Disciplining Information Systems: Truth and its Regimes

Transcription

1 (2003) 12, & 2003 Palgrave Macmillan Ltd. All rights reserved X/03 $ RESPONSE Disciplining Information Systems: Truth and its Regimes Lucas D. Introna 1 1 Centre for the Study of Technology and Organisation, Lancaster University Management School, Lancaster, LA1 4YX, UK Correspondence: Lucas D. Introna, Centre for the Study of Technology and Organisation, Lancaster University Management School, Lancaster, LA1 4YX, UK l.introna@lancaster.ac.uk Abstract This paper provides a critical discussion of the ongoing concern of Information Systems (IS) academics on the status of IS as an academic discipline. The focal claim of the argument is that the status of IS as an academic discipline is not an ontological or an epistemological question but rather a political one. In defending this claim the paper will draw on the work of Foucault, in particular his discussion on the relation between power and knowledge. The paper will conclude that the advice given by Paul (2002) and others may be appropriate but for very different reasons. It will claim that it is equally important, if not more so, to know the reasons why claims to legitimacy may succeed or not, for such claims to be effective as a political programme. (2003) 12, doi: / palgrave.ejis Introduction The paper on the status of Information Systems (IS) as an academic discipline by Paul (2002), co-editor of the European Journal of Information Systems (EJIS), which was also published in EJIS, is stimulating and interesting in many respects. His comments on the general concern about the status of IS are not new. It follows a long tradition in which the paper by Banville and Landry (1989), for example, has now become a canonical piece. Ever since the early 1980 s, there has been a steady stream of papers grappling with this seemingly perplexing problem: is IS an academic discipline? and if not what can it do to become one?. The answer to the first part has mostly been that IS is still young, immature and emerging, that is, it is moving towards being an academic discipline. On this there seems to be general agreement. The answer to the question of how it can best become one is much more diverse. There are those who feel that the field can only legitimise itself as an academic discipline by becoming more rigorous and more relevant (Keen, 1991; Benbasat & Zmud, 1999), or by having a cumulative tradition (Keen, 1980). There are those who feel that the field should define its core phenomena or deep structures in a way that will be distinct (Weber & Wand, 1995) or stick to its core as already defined by the Association of Computing Machinery curriculum recommendations (Nunamaker et al, 1982; Chand, 1989). Others suggest that although it is important to continue to define and defend a core, we should remain open to the possibility to include a rich set of intersections with other disciplines (Davis, 2000). Fitzgerald & Adam (1996; Adam & Fitzgerald, 2000) provide a useful summary of the issues that the IS field needs to overcome in its march towards becoming a legitimate academic discipline : (a) A lack of first principles as a stable and widely accepted conceptual foundation.

2 236 Disciplining IS Lucas D. Introna (b) An identity crisis as seen in the location and naming of departments. (c) A lack of a cumulative tradition. (d) An absence of barriers to entry in the field. (e) The enormous potential breadth of the area of the field. (f) A reference indiscipline problem, that is, inappropriate use of theory from other disciplines. (g) A trend towards divergence rather than convergence in research in the field. This paper will comment on Paul s paper, but also more broadly on the questions posed by those before him. In doing this, the paper will first comment on the question of the constitution of an academic discipline and, second, on the particular manner in which Paul presents his views of where the field of IS is and how it could achieve disciplinary status. In both of these levels, the discussion and arguments will draw on the work of Foucault in particular his view on the relation between knowledge and power. On becoming disciplined In the background of Paul s argument, as well as the authors outlined above, there lurks a view that truth or the search for truth operates, at least in its ideal form, outside of the play of power and politics. For example, Paul concludes: Intelligent thinking is what will advance IS. In my opinion there are no short cuts. Maybe, but who will recognise intelligent thinking as such it is not self-evident. To the corporate sponsor of research, intelligent thinking may be the sort of thinking that helps the organisation compete and succeed. For the environmentalist, that same sort of thinking might be utterly absurd and shortsighted. What might be intelligent thinking in Computing may not be in the Social Sciences it may even be seen as reductionistic and perhaps simplistic. The view that will ultimately prevail as exemplars of intelligent thinking is a political question of funding, institutional arrangements, and the like. The truth, as Foucault said, is ultimately a political matter. Thus, the focal position of this paper is that the status of IS as an academic discipline is not an ontological question or an epistemological question as such, but rather a political one from the start. This claim needs further elaboration. When claiming that it is not an ontological issue, the claim is that the phenomena that people in IS study (IS in organisational contexts for example) do not by their very nature (their ontology as such) inform or compel the people studying them to organise themselves, or be organised, in any particular institutional way. After all, being an academic discipline is a status conferred by institutional practices such as the ability to form departments, appoint chairs, organise conferences, edit journals, etc. Nevertheless, there is obviously some historical link between the ontological view of the phenomenon being studied and the traditions that govern the particular institutional arrangements at universities, as is visible in faculty and school structures. For example, one might believe that an IS is essentially note this is an ontological statement a physical system (hardware and algorithms) for the efficient processing of data objects. Alternatively, one might take a radically different view and believe that an IS is essentially a social system for the purposes of organizing of which its physical state is just a practical issue to be resolved. Depending on the ontological view that one takes, certain methods of study may be seen (by a group of people studying them as well as those validating this study) as more relevant and others less so. Once again we must note that the phenomena we study do not tell us or compel us how we ought to study them as such they certainly constrain us, but they also allow us quite a bit of interpretive flexibility. Now it often happens that the people that take IS as being x or y systems tend to choose methods they believe appropriate for the job, and they tend to seek out others that have a similar view. These groups of people then tend to organise themselves (or get organised) into intellectual fields, eventually becoming located in departments, research groups, schools, faculties, etc. For example, being located in a Management School rather than in a Computing School makes certain type of research possible and others not, and vice versa. These groups tend to align themselves to other groups in order to legitimise themselves (as being a respectable field of intellectual pursuit). In every alignment there is always a political cost certain ways of conceptualising and researching the phenomena of study will simply not go down as relevant, appropriate, etc. As these groups succeed to grow, they tend to develop programmes of teaching and research, organise conferences, create and publish journals, and so forth. If they succeed to build up a sustainable intellectual or academic infrastructure of departments, research programmes, conferences, journals, associations, etc, they may eventually become acknowledged as being a discipline. Thus, being an academic discipline is first and foremost a political achievement, not an ontological or epistemological one. They are not a discipline because they got it right found the distinctive core descriptions, theories, and methodologies but because they have convinced enough people, and aligned themselves with enough people that matter. The reference to those that matter is a reference to the sort of people who are in a position to make decisions about funding, promotions, publications, and so forth. This very brief discussion of research and its institutional entanglements already suggests the intimate link between knowledge claims and politics (truth and power in Foucault s terms). In the next section, a brief summary of Foucault s views of the link between power and knowledge will be presented. This will be followed by a more detailed commentary on Paul s paper in terms of Foucault s notion of regimes of truth.

3 Disciplining IS Lucas D. Introna 237 Knowledge/power and regimes of truth In opposition to the modern view that knowledge is produced in a zone where power is suspended, Foucault (1977) argues that any attempt to separate power and knowledge is futile since the production of knowledge is always and already political. To attempt to separate knowledge and power would be to claim that we could separate statements of fact from the values and mechanisms that already constitute them as such. Latour (1987, 1993; Latour and Woolgar, 1986 [1979]) has convincingly argued that what we find in practice is that facts do not speak for themselves. Facts are produced as facts because we value them as such. It is institutional mechanisms and practices that give facts a voice in the first instance. They become constituted as facts through processes, procedures and discursive practices that produce them and are likewise produced by them. For example, in the modern scientific regime of truth, we value scientific method and therefore we judge its outcomes to be facts. We do not value intuition and therefore we judge its outcomes to be speculation. One could say that facts are merely legitimate value choices accorded that status through the prevailing institutional mechanisms. Thus, for every recognised fact (or set of facts), one could always, in principle, find the institutional mechanisms that accord it that status and which is itself dependent on that status. For example, the valuing of profit becomes sedimented as accounting or economic facts in the income statement and balance sheet of the company, which are themselves necessary to sustain that value. Through the rituals of accounting practices and stock exchanges, the income statement and balance sheet become constituted as truth at the expense of equally valid alternative values, such as environmental or employee concerns. Nietzsche (1967) writes in Will to Power: But what is truth? Perhaps a kind of belief that has become a condition for life. (p 248) or as Foucault later articulates it, truth simply means beliefs that have become a condition for that institution to sustain itself as that which it believes it is. Every attempt to secure knowledge in a zone outside of power will itself already be a resource for, and outcome of, power. For we should, as Foucault argues, admit rather... that there is no power relation without the correlative constitution of a field of knowledge, nor any knowledge that does not presuppose and constitute at the same time power relations (Foucault, 1977). The linking of power and knowledge through discourse gives rise to what he calls regimes of truth. A regime of truth is the institutional infrastructure for the production and circulation of truth claims. Foucault claims that truth is produced in and through institutionalised discursive practices. Discourse here is understood as a particular way of talking, of making statements, about the world. As such, truth is understood as claims about the world that is proffered as valid claims within a particular regime of truth. We must note here that Foucault is talking about truth here as knowledge claims we expect others to accept as such. We can obviously say what we want and believe what we want vis-á-vis truth; truth can be whatever we want it to be for us. However, when we expect others to accept these statements and views as true, we find ourselves within a regime of truth. Foucault (1977) argues that each institution or society has its regime of truth, its general politics of truth. A particular regime of truth is constituted through a set of mechanisms and discursive practices that legitimises claims and is itself dependent on the legitimacy of those claims. In Power/Knowledge, Foucault (1977, p 131) mentions the following mechanism and practices that constitute a particular regime of truth: (1) Types of discourse that it accepts and makes function as true: Clearly, not all discourse in institutions functions as truth claims. For example, when I present this paper at an academic conference, I am proffering truth claims in a way I am not when chatting to a colleague in the passage about the ideas in the paper. Every institution has its institutionally designated moments or occasions where truth claims can be proffered and legitimised as such. (2) Mechanisms and instances that enable one to distinguish true and false statements: Once we occasion certain discursive moments as instances of truth making, we must set in place the mechanisms for identifying true statements from false statements. How will it be verified? For example, a paper in a peer-reviewed journal is reviewed by expert reviewers who are charged with the task of making the distinction on behalf of the academic discipline. (3) Means by which each is sanctioned: Truth is not produced if it is not sanctioned. Presenting the paper to my peers at an academic conference and having it published in the proceedings is one of the practices to sanction the truth claims I am proffering. (4) Techniques and procedures accorded value in the acquisition of truth: In distinguishing true from false statements, and in sanctioning these, careful attention is paid to the techniques and procedures used in the construction of these truth claims our methodology. If I would claim that you must accept what I am saying because I had a vision in the night in which an angel appeared to tell me that this or that is the case, I would expect you to think I have lost the plot. Nevertheless, in a religious institution, this may be considered to be a valid procedure for the acquisition of truth. (5) States of those who are charged with saying what counts as true: In institutions, the charge of acknowledging truth claims is carefully distributed and controlled. The auditor is charged to make claims about the financial status of the organisation in a way that the marketing manager is not. If the auditor claims that the company is in financial trouble, then this is taken seriously as valid statements about the financial state of affairs. However, to make these statements, the

4 238 Disciplining IS Lucas D. Introna auditor must be at least a certified accountant. Likewise, we normally require professors or senior academics as examiners of PhD dissertations. In regimes of truth [t]ruth is linked in a circular relation with systems of power which it induces and which extend it (Foucault, 1977 p 133). Power, through micro-practices and mechanisms of meaning, membership and discipline (Clegg, 1989), structures and restructures practices in which truth claims about the world are made, in a discontinuous and diffused manner. Owing to the non-egalitarian and diffused nature of the power relationship, such discourse gives rise to a particular regime of truth. As such, it is more likely that a regime of truth emerges (even by accident) rather than being an intentional strategy or a plot. There is never just one location, point or person that holds all the strings. Nevertheless, every claim to truth whatsoever always implies a regime of truth to accord it that status. Some regimes of truth may be subtle and very mobile and other more explicit and more rigid. For example in the family, the parents have the ability to make truth claims in a way that the children cannot. This is because of the particular relations of power enabled by their access to resources of meaning, membership and discipline such as the sanction of society of the role of parents as guardians. Yet, this regime of truth is continually open for dispute and manoeuvring in a way that a court of law is not. Thus, ultimately the claim that IS is a discipline or not is a truth claim made within a particular regime of truth the institutions expected to legitimise such claims other academics or disciplines in universities, faculties and schools. It is irrelevant whether we put our claim to each other or to the public at large. That is not where we seek academic legitimacy in the first instance. On making truth claims: about being a discipline (or not) Having in mind Foucault s work discussed above, the discussion and analysis will now return to Paul s claims as well as the claims by the other authors concerned about IS as an academic discipline. The aim is to show how these claims are themselves already configured by and within regimes of truth. The purpose is not to question these practices or claims as such even this critique operates within such a regime but rather to demonstrate in very concrete terms the way in which power and knowledge are intertwined in the case of Paul s paper and in the broader case of the discourse about the status of IS as an academic discipline. In order not to be repetitive, any reference made to a particular mechanism or practice of the regime of truth in operation will be indicated by using a square bracket and a number that correspond to the discussion above such as [1] for types of discourse that it accepts and makes function as true or [3] for means by which each is sanctioned, and so forth. This analysis starts by noting that Paul s paper was published in the EJIS. As a paper published in an international peer-reviewed journal, the author is expecting the reader to consider its claims as truth claims and not just as a light-hearted chat or mere speculation [1], especially since it is not published as an editorial or an opinion piece. One must then assume that it was at least reviewed by the Editor of the issue O Keefe [2] who is indeed charged with making such decisions as the Coeditor of EJIS [5]. The claims were legitimised and sanctioned as such by the fact that it was published in EJIS [3]. The paper had also previously being sanctioned as it is based on a Keynote address to the 11th Annual BIT Conference in Manchester, 30 and 31 October 2001 [1, 2, 3, 5]. A critical reader may suggest that there is still an important element missing: what about his methods or evidence to back up the claims he is making [4] as the paper does not have an obvious research methodology. Paul deals with this matter in his introduction by stating the following: (a) he has been an academic and consultant for the last 30 years; (b) he has led the initiative in turning the Department there [at Brunel] into a major teaching and research IS Department (40 academic staff and growing, over 1000 students including of the order of 80 PhD students and in the recent UK Government s Research Assessment Exercise (RAE) held about every 5 years, scoring at the second highest level on a 7 rating scale) (p 174). Who could possibly disagree with somebody with such a formidable record (and it is)? Especially as the evidence being proffered directly supports the claims, that is, he knows what needs to be done to become a legitimate field of teaching and research in an academic institution. Not only this, his track record was also sanctioned by an independent agency, the Research Assessment Exercise [5]. Paul further claims that for me, an academic discipline is a subject plus a body of knowledge. By the latter I mean some rules or laws or evidenced guidelines or even practical universally applicable results (p 174, emphasis added). In making this claim on the nature of a body of knowledge which is an epistemological claim he is aligning himself with a particular regime of truth; one that values generality, universality and practicality, based on evidence, rather than to a regime that values specificity, idiosyncrasy (otherness) without regard to practical value or applicability. This is obviously a good move to make when you make these claims within a journal that shares these values and as an active academic in a university context where such values dominate [2, 4]. He may or may not personally believe this (although as a selfdeclared pragmatist he probably does). This is not the point. Clearly a claim to the contrary would simply not have been politically feasible and somebody in his position would know this all too well. He continues to claim that IS lack such a body of knowledge and therefore tend to exaggerate their work, make unjustified claims about other disciplines, and suffer from an inferiority complex which makes them believe they need to depend on a label provided by another academic discipline (p 174). Paul s claim about the need for a

5 Disciplining IS Lucas D. Introna 239 body of knowledge as well as his claim about the lack thereof in IS suggests that Paul is either saying this because it is politically correct (which I doubt) or because he genuinely believes that being a discipline is an ontological and epistemological problem. In such epistemological terms, the claim of a lack of a body of knowledge might be challenged by many (Davis, 2000) without necessarily becoming convinced to the contrary, as he suggests. One obvious strategy to argue against his claim, which one would expect his opponents to use, is to suggest that there is indeed a significant implied body of knowledge in the form of departments, research and teaching programmes, journals, conferences and associations of these there are so many that it would take a number of pages to list them all. One need only browse the Association of Information Systems web pages to appreciate its extent. It would be extremely difficult for Paul to argue that all of these academics, departments, teaching and research programmes, conference publications and journal publications do not constitute a relatively distinct and discernable body of knowledge. Clearly, every department came about as a result of debates where those present were able to persuade those that made the decisions that they represented something sufficiently distinct as to be allowed to form a separate department, likewise with conferences, journals, research programmes, etc. The particular arguments they used are not as important as the fact that they have been spectacularly successful in convincing many to accept their claims that they do indeed represent a distinct body of knowledge. It would be very difficult for Paul to make the argument that they have all been ontologically and epistemologically misguided this is unless he shifts to a political point of view, which would undermine his original position of defining a body of knowledge epistemologically. Indeed, it seems that the political project of turning IS into an academic discipline has been very successful indeed. This point will be taken up again later on. One should also mention that using references to other disciplines already recognised as legitimate is a very common political strategy, which may of course turn out to be a bad move if done inappropriately. It seems that a lot of the typical inappropriate behaviour that he ascribes to the IS researchers can easily be understood if one takes a political view of knowledge rather than an ontological or epistemological view. In the section on research methods and methodologies, Paul attempts to steer carefully between the political and the epistemological position. He both claims that methodologiesy are in my experience only carried out on paper (sometimes retrospectively) in order to satisfy quality compliance (p 175), which is a political statement, and but surely research is about the application of the researchers intelligent thought processesya [particular] research method may be appropriate if one knows the research task well enough and maybe there are no established methods, and so argumentation and evidence will have to suffice (p 175), which is an epistemological statement. If this last statement is to be taken seriously, why do editors (like Paul) insist that papers published in peer-reviewed journals must include an account of the research methods used? The scope of EJIS calls for first rate research articles by academicsy. What will the interpretation of first rate research articles be? Would the typical academic reader assume anything other than a paper based on well-established and excepted research methods? [4]. Paul might object and claim that it is not required for EJIS as such, but one would be hard pressed to find any papers in EJIS without an explicit discussion or reference to a research method except of course the paper now being subjected to this analysis. Why this contradiction? As EJIS and its editors and reviewers also exists within a regime of truth, it cannot escape (even if one of its editors may wish to do so). The authors submitting work are disciplined by their supervisors, by the reviewers, by tenure and promotion committees, and many others, which already have the status of deciding what counts as true [5]. Allan Lee, the past editor of the MISQ, once commented that although he as the editor explicitly requested good papers using interpretative methods (which he wanted to promote), he did not receive much but was rather continually swamped by positivist research papers exactly the sort of research that is seen as legitimate in many of the departments where potential contributors were located. The editor both draws upon the regime of truth for his/ her power and is also disciplined by it. It is exactly these ongoing disciplining practices of the regime of truth that make a field a discipline rather than a mere collection of loosely related researchers. Concluding discussion From the analysis above, it seems that there is an interesting ambiguity in Paul s paper as well as in many of the previous commentators on the status of IS as a discipline. On the one hand, there is a firm belief that the matter (of status as an academic discipline) is essentially an ontological and epistemological one that is, is there a distinctive core, established research methods, a cumulative tradition, etc. On the other hand, they frequently make statements and comments that seem to suggest the contrary, suggesting an awareness of the political manoeuvring required. However, these political realities are often seen, minimally as a nuisance, or ultimately as an unnecessary corrupting influence in a debate that ought to free of power and politics. Foucault suggests that the questions of ontology and epistemology are already political questions from the start. Our phenomena do not speak for themselves. Thus, the debate on ontology and epistemology (core, methods, traditions, etc) will only be effective if it is framed as an appropriate political programme. This is not to say that anything goes. Quite to the contrary, it rather means that it is only within the disciplinary practices of

6 240 Disciplining IS Lucas D. Introna the regime that such a status can be conferred. There simply is no external, objective and neutral foundation, criteria, or truth as such that we can appeal to or call upon in search of academic status. It is only through the disciplinary practices of the regimes of truth that we become a discipline. This means that we need to understand and recognise the regimes of truth within which we operate, and seek status, if we want to become disciplined and recognised by them as such. Certainly, all experienced researchers know that you need to be very careful in choosing where you submit your work for recognition. The problem is that we tend to see these practices as distorting forces rather than the very practices that constitute communities of researchers as such and enable them to continue to be what they already are. At a different level, one could also ask the question: why this anxiety about being an academic discipline? Is it just a matter of respectability or has it to do with survival. If it is rather about survival, then one could question whether this is the right strategy. Many very respectable Mathematics and Philosophy departments are being closed down or downgraded as sections within other larger departments. In many respects, IS departments have enormous resources for political manoeuvring, such as the income they generate. In a world where education is becoming more and more market driven, they may have increasing room for manoeuvre. However, the discipline cannot escape the regimes that already discipline other disciplines, the university and ultimately society at large. Clearly, the automatic legitimacy of the scientific project is no longer taken for granted in the face of so many controversial findings (BSE, autism and MMR, cloning, etc). Regimes are shifting. Maybe being an academic discipline may become a distinct disadvantage? References ADAM F and FITZGERALD B (2000) The status of the IS field: historical perspective and practical orientation. Information Research 5(4), available at: BANVILLE C and LANDRY M (1989) Can the field of MIS be disciplined. Communications of the ACM 32(1), BENBASAT I and ZMUD R (1999) Empirical research in information systems: the practice of relevance. Management Information Systems Quarterly 23(1), CHAND DR (1989) Some observations on information systems as an academic discipline. CIS Educator Forum 1(3), CLEGG SR (1989) Frameworks of Power. Sage Publications Ltd, London. DAVIS GB (2000) Information systems conceptual foundations: looking backward and forward. In (Basskerville R, Stage J, DeGross JI (Eds), Organizational and Social Perspectives on Information Technology pp 61 82, Proceedings of the IFIP TC8 WG8.2 International Conference, Aalborg, Denmark, 9 11 June, Kluwer Academic, Boston. FITZGERALD B and ADAM F (1996) The future of IS: expansion or extinction?, Proceedings of the First Conference of the UK Academy for Information Systems, Cranfield University, UK, April 1996, 15 pp. FOUCAULT M (1977) Truth and Power, Power/Knowledge: Selected Interviews & Other Writings (Gordon C Ed.) Pantheon Books, New York. pp KEEN P (1980) MIS research: reference discipline and a cumulative tradition. In Proceedings of the First Conference on Information Systems, Philadelphia McLean E (Ed.) pp KEEN P (1991) Keynote address: relevance and rigor in information systems research. Information Systems Research: Contemporary Approaches and Emergent Traditions, in Nissen H, Klein H and Hirschheim R, (Eds), pp 27 49, Elsevier Publishers, North-Holland, Amsterdam. LATOUR B (1987) Science in Action: How to Follow Scientists and Engineers through Society. Harvard University Press, Cambridge, MA. LATOUR B (1993) We have never been Modern. Harvester, New York. LATOUR B and Woolgar S (1986 [1979]) Laboratory Life: The Construction of Scientific Facts. Princeton University Press, Princeton. NIETZSCHE F (1967) The Will to Power. Vintage Books, New York. NUNAMAKER Jr. JF, COUGER JD and DAVIS GB (1982) Information systems curriculum recommendations for the 80s: undergraduate and graduate programs. Communications of the ACM 25(11), PAUL R (2002) (IS) 3 : is information systems an intellectual subject? 11, WEBER R and WAND Y (1995) On the deep structure of information systems, Information Systems Journal 5(3),