The struggle to redefine the relationship between 'knowledge' and 'action' in the academy: some reflections on action research.


John Elliott, Centre for Applied Research in Education, School of Education and Professional Development, University of East Anglia



Action research and the theory-practice relationship


'Action-research might be defined as 'the Study of a social situation with a view to improving the quality of action within it' (Elliott 1991 p 69).


This definition appeared in my book 'Action Research for Educational Change' (1991) and is widely cited in books and papers on action research.  Rather than feeling pleased about this, I find myself annoyed and irritated.  Why?  Because I feel that the authors neglect my attempts to redefine the relationship between theory and practice in terms of the idea of action research.  At times they appear to be using my definition to place a tight boundary between action research aimed at the improvement of practice and research aimed at the construction of theory.  The drawing of such a tight boundary is often based on the assumption that the practical knowledge which stems from action research is non-theoretical in character because its value is entirely instrumental to the task of improving practice as a means to an end.  Such an assumption implies that the pursuit of practical knowledge through action research is for the sake of practical goals that can be defined independently and in advance of the action research process, whereas research aimed at the construction of theory is the pursuit of knowledge for its own sake.  Conceived in such instrumental terms, practical knowledge has no value in itself, and is set against theoretical knowledge regarded by those who pursue it as valuable in itself. My own work was being selectively appropriated to legitimate a conception of action research which privileged practice over theory, whereas I had seen it as an attempt to redefine the relationship between theory and practice in a way which dissolved the dualism.


In the late '90s, I directed a study of action research carried out in the context of post-graduate courses for teachers within the UK, and discovered that it was predominantly conceived inside academic institutions as the production of instrumental knowledge aimed at underpinning improvements in practice in schools and other educational organisations (see Elliott, MacLure & Sarland 1996).


One obstacle to dissolving the dualism between theory and practice is the idea of a 'theory' as a generalisable representation of events and occurrences.  From such a standpoint, theory generation implies a large-scale study of samples and the exclusion of small-scale studies of particular events and situations.  Hence my definition of action research as 'the study of a social situation with a view to improving the quality of action within it' will be read as an account of a form of small-scale research carried out in particular settings, such as a single classroom or a school, with a view to generating a highly particularised and therefore non-theoretical representation of action.


Earlier in my book 'Action Research for Educational Change', I defined action research in similar terms to the above, but said rather more about the relationship between its practical aim and the production of knowledge.


The fundamental aim of action research is to improve practice rather than to produce knowledge.  The production and utilisation of knowledge is subordinate to, and conditioned by, this fundamental aim (p.49).


As I shall explain more fully later, I was trying to signify the primacy of the practical standpoint as a context for knowledge generation.  I was saying that in the process of action research, knowledge is produced and used in the process of improving practice.  It is constituted by the intention of an agent to change a situation, an intention that is continuously modified in the course of action as the agent's knowledge of the situation develops.  I certainly did not wish to imply that the production and use of knowledge is simply a means to the realisation of a practical intention that can be formed independently and prior to it.  Or that this kind of knowledge would have no value in itself and lack any theoretical significance.  However, I see now that my words can be read in a way that implies a privileging of practice over theory.  Just as to privilege theory over practice implies the exclusion of the practical standpoint, so to privilege practice over theory excludes the theoretical standpoint.


Action research that privileges practice over theory does not dissolve the theory-practice dualism by linking theory to practice.  It simply excludes the theoretical standpoint.  In doing so, it is shaped by the same assumptions which shape forms of educational research that privilege theory over practice; namely, that 'theory' consists of generalisable representations of events, and is generated by activities that in themselves are dissociated from the practical intentions of human agents.  In failing to challenge these assumptions, much of what counts as action research in the field of education fails to dissolve the dualism between theory and practice.  It simply sets up a tension inside the academy with those forms of educational research that privilege 'theory'.  Educational action research is pitted against educational science, and as such confined to a lowly status in the academic hierarchy of knowledge as a minor 'sub-discipline' in the field of educational research. 


The shared assumptions outlined above positively shape the conduct of educational science whereas they negatively shape the conduct of educational action research.  They effectively exclude action research from the domain of public knowledge and confine it to the domain of private knowledge.  In terms of these assumptions, public knowledge is defined from a standpoint which privileges theory over practices.  From this standpoint, what counts as public knowledge is determined by considerations concerning the validity and truth of theoretical propositions rather than considerations concerning their practical usefulness.  The latter may be important to address but they are extrinsic to the activities of knowledge production.  In the UK, educational researchers are being asked to address the relevance of their research to potential users before they design it, and to play a more active role in disseminating their findings to the public.  Although researchers may regard such considerations as important they are viewed as quite distinct from methodological considerations about the conduct of the research itself. 


What counts as public knowledge generally determines what gets published.  Academics who wish to support action research with teachers and other professional practitioners (eg nurses and social workers) tend in the main to publish accounts of the research process and methodology.  The knowledge outcomes are often not deemed to be of sufficient status to report and find acceptance in prestigious academic publications.  Academic action researchers tend to find themselves marginal players in the educational research establishment.  Most of them go along with this.  They compensate by identifying with communities of practitioners and may acquire the status of 'big fish' in the small action research pool inside the academy, but they leave the domain of educational research essentially intact and unchallenged. 


Currently the most influential challenge to this domain inside the academy is stemming from the ideas of poststructuralist thinkers like Derrida, Lacan, Lyotard and Foucault (see Belsey 2002).  From the perspective of poststructuralist educational researchers, such as Stronach and MacLure (1997), inasmuch as the idea of action research privileges practice over theory it is trapped in the patterns of dualistic thinking that characterise the western tradition of enlightenment thought established by the philosophy of Descartes.  For example, MacLure (1995) has applied the poststructuralist methodology of 'deconstruction' to the texts created by action researchers in the field of education.  I will now examine the poststructuralist challenge to enlightenment thinking with a view to asking what its implications are for the theory-practice problem and the idea of action research as a resolution of this problem.  



The poststructuralist challenge and the theory-practice relationship


It is often assumed that theorising is a mental activity and action a physical activity.  In this mind-body dualism resides the problem of theory and practice.  From a theoretical standpoint the 'self' is a thinking subject that construes the world as an object of contemplation rather than an object of change.  Descartes' 'Cogito ergo sum' (see 1968) established 'the self' as a substance whose essence is thinking and therefore the primacy of the theoretical over the practical standpoint.  From the standpoint of the 'Cogito', reasons for action have their source outside the context of the practical affairs of everyday life in the contemplative knowledge of the 'thinking subject'.  Such knowledge can therefore be applied to practice but not derived from it.  The 'Cogito' has defined the relationship between theory and practice in the western enlightenment tradition and shaped the process of knowledge production within the academy.  In doing so it challenged traditional authority on matters of belief and constituted a declaration of independence.  As the Scottish philosopher John Macmurray (1957 pp.75) explained, if to think is my essential nature then 'I have the right and the duty to think for myself, and to refuse to accept any authority other than my own reason as a guarantor of truth.' This logic was radically challenged by Macmurray himself as well as by the poststructuralist and postmodern thinkers on the European continent during the latter half of the 20th Century.


Poststructuralist thinkers elaborated on the work of Saussure (1916, trans 1974) and brought the idea of the substantial self whose essence is thinking into question, and along with it the idea of reason as a guarantor of truth.  According to Saussure, 'meanings' such as theories about the world do not originate from a 'thinking self'.  The latter is a product of the meanings individuals learn from their culture, and that originate in its symbolic systems or discourses.  The words and other symbols that make up a language do not refer to meanings that exist outside the language itself.  They neither represent an objective order of things in the world or the ideas of a thinker that exists independently of their use within the culture.  Meaning resides in the sign, not beyond it.  It is differential rather than referential (see Belsey, 2002, p.10) in the sense that it is culturally differentiated and has no existence beyond the words and symbols that signify it.


Poststructuralist thought deconstructed a conception of theoretical knowledge in terms of a thinking subject, construed as the 'essential self', contemplating independently existing objects in the external world.  If the thinking subject is the effect of learning the trajectories of meaning embedded in the symbolic systems of the culture, then it does not exist as an unconditioned consciousness.  The subject is decentered as the origin of thought.  It thinks only what it is permitted to think within the culture it is conditioned by.  The world it 'knows' is therefore a culturally differentiated one rather than an objective world that exists independently of the knower.  One cannot even talk intelligibly of the decentered subject possessing personal knowledge  for this presupposes a culturally unconditioned consciousness or 'self'.  If the objects of knowledge are culturally differentiated and the knower is the effect of culture, then individuals are not in a position to construct purely personal knowledge.  What they believe is always what their culture permits.  Poststructuralism, through its method of deconstruction, dissolves the binary opposition enshrined in Descartes' Cogito between "the knowing subject in here and the objects of its knowledge out there" (see Belsey 2000, pp.72-73). 


Foucault in particular pointed out the implications of this decentered vision of the subject for the way power operates in society (see 1979a & 1979b).  Learning and maintaining the ways of thinking about the world differentiated by the culture, its theoretical and normative discourses, involves submission to the authorities responsible for their transmission and maintenance.  For Foucault all social relations connecting the individual to social institutions are relations of power.  Power is not a thing some individuals have and others do not, that can be gained or lost.  Rather it defines the relation between all individuals and their culture, including those authorities who are responsible for the transmission and maintenance of that culture.  The latter exert power in their relations with others by virtue of their own compliance to the culturally differentiated meanings circulating within the society.  According to Foucault, this relational conception of power implies the possibility of resistance.  Individuals can always refuse to conform, although usually at a price, and create reverse discourses to maintain their resistance to the dominant ones operating in the society.  Power relations are a site of struggle and conflict.  One might indeed interpret the action research movement in such Foucaultian terms as a reverse discourse of resistance to the prevailing discourse of research in the academy; namely one which privileges theoretical knowledge over practice. 


From the poststructuralist perspective, 'theories' are not a rational foundation for ordering practical affairs.  In learning to apply them to our practices, we are not grounding those practices in objective truths about the objects of our experience, but securing their compliance with culturally differentiated systems of meaning that tell us what to think about what we are doing.  Theoretical discourses, understood as systems of culturally differentiated meanings circulating in society, constitute resources for exerting epistemic sovereignty over our practical thinking.  The increasingly policy-driven 'evidence-based practice' movement in the UK (see Hargreaves 1997) that holds professional practitioners (eg doctors, nurses, social workers and teachers) accountable for the extent to which they ground their practices in research evidence, is an attempt by the state to get them to base their practical judgements and decisions on the generalisable representations of good practice that are produced by research.  From a poststructuralist point of view this movement can be interpreted as an indirect and 'soft' attempt to exert a form of epistemic sovereignty over the practical thinking of practitioners in the guise of fostering rational practices.


If, in applying theory to practice, social practitioners such as teachers are managing their own compliance with culturally determined systems of meaning, how are we to understand the practices shaped by this process?  Descartes' 'Cogito' assumes a sharp division between mind and body.  Whereas the thinking and reasoning mind is the essence of the self, the body is simply an organism it possesses (see Belsey 2000 p.66).  When left to respond to its environment on the basis of its own physiological make-up, the movements of the body are entirely independent of the reasoning activities of the mind.  However, the thinking and reasoning mind can exert a measure of control over the physical movements of the body as a means of achieving practical ends that transcend the survival needs of the organism.  From the standpoint of the 'Cogito', the physical movements of the body (behaviour) are transformed into the practices of a human agent (actions) by the capacity of the mind to impose some form of rational order on them.  The poststructuralist challenge to the 'Cogito' nullifies this account of social practices as the effect of rational human agents on the movements of the body and construes social practices as reactions on the part of the human organisms to stimuli in the cultural environment, motivated by their survival needs.  Such reactions will involve consciousness but it will take a different form from consciousness conceived in terms of an agent having reasons for action.  As Macmurray (1957 p.167) points out, conscious reactions to environmental stimuli stem from motives connected to the organisms survival needs, rather than reasons for action.  The initiative for such behaviour lies with the stimulus as opposed to a human agent, whereas the initiative for action lies with an agent who determines it in the light of their knowledge.  From the perspective of poststructuralist theory, social practices are conceived in terms of adaptive responses on the part of human organisms to cultural stimuli rather than in terms of self-initiated actions.  Viewed in such terms, the activity of applying theory to practice depicts not so much the process by which human agents rationally determine their actions in the world, as the process by which human organisms consciously adapt their behaviour in response to cultural stimuli.


The poststructuralist challenge, as I have argued, acknowledges the possibility of resistance to the cultural conditioning it depicts.  Human beings can inhibit the tendency to adapt to their cultural environment in the required ways but only at the risk of their survival.  They can transgress and disrupt hegemonic discourses and even establish reverse discourses .  However, might such resistances be simply interpreted as negative reactions to cultural stimuli - failures on the part of certain human organisms to adapt appropriately to the prevailing hegemonic discourses within the cultural environment - rather than forming a basis for free action?  I shall return to this question a little later.



Hannah Arendt and the philosophy of action


It is interesting to look at the view of social practice implicit in postmodern deconstructions of the prevailing discourses in western societies in the light of Hannah Arendt's account of the The Human Condition (1958).  As Canovan (1974 p.54) points out, Arendt focuses her philosophy on describing and evaluating the various forms of human activity, rather than focusing, like most western philosophers have done, on evaluating the products of human thought.  Human activity she claimed had not been sufficiently thought about and " its modes not clearly articulated" (Canovan p.54).  I would argue that such philosophical neglect also extends to poststructuralist thinkers.  Their deconstructions of western enlightenment thought appear to leave us with a view of social practices as forms of cultural conditioning, but they are less than clear about the extent to which alternative modes of activity are possible. 


Arendt distinguishes three basic modes of human activity: Labour, Work and Action.   'Labour' is activity dictated by what is required to sustain life.  It is basically life lived under the domination of biological necessity, although Arendt reluctantly acknowledges that in the modern world what is experienced as necessary to sustain life has been extended to cover the consumption of material goods that go beyond the basic necessities of living (see Coulter 2002 p.195).  Activities of labour involve endless repetition.  They are not directed to some end determined by an agent.  They focus on means rather than ends.  If labour has an 'end' it is simply the perpetuation of life, the successful adaptation of human organisms to their environment, in an endless cycle.  The poststructuralist perspective on social practices in western societies appears to render them  predominantly activities of 'Labour' in the Arendtian sense of this term.


'Work', according to Arendt, involves the creation of enduring objects or artefacts for use rather than consumption to satisfy basic needs (see Canovan 1974 p.56 & Coulter 2002 p.197).  Unlike 'Labour' such activities have a beginning and a finite ending consciously determined by the workers themselves.  Moreover, workers deploy their particular talents and abilities to create their 'works'.  'Work', in the Arendtian sense of the term, calls forth the generative capacities of human beings and, in doing so, as Canavan (1974 p.56) points out, "is characteristically human as labour is not." From an Arendtian perspective theories or ideas can be regarded as the products of human work.  They form part of a cultural environment that human beings create for themselves.  Once created, cultural artefacts like theories and ideas stand over against human beings to define their world.  Poststructuralist theory only leaves space for conceiving culture as that which stands over against human beings.  From this point of view 'the self' is an effect rather than an originator of culture.  In destroying 'the self' conceived as a thinking subject passively mirroring an objective world from a contemplative standpoint, poststructuralist theory has difficulty in conceiving of any location for 'the self' other than as an effect of culture.  By focusing on human activity and its distinct modes, rather than thinking as such, Arendt is able to explore alternative locations for 'the self' to those of the purely intellectual standpoint and that of an organism reacting to an environment that is set over and against it.


Arendt's third mode of human activity is that of 'action', a category she deploys to vindicate her belief in human freedom (see Canovan 1974 p.58, Coulter 2002 pp.198-203).  'Action' involves initiating change in a social situation to bring about something new in the web of social relationships that constitute it.  The consequences of 'action' for the agent and those effected by them, where they will lead, cannot be entirely foreseen in advance.  'Action' therefore becomes a matter of continuous negotiation with others through the construction in process of 'transient accounts' as it unfolds in the process.  The full story of 'action' can only be pieced together after the event. 


Since for Arendt 'action' is inextricably linked to communication with others considered as equals, it occurs in public rather than private space, which she regards as the realm of freedom.  In this sense it is intrinsically 'political', and is not to be confused with the political organisations human beings establish for the purpose of perpetuating their natural biological needs.  The sphere of 'action' transcends the hierarchical or sovereign relation between governments and their subjects (Canovan 1974 p.68). 


In the activity of 'labour', human beings are bound by biological necessities and therefore do not engage in them freely.  Even in the activity of 'work' their freedom is restricted by the object it aims to create.  It is only in 'action' - an activity that changes a human situation by initiating something new - that human beings experience unconstrained freedom.  This is because in 'action', in exercising agency to effect change, human beings reveal their unique individuality to themselves and others.  This is not 'a self' that they are aware of prior to acting.  Human beings learn who they are from their 'actions' in the human world (see Canovan 1974 p. 59).  From an Arendtian perspective 'the self' is located in its 'actions' and the experience of agency which accompanies them. 


Since for Arendt 'action' is always carried out in the company of others conceived as free and equal individuals it possesses the twin qualities of plurality and natality.  In 'action' the agent takes into account the unique points of view that others hold towards the situation in question.  This is not the same as acting on the basis of a negotiated consensus.  In 'action' the agent reveals his or her own distinctive view of the situation, but it is developed in communication with others and accommodates or 'invoices' (my term) their own distinctive outlooks.  It is in this sense that Arendt regards 'action' as plural.  The more an agent accommodates the plural voices of others, the more his or her activity constitutes 'action'.   The concept of natality as a quality of action is used by Arendt to contrast 'action' with mere role governed behaviour.  In 'action' conditions are created that enable the agent and others to reveal their individuality and uniqueness by starting something new and, in doing so, to transcend what is merely required of them in their roles in life.  If 'action' has an aim, it is to enlarge the space in which human beings can relate to each other as unique individuals in the situation.  Such an aim is not the intention to produce an outcome or result, but a value built into the process of action itself.


In articulating these distinctive modes of human activities Arendt perhaps achieves what poststructuralist theory fails to; namely, an alternative social location to the 'Cogito' for the existence of 'the self', other than as a mere effect of culture.  For Arendt 'the self' only exists in 'action'.  However, one might argue that the possibility of 'action' in her sense of the term is what Foucault alludes to when he talks about resistance  and the struggle of power, and indeed what Derrida (1995) is attempting to articulate when exploring the possibility of an ethics of deconstruction  in his later work (see Belsey 2002 p.90).  The fact that we live in a culturally differentiated world does not exonerate us, Derrida argues, from the responsibility to acknowledge this in the way we live.  Such an acknowledgement may leave no certain foundations for living, but it does leave what he calls 'messianicity', not the hope of realising some utopean or fixed vision of the future but of a different future (see Belsey p.91).  Within such a postmodern 'acknowledgement' of the possibility of new beginnings for human beings lies the 'seeds' of an Arendtian view of 'action' and 'the self' as agent.


I am struck by the parallels between Arendt's account of action and my own account of 'educational action research'.  Interestingly Coulter (2002 pp.189-206), drawing on Arendt's categories, finds few examples of 'action' research reported in his review of papers published in the Educational Action Research Journal compared with 'labour' and 'work' research.


I have always stressed the importance of viewing 'education' as an activity directed by process values rather than objectives which refer to extrinsic outcomes of the activity.  Also I have attempted to locate action research in the context of teachers attempts to effect changes in the conditions governing life in classrooms and schools for themselves and their students.  Again, in researching educational practice to effect change I have argued that teachers and their collaborators should gather multiple perspectives on the situation in question from their colleagues, students and even parents in the form of triangulation data.  Finally, the value Arendt places on 'action' in particular human situations, as the context in which human beings realise their freedom and dignity, makes her sceptical about the value of sociological theory couched in the form of generalisable representations of events.  She views such 'representations' as potential devices for social control and centralising power within the state.  I have argued, consistently with Arendt's position, that action researchers may use such 'representations' as resources to inform their understanding of particular aspects of the situation they face as agents of change, but they should not treat them as 'law-like' generalisations which offer firm prescriptions for what to do.  They need to be integrated into a more personal holistic understanding of the situation forged by the agents of change themselves in the course of 'action'.  We may refer to such understanding as a theory of the situation.


It is to the articulation of such a conception of 'theory', one that is largely hidden from the poststructuralist thinker's gaze, that I shall now turn in the next section.  In doing so, I will draw heavily on John Macmurray's  'The Self as Agent' (1957).  His standpoint on the location of 'the self' in action is remarkably consistent with Arendt's philosophy of action.



Theorising from the standpoint of action


In this section, I will argue that action research need not exclude the development of a theoretical representation of action, albeit a highly particularised one.  One can provide a meaningful account of action research as a process of theorising about a practical situation.  This will involve challenging the assumptions that the term 'theory' exclusively refers to generalisable representations of events, which can only be produced under conditions that are dissociated from the intentions of agents to effect change in practical situations.  In challenging these assumptions, I hope to demonstrate that improving the quality of action in such situations involves the development of theory.  I have elsewhere tended to use the term 'situational understanding' (see Elliott 1993) to demarcate the theoretical outcomes of action research from theory construed as generalisable representations of events and occurrences. 


My account of action research includes rather than excludes theoretical activity as an aspect of the practical.  In doing so it dissolves the dualism between theory and practice.  Few have articulated the position I shall argue for better than Macmurray.  I will begin with the following extract from 'The Self as Agent':


Action---  involves knowledge as its negative aspect.  The carrying out of a practical intention therefore involves a development of knowledge – or if you will, a continuous modification in the representation of the Other- as its negative aspect.  This indeed is the primary source of that knowledge which comes unsought with the growth of experience' (p.179).


Here the use of the term 'negative' to refer to an aspect of action should not be construed as an undesirable characteristic to be excluded from action.  For Macmurray, 'Practical activity includes theoretical activity, of necessity in its constitution' (p.180).  The latter therefore is secondary to the primacy of practical activity and derivative from it.  It is in this sense that it constitutes the negative aspect of action.  This in no way implies that knowledge is simply instrumental to action that can be defined independently of it.  Macmurray defines 'action' as 'a unity of movement and knowledge' (p.128).  Therefore, he argues, 'Knowledge is that in my action which makes it an action and not a blind activity' (p.129). 


Donald Schon's idea of 'reflection-in-action' echoes Macmurray's account of knowledge in action, although his influential book 'The Reflective Practitioner' (1983) makes no reference to Macmurray's work.  However, Macmurray's account of the growth of 'knowledge-in-action' as depicted above does not in itself add up to an account of action research.  What is missing is any reference to the intention to seek knowledge of a situation through systematic and self-conscious inquiry (which bears some resemblances to Schon's idea of 'reflection-on-action').  Since this intention must be viewed as the negative aspect of a broader practical intention to change a situation, it would imply that the action undertaken to effect change was developed systematically and self-consciously.  Action research may be viewed as a systematic form of action in which the theoretical intention to 'modify the representation of the Other', to use Macmurray's terms, arises as the negative aspect of a positive intention to systematically and self-consciously bring about some change in 'the Other', understood as a practical situation for an agent.  From this perspective, it is inappropriate to treat educational action research as merely a minor sub-discipline within a broader domain of educational research.  It implies a radical reconceptualisation of the domain itself. 


Such a position would assert the primacy of the practical and embrace the proposition 'I act therefore I am'.  This implies, as Macmurray argues in 'The Self as Agent', that the self exists only as an agent in a practical situation, who acts with the intention of changing it in some respect.  Can we talk sensibly about theorising from the standpoint of practice as opposed to the intellectual standpoint of the 'Cogito'?  Like Macmurray (p.85) I believe we can.  Indeed the idea of action research embraces this belief (see Elliott & Adelman, c 1996).


To reflect about the world from the purely intellectual standpoint of the 'Cogito' excludes any reference to the self as an agent in action intent on changing the world, since this standpoint presumes that the self is the substance of a mind that thinks about the world independently of any action to change it.  Macmurray succinctly summarises the ideal of this intellectual mode of reflection, one which still shapes our educational system in the west and what counts as research in the academy. 


---a pure activity of thought which is cool, passionless and completely disinterested, seeking truth for its own sake, with no eye to the practical advantage for the seeker or for anyone else (p.192).


It is impossible for the knowledge produced by this type of reflection to make any direct link with the experience of those who want to effect change in the world.  Any link to the action context must be indirectly determined by agents.  Macmurray (p.192-193) argues that since the intellectual mode of reflection suppresses any feelings the observer of a situation may have towards it, and abstracts features in it which make no reference to the practical valuations of participants as they seek to effect change in it, the knowledge produced can have no practical value in itself other than as a means to an end.  From the practical standpoint the knowledge yielded by the intellectual standpoint can only have instrumental significance at best.  It is always knowledge of the World-as-means and takes the form of generalised representations of facts about the world in the form of 'formulae which express the recurrent patterns of continuance in experience' (p.198).   If Macmurray is correct, then we cannot argue that the Knowledge generated from the intellectual standpoint in the academy is useless knowledge if agents can find a use for it in deciding on the means they will adopt to realise their intentions.  However, if one accepts the postmodern critique that the intellectual standpoint masks a will to power and that the 'knowledge' it produces invariably serves the interests of those who wish to coerce and control the activities of others, then one might question its usefulness to ethical agents like teachers who wish to effect change in ways which respect the agency of their students.  See for example my analysis of the control values that shape much of what counts as 'school effectiveness' research (Elliott 1996).


Macmurray contrasts the intellectual mode of reflection with the emotional mode.  In the latter mode, although reflection involves a suspension of action it adopts the standpoint of the agent and proceeds 'as though we were in action'(p.86).  In emotional reflection, adopting the practical standpoint does not exclude the theoretical.  Since it is this mode of theoretical reflection which lies at the heart of the action research process (see Dadds 1995), let me now summarise Macmurray's account of it (pp.198-202).


1.                  When reflection proceeds as though we were in action it does not abstract from the agent's feelings about the situation.  Action is motivated by a feeling of dissatisfaction with a situation and terminated when the agent feels satisfied that the situation has been improved.  Reflection involves understanding what makes the situation an unsatisfactory one for the agent, discriminating the possibilities of action in it, and selecting one of these possibilities for realisation in action.  Valuation is integral to this mode of reflection.  There is a unity of understanding the situation and the valuation of it (see also O'Hanlon 2002).  As Macmurray puts it, 'The world is known primarily as a system of possibilities of action' (p.191).  Valuation and Knowledge are the positive and negative aspects of forming and sustaining an intention to change a situation from an unsatisfactory to a satisfactory state.  Without them action would be impossible, and in some situations they require a prolonged period when action is suspended for the sake of reflection about the situation from the standpoint of the agent.


2.                  Emotional reflection seeks to determine a situation as an end in itself.  In constructing a representation of a possibility for realisation in action, it expresses a valuation of what is represented as something to be enjoyed for its own sake and not for the sake of accomplishing some further objective.  Such a representation will constitute an image of a particular situation yet to be realised.  Emotional reflection therefore moves towards a greater particularisation of the representation of the possibility of action (see, for examples in the context of teacher-based action research, Elliott & MacDonald 1975).  This contrasts with the intellectual mode of reflection which seeks generalisable representations of the events and occurrences it selects for attention.  It constructs knowledge scientifically.  Emotional reflection constructs knowledge aesthetically.  Both are activities of knowing and forms of research.  Within the intellectual mode of reflection 'theory' refers to generalisable representations of the world while within the emotional mode it refers to a representation of a possibility for realisation in action within a particular situation.  However, this does not rule out the discernment of similarities as well as differences through a comparison of cases.  Such discernment will take the form of general insights into the problems of effecting change in relation to a practice such as teaching.  Action research does not rule out the development of overlapping theories that yield shared insights into the possibilities for action (see, for example, Ebbutt & Elliott 1985).



Concluding remarks


Action research resolves the theory-practice problem by theorising from the standpoint of the agent in a situation s(he) feels to be unsatisfactory.  It need not simply involve the agent who wants to effect the change.  Educational researchers in the academy can collaborate with an educational agent by adopting his/her practical standpoint as though they were in the action context.  Educational action research need not be exclusively practitioner research.  The fact that it is so often construed as such by educational researchers, suggests that they are viewing it as a low level, non-theoretical activity from an intellectual standpoint. 


As an emotional mode of systematic reflection, educational action research constitutes an art rather than a science and constructs knowledge aesthetically in unity with the activity of valuation.  However, this does not make it any less theoretical.


So how can one explain the resistance in the academy to educational action research?  I can only conclude that it is a resistance to educational change effected by teachers.  The widespread involvement of teachers as active agents in changing educational situations would reduce the power exerted by academic researchers - perhaps on behalf of the centralising power of the state - over what is to count as knowledge about their practice.  This because theoretical knowledge from the standpoint of educational action is meaningless and valueless if it cannot be validated in action as knowledge of the aims of education, conceived as possibilities for action in a particular situation.


In discussing Arendt's distinction between 'Action' and 'Making', Joseph Dunne (1993 p.89-90) highlights her concern about the extent to which the products of 'making' in the sphere of science and technology were increasingly deployed as standards of technocratic efficiency to shape human behaviour.  Through her eyes, he points out, the passive adaptation of citizens to the products of science and technology leads to an increasing intolerance of 'action'.  This, in my experience, is precisely what is happening with respect to the teaching profession.  Governments hold teachers and other public service professionals accountable in terms of 'quality assurance' systems that equate 'standards' with 'value-for-money'.  It is the task of educational researchers to 'make' knowledge, in the form of 'generalisable representations' that can be deployed as means-ends rules, to maximise the performativity of teachers in delivering 'value-for-money'. In embracing this task, with national research assessment exercises providing incentives for doing so, mainstream educational researchers will tend to be intolerant of too much 'action' in teaching, and of a form of research which supports it.  In this context, action-research constitutes a reverse discourse that offers teachers an alternative future.





Arendt, H (1958) The Human Condition, University of Chicago Press: Chicago.


Belsey, C (2002) Post Structuralism: a very short introduction, Oxford University Press: Oxford.


Coulter, D (2002) Defining Action in Action Research, Educational Action Research, Vol 10, No 2.


Dadds, M (1995) Passionate  enquiry and school development: a story about teacher action research, Falmer Press: London


Derrida, J (1995) The Gift of Death, University of Chicago Press: Chicago.


Descartes, R (1968) Discourse on Method, Penguin Books: London


Dunne, J (1993) Back to the Rough Ground: practical judgement and the lure of technique, University of Notre Dame Press: Notre Dame, Indiana.


Ebbutt, D & Elliott, J (1985) Issues in Teaching for Understanding, Longmans: York.


Elliott, J (1991) Action Research for Educational Change, Open University Press: Milton Keynes (published in Spain by Morata: Madrid).


Elliott, J (1993) Professional Education and the Idea of a Practical Educational Science, in Elliott, J (Ed.) Reconstructing Teacher Education, Falmer Press: London.


Elliott, J (1996) School Effectiveness Research and its Critics: Alternative Visions of Schooling, The Cambridge Journal of Education, Vol 26, No 2


Elliott, J & MacDonald, B (1975) People in Classrooms, CARE Occasional Publications No 1, Centre for Applied Research in Education, University of East Anglia: Norwich UK


Elliott, J & Adelman, C (1996) Reflecting Where the Action Is: an account of the Ford Teaching Project, in O'Hanlon, C (Ed.) Professional Development through Action Research in Educational Settings, Falmer Press: London (originally published in 1973 in Education for Teaching)


Elliott, J, Maclure, M, Sarland, C (1996) Teachers as Researchers in the Context of Award Bearing Courses and Research Degrees, UK Economic & Social Research Council End of Award Report No R000235294.


Foucault, M (1979a) Discipline and Punish, Penguin Books: London


Foucault, M ( 1979b) The History of Sexuality, vol.1, Allen Lane: London


Hargreaves, D (1997) In Defence of Research for Evidence-based Teaching: a rejoinder to Martin Hammersley, British Educational Research Journal, Vol 23, No 1


MacLure, M (1995) Post-Modernism: a postscript, Educational Action Research,

Vol 3, No 1


Macmurray, J (1957) The Self as Agent, Faber & Faber: London


O'Hanlon, C (2002) Reflection and Action in Research: is there a moral responsibility to act?  Theory and Practice in Action Research:  some international perspectives, Symposium Books: Oxford


Saussure, F de (1974) A Course in Linguistics, (transl) Wade Baskin, Fontana Books


Schon, D (1983) The Reflective Practitioner: how professionals think in action, Temple Smith: London


Stronach, I & MacLure, M (1997) Educational Research Undone: the post-modern embrace, Open University Press: Buckingham & Philadelphia