Curating creative communities of practice: the role of ambiguity

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 freedom 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 articulated 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 importance 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 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 communities. These, with few exceptions, are activity oriented and often follow a pattern of procedures that include developing clear objectives and purpose, creating a clear plan or infrastructure, selecting strong leaders, facilitating dialogue and so forth. Recommendations are highly pragmatic and belie a notion that exchanging knowledge is straightforward 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, informality 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.

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 participation and reification. Participation suggests both action and connection; more specifically, 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 characterised 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 experience into “thingness”’ (Wenger 1999: 58). Viewed socially, it is the moment that 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 formalised 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).

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 application in all manner of contexts and with a range of formulaic approaches leading to a deterioration of its original qualities that emphasised situatedness, social interaction, 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 mirrors 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 appropriate infrastructure for its execution, the selection of strong leaders, and the facilitation of dialogue and collaboration. Knowledge is perceived as accessible and easily transferable, diminishing the dimension of practice in CoPs and emphasising community 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.

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, locations, 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 components; 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 relevant 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 limited 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 improvised 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 components 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 structuring 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 indirectly seeding future collaboration’. He distinguishes between structuring and seeding, 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. Likewise, 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, he argues, can be indirectly given form through a focus on symbolic monuments, infrastructural instruments and conceptual points of focus.

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 participants 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 prescription 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 structures 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’.

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 developed. 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.

This section explicates the relationship between ambiguity with the channelling factors of problem loyalty, peer recognition, slack space and alignment as elucidated 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).

By investigating the relationship between ambiguity and the four factors of peer recognition, 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 corroborates Amin and Roberts’ (2008) description of the importance of ambiguity as a condition 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 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, ambiguity 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 alignment was necessary to balance out the excesses of ambiguity when it became destructive to group dynamics.

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, comparable 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 ensuring the existence of a wider support network.

Taking the experiences of BIO50 participants, mentors and curators, we can assemble 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 collaborative 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.