UNCERTAIN KNOWLEDGE, UNCERTAIN TIMES

ASAC 2005 Toronto, Ontario Teresa Marcon Abhijit Gopal Richard Ivey School of Business The University of Western Ontario UNCERTAIN KNOWLEDGE, UNCERTAIN TIMES Different research traditions co-exist in the information systems domain. We explore the tensions between these in the context of the history of the human sciences and the problem of representation (Foucault 1994). Against this background, we articulate elements of difference between research traditions toward a deeper understanding. Introduction In our quest for knowledge, we are fortunate in our time to stand on the shoulders of those who have gone before. A growing body of theoretical, methodological and empirical work informs research in the information systems (IS) field. Yet, paradoxically perhaps, as a community we have never been more aware of the uncertainty that surrounds our knowledge than in recent times. This is not only because the world around us is changing at a seemingly unprecedented rate (e.g. Beaudan 2002; Colett, 2004), challenging existing theories and practices, but also because we have developed a keener appreciation for the precarious foundations of our knowledge. The research community in IS has welcomed into its fold an expanding range of methodological perspectives from different paradigmatic positions (e.g. Ciborra and Hanseth, 1998; Crabtree, 2004; Doolin and Lowe, 2002; Gopal and Prasad, 2000; Orlikowski, 2000). Such perspectives, often referred to as qualitative research methodologies (Denzin and Lincoln, 1994), begin from a different set of ontological and epistemological commitments to arrive at alternative explanations of phenomena of interest to IS researchers. This shifting of the paradigmatic ground (Burrell and Morgan, 1979) offers insight along with the reminder that all knowledge is dependent on a frame of reference or a set of assumptions that cannot be measured against an absolute standard of truth (Kuhn, 1970; Foucault, 1994). In confrontation with the claims to knowledge of multiple perspectives, the field has from time to time engaged in a healthy debate as to what this means and how we should proceed (e.g. Benbasat and Weber, 1996; Lee, 1999). We have attempted to articulate the relationship between different modes of research, seeking to find points of continuity (e.g. Lee, 1991) and difference (e.g. Klein and Meyers, 1999) and ways of working together from our respective positions (e.g. Robey, 1996; Lewis and Grimes, 1999). The interaction overall has been productive rather than 248

divisive. Nevertheless, perhaps a measure of discomfort has accompanied such confrontations from time to time. It is perhaps this discomfort that prompted Ron Weber (2004), until recently editor in chief of MIS Quarterly, to protest in an editorial statement that the differences between positivism and interpretivism are spurious once we take a careful look, and that perhaps it is time we stopped drawing lines of division and moved on to more interesting and fruitful debates. We concur with Weber (2004) that it is important to attend with care to the differences between paradigmatic positions, avoiding naïve stereotypes. We disagree however with the contention that there are no differences between positivism and interpretivism. In this paper, we respond to Weber s argument and attempt to begin to articulate what we believe are important differences between research traditions. In our opinion, such differences cannot comfortably be collapsed in spite of the desire we also share to work toward a collective enterprise. Before presenting our view, we outline the substance of Weber s argument, with particular attention to the frame within which the confrontation between positivism and interpretivism is articulated. Subsequently, we propose an alternate frame and consider differences between methodological positions on this basis, attending in particular to the question of representation. In speaking in defense of difference, we seek to respond to Weber s call for a deeper understanding of the modes of knowledge in use in our field in hope that such an understanding may help us to deal constructively with the tensions that the co-existence of these different modes may engender. A Case of Spurious Difference? Are positivism and interpretivism so very different, asks Weber (2004)? In posing this question, Weber enjoins us to move beyond what has become the rhetoric of positivism versus interpretivism (p. iii), something that has achieved the status of folklore (p. iii), and take a careful look. Weber argues that the alleged differences between positivism and interpretivism are spurious (p. xi): they are based on a naïve view of positivism that does not take account of the manner in which our collective understanding of knowledge production has changed following the meta-theoretical reflections of philosophers of science such at Thomas Kuhn (1970). When we juxtapose positivism and interpretivism along ontological and epistemological dimensions, as is often done, Weber suggests that we move beyond simplistic comparisons that associate positivism with a correspondence theory of truth and the possibility of an unproblematically objective representation of the world. Weber (2004) argues that both positivists and interpretivists recognize that theories are social constructions that provide useful ways of understanding our world rather than truths. And while positivists assume a world that has an independent existence, they recognize that they cannot ever know reality with any degree of certainty, having no unmediated access to it. Both positivists and interpretivists acknowledge that our representations, those facts we seek to uncover, are inevitably theory laden. Objects are conceived either as subject to perception or as the repository of meanings we ascribe to them; they are inevitably coloured by the biases we carry with us into the research domain. Moreover, since Heisenberg, positivists have been aware that our interaction with the objects we study inevitably changes them. The interdependence of subject and object is not the privileged insight of interpretivism but is equally shared by 249

positivism. In light of this unavoidable interdependence, both positivists and interpretivists strive toward an always imperfect objectivity, even if objectivity is understood in somewhat different terms in their respective domains. According to Weber, for positivists, to be objective entails minimizing bias. Interpretivists, on the other hand, seek objectivity in representations that reflect the consensus of research participants as they point to a subjective reality that is intersubjectively shared (i.e. objective). While in the interpretivist perspective an intersubjectively shared world that is a construction replaces a reality conceived as independent of the observer, such differences do not materially alter the aims of either kind of research. Viewed against a Kuhnian understanding, positivism and interpretivism would seem to share a great deal of common ground, perhaps far more than is often suggested. They are perhaps only different theories about the world that embody different assumptions and may favour the application of different methods. Both question our ability to arrive at truth and both recognize the impossibility of a truly objective science, striving nevertheless for a measure of accountability. And while criteria for evaluation are more clearly defined in the positivist domain than in the interpretative realm, both interpretivist and positivist research gain their legitimacy through a consensual process that is predicated on the judgment of the research community (Golden-Biddle and Locke, 1993). When we abandon a naïve depiction of positivism, commonality replaces difference (Weber, 2004). Weber suggests that the rhetoric of positivism versus interpretivism (p. iii) may have served a useful purpose at one time, opening a space in the IS field for alternative research traditions but do such juxtapositions have anything to offer the research community today? Perhaps it is time to recognize the commonalities and move on to explore more interesting questions, such as what motivates researchers to apply particular methods to a problem. We concur with Weber (2004) that the field has little to gain by juxtaposing different research traditions based on naïve conceptions of either positivism or interpretism. We challenge, however, Weber s contention that the differences between positivism and interpretivism are spurious. Prior to challenging Weber s conclusion, we question the neutrality of the grounds upon which his argument is made. The Importance of Frames Any comparison between different modes of research takes places within a particular framework: a way of understanding that forms the ground against which a comparison can be made. Such grounds are never neutral. Rather, they reflect particular world views and assumptions or systems of thought (Foucault 1994) which inevitably colour the understanding that emerges. The importance of attending to the frame of reference for a particular claim as well as to the claim itself may be quickly grasped by inspecting Figure 1. The small square in the centre of figures 1a and 1b is the same colour. Yet, framed by black as in figure 1a, the middle square appears to be lighter in colour than the middle square of figure 1b which is positioned in a white frame. As a further illustration of why the grounds for a comparison matter, consider what 250

happened when social studies researchers responded to Kuhn s (1970) call for a sociology of scientific practices. Arguing from a broadly constructionist position, sociologists from Latour (e.g. 1988; 2000) to Garfinkel (e.g. Garfinkel et al, 1981) to Bijker (e.g. Bijker et al, 1987) produced accounts of the work of scientists that depicted them not only as constructors of theories and useful instruments, but also as inventors of facts, objects and phenomena, from nebulas (Garfinkel et al, 1981) to bacteria (Latour, 1988). The backlash from the scientific community, manifested in the so called science wars (Latour, 2000), puzzled these researchers: had they not only just rendered into their accounts what scientists themselves had shown them and told them about without in any way devaluing the scientific enterprise (Latour, 1993)? Not all scientists agreed. Indeed, accounts of human practices, including those of researchers, are never neutral. The depictions they produce are inevitably coloured by the grounds upon which they are constructed and the language in which they are articulated. Such grounds reflect deeply held beliefs and commitments. When producers of accounts and their readers do not share the same background of beliefs, assumptions and commitments, such accounts may result in misunderstandings and a heated exchange of opinions. Figure 1 1 Gestalt Theory Boxes (a) (b) (Source: American Heritage Dictionary, 2005) It is not our intention to engender either misunderstandings or a heated confrontation among researchers in IS that favour different modes of research. Yet, we believe that in arguing within the confines of a Kuhnian science, Weber (2004) appropriates interpretivism into a frame which shapes the contours of the inquiry and colours the understanding that emerges in a way that erases what we believe are important differences between interpretivism and positivism. To attempt to bring such differences to light, we begin by articulating a new frame of reference based on an epistemology of the human science as proposed by Michel Foucault (1994). Against this background not a neutral one but one that is more sympathetic to the interpretivist position we 1 We are adopting this well known optical illusion from the field of psychology as illustration aid only. We do not proceed to frame our argument in terms of perceptions in a psychological sense. 251

attempt to highlight elements of differences between positivism and interpretivism, attending to both these positions in a non-naïve way. We wish in this manner to contribute to a deeper understanding of different modes of research that may help us live with our uncertain knowledge and with the differences that arise in the confrontation between research traditions in our field. Information Systems Research as a Human Science Thomas Kuhn (1970) looked to the history of noble sciences such as physics to unravel the mystery of scientific discoveries and scientific progress. Of interest to Michel Foucault (1994) instead were those yet unexamined and perhaps less noble disciplines whose history spans a mere two hundred years: the human sciences. First published in 1966, four years after the release in book form of Kuhns (1970) The Structure of Scientific Revolutions (originally published in 1962), The Order of Things (Foucault 1994) attempts to uncover through archaeological analysis the epistemological grounds of the human sciences. This effort parallels the aims and goals of a long tradition of investigations into the philosophy of science (e.g. Chalmers, 1999) but, unlike such accounts which look to the history of physics or mathematics for empirical examples, Foucault s (1994) investigation focuses specifically on the domain of the human sciences. As such, it may be particularly relevant to the IS field which we would place among the human sciences as will be evident below. Arising at the beginning of the nineteenth century, the human sciences take as their object man as an empirical entity (Foucault 1994) 2. Unlike other kinds of empirical entities, though, man has a double role: man is both an object of knowledge and a subject that knows (p. 312). Whereas other sciences biology, economics and linguistics, for example are concerned with man only as an object, with what man is by nature (p. 353), the human sciences are concerned also with man as a knowing subject. They inquire into what enables man to create representations of entities encountered in the course of his life and which permit him to function in the world. They are concerned moreover with what allows man to create representations of his own life as a self-reflexive being. Thus, man is that living being who, from within the life to which he entirely belongs and by which he is traversed in his whole being, constitutes representations by means of which he lives, and on the basis of which he possesses that strange capacity of being able to represent to himself precisely that life (p. 352). In the study of man, representation then is of key importance and this in two senses. First, because man engages in representation as an integral part of his life. As human beings, social scientists of course also engage in representation in their personal lives as also in their scholarly work. Second, representation is important to the human sciences because the possibility of representation is in itself a pre-condition both for man s activities and for the activities of social scientists. As a pre-condition, or a condition of possibility (Foucault 1994) for the human sciences, the question of representation belongs to the epistemological ground that guides the 2 Following Foucault, we adopt the masculine term to denote humankind. We do so for the sake of simplicity in integrating Foucault s own language into our discussion, while acknowledging the problematic nature of adopting this terminology. 252

rationality of the human sciences, that is, to the underlying system that gives representation its force and its legitimacy. To address the question of what constitutes the system of representation for the human sciences, we must first consider what representation entails. Representation relies on symbols or signs. It entails the establishment of a relation between a sign (a word) and a signified (an object or idea). Foucault (1994) set out to investigate the form of this relation as it changed over time across a variety of disciplines, including the human sciences psychology, sociology and literature and the sciences of life, labour, and language (p. 351) biology, economics and linguistics which he sets apart from the human sciences. His method is comparative and archaeological: it is concerned with uncovering the system of thought or episteme (p. 312) that pervades a particular time in history and in searching for its emergence across a variety of domains of knowledge. It also entails looking for transitions from one system of thought to another. Such transitions may at times permit the emergence of new domains of knowledge and this was indeed the case with the human sciences. The human sciences arise around the beginning of the nineteenth century as the epistemological grounds shift from those of the Classical to those of the Modern period (Foucault 1994). In the Classical age which spans the seventeenth and eighteenth centuries and pre-dates the emergence of the human sciences, science revolves around a project of classification that attempts to organize nature according to similarities and relations within an immense and continuous hierarchical table. The relation of the sign to the signified of words and things (Kendall and Wickham, 1999) is assumed to be one of direct correspondence. Efforts in this period were directed at formalizing concepts and providing descriptions exhaustive enough to point precisely to specific entities. Representation was contained entirely within the sign (the word, the concept) itself and the elimination of ambiguity was the aim. Since the nineteenth century, in a period that Foucault (1994) calls Modern, we recognize in the relation between signs and the objects they point to a more complex arrangement which is triadic (sign [ ] object), rather than dyadic (sign object). The exact nature of that which mediates the relation between the sign and the object varies according to particular perspectives (e.g. consider, for example, differences in the semiosis of Saussure (Culler, 1976) and the semiotics of Pierce (Thompson et al, 1989)). Irrespective of the precise system that is posited, this mediation is predicated on language: a language that is not only a grammar, that is, a set of logical relations between parts of speech, but also the site of a constant fragmentation, spreading and destabilization of meaning (Foucault 1994). On this basis, we can perhaps understand the modern preoccupation with language, narrative and representation, not only within those domains that would naturally need to take an interest in it linguistics, or literature, for example but also across the entire spectrum of the human sciences (Alvesson and Karreman, 2000) and in management research (e.g. Boje, 1995). It is against the epistemological underpinnings of a problematic of representation that the human sciences arise and function. Because of the importance of representation and the complex relation between signs and signifieds, the human sciences maintain a relation not only to those disciplines that take man as their object i.e. biology, economics and linguistics but also to philosophy as the discipline that takes man as its subject. The double role of man as both subject 253

and object thus places the human sciences in a unique position vis a vis other domain of knowledge. Foucault (1994) places the human sciences in a complex epistemological space (p. ix) delimited by the three axis of mathematics, philosophy and the sciences of life, wealth and language (Figure 2). Like other sciences, the human sciences maintain a relation to mathematics though primarily as a tool. More importantly, they are located in the vicinity of the sciences of life, wealth and language (biology, economics, linguistics) from which they draw their models, concepts and theories and which they concurrently challenge by reminding them that man is more than an object. They are further located in relation to philosophy with which they exist in constant confrontation: philosophy questions the naiveté of foundations of the human sciences and aims to provide it with such a foundation. At the same time, philosophy is threatened by the encroachment of the human sciences into its territory and the appropriation of its philosophical apparatuses. It is from this complex and tensioned epistemological space which is predicated on the possibility of representation that the human sciences exist: in the volume of the cube depicted in Figure 2. Foucault (1994) argues that indeed it is not the complexity of the subject matter addressed by the human sciences that accounts for their constant wavering, fuzziness or multiplicity. Rather, the uncertainty that surrounds the human sciences is a result of the complex epistemological space in which they are located. Figure 2 The Epistemological Space of the Human Sciences Biology, Economics, Linguistics Mathematics Philosophy A common foundation in a system of representation points in the direction of sameness across knowledge domains, rather than difference. Though Foucault (1994) would suggest that all 254

knowledge within a particular period, such as the present, shares a common episteme on the question of representation, this is not equivalent to saying that all modes of knowledge share the same rationality in their entirety. Foucault recognizes in the Modern period the co-existence of different forms of knowledge not only across disciplines, but also within a particular discipline (Foucault 2002). Thus, within the human sciences, there are those investigations that pursue a more positive sort of knowledge, taking their models primarily from biology, economics, and linguistics. Other efforts tend toward a more philosophical or reflexive sort of knowledge, looking to philosophy for guidance (Rajan 2002). These two types of knowledge co-exist today. The tensions that Foucault identifies in the relation of the human sciences to other disciplines suggest points of contention and difference that have persisted throughout the lifespan of the human sciences. In the field of IS, as also in psychology or sociology, such tensions are replicated within disciplinary boundaries among researchers who share a domain of problems but frame their questions and answers within different modes of knowledge. Eradicating those tensions and collapsing difference into a common framework, as Weber attempts to do, may not be an achievable aim in light of the complex nature of the interactions. Moreover, it may not be a desirable aim in so far as it may obscure what Foucault (1994) suggests is the episteme of our age: the shifting ground of language and representation. In the following section, we attempt to begin to make visible through an example the way in which the question of representation, and the differences it engenders, is manifested in different modes of knowledge that co-exist in the field of IS. In Defense of Difference We have noted above that the human sciences address themselves to man as an entity that is both the object of inquiry and the subject who engages in representation (Foucault 1994). On an epistemological level, representation in the modern age is a complex affair. It involves a triadic structure wherein the relationship between representation and that which is represented cannot unproblematically be fixed. We live in the present time with both these tensions: the tension of a human who has a dual nature, being both subject and object, and the tension arising from the shifting relation between the sign and the signified. Against this backdrop, two kinds of knowledge co-exist in the human sciences. We will take positivism as representative of a kind of knowledge that aims at positivity and interpretivism as evincing more of an interest in a philosophical kind of knowledge. 3 We will use a simple example in order to begin to tease out how positivist and interpretative approaches might accommodate two of the thorniest issues in the human sciences which we identified in the above paragraph: the doubleness of man and the issue of representation. That positivism and interpretivism attend to both these issues, as will become evident below, attests to the compelling 3 It is quite possible that some forms of interpretivism may more actively pursue positive rather than philosophical knowledge. Concurrently, one should also consider whether some positivist theories might not be more in tune with a philosophical rather than a positive type of interest. We leave consideration of such questions for another occasion, merely noting at this time that such possibilities exist. 255

quality of Foucault s analysis of the human sciences. How they do so offers some interesting hints as to where some differences between them may lie. We will take up by way of example the generic task of studying a mountain. We choose this example because it evokes a story that we have often heard as an explanation for a multiparadigmatic approach to research. The mountain is a metaphor for an object or phenomenon which is of interest. It is often stated that researchers approach the mountain from different angles, seeing a different piece of it or looking at the same vista from a different perspective. Researchers may also wear different shades of sunglasses which filter the light in particular ways, bringing particular aspects into relevance. The cumulative efforts of many researchers approaching the mountain from different sides, viewing it from different angles and through different filters produce multiple accounts of the same phenomenon and enhance our collective understanding. Weber s (2004) own position would seem to fit comfortably within these confines. This account is potentially problematic for us on two counts. First, it adopts a realist view in positing that there is a mountain that is objectively out there and which we collectively study. Those arguing from non-realist perspectives, to which we are sympathetic, would hold in abeyance the question of whether there is or is not an object out there that is a mountain. We would reframe the metaphor in terms of a variety of researchers taking an interest in something that has been called a mountain. Second, an inquiry within the field of IS, which we place among the human sciences, would take an interest not only in the mountain itself but also in the human beings who represent the mountain to themselves or others and who are capable of reflecting on this within the broader context of their lives. Included in this category, of course, are the researchers themselves. We will now consider an example of a positivist and an interpretative approach to studying a mountain in turn. Taking a socio-psychological approach as an example, a positivist researcher might inquire into the types of beliefs about the mountain that are held by human beings. S/he may further posit factors that would influence such beliefs, including properties of the mountain itself (which research subjects are assumed to perceive in a largely unproblematic way), traits, emotions, the opinions of significant others, etc. In inquiring into beliefs and related constructs, the researcher is inquiring into the objective properties of human beings, following in large part a biological model that considers the human as an object with certain properties and functions. At the same time, by asking people to complete questionnaires or answer questions verbally the researcher would be attending to the representations of the mountain offered by people, a focus that allows human beings to participate as subjects who know (Foucault, 1994, p. 312). This is largely a consequence of the choice of measurement instrument. Indeed, the fact that human beings have shifted from being objects to being subjects may in large part escape from notice if it is not somehow part of the theory being employed. A socio-psychological view therefore, as one possible theoretical perspective within a positivist mode of research, would tends to emphasize man s role as object rather than as subject 4. 4 Indeed, Nietzsche (1964) charged psychology with turning the subject into an object. 256

The question of the shifting grounds of representation, for its part, would not surface explicitly in this kind of research. The development of measurement instruments in survey research, however, attempts to address the problematic of representation by seeking to stabilize the connection between constructs and measures. This task relies on various techniques: item generation, selection, validation, and statistical tests of validity. Here the doubt that accompanies the problematic relation of signs (constructs) to signifieds (the objects they point to) is largely absorbed into techniques for judging the adequacy of measurement instruments. We see from this simple hypothetical example that a positivist examination of a phenomenon would take account of both the doubleness of man and what we might term the problematic of representation. It would do so, however, by emphasizing man s role as object rather than as subject and by trying to minimize the fragmentation of meaning through elements of method. Interpretivism is generally associated with the phenomenological tradition (e.g. Burrell and Morgan, 1979; Weber, 2004) 5. Phenomenology proposes a way of bringing a phenomenon into clarity through a particular approach to investigation that attempts to bracket pre-conceptions (Husserl, 1982) and considers entities in the context of a totality or ensemble (Heidegger, 1962), relating the part to the whole and the whole to the part 6. As such, phenomenology is more like a methodology for investigating phenomena rather than a theory that describes the world. In the process, however, phenomenology also attends to what is uncovered through such an investigation. Phenomenology (Heidegger, 1962) seeks to understand beings, including entities and human beings themselves, in their essence, attending to what is hidden rather than to surface manifestations. A phenomenological investigation of the mountain would take an interest not in a set of characteristics or properties of the mountain but in the essence of it, or in mountainess. Concurrently, it would take an interest in the being who senses the mountain. It would inquire into how it is possible for a conscious subject to know the mountainess of the mountain. The meanings that people ascribe to the mountain would be of interest to phenomenology in so far as they point to a deeper domain of essences. Phenomenology posits a consciousness that senses the world and interprets it, ascribing meaning to it. This is a knowledgeable subject capable of representing other entities as well as of him/herself. As such, phenomenology privileges the role of man as subject, though it may take man as an object of investigation 7. Though phenomenology does not address the question of language and representation directly, a search for essences or hidden meanings reflects a lack in the signifier (or sign) itself which fails to contain the meaning of a phenomenon. To name the 5 In this example, we draw primarily on the existential phenomenology of Heidegger (1962) though Husserl s (1982) phenomenology could put to the task instead. Given space constraints, we limit our interest to a single strand of phenomenology. 6 The tradition of hermeneutics (Gadamer, 2000) further developed this insight. 7 Heidegger most notably did so in investigating the question of being (Heidegger, 1962). Yet, in so doing, he did not reduce the subject to an object. 257

mountain as mountain is not sufficient for the task of arriving at its essence. The search for essence attempts to capture the meaning of the mountain, eliminating the multiplicity of descriptions of it that might be proffered. More importantly, phenomenology shows us that to know something as mountain is a very complex activity. Though it attends to representation in so far as it tells us about what an object is in its essence, the fundamental contribution of phenomenology is to show us how we come to know a phenomenon as a particular kind of thing. In this sense, phenomenology represents a different kind of knowledge that takes a more reflexive turn or interest. We have attempted to begin to contrast two kinds of knowledge that co-exist in the field of IS. Both engage what Foucault identifies as the crucial tensions of our present age, yet they do so in different ways. Keeping such continuities in mind, a positivist perspective, takes the human principally as its object whereas a phenomenological view tends to focus on human beings as subjects. Confronted with a complex and problematic relation between representations and that which they seek to represent, a positivist perspective may largely subsume such concerns into matters of method. A phenomenological view, while attempting to close meaning around a phenomenon in its essence, seeks to expose to light the process through which such attempts are made. In doing so, it seeks a different kind of knowledge and entails differences that we believe should not be erased. Some Conclusions We have argued in this paper that there are differences between positivism and interpretivism that cannot comfortably be collapsed. To understand the nature of such differences, we have drawn on Foucault s (1994) archaeology of the human sciences. This has allowed us to provide a framework to explain the tensions that surround research in the human sciences and to point to fundamental epistemological issues that shape current forms of knowledge. In the final section, we have relied on an example to attempt to illustrate some differences we perceive between different modes of knowledge. Much more, of course, could and should be done to elucidate these issues. We believe preserving an understanding of differences between research methodologies is important at least for the reason that it affords us the opportunity for a constant questioning and evaluation of our practices. As knowledgeable researchers with an appreciation not only of the ontological and epistemological grounds upon which we carry out our work but also of the social practices that are entailed in the process, we are aware of the uncertain nature of our knowledge. To add to this, we have suggested that this uncertain knowledge is accompanied by the tensions and difficulties of living in an uncertain age, marked by the two issues of the doubleness of man and the problem of representation (Foucault, 1994). Perhaps all knowledge is uncertain and all times are uncertain. We have sought to bring to light the particular uncertainties that are characteristic of our time and the tensions that surround our collective enterprise. Within the tensioned space occupied by the human sciences, different modes of knowledge co-exist. We are suggesting that rather than attempt to bring these modes of knowledge into a common frame, we might through understanding learn to live in this tensioned space: to live with ambiguity and with difference. 258

References Alvesson, M., and D. Kärreman. "Taking the Linguistic Turn in Organizational Research: Challenges, Responses, Consequences." Journal of Applied Behavioral Science 36, no. 2 (2000): 136-158. American Heritage Dictionary, 2005 Available online at http://www.bartleby.com/61/imagepages/a4illusn.html. Accessed January 30, 2005 Beaudan, E. "Leading in Turbulent Times." Ivey Business Journal May/June (2002): 22-26. Benbasat, Izak, and Ron Weber. "Research Commentary: Rethinking "Diversity" in Information Systems Research." Information Systems Research 7, no. 4 (1996): 389-399. Bijker, Wiebe, Thomas P. Hughes, and Trevor E. Pinch, eds. The Social Construction of Technological Systems. Cambridge, MA: MIT Press, 1989. Boje, D. "Stories of the Storytelling Organization: A Postmodern Analysis of Disney as 'Tamara- Land'." Academy of Management Journal 38, no. 4 (1995): 997-1035. Burrell, Gibson, and Gareth Morgan. Sociological Paradigms and Organizational Analysis. London: Heinemann, 1979. Chalmers, A.F. What Is This Thing Called Science? Indianapolis: Hackett Publishing Company, 1999. Ciborra, C. U., and O. Hanseth. "From Tool to Gestell: Agendas for Managing the Information Infrastructure." Information Technology & People 11, no. 4 (1998): 305-327. Collett, Stacy. "Guiding Staff through Turbulent Times." ComputerWorld (2004). Available at http://www.computerworld.com/managementtopics/management/story/0,10801,88692,00.html. Accessed January 30, 2005 Culler, J. "Saussure's Theory of Language." In Saussure, ed. J. Culler, 18-52. Hassocks: Harvester Press, 1976. Crabtree, A. "Taking Technomethodology Seriously: Hybrid Change in the Ethnomethodology- Design Relationship." European Journal of Information Systems 13, no. 3 (2004): 195-209. Denzin, N. K.., and Y. S. Lincoln. "Introduction: Entering the Field of Qualitative Research." In Handbook of Qualitative Research, ed. N. K.. Denzin and Y. S. Lincoln, 1-17. Thousand Oaks, CA: Sage Publications, Inc., 1994. 259

Doolin, Bill, and Alan Lowe. "To Reveal Is to Critique: Actor-Network Theory and Critical Information Systems Research." Journal of Information Technology 17 (2002): 69-78. Foucault, Michel. The Order of Things: An Archaeology of the Human Sciences. London: Vintage Books, 1994 Foucault, Michel. The Archaeology of Knowledge. London: Routledge Classics, 2002. Gadamer, Hans-Georg. Truth and Method. New York: Continuum, 2000. Garfinkel, H., M. Lynch, and E. Livingston. "The Work of a Discovering Science Construed with Materials from the Optically Discovered Pulsar." Philosophy of the Social Sciences 11 (1981): 131-158. Golden-Biddle, Karen, and Karen Locke. "Appealing Work: An Investigation of How Ethnographic Texts Convince." Organization Science 4, no. 4 (1993): 595-616. Gopal, Abhijit, and Pushkala Prasad. "Understanding GDSS in Symbolic Context: Shifting the Focus from Technology to Interaction." MIS Quarterly 24, no. 3 (2000): 509-546. Heidegger, Martin. Being and Time. New York: Harper, 1962. Husserl, Edmund. "Husserl's Inaugural Lecture at Freiburg Im Bresgau." In Introductions to Phenomenology: Interpretation and Critique, ed. W. R. McKenna, 9-17. Den Haag, The Netherlands: M. Nijhoff, 1982. Kendall, Gavin, and Gary Wickham. Using Foucault's Methods. London: Sage Publications Ltd., 1999. Klein, Heinz K., and Michael D. Meyers. "A Set of Principles for Conducting and Evaluating Interpretive Field Studies in Information Systems." MIS Quarterly 23, no. 1 (1999): 67-94. Kuhn, Thomas. The Structure of Scientific Revolutions. Chicago: University of Chicago Press, 1970. Latour, Bruno. The Pasteurization of France. Cambridge, MA: Harvard University Press, 1988. Latour, Bruno. We Have Never Been Modern. Hemel Hempstead, Maryland: Harvester Wheatsheaf, 1993. Latour, Bruno. "When Things Strike Back: A Possible Contribution of 'Science Studies' to the Social Sciences." British Journal of Sociology 51, no. 1 (2000): 107-123. Lee, Allan S. "Integrating Positivist and Interpretative Approaches to Organizational Research." Organization Science 2, no. 4 (1991): 342-365. 260

Lee, Allan S. "Rigor and Relevance in MIS Research: Beyond the Approach of Positivism Alone." MIS Quarterly 23, no. 1 (1999): 29-34. Lewis, M. W., and A. J. Grimes. "Metatriangulation: Building Theory from Multiple Paradigms." Academy of Management Review 24, no. 4 (1999): 672-690. Nietzsche, Friedrick. The Genealogy of Morals : A Polemic. New York: Russell & Russell, 1964. Orlikowski, Wanda. "Using Technology and Constituting Structures: A Practice Lens for Studying Technology in Organizations." Organization Science 11, no. 4 (2000): 404-428. Rajan, Tilottama. Deconstruction and the Remainders of Phenomenology: Sartre, Derrida, Foucault, Baudrillard. Stanford, CA: Stanford University Press, 2002. Robey, Daniel. "Diversity in Information Systems Research: Threat, Promise, and Responsibility." Information Systems Research 7, no. 4 (1996): 400-408. Thompson, C. J., W. B. Locander, and H. R. Pollio. "Putting Consumer Experience Back into Consumer Research: The Philosophy and Method of Existential-Phenomenology." Journal of Consumer Research 16, no. 2 (1989): 133-146. Weber, Ron. "The Rhetoric of Positivism Versus Interpretivism: A Personal View." MIS Quarterly 28, no. 1 (2004): iii-xii. 261