Introduction

King (2006), in the Encyclopedia of Knowledge Management, defines knowledge sharing as 'the exchange of knowledge between and among individuals, and within and among teams, organisational units, and organisations' (p. 498). He distinguishes between knowledge sharing and knowledge transfer: 'transfer implies focus, a clear objective, and unidirectionality, while knowledge may be shared in unintended ways multiple-directionally without a specific objective' (p. 493). However, he acknowledges that the terms are often used interchangeably. In this paper, this latter convention is followed, and knowledge sharing is considered to encompass both well-focused and less focused knowledge transactions.

Knowledge sharing occupies a central position in the field of knowledge management. Indeed, for some (e.g. Bosua & Scheepers, 2007), it seems almost the single key issue of the field. It is a crucial enabler of the processes described by Nonaka & Takeuchi (1995) in their influential SECI spiral of knowledge creation. Yet, as Berends (2005) observes, 'current literature predominantly treats knowledge sharing as a black box' (p. 97).

This paper reviews some issues associated with knowledge sharing, and identifies what the author considers to be potentially fruitful avenues of future research. Four areas, in particular, are explored: the common distinction between tacit and explicit knowledge; the use of stories as knowledge sharing tools; the role of communities of practice in knowledge sharing; and organisational learning and memory.

Beyond the tacit–explicit dichotomy

Conventional treatments of knowledge management frequently invoke the distinction between tacit and explicit knowledge (see, e.g. Hislop, 2005). Explicit knowledge, being encodable and expressible, is in principle straightforward to share, while tacit knowledge is not. Exactly how difficult it is to convert tacit knowledge into explicit, shareable, knowledge is subject to debate (Gourlay, 2006). Some argue that such conversion is impossible, others argue that it is merely difficult or impractical. Wilson (2002), in a robust critique of the field of knowledge management that deserves wider exposure, observes that the common understanding of tacit knowledge popularised by Nonaka & Takeuchi (1995) appears to be based on a misunderstanding or distortion of Polanyi's idea of tacit knowing. For Polanyi (1967), tacit knowing describes the way in which, for an individual, the sensed world is integrated and interiorised into a subjectively known world. Such tacit knowing, which underlies all knowledge and all skill, is, by its very nature, hidden and inexpressible. Wilson suggests that the non-explicit knowledge which is often the central concern of knowledge management is better understood as implicit knowledge: knowledge that can, with varying degrees of effort and ingenuity, be made explicit. Collins (2007) illustrates Polanyi's point when he notes that, contrary to popular belief, it is both possible to specify explicitly how to ride a bicycle, and to create a machine that successfully implements such a specification. Under normal conditions, however, humans cannot use these rules to discover how to ride a bicycle; they have to develop their own tacit knowledge. This is a key point of this characterisation of tacit knowledge: it cannot be shared, but must be developed within the individual. When we make reference to sharing tacit knowledge, we must in reality be referring to setting up conditions in which individuals can develop their own tacit knowledge. We do not explain to our daughters and sons how to ride their bicycles; we provide conditions in which they can develop this know-how themselves.

Collins (2007) suggests that a sieve possesses tacit knowledge of its own ability to distinguish between the materials that it can separate. In this case, the tacit knowledge is built into the structure of the sieve. More generally, it might be suggested that any entity possesses the tacit knowledge to be itself. A pebble is expert at being a pebble. This might seem to be an alarmingly animistic view (the authors recall this suggestion being met with polite bemusement at a conference some years ago), and some might seek a less provocative term, but so far the author has not been able to identify a satisfactory one. The problem here appears to lie with the common and intuitively reasonable axiom that knowledge must be possessed by a thinking, conscious, knower.

According to the argument being pursued here, tacit knowledge is ubiquitous. All entities possess it. It may be intrinsic, as for the pebble. It may be deliberately introduced, as for the sieving property of the sieve: in this case, the sieve is designed accordingly to decidedly non-tacit understanding that is embodied in the structure of the sieve – it is tacit knowledge only for the sieve itself. In living organisms that can learn, and in humans in particular, tacit knowledge can develop within the entity. In contrast, explicit knowledge is encountered far more rarely. Only humans (and, possibly, some machines and, in rudimentary form, some higher forms of animal life) can produce explicit knowledge (as observed by Tallis, 1999). However, an implication of Polanyi's characterisation of tacit knowing is that explicit knowledge is not separable from tacit knowledge: to be known by an individual, explicit knowledge requires foundation in appropriate tacit knowledge (an argument, which might undermine the possibility of machine knowledge).

The above discussion has various implications. One is that much of the effort to develop ways of transforming tacit knowledge into explicit knowledge and back again, to drive Nonaka and Takeuchi's (1995) SECI spiral, is more properly seen as directed towards transformations between explicit and implicit knowledge. But, though this may be valuable, it will not be enough. Real know-how is embedded in tacit knowledge, and such knowledge, if not already possessed, must be developed by the individual himself or herself. Thus, what is required is facilitation of personal tacit knowledge development. It is suggested that the very best way of tacit knowledge development is by first-hand experience: learning-by-doing. However, the provision of such experience is frequently impractical. Under such circumstances, the simulation, to a greater or lesser extent, of such experience is an alternative. The simulation of experience may be achieved in a variety of ways, each offering both pros and cons, such as training exercises and thought experiments. The next section discusses one of the most pervasive of methods for providing simulated experience: the use of stories.

Stories: vicarious experience

Many authorities have emphasised the importance of stories in sharing knowledge, and in particular in sharing tacit knowledge. In the light of the argument presented in the previous section, we may ask: do they really mean tacit knowledge, or are they simply referring to implicit knowledge? The answer is probably: both.

Stories can clearly allow the sharing of explicit knowledge, though whether this is always a more efficient way of sharing such knowledge than other methods is debatable. The advantage offered by stories is much clearer in the way in which they allow implicit knowledge to be shared, by invoking the framework of assumptions possessed by their audience. Thus, a fairly sparsely told story may nevertheless invoke a richer knowledge in listeners by encouraging them to fill in the gaps (interpolate) and go beyond the story (extrapolate). For example, I might tell a story in which I recount briefly how I travelled by rail from Southampton to Edinburgh. Listening to my recalling that I travelled from London by the East Coast Main Line might lead our audience to conclude (correctly) that I crossed London from Waterloo Station to King's Cross Station. But members of the audience who were unaware that Southampton and Edinburgh were served by different London termini (because I had left this implicit in my story) might have real difficulty if they attempted to duplicate my journey.

This example serves as a warning for those who would use storytelling to share implicit knowledge. The knowledge, which might be shared becomes less predictable if the background understanding of the audience is not as the teller expects. In many circumstances, the background understanding of an audience might be genuinely impossible to gauge. Parkin (1998) suggests that the use of 'universals' in stories can lead to the right message getting across (she uses the example of a stream as a metaphor for the future development of an organisation); while some knowledge is undoubtedly more universal than other, it is suggested that the notion that there are any true universals be treated with caution.

The mechanism by which implicit knowledge is shared in storytelling is by engineering the development and surfacing of that knowledge in the mind of the receiver. Generally, the intent is that the knowledge that is surfaced in this way replicates that of the teller (as could be tested if both receiver and teller were to make the knowledge explicit), and use of the term sharing still seems appropriate. However, as noted in the previous section, non-tacit knowledge is not separable from tacit knowledge, and sharing non-tacit knowledge with a recipient will change the tacit knowledge of the recipient. Since tacit knowledge is both personal and inexpressible, it might seem more appropriate to refer to tacit knowledge generation rather than sharing.

Why are stories effective knowledge sharing devices? Many authors suggest reasons: Parkin (1998) cites the focus of Evans & Evans (1989) on concretizing, assimilation, and structurizing, and Ortony (1993) on identification of compactness, vividness, and expression of the inexpressible. Davenport & Prusak (2000), following Weick (1995), point to the way in which stories 'embody experience and apply it to future expectations, their basis in feeling and thought, their essential humanness' (p. 82). The position adopted in this paper is that the effectiveness of stories is related to the way in which they provide a context in which the (real or imagined) experience of their tellers can be experienced vicariously by their receivers. The vicarious experience that stories provide accounts for their power in enabling the sharing/generation of not just tacit, but also explicit and implicit, knowledge. A story is not the experience, but in the hands of a skilled storyteller it may provide a good second-best. Generalising, it is contended that it is the narrative quality of teaching methods that is the driver of their effectiveness as educational vehicles – narrative threads underlie case studies, training exercises, and even well-structured lectures, although arguably it is in stories that they are encountered in their purest form. 'Living' a narrative provides a natural context, which encourages the assimilation and accommodation (Piaget, 1972) of all types of knowledge. The intensity of vicarious experience that an educational vehicle, such as a story, provides is surely related to qualities such as vividness and accessibility.

The judicious use of stories as knowledge sharing/generation instruments implies an imposition by the storyteller of an appropriate degree of control lest the 'wrong' knowledge be shared (i.e. the story is misunderstood, or just not fully understood). This is generally due to a mismatch between the story (and the context in which it is told) and the receiver – as in the rail-travel example cited above, where inappropriate assumptions about the frame of the receiver might have been made. Connell et al. (2004) found it useful to classify the purposes of storytelling by adapting a framework originally devised by Bowen (1978) for the classification of games according to their purpose. (It is perhaps unsurprising that Bowen's classification lends itself to this adaptation, since games have a strong narrative content.) Stories intended to educate, in order to convey the required knowledge, require control of the storytelling context, and tailoring to the particularities of the receivers. Stories intended to enrich the receivers are also created to correspond to their particularities, but because no specific message is to be conveyed, control is relaxed and receivers are free to make what they wish of the stories. Stories intended to produce a particular effect need to be delivered in a controlled way, and individuals for whom those stories will work in the intended way are selected appropriately. Stories for entertainment are not controlled, but their receivers are, similarly, selected (very often, self-selected).

While much literature supports the use of stories as knowledge sharing/generation tools, it is suggested that systematic research on the effects of stories has yet to be done. Such research might explore the effects of changing the conditions under which stories are delivered to their receivers, as well as more specific hypotheses, which might be suggested by the literature. Denning (2004), for example, suggests that to be effective, the burden of a story must be true – which is understood to mean that it is credible, both to its teller and its recipients. Developing methods for investigating the use of stories rigorously is, however, challenging.

The context of knowledge sharing: communities of practice

The knowledge management literature recognises the importance of communities of practice in the creation and maintenance of knowledge within organisations (Brown & Duguid, 1991; Hislop, 2005), placing some emphasis on the role of communities of practice as the 'missing link' between, on the one hand, individual learning, and, on the other hand, organisational learning. Exactly what communities of practice are conceived to be, however, has shifted since the concept was first introduced by Lave & Wenger (1991). Their original emphasis was on the role of communities of practice in integrating individuals into the life of a community bonded by a particular domain of interest. Subsequently, the perceived role of communities of practice has been expanded into a more general one of domain knowledge management, and this expansion has been accompanied by a shift from understanding communities of practice as purely 'natural' phenomena, to viewing them as 'cultivatable' (Wenger et al., 2002) or even 'creatable' (Swan et al., 2002).

Bogenrieder & Nooteboom (2004) have characterised communities of practice as but one type of 'learning group', while Engestrom (2001) notes that in some contexts relatively stable communities of practice cannot be usefully identified. In one way or another, such treatments have recognised that the groupings in which learning takes place are varied. Recognition of the potential variety in natural, cultivated, or artificial groups of learning individuals led Klein et al. (2005) to identify what they considered to be two important dimensions of communities of practice. They argued that the degree to which a community exhibits stratification – that is, different levels of knowledge expertise, which may or may not correspond to formal organisational ranks or titles – is an important determinant of community life. In stratified communities, experts tend to share their knowledge with novices, rather than vice versa. In more egalitarian communities, there is more equal sharing. An expected consequence is that stratification leads to slower evolution of knowledge within the community. The other dimension Klein et al. identified is that of knowledge activity. At one extreme, a community of practice may be concerned with sharing knowledge – in activities directed at allowing members to share/generate knowledge among themselves. (Storytelling is such an activity.) At the other, the concern of the community may be with nurturing knowledge – by providing contexts in which members may learn from first-hand experience. A community of musicians, for example, by offering opportunities for performance within the community, allows the development of, in particular, musical presentation skills (Davidson, 2002).

The positioning of a community of practice on the activity dimension has implications for the type of knowledge with which the community deals. Within a knowledge sharing community, the shared knowledge tends to be founded in the explicit (although it is likely to invoke implicit knowledge, and will have tacit effects). Within a knowledge nurturing community, non-explicit, and, in particular, tacit, knowledge is typically central to the nurturing activity. Musical presentation skills, for example, have a strong tacit component. A community of musicians that shared stories of performance would no doubt be valuable to many of its members, but this activity would not be a substitute for the opportunity of actual performance.

It is argued, therefore, that different situations require different types of community of practice. This is especially important in a climate where knowledge managers are increasingly thinking in terms of cultivating or even creating communities of practice, rather than simply letting them develop naturally, and, heeding Brown & Duguid's (1991) warning, avoiding damaging them when they do develop. If organisations are going to invest resources in engineering communities of practice, then they need to consider what those communities are for and how to create communities that match their intended purposes. Some preliminary guidelines are suggested by Klein & Connell (2007). However, there is much scope for further research in this area, underpinned by well-explored case study work.

Organisational learning and memory

Conventional treatments of organisational learning tend to relate it to individual learning. A fine example is Kim's (2004) OADI-SMM model, which relates a cyclical model of individual learning based on experiential learning theory (Kolb, 1984) to March & Olsen's (1975) model of organisational learning. Central to Kim's approach is the characterisation of organisational learning as development of shared mental models within the organisation. Such approaches emphasise individual learning as the root of organisational learning. A new perspective on organisational learning, however, has, in turn, placed individual learning in the social context of a group of individuals: the community of practice (as discussed in the previous section). This perspective not only illuminates the social aspects of how individuals learn, but provides insight into how learning spreads throughout a community and becomes organisational.

The knowledge that an organisation accrues can be regarded as the content of organisational memory (Walsh & Ungson, 1991). Klein et al. (2008) have contrasted two views of organisational memory. Characterising memory as a repository of accumulated commodified knowledge leads to a rather static view of memory, which tends to impede organisational flexibility and adaptability. In contrast, an understanding of memory as linked to the ongoing activities of organisation promotes flexibility. Klein et al. suggested that such an organisation is 'Markovian', having no memory beyond that which is implicated in current organisational activity. In this view of organisational memory, knowledge is 'living' and accessible to what might be termed the organisational 'consciousness'. An organisation which adopts this image of memory as a guide to the way in which it operates needs to address its ability to learn in 'real time': for its members to be able to make sense of a changing environment, and for their sense making to spread rapidly throughout the organisation, or at least to those parts where it is needed. This places a burden upon the communities of practice or learning groups that comprise the organisation. In such organisations, the need to foster and promote such communities is particularly strong. It is suggested that the ability of an organisation to operate flexibly in a rapidly changing environment is crucially linked to the health of its communities of practice. It would be interesting to test such a hypothesis empirically.

Conclusions

In this paper, the author has set out a personal view of the field of knowledge sharing, and identified what he considers to be some of the more interesting issues associated with it. Despite the wealth of literature on knowledge sharing that now exists, understanding of the field is broad-brush: the details lack definition and are sometimes absent altogether. There is much potential for further research in the area, and particularly for research with a strong empirical foundation that is likely to increase the resolution of our picture of knowledge sharing. This paper has suggested some possible lines along which such research might proceed.

However, the methodological difficulties in pursuing such research should not be underestimated. The importance of empirical work as the basis for understanding should be emphasised. Most of the empirical material in the area to date has taken the form of case studies. The value of such research is unquestionable, but there is room for other kinds of research, of a more systematic nature, to complement the naturalistic research that dominates the field, such as controlled experimentation (for a more detailed argument along these lines, in a somewhat different context, see Klein, 1994). It is recognised that to some, these remarks amount to cross-paradigmatic heresy. The author takes a more relaxed view.

Understanding the realities of how knowledge works in organisations has some particular characteristics that make it extremely challenging. Tacit knowledge, being inexpressible, cannot be identified directly. How can research on the mechanics of tacit knowledge rise above mere speculation? We may be facing the same problem as those physicists who, exploring the fundamental nature of matter and the universe, formulating hypotheses that are apparently empirically untestable, have been accused of working in religion rather than science. But who said it was going to be easy?

References

1. Berends H (2005) Exploring knowledge sharing: moves, problem solving and justification. Knowledge Management Research & Practice 3, 97–105. | Article |
2. Bogenrieder I and Nooteboom B (2004) Learning groups: what types are there? A theoretical analysis and an empirical study in a consultancy firm. Organization Studies 25, 287–313. | Article |
3. Bosua R and Scheepers R (2007) Towards a model to explain knowledge sharing in complex organizational environments. Knowledge Management Research & Practice 5, 93–109. | Article |
4. Bowen KC (1978) Research Games: An Approach to the Study of Decision Processes. Taylor & Francis, London, UK.
5. Brown JS and Duguid P (1991) Organizational learning and communities-of practice: towards a unified view of working, learning and innovation. Organization Science 2, 40–57.
6. Collins H (2007) Bicycling on the moon: collective tacit knowledge and somatic-limit knowledge. Organization Studies 28, 257–262. | Article |
7. Connell NAD, Klein JH and Meyer E (2004) Narrative approaches to the transfer of organisational knowledge. Knowledge Management Research & Practice 2, 184–193. | Article |
8. Davenport TH and Prusak L (2000) Working Knowledge: How Organizations Manage What They Know. Harvard Business School Press, Boston, Massachusetts, USA.
9. Davidson J (2002) Developing the ability to perform. In Musical Performance: A Guide to Understanding (RINK J, Ed), pp 89–101, Cambridge University Press, Cambridge, UK.
10. Denning S (2004) Squirrel Inc.: A Fable of Leadership through Storytelling. Jossey-Bass, San Francisco, CA, USA.
11. Engestrom Y (2001) Expansive learning at work: toward an activity theory reconceptualization. Journal of Education and Work 14, 133–156. | Article |
12. Evans G and Evans R (1989) Cognitive mechanisms in learning from metaphors. Journal of Experimental Education 58, 5–19.
13. Gourlay S (2006) Towards conceptual clarity for 'tacit knowledge': a review of empirical studies. Knowledge Management Research & Practice 4, 60–69. | Article |
14. Hislop D (2005) Knowledge Management in Organizations: A Critical Introduction. Oxford University Press, Oxford, UK.
15. Kim DH (2004) The link between individual and organizational learning. In How Organizations Learn: Managing the Search for Knowledge. (STARKEY K, TEMPEST S and MCKINLAY A, Eds), 2nd edn, pp 29–50, Thomson, London, UK.
16. King WR (2006) Knowledge sharing. In The Encyclopedia of Knowledge Management (SCHWARTZ DG, Ed), pp 493–498, Idea Publishing Ltd, USA.
17. Klein JH (1994) Cognitive processes and operational research: a human information processing perspective. Journal of the Operational Research Society 45, 855–866. | Article |
18. Klein JH and Connell NAD (2007) The identification and cultivation of appropriate communities of practice in higher education. In Communities of Practice: Creating Learning Environments for Educators (KIMBLE C and HILDRETH P, Eds) Information Age, Greenwich, CO, USA.
19. Klein JH, Connell NAD and Meyer E (2005) Knowledge characteristics of communities of practice. Knowledge Management Research & Practice 3, 106–114. | Article |
20. Klein JH, Connell NAD and Jasimuddin S (2008) Who needs memory? The case for the Markovian organisation. Knowledge Management Research & Practice 5, 110–116.
21. Kolb DA (1984) Experiential Learning: Experience as the Source of Learning and Development. Prentice-Hall, Englewood Cliffs, NJ, USA.
22. Lave J and Wenger E (1991) Situated Learning: Legitimate Peripheral Participation. Cambridge University Press, Cambridge, UK.
23. March JG and Olsen JP (1975) The uncertainty of the past: organizational learning under ambiguity. European Journal of Political Research 3, 147–171. | Article |
24. Nonaka I and Takeuchi H (1995) The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation. Oxford University Press, New York, USA.
25. Ortony A (Ed) (1993) Metaphor and Thought. Cambridge University Press, Cambridge, UK.
26. Parkin M (1998) Tales for Trainers. Kogan Page, London, UK.
27. Piaget J (1972) Intellectual evolution from adolescence to adulthood. Human Development 15, 1–12.
28. Polanyi M (1967) The Tacit Dimension. Routledge & Kegan Paul, London, UK.
29. Swan J, Scarbrough H and Roberston M (2002) The construction of 'communities of practice' in the management of innovation. Management Learning 33, 477–496. | Article |
30. Tallis R (1999) On the Edge of Certainty: Philosphical Explorations. Macmillan, Basingtoke, UK.
31. Walsh JP and Ungson GR (1991) Organizational memory. Academy of Management Review 16, 57–91. | Article |
32. Weick KE (1995) Sense making in Organizations. Sage, Thousand Oaks, CA, USA.
33. Wenger E, Mcdermott R and Snyder WM (2002) Cultivating Communities of Practice: A Guide to Managing Knowledge. Harvard Business School Press, Boston, MA, USA.
34. Wilson TD (2002) The nonsense of 'knowledge management'. Information Research 8(1), paper no. 144 (Available at http://InformationR.net/8-1/paper144.html).

Acknowledgements

I thank my colleague Professor Con Connell, Director of the Centre for Narrative Studies at the University of Southampton, for his support and encouragement in the preparation of this paper, and his highly valued collaboration in much of the work preceding it.

About the author

Jonathan H. Klein is a Senior Lecturer in the School of Management at the University of Southampton. He is Director of the School's MSc Programme in Knowledge & Information Systems Management. His research interests include the internal characteristics of communities of practice, storytelling as a knowledge sharing mechanism, and the nature of organisational memory. He is a participant in the RIGHT (Research into Global Healthcare Tools) project, a multi-university project supported by the EPSRC.