Journal of Management Studies 43:3 May 2006

0022-2380

Limits to Communities of Practice

Joanne Roberts
University of Durham Business School

abstract The purpose of this paper is to critically explore the communities of practice
approach to managing knowledge and its use among management academics and
practitioners in recent years. In so doing, the aim is to identify the limits of the approach
in the field of knowledge management. The paper begins with a brief description of the
communities of practice approach. This is followed by a review of critiques of the
approach evident in the management literature. A number of further challenges are
then elaborated. The limits of communities of practice are subsequently discussed and
brief conclusions drawn.

INTRODUCTION
Since being identified as a mechanism through which knowledge is held, trans-
ferred and created (Brown and Duguid, 1991; Lave and Wenger, 1991; Wenger,
1998), the communities of practice approach has become increasingly influential
within management literature and practice.[1] Originally developed by Lave and
Wenger (1991) in a study of situated learning, the communities of practice
approach is currently being used to analyse and facilitate knowledge transfer in a
wide range of organizational environments. The purpose of this paper is to criti-
cally explore the communities of practice approach to managing knowledge and its
use among management academics and practitioners in recent years. In so doing,
the aim is to identify the limits of communities of practice in the field of knowledge
management.
An increasing number of studies in the management literature have provided
critiques of the communities of practice approach (e.g. Contu and Willmott, 2003;
Fox, 2000; Handley et al., 2006; Marshall and Rollinson, 2004; Mutch, 2003).
However, I will argue that this literature is in need of further development because
some challenges have not yet been fully explored. In this way, I hope to encourage
Address for reprints: Joanne Roberts, University of Durham Business School, Mill Hill Lane, Durham
City DH1 3LB, UK ( joanne.roberts@durham.ac.uk).

© Blackwell Publishing Ltd 2006. Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ,
UK and 350 Main Street, Malden, MA 02148, USA.
624 J. Roberts
critical reflection on the communities of practice literature, and, through this, the
development of an enhanced understanding of the creation and transfer of knowl-
edge through practice.
The paper begins with a brief description of the communities of practice
approach. This is followed by a review of critiques of the communities of practice
approach, beginning with those evident in the management literature before a
number of further challenges are elaborated. The limits of communities of practice
are subsequently discussed and brief conclusions drawn.

WHAT ARE COMMUNITIES OF PRACTICE?

The concept of communities of practice was originally developed by Lave and
Wenger (1991) in a study of situated learning in the context of five apprenticeships:
Yucatec midwives; Vai and Gola tailors; naval quartermasters; meat cutters; and
non-drinking alcoholics. Lave and Wenger (1991, p. 98) argue that a community of
practice, which they define as ‘a system of relationships between people, activities,
and the world; developing with time, and in relation to other tangential and
overlapping communities of practice’ is an intrinsic condition of the existence of
knowledge. Brown and Duguid (1991, 1998), drawing on the work of Orr (1996)
among others, have further developed the approach. Moreover, through a study of
an insurance claims processing office, Wenger (1998) developed a detailed under-
standing of the dynamic operation of communities of practice. The communities of
practice approach focuses on the social interactive dimensions of situated learning,
a subject that has received attention from a variety of other organizational
researchers (see, for example, Barley and Orr, 1997; Blackler, 1995; Boland and
Tenkasi, 1995; Carlile, 2002; Gherardi, et al., 1998).
According to Wenger (1998, p. 55), within communities of practice, meaning is
negotiated through a process of participation and reification. Wenger (1998, p. 58)
defines the concept of reification as the process of giving form to experience by
producing objects. ‘Any community of practice produces abstractions, tools,
symbols, stories, terms, and concepts that reify something of that practice in a
congealed form’ (Wenger, 1998, p. 59). Such forms take on a life of their own
outside their original context where their meaning can evolve or even disappear.
For Wenger (1998) communities of practice are important places of negotiation,
learning, meaning, and identity. Wenger (1998, pp. 72–84) identifies three dimen-
sions of the relation by which practice is the source of coherence of a community.
Firstly, members interact with one another, establishing norms and relationships
through mutual engagement. Secondly, members are bound together by an under-
standing of a sense of joint enterprise. Finally, members produce over time a shared
repertoire of communal resources, including, for example, language, routines, arti-
facts and stories. Furthermore, Wenger (2000, pp. 227–8) distinguishes between
three modes of belonging to social learning systems. Firstly, engagement is achieved
© Blackwell Publishing Ltd 2006
 Limits to Communities of Practice 625
Table I. The characteristics of communities of practice

Key characteristics of a community of practice

• Sustained mutual relationships – harmonious or conflictual

• Shared ways of engaging in doing things together
• The rapid flow of information and propagation of innovation
• Absence of introductory preambles, as if conversations and interactions were merely the
continuation of an ongoing process
• Very quick setup of a problem to be discussed
• Substantial overlap in participants’ descriptions of who belongs
• Knowing what others know, what they can do, and how they can contribute to an enterprise
• Mutually defining identities
• The ability to assess the appropriateness of actions and products
• Specific tools, representations, and other artifacts
• Local lore, shared stories, inside jokes, knowing laughter
• Jargon and shortcuts to communication as well as the ease of producing new ones
• Certain styles recognized as displaying membership
• A shared discourse reflecting a certain perspective on the world

Source: Compiled from Wenger (1998, pp. 125–6).

through doing things together, for example, talking and producing artifacts. Sec-
ondly, imagination involves constructing an image of ourselves, of our communities,
and of the world, in order to orient ourselves, to reflect on our situation, and to
explore possibilities. Finally, alignment involves making sure that our local activities
are sufficiently aligned with other processes so that they can be effective beyond our
own engagement.
The existence of a community of practice may not be evident to its members
because, as Wenger (1998, p. 125) notes, ‘a community of practice need not be
reified as such in the discourse of its participants’. Nevertheless, he argues that a
community of practice does display a number of characteristics including those
listed in Table I.
Communities of practice are not stable or static entities. They evolve over time
as new members join and others leave. Communities of practice as defined by Lave
and Wenger (1991) cannot be formed. For example, a business can establish a team
for a particular project, which may, in time, emerge as a community of practice.
But management cannot establish a community of practice. What it can do is
facilitate the spontaneous emergence of communities of practice and support those
communities of practice that do develop. As Brown and Duguid (2001a) suggest,
managers can seek to structure spontaneity; in particular, they have a role to play
structuring fragmented practice across their organization. On the one hand, man-
agers have a role supporting the development of communities of practice. On the
other, they can encourage alignments of changing practices between communities,
thereby assisting the transfer of knowledge across the organization (Brown and
© Blackwell Publishing Ltd 2006
626 J. Roberts
Duguid, 2001a). More recent contributions suggest that communities of practice
can be cultivated and leveraged for strategic advantage (Saint-Onge and Wallace,
2003; Wenger et al., 2002). In line with this view, an increasing number of
consultancy firms are offering to improve their clients’ abilities to manage knowl-
edge creation and dissemination by identifying or establishing communities of
practice.[2]
Managers are increasingly seeking to develop and support communities of
practice as part of their knowledge management strategies and, for some, commu-
nities of practice are viewed as a supplementary organizational form (Wenger and
Snyder, 2000; Wenger et al., 2002), which can create value and improve perfor-
mance (Lesser and Storck, 2001). Furthermore, Swan et al. (2002) suggest that the
notion of communities of practice can be used as a rhetorical tool to facilitate the
control of professional groups over which managers have little authority. Addition-
ally, much research concerning the transfer of knowledge and information in
virtual organizations has been influenced by the communities of practice literature
(see, for example, Johnson, 2001; Pan and Leidner, 2003; Smeds and Alvesalo,
2003).

COMMUNITIES OF PRACTICE: A CRITIQUE

The communities of practice approach is not without its weaknesses and limita-
tions. In their book, Cultivating Communities of Practice, Wenger et al. (2002, p. 141)
devote a chapter to what they refer to as the ‘downside’ of communities of practice
arguing that the ‘very qualities that make a community an ideal structure for
learning – a shared perspectives on a domain, trust, a communal identity, long-
standing relationships, an established practice – are the same qualities that can
hold it hostage to its history and its achievements’. Taking a broader perspective,
this section explores a number of unresolved issues and difficulties evident in the
communities of practice approach. In so doing, it provides the basis for the
discussion of the limits to the communities of practice approach to knowledge
management in a subsequent section. Issues concerning power, trust and predis-
positions, which have already been explored in the academic literature, will be
reviewed in the first part of this section. I will then consider a number of further
challenges to the communities of practice approach related to their size and spatial
reach, the nature of community and the accelerated business environment.

Extant Critiques
Power. An understanding of the power dynamics of communities of practice is
essential to the development of a full understanding of knowledge creation and
dissemination. Power is the ability or capacity to achieve something, whether by
influence, force, or control. While meaning may be negotiated within communities
© Blackwell Publishing Ltd 2006
 Limits to Communities of Practice 627
of practice, it is vital to recognize the role of power in this process. Communities of
practice will include members of varying standing in terms of experience, expertise,
age, personality, authority within the organization and so on. Indeed, power may
be evident in terms of the degree of participation. In the context of the appren-
ticeships explored in Lave and Wenger’s (1991) study of situated learning, new
community members move from the periphery to a position of full participation as
they develop their knowledge and learn from skilled practitioners. Those members
who have full participation will have a greater role and therefore are likely to wield
more power in the negotiation of meaning. However, in a broader organizational
context peripheral community members may not necessarily develop beyond a
position of peripheral participation. Meanings may continue to be merely a reflec-
tion of the dominant source of power. Although Lave and Wenger (1991) do note
the significance of power in shaping the legitimacy of peripherality and participa-
tion, they fail to explore the implications of the distribution of power when dis-
cussing their case studies of communities of practice, and considerations of power
are absent or relegated to footnotes in Wenger’s (1998, 2000) later work (Contu
and Willmott, 2000, 2003). As Marshall and Rollinson (2004, p. S74) suggest, Lave
and Wenger’s (1991) account of the negotiation of meaning can be misinterpreted
as ‘excessively quiescent and consensual’, while in reality such activities are plagued
by misunderstandings and disagreements. Seeking to prevent such misinterpreta-
tions Contu and Willmott (2003) develop Lave and Wenger’s (1991) analysis of
power through an examination of Orr’s (1996) ethnographic study of Xerox
photocopy repair technicians. They reinterpret the practices of these technicians in
terms of relations of power in a corporate hierarchy. Pressures from internal
sources such as directors and experts as well as from outside the organization can
inhibit the will and ability of workers to engage effectively in the negotiation of
meaning (Coopey and Burgoyne, 2000, p. 877).
Yanow (2004) explores issues concerning recognition and power in relation to
local knowledge versus expert knowledge. While workers may be full participants
in their own community of practice, their knowledge, despite its relevance for the
formulation of strategy, is not necessarily recognized within the formal organiza-
tional hierarchy where, so called, expert knowledge, acquired from external con-
sultants, is preferred. Similarly, for Blackler and McDonald (2000, p. 848), the
‘dynamics of power, mastery and collective learning are inseparable’. For, as
Foucault (1979, p. 27) notes, power and knowledge imply each other: ‘ . . . there is
no power relation without the correlative constitution of a field of knowledge, nor
any knowledge that does not presuppose at the same time power relations’.
Making use of the work of Foucault and Actor Network Theory (ANT), Fox
(2000) provides a useful critique of power relations within communities of practice.
According to Fox (2000, p. 860), from an ANT perspective, learning ‘is seen as an
outcome of a process of local struggle and that struggle is many-faceted involving
the self acting upon itself, as well as upon others and upon the material world’.
© Blackwell Publishing Ltd 2006
628 J. Roberts
ANT also provides greater insight into how members of a community of practice
work with and interact with other humans and non-humans, such as information
and communication technologies (ICTs), recognizing that both human and non-
human materials have resistive agency (Fox, p. 863).
An organization’s overall power structure may be reflected in the power rela-
tions within its communities of practice. For instance, in decentralized network
type organizations, where power is distributed, one might expect to find greater
diversity in the voices actively shaping and negotiating meaning, which is to say
that there will be a greater variety in the possible range of knowledge created and
shared. Whereas in hierarchical organizational structures where power is central-
ized, negotiation may be limited to key figures of authority within the organization,
the voices of members of a community may be somewhat muted. Alternatively,
communities of practice have the potential to provide a place free from the power
construct evident in the formal organizational structure, offering a space for experi-
mentation and creativity.

Trust. Without trust, members of a community of practice may be reluctant to

share knowledge. While recognizing that defining and comprehending trust is
complex and problematic, Lazaric and Lorenz (1998, p. 3) argue that three
conditions are common in definitions of trust and together they provide a basis for
a general definition. Firstly, trust is identified with an agent’s belief rather than with
his or her behaviour or action. Secondly, trust refers to beliefs about the likely
behaviour of another, or others, which matter for the trustor’s decision-making.
Finally, trust pertains to situations where the complexity of the relationship, or the
fact that it is marked by unanticipated contingencies, precludes having recourse to
complete contingent contracts with third party enforcement.
The presence of a relationship of trust between individuals indicates an ability
to share a high degree of mutual understanding, built upon a common appre-
ciation of a shared social and cultural context. Trust, familiarity and mutual
understanding, developed in their social and cultural contexts, are prerequisites
for the successful transfer of tacit knowledge (Roberts, 2000). Indeed, empirical
evidence suggests that trust leads to higher levels of openness between
co-operative partnerships, thereby facilitating effective knowledge transfer
(Wathne et al., 1996).
Andrews and Delahaye (2000) develop the idea of the psychosocial filter that
ensures that knowledge-sharing decisions are made in a thoughtful and deliberate
manner. This filter, which mediates the process of knowledge importing and
knowledge sharing, includes individual perceptions of approachability, credibility
and trustworthiness. Trust within and between organizations can assist the dissemi-
nation of knowledge. However, as noted above, it is important to recognize that
power shapes social interaction and perceptions concerning its use will influence the
degree of trust among those engaged in knowledge transfer (Roberts, 2000).
© Blackwell Publishing Ltd 2006
 Limits to Communities of Practice 629
In relation to UK organizations, Coopey (1998, p. 366) notes, ‘the ideology and
practices that constitute “management” tend to undermine the foundations on
which trust is built, and hence the processes through which people become com-
mitted to an enterprise, and those through which they learn and innovate.’ This
would suggest that UK organizations are less likely to benefit from the process of
knowledge creation and transfer through social interaction. The nature of work
relations will play an important role determining the success of a community of
practice as a knowledge management tool. Adversarial relations between workers
and management with low levels of trust and strong hierarchical control may fail
to support effective communities of practice. Competition between workers is likely
to discourage the collaborative efforts required in the establishment and mainte-
nance of successful communities of practice. Indeed, communities of practice may
be better suited to harmonious and trusting organizational environments in which
workers are given a high degree of autonomy.

Predispositions. According to Wenger (1998), meaning is negotiated within commu-

nities of practice. However, Bourdieu’s notion of habitus challenges this view
(Mutch, 2003; Gherardi et al., 1998). Habitus is produced from ‘[t]he conditioning
associated with a particular class of conditions of existence’ (Bourdieu, 1990, p. 53).
While Bourdieu’s concept of habitus consists of modes of thought that are uncon-
sciously acquired, resistant to change, and transferable between different contexts,
the communities of practice literature, by contrast, focuses on changes brought
about through practice (Mutch, 2003, p. 388). Mutch (2003) also highlights Bern-
stein’s (1971) distinction between ‘restricted’ and ‘elaborate’ codes developed to
explain the discrepancies between the performances of apprentices from different
social backgrounds. A restricted code relates to a specific context, with a use of
language which rests heavily on shared assumptions about context. This type of
code is limited in the terms and concepts used and is lacking in flexibility and range
of use. An elaborate code, by contrast, would be appropriate to contexts where
assumptions are not shared and where language has to make explicit its claims
(Mutch, 2003, p. 393). Communities that use elaborate codes will be more open to
learning and knowledge creation compared to those using a restricted code.
The existence of habitus and/or specific codes suggest that communities of
practice may well be predisposed to the absorption and creation of certain knowl-
edge and the negotiation of particular types of meaning to the detriment to other
possible interpretations. Meaning is then mediated through predispositions. Indi-
viduals have specific preferences and predispositions; when they join communities
these do not disappear, although they may be moderated. Over time communities
develop preferences and predispositions that will influence their ability to create
and absorb new knowledge.
Communities of practice may become static in terms of their knowledge base and
resistant to change. Knowledge that is aligned with the specific predispositions of a
© Blackwell Publishing Ltd 2006
630 J. Roberts
community and supports the identity and current practices of its members is likely to
be adopted more readily than knowledge that challenges current identity and
practices. Indeed, ways of doing things can become institutionalized within routines
(Nelson and Winter, 1982). Moreover, as Fox (2000, p. 860) notes, ‘community of
practice theory tells us nothing about how, in practice, members of a community
change their practice or innovate’. Given predispositions, the development of
knowledge within a community of practice may become path-dependent as new
knowledge reinforces an existing preference or predisposition. Radical change may
be very difficult to bring about within existing communities, and may be more easily
introduced through the destruction of old communities and the emergence of new
ones. Grabher’s (2004) comparison between the ecology of projects in the computer
software and advertising sectors lends support to this view. He found that project
groups in the computer software businesses studied, which give priority to the
accumulation and sedimentation of knowledge, tended to remain stable over time in
terms of membership, while the project groups in advertising, which prioritize
creativity, change membership regularly in order to retain high levels of creativity
through the recombination of different knowledge. These findings would suggest
that certain activities may be more suited to communities of practice than others.
While communities of practice may support the accumulation of incremental
knowledge developments, they may reduce the scope for radical innovation.

Further Challenges
Size and spatial reach. Communities of practice were originally presented as spon-
taneous, self-organizing and fluid processes (Lave and Wenger, 1991). However, in
later work Wenger (2000), among others (Wenger et al., 2002; Saint-Onge and
Wallace, 2003), suggests that they are not only amenable to manipulation by
organizational designers but can be applied in a wide variety of organizational
contexts. For instance, Wenger et al. (2002), consider communities of practices in
large multinational organizations, including Shell Oil Company, Daimler
Chrysler, Hewlett Packard Company, McKinsey and Company and the World
Bank. Not only are communities of practice applied in large multinational orga-
nization but some are also identified as having very large memberships. For
example, Shell Exploration and Product International Ventures includes a globally
distributed community of more than 1,500 members (Wenger et al., 2002, p. 115).
While it may be possible to identify communities of practice in both small groups
of people working in close proximity and in globally distributed communities of
1,500 people, there is surely a significant difference between these two types of
communities of practice. Is it really possible to apply exactly the same principles to
these two communities?
In some senses large distributed communities can be viewed as a collection of
communities of practice. According to Wenger (1998), a specific community of
© Blackwell Publishing Ltd 2006
 Limits to Communities of Practice 631
practice can be part of any number of constellations of practice, which arise from
interactions among practices involving boundary processes. Wenger (1998, 2000)
identifies a number of boundary processes through which knowledge can be
transferred including brokering, boundary objects (Star and Griesemer, 1989),
boundary interactions and cross-disciplinary projects. For instance, elements of
styles and discourses can travel across boundaries (Wenger, 1998, p. 129), diffusing
through constellation they can be shared by multiple practices and create forms of
continuity that take on a global character. But such styles and discourses may be
integrated into these various practices in very different ways.
The boundaries between communities of practice are not fixed, but flexible,
continuously shifting, porous in nature and difficult to identify. Although commu-
nities may originate in a local context, sustained and repeated interaction facilitated
by various boundary processes may create new spatially extensive communities and
constellations (Coe and Bunnell, 2003, p. 446). For instance, in a study of commu-
nities of practice in a high-technology firm, Teigland (2000, p. 143) identified the
importance of Internet communities that exhibited many of the characteristics of
communities of practice but the individuals involved have typically never met.
Technological developments in transportation and ICTs are increasing the
scope of engagement, but, as Wenger (1998, p. 131) argues, these developments
involve trade-offs that reduce participation in the complexity of situations and their
local meanings. Amin (2002) suggests that organizational or relational proximity,
achieved through communities of practice, may in reality be more important than
geographical proximity. Relational proximity, usually achieved through face-to-
face interaction may also be achieved through ICTs and the mobility of individuals
(Coe and Bunnell, 2003, p. 445). Indeed, Sole and Edmondson (2002, p. 32),
noting the importance of the mobility of people in multi-site teams, claim that
‘. . . dispersed teams may be successful . . . because they have enhanced awareness
of a greater breadth of situated knowledge from which they are . . . better posi-
tioned to learn’.
Constellations of practice, together with other concepts such as fractal structures
for global communities (Wenger et al., 2002, p. 127), help to incorporate spatially
dispersed, virtual, or distributed communities and very large communities.
However, Brown and Duguid (1991) argue that all but the smallest of organizations
should be regarded as communities of communities of practice. They make use of
the term networks of practice to describe relations among members which are
significantly looser than those in a community of practice (Brown and Duguid,
2001b, p. 205). While members of such a network are able to share knowledge
most of them will never know or meet one another.
There is a need to differentiate communities of practice in terms of size and
spatial reach as it is not possible to expand all communities beyond certain limits.
For example, Thompson’s (2005) empirical study of a large global IT hardware
and services organization suggests that there are lower and upper structural and
© Blackwell Publishing Ltd 2006
632 J. Roberts
epistemic parameters for the community of practice investigated. While certain
features may be common to all communities of practice, others may be sensitive to
their scale and geographical spread or their nature and purpose.

W(h)ither the community? The use of term community in the communities of practice
approach has been recognized as somewhat problematic (Contu and Willmott,
2003; Handley et al., 2006; Lindkvist, 2005), this is in part due to the connotations
that it carries with it. For instance, community is traditionally viewed as a warm,
comfortable cosy place characterized by a common understanding (Bauman,
2000). In relation to the communities of practice approach Wenger (1998, p. 5)
defines community as ‘a way of talking about the social configuration in which our
enterprises are defined as worth pursuing and our participation is recognizable as
competence’. As a social configuration a community of practice will reflect the
wider social structures and institutions, or lack of them, evident in the broader
context within which it is situated.
Accordingly, those societies that have strong social structures may well have
stronger and more effective communities of practice in the business environment.
In the USA or the UK, where the pursuit of neo-liberalism with its emphasis on the
market and the individual has eroded the sense of community over the past 25
years, communities of practice may be a less effective means through which to
organize knowledge creation and transfer. While communities in the business and
work environment are attracting increasing attention from managers, academics
and policy makers concerned with leveraging knowledge for competitive advan-
tage, we are simultaneously witnessing the individualization of society (Bauman,
2000).
Is the community in the work environment a substitute for social communities?
In societies that value the individual over the community are workers being
socialized to be effective participants in work-based communities? Those societies
that value community over the individual may be more effective in the use of
communities of practice in the business environment. Differences in socio-cultural
characteristic such as this do impact on the optimal organizational form. For
instance, East Asian business networks and large Anglo-Saxon corporations are
different organizational forms that have emerged in distinct socio-cultural envi-
ronments (Castells, 2000). The demise of community in the social context does not
bode well for the adoption of community type work organizational structures in the
business environment. In addition, communities take time to develop, yet time is
rarely available in the current business environment.

Fast versus slow communities? The contemporary business environment is character-

ized by increasing complexity and the intensification of competition as the forces of
globalization reach out across every region of the economic, political and social
world. Alongside this is a growing demand for continuously improving business
© Blackwell Publishing Ltd 2006
 Limits to Communities of Practice 633
performance resulting from the increasingly short-term perspective of Western
financial markets, together with rapid technological change which is transforming
the competitive landscape. The current era is consequently experiencing the accel-
eration of business activity. This trend it reflected in a range of popular manage-
ment literature encouraging the adoption of acceleration (see for example, Davis
and Mayer, 1998; Gates, 1999).
I would suggest that communities within business organizations will have diffi-
culty forming when the pace of change is accelerating. For instance, restructuring,
downsizing and outsourcing, which are common occurrences in the current busi-
ness environment, are hardly recipes for the development of trusting communities
(Littler et al., 2003). So, how appropriate is the community of practice as a means
of managing knowledge in an era driven by acceleration? Communities of practice
suggest a degree of trust and mutual understanding both of which require time to
develop. We might then regard the traditional community of practice to be a ‘slow
community’. However, in the era of ‘fast capitalism’ groups emerge and dissolve
rapidly. They may have very specific tasks which when complete makes the group
obsolete. Members may not share a joint enterprise, each having a specific aim the
achievement of which feeds into the overall purpose of the group. Such groups may
depend on ‘swift trust’ (Meyerson et al., 1996). Clearly such groupings are not
communities of practice in the traditional sense, but the existence of such ‘fast
communities’ suggests that the communities of practice considered here may be
relevant only to those organizations or parts of organizations that have the luxury
of sustained engagement. They may only relate to areas within large organizations
which are protected from the vagaries of the short-term competitive environment
in order to focus on longer-term concerns.
Lindkvist (2005) develops an alternative, though complementary, view of com-
munities of practice in the form of collectivities of practice, to refer to temporary
groups or project teams concerned with knowledge creation and exchange. These
collectivities of practice are temporary and, like fast communities, they are estab-
lished quickly. Lindkvist (2005, p. 1200) notes that such groups or teams will not be
sufficiently developed to qualify as communities of practice, nevertheless ‘groups
with members that embrace a collective goal and have a good representation of
what others know, may thus, based on quite a minimalist base of shared knowl-
edge, develop a pattern of interaction and the collective competence needed’.
While communities of practice depend on shared enterprise, mutual engagement
and shared repertoire, collectivities of practice rely on individual knowledge,
agency and goal-directed interaction. There are then different types of knowledge
creating and transferring communities.
While communities of practice within business organizations may be more
difficult to develop and sustain in a rapidly changing business environment, com-
munities of practice that exist independently of business organizations may take on
an increasingly important role in the creation and transfer of knowledge. For
© Blackwell Publishing Ltd 2006
634 J. Roberts
instance, in the film industry individuals come together to create a film and once
this is achieved they disperse, yet they remain members of film making community
even when they are no longer employed by a film producing business organization.
The shared enterprise, mutual engagement and shared repertoire of the film
making community are brought to bear in the temporary project of the production
of a film, but it is in these extra-organizational communities of practice that new
members gain legitimate peripheral participation and over time become full par-
ticipants. Employment opportunities will come and go but membership of a com-
munity of practice becomes an important constant in the lives of certain workers in
the current accelerated business environment.

LIMITS TO COMMUNITIES OF PRACTICE

This section provides some reflections on the key limits to the communities of
practice approach as a knowledge management tool and offers some suggestions
for further research with a view to developing a more robust approach. To begin
with, it is evident from the discussion of issues of power, trust and predispositions
that a community of practice does not develop and function in a vacuum. The
context within which a community of practice is embedded is a major factor
determining its success as a means of creating and transferring knowledge. More-
over, broad socio-cultural factors may or may not be conducive to the success of
communities of practice as a knowledge management tool. National competi-
tiveness deriving from knowledge creating and sharing capabilities may then vary
depending on nation-specific socio-cultural characteristics, such as levels of trust
or the relative position of the individual versus the community. For instance,
while the individualization of society, and the consumerism that it promotes, is of
benefit to business in terms of the creation of markets and the stimulation of
demand, it is less conductive to collaborative working practices. Paradoxically,
community is becoming more important on the supply side of markets in terms
of the organization of production and the creation of new products. The com-
munities of practice as a tool of knowledge management is likely to be more
successful in those regions and nations that have a strong community spirit com-
pared to those nations that have a weak community spirit. For instance, in rela-
tion to Hofstede’s (1991) study of national culture, we might expect that a nation
characterized by collectivism might find that the community of practice is a more
effective knowledge management strategy than nations characterized by indi-
vidualism. If this is the case, then a country wishing to excel in knowledge cre-
ation and dissemination through the development of communities of practice will
need to promote community in the wider society through education and social
infrastructures.
One of the strengths of the communities of practice approach is that it can be
applied in a wide range of organizational settings. However, this can also be viewed
© Blackwell Publishing Ltd 2006
 Limits to Communities of Practice 635
as a weakness, since it may encourage its inappropriate application. For instance,
Wenger et al. (2002) suggest that the communities of practice approach can
usefully be applied to civil society and the public sector. Indeed, it is already being
promoted in the public sector, for instance, as part of a knowledge management
tool kit developed for the UK’s National Health Service. In contrast, Kimble and
Hildreth (2004, p. 5) question whether communities of practice are always suitable
for the business setting, arguing that their interests may not be aligned with those
of the organization and ‘because they are self-managed and self directed, their
contribution to the organization will always be uncertain’. They go on to argue that
workers increasingly operate in an individualistic world of weak ties where
resources are frequently obtained through personal networks and individual rela-
tionships rather than through organization based communities. Individuals belong
to a variety of communities of practice some internal to their work organization
while others will be external arising from their personal and professional networks.
Hence, I would suggest that for business organizations to fully leverage their
knowledge capacities they must seek to harness communities of practice that are
both within and beyond their organizational boundaries. Given that knowledge is
transferred through social interaction within communities, then businesses need to
pay particular attention to their recruitment and training policies to ensure that
they maintain an appropriately skilled workforce to maximize the benefits of
communities of practice.
A further limitation to the approach may relate to its relevance for small and
medium sized organizations. Communities of practice require cultivation if busi-
ness organizations are to fully exploit their benefits (Wenger et al., 2002), they will
not flourish in inhospitable organizational environments. But are small and
medium-sized organizations able to spare the necessary resources to cultivate
communities of practice? Very small organizations and professional practices,
where communities form spontaneously, may be highly amenable to the use of the
communities of practice approach. However, while very small and large organi-
zations can use communities of practice to encourage knowledge creation and
sharing, due to resource limitations small and medium sized firms may be less able
to exploit these methods of knowledge management.
It may be possible to witness the development of a divide between workers in
terms of those who work primarily within communities of practice and those that
work within hierarchically controlled structures. In a sense, a divide already exists
between the knowledge worker and the unskilled worker. The nature of the
appropriate knowledge management structure will depend on the nature of the
knowledge being applied in the production process. In highly standardized pro-
duction systems knowledge is codified and easily transferred and applied, workers
are not required to use their own discretion. In such environments the production
process can be hierarchically organized. However, when knowledge is tailored in a
customized production system then a more decentralized organizational structure
© Blackwell Publishing Ltd 2006
636 J. Roberts
would be more appropriate, allowing workers more autonomy to use their tacit
knowledge to satisfy client requirements. Communities of practice may be most
suited to such contexts.
Perhaps, in relation to the development of our understanding of knowledge
creation and diffusion we need, as Brown and Duguid (2001b, p. 203) suggest, to
focus more on the practice rather than on the community, for ‘[t]he appeal of
community has tended to obscure the importance of practice’. Alternatively, as
Contu and Wilmott (2003) argue, a renewed emphasis on situated learning theory
may be of value. In addition, it would be helpful to examine the nature of
communities of practice of various sizes, in different sectors, and in a variety of
socio-cultural contexts. In so doing, it may be possible to develop a detailed
classification of types of communities of practice together with the appropriate
organizational context for their successful application.

CONCLUSION
Several directions for further research emerge from the discussion of communi-
ties of practice provided in this paper. Firstly, given that the broad socio-cultural
environment will impact on the success of the community of practice as an
approach to knowledge management, research needs to take account of this.
Included within the broad socio-cultural environment is the relative weight given
to the individual versus the community. A comparative investigation of commu-
nities of practice in very different socio-cultural environments such as the USA
and China would shed light on the impact of this broader context. A second area
for research concerns the organizational context. How do communities of prac-
tice interact with the formal structure of an organization? In which organiza-
tional contexts is the communities of practice approach the most appropriate
knowledge management tool? Furthermore, the boundaries of a community of
practice may not reflect organizational boundaries. Hence, an appreciation of
the interaction between formal organizations and extra-organizational commu-
nities of practice is required. For instance, how can organizations leverage their
access to such communities to build their internal knowledge assets? Third, an
understanding of the variations in the prevalence and success of communities of
practice in organizations of different sizes and in diverse sectors would be helpful.
As also would an appreciation of the variations between communities of practice
of disparate sizes and spatial distributions. Finally, there is a need to refocus on
Lave and Wenger’s original conceptualization communities of practice as a
context for situated learning. An area that is a given further consideration by
Handley et al. (2006; this issue), who give particular attention to issues of par-
ticipation, identity and practice.
Despite the limitations highlighted here, it is important to remember that the
communities of practice is, in a sense, still an evolving approach to knowledge
© Blackwell Publishing Ltd 2006
 Limits to Communities of Practice 637
management. Over the coming years, as communities of practice are applied and
studies in an increasing number of organizational contexts, we will gain a deeper
understanding of the strengths and weaknesses of the approach. Notwithstanding
its current limitations, the communities of practice approach does provide us with
a means to explore the transfer of tacit knowledge in a social context. In this sense,
it provides a valuable alternative to the knowledge management tools focused on
the codification of knowledge. However, the community of practice is one of a
number of knowledge management tools and each organizational environment
requires a different set of tools. Additional mechanisms must be developed to
manage tacit knowledge since it is clear from the discussion above that the com-
munity of practice is not always an appropriate knowledge management tool.

NOTES
[1] This is evident in the growing number of articles written on the subject each year. For instance,
a search of the EBSCO Business Source Premier reveals a rise from four articles in 1993 to 63 in 2003,
and 75 in 2004. EBSCO Business Source Premier provides full text for nearly 7,600 scholarly business
journals and other sources, including full text for more than 1,125 peer-reviewed business
publications.
[2] The French firm Knowings is an example of a consultancy promoting the community of practice
as a knowledge management tool, details available at www.knowings.com (last accessed 30
November 2005).

REFERENCES
Amin, A. (2002). ‘Spatialities of globalisation’. Environment and Planning A, 34, 385–99.
Andrews, K. M. and Delahaye, B. L. (2000). ‘Influences on knowledge processes in organizational
learning: the psychosocial filter’. Journal of Management Studies, 37, 6, 797–810.
Barley, S. and Orr, J. (Eds) (1997). Between Craft and Science: Technical Work in the United States. Ithaca,
NY: Cornell University Press.
Bauman, Z. (2000). Liquid Modernity. Cambridge: Polity Press.
Bernstein, B. (1971). Class Codes and Control, Volume I: Theoretical Studies Towards a Sociology of Language.
London: Routledge & Kegan Paul.
Blackler, F. (1995). ‘Knowledge, knowledge work and organizations: an overview and interpretation’.
Organization Studies, 16, 6, 1021–46.
Blackler, F. and McDonald, S. (2000). ‘Power, mastery and organizational learning’. Journal of
Management Studies, 36, 4, 833–51.
Boland, R. J. and Tenkasi, R. V. (1995). ‘Perspective making and perspective taking in communities
of knowing’. Organization Science, 6, 4, 350–72.
Bourdieu, P. (1990). The Logic of Practice. Cambridge: Polity Press.
Brown, J. S. and Duguid, P. (1991). ‘Organizational learning and communities of practice: towards
a unified view of working, learning and innovation’. Organization Science, 2, 40–57.
Brown, J. S. and Duguid, P. (1998). ‘Organizing knowledge’. California Management Review, 40, 3,
90–111.
Brown, J. S. and Duguid, P. (2001a). ‘Structure and spontaneity: knowledge and organization’. In
Nonaka, I. and Teece, D. (Eds), Managing Industrial Knowledge. London: Sage, 44 –67.
Brown, J. S. and Duguid, P. (2001b). ‘Knowledge and organization: a social-practice perspective’.
Organization Science, 12, 2, 198–213.
Carlile, P. R. (2002). ‘A pragmatic view of knowledge and boundaries: boundary objects in new
product development’. Organization Science, 13, 4, 442–55.

© Blackwell Publishing Ltd 2006

638 J. Roberts
Castells, M. (2000). The Rise of the Network Society, 2nd edition. Oxford: Blackwell.
Coe, N. M. and Bunnell, T. G. (2003). ‘“Spatializing” knowledge communities: towards a concep-
tualisation of transnational innovation networks’. Global Networks, 3, 4, 437–56.
Contu, A. and Willmott, H. (2000). ‘Comment on Wenger and Yanow. Knowing practice: a “delicate
flower” in the organizational learning field’. Organization, 7, 2, 269–76.
Contu, A. and Willmott, H. (2003). ‘Re-embedding situatedness: the importance of power relations
in learning theory’. Organization Science, 14, 3, 283–96.
Coopey, J. (1998). ‘Learning to trust and trusting to learn’. Management Learning, 29, 3, 365–82.
Coopey, J. and Burgoyne, J. (2000). ‘Politics and organizational learning’. Journal of Management Studies,
36, 4, 869–85.
Davis, S. and Mayer, C. (1998). Blur: The Speed of Change in the Connected Economy. Oxford: Capstone.
Foucault, M. (1979). Discipline and Punish: The Birth of the Prison. Harmondsworth: Penguin.
Fox, S. (2000). ‘Communities of practice, Foucault and actor-network theory’. Journal of Management
Studies, 37, 6, 853–67.
Gates, B. (with Hemingway, C.). (1999). Business @ The Speed of Thought: Succeeding in the Digital Economy.
London: Penguin.
Gherardi, S., Nicolini, D. and Odella, F. (1998). ‘Towards a social understanding of how people learn
in organizations’. Management Learning, 29, 3, 273–98.
Grabher, G. (2004). ‘Temporary architectures of learning: knowledge governance in project ecolo-
gies’. Organization Studies, 25, 9, 1491–514.
Handley, K., Sturdy, A., Fincham, R. and Clark, T. (2006). ‘Within and beyond communities of
practice: making sense of learning through participation, identity and practice’. Journal of
Management Studies, 43, 3, 641–53.
Hofstede, G. (1991). Cultures and Organizations: Software of the Mind. London: McGraw-Hill.
Johnson, C. M. (2001). ‘A survey of current research on online communities of practice’. Internet and
Higher Education, 4, 45–60.
Kimble, C. and Hildreth, P. (2004). ‘Communities of practice: going one step too far?’. Proceedings 9e
L’AIM, Evry, France, May.
Lave, J. and Wenger, E. (1991). Situated Learning: Legitimate Peripheral Participation. Cambridge: Cam-
bridge University Press.
Lazaric, N. and Lorenz, E. (1998). ‘Introduction: The learning dynamics of trust, reputation and
confidence’. In Lazaric, N. and Lorenz, E. (Eds), Trust and Economic Learning. Cheltenham, UK,
Northampton, MA: Edward Elgar, 3.
Lesser, E. L. and Storck, J. (2001). ‘Communities of practice and organizational performance’. IBM
Systems Journal, 40, 4, 831–41.
Lindkvist, L. (2005). ‘Knowledge communities and knowledge collectivities: a typology of knowledge
work in groups’. Journal of Management Studies, 42, 6, 1189–210.
Littler, C. R., Wiesner, R. and Dunford, R. (2003). ‘The dynamics of delayering: changing manage-
ment structures in three countries’. Journal of Management Studies, 40, 2, 225–56.
Marshall, N. and Rollinson, J. (2004). ‘Maybe Bacon had a point: the politics of interpretation in
collective sensemaking’. British Journal of Management, 15, S71–S86.
Meyerson, D., Weick, K. E. and Kramer, R. M. (1996). ‘Swift trust and temporary groups’. In
Kramer, R. M. and Tyler, T. R. (Eds), Trust in Organizations. Thousand Oaks, CA: Sage, 166–95.
Mutch, A. (2003). ‘Communities of practice and habitus: a critique’. Organization Studies, 24, 3,
383–401.
Nelson, R. R. and Winter, S. G. (1982). An Evolutionary Theory of Economic Change. Cambridge, MA and
London, UK: The Belknap Press of Harvard University Press.
Orr, J. E. (1996). Talking about Machines: An Ethnography of a Modern Job. Ithaca, NY and London, UK:
IRL Press, an imprint of Cornell University Press.
Pan, S. L. and Leidner, D. E. (2003). ‘Bridging communities of practice with information technology
in pursuit of global knowledge sharing’. Journal of Strategic Information Systems, 12, 71–88.
Roberts, J. (2000). ‘From know-how to show-how: the role of information and communication
technologies in the transfer of knowledge’. Technology Analysis and Strategic Management, 12, 4,
429–43.
Saint-Onge, H. and Wallace, D. (2003). Leveraging Communities of Practice for Strategic Advantage. London
& New York: Butterworth Heinemann.
Smeds, R. and Alvesalo, J. (2003). ‘Global business process development in a virtual community of
practice’. Production Planning & Control, 14, 4, 361–71.

© Blackwell Publishing Ltd 2006

 Limits to Communities of Practice 639
Sole, D. and Edmondson, A. (2002). ‘Situated knowledge and learning in dispersed teams’. British
Journal of Management, 13, S17–34.
Star, S. L. and Griesemer, J. R. (1989). ‘Institutional ecology, “translations” and boundary objects:
amateurs and professionals in Berkeley’s Museum of Vertebrate Zoology, 1907–39’. Social Studies
of Science, 19, 3, 387–420.
Swan, J., Scarbrough, H. and Robertson, M. (2002). ‘The construction of “communities of practice”
in the management of innovation’. Management Learning, 33, 4, 477–96.
Teigland, R. (2000). ‘Communities of practice in a high-technology firm’. In Birkinshaw, J. and
Hagström, P. (Eds), The Flexible Firm: Capability Management in Network Organizations. Oxford:
Oxford University Press.
Thompson, M. (2005). ‘Structural and epistemic parameters in communities of practice’. Organization
Science, 16, 2, 151–64.
Wathne, K., Roos, J. and von Krogh, G. (1996). ‘Towards a theory of knowledge transfer in a
cooperative context’. In von Krogh, G. and Roos, J. (Eds), Managing Knowledge: Perspectives on
Cooperation and Competition. London: Sage.
Wenger, E. (1998). Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge
University Press.
Wenger, E. (2000). ‘Communities of practice and social learning systems’. Organization, 7, 2, 225– 46.
Wenger, E. C. and Snyder, W. M. (2000). ‘Communities of practice: the organizational frontier’.
Harvard Business Review, January–February, 139– 45.
Wenger, E., McDermott, R. and Snyder, W. M. (2002). Cultivating Communities of Practice: A Guide to
Managing Knowledge. Boston, MA: Harvard Business School Press.
Yanow, D. (2004). ‘Translating local knowledge at organizational peripheries’. British Journal of
Management, 15, S9–25.

© Blackwell Publishing Ltd 2006