Clash of the Titans: Temporal Organizing and Collaborative Dynamics in the Panama Canal Megaproject

Introduction

Large-scale, global projects require inter-organizational collaboration (Jones & Lichtenstein, 2008) across national, cultural and political boundaries (Orr & Scott, 2008) between public, private and third sector organizations (Ainamo et al., 2010) over a finite period. Global projects, defined as temporary endeavours where multiple actors seek to optimize outcomes by combining resources from multiple sites, organizations, cultures, and geographies through a combination of contractual, hierarchical, and network-based modes of organization (Scott, Levitt & Orr, 2011, p. 17), potentially constitute highly unstable and complex, potentially conflict-ridden contexts for collaboration. Expert employees of diverse permanent organizations are assembled to carry out a specific project; assembly is project specific, typically with a limited history of working together and limited prospects of collaborating in the future (Meyerson, Weick & Kramer, 1996). Usually, roles are not highly prescribed or formally fixed but rather ambiguous (Morrison, 1994). Viewed from the outside, the long timelines, sophisticated organizational arrangements and sheer size of global projects may give the appearance of a relatively stable, almost permanent organization. From within, however, a large-scale project appears as a fleeting constellation of multiple, interrelated subprojects and several stakeholders collaborating in shifting alliances (Van Marrewijk, Clegg, Pitsis & Veenswijk, 2008).

Potentially, large-scale global projects are also conflict-ridden contexts for project partners, because they involve a geographically dispersed multiplicity of stakeholders, often with conflicting interests, working across ‘institutional differences’ between project partners; i.e. differences in regulations, political systems and culture (Scott et al., 2011). For example, in the Apollo project there was a difficult interface between the multiple permanent organizations assembled for the project (Wilemon, 1973). Differences in national, organizational, professional and project cultures and concomitant loyalties and interests influence the success of such projects. Performance and collaboration between project partners is often highly problematic (Kramer, 2009). Seeking to address the many interests that are at stake through contractual arrangements and strict governance regimes (Miller & Hobbs, 2005) designed to ensure consistent and predictable delivery cannot fully capture or remove the complexity of organizational collaboration in large-scale, global projects (Van Marrewijk et al., 2008). Due to global projects’ large budgets, high public profile and strong and lasting impact on their environment and society, project partners’ collaborative relationships are under constant pressure.

Under such circumstances, project partners may be motivated to overcome differences, to clearly define roles, responsibilities and hierarchical relations and to establish firm relationships across institutional divides. Orr and Scott (2008) suggest that project partners resolve differences, going through phases of ignorance, sensemaking and ‘response’. In a similar vein, Clegg, Pitsis, Rura-Polley and Marosszeky (2002) described how project partners invested much effort in socialization at the initial stage of the project to develop an ‘alliance culture’. The employees subsequently engaged in a number of cross-boundary coordination practices that made their work visible and legible to each other, such as switching advocacy roles for various elements of the project so that members of the leadership team became advocates for areas in which they had no expertise: such strategies enabled ongoing revision and alignment of leaders’ views of others’ roles and identities (Clegg et al., 2002). In another example freelance expatriates mitigated differences between client organization and international contracting firms through role reallocation, education and translation (Mahalingam, Levitt & Scott, 2011). Frequently, principals hire agents to manage and guard the project execution and objectives (Turner & Keegan, 2001).

These studies confirm findings from research into project work and temporary organizations more broadly. Drawing on a study of film projects, Bechky claims that temporary organizations are not ephemeral and unstable but instead manage to maintain continuity across different projects by relying on structured role systems perpetuated through ‘practices of enthusiastic thanking, polite admonishing and role-oriented joking’ (Bechky, 2006, p. 3). Other research on filmmaking and theatre productions also shows how employees take up distinct roles to perform flexible tasks under time pressure (Bechky, 2006; DeFillippi & Arthur, 1998; Goodman & Goodman, 1976), attempting to create permanence in temporary project work, using social mechanisms creating local networks through socialization, reciprocity and reputation (Jones, Hesterly & Borgatti, 1997).

Existing studies cast light on practices that establish, secure or restore stability in temporary organizations. In doing so, organizational actors are primarily seen as seeking to forge order, continuity and consensus out of chaos. Members of temporary organizations are not necessarily consensus-seekers, however. Complexity, ambiguity and uncertainty do not merely constitute exogenous conditions for collaboration but can drive collaboration in temporary organizations. An intricate web of shifting relationships and divergent interests between partners in large-scale global projects may give rise to disagreement, discord and power struggles between project partners (Clegg & Kreiner, 2013; Flyvbjerg, Bruzelius & Rothengatter, 2003; Mahalingam et al., 2011), a view gaining in traction, albeit substantively under-researched and under-theorized (e.g. Ivory & Alderman, 2015). In this paper we draw on a conflict view as well as order view (Burrell & Morgan, 1979) by zooming in on political struggles over control and hierarchy in a large-scale global project.

The discussion above prepares a central research question: How do project members negotiate their roles, responsibilities and hierarchical relations in the collaboration between principal and agent in a large-scale global project? Answering this question, we use data from a one-year ethnographic study of the project organization of the Panama Canal Expansion Program (PCEP). The operator, Autoridad del Canal de Panamá (ACP), initiated the PCEP in 2006 to expand and modernize the Panama Canal with an estimated budget of $5.25 billion. The ACP, the largest employer in Panama, with 10,000 employees, hired the US-based CH2M Hill (from here on: CH), a global leader in programme management, with 26,000 employees. These local and global Titans jointly managed the PCEP execution to guarantee delivery within budget, scope and before the targeted completion date. The study demonstrates that instead of the usual hierarchical positioning between the PM (CH) as principal and ACP as agent (Turner & Keegan, 2001), a temporary and unexpected role of coach and mentor was assigned to CH. The emergent ‘order’ appeared to be a profoundly diffuse hierarchy giving rise to ACP and CH constantly renegotiating roles, responsibilities and relations. We distinguished three different faces of the collaborative process, each offering a conflicting view of the roles, responsibilities and relations of ACP and CH in the project: (1) the Titans’ attempts to establish a collaborative order by seeking common ground, engaging in harmony-seeking practices and optimistic, egalitarian talk of growing trust and emerging ‘love’ in the ‘marriage’ between ACP and CH; (2) CH consultants’ contestation of emerging roles and hierarchic relations between CH and ACP, claiming the project was a break with established traditions in project work that cast CH as agent in full control of the project; (3) ACP managers’ mirroring response as they complained about not being in control, contesting CH’s dominant position by claiming a temporal schism with their tradition of leading canal projects and operations.

The paper’s contribution is not to show how order is being maintained in the potentially disorderly world of temporary organizations but instead to show how project leaders constantly renegotiate the emergent ‘order’. Existing research into temporary organizations shows a bias towards studying the establishment and maintenance of consensus and continuity through resolving institutional differences (Scott, et al., 2011), forming alliance cultures (Clegg et al., 2002), constructing an alluring prospect in a future perfect (Pitsis, Clegg, Marosszeky & Rura-Polley, 2003), or through practices of thanking, admonishing and joking (Bechky, 2006). Exactly how conflicts and negotiations over role structures between project members are differently interpreted by and fought out between project partners is missing. Project partners do engage in practices of maintenance and conflict resolution, such as through talk of ‘marriage’ and ‘love’, yet, our findings indicate that such consensus-seeking discourse does not fully resolve institutional differences, nor does it create clarity or permanence in the formal roles and hierarchic relations.

The case we present suggests, in contrast to Grabher (2002) and Bechky (2006), that a temporary organization may constitute a context where order and ‘permanence’ are not self-evident, particularly when collaborating actors design an unusual and diffuse hierarchy. Such an unstable context may gradually build up tensions through struggles over emerging roles and hierarchic relations, while hollowing out mutual trust, finally escalating time delays, budget overrun or scope changes.

The paper starts with a discussion of views of ‘order’ and ‘conflict’ underpinning research into temporary organizing. The presumed precariousness of temporary organizations seems to have led organizational scholars to orientate their research towards explaining how members of temporary organizations establish and maintain clear role structures and harmonious relations in the face of precariousness (thus adopting an ‘order view’) rather than focusing on conflict-ridden negotiations and power struggles (a ‘conflict view’). We then explore the idea of conflict and hierarchy in large-scale global projects in infrastructure. After explaining our research methodology, the research findings are presented in detail, followed by discussion of the implications for theorizing and studying order and conflict in temporary organizations, and organizing in large-scale global projects specifically.

Order and Conflict in Temporary Organizing

Temporary organizations are established to deliver some specific task, to achieve a particular goal, or to organize an event or accomplish a project in a specific time with a clear ending, after which they cease to exist (Grabher, 2002). As a specific form of organizing they have become an object of theoretical reflection and debate. Explorations of temporary organizing’s deviations from more enduring forms of organizing address a fundamental issue in the theorization of organizations and organizing, i.e. the endurance of organizational practices as they form and reform in different spaces and times (Bakker, 2010; Lundin & Hällgren, 2014; Lundin & Söderholm, 1995). Temporary organizing offers excellent opportunities to study how organizational actors constitute practices and processes of organizing in their everyday working lives, securing the transfer of experience and expertise and establishing and maintaining stable collaborations (an order view) as well as how they innovate or challenge existing roles, routines and prescriptions in everyday work (a conflict view). Indeed, a growing interest in research on temporary organizations has been apparent over the past few decades (Bakker, 2010; Bechky, 2006; Grabher, 2002, 2004; Kenis, Janowicz-Panjaitan & Cambre, 2009; Lundin & Hällgren, 2014; Orlikowski & Yates, 2002; Sydow, Lindkvist & DeFillippi, 2004). The growing interest has resulted in a diverse body of academic studies, including a focus on temporality in theatrical production (Goodman & Goodman, 1976), film and television production (Bechky, 2006; DeFillippi & Arthur, 1998), engineering projects (Wilemon, 1973) as well as project management (Lundin & Söderholm, 1995; Lundin & Steinthórsson, 2003; Winch, 2014), sharing a theoretical interest in how members of temporary organizations organize collaborative relations.

The literature on temporary organizations tends to view organizing as the process through which people establish and maintain collaborative relations (Bechky, 2006). The assumption is that a temporary organization constitutes an inherently and extremely transient context which motivates members to create and organize some measure of collaborative order and permanence. Temporariness thus prompts researchers to ask how organizational actors produce and perpetuate collaborative roles and relations across time and space. While insightful in itself, this research privileges order and permanence at the expense of offering an understanding of ongoing negotiations and transient relations, adopting an ‘order view’ rather than a ‘conflict view’, primarily concerned with ‘explaining the status quo’ (Burrell & Morgan, 1979, p. 14). Viewed from a conflict perspective, organizing is instead a process infused with overt or covert power struggles in which people attempt to impose and sustain or resist and overthrow an emerging order. Collaborative arrangements and hierarchic positions are then seen as potentially contested, something to fight for and to fight over, a concrete stake in negotiations between actors with conflicting interests (e.g. Mahalingam et al., 2011). What is claimed as usual or unusual, acceptable or unacceptable in project roles and relations figure as symbolic resources for struggles and contestation between collaborating stakeholders.

Such contestation and negotiation remains under-explored in studies of temporary organizations (some notable exceptions are Georg & Tryggestad, 2009; Kramer, 2009; Van Marrewijk, 2015). Yet, trading in an order view for a conflict view would be unwise. While opening up a new field of vision for studying temporary organizing, it would be no less myopic, constraining analysis within one view, excluding the possibility of analysis transcending the limitations of a single perspective (Willmott, 1990). Building on the assumption that conflict and order may exist ‘in tandem’ (Young, 1989, p. 188), temporary organizing can be viewed as the process through which organizational actors establish and maintain, as well as challenge or change, collaborative roles and relations. Consequently, scholars have questioned the differentiating characteristics separating ‘temporary’ from ‘ordinary’ organizations (Lundin & Hällgren, 2014; Lundin & Söderholm, 1995; Lundin & Steinthórsson, 2003), suggesting that the permanent and temporary may well coexist in projects (Müller-Seitz & Sydow, 2011). We will analyse how project members establish and maintain order while contesting and struggling over emerging collaborative practices negotiating roles, responsibilities and relations in the principal–agent collaboration.

Collaborating in Large-Scale Global Projects in Infrastructure Construction

Combining a conflict with an order lens is particularly relevant for studying complex social dynamics of collaboration in large-scale global construction projects, where there is considerable pressure on stakeholders to establish a working consensus and workable relations, as well as considerable potential for politicking and negotiating between them. Large-scale global projects in infrastructure construction constitute a context that is complex, uncertain and ambiguous (Van Marrewijk et al., 2008), because, first, a culture of temporariness (Jones & Lichtenstein, 2008). Although large-scale projects tend to have a long time frame, they comprise many sub-projects that deploy specific capabilities and set intermediate goals with related sub-project endings. The limited time for task execution and the frequent and intentional dismantling of project teams at pre-set times produces the core of temporary organizing (Söderlund, 2004). Second, a large number of partners, interest groups, supporting and opposing citizens as well as multiple other stakeholders participate in the project. In construction work, project work also often involves establishing relations between permanent and temporary organizations (Lundin & Hällgren, 2014). Usually, a permanent organization organizes and assigns large-scale construction work to a temporary project organization and hires a host of permanent organizations to accomplish particular tasks. Internal life in temporary organizations is thus inextricably interwoven with permanent organizational provision of key resources of expertise, reputation and legitimation (Grabher, 2004).

Given fleeting tasks, relations and partners, the complexity of relations between multiple stakeholders, and lack of clarity and agreement concerning project goals and their achievement, collaboration between project partners in global infrastructure construction projects is critical, difficult and laborious (Williams, 2002), frequently resulting in underperformance or failure (e.g. Flyvbjerg et al., 2003; Scott et al., 2011). Oddly, however, the few studies that detail actors’ day-to-day organizing and lived experience of such temporary, inter-organizational forms of collaboration tend to focus on successful projects (Van Marrewijk, 2015). Collaboration between different groups in project work is often taken as a natural part of the working process and, insofar as it is a source of difficulties it is viewed positively, as learning episodes for participants, related industries and occupational communities (Grabher, 2002; Lundin & Hällgren, 2014; Lundin & Söderholm, 1995; Winch, 2014). These studies prioritize order and harmony over negotiation and contestation, exploring successful collaboration at the expense of offering insights into the conflicts and negotiations that go on beneath the surface, behind the scenes.

Collaboration in global projects can be conflict-ridden and politicized, yet, as Clegg and Kreiner (2013) point out, the existing literature does not offer analyses in terms of power relations. Power relations focus on differential capacities to achieve variably weighted desiderata (Clegg & Kreiner, 2013). From this perspective, projects can be perceived as temporary organizational entities constructed from and constituting relations of power (DeFillippi & Arthur, 1998; Meyerson et al., 1996). At the same time, these relations are (often ambiguous) objects of ongoing construction and contestation. Partners in a project are variably able to exploit the ambiguity that characterizes collaboration in temporary organizations (Davenport & Leitch, 2005; Ivory & Alderman, 2015). We will focus on the various ways members of a temporary organization engage in both conflict- and consensus-seeking practices, concentrating on the everyday work and politics occurring between the spaces of ‘permanent organizations’ operating as part of ‘temporary organizations’.

Our case study deals with two relatively stable and permanent Titans who chose to work together temporarily in an inter-organizational project. A perceived discrepancy between expected and actual roles and relations triggered processes of formal and informal negotiation to reduce discrepancies (Morrison, 1994) and to resist or counteract the expectations of the other party (Courpasson, Dany & Clegg, 2012). Wider contexts provide interpretive frameworks for use by diverse organizational agents and agencies in large-scale global projects (Jones & Lichtenstein, 2008; Lundin & Söderholm, 1995). As we will show, the ambiguity attending the hierarchic relations between members of ACP and CH is grounded in national, cultural, contractual and (inter-)organizational contexts contributing to the emergence of divergent interests and sentiments. Drawing close to such a complex case of temporary organizing allows a more grounded understanding of collaboration in a large-scale international project.

Methodology

Ethnographic research of the PCEP informs a complex case study analysed to generate in-depth knowledge for theory building (Welch, Piekkari, Plakoyiannaki & Paavilainen, 2011). In-depth understanding captures the collaborative (and non-collaborative) behaviour of project participants in a socially complex setting (Kenis et al., 2009, p. 265). Single cases provide excellent contextual understanding of organizations as temporary phenomena (Lundin & Steinthórsson, 2003). The ‘actualities’ of the PCEP project, its lived experiences and daily practices (Ybema, Yanow, Wels & Kamsteeg, 2009) were explored in one year’s fieldwork. From July 2009 to July 2010, the fieldworker (the third author) observed daily practices and documented research participants’ lived experiences.

The contribution of ethnographic fieldwork in organization studies is widely recognized but underutilized (Ybema et al., 2009). Moore (2011) lists five contributions of ethnography to organization studies: it makes it possible (i) to compare different groups’ perspectives; (ii) to acknowledge ambiguities; (iii) to focus on explanation, categorization and sensemaking; (iv) to offer insight into the tacit aspects of processes of cultural negotiation, and finally, (v) to appreciate the uniqueness of specific situations. These attributes make ethnographic fieldwork suitable for studying the PCEP employees’ roles and relations and their daily practices of collaboration.

Data collection

Data collection involved observation and participation. Observation provided direct experiential and observational access to the insider’s world of meaning (Ritchie & Lewis, 2003, p. 35). Participation enabled access to otherwise inexplicable routines and activities. Observations were made of daily work routines, workshops, celebrations and meetings at every organizational level, as well as informal gatherings, such as lunch and coffee breaks and hallway conversations. Interest in the lived experience of the actors in large-scale projects led to the use of situated participant observation (Yanow, 2006), a method providing data on how practices actually comes about in situ, how they are produced, reproduced and negotiated. Observations typically involved three hours of fieldwork each working day; the field researcher always carried a small notebook to make sure information was directly registered. Observing, listening and querying project participants and their conversations provided information about everyday organizational life and emerging practices of collaboration.

Apart from observation and participant observation, the fieldworker conducted 47 in-depth interviews, 28 with ACP representatives and 19 with CH employees (see Table 1). Interviews benefit the systematic collection of people’s experience, interpretation and feelings without losing flexibility and spontaneity (Ritchie & Lewis, 2003). Interviews focused on employees’ roles and relation to other project members, discussing practices of collaboration, coordination, interaction and socialization. Although the project’s official language was English, in most interviews ‘Spanglish’ (a combination of English and Spanish) was spoken because many respondents were native Spanish speakers. All interviews were audio recorded, with only three exceptions captured in notes, while all audio records were transcribed literally.

OPEN IN VIEWER

Table 1.

Number of interviews per hierarchical level, per organization.

	ACP	CH
Senior Management	4	2
Middle Management	7	9
Operational Staff	17	8
Total	28	19

Finally, a documentary study collected historical, economical and political information on the PCEP to provide knowledge of the contextual framework (Lundin & Steinthórsson, 2003). Table 2 provides insight into the type of documents gathered.

OPEN IN VIEWER

Table 2.

Type of documents per organization.

Organization	Document types
ACP	Website, intranet, emails, company brochures, newspaper articles and cartoons, presentations, policy documents, annual reports, organization charts, contracts, tender documents, studies on Panama Canal, historical photos, maps and articles, videos
CH	Website, company booklet, presentations, studies, ACP agreement documents

Context

Panamanian ACP and US-based CH contracted to work together in a large-scale temporary project. As the relationship formed and unfolded over time, assumptions about roles, responsibilities and relations became increasingly vexatious. The hierarchic relations between the two Titans is grounded in wider institutional contexts, such as (a) (dis)continuities in national histories, (b) established (inter)organizational practices in large-scale construction projects, and (c) particular inter-organizational arrangements and contractual agreements that in this case constituted ambiguous power relations between ACP and CH and contributed to the emergence of divergent and conflicting interests and sentiments.

Historical context: shifting USA–Panama power relations

The history of USA–Panama relations in the Panama Canal Zone cast a shadow on the PCEP project. Although the Panama Canal had come under Panamanian control in 1999 after almost a century of US sovereignty over the Canal Zone (see Table 3 for a historical overview), the PCEP reintroduced ACP’s dependency on foreign expertise. The PCEP was initiated to maintain the Panama Canal’s competitiveness, to increase its turnover and capacity and to make it a more productive, safe and efficient work environment, entailing dry-land excavation of a massive amount of land and the deepening and widening of the Panama Canal and its navigation channels. The key component was the design and construction of the Atlantic and Pacific locks, the so-called Third Set of Locks (alongside the east and west reaches of the Panama Canal). Owners and operators of infrastructure devices such as dams, roads and railways generally do not have the experience to execute large-scale infrastructure projects (Winch, 2014), certainly not of the size and scope of the PCEP; thus, ACP created a temporary project organization in which ACP staff and a newly hired team of experts (referred to as the ‘Programme Manager’) would collaborate in the construction of the new set of locks. Formally, the project organization would reside under ACP as a separate Department of Engineering and Programme Management.

OPEN IN VIEWER

Table 3.

Time line of events and moves of participants in the PCEP.

Events	Date	Moves of Principal and Agent
Start of Canal construction	May 1, 1904	US finances the construction
Opening of Canal	Aug 15, 1914	Canal operations under US command
Signing of Torrijos-Carter Treaties	Sept 7, 1979	ACP gradually taking over Canal control
Canal transfer ceremony	Dec 31, 1999	ACP receives full control
Approval of PCEP	Oct 22, 2006	National Assembly and referendum gives ACP power to conduct PCEP
Finding project financers	Spring 2007	Pressure of project financers on ACP to hire an experienced agent for support
Invitation to Bid	June 1, 2007	ACP asks for an agent for the PCEP project
	Aug 16, 2007	US-based CH is hired by ACP
Flying in CH core team members	Aug 23, 2007	CH team members unknown to the required roles
Request for Qualifications for construction of new set of locks	Aug 28, 2007	CH team members feel uncomfortable in the role they are expected to carry out
	Sept 2007	CH realizes their role is to be limited to 33 people who act as advisors
Early Warning Notice	Oct 2008	To ACP’s irritation, CH issues a notice as ACP is likely to increase the cost and to delay the programme
Organization of workshop	Dec 2008	CH and ACP agree to form a strong partnership
The Gamboa Workshop with all CH members and their counterparts	Jan 2009	ACP communicates that success relies on teamwork and revitalizes the slogan ‘One Team, One Mission’. To CH this slogan has no meaning
Diverse integration activities	Feb 2009	ACP and CH work on their collaboration
PCEP bidding ceremony	July 8, 2009	GUPC presents ‘best value’ proposal
Formal start of construction of Third Set of Locks by GUPC	Aug 25, 2009	CH takes the lead in meetings and site offices. ACP claims authority and complains about lack of learning
Dividing of authority	Spring 2010	ACP takes care of internal politics, while CH takes the lead in content issues
Reclaiming of authority	Jan 2011	ACP employees gradually replace CH consultants
GUPC and ACP dispute about price compensation	Dec 30, 2013	GUPC issues notice of suspension of works in Third Set of Locks project
Strike of GUPC	Jan-Feb 2014	ACP postpones opening of the expanded Panama Canal to April, 2016
	Feb 20, 2014	GUPC restarts work on Third Set of Locks project
Planned end date of project	Aug 15, 2014	100 years after the opening of the Panama Canal
CH Project Manager leaves project	Mar 20, 2015	ACP does not backfill the position. CH remains available to provide advice

Findings – Titanic struggles

Members of the Panamanian ACP and the US-based CH engaged in a variety of collaborative practices, aimed at harmonizing relations, contesting ACP’s dominancy or challenging CH’s superior position. The reciprocal harmonization or contestation of roles and relations that emerged during the project work thus had three different, contradictory faces: (1) the Titans’ optimistic and harmony-seeking talk on trust and marriage; (2) CH disputing ACP’s control over the project; (3) ACP disputing CH’s control over PCEP.

Harmony-seeking: the Titans’ marriage

Particularly in the initial stages of the project, when global project members, equipment and materials arrived in Panama and housing, offices and organization were established, participants sought collaboration, willing to build non-hierarchical cross-boundary relationships. Project members organized social events, joint activities and social gatherings at work to socialize and strengthen relationships (see Table 4).

OPEN IN VIEWER

Table 4.

Harmonization in the ACP–CH collaboration in the Panama Canal Expansion Project: ACP and CH members’ social and discursive practices building harmonious, egalitarian relations (first face).

Social practices	Relational talk	Temporal talk
Social events and joint activities to get to know each other and to strengthen relationships. Examples: birthday parties, Halloween, Secret Santa, Friday afternoon drinks, joint breakfast and baseball team playingTraining and transferring of project knowledge.Illustrating quotes: ‘ACP are used to running an organization which is more operations and maintenance than building billion dollar projects.’ (interview CH employee 10.2009)‘For me it was a very good experience, because they have brought a lot of systemization of information.’ (interview ACP employee, 10.2009)	Soft talk of ‘marriage’, joint interests, trust, synergy, etc.Illustrating quotes: ‘What we have to do is pull the soft skills together to create a synergy’ (observation of a meeting 7.9.2009);‘Look at this marriage: ACP and CH … We’re going to work together and get things going.’ (CH manager, 3.2010)Pragmatic talk of a clear division of labour.Illustrating quotes: ‘[It is] a very hard relationship to work with.’ (Interview CH employee, 09.2009)	Talk of ‘a new start’: a bright future and no past. Illustrating quotes: ‘We’re getting to the point that we like each other and we kid around.’ (ACP Project Director 28.5.2010);‘I have been working 21 years at ACP, and it is the first time that a consultant is working with us to control a project.’ (interview ACP employee, 10.2009)Active use of the past, constructing a historical bridge: reinstating a slogan from the previous period of collaboration between Americans and Panamanians (1979–2000): ‘One team, One mission.’‘If it worked then, it could work now.’ (Interview ACP employee, 8.2010)

The relational discourse that dominated this early stage also sought to establish harmonious relations. When discussing the project organization, project members frequently spoke of a ‘marriage’ between ACP and CH, albeit one that was arranged: ‘I don’t think there is such thing as love yet. I think it’s premature. … We’re getting to the point that we like each other and we kid around. But, you know, that’s the first step’ (ACP Project Director, 28 May 2010). It was not only the optimism of the start-up phase; it was also the pressure to perform and if the temporary bond broke prematurely there would be existential risks for the partners beyond their temporary organization. The premature end of the ‘temporary’ organization would mean the end of the ‘permanent’ organization:

Look at this marriage: ACP and CH … We’re going to work together and get things going. That marriage is going to have to work. Because if it doesn’t, if they don’t get this thing working, it can sink the companies. (CH manager, March 23, 2010)

Rather than romanticizing the collaboration between ACP and CH as a project of love, project members applied the marriage metaphor to suggest that the project demanded that each partner in the collaboration invest time, money and effort reciprocally. Collaboration in the project involved establishing an enduring bond to create synergy:

CH Programme Director:

It required a lot of training on both sides of the fence and getting the right skills to collaborate. … It’s like a marriage; personalities have to match and that asks for mutual respect, professionalism, and the same goals for each team.

CH Project Manager:

What we have to do is pull the soft skills together to create a synergy. (Observation of a meeting, September 7, 2009)

The marriage metaphor was used prescriptively to explain what was needed to make the collaboration work. ‘Trust is essential,’ explained an ACP Programme Director: ‘If you have a problem with your wife, talk about it… don’t keep it for yourself’ (Interview, May 18, 2010).

In terms of temporal discourse, optimistic talk of the start of new collaboration abounded in the initial stages of the project. Project members made no reference to the past, while a joint future was presented as full of promise. Increasing conflict besmirched this bright future. When conflict became more prominent, ACP management started to promote an image of harmony, revitalizing and systematically promoting the old slogan ‘One Team, One Mission’ in internal and external communication. From 1979 to 2000, this slogan had been used to set the stage for a smooth transfer of the Canal to Panama and a seamless transition for canal custumers. Invoking this slogan, collaboration in the project was placed in a long-standing tradition of coordinated and harmonious relations. At various moments project participants reproduced the harmony-seeking, egalitarianism language also used in the Invitation to Bid, which depicted ACP–CH relations in non-hierarchical, consensual terms. Yet, despite attempts to promote the image of the ACP–CH tandem as one team, a married couple or as mutually complementing, project participants also articulated alternative views of collaboration and the role and relations between the project partners, based on workaday experiences.

CH’s contestation of ACP’s dominancy: ‘Why are we here?’

The second face of the collaboration between ACP and CH contradicted the first, suggesting relations that were neither harmonious nor egalitarian. We first discuss CH consultants’ views (under the second face). For the consultants selected by CH to work on the project, the roles, responsibilities and relations in the PCEP ran counter to their expectations, constituting a radical break with the past. Having led projects such as the Tsunami Reconstruction Program in Sri Lanka and the Maldives, the expansion of the Haifa Port in Israel and a wastewater treatment programme in Egypt, they came to Panama with ‘normal’ expectations of working in a standard project management operation, taking over leadership and responsibility, managing contracts and handling daily procedures and processes. However, the project breached normal client–consultant relational expectations. The division of roles and positions between CH and ACP staff laid down in ACP’s Invitation to Bid and the signed contract did not live up the standard set by the past.

Based on their former experience in programme management services, CH consultants assumed that their roles in the PCEP would not be different from any of their roles in previously conducted projects. In their standard role they assumed power and authority to act:

Contract oversight, contract administration, consultation, advice, but we are not responsible for directing or deciding, we are not, and that is very hard for our people. Most of our people here are very senior people, they have been around for twenty, thirty or some even forty years and they are like; ‘Well, then why are we here?’ ‘Well, because they want us here.’ (Interview, May 2010)

After the initial meetings CH consultants realized that ACP’s expectations of programme management services differed from common practice in this field: ‘There were no intentions to give leadership out of hand and we were treated as though we were staff’ (CH Programme Director, interview May 2010). In terms of role division and distribution of power, the PCEP thus constituted a temporal disruption of taken-for-granted and normal practices for CH consultants in project work (for a summary of social practices and relational and temporal talk of the second face of the collaboration, see Table 5).

In this role we stand to the side, we observe the contractor and we … we will try and document for the owner in case the project, or some phase of the project really goes south. We’re in a position that if nothing goes wrong, then we’re heroes; if something does go, something they’re going to ask is why, why didn’t we foresee it? (interview, June 2010)

OPEN IN VIEWER

Table 5.

CH consultants’ contestation of roles and relations in the ACP–CH collaboration in the PCEP through highlighting ACP’s power-enhancing practices and self-serving relational/temporal talk (second face).

Social practices	Relational talk	Temporal talk
CH consultants highlight their own subordination to ACP and exclusion from decision-making processes.Illustrating examples: ACP’s practice of arranging Spanish-speaking meetings (e.g. staff meetings on personnel performance, meetings with the government, local agencies, stakeholders or internal departments)ACP’s writing of internal reports in Spanish.Informal networks needed to get things done within ACP.	Being subordinated to ACP as formal decision-maker, CH consultants struggle to adjust to a supportive role, which does not fit their status, experience and self-image. Illustrating self-congratulating, other-diminishing talk:‘CH managers here all want to be leaders, but… we overlook, teach and manage the business, but we don’t lead the operation’ (CH Project Manager, informal conversation, 23.11.2009)‘We are Type-A personalities. ACP employees are more used to follow. That clashes’ (CH Project Manager 18.10.2010)‘[It is] a very hard relationship to work with.’ (Interview CH employee, 09.2009)‘If you have a mentor–protégé relationship, it’s important that the protégé wants to be mentored by the mentor.’ (Interview CH employee 09.2009)	Talk of a sharp temporal contrast, a breach with usual consultant–client relations, a (frustrating) present is measured against the past that serves as a positive standard. Illustrating positive-past, negative-present talk:‘We are all used to run big projects and now we are placed under managers.’ (CH Project Manager 18.10.2010)‘Most of the time they [clients] are just kind-of… leaned back. And we do it [but not on this project].’ (interview CH employee 3.2010)‘This is so much different than being a real PM.’ (interview CH employee 6.2010)‘We are not responsible for directing or deciding. And that is very hard for our people.’ (Interview CH employee 5.2010)‘This is not a normal relationship for a PM. It is usually that we are running the show.’ (Interview CH employee 10.2009)

CH consultants were ill prepared in terms of expectations for the specifics of the PM’s roles. Being accustomed to taking the lead and controlling the execution of a project, CH consultants had to learn to accept a radically new role as ‘trainer on the job’. They found their formal role in the project unusual and difficult. According to others, CH underestimated or even deliberately ignored potential problems. In the car on their way to lunch, the fieldworker asked a CH employee whether he thought CH read the contract well before signing it:

I’m confident they did not read the contract in detail. How could there otherwise be so much confusion about why we are down here and what our tasks are? Having the contract for the Expansion was much more important than its content. Our company needed it… It’s status, Karen, it’s status. Winning this tender had more value than figuring out how to execute it. This project is so important for us Americans and hey, we are making a hell of a lot of money here! (6 October 2010)

Despite the financial compensation, CH consultants constantly wrestled with their subordinated role in the project. Making a lot of money did not compensate for playing second fiddle and having to wait for ACP’s initiative as Kevin, a CH employee, explained:

There is not much work to do for me here, but ACP approved, and the only way CH can make money is by writing the hours. I feel ashamed though. I make a lot of money for doing nothing, and I don’t even like it… He [his boss] told me to focus on my senior status: when they have a question, I can help them. But that does not happen 8 hours a day. (Informal conversation, 8 June 2010)

Even though CH managers headed the organization charts in tandem with an ACP manager, apparently making them formal equals in authority, the former lacked the informal authority of their ACP colleagues. CH managers had superior expertise and, in their perceptions, senior status, but were not allowed to make decisions. Unused to this, frustrated when their advice was not followed or not even sought, they had to resign themselves to ACP decisions. In CH’s accounts, ACP constantly pushed plans through without heeding CH consultants’ advice.

If you have a mentor–protégé relationship, it’s important that the protégé wants to be mentored by the mentor. And that goes beyond respect… We have not the easiest relationship to work with. [It is] a very hard relationship to work with. (Interview CH employee, September 2009)

On occasions CH consultants could not participate in debates because ACP kept them away from ‘Spanish-speaking’ meetings, using language difference as a tool to exclude their counterparts.

I only heard about the meeting with [the committee] afterwards, but felt that I should have been involved too. I mean, we [ACP counterpart and me] share the same position, why did you not invite me? ‘Because these meetings are in Spanish anyway,’ is what he [ACP director] said. I could not believe my ears! (Informal conversation, May 2010)

Such exclusion happened frequently. Staff meetings on personnel performance, for instance, were held among ACP staff only. Meetings with the government, local agencies, stakeholders or internal departments were also Spanish speaking and many internal reports were written in the local language.

According to CH consultants, a supportive role in the PCEP project did not fit their ‘personality’: ‘We have many internal clashes. We are all used to run big projects and now we are placed under managers. We are Type-A personalities [high-achieving, status-conscious]. ACP employees are more used to follow. That clashes too!’ (CH Project Manager, 18 October 2010). A CH Project Controls Manager self-critically acknowledged that collaboration was often difficult ‘because of egos’: ‘CH managers here all want to be leaders, but we are not here to take the lead. Our task is simple: we overlook, teach and manage the business, but we don’t lead the operation’ (informal conversation, 23 September 2009). The CH consultants’ version of the ACP–CH collaboration cast them as struggling to adjust to a supportive role, being subordinated to ACP as the formal decision-maker, a continuous frustration because it did not fit their perceived status, experience and self-image.

ACP’s contestation of CH’s dominancy: ‘Somos la sombra de CH!’

CH consultants constituted their project relations as subordinated and excluded from power circuits, claiming that ACP ultimately pulled the strings on the project. Paradoxically, ACP members made the exact opposite claim: they maintained that CH consultants were in control, keeping ACP out of the decision-making process. This was the third face of the collaboration in the PCEP. Roles and relations of ACP and CH members were not egalitarian and harmonious (the first face), nor did they constitute a formal hierarchy in favour of ACP (the second face); in the eyes of ACP members they themselves were subordinated to ‘the Americans’ in day-to-day work. Although the Invitation to Bid had clearly outlined a supportive role for CH consultants, and ACP managers held powerful positions in the organizational hierarchy, formal meetings in which CH consultants participated reversed the pattern. Here, CH managers enacted their familiar leader roles, effectively renegotiating their position in the project. Here CH was often in a superior position vis-a-vis ACP (see Table 6).

OPEN IN VIEWER

Table 6.

ACP staff members’ contestation of roles and relations in the ACP–CH collaboration in the PCEP through highlighting CH’s power-enhancing practices and self-serving relational/temporal talk (third face).

Social practices	Relational talk	Temporal talk
ACP staff members highlight their exclusion from decision-making processes by pointing at CH consultants’ practices of holding CH-only meetings, and chairing and dominating English-speaking meetings with ACP.Illustrating quote: ‘They were reserved, very guarded, would have their meetings within CH and then come and talk to us.’ (interview ACP employee 11.2010)Exploiting their expertise.Illustrating quote: ‘The problem with the CH employees is that they want to impose their way of thinking, and they don’t want to just to share the information sometimes, they don’t give all the details.’ (interview ACP employee 12.2010)	ACP staff members critique their subordination to CH’s superior position.Illustrating quotes: ‘In the current situation we feel low, and we take a step backwards. Somos la sombra de CH! [We are the shadow of CH!]’ (Informal conversation 10.2009)‘With [CH] it’s “you’re my client, I’m the owner”-type of feeling.’ (ACP Senior Manager, 6.2010);‘They’re here to make money. And we wanted here a partner to help us.’ (ACP Senior Manager, 10.2010)Reflexive, self-critical talk and accepting to be ‘apprentices’ to CH: ‘We have a lot of “ego-persons” here. So it’s better to have an outsider to manage this complicated project.’ (ACP employee interview 3.2010);‘It’s just a matter of learning and to be humble.’ (Interview ACP employee 10.2010)	Implicit discontents with CH in calls to resign oneself in a subordinate role because ‘this is all new’, pointing out a past/ experience that is missing.Illustrating quotes: ‘We need to recognize that we don’t have the know-how on how to deal with it [new contract].’ (Interview ACP employee 3.2010)‘But, one of our goals was that the ACP people get experience from somebody that has done this before. There would be no added value to have hired them to do 100 per cent of the work.’ (Interview ACP employee 6.2010)‘You have your mind set on something like you are going to do this today, but then he [CH consultant] comes and changes it. He does it on purpose. This is like training. He wants you to be a robot, you know?’ (Interview ACP employee 12.2010);‘When we ring them up somebody [CH employee] said: “Well I didn’t come here to mentor anybody”.’ (Interview ACP employee 12.2010)

An excerpt from the fieldworker’s field notes illustrates this side of the ACP–CH relation:

Most people were already seated when I [the fieldworker] arrived. Not enough chairs were available, so more chairs were taken from a nearby office. When I finally found a place to sit, I noticed that the CH consultants were all seated at the table, while their ACP counterparts formed a second ring around the table. The setting seemed to emphasize a hierarchy in the relationship between the ‘CH chaperons’ and their ‘ACP apprentices’. The set-up looks like a theatre setting, I said to myself. Attendees formed a U-shape facing to the left end of the table, where ‘the stage’ was. From here, the CH Program Director started the meeting. Being seated on ‘the stage’ underlined his leadership over the meeting. In the discussions that unfolded in the meeting CH consultants dominated. The ACP employees were seated further away from the table, listening to what was said and taking notes. Reluctantly, an ACP employee asked for clarification. Often, participation from ACP’s side only came to the fore when their opinion was asked directly or when questions related to the ACP organization were raised. Being seated in the second row, it looked like the ACP attendees were hiding behind the consultants, showing a hesitant attitude, whereas the consultants had no trouble speaking up, as masters teaching their trade. (Field notes, September 2009)

In relational discourse, ACP employees often felt subordinated to CH consultants. Although CH employees had no decision power formally, in practice they had considerable control over decision-making processes. Dominant in their behaviour, CH consultants often played a leading part in meetings (observations weekly meetings, September 2009–March 2010). Formally they had little authority but CH consultants successfully played off their knowledge and expertise. In meetings in project site offices, such as in Cocolí, CH also often pulled the strings, as one ACP employee explained in one of many similar accounts:

CH is supposed to be advising us, but all I see in Cocolí is that they are very much in control. They rule the office. And when I have a question, I’d rather go to a CH person, because at least he knows the answer. ACP managers in the same position always need to verify with their counterpart! … On paper [the ACP manager] is the boss and, yes, he signs, but in reality it’s his CH counterpart who is making the decisions. … We need a different approach: CH should be in the second row in meetings. They are advising. In the current situation we feel low, and we take a step backwards. Somos la sombra de CH! [We are the shadow of CH!] (Informal conversation, October 2009)

To the ACP employees, the roles and relations within the PCEP were mostly new and temporary while continuing to work for ACP on permanent unchanged conditions of employment, often living in the same area as before. Working in a new location in a temporary project organization, they were responsible for construction of a new set of locks with a temporary increase of salary and work pressure, working in a new team under new supervisors, collaborating with and learning from US consultants. Consequently, work conditions, work environment, the work itself, as well as the mentoring-on-the-job by foreign experts, were all new to them: ‘We acknowledge that we don’t have experience in this large project or something similar and they do’ (interview with ACP employee). For ACP staff members, the project also offered new prestige. Each department had selected its best employees to work on the project; some had applied because it would be a promotion. Within ACP, selected employees were seen as ‘the chosen ones’ (ACP employee, informal conversation, July 2009). So, in many ways, working on the PCEP constituted a break with the past.

For the Panamanian staff, however, collaborating with ‘Americans’ was not entirely new. Although ACP members were aware of being dependant on CH experience and expertise for the project’s success, they started off with an adverse attitude towards the CH consultants. ‘They are concentrating on the image and success of their company. And, well, I am not pretty sure whether they have the same commitment with the program’ (ACP Ass. Project Site Manager, 9.2009). The need to converse in English when meeting with the Americans placed the locals at a disadvantage. Using Spanish as a reason for excluding Americans from meetings was not only tactical but also reaffirmed changed local relations of mastery in Panama over the Canal. CH did not keep to their roles as advisors and trainers, taking over leadership of the PCEP. ‘They [CH consultants] are very much in control. They rule the office…’ (informal conversation 10.2009). Several ACP staff members saw confirmation of their presentiments in the way US counterparts handled operations. ‘I’m not learning a lot from you guys hiding in a room, coming to a conclusion… The idea is that we all brainstorm together and we all learn… So that was kind of shocking to me, and they still work like that’ (interview ACP employee 2010). They were disgruntled, cast ‘in the shadows’ because of the ‘American’ style of working and hierarchical attitude (ACP employee, interview May 2010).

Not all ACP staff members were unhappy with Americans running the show. Some saw clear benefits in the presence of CH consultants on the project: ‘I also think that we right now do not have the experience in a multibillion dollar project’ (interview with ACP employee 3-2010). The temporary organization reflected the historical American–Panamanian relationship and several ACP employees expressed no interest in learning from the consultants, feeling comfortable in having ‘the Americans’ run the programme, finding it ‘refreshing to have them back’ (ACP engineer, interview October 2010). Others found CH’s presence beneficial, mostly because a majority of the regulations, processes and values within the ACP originated from the era of American control and thus were well known to CH consultants. Perhaps reminding their colleagues (or themselves) to remain self-critical, many ACP respondents underscored that the ACP employees lacked sufficient knowledge and experience in the management of large-scale projects:

This complicated project is something new for ACP. We need to recognize that we don’t have the know-how on how to deal with it. We have been doing excavation works, but this is the first locks project that we are going to face in this generation. (ACP employee, interview March 2010)

Some participants saw completing the project without CH’s assistance as impossible for the ACP (field notes, April 2010). Being organizational ‘apprentices’ on the project was not easy: for some, it was a lesson in humility:

So it’s just a matter of learning and to be humble enough to understand that every person, no matter who he is or who she is, can teach us something. But sometimes in ACP we think that we are almighty. But that’s not true. (ACP employee, interview October 2010)

In practice, CH consultants and ACP staff had to put up with each other, giving in to the other’s decisions, expertise or demands, losing discretion. As this was not always compatible with ACP members’ self-confidence – ‘sometimes in ACP we think that we are almighty’; ‘we have a lot of “ego-persons” here’ – it was hard for them to accept. The ACP Managing Director, acknowledging that the design of the collaboration with CH was unconventional and caused problems, expounded the original rationale. Given the project’s goals, it was the only way, he thought:

We have chosen the hardest way to execute a project. It would have been a lot easier when it would have only been ACP, or only CH … it was too high a risk to do it only with ACP people who have never done something like this. So… [pause] Oh well, we’re working it out. We’ll make it work. (Interview, June 2010)

Discussion

Collaboration in the Panama Canal Expansion Project between the two Titans CH and ACP – large high-status organizations with status-conscious members – involved engaging in harmony-seeking, hierarchy-evading and relation-improving practices to establish a working consensus, as well as in more discordant practices to contest and renegotiate hierarchic positions. The project partners’ collaboration had three faces, each showing different roles, responsibilities and relations. Early on in the project the disruption of role expectations and emerging conflicts focused both principal (ACP) and agent (CH) on overcoming differences and harmonizing relations. These findings align with earlier studies (e.g. Clegg et al., 2002; Morrison, 1994; Scott et al., 2011) observing harmony-seeking actors trying to maintain consensus. However, over time the discontentment with roles and hierarchic relations and the negotiations over mutual positioning shifted the focus to differentiation, contestation and pessimism, resulting in two additional, contrasting views of collaborative roles, responsibilities and relations. Ironically, each project partner claimed the other was ‘in control’ of the project, out of frustration with their own lack of control or in order to challenge the other’s dominancy. In the collaborative dynamics between the two titans in the Panama expansion project, conflict and order thus existed in tandem or, perhaps, in tension. One appeared to trigger the other. When, for instance, discontents and contestation surfaced, management responded by launching a harmonizing mission statement. Importantly, it was through addressing underlying tensions and ongoing negotiations that we came to fully understand the collaborative dynamics, highlighting that an order view may only partially capture collaborative practices. The findings in this study thus stretch our understanding of collaboration in temporary organizations by combining a ‘conflict view’ in our analysis with the ‘order view’ that currently dominates the literature (e.g. Bechky, 2006; Grabher, 2004; Scott et al., 2011).

In order to bring into view ‘conflict’ and ‘order’, we adopted an actor-centred, context-sensitive approach (Ybema et al., 2009), describing both the micro-dynamics and the wider cultural, contractual and organizational contexts of the ACP–CH collaboration (cf. Bechky, 2006). First, to describe the micro-dynamics our focus on social practices and relational and temporal discourse proved to be analytically helpful. Project partners harmonized collaborative roles and relations through social practices by, for instance, organizing workshops and festivities and informally dividing responsibilities between themselves (ACP taking care of internal politics; CH taking the lead in content issues). Through various forms of harmony-seeking relational talk, ACP and CH members sustained and strengthened collaborative relations, smoothed out hierarchical differences and remedied potential frictions; for instance by deploying the metaphor of ‘marriage’ and introducing the hierarchy-denying slogan ‘One team, One mission’. Finally, both CH and ACP members also attempted to harmonize collaborative roles and relations by engaging in specific forms of temporal talk; for instance, invoking optimistic images of a bright future and thus presenting the project as an opportunity to start from scratch, from an empty past and with a full future ahead. This way, they infused the project with a sense of hope and harmony.

Project partners had alternative and antagonistic ways to contest project roles, responsibilities and relations (the collaboration’s second and third faces). ACP and CH members challenged existing or emerging collaborative roles and hierarchical relations by engaging in particular social practices; for instance, hijacking the decision-making process by holding private meetings, excluding their project partner. In antagonistic relational talk they built binary oppositions between self and other to denigrate and diminish ‘the other’ while empowering the self, typically articulating their qualities and competence and critiquing the other’s power-pursuing practices. In antagonistic temporal talk project partners critiqued the present by claiming a historical rupture with the past. By referring to original intentions or ‘normal’ practices, they framed the present situation as unintended or abnormal. Therefore, when Cattani, Ferriani, Frederiksen and Täube (2011) state that project-based organizations have neither past nor future they are incorrect: for stakeholders the traditions of the past and the promise of the future may haunt the here-and-now. In sum, by engaging in particular social and discursive (temporal and relational) practices, they legitimized or delegitimized, sustained or undermined, particular roles, responsibilities and relations.

Second, we show that organizational actors’ day-to-day practices were embedded within wider contexts of contractual arrangements and cultural differences, embodied in language practices of Spanish versus English, and embrained in memories of past projects (CH) and past slights as subaltern partners (ACP). A particularly important context for the negotiations was the ambiguity of the hierarchy between ACP and CH. The power relation was remarkably diffuse. Normally, an agent is ‘in the lead’ and has delegated responsibility, appropriate expertise and requisite authority founded on both task and status structures (Turner & Keegan, 2001). In this case, however, the principal maintained its contractual right to formal authority albeit assigning itself the role of novice or apprentice, learning from its counterpart, being ‘chaperoned’ by CH as a mentor, roles and relationships radically new to both parties. CH’s preferential expectations of being ‘in the lead’ complicated the situation further. As a result, the constitutive rules of ACP ‘ownership’ and ‘learning’ were imperfectly enacted with ensuing tensions. At times, ACP made decisions with CH coming second best. At other times, CH made decisions and ACP, despite its formal authority, signed the paperwork to formalize decisions in a less than strategic ambiguity (e.g. Davenport & Leitch, 2005; Ivory & Alderman, 2015). This mixture of different hierarchies at play and participants holding competing views of the appropriate or preferred power relation weakened control over the execution of the PCEP project. After the period of fieldwork, the Third Set of Locks came to a standstill in January 2014 because of a conflict with GUPC over the contract. After weeks of public wrangling, GUPC and ACP agreed upon large cash injections by all partners to resume construction work, causing a budget overrun and a delay in completion of nearly two years.

Given the ambiguities of the formal hierarchy and expertise status structure embedded in the historical, (inter)organizational and contractual context of this project, it is perhaps not surprising that collaboration between the principal (ACP) and the agent (CH) became highly dynamic, complex and conflict-ridden. In situations where issues of status and hierarchy are unresolved and open to interpretation, increasingly the case in complex project networks (Lundin et al., 2015), the seeds of subsequent collaboration marked by controversy and discontent alongside harmony-seeking efforts may be sown. Where roles and relations of principal and agent are unclear, similar conflict-ridden negotiations may emerge; as Clegg and Kreiner (2013) suggest, organizational politics frequently dominate projects. The data may well be typical of dynamics in new project phases, when new combinations of project partners start to collaborate or when contractual arrangements differ from traditional roles and relations as in innovative public–private projects (Van den Ende & Van Marrewijk, 2014). National histories and cultural identities may impose additional challenges for project governance practices between principal and agent.

The literature on temporary organizations tends to set up a dichotomy between ‘temporariness’ and ‘permanence’ that provides a pragmatic way of distinguishing organizational forms and different literatures, a potentially problematic dichotomy that runs the risk of reifying permanence and temporariness as objective states of being (e.g. Winch, 2014). Grounding an analysis in the actual practices of day-to-day organizing, as we did in this paper, implies conceptualizing organizing as ongoing social, cultural and material accomplishments. From this vantage point, ‘permanence’ and ‘temporariness’ become something to be achieved, constituted and sustained or challenged and overthrown by the work of actors (Ybema, 2014). Rather than reifying these as objective realities, organizational actors may indeed use ‘permanence’ and ‘temporality’ as symbolic sources of power. Organizational actors attempted to achieve, sustain or challenge collaborative practices, thus constantly constituting or contesting permanence or temporariness.

Agent and principal interpreted the project architecture in indexical ways, in terms of prior dispositions and past experience. For CH, used to being in the command seat, it meant simply working according to well-tried and deeply familiar project principles and expecting everyone to be or to rapidly become familiar with them. For ACP, experienced in running the canal since its repatriation, the alliance was a sub-contract and ACP personnel naturally set the requirements and ensured continuity of command and control. Both CH and ACP members framed the actual experience of collaborating in the project organization as discrepant from their expectations, triggering constant negotiation over role enactment. Each of the project partners thus deployed the past as a symbolic resource to secure their leading role in the project or to challenge their partner’s claims, thus sustaining preferred versions of permanence or ruling out alternative versions.

Conclusion

Three faces of collaboration were distinguished: (1) both project partners’ harmony-seeking practices and optimistic talk on trust and marriage; (2) CH’s contestation of ACP’s superior position in the formal hierarchy; (3) ACP’s contestation of CH’s dominant position. To conclude this paper, we highlight three wider theoretical contributions this paper makes and we offer a practical suggestion.

By studying the conflict-ridden dynamics of collaboration between partners in a large-scale international project we show first how organizational actors negotiate hierarchy in situ by engaging in a variety of collaborative practices and relational and temporal talk aimed at harmonizing relations or contesting the emerging hierarchy. Hierarchy evolved as a symbolic site for struggle. Grabher (2004), Bechky (2006) and Kramer’s (2009) understanding of temporary organizations as based on enduring, structured role systems whose details are negotiated in situ is supported. However, unlike studies that focus on order being achieved in the potentially disorderly world of temporary organizations (e.g. Bechky, 2006; Pitsis et al., 2003; Scott et al., 2011) we show how persistent ambiguities in the roles, responsibilities and hierarchic relations trigger more conflict-ridden in situ negotiations over expected roles of principal and agent. A focus on social practices and relational and temporal discourse brings into view the day-to-day processes of harmonization and contestation.

In addition to the previous point, our findings may also extend academic debate on complex mega-projects’ governance structures (Miller & Hobbs, 2005; Müller, 2012; Sanderson, 2012). The roots of the conflict described herein lie not only in the governance structure but also in the micro-practices that emerged. These practices weakened control over the execution of the PCEP project. We need to understand post-contract governance processes, especially when contractual arrangements, intercultural histories and organizational traditions give rise to ambiguous and potentially conflicting interests, cultural identities and expectations. By producing an ethnographic account of social and discursive practices we gained insight into how post-contract processes may shape projects beyond the norms of constitutive contractual frames (Sanderson, 2012).

Third, there is heuristic value for future research into processes of temporary organizing in viewing permanence and temporariness as symbolic accomplishments or contested categories. We have shown that ACP and CH members, for instance, sought to create permanence for their own position and to undermine that of the other. In negotiations, when positions are claimed or contested, permanence and temporariness become concrete stakes, conceived as political projects to sustain or oppose the legitimacy of a hierarchical position, a particular role, or an established routine. Organizational actors invoke particular imaginations of their past, present and future and discursive constructions of (dis)continuity, thus inventing or inverting a tradition or a transition (Ybema, 2014).

Finally, to practitioners the findings presented in this paper help to better prepare for temporary collaboration in complex infrastructure projects. In our case, agreements on roles, relations and collaboration philosophy made in the tender phase were hardly known and poorly understood by project employees in the execution phase. As this not a unique case (Van Marrewijk, 2015), reflection upon the context and situatedness of temporary work is needed to align mutual expectations and to stimulate learning between principal and agent. Otherwise, principals and agents may fall back on, and fight over, established work practices and preferred hierarchical positions.