Kaethler Journal of Organization Design (2019) 8:10

https://doi.org/10.1186/s41469-019-0051-z

CASE STUDY Open Access

Curating creative communities of practice:

the role of ambiguity
Michael Kaethler

Correspondence: michael.kaethler@
kuleuven.be Abstract
KU Leuven, Kasteelpark Arenberg 1,
3001 Leuven, Belgium Communities of Practice (CoPs), it is argued, are loci for creativity, innovation and
problem-solving. Instigating a CoP and harnessing this creative energy from an
external position (be it institutional or individual) is, however, problematic. Literature
surrounding CoPs emphasises the delicate manner in which they are formed and
sustained. Those instigating these communities from an external position, such as
curators, managers, or educators, do so at the risk of undermining some of CoPs’
fundamental qualities. Namely: the fluid social relations, the level of informality and
the processes of self-selection and moderation that characterise CoPs. Asking the
question ‘how, if possible, can one instigate creatively-oriented CoPs?’—in particular
those composed of experts working within limited timeframes—this paper analyses
eleven newly formed groups partaking in an experimental design biennial (BIO50)
organised to foster collaborative learning and practice. The study focuses on the
relationship between a curatorial structure with high degrees of ambiguity and
participant collaboration. The paper provides practical implications and theoretical
elaborations for those seeking to organise creative collaborations.
Keywords: Curating, Ambiguity, Creativity, Collaboration, Community of Practice

Introduction
The Icelandic singer Bjork’s oft-quoted lyric: ‘I thought I could organise freedom, how
Scandinavian of me’1 summarises succinctly the tension between organisation and free-
dom that is inherent in organisational and managerial pursuits. In many of today’s
creative hubs, from silicone valley offices to faculty lounges at art schools, managers,
educators and curators are asking a similar question: ‘how can we organise creativity,
or more specifically, creative exchange and learning?’ Much of creativity remains in the
knowledge realm of the tacit (Polanyi 1966) and is bound up in practice and poorly ar-
ticulated in formal language; creative knowledge is also referred to as sticky (Szulanski
2002) or intangible knowledge. The field of organisation studies has for some time
made compelling arguments that such knowledge can be transferred in Communities
of Practice (CoPs). This view, based on a social learning theory, emphasises the import-
ance of collaboration, of learning collectively through participation in a shared activity,
and of the development of a common repertoire (Wenger 1999; Cox 2005). These
activities form the basis for a transfer of a broad spectrum of knowledge(s) and can act
as a locus for creative and innovative problem-solving (Mørk et al. 2008; Cook and
Yanow 1993; Lindkvist 2005). However, the few studies detailing attempts at instigating
© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International
License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium,
provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and
indicate if changes were made.
Kaethler Journal of Organization Design (2019) 8:10 Page 2 of 17

CoPs from an external position (such as that taken by an educator, manager or curator)
have shown it to be problematic due to the informal, self-regulated, non-canonical and
spontaneous nature of these communities. This brings us back to the tension between
organisation and freedom (Thompson 2005; Contu and Willmott 2003; Brown and
Duguid 1991; Wenger 1999; Breu and Hemingway 2002).
This tension appears to have been ignored in much of the practitioner literature on
forming CoPs. In the 28 years since Lave and Wenger (1991) introduced CoPs, there
has been an abundance of tips, tricks, guides and how-tos, on forming such communi-
ties. These, with few exceptions, are activity oriented and often follow a pattern of pro-
cedures that include developing clear objectives and purpose, creating a clear plan or
infrastructure, selecting strong leaders, facilitating dialogue and so forth. Recommenda-
tions are highly pragmatic and belie a notion that exchanging knowledge is straightfor-
ward and procedural. In other words, there is an underlying assumption that a CoP can
be structured with the right set of managerial actions. Notions of self-regulation, infor-
mality and spontaneity are largely absent or trivialised.
This paper seeks to nuance this procedural approach by focusing on how introducing
certain qualities (as opposed to activities) such as ambiguity into an organisational
approach can create the basis from which a CoP can emerge and thrive. This research
intends to provide useful practical material as well as a theoretical elaboration to those
taking an organisational role in creative contexts who are seeking innovative methods
to inspire collaborative learning and practice. To do so, the article empirically situates
the discussion of CoPs in the world of design, where collaborative learning plays a
significant part in the development of creative solutions. The case study focuses on the
experimental curatorial methods employed at the 50th anniversary of the Slovenian
Design Biennial.
Following the introduction, the paper is broken down into the consequent order.
Section 2 elaborates on CoPs and their relevance for creative collaborative production.
Section 3 highlights the debates surrounding the instigation of CoPs and analyses the
types of conditions that can underlie and support their emergence. Section 4 presents
the data source and methodology. Section 5 discusses the findings of the empirical
research, and Section 6 provides concluding reflections on these findings.

Communities of practice: creativity, knowledge and organisation

According to Wenger (1999), a CoP is a joint enterprise which is continually being
renegotiated by its members; it functions through binding relationships of mutual
engagement that coalesce into social entities; and it produces a shared repertoire among
its members that acts as a communal resource and is composed of elements such as
sensibilities, artefacts and/or vocabulary. Wenger understands practice to be an act of
negotiation of meaning (and with it, identity) with the interlinked processes of partici-
pation and reification. Participation suggests both action and connection; more specif-
ically, Wenger refers to participation as ‘a process of taking part and also to the
relations with others that reflect this process’ (1999: 63). As such, participation is char-
acterised by the possibility of mutual recognition or as Handley et al. (2006) stress, by a
sense of belonging with a progression towards full participation. Reification then is ‘the
process of giving form to our experience by producing objects that congeal this experi-
ence into “thingness”’ (Wenger 1999: 58). Viewed socially, it is the moment that
Kaethler Journal of Organization Design (2019) 8:10 Page 3 of 17

knowledge of a group is synthesised into a material and symbolic system (Gherardi and
Nicolini 2000, Moulaert and Van Dyck 2012).
CoPs are fundamentally self-organising and self-moderating systems (Wenger 1999),
which like most innovative endeavours have a tendency to reject canonised and forma-
lised received wisdom in order to pursue creativity (Brown and Duguid 1991; Litchfield
et al. 2015). Indeed, CoPs are often understood to be a locus of creativity and
innovation and offer explanations for the relationship between practice, learning and
innovation (Mørk et al. 2008; Cook and Yanow 1993; Lindkvist 2005).

Structuring or seeding communities of practice

Despite considerable literature foregrounding the difficulties or questionability of giving
intentional form, or ‘structuring’, a CoP (Thompson 2005; Contu and Willmott 2003;
Wenger 1999; Cox 2005; Breu and Hemingway 2002), popular literature on workshops
and collaboration are abuzz with proposed solutions. These, upon closer inspection,
neglect important social learning facets of CoPs such as informality, the non-canonical
or spontaneity. Additionally, many tend to take on characteristics that Wenger (1999)
originally cautioned against, such as hybrid working groups (defined by a task and not
by knowledge), functional units (defined by a charter and not by action) or networks
(defined by relationships as opposed to the act of doing).
‘How-to’ guides for structuring a CoP tend to oscillate between prescriptive
actions and generic recommendations. In 2002, Wenger, McDermott and Snyder
published a practitioner’s guide for ‘Cultivating Communities of Practice’. The book
rode the wave of interest in CoPs but was controversial and heavily criticised as
commodifying, oversimplifying and fundamentally redefining the concept of CoPs
(Cox 2005: 533). It was a departure from Wenger’s earlier work: attempting to
appeal to the broadest possible audience and to provide clear guidelines, bypassing
the inherent tensions that come with structuring. Perhaps it is for this very reason
that the guide found a wide appeal among practitioners and has been influential in
shaping organisational practices. The surge of interest in CoPs has led to its appli-
cation in all manner of contexts and with a range of formulaic approaches leading
to a deterioration of its original qualities that emphasised situatedness, social inter-
action, materiality and so forth (Amin and Roberts 2008).
A study conducted by the author of 20 websites that offer advice or instructions on
creating, cultivating, forming or building CoPs shows that the field is dominated by a
highly procedural understanding of how a CoP can be formed; the advice closely mir-
rors the Wenger et al. (2002) guide. The study found that little to no consideration is
given to the difficulties and tensions of instigating a CoP. Recommendations emphasise
the need for ensuring a clear objective and definite purpose, a strategic plan and appro-
priate infrastructure for its execution, the selection of strong leaders, and the facilita-
tion of dialogue and collaboration. Knowledge is perceived as accessible and easily
transferable, diminishing the dimension of practice in CoPs and emphasising commu-
nity as a group connected through a verbal exchange of thoughts, ideas and feelings.
None of the 20 sites reviewed warned of the difficulties or complications in forming
such a community, of the need for creative freedom, spontaneity, non-canonical
problem-solving and informality outside of managerial reach.
Kaethler Journal of Organization Design (2019) 8:10 Page 4 of 17

These findings contradict Wenger’s (1999) earlier work, which emphasises that one
cannot reproduce the structural components of a CoP, such as the types of actors, loca-
tions, objectives or actions and expect a similar set of social relations and coalescence
to emerge. In other words, it is one thing to provide support for the structural compo-
nents; it is another to facilitate the types of exchanges within the community—such as
identification with the goals, a sense of ownership and wanting to contribute to
bottom-up interactions.
Wenger (1999: 229) sums this up:

Communities of practice are about content—about learning as a living experience of

negotiating meaning—not about form. In this sense, they cannot be legislated into
existence or defined by decree. They can be recognized, supported, encouraged, and
nurtured, but they are not reified, designable units. Practice itself is not amenable to
design. In other words, one can articulate patterns or define procedures, but neither
the patterns nor the procedures produce the practice as it unfolds… Learning cannot
be designed: it can only be designed for—that is, facilitated or frustrated.

Brown and Duguid’s (1991) influential study on Xero repairmen, a key contribution
to the discourse on CoPs, emphasises its bottom-up nature. It tells a story that stresses
the informality at the heart of CoPs, one that addresses problems that are deemed rele-
vant but does so in a counter-cultural or non-formalised way. An underlying factor is
the argument that canonical accounts of work are inevitably flawed, inflexible and lim-
ited and as such require local and situated understanding and action for the completion
of the task (Cox 2005). Brown and Duguid are essentially arguing for collectively im-
provised knowledge, emphasising ‘shop floor’ innovation through informal exchanges.
This provides little in the way of practical guidance or quick tips for CoP formation
other than elusive qualities such as spontaneity.
With a focus on informality, self-regulation and self-management, structural com-
ponents of a CoP tend to be mediatory—taking the form of what Wenger (1999)
includes as artefacts, documents, terms and concepts. Thompson’s (2005) empirical
findings echo Wenger’s (1999) theories on the use of boundary objects as structur-
ing mechanisms including instruments, monuments and points of focus. As
Thompson points out, if these boundary objects are significant as mediators for
CoP growth and development, then working with boundary objects can possibly
yield results in influencing the instigation and development of CoPs. However,
based on his research on a web-design agency, Thompson (2005: 162) emphasises
that such objects must be used in a ‘“non-prescriptive way” in the hope of indir-
ectly seeding future collaboration’. He distinguishes between structuring and seed-
ing, controlling the present verses influencing future collaborations. His research
exhibits the fragility of a CoP, as a delicately formed network of relationships,
which could come crashing down given a change in conditions.
Thompson (2005: 162) argues that an emphasis on ‘culturally symbolic infrastructure
can set the emergent properties for subsequent interaction’. This however, is premised
on an infrastructure that supports interaction and strong personal identification. Like-
wise, imposing structural constraints is to ignore the fundamentally social and informal
nature of its emergent social dynamics and will fail to materialise a CoP. As such CoPs,
Kaethler Journal of Organization Design (2019) 8:10 Page 5 of 17

he argues, can be indirectly given form through a focus on symbolic monuments, infra-
structural instruments and conceptual points of focus.

Knowledge types, organisational dynamics and conditions for emergence

Amin and Roberts’ (2008) important survey of the existing literature on CoPs high-
lights the distinction between four different types of knowledge in action—craft/task,
professional, virtual and epistemic/creative—and how these affect social interaction,
innovation and organisational dynamics. Epistemic/creative knowledge CoPs, which
this paper focuses on, emphasises short-term collaboration involving participants who
come together with the purpose of experimenting with new knowledge in order to un-
leash creative energy. Members of these types of communities tend to be autonomous
and specialised and have substantial egos and high expectations; these environments
are characterised by high turnovers, rudimentary rules and tight deadlines (Amin and
Roberts 2008; Lindkvist 2005). This typology commonly takes place in creative organi-
sations, academies or across a network of specialists.
Amin and Roberts’ typology of epistemic/creative knowing2 highlight three common
underlying qualities (degrees of uncertainty, variety, and ambiguity) and four common
factors that channel uncertainty, variety and ambiguity towards creative openings (peer
recognition, problem loyalty, slack space and alignment mechanisms). These are identi-
fied as critical elements found across a range of practices involved in this epistemic/cre-
ative type of knowledge community and open up another perspective on seeding
structures. Instead of focusing on boundary objects, it specifies the prospective qual-
ities—uncertainty, variety or ambiguity—around which such objects emerge. These
have the potential to produce conditions where individuals converge and consolidate
around certain factors such as peer recognition or problem orientation. Amin and
Roberts do little more than identify these factors, leaving it to others to expand upon
their observations. For the sake of brevity, this paper focuses on ambiguity, which
involves elements of uncertainty and to a lesser degree variety, in its role as a force for
establishing creative communities of practice.

Ambiguity
For many in the organisational and management fields, ambiguity is perceived as an
‘ugly’ quality (Alvesson and Sveningsson 2003), a barrier for pursuing collective action
(Jarzabkowski et al. 2010), and something to eschew especially in the context of
time-limited and market-oriented transactions. Ambiguity, according to the Stanford
Encyclopaedia of Philosophy (Spring 2016 Edition), is composed of two meanings: (1)
uncertainty or dubiousness and (2) a sign bearing multiple meanings. If your goal is
clarity, then ambiguity is counter-productive. However, other fields have applied ambi-
guity as a tool or method for achieving a variety of objectives. Ambiguity enables the
co-existence of multiple meanings, which may prove plausible and are not resolvable
through clear definable rules. The room for interpretation afforded by the presence of
ambiguity can enhance or deepen the act of personal engagement (such as in the arts:
poetry, literature, visual arts); it can also enable differing parties to come together as
described in Eisenberg’s ‘strategic ambiguity’ which is now used across a range of fields
such as political science, corporate communication, and urban activism (Kaethler et al.
Kaethler Journal of Organization Design (2019) 8:10 Page 6 of 17

2017). In order to explore ambiguity as a condition for seeding, the paper empirically in-
vestigates its influence on the four factors of peer recognition, problem loyalty, slack space
and alignment, as emphasised by Amin and Roberts as channels for creative openings.

Case study: Slovenian Design Biennial (BIO50)

The motto for the 50th anniversary of the Slovenian Design Biennial (BIO50) was
‘3,2,1…test’. The event lived up to its motto: it was a design expo aimed squarely at
experimentation. BIO50 was organised around 11 loose thematic categories and partici-
pants came from a variety of backgrounds, levels of expertise and cultures; hierarchical
structure was undefined and required outputs were vague and unclear.
The curatorial structure focused on collaboration without providing an explicit pre-
scription of how collaboration is made manifest. One of the two curators, Jan Boelen,
states, ‘If design is the answer, what is the question? The Biennial is a question. They
[the participants] have the tools and we are facilitating them with mentors and struc-
tures so that they can develop their own stories.’ Boelen views the collaboration of
experts through a learning prism. Specialists are brought together to merge fields of
knowledge through periodic exchanges involving collective as well as autonomous
reflection—in a pendulum of individual and collective learning. To do this, he argues,
‘You must adopt a position of ignorance, recognising not what you know but what you
don’t know’. In this regard, the structure of BIO50, he claims, is a ‘learning structure
for the participants and for me, as a curator’.
The curators provided a structure of openness. The 11 rather ambiguously defined
thematic categories (including themes such as Designing Life, Knowing Food, Walking
the City, Observing Space, and Nanotourism) were given autonomy and freedom with
little intervention from the curators. Boelen’s curatorial partner, Maja Vardjan, referred
to their level of involvement as ‘commentators in the process, not intervening in the
content of the work’. These groups were obliged to present something at the end of 6
months for the opening of BIO50. How they were structured or the internal workings
were a matter for the individual group and the mentor(s) to decide. One hundred
seventy-five participants initially joined with 21 mentors. Over the 6 months, 52 partici-
pants dropped out and 18 additional participants joined.
According to the curators, this model was deemed a success. It was noted that 1 year
after BIO50, a number of the projects from across the 11 themes continued to be devel-
oped. Some of the BIO50 designs have since circulated events and exhibitions around the
world, while former partnerships and collaborations have in the meantime birthed new
projects and collaborations. The model has been re-applied for exhibitions in the Museum
of Architecture and Design (MAO), Ljubljana, and the Istanbul Design Biennial.

Research methodology
The question, ‘how does one curate creative collaboration?’ was the starting point for
this research. After discussions with the curator of BIO50, which led to a better grasp
of the organisational structure of the biennial, the author spent several days immersed
in the exhibitions and in an informal conversation with participants. Researching
collaboration is not without difficulties; it involves considerable assumptions on what is
considered a ‘successful’ collaboration and what are the corresponding determining
Kaethler Journal of Organization Design (2019) 8:10 Page 7 of 17

factors, most of which remain unobservable and intangible. To capture this, the
researcher used in-depth interviews as the central methodological approach to examine
the social, epistemic and structural components at BIO50.
The author selected a loose sample of respondents from 7 of the 11 themes with a
minimum of 2 participants or mentor; for the other 4 groups, the author used a more
intense selection process, interviewing 70–100% of the participants. The 4 groups were
chosen according to two criteria—size and perceived level of successful collaboration—
in order to compare and contrast findings. The four thematic groups (large/small, per-
ceived success/failure) included Affordable Living and Nanotourism (large groups) and
Designing Life and Engine Blocks (small groups).
In total, 51 interviews were conducted with the 42 participants, 7 mentors and the
two curators. These interviews ranged from 35 min to 2.5 h in length, with an average
lasting approximately 1 h. Interviews were initially semi-structured; questions ranged
from very practical to experiential. As an idea of each group’s collaborative experience
grew clearer, questions became more exploratory and probing. Questions were used to
prompt the respondent to tell the story of their experience within the group from day
one to the present (shortly after the final exhibition) with as little interjecting as pos-
sible from the interviewer.
As this paper focuses on the social and situated nature of knowledge, it naturally
follows that this be applied to its research methodology. Interviews can elucidate how
individuals perceive the world and how they choose to communicate it (Silverman
1985). The author recognises the activeness of interviews and that since meaning is
socially constituted, interviews provide the site and occasion for producing knowledge
(Silverman 1985). Both parties, the researcher and respondent, are thus responsible for
creating this knowledge as active parties in a two-way relationship. As Gubrium and
Holstein (1997:106) argue,

Meaning is not merely elicited by apt questioning, nor simply transported through
respondent replies; it is actively and communicatively assembled in the interview
encounter. Respondents are not so much repositories of knowledge—treasuries of
information awaiting excavation—as they are constructors of knowledge in
association with interviewers.

As the respondents are co-creators of this knowledge, the findings and discussions in
this article were fed back to the respondents for further comment, discussion or
suggestions. This open loop enabled a greater reflexivity on their individual process
within the wider set of experiences documented in this article.
Interviews were placed within the context of each group’s design production and
process. Not only was the experience discussed, it was situated within the multiple
other data forms such as the final exhibition, the tools for group communication (Face-
book group chats, Google hangout, etc.) and boundary objects such as concept notes,
sketches and models. These helped ground the discussions.
The interviews aimed at identifying how the group functioned as a whole, breaking
down the social and structural elements and describing the process of collaboration
within the group such as knowledge transfers or group alignment (the latter being
more difficult to articulate). The respondents led with their narrative of the experience
Kaethler Journal of Organization Design (2019) 8:10 Page 8 of 17

and the author intervened from time to time to enquire about structural elements;
these were identified using follow-up questions such as ‘how was it done, why was it
done that way, how was it organised?’ At the end of the interview, the author would en-
courage discussions on the nature of collaboration and design. The author was inspired
by Douglas’ (1985) creative interviewing, which seeks to uncover more than just opin-
ion through an openness to mutual disclosure by the interviewer expressing some of
his/her feelings on the subject. This approach helped place respondents’ narratives of
BIO50 within the broader perspective of design collaboration.
Notes and voice recordings were taken during the interviews. These were compiled
and sifted through; the relationship between features such as social interactions, bound-
ary objects, organisational structure, roles and rules was analysed in order to identify
structural components and epistemic dynamics. These were compared and contrasted
with broader experiences of collaboration, group dynamics and the overall outlook on
the experience and quality of cultural production. For the sake of anonymity, the
discussion on findings will only make reference to prevalent themes.

Discussion on findings
This section explicates the relationship between ambiguity with the channelling
factors of problem loyalty, peer recognition, slack space and alignment as eluci-
dated by Amin and Roberts (2008). Based on these findings, I discuss the merits
and constraints of uncertainty, variety and ambiguity in relation to these factors
for seeding Creative CoPs (CcoPs).

Problem loyalty
The development of loyalties to a shared problem or goal orients actions and attitudes
and provides what Lindkvist (2005) describes as a relevance structure based on
self-organised discovery. Problem orientation provides a platform for shared identity
and in some cases can lay the foundation for deepening social ties while still allowing
for a high turnover of participants. Wenger (1999) argues that the act of participating
in the common problem forges strong social bonds. Brown and Duguid’s (1991)
research provides an account where problem-solving through situated approaches and
reformulating canonical methods is the pinnacle for creative collaboration.
At BIO50, forging loyalty to a common problem through ambiguity was evidenced in
several groups’ attempts at defining a shared theme. The group Nanotourism spent
much of their collaboration on defining what this term meant in theory and practice.
This was done through shared exploration and co-definition. The group was broken
down into 5 subgroups, each working separately in 5 different cities on projects related
to the evolving concept of Nanotourism. The sub-groups explored what it could mean
based on initial plenary discussions and consistently reverted back to the main group
with reflections on their experience. ‘Homework’ was regularly assigned by the mentors
requiring all participants to submit sketches, ideas, and documentation of progress,
which would feed into the larger discussion on Nanotourism. In this way, loyalty to a
common problem was the starting point for the group; its very identity was in a state
of being defined. The abstract and elusive term Nanotourism acted as an anchor point
Kaethler Journal of Organization Design (2019) 8:10 Page 9 of 17

and boundary object around which the group pivoted and exchanged their ideas, while
simultaneously forging a common identity.
The act of co-defining became a central point of focus and gave meaning to participa-
tion. The process of co-exploration through word and deed replaced what one partici-
pant called ‘the tyranny to produce’, stating that ‘the collaborative act of defining made
the entire BIO50 thing worthwhile’. This process-driven attitude, which had been
fostered by the curators, trickled down into some of the mentors’ approaches. For
Nanotourism, one of the mentors stressed that ‘it was not the theme that defined the
process but the process that defined the theme; getting the process right was crucial to
the project.’ Discussions across the sub-groups of Nanotourism turned up the same
thing over and over again: the importance of co-defining a concept, especially one
which had no pre-existing meaning—conferring ownership and building a shared iden-
tity. In other words, positioning the ‘problem’ in ambiguity forged social and cognitive
ties. After the fact, many of the participants in that group continued to pursue Nano-
tourism as a concept in their practices.
The experience of co-defining was shared by the smaller group Designing Life, where
the mentor structure offered a horizontal organisational dynamic, one with a high
threshold of diversity in backgrounds and nationalities. For the first while, the group
struggled to congeal around a shared vision and to understand each other’s ideas across
disciplinary boundaries. Coming to terms with the thematic category of Designing Life
forced the members into somewhat uncomfortable synergies, which would not have
happened otherwise. The diversity of opinions and ideas within the group was particu-
larly evident when they were tasked to narrow down the question they wanted to
respond to, showcasing the realisation of a need to find a shared purpose. As one mem-
ber stated, ‘It was very challenging to solidify all the opinions [of the group]... into a
question. It was like our differences were tied to a single topic.’
Both Nanotourism and Designing Life are examples where negotiating the ambiguity
of their project conferred a sense of solidarity, ownership, and collaboration. These
groups were able to overcome significant cultural and disciplinary boundaries to com-
municate ideas based on different ways of knowing—including scientists, anthropolo-
gists and designers. The ambiguity forged the basis for a collective learning experience
in which knowledge bases were linked together through the act of a shared learning
experience. As such, they provided the conditions for shared learning and reification.
Conversely, the thematic group of Affordable Living started the process off with a
pre-determined idea of the project, programmed by the mentors and without the con-
sent of the participants. Before long, frustration and disinterest began to set in. Some
participants spoke-up and the group was subsequently re-structured, opening up new
avenues to address the original theme. Despite reverting back to a more ambiguous
position, a number of participants voiced that it came too late; the sense of community
had failed to materialise.

Peer recognition and trust

Community is formed not only by mutual recognition of those on the ‘inside’ of a
community (Wenger 1999) but also through a recognition of the qualities that
those individuals possess as being of value to the community. In CCoPs, an expert
Kaethler Journal of Organization Design (2019) 8:10 Page 10 of 17

working autonomously does not have time to establish professional trust through
time-bound means in collaborations and relies instead on peer recognition (Amin
and Roberts 2008).
Ambiguity, in the case of BIO50, was fomented through the loose framing of groups’
objectives, the desired processes and even the theme in which they were formed—
vague and under-defined. As explained in the ‘Problem loyalty’ section, each group
sought to develop a collaborative relationship whilst interpreting and developing their
own perspective of the biennial. The variety of participants in each group with unique
cultural and professional perspectives rendered the intentionally ambiguous exhibition
as a powerful force for two opposing experiences—group solidification and atomisation.
Ambiguity, without structure or guidance, was responsible for derailing peer recogni-
tion and trust within heterogeneous groups.
Initial acceptance of the multiplicity of specialities present in groups was appreciated
and praised by participants as ‘exciting’ or ‘challenging’. A positive and excited spirit
existed around transdisciplinarity and multiculturalism but it grew increasingly difficult
to integrate different approaches and methods without clear objectives, approaches and
modes of operation in the face of looming deadlines. Eventually, frustration and frac-
tures began to emerge and participants looked to the mentors and curators for clarity
in order to tame the systemic ambiguity of the biennial.
Without mentors safeguarding inclusion and promoting peer recognition, ambiguity
resulted in hierarchies within groups. As this occurred, certain disciplinary logics or
cultural groups came to disregard or instrumentalise others. One example is when a
biologist was responsible for contributing a particular piece of specialised research but
another member of the group from a design background did not trust the quality of
that research (despite the qualifications of the biologist) and without explanation simul-
taneously attempted to duplicate the research. This undermined both individual and
group trust. On a larger scale, this was apparent in another group, which saw a split
between local Slovenians and international participants. Unclear expectations from the
start as to the ‘purpose and focus of the theme’ resulted in a significant schism. This
group, despite later attempts to reign in misunderstandings, lost over one third of its
participants. New members already known to some in the group and sharing the same
approach and background were invited to join in order to shore up efforts for the final
exhibition—in effect replacing diversity with homogeneity. In the end, there was a lack
of trust among members, leading many respondents to claim that there was ‘no collab-
oration’ whatsoever.
In a context of distinct cultural and professional differences ambiguity provided nov-
elty and potentiality but under the pressure to perform peer recognition and trust were
quickly replaced by hierarchy and exclusion. Without proper channelling such as estab-
lishing clarity of purpose and general direction, ambiguity, in a context of heterogeneity
shows itself to be a factor that is potentially socially destructive, leading to collaborative
breakdown, individualistic approaches, hierarchies and schisms.

Slack space
Slack space is unsanctioned space, which as Thompson (2005) states allows a group to
‘consciously cultivate informality’. Amin and Roberts (2008, 362) note that ‘informality,
Kaethler Journal of Organization Design (2019) 8:10 Page 11 of 17

iterative purposefulness, and productive idleness’ are commonly found in groups from
scientists to artists. This involves elements of freethinking, imaginative play, ‘serendip-
ity’ and so forth.
Collective slack space thrives within conditions of ambiguity, whereby the drive for
efficiency is moderated by the need for creativity and innovation. Being able to move
between the professional and the social and to bond and interact in a different manner
was perceived by group members as a unifying force and at times a pinnacle moment
for pooling creativity. Ambiguity here was experienced as the freedom to determine the
‘house rules’ of a group, how they could organise their own processes and practices.
The extent to which slack space was formally directed or emerged spontaneously
differed across groups with various attributing factors including peer recognition
and problem loyalty. As such it is difficult to study slack space as it is both exogen-
ous and endogenous to the group. The social interactions that occurred in the inner
workings of the group were factors for developing the desire to spend unsanctioned
time together.
One group’s slack space consisted of simply having drinks in the evening after the
day’s meeting. ‘Beer after meetings changed how we worked [formally]’, reflected a par-
ticipant. Another group, which already had a history together, attributed their creative
process to a period of sitting in a living room smoking cigarettes. Such events were not
sanctioned from the organisers and there was no intentionality in the structure for
these; instead, they emerged naturally out of a desire for time together outside of the
working rhythm.
The autonomy with which groups worked enabled them to direct their modus oper-
andi. Some groups emphasised slack space as part of their creative process while others
kept it to a minimum, treating the process as a professional project constructed around
professional relationships. Informal bonding was cumbersome to realise in large groups
where there was a sense of conviviality without the time to develop a sense of unison
within the group. For example, the largest of the groups, Affordable Living, reported
nearly no slack space and little to no social bonding outside of work routines resulting
in only a superficial knowledge of each other. This is the same group that reported a
negative sense of inter-participant hierarchy and very limited collaboration.
Slack space was particularly important for groups who were interacting extensively
through virtual means. However, despite attempts to create informal ‘hangouts’ online,
such as the Knowing Food group, preparing a dish based on the same key ingredient
and then eating together networked via webcams and Google hang-out, these were
unanimously perceived as lacking in social character. As it was pointed out, ‘one day
together [in person] is like three weeks virtually [connected]’. Another participant spe-
cifically referring to virtual meetings emphasised, ‘it’s not nice sitting for two hours in a
meeting, it’s a lot worse sitting for two hours in a Skype meeting’. Discussions with
respondents on the topic of virtual communication were overwhelmingly negative; only
one group had a positive experience, which they explained was due to some members
having a pre-existing virtual relationship of collaboration together. Many felt that it
was difficult to have informal or casual ‘banter’, feeling the pressure to get straight to
the point and not linger unnecessarily. One respondent, speaking of a group trip to
London and referring back to virtual communication, alleged, ‘before London it was like
talking into a dark room.’
Kaethler Journal of Organization Design (2019) 8:10 Page 12 of 17

Alignment
Alignment is intended to provide cohesion and mutuality through managing disson-
ance or creating coordination (Regeer and Bunders 2003; Wenger 1999). As covered by
the work of Wenger (1999), Thompson (2005), and Nonaka and Takeuchi (1995),
boundary objects, entities that hold and link the interest of diverse groups, provide
powerful sources of knowledge alignment (Nerland and Jensen 2012; Lindkvist 2005).
These are important for sense-making, for distributing ‘know-how’ (as opposed to
know—that), and for facilitating communication on complex issues (Star 2010; Bucciar-
elli 2002). Similarly, Wenger’s emphasis on the dual importance of participation and
reification are pivotal in aligning actors within a community. Experiences at BIO50
confirm this. Regular assignments by mentors to provide sketches, concept notes,
models or blueprints were extremely powerful tools for aligning differences in the
group. As one participant exclaimed, ‘the way you work with people [in the group]
becomes embedded in the artefact.’ Theoreticians like Karin Knorr Cetina (2007) would
add that the artefact becomes embedded in the way you work with those people to
create a culture around a practice. Indeed, boundary objects are recognised as a crucial
part of the design (Carlile 2002; Bucciarelli 2002).
As there already exists considerable literature on boundary objects for the purposes
of alignment, including Thompson (2005) and Wenger (1999), the rest of this section
focuses on alignment through mediatory roles such as the role of convenor, moderator
or broker in conditions of ambiguity. While not a precondition, a mediatory role has
an important part in fostering collaboration (Wood and Gray 1991). Wenger (1999:
109) acknowledges the complex and important job of the broker as involving processes
of ‘translation, coordination, and alignment between perspectives’ and goes on to say ‘it
requires the ability to link practices by facilitating transactions between them, and to
cause learning by introducing into a practice elements of another’. Breu and Heming-
way (2002), however, caution that beyond informal inter-group alignment, CoPs are
predominately self-organising and self-determining. In the case of BIO50, convenors
are evident on several levels—curator, mentor, and in some cases participants. Each of
these provides different traits; the curator as legitimising, the mentor as guiding, and
the participant, in cases where he/she took a role within their group or sub-group, as
broker. These roles can be fluid and reflect the nature of the individual’s influence in
the broader organisation. Amin and Roberts are relatively silent on this, mentioning
‘meta-coding’ and ‘visionaries’.
The infrastructure of BIO50 provided the institutional space for a social organ-
isation to autonomously develop under a common theme and purpose. Mentors
were given little to no instruction or mandate as to how involved they could or
should be in the overall work. It was a laboratory setting with broad parameters
in which social organisations emerged through processes of shared learning and
exchange. With such high degrees of ambiguity there also exists an increased
likelihood of dissonance, chaos and confusion. Mentors took on different roles:
some applied strict sets of instructions and deadlines while others took on the
role of observers, accepting disorder as part of the creative and collaborative
process. In at least three of the groups, particularly in situations of communica-
tion breakdowns, a participant took a mediatory role, acting as a broker between
participants and mentors.
Kaethler Journal of Organization Design (2019) 8:10 Page 13 of 17

The curators provided the space for the mentors and participants to find their way
and the mentors either structured this space or let it be structured by the emergent
group dynamics. Large groups and groups with a high variety of skills and interests re-
lied more heavily on mentor-led directions for alignment and sought out this mediatory
role in order to bring together different interests. Small groups tended to develop more
horizontally with mentors acting as participants, such as Engine Blocks. It is unclear,
however, what type of interaction-effect occurs between ambiguity and role specifica-
tion for improved commitment. The importance of setting parameters and expectations
helped offset some of the negative effects of ambiguity such as frustration and group
atomisation. The larger the group, the more these parameters needed to be re-enforced.
Mentors played the role of moderating the degree of ambiguity within the groups by
providing clarity, setting certain markers or goals, affirming processes and encouraging
novelties. For example, the difficulties in establishing peer recognition and trust, as dis-
cussed in 5.2, in circumstances with high degrees of ambiguity were considerably offset
with even minimal mentor interventions including slight clarification of goals, provision
of direction and/or re-assurance that the process was moving forwards in the intended
direction. Likewise, mentors were able to confer authority on productive markers such
as keeping to timelines, instituting assignments and fostering a culture of collective
inclusion. This held particularly true in groups with more than 10 participants.
Positions of authority reflected certain values and inspiration to the participants. The
curators represented an ethos and culture for the Biennial. When asking participants
about the extent of their interaction with the curators 80% of participants declared that
they had very little or none at all. However, when asked about the role of the curators,
it was felt that they reflected a sense of order and meaning amidst the chaos occurring
in many of the collaborative situations. One participant spoke of the feeling of ‘security’
knowing that there was a ‘bigger plan’, embodied by the curators and also by the institu-
tional support of the hosting Museum of Architecture and Design (MAO). In this
sense, curators employed a dual strategy with one hand maintaining a structure of am-
biguity with its associated discomfort and confusion and on the other hand transmit-
ting a sense of confidence and reassurance.

The implications of ambiguity in designing Creative Communities of Practice (CCoPs)

BIO50 was designed as a testing ground, a laboratory for collaborative design. Failure was
always on the table and was considered an acceptable risk in the larger picture. Allowing
ambiguity to be a guiding condition for collaboration entailed embracing the risk but
trusting that negotiating this would in the end bring about effectual collaboration and
commitment through shared learning and practice. Ambiguity produced a substantial ef-
fect on group formation, exchange and co-production. This accelerated group dynamics
leading to close collaboration in some cases and fragmentation in others.
The findings provide the groundwork for a framework for collaboration where ambi-
guity is supported through association with group members and with the design prob-
lem (‘Problem loyalty’ and ‘Peer recognition and trust’), informal exchanges away from
the work environment (‘Slack space’) and role clarity and alignment within the groups
(‘Alignment’). This framework contributes to a collaboration that facilitates the
exchange of creative knowledge(Fig. 1).
Kaethler Journal of Organization Design (2019) 8:10 Page 14 of 17

Fig. 1 Interacting factors for establishing CCoPs. Ambiguity when filtered through the interacting factors of
social and problem-oriented associations, informal exchanges and role clarity and alignment can foster the
development of creative communities of practice

Notwithstanding that the factors of peer recognition, problem loyalty, slack space and
alignment are under separate headings they are deeply intertwined and connected. For
example, slack space sustains peer recognition, while peer recognition and problem
loyalty feed into the desire for slack space. Likewise, alignment is crucial for moderating
the negative effects of ambiguity across peer recognition and problem loyalty. These
factors stress the social and human-interactive variables that are so crucial in establish-
ing transfers of knowledge, in particular in domains where knowledge is difficult to
articulate or share such as in the arts or design. As articulated by Gherardi et al.
(1998), CoPs emphasise that every practice is dependent on social processes through
which they are sustained and perpetuated and that learning takes place through the
engagement in that practice. Structuring design collaboration to facilitate this evidently
requires certain conditions for social coalescence. Ambiguity provides circumstances
that galvanise group dynamics, challenging groups to move beyond exchanging infor-
mation as a matter of business as usual to developing close ties based around shared
learning through practice. These overlapping and interrelated factors offer insight into
the somewhat precarious conditions for sustaining a CCoP, whereby ambiguity provides
a certain amount of informal and non-canonical oriented exchanges while simultan-
eously requiring an alignment for direction and support.
The act of collectively negotiating ambiguous elements such as task, method,
approach, objective and even basic definitions related to a group’s theme had strong
ramifications: it determined how well individuals could collaborate and share types of
knowledge that resist easy transfers. It provided the conditions for informal social
organisation and collective learning through disentangling, interpreting and ultimately
allowing for disambiguation. Depending on factors such as support structures and
alignment, for some groups ambiguity was a powerful and stimulating force for collab-
oration, in others a source of frustration and social irritation which resulted in large
numbers of dropouts and isolated work culture.

Discussion and conclusions

By investigating the relationship between ambiguity and the four factors of peer recog-
nition, problem loyalty, slack space and alignment, it is possible to draw out a number
of findings on the nature of the emergence of CoPs. First, the empirical data corrobo-
rates Amin and Roberts’ (2008) description of the importance of ambiguity as a condi-
tion for epistemic/creative knowledge communities. Secondly, the findings help
substantiate Thompson’s (2005) conclusions on the importance of seeding as opposed
to structuring CoPs. This directly challenges the procedural approach to forming a CoP
being espoused across managerial-oriented websites. Third, and importantly, the
Kaethler Journal of Organization Design (2019) 8:10 Page 15 of 17

findings provide the basis to develop tools or principles for action for those involved in
collaborations that are keen to seek out creativity and innovation through collaborative
learning and action.
Ambiguity proves to be a powerful but unwieldy tool for seeding. It was often the
underlying quality that congealed the group by activating factors of peer recognition,
problem loyalty, and slack space. For example, ambiguity was instrumental in building
a sense of ownership, inspiring exploration, and negotiating differences. Likewise, ambi-
guity aided the development of peer recognition, social ties and a sense of otherness.
However, it was not always such a rosy story; the process of carefully executed align-
ment was necessary to balance out the excesses of ambiguity when it became destruc-
tive to group dynamics.
Reflecting on Thompson’s argument for an indirect seeding of CoPs, instruments,
monuments and points of focus appear insufficient without identifying the conditions
in which these are placed. Boundary objects, for example, have the power to consoli-
date knowledge. But in contexts with high degrees of ambiguity, such as with BIO50,
these objects act as strong conveyors of both knowledge and shared identity. Likewise,
Thompson’s ‘symbolic monuments, infrastructural instruments and conceptual points
of focus’ are identifiable in all the groups studied at BIO50 but are only activated in
those which sustained a degree of ambiguity throughout the 6 months. Where it was
present, it stimulated strong social responses, either towards collaboration, such as
co-defining themes or quite the opposite with a me-first attitude.
The importance of a mediator or broker to moderate the volatility of group dynamics
proved to be an important and delicate part of the equation. Echoing Thompson and
Wenger, there is a fine line between seeding and controlling: the former facilitates and
the latter frustrates. Examples from BIO50 demonstrate that without the presence of
some type of broker, such as a mentor or lead participant, ambiguity can overwhelm
the collaborative process and fracture social learning and practice. However, compar-
able to Thompson’s assertions on the delicacy of managerial over-step in promoting
collaboration towards a desired end, in the BIO50 case, delicacy was needed to find a
balance for regulating the degree of ambiguity by providing subtle parameters and en-
suring the existence of a wider support network.
Taking the experiences of BIO50 participants, mentors and curators, we can as-
semble a compelling story of experimentation and a willingness to take risks and
accept failure. It provides us with 11 cases where different degrees of collaboration
took shape over a short amount of time without significant external intervention.
Just as there were many successes, as is evident in factors like peer recognition and
problem loyalty, so too were there intermittent failures. From both, we can begin to
draw conclusions for conducting future endeavours, looking to the discoveries from
BIO50 as starting points for new explorations in creating short-term intensive col-
laborative learning communities. Among these is an awareness of how meaningful
collaborations, as with CoPs, can be strategically influenced by the conditions in
which they occur and that these conditions can, with delicate hands, be curated. For
organisational domains involved in related practices, it raises the question of how to
apply ambiguity in different environments, such as the classroom, studio, creative
organisations, and of how to manage the risk between its potential to consolidate or
atomise collaboration.
Kaethler Journal of Organization Design (2019) 8:10 Page 16 of 17

Endnotes
1
Hunter, from the album Homogenic, 1997.
2
To be distinguished from ‘epistemic communities’, which denote groups that work
together to master theories, codes and tools of a common practice relying primarily on
professional (not personal) ties and a non-experiential set of learning practices with no
need for geographical proximity. It also differs from Lindkvist’s (2005) ‘collectivities of
practice’, which focuses on accessing distributed knowledge and not on shared learning.
Abbreviations
BIO50: Slovenian Design Biennial (50th anniversary); CCoP: Creative Community of Practice; CoP: Community of
Practice

Acknowledgements
The author would like to thank the numerous BIO50 participants and organisers for their time, observations and
willingness to contribute to this research; especially to Jan Boelen for providing access and continued insight into the
curatorial process. He would also like to thank the invaluable input from the anonymous reviewers and in particular
from the editor Mikolaj Jan Piskorski. Lastly, much gratitude goes out to Frank Moulaert and Hilde Heynen who, from
the very start, supported this curious piece of research.

Funding
The author receives funding from the European Union’s Seventh Framework Programme for research, technological
development and demonstration under grant agreement No 608299.

Availability of data and materials

Interview data will not be shared for reasons of privacy.

Authors’ contributions
The author read and approved the final manuscript.

Authors’ information
N/A

Competing interests
The author declares that he has no competing interests.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Received: 22 February 2019 Accepted: 6 May 2019

References
Alvesson M, Sveningsson S (2003) Good visions, bad micro-management and ugly ambiguity: contradictions of (non-)
leadership in a knowledge-intensive organization. Organ Stud 24(6):961–988
Amin A, Roberts J (2008) Knowing in action: beyond communities of practice. Res Policy 37:353–369. https://doi.org/10.1016/
j.respol.2007.11.003
Breu K, Hemingway C (2002) Collaborative processes and knowledge creation in communities-of-practice. Creat Innov Manag
11(3):147–153. https://doi.org/10.1111/1467-8691.00247
Brown JS, Duguid P (1991) Organizational learning and communities-of-practice: toward a unified view of working, learning,
and innovation. Organ Sci 2(1):40–57. https://doi.org/10.1287/orsc.2.1.40
Bucciarelli LL (2002) Between thought and object in engineering design. Des Stud 23:219–231. https://doi.org/10.1016/s0142-
694x(01)00035-7
Carlile PR (2002) A pragmatic view of knowledge and boundaries: boundary objects in new product development. Organ Sci
13:442–455. https://doi.org/10.1287/orsc.13.4.442.2953
Contu A, Willmott H (2003) Re-embedding situatedness: the importance of power relations in learning theory. Organ Sci 14:
283–296. https://doi.org/10.1287/orsc.14.3.283.15167
Cook SDN, Yanow D (1993) Culture and organizational learning. J Manag Inq 2:373–390. https://doi.org/10.1177/
105649269324010
Cox A (2005) What are communities of practice? A comparative review of four seminal works. J Inf Sci 31:527–540. https://
doi.org/10.1177/0165551505057016
Douglas JD (1985) Creative interviewing, vol 29. Sage Publications, Inc, New York
Gherardi S, Nicolini D (2000) The organizational learning of safety in communities of practice. J Manag Inq 9:7–18. https://doi.
org/10.1177/105649260091002
Gherardi S, Nicolini D, Odella F (1998) Toward a social understanding of how people learn in organizations. Manag Learn 29:
273–297. https://doi.org/10.1177/1350507698293002
Gubrium JF, Holstein JA (1997) The new language of qualitative method. Oxford University Press on Demand, Oxford
Kaethler Journal of Organization Design (2019) 8:10 Page 17 of 17

Handley K, Sturdy A, Fincham R, Clark T (2006) Within and beyond communities of practice: making sense of learning
through participation, identity and practice. J Manag Stud 43:641–653. https://doi.org/10.1111/j.1467-6486.2006.00605.x
Jarzabkowski P, Sillince JA, Shaw D (2010) Strategic ambiguity as a rhetorical resource for enabling multiple interests. Hum
Relat 63(2):219–248. https://doi.org/10.1177/0018726709337040
Kaethler M, De Blust S, Devos T (2017) Ambiguity as agency: critical opportunists in the neoliberal city. CoDesign 13(3):175–
186. https://doi.org/10.1080/15710882.2017.1355002
Lave J, Wenger E (1991) Situated learning: legitimate peripheral participation. Cambridge University Press, Cambridge
Lindkvist L (2005) Knowledge communities and knowledge collectivities: a typology of knowledge work in groups*. J Manag
Stud 42:1189–1210. https://doi.org/10.1111/j.1467-6486.2005.00538.x
Litchfield RC, Gilson LL, Gilson PW (2015) Defining creative ideas. Group Org Manag 40:238–265. https://doi.org/10.1177/
1059601115574945
Mørk BE, Aanestad M, Hanseth O, Grisot M (2008) Conflicting epistemic cultures and obstacles for learning across
communities of practice. Knowl Process Manag 15(1):12–23. https://doi.org/10.1002/kpm.295
Moulaert F, Van Dyck B (2012) Framing social innovation research: a sociology of knowledge (SoK) perspective. Social
innovation: collective action, social learning and transdisciplinary research. SAGE, London
Nerland M, Jensen K (2012) Epistemic practices and object relations in professional work. J Educ Work 25(1):101–120. https://
doi.org/10.1080/13639080.2012.644909
Nonaka I, Takeuchi H (1995) The knowledge-creating company: how Japanese companies create the dynamics of innovation.
Oxford university press, Oxford
Polanyi M (1966) The logic of tacit inference. Philosophy 41(155):1–18. https://doi.org/10.1017/S0031819100066110
Regeer B, Bunders J (2003) The epistemology of transdisciplinary research: from knowledge integration to communities of
practice. Interdiscip Environ Rev 5:98. https://doi.org/10.1504/ier.2003.053901
Silverman D (1985) Qualitative methodology and sociology: describing the social world. Gower Pub Co, Farnham
Star S (2010) This is not a boundary object: reflections on the origin of a concept. Sci Technol Hum Values 35(5):601–617.
https://doi.org/10.1177/0162243910377624
Szulanski G (2002) Sticky knowledge: barriers to knowing in the firm. Sage, New York
Thompson M (2005) Structural and epistemic parameters in communities of practice. Organ Sci 16(2):151–164. https://doi.
org/10.1287/orsc.1050.0120
Wenger E (1999) Communities of practice: learning, meaning, and identity. Cambridge university press, Cambridge
Wenger E, McDermott RA, Snyder W (2002) Cultivating communities of practice: a guide to managing knowledge. Harvard
business press, Harvard
Wood DJ, Gray B (1991) Toward a comprehensive theory of collaboration. J Appl Behav Sci 27(2):139–162. https://doi.org/10.
1177/0021886391272001