Revealing gaps in the project management literature: An application of multi-paradigm review to explanations of megaproject governance

Sanderson, Joe Working Paper Revealing gaps in the project management literature: An application of multi-paradigm review to explanations of megaproject governance Birmingham Business School Discussion Paper Series, No. 2013-02 Provided in Cooperation with: Birmingham Business School, University of Birmingham Suggested Citation: Sanderson, Joe (2013) : Revealing gaps in the project management literature: An application of multi-paradigm review to explanations of megaproject governance, Birmingham Business School Discussion Paper Series, No. 2013-02, University of Birmingham, Birmingham Business School, Birmingham, http://epapers.bham.ac.uk/1722/ This Version is available at: https://hdl.handle.net/10419/202646 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-sa/2.5/ Birmingham Business School Discussion Paper Series Title: Revealing gaps in the project management literature: an application of multi-paradigm review to explanations of megaproject governance Author: Joe Sanderson Discussion Paper: 2013-02   1  Revealing Gaps in the Project Management Literature: An Application of MultiParadigm Review to Explanations of Megaproject Governance Joe Sanderson Birmingham Business School [email protected] Abstract A multi-paradigm literature review methodology, ‘interplay’, is applied to alternative explanations of megaproject governance and performance. A two-fold categorisation of explanations, functionalist and interpretivist, is employed. The key insight is that despite important differences in epistemological orientation these two categories of explanation are essentially ‘performative’, which is expressed through a shared acceptance of the notion of actor farsightedness. This means that governance in megaprojects is primarily understood as static, convergent and patterned forms of organization (made order), while governance as discontinuous, divergent and fluctuating micro-processes of organizing is ignored. Having identified this explanatory gap, the article concludes with a call to refocus project governance research to include proper consideration of the multiple processes of organizing through which actors use, reproduce and transform governance as made order. Keywords: multi-paradigm review; paradigm interplay; functionalism; interpretivism; megaprojects; governance.   2  This discussion paper is copyright of the University, the author and/or third parties. The intellectual property rights in respect of this work are as defined by The Copyright Designs and Patents Act 1988 or as modified by any successor legislation. Any use made of information contained in this paper must be in accordance with that legislation and must be properly acknowledged. Copies of the paper may be distributed and quotations used for research and study purposes, with due attribution. However, commercial distribution or reproduction in any format is prohibited without the permission of the copyright holder.   3  Introduction Project governance has been defined as ‘the complex process of steering multiple firms, agencies and organizations that are both operationally autonomous and structurally coupled in projects through various forms of reciprocal interdependencies’ (Miller and Floricel, 2000: 135). This article uses multi-paradigm literature review to provide a comparative commentary on a sample of the wide variety texts addressing the topic of project governance. Multiparadigm review is chosen as a means of examining the literature in a particular research domain to consider the impact of researchers’ ‘underlying and often taken-for-granted assumptions on their understandings of organizational phenomena’ (Lewis and Grimes, 1999: 673). Moreover, multi-paradigm reviews ‘may distinguish the value and limits of divergent perspectives’ (Lewis and Kelemen, 2002: 260) and thereby enhance our understanding of particularly complex and ambiguous phenomena. There has been a noticeable shift in the project management literature in recent years away from a narrow focus on the technical tasks necessary to deliver project outcomes towards a much greater interest in how the interactions between the multiple actors responsible for undertaking those tasks are organized and coordinated , in other words project governance (see, for example, Clegg et al., 2002; Flyvbjerg et al., 2003; Miller and Lessard, 2000; Pitsis et al., 2003; van Marrewijk et al., 2008; Williams et al., 2009; Winch, 2001, 2009). Accompanying this shift, there has also been a more general increase of interest in projects as a means of organizing work. Projects and project management have been seen as a better way to achieve effective intra-organizational integration and optimal resource utilization (Cleland, 1997), to stimulate knowledge sharing, learning and creativity (Hansen et al., 1999; Hobday, 2000; Silver, 2000) and to control activities in turbulent environments (Ekstedt et al., 1999). Some have even described the expanding influence of projects as a   4  form and a process of organizing as the ‘projectification of society’ (Lundin and Söderholm, 1999; Midler, 1995). This extension of project concepts and practices to the organization of work sensitizes us to the need to better appreciate the meanings and limitations of these concepts and practices (Söderlund, 2011). Rather than looking at the notion of projectification in general, however, our concern here is with governance in the context of an increasingly popular project form, the megaproject (Flyvbjerg et al., 2003; van Marrewijk et al., 2008; Williams, 2009). This term describes very large-scale projects with a number of common characteristics deemed to create exceptional challenges for those involved. These characteristics include high levels of complexity along various dimensions (Remington and Pollack, 2007), the potential for significant conflicts of interest between the wide variety of public and private sector stakeholders (Alderman et al., 2005; Clegg et al., 2002), and the need to make decisions and to act under conditions of uncertainty as well as risk (Atkinson et al., 2006; Loch et al., 2006). There is ample evidence that these challenges are proving somewhat intractable, leading in many cases to substantial cost overruns, delays in completion and failure to deliver against the objectives used to justify projects (see, for example, Atkinson, 1999; Flyvbjerg, 2009; Flyvbjerg et al., 2002, 2003; Miller and Lessard, 2000; NAO, 2006, 2009; Williams, 2009). This article considers how a range of different writers explain the significant performance problems exhibited by many megaprojects, and examines their arguments about what is necessary to achieve better performance. It does not therefore engage with the small, but growing critical management literature on projects (cf. Hodgson and Cicmil, 2006), which tends to focus more on questions of emancipation than on performativity (Fournier and Grey, 2000). Rather than review explanations individually, however, the aim is to provide a   5  broad categorization of different types by using the notion of organization theory paradigms as an organizing principle and multi-paradigm literature review as a methodology. Rather than simply providing the substantive content of a literature review, then, the article examines and illustrates the analytical value of adopting a multi-paradigm approach. Other writers have pursued similar objectives in their reviews of the broader project management literature (Pollack, 2007; Söderlund, 2011), but the focus of these works has primarily been on the descriptive clustering of various contributions into distinct schools of thought. There is an implicit acknowledgement of paradigmatic differences between schools in terms of their ‘interest in describing or prescribing, and whether the published papers are primarily based on inductive or deductive research approaches’ (Söderlund, 2011, p. 165), but the implications of these differences, in terms for example of the scope for paradigm crossing, have not been fully explored. More importantly for the argument developed in this paper, however, there is no explicit reflection on what might be missing from the project management literature. Instead, the different schools of thought are described in terms of their epistemology, research foci, methodologies and dominant ideas. The possibility of shared lacunae, conceptual and empirical blind spots, remains unexplored. It is here that this paper makes one of its major contributions. Perhaps the best known framework of organization theory paradigms (Burrell and Morgan, 1979) and the vigorous debate that it has stimulated are discussed in the next section. The strengths and limitations of multi-paradigm inquiry are also considered, and the review methodology adopted by this article, known as ‘paradigm interplay’ (Schultz and Hatch, 1996), is described and justified. Two broad types of explanation, functionalist and interpretivist, are identified and discussed in the subsequent section.   6  The key insight is that despite important differences in their epistemological orientation these two types of explanation are essentially ‘performative’ (Lyotard, 1984) in that their intention is to better describe the dynamics within megaprojects so that they can be controlled more efficiently and effectively. A key expression of this interest in performativity is a shared acceptance, albeit conceived in different ways, of the notion of actor farsightedness. It is concluded that this allows these explanations to emphasize governance in megaprojects as static, convergent and patterned forms of organization, what Chia and Holt (2009) refer to as made order, and to ignore governance as discontinuous, divergent and fluctuating micro-processes of organizing (Law, 1992). By revealing this explanatory gap, the article recognizes the value of adopting the multi-paradigm notion of a ‘stratified ontology’, in which reality is at once made and in the making (Lewis and Kelemen, 2002; Reed, 1997). A call is made to refocus project governance research to include proper consideration of the multiple processes of organizing through which actors use, reproduce and transform governance as made order. Paradigms in Organization Theory and Multi-Paradigm Review The Paradigm Debate The notion that researchers studying organizations operate within different paradigms has a relatively long pedigree, going back over thirty years to the work of Burrell and Morgan (1979). They defined paradigms as clearly delineated sets of ideological and epistemological assumptions that guide various modes of organizational analysis. Their framework, illustrated in Figure 1, employs two dimensions to identify four paradigms of organization research: functionalism, interpretivism, radical structuralism, and radical humanism.   13  they note that restricting their discussion to two paradigms helps them ‘to develop the interplay strategy in an explicit way’ (Schultz and Hatch, 1996, p. 531). For the same reasons, this article will also use these two paradigms in developing its review. This will, of course, restrict our attention to those parts of the megaproject governance literature with a regulatory or performativity focus. This is not to deny that there are researchers applying a radical change or critical management focus to their work on projects (see, for example, Bresnen, 1996; Gill, 2002; Hodgson, 2002; Buckle and Thomas, 2003; Hodgson, 2004; Thomas, 2006; Lindgren and Packendorff, 2006). Rather, the review presented here makes a conscious analytical choice to restrict its discussion to the functionalist paradigm, where the bulk of the literature on projects and megaproject governance is located, and the interpretivist paradigm, which has provided a growing number of more sociologically grounded accounts since the mid 1990s (Pollack, 2007; Söderlund, 2011). Also, functionalist research represents the author’s home paradigm and is included in keeping with a desire to be reflexive. The basic elements of paradigm interplay are illustrated in Figure 2. Three key contrasts are identified on the basis of the Burrell and Morgan (1979) framework. These are in terms of the analytical framework, the mode of analysis, and the analytical processes applied by a researcher operating in each paradigm.   14  Figure 2: Contrasts and Connections between Functionalism and Interpretivism  Functionalist research is carried out on the basis of a predefined and universal analytical framework. Interpretivist research, by contrast, is more about examining specific organizational contexts without prior theorization and allowing useful conceptual categories and constructs to emerge. Turning to the mode of analysis, functionalist research is targeted at gathering and categorising empirical data in terms of predefined variables, and then mapping the causal relationships between those variables to see if hypothesized relationships hold. Interpretivist research, by contrast, has a more associative mode of analysis, which explores the creation of meanings by actors in organizations and the ways in which those meanings can be associated and interpreted. Finally, functionalist research is about arriving at CONTRASTS CONNECTIONS Functionalism Interpretivism Predefined Categorical Convergent Emergent Associative Divergent Underlying Pattern Underlying Essence Static Representation Source: Adapted from Schultz and Hatch (1996)   15  an understanding of the underlying causal mechanisms at play by simplifying and clarifying complexity. In interpretivism, by contrast, the processes are about expanding and enriching the analysis by ‘constantly seeking more interpretations and making new associations’ (Schultz and Hatch, 1996, p. 539). The three connections identified in Figure 2, flow from the application of a postmodernist critique to these two modernist paradigms. Postmodernists assume that ‘human existence is fragmented and discontinuous’ (Schultz and Hatch, 1996, p. 540), and that there is therefore no underlying sense of pattern or order to be discovered by researchers. In modernist paradigms, by contrast, researchers are concerned with discovering underlying patterns in order to ascribe causal relationships (functionalism) or to bring meaning to complex organizational phenomena (interpretivism). Postmodernists also argue that research should not be about a search for some hidden underlying essence, but rather it should focus on the ‘superficial and the unexpected’ (Burrell, 1988, p. 225), because that is all there is. Research in the functionalist and interpretivist paradigms, by contrast, is about a search for the essence of organizational phenomena, ‘the underlying assumptions or meanings believed to order human experience’ (Schultz and Hatch, 1996, p. 541). Finally, postmodernism stresses the temporary, discontinuous and indeterminate character of organizational phenomena. It has an ontology of becoming rather than being, which focuses attention on processes as opposed to organizational structures and entities (Cooper and Burrell, 1988). Functionalist and interpretivist researchers, by contrast, tend to offer more or less static representations of organizational phenomena that ignore the flux and discontinuity of everyday organizational life. Where change processes are included in an analysis they are represented either as a relatively predictable movement from one static state   16  of being to another or as interrelated, cyclical and ordered processes of sense-making and meaning creation. Paradigm Interplay in the Literature on Megaproject Governance Text selection We now use the paradigm interplay technique to review a sample of texts focused on megaproject governance. In contrast to those papers reviewing much larger numbers of texts to provide an overview and categorization of the broader literature on project management (see, for example, Crawford et al., 2006; Pollack, 2007; Shenhar and Dvir, 2007; Söderlund, 2011), the objective here was simply to choose example texts addressing megaproject governance from within the functionalist and interpretivist paradigms to demonstrate the application and value of multi-paradigm literature review. The selection process was a version of snowball sampling, where an initial set of key data sources suggest and lead onto further data sources (Scarbrough et al., 2004). The snowball sampling process began with the comprehensive literature review paper by Söderlund (2011), which identifies seven schools of project management research. Three of these, the Governance, Contingency and Behaviour Schools, were identified as being particularly relevant based on their main research focus and their implicit epistemological orientation. All three of these schools are concerned with questions of how projects are organized, coordinated or governed. According to Söderlund (2011), papers within the Governance and Contingency Schools, favour the testing of hypotheses derived deductively from predefined theories, which suggests that they are functionalist in their epistemological orientation. Literature in the Governance School draws primarily on agency or transaction   17  cost theories (see, for example, Hart and Moore, 1999; Williamson, 1996), while the Contingency School is grounded in theories linking organizational design and structure to environmental and task characteristics (see, for example, Galbraith, 1973; Lawrence and Lorsch, 1967). Papers categorized within the Behaviour School, by contrast, are concerned with documenting the emergent processes within projects of ‘trust-building, problem-solving, sense-making and learning’ (Söderlund, 2011, p. 162). This suggests that they are broadly interpretivist in their approach to research. Based on the application of a number of search terms to titles, keywords and abstracts, four key articles were identified from amongst those categorized by Söderlund (2011) in the Governance, Contingency and Behaviour Schools. The search terms used were: ‘megaprojects’, ‘large-scale projects’, ‘major projects’, ‘complex projects’, ‘governance’, ‘risk’, ‘uncertainty’, and ‘performance’. Performance was included as a search term to focus attention on those texts with a regulatory bias and to screen out the critical management literature. The author names and reference lists of these four articles were then used to identify further relevant texts. This process of snowball sampling continued through two further iterations until it was decided, based on evidence of substantial cross-referencing and the logic of conceptual saturation (Guest et al., 2006), that a continued search was unlikely to generate further new leads. The final result of this sampling process is shown in Table 1, where the selected texts are categorised into three broad strands of research associated with the Governance, Contingency and Behaviour Schools respectively. Arguments and Conclusions1  1ThissectiondrawsonTheAuthor(2011).   18  Before moving to a detailed discussion of paradigm contrasts and connections, the core arguments and conclusions advanced by each strand of research are briefly described. Table 1 presents a summary. The first strand of functionalist research argues that performance is often disappointing, because non-viable projects are so regularly undertaken (Davidson and Huot, 1989; Flyvbjerg, 2009; Flyvbjerg et al., 2002, 2003, 2005; Wachs, 1989, 1990). It suggests that those actors with a vested interest in seeing projects undertaken, typically politicians and contractors, engage in strategic rent-seeking behaviour to get projects approved and to win associated contracts. This takes the form of systematically under-estimating project costs, over-estimating project benefits and being over-optimistic with project scheduling. These under and over-estimates are not seen as an honest mistake or a function of poor technical skills and inadequate data. Rather, they are attributed to straightforward ‘deception and lying as tactics aimed at getting projects started’ (Flyvbjerg et al., 2003, p. 47).   19  Table 1: Alternative Explanations of Megaproject Governance and Performance Functionalist Explanation 1 Weak Institutional Safeguards (Governance) Functionalist Explanation 2 Misaligned and Underdeveloped Governance (Contingency) Interpretivist Explanation Diverse Project Cultures and Rationalities (Behaviour) Arguments Conclusions Sample Texts Weak institutional safeguards and warped incentives mean that project promoters and contractors regularly engage in intentional rent-seeking behaviour (underestimating costs, over-estimating benefits) to get non-viable projects approved Better performance depends on legal requirement for thorough ex ante risk analysis and management plan; limit role of politicians to formulating and auditing public interest objectives; various ex ante measures to improve accountability of project decision-making Davidson and Huot 1989; Flyvbjerg 2009; Flyvbjerg et al. 2002, 2003, 2005; Wachs 1989, 1990 Performance problems result from misaligned or underdeveloped governance arrangements incapable of handling the emergent turbulence and opportunistic behaviour inevitably associated with me g a p ro j ects Better performance requires conscious design and creation at the front-end of the project of mechanisms that enhance ex post governability; mechanisms must be appropriate to the particular context of the project  De Meyer et al. 2002; Loch et al. 2006; Miller and Hobbs 2009; Miller and Lessard 2000; Morris 2009; Winch 2001, 2009; Winch et al. 2000 Projects are subject to processes of social construction and characterized by diverse and often competing cultures and rationalities – performance problems result from normal day-to-day management p ractice Good performance facilitated by ‘future perfect thinking’ and conscious design and creation at the front-end of the project of a shared culture supported by governance mechanisms to encourage collaborative and coordinated behaviour Alderman et al. 2005; Atkinson et al. 2006; Clegg et al. 2002, 2006; Pitsis et a l . 2003 ; van Marrewi j k et a l . 2008   20   These authors argue that such rent-seeking behaviour is encouraged, because the incentives to produce over-optimistic estimates of project viability are very strong and the disincentives relatively weak. Given the very lengthy time-frames that apply to megaproject development and implementation, there is a lack of proper accountability for project promoters, typically politicians, because they are often not in office when the actual viability of a project can be assessed. Getting a project approved will deliver significant political capital in the short-term, however. Similarly, it is argued that the accountability of contractors for their behaviour is weak, because the contractual penalties for producing over-optimistic tenders are often low compared to the potential profits involved (Davidson and Huot, 1989; Wachs, 1990). The key solution to project performance problems, then, is to create well defined policies and procedures and to embed them in institutional structures that strengthen the accountability of key project actors. This might include a legal requirement for a thorough ex ante risk analysis and management plan, limiting the role of politicians to formulating and auditing public interest objectives, and various ex ante measures to improve the accountability of project decision-making (see, for example, Flyvbjerg et al., 2003, pp. 107124). The second strand of functionalist research argues that the underperformance of many megaprojects is best explained by the presence of inappropriate or underdeveloped governance arrangements that are incapable of handling the turbulence inevitably associated with these endeavours (De Meyer et al., 2002; Loch et al., 2006; Miller and Hobbs, 2009; Miller and Lessard, 2000; Morris, 2009; Winch, 2009; Winch et al., 2000). The complexity, scope and scale, and the long time frames of megaprojects are seen as major reasons for the significant turbulence experienced in most cases. Turbulence is seen to originate either from   21  exogenous events in the broader macroeconomic, political, social and natural environments, or from endogenous events within and between the organizations directly involved in a project t. In the latter case, emphasis is placed on contractual disputes and the breakdown of partnerships or alliances. The principal set of solutions focuses, therefore, on the ex ante design of governance mechanisms. This is described as ‘planning for the journey rather than planning the journey’ (Miller and Lessard, 2000, p. 203). Broadly speaking, these governance mechanisms are intended to build stronger, more cooperative and more flexible relationships between project participants. Examples might include an alliance ownership structure, combining balanced equity positions with a strong leader; financial guarantees from government to support project financiers; rendezvous clauses making it possible to renegotiate contracts; integrated project teams with financial incentives to stimulate innovation; and multiple sources of finance to diversify dependencies (Loch et al., 2006; Miller and Floricel, 2000; Miller and Hobbs, 2009). There is also an explicit recognition that the governance mechanisms selected and designed must be appropriate to the particular context and characteristics of a project (De Meyer et al., 2002; Miller and Floricel, 2000; Winch, 2009). Finally, the interpretivist strand of research argues that megaprojects are characterized as a matter of course by multiple and diverse cultures and rationalities rather than by a singular, shared rationality as is assumed by more orthodox, functionalist perspectives (Atkinson et al., 2006; Clegg et al., 2002, 2006; Pitsis et al., 2003; van Marrewijk et al., 2008). This means that different actors understand inputs to and outputs from the project in very different, incomplete and often competing ways. For example, the contractual documents and other boundary objects used to define and coordinate the roles and responsibilities of the project actors are often highly ambiguous in meaning (Alderman et al.,   22  2005) and provide substantial scope for ‘language games’ leading to ‘contested action in complex, inter-organizational and professional disputes’ (van Marrewijk et al., 2008, p. 592). The performance problems of many megaprojects are an almost inevitable result, then, of the normal day-to-day practice of managers trying to cope with an organizational environment that is complex, ambiguous and often highly conflictual. The focus here is said to be contextually grounded, looking in detail at actual practice within project organizations. As van Marrewijk et al. (2008, p. 592) comment, ‘[t]his approach recognises that project environments are subject to processes of social construction, in which participants construct a more or less stable working environment for themselves’. It is concluded that where project participants are able to construct a relatively stable environment, which promotes peaceful cooperation between differing cultures, there is a greater chance of good project performance. An example of one such stable working environment is identified in research on a project to build a 20 kilometre long tunnel under the area north of Sydney Harbour in the run up to the Olympic Games in 2000 (Clegg et al., 2002, 2006; Pitsis et al., 2003; van Marrewijk et al., 2008). This research links the broadly successful delivery of the project to the decision at the beginning of the process to create ‘a project culture that was explicitly designed and crafted to encourage shared behaviours, decision-making, and values’ (Pitsis et al., 2003, p. 576). A number of governance mechanisms were used to underpin this project culture, including a formal statement of key values as a basis for resolving disputes internally and a risk/reward regime based on monetized key performance indictors (KPIs). The institutional element of project governance – the Project Alliance Leadership Team (PALT) – was explicitly created to be legally and spatially separate from the four parent organizations. The intention was ‘to produce a designer culture for the project rather than have it as an arena   29  to be able to use beliefs or expectations grounded in historical practice to estimate their likelihood (Knight, 1921; Samset, 2009). It is on the basis of such estimates, that the ex ante design of governance mechanisms relies. This is not deny that some contributors to these functionalist explanations do recognize the existence of some unforeseeable uncertainty, where past experience does not exist or is not a good guide to the future (see, for example, De Meyer et al., 2002; Loch et al., 2006). The key point is that the conceptions of project governance promoted by these explanations are grounded primarily in assumptions that what we can know about the future is characterized either by risk or by foreseeable uncertainty. The interpretivist explanation, by contrast, suggests that the successful completion of the project in the specific case considered was only partly a function of a consciously predesigned culture intended to encourage collaborative working. A significant part of the explanation focuses on the notion of future perfect strategy, which emerged as the research progressed and the case data were interpreted. Future perfect strategy was understood by the researchers as being significantly different to the kind of scenario planning techniques often associated with projects facing high degrees of ambiguity and uncertainty. It was not based on the use of past experience to identify and plan for the achievement of possible future scenarios in the project. Rather, the strategy ‘comprised imagining a future and then seeking to realize it, subject to constant revision’ (Pitsis et al., 2003, p. 576). This suggests that, in the context of this project at least, the interpretivist explanation conceives of knowledge about the future as being significantly characterized by unforeseeable uncertainty, which is in tune with the socially constructed nature of knowledge in this paradigm. The future is seen as a created process shaped by the nature and pattern of decisions made now and in the future (Froud, 2003; Minsky, 1996). The interpretivist explanation does recognize that governance mechanisms can be pre-designed, as the designer   30  culture in the case demonstrates, and that this process of institutionalization might shape the future of the project in a broad sense. The emphasis is much more, however, on the need to continuously revise activities as the project emerges in unforeseeable ways over time. Connections between Functionalist and Interpretivist Explanations We turn now to the other component of paradigm interplay, to discuss what is revealed by an examination of connections between these different types of explanation. As Schultz and Hatch (1996) did, we use a postmodernist style of thinking to establish and examine these connections. It should be emphasized that the author is not a postmodernist, but is pragmatically assuming this perspective in line with the reflexive philosophy of multiparadigm review. Connections are identified on the basis that both functionalist and interpretivist explanations are concerned with identifying an ordered pattern and an underlying essence in the failures and successes of megaprojects. We also identify a connection based on the common use of static representations of project governance. The core argument put forward on the basis of these connections is that there is an acceptance within both paradigms of the notion of actor farsightedness. Table 3 summarizes the key aspects of the discussion. Table 3: Connections between Functionalist and Interpretivist Explanations   31  Connecting Assumptions Functionalist Explanations Interpretivist Explanation Underlying Pattern Project failure explained by an underlying pattern of weak or misaligned governance mechanisms/institutions, and success explained by a reverse pattern Failure associated with conflict between different, but coherent cultures/rationalities within the project domain, and success associated with an integrative culture promoting collaborative working Underlying Essence Actor farsightedness based on a future characterized largely by risk and knowable uncertainty Actor farsightedness in the project domain, with certainty created in situ on an iterative basis Static Representation Governance as predesigned institutions, mechanisms, contracts, coalitions, alliances, relationships, and risk management plans with inherent flexibility Governance as designer culture, project alliance leadership team, statements of vision and values, KPIs, and future perfect strategy based on coherent stepwise revision Framework adapted from Schultz and Hatch (1996)  The suggestion that both types of explanation have a modernist concern with identifying ordered and coherent patterns is evidenced by the way they discuss the reasons for megaproject success and failure. Functionalist explanation 1 sees project failure as a function of coherent and strategically organized ex ante opportunism facilitated by weak institutional/contractual safeguards. The theoretical explanation or grand narrative (Lyotard, 1984) of failure is one of regular and systematic cheating by project sponsors and other key actors with an interest in seeing projects approved. The narrative of improvement, a call to   32  design better institutional safeguards, is similarly coherent and ordered. Functionalist explanation 2 provides an equally patterned picture of project failure, seeing it as a function of poor or misaligned governance. The grand narrative suggests that project actors make poor governance choices, but that they can learn to do better. Finally, the interpretivist explanation sees failure as a function of the inherent difficulties posed by the social construction of projects, and the almost inevitable conflict that occurs between the different cultures and rationalities that emerge from that social construction. At first sight, this explanation appears much less concerned with order and coherence given its emphasis on conflict as a normal part of organizational life. Even here, though, there is a search for pattern. A project a may contain different and competing cultures, but each is seen as coherent. Conflict may occur, but the rules of engagement and resolution are ordered and coherent. Power resources or an appeal to collaborative principles and values will settle the matter (Clegg et al., 2002; van Marrewijk et al., 2008). Looking at the way these explanations discuss megaproject failure and success, we can also identify a modernist interest in treating observable phenomena as a reflection of a deeper underlying essence. In functionalist explanation 1, for example, project failure is not simply about the observable existence of weak institutional safeguards, which facilitate opportunistic behaviour. It is also about that deeper characteristic of the human condition that enables actors to undertake regular and systematic cheating. This explanation suggests that project sponsors and others are able to calculate the likely future path of a project, based on data from previous similar projects, and therefore could produce an appropriate and comprehensive plan of action and associated governance structures and contractual documents to ensure a successful outcome. Knowledge about the future is primarily characterized by risk and foreseeable uncertainty. The reason for failure, therefore, is that   33  these key actors consciously choose not to use this ability and instead opportunistically misrepresent the future path of the project, in terms of costs, benefits, and schedule, to serve their own interests. The underlying essence for this explanation, then, is actor farsightedness. Evidence of an appeal to the same underlying essence can be seen in the second functionalist explanation. As in the first, knowledge about the future is characterized primarily by risk and foreseeable uncertainty. This suggests that actors should be able to look ahead and discern potential sources of turbulence and design appropriate governance mechanisms to deal with them. Project failure, it is argued, is not simply about the observable existence of misaligned or inappropriate governance arrangements, which cannot cope with the various sources of turbulence that emerge over time. It is also about the fact that governance design is a management choice, which can be undertaken well or badly. When it is done badly, and the project fails, this is taken as a failure of actor farsightedness. Project success, which is explained by the observable existence of aligned and appropriate governance, is essentially therefore a function of farsightedness. Perhaps more surprisingly there is also evidence of an appeal to actor farsightedness in the interpretivist explanation, which might be expected to reject this underlying essence given its assumption that knowledge about the future is socially constructed and therefore significantly characterized by unforeseeable uncertainty. What we see here, however, is that actor farsightedness remains essential to this explanation’s grand narrative, albeit differently conceived. At the observable level, the failure of projects is seen as a function of conflict between different cultures and rationalities. Project success, by extension, is linked to the creation of a mechanism by one or more actors that mediates between these competing cultures and encourages (positive) collaboration instead of (negative) conflict. Looking beyond the observable level of explanation, we suggest that the very existence of such a   34  mechanism requires an essence of actor farsightedness. Rather than being conceived in a temporal sense, however, farsightedness is conceived here as being able to look comprehensively across the project domain in order to recognize and understand the range of different cultures and rationalities. Such recognition and understanding would seem to be the prerequisites of creating an effective integrative governance mechanism. The explanation of future perfect strategy appeals to the same essence of actor farsightedness in the project domain. Imagining a future outcome and then seeking to realize it, subject to constant revision, suggests an appreciation, continually updated, of what is and what is not happening within the project domain. The final modernist connection between these functionalist and interpretivist explanations is their reliance on static representations of governance. Governance is represented largely in terms of structural characteristics of coherence, pattern and order, seeing it as a form of organization. The idea of governance as dynamic and potentially disordered processes of organizing is downplayed or completely ignored. So, for example, the functionalist explanations talk about governance in terms of mechanisms, institutions, relationships, alliances, contracts, coalitions, risk management plans and so on. There is a strong emphasis on the notion of governance design and an assumption that much, if not all, of this design work can be, and indeed should be, done at the front-end of the project before turbulence and opportunism arise (Flyvbjerg et al., 2003; Loch et al., 2006; Miller and Lessard, 2000; Williams et al., 2009). The possibility of changes or discontinuity in project governance is downplayed by references to flexibility by design or infusing governability (Miller and Hobbs, 2009). This ex ante governance design relies heavily, of course, on the underlying essence of actor farsightedness.   35  The interpretivist explanation also represents governance in largely static terms, with its references to a coherent designer culture embodied in the project alliance leadership team (PALT), a list of value statements, and monetized key performance indicators (Clegg et al., 2002; Pitsis et al., 2003). As a potential counterpoint to this, the researchers’ description of future perfect strategy as a ‘process [that] comprised imagining a future and then seeking to realize it, subject to constant revision’ (Pitsis et al., 2003, p. 576) does suggest some interest in representing project governance in processual terms. On closer inspection, however, it becomes clear that this process is being described in essentially static terms, as an orderly and coherent series of steps towards the imagined end goal. Problems emerge, discussions happen, a revising step takes place. As Schultz and Hatch (1996, p. 543) observe, interpretivists evince a strong interest in processes of sense making and meaning creation within organizations, but, as in this case, they rarely examine ‘the ruptures, discontinuity, and fragmentation’ involved. Chia (1995, p. 587) makes very much the same point when he suggests that ‘when modernists talk about process, they usually mean the various stages/states of isolatable events/conditions which lead towards an achievement or outcome’ [emphasis in the original]. Again, this coherent and ordered representation relies on an essence of actor farsightedness in the project domain. Conclusion: Seeing Beyond Governance as Made Order Through a literature review based on paradigm interplay (Schultz and Hatch, 1996), this article has discussed a number of important contrasts and connections between two bodies of megaproject governance research, one functionalist and one interpretivist. The use of paradigm interplay as a review technique enables significant reflexivity through a   36  combination of paradigm bracketing and bridging. Bracketing is about considering the implications of different epistemological assumptions for the kinds of understanding which are made possible by different paradigms. In this way, paradigms are seen as partial sensemaking heuristics, focusing our attention on certain facets of complex organizational or social phenomena and obscuring others (Lewis and Kelemen, 2002). Bridging challenges the notion of incommensurability and seeks out the assumptions, concepts, and constructs that paradigms have in common (Lewis and Grimes, 1999). Paradigm interplay applies a postmodern style of thinking (Chia, 1995) to identify these connections, but retains an acceptance of the value of paradigm distinctions. The aim is to uncover which facets of a phenomenon are ignored or not adequately explained by either paradigm as a consequence of these shared assumptions and constructs. The key common assumption identified by the literature review is that actors engaged in megaprojects are farsighted. In a broad sense, this means that all three explanations accept that actors can and should prepare for the future before it has happened. Having said that, conceptions of what it means to be farsighted do differ between the functionalist and interpretivist explanations. This is unsurprising given their different epistemological assumptions about what can be known about the nature of the future. Farsightedness for the functionalist explanations is conceived in a temporal sense. So, functionalist explanation 1, which assumes that future events and project outcomes are primarily risky and therefore susceptible to detailed analysis and contingency planning, argues that actors can and should undertake comprehensive ex ante contracting and thereby control the future. Functionalist explanation 2 assumes that project outcomes are in most cases foreseeably uncertain. This means that decision-makers lack the necessary reference class data to undertake a calculation of statistical probability, but experience and judgement   37  still allow actors to be farsighted enough to know the range of possible future events, to rank them based on subjective probability, and therefore to prepare appropriate governance arrangements ex ante to manage those events ex post. The interpretivist explanation appears to present an interesting paradox. On one hand it assumes that projects take place in an environment of unforeseeable uncertainty – knowledge about future outcomes is socially constructed and therefore unknowable in the present. Farsightedness thus appears to be ruled out. On the other it suggests that ex post problems can be more effectively addressed if a collaborative project culture is consciously designed ex ante, and future perfect thinking is used to imagine project outcomes which are then achieved through stepwise revision of activities. This seeming paradox is resolved, then, by an acceptance of actor farsightedness, but conceived in this case as project domain farsightedness. Given this shared assumption, all three of these explanations are able to conceive governance in megaprojects as static, convergent and patterned forms of organization, what Chia and Holt (2009) refer to as made order, and to ignore governance as discontinuous, divergent and fluctuating micro-processes of organizing (Law 1992). The argument here is that an assumption of farsightedness provides a convenient conceptual short-cut, which bypasses a serious consideration of the perhaps haphazard, transient and accidental processes through which actors use, reproduce and transform this made order on an on-going basis. There is, of course, recognition that actors might consciously choose to renegotiate the terms of governance during a project, but this simply represents the replacement of one made order with another.   38  The core conclusion, then, is that future research must give greater attention to governance as micro-processes of organizing within projects. Structural forms of governance dominant the functionalist explanations reviewed here. The interpretivist explanation evinces a concern with ‘everyday organizational life’ (Pitsis et al., 2003, p. 588), but presents an ordered, coherent and static picture of that life. Both types of explanation operate on the basis of a modernist ontology. This is not a call, however, simply to replace a modernist ontology of being with a postmodern ontology of becoming (Cooper and Law, 1995). As Linehan and Kavanagh (2006, p. 52) insightfully comment, ‘each ontology is necessarily partial…an excessive attachment to one or the other leads to a privileging of some questions, methods and interventions, and a marginalisation of others.’ Rather, this article supports the accommodating philosophy of multi-paradigm inquiry, valuing alternative paradigm lenses for their capacity to ‘reveal seemingly disparate, but interdependent facets of complex phenomena’ (Lewis and Kelemen, 2002, p. 258). Söderlund (2011, p. 168) reaches a similar conclusion, arguing that the pluralism of project management research needs to be embraced ‘to illuminate the complex actuality of projects and project management practice.’ Similarly, Pollack (2007, p. 272) concludes that a ‘wider variety of paradigms employed within the field increases the ways in which existing techniques are understood.’ These papers seem to assume, however, that each piece of the jigsaw is already available and that they need only to be properly assembled to provide a holistic understanding of projects. The key difference and contribution here is that this paper has used paradigm interplay to reveal what is, crucially, left unsaid by the literature on megaproject governance.   45  & C. Midler (Eds.), Projects as Arenas for Renewal and Learning Processes. Boston, MA: Kluwer Academic Publishers. Lyotard, J-F. (1984). The Postmodern Condition: A Report on Knowledge. St Paul, MN: University of Minnesota Press. Midler, C. (1995). ‘Projectification’ of the firm: the Renault case. Scandinavian Journal of Management, 11, 363-373. Miller, R. and Floricel, S. (2000). Building governability into project structures. In R. Miller, & D.R. Lessard (Eds.), The Strategic Management of Large Engineering Projects: Shaping Institutions, Risks and Governance (pp. 131-150). Cambridge MA: MIT Press. Miller, R. and Lessard, D.R. (2000). Public goods and private strategies: making sense of project performance. In R. Miller, & D.R. Lessard (Eds.), The Strategic Management of Large Engineering Projects: Shaping Institutions, Risks and Governance (pp. 19-49). Cambridge MA: MIT Press. Mingers, J. (1997). Multi-paradigm methodology. In J. Mingers, & A. Gills (Eds.), Multimethodology. New York: Wiley. Minsky, H.P. (1996). Uncertainty and the institutional structure of capitalist economies. Journal of Economic Issues, 30, 357-368. Morris, P.W.G. (1994). The Management of Projects. London: Thomas Telford. Morris, P.W.G. (2009). Implementing strategy through project management: the importance of managing the project front-end. In T.M. Williams, K. Samset, & K.J. Sunnevåg (Eds.), Making Essential Choices with Scant Information: Front-End Decision Making in Major Projects (pp. 39-67). Basingstoke: Palgrave Macmillan.   46  NAO (2006). Ministry of Defence: Major Projects Report. London: HMSO. NAO (2009). Commercial Skills for Complex Government Projects. London: HMSO. Parker, M. and McHugh, G. (1991). Five texts in search of an author: a response to John Hassard’s multiple paradigms and organizational analysis. Organization Studies, 12, 451-456. Pfeffer, J. (1993). Barriers to the advance of organizational science: paradigm development as a dependent variable. Academy of Management Review, 18, 599-620. Pfeffer, J. (1997). Mortality, reproducibility, and the persistence of styles of theory. Organization Science, 8, 681-686. Pitsis, T.S., Clegg, S.R., Marosszeky, M. and Rura-Polley, T. (2003). Constructing the Olympic dream: a future perfect strategy of project management. Organization Science, 14, 574-590. Pollack, J. (2007). The changing paradigms of project management. International Journal of Project Management, 25, 266-274. Reed, M.I. (1985). Redirections in Organizational Analysis. London: Tavistock. Reed, M.I. (1992). The Sociology of Organizations: Themes, Perspectives and Prospects. New York: Harvester Wheatsheaf. Reed, M.I. (1997). In praise of duality and dualism: rethinking agency and structure in organizational analysis. Organization Studies, 18, 21-42. Remington, K. and Pollack, J. (2007). Tools for Complex Projects. Aldershot: Gower.   47  Scarbrough , H., Swan, J., Laurent, S., Bresnen, M., Edelman, L. and Newell, S. (2004). Project-based learning and the role of learning boundaries. Organization Studies, 25, 1579 – 1600. Scherer, A.G. (1998). Pluralism and incommensurability in strategic management and organization theory: a problem in search of a solution. Organization, 5, 147-169. Schultz, M. and Hatch, M.J. (1996). Living with multiple paradigms: the case of paradigm interplay in organizational culture studies. Academy of Management Review, 21, 529557. Shenhar, A. and Dvir, D. (2007). Project management research: the challenge and opportunity. Project Management Journal, 38, 93-99. Silver, C.A. (2000). Where technology and knowledge meet. Journal of Business Strategy, 21, p. 28. Söderlund, J. (2011). Pluralism in project management: navigating the crossroads of specialization and fragmentation. International Journal of Management Reviews, 13, 153-176. The Author (2011). Risk, uncertainty and governance in megaprojects: a critical discussion of alternative explanations. International Journal of Project Management, doi: 10.1016/ j.ijproman.2011.11.002. Thomas, J. (2006). Problematizing project management. In D. Hodgson, & S. Cicmil (Eds.), Making Projects Critical (pp. 90-107). Basingstoke: Palgrave Macmillan.   48  van Marrewijk, A., Clegg, S.R., Pitsis, T.S. and Veenswijk, M. (2008). Managing publicprivate megaprojects: paradoxes, complexity and project design. International Journal of Project Management, 26, 591-600. Wachs, M. (1989). When planners lie with numbers. Journal of the American Planning Association, 55, 476-479. Wachs, M. (1990). Ethics and advocacy in forecasting for public policy. Business and Professional Ethics Journal, 9, 141-157. Weaver, G.R. and Gioia, D. (1994). Paradigms lost: incommensurability vs. structurationist inquiry. Organization Studies, 15, 565-590. Weick, K.E. (1999). Theory construction as disciplined reflexivity: tradeoffs of the 90s. Academy of Management Review, 24, 797-807. Williams, T.M. (2002). Modelling Complex Projects. London: Wiley. Williams, T.M. (2009). Decisions made on scant information: overview. In T.M. Williams, K. Samset, & K.J. Sunnevåg (Eds.), Making Essential Choices with Scant Information: Front-End Decision Making in Major Projects (pp. 3-17). Basingstoke: Palgrave Macmillan. Williams, T.M., Samset, K. and Sunnevåg, K.J. (Eds.) (2009). Making Essential Choices with Scant Information: Front-End Decision Making in Major Projects. Basingstoke: Palgrave Macmillan. Williamson, O.E. (1985). The Economic Institutions of Capitalism. New York: The Free Press.   49  Williamson, O.E. (1996). The Mechanisms of Governance. Oxford: Oxford University Press. Willmott, H. (1993). Breaking the paradigm mentality. Organization Studies, 14, 681-719. Winch, G.M. (2001). Governing the project process: a conceptual framework. Construction Management and Economics, 19, 799-808. Winch, G.M. (2009). Managing Construction Projects, 2nd edn. Oxford: Wiley-Blackwell. Winch, G.M., Clifton, N. and Millar, C. (2000). Organization and management in an AngloFrench consortium: the case of the Transmanche link’, Journal of Management Studies, 37, 663-685.  Birmingham Business School University House Edgbaston Park Road Birmingham B15 2TY, United Kingdom www.birmingham.ac.uk/business