Part I
Framing and Definition
John C. Morris and Katrina Miller-Stevens
Introduction
Thirty years ago, Barbara Gray (1985) published an article entitled “Conditions Facilitating Interorganizational Collaboration” in the journal Human Relations. Drawing heavily from the literature on interorganizational theory, Gray (1985) argued:
There is a growing need to promote collaborative problem solving across various sectors of society…. Organizing such collaborative efforts requires focusing on the interorganizational domain or set of interdependencies which link various stakeholders rather than on the actions of any single organization.
(p. 911)
Gray’s research was focused not only on an interorganizational environment, but an inter-sectoral environment as well. More traditional conceptions of the interorganizational space tended to engage either public-sector settings, in which legal authorities entrusted to organizations determined the kinds of interactions possible (see Barnard, 1938; Simon, 1997; Weber, 1968), or the private sector and market-based systems that were governed entirely by free market forces (see Buchanan & Tullock, 1962; Lawrence & Lorsch, 1978). Other scholars (e.g., Niskanen, 1994; North, 1990; Olson, 1965; E. Ostrom, 1990; V. Ostrom, 1989) argued that market values and processes should be applied to the public sector, but there was no body of literature suggesting a truly cross-sectoral approach to problem-solving. In this sense, Gray was truly an intellectual pioneer.1
Gray identified four ways in which her work was distinct from previous work. First, Gray (1985) focuses her analytical attention on the set of relationships present across an interorganizational domain, rather than focusing on a central (or referent) organization. For Gray, all of the relationships in the system are important, not just those involving the referent organization. Second, her work focuses on “underorganized domains” (Gray, 1985, p. 912); those that are not already engaged in highly structured networks or structures. This allows Gray to address the conditions that might allow the development of collaborative structures.
Third, Gray is concerned with settings in which “wicked problems” (Rittel & Webber, 1973) exist. In Gray’s (1985) terms, these are “… domains which cannot be satisfactorily managed by a single organization or by an oligopoly” (p. 913). Wicked problems are thus problems that defy solutions (or even definitions), and that cannot be addressed satisfactorily by single organizations. Finally, Gray’s work examines the set of relationships and behaviors that develop within the setting, from the premise that the set of relationships present defines the domain, rather than assuming that the type of domain present controls the behaviors and relationships present (Gray, 1985, p. 913). From the perspective of hindsight, Gray’s early work arguably proved to be the foundational piece in this stream of literature. She followed the publication of her article four years later with her book Collaborating (1989), which more fully developed the arguments and propositions offered in the initial article, and offered a series of case study vignettes to illustrate her arguments.
Gray’s work opened a floodgate of new research, as scholars began to explore more fully the theoretical and practical implications of her ideas. Early in this process, the scholarship began to move away from the explicitly interorganizational underpinnings of Gray’s work, as many scholars approached collaboration as a study of institutional networks. The relatively informal nature of collaboration (when compared to more traditional authoritative interorganizational conceptualizations) lent itself well to a network approach. In this respect: relationships were not governed by an established set of rules; barriers to entry and exit were largely nonexistent; traditional public-sector accountability mechanisms, especially for both resources and outcomes, were inappropriate; and traditional conceptions of leadership were not appropriate for the setting. Personal relationships and connections were important, and power and authority were likely to be shared equally among participants. For these reasons, much of the scholarship in these early years was written by those with a strong interest in network theory.
The literature has progressed significantly in the intervening years, and scholars have brought a number of different perspectives to bear on the study of collaboration. Many articles have been written from the perspective of collaborative management (Koontz & Thomas, 2006; Leach, 2006; McNamara, 2012; Selin & Chavez, 1995; see also O’Leary & Bingham, 2009), accountability mechanisms (Bardach & Lesser, 1996; Page, 2004; Romzek, LeRoux, & Blackmar, 2012), goal-setting (Gray, 1989; Wood & Gray, 1991), collaboration as governance (Agranoff & McGuire, 1999; Ansell & Gash, 2007; Emerson, Nabatchi, & Balogh, 2011; Imperial, 2005; Provan & Kenis, 2007), and others. Throughout this period some questions about collaboration have been settled, but many more are left unanswered, and even unasked. The purpose of this volume is to examine the state of knowledge in collaboration theory and practice; we seek to provide insight into existing ideas and theories of collaboration, determine the state of knowledge in the study of collaboration, identify some of the unasked questions in the field, and offer some thoughts about questions that are at the cutting edge of collaboration research. Finally, we offer some cases to illustrate how these issues translate to help us understand collaborative processes within the public sector.
Five Themes
We begin this process by developing five overarching themes that thread their way through the chapters in this volume. Taken collectively, these chapters examine these themes in some detail, seek to identify the current state of knowledge inherent in these themes, and identify issues for future research.
Definitional Clarity is a Challenge
In the spring of 2012 we offered a graduate seminar in collaboration. As an exercise in definitional clarity, we invited students to review a group of about 30 journal articles in collaboration. The articles ranged from Gray’s (1985) work to articles published in 2012, and many were chosen because they represented important contributions to the body of literature. The purpose of the exercise was to note all of the definitional elements included in each of the articles. While there were a number of elements that were included by subsets of authors, the list included more than 60 different definitional elements (many of which were included by only one author). No single element was included in more than one-third of the articles reviewed.
The wide variability noted in this vignette is endemic in the academic literature. At the same time, there are reasonable arguments both for and against definitional clarity. Definitional clarity can lead to a greater degree of both certainty and specificity in terms of the phenomenon under study. If those engaged in the discussion share a common definition, effective communication and shared knowledge are optimized. Greater definitional clarity also enhances learning, especially for students new to the literature. To the extent that academics are clear on a definition, the lessons and knowledge passed to practitioners can be more effective.
The same issue was noted by Wood and Gray (1991) more than 20 years ago. In their introduction to a journal symposium on collaboration (Wood & Gray, 1991), they wrote that:
Definitions are crucial to theory building. A general theory of collaboration must begin with a definition of the phenomena that encompasses all observable forms and excludes irrelevant issues. We began our work on these special issues assuming that a commonly accepted definition of collaboration existed and that we could move quickly beyond this primal task. Instead, we found a welter of definitions, each having something to offer and none being entirely satisfactory by itself.
(p. 143)
On the other hand, one may reasonably argue that firm definitions of complex and multifaceted human interactions suppress the ability to detect nuances in observed behaviors that might otherwise prove to be empirically or theoretically important. A definition explains what a ‘thing’ is, but it also defines what it is not. It may also be argued that the study of collaboration is still in its infancy, and that we do not yet understand the behavior well enough to arrive at a clear, useful, and relatively universal definition. Finally, singular definitions tend to support the status quo at the expense of creativity and scholarship outside of “normal science” (Kuhn, 1996).
Regardless of one’s position on this question, the empirical evidence suggests that we do not have a clear and widely accepted definition of the phenomenon under study, even a quarter-century after Wood and Gray’s (1991) observation. The implications of this fact are an important theme of this book, whether it is addressed as an issue of theoretical development, or whether implications for practitioners are brought into focus.
Collaboration is Constantly Evolving
As noted earlier in this chapter, the study of collaboration is a relatively young enterprise. Much like the early development of the literature in implementation studies (see O’Toole, 1986), one may argue that collaboration research has seen steep growth in the number of studies, models, and theories published. These efforts may seem somewhat haphazard and undifferentiated at first, but like the implementation literature, one may detect certain characteristics in the literature. While other chapters in this volume will address those characteristics in more detail, our initial point is that the steep growth in the number of collaboration studies published, coupled with the definitional challenges noted above, indicates a field of study in a state of constant and rapid development. As we refine our theories and models, we move collectively toward a more complete understanding of this form of interaction.
The same may be said of collaboration in practice. While the nature of the gulf between theory and practice in public administration is beyond the scope of this discussion, we understand intuitively that the practitioner’s view of the world (and the terms in which they describe that world) does not always match with the views and terms found in academia. Practitioners have been employing ‘collaboration’ in the workplace for many years, yet there is no more agreement among practitioners as to the definition and operation of ‘collaboration’ than there is among academics. Moreover, the term ‘collaboration’ may carry a different perception than ‘coordination’ (for example: ‘working together’ vs. ‘directing’), leading practitioners to adopt different terms for the same activity, or, conversely, the same term for different activities. Still, as collaboration becomes more commonly taught in Master of Public Administration (MPA) programs and as those graduates enter the public workforce, it is more than likely that the different approaches to collaboration will find their way into both the practice and the lexicon of practitioners. The effect will be that the nature of collaboration, as practiced in the workplace, will evolve as well.
Collaboration can be Understood as Both Organizational Process and Structure
Distilled to its essence, collaboration is necessarily a group activity involving two or more people. Collaboration is an interaction that takes place between people, or organizations, or both, in a wide range of settings. In this regard, we suggest that Gray (1985, 1989) provides a useful basis for collaboration. Likewise, more recent work that treats collaboration from a distinctly network perspective is also useful. Within these arenas, one may identify literature that seeks to classify collaboration as one of several forms of organizational interaction (e.g., Imperial, 2005; McNamara, 2012). Most of these works tend to focus specifically on an interorganizational domain, much as Gray did. The underlying assumption is that collaboration takes place in an explicitly interorganizational environment, and the research questions tend to address questions of resource-sharing, goal-setting, and outputs or outcomes of the observed activity.
Other streams in the literature treat collaboration as an organizational process. Agranoff and McGuire’s (1999) foundational piece on collaborative management has spawned a great deal of additional work designed to understand how collaborations are administered. Specific streams in this arena focus on goal-setting (Gray, 1989; Wood & Gray, 1991), leadership (Agranoff & McGuire, 2003; Weber & Khademian, 2008), accountability (Bardach & Lesser, 1996; Page, 2004), and decision-making (Glass, 1979; Kaner, 1996; Smith, Nell, & Prystupa, 1997), among others. Much of the empirical work in this tradition is case based and exploratory, and is often concerned with drawing comparisons between traditional public management and collaborative management.
Not all Collaboration is Equal
A relatively new stream in the literature revolves around attempts to classify different forms of collaboration, usually along a continuum of action (see Imperial, 2005; Keast, Brown, & Mandell, 2007; McNamara, 2012). Generally conceived as an ordinal scale, theorists attempt to place different forms of interaction on a scale based on a set of characteristics. The underlying idea is that these interactions...