Communities of Practice: Learning by Belonging, Not by Being Told

AA new tailoring apprentice doesn't start by cutting cloth. They start by sweeping the floor, fetching thread, and watching. Then they're given buttonholes — a real, useful, low-stakes piece of a real garment — long before anyone lets them near a sleeve. Jean Lave's fieldwork among tailors in Liberia, later generalised with Etienne Wenger into a general account of how expertise is actually acquired, treats this sequence as the rule rather than the exception: apprentices across wildly different trades — midwifery, butchery, naval quartermastering, recovery from alcoholism — learn by moving gradually from small, safe, peripheral tasks toward full participation in a community's real work, not by being handed a manual and tested on it¹. The name Lave and Wenger gave this movement, legitimate peripheral participation, is the organising idea behind most of what a workplace actually calls "training."

Legitimate Peripheral Participation FoundationalKnowledge that endures for decades — core principles

EachEach of the three words is doing work. Peripheral because the apprentice starts at the edge of the real task, not the centre — buttonholes, not sleeves. Participation because the edge task is still real: a genuine buttonhole on a genuine garment someone will actually wear, not a simulation built to be thrown away afterwards. Legitimate because the community itself recognises the newcomer's peripheral contribution as a real, valued part of the work, which is what gives the newcomer a path inward rather than leaving them permanently stuck at the edge. Strip any one of the three away and the mechanism stops working: a task that's peripheral but not legitimate is busywork; a task that's legitimate but not peripheral is thrown a novice before they're ready; participation without any path from periphery to centre is just cheap labour wearing an apprenticeship's name.

Identity Is Part of What's Learned FoundationalKnowledge that endures for decades — core principles

WhatWhat moves along this path isn't only skill. Lave and Wenger's claim is that becoming a tailor, a midwife, or a quartermaster is inseparable from coming to see yourself as one — competence and identity develop together, not in sequence. This site's own page on the social structure inside a rhizomatic learning network develops the mechanism by which a leaderless community converges on shared vocabulary at all: Etienne Wenger's later account of a community of practice as built from participation (the lived business of doing the work together) and reification (producing the objects, terms, and concepts that give shared experience a settled, nameable form)². That page's own worked example is a piece of jargon nobody formally coined; this page's worked example is a person — an apprentice becomes legible as "a tailor" through the same process, reified not into a term but into a recognised role inside the community that trained them.

Why Manuals Routinely Miss This FoundationalKnowledge that endures for decades — core principles

IfIf expertise travels this way, a procedure manual was never going to be the main channel it travels through, and the gap isn't a drafting failure so much as a structural one. Michael Polanyi's phrase for the territory a manual can't reach is tacit knowledge: "we know more than we can tell"³. This site's lecture on software engineering as practical anthropology develops the same point for requirements work specifically: a domain expert's procedure manual is a fossil of what they managed to say out loud on a particular day, not a complete record of what they actually do, and the gap between the two is reliably where the organisation's working knowledge of itself actually lives. A tailoring apprentice absorbs which fold of cloth signals a problem, and how much tension a seam should hold, the same way a claims handler absorbs which fields to double-check — by doing the work alongside someone who already can, not by reading a description of it. What happens to this kind of knowledge once the person carrying it leaves is this cluster's own next question.

Strongest Objection: Communities of Practice Can Gatekeep as Easily as They Teach FoundationalKnowledge that endures for decades — core principles

LegitimateLegitimate peripheral participation quietly assumes the community wants the newcomer to move inward. Nothing about the mechanism guarantees that. Alessia Contu and Hugh Willmott's critique of the original framework argues it under-attends to power: who gets access to legitimate peripheral tasks in the first place, whose contribution is actually recognised as legitimate rather than quietly discounted, and whose "periphery" is a genuine path toward the centre rather than a holding pattern the community has no intention of ever promoting someone out of⁴. A workplace that keeps certain employees permanently on peripheral tasks — not because the tasks are a deliberate first rung, but because the community never intends to let that person climb further — is running the same surface mechanism as a genuine apprenticeship while producing a very different outcome. The difference isn't visible from the task assignment alone; it's visible only in whether the path inward is actually open.

Provisional Conclusion FoundationalKnowledge that endures for decades — core principles

ExpertiseExpertise moving through legitimate peripheral participation, rather than through explicit instruction, is a well-evidenced account of how organisations actually transmit their working knowledge — and it explains, better than any claim about bad documentation does, why a new hire's first few weeks of shadowing someone experienced routinely teaches more than the induction manual they were handed on day one. Whether a given community's periphery is a genuine path inward or a permanent holding pen is a separate question the mechanism itself can't answer, and treating every instance of "learning by doing alongside an expert" as automatically benign skips past exactly the question Contu and Willmott raise.

Questions for Further Thought

• How would a newcomer tell, from the inside, whether their peripheral tasks are a deliberate first rung or a holding pattern the community has no intention of promoting them out of?
• If identity and competence develop together, what happens to someone who gains the competence but is never granted the identity — recognised as capable, but never as a full member?
• Does legitimate peripheral participation scale to a fully remote or asynchronous team, where the newcomer may rarely be in the same room as the work they're meant to be absorbing by watching?

Related Topics

• Inside the Rhizome: Four Social Groupings, and How Each Should Count — the participation/reification mechanism this page applies to a person's own developing identity rather than to a field's shared vocabulary.
• Software Engineering as Practical Anthropology — the tacit-knowledge argument this page's "why manuals miss this" section builds on directly.
• Can an Organisation Know Something No One in It Knows? — the organisational-knowledge claim this page explains the transmission mechanism for.
• The Half-Life of Organisational Knowledge — what happens to tacit knowledge absorbed this way when the person carrying it eventually leaves.

Further Reading

• Lave, J., & Wenger, E. (1991). Situated Learning: Legitimate Peripheral Participation. Cambridge University Press.
• Wenger, E. (1998). Communities of Practice: Learning, Meaning, and Identity. Cambridge University Press.
• Polanyi, M. (1966). The Tacit Dimension. University of Chicago Press.
• Contu, A., & Willmott, H. (2003). Re-embedding situatedness: The importance of power relations in learning theory. Organization Science, 14(3), 283–296.

References