolically, experts are visible in cities as "yuppies" -young urban professionals -a wide and heterogeneous status group courted by municipal authorities, especially after Richard Florida hailed the regenerative potential of this "creative class".
The new specialties, although they cannot easily claim public esteem, receive high salaries, especially in management, finance and information technology. Within the established professional groups, young lawyers, doctors, social workers, journalists, engineers, although likely to be paid less than in finance or IT, may also be seen as part of the "young professional" strata; also, their ranks risk to be swollen by the overproduction of credentials. Yet, despite the demographic risk of excess supply, the professional-managerial census category has been much better able to weather the Great Recession than other categories of workers, adding to the "common man's" resentment.
I have argued that a profession's expertise confronts a "lay public… constituted by the combined effect of a 'personalized' experience of [free and compulsory] education and a distracted and impersonal relation to the media" (Larson, 1990: 37) and I was wrong, almost 30 years ago, about the relative effect of each, with all but "impersonal" media proving much more engaging than the education most people receive. This lay public also inhabits discursive fields that we can identify by topic or purpose. There, in our societies, lay people should "listen" to credentialed experts, who always tend to occupy the core regions of discourse production. Politicized mistrust means that non-experts increasingly want to talk back. So, in sum, experts still have the authority to speak, but we must increasingly ask what happens if it is not recognized. Anti-vaccine movements are but one early and still-contained example of possible battlefields.
It is a fact that many experts only speak to very specialized clients, while others may be brought to defend their discursive authority in front of "lay publics". This may well be an important differentiation among expert categories. Now, internal differentiation makes it difficult for us to continue treating the category "profession" as corresponding to any kind of community. Especially in France, the sociology of professions had emphasized the normative importance of shared professional training and work experiences, as generators of professional identity. Identity, supported by a common institutional base, was like a socio-psychological mainstay of the professional order, a complement of branding and a support for trust. The public had an idea of what professionals should be like. But the great variety of work situations and careers and the possible transition away from lifelong work pose the problem of what kind of identity expert workers can develop. We used to believe that you could defrock a priest, suspend a physician or disbar a lawyer without erasing what they were, or what they appeared to be. We must ask if that is still conceivable, let alone if it ever was true: what happens to the notion of "calling" and to any community of knowledge, when marriage to one's work is expected to turn into so many amiable divorces, into the self-promotion of careers that move across different forms of employment, into always unstable connections with colleagues and supervisors? Is shared knowledge enough to form communities of practice?
From the perspective of our discipline, the dilution of professions into expertise blurs boundaries between the sociology of professions and the sociology of work, at one level, and the sociology of knowledge, at another level. The blurring of boundaries leads to reflexive work, work that asks itself what the study of professions has been, is and should be, like this essay, or the recent book edited by Andreas Liljegren and Mike Saks, Professions and Metaphors. For Liljegren, two key spatial metaphors inform the field: in the metaphor of hierarchy, subordination is the central dimension, while the landscape metaphor, or its equivalent, the map, highlights boundaries that are both social and symbolic and emphasizes the importance of boundary work (Liljegren, 2017).
Of course, the two metaphors are not mutually exclusive, not only because one is vertical and the other horizontal but also because which one we choose depends on the profession we are considering and the work setting (or the sector of the social division of labor) where we place our object. For instance, architects do not appear to be a dominant profession and never did historically; their main collective task as an organized profession seems to be patrolling and defending boundaries against encroaching professions, like engineering or interior design. Yet, in many countries, construction documents must be signed by a licensed professional: architects are either prominent (therefore superordinate to the other trades and occupations involved in construction) in this category or compete with other licensed trades, which can be minor professions like surveyors in Italy.
The landscape metaphor -especially considered as a map -is indispensable in understanding what I have called the construction of a lay public. In what "landscape" (certainly not a system) of occupations does a broadly conceived lay public situate, if it can do so at all, both clearly-established professions and new groups, as they claim professional status and guard entry into their ranks? Indeed, it is hard to believe that the public has a conception of what specialists in diagnostic imaging or big data development do, or what they are, but then most people would be hard-pressed to say what lawyers or physicians actually do, although they presumably know what they are. I think it is important that we follow how the absence or presence of a cognitive map is related to the construction of a lay public, and to trust. Or, in other terms, that we show how a sociology of knowledge relates to the study of occupations and the sociology of work.
I tend to agree that it "no longer seems important to draw a hard-definition line between professions and other (expert) occupations" (Evetts, 2006: 519). Yet some questions may well be more important than ever, especially for occupations that deal with risk and uncertainty by institutional arrangement: what are experts for, what are they expert in, who should the public trust and why, and most emphatically, to whom are experts accountable? These questions are relevant for all expert occupations and it remains to be determined empirically whether the well-established professions have better ways, and better tools for answering them than newer, still little-known specialties. Here, the denomination of "profession" and the positive connotations of professionalism may be serving a function different than what we expected historically.
Many sociologists have reappraised professionalism as an effective form of control of knowledge work, although they do not intend to revive the old assumption of a conflict between profession and bureaucracy. As knowledge-based organizations seek to move away from ineffective forms of bureaucratic control, adapted forms of professionalism help in the management of knowledge workers, who must nonetheless retain full autonomy in their work. I am no longer willing to claim that professionalism can spread as a sort of false consciousness among subordinate workers. However, in the 1990s, researchers like Valerie Fournier (1999) and Gerard Hanlon (1998) considered professionalism a "technology of the self", which facilitates control in organizations with leaner budgets, growing caseloads or demanding clients. Beside control, since advanced capitalism increasingly depends on its knowledge workers, trust is a cheap and effective means to coordinate knowledge-intensive organizations: in support of this thesis, Paul Adler detects the positive emergence of knowledge communities in new forms of post-bureaucratic management. They may arise from the bottom-up, and be facilitated from the top.
Negative or positive, "professionalism from above" brings up the question of where expertise resides, in individual producers, or in the very organizations where they work, since the latter are indispensable to make it possible. Today, a large and increasing number of knowledge-workers are employed in heteronomous organizations, governed by professional managers, like school principals and superintendents, or hospital administrators, or personnel managers of various kinds, and not by colleagues like themselves. How they govern and monitor their knowledge and their performances varies from occupation to occupation, and from work setting to work setting. What power the practitioners derive from the collectivity of profession remains to be empirically determined in each case. In recent work, I have begun to consider school teachers, knowledge workers who enjoy high public esteem, yet have often failed in their efforts to gain autonomy and respect in the knowledge-based organizations where they work. In their case, at least in the United States, professionalization and status gains depend much more clearly on the political power of their unions than on credentials (Larson, 2014). A very different case is the profession of architecture, in which I have long been interested (Larson, 1993). In this weak profession, the elite's discursive authority, supported by schools and critics, maintains an illusion of coherence in the public image of architecture. For the diverse mass of practitioners, the discourse is at best a focus of distracted interest. For the public, discursive authority creates a public professional identity. The architectural "aesthetes" may seek to control the profession's discursive authority but they have been forced to extend its paradigms and recognize the silent power of what is built, which does not depend on the profession's authority.

Conclusion
In conclusion, the dilution of professions into a general category of expert occupations, many of which are poorly defined and poorly understood, has contributed to create resistances. Resistance may acquire dangerous political expressions even as it illuminates the lack of dialogue, the imposition of standards, the deafness to ordinary concerns of which many experts fail to take consciousness. We are far today from the time when Eastern European intellectuals wondered about the intellectuals' road to class power, and far from a political project of the new class. Yet, for us sociologists of professions, bringing to bear on our oft-embattled subjects not only the light of the sociologies of work, knowledge and culture, but also the pressing questions of political sociology and contemporary politics, our work continues. In many ways, it has just begun.