De‐Essentializing the Knowledge Intensive Firm: Reflections on Sceptical Research Going Against the Mainstream

This paper provides an updated summary and discussion of my 1993 JMS paper ‘Organization as rhetoric: knowledge‐intensive firms and the struggle with rhetoric’, and relate this to my own and others' later work in the area explored in the paper: knowledge‐intensive work and discourse. A key aspect of knowledge work is the ambiguity of what it stands for, what people working with ‘knowledge’ are doing, and what they accomplish. This fuels identity uncertainties, which is also addressed and reflected upon in the paper. As part of my broader reflections, I also provide a brief overview of some methodological ideas on how to do problematization: a key theme in the original piece but which is more generally a research approach that may lead to contributions that generate interest.


INTRODUCTION
The Alvesson (1993) paper offered a reinterpretation of knowledge-intensive firms (KIFs) and knowledge work. I argued for a move away from a focus on the essential qualities and the functional value of these organizations and their knowledge to instead an acknowledgement of the uncertainties involved and the significance of persuasive claims about this knowledge and its contributions. Later in my research I expanded this line of inquiry to also include self-persuasions (identity work) as a significant theme and concern for these firms. This is in particular salient in the professional service sector, as a way of compensating for the lack of any clear product and difficulties in demonstrating a specific result or output of the work. This paper summarizes, extends and reflects upon my subsequent work in this area. The paper starts with an account of my move into, and continuance of, research on KIFs. I then summarize my 1993 paper and its main points about ambiguity, followed by a section where I extend some ideas on how many KIFs can be seen as being in the 'institutionalized myths' business and how this, as emphasized by Green and Li (2011), motivates a focus on rhetoric. A key aspect is the 'lack' of substance and the notorious ambiguity of knowledge work. This easily undermines straightforward identity constructions and reinforces the pressure on people to do intensive identity work, which the paper then targets. A brief empirical case of a management consultancy project illustrates my approach and provides some empirical backup. The paper ends with a brief section on a particular methodology of problematization -an approach used in the 1993 paper and subsequently further developed.

BECOMING AND BEING A KIF RESEARCHER
Over the years I have taken an interest in a large number of theoretical and methodological issues, but have to some extent retained a focus in my own and co-workers' empirical studies on what may loosely be described as KIFs. That this became the case was partly by accident. During the corporate culture boom in the mid 1980s I started to work with anthropologically oriented views in order to develop a cultural framework. This kind of approach, for me, means that you try to take not too much for granted but look at organizational phenomena as exotic, arbitrary, and strange, rather than rational, functional, and self-evident. The natives' talk about knowledge', 'leadership', etc. then calls for an exploration of meanings ('what do the natives think they are up to?'), rather than being taken for granted as referring to some self-evident quality or practice. In a critical and reflexive spirit 'natives' here may also include academics: popular interests and broad truth claims often motivate a degree of doubt and critical challenge. Having worked with interpretive and sceptical ideas, including Critical Management Studies, I decided it was time to do an empirical study. Given my more theoretical and critical interests, the idea of extensive empirical work did not appeal to me; life is short and there are so many empirical studies that do not seem to lead to much in the way of interesting results. At the same time cultural studies call for some depth and richness, so I picked a firm that appeared to be open-minded and accessible, in the form of an IT consultancy firm which had appeared on television when its rather spectacular corporate building was inaugurated. The firm seemed to like to exhibit itself and I quickly negotiated access for a kind of 'semi-ethnography' ('it is good that the firm is written about', the Vice President told me). The study was more interesting and productive than I had anticipated and led to a fairly thick book, covering a wide set of aspects and themes from a cultural-symbolic perspective (Alvesson, 1995). I framed the study to be about the management of knowledge-intensive firms; this led, to my surprise, to 1000 copies being sold the first year in Sweden -indeed a rare accomplishment for this kind of book (the case study firm itself bought 100 copies.) I sometimes wonder how many actually read it. There was a booming interest in Sweden in KIFs during the late 1980s, partly overlapping and preceding the general interest in knowledge. As a follow-up to this study, I thought I should start a larger research project, working with PhD students and others, and I applied for and received a grant for studying the category of 'knowledge-intensive firms'. I thought this was a rather good focus: it involved an interesting type of firm and of people (not dissimilar to us academics), and also more generally an appropriate category in terms of combining breadth and variation, with some shared characteristics that would allow for comparisons and synergies between various studies. We have since, within our expanding research group in Lund, studied management consultancy firms, newspapers, high-tech companies, pharmaceuticals, and advertising agencies. Research is based on ambitious case studies (interviews and repeat interviews with people at all levels in the firm, observations of significant and revealing events, as well as general 'hanging around', trying to sniff out everyday work -although observing people sitting in front of computers is not always so informative).
In my 1993 piece I drew partly on experiences based on my early empirical studies of mainly the IT consultancy firm and an advertising agency (Alvesson, 1994;Alvesson and Köping, 1993). Empirical projects were used less as a basis for reporting findings than as one of many sources of inspiration for the paper. Equally important were, first, an interest in the ambiguity of most complex phenomena in organizations, partly encouraged by a move in organizational culture studies from consensus and integration towards ambiguity and fragmentation thinking (e.g. Martin and Meyerson, 1988), triggered or at least reinforced by the increasing popularity of post-structuralism from the late 1980s. Also the general experience of doing close-up studies is that organizations are much more ambiguous than 'mainstream', pattern-seeking research indicates. In addition, I also felt that it was important to acknowledge and incorporate an awakening attention to the significance of persuasion and language use, partly related to constructionist ideas and the linguistic turn in social science (and to some extent overlapping with but hardly the same as the interest in post-structuralism and its emphasis on the indecidabilities of representations). Furthermore, the 1993 paper was also led by an interest in images as an increasingly significant subject for management action, partly associated with a cultural interest in meaning, but also with a weakening of the role of material, substantive practices and results in organizations and a stronger emphasis on efforts to manage images about reality, rather than the latter per se. Finally, the work was also guided by an ambition to sceptically (ironically more so than critically) explore new phenomena that attract a positive interest and an inclination of academics to follow and reinforce fashions and reproduce dominant 'truths' and ideologies. Behind this we find some interest in ideological critique as in Critical Management Studies, but perhaps even more generally it pointed to my interest in all kinds of scepticism towards authors and fashions that otherwise paint a harmonious view of social developments and orders. Here a critically framed version of institutional theory that points to mainstreaming mechanisms plays an important role. (I also have an interest in less mild forms of critical research, expressed in a range of work on Critical Management Studies, but this is only characterizing some of my work, and is not so salient in my studies on knowledge and KIFs.) These four theoretical 'streams' of interest, in combination with a strong ambition to do open, ambitious studies, where empirical material can inspire a rethinking of frameworks and ideas, have continued to characterize my work. I will reconnect to these themes after the following summary and commentary on my 1993 paper.

THE 1993 PAPER
The idea of the knowledge-intensive organization (knowledge-based company) rapidly attained interest during the early 1990s (Blackler et al., 1992;Starbuck, 1992). An indication of this attention and perhaps ongoing interest is the selection of my 1993 and Starbuck's 1992 papers as 'classics' amongst the JMS papers published over the years. Starbuck (1992) coined the term knowledge-intensive firm and the acronym KIF is often used. The focus on a particular kind of organization overlapped with a more general interest in knowledge and knowledge management, with links also to themes such as core competence and capability attracting considerable attention (Kärreman, 2010). In the 'knowledge movement' there is a 'shared conviction that the management of knowledge of whatever kind has become a critical issue for competitive advantage, international strategy, the building of resources, the boundaries of firms and many other issues' (Foss, 2007, p. 31).
From a functional point of view, areas of knowledge develop as a response to problems and possibilities. A knowledge-intensive organization is thus a firm that can produce exceptionally good results through the help of outstanding expertise. Starbuck (1992) suggests, for example, that 'to make the KIF a useful category, one has to require that exceptional expertise make important contributions'. Ekstedt (1990, p. 21) claims that 'knowledge companies . . . supply the large corporations and the various public agencies and other institutions with the knowledge they need to solve problems of various kinds'. He believes that the 'knowledge factor becomes increasingly important . . . and will gradually take over the influential position previously enjoyed by real capital' (p. 21). Similarly, Grant (1996) claims that 'the critical input in production and primary source of value is knowledge ' (p. 112).
This booming interest can be seen as reflecting actual changes in society, work, and organizations which supposedly means that 'knowledge' becomes more important, that the number and significance of 'knowledge-intensive' organizations increases, and/or that 'knowledge-intensity' in organizations and work in general increases in the modern 'knowledge economy'. Such a thesis is less self-evident than it may appear at first glance. This is in particular the case if one does not reserve the word 'knowledge' only for formal, theoretically based versions of knowledge as a base for competence and applications but defines it more broadly to also include knowledge for craftsmanship and other skills. It is not self evident that running a farm two hundred years ago called for less knowledge and skills than being a software programmer today. That a limited number of high-tech, IT, and life science firms are of increasing significance for the economy does not necessarily motivate the thesis that we on a more general level need to be much more intelligent, creative, and knowledgeable in contemporary Western society, nor that we need to multiply the higher education sector and raise the norm for the number of years in schools before people can secure access to non-routine jobs (and to some extent any kind of job). Even if there is a broad consensus about the significance of knowledge, this does not mean that one needs to join the masses and view this framing as the necessarily best one to make sense of contemporary business. Perhaps it is less the level or intensity than the nature and forms of knowledge that have changed.
'Objective' changes do not stand in an one-to-one relationship with words, concepts, and proposed images, and it is not self-evident that we need a concept or category of the 'knowledge-intensive' in management and organization theory, at least not as a way of mirroring the 'nature' of the contemporary organizations. As with any effort to describe something non-trivial, the representation 'knowledge' or 'knowledge-intensive' is a construction and it is far from unproblematic. As with most popular constructions these labels are persuasive, even seductive. The strong symbolic value of knowledge easily creates biases in talk about it, which motivates an extra dose of scepticism to accounts of it. Perhaps people like to use the term -in mass media, academia, and business practice -because it sounds good and appealing and can boost status and self-esteem as well as lead to support from others if one is in, or an active supporter of, the knowledge industry. A celebratory tone needs to be balanced against more sceptical voices, so as to resist the temptation to be seduced by ideologically appealing constructions of a society and its institutions. As with any new and rapidly fashionable area of academic or general interest, one can take a distanced view and ask: What goes on here? Are there reasons to ask critical questions about something that makes sense and 'everybody' seem to agree upon? There is, as Schreyögg and Geiger (2007, p. 177) note, a 'striking discrepancy between the great importance nowadays attributed to knowledge (knowledge economy, knowledge resources, knowledge societies, knowledge-intensive firms) on the one hand and the vague and blurring conceptualizations of knowledge on the other hand' (see also Scarbrough and Burrell, 1996). This discrepancy -that has remained over the yearscalls for our suspicion.
Rather than assuming that knowledge is a reliable functional resource, being the primary ingredient of what KIFs do and that this knowledge work leads to good results, one could instead point at the ambiguities involved, i.e. uncertainties around knowledge, its role for KIFs, and what is being accomplished. Although one could argue that rhetoric is in general a key aspect for the accomplishment of organizations and institutions (Green and Li, 2011), my point is that is especially significant in 'high-ambiguity' contexts like most KIFs, as the degree to which practices, products, and services speak 'for themselves' is very low.

The Ambiguity of Knowledge
One strong reason for critical questions is the uncertainties around what people actually mean by knowledge. Many authors acknowledge that knowledge is very difficult to define, but nevertheless treat it as a robust and substantial capacity which can bring about 'good results'. The way out for a lot of people claiming the significance of knowledge, but then being unable to define or delimit it, is to run to dividing 'it' up in various forms. Four-fielders (2 ¥ 2 box models) in particular are popular: one mixes explicit and tacit as well as individual and organizational and then four forms of knowledge are identified. Whether this clarifies matter is questionable; there are, for example, few knowledge areas where advanced knowledge simply can be made fully explicit and stored, and on the other hand not much knowledge is so tacit that it totally escapes verbalization (Alvesson, 2004). Irrespective of the value of the four-fielders, the meaning of knowledge is still not revealed or clarified.
Sometimes researchers interested in knowledge acknowledge that knowledge is hard to define and correspondence (truth) ideas are neglected in favour of a more pragmatic, process-oriented view on knowledge, emphasizing not what is true but what works.
Knowledge is sometimes claimed to be about 'the processes of resolving the uncertainties of our condition' (Spender, 1998, p. 240). A similar approach also views knowledge in terms of its productive effects: Although we may not be able to judge the knowledge itself, we can certainly see the results of the knowledge . . . In any operation, different individuals using the exact same machinery may produce very different outputs, just as skiers or tennis players vary in performance using the same equipment. (Leonard and Sensiper, 1998, p. 126) Given a less all-embracing and tautological view on knowledge, one may allow space also for elements other than 'knowledge' making a difference, e.g. motivation, physical training, and muscles. A key factor behind sport performance is presumably age: a coach is typically more 'knowledgeable' than the athlete being trained, but the former can normally not compete with the athlete, as ageing weakens physical ability.
At stake is what works. Managers, professionals, firms, and others may believe that a specific form of knowledge works for them, but if these beliefs can be validated somehow one could be able so say something about what works beyond someone being reassured about what they do and about their own competence.
Against uncomplicated and reified understandings of knowledge, one can point to the uncertainties and controversies characterizing a lot of science (Brante, 1988), as well as to the fact that whatever the relative degree of rationality characterizing science and formal knowledge, people in their functioning are much less rational (Fores et al., 1991). These two uncertainties make the impact of the 'knowledge-factor' or esoteric expertise much less clear-cut in practice. A third is Foucault's (1980) point that knowledge can not be separated from power and knowledge; instead it orders and produces rather than mirrors or gives us pure and innocent knowledge of the world 'out there'.
This does not mean that all forms of knowledge are productively seen as uncertain and far from unproblematic, but when we are addressing knowledge-intensive firms we are normally interested not in those natural laws or other forms of knowledge that we all agree upon and that can be treated as valid (such as the earth is round, smoking increases the risk of cancer, Elvis is dead), but in more specialized, sometimes esoteric knowledge, often difficult to assess. What architecture firms, pharmaceutical companies, management consultancies, law firms, advertising agencies, and similar organizations do and produce is seldom clear-cut in terms of how to assess the knowledge that is involved.
However, not only is a great deal of the knowledge drawn upon in KIFs in itself ambiguous, but also what role this 'factor' plays in most KIFs.

Ambiguity of What KIFs Do
One could of course argue that even though it may be difficult to specify the knowledge involved and demonstrate its value, engagement in knowledgeable action is still the core of what people in KIFs do. But this is not self-evident. Often formal and other forms of distinct knowledge seem to play a quite limited role in their work. A study of psychologists and architects indicated that these had problems in finding examples of theoretical knowledge and other higher education outcomes directly being used in their work (Svensson, 1990).
In a case study of an IT consultancy firm it was found that managers and employees often downplayed the role of technical expertise in their work (Alvesson, 1995). The work tasks varied a lot and often people were assigned to jobs for which they had very little formal education or relevant experience. Of course, this capacity to adapt to various contexts and tasks is an important part of consultants' skills, but it is somewhat different from the application of a specialized set of knowledge. Being flexible and willing to learn something new are key qualities not well captured by the signifier knowledge. A management consultant I interviewed had a PhD in a specific area, but two years after he started in a large consultancy firm, he had still not worked within this area, but been allocated to projects where he had no specific expertise. Apparently consultants were often plugged in where there was a need for more manpower. In many cases, consultancy can be seen as adding qualified extra manpower when there is a change process, less because unique expertise is needed than because a focused, hard-working, and compliant workforce provides an additional resource. While some authors view the knowledge economy as characterized by a replacement of capital and labour with knowledge as a key production factor (e.g. Ekstedt, 1990), much consultancy may be about the outsourcing of labour -using the market rather than employment for managerial and administrative labour -as much as signalling the need for advanced expertise.
To say that the work of professionals and other knowledge-workers is only or even mainly the direct or 'creative' application of a systematic, institutionalized body of formal knowledge or esoteric expertise may be idealized and misleading. Knowledge -in the sense of using an advanced framework, backed up by science or other institutionalized, validated forms of knowledge -is thus not necessarily so significant in ostensibly 'knowledge-intensive' work.

The Ambiguities of the Work Results
A third vital aspect of a lot of knowledge-intensive work concerns the ambiguities of the results produced. Persistent uncertainty is per definition a part of the area in which most professionals and other knowledge workers are operating. In an ideal world, if there is a problem or a state of suboptimal performance, then knowledge is put into action and the problem is solved or a better state of affairs is accomplished. In many areas of knowledgeintensive work, like advertising, strategy consultancy, recruitment, and leadership training, this is difficult to assess. Also, in many technical situations, it is difficult to assess the outcomes of knowledge work. In a study of the Eurotunnel, consulting engineers described themselves as saving the project from chaos while many of the permanent staff saw them as of little help, trying to insert costly arrangements and doing things only to justify their high fees (Henriksson, 1999). Actually, in management consultancy, for example, it is very rare, even with an effort, to assess and learn from the results (Ernst and Kieser, 2003).
To sum up, it can be argued that significant for KIFs are the ambiguities characterizing (a) their claimed core product (knowledge), (b) what they are doing (working with 'knowledge' compared to behaving in ways that are loosely connected to this quality), and (c) the results of their work, and its -mythical -meaning. Knowledge-intensive organizations are thus 'ambiguity-intensive', i.e. clarity and order are not the best words for providing accounts of the work and contributions of KIFs. Saying this is not contradictory to acknowledging that ambiguity exists in all organizational life, and KIFs do not have a monopoly on this label. Of course there are some KIFs and some activities in these where ambiguity is less of an issue. Focusing on ambiguity may, however, be particularly fruitful in the study of this type of organization.

KIFs IN AN INSTITUTIONALIZED AND IMAGE-SENSITIVE CONTEXT
Institutionalized assumptions, expectations, recognitions, reputation, and images matter strongly for how the products of KIFs are perceived. Generally, we live in an increasingly image-sensitive society and economy, where branding plays an increasing role (e.g. Biehl-Missal, 2011;Kjaergaard et al., 2011;Klein, 2000). Mass-media representations, and orchestrated efforts to manage these, matter greatly for assessments and choices of groups lacking rich information and experiences of these firms. Job seekers, clients, and consumers are guided by more or less loose impressions. Substantive qualities -activities, accomplishments -may be disconnected from or standing in a loose relationship to images (Alvesson, 1990, forthcoming). Many KIFs score high on image-sensitivity.
Given these ambiguities, the success of a KIF is more contingent upon more or less loose beliefs about them being able to offer something specific to clients. Instead of an objectivistic and functionalist understanding of knowledge and knowledge-intensity, a social constructivist or institutionalist position can be explored (Meyer and Rowan, 1977;Scott, 1987).
The idea of Meyer and Rowan (1977) that 'institutionalized products, services, techniques, policies, and programs function as powerful myths' which many organizations adopt 'ceremonially', appears to have a particular significance for understanding workers and organizations that tend to be labelled 'knowledge-intensive'. This approach means that functional thinking and the 'market knows best' argument is questioned. As with all theories, they sometimes work well, sometimes less so in relation to specific social and economic conditions and empirical variations. My suggestion is that the institutionalized myth perspective is especially valuable in understanding parts of the KIF sector, especially professional service firms. Sharma (1997) makes the point that the market for professional services is highly inefficient -the quality of the services is not easy to assess and there is seldom any systematic and comprehensive evaluation or communication of experienced outcomes between clients, so it is difficult for the market to react 'rationally' (Ernst and Kieser, 2003). This increases the 'space' for ceremoniality, without feedback from poor performances.
Institutionalists (DiMaggio and Powell, 1983;Meyer and Rowan, 1977) would argue that clients use consultants mainly because others do -using consultants means legitimacy and/or the reduction of cognitive uncertainty. One could add here that managers may also have political motives for the use of consultants (e.g. Bergh and Gibbons, 2011;Heller, 2002). Arguably, there are thus institutional as well as political elements reducing the force of the market logic. Rather than seeing people uncritically buying into institutionalized myths as cultural dopes, there is presumably often a calculation of the benefits of doing so, including political considerations. The selective 'consumption' -or exploitation -of institutions (myths) typically reflects a combination of conformist (taken for granted), economic, and political pressures and motives. Meyer and Rowan (1977) argue that formal organizations 'are driven to incorporate the practices and procedures defined by prevailing rationalized concepts of organiza-tional work and institutionalized in society' so that they 'dramatically reflect the myths of their institutional environments instead of the demands of their work activities' (p. 304).
A key aspect here is what is fashionable. During an early phase a strong degree of agency and innovative intensive rhetorical work is typically needed to convince audiences about the value of the new product or service. In Green and Li's (2011) terminology, conscious symbolic action prevails over unconscious material motion in the institutional work done by the agents. This means producing and anchoring what is to become new institutional products in the taken for grantedness of a field. But for other 'myth-providing' KIFs other than early starters, copying, varying, and refining the rhetoric may be sufficient and even the best way of profiting from a new fashion. Often something novel is defined as superior to something that is not. Consultancy services sometimes are delivered simply through relabelling what has been done before, but sometimes new services are developed and old ones are abandoned because it is vital to comply with new truths of what is good and what is obsolete. This process is typically non-rational or even irrational, from an economic point of view (Benders and van Keen, 2001). Still, 'knowledge' is needed, but the value of this in terms of efficiency is often questionable -or at least difficult to assess. Harmony with what is (on its way to become) institutionalized becomes more important than demonstrated efficiency, especially as the high degree of ambiguity prevents the reliable assessment of what contributes to efficiency. Sometimes there is a direct contradiction between making things look good and doing something that is more likely to increase efficiency or solve specific problems (Prasad et al., 2011), but as already said, the latter is often difficult to assess.
Rationalized myths exercise a strong impact on formal organizations which are compelled to respond to these through developing the 'right' structures including professions, programmes, and technologies in organizations. Meyer and Rowan's ideas have special relevance for the 'knowledge-intensive' industries. Most of their examples can be attributed to this category. Knowledge-intensive organizations -with a few exceptionscan thus be viewed as both surfers on and providers of the institutionalized myths. Sometimes they are also producers of them. Generally they are also carriers of them, i.e. the rhetoric of the myths has a strong grip also on those producers and providers. Surfing on a myth is much easier if one takes the truth of the myth for granted. Let us briefly explore these four aspects. As surfers, KIFs can rely on a strong market demand. Professional services and new knowledge products are asked for as these are prescribed in the institutionalized environment and KIFs can surf on an institutionally driven market demand rather than need to prove their added value. We have here a mythdriven demand. KIFs also provide specific services that make it possible for client firms to develop their own correct structures and technologies, such as equal opportunity and diversity programmes, vision statements, human resource management (HRM) procedures, psychological tests, and questionnaires. In principle many companies may be able to do parts of this on their own, but the work called for typically means that extra support is needed. In addition, this calls for the right 'expertise' in order to be fully persuasive and many KIFs here operate on the 'credibility market'. At least in publically exposed organizations -dependent on creating the right impression and vulnerable to exposure from mass media, unions, and interest groups -this is vital. The actual delivery can then, following Meyer and Rowan (1977), be seen as myth-compliant products/service. KIFs sometimes are producers of institutionalized myths, in particular large professional service firms. Large consultancy firms develop new knowledge products that sometimes are widely dispersed and become institutionalized, or commodified and colonializing (Suddaby and Greenwood, 2001), such as the McKinsey sponsored idea of the centrality and necessity of having a unique corporate culture (Peters and Waterman, 1982). In other words, KIFs are, often in collaboration with business schools and universities, eager to offer recipes that may become institutionalized. Finally, KIFs are also carriers of myths in the sense that the pressure within these firms to buy into the products and services offered, in combination with a self-serving bias, lead to a strong acceptance of the myths they disseminate. I frequently meet consultants, educators, and others who are absolutely convinced that their specific model for making organizations more humane and effective will produce really good results. Although people with cynical and instrumental attitudes exist, persuading others is often much easier if you are persuaded yourself. Being a devoted myth-carrier is then a resource for much KIF work.
Knowledge-intensive service organizations thus become vital symbols for client organizations' elaboration of rules and their requirements for rationality. The well-run company utilizes expertise from recognized knowledge-intensive firms for education, personnel recruitment, management development, IT development projects, market research, advertising, auditing, managerial advice, strategic planning, and so on. External as well as internal faith is thereby created. Executives in many organizations face normative as well as cognitive, and sometimes also legal pressures for adapting specific forms of knowledge: new strategic formulas, organizational reorganizations, for example (Nicolai et al., 2010). KIFs are thus often traders in myths.

THE CLAIMS OF KIFs: THE EMPHASIS ON DISCOURSE AND RHETORIC
The notoriously ambiguous character of (a significant part of ) knowledge work and the organizations (claiming to) be mainly doing this means that the demands of the agents involved, in terms of providing convincing accounts, regulating impressions, and images, are central. As knowledge does not speak for itself and as its value can seldom be simply demonstrated, someone is needed in order to speak for it. Knowledge or core competence is still vital; only being capable of speaking persuasively is not sufficient, but to be socially recognized as an expert is a prerequisite for any 'essential' form of expertise playing a role in business. Part of the necessary 'knowledge' is about how to act in an 'expert-like' way: by talking, dressing, and behaving in a convincing way (see also Biehl-Missal, 2011). The persuasive or rhetorical element is then vital. Being perceived as an expert is as crucial as actually being one. This means that organizational and professional discourse, including jargon, becomes important and may revolve around institutionalized vocabularies and discourses that express knowledge claims. Discourse is vital not only in order to address the external world, but also in order to constitute organizational members as confident and credible knowledge workers.
The view that is proposed here inspires a broader and more suspicious understanding of what KIFs and their workers do as well as a focus on their activities at a practical level.
Without denying that knowledge may be a functional resource that is directly applied in work, it is misleading to talk about 'pure functionality' decoupled from the discursive constitution. The adaption of the myth perspective means that other functions of knowledge and knowledge talk become central. Knowledge plays other roles, such as (a) a means for creating community and social identity through offering organizational members a shared language and promoting their self-esteem, (b) a resource for persuasion in, for example, marketing, PR work, and interactions with customers, (c) a way to provide the company with a profile (an intended image targeted at the market), (d) a means to create legitimacy and good faith with regard to actions and outcomes through the more or less subtle use of knowledge rhetoric, and (e) a basis for providing some sense of order and trust through obscuring uncertainty and counteracting reflection. This last point indicates that 'knowledge' and 'knowledge-work' may lead to the opposite of what it claims, to ignorance and uncritical attitudes (Alvesson and Spicer, 2011).
It is, of course, important to acknowledge efforts to distinguish between bad and good knowledge and reserve the latter term for what is assessed to be valid and valuable. One can imagine ambitious efforts to identify distinctive and exclusive qualities of knowledge. Schreyögg and Geiger (2007) suggest a Habermasian communicative framework, 'one that allows for drawing distinctions between high quality and low quality knowledge and informs us of what knowledge is and what it is not' (p. 81). This is interesting, and when clarification and consensus on what is valuable is arrived at after careful scrutiny, one could imagine there being knowledge that is clearly defined and functioning as a reasonably robust source of support, counteracting a high level of ambiguity. However, one can also imagine the opposite outcome, careful efforts to clarify high-quality knowledge within a community of communication, leading to dissensus and conflict and thus a reinforced sense of ambiguity. This is a not unlikely outcome of communicative action (Deetz, 1992) and efforts to accomplish decision rationality (Brunsson, 1985). A fully open and critical scrutiny of the forms of knowledge -of management, HRM, legal issues, new technologies, predictions of the future -that are objects for rapid promotion and sales in the market may result in uncertainty and doubt, rather than the opposite. The mythical qualities of the knowledge involved may be deconstructed and weakened, jeopardizing the rhetorical basis for KIFs -as an integrated base for persuasion. Internal doubt weakens the KIF as a system of persuasion and the external effects may be problematic. Risks of internal doubt and less self-confident external communication may discourage the ambitions and success of careful scrutiny, encouraging actors to hide rather than reveal and solve the ambiguity that is a key aspect of most forms of knowledge. In many businesses where strong persuasion and rapid moves on the market are needed, there may be limited space for carefully assessed forms of knowledge. I am not suggesting that this scrutiny-reducing and 'myth-preserving' ingredient is the only or dominating outcome, but the value of knowledge and identification of high-quality versions of it is not easily resolved in an area of complexity and ambiguity.

IDENTITY AT STAKE
Many of the key aspects of the 1993 paper indicated an unsettling existential insecurity for people in KIFs, in particular in professional service firms like management consul-tancy organizations. This goes against traditional understandings of professionals and professional organizations, seen as characterized by superior knowledge, a homogenous, protected field of work, and high status. Today these qualities are often called into doubt: knowledge and professionalization claims are abundant but increasingly contested, and most professionals work in large firms or organizations in which they are subordinates rather than autonomous professionals in a classical sense. The number of groups with more or less strong claims of professionalism is increasing, leading to inter-group conflicts, reduced status associated with being a 'profession', and general suspicion. In the increasingly expanding field of commercial professional services, fashionable 'cuttingedge' forms of knowledge deviate from traditional, more conservative knowledge claims of established professions. The high level of ambiguity of many contemporary KIFs is crucial here.
Existential anxiety comes from a lack of reliable stabilizers confirming a positive and coherent sense of self (Knights and Willmott, 1989;Sennett, 1998). Crucial here is the difficulty in demonstrating any clear results (Clark, 1995). The relationship between what is being espoused -in marketing and prescriptive texts -and what is actually being done may be loose (Benders and van Keen, 2001;Bloomfield and Danieli, 1995). Furthermore, consultants' relations with client personnel may deviate heavily from functionalist models of the professional helper (Alvesson and Johansson, 2002). The ideal of 'client orientation' is sometimes espoused (e.g. Anderson-Gough et al., 2000;Löwendahl, 1997), but what client relations may look like in specific projects may be another matter (Kosmala and Herrbach, 2006), and 'client orientation' may be translated to 'whores in stripes', by those suspicious about which client the orientation and loyalty is directed to, and how this is being expressed ( Jackall, 1988). Units or actors of the client organization are often in plural, shifting and heterogeneous, and are also in one and the same overall project ). Many authors (e.g. Alvesson, 2004;Czarniawska and Mazza, 2003;Sturdy, 1997) emphasize the uncertainty and anxiety of consultants in complex and uncertain work situations. Kitay and Wright (2004) suggest that 'the image of the confident consultant/gullible manager may be balanced by the alternative of tough client/vulnerable consultant' (p. 3). Sometimes consultants as representatives of management interact with subordinates negative to proposed changes and thus are forced to balance between different interests (Whittle et al., 2008). Given the sometimes conflicting and contested nature of consultancy work, there is a shortage of a clear confirmation of a positive self-understanding. This view means that issues around selfesteem and identity construction become problematic for people involved in knowledge work, lacking reliable knowledge, measurable results, and alignment of talk and practice in support for 'positive subjectivity'. Fragmented and vulnerable identities may follow. This leads to existential insecurity and a pressure to actively construct and reconstruct a clear and positive sense of self. How consultants deal with identity themes then becomes important to address in understanding consultancy projects.
One can argue that people in organizations often engage in identity work, aiming to achieve a feeling of a positively valued self-identity as well as a basis for social relations, which is necessary for coping with work tasks and social interactions (Alvesson and Willmott, 2002;Kjaergaard et al., 2011;Knights and Willmott, 1989;Pratt et al., 2006;Sveningsson and Alvesson, 2003). Self-identities are constituted, negotiated, reproduced, and threatened in social interaction, in the form of narratives, and also in material practices. Subtle and not so subtle communication is central. As consultants and other knowledge-intensive workers often invest a lot of themselves in their work and careers, the identity aspect becomes heightened and at stake. People engage in identity work when the routinized reproduction of a self-identity in a stable setting is discontinued. It is triggered by uncertainty, anxiety, questioning, or self-doubt (Collinson, 2003;Knights and Willmott, 1989). The lack of stable and secure confirmation by significant others is a major driver of identity work, aiming to lead to a restoration of a non-confirmed identity. As Deetz (1998) writes, 'the largely fluid nature of anything external to interactional accomplishments provides for very active symbolic labour ' (p. 157). This symbolic labour takes the forms of discourse, narratives, and uses of specific tropes (Green and Li, 2011).
Many consultancy firms and other KIFs work intensively with various forms of identity regulations in order to provide a set of symbolic stabilizers for the employees (Alvesson, 2004;Alvesson and Empson, 2008). Even seemingly rational structures and procedures such as ambitious HRM programmes can be seen more as supporting a system for normative identity regulation rather than providing accurate feedback, supporting career development and fair promotion (Alvesson and Kärreman, 2007). This became clear in a study of HRM in a large management consultancy firm. Here senior and junior people praised the firm for its superior way of assessing, developing, and promoting people, investing a great deal of resources and time in making this work. On the surface the system was impressive and employees referred to it as proof of their quality and development, and thus of the competence of the firm. But a closer look showed strong cracks in the systems and its specific workings. The HRM system may appear to effectively deal with uncertainty, but simultaneously it hides and reinforces it as the aimed for consistency and clarity is hard to accomplish. High ambitions and claims for ambition, coherence, and rigour open up space for deviations and contradictions. Clear criteria and prescribed procedures and assessment frameworks are, for example, undermined by politics, situation-specific home-made criteria, shortage of time, idiosyncratic assessments, and the need for flexibility. Often people that claimed the superiority of the firm's promotion policies and practices could not understand specific promotion decisions ('how in the hell could he get this position?') and noted many deviations from prescribed procedures, such as inflated rankings and feedback sessions being little more than self-assessments polished by the manager. But these experiences of a mismatch between policy and reality were compartmentalized or seen as an exception and the 'official' discourse of the ambitious and effective HRM system was rather used as an identity resource and as an ingredient in image management (Alvesson and Kärreman, 2007). In other words, HRM as an internal system for identity regulation created suitable subjects, with good rhetoric skills to persuade external stakeholders about their and their firm's quality. In the language of Meyer and Rowan, we can say that certain institutionalized myths (knowledge used for rational recruitment, development, promotion, and staffing of human resources) were internalized and used to persuade people, internally as well as on the market, about the value of the institutionalized myths (consultancy knowledge) provided. Rhetoric is important here, but so are the individuals (with bodies and brains) recruited as well as an assortment of practices and arrangements that gives symbolic as well as material support for various claims and persuasive moves. Apart from technical expertise, social skills and networks are important (Alvesson, 2004;Starbuck, 2010) and these resources can not be reduced to be only about language use and rhetoric. There is, as I think Green and Li (2011) exemplify, a risk of overemphasizing language and rhetoric here (Alvesson and Kärreman, 2011).
The key themes of the 1993 paper -ambiguity, rhetoric, and image -and the added theme of identity have been further addressed and developed in a number of my recent publications (e.g. Alvesson, 2004). I will here present some empirical findings, based on a case study of a management consultancy project, in which we interviewed both consultants and client managers.

THE IMPORTANCE OF POWERPOINT PRESENTATIONS: A CASE OF MANAGEMENT CONSULTANCY
The case is about administrative rationalizations involving a group of consultants from a global consulting firm (BC) as well as a number of managers in the client firm, the national subsidiary of a large life science firm (Alvesson and Sveningsson, 2011). Striking in a set of interviews were the different accounts of what happened in the project as well as what were the outcomes, indicating the high level of ambiguity of (many) consultancy projects.
The senior consultant, Farringdon, emphasizes the general value of the project. He refers to huge savings for the client firm and a positive image as strong indications of a successful assignment. He ascribes force ('we pushed') and boldness ('revolutionizing solutions') to himself and his colleagues, while the people from the client are seen as the opposite; as weak and unable to act: The client was too cautious, nothing happened, nobody from management pushed. The reason was their history. HT was a spoiled organization, a little fat . . .
The positive results are then mainly the accomplishments of the consultants, while the non-realized potential is attributed to the lack of drive and confusions of the client people. Client people are seen as morally degraded ('spoiled', 'fat'). Junior consultants partly diverge from this view. They did not think that huge savings were enabled, and emphasized that not many of the ideas and plans were actually implemented due to a shortage of personnel and time on the client's side, which in turn partly disconfirmed the senior consultants' opinion about the results. They did, however, like Farringdon, emphasize their own really good ideas and suggestions which, if implemented, would have made substantial savings possible. The narratives then involved strong persuasion of themselves as well as their audience (researchers) about their professionalism, drive, competence, and (potential) contributions.
The client managers, on the other hand, were not so impressed by the ideas and suggestions of the consultants, promising cost savings, but neglecting or underestimating implementation costs for the change: If we make a change that costs us 30 million it's not profitable.
De-Essentializing the Knowledge Intensive Firm BC thinks in terms of concepts. But doing an analysis in order to document what consequences this is having if we implement the changes, that is an assignment for them. I am dissatisfied with what BC has delivered. It's a problem. BC doesn't take responsibility. They think conceptual thoughts about how it could work but haven't gone out and acted in the business.
The business case is thus subject to diverse assessments, ranging from huge potential savings to a very costly investment, if implemented. Another client manager, Hodges, refers to symbolic performances involving a lot of rhetoric: They're professionals when it comes to making PowerPoint presentations, no one can object to that. They presented more overheads in order to push our system aside than the total amount of our systems documentation.
Hodges says the consultants were preoccupied with marketing and trying to promote themselves. Rhetorical skills were thus recognized, although not appreciated, as the consultants failed to communicate an image of themselves as competent deliverers of results. Client managers were hostile to the consultants' sales and promotional efforts, seeing their skills in PowerPoint presentations as camouflaging, indicative of weaknesses in terms of having any valuable knowledge and contribution to deliver end results. This indicates the limitation of the significance and range of an exclusively rhetorical approach, like the one suggested by Green and Li (2011), and emphasizes the need for careful studies of processes, interactions, and their reception (see Alvesson and Kärreman, 2011, for an overall critical discussion.) However, for the senior client manager, responsible for the change project and the one who had contracted the consulting firm, their skills in presentations were seen as something that led to the project being seen as positive by top management, despite lack of specific results.
. . . what have come out of this are the presentations that Farringdon and Blake [senior manager in client firm] had for management in London [and] everybody thought that this was an amazingly good and well-accomplished project. But that everything can't be accomplished has political reasons.
To go further and investigate what 'actually' happened in the rationalization project and the exact role of the consultants and the client managers is difficult, as the interviewees gave rather different accounts. It is tricky to 'objectively' determine the skills, deliveries, concept-focus, inertia, and lack of team leadership of those involved in the process, as the diversity of assessments and ascriptions of credit/blame vary heavily between those involved, underscoring the ambiguity of the consultants' work and the outcome of this.
Blaming/crediting involves drama and vivid experiences which contribute to 'effective' identity work. The consultants' identity work, in terms of securing a sense of self, seems to be the expression of ambition, drive, consistency, knowledge, and ability to participate in radical changes. They also indicate that they are clear, professional, active, result-and long-term oriented, in contrast to the client that they construct as being unclear, passive, non-knowledgeable, and unreliable. The construction of the 'fat', 'spoiled and passive' other, i.e. the client people, allows for the consultants to view themselves as ambitious, energetic, and efficient. These are discourses that have a 'semi-institutionalized' nature and are therefore credible resources available for use in the identity construction. In addition to my main point in the 1993 article, about KIFs as 'systems of persuasion', one can here add 'systems of self-persuasion', adding a flavour to understandings of organizational culture and identity. Rhetoric here is a combination of the knowledge workers using language but also being used by (institutionalized) discourse in producing and reproducing a persuasive view of themselves (Green and Li, 2011).
The ideal consultant comes forth more magnificently against the background of an other, the (typical) manager, here constructed as spoiled, soft, irrational, fat, and incapable of running a competitive business. From the view of the other, that is, the client manager, we see a similar, although opposite, discourse, about the superficial, ignorant, greedy, conceptual, and PowerPoint using consultant, to be contrasted with the substantive, knowledgeable, trustworthy, and result-oriented manager.
The absence of clear confirmation of people's skills and accomplishment is a threat to experiences of security, self-confidence, and coherence. This may trigger strong efforts to (re-)create a sense of a valued, competent and coherent self, e.g. to do intensive identity work. The ambiguity is however also a 'resource' in that it makes it possible for the people involved to move between criteria, allowing multiple versions of successes and failures, thus facilitating identity work. These moves increase the ambiguity as they hamper the negotiation of shared meanings. The unclear character and mixed perceptions of consultancy projects thus seem to (a) produce a need for and (b) provide symbolic resources for identity work (Alvesson and Sveningsson, 2011).
The case illustrates some of the key themes of the conceptual 1993 paper. We find considerable ambiguity in terms of what the consultants bring into the project, what they actually do, and what the results are, if any. Those involved use discourses on clients and consultants in order to produce a persuasive picture of themselves, producing legitimacy and identity for their knowledge, contributions, and character. While mainly used 'internally' (identity-focused) in the specific study, indirectly there are implications for image management. In the consultancy case, stories about client people being slow, cautious, spoiled, and lagging behind in mobilizing resources pave the way for the need for the consultancy firm. These stories, as well as the carefully prepared PowerPoint presentation -a common form for the business rhetoric of today -do seem to work better in relationship to client top management than clients involved in the practical work. The suspicion of people focusing too much on rhetoric and the risk of a backfire is worth noting, calling for much more careful attention to how people actually respond to discourse and rhetoric than is common within the field (Alvesson and Kärreman, 2011).

PROBLEMATIZATION: CHALLENGING ASSUMPTIONS AS A KEY TASK IN RESEARCH
The (re)interpretation of key issues around knowledge in at least a great deal of knowledge-intensive businesses exemplifies a problematization of a theoretical field . The functionality of knowledge and institutions believed to be based on as well as working with superior knowledge, leading to the creation of good results, is targeted for scrutiny and reassessment. This has lead to a basic reframing of understanding and a shift in focus from knowledge as vague, black-boxed, or weakly unpacked (e.g. through four-fielders or other categorizations of types of knowledge), but nevertheless a reliable and functional phenomenon, to an emphasis on claims of knowledge. Rather than to speculate about the content and effects of the 'black box of knowledge', it is better to study the accessible phenomenon of discourses on knowledgeintensiveness and knowledge-intensive work and organization. In most cases where this discourse is in play, it is employed in a persuasive way: knowledge is good, valuable, functional, and a resource that needs to be supported.
I have throughout my career worked with a sceptical distance to self-evident truths and fashionable ideas. This includes not just positions and areas that by critical and interpretive researchers are typically seen as the usual suspects, e.g. functionalism, managerialism, positivism, and technocracy, but also other approaches that one spontaneously may feel belong to the other and right camp, including in my case approaches like organizational culture, identity, gender, postmodernism, and discourse analysis. It is here important not just to associate oneself with and apply a specific framework, making problematization predictable, but also to target one's original framework to investigate and challenge its assumptions, thereby coming up with new ideas rather than reproduce and reinforce established ones. This means an ambition to not move with the mainstream -nor necessarily to embrace an established anti-mainstream school either -but to follow the development at close range and investigate whether a critical view on its assumptions and vocabulary may lead to alternative, perhaps more unexpected and interesting conceptualizations. We should always hesitate a bit before we jump on what is new and fashionable, thus embracing uncritically its basic assumptions. Over the years I have tried to develop a mildly positive interest in and openness to inquiries in various knowledge fields at the same time as I have sceptically asked myself -and the field -what goes on here? What are these people (us) in this discipline up to? Combining a genuine interest with an effort to seriously consider a counter-position is the core of what I see as reflexivity (e.g. Alvesson and Sköldberg, 2009;.
More recently Jorgen Sandberg and I have developed a methodology for the problematization of assumptions in a field, the ambition being to develop new, unexpected, and challenging research questions and framings of projects. We see a strong trend in research to be more rigorous, incremental, narrow, and predictable than imaginative, surprising, and provocative. There are many complaints about a shortage of really interesting and novel work, and investigations show that organization studies and other areas are heavily characterized by gap-spotting, i.e. researchers more or less accept existing literature as a given and as a point of departure, and in turn try to identify and fill in the gaps . This seldom leads to exciting ideas, research questions, or results. An alternative is to not reproduce assumptions, but to critically explore and challenge them, and develop alternative ones . The 1993 paper in JMS in a sense exemplifies this approach, and the interest in the paper and some of my similar other efforts have reinforced my conviction that we need more assumption-challenging research efforts. Any new -or existing, well established -research field then offers opportunities not only for ample gap-finding and -filling, but also new ideas based on the critical challenging of what is taken for granted. The broadly held assumptions of functionality, centrality, and result-guaranteeing quality of 'knowledge' in management consultancy and other knowledge-intensive firms and occupations held by most researchers in the 1990s offered good points of departure for the rethinking and reconceptualization of key aspects of this kind of firm and the work involved.

SUMMING UP AND A PROPOSAL
The great interest in knowledge, KIFs, and the booming of claims about their extraordinary significance includes some ingredients that appear to characterize many popular areas of research and publication: (a) the key concept is used broadly and either nondefined or vaguely defined, allowing it to work as an all-embracing sweep; (b) despite this all-embracing and ambiguous nature, it is somehow still addressed as a robust and reliable force, mechanism or resource; (c) it is vital for accomplishing any kind of good outcome; (d) the framing includes some positive performative effect or the phenomenon tends to imply a positive trajectory (e.g. knowledge sharing means that people can do a better job, a positive result is tautologically viewed as indicative of knowledge having been used); (e) users of it are very reluctant to acknowledge any limitation to or boundary condition for its analytical or practical usefulness (e.g. the costs of or inherent contradiction of knowledge management are neglected); and (f) the general situation and development of the economy or society provides a scene that is perfect in making the key concept play a star role.
Knowledge is in good company with many other themes. Organizational culture and identity has been mentioned, but one could also add branding, innovation, entrepreneurship, quality, and strategy, as well as leadership. All authors of course agree that there may be something wrong with a specific culture, identity, quality programme, leadership, or strategy, but the majority would claim that you do not need less of it or something else, but a 'better' culture, identity, quality programme, leadership, or strategy.
Despite a couple of decades in focus, knowledge is still seen as crucial, but it is often black-boxed and thereby mystified. Its nature and ways of understanding are still debated (Spender and Scherer, 2007). I read this as a sign of the problems with essentialistic understandings, supporting a more sceptical view that emphasizes the rhetorical nature of the role of knowledge and knowledge claims in business and organizations.
To address rhetorically powerful and intuitively appealing but broad, vague, and possibly empty concepts like knowledge, leadership, and quality carries advantages for researchers, such as appearing to deal with really important and impressive topics. People are happy to gather under these labels, often functioning much better as signposts for researchers to organize themselves and boost an image than as a starting and anchoring point for interesting intellectual work (Astley, 1985). We live in an age of image and grandiosity, where what we are doing and standing for should appear as highly impressive (we focus on knowledge not information, strategy not planning, leadership not managerial work or supervision, human resources not labour) (Alvesson, 1990, forthcoming). In this age, where the economy is increasingly about intangible work involving persuasion (marketing, branding, mass media communicated images) rather than creating products and services with unambiguous use value, intellectual work easily also becomes adapted to and skilfully employing institutional forms of rhetoric. This is hardly new, but the pressure for effective issue-selling and image-management has its unfortunate effect on scholarly activities.
This overall context increases the need for problematization projects that are not just accepting the inevitable nature of rhetoric in research but also target forms of it for critical scrutiny. The dominant view on the subject matter can be targeted by a problematization effort where key ingredients are scrutinized and alternative assumptions expressed . In my 1993 article, I addressed the ('essentialist') ideas of KIFs being characterized by (a) the possession of robust knowledge and (b) with the deployment of this knowledge being a core quality of the work being done and which then often is seen to (c) lead to good results. I also discussed how (d) a market logic brings a rationale for and quality assurance of KIFs and (e) the general knowledgeintensiveness of the knowledge economy is the main driver behind the rise and expansion of KIFs as a core category of contemporary firms and, even more, of the future. As a formula for problematization, I guess points (a) to (e) may also be of some relevance for many other problematizing projects.
In the 1993 piece, alternative assumptions were formulated and a counter-picture offered. These counter-ideas can be understood either as a superior (more insightful) understanding or a supplementary one (broadening the spectrum of aspects worth considering), or as a key ingredient in dialectial thinking. If we take the last option seriously, then research guided by a repertoire involving both a version of the mainstream view and the problematizing one could open up for interesting studies with no guaranteed outcome. Such studies unfortunately seem to be rare. Too much empirical inquiry seems to be caught by a given framework and vocabulary bringing about predictable results. While we can never, in an empiricist mode, assume that raw 'data' gives us the truth or be treated as external to, and used for testing theory (all data are impregnated by theory), one may try to use empirical studies for getting critical feedback to ideas and as stimulation for rethinking assumptions and convictions. Studies bearing in mind both conventional 'essentialist' ideas about expertise producing solutions, as well as the counter-picture in the form of a constructionist one, where KIFs are seen as 'systems of persuasion', surfing on and providing institutionalized myths, could be interesting to carry out. Following key people in KIFs in their development, confirmation, communication of, and working with knowledge -or whatever they are working with (sweat, tears, rhetoric) -and their interactions with clients/customers, with a variety of ideas and critical imagination in mind, is still a research topic open for grabs, despite a couple of decades of interest in KIFs and knowledge and hundreds of studies on the topic.