Academic journal article International Journal of Action Research

Constructive Practice and Critical Theory: The Contribution of Action Research to Organisational Change and the Discourse on Organisations

Academic journal article International Journal of Action Research

Constructive Practice and Critical Theory: The Contribution of Action Research to Organisational Change and the Discourse on Organisations

Article excerpt

This article presents a new perspective on the question of how action research may contribute to improving the discourse on organisations. The three first sections deal mainly with some important features of action research, following from action research methods used in projects that comprise organisational change. On the basis of a distinction between practical discourse and theoretical discourse, the point is made that while descriptive research (like organisation theory) takes place mainly as a theoretical discourse, action research also enters the arenas of practical discourse. What kind of knowledge is required, and what kind of experience is made in practical discourses, is elaborated by one example of an action research from a Norwegian international corporation. It is argued that in order to cause practical change, the power of knowledge is dependent on the power of judgment.

On this basis, the three last sections deal with this question how knowledge and experience from action research may contribute to the improvement of organisation theory. Initially, a short historical account on the development of organisation theory is presented. It is shown that the split of the discourse on organisations, into a theoretical and practical discourse, has had some unintended and unnoticed consequences as regards the style of writing in organisation theory. This style of writing has resulted in a discourse on organisation which is rich in very general perspectives and concepts, but which nevertheless remains too poor in content. Thus, the conclusion is that for the time being, one of the most important contributions from action research to the discourse on organisations will be to make organisation theory become subject to criticism that may provoke changes in the style of writing organisation theory.

Key words: Organisation theory, practical discourse, theoretical discourse, power of judgment, style of writing

1. Introduction: On the difference between action research and descriptive research

Most kinds of social science somehow have to cope with a challenge which is conceptualized in many ways: System vs. actors, structure vs. event, structure vs. process, etc.. All these concepts, not to say dichotomies, aim at designating the static and respectively the dynamic aspects of social phenomena. The need to apply/make use of these kinds of concepts may, in a simplified way, be explained by the fact that when we write about some social phenomenon in order to understand it (i.e. in order to create a piece of scientific knowledge), it will not suffice just to 'say what happens/has happened'. Everything that happens to people, everything people do, does not simply happen or is simply done. All events take place within frames, whether these are acknowledged or not by the actors.

This is common wisdom for social scientists, but perhaps not for common people, at least not to the same extent, or in the same way. However this may be, social scientists usually consider their awareness of the social structures within which people act as being among the strengths of social science as such. Thus, the focus on structural features of social phenomena, structural explanations etc, is quite strong in many of the disciplines in the social science.

In the field of organizational studies this kind of structural bias has been noticed for a long time, and it has been criticized from a number of different angles. At the most general level, the point is made that the social structures within which humans act are themselves created by human actions. Another kind of critique, addressing the conceptual practice of social scientists, rather than the practice of people who make society, questions the kinds of theorems, dogma or phrases which say that people act within frames. How can we decide, who is to decide, or from what position to decide, what is inside and outside the frames? But there are also other kinds of frames at work, which are not always noticed by the social scientists: the theoretical framework of social science itself. And perhaps it may also happen, that this framework works to put social science 'outside' the real dynamic, and thus the structure of the social phenomena under study?

The theoretical framework is made up of concepts, models and perspectives that in turn are made up by words. They have come into public existence by means of the words written in publications. And, as we know, words may refer to things, but words are not identical with the things they refer to. That is what Hegel wanted to remind us of, when he once wrote: "The word 'dog' cannot bark". Of course, we know this very well. That is why it seems to be necessary to be reminded of it again - and again: The fact that we use words when we talk and write about any social phenomenon is so obvious that usually we do not pay any attention to it. The consequence is that all the time we run the risk of mistaking a word for a phenomenon.

You may think: no researcher is that stupid. But this is not a question of being intelligent. Let me present an example of this kind of mistake: our use of the word Organisation'. About this, Karl Weick once wrote: "The word organisation is a noun and therefore a myth. If you look for an organisation, you will never find it. What you will find is events that are linked to each other, events that transpire through walls, corridors, time schedules etc.. It is these events which we wrongly turn into a concrete substance, when we talk about an organisation. Instead of talking about organisations, we should talk about organizing" (Weick 1975).

Weick's point is that we may, without really noticing it, turn what takes place as events into what seems to be a structure. In most approaches to organisation theory the word Organisation' is so common in use that it has become an almost self-evident term, and in turn this use works as evidence for the existence of the phenomenon. In organisational studies, a number of approaches have been developed during recent decades to cope with challenges like this. Among these are different ways to apply some kind of narrative approach: various strategies of 'telling the story'. Also action research approaches often represent, explicitly or implicitly, a kind of critique of overly structural approaches in organisational studies.

However, the structural bias of organisational studies is not criticized by action researchers just for theoretical reasons, but also for practical ones. Structural analyses of organisational phenomena may very well present the main reasons why things work the way they do, why the situation is as it is, but such analyses do not necessarily present much knowledge on how to change the situation, how to improve it, how to make things work differently. To come to knowledge of some structural aspects which condition the way people act in particular organisations, does not automatically lead to a way, or a possibility, to act to change these general conditions.

Organisational studies often seem to be performed as if their addressee is a very determined collective actor, who is capable of acting in order to change general structures as soon as knowledge of what changes ought to be made is provided. But even though it is usually possible to create some general knowledge, there are usually no corresponding possibilities of some 'general action'. Organisational change usually takes place by means of very differentiated forms of actions, as a practical process performed by a great number of actors coping with a multitude of details. In short, changing organisational structures takes place by a multitude of organisational events.

As we have seen, Weick poses the problem strikingly. But what about his suggestion for a solution? If he is right that organisations in practice are made up by events, will it help to talk about it in terms of verbs rather than nouns? Will it help to condemn the word organisation and use the word organizing instead? I do not think so. Instead of talking about organizing, we have to do some organizing. Because, as we know, the word Organize' cannot organize. Weick's theoretical point is quite right, but to make his point have some practical impact, we have to make some practical steps. This is exactly the point where action research differs most radically from descriptive research. To clarify this point, I will start by making a simple, pragmatically conceived distinction between practical discourse and theoretical discourse.

To have theoretical knowledge of something means to understand it. To make practical changes in something means to do something with it. This simple fact may help us to make a quite useful distinction between what I will call respectively theoretical discourse and practical discourse: a theoretical discourse is undertaken in order to understand something, while a practical discourse is undertaken in order to do something.1 By differentiating between theoretical discourse and practical discourse, the main difference between action research and descriptive research comes to the fore: The arena of descriptive research is almost exclusively the theoretical discourse, while action research also enters the arenas of practical discourse, that is, we engage in some kind of dialogue and cooperation with the actors in the field.

The participation of researchers in practical discourses may take place in many ways. There are many approaches, many positions. But, across the different positions that action researchers take in this question, there is a challenge common to all: how can we get into a position to actually carry out action research in order to promote democratic development in practice? Again, the conditions around the globe are quite different. Within working life, the quality of the relations between the social partners is quite important. In Scandinavia these relations are generally known to be relatively good. In Norway the labour market parties have, for 40 years, given their policy support to action research projects within working life. Part of the aim of this action research has been to contribute to the democratization of working life.

However, policy support is not equal to financial support. And even though there is a general agreement at the top level between the labour market parties, the decision to involve themselves in action research projects is made at the enterprise level, by local management. Among the actors of working life, not least among the most powerful ones, that is, the owners and the management, the interest in contributing to the democratization of working life may be rather modest. In the West, capitalism and democracy may be regarded as true-born twins, but they are not identical twins. The relationship between political democracy and economic democracy is one of tensions. The interest for these questions seems to be far greater among social scientists and social researchers than among the actors of working life. Against this background: does the knowledge we have as action researchers constitute a kind of power of knowledge, strong enough to not only match the power of the actors of the basic units of capitalism, the enterprises, but to contribute to practical change?

2. The power of knowledge and the power of judgment: An example of action research

I will examine this question by means of an example: Norske Skog Folium is a paper mill with about 400 employees; this paper mill is part of an international corporation with about 9.400 employees. The union leader at this paper mill knew our (WRI) approach from a presentation we had given at a conference.2 Our approach, emphasizing broad participation in development processes as a strategy of democratization, appealed to him. He persuaded the top manager at this paper mill to have a meeting with us.

The development process we were invited to advise them on was one of organisation development. In order to survive as a paper mill producing paper of special qualities, the performance of the production had to be improved. On this the top management and the union leaders were in agreement. However, not surprisingly, they had different views on how to reach this goal. Both views concerned the organisational structure. To simplify a bit, the union was concerned mainly with making the organisational structure less hierarchical, and to increase the involvement of the employees in questions pertaining to their own workplace and questions of improvement and development. One condition for obtaining this they considered to be removing the foremen. The management was concerned mainly with creating teams among the workers on the shop floor, in order to improve the performance of the employees.

The union leader was not really negative to team-organisation; his view was that removing the foremen was also a necessary condition in order to make team-organisation work in practice. Neither was the management totally negative to removing the foremen. In addition, both the union leader and the management were aware that among the rank and file at the shop floor, there were lots of negative attitudes towards the concept of team-organisation.

Now, such was the different views of these powerful actors at this enterprise. What, then, about the power of knowledge? How could scientific knowledge help to mediate between these two views? As you have noticed, the local parties were not in diametrical opposition to each other, but their opinions were different enough to block any practical progress: they were not able to agree on what to do or how to proceed. Their expectation of us was that we should advise them as to what kind of new organisational structure would be the smart one, and advise them on how to get there.

In fact, our judgment was that the best solution would be to both remove the foremen and to create a team-based work organisation; in addition we thought they should also create a more team-based management. However, to present our thoughts in the form of a general model of a new organisational form, would not have been a very smart thing to do. The result might very well have been that the union leaders and the management had come to an agreement about our suggestion as a model for a new organisational structure. Nevertheless, the problem of how to proceed, in order to make the new structure also really lead to a better performance, would have remained - as the main problem. This is because, to our knowledge, a new, less hierarchical and more democratic structure, does not automatically lead to a better organisational performance. Of crucial importance is the very process by which the organisational changes are made.

This was our judgment, based on our - let us call it scientific - knowledge. But, we also knew that simply to present this knowledge to them (union/management), would not work. The power of knowledge does not work that way. Since theories are made up by words, any knowledge of any phenomenon is a certain kind of interpretation of this phenomenon. And, even though the local actors may lack knowledge of the kind of organisational theory that we have as researchers, they do not lack theories. They simply have other theories, in the form of their own interpretations. In other words: Solid knowledge is not a solid. Thus, there is no open space in the heads of the local actors into which our knowledge simply may be 'poured in'. In order to make our knowledge, our theories or interpretations, relevant to the local actors, we have to relate to their interpretations.

Therefore we have to get knowledge of their interpretations. These interpretations usually are manifold, rich in content and not to be found in a database from where they may be printed out and read. For us to "read" their interpretations, and for them to "read" ours, we have to enter into some kind of dialogue with each other. This means that the way we contribute with our theoretical knowledge cannot simply take the form of a theoretical discourse, in which we try to make the local actors come to an agreement with us on a common understanding of what is the situation. Rather, our contributions have to be presented within a practical discourse, where we try to find answers to the question about what they can do (in this situation).

In this sense we may say that the power of knowledge is dependent on our ability to enter into a practical discourse, in the form of a dialogue whose outcome to a large extent is dependent of the 'power of the better arguments'. I make this allusion to Habermas' well-known phrase deliberately. As researchers, we are not in the position to make decisions (on what to do). The management has the exclusive right of making decisions, eventually in some kind of cooperation with the union. Our chance to influence these decisions is given by the quality of the arguments we are able to provide during the dialogue. Our status as researchers, as representatives of a scientific community, does not make our arguments any stronger. Confronted with the actors that hold the powerful positions in capitalist enterprises, we do not occupy a strong position. We are not stronger than the strength of our arguments. Thus, what is decisive is the content of our arguments.

The content of our arguments, then, is not to be simply derived from our theoretical knowledge of organisational structures, processes of organisational change etc. One important reason is that the dialogue between the parties at the enterprise and us often takes on a course that is not possible to foresee. Suddenly we find ourselves discussing issues that, to be dealt with properly, require life experience quite as much as theoretical knowledge. Another reason is that sometimes we have to really create arguments, invent new arguments, on location, so to speak. That is, we have to create new arguments on the basis of the understanding that we develop in the course of the dialogue with the local actors.

This means that in order to try to create practical results that are in accordance with our theoretical knowledge, we have to admit that the power of knowledge is not powerful enough, if left to itself. The power of knowledge has to be supplemented by the power of judgment. It is necessary to have a sensible power of judgment, in order to understand what takes place in the particular dialogue we participate in, and on that basis to create those particular arguments that are apt in this dialogue. To put it into a formula: the power of knowledge is dependent on the researchers' ability to participate in practical discourses based on of the power of the better arguments, and this ability is in turn dependent on a sensible power of judgment. For these reasons, what will be the content of the better argument will differ from case to case.

In the case of Norske Skog, the kinds of arguments that turned out to be most powerful were these:

1) Instead of recommending any particular theoretical model of the future organisational structure, we recommended them to agree upon what kind of development process to organize.

2) Among the rank and file there were lots of sceptical attitudes towards teams. A suitable way to cope with this would be to invite all employees to participate in well organized dialogues to give their answers to three questions: a) What do they consider to be the main problems of the factory? b) In what way might team-organisation be helpful in solving these problems? c) If a team-based work organisation should be developed and implemented, how then is it best to proceed?

3) On the basis of these dialogues, in which practically all employees (plus management) should participate, the question of how to organize the process of organisational development could then be decided upon.

Last, but not least, I have to mention that in addition to the arguments we gave them, we also gave them ¢ promise'. We offered to take care of organizing the dialogues based on broad participation. As a part of accepting our arguments, they also accepted this offer. Thus, in order to create a process of democratic development, not only theoretical resources were required, like the power of knowledge, the power of the better arguments and the power of judgment. Also our practical skill in the organizing of development processes, based on broad participation, was required. In other words, for action researchers it is not sufficient to have theoretical knowledge of the concept of practical discourse. It is necessary also to have the skill to participate both in the organizing of and in the performance of practical discourses.

Now, to make a long story short: the management made their decision on a process based on broad participation, and we kept our promise to participate in organizing it. This process included three dialogue conferences, comprising 200 participants; a subsequent decision on developing models of teamorganisation in a number of project groups; establishing a steering committee; working in project groups; new dialogue seminars on theoretical models of teams; creation of teams; selecting team leaders and, in the end, to use the ordinary management vocabulary, the 'implementation' of teams. We organized the dialogue conferences and participated in various kinds of critical events along the process, like meetings in the steering group, in some of the project groups, etc. In short, we participated both in the organizing of and in the performance of the practical discourse that was undertaken at Norske Skog.3

3. What kind of knowledge is required in practical discourses?

Some of this kind of project work is well known to everybody who has experience of this kind of processes; it is rather trivial and relatively easy to perform. Therefore, I would like to highlight a less well known, or perhaps less acknowledged, aspect of this kind of processes. This aspect concerns the relation between dialogue and action. From words to deeds: that is the logic of this relation. But very often this logic is conceived in a too linear way. One thinks that in order to be able to take the step from dialogue to action, the dialogues have to generate a common understanding. Such a common understanding, a kind of consensus, is considered necessary as a common ground for action.

But this conception may very well be a misconception, or a misunderstanding. As a matter of fact, a very common misunderstanding, common both to practitioners and to theorists who are experts in organisation theory, is the following: They think that if the future solution on organisational problems becomes sufficiently detailed and then agreed upon, the practical steps to create this solution will follow almost automatically. But, what is called a "solution" is, however detailed, still a solution on paper, that is, a theoretical solution. Both theorists and practitioners happen to make mistakes here: The words by which the solution is conceptualized are treated as if they were identical with the practical solution.

But the real phenomena are different from the words by which they are conceptualized. Thus, regardless of the degree of details agreed upon in the theoretical draft of a solution, the practical steps to be taken do not follow automatically, as if the real action is simply a practical consequence of the theoretical conclusion. This kind of mistake as regards the real status of a solution on paper is more likely to happen within a theoretical discourse than within the practical discourse. But also practitioners may easily be victims of the temptation to enter into theoretical discourse. And this may very well happen while they are in the midst of performing a practical discourse.

Therefore, more than often we as action researchers have to make points like the one I have just made, to those we collaborate with. Not as a theoretical point, presented as a piece of knowledge; literally we have to find practical answers to theoretical disagreements among the practitioners. That is, at certain phases in the process of organisational development, we have to point out that what will make progress is not to agree upon some particular solution. What will make progress is to agree on what practical steps to be taken next, in order to organize, or to continue, the process by which the practical solution is to be generated.

For these reasons the point is not to create a common understanding by means of dialogues. Rather the point is to create common dialogues, a common practical discourse within the organisation(s). What is needed in order to develop more democratic work organisations is not a common theoretical understanding of what a democratic organisation might look like. What is required is a common practical discourse, which allows for everybody to participate, and which allows for different understandings and interpretations to come to the fore. Thus, the really big challenge in this kind of democratic organisation development is to organize those who perform the action, the local actors, in practical discourses.4

To obtain this, researchers cannot act as if we are some kind of distant observers of theoretical knowledge. We have to act as one among other kinds of social actors that are engaged in the practical change of working life. What kind of practical discourses may be organized on this mutual basis will depend on a lot of specific circumstances, from the institutional character of working life and work life research, to the personal idiosyncrasies of the participating researchers.

It goes without saying that the relationship between theoretical and practical discourse is not to be organized in a strict instrumental way. In the Scandinavian action research tradition we have, together with the parties of working life and cooperating enterprises, made lots of theoretical and practical efforts to organize practical discourses of a kind that allows for broad participation; in principle everybody (all) concerned shall be invited to participate. Since most people are at job to perform their work, not to perform dialogues, you can imagine that to create a practical discourse that allows for everybody to participate requires quite carefully organized devices. The totality of events, arenas and forums that together form a practical discourse that allow for everybody to have a saying in an OD process, may be compared to the forums of public debate at the societal level, the public sphere. In this sense, the practical discourse is organized as a kind of internal public sphere within the enterprise.

Notwithstanding all the differences between the public sphere of society and the internal public sphere within enterprises, there are some important similarities. The public sphere is a medium for reasoning, and thereby it is a source of legitimacy. Whatever one says, whatever one claims to be true, somebody else can contest it. The decisive factor is not who makes the claim, but the reasons put forth to support it, respectively contest it. And questions of what to do with some phenomenon or problem may turn out to be also a matter of how to understand the problem, what would be a legitimate interpretation of the matter. For these reasons, the public sphere is obviously a 'medium' where theoretical and practical discourses may be linked.

This is also the case with the internal public sphere in enterprises. What to do with some issue is indeed dependent on how to understand it. This implies that also the so-called practitioners from time to time have to undertake some kind of theoretical discourses, in order to come to a better understanding of the issues at stake in their process of practical development. Or, to phrase it otherwise: the practical discourse also has its 'moments' of theoretical discourse.

These moments, and these issues, obviously may be linked to the theoretical discourse of researchers. But what kind of 'input' is required from the theoretical discourses of the research community is not that obvious. Sometimes, some 'piece of knowledge' that is commonplace within the research community is needed. But, it may also be a need for knowledge of a kind that has quite other sources than the field of management and organisation studies: the social sciences in general, as well as the humanities.

Art and literature are also sources of knowledge, especially the kind of knowledge that is required to provide new interpretations of seemingly well-known phenomena. This kind of knowledge may turn out to be quite as actionable as the huge amount of knowledge that is produced by the standard procedures of management and organization studies. Because, as I have argued, the crucial point about creating actionable knowledge is not at all to avoid the creation of general knowledge, neither to dismiss theoretical discourses. The crucial task is to organize 'local orders of discourse' in ways that make it possible for any relevant, whatever general, knowledge to come to work in local settings.

To organize public spheres within private enterprises is just one way to cope with this task.5 As I have tried to demonstrate by the example from Norske Skog and the perspectives presented thereby, this approach indeed makes the researchers involved come closer to the events than is the case by most other research methods. As action researchers we are not just close to the events; we are literally part of the events, we even participate in creating the events. Knowledge of organisational structures normally is required as an integral part of action research methods as methods for generating both new ways of working and new kinds of knowledge. However, the kind of knowledge of organisational structures and dynamics required, in order to support/promote organisational change in processes that are based on extended participation from both management and employees, also comprises a kind of theoretical/practical competence, concerning challenges/questions like:

- which kind of knowledge to apply in the particular situation/context,

- how to apply this knowledge in a way that makes it work in the particular context,

- what kind of new theoretical points or practical devices have to be created in order to make progress in each particular context?

From the experience of coping very specifically in different kinds of action research projects with these kinds of challenges, action researchers contributing to organisational change will gain a lot of knowledge that may also contribute to the discourse on organisations, by writing publications that address this discourse. But still the question is: does this closeness to the events produce a better ability to present a new and better understanding of organisational structures and organisational events? There are action researchers who argue that the closeness to the events, in particular by means of the common responsibility for organisational change between action researchers and practitioners, represents a kind of privileged methodological position compared to other research methods. From this kind of perspective, narratives from action research projects may appear an adequate mode of writing publications. However, also within the community of action researchers a broad spectre of arguments has been presented against this perspective (Greenwood 2002; Gustavsen 2003a; Pålshaugen 1996).

My perspective on what might be the most important kinds of contributions from action research to organisation theory is quite different, as will be seen from the following sections. My perspective is based on a critical diagnosis of certain problematic features that pertain to the discourse on organisations, concerning the style of writing. Therefore, to be able to present my perspective I have to start with some considerations on the emergence and development of organisation theory as an academic discipline (section 4&5). On this basis, my answer to the question of in what way knowledge and experience from within the events may contribute to an improvement of the discourse on organisations will be developed in the final section 6.

4. The genesis of general writings on organisations

During the last century, in particular after WW2, theories of organisation have been proliferating. The more or less explicit rationale of all this theorizing may be stated in rather simple terms: the better we understand organisations, the better organisations we will be able to create. However, after more than half a century of experience with organisation theory as a kind of 'cross-disciplinary discipline', the interrelationship between the features of organisation theory and the organisations of our society is not easy to state in simple terms. In fact, the multitude of experiences of the troublesome relation between organisation theory and what is supposed to be its object of study, organisations, have given rise to an increased interest in making organisation theory (OT) an object of study in itself. This is the explicit purpose of the recent Oxford Handbook in Organization Theory, which is therefore sub-titled 'Meta-theoretical Perspectives'. The object of analysis and debate in this book "is not a set of organisational phenomena, but OT itself (Tsoukas/Knudsen 2003: 1).

This book is of particular interest also because quite a number of those who have been among the outstanding contributors to the development of organisation theory throughout the last century here contribute their theories and perspectives on what kind of phenomenon organisation theory is. Not surprisingly, the fact that OT is a phenomenon that comes to existence by writing is addressed by more than one of the authors. 'The Styles and the Stylists of Organisation Theory' is the topic of an article by B. Czarniawska, and the question of 'whom do we write for' is addressed both by the editors and by G. Burrell. However, the interconnection between these two questions deserves more attention than is given in this book, for reasons I will explain.

My thesis is that the question of 'to whom do we write', or 'whom do we address', is a constituent part of the texts that we write on organisation theory and organisation studies. Not in the sense that there is a limited number of singular addressees to which each specific text is directed. Rather, the point is that there are just not any particular, empirical addressee(s), but some imagined addressees (or group/community of addresses) we have in mind when writing. We write publications to make them public, and what we write and how we write is affected by how we imagine our readers and their reading.

Since the question of whom we write for has been shifting throughout the times, we should begin at the beginning, that is, the beginning of organisation theory as it is presented in this book by W. Starbuck, in his article on 'The Origins of Organization Theory'. He has studied texts on management and organisation from remote times and civilisations and has made an interesting observation:

The history of organization theory contrasts with the history of managerial thought. When people began to compose texts about organized activities, between 2,000 and 3,000 years before the Christian era (BCE), they focused on managerial practices rather than on organizations as such. Several writers proposed general principles for managerial practices before 1000 BCE, so one can say that theories about managing have existed for at least 3,000 years. These writings often said nothing about the organizational contexts in which managing was to occur. When the writers did make statements about organization, they did not generalize. They wrote about specific organizations ... This bifurcated pattern persisted over the next 3,000 years. ... I have been able to find only a few authors who proposed generalizations about organizations as distinctive social systems before the late nineteenth century (Starbuck 2003: 143, 147/148)

The reason for this persistent, bifurcated pattern Starbuck finds by applying a simple perspective of sociology of knowledge: "Evidently, people saw the results of organized activities as being products of actions taken by individuals, so they formulated prescriptions about actions individuals should take" (Starbuck 2003: 148).

But there is more to be said about this 'persistent pattern' of writing than to say it depends on 'a way of seeing'. It depends more literally on a way of talking. As we remember, Starbuck does not make the claim that organisational matters were not written about at all. What he tells us is that these texts describe only one specific organisation and do not make any generalization. Even when such a text states principles, it does not make any suggestion that these principles "also apply to other organisations" (Starbuck 2003: 148). This is obviously not just because the writer does not regard the object of his description as an organisation, it is also due to who is regarded the addressee of the text.

In the case of a Chinese text known as The Officials of Chou, which was written around 1100 BCE and among others contains "job descriptions for the many officials in the king's service, ranging from the prime minister to household servants", the addressees are obviously not other (royal) leaders, but the leaders in this singular 'organisation'. The text is probably written in order to maintain its procedures, rules and structure, in short, its way of working. Both the leaders and the officials may change, but the organisation is to be maintained, thus the specific descriptions are made to serve as specific prescriptions.

This sheds some light on the relationship between the use of generalizations in writing and the users of the writing, the addressees. The context of The Officials of Chou is singular, known both to the writer and the reader, and therefore the descriptions (of jobs, procedures) can work as prescriptions, or even instructions, on how to act. The conditions for making use of the text as work instructions, is mainly knowledge of the specific organisation for which the text is written. For such a text to be of any use in another organisation, this organisation would have to be not just a similar one, but an identical one. Therefore, even when the text states principles of how to act, the question of application of these principles outside the context of which it is written is not addressed.

This kind of writing is, as demonstrated by Starbuck, in great contrast to the writing on principles of leadership. The addressees of these texts are not a limited number of individuals within one singular context, known to both the writer and the reader. The written text addresses readers across specific contexts of action. Thus, the context of use of any of these texts is not identical to all readers; it is a number of different contexts with certain similarities that form the background of both the writing and the reading of texts on leadership. And what is regarded similar across various contexts which are not identical and thus different, is exactly such common aspects which come to be considered general features, of which generalizations can be made in writing.

The question then, is: how is it possible to compare different realities and find out about differences and similarities? By what means, or by what medium, are such different contexts brought in contact with each other? Such questions may give raise to enormous problems of both philosophical and empirical kinds, and I do not pretend to be able to give a full answer. But an unavoidable part of the answer on what made it possible to bring different contexts together and make comparisons, is that ancient leaders had experiences from different contexts, and that the medium to compare these was by means of linguistic communication, mainly oral discussions, gradually supplemented by written texts. Thus, the medium for comparison was, and still is, language.

With this simple fact in mind we may realize more of what is at stake in the 'persistent, bifurcated pattern' which Starbuck discovered. In the ancient times Starbuck has focussed on, writing general statements on leadership obviously made sense, while writing general statements on organisations obviously did not. Why? The answer to this question may make us become more aware of the conditions for writing organisation theory today. Therefore, even though the answer is rather obvious, I will present it some more in detail than otherwise would have been necessary.

The main preconditions for writing on leadership in ancient times, we have to assume, were two: Firstly, that there existed a kind of oral discourse on leadership, which provided the vocabulary required for writing texts. Secondly, that this discourse was a kind of practical discourse, in the sense that the participants who made up this discourse mainly were leaders practicing some kind of leadership, discussing how to perform leadership, on the basis of their personal experience. In such a discourse, to be of common interest for the participants, important common features of performing leadership will have been among the crucial issues. Thereby linguistic expressions that refer to common aspects of leadership are generated, be it in the form of phrases, maxims, concepts or lines of reasoning. In short, a vocabulary for conversations on common aspects of leadership is created.

This makes clear that the existence of a common discourse among people experienced in leadership is the precondition for these texts to be read in a way that makes the reader get a specific understanding of the general terms, which in turn is the condition for the texts to be useful, i.e. applicable in practice. In this sense, this ancient discourse on leadership is indeed a practical discourse: The discussions, writing and reading are undertaken in order to figure out how to perform leadership, not in order to come to a general understanding of what kind of phenomenon leadership is. The texts are written within a practical discourse. It is worth noting that their character of being general does not by itself make these texts constitute any theoretical discourse.

Obviously, there were no people who gathered to form a corresponding discourse on organisational matters or issues at that time and place in history. Thus there was no common vocabulary, no possibilities of communicating on similar experiences across different organisational contexts. Such a discourse did not really take shape until from the late nineteenth century on. The kind of people who formed this emerging discourse were not just leaders and managers, but also engineers, middle management and others who experienced working life from within private firms, public administration etc.. These different kinds of people made various kinds of experiences from different kinds of organisations. Among others a common element was the experience that the way the work was organised affected the performance of those who worked in the organisation.

By discussing these kinds of experiences a certain vocabulary, or rather certain vocabularies, for addressing organisational issues was gradually developed. In these early phases persons with practical experience from organisations, be it private or public ones, were among the main contributors. Y. Shenhav has been doing extensive research on this period in USA, and he argues that around the turn of the former century there

existed a systematic discourse about organizations in the United States. This discourse, that was already categorized in 1912 as 'a smaller sister of sociology, as a science of human nature'6 ... was infused into American sociology 'from below' during the first half of the twentieth century. This engineering/managerial discourse invented the 'organization' as a reified epistemological concept and celebrated the idea that it is worthy of intellectual and academic attention (Shenhav 2003: 186).

One implication of the emergence of this discourse of organisation was that managers became increasingly aware that the performance of leadership was just one of the main 'parameters' or conditions that influenced the performance of the employees. The other main parameter was the organisation, organisational forms and procedures. Hence the need to understand the matters of organisation in order to form the best kind of organisations and organisational performance.

It is well known that among the prominent contributors to organisation theory in this early period, before WW2, we find people highly experienced in management like F. W. Taylor, H. Fayol, C. Bernard and others. But it is also worthwhile noting that a large part of the writings on organisations in this early period was by authors who were less prominent than those just mentioned, but who nevertheless to a large extent were people with their main professional experience from work organisations, both in industry and other sectors, even public administration. Based on their various experiences, these authors did

look for properties common to organizations of all kind. ... they used the plain language of managers, they rarely attempted to compare organisations from different eras, and they focused their thinking on how to make organizations more effective: How should organizations be organized? Their prescriptions resemble 'The Officials of Chou', but unlike the authors of 'The Officials of Chou' they sought to generalize7 across many organizations ... (Starbuck 2003: 167).

We may now see one important reason why it lasted about 3 millennia from the general writings on management to the occurrence of texts making generalizations on organisations. Not until the turn of the former century there had been established something like a discourse on organisation, which provided the vocabulary necessary for writing texts on this issue. And this vocabulary was, just analogous to the ancient vocabulary on leadership, generated within a mainly practical discourse: people who were rich in experience from lots of different organisations performed a discourse on what were the important common features of organisations, in order to make them better. Thereby it became possible to write in general terms on organisations, and to have these writings make sense to the audience of readers.

Until WW2, the discourse on organisation, by which indeed general perspectives and theories of organisation were developed, by and large remained a practical discourse. This is not just because such a large part of the authors was experienced with organisational life, it was also because the large part of this literature was written in order to answer questions of what to do about organisations. Like in the United States, the authors were "searching for general properties of organisations that could lead to prescriptions" (Starbuck 2003: 169). By these efforts those authors created lots of theories, perspectives, and concepts that enriched the vocabulary of organisational discourse. Many general aspects were discussed: the need of functional differentiation, coordination and hierarchy, procedures, relations between line and staff, spans of control, etc.

The texts were written to help create organisations with better, that is, more effective performance. The focus on structure, on organisational forms, became gradually more important: "...a gradual change took place, and documents written during the 1920s began to view organizations as integrated systems and to discuss the structures of this system" (Starbuck 2003: 167), not just because the structure appeared to be the most important single feature/property of organisations. The focus on organisational forms was due to the fact that the addressees of these texts were those who were in charge of forming organisations, of influencing the decisions on what structural form organisations should have: managers, engineers, consultants. The discourse on organisations was indeed a practical discourse.

5. The emergence of a theoretical discourse on organisations, and the consequences for the style of writing organisation theory

Thus, during the first half of the twentieth century, it would not make sense to claim that the discourse on organisations was divided into a practical discourse and a theoretical discourse. This divide did not happen until after WW2. Before WW2 it was usually simply assumed that the writers were experienced in the topics they wrote about. What counted was the content of the book or article, the quality of its reasoning and suggestions. Nobody considered it necessary to give an account for the empirical foundations of the text. But after the WW2, this situation changed:

The late 1940s and early 1950s saw two changes in the character of writings about organization theory ... For one thing, the two streams of thought abut organization - sociological writings about bureaucracy and managerial writings about organizational effectiveness - discovered each other. For another thing, authors began to speak about the empirical bases for their theories (Starbuck 2003: 171).

With the entry of writings based on research projects in the discourse on organisations, two kinds of changes occurred that created new kinds of relations between the discourse of organisations and the phenomena that this discourse is about.

Firstly, the base of direct experiences from organisations as a foundation for writing was narrowed down. This is because the experience from managing organisations and working in organisations was gradually substituted by 'second hand' experiences in the form of empirical data, provided by the expanding number of methods for undertaking empirical studies. These methods, as we know, by and large consist of various techniques for either looking at/listening to people or communicating with them (orally or in writing), in order to construct written fragments labelled and stored as 'data'. Thereby direct, personal experience from organisations as real phenomena is in effect downgraded, and the virtual reality of written fragments claming to represent experience (hence the label 'empirical data') is upgraded, notably, within the emerging research-based discourse on organisations.

Secondly, this narrowing down of experience is 'compensated' by widening considerably the vocabulary of discourse on organisations; not just sociological but also anthropological and psychological concepts, terms and perspectives entered the discourse on organisations. In the wake of these two kinds of changes the discourse on organisations underwent significant transformations, and its divide into one theoretical and one practical discourse took place. The theoretical discourse took on the shape of what we today, like the editors of the above-mentioned Handbook, call 'organisation theory and organisation studies'. The practical discourse per definition remained mainly within working life, but was by no means unaffected by the growth of the theoretical discourse, that is, the establishment of OT as an academic enterprise in universities, business schools and research institutes. Neither were these two discourses totally separated; an increasingly complex pattern of interfaces was an integral part of this development.

The actual complexity of this interface seems to be not fully recognized by organisational theorists. In accordance with a very persistent figure of thought, or manner of thinking, within the communities of social science and research, organisational research included, the relation between the theoretical and practical discourse is considered a hierarchical relation. The theoretical discourse resides over the practical discourse. Against this background it seems that organisational theorists hope that the knowledge of OT sooner or later will move down the hierarchy until it reaches the actors within organisations, as if knowledge might be transferred by some irrigating system.

My characterization is of course exaggerated. Nevertheless, this kind of notion may be some of the reason why there is one significant transformation which seems to have avoided any attention in the writings on the development of organisation theory. What I have in mind is that as the theoretical discourse on organisations, organisation theory, gradually developed after WW2, research based writings on organisation theory increasingly show signs that there are simultaneously two addressees of these texts: The main addressee continue to be the actors in charge of forming organisations within working life, management, engineers, personnel management etc., but in addition also other organisation theorists (or organisation theories) become important addressees.

The reason is, a bit simplified, that with the growth of organisation theories based on empirical research any author also enters into a competition with other, different research approaches. Of course, there had been competing theories also earlier, but with the introduction of research methods and scientific norms of presentation, it became commonly assumed among the competitors that the best scientifically based theory would be the legitimate 'winner'. It was commonly assumed that any winning price for any such theory was just a 'travelling trophy' - the logic of research and science deem this activity to be work in progress.

Thus, to undertake research projects on the basis of empirical methods in order to create a comprehensive, general theory of organisations became regarded the best way to create actionable knowledge of management and organisations. This general perspective became rather soon commonly accepted among the performers of organisation studies. However, there was no corresponding common acceptance of neither methods nor theories to be used. Thus the competition between different approaches and 'schools' of organisation theory and organisation studies about which of them provided the best general theory of organisations was established as the one main horizon of almost all the various research approaches.

This competition between theories became not only the main horizon, but also the main driving force. Since it turned out in practice that any research programme, approach or 'school' in organisation theory was able to cover only certain aspects of the phenomenon of study, each new approach would rather soon be contested by another new approach, which claimed to cover those aspects neglected or let out or undiscovered by the former approach. However, common for each new approach was the ambition of creating a general, comprehensive theory of organisations, being able to cover all main aspects of organisations, ambitions that were not fulfilled, but which paved the way for new, not less ambitious approaches. In the 50s and 60s, this was the logic of the 'rise and fall' of a number of theoretical approaches of organisation studies and organisation theory.

During the 70s and 80s it was gradually acknowledged that the new approaches not necessarily had to replace the prevalent one(s); the concept of scientific progress was supplemented by the concept of pluralism: Different approaches or schools might coexist in a (not too) friendly competition. This 'competition' between research approaches has deeply affected the construction of and the content of organisation theory from the second half of the former century until now. Since long the primary addressee of the writings on organisation theory have become other theorists, other writers, in short, researchers and academicians. At the same time the style of writing has continued to be one which presumes that the real, final addressee are the practitioners 'in charge' in organisations, those who are in a position to form and change organisations. The upholding of this 'double addressee' is, though largely unnoticed, conditioning the kind of writing that still dominates organisation theory.

The rationale of writing organisation theory, 'to create a better understanding of organisations in order to create better organisations' is thus maintained, in spite of the fact that 'a better understanding' no longer refers to the understanding of the practitioners performing their work within organisations, but rather to the understanding of those who are performing the theoretical discourse on organisations, the theorists. In practice, writing in order to improve the 'understanding of organisations' has become writing scientific publications which try to improve the kind of theoretical understanding of organisations that the author regards as the more or less prevalent, and, not a bene: more or less unsatisfying dealt with, within the theoretical discourse on organisations. Thus, this unacknowledged 'double addressee' has led to a style of writing which is based on the (tacit) assumption that a comprehensive, general theory of organisations and organisational phenomena is inherently actionable.

This situation, the contemporary one, is very well reflected in the Handbook of Organization Theory. E.g., in the coexistence of organisation theory as a "Positive Science" (L. Donaldson), as an "Interpretive Science" (M. J. Hatch & D. Yanow), as a "Critical Science" (H. Willmott) and as a "Post-modern Science" (R. Chia). Organisational theorists no longer believe that there will be 'one best theory', nor do they believe in just one direction of theoretical progress. Pluralism has become part of the contemporary 'normal science' in organisation theory. But this pluralism is not very plural as regards theoretical perspectives on the question of practical application of organisation theory. This problem is 'solved' by a certain style of writing, which is based on the unnoticed confusion of the addressees of the writing, as I have sketched above.

The main problem with this style of writing is not that the aim is to create general theories. The problem is the very content of the theories, or rather the lack of content; they tend to be too abstract, too poor in content. The assumption that knowledge based on scientific methods and scientific concepts is inherently actionable has made the writing of organisation theory being imprinted by a peculiar mixture of constraints and freedom: on the one hand the repertoire of research methods is limited by what counts as methods for generating empirical data. This limitation also limits the multitude of kinds of real organisational experience that should have been at work in the process of creating organisation theory. On the other hand, there is in principle no limitation of the new theories and concepts that can be applied as frameworks and perspectives by which the sparse data may be interpreted. By this particular mixture of constraints in methods and freedom of theories that govern the 'normal science' in organisation theory, we today witness an ever growing proliferation of theoretical perspectives being applied to rather tiny empirical data.

6. The contribution from action research to the theoretical discourse on organisations: critical theory

On the basis of my critical diagnosis of the development of contemporary organisation theory, it should now be possible to develop an answer to the question whether, and how, action research may contribute to make it better in the future. If I am right that the main problems of organisation theory adheres to the style of writing in the sense I have described above, it is obvious that the contribution from action research cannot be to write just narratives from the field. Such narratives would easily be read as just another empirical study of organisations. In this way, narratives from action research may as well work to sustain the common style of writing organisation theory. However, it is the very style of writing, the conceptual schemata and the general form of the theoretical frameworks that flourish within organisation theory, which has to be challenged. Thus, the most important kind of contribution from action research to the improvement of organisation theory seems to be to present the kind of general criticism that on the basis of the experience and knowledge from action research appears to be most pertinent. In this final section I will present what I consider to be the main topics of such a general criticism.

To begin with, I have to emphasize one thing: the 'moral' of my description of the split of the discourse on organisations into a practical and a theoretical discourse was not intended to be that the problems of today might be solved by returning to the good, old days when most writings on organisations were based on the personal experience of professional managers etc. In fact, one of the reasons why professional academics and researchers entered into the discourse on organisation after WW2 was that they were quite dissatisfied with the content of the discourse dominated by 'amateurs' who wrote texts in which they tried to formulate general principles of organisations, based on their experience as managers etc.. Having read some of these texts, I find this discontent quite understandable.

This discontent was very clearly expressed by one of the founding fathers of OT as a multidisciplinary discipline, Herbert Simon: "In particular, Simon attacked the idea that principles of organization and management are useful, and he pointed out that every principle seemed to contradict an equally plausible principle. He asserted that a more 'scientific' approach to the study of organizations would eliminate the contradictions" (Starbuck 2003: 171). The title of the article in which he first expressed these ideas is "The Proverbs of Administration" (Simon 1946). His analogy with the theoretical generalizations of the 'amateurs' of organisation theory and proverbs is quite illuminating; not just for how Simon conceived these generalizations, but also for his conception of organisation theory as scientific.

As we know proverbs may be regarded as condensed, general expressions of a multitude of life experience, but as we also know, there are numerous examples of proverbs contradicting each other. The project of erasing contradictions between different general statements, theorems and principles and to create a comprehensive, general theory which consists of empirically based generalizations that do not contradict each other, is based on the assumption that in order to guide action, a theory cannot consist of statements that contradict each other. If a theory contains contradictory statements, how can we know which is true, and which one to act in accordance with? A general theory free of contradiction will solve this problem, and thus work better to guide action. This conception of the usefulness of a general theory was approved not only by organisation theorists, but by many social scientists and social researchers as well, and to a great extent it still is.

What is neglected in this conception is that for any general statement, principle or theoretical knowledge to guide specific action, the person(s) who want(s) to apply it cannot rely on a general theory/knowledge of what situations it applies to. The relation between general knowledge and specific situations goes the other way round: the actor (each person) has to judge in what situation it may be applied. The ability to make such judgments is equal to the power of judgment one has. Proverbs are created on the base of life experience, and they are created in the confidence that most people are sufficiently experienced to have developed a power of judgment strong enough to apply the right kind of proverb to the right kind of situation. The existence of proverbs that, if compared as general statements regardless of any specific, practical context contradict each other, is literally a theoretical problem; it represents no problem for actors who are able to judge what proverb to apply to what situation/context. That is, people with a sensible power of judgment.

Exactly the same goes for the ability to apply the kind of knowledge that is required in the kind of situation or kind of event one is into (at the moment). No theory, however comprehensive, coherent and rich in content can contain a kind of 'complete map' of the situations of its possible relevance for action. Theoretical knowledge has to be put into use, no knowledge is inherently actionable. That is, knowledge has to be put into use within some particular event by somebody partaking in this event. In order to judge what knowledge is suitable to apply in each particular situation it will be necessary to make use of one's power of judgement. As we have commented upon earlier, that is an effort which requires more than invoking one's repertoire of theoretical knowledge of organisations.

The power of judgment is dependent on many kinds of experience and many kinds of knowledge, which cannot be organized within one theoretical perspective where the relations between the various kinds of knowledge and experience are structured in one strict logical order. Making use of the power of judgment goes beyond the realm of logic. Making use of the power of judgment is literally a making: it means to perform an act: the judgement is to be made, and it is always a risky business in the sense that no criteria, no logic or no theory can in beforehand tell what will be the right judgement. A sensible power of judgment is shown by the quality of the judgment(s) made.

This little exercise perhaps reveals the hidden assumption that seems to have motivated the project of creating comprehensive, general organisation theory free of contradictions in order to guide action: Such general theories will/may reduce the dependence on peoples' power of judgment; the theory itself would, if worked out sufficiently comprehensive, logically coherent and empirically validated, contain in itself its own criteria for practical application. Even though the theoretical approaches of today are not as rigidly conceived in this respect as was H. Simon's initial project, we may recognize a very closely related kind of disbelief in, or dissatisfaction with, making the question of practical application of organisation theory being dependent on the power of judgment. This disbelief is of course not articulated explicitly, but we may recognize this belief the practice of writing: organisation theories that claim to satisfy the criteria of a good, reliable and valid scientific text are presented as if they were inherently actionable.

The most common way of imagining this presumed actionability of theories goes like this: The main building blocks of theories are concepts. As we know concepts refer to only general aspects of any phenomenon, and so do also the concepts used in organisation theory. It is usually imagined that for a general theory to be practically applicable in some local context, the more concrete elements of the specific context/situation have to be put into the general theoretical framework in accordance with the conceptual order of the general theory. Thereby the relationship between the general concepts and the specific context is imagined established, and thereby the scientifically based knowledge is considered specific enough to be actionable. When the local actors know what the theory means in their case, they can act upon it.

However, this kind of relationship between the general concepts and the specific context is established only in theory, that is, in the imagination of the authors of these kinds of theories. In the real world it is not possible to put the concrete elements of the local context and the local events into the conceptual framework of the theory; that is possible only in the virtual world of theory, or imagination. As I have described earlier, in practice it goes the other way round: For a theory to become actionable knowledge, the elements of the theoretical framework, the concepts, have to be put into the local context, by means of some kind of communication, some kind of conversation/discourse.

As the meaning of words and concepts is no inherent feature of the term but dependent on the way words are used, and the context within which they are used, no conceptual meaning can simply be transferred into a new discourse, as if it was a fixed entity. The meaning will somehow be reconstructed; it takes on a particular meaning in the particular discourse where the concepts are in use8. Thus, the concepts put into use in a practical discourse may very well turn out to be actionable, but which one and in which way is never to be predicted. What can be predicted, on the basis of experience and knowledge from action research projects, is that only some concepts and certain parts of the theoretical framework will show up as relevant in the practical discourse. Also, as I have reported earlier, knowledge and concepts that are not all part of one particular theory, will have to be applied. This means that the conceptual framework as a whole, as a general theory, no longer really works as a framework. The local reconstruction of a specific understanding of specific organisational phenomena in this sense works as a deconstruction of the general understanding that is represented by organisation theory.

Thus, experience and knowledge from action research does not only subvert the common belief in most organisation theory, that general knowledge, if it is systematically developed and presented in a conceptual order free of inner contradictions, in principle is actionable knowledge. Also the common belief in the need for a theoretical framework in order to undertake organisation studies may be subverted. The need for a conceptual framework is commonly considered necessary because the most important general concepts are regarded as representing the most important general aspects of the phenomena in question, the organisation(s). The conceptual model, the theoretical framework, is regarded nothing less than a general model of the most important features or aspects of the phenomenon of study. The common belief is that without any such conceptual framework, in which the logical order of the relations between the concepts in the framework represents the real order of the relations between the general aspects of the phenomenon, the researcher will be not able to grasp the essential general aspects of the phenomenon of study.

On the basis of my knowledge from the practical discourses of organisational life that we are confronted with in action research projects, I look upon this belief in disbelief. By our participation in practical discourses in numerous action research projects, we regularly experience how little of the vast literature on organisation theory is relevant in efforts of organisational change, of creating better organisations and better organisational performance. This is not only due to the fact that lots of other kinds of knowledge and experience are relevant than knowledge of the main subjects of organisation theory. Quite as much it is due to fact that the quality of the concepts of organisation theory is insufficient. The main problem is not that the concepts are of a general nature; the main problem is simply that they are too poor in content, too abstract.

The concepts that are circulating within the theoretical discourse of organisation theory suffer from the lack of being enriched and developed by direct, personal experience from the practical discourse and the real events that take place within organisations. The theoretical discourse of organisation theory considers itself as self-sufficient with knowledge of organisational phenomena developed by means of the methods of constructing empirical data. In this way the theoretical discourse on organisations literally abstracts bits and pieces from the practical discourse and the organisational events, it is no surprise that the knowledge generated appears to be rather abstract.

In this way empirical studies mainly have the function of giving sense to the general concepts within the theoretical discourse, rather than having the function of giving a richer understanding of the real events that take place in organisations. With the proliferation of new theoretical approaches and concepts, organisation theory has in the last decades developed many different ways to express more or less similar and rather well known aspects of organisational phenomena. To put it bluntly, both to experienced organisational theorists and to practitioners, many attempts to make renewals in organisation theory appear as an exercise in expressing common insights by uncommon terms. Organisation theory thus runs the risk of trivializing itself, like a funny dressed person who appears boring as soon as s/he starts speaking.

What is worse is that organisation theory by this research practice continues to run the risk of mistaking a word for a phenomenon. Since organisational structure is no longer the main issue in organisation theory, Weick's warning about the risk of making "events into structure" is perhaps no longer the most important warning to make. However, the same risk pertains to other features of organisations. There are lots of other aspects than the organisational structure that has become subject of conceptualization, and the awareness of the importance of organisational processes and events has been increasing (Czarniawska 2006).

But also the concepts developed to cover other general aspects of organisational events nevertheless tend to take a structural turn, in the sense that they together make up the structure of the theoretical framework that is the main feature of any theoretical approach of organisation studies. Since any theoretical framework tries to cover the main features and events of organisations, the creation of such frameworks always implies transforming abstracted aspects of events into concepts. Therefore we are permanently confronted with the risk Weick warned us against: to make events into a concrete substance by means of the concepts we use. Thus, it is an open question whether such writing will be an example of taking a word for a phenomenon. It will depend partly on the quality of the text, partly on the quality of the reading. In any case, any concept is so to say referring only to the surface of the real events, or rather: part of the surface. In order to judge whether this is the essential part, as is presumed by the construction of theories, concepts and theoretical frameworks, some kind of confrontation with 'the real thing' is required.

However, as long as the concepts and methods of descriptive organisation studies work to keep organisation theory producing a theoretical discourse that preserves itself from getting comprehensive experience from the practical discourse within organisations, from within the events, such kinds of confrontations are rare, to say the least. The need for comprehensive experience from within the events is substituted by the quest for a comprehensive theoretical framework as a necessary precondition for performing organisation studies and generates organisation theory. But since this framework works to keep the researchers away from any criteria for validation beyond those that are to be found only within the theoretical discourse, any theoretical framework may as well work as a shelter that makes researches partially blind or deaf to grasp the essential aspects of the organisational phenomena they are confronted with, as it may help to grasp them. For this reason, the most apt contribution from action research to the theoretical discourse on organisations seems to be to repeat the question presented at the beginning of this article: Maybe the theoretical framework of organisation theory works to put organisational research 'outside' the real dynamic of organisations?

To those who like myself fear that the right answer to this question would too often be 'yes', it may be of some help to remind that the theoretical framework we find within organisation theory is not upheld by strong theoretical arguments and unquestionable scientific standards. Rather, it is upheld by a practice, that is the practice of writing and the practice of performing organisation studies. This practice has, as we have seen, changed considerably throughout history, and I have implicitly argued that considerable changes are still needed. Neither the earlier practice nor the contemporary practice is governed by any kind of collective or common reason. Like any scientific community, we have no ultimate principle to rely on except for the principle that the scientific discourse is a public discourse. The necessity of writing publications and the freedom of advancing public criticism are the main principles, or mechanisms, by which improvements in the process of knowledge generation may occur. In other words: there is no other principle that warrants the quality (and the truth) of scientific knowledge other than the principle of the scientific discourse being public.

The main reasons for the need of advancing critical perspectives in the contemporary discourse on organisations I have presented throughout this article, by diagnosing some of the main general characteristics of the development of organisation theory. To recapitulate: Firstly, the split of the discourse on organisations in a practical and a theoretical discourse. Secondly, the style of writing that thereby emerged, where organisation theory is written both as if the addressee is some general actor and with other organisation theories as the real addressee. Thirdly, the predominant belief that a comprehensive theory of organisation is inherently actionable, and, last but not least, the thereby connected belief that working out comprehensive theoretical frameworks is a necessary condition for undertaking organisation studies. All these features have to a large extent worked to prevent researchers from using research methods which may give them rich experiences from within the practical discourse of organisations. The result of all this is a discourse on organisations which is rich in theoretical frameworks but rather poor in content.

On the basis of experiences and knowledge from action research where direct involvement in the practical discourse within organisations is an integral part of the research projects, action researchers have a unique potential for working out a deeply needed criticism of the above sketched predominant features of organisation theory. However, this potential is of little value if such criticism remains just part of an oral culture within the community of action researchers. In that case the value of this potential for critique may even be negative in the sense that it works to support prejudices against general theories as such, among action researchers. Thus, in order to take advantage of this potential for critique, the criticism has to be worked out and presented publicly. Only when presented as a publication within the discourse on organisation theory, the value of this potential can be tested, and contested. A criticism which does not make itself publicly available for criticism is no critical theory.

Thus, against the background of the lines of reasoning presented in this article, I will conclude by suggesting in general terms what might be a strategy for action research to contribute to the improvement of organisation theory: a combination of advancing methods for constructive practice within the practical discourse of organisations, and advancing critical theory within the theoretical discourse on organisation theory. The aim of this article has been to present one example of this strategy.

1 For an elaboration of this conceptualizing, cf. Pålshaugen 2004.

2 This conference was arranged by the Norwegian Research Programme "Value Creation 2010" (VC 2010) in order to enhance the possibilities of creating new kinds of collaboration between the parties of working life and work life research milieus on research projects aimed at enterprise development and organisational change. Our collaboration with Norske Skog is part of this research programme. For an overview of the research conducted within this programme, see Gustavsen 2003b.

3 For a more comprehensive account on this case - both from a practical and theoretical perspective - see Amble/Palshaugen 2005.

4 Practical discourses are by no means democratic in themselves. Numerous specific questions concerning unequal distribution of power, differences in communicative competence etc. have to be coped with in the design and performance of practical discourses. For one example of how to cope with these questions, see Pålshaugen 1998.

5 For a more elaborated and historically situated presentation of this topic, cf. Pålshaugen 2002.

6 Shenhav quotes from Engineering Magazine, January 1912, p. 481

7 My italics ØP

8 For elaborations of this argument, see Gustavsen 2001; Pålshaugen 2001.



Amble, N. / Pålshaugen, Ø. (2005): Bedriftsutvikling som samspill mellom teori og praksis: Hvem kan hva, om hvordan? (Paper presented at Thue HSS05-conference, Tønsberg June 1-3.

Burrell, G. (2003): The Future of Organization Theory: Prospects and Limitations. In: Tsoukas/Knudsen 2003.

Czarmawska, B. (2003): The Styles and the Stylists of Organization Theory. In: Tsoukas/Knudsen 2003.

Czarniawska, B. (ed.) (2006): Organization Theory. Volume 1: Central Topics Cheltenham: Edward Elgar Publishing.

Greenwood, D. (2002): Action Research: Unfulfilled promises and unmet Challenges. In: Concepts and Transformation, 7(2): 117-139.

Gustavsen, B. (2001): Theory and practice: The Mediating Discourse. In: P. Reason/H. Bradbury (eds.) Handbook of Action Research: Participative Inquiry and Practice. London: Sage Publications.

Gustavsen, B. (2003a): Action Research and the Problem of the Single Case. In: Concepts and Transformation, 8(1): 93-99.

Gustavsen, B. (2003b): Bedrifter og regionale partnerskap. Erfaringer fra verdiskaping 2010. (Enterprises and Regional Partnerships. Experiences from Value Creation 2010). Oslo: Norwegian Research Council.

Jones, C./Munro, R. (2005): Contemporary Organization Theory. Oxford: Blackwell Publishing/The Sociological Review.

Pålshaugen, Ø. (1996): This is not the whole story - on the demand for scientific reports from action research. In: Gustavsen, B./Toulmin, S. (eds.) (1996): Beyond Theory: Changing organizations through participation. The Hague: John Benjamins: 137-158.

Pålshaugen, Ø. (1998): The End of Organization Theory? Amsterdam: John Benjamins Publishing Company.

Pålshaugen, Ø. (2001): The Use of Words: Improving Enterprises by Improving their Conversation. In: P. Reason,/H. Bradbury (eds.) Handbook of Action Research: Participative Inquiry and Practice. London: Sage Publications.

Pålshaugen, Ø. (2002): Discourse Democracy at Work: On Public Spheres m Private Enterprises. In: Concepts and Transformation, 7(2): 141-192.

Pålshaugen, Ø. (2004): How to do Things with Words? Towards a Linguistic Turn in Action Research. In: Concepts and Transformation, 9(2): 181-203.

Shenhav, Y. (2003): The Historical and Epistemological Foundations of Organization Theory: Fusing Sociological Theory with Engineering Discourse. In: Tsoukas/Knudsen, C. (2003).

Simon, H. (1946): The Proverbs of Administration. In: Public Administration Review, 6: 53-67.

Starbuck, W. (2003): The Origins of Organization Theory. In: Tsoukas/ Knudsen (2003).

Tsoukas, H./Knudsen, C. (2003): The Oxford Handbook on Organization Theory. Oxford: Oxford University Press.

Weick, K. (1975): Making Sense of Organizations. Oxford: Blackwell Publishing.

[Author Affiliation]

About the author:

Dr. Øyvind Pålshaugen is Research Director at the Work Research Institute, Oslo. His latest publications in English are: "Reading and Writing as Performing Arts - at Work." In: Göranzon, B./Hammarén, M./Ennals, R.: Dialogue, Skill/Tacit Knowledge. London: John Wiley/Sons, 2006, 216-228; "How to do things with words: Towards a linguistic turn in action research?" In: Concepts and transformation, (2004), 9(2): 181-203.

Author's address

Work Research Institute

Pb. 6954 St.; Olavs Plass

0130 Oslo, Norway



Author Advanced search


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.