David Snowden
Cognitive Edge, ENG
Knowledge management is a difficult and challenging topic that has been subject to oversimplistic approaches from a variety of authors and technology vendors. It is fashionable to reference its 2,500-year-old philosophical origins in epistemology, which date back to the first use of the phrase “justified true belief” in Plato's Theactetus. It is equally fashionable to claim that the nature of knowledge is such that it cannot or should not be managed. Both statements are misguided and diversionary for related reasons.
Philosophy is concerned with the nature of what it means to “know,” using the specialist language of epistemology, informed and burdened with thousands of years of context, for day-to-day management action either trivializes that language or becomes too academic. The needs of an organization to stem the outward flow of intellectual capital and more effectively deploy what has been called the only sustainable source of competitive advantage (Stewart, 1997) are not best served by the trivialization of “philosophy made simplistic.” Equally, to say that knowledge management is a mere oxymoron is at best an abrogation of responsibility, only available to those who have ceased to manage and now merely advise. At worst, it misrepresents management as an exercise of authority within a bureaucratic command-and-control environment: a very restricted use of the word that enters the English language from Latin via the French for the ability to ride a horse in dressage. Knowledge is something that modern organizations, both commercial and governmental, have to manage in the here and now; the question is how.
Modern knowledge management starts with Nonaka and Takeuchi's 1995 book The Knowledge Creating Company, although Nonaka outlined the ideas in an article of the same title in the Harvard Business Review four years earlier (Nonaka, 1991). The 1995 book created the interest from which many conferences, a few good, and far too many superficial, books and articles have arisen. Key was its introduction of the distinction between tacit and explicit knowledge in the much misused SECI (socialization, externalization, combination, internalization) model, highly derivative of, but not completely true to, ideas originally put forward in Polanyi's 1962 Terry Lectures at Yale University (Polanyi, 1983).
The context of the 1995 publication, in contrast with that of 1991, was that the limitations of the then dominant “fad,” business process reengineering (BPR), had started to become evident. Knowledge was managed before and during BPR, but the word was not problematic in day-to-day business use. One of the main drivers of BPR was the desire to capture and imbed knowledge into processes in the interests of efficiency, and in most cases to legitimize or excuse significant cost reduction through downsizing. This is not the time or the place to argue the merits or otherwise of BPR; what is relevant is that the attempts to imbed knowledge into processes made organizations increasingly aware of the human or tacit components of knowledge and the problems and issues associated with rendering that knowledge explicit. Nonaka's SECI model was seized on, whatever its intent, as providing a means by which the said tacit knowledge could be rendered explicit.
At the same time, we had the first examples of attempts to produce accounting standards for intellectual capital, most notably at Scandia under the direction of the world's first chief knowledge officer (Edvinsson & Malone, 1997), coupled with the growth of balanced scorecards and the first scalable collaboration software in Lotus Notes.
This combination of the intuitively attractive SECI model, a major company managing knowledge within the accepted conventions of accounting, and the growth of technology enablers gave rise to knowledge management and set its agenda.
Looking back in a decade or so, we will probably see 1995 and the growth of knowledge management as the moment signaling the bounding of Taylorism, just as quantum mechanics and the uncertainty principle bounded the universal assumptions of Newtonian science, on which Taylorism or “scientific management” was predicated. At any one time, there is a limit to the range of ideas or concepts that managers can hold within their attention zone of active awareness and conscious management. This is not to say that other things are not being managed, but there are always a limited number of erogenous zones within the body politic of the organization. These zones receive attention, new investment, and focus until their value is understood, its limits defined, and the practices internalized as part of the day-to-day unconscious activities of the organization; at this point their capability to stimulate interest fails.
Until the period from 1995 to the present day (and for some organizations a few years yet), the erogenous zones were often challenging and frequently complicated, but benefit could be obtained through the application of Taylorist principles. Business schools, the rapid and sometimes parasitical growth of management consultancies, and the increasing number and volume of innovative technology developments reflected this in an accepted process of “fads,” itself a manifestation of entrainment. An HBR article or a book by a guru would define a new area of attention, usually at the boundary limits of the previous fad. Conferences and a popular journal or two would create the necessary level of interest, with aspirational presentations and articles by industry practitioners providing the critical mass for a phase shift transition from interest to investment. At this point, a simple model or easily grasped concept or saying would become commonplace, and standard recipe-book approaches based on reductionist thinking would be put in place.
For a time this appeared to be happening with knowledge management, and for certain technology-based solutions it is firmly established. However, Nonaka's separation of tacit from explicit knowledge brought the commonplace discourse of managers into domains of cultural ambiguity, human irrationality, and “knowing,” in which the level of interdependency and interaction rendered the “awareness zone” complex rather than complicated. The increasingly global nature of organizations, enabled by technology and to a lesser extent the growth of the internet, increased network connections to the point where the old infrastructure of BPR, the balanced scorecard, and much systems theory, including the learning organization (Senge, 1990), started to break down and the space opened up for a new organic metaphor of management theory informed by complexity thinking.
The deficiencies of the SECI model in practice are becoming evident (Snowden, 2000a). In particular, organizations are increasingly realizing that there is a body of tacit knowledge that cannot be made explicit, and that even much of what can be made explicit shouldn't be, on grounds of either cost or flexibility (Snowden, 1997). It is also becoming accepted that the reductionist assumptions of intellectual capital measurement systems cannot account for the difference between asset and market value, and that technology cannot fully replace the need for human interaction. Finally, and of most interest, there is the increasing realization that much knowledge is held collectively within communities, and cannot be represented as the aggregation of individual knowledge. This has immense consequences for reward and management systems: we are moving slowly and steadily to a recognition that the modern organization is a complex network of tribes, rather than a feudal landscape in which ownership of budget replaces land as the means of enslavement.
Complexity theory was also starting to appear on the radar of industry practitioners at this time, but its links to knowledge management were not immediately evident. However, as early as 1998 virtual conferences were being set up on “organic knowledge management.” Many of the early KM practitioners had very quickly reached the limits of what could be achieved by treating the organization as a machine, the basic metaphor of Taylorism. Four examples will illustrate this:
Gradually, we are seeing a new pattern of knowledge management practice emerging, in which the organization is treated as a complex ecology. The role of the manager is as gardener or game warden, not mechanic or big game hunter; the consultant becomes a mentor or enabler of descriptive self-awareness rather than the purveyor of prescriptions to manage the symptoms of corporate failure. However, little of this work is properly rooted in a coherent set of concepts. As such, it remains new wine in old wineskins, with all the consequent problems of contamination and leakage. Even some of the more organic practitioners and academics constrain their written words and practices to a Taylorist model of respectability; itself a sin of both omission and commission from which many complexity writers are, regrettably, not exempt.
In parallel with this shift from mechanical to organic, we see an increasing awareness of complexity theory in organizations. The last few years have seen an increasing number of books seeking to popularize complexity theory in the context of management science. Too many of these are trivial, based on inadequate understanding and attempts to relabel existing industry practices as “complex” in an attempt to take the guru spot of an emerging fad. Others provide real insight in a form that is readable to a more general population than the research community (e.g., Axelrod & Cohen, 1999). However, most good complexity writing is descriptive, reflecting its research roots. For Taylorist thinking a description can readily lead to a prescriptive model: a hypothesis from the description is tested, and a prescriptive and purportedly universal model or recipe book created. Nevertheless, this is not true for the cases described as complex; here we are identifying underlying principles or concepts that result in practices. The practices themselves cannot be generalized into universal models, as they result from the application of the principle but are not principles in themselves. In a complex world, best practice is too context specific for universal application; it is always past practice. Unfortunately, the market has been habituated to imitation of best practice arising from models generalized from the practices of a sample of organizations, ideally blue chip, that are used to validate new “fads” and provide security blankets. For understandable reasons, many workers in complexity theory are conforming to this model.
If we are to inform organic knowledge management by complexity, to provide it with the necessary conceptual underpinning, then we need to shift market awareness from imitation of generalized practice to the application of basic concepts. Management has to realize that each solution will be unique, but the underlying principles remain the same: we start intervention-based journeys open to discovery, rather than determining goals focused entirely on exploitation of what is currently known. Expressed another way, organizations have to manage on the basis of ambiguity and uncertainty if they are to take true advantage of the complex system formed by their intellectual capital. This is likely to be bottom up, rather than top down.
An important way to achieve this step change in thinking is to take an existing knowledge management issue and associated organic practice, reviewing it in the context of complexity theory, and applying the revisions in a visible way that on articulation enables a shift in thinking and understanding. By way of illustration, the remainder of this article will examine enhancements to, and the consequent transformation of, communication issues within organizations through the use of story, enhanced by thinking from complexity.
Story allows the communication of complex ideas in a simple, memorable form (Snowden, 2000b, 2000c). It also provides a highly effective means of mapping knowledge within the organization and embedding sustainable lessons learnt (Aibel & Snowden, 1998). Some approaches to the construction of stories rely on identifying examples of good and bad behavior and, through a process of exclusion and refinement, create memorable stories that can lead to behavior modifications (Denning, 2000). Others use narratives to create a greater understanding of the organization through the construction of story taxonomies (Gabriel, 2000), while some of the more creative applications construct fairy stories to create a metaphorical environment to enhance understanding and mutual respect (Spark Team, 2000). These three approaches rely to a greater or lesser extent on the role of an expert interpreter and story writer/creator. They are either fictional or in varying degrees purport to tell the truth in a compelling manner.
All such methods suffer from the danger of emerging anti-stories: stories that evolve very quickly in organizations as a cynical reaction to an official story of “goodness” that exceeds the limits of the believable or politically acceptable. For factual stories, the need to make the story compelling requires a degree of emphasis and selection, which is fertile ground for the emergence of anti-story. For fantasy, the medium itself is likely to induce cynicism; there are a limited number of executives prepared to kiss frogs and save fair maidens with any degree of seriousness, whatever the literary merits of the story.
Work under the author's direction approached story from a different perspective. The early emphasis was on the use of story circles, a facilitated technique to provide the raw material for extracting “knowledge disclosure points” in the form of decisions, judgments, problem resolution, etc. (Snowden, 1998, 2000a), from which knowledge assets could be derived and cataloged. Story circles typically lasted for a half to one day and used a variety of techniques (Snowden, 2000b), including fiction to provoke and elicit anecdotes from the participants that, considered as a whole, express the learning, experience, and culture of the group. The gathering of anecdotal material provided highly valuable content in its own right, aside from its use as a disclosure medium. It was soon evident that this offered the raw material for story construction; indeed, anecdotal material was used to persuade executives to action in several assignments.
From this point, development of story-construction techniques could easily have evolved in the same direction as the three approaches identified above, had it not been for a moment of serendipity. A large project requiring significant behavior change in a rule-bound organization with strong private networks coincided with early study and conference attendance to scope a research project into the application of complexity science to organizations. The eureka moment took place in a tortuous workshop session in an airless room in Paris.
The project in question was a lessons-learnt program in international sales effectiveness. Several story circles had been conducted in a variety of countries with failed, failing, and successful bid teams. A depressing pattern was emerging: where teams were successful in winning business, they ignored the processes designed to mitigate risk, provide management control, and ensure responsible pricing. On the other hand, teams who followed the process generally failed, and often spent the entire gross margin of the contract on the bidding process itself. This was not going to be a palatable conclusion to the project sponsor, hence the depression.
In an attempt to create some meaning, anecdotal material from successful projects was sorted into two groups: contracts won that were both profitable and had created few significant problems in implementation; and contracts that everyone would, in retrospect, prefer to have lost. Each anecdote was printed in full or summary on a single sheet of paper and the two classes of material were tacked to opposite walls. Workshop participants were encouraged to walk and talk about the material. Hypothesis, speculation, challenge, discussion, and disputation created a rich information flow and suddenly a pattern became clear, self-evident to all; to this day, most participants claim the original insight and probably all are correct in their claims. Successful bids, which were not the subject of bitter recrimination years later, had all followed the spirit behind the process. Even if the formal risk assessment had been downplayed, or the risk premium cut without proper authority in order to win a bid, risk had been actively discussed. Further discussion and refinement revealed that most bid teams who were successful in winning sustainable business were operating from some simple, unarticulated heuristics that were held in common across bid teams with little or no contact, but with many common experiences.
Once they had been articulated, these heuristics were capable of rational explanation, but only with the benefit of hindsight. For example, one heuristic was that the risk premium could be halved with complete safety. This was easy to remember and operated in many cases; study of the bid process indicated a series of review stages in the deduction of the risk premium in which the actual practice of each review was to add a safety margin to cover themselves. The net effect was that the premium was doubled by the formal process, and then halved back to its original level by human wit and ingenuity.
This valuable insight would have remained unique to that project, had it not been for the coincidence that the project leader and the initiator of the complexity project was the author, with both activities concurrent. It appeared that there was a strong correlation between the heuristics and the rules governing flocking in Boid's algorithm. In addition, the workshop process had increased information flows to painful levels, but had in consequence seen a breakdown of existing perceptions and beliefs and resulted in new insights that emerged from the active discourse of informed participants. Critically, no expert had analyzed the material and drawn conclusions; meaning had arisen from the community itself, but only where the environment had been changed to create discomfort and disruption.
Subsequent work in a variety of projects, this time informed in advance by complexity, validated the original insights. Once a critical mass of anecdotes, in practice between 30 and 40, had been gathered, increasing information flow between agents in a workshop environment would lead to the emergence of articulated organizing principles, generally expressed as rules, values, or beliefs. This process was assisted if it took place in a performance space with physical activity, movement, and active, often contentious, dialog. By increasing the information flow to the point where current perceptions or infrastructure broke down, organizing principles could be articulated: emergent properties of a complex system. In addition, the process of emergence involved a degree of phase shift: dialog would appear meaningless for long periods and then suddenly meaning would emerge in the form of a memorable phrase. However, there were still issues: sometimes the identification of an organizing principle was difficult, and not always consistent. There seemed to be a step missing in the process.
Now we reach the second moment of serendipity: a keynote address on the use of story at a congress of librarians in Derbyshire, UK. Librarians are interesting people: they are curious, collect eclectic knowledge, and are willing to share it. In this case the inspiration came from Dr. Judy Palmer, director of the Health Care Libraries Unit at John Radcliffe Hospital, Oxford. She had little knowledge of what she was unleashing when she recommended the Mulla Nasrudin stories from the Middle East, collected by Idries Shah (1985). These stories follow a format common to many storytelling cultures in that they use archetypes to explore aspects of human interaction. Other examples include Greek myths, where each god represents an extreme form of human behavior; and aboriginal stories in Australia, where individual animals display specific aspects of human interaction. Archetypal stories, among many other things, allow conversations to take place about aspects of human behavior that cannot be talked about directly. This is one of the uses of the Mulla Nasrudin stories in Sufi society: if you do something stupid, you don't tell people about it, you make up a story in which the Mulla did it. The form, structure, and characters are well known and an amusing story will spread quickly and naturally within the community, dispersing knowledge with it. Mulla stories continue today, with warning stories about the Mulla meeting British Immigration at Heathrow Airport. Peter Hawkins of Bath Consulting in the UK has written, but not published, a series of Mulla stories for today's managers. The popular Dilbert cartoons are a more modern version of this story form; pinned to walls, emailed, or posted anonymously they provide a powerful, amusing learning mechanism, with bite!
The Mulla Nasrudin stories evolved over many years; for every cartoonist who succeeds there are many who fail. If we want to use archetypes in organizational story, we cannot wait for evolution, nor can we experiment with many options until one succeeds. We need to be able to produce archetypes that resonate with the organization in an efficient and timely manner. Again, the same workshop techniques that were used for organizing principles come into use, but with the addition of a cartoonist and possibly some actors, depending on the planned use of the archetypes. The anecdotal database is once more key: workshop participants converse about the anecdotes, their meaning and relevance, and as this talking increases and more individuals connect with other individuals in multithreaded conversation, characters start to emerge. The cartoonist is there to funnel this dialog into a set of archetypal characters by drawing what he or she hears and then redrawing in dialog with the workshop participants. Far more easily than with organizing principles, a phase shift takes place and the archetypes emerge from the discourse between the participants, focused by the cartoonist. Importantly, an “expert” does not “analyze” the material or use archetypes previously identified as “appropriate” or “best practice” for that industry sector. That would be a Taylorist approach, old wineskins for new wine. Archetypes emerge from the discourse of the community and thus resonate with that community when they are used.
Now we can return to the difficulties and inconsistencies in identifying organizing principles. With the archetypes we have created an additional and useful agent. The archetypes have already focused the extreme aspects of the community and are represented by vivid cartoons. We can create a new discourse between and about the archetypes. Here, actors can be used to enhance the dialog. Actors are not inhibited in role-play, while most members of an organization are. Someone engaging in a role-play also tends to formulate strong opinions about what they said, regardless of what is heard. By allowing workshop participants to coach an actor in role-play, the inhibition is removed and hearing is not impaired. Stories are created about the archetypes, coaching actors into impromptu plays in which they characterize and represent the archetypes. This additional level of discourse provides the organizing principles in the event of difficulties or inconsistencies.
Some of the uses of organizing principles are self-evident; they are a measure of culture and an indication of the nature of the community. The form of their expression is also valuable. Rules, values, and beliefs are different things and have different implications for knowledge creation and community formation.
In the context of organizational storytelling, the value of the organizing principles, and to a lesser extent the archetypes, is in preventing anti-stories. As stated, an anti-story is generally a cynical and spontaneous reaction to a script that is too far away from the reality of life within the organization concerned, or where the powerful act in a hypocritical manner. All organizations have anti-stories, ranging from initiative-weary cynicism to self-righteous indignation. Most internal communication within an organization attempts to create a script based on idealized behavior. This is also a strong tendency in some of the other approaches to storytelling identified earlier. The temptation to tell things how they should be is, or becomes, irresistible.
The issue with organizing principles is that they rarely reflect the official descriptions of organizational culture, but official messages presume the official perception of culture. Now that we know the organizing principles we can manage the story, by controlling the shift in organizing principles. Reinforcing an existing value or achieving a change or modification of a rule are possible without anti-story. Small incremental changes work; the catastrophic changes that are occasionally necessary are more unpredictable in their outcome and more likely to generate anti-story.
Archetype-based stories are very powerful for a variety of purposes. One of the most obvious is that of the original Mulla Nasrudin scenario, namely, as an indirect confessional device. Because the archetypes are drawn from the community, they resonate in day-to-day use and can be incorporated into lessons-learnt programs. Encouraging teams to relate stories using the archetypes as well as telling the “official” story creates a more complex and valuable learning environment. If the same archetypes are incorporated into training programs, induction programs, and the like, the confessional device is institutionalized within the organization. This can produce drastic reductions in training times and increased retention of learning through reference to the archetypes and their stories (Snowden, 2000d).
Another use, linked strongly to concepts of self-organization and descriptive self-awareness, is of archetypes as an indexing tool for oral histories. These are valuable means by which the experience of past employees can be captured in a memorable form; they allow induction times to be shorted by giving employees access to stories that it might otherwise take them months or years to accumulate. They allow employees at all levels to explore alternative views, or investigate likely responses to a situation from different viewpoints. While several companies have started work on oral histories, they have allowed themselves to be constrained by Taylorist thinking. The general tendency is to construct templates and interview guidelines and then go out and gather the material. The reasons for this are understandable: there is a concern about how to index and catalog the database; a desire to ensure that all necessary information is gathered; a need to structure incoming data so that it is easy to analyze. All of this misses both the point and the opportunity. An oral history is a rich repository of stories told and retold across generations. If we look at the evolution of standard stories in communities, we see that they evolve through constant telling and retelling to a form that stands up and delivers a desired message. This cannot be anticipated or designed; it has to evolve.
Designing an environment for that evolution to take place is simple, if we treat the problem as complex rather than complicated. First, storage is cheap and stories can be captured pervasively. Video booths, tape recorders given to key staff before they leave for home each night and collected the next day, children asked to enter a competition to interview their parent about their work and present it to the company: the list is endless if one is creative. Once a reasonable sample of these stories is available, in a workshop environment we create discourse levels to the point at which a limited number of archetypes emerges, say five. Each incoming oral history is then self-classified using a simple point allocation; for example, the storyteller allocates 100 points over no more than three archetypes.
Given this approach, we can take any number of incoming oral histories without involved structured interviews or similar. Now, in using those histories in, say, a web environment, we present the archetypes on the screen, preferably in cartoon format, and under each archetype we place a slider bar. By manipulating the slider bar the user of the system can navigate across a rich database of experiences, discovering material anew on each occasion. Decision makers can look at the stories being told by groups of individuals who categorize themselves in a similar way. We have created a self-organizing learning ecology, though a simple process that will itself give rise to complex behavior.
Complexity-based thinking, whether through direct action or metaphor, is a fundamental shift in the way we think about organizations. It is not the latest in a set of fads or concepts that extend and develop the Taylorist philosophy. Instead, it bounds Taylorism, limiting it to the execution of stable and structured initiatives, just as Newtonian science was bounded but not invalidated by the discoveries of modern physics. This is very difficult for individuals moving into this field, whether practitioners or academics, or the increasing population of individuals who straddle both domains. Academic life and the day-to-day practices of consultancy companies are firmly established in the norms and paradigms of scientific management. The great and understandable temptation is to dress up the new ideas in the models of the old: to put new wine into old wineskins. This may be the only way to get funding, or it may be a necessity for survival. However, the general pattern of human history is that new ideas gain currency only after a degree of martyrdom, or at least the courage to risk being condemned for heresy.
Much knowledge management practice, and the associated failures directly attributable to Taylorist assumptions, has provided an awareness at a high level in many organizations that something is wrong. The old models do not work, or work in different ways. Planned and structured interventions result in unanticipated and surprising consequences. Knowledge extracted from an employee and embedded in a database is discovered to be unusable without the original creator. Investments in large, sophisticated document databases in which experts codify their knowledge provide no return or benefit; they are just used to find contact names. The list is legion, depressing, and hope bearing at the same time. It is no longer necessary to apologize for thinking in terms of complex rather than complicated systems.
If this change is to be consolidated and built on, there are several important changes necessary in the way we think about and intervene in organizations:
Major shifts in thinking are rare. The connectiveness of the web, the breakdown of mechanistic knowledge management, globalization, and all the words beginning with e: all of these signal a change in thinking at least as great as the switch from medieval to renaissance society. A change of this magnitude will always mean that the inquisition of academic and business orthodoxy will offer the Galileo option to pioneers, but it should be resisted. However painful the alternative, putting new wine into old wineskins always results in leakage and spoilage.
Aibel, J. & Snowden, D. J. (1998) “Intellectual capital deployment: A new perspective,” Focus on Change Management, 47(September): 15-20.
Axelrod, R. & Cohen, M. (1999) Harnessing Complexity: Organizational Implications of a Scientific Frontier, New York: Free Press.
Denning, S. (2000) The Springboard, Oxford, UK: Butterworth Heinemann.
Edvinsson, L. & Malone, M. (1997) Intellectual Capital: Realizing Your Company's True Value by Finding its Hidden Brainpower, New York: HarperBusiness.
Gabriel, Y. (2000) Story Telling in Organizations: Facts, Fictions, and Fantasies, Oxford, UK: Oxford University Press.
Nonaka, I. (1991) “The knowledge-creating company,” Harvard Business Review, Nov-Dec: 96-104.
Nonaka, I. & Takeuchi, H. (1995) The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford, UK: Oxford University Press.
Polanyi, M. (1983) The Tacit Dimension, New York: Doubleday.
Senge, P. (1990) The Fifth Discipline: The Art and Practice of the Learning Organization, New York: Doubleday Currency.
Shah, I. (1985) The Exploits of the Incomparable Mulla Nasrudin/The Subtleties of the Inimitable Mulla Nasrudin (2 vols.), London: Octagon Press.
Snowden, D. (1997) “A Framework for Creating a Sustainable Programme in Knowledge Management,” in Business Guide to Knowledge Management, S. Rock (ed.), London: Caspian Publishing.
Snowden, D. (1998) “Thresholds of acceptable uncertainty: Achieving symbiosis between intellectual assets through mapping and simple models,” Knowledge Management, May.
Snowden, D. (2000a) “Organic knowledge management Part I: The ASHEN model: An enabler of action,” Knowledge Management, 3(7, April): 14-17.
Snowden, D. J. (2000b) “The art and science of story or 'Are you sitting uncomfortably?' Part 1: Gathering and harvesting the raw material,” Business Information Review, 17(3): 147-56.
Snowden, D. J. (2000c) “The art and science of story or 'Are you sitting uncomfortably?' Part 2: The weft and the warp of purposeful story,” Business Information Review, 17(4).
Snowden, D. J. (2000d) “Story Telling and Other Organic Tools for Chief Learning Officers and Chief Knowledge Officers,” in D. Bonner (ed.) In Action: Leading Knowledge Management and Learning, London: ASTD (www.astd.org).
Spark Team (2000) Story Telling, Stories and Narrative in Effecting Transition, Spark Press, sparkteam@sparknow.net.
Stewart, T. (1997) Intellectual Capital: The New Wealth of Organizations, New York: Doubleday.