The Metaphor of Networks in Learning: Communities,
Collaboration and Practice
Chris Jones1 and Liliane Esnault2
Lancaster
University1, EM Lyon2
c.r.jones@lancaster.ac.uk, esnault@em-lyon.com
This paper explores the use of the network metaphor and the way in which
it relates to Computer Supported Collaborative Learning (CSCL) and Communities
of Practice. The idea of networked learning stresses the interaction of
learners, tutors and their resources through networks. The arguments put in
this article are firstly that learning technology needs to take account of the
wider debate about networks and secondly that research in this field needs to
address the theoretical and practical issues raised by advances in the field of
networks. A further argument made is that the idea of the network can act as a
unifying concept allowing us to bring together apparently disparate elements in
the field of e-learning.
Networks, networked learning, communities of practice, collaboration,
CSCL.
This paper argues that using the metaphor of networks can assist us to
conceive of the broad context in which learning and education take place in a
society reliant on computer networks. The paper will try to provide some strong
indications of how a research programme could be elaborated and some
suggestions for practical guidance and applied outcomes arising from the
network metaphor. Network analysis provides a useful focus for analyzing the
patterns of growth and interaction in a wide range of fields.
The Centre for Studies of Advanced Learning Technology (CSALT) group at
Lancaster University has been associated with the following definition of
networked learning.
Networked learning
is learning in which information and communication technology (C&IT) is
used to promote connections: between one learner and other learners, between
learners and tutors; between a learning community and its learning resources.
The key element of this definition is the term connections. The notion of
learning emphasized in this definition is a relational view in which learning
takes place in relation to others and in relation to an array of resources.
Networked learning doesn’t privilege the type of relationships between people
or between people and resources and in this it differs from two of the most
popular approaches to the use of computers and networks in an educational
setting CSCL and communities of practice. For CSCL, however defined, the
relationship is one of cooperation or collaboration and in terms of communities
of practice the relationship is one that implies both the closeness of
community and a certain unity of purpose. This paper will argue that networked
learning is a more appropriate term to describe the types of relationship
emerging in educational settings and within the broader scope of a networked
society.
Several terms have been used to identify the emergence of new or revised
pedagogies related to the introduction of computer technologies in education.
Of these perhaps CSCL has proved the most attractive and developed. CSCL has
been used as a term in its own right (Koschmann 1996) and as an acronym for
both Computer Supported Cooperative and Computer Supported Collaborative
Learning. In all of these uses the term has signified a practical interest in
computers and group activity and a privileging of social and situated views of
learning. Koschmann for example returned to this theme five years later:
CSCL research has the advantage of studying
learning in settings in which learning is observably and accountably embedded
in collaborative activity. Our concern, therefore, is with the unfolding
process of meaning-making within these settings, not so-called “learning
outcomes”. It is in this way that CSCL research represents a distinctive
paradigm within IT. By this standard, a study that attempted to explicate how
learners jointly accomplished some form of new learning would be a case of CSCL
research, even if they were working in a setting that did not involve
technological augmentation. On the other hand, a study that measured the
effects of introducing some sort of CSCL application on learning (defined in
traditional ways) would not. (Koschmann 2001)
This approach is a way of seeing, a
theoretical lens or paradigm, rather than simply a concern with the practical
application of new network and computing technologies. The problem lies in the
admission that this paradigm is not specific to the technological setting
rather it is a theory of learning in general. The claim made here is that
social and situated views of learning whilst compatible with networked learning
are not and should not be a pre-condition. The authors of this paper hold
social and situated views of learning but we argue that the field can be open
to study from other perspectives. Networked learning begins from the setting
itself and does not imply any one particular paradigm of learning.
The idea of CSCL, although Tim
Koschmann uses it without unpacking the acronym into its parts, implies the
computer is the focus of attention. In Networked Learning this focus is
displaced from the particular device, the computer, to the non-specific
location of the network. The network is non-specific because it does not reside
in any particular device or location. It is rather like the idea of the
university itself in that you can illustrate the network by pointing to
particular elements but the network is a collective term expressing a set of
persistent relationships over time and analytically above its component parts.
The term CSCL also draws attention to particular kinds of relationship, those
of cooperation and collaboration. These terms are not neutral and although they
draw on etymological roots that simply indicate working together they suggest a
moral imperative for close forms of coordination and cohesion rather than
looser relationships. Conflict is identified as a potentially productive part
of a learning process, but though some Piagetian influenced proponents of CSCL
recognize, this it is not a central characteristic of CSCL (see Dillenbourg
1999). A further feature of a network understanding of learning is that it
draws attention to the potential strength of weak ties. The network metaphor
doesn’t privilege the closeness of community rather it serves to encompass all
kinds of links and relationships. More generally the form of networked society
in which networked learning takes place has been described by Castells
as one of 'networked individualism'
(1996, 2001) not one of close community.
Communities of practice are a generalized form of learning practices
found in a variety of societies. The idea of Communities of Practice, unlike
CSCL was not linked to new technologies in its origins; rather it emerged from
anthropological studies that were often in pre-modern social systems such as
tailors in Liberia. Communities of practice in their turn rest on the
apprenticeship model that Jean Lave generalized in terms of learning as
legitimate peripheral participation. Communities of practice involve a process
of relatively close engagement in a community with the distinguishing feature
being the sharing of practice. Shared practice in turn requires members to have
the time and space to collaborate (Lave and Wenger 1991; Wenger 1998). A number
of questions arise from the translation of the idea of communities of practice
into networked environments. Community as a term is loaded, like the terms
cooperation and collaboration with a preference for close ties. Both Wenger
(1998) and Brown and Duguid (2001) note that there are relationships between
certain types of network and communities of practice. Wenger notes in a number
of footnotes to the main text the work of Wellman and Berkowitz (1988) and
suggests that a community is similar to a network composed of strong ties
(Wenger 1998 p283). Wenger distinguishes networks from communities of practice
in a number of ways.
The key divergence between networks and communities of practice for
Wenger lies in the emphasis placed on structural properties and process. Wenger
emphasizes process elements and suggests that the key distinguishing element of
communities of practice when viewed from a more structural point of view is
that they are cohesive, sharing historical processes that are composed of
strong ties.
Early sociologists of modern society distinguished, in a variety of ways
between pre-modern communities and the modern scientific-rational structures of
the city and industry. The classic formulation of this was the contrast between
Gemienschaft and Gesellschaft by Tonnies in the 19th Century. Gemeinschaft was close community based
on kinship, identity and place whereas Gesellschaft
was a rational calculating and contractual form of relationship. Brown and Duguid reference Tonnies and the roots of
sociological debates on community, but oddly in the main text they only mention
gemeinschaft (Brown and Duguid 2001
p202). This omission is odd because Brown and Duguid are discussing modern
business organization. The social forms of late capitalism are not obviously
communal in form and in general the process of individualization has continued
in the modern period. In a footnote Brown and Duguid note that gemeinschaft and gesselschaft are not temporally successive ways of organizing
society or even mutually exclusive but it is clear that the emphasis place on gemeinschaft in their work is an
emphasis on types of community most common in pre-modern society and as Fox
(2002) has noted such communities only exist in the “interstices” of modern
organizations. Communities of practice of the type identified by Lave and
Wenger are not the central organizational forms in contemporary social systems,
they are in many ways marginal, inhabiting the informal regions that survive
outside of and often in spite of mainstream relationships. Communities of
practice have become popular in modern business and management as a corrective,
identifying self-activating unit within large organizations that are capable of
significant self-management. Communities of practice are not a neutral
description of aspects of modern societies they are mobilized by modern
business organizations to reduce the administrative load on formal structures
and reposition responsibility on informal structures and organization. In a
similar way educational use of Communities of practice puts additional
responsibilities upon students who are required to develop, monitor and control
their own groups and relieves the burden from the formal teaching structure
that adopts a more facilitative or moderating role.
Network theory would suggest that the strong notions of community
contained in communities of practice might ignore the importance of the
'strength of weak links'. The idea of weak ties has recently been applied in
relation to communities of practice (Rosson 2003). Rosson argues that the
social nature of Internet use by people who act as weak links, in her terms
‘bridges’, suggests that the Internet is used by them for maintaining relations
and increasing face-to-face interaction. Weak ties are in this view an enabling
factor in social activism and the building of ‘social capital’. The educational
focus in networked learning has often been on strong links and the emphasis on
community may have made less visible the many necessary but weak connections
that make the network idea so powerful. The nature of networked learning is
such that whether the network is used for distance or largely place based
learning the participants do not have to be co-present. The student cohort in a
networked course may well have weaker ties with each other and with the tutor
than might be expected in terms of a community of practice. Student cohorts
often do not have a history and may never become cohesive units. For these
reasons we argue that a network analysis might be more appropriate.
Another key issue in communities of practice and social practice theories
of learning has been knowledge sharing (Osterlund and Carlisle 2003). Osterlund
and Carlisle claim that a relational thinking lies at the heart of social
practice theories. They claim that subjects or social groups only develop their
properties in relation to other subjects or social groups. In particular
“social objects derive their significance from the relations that link them
rather than from the intrinsic features of individual elements.” (Osterlund and
Carlisle 2003 p3). This relational view borders on a network description that
privileges the links rather than the properties of individual nodes. The
authors go on to claim that what they call practice theory goes beyond other
theories by not only looking at the “recursive dynamics of a given relation but
places everyday practice as the locus of the production and reproduction of
relations.” (ibid p3). The issue of knowledge sharing is developed further in
relation to the idea of networks of practice below.
Etienne Wenger has noted that the intensity of interaction between people
distinguishes between a community of practice and a personal network (Wenger
1998 p126). He goes on to note that
Some configurations
are too far removed from the scope of engagement of participants, too broad,
too diverse, or too diffuse to be usefully treated as a single community of
practice. (Wenger 1998 p126)
Wenger notes this limitation is not one solely of scale, affecting both
large and small configurations. To describe this type of broad and diffuse
configuration Wenger uses the term constellation, specifically drawing
attention to the use of the term grouping stellar objects even though they may
not be close to each other. Constellations of practice are used by Wenger to
describe relations that link a community of practice with any number of
constellations. In part Wenger claims communities of practice define themselves
by negotiating their place within a variety of constellations. Constellations
of practice in this account are intimately connected with the negotiation of
boundaries and interactions among practices. They are engaged in the ‘export’
of practice, allowing detachment from any specific enterprise. The exported
styles and discourses whilst not practices themselves provide resources that
can be used in the context of practice.
A particular question that arises from Wenger’s analysis is the nature of
the geography of place in communities of practice. The claim made is that the
introduction of constellations of practice adds notions of locality, proximity
and distance to the ideas of boundaries and peripheries. The setting out of
these issues indicates the potential importance for networked learning. Issues
of locality, proximity and distance are central issues in networked learning
but in terms of communities of practice they are issues little explored despite
Wenger’s careful provision of the constellation idea to accommodate such
issues. A related development of communities of practice has been popularised
by Brown and Duguid (2001) as networks of practice.
Brown and Duguid note that the appeal of the term community has tended to
obscure the importance of practice. They wonder whether if Lave and Wenger had
spoken about cadres or communes of practice this would have been as powerful or
as widespread a notion. By practice Brown and Duguid mean “undertaking or
engaging fully in a task, job, or profession.” (Brown and Duguid 2001 p203). A
second question Brown and Duguid raise is in relation to the relationship
between communities of practice and other forms of social alignment. They
comment that communities of practice can seem indifferent to other forms of
social alignment; in particular they can seem a “social monad – a fundamental
building block” (Brown and Duguid 2001 p203). This atomic view of communities
of practice obscures the social heterogeneous nature of communities of practice
and could be thought of as ‘communities of communities of practice’. For Brown
and Duguid a particular reason for the importance of this issue is that it
touches on the question of ‘disembedding’ and ‘reembedding’, and that new
technologies allowing communication across time and space.
This issue echoes Wenger’s concern with the export of repertoires and
styles. Central to both accounts is the need for conditions at both ends of an
exchange to allow for a flow of information and knowledge and the disembedding
or export and reembedding or import of discourses and repertoires originating
in one practice to be incorporated in another. This issue, in another tradition
referred to as transfer, is central to education and networked learning. It is
the difficulty of dealing with this in the classic articulation of communities
of practice that suggest the adoption of the metaphor of networks. Brown and
Duguid use the term networks to indicate loose epistemic groups and note that
most people in such a network will never know, know of, or come across one
another. The argument of this paper is that the term should have a more general
currency, that networks should cover not only very distant relationships but
also relationships that have varying degrees of proximity but do not have the
degree of cohesion required for a community. The use of the term network allows
for scalability in analysis as networks can have a nested character. Each node
in a network can itself be a network; the atomic nature of the community
metaphor can be replaced by a fractal geometry allowing for an infinite
repetition of similar but not exact forms in various scales. Finally the
network metaphor does not privilege any one particular view of relationships,
and this may be especially important for learning.
There has been a significant focus in research and practical work related
to CSCL and networked learning on the role of the moderator or facilitator. As
an illustration of the ways in which the network metaphor can be applied to the
practical management of networks we explore the role of the animator. For
purposes of clarity we have adapted the French language term
animateur/animatrice to indicate this type of role within a network. We want to
indicate the place of the network animator and how this description might be
applied to a role within a network and relate this to the more common terms
moderator and facilitator. The paper will try to abstract elements in a network
animator’s role and from experience in networked organizations suggest
necessary and desirable characteristics for successful animation within a
networked learning setting. Issues examined will include the need for
explicitness about network relations and the degree and type of regulation
required within a networked system.
The animator’s role can be taken by an individual or distributed across a
number of individuals. The animator role can be critical to a networks
functioning. The animator role is not a minor function for a network and could
be considered an essential component for designed systems that rely on
systematic rather than simply organic network development. This would apply to
most networked learning courses that rely on the development over time of a
dense set of interrelationships between students, students and their tutors and
students and the resources that are required for learning. For networks to grow
organically members of the network have to give something to others in the
network. The essential relationship is reciprocity. This does not mean all
persons in a network need to be highly active. It is quite possible that a
proper relationship in a network could be providing an interested audience for
others. It does mean that members of a network must be self-activating;
networks cannot develop if all await the others to begin. Networks are dynamic
systems that rely on self-reinforcement, without sustaining activity networks
become moribund, the non-animate links and connections may remain, skeleton
like, but without activity the flows across the network die out.
Networks develop in a learning context bringing together people from a
wide variety of backgrounds. Students bring different backgrounds that can
imply hierarchic positions. These can flow from broader cultural influences
including reputation, force of personality, expertise, and professional
position. Part of an animator’s role can be to manage these issues of power in
ways that enable the network to function in ways that enhance learning. This
might mean identifying those persons or elements in a network that constitute a
potential risk and those other elements that positively assist the network.
Such a role mirrors the social functions identified for moderators and
facilitators but has a distinctly less positive gloss. The network animator may
be concerned to inhibit certain features and to mobilize others to act as a
counterweight. The animator role may not be taken up by any one individual but
it is important that this coordinating activity is taken up somewhere in the
network. A further example of the coordinating role is that of integration.
Networks are complex systems and different elements in a network will not have
an overall view. The production of an overall sense of the network and its
activities is an ongoing process and a network may not need to or be able to
generate a global view of the network itself. No singular view may develop and
networks lend themselves to multiple representations and integration does not
mean the imposition of any singular view. The animator role is to try and
encourage such integration and the distribution of information across the
network. It is also to encourage both process and practice and the reification
of network activity into products of one kind or another. Network activity
without reification can be highly ephemeral.
The monad or atomic view of communities of practice as the foundational
building block contains within it definite problems when applied to networked
learning. The idea of constellations of practice conjures up the image of a
series of Venn diagrammes with each community of practice represented by a
circle and overlaps of circles indicating the areas in which communities of
practice overlap with wider constellations of practice. The network of practice
metaphor allows for the linking of much looser groupings but doesn’t follow
this down to the problem of each community being represented as a monad or
basic atomic unit. Network analysis allows for the node in one network to be a
part of or an entire other network. Networks are porous by nature, including
long weak linkages as well as strong ties. The community of practice metaphor
privileges the idea of strong ties and a cohesive view of community. For these
reasons we argue that networked learning provides a better metaphor than
communities of practice for technologically enhanced learning. We also argue
that the term networked learning better captures the setting for learning in a
networked society and using computer networks. The term CSCL privileges
cooperation and collaboration above other relationships and directs attention
to the computer itself and not the network. We argue this is something like a
category mistake, as for example when pointing to a campus building and asking
if that is the university. The network may be accessed through computers and
other devices but it is a complex self-organizing system that cannot be
identified as one of its component parts.
Benedict Andersen spoke in terms of imagined communities (1983). He noted
that it was highly unusual for all the members of a community to know one
another and that it was an act of imagination that allowed members of a
community to identify with each other. In education the use of the communities
of practice metaphor has led many to imagine communities. Some communities of
practice emerge naturally within modern organizations and in education
including networked learning environments. They are not the predominant form
for networked learning or for modern organization. There is a danger of course
that something similar might happen using the network metaphor, that by talking
in terms of networked learning we actively constitute them in our practice. We
see this as a danger and agree with John Law that it is possible:
that we are in the
process of uncritically reproducing some
kind of dominant ideology. We are reproducing the ways in which the current
orderings of the world like to represent themselves. (Law undated draft para No
1)
As such networked learning could become part of a hegemonic discourse not
simply in educational terms but as part of wider debates concerning the nature
of social processes, power and culture.
Networks provide a framework for analyzing the new relationships that
move beyond simple geography in networks. Andreas Wittel has suggested a move
in ethnographic research that takes account of this shift from geographic
metaphors.
Networks are still strongly related to geographical space - like field.
Unlike field, a network is an open structure, able to expand almost without
limits and highly dynamic. And even more important: A network does not merely
consist of a set of nodes but also a set of connections between nodes. As such,
networks contain as much movement and flow as they contain residence and
localities. (Wittel 2000 paragraph [5])
A final claim made in this paper is that the use of the network metaphor
in learning technology helps us to connect ourselves to wider social debates
about the networks and helps us to think about the fundamental nature of the
network and the patterns of activity associated with it. These debates around
the network society (Castells 1996, 2001) and social applications of network
analysis in terms of actor network theory (Fox 2002, Law undated) relate to the
modeling of networks and descriptions of phenomena in terms of nodes and the
links between them. Network analysis is interested in the ways in which
transfers can take place across a network, for example whether the network is
traversable or not. The importance of this field of study is that it holds out the
prospect of developing mathematical ways of describing networks that may prove
to be robust across a broad range of phenomena. A number of texts aimed at lay
readers, originating in the mathematical and physical science traditions, have
begun to examine phenomena from a wide range of areas, including social and
biological domains (Barabasi 2002, Buchanan 2002). Our conclusion is that
networked learning provides a more persuasive metaphor for learning using
modern network technologies than either CSCL or communities of practice.
This paper was written as part of the work conducted for the EU project
EQUEL and the work and discussions in SIG 6 in particular.
Anderson, B. (1983) Imagined Communities. London:
Verso
Barabasi, A-L. (2002) Linked:
The New Science of Networks. Cambridge, MA: Perseus Publishing.
Brown, J.S., and Duguid, P. (2001) Knowledge and
Organization: A Social-Practice Perspective. Organization Science. Vol
12 (2) pp198 – 213.
Buchanan, M. (2002) Nexus:
Small Worlds and the Groundbreaking Science of Networks. New York &
London : W.W. Norton & Co
Castells, M. (1996). The
Information Age: Economy, Society and Culture Volume 1. The Rise of the Network
Society. Oxford: Blackwell.
Castells, M. (2001) The
Internet Galaxy: Reflections on the Internet, business, and Society.
Oxford: Oxford University Press.
Dillenbourg, P. (Ed) (1999) Collaborative Learning:
Cognitive and Computational Approaches. Amsterdam: Pergamon.
Fox, S. (2002) Studying
Networked Learning: Some Implications from Socially Situated Learning Theory
and Actor-Network Theory. In Steeples, C. and Jones, C. (eds.) Networked
Learning; Issues and Perspectives. London: Springer.
Koschmann, T. (Ed.). (1996). CSCL: Theory and Practice of an
Emerging Paradigm. Mahwah, NJ.:
Lawrence Erlbaum Associates.
Koschmann, T. (2001). Revisiting the paradigms of
instructional technology. In G. Kennedy, M. Keppell, C. McNaught & T.
Petrovic (Eds.), Meeting at the Crossroads. Proceedings of the 18th
Annual Conference of the Australian Society for Computers in Learning in
Tertiary Education. (pp. 15 - 22). Melbourne: Biomedical Multimedia Unit, The
University of Melbourne. Available online at:
http://www.ascilite.org.au/conferences/melbourne01/pdf/papers/koschmannt.pdf
[Accessed 28th January 2004]
Lave, J., and Wenger, E. (1991). Situated Learning: Legitimate peripheral participation. Cambridge:
Cambridge University Press.
Law, J. (Draft) 'Networks, Relations, Cyborgs: on the
Social Study of Technology', published by the Centre for Science Studies and
the Department of Sociology, Lancaster University. Available online at:
http://www.comp.lancs.ac.uk/sociology/soc042jl.html
[Accessed 24th January 2004]
Osterlund, C., and Carlisle, P. (2003). How Practice
Matters: A Relational View of Knowledge Sharing. . In M. Huysman, Wenger, E.,
and Wulf, V. Communities and Technologies. Dordrecht: Kluwer pp1- 22
Rosson, M.B. (2003). Weak Ties in Networked Communities. In
M. Huysman, Wenger, E., and Wulf, V. Communities and Technologies.
Dordrecht: Kluwer pp265 - 286
Wenger, E. (1998). Communities
of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge
University Press.
Wittel, A. (2000). Ethnography on the move: From field to
net to Internet [23 paragraphs]. Forum Qualitative Sozialforschung/Forum:
Qualitative Research [On-line Journal], 1 (1). Available at: http://qualitative-research.net/fqs
[accessed 24th January 2004]