Boundaries and objects: the materiality of interdisciplinarity

Let us start with the first phase: “Interdisciplinary fields detach a category as subject and object from existing disciplinary frameworks, loosening boundaries and stimulating trading zones.”

As in all types of collaborative team efforts (see e.g. West 2002), also in collaborative research, members start off with the team engaging in joint definition of the goals of a project, attempt to define the core problem and questions, and design a framework for coordinated inputs with ongoing communication and interaction (Klein 2008b). In interdisciplinary collaboration, the definition of the goal and problem itself is often problematic. As Klein states, the problem needs to be first “detached” from the various existing frameworks, which originate from the disciplinary domains of the participants.

First and foremost, this involves the “loosening of boundaries” that Klein refers to.

Boundaries are a crucial and critical element in all cross-functional and interdisciplinary collaboration. Within the practice-based studies, Erden et al. (2014) note that a considerable body of research concentrates on what they call the theme of “practice boundaries and coordination of work”. More specifically, research has focused on the differences and boundaries between social practices that result from adopting the practice-based view of knowledge in itself – viewing knowledge as localized, plural and situated. Carlile’s work (2002, 2004) has been crucial in increasing our understanding of how knowledge itself is a barrier to knowledge sharing and creation. He has developed a framework that depicts three kinds of knowledge boundary – syntactic, semantic and pragmatic – and suggests the ways in

which to overcome these boundaries. A particular prominence in his research has been given to the concept of boundary objects; that is studying the knowledge processes – more particularly that of knowledge sharing - that unfold around material artefacts, and facilitate the sharing of knowledge across the knowledge domains.

The term “object” – is used in literature in a variety of ways, some of which I will explore in this chapter. In its simplest definition within this particular context of collaboration²³, objects are “the collection of artifacts that individuals work with—the numbers, blueprints, faxes, parts, tools, and machines that individuals create, measure, or manipulate (Carlile 2002, 446)”. To broaden the definition, Star (2010, 603) states: “anobject is something people act toward and with. Its materiality derives from action, not from a sense of prefabricated stuff or

‘‘thing’’-ness. So, a theory may be a powerful object.” The above implies that what is an object is not bound to its physical substance – rather, it is the action it potentially invokes that makes it “an object”. “Boundary object” as a consequence is an object that acts in the capacity of being the “stuff of action” (ibid.) in creating a shared space across boundaries.

Empirically, for example Bechky (2003) in her ethnographic study on the work practices of engineers, technicians and assemblers on the production floor, found that in cases where the participants come from strongly contextualized and situated occupational communities, very concrete and tangible means were needed in order to achieve knowledge sharing between these participants. The required tangibility, however, may pose additional challenges, as

“people are not always able to create tangible definitions of their problems… In attempts to share knowledge across occupational boundaries, written and verbal explanations frequently failed to make meanings clear. Because their languages emerged from different contexts, members of different groups had a difficult time finding common ground on which to base their conversation. This common ground was more frequently found in a tangible object, which provided a concrete referent that individuals could manipulate to embed the understandings of others into their own understanding of their work context.” (Ibid. 2003, 327.)

23 Of the more general use of the term, Star writes (2010, 603): ”In common parlance an object is a thing, a material entity composed of more or less well-structured stuff.”

Bechky was describing the importance of boundary objects, which indeed has become a

“house-hold concept” ever since Carlile introduced it in 2002, becoming an almost “catch-all”

notion of explanation to all things object-like (Nicolini et al. 2010, 614).²⁴ However, to be more precise, Carlile re-introduced a concept originally developed by Star in 1988. Star developed the model of boundary objects (to full extent with Griesemer in 1989) that theorizes on objects that are shared and shareable across different problem solving contexts, and in fact her concept was intended to explain the possibility of cooperation of groups working without consensus – not to be applicable in all situation of collaboration. In addition, boundary objects involve three dimensions: interpretive flexibility, that is their dependence on the use and interpretation of the object; the material / organizational structure of the objects;

and their scale / granularity.²⁵

Most importantly to my research, Star (2010) points out that boundary objects are “a sort of arrangement that allow different groups to work together without consensus” (ibid. 602). The context from which her initial framing of the concept arose was in fact interdisciplinary scientific research, which she characterizes as co-operative work without consensus, yet still succeeding. “From my own fieldwork among scientists and other cooperating across disciplinary borders, and two historical analyses of heterogeneous groups who did cooperate and did not agree at the local level, it seemed to me that the consensus model was untrue.

Consensus was rarely reached, and fragile when it was, but cooperation continued, often unproblematically. How might this be explained? (Ibid. 604).” For Star, boundary objects arise from the information and work requirements as perceived locally and by groups wishing to cooperate, and what is important is how practices structure, and language emerge, for the doing of things together. Thus in interdisciplinary contexts, boundary objects emerge from the needs of the people in order to do things together without consensus. They “sit in the middle” (ibid. 608) in order to create the shared context. In fact, Star explains that for her, the term boundary is used to mean a shared space, where the common objects form the

24 The concepts widespread use has prompted Star to name her article “This is Not a Boundary Object:

Reflections on the Origin of the Concept” (2010).

25 Star (2010) argues that in fact almost all work on boundary objects concentrates on the interpretive flexibility dimension, neglecting the other two.

boundaries between groups through their flexibility and shared structure, being the “stuff of action” (ibid. 603). In this sense, boundary objects themselves create boundaries as the shared space makes the joint action possible.

The basic premise in most research on boundary objects is that the practices of interacting with artefacts, technology and tools are constitutive of knowing and learning within and between organizations. For example the research of Bechky (2003), Orlikowski (2007) and Carlile (2002) show how learning and knowing indeed does not happen just in the mind, but in interaction with the material environment (represented through the objects). However, Gärnter (2013) points out that there is a tendency to downplay the role of the human body in these interactions (for example assuming that the interactions physically can happen in the first place), as well as emphasizing the here-and-now quality of interactions. They do not take into account how material interactions are also themselves learned over time as embodied skills. Erden et al. (2014) have further noted that boundary objects are not static tools, as much research takes them to be, but dynamic. Star (2010) herself points this out: boundary objects are dynamic organic infrastructures that arise from the needs of the people cooperating, and that they grow, evolve and indeed – die. She sees that the dynamics involved is core to the whole notion of boundary objects, and which often has been ignored in the concept’s use.

The dynamic involved has to do with the ill-structured nature of the boundary objects. There is an anomaly, a problem that seems to defy existing categorizations. It something that does not “fit”: “Our experience and our language frequently delineate ‘in-between’ phenomena…

Characterizing what lies ‘in-between’ things, or that which is unusual, new, or which simply just ‘doesn’t fit’” (Aram 2004). Star calls these “residual categories” (Star 2010, 615); and as in the quote from Klein in the beginning of this Chapter, they are detached from the existing disciplines and such. These anomalies draw people together – often from the “outside” of the actual categories, and as cooperative work starts to emerge, boundary objects are generated.

The group, however, does not have a consensus – or a fragile one at best -– on fundamental issues such as evaluation, measurement, validity and the like. The problem – and the later emerging processes of solving it - is given a very ill structured and sketchy form that serves as the basis for creating a shared space and basis for communication between the different

domains within the group, being however recognizable enough to serve the different information requirements and allowing for the translation to take place. When needed, this object is “reversely detached” from the interdisciplinary practice and space, and worked on in the disciplinary domains, making it more structured and specific (but not in the interdisciplinary sense). The group is able to tack back-and-forth between these types of objects, the structured and the ill structured, in order to satisfyingly resolve the problem at hand. Over time, if the boundary object reaches a critical mass, or becomes of broader (political, economical, societal) interest, and if the ill/well-structured forms of it collapse into one, the object becomes standardized and part of some domains infrastructure. Until the next residual category or anomaly comes up, “the others” take it up – and a new boundary object is born.

One more concept that emerged from both Klein’s (2008b) description opening this Chapter as well as practice based research on cross-boundary coordination is that of trading zone.

Kellogg, Orlikwoski & Yates (2006) studied a fast-paced digital marketing organization (a post bureaucratic organization in their terms) in order to examine how members of different communities coordinate their work across boundaries. They draw from a perspective developed in a study by Galison (1997) on how distinct communities within physics were able to align their activities without homogenizing the inherent diversity of their communities.

This research found that despite the fundamental differences members were able to work out coordination practices in “exquisite local detail, without global agreement” (Galison 1997, quoted in Kellogg et al. 2006, 39). This is strongly aligned with Star’s notion of cooperation without consensus. The research proposed that this type of cross-disciplinary interaction could be described as a “trading zone”, where the local coordination of ideas and actions may take place despite differences in the disciplinary purposes, norms, meanings and evaluative criteria. They further propose that this trading zone is a coordination structure that is enacted by organizational members, allowing the continuous practices to make one’s own perspective visible happen. Kellogg et al. conclude that in a high-paced, volatile and uncertain environments such as digital media, boundary objects may not be effective, as thoughts and ideas change too rapidly for them to be “inscribed within objects” (ibid. 41).

This opens up few questions regarding boundaries and specially boundary objects. If, as Kellogg et al. demonstrated, in some types of collaboration there is evidence of something like objects that facilitate the working of together, but they are not “boundary objects”, what other types of objects might emerge in interdisciplinary collaboration? And what about the actual reason for the cooperation in the first place, as in “why these participants in this particular time and space have gathered together”? Star (2010) and Klein (2008b) both refer to this in that they both see that some “anomaly” or “detaching a category as subject and object from existing disciplinary frameworks” as the starting point of interdisciplinary collaboration. But what is the anomaly or the detached object, and what is its role in practices?

I found important additional insights from the plural approach to objects developed by Nicolini et al. (2012). They have taken up on the tendency to ascribe the status of “boundary object” to all objects, and instead have looked at the multitude of roles objects in fact play in supporting interdisciplinary²⁶ collaboration, following Star (2010) in understanding objects as something people act toward and with.²⁷ They develop a pluralistic theoretical approach to studying objects in this type of setting, drawing - in addition to the concept of boundary objects - from both the notion of “epistemic things” introduced by Knorr Cetina (1999), as well as the role of the object(ive) that motivates collaborative work from activity theory. They also include Orlikowski’s (2007) notion of the “scaffolding” or infrastructure that objects provide in daily work practices. ²⁸ This is also to what Star (2010) refers to in objects becoming “standardized” and becoming part of the categories and structures of a particular domain.

26 Nicolini et al. use the term cross-disciplinary, without defining it further. For the sake of being consistent, I dare to use interdisciplinary also in the conjunction of their work (excluding naturally citations).

27 Also Ewenstein & Whyte (2009) have taken a more holistic and multidimensional approach to looking at objects (architectural drawings in their empirical case) in order to see the roles they play in knowledge-intensive work. They employ the concepts of boundary objects, epistemic objects and technical objects, taking a cultural approach to analyse their situated meaning.

28 Nicolini et al. (2012, 614) argue that although all four perspectives are distinctive, they share a number of common assumptions that make it possible for them to be used together: seeing collaboration as practical accomplishment; the mediation of social action by artefacts; groups and communities are results of organizing work, not a given; social structures are both mediums and outcomes of human activities; that action and environment all mutually constituted; and that human actors not only are driven by rational considerations but also by emotions, desires and passion.

Nicolini et al.’s plural approach includes important elements that I have found to be relevant and holding explanatory power to the empirical material of my ethnography. First, they apply the idea of the epistemic object in order to shed light on “why people make the effort to search for alignment to begin with” (ibid. 618). Thus the object of investigation – i.e. the anomaly or other – that is in the process of being materially defined is an epistemic thing or object that embodies what one does not yet know. They are open-ended and are the source of interest and motivation for the participants. Knorr Cetina (1999) has explored where the power of these epistemic objects arises from, and Nicolini et al (2012, 618) elaborate on her findings. First, epistemic objects arouse “wanting” and a lack of fulfillment that generates energy and emotional investment in the participants as well is the group that is forming around the epistemic object. It is thus the “push” or “drive” that sets the wheel on motion. Based on their ethnography, Nicolini et al. aptly describe how “the epistemic object triggered a form of desire and attachment that had a libidinal, rather than calculative origin (ibid. 619).²⁹ Epistemic objects also provide the engine of solidarity for the group, and from their material Nicolini et al. “find” a lead scientist that “was, in a real sense organizing by desire – jelling the group by stimulating attachment and a desire to know (ibid. 619)”.

The nature of the epistemic object also influences the ways of working, as it proposes some structure on the “lack of knowing” related to it. It thus guides the participants towards some forms of “collective obligations” which then turns the loose group into a “proto-community”

(Ibid. 619)³⁰. The group may thus engage in practices that are indeed interdisciplinary, and community-like, but most importantly, they are not stemming from the socialization processes of the community (as in the communities-of-practice literature, see next Chapter). Rather they are induced by the epistemic object and the community-like practices stem from the promises it holds for each participant.

However, the above paints a rather harmonious picture of interdisciplinary collaboration. The object of collaboration is in most cases at least partly given (the “official” reason for the

29 I am willing to argue that most organizations are not prepared for the ”libidinal desire” that their product or service innocation work might generate, let alone attempting to nurture it. Of course, more sadly, most innovation work might not generate any such desires in the first place. Therein may lie a bigger challenge altogether.

30 The notion of ”proto-community” is an interesting one, even though Nicolini et al. elaborate it no further.

group to be formed, as in a development project of a new product or service). Nicolini et al.

draw further from the cultural historical activity theorists in order to explore how the partly given nature of the object dictates what types of skills and functions should be involved, what the division of work might be like, and what is the position of each member within the group.

In some cases participation may even be forced. Thus the object is also potentially conflictual and heterogenous and “the mutuality performed by the object is far from a smooth fusion of intents and goals (ibid. 622)”. They name these activity objects.

As the collaboration progresses, the objects that facilitate the work across boundaries become pronounced. It is in this function that Nicolini et al. place boundary objects. They follow Star

& Griesemeyer (1989, 393) in seeing them as “flexible artifacts that inhabit several intersecting social worlds”. They provide a shared medium allowing the representation of particular knowledge to be such that the “other side” of a particular boundary can understand it as well. They also provide concrete means to uncover the differences that the different perspectives might have. Boundary objects also act as anchors around which the emergent shared meanings can happen. Boundary objects are complimented by boundary-spanning activities, that is the face-to-face interactions that supporting the role of the objects. (Nicolini et al. 2012, 616.)

Finally, Nicolini et al. turn to the bulk of objects that surround us in our work practices. They are the ones that are in the “shadow of other practices (ibid. 622)”, and form the mundane support of everyday life (Orlikowski 2007), or “material infrastructure” as defined by Star and Ruhleder (1996). Nicolini et al. further distinguish (after Hanseth & Lundberg 2001) between work-oriented infrastructure and service infrastructure that support interdisciplinary collaboration. Work-oriented infrastructure refers to the various objects that help initiate and support the collaborative work practices, and have a relational property that emerges from the practices – i.e. the objects are infrastructure only if they become used and enacted in practices. Typically these include the various ICT-systems people use (e-mail, chats, social media platforms, collaboration software, databases etc.) and documents (e.g. project documents). Service infrastructure on the other hand is more “sunk-in” and invisible in the sense that they are more removed from the day-to-day practices, even though work could not happen without them. These include the physical space where collaboration takes place and

the (material) services of the space such as electricity, photocopiers, coffee machines etc.

These objects are rarely considered in themselves, they are taken for granted – until for example the photocopier or coffee machine breaks down!

These objects are rarely considered in themselves, they are taken for granted – until for example the photocopier or coffee machine breaks down!