Communities of Practice

A definitional review. Citations follow IEEE style; see References.

Abstract

A community of practice is a group whose members learn a shared practice by taking part in it. The concept was coined in the study of apprenticeship, where it named the social arrangement — rather than the master–novice pair — through which craft knowledge is actually transmitted. Within a decade it had been reappropriated as an instrument of corporate knowledge management, and the term now carries two substantially different meanings. This article sets out the definition, the lineage, the discriminating features, that shift from a descriptive to a prescriptive concept, and what the literature does and does not establish.

I. Definition

The most widely circulated definition is the authors’ own summary formulation: communities of practice are “groups of people who share a concern or a passion for something they do and learn how to do it better as they interact regularly” [3].

The same source is explicit that the definition allows for, but does not assume, intentionality: learning may be the reason the group convenes, or it may be an incidental outcome of members’ interaction [3]. Three characteristics are then treated as jointly necessary [3]:

• the domain — a shared area of interest that implies commitment and a competence distinguishing members from non-members. It need not be an expertise recognised outside the group;
• the community — members engage in joint activity, help each other and build relationships, which is why a shared job title, a shared employer or a website is not by itself a community of practice;
• the practice — members are practitioners, developing a shared repertoire of experiences, stories, tools and ways of handling recurring problems, which takes sustained interaction.

It is the combination of the three, developed in parallel, that constitutes the community [3].

II. Origins

The concept originates in learning theory rather than in management. The anthropologist Jean Lave and Etienne Wenger coined the term while studying apprenticeship as a learning model, on the finding that apprenticeship is not principally a relationship between a novice and a master but a wider set of social relations in which learning happens largely with journeymen and more advanced apprentices. The term was coined to name the community that acts as a living curriculum for the apprentice [3].

The founding monograph, Situated Learning: Legitimate Peripheral Participation [1], advanced a theory of newcomer learning as continuous, active, engaged, situated and identity-forming, set against the then-dominant cognitive account of learning as transmission [4]. Its central proposition is that learning is more than acquiring knowledge — it is a change of identity — and that peripheral participation, active involvement in the practice at its margins, is the mechanism [4].

The comparative literature stresses that the 1991 concept is subtler than its common citation. A community of practice is never precisely defined in that book; it is explicitly not a “primordial culture sharing entity”, and the word community is not taken to “imply necessarily co-presence, a well-defined, identifiable group or socially visible boundaries” — passages quoted from pp. 42 and 98 in [4], which was read in full; the book itself was not consulted for this article. The reading Cox endorses is that naming a community of practice is not a way to postulate a new informal grouping inside an organisation but a way to emphasise that every practice depends on the social processes that sustain it [11], as quoted in [4].

Two further works are routinely conflated with the founding one. Brown and Duguid’s 1991 article carried the concept into organisational settings, stressing improvisation where canonical accounts of work prove inadequate to get the job done [5], as characterised in [4]. Wenger’s 1998 book [2] supplied the first clear definition — a group cohering through mutual engagement on an appropriated (or “indigenous”) joint enterprise, developing a shared repertoire — and turned the analytic focus onto identity, trajectories of participation and the strain of belonging to several communities at once [4]. Cox’s central observation is that although the four seminal works share a view of meaning as locally and socially constructed, they diverge markedly on community, learning, power, change, formality and diversity, so that a concept central to one is often absent from another [4].

III. What distinguishes a community of practice

The discriminating criterion is joint practice, not shared interest, shared employment or shared occupation.

• Against a community of interest: members must be practitioners developing a repertoire, not people who merely like the same thing [3].
• Against a team or a network: membership is voluntary and the boundary is set by the enterprise rather than by the organisation chart; communities of practice “don’t substitute teams or networks or other joint enterprises” [3].
• Against an occupational or professional community: this is the distinction most often lost. As originally conceived the term meant people working on a common enterprise; increasingly it is used for people working on a merely similar one, which is better described as an occupational community [4]. The confusion is longstanding — the ethnography that supplied Brown and Duguid’s evidence [12] used “occupational community”, not “community of practice”, and did not cite Lave and Wenger at all [4].

IV. The shift from analytic to prescriptive

This is the single most consequential fact about the concept, and it is well documented.

A. What changed

Cultivating Communities of Practice [6] is a practitioner handbook on forming and managing informal cross-functional groups to improve organisational performance. It redefines a community of practice as groups who share a concern or passion about a topic “and who deepen their knowledge and expertise in this area by interacting on an ongoing basis” — quoted from p. 4 of [6] in [4].

Cox’s assessment is that this is “genuinely a different concept”, not a change of tone: the new definition describes a group interested in the same thing rather than one tied together in accomplishing a common enterprise, and its purpose is specifically to learn and share knowledge rather than to get the job done [4]. The prescription changes correspondingly — no longer making space for workers to appropriate a joint enterprise, but creating groupings of people working on parallel rather than joint practices [4].

B. Why it happened

The concept was taken up fastest in business, on the recognition that knowledge is a strategic asset and that early knowledge-management efforts focused on information systems had produced “disappointing results” [3]. Communities of practice offered a people-centred alternative, and Wenger’s own account describes the trajectory plainly: from “humble beginnings in apprenticeship studies, the concept was grabbed by businesses interested in knowledge management and has progressively found its way into other sectors” [3]. The same evolution is the subject of a dedicated critique in the implementation-science literature [9], of which only the abstract could be obtained for this article.

C. The ethical shift

Cox characterises the move as a shift from revealing the value of what people know — especially in seemingly routine or mechanical jobs — to designing a tool for managing knowledge workers and experts in blue-chip companies [4], and notes that Wenger’s writing had already been observed to be more “performative” than “analytic” [7], as cited in [4]. The critical literature had earlier proposed that there is an affinity “between the dilution and selective adoption of Lave and Wenger’s thinking and its ideological compatibility with dominant managerial values” — quoted from p. 284 of [7] in [4].

The practical consequence for a reader is procedural: because the versions differ this much, any use of the term has to state which version it means [4].

V. Scope of adoption

Applications are claimed across business, organisational design, government, education, professional associations, international development and civic life, and the authors state that there is “hardly any organization of a reasonable size” without some form of communities-of-practice initiative [3]. In education the first applications were in teacher training and in connecting isolated administrators [3]. Adoption is also reported under other names — learning networks, thematic groups, tech clubs — and at very different scales, formal and informal, co-located and distributed [3].

A systematic review of the business and health-care literature published between 1991 and 2005 identified 31 primary studies, 18 in business and 13 in health care [8]. It found that although most health-care studies cited Wenger’s framework, the actual structure and function of the groups “varied greatly”, spanning clinical apprenticeships, journal clubs and virtual networks, and that four characteristics recurred to varying degrees: social interaction, knowledge sharing, knowledge creation and identity building [8].

VI. Limitations

Power and conflict are under-theorised. The founding account offers one hook for conflict — the generational tension over legitimating participation — but this is judged inadequate to explain the power forces inside a community, let alone those structuring it from outside, and it does not address conflict among old-timers or among newcomers [4]. Brown and Duguid’s internally egalitarian community loses sight of conflict altogether [4]; in Wenger’s 1998 book, power “is not a central concern” and conflict is treated mainly as internal to identity [4]. A dedicated critique of the approach’s limits for knowledge management organises itself around exactly these variables [10], though only its abstract could be obtained here.

“Community” imports a harmony the theory does not claim. Wenger’s own first indicator of a community of practice is “sustained mutual relationships — harmonious or conflictual” (p. 125, quoted in [4]), and the authors’ later primer answers the assertion that communities of practice are harmonious places with “Maybe”, warning that a totally conflict-free community should raise concern about groupthink or silenced voices [3]. Yet the word itself pulls readers toward “a rather large, helpful and friendly, bounded group” — a view Wenger warns against [4]. Cox’s sharpest warning is that the theory is “particularly dangerous when it presents a vision of the community of practice without conflict”, because a harmonious ideal can become a norm imposed on participants, and criticism or disagreement can then be labelled anti-social [4].

The theory says little about structure. Because the concept is “in a sense … a theory of pure agency”, it under-weights the degree to which a community’s life is shaped by the task and by formal controls [4]. Cox lists eight features of contemporary work that would inhibit the sustained mutual engagement the concept requires — frequent reorganisation, temporary staffing, tight management, individualised work, competitive environments, time pressure, spatial fragmentation and heavily mediated interaction — with the implication that genuine communities of practice may be rare [4]. Nor does the theory explain why appropriating a task produces a community that gets the job done rather than colleagues bending rules or organising around their own interests [13], as quoted in [4].

Effectiveness is not established. The systematic review found no health-sector study meeting its criteria for quantitative analysis, so the effectiveness of communities of practice in that sector “remained unclear”, and concluded that there is no dominant trend in how the concept is operationalised — which in turn makes such groups hard to identify and develop [8]. Its own stated limitations are that it covered publications only to 2005, excluded groups not explicitly labelled communities of practice, and appraised no qualitative study for quality [8].

Ambiguity is load-bearing. Cox’s overall verdict is that the ambiguity of both community and practice is the source of the concept’s reusability, allowing it to be reappropriated for different academic and practical purposes — which is simultaneously why it has lasted and why citing it without specifying a version conveys very little [4].

Readers should note that the transmission mechanism the concept describes — learning by taking part in a practice under the guidance of more experienced participants — is theorised from the instructional side by cognitive apprenticeship, whose framework names community of practice as one element of its social dimension. The two literatures are contemporaneous and frequently invoked together; they are not the same claim.

References

[1] J. Lave and E. Wenger, Situated Learning: Legitimate Peripheral Participation. Cambridge, U.K.: Cambridge University Press, 1991. Not obtained; cited for attribution only.

[2] E. Wenger, Communities of Practice: Learning, Meaning, and Identity. Cambridge, U.K.: Cambridge University Press, 1998. Not obtained; cited for attribution only.

[3] E. Wenger-Trayner and B. Wenger-Trayner, “Communities of Practice: A Brief Introduction,” Jun. 2015. [Online]. Available: https://www.wenger-trayner.com/wp-content/uploads/2022/06/15-06-Brief-introduction-to-communities-of-practice.pdf

[4] A. M. Cox, “What are communities of practice? A comparative review of four seminal works,” Journal of Information Science, vol. 31, no. 6, pp. 527–540, 2005, doi: 10.1177/0165551505057016. (Read in full via the author’s accepted manuscript, White Rose Research Online.)

[5] J. S. Brown and P. Duguid, “Organizational learning and communities of practice: toward a unified view of working, learning and innovation,” Organization Science, vol. 2, no. 1, pp. 40–57, 1991. Not obtained; characterised here from [4].

[6] E. Wenger, R. McDermott, and W. M. Snyder, Cultivating Communities of Practice. Boston, MA, USA: Harvard Business School Press, 2002. Not obtained; quoted here from [4].

[7] A. Contu and H. Willmott, “Re-embedding situatedness: the importance of power relations in learning theory,” Organization Science, vol. 14, no. 3, pp. 283–296, 2003. Not obtained; quoted here from [4].

[8] L. C. Li, J. M. Grimshaw, C. Nielsen, M. Judd, P. C. Coyte, and I. D. Graham, “Use of communities of practice in business and health care sectors: a systematic review,” Implementation Science, vol. 4, art. 27, 2009, doi: 10.1186/1748-5908-4-27.

[9] L. C. Li, J. M. Grimshaw, C. Nielsen, M. Judd, P. C. Coyte, and I. D. Graham, “Evolution of Wenger’s concept of community of practice,” Implementation Science, vol. 4, art. 11, 2009, doi: 10.1186/1748-5908-4-11. Abstract obtained only; not read in full.

[10] J. Roberts, “Limits to communities of practice,” Journal of Management Studies, vol. 43, no. 3, pp. 623–639, 2006, doi: 10.1111/j.1467-6486.2006.00618.x. Abstract obtained only; not read in full.

[11] S. Gherardi, D. Nicolini, and F. Odella, “Toward a social understanding of how people learn in organizations,” Management Learning, vol. 29, no. 3, pp. 273–297, 1998. Not obtained; quoted here from [4].

[12] J. E. Orr, Talking about Machines: An Ethnography of a Modern Job. Ithaca, NY, USA: Cornell University Press, 1996. Not obtained; characterised here from [4].

[13] S. Gourlay, “Communities of practice: a new concept for the millennium, or the rediscovery of the wheel?,” 1999. Not obtained; quoted here from [4].