Community of Practice

Abstract

A community of practice, which MeSH classifies under social learning, is a group of people who share a concern for something they do and learn to do it better through regular interaction. Introduced by Jean Lave and Etienne Wenger, the concept reframed learning as participation in a social practice rather than the transfer of knowledge into an individual head: a newcomer learns by moving from the periphery of a community toward full participation, absorbing its shared repertoire along the way. This article traces the idea from its roots in situated cognition, through Wenger's three constitutive dimensions — domain, community, and practice — to the knowledge-management movement it inspired and the critiques that followed. Three interactive demonstrations let the reader trace a participation trajectory, classify what is and is not a community of practice, and watch knowledge spread through shared engagement.

Keywords: community of practice, situated learning, legitimate peripheral participation, tacit knowledge, knowledge management

What a Community of Practice Is

A community of practice is a group of people who share a concern or a passion for something they do, and who deepen their knowledge and expertise by interacting on an ongoing basis (#ref-wenger-mcdermott-snyder-2002). The phrase was coined by Jean Lave and Etienne Wenger to name a phenomenon they had observed in studies of apprenticeship: that becoming a tailor, a midwife, or a quartermaster is less a matter of acquiring information than of becoming a participating member of a community that practices the trade (#ref-lave-wenger-1991). The unit of analysis is neither the individual learner nor the body of knowledge, but the community in which the two meet.

Defining the concept this way carries three commitments that organize the rest of this article. First, it is a theory of situated learning: competence is inseparable from the setting and the social practice in which it is used, not a portable possession carried from context to context (#ref-brown-collins-duguid-1989). Second, membership is a trajectory — one joins at the periphery and moves inward — so a community of practice is as much about identity and belonging as about skill (#ref-wenger-1998). Third, because the concept travelled quickly from the ethnography of apprenticeship into management, education, and health care, it has been stretched to cover groups that differ in important ways, and much of the later literature is an effort to say what it does and does not mean (#ref-cox-2005).

The reframing is easiest to see as a change in the structure of learning, shown in Figure 1. A transmission model routes knowledge one way, from a single source to each receiver; a community of practice distributes it across the relationships among the members, where knowing is built and rebuilt in interaction.

Figure 1

Two Structures of Learning: One-to-Many Transmission versus a Community of Practice

Situated Learning and Legitimate Peripheral Participation

The concept's foundation is the claim that learning is situated. In the cognitive tradition that preceded it, knowledge was modelled as a structure in an individual mind, and learning as the acquisition and storage of that structure. Lave and Wenger's anthropology of apprenticeship pointed the other way: in the trades they studied, knowing how to do the work was bound up with the tools, the setting, the division of labour, and the relationships of the workplace, and could not be abstracted from them without loss (#ref-lave-wenger-1991). The parallel argument from cognitive science held that concepts taught divorced from authentic activity remain inert — usable on a test but not in the situations the test was meant to stand for (#ref-brown-collins-duguid-1989).

Their central mechanism is legitimate peripheral participation. A newcomer is given real work to do, but work that is peripheral — low-risk, consequential enough to matter, simple enough to be survivable — and is granted legitimacy as a genuine, if junior, member of the community. Over time the newcomer takes on more central and more demanding tasks, and in doing so absorbs the community's shared repertoire of stories, tools, routines, and standards. Learning, on this account, is not a precondition for participation but its product: one learns by participating, moving from the edge toward the centre. The first demonstration lets the reader trace that trajectory and see what a participant does at each stage.

The Three Dimensions: Domain, Community, Practice

In his 1998 monograph Wenger developed the idea into a full social theory of learning and set out the three dimensions that jointly constitute a community of practice (#ref-wenger-1998). The domain is the shared area of interest that gives the community its identity and distinguishes members from non-members — the thing the community is about. The community is the group of people who interact, build relationships, and help one another around that domain; a shared interest with no mutual engagement is not yet a community. The practice is the shared repertoire the members develop over time — the cases, tools, vocabulary, routines, and tacit ways of doing things that mark a competent practitioner. A community of practice exists only where all three are present at once.

The force of the definition is in what it excludes, and Table 1 makes the contrasts explicit. A group that shares a domain but has no sustained interaction is a field or a discipline, not a community. A group that interacts warmly but develops no shared practice is a network of friends or an interest club. A group that shares methods but no common domain is a professional toolkit, not a community. This is also the main source of confusion in the applied literature: because real groups vary in how strongly each dimension is present, the single term gets applied to teams, project groups, and whole professions that satisfy the criteria only loosely (#ref-cox-2005). The second demonstration lets the reader switch each dimension on and off and watch the label change.

From Apprenticeship to Knowledge Management

The concept's second life came when organizations recognized their own informal groups in it. Wenger, Richard McDermott, and William Snyder reframed communities of practice as an asset that a firm can deliberately cultivate to steward knowledge that does not live in any document or database — the practical know-how distributed across experienced staff (#ref-wenger-mcdermott-snyder-2002). On this reading a community of practice is the social structure best suited to developing and sharing tacit knowledge, because tacit knowledge is learned by participation and conversation rather than by being written down.

This move was enormously influential and also changed the concept. In the original ethnography, communities of practice were found, not made; they emerged wherever people did shared work. In the management literature they became something to be designed, sponsored, and measured, which sits uneasily with the claim that they are self-organizing and voluntary. A systematic review of how the concept was used in health services found it defined and operationalized in strikingly inconsistent ways, with Wenger's own formulation evolving across his writings (#ref-li-2009). The tension between the community as an emergent social form and the community as a managed instrument runs through every applied use of the idea.

Thinking Together: What Makes a Community Work

If a community of practice is not reducible to a roster of members or a shared folder of documents, what is the active ingredient? Igor Pyrko and colleagues argue that it is thinking together: the process by which practitioners mutually engage with real, indeterminate problems and, in doing so, make each other's tacit knowledge usable (#ref-pyrko-2017). On this account the community is enacted, not merely inhabited. Two people who are nominally in the same community but never think through a shared problem together are not, in the sense that matters, practising together at all; conversely, the moments when a community genuinely functions are the episodes of joint inquiry in which knowing is mobilized.

This refocuses the concept from structure to activity and answers a long-standing puzzle: why some officially chartered communities of practice never come alive while informal ones thrive. The difference is not in the domain statement or the membership list but in whether real thinking-together occurs. It also connects the idea back to its situated-learning roots — tacit knowledge becomes accessible to a newcomer precisely in the shared engagement with concrete problems, which is what legitimate peripheral participation provides. The third demonstration shows how knowledge becomes mobilizable as mutual engagement across a community increases.

Critiques and Limits

The concept's reach has drawn sustained critique, most of it aimed not at discrediting it but at marking its boundaries. Joanne Roberts argues that the original formulation underplays power: communities have insiders and gatekeepers, participation can be withheld, and the harmonious picture of newcomers welcomed inward ignores the politics of who is allowed to learn and whose practice counts (#ref-roberts-2006). Scale is a second limit — the intimate, face-to-face dynamics that define a small apprenticeship community do not straightforwardly extend to the large, dispersed, or virtual groups the label is now applied to.

Ash Amin and Roberts pressed this further, arguing that lumping craft work, professional expertise, high-technology innovation, and online communities under one term flattens real differences in how knowing actually happens in each (#ref-amin-roberts-2008). Their point is diagnostic: knowing in action takes different forms, and the single phrase 'community of practice' can obscure more than it reveals. The comparative review of the founding texts reaches a compatible conclusion, showing that even the four seminal works use the term in materially different senses (#ref-cox-2005). Taken together the critiques do not retire the concept; they convert it from a universal description into a model whose fit must be argued for in each case.

Worked Example

The claim that learning is distributed across a community, rather than transmitted from a single teacher, has a structural consequence that can be made quantitative. Model a community as a graph in which a learning relationship is an edge between two members who engage with each other's work. If learning happens through mutual engagement between pairs, a community of n fully engaged members supports C(n, 2) = n(n − 1) / 2 distinct dyadic relationships. For a community of n = 12, that is (12 × 11) / 2 = 66 potential learning relationships.

Contrast this with a transmission model, in which a single instructor pushes knowledge out to each of the other n − 1 learners and the learners do not engage one another. That structure — a star — supports only n − 1 = 11 channels. The ratio of the two is [n(n − 1) / 2] / (n − 1) = n / 2, which for n = 12 equals 6: a community that learns through mutual engagement makes roughly six times as many learning relationships available as a one-to-many transmission of the same size. Because the ratio is n / 2, the advantage grows linearly with the size of the community — doubling the membership doubles the relative gain of participation over transmission. The arithmetic is a crude idealization, but it captures the formal reason Lave and Wenger's reframing matters: moving the locus of learning from a teacher to a community does not add a channel, it changes the order of growth of the relationships through which knowing can spread.

Discussion

The ideas here form a single arc. Situated-learning theory supplies the premise that knowing is inseparable from the social practice in which it is used (#ref-lave-wenger-1991), (#ref-brown-collins-duguid-1989). Wenger's three dimensions turn that premise into a definition precise enough to exclude teams, networks, and disciplines that merely resemble communities of practice (#ref-wenger-1998). The knowledge-management literature shows the concept's practical payoff and, in the same motion, exposes the strain of treating an emergent social form as a managed instrument (#ref-wenger-mcdermott-snyder-2002), (#ref-li-2009). The thinking-together account locates the concept's active ingredient in joint inquiry rather than in membership (#ref-pyrko-2017). And the critiques mark where the model's fit must be argued rather than assumed (#ref-roberts-2006), (#ref-amin-roberts-2008), (#ref-cox-2005).

What the concept does not license is the assumption that assembling people with a shared interest produces learning on its own. A domain statement and a membership list are not a community of practice; the practice has to be enacted, the participation has to be legitimate and genuinely peripheral, and the mutual engagement has to actually occur. The concept's durability comes from naming a real and common social form; its frequent misuse comes from applying the name to groups that satisfy its criteria only in part. The mature position treats 'community of practice' as a model with identifiable constitutive conditions, to be tested against a group rather than asserted of it.

Current Directions

The most active front widens the frame from the single community to the landscape of practice a practitioner moves through. Pyrko and colleagues extend the thinking-together account to this larger scale, showing how a professional's knowledgeability is composed across the many communities they touch rather than residing in any one, and how brokering knowledge across community boundaries becomes a competence in its own right (#ref-pyrko-2019). This reframes identity as a trajectory across a landscape, not merely inward within one community. In parallel, a comprehensive stock-taking review has organized three decades of research, distinguishing the concept's incompatible uses, consolidating what is robustly known, and setting an agenda for where it should go next (#ref-nicolini-2022). A recurring theme in the recent work is the status of virtual and dispersed communities, whose mediated interaction strains the face-to-face assumptions of the original formulation, and the continuing effort — begun with the founding interview-based clarifications (#ref-farnsworth-2016) — to state precisely what the social theory of learning claims. The direction of travel is away from asking whether a group is a community of practice and toward specifying how knowing is actually accomplished within and across the communities a person inhabits.

Key Researchers

John Seely Brown

(former Chief Scientist, Xerox; former Director, Xerox PARC). With Allan Collins and Paul Duguid, author of the situated-cognition argument that knowing is inseparable from authentic activity and culture — the cognitive-science foundation the community-of-practice concept grows from. Wikipedia · Website

Jean Lave

(Professor Emerita, University of California, Berkeley). Social anthropologist who, with Etienne Wenger, introduced the community-of-practice concept and legitimate peripheral participation; her studies of apprenticeship reframed learning as participation in a community rather than the transmission of knowledge into a head. Wikipedia

Davide Nicolini

(Professor, Warwick Business School). A leading practice-theory scholar and lead author of the 2022 Academy of Management Annals review that takes stock of three decades of community-of-practice research and sets its forward agenda. Faculty

Igor Pyrko

(University of Bath, School of Management). Lead author of the thinking-together studies and co-author of the 2022 stock-taking review; recast what binds a community of practice as mutual engagement with real problems rather than shared membership alone. ORCID

Barbara Rogoff

(UCSC Foundation Distinguished Professor of Psychology, University of California, Santa Cruz). Developmental psychologist whose work on guided participation and apprenticeship in thinking gave the community-of-practice idea its account of how newcomers learn through graduated participation in culturally organized activity. Faculty

Etienne Wenger-Trayner

(Independent researcher and consultant). Co-originator of the concept and author of the 1998 monograph that built the full social theory of learning; later developed the landscapes-of-practice framework extending the idea beyond any single community. Google Scholar · Website

Glossary

Apprenticeship.

A traditional form of situated learning in which a newcomer acquires a trade by working alongside skilled practitioners; the ethnographic starting point from which the community-of-practice concept was drawn.

Boundary object.

An artefact — a document, tool, or shared representation — that lets different communities coordinate around it while each interprets it in its own terms.

Brokering.

The work of carrying practices, ideas, or knowledge across the boundary between one community and another; a competence that spans a landscape of practice.

Community of practice.

A group that shares a domain of interest, interacts as a community, and develops a shared practice; all three dimensions must be present at once.

Domain.

The shared area of interest or concern that gives a community of practice its identity and distinguishes members from non-members.

Knowledgeability.

A person's competence composed across the many communities of a landscape of practice, rather than mastery residing within a single community.

Landscape of practice.

The wider terrain of multiple, interconnected communities of practice through which a practitioner moves over a career.

Legitimate peripheral participation.

The process by which a newcomer learns by doing real but low-risk work at the edge of a community and moving toward full participation as competence grows.

Mutual engagement.

The sustained interaction around shared work through which members of a community build relationships and a common practice.

Practice.

The shared repertoire a community develops — the cases, tools, vocabulary, routines, and tacit ways of doing things that mark a competent member.

Shared repertoire.

The accumulated resources of a community — stories, routines, artefacts, and vocabulary — that constitute its practice and are available to members.

Situated learning.

The view that learning is inseparable from the activity, context, and culture in which it occurs, rather than an abstract structure acquired independently of use.

Tacit knowledge.

Know-how that is difficult to articulate or write down and is learned chiefly through participation and shared practice rather than instruction.

Thinking together.

The mutual engagement with real, indeterminate problems through which practitioners make each other's tacit knowledge usable; what makes a community of practice function.

Frequently Asked Questions

What is a community of practice?

A community of practice is a group of people who share a concern or passion for something they do and learn to do it better as they interact regularly. It is defined by three things together: a shared domain of interest, a community that engages around it, and a shared practice the members develop over time.

Who introduced the concept?

Jean Lave and Etienne Wenger introduced the term in 1991, drawing on anthropological studies of apprenticeship. Wenger developed it into a full social theory of learning in a 1998 monograph, and later work with Richard McDermott and William Snyder carried it into organizational knowledge management.

What are the three dimensions of a community of practice?

The domain is the shared area of interest that gives the community its identity. The community is the group of people who interact and build relationships around that domain. The practice is the shared repertoire of cases, tools, and routines the members develop. A community of practice requires all three at once.

What is legitimate peripheral participation?

It is the way newcomers learn within a community of practice. A newcomer is given real but low-risk work at the periphery and is accepted as a genuine, if junior, member. As competence grows they take on more central tasks, learning the practice by participating in it rather than by being taught it in the abstract.

How is a community of practice different from a team or a network?

A team is assembled to deliver a specific output and dissolves when the task is done; a network is a web of relationships that need share no common practice. A community of practice is organized around learning to do shared work better, is often informal and voluntary, and persists as long as the practice holds the members' interest.

Can organizations deliberately create communities of practice?

Organizations can cultivate the conditions for them, by sponsoring time, connecting people, and legitimizing the work, but they cannot simply mandate one into being. Because the social form is emergent and voluntary, a chartered community with a membership list will not function unless genuine mutual engagement and shared practice actually develop.

What makes a community of practice actually work?

Research points to thinking together: members mutually engaging with real, difficult problems and, in doing so, making each other's tacit knowledge usable. A community that never engages in joint problem-solving remains a community of practice in name only, which is why many formally chartered ones never come alive.

What are the main criticisms of the concept?

Critics argue that the original formulation underplays power and conflict, that its intimate dynamics do not scale cleanly to large, dispersed, or virtual groups, and that applying one label to craft, professional, technological, and online groups flattens real differences in how knowing happens. The critiques bound the concept rather than discard it.

References

Amin, A., & Roberts, J. (2008). Knowing in action: Beyond communities of practice. Research Policy, 37(2), 353–369. https://doi.org/10.1016/j.respol.2007.11.003

Brown, J. S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18(1), 32–42. https://doi.org/10.3102/0013189X018001032

Cox, A. (2005). What are communities of practice? A comparative review of four seminal works. Journal of Information Science, 31(6), 527–540. https://doi.org/10.1177/0165551505057016

Farnsworth, V., Kleanthous, I., & Wenger-Trayner, E. (2016). Communities of practice as a social theory of learning: A conversation with Etienne Wenger. British Journal of Educational Studies, 64(2), 139–160. https://doi.org/10.1080/00071005.2015.1133799

Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge University Press.

Li, L. C., Grimshaw, J. M., Nielsen, C., Judd, M., Coyte, P. C., & Graham, I. D. (2009). Evolution of Wenger's concept of community of practice. Implementation Science, 4(1), 11. https://doi.org/10.1186/1748-5908-4-11

Nicolini, D., Pyrko, I., Omidvar, O., & Spanellis, A. (2022). Understanding communities of practice: Taking stock and moving forward. Academy of Management Annals, 16(2), 680–718. https://doi.org/10.5465/annals.2020.0330

Pyrko, I., Dörfler, V., & Eden, C. (2017). Thinking together: What makes communities of practice work? Human Relations, 70(4), 389–409. https://doi.org/10.1177/0018726716661040

Pyrko, I., Dörfler, V., & Eden, C. (2019). Communities of practice in landscapes of practice. Management Learning, 50(4), 482–499. https://doi.org/10.1177/1350507619860854

Roberts, J. (2006). Limits to communities of practice. Journal of Management Studies, 43(3), 623–639. https://doi.org/10.1111/j.1467-6486.2006.00618.x

Wenger, E. (1998). Communities of practice: Learning, meaning, and identity. Cambridge University Press.

Wenger, E., McDermott, R., & Snyder, W. M. (2002). Cultivating communities of practice: A guide to managing knowledge. Harvard Business School Press.