A Philosophical Perspective on Knowledge Management

Lewis Hassell, Ph.D. College of Information Science and Technology Drexel University Philadelphia, PA 19104 lew.hassell@cis.drexel.edu

Abstract: Knowledge management claims to be the successor of various trends in the business world, including, but not necessarily limited to information resources management, business process reengineering, management information systems, and organizational memory. A number of definitions have been proposed for it. This paper takes a philosophical look at knowledge management and questions the implications of these definitions for and the possibility of knowledge management. 1 Conclusion

It may seem odd to put the Conclusion section first. I do this because many of the readers will be non-philosophers. In recognition of this, I am attempting to spare these readers the pain of trying to keep straight the circumlocutions in which philosophical argumentation sometimes seems to engage. My conclusions, then, are: 1) There is no knowledge outside of experience. 2) Knowledge is therefore always embodied i.e., it is always in a physical human being1. 3) Experience is always the experience of some (rational) individual in a society. 4) Knowledge therefore is associated with a social group. 5) (Computerized) knowledge management, therefore, can only ever be global information management. By global information management I mean the widespread distribution of structured, semi-structured, and unstructured organizational data. A number of implications for knowledge management may thus be derived from our conclusions: 1) Attempts to capture real knowledge are doomed to failure. 2) Simply providing full text documents or semi-structured documents (like those found in Lotus Notes or OVAL [ML92]) can lead to less productivity (indeed, less knowledge) unless great care is given to properly filtering and categorizing the information before presenting it to employees. 3) While vendors and other practitioners who have a vested interests in providing knowledge management solutions are perfectly entitled to continue to call what they do knowledge management, it would be more productive to focus on what one can do, which is facilitate the communication of data or information (which is structured data) within an organization.
1 If god(s), angels, etc. have experiences, it is qualitatively different from human experience, assuming they have no bodies. If they do have bodies, a great many theologians will be sadly disappointed.

4) Real knowledge management must include instilling loyalty in the employee that can only come from a sense of legitimacy that modern capitalism and its social structures sorely lack. 5) But even proper management and proper organization of information and documents will probably not get you what you want anyway. What those who look to the benefits of knowledge management are really looking for is to foster creativity, which requires going beyond management to leadership. Hopefully, with my conclusions thus presented, my arguments will be easier to follow. 2 Introduction

The idea of managing knowledge might have been considered somewhat amusing to (at least some) philosophers in earlier times. Many of these throwbacks would probably have conceived managing knowledge to mean managing truth. This, is clearly not what is meant by advocates of knowledge management. What, then, are we attempting to manage, and what activities are involved in managing this that is called knowledge? Reading the literature of knowledge management advocates, one might almost think that the goals of knowledge management seem akin to those of education or culture or perhaps both. The Greek concept of paideia2 comes to mind [Ja70]. But can this be what the purveyors of knowledge management systems mean? One would think not. Paideia carries with it a broader time frame, befitting the passage of multiple generations. What the business community seems to mean by knowledge management might span the passage of multiple department heads, but probably not even that. Paideia entails a mutual commitment of the members of the community to some broader, higher, mutually advantageous purpose. The purpose of knowledge management seems to be more on the order of supplying a short-term competitive advantage to a business or other organizational entity. The commitment to the individuals comprising those organizational units is somewhat dubious, to which the many recent news stories involving corporate greed will attest. In what follows, I will take a (perhaps jaundiced) philosophical look at knowledge and management in hopes of finding out what is meant by knowledge management. If I went no further, then this would be as idle an endeavor as medieval theologians debating how many angels can fit on the head of a pin. What I hope to achieve by the descriptive process is to indicate some prescriptions about what should be considered knowledge and the domain of knowledge management. Philosophy may thus be the discipline that keeps everyone honest. It realizes that words have consequences and it demands a thorough accounting of those words. The larger implications of the business communitys attempt to manage their knowledge assets for political and social philosophy should also be considered, though a thoroughgoing discussion is beyond the scope of this paper. 3 What is knowledge?

One may think of paideia as culture, for a start. It encompasses a consciousness of (shared) values that govern human life and an approach to educating the society in those values.

First things first: philosophy (at least the Western philosophical tradition) tells us that there is no knowledge outside the possibility of experience [Ka65]. And experience is the experience of a sentient being (so, at this time, computers dont qualify). While there is still lively debate in philosophical circles concerning the exact nature of experience, for present purposes I may consider it to mean that knowledge occurs within a context of social activity, or what continental philosophers call a Life World [see, for instance, Hu70]. Furthermore, according to Paul Feyerabend, knowledge is essentially idiosyncratic. Knowledge lies not in discrete elements that are provably correct, but in the interplay of sometimes conflicting elements [Fe75,Fe99]. Thus knowledge has a curious characteristic much like a radioactive isotope, it has a half-life. The fact that the earth revolves around the sun is just that a fact. It is a datum. At some time in the past, it may have qualified as knowledge (and as Copernicus discovered, knowledge can be dangerous), but not now. Knowledge is thus something that matters. Knowledge, as the experience of a human being, means that there is no such thing as disembodied knowledge. Intelligence is not based on deductions made from rules overlaid on some massive database. Instead, it is a way of acting within an environment or community. (Indeed, the test Turing should have required is to see whether a computer can act, undetected in a society.) Decades of experiments with databases and expert systems have failed to come even close to human intelligence [Dr72, Dr92, Dr01). Even the purest form of knowledge, mathematics, is still (arguably) based on the embodied mind [LN00]3. One doesnt even have to be a philosopher to realize that knowledge is not just some elaborate data that can be shoved into a data (or knowledge) base. In the California Business Review, McDermott [Mc99], too, pointed out that knowledge belongs to communities. But what does it mean to have a community? Is this community simply a group of knowledge workers? What about the clerks, secretaries and other lower level workers? Surely they cannot be knowledge workers or can they? Are they part of the community? Do they support and add to its activities? Where does knowledge begin and non-knowledge end? A number of philosophers (among them Cairns [Ca00] and de Sousa [So87]) have argued convincingly that emotions may manifest reason as much as belief. (If the reader is unmoved by philosophical argumentation on this matter, she or he is referred to the neurologist Antonio Damasios discussion of Phineas Gage in [Da94].) If emotion does manifest reason, then emotion, too, is part of knowledge. How does this impact knowledge management? Certainly, computers would be of dubious value here, regardless of the supposed great advances that have been made in the field of affective computing [Pf00] There is yet another aspect to knowledge that philosophers in the continental tradition refer to as authenticity [Hu70, He62]. Authentic knowledge is the gold standard of knowledge. Authentic knowledge grasps the object of knowledge itself. Inauthentic experience is more akin to hearsay. It is second hand knowledge. Commenting on Heideggers Being and Time, Dreyfus [Dr91] notes, Thus nongenuine inauthentic existence can take the form of a spiritual trying-to-be-saved-inevery-possible-way by, for example, practicing yoga, transcendental meditation, Greek bodybuilding, and Christian love all at once (p. 193). We believe this describes the knowledge management movement quite well. While the knowledge gathered by a knowledge management system may have been authentic for the person who put it there, it is certainly not authentic for the person who finds it in the system. The answer to this problem might seem to be to include as much
3

The problem of psychologism must be left for another day.

of the context surrounding the knowledge as possible. This is a slippery slope as aptly demonstrated by the sociologist Garfinkel [Ga67]. Thus, knowledge is not just organized stuff (information, call it what you will). Knowledge is even more than just recognizing patterns in information, which is what vendors of knowledge management products would like you to think. Conceivably, one day, we will produce computers But for that are capable of recognizing complex patterns in piles of data or information4. knowledge to occur, they will also have to know when not to apply the information, to make exceptions, to reason. Furthermore, they will have to collaborate with others who can help. In short, they will have to be able to do the sort of thing that any good secretary can do today. Making implicit or tacit knowledge explicit seems to be another goal of knowledge management systems [Kr02]. But the difference between implicit and explicit knowledge is not that one is in your head and one is recorded. Implicit knowledge is that to which one cannot give ready voice. It may be knowledge that you do not know that you have until you find it. This is not to say that elements (we put this in term in quotes because we wish to stress that it is an abstraction) of implicit knowledge are explicit. They are. The process in which these elements combine and morph, however, is a creative process that does not readily lend itself to capture. In his classic study of how epic oral poets sing their poems, A. B. Lord [Lo60] notes that every time a singer sings his song, he re-creates the song using traditional components. These traditional components comprise the elements upon which the creative process is exercised. This seems to be closer to what modern business is looking for, and it cannot be bought on the cheap. Moreover, if Feyerabend (see above) is correct, that knowledge is the interplay of discrete elements, then you alter knowledge either by altering the elements or by altering the interplay of them. Thus extracting the knowledge from someones head is doomed to failure. The best one can hope to do is order these discrete elements in one or more reasonable (whatever that means) arrangements for later retrieval. But this is not what seems to be meant by knowledge management. Given the above, we can then conclude that knowledge cannot come from simple information transmission mechanisms but must come from a culture, which implies social organization, commitment, and common values. And if providing a quick and ready reference to expertise (implicit or explicit) is what knowledge management is trying to achieve, then we are in fact trying to build an electronic community. Community, however, requires more than email (or any other, even more elaborate form or groupware). 4 What is management? Philosophers have not been very active in the discussion of what is meant by management, but lets take a stab. Websters defines to manage as to direct or conduct business affairs. That is the definition of the word as an intransitive verb. As a transitive verb, the definition is much more to the point: to direct or control the use of; to exert control over; to make submissive to one's authority, discipline, or persuasion. By virtue of what authority does an organization have the right to exert control over anything that is essentially the property of a human being (i.e., essentially embodied in the human being)? Certainly, organizations can exert control over documents, but knowledge management is (supposedly) much more than document management. Certainly, organizations can exert control over their patents, trademarks and trade secrets, but knowledge management is (supposedly) much more (voluminous) than that.
4

Dont look forward to this any time in the near future. Computers are still terrible at simple pattern recognition.

Furthermore, we must be circumspect in our desire to dominate. There are those who believe that if nature rails at our attempt to dominate her, this can be solved by a little more domination and manipulation. Global warming, regardless of whether President Bush believes in it or not, is an excellent example that this sort of dominating, manipulating sort of approach to the world may not work for long. Even if we can exert control over our material assets make them submissive to our will this just sounds a bit odd, if not unpleasant, when directed to our knowledge assets. Is there any aspect of our life physical, spiritual, social which modern business does not want to dominate and control? Perhaps it is time to reread Marx. If Karl Marx is unacceptable, then perhaps Michael Useem ([Us01], [Us97]) will better suit the taste of capitalists. Useem is a Wharton professor who discusses the importance of leadership, as opposed to management. Management of a business is simply running it. When we manage we discharge our responsibilities to the organization. Management concerns the day-to-day activities of the business. Leadership, on the other hand, is creative. Leadership adds value. Leadership exceeds goals. Leadership is the effective use of power for the greater good ([Us01] p.3). And if we read what Useem says about leadership, it has two other important qualities. Leadership is difficult, and leadership is a distinctly human activity. This is not to say that leadership is only a personal activity. There is also organizational leadership. Organizational leadership is the ability to meld and exploit a team of associates ([Us97] p.80). Organizational leadership makes people want to excel. Organizational leadership includes the exercise of change and development of other people Leadership, then, can be viewed as leveraging what you are given to achieve far more. ([Us97] p.91). It would seem that philosophers have been silent about management simply because they felt it mundane (thus betraying a bit of philosophical snobbery). They have written no end on virtue and character. But these are elements constitutive of leadership. Leadership also inherently falls under the concept of praxis. In the Nichomachean Ethics, Aristotle [Ar62] tells us that if we want to know how to live the good life, we need to ask the man of practical wisdom. So, too, if we want to know what it means to be a leader, we must look at the lives and conduct of leaders. 5 Knowledge management or leading up with technology?

So, one more time, what does knowledge management mean? According to whatis.com, Knowledge management is the name of a concept in which an enterprise consciously and comprehensively gathers, organizes, shares, and analyzes its knowledge in terms of resources, documents, and people skills. The confusion over the difference between information and knowledge is exemplified in the next paragraph, where they say that Knowledge management involves data mining and some method of operation to push information to users. Knowledge management involves pushing information to appropriate users? Somehow, this seems less lofty a goal than many in the knowledge management community would like. Are we then to conclude that knowledge management is information management plus email? So, we can at least safely say that we want to push this data/information/knowledge to knowledge workers? As the reader can guess, the definition of a knowledge worker is as nebulous as the definition of knowledge in knowledge management. Programmers are often referred to as archetypal knowledge workers. Factory workers, presumably, are not knowledge workers. Nor are farmers. But what about programmers who dont keep their skills up to date, or arent imaginative in their approach to algorithm construction, are they knowledge workers? And what of the factory worker who pours over a problem on the factory floor, even dreaming about it, and one day a

solution dawns on her is she a knowledge worker? Or the farmer who uses satellite and advanced agricultural information from the government to produce more wheat is he a knowledge worker? And what about our ancestors who invented the wheel, or a reliable way to produce fire were they knowledge workers? Furthermore, as pointed out above, knowledge workers dont exist in a vacuum but in a social network much broader than their own narrow field or discipline. It has also been said that the fundamental characteristics of management in knowledge management is gathering, organizing, refining, and disseminating. But are we gathering, organizing, refining, and disseminating knowledge or information? If information, then when does it become knowledge? If knowledge, when did it become such? In any case, what one would need in order to have real knowledge is an electronic Lebenswelt, or better yet, electronic paideia i.e., an extended culture with a system of values and education in those values. There is precious little discussion in knowledge management literature, however, of the development of this type of community. Where is the discussion of shared values, of common commitment? One suspects it left when the focus was put on the computer. Or maybe it was when employees became information assets. Or maybe when those few at the top of an organization began to make hundreds of times what those at the bottom did. A great many knowledge management systems seem to be textual databases, or semi-structured document databases (e.g., Lotus Notes or Oval [ML92]). As such, they are based on some sort of text or written expression [Da01]. Communication between people often involves verbal communication. Research [Ha95, HC96, Ho02] has shown, however, that linguistic expression is considerably different in written versus oral discussion of the same subject domain. The impact of these differences has yet to be adequately explored by the knowledge management community. On a related note, an interesting debate has occurred in the linguistics community about what text even means. Specifically, does text communicate authorial intentions, or simply transmit information [DB01]? Those who contend that text does communicate authorial intentions center their argument on the production of fiction [see, e.g., Gi01]. We must wonder if this applies to nonfiction, particularly business text, where the author(s) may or may not be known to the reader. Gerrig and Horton [GH01] argue that communication is not dichotomous (communicating authorial intention or not) but continuous. Text may be communication, but it certainly is not the same as spoken language. The differences can be crucial. According to the Gartner Group [cited in SK00], knowledge management is a discipline that promotes an integrated approach to identifying, capturing, evaluating, retrieving, and sharing all of an enterprises information assets. This includes not just paper-based and electronic data and documents, but also uncaptured expertise and experience in individual workers (p. 3). This would imply that the cognitive (conscious?) life of the worker is somehow the property of the employer. (Karl Marx would be fascinated, though perhaps not shocked, at this.) While it is reasonable to believe that an employer contracts with the employee to perform certain functions, and thus owns the result (artifact) of the process or activity, the ownership of ones expertise and experience is troubling to those of us who do not yet recognize the deification of Adam Smith. 6 Final thoughts

What I have said is not somehow peculiar to philosophers or a particularly philosophical stance. In the Harvard Business Review (hardly the bastion of wild-eyed philosophers), John Seely Brown [Br02] points out that work practices must be developed as one develops products. Furthermore, as

discussed above, Useem reminds us that organizations do not move forward simply by efficiently managing what they have. Organizations move forward (particularly in tough and perilous times) through leadership. According to Useem, leadership involves improving and perhaps even transforming what we have inherited from others ([Us98] p. 263). This is what the advocates of knowledge management are claiming for their product (discipline?). Efficiently moving data, or documents, or whatever back and forth will never result in effectively using the knowledge that is embedded in an organization. Someone with a managers mentality will not become a leader. They will never inspire and motivate their employees to integrate or to borrow a term from Hegel, synthesize past and present knowledge to create new value for the company. And the computerized knowledge repositories are little more than electronic libraries. Mohanty [Mo00] argues that the creation of philosophical texts is not sufficient for philosophy. All that without the internal activity of thinking is not genuine philosophy, but a body of doctrines (38). So too with the more mundane fields of human endeavor, and knowledge management. We are constantly told (e.g., Drucker, pick a book) that the old rules dont apply and that we are living in a new age. It is interesting that the old rules stopped applying about the time that we stopped applying the old rules: respect for workers who possess knowledge, who do a good job, who work hard. Of course, we were also told that the old rules of the stock market profitability, sound business models, etc. dont apply any more. I will assume that none of the readers are planning on retiring on their dot com stock portfolio. If we can assume, then, that knowledge is sense impressions (or data, or whatever) within a structure as interpreted within the experiential background (let us call it a Lebenswelt), then real knowledge management is not possible without a wildly totalitarian regime. Polanyi [Po70] correctly argues, Apart from meaningless sense impressions there is no experience that abides as a fact without an element of valid interpretation having been imparted to it (p.89). Interpretation occurs against a backdrop of context as it occurs within a society. Suffice it to say that what I have argued for implies that if you spend enormous amounts of time and money creating a knowledge repository, you will be sadly disappointed. What you must spend time and money on is your work force. Humans are not replaceable pawns in the process of knowledge production. They embody the knowledge indeed, they are the knowledge. Useful, even provocative, bits of information can help them, but that is all. The best the knowledge management community can do is to provide information that is well categorized [e.g., Sm95, Sm98] something easier said than done. You are of course free to redefine knowledge and then declare that you have created a knowledge management system. That, however, will not change the underlying truths. Knowledge resides in the human being not simply the human brain, but in the whole human, social being. Recognizing that people embody knowledge, some organizations, in the name of knowledge management, try to facilitate directing people with questions to people with answers. Without the creation of a true community, this can only be knowledge herding. Community true community implies shared values, a shared sense of purpose, shared rewards and motivations. Twenty years ago, those of us who were building databases discussed the problems of getting people to populate our databases. What would motivate a person to take what they know that which makes them valuable to an organization and share it with others? If you answer that question, you will find the answer to productivity, indeed, creativity. My bet is, however, that the answer will have little, if anything to do with a computer.

Bibliography [Ar62] [Br02] [Ca00] Aristotle (1962), Nichomachean Ethics, translated by M. Ostwald, The Liberal Arts Press, Inc. Brown, J. S.: Research That Reinvents the Corporation, Harvard Business Review. August 2002, 80(8). Cairns, D.: Reason and emotion. Husserl Studies, 17, 2000; 21-33.

[Da94] Damasio, A. R. (1994), Descartes Error: Emotion, Reason and the Human Brain, Quill, New York. [Da01] Day, R. E.: Totality and representation: A history of knowledge management through European documentation, critical modernity, and post-Fordism. JASIST, 52(9), 2001, 725735. [De87] de Sousa, R. (1987), The Rationality of Emotion, MIT Press. [DB01] Dixon, P.; Bortolussi, M: Text is not communication: A challenge to a common assumption, Discourse Processes, 31(1), 2001; 1-25. [Dr72] Dreyfus, H.: What Computers Cant Do. Harper and Row, New York, 1972. [Dr91] Dreyfus, H.: Being-in-the-World. MIT Press, Cambridge, 1991. [Dr92] Dreyfus, H.: What Computers Still Cant Do. MIT Press, Cambridge, 1992. [Dr01] Dreyfus, H.: On The Internet: Thinking in Action. Routledge, New York, 2001. [Fe75] [Fe99] Feyerabend, P.: Against Method. NLB, London, 1975. Feyerabend, P.: The Conquest of Abundance. University of Chicago Press, 1999.

[Ga67] Garfinkel, H.: Studies in Ethnomethodology. Prentice Hall, 1967. [GH01] Gerrig, R. J.; Horton, W. S.: Of texts and toggles: Categorical versus continuous views of communication. Discourse Processes, 32(1), 2001, 81-87. [Gi01] Gibbs, R. W. Jr.: Authorial Intentions in Text Understanding. Discourse Processes, 32(1), 2001, 73-80.

[Ha95] Hassell, L.: Media, Speech Act Theory and Computer Supported Cooperative Work. Drexel University, Unpublished doctoral dissertation, 1995. [HC96] Hassell, L.; Margaret, C.: Indirect Speech Acts and Their Use in Three Channels of Communication. In Proceedings of LAP96, Springer-Verlag, 1996. [Ho02] Holtgraves, T.: Language as Social Action, London: Lawrence Erlbaum Associates, 2002. [He62] Heidegger, M.: Being and Time. Trans. by J. Macquarrie and E. Robinson. Harper and Row, New York, 1962. [Hu70] Husserl, E.: Logical Investigations. Second Edition. Trans. by J.N. Findlay, Routledge and Kegan Paul, New York, 1970. [Ja70] Jaeger, W.: Paideia: The Ideals of Greek Culture. Oxford University Press, 1970. [Ka65] Kant, I.: Critique of Pure Reason. Trans by N. K. Smith, St. Martins Press, 1965.

[Kr00] Kreiner, K.: Tacit knowledge management: the role of artifacts. The Journal of Knowledge Management, 6(3), 2000, 112-123. [LN00] Lakoff, G.; Nez, R.E.: Where Mathematics Comes From: How the Embodied Mind Brings Mathematics into Being. Basic Books, 2000. [Lo60] Lord, A.B. (1970), The Singer of Tales, Atheneum, New York. [ML92] Malone, T.W., Kum-Yew Lai and Fry, C., 1992, "Experiments with OVAL: A Radically Tailorable Tool for Cooperative Work," in Proceedings of the Conference on Computer Supported Cooperative Work, ACM Press. [Mc99] McDermott, R.: Why information technology inspired but cannot deliver knowledge management. California Management Review, 41(4), Summer, 1999, 103-117. [Mo00] Mohanty, J.N. (2000), The Self and Its Other: Philosophical Essays, Oxford University Press. [Pf00] Pfeifer, R. (2000), On the role of embodiment in the emergence of cognition and emotion, in Affective Minds, Edited by G. Hatano, N. Okada, and H. Tanabe, Elsevier, New York, 43-57.

[Po70] Polanyi, M.: Science, Faith and Society. University of Chicago Press, 1970. [SK00] Srikantaiah, T.K.; Koenig, M.E.D. : Knowledge Management for the Information Professional. ASIS Monograph Series, Information Today, Inc., Medford, New Jersey, 2000. [Sm95] Smith, B.: Formal Ontology, Common Sense, and Cognitive Science. International Journal of Human-Computer Studies, 43, 1995, 641667. [Sm98] Smith, B.: Basic Concepts in Formal Ontology. In (Guarino, N., Ed.): Formal Ontology in Information Systems, IOS Press, 1998. [Us98] Useem, M.: The Leadership Moment. Random House, New York, 1998. [Us01] Useem, M.: Leading Up: How to Lead Your Boss So You Both Win. Random House, New York, 2001.