COMMUNITIES OF PRACTICE

What is a Community of Practice?

Jean Lave (Lave, 1991; Lave & Wenger 1991) is credited with first introducing the term of Communities of Practice (CoP) (Wenger, 1998). Lave and Wenger (1991) described a CoP as:

…a set of relations among persons, activity, and world, over time and in relation with other tangential and overlapping communities of practice (p. 98).

They illustrated it with five examples of apprenticeship, emphasising that their notion of apprenticeship is not restricted to the historical idea of apprenticeship in a trade. Rather, they view it as a form of socialisation into a community, where the newcomer gradually becomes a legitimate member of the CoP by learning the practice, the language, and the conventions of the community by having access to and interacting with established members. The key point here is the emphasis on the social as opposed to information and knowledge. The learning a newcomer undertakes is situated in and cannot be separated from the practice of the community. In explaining this, Lave and Wenger emphasise the context of the community for learning and knowledge, using the example that even if a general rule is known, it does not necessarily mean that it can be applied in relevant specific circumstances; that is, the "power of abstraction" is situated in the culture of the community.

Their "apprenticeship" examples are from different traditions and cultures and take different forms—formal arrangements, informal arrangements, and a community where membership was purely voluntary—but in all of them what is common is the importance of the access that newcomers have to old-timers who are long standing members of the community and from whom newcomers will learn. Their five examples were Yucatec midwives, Vai and Gola tailors, quartermasters, butchers, and non-drinking alcoholics. For the tailors, there was a formal arrangement between the apprentice and a single master; the quartermasters were part of a formal arrangement; the apprenticeship of the butchers followed the historical notion of apprenticeship more closely, but it was also the one that did not work as it should because of the lack of access to the old-timers. The apprentices were taught specific tasks, that is, the hard aspects of their domain knowledge, but in being denied access to interaction with old timers, they were not being given the opportunity to develop the softer aspects of their knowledge in the context of the community. The Yucatan midwives had an informal arrangement—in many cases the newcomer was the daughter of an existing midwife who would accompany her mother. The nondrinking alcoholics also had a purely voluntary "apprenticeship." The process is from alcoholic to non-drinking alcoholic. An "apprentice" will attend meetings several times a week. There will be other people at the meetings, who are at different stages in the process and who, together, make up the community. In general meetings, the "old-timers" tell stories of their lives as alcoholics. These stories can have taken years to reach the stage of polish at which they are presented. In discussion meetings, the focus tends to be on one smaller part that will eventually be part of the whole of the life story. There are also 12 steps through which the newcomer must go. These steps basically represent the process from peripheral to full participation in the community. For the absolute newcomer, the first participation, apart from attending the meeting, may be as simple as a declaration of intent to abstain from drinking in the next twenty-four hours, in the form of picking up a white chip at the end of a meeting. From there, the newcomer will develop his/her story and pass through the "12 Steps," which culminates in being recognised as an old-timer and trying to persuade someone else to become a member of the community.

All of Lave and Wenger's (1991) examples show this difference between newcomer and old-timer and the process of moving from one way of being to another. The members achieve this shift in state by learning the practice of the community. They do this by participating in the practice of the community, gradually moving from peripheral to full participation. This process is what Lave and Wenger term Legitimate Peripheral Participation (LPP).

Legitimate Peripheral Participation

For Lave and Wenger (1991), LPP is the defining characteristic of apprenticeship as a form of learning. Newcomers learn the practice of the community by being situated in the practice and by having access to established members. LPP is part of the process by which a newcomer becomes an established member of a CoP.

In CoPs, newcomers learn from old-timers by being allowed to participate in certain tasks relating to the practice of the community and gradually move from peripheral to full participation in the community. For example, the daughter of the Yucatec midwife may start simply by knowing what the life of a midwife involves, such as having to go out at all hours, the kinds of herbs that need to be collected, etc. The midwife may then start taking her daughter with her on visits. The daughter will hear stories of cases. She may then proceed to running errands and taking messages. Eventually, she may assist at a birth or give a massage to an expectant mother. As time goes on, if she has decided to pursue this line of work, she will gradually take over more and more of the workload.

Lave and Wenger (1991) saw a Community of Practice as "an intrinsic condition for the existence of knowledge" (p. 98). They saw the learning that took place in such communities not as narrow situated learning where instances of practice are simply replicated but "learning as Legitimate Peripheral Participation" (p. 34). LPP is not merely learning situated in practice but learning as an integral part of practice: learning as "generative social practice in the livedin world" (p. 35).

LPP is complex and composite in character. Lave and Wenger (1991) state that each of its three aspects—legitimation, peripherality, and participation—is indispensable in defining the others and can not be considered in isolation. Legitimation and participation define the characteristic ways of belonging to a community, while peripherality and participation are concerned with location and identity in the social world. Although Lave and Wenger stress the composite character of LPP, it is useful as an analytical convenience to consider the three components and their relationships separately.

Legitimation is the dimension of CoPs that is concerned with power and authority relations in the group. In Lave and Wenger's examples, legitimation does not necessarily have to be formal. For example, for quartermasters, tailors, and butchers, there is a degree of formal legitimacy that comes from hierarchy and rank, but for the midwives and alcoholics, legitimacy is more informal. For example, the alcoholics gain legitimacy, as the stories they tell of their experiences become more mature and closer to those of an old-timer. The legitimation bestowed by a story was also noted by Orr (1997), who described the stories as artefacts for sharing experience.

Once war stories have been told, the stories are artefacts to circulate and preserve. Through them experience becomes reproducible and reusable…They preserve and circulate hard won information and are used to make claims of membership or seniority within the community…They also amuse, instruct and celebrate the tellers' identity as technicians. Such tellings are also demonstrations of one's competence as a technician and therefore one's membership in the community (p. 26).

Peripherality is neither a physical concept as in core and periphery nor a simple measure of the amount of knowledge that has been acquired. Lave and Wenger (1991) use the terms peripheral and full participation to denote the degree of engagement with and participation in the community, but note that peripherality "must be connected to issues of legitimacy of the social organisation and control over resources if it is to gain its full analytical potential" (p. 37).

For Lave and Wenger (1991), it is participation that provides the key to understanding CoPs. CoPs do not necessarily imply co-presence, a welldefined or identifiable group, or socially visible boundaries. However, CoPs do imply participation in an activity about which all participants have a common understanding of what it is and what it means for their lives and community. The community and the degree of participation in it are in some sense inseparable from the practice.

Expanding the Model

Lave and Wenger (1991) emphasised that CoPs were not restricted to an apprenticeship model, and other researchers have attempted to extend the concept in order to apply CoPs in a KM setting, that is, by applying the concept to commercial organisations and regarding them as a new organisational form. Some researchers have even used the term to refer to the wider community of a specific group of practitioners. For example, Hutchins and Klausen (1991) refer to the Community of Practice of pilots and their shared knowledge. Classifying such a large community as a CoP gives us two tiers of CoPs—the wider community (for example, of pilots) and more compact CoPs (for example, a team of pilots who work for a particular airline and who are based at a particular airport). Clark (1996) provides a basis for both types of CoP with the notion of Common Ground described earlier. When using language within a community, the Common Ground forms part of the basis of the communication and can manifest itself in conventions within the community. Common Ground will also include terms and jargon—a language shared by members of a particular community.

Seely Brown and Duguid (1996) have noticed similar aspects when exploring the role of CoPs in learning. Their examples of learning how to become a physicist or a football player do not only entail learning formulae and plays. The newcomer must "learn how to act as one, talk as one, be recognised as one—it's not the explicit statements, but the implicit practices that count" (p. 13).

The people who fail to learn the language and practices of the community, that is, those who fail to become a member, are those who simply learn the outside information (that is, the harder aspects) and thus give themselves away as outsiders. Indeed, it is often only knowing the explicit statements that betray an outsider. This does, however, show the importance of the Common Ground, in the form of developing the softer aspects of knowledge in the context of the community.

Common Ground alone, however, is only one constituent of a CoP—in extending the concept, different characteristics have been identified and many definitions have been suggested. Manville and Foote (1996) place their version of CoPs firmly in the organisational context and emphasise the informality that may be found in CoPs. The CoP does not have to be an official grouping. Membership may be totally voluntary. They also allude to the tackling of common problems and refer to the knowledge that can be created within the group through the tackling of those problems. Concei o, Gibson, and Shariq (1997) also restricted their version of a CoP to an organisational context, referring to the knowledge that is to be found in a CoP and observing that CoPs can be cross-functional, in that they are not restricted to specific business functions. They cross the borders between the different parts of the organisation.

Seely Brown and Duguid (1991, 1998) took the notion of CoPs and applied them to a study of technical photocopy technical repairers undertaken by Orr (1990). They describe the copier repairers as a CoP and relate how narration, in the form of war stories (as described in Chapter II), was used to help solve problems and create new knowledge. The story might be told at a later stage and become part of the community's stock of knowledge. Lave and Wenger (1991) explain the importance of stories by relating them to knowledge being situated in the context of a community.

…any "power of abstraction" is thoroughly situated, in the lives of persons and in the culture that makes it possible. On the other hand, the world carries its own structure so that specificity always implies generality…That is why stories can be so powerful in conveying ideas, often more so than an articulation of the idea itself. What is called general knowledge is not privileged with respect to other "kinds" of knowledge. It too can be gained only in specific circumstances. And it too must be brought into play in specific circumstances. The generality of any form of knowledge always lies in the power to renegotiate the meaning of the past and the future in constructing the meaning of present circumstances (p. 34).

In Orr's (1990) example, the story is heard being re-told in the lunchroom and Seely Brown and Solomon Gray relate how technical repairers gather round the coffee pot to swap stories. They point out that whereas many people might regard this as "dead" time, the repairers were in fact carrying out valuable work—as field service is a social activity, they were swapping and producing insights as to how better to repair the machines. Davenport and Prusak (1998) widen this to ad hoc conversation in general, showing that informal ad hoc communication is an essential part of interaction within a community, during which valuable work gets done. They observe that when people are having a conversation at the water cooler, the coffee machine, or in the company cafeteria knowledge transfer is often taking place. Unfortunately, some managers often regard this as wasted time, but in general most of the conversations will be about work—current projects, solving problems, asking advice, etc.

The extension of the concept of a CoP has led to some inappropriate usage of the term. For example, Lindstaedt (1996) referred to them as project teams and interest groups, and Sandusky (1997) equated them to business functional units. It is wrong to define CoPs in these terms, for project teams, interest groups, and business function units are formally constituted groups. CoPs per se are not formally created. It may be the case that a formally constituted group, such as a project team, develops into a CoP by virtue of the relationships between its members, but this only shows that such a group can develop into a CoP, not that all such groups are CoPs.

Putting Together a Covering Definition

In order to draw the CoP characteristics together it is perhaps best to regard the term Community of Practice as an umbrella term, covering a range of groups, some of which might have more of some characteristics than others.

Figure 1 shows that there are a number of characteristics that would appear to be present in most CoPs to a greater or lesser degree:

Figure 1: Community of Practice Characteristics

Common Ground: Clark (1996) classified and defined cultural communities by the use of Common Ground—the knowledge, beliefs, and suppositions shared by members of the community. As described in Chapter III, this provides us with two tiers of CoP—at a higher level, for example, a community of dentists who share a common language, background, and experiences, and at a lower level, there is the smaller CoP. In the case of the dentists, it may be a group of dentists who work in a particular practice and who have evolved into a CoP. They may have newcomers, practicespecific jargon, and nicknames. They may have developed strong working relationships and have a common goal for the practice towards which they are working. They will also have practice-specific Common Ground, for example, relating to particular incidents or tricky operations within the practice.

The members of the CoP have some sort of shared or background knowledge. This is generally, but not necessarily, related to the common interest of the group. Nardi and Miller (1991) term this the "domain knowledge." Of more interest here is the soft knowledge that forms part of the Common Ground.

Common Purpose/Motivation: The members of the group will feel that they have a common purpose, which will give the group an internal impetus. The motivation of the members of a CoP is also a key element of what makes it a CoP. There is an internal motivation that provides the group with its impetus. Even when the group is formally constituted, part of what makes the group a CoP will be the internal motivation that drives the group rather than any external pressures. The CoP members are self-motivated in what they do within the CoP.

Legitimate Peripheral Participation: (Lave, 1991; Lave & Wenger 1991) As described earlier, LPP is the process by which a newcomer to the group gradually becomes an established member of the CoP. In Lave and Wenger's (1991) examples, this was based on a form of apprenticeship. Even in CoPs where a newcomer already has a degree of domain knowledge, there will be group- and organisation-specific aspects that the newcomer has to learn as (s)he works him/herself into the group. This might be, for example, the particular practices of the group and organisation, "the way things are done here," or relevant people to contact.

Fluidity/Regeneration: The fluidity of the CoP refers to the arrival of newcomers and the eventual departure of old-timers. Newcomers arrive, learn the language, domain knowledge, and practices of the group, go through the process of LPP, and themselves become acknowledged oldtimers. This is how the CoP regenerates itself and ensures its stock of knowledge is not lost.

Evolution: A CoP will go through some sort of evolution. It may be simply that the CoP owes its very existence to evolution in that it developed through a group of people having a common interest. On the other hand, the group may originally have been formally created but has developed or evolved into a CoP as a result of the relationships, inner momentum, and feeling of community that have developed into an identity of its own over time.

Relationships: The relationships are a key part of what makes a CoP. It is possible for a team to become a CoP as informal relationships begin to develop and the source of legitimation changes in emphasis. The relationships that develop are central to developing the sense of trust and identity that defines the community.

Community/Identity: The internal motivation and the development of relationships contribute to a feeling of community and identity. The members feel they belong to the community. They may even go as far as giving the group a name.

Narration: Narration takes the form of story telling and is seen by some commentators as a key means by which CoP members share knowledge (Goldstein, 1993; Orr, 1990; Sachs, 1995). This was shown as a central part of the transition from newcomer to old-timer in Lave and Wenger's (1991) community of non-drinking alcoholics. Orr (1997) also showed that the qualities of the stories bestowed members of his copier repair engineers with a form of legitimation and confirmed their status in and membership of the community.

Dynamism/Creation of New Knowledge: The dynamism aspect relates to the social distribution of the knowledge in the group. Over a period of time, as the group's work progresses, the members will learn different things at different paces. There will be members of the group who have specialised knowledge in certain aspects, that is, they are "gurus" (Star, 1995). Other members will be able to consult them for help, but as time passes, the social distribution of the knowledge will shift – it is dynamic, not static. This dynamic process can also create new knowledge as demonstrated by Orr's (1990, 1997) photocopier technicians. In swapping stories and bouncing ideas off each other, the two technicians finally arrived at a solution that was then added to the community's stock of knowledge.

Informal: It is often the case in a CoP that there is no hierarchy and the group has no specific deliverable. The group is run on informal lines. This has close links with the legitimation aspect (in LPP). In such an informal group, the legitimation is bestowed by the informal relationships and not by an externally imposed rank. In the case of an officially created group that has evolved into a CoP, the official hierarchy of the original group will still be present and functioning, and such a group may still have deliverables. This factor may then be less essential than some of the other aspects. It is particularly interesting, however, when the informal legitimation and externally imposed rankings clash, as in the Chief Petty Officers of Hutchins (1995a) having to "break in" the new officers.

Unofficial: In many cases a CoP is not formally created by an organisation. The example of the technicians in Orr (1990) is a good example – the company was unaware of the technicians' ways of working and collaborating. It can also often be found that membership in a CoP is voluntary, possibly because the CoP is unofficial. This does not, however, preclude voluntary membership of an official group or team.

Similar Jobs: This characteristic is where a CoP is situated in an organisational context and refers to the identification of CoPs. In many cases a CoP may form around a particular job, the job is the practice—as in the example of photocopier repairers provided by Orr (1990), or, on a wider level, the ophthalmologists suggested by Clark (1996). The fact that CoPs may form around similar roles or jobs in an organisation provides a starting point when looking for potential CoPs, particularly in a distributed environment where they might not be so obvious. There will be individuals doing similar jobs and closer contact would be beneficial to them.

In a Community of Practice it is important to emphasise the social aspects, which feature heavily. Several of the characteristics described above point to the importance of the social features. A CoP cannot be created, merely facilitated, for without the internal motivation and the relationships that the members develop, a group will fail to evolve into a CoP. This is demonstrated particularly well by the Legitimation aspect where the legitimation of a member in the community comes not from some formal and externally imposed hierarchy but by his/her gaining the acceptance of the other members.

The relationships developed between the members are also important, for the socially constructed softer aspects of knowledge are to be found there. This raises the question of what impact this will have on the knowledge loss problems posed by outsourcing and downsizing. CoPs cope with the loss of members by a process of regeneration—as old-timers leave they are replaced by newcomers moving to full participation. However, if the whole community were to be lost in a program of staff reduction, knowledge would be lost which could not be captured and stored for the use of the organisation.

Managing CoPs

The extension of the concept (as in Figure 1) has also led to some consultancies trying to formalise CoPs to the extent where consultants are told to go and create CoPs. CoPs per se cannot be created, or "managed" in the traditional sense of closely controlled, because they depend to such a great extent on the social issues. Perhaps a better approach is one of facilitation, enabling, and coaching. For example, it is not possible to say to a group of people that it is now a CoP and that it should go away and work as one. A group can be supported with facilities that facilitate interaction, and it may be that as relationships in the group develop, the group develops into a CoP. By the same token, a CoP per se cannot be managed—however, there will be the case where a formally managed group develops into a CoP. In that case, it is the formal group that is being managed; it just happens that that group is a CoP. The management was put in place for the original group, whether it was a CoP or not. Wheatley (1992), in trying to move away from rigid structures, emphasises the importance of relationships, interactions, and of linking people but acknowledges that precise outcomes cannot be pre-determined.

O'Dell and Jackson Grayson (1998b) see this from an ICT point of view and report on how Chevron has tried to nurture and support CoPs when they have appeared. Support and facilitation does not, however, have to be technological. Stewart (1996) recognised the value of IT but also of simpler support. He explained that managers can help CoPs in a number of ways. In the first place, simply recognising CoPs and their importance is a step forward. Once CoPs have been found, they can be supported initially by letting them build an intranet, or use a conference room, or by funding a get-together.

How do the Ideas of CoPs Inform Soft Knowledge?

LPP allows the development of both hard and soft knowledge. Hard knowledge can be articulated and may be exemplified by tasks that can be demonstrated. Soft knowledge is that which the newcomer cannot learn simply by demonstration or instruction; for example, in a craft situation the newcomer will learn to develop a "feel" for the material to be able to tell when the artefact being worked on "feels" right. (S)he can learn this through LPP, through working on tasks with the old-timers. Soft knowledge also goes further than this—it includes learning the language and unspoken conventions of the community. These softer aspects are learned through being socialised into the community through interaction with the existing members. The stories provide a good example, as well as demonstrating legitimation in a CoP. The utterance of the story itself is an articulation of the teller's hard knowledge. The quality of the story will depend on the soft knowledge that has been developed through interaction with the other members of the community. The members' soft knowledge is also necessary for understanding the story. An outsider or newcomer who has not yet developed the soft knowledge would not gain the same understanding as an old-timer who may also be in a position to make new inferences and interpretations from the story.

Part of the soft knowledge developed in a CoP, however, is still knowledge from a representationist/cognitivist view. This goes against the simple distinction expressed earlier that hard knowledge equates with the cognitivist/representationist and soft knowledge equates with the constructionist viewpoint. What this shows is that it is problematic to seek soft knowledge on its own. Knowledge is not made up of opposites. Regarding knowledge in these terms appears to be a false dichotomy. On the contrary, rather than regarding knowledge as opposites, it should perhaps be regarded as consisting of complementary facets. It needs to be regarded as a duality. Therefore, in order to move forward we will not regard knowledge as opposites but as a duality with hard and soft aspects; that is, all knowledge is to some degree hard and soft, simply the balance varies (Figure 2). The softer aspects of knowledge are those that cannot be captured, the harder aspects are those that can be articulated, captured, and stored.

Figure 2: Soft-Hard Knowledge Duality

This has a number of benefits; for example, it uses the "knowing" idea of Cook and Seely Brown (1999) as a different view of knowledge that cannot be captured, but it shows this in a simpler form. "Knowing" is one of the softer aspects of knowledge. The duality is not exclusive; it caters to those types of knowledge described earlier—those "types" of knowledge that cannot be so easily captured are some of the softer aspects. It can apply to those types of knowledge that are difficult to express or capture even if they are being viewed from a representational perspective, but it can also cover knowledge viewed from the constructionist perspective as softer aspects. Most importantly, it shows us that both perspectives are needed and must be taken into account for managing knowledge.

Viewing knowledge as a duality could explain some of the failures of KM initiatives. When the harder aspects are abstracted in isolation, the representation is incomplete—this is analogous to copying one side of a banknote and regarding it as a complete banknote, however, the softer aspects of knowledge must be taken into account as well. Hargadon (1998) gives the example of a server holding past projects, but developers do not look there for solutions. As the developers put it, "It's all in people's heads," that is, such solutions only represent the harder aspects of the knowledge. For a complete picture, the softer aspects are also necessary. Even the knowledge "in people's heads" is not sufficient; there is also the interactive aspect of Cook and Seely Brown's (1999) "knowing" that must be taken into account. This is one of the key softer aspects. As they explain it:

We act within a social and physical world, and since knowing is an aspect of action, it is about interaction with that world. When we act we either give shape to the physical world or both. Thus "knowing" does not focus on what we possess in our heads; it focuses on our interactions with the things of the social and physical world (p. 388).

Including the management of these softer aspects of knowledge in KM poses different challenges. Certainly there should be a shift from capturing and leveraging knowledge to supporting learning and the sharing of softer knowledge. The harder aspects of knowledge are managed by capturing them, storing them, and then disseminating them, for example in training sessions, books, or reports. The management of the softer aspects, however poses different problems—how can they be managed if they cannot be captured and there is no possibility for face-to-face interaction; that is, if individuals want to share such knowledge across time and space? Technology does have a role to play, but the emphasis needs to shift from trying to package knowledge as an object to using technology in a different context to try to share experience. This is a view supported by Davenport and Prusak (1998), who emphasise the communications aspect of ICT, with their potential to link people: "the more rich and tacit knowledge is, the more technology should be used to enable people to share that knowledge directly. It's not a good idea to try and contain or represent the knowledge itself using technology" (emphasis added) (p. 96).

Recognising this, Junnarkar and Brown (1997) point to the potential of technology such as tele- and video-conferencing. The technology currently receiving the most attention for supporting the sharing of softer knowledge is the intranet^[5], but at the moment even intranets seem to mainly be used for the sharing of the abstracted hard aspects of knowledge in the form of reports and documents.

Clearly IT itself is not enough and some emphasise the need for physical contact (Leonard & Sensiper, 1998). However, even with physical contact, there are still difficulties involved in the sharing of softer knowledge, for example, the importance of speaking the same language, not only the same national language but the same professional background language as well. Drucker (1992) gives the example of the American civil servant who would be at ease discussing bureaucratic issues with a Chinese counterpart, but who would be lost if he had to sit in a marketing meeting of a major retail organisation.

Developing the notion of the soft/hard duality indicates an area needing further exploration: the development and management of the soft aspects of knowledge needs joint activities and physical proximity for interaction; however, the increase in globalisation is resulting in an increased need to share knowledge in a physically distributed environment. This therefore raises the question as to how the softer aspects of knowledge can be managed in the physically distributed global environment.

To some the answer to this problem would be simple—the softer aspects of knowledge could be made hard, making its management a trivial matter. For example, are the war stories of the photocopier repairers (Orr, 1990) not soft knowledge made hard? This view falls into the same trap as traditional KM and attempts to view the softer aspects of knowledge from a representationist perspective. It also fails to recognise the essential feature of the duality, namely, that all knowledge is soft and hard. It is only the balance that differs and that only some aspects of knowledge can be hard. The war stories are only one aspect of the knowledge—without the softer aspects, they are of limited value. If they were taken outside the community, that is, removed from their context, people hearing them would obtain only very limited benefit. The softer aspects of knowledge are required to maximise the value of the stories; for example, the community members have worked closely with each other, have learned the language of the community and have interacted together. Interacting with each other and telling the stories enables different interpretations to be made and new knowledge to be created. Viewing knowledge as a duality helps to explain some other discrepancies in KM: the expert systems described in Chapter II failed because they concentrated solely on the harder aspects of knowledge. Ignoring the softer aspects meant the representation was incomplete, and the system could not be transplanted to a different environment. Another inconsistency was indicated in Chapter II: Seely Brown and Duguid (1998) differentiated between know-how and know-what. It was pointed out, however, that know-how can have an explicit component in the form of procedures that help people perform a task. Viewing this through the soft/hard lens explains the apparent contradiction. A procedure is a codified form of some of the harder aspects. Someone who had been doing a procedure for some time, that is, doing the activity, working the practice, would reach the point where (s)he would not need to rely on the procedure, could circumvent it if a work-around was needed, and would possibly be able to improve the procedure.

Some successes were also described in Chapter II, namely Dow Chemical's screening of the patents. However, using the soft/hard lens shows that it is not simply the fact that Dow has the patents (that is, the harder aspects of knowledge) but what it did with them that brought the success.

Viewing knowledge as a soft-hard duality poses the question of how this could be managed. We have noted that CoPs are groups where the softer aspects of knowledge are created, nurtured, and sustained. Recent work done in the area of CoPs provides an interesting avenue for exploration.

Recent Work in Communities of Practice

In 1998, Wenger brought the theory of CoPs up-to-date. He revisited the earlier (Lave & Wenger, 1991) work in the light of moving CoPs into a business environment. He based his study in a commercial organisational environment (in a claims processing department). He divided the book into two parts. In the first part, he concentrated on providing a series of characterisations of the concept of Communities of Practice. The essential part of this is practice. In the second part, he examined the concept of identity. This is the development of the members of the community as they join the community and participate in its practice.

Practice: Wenger considers "practice" to be much more than the everyday practice of a community. He prefers to explore practice as meaning; in particular, " practice is about meaning as an experience of everyday life" (p. 52). He states that it is this meaning as an experience that interests him and that it is located in a process that he calls the "negotiation of meaning." This negotiation of meaning involves the interaction of two processes that form a duality: participation and reification. This duality is central to bringing the concept of CoPs up-to date. Participation is one of the elements of LPP (Lave & Wenger, 1991), but while Wenger does not ignore legitimacy and peripherality, it is participation that he extracts as being key, showing it to be one of the constituent processes of his negotiation of meaning.

For Wenger (1998), participation is more than engaging in some activities with certain other people. Learning is social participation; that is, it is a fuller process whereby people are active participants in the practice of a community through which they can develop identities in relation to that community. Wenger's description of participation is an excellent example of how CoP members could develop softer aspects of their knowledge. For him, participation describes:

the social experience of living in the world in terms of membership in social communities and active involvement in social enterprises…Participation…is not tantamount to collaboration. It can involve all kinds of relations, conflictual as well as harmonious, intimate as well as political, competitive as well as cooperative (pp. 55-56).

A key aspect of participation for Wenger (1998) is mutuality—there must be reciprocal recognition. This does not necessarily imply respect and equality. He gives the examples of parents and children, and workers and direct supervisors. The relations between these people are mutual because the participants "shape each other's experiences of meaning" (p. 56). It is through the mutual interaction that CoP members gradually learn more and more about being a member of the community and thus develop the softer aspects of knowledge that cannot be directly articulated.

As knowledge has both softer and harder aspects, so participation and reification are also inextricably linked. Having concentrated on the participation aspect of LPP, Wenger (1998) emphasised that it remains undefined without the other constituent process that makes up the negotiation of meaning: reification. This is taken to mean giving concrete form to something that is abstract. He uses the concept of reification to refer to the process where experience is given a "form," somehow made concrete. He uses the term to cover a whole range of processes that include "making, designing, representing, naming, encoding and describing, as well as perceiving, interpreting, using, reusing, decoding and recasting" (p. 59).

He explains that any CoP will produce artefacts, tools, stories, procedures, and terms that reify something of its practice. Like the shared artefacts in DC, the artefacts reified here have knowledge embedded in them; however, the duality shows that it is the harder aspects that have been reified.

Wenger (1998) emphasises that these two constituent processes are separable analytically but are a duality. One cannot replace the other. Participation is undefined without reification and vice versa. They are not a dichotomy; they are a complementary pair in that they differ in how they affect the negotiation of meaning. In participation, mutuality is essential as members of a community recognise themselves in each other. In reification, however, our meanings are projected on to the world and attain an independent existence.

Figure 3: The Duality of Participation and Reification (Wenger, 1998, p. 63)

Another key point in the Participation-Reification duality is the proportion of each of the constituent processes. They need to be in a correct proportion so that one will balance the shortcomings of the other. The correct balance is necessary to achieve coordinated practice through Common Ground. Wenger (1998) points out:

If participation prevails—if most of what matters is left unreified—then there may not be enough material to anchor the specificities of coordination and to uncover diverging assumptions. This is why lawyers want everything in writing. If reification prevails—if everything is reified, but with little opportunity for shared experience and interactive negotiation—then there may not be enough overlap in participation to recover a coordinated, relevant, or generative meaning. This helps explain why putting everything in writing does not seem to solve all our problems (p. 65).

The participation/reification duality maps very closely to the hard knowledge/soft knowledge duality proposed earlier, particularly with regard to the proportions. Knowledge was said to be a soft/hard duality in that all knowledge is hard and soft—it is simply the proportions that may differ. This is also true of participation and reification. If knowledge is predominantly soft, then the participation proportion of the duality will be higher. Conversely, the harder the knowledge the greater the reification proportion. This provides an indication as to why the capture/codify/store approach fails to provide a whole KM solution. The capture/codify/store is reifying the harder aspects of knowledge and not accounting for the softer aspects, that is, the participation part of the duality.

Wenger (1998) also takes the reified artefact beyond the boundary of the CoP, because CoPs do not exist in isolation, and in many cases a lot of the work has to be done in coordination with other communities and groups. As the artefact bridges the boundary between communities, it functions as a "boundary object" (Star, 1989; Star & Griesemer, 1989). Boundary objects can be artefacts (for example, a document) around which CoPs can organise their interconnections.

The role of the artefact as connecting different communities provides interesting possibilities. Wenger (1998) refers to this role of the artefact as a "reificative connection" and explains that, as such, artefacts can help us overcome the time and space limits of participation. As examples, he observes that we cannot be everywhere, but we can read newspapers. Similarly he points out that we cannot live in the past, but we have monuments left by people from long ago.

However, he does also point out the limitations of the artefact alone. There are the possibilities of different interpretations, as reification must accommodate different viewpoints. Therefore, it is preferable to have people and artefacts travel together to take advantage of the complementarity of reification and participation—it has already been pointed out that an artefact by itself has a meaning projected on to it. As it functions as a boundary object, the receiving community will make its own interpretations. Having people and artefacts travel together redresses the proportions between reification and participation and enables a more fruitful negotiation of meaning.

In addition to practice as meaning supported by reification and participation, Wenger (1998) also explores practice as the property of a community, pointing out that it has three dimensions: mutual engagement, joint enterprise, and shared repertoire.

Figure 4: Dimensions of Practice as the Property of a Community (Wenger, 1998, p. 73)

Mutual Engagement: Mutuality is a key aspect of engagement. Wenger (1998) explains that practice exists because people engage in actions, and they have to negotiate the meanings of these actions with each other. Therefore, he states that practice is not to be found in books or tools (although it may involve many kinds of artefacts). Practice is to be found in a community "and the relations of mutual engagement by which they can do whatever they do" (p. 73).

Joint Enterprise: Joint enterprise is key to keeping a community together. Wenger (1998) explains that it is not simply a stated goal; rather, relations of mutual accountability are created between and among the members of the community and this becomes part of the practice. The joint enterprise is also defined by the members of the community as they are in the process of pursuing it.

Shared Repertoire: Over time a CoP will create resources for negotiating meaning as the participants pursue the joint enterprise. These resources can include procedures, routines, tools, stories, concepts, and artefacts that the community has produced and that have now become a part of the practice.

These areas provide further support to the soft/hard duality. The harder aspects are clearly visible in the shared repertoire, and the softer aspects of knowledge are necessary for the success of the joint enterprise and mutual engagement.

Identity

Participation and reification also play a part in the second part of Wenger's (1998) book where he examines the concept of identity. This is the development of the members of the community as they join the community and participate in its practice. Identity and practice are closely linked. Wenger observes that to develop a practice needs the members of the community to engage with each other and acknowledge each other as members of the community; that is, it involves the members each negotiating ways of being a person in the context of the CoP.

Identity is partly defined in terms of the relations that develop in a community. Engaging in the practice of a community develops relations that constitute a form of legitimation—it defines a member's place in the community. It defines, among other things, who has specific expertise, who is central, who is peripheral. For Wenger (1998), identity is defined socially in practice because it is a result of participation in communities. However, participation and reification are a duality and a balance of both is needed: "An identity, then, is a layering of events of participation and reification by which our experience and its social interpretation inform each other" (p. 151).

People learn how to work, they learn how to engage with each other, and as they do so they can play a part in the relations that define the community. They also learn how to interpret and use the community's repertoire of practice. This includes a community's artefacts, actions, and language and makes use of the community's history in the form of references and memories. Identity then, like legitimation, is a form of competence that is constantly being renegotiated over time.

Having brought CoPs up-to-date Wenger (1998) provides a list of possible characteristics that he feels define a CoP. Many of them overlap the characteristics identified in Figure 1^[6] but some points add to the concept, for example:

• When members offer a description of who "belongs," there is substantial overlap in the descriptions.

• Introductory preambles are non-existent. It is as if conversations and interactions are simply continuing an ongoing process.

• The members know what each other knows, what they are able to do, and how they can make a contribution.

Wenger (1998) is of the same view that it is not possible to create a single, tight definition, but rather it is preferable to use the term CoP as an umbrella term. Making a definition too broad would devalue the concept, yet having too narrow a definition would reduce its usefulness.

It is not necessary, for instance, to develop a simple metric that would yield a clear-cut answer for each of the social configurations…by specifying exact ranges of size, duration, proximity, amount of interaction or types of activities (p. 122).

There is however a limitation in the descriptions of CoPs, even when brought up-to-date by Wenger (1998)—that Communities of Practice are essentially a co-located phenomenon. It is only recently that much work has been done in the development of virtual CoPs. We have seen in Chapter II that the internationalisation of business is impacting an ever-increasing number of organisations and causing more and more work to be undertaken in a distributed, often international, environment. Lave and Wenger (1991)

pointed out in passing that CoPs do not necessarily have to be co-located; that is, the term community does not automatically imply co-presence. However, all of the examples were co-located. Wenger (1998) goes slightly further. He states that geographical proximity alone is not sufficient to develop a practice, but that it can be advantageous to have physical proximity. Wenger tackles the issue of locality of practice and within this context he introduces the notion of a constellation of communities of practice able to move out of physical colocation. Initially, he gives the examples of large communities such as speakers of a language, a city, or a social movement, and some smaller ones such as a factory, an office, or a school. He acknowledges the movement to a global environment and states that these configurations are better viewed as constellations of interconnected CoPs.

Work that has been done in the field of virtual CoPs has concentrated on on-line communities that regard themselves as CoPs. In this book, we are more concerned with organisational CoPs that have to function in a distributed international environment where knowledge (both soft and hard) needs to be shared across physical and temporal distance. It is not clear what effect such an environment will have on Wenger's reification/participation duality. This is an important question and we will explore it further in Chapter IV.

^[5]An intranet is an internal network which is contained within an organisation. It can be used for collaboration, to share files and use websites as it typically uses Internet protocols and looks like a mini, private version of the Internet.

^[6]The full list of CoP characteristics offered by Wenger (1998, pp. 125-126) is:
There are sustained mutual relationships. They do not have to be harmonious—they may be conflictual
There are shared ways of doing things together
There is a rapid flow of information and also a rapid propagation of innovation
Introductory preambles are non-existent. It is as if conversations and interactions are simply continuing an on-going process
Problems to be discussed are set up very quickly
When members offer a description of who ‘belongs’ there is a lot of overlap in the different descriptions
The members know what each other knows, what they are able to do and how they can make a contribution ‘mutually defining identities’
The appropriateness of actions and products is easily assessed
There are specific tools, representations and other artefacts
There is a shared background—stories, insider jokes
The members have their own language in the form of jargon
There are certain styles which indicate membership
A certain view of the world which comes out in ‘a shared discourse’