Power and Participation in the Production of Boundary Objects
Robyn Thomas[1]
CardiffBusinessSchool
Cardiff, UK
CF10 3EU
Email:
Tel: +44-29-2087-5724
Leisa D. Sargent
Department of Management & Marketing
University of Melbourne
Parkville
Victoria, 3010
Australia
Email: lsargent unimelb.edu.au
Tel: +61-3-8344-5576
Cynthia Hardy
Department of Management & Marketing
University of Melbourne
Parkville
Victoria, 3010
Australia
Email:
Tel: +61-3-8344-3719
Power and Participation in the Production of Boundary Objects
Abstract
The research on boundary objects emphasizes how they allow actors from different “social worlds” to reconcile different meanings and produce shared understandings, in turn, enabling cooperation and the transformation of knowledge. Research has also indicated that this process should involve the participation of all the relevant groups in the co-construction of meaning; otherwise the designated artifact may fail to become a boundary object-in-use. Our study examines how meaning was negotiated in the case of a particular artifact – a culture toolkit – which was intended to secure cooperation among members from different parts of a telecommunications company in the development and implementation of a new customer-oriented culture. By conducting a detailed, realtime analysis of how the meaning of this designated boundary object was negotiated in a workshop, we show how different patterns of participation in negotiating meaning influence whether an artifact becomes a boundary object-in-use. Further, we show how these patterns of participation arise from the different strategies used by senior managers and their subordinates to exercise power associated with their respective vertical positions.
Key Words: Boundary objects, negotiation of meaning, power
Introduction
Over the last 25 years, the concept of boundary object – as delineated by Star and Griesemer (1989) and associated with scientific research – has transcended its own boundaries, becoming employed in the study of areas such as project management, organizational learning and knowledge management. The concept is attractive to management theorists because it represents an important means of achieving collaboration. Boundary objects can be used to solve “heterogeneous problems” through the way in which they “sit in the middle” of a “group of actors with divergent viewpoints” (Star, 1988: 46). Star & Griesemer (1989) examined how boundary objects such as specimens, maps and fields notes helped to bring about cooperation among professional scientists, amateur collectors, trappers, and university administrators in setting up a museum of vertebrate zoology. Since then, awide range of different artifacts have been studied – Gantt charts, engineering drawings, physical prototypes, maps, databases, accounting systems, and computer software – for their ability to bring about cooperation in business settings, especially among members of different horizontal functions.
This research emphasizes the importance of meaning i.e., the boundary object allows actors from different “social worlds” to reconcile their different meanings and produce shared understandings which, in turn, enables them to cooperate(Bechky, 2003a). This process occurs as diverse groups “fill in content and interpretations [of the boundary object] and negotiate as they see fit” (Yakura, 2002: 968). In other words, a designated artifact becomes a boundary object-in-use (Levina & Vaast, 2005) only if its meaning is negotiated through a process in which all the relevant groups participate (Star, 1988; Star & Greisemer, 1989; Carlile, 2004; Sapsed & Salter, 2004). In order to learn more about how the negotiation of meaning affects whether an artifact becomes a boundary object-in-use, we conducted a detailed, real-time analysis of the negotiations around a designated boundary object. The artifact in question was a culture toolkit, comprising a brochure, video and workshop template, and intended to secure cooperation among members from different parts of a telecommunications companyin the development and implementation of a new customer-oriented culture.
Our findings suggest that different patterns of participation among a range of actors in these negotiations help to explain whether an artifact becomes a boundary object-in-use. In our case, when participation was distributed and a range of participants contributed to the co-construction of meaning – which we refer to as “processual plasticity’ – the artifact became a boundary object-in-use. Our study suggests that this processual plasticity depends upon the use of strategies by both superiors and subordinates in exercising power to influence the negotiation of meaning. Depending upon the strategies employed by superiors and subordinates, this processual plasticity can be lost.
Our study makes a number of contributions. First, it indicates that an artifact cannot simply be designated as a boundary object; if it is to be a boundary object-in-use, it must be continually “worked on” by all participants all the time in order to maintain processual plasticity. By identifying different strategies used by senior managers and their subordinates, we also show how processual plasticity requires senior managers to redress the power advantages embedded in their superior vertical position; while middle managers must also exercise or exploit the power associated with their position in the hierarchy. In this way, we extend the work of Levina and Vaast (2005) to show how a boundary object-in-use results from the interplay between senior and subordinate members, and not just the “top-down” work of designated or informal boundary spanners. Second, we identify strategies that disrupt processual plasticity and result in meaning becoming reified, and show how such strategies can escalate during the course of the negotiations. At this point, the artifact no longer acts as a boundary object, but instead serves as a means of control as senior managers, consciously or unconsciously, use their position in the hierarchy to undermine the very artifact intended to function as a boundary object. In this way, our study builds on previous work which has found that the failure of boundary objects is often explained by horizontal, functional and occupational power relations i.e., as a result of competition among functions or departments (e.g., Bechky, 2002b; Carlile, 2004: Henderson, 1998) to show the effects of vertical power relations and direct line authority. Finally, we build on Carlile’s (2002: 452) work that shows that a boundary object that is effective at one stage “can falter when taken to another setting.” We show how the same artifact can slip in and out of functioning as a boundary objecti.e., an artifact can function as a boundary object-in-use and serve as a means of control, depending on the way in which actors participate in the negotiation of meaning.
The remainder of this paper is structured as follows. We first examine the organizational and management literature on boundary objects. We then introduce our case study and explain our methods of data collection and analysis. Third, we present our findings.Finally, we discuss their implications for research and practice.
The Emergence and Development of Boundary Objects
Research on boundary objects initially examined the way in whichscientific work is conducted by diverse groups of actors, such as researchers from different disciplines, amateur and professional scientists, as well as administrators and technicians. Objects, methods, findings and theories mean different things in these different worlds, and as a result, different meanings – and tensions – must be reconciled if cooperation is to ensue. “Simply put scientific work is heterogeneous. At the same time science requires cooperation” (Star Griesemer, 1989: 387). The same is true of management, which requires collaboration among actors from different groups, departments and organizations (Gerson Star, 1986). Like science, management inhabits “multiple social worlds” and requires “intersectional work” because “new objects and methods mean different things in different worlds [and] actors are faced with the task of reconciling these meanings if they wish to cooperate” (Star Griesemer, 1989: 388). For example, individuals from different parts of a company may have very different ideas regarding, for example, who the customer is (Sturdy Fleming, 2003), what constitutes innovation (Dougherty, 1992), or what project deadlines involve (Yakura, 2002). Accordingly, they create their own representations, investing them with meanings that suit their needs and reflect their backgrounds. However, these representations must be reconciled if individuals are to cooperate (Gerson Star, 1986) and collective action is to ensue (Hardy, Lawrence Grant, 2005).
Boundary objects provide an important means of translating such multiple, overlapping but divergent representations in ways that accommodate both diversity and synergy because they “inhabit several intersecting social worlds and satisfy the informational requirements of each of them” (Star Griesemer, 1989: 393). In this way, they provide a locus for communication, conflict, and coordination” (Yakura, 2002: 968), and create “the common ground that leads to shared understandings” (Bechky, 2003a: 326). They allow “actors with diverse goals” to work together (Briers Chua, 2001: 241-2) and promote “the sharing of knowledge in practice between diverse groups” (Sapsed Salter, 2004: 1515). Given their potential, it is not surprising that boundary objects have sparked considerable interest in areas such as project management and new product development, where cooperation among diverse groups is deemed vital. The interest in boundary objects is also linked to the growing use of practice-based approaches in knowledge management and organizational learning, according to which information is not universal, individual and explicit; rather, it is situated, social and tacit (Gherardi, 2000; Yanow, 2000). Instead of conceptualizing learning as something that goes on inside peoples’ heads, knowledge is produced as actors participate as members of “communities of practice” (Brown Duguid, 1991; Lave Wenger, 1991), collectively making sense of information in ways that enable action (Weick, 1995). Boundary objects help in this process by tying together individuals who come from different departments, functions or locations.
Participation and Boundary Objects
The artifacts that constitute boundary objects have no intrinsic meaning– artifacts are “experienced differently by different individuals and differently by the same individuals depending on the time or circumstance” (Orlikowski, 2000: 408).They require interpretation (Yakura, 2002) and are “brought to life … through social interaction” among a range of different actors (Briers Chua, 2001: 240). It is this looseness or ambiguity of meaning that enables them to function as boundary objects – by allowing different groups to read into them meanings that make sense to them (Bechky, 2003a).
Boundary objects may provide informational support but denote no intrinsic meaning. They are, in this sense, empty vessels to be filled with whatever is the preferred local beverage. Boundary objects facilitate the reading of alternative meanings by different groups (Sapsed Salter, 2004: 1519).
In this way, boundary objects are “plastic enough to adapt to local needs” and “robust enough to maintain a common identity across sites” (Star Griesemer, 1989: 393); they are “weakly structured in common use” and “strongly structured in individual use”; they are “concrete and abstract”, “specific and general” and “conventionalized and customized” (p. 408).
These tensions enable boundary objects to supply “concrete means of representing different functional interests and facilitating their negotiation and transformation” (Carlile, 2004: 559). In other words, there must be sufficient convergence of meaning to make the boundary objectrecognizable to the different groups – and able to serve as an effective means of communication across boundaries; but, at the same time, ambiguity and diversity of meaning are also crucial. Ambiguity allows different groups to read preferred or familiar meaningsinto the boundary object, making cooperation with other actors appealing; while the multiple, diverse meanings that come from the involvement of different groups help to generate new knowledge and learning. Some form of convergence in relation to these diverse meanings is also necessary to produce the shared understanding and common ground (Bechky, 2003a). In this way, effective boundary objects establish a “shared language for individuals to represent their knowledge; provide the means for individuals “to learn about their differences and dependencies across a given boundary”; and facilitate “a process of transforming current knowledge (knowledge that is localized, embedded, and invested in practice) so that new knowledge can be created” (Carlile. 2002: 452-3).
Star and Griesemer (1989: 412) emphasize that, in the case of boundary objects, meaning should be collectively negotiated because each social world “has partial jurisdiction over the resources represented by that object.”In other words, for a boundary object to be effective in sharing learning, generating knowledge and securing cooperation, actors from all the worlds that it is intended to link must be involved (Carlile, 2004). If the boundary object is neglected by one of the communities it is intended to bridge, it is likely to fail (Sapsed & Salter, 2004). The production of boundary objects therefore depends on distributed participation – where meaning is developed through all participants actively taking part in the negotiations around meaning – and a connection to or mutual recognition of other participants (cf. Handley et al., 2006).
Widely distributed participation can, however, be difficult to achieve. For example, work on communities of practice has noted that some actors may limit the participation of others (Handley et al., 2006) and that individuals may even voluntarily constrain their own participation (Roberts, 2006). Given that participation in the negotiation of meaning is equally central to the conceptualization of boundary objects, our first research question is: what different patterns of participation occur in the negotiation of meaning of boundary objects; and how do they arise?
Power and Boundary Objects
In advocating the collective negotiation of meaning, Star and Griesemer (1989: 389) argue that each participant “must maintain the integrity of the interests of the other audiences”. However, the negotiation of meaning is widely recognized to be a process in which actors exercise power (Deetz & Mumby, 1990; Hardy Phillips, 2004). Recent work on boundary objects has also drawn our attention to the need to consider power (Boland & Tenkasi, 1995; Carlile, 2002; Levina & Vaast 2005;Sapsed & Salter, 2004). Henderson (1998: 146) notes that, because boundary objects can be read in different ways, they have “political possibilities … multiple readings can both facilitate collaborative work and contribute to conflict among collaborators.” Even the apparent resolution of different meanings “does not mean consensus. Rather, representations, or inscriptions, contain at every stage the traces of multiple viewpoints, translations and incomplete battles” (Star Griesemer, 1989: 413). The development and implementation of new knowledge often has negative consequences for other organizational members, and therefore the success of a boundary object is linked to political as well as practical skills (Carlile, 2004); and boundary objects may be used, not purely for technical purposes, but also “as a means of representing and instigating difference and conflict” (Bechky, 2003b: 724).
Most of the empirical work to date has focused on the effects of power related to the horizontal or functional relations among different specialist groups (e.g., occupations, professions, and departments) as they compete for control over tasks and knowledge. As Henderson (1998: 146)has noted in relation to differences between research and development and manufacturing functions: “Diverse orientations can also potentially lead to out-and-out conflict when readings are based not only on different specialty orientations and background experience, but also on departmental politics and distrust”. Carlile (2004: 566), in noting the dominance of one particular form of boundary object, attributed it to power and status differentials between marketing and safety functions: since the two groups did not occupy “politically equal positions in representing their knowledge to each other” (Carlile, 2004: 565), they did not exert the same influence over the boundary object. Bechky (2003b) has examined power relations between occupational groups and found that the greater status, authority, and expertise of engineers in relation to technicians and assemblers led to situations where, rather than acting as a boundary object, artifacts were used to exercise control over work, solidify occupational status, and maintain jurisdiction over tasks.
In being used in organizations, boundary objects do not only bridge horizontal or functional boundaries, they also bridge vertical ones, involving direct line authority. For example, the particular artifact that is designated to be a boundary object is typically selected by actors who “hold positions of power” (Levina & Vaast, 2005: 341). Henderson (1998), in describing the top-downapplication of TQM in a call centre, noted that “politics in the form of management prerogatives can be built into” boundary objects,with adverse effects on “creativity and innovation.”Revisiting Yakura’s (2002) work, Sapsed and Salter (2004) argue that while timelines may offer potential as boundary objects because of the different meanings that can be read into them, they are also used as a form of managerial control. Despite their potential importance, vertical power relations between senior managers and their direct subordinates have not so far been studied systematically in the case of boundary objects, even though these “relational dynamics” (Bechky, 2003b: 749) are likely to influence the nature of participation in the negotiation of meaning and may differ from those associated with horizontal, functional relations. In addition, these relational dynamics are reciprocal and not simply “top down.” Therefore the use of power by both superiors and subordinates must be examined for its effects on the designated boundary object. Hence, our second research question is: do groups draw on vertical power relations to influence the negotiation of meaning of boundary objects and, if so, how?
Methods
The Case Study
Our study concerns the use of boundary object during a company-wide culture change program at UTel[2] – a pioneer in licensing open-standard GSM/GPRS, EDGE and WCDMA[3] technology to manufacturers of mobile phones and other mobile communication devices. At the time of the study, UTel employed approximately 1,500 employees, most of whom were located at the European Head Office, with the remainder working in sites in UK, Europe, Asia and North America. The company had been formed in 2001, when it was “spun off” from being an internal division of GlobalTel, a global telecommunications company that was undergoing major restructuring and redundancies at the time as a result of low cost competitors and lower growth in the mobile phone market worldwide. As an independent organization, UTel changed from being part of a much larger company that made mobile phones for the end-user to one that sold “knowledge” to mobile phone manufacturers. Its success now hinged on the replacement of the existing engineering focus, which stemmed from its previous position as a unit that interacted primarily with other divisions within GlobalTel, with a customer oriented culture that focused on the new customer – no longer the end-user of the phone but other companies who manufactured and sold phones.