A pragmatist perspective on front­end project organizing: the case of refurbishment of the Palace of Westminster

This study examines how actors organized collective efforts to initiate the project and navigated their way forward, even though the ultimate outcome was not clearly defined. To address this question, we explore the potential of the philosophical tradition of pragmatism. This approach foregrounds duality, recursiveness and temporality of collective activities and offers a new and compelling way to understand and address the challenges actors face in organizing and managing the front- end. By accounting for both the situatedness of actions in the wider social and relational contexts, and by connecting the flow of present experience to the interpretations of the past and future, pragmatism holds the potential for integrating theory the actuality of lived experience in its continuous unfolding while ac- counting for actors ’ transformative agency. By drawing on a real-time longitudinal study of conception of the program of the Restoration and Renewal (R & R) of the Palace of Westminster, we show how participants face the challenge of understanding and managing complex tensions that continually arise due to a duality of mobilization (between seeking consensus and expanding divergent possibilities) and a duality of transformation (between forming a bold vision of the future and translating abstract goals into concrete actions). To tackle the challenges, participants create strategic accounts that are stable enough to be practically feasible in current circumstances, but also sufficiently adaptable to pursue future possibilities in ways that challenge prevailing approaches. By showing how participants cope with these challenges by creating spaces of experimentation and constructing flexible boundaries this study contributes to the literature on management of project ’ s front-end.


Introduction
The "front-end" of a project where key decisions on project outcomes and execution are made is the critical period determining project success or failure (Edkins et al., 2013;Miller & Lessard, 2001;Morris, 2009;Samset & Volden, 2016). Although recent literature on the front-end offers viable alternative to rational choice models, however, research has typically focused on the "content" of front-end management including: issues regarding the definition and conception, what critical factors needs to be considered, what roles different actors play, what are the attributes of governance and strategy, and what defines a successful stakeholder engagement or risk management (See Williams et al., 2019 for the review). This focus on the content of this stage rather than how content is produced or used neglects to consider the relational and processual dimensions of the front-end organizing processes (Geraldi & Söderlund, 2016;Zerjav et al., 2021).
Following the call to understand the actuality of projects , this study conceives the front-end development as an inherently dynamic and collaborative process. During the early stage of development, participants with diverse backgrounds, identities and interests come together in an attempt to collectively construct strategic frameworks regarding what the future of the project should look like and what paths should be taken to achieve a desired outcome (Alimadadi et al., 2021). The transformative objectives and diversity of participants create a highly uncertain and fuzzy context which makes coordinating actions and realizing future objectives difficult (Martinsuo et al., 2019). The front-end development is therefore rife with tensions stemming from the interplay of seemingly opposing imperatives: between seeking consensus and divergent interests, and between transformative value of the outcomes and translating those goals into concrete actions ("means versus ends") (Gil, 2017;Samset & Volden, 2016).
As various project participants engage in processes of decisionmaking and organizing, they must cope with these tensions and navigate a path forward. In this context, it is through negotiation, collective sense-making and reflection that decisions are made (Weick, 1995;Williams & Samset, 2010) and order emerges through creative acts, interactions and local orchestration of relationships (Chia, 1994;Langley et al., 2013). This collective and future-oriented nature of the front-end lays the groundwork for the present study. Although existing research broadly acknowledged the importance of these processes and practices Geraldi & Söderlund, 2016;Davies et al., 2018), we still lack understanding of how actors act creatively and collectively to resolve conflicting imperatives in the front-end of projects and of how these activities unfold over time as participants attempt to construct a path of action toward a realizable project outcome. Therefore, the study focuses on the following research question: How do actors participating in the front-end of projects creatively resolve conflicting imperatives and construct a path of action to move forwards?
To address this question, we look beyond the dominant conception of the front-end management to explore the potential of the philosophical tradition of pragmatism as an alternate theoretical perspective. By shifting from rational approaches to focus attention on project actuality Davies et al., 2018;Tywoniak et al., 2021), pragmatism foregrounds duality, recursiveness and temporality of collective activities and offers a new and compelling way to understand and address the challenges actors face in organizing and managing the front-end (Buchan & Simpson, 2020;Cicmil, 2006;Tywoniak et al., 2021).
The study draws on a real-time longitudinal study of conception of the program of the Restoration and Renewal (R&R) of the Palace of Westminster. The R&R program provided an opportunity to observe the very early stage of inception when a so-called "shadow organization" was established to prepare for the execution of a large, complex program when the organization becomes "substantive". As participants had a fuzzy, ill-conceived understanding of project outcomes and no formally agreed way of tackling it, they faced the challenge of attempting to predict an uncertain future and understanding various alternatives for achieving it. Yet, the urgency of the situation prompted them to find a viable way of moving forwards. To the best of our knowledge, no previous work has been carried out to specifically study the unfolding of these activities in real-time to examine the experience of projects participants as they confront and cope with various challenges in practice (Brunet et al., 2021;Buchan & Simpson, 2020). Hence, this study contributes to the literature on management of project's front-end by examining how actors organized collective efforts to initiate the project and navigated their way forward, even though the ultimate outcome was not clearly defined. The paper identifies two main challenges facing participants and illustrates how they tackle the challenges through creating strategic accounts that are stable enough to be practically feasible in current circumstances, but also sufficiently adaptable to pursue future possibilities in ways that challenge prevailing approaches by creating spaces of experimentation and constructing flexible boundaries.
The remaining sections of the paper are structured as follows. First it reviews a selective body of literature on the challenges of organizing the front-end of projects and recent development of research on project actuality. The paper continues by introducing pragmatism as an overarching conceptual framework that can help to advance our understanding of the challenges of organizing the front end and how they can be managed. The R&R case is presented, before proceeding to the analysis, focusing on the main difficulties that participants faced and how they managed to overcome those challenges.

Theoretical background
The following section is structured in three parts. To outline the challenges of organizing front-end, we begin by reviewing the extant literature to highlight two main challenges. Second, the section provides an overview of the stream of research of project actuality with a particular emphasis on project-as-practice approaches. We then turn to research in pragmatism to elaborate on the key implications of this approach for studying the front-end. The aim is to explore how the growing body of literature informed by pragmatist perspective may help to illuminate the actuality of managerial practices in the front end.

The challenges of organizing front-end
While there is no wide-ranging agreement on the definition of the project front-end, this study considers the front-end as the preliminary emergence phase, the end of which is often marked by the final sanction to authorize the project (Williams et al., 2019). In the case of large public projects, the formal decision to initiate a project is quickly followed by the establishment of informal bodies to get things underway (Samset & Volden, 2016). Project work, particularly at the front-end, is inherently complex and messy social processes (Williams et al., 2019;Williams et al., 2009). In such circumstances, well-planned and rational-oriented actions may not apply. Drawing on the extant literature, two main challenges facing participants as they embark upon initiating a large project are discussed below.

Means versus ends
Projects are a means for achieving a particular outcome in response to a combination of opportunities and problems (Williams et al., 2019). In the context of the front-end, attention has been given to the choice and clarity of conceptual solutions (Samset, 2009;Samset & Volden, 2016), and consistency of those solutions with the wider and longer-term social benefits (Zerjav et al., 2021). The literature offers insight into challenges of maintaining alignment between organizational strategy and the project concept (Williams & Samset, 2010;Williams et al., 2019). However, the attention often focuses on "identifying", "evaluating" and "choosing" a viable solutions ex post, rather than studying the unfolding and emergent qualities of initial processes of sensemaking, problematization of the present and projection of the future outcomes. Smith and Winter (2010:53) argue that attempts to produce "something properly defined that could be achieved in the time allowed" at the outset could come at the expense of deterring of creativity and restraining of the transformative potentials of a project.
Few studies have critically questioned the prevailing "stage gating philosophy" and explored the processes and practices of generating alternative ideas, imagining the possible project outcomes, and constructing paths to reach projected futures. The process of imagining a future and then seeking to realize it is captured by project narratives or future perfect thinking in the front-end (Pitsis et al., 2003). Winch and Sergeeva (2021) identify how the forward-looking projection of ends combines with a visualization of the means by which that projected future may be accomplished (Weick, 1979, p. 198). Pitsis et al. (2003) show how project participants use "future perfect" thinking to construct a path of action toward a realizable end. By projecting non-existent phenomena into an imagined future and then working backwards to the present, actors engage in a retrospective construction of paths that should be taken to achieve a desired outcome.
For future perfect thinking to be effective, however, actors need to know where they want to go. Yet, the transformational nature of large projects requires participants to pursue alternative future possibilities in ways that challenge and move beyond current arrangements (Alimadadi et al., 2021;Gil & Fu, 2021;Zerjav et al., 2021). This is often highlighted by the considerable socio-economic, cultural, and environmental impacts of projects characterized as complex, ambiguous, and uncertain. But the transformative nature of projects can be a barrier preventing participants from moving forward with well-defined ends and practically actionable and viable strategies. A key challenge in the front-end, therefore, is striking a balance over time between assessing what is realistically achievable and envisioning transformational goals.

Hierarchical versus consensus-oriented coordination
Organizing large projects is a collective effort in which individual and organizations with diverse interests come together to work jointly on a shared task for a limited period (Bakker, 2010;Winch, 2014). Zerjav et al. (2021) highlight the political, value-driven and negotiated nature of decision making in the front-end. The subjective stakeholder interpretation and plurality of what is deemed to be worthy (Martinsuo et al., 2019) contribute to the "fuzziness" of the front-end: a poorly structured context 'where objectives are not clear, where different constituencies have conflicting aims, and where the way forward requires vision and leadership as well as hard analysis and design' (Morris, 2002, p.88). During this phase, the consequences of decisions will be highest (Williams & Samset, 2010) and processes, structures, authorities, and accountabilities are established to initiate and complete the project as planned (Biesenthal & Wilden, 2014;Samset & Volden, 2016;Turner, 2009). This phase finishes when the desired outcome and the project participants' roles and relationships are specified, formalized, and agreed upon.
Centralized decision making is, however, often impractical and inefficient during the early stages (Müller et al., 2017) because critical knowledge and resources are distributed among many participants who can alter the project's strategic direction. Traditionally, for example, the client/sponsor and the contractor/deliverer play the main roles in large projects. The former is responsible for specifying the right project (what), while the latter is responsible for delivering the project as specified, on time and budget (how) (Merrow, 2011). The inherent interdependencies between 'what' and 'how' of the project at these early stages however requires a consensus-based coordination. Participative engagement is necessary to ensure that all stakeholders are involved and committed to projected benefits generated by such endeavours (Gil & Fu, 2021;Smith & Winter, 2010). Outcomes are achieved through a series of open-ended interactions and negotiations amongst actors with diverse interests engage in processes of legitimation and contestations (Zerjav et al., 2021). Yet, hierarchical modes of coordination, manifested by the predisposed division of labour between the assigned roles and expected responsibilities, can constitute a major challenge to consensus-based approaches necessary to align actions and objectives (Smith & Winter, 2010). Table 1 summarizes the two challenges discussed above.

Project actuality research
In their influential work,  questioned the normative and rational underpinnings of project management and invited researchers to use other, more critical perspectives, that can capture the actuality of projects as they happen in situation and in action (see also Cicmil &Winter et al., 2006). Advocating pragmatic philosophical considerations, scholars suggested more attention to issue such as actors' sensemaking process, the experience of emotions and feelings that drive action, how actions unfold over time with broader strategic consequences for projects .
In response, a small yet growing body of research adopting a practice-based perspective has emerged in recent years; mainly inspired by the work in strategy-as-practice (SAP) (Jarzabkowski & Wilson, 2006;Orlikowski, 2000;Whittington, 2006). Advocates of SAP examine the 'micro-activities involved in the social accomplishment of strategy' (Jarzabkowski & Balogun, 2009: 1258Jarzabkowski & Spee, 2009;Jarzabkowski et al., 2007). Drawing on Whittington's (2006), a nascent literature that explores projects-as-practice privileges Praxis (actions of a project manager or project worker), Practices (routinized types of behaviour and tools that are used in project work), and Practitioners (actors that are involved in project work) as their key research foci (e.g. Blomquist et al., 2010;Hällgren & Söderholm, 2011). By turning to questions of how project as an organizational form is enacted in practice rather than in design, and by focusing on situated action, these studies have contributed to our understanding of the daily activities in which people engage when they are taking part into what is labelled "a project" (Clegg et al., 2018;Sergi, 2012).
Generally, project-as-practice share a common concern with process studies (e.g. Brunet et al., 2021;Langley et al., 2013) in how strategies and governance frameworks emerge in practice. However, this research tradition often attributes emergence and development of projects to the micro 'doing' of actors in formal project planning settings (Jarzabkowski et al., 2007;Johnson et al., 2007). Buchan and Simpson (2020), however, make an important distinction between projects-as-practices (noun), which is concerned with the various routines and entities that are engaged in projects, and projects-as-practice (verb), which attends to the processual nature of project as socially engaged emergent processes, rather than identifying a number of practices done 'inside' a project (See also Sergi et al., 2020). While a dominant "representationalist approach" may to some extent answer the call by Cicmil and colleagues to study project management from a more critical and interpretivist stance, it nonetheless fails to engage with those aspects of projects that are inherently relational, dynamic, and emergent (Buchan & Simpson, 2020;Lorino et al., 2011). Recently scholars have proposed incorporating process thinking into the agenda set by researchers interested in project actuality to explore the largely overlooked emergent nature of projects (Buchan & Simpson, 2020;Sergi, 2012;Sergi et al., 2020).
Process thinking in projects foregrounds the situate activities within a broader timeline as well as how projects emerge over time as participants make connections between multiple project experiences (Burgelman et al., 2018;Langley et al., 2013;Brunet et al., 2021). With its focus on emergence and unfolding, process thinking is especially well-suited to examining the front-end of a project. Specially, the pragmatist approach as one type of process theory (See Simpson & den Hond, 2021) complements the insights from the project-as-practice studies by attending to how tensions between imperatives of means-ends and hierarchical versus consensus-oriented coordination surface and are experienced by participants over time as they engage in project work. Furthermore, pragmatism finds inquiry, reflection and experimentation-rather than practical coping-instrumental for agency and for the transformation of practices (Lorino, 2018;Miettinen et al., 2012). In their recent article, Buchan and Simpson (2020) draw on Dewey' Inquiry lens (1938) to show some of the complexities of experiential process as managers engages with doubt in order to bring about situational transformation. These qualities are well-suited for studying the dynamics of organizing front-end. For example, Smith and Winter (2010) argue that projects are shaped and formed as practitioners, through sensemaking, trial and error, and muddling through, develop reflective capacity that allows them to learn, operate and adapt effectively in the front-end of projects. Ultimately, by accounting for both the situatedness of actions in the wider social and relational contexts, and by connecting the flow of present experience to the interpretations of the past and future, pragmatism holds the potential for integrating theory the actuality of lived experience in its continuous unfolding while accounting for actors' transformative agency. The following section outlines some of the core principles that can be useful in analyzing the front-end activities.

Pragmatism as a conceptual lens for studying front-end
The philosophical tradition of American pragmatism is concerned with the practical consequences of action and places problem solving at the center of its understanding of human action (Farjoun et al., 2015). This focus on how problem solving is accomplished in practice is particularly useful for understanding and analyzing the organizational challenges in the front-end (Lalonde et al., 2010;Tywoniak et al., 2021). What the problem or opportunity is and what task is to achieve is "undetermined or not wholly given" (Dewey, 1985;cf. Emirbayer & Mische, 1998). Instead, the definition of the problem is not neatly defined and is itself an essential part of its possible resolution (Dewey, 1933). Means and ends are not always clearly determined prior to action. Actors are active experimenters, finding out the various lines of possible action and making various combinations of selected elements of habits and impulses to find a way forward (Dewey, 1922). Pragmatism highlights activities of sensemaking, deliberation, experimentation, and reflection as the maxim of action (Ansell, 2011;Elkjaer & Simpson, 2011) and offers an insightful lens for studying the front-end of projects because, as Engwall (2002) observed, it is through practical actions that expectations regarding a project's future outcomes are transformed. The meaning of the objectives is reinterpreted and their frame of action as well as the scope is often revisited as the project progresses. As outlined below, three principles of pragmatism are useful in understanding the front-end organizing. Table 2 shows an overview of these principles.

Dualistic tensions
Front-end organizing is replete with dualistic tensions, such as between means and ends, and between consensual and hierarchical coordination. Advocating a pragmatist approach in studying complexity in project, Tywoniak et al. (2021) associate the lack of sophisticated theories in project studies with the reductionist and simplistic theorizing, often generated by dualistic theorizing based on distinct categories and clear-cut boundaries. Pragmatism, on the contrary, conceives categories as overlapping and boundaries as elastic. It promotes theorizing that perceives "reality" as dynamic constellations of things connected in co-constituting and non-dualistic cycles (Geraldi et al., 2021;Joas, 1996;Whitford, 2002). Categories such as means and ends, and theory and practice, are seen as both contradictory and complementary, recursively linked and mutually constitutive (Farjoun et al., 2015). In the context of the front end, pragmatists would not presume "fuzziness", but pay closer attention to how multiple meanings associated with a certain category could change and established boundaries might shift over time (Glynn & Navis, 2013).

Recursiveness
Through its recursive logic, pragmatism offers a novel perspective on progression and complexity in the front-end phase (Tywoniak et al., 2021), paying attention to stimulus and response processes that relate back on themselves in a co-evolving, looping fashion. Preferences are not given but shape, and are shaped by, action. Individuals are capable of reflecting on their own actions, revising habits and enacting the environments to which they later adapt (Commons, 1934;Weick, 1979). In the recursive cycle, a continuously unfolding social process of inquiry occurs through which meanings are constructed and uncertainties are addressed (Buchan & Simpson, 2020;Dewey, 1938). Actors confront problems and generate hypotheses by combining novel and existing knowledge, leading to new beliefs and habits as well as imaginative leaps (Farjoun et al., 2015;Fontrodona, 2002).
In a project surrounded by uncertainty, participants hypothesize chains of means and ends to choose a course of action, and later adjust actions through reflection and learning, rather than pursuing a predetermined set of preferences and beliefs. Experimental learning highlights the capacity of actors to reflect on their own habitual knowledge and deal more effectively with other similar challenging situations (Lalonde et al., 2010). In contrary to the often linear portrayal of front-end practices, the cyclical nature of processes emphasizes the ongoing, iterative, and cumulative processes of co-creation and context of action.

Future-orientation
Pragmatists adopt a processual view that appreciates the temporal and emergent aspects of social life (Simpson & Den Hond, 2021). The constitution of action is inherently temporal: participants engage with the future and make decisions in the present, while building on the past (Kaplan & Orlikowski, 2013). The notion of projectivity captures the future-oriented dimension of action (Mische, 2014). The front-end of a project is where the projective dimension of human agency (Emirbayer & Mische, 1998) is at its highest, constituting a site of deliberation and "hyperprojectivity" (Mische, 2009).
The dominant view in project management emphasizes a "linear" conception of the future (Lindahl & Rehn, 2007) and depends on a "clock-time" as a structuring device, with considerations of the pre-determined duration (Bakker et al., 2016). Clock-time provides external order that renders project organizing amenable to mechanistic control and coordination (Adam, 1994;Reinecke & Ansari, 2015). As actors engage in collective endeavours, they experience a continuous tension between the future determined by clock-time, instantiated in project plans, milestones and deadlines, and the flow of event-time that is an emergent property of the activities carried out in practice (Langley et al., 2013).
A pragmatist approach accounts for these experiences, as participants simultaneously influence and are influenced by emerging futures (Simpson & Den Hond, 2021). Rather than any deterministic form of progression, pragmatism considers action inspired by what actors conceive to be the future implications of their action. As such, this approach give primacy to abductive reasoning (Mantere & Ketokivi, 2013;Peirce, 1878). Whenever encounter with uncertainty and "doubt" about what to do next, abductive reasoning challenges existing Table 2 Implications of pragmatism for studying the front-end Principles of pragmatism Implications for studying the front-end

Dualistic tensions
Categories such as means and ends, and theory and practice, are seen as both contradictory and complementary, recursively linked and mutually constitutive. Implication: Front-end organizing is replete with dualistic tensions between means and ends, and between consensual and hierarchical coordination. Pragmatism's non-dualistic stand focuses attention on how multiple meanings associated with a certain category could change and established boundaries might shift over time.

Recursiveness
Processes and relationships emerge and evolve in recursive cycles. Attention is focused on stimulus and response processes that relate back on themselves in a co-evolving, looping fashion. Implications: Pragmatism, with its recursive logic, accounts for how participants, to cope with uncertainty and fuzziness of the front-end, hypothesize chains of means and ends to choose a viable course of action, and how they might later adjust actions through reflection, experimentation, and learning, rather than pursuing a predetermined set of preferences and beliefs.

Future-orientation
The constitution of action is inherently temporal: participants engage with the future and make decisions in the present, while building on the past. Implications: The front-end of a project constitutes a site of deliberation and hyperprojectivity. Rather than a linear conception of the future, pragmatism accounts for the experience of continuous tension between the future determined by clock-time (plans and deadlines) and flow of event-time that as participants conceive the future implications of their action.
assumptions, and creatively generates new hypotheses by combining existing knowledge and novel ideas, which in turn can lead to transformative thinking about the future (Emirbayer & Mische, 1998;Farjoun et al., 2015).

Methodology
This study is part of a larger program of research investigating the dynamics of organizing in a pluralistic setting. The front-end of projects is one of exemplary organizational settings where various organizational and social actors with diverse background, identities and often conflicting perspectives work together to reach agreement on how to create a frame of action capable of delivering the project.
The overall research approach is inspired by process thinking (e.g. Langley et al., 2013), conducting qualitative investigation into how situations and issues unfold over time, what emerges out of the inherent contingencies of action and what plays a part in the unfolding of events (Brunet et al., 2021;Sergi et al., 2020). This is achieved by paying close attention to what is uniquely experienced by the actors involved in projects (Fachin & Langley, 2017). From this standpoint, research does not seek to generalize about the population, rather to achieve transferability and reflexivity (Sergi et al., 2020). Therefore, the present study is concerned with identifying the underlying mechanisms that drive the development of the front-end and possess relatively enduring properties and powers (Reed, 2009, p. 60). However, the effects and consequences of these mechanisms are highly context dependent and can only be displayed with regard to the specific spatial and temporal context. Edwards et al., 2014). Hence, what in traditional case study approach would be regarded as "idiosyncratic detail" to be removed from the explanation (Eisenhardt, 1989), now becomes part of the fabric of an explanatory account (Piekkari et al., 2011).
To this end, few studies have such privileged access to the inception of a project, so our study of the front-end phase provided us with an exceptional opportunity to study project participants in situ with their peers (Jarzabkowski et al., 2013). Being involved so early in the R&R project, when various actors such as contractors, sponsor team, and consultants were just joining the program, allowed us to focus on the real-time activities and discussions between participants as they were attempting to understand the problem and find an agreed-upon way of tackling it. Real time longitudinal process research offers greater possibility to capture the ongoing development of the front-end in all their richness, while avoiding hindsight bias and unwitting rationalization of past events (Bizzi & Langley, 2012). In this vein, this study offers opportunities to challenge some of the more linear and variance-based conceptualization of the front-end organizing.
Moreover, the immense symbolic, political, and social value of our case-the reconstruction of the home of British Parliamentary democracy-as well as the importance of the program as a high-value public infrastructure project allowed us to gain a deeper understanding of the dynamics of front-end organizing when conditions were highly strategic, urgent, and uncertain.

Data collection
A real-time longitudinal study was undertaken between November 2018 and June 2019. The author spent an average of two days a week (almost 400 h) in the field. During this period, to build reliability, gain in-depth insight and capture real-time "in the making" phenomenon (Brunet et al., 2021;Fachin & Langley, 2017), multiple sources were used: observations of various strategic panels and workshop, in-depth interviews, archival and field material. The research was also informed by collection of independent media coverage, reports by industry experts and academics, online publications. While this extensive database served as a source for understanding the context of our argument here, this study specifically drew upon a detailed analysis of observational data where the overall strategy for the R&R program was discussed and planned, including six strategy workshops and selected 18 key interviews with the leadership team. Table 3 provided an overview of main data sources.
The six strategy workshops (between a half-day and two-days in duration) involved all members of the leadership team and on occasion some other senior attendees. During the sessions additional data was gathered, including presentations and other artefacts (exhibits, diagrams, lists and figures) created by the participants during the workshops. Direct observations allowed the field researcher to study participants in situ, while detailed accounts of the events obtained from the interviews furnished a deep appreciation of the work based on situated interactions (Langley et al., 2013). All interviews for the present study were conducted in person, audio recorded (except three) and transcribed.
During data collection, the author had numerous informal discussions with managers, usually after meetings or workshops, which helped the author understand specific problems raised in the sessions and appreciated actuality of the situation, such as practical concerns and various local repertoires and lexicons that all members had to learn (Nicolini, 2009;Van de Ven, 2007). Importantly, it enabled the author to compare what was said and done in situ (during the meetings) with managers' view of what just happened after the meetings. Discussing interpretations and the analysis with practitioners provided an opportunity to reflect on the validity of the theoretical ideas, as discussed in the section below, and their relation to practice.

Data analysis
Although this study was in the beginning exploratory and inductive, the analysis became abductive in orientation, involving development of specific theoretical ideas alongside increasingly targeted analysis of the empirical material (Alvesson & Kärreman, 2007;Mantere & Ketokivi, 2013;Swedberg, 2014). The analysis was conducted in multiple stages.
Stage 1. Based on both real-time data and retrospective data from interviews and reports, we developed an overall chronological summary or our researchers' narrative (Langley, 1999) to connect various events and create a coherent narrative from the "lead-up" to the inception of the R&R program in 2018, up to the end of the field work in June 2019. This helped to develop an overall understanding of the key themes in the unfolding of the front-end phase as well as the roles of various project participants in this process. The notion of "urgency" became increasingly apparent as one of the main characteristics of the setting because participants were under increasing pressure to tackle the issue of the crumbling buildings as quickly as possible and there was little consensus or clarity around what the Palace should look like in the future. These issues, although manifested differently in different meetings and workshops, persisted over time, against the backdrop of attempts to reach consensus about how to proceed with the work. As leaders made decisions on one issue, a new issue emerged. Project participants experienced ongoing tensions and in various occasions expressed that they are constantly grappling with these issues, which oriented our attentions towards patterns of response over time, rather than focusing on individual instances.
Stage 2. The received literature on the project's front-end often portray this process as a stage-gated, linear progression where uncertainty is reduced and clarity of concept and formality of structures increase over time (e.g. Edkins et al., 2013). The findings, however, showed the recursive nature of this process, wherein similar issues (such as creating an effective governance framework, planning the activities, and envisaging the future goals and objectives) were surfaced and reopened over time, despite various attempts by project leaders to proceed to the next stage. This marked a key moment in an abductive process, where a surprising empirical phenomenon is observed in light of the interpretive rulei.e., extant literature on the front-end (Alvesson & Kärreman, 2007;Van de Ven, 2007).
To create a novel understanding of the front-end development following this discovery, or "recovery of understanding" (Alvesson & Kärreman, 2007;Peirce, 1878), this research drew on the corpus of work by classical pragmatism and the recent conceptual development in organization and management literature (e.g. Farjoun et al., 2015;Simpson, 2009) to gain a deeper understanding of the challenges experience by project participants. It was useful to focus on the tensions experience by participants stemming from interplay between collective attempts to move forward and the experience of uncertainty and ambiguity that surrounded these attempts, with particular emphasis on decision-making and strategizing practices. After some iterations, it became increasingly apparent (as expressed in various workshops and meetings) that participants grappled with two main challenges: (1) reaching consensus around what the ultimate objective of the project is, i.e., what Palace should look like in the future, and (2) how to develop a plan of work and a governance framework that enable them to deliver the program.
The analysis therefore focused on specific events and occasions to identify the situations that brought those tensions to the fore, the participants' experience with the subsequent challenges in their daily practice, and the unfolding of events over time. In doing so, two analytical strategies were applied. First, the author focused "up close" with day-to-day practice (Van de Ven, 2007, p.10, 77-79), for example, by carefully examining various exercises for envisioning the program outcomes, paying particular attention to the articulation of goals and how participants interacted. Second, in parallel to up close engagements, the analysis followed Van de Ven's (2007) advice to study the research problem "from afar" by discussing the prevalence of the research problem in a wider context and considering how various events corresponded to different patterns of behavior. Drawing on insights from pragmatism (e.g. Emirbayer & Mische, 1998;Farjoun, 2010;Farjoun et al., 2015), the research substantiated the tensions in the flow of activities of organizing and strategizing. As mentioned earlier, these challenges and how participants dealt with them were not covered by the extant project management literature on the front-end. For example, the shared sense of urgency during the conception of the project created an impetus for action, but also masked the divergent opinion of decision makers, which would surface again at a later stage. Similarly, the findings showed that the ambitious vision of the program inspired participants and ensured their sustained engagement with the project, however, abstract and imaginary ideas and goals made it difficult for them to create an agreed upon frame of action. Further iterations between theoretical conjectures (e.g. conceiving dynamic constellations of events connected in co-constituting and non-dualistic cycles) and empirical observations led to conceptualization of two main challenges during the front end of the program: dualities of mobilization and transformation.
Stage 3. The research then turned to consider to how project participants collectively tackled conflicting imperatives in the front-end and constructed a path of action forward. Evidence was gathered about patterns of strategies and responses that centered on activities and practices around planning (including articulating the future visions, objectives and goals), organizing and developing (governance) frameworks for action. In this process, it was possible to identify how participants developed their understanding of the challenges as they alternated between emphasizing the urgency of the work to mobilize action and ambitious and transformative vision of the project to inspire and sustain participation. A shift in their approach became apparent as participants came to realize the contradictory nature of the tensions. It was only at this stage that what could a shift to what could be considered as a pragmatic inquiry and abductive reasoning was observed. The findings clearly indicate how, although over a course of a few months, leaders started to questions and reflect on the taken-for-granted practices of project management. In response they changed the governance framework to more hybrid form that could accommodate both sides of the dualities. The case narrative was structured to reflect these processes.
Stage 4. The final stage closely examined the process dynamics to understand the shifts from various challenges, considering how various organizing and strategizing practices, governance mechanism, and decision patterns could describe overall processes. Insights from pragmatisms and related literature (e.g. Denis et al., 2011;Smith and Besharov, 2014) inform relationships between constructs. The data and literature were incorporated in a final process model as graphically depicted in the Discussion section.
Finally, the author was also fortunate to be able to have several meetings with the leaders of the R&R to share interpretations and exchange ideas. These discussions have been very valuable in the evaluation of alternative explanations and verification of the interpretations presented in the present study.

Introduction to the case study
Following the fire of 1834, the Palace of Westminster was designed to be the purpose-built home of the British Parliament. Since its completion in the 1860s, the Palace has remained one of the world's most famous and symbolic buildings. However, the building has experienced a long period of decay and dilapidation. Despite the evident need for renovation of the Palace by the early 2000s, the question of whether a major restoration should take place remained controversial and divisive. Those against a major refurbishment were concerned about wasting taxpayer's money when other public projects might be more worthwhile or worried that any delay would leave Parliament located semi-permanently outside the Palace. Those in favor of major works, by contrast, were not just concerned about preserving the building's heritage status, they believed reconstruction was a chance to transform Parliament to keep pace with future demands and create a Palace of Westminster for a new era. Irrespective of whether they were against or for a major refurbishment, those involved wanted to minimize disruption to the business of the Parliament. However, there were at least 40 reports of fire between 2008 and 2012 and a growing number of Parliamentarians recognized that the situation could no longer continue.
In January 2018, after more than two decades of debates, surveys and feasibility studies, Parliament voted for the "R&R" and decided that both Houses of Parliament would entirely move outor "fully decant"and return on completion of the work. Following the approach used to deliver the London 2012 Olympics and Crossrail (London's new Elizabeth railway line), Parliament agreed that a two-tier governance structurewith a Sponsor Board and Delivery Authorityshould be established when the program become "substantive". As soon as Parliament's intention to legislate became clear, the two organizations were established in "shadow" form to prepare for the launch of a large and complex multi-billion pound program. This study was motivated by the need to understand how members engage in the front-end preparatory phase, what challenges they face, and how they overcome those obstacles and create a path forward.

Findings
Although the phases described below overlap in reality, they are presented as independent phases to highlight the two distinct, yet interrelated, challenges participants faced during organizing the frontend of the R&R program. In the final section of findings, we present how participants could cope with these challenges.

Challenge#1: What is the job? The risk of entrapment and delays
The first challenge identified in the case denotes the uncertainty surrounding the understanding of the objective of the program. The subjective stakeholders' interpretations about the future of the Parliament constituted a plurality of alternative possible states, rather than clearly-defined objective.
Between 2012 to 2016, drawing on over a decade of evidence and reports arguing for major works, a series of influential reports were published by the Parliament authorities and external expert consultants. These reports, despite being ineffective in rendering any decision, laid the foundations for what to later become the R&R program by exploring the possible scope of the program beyond the restoration work and setting out possible (hypothetical) outcomes. See Table 4 for the detail description of the reports.
Despite the lack of clarity of objectives of such a program of work, the Joint Committee's report was a galvanizing moment for the inception of the R&R program. It showed for the first time that the work could be accomplished without disrupting the work of Parliament. There were two decant options for moving both Houses of Parliament and their staff into temporary accommodation not far from the Palace. The report also suggested a two-tier governance structure (Sponsor Body and Delivery Authority) -similar to the arrangement used for recent high-profile projects in the UK.
At this point, the committee realized that little or nothing would happen unless they were able to create a momentum needed to support the proposed plans. A series of briefings and presentations provided evidence for the program of work, including taking politicians, senior staff and other key stakeholders on tours of the basement of the building. By showing people the actual state of the building's disrepair, the Recognizing that a "doing-nothing" approach was no longer feasible. The report drew on the three outcome levels from the IOA, but it went further by outlining some of the potential areas for improvement that did not easily fit into the three outcome categories, such as rethinking "public access and visitors' facilities", "accessibility", and "working environment". Highlighted a "a one-off opportunity" to renew and transform the Palace of Westminster into a home fit for a 21st Century Parliament, while preserving the best aspects of its fine Victorian heritage.
committee were able to raise awareness and promote the immediate need for a major undertaking. There was a growing acceptance that it was becoming increasingly unsafe and in need of urgent repair. As one director explained, the continuation of the status quo without remedial work threatened to disrupt critically important Parliamentary business such as the debate and vote for the UK's withdrawal from the EU. A sense of urgency was reinforced and amplified by the press and media publications (e.g., The Guardian, BBC, The Times), who warned about the risk of a catastrophic incident and added impetus to the growing momentum for the R&R program to proceed as rapidly as possible.
In January 2018, after more than two decades of preparation, the House of Commons voted (by a majority of only 16 votes) for both Houses to fully decant during construction and return on completion of the work. This event is critical for understanding the unfolding of the front-end processes. First, the sense of urgency stimulated a pressure that "something needs to be done" and a decision reached, after decades of delays and postponements. The narrow margin in the vote, once again, indicated the divide between those who voted against the R&R programemphasizing the huge costs of a major refurbishment and were uncomfortable with moving Parliament to another locationand those in favour who emphasized the risks of a "looming disaster" and "catastrophic failure". Second, the backbench amendment that gained the vote was essentially foregrounding the discussion around the issue of full-decant, while what the program was set out to deliver (i.e., objective of the project) remained unclarified in the background. As one director explained, insufficient attention was paid to identify desirable outcomes for the program: "we knew something needed to be done, we didn't quite know what it's going to be." After the approval of the R&R program, work began on establishing the two shadow bodies responsible for R&R program preparations. Established in 2018, the shadow sponsor body was comprised of Parliamentarians and senior staff representing the interests of the House of Commons and House of Lords. It was responsible for appointing the shadow delivery body, including BDP, an architectural practice, Jacobs, a large program management firm, independent consultants and an inhouse team including architects, engineers, heritage and conservation experts and maintenance staff. Within their pre-assigned roles, the two shadow bodies were jointly responsible for establishing substantive organizations with the capabilities needed to start delivering the program in 2020.
The Palace building had to be designed to adapt to the future needs and requirements of the Parliament. To this end, BDP carried out a survey and asked some of the world-renowned leading experts about the technologies likely to be incorporated in the building, and what the future is going to look like. A manager explained that they soon learned that experts could not look beyond 5 to 10 years ahead, and beyond that it was just speculations. While there are many concepts "out there" of what might happen, he explained, to concretely working out what the future is going to look like is extremely difficult, if not impossible.
Between Summer 2018 and Spring 2019, in an attempt to define and clarify the outcomes of the program, a series of collaborative workshops brought together Parliamentarians and members of the shadow Sponsor Board, to gain a clearer understanding of the purpose and future direction of the program. Participants in the workshops developed four "outcome levels" from level 0 to 3. Outcome Level 0 was the "base case" required to maintain the current state. Outcome Level 1 referred to the strict minimum requirement "needed" to improve the state of the place. Outcome level 3 identified the maximum improvement imaginable. While participants were easily able describe outcome levels 1 and 3, they found it more difficult to envisage outcome 2 -Parliament renewed because it described a feasible, realizable and yet aspirational end state.
Guided by the R&R vision statement -"to transform the Houses of Parliament to be fit for the future as the working home for our Parliamentary democracy, welcoming to all, and a celebration of our rich heritage" (National Audit Office, 2020: 18) -participants in the workshops developed six strategic themes: "Time & value for money", "Sense of History", "Functionality and design", "Health, safety and security", "Accessibility and inclusions", and "Sustainability". Attempts to consider the purpose and objectives of the program helped to inspire and galvanize various stakeholders. Participants were asked to imagine more concretely the outcome they wanted to achieve, using various creative practices (e.g. drawings, analogical thinking, and brainstorming) to interpret, articulate and produce an image of the program's end state. They were invited to develop categories and statements to describe desired outcomes of the program such as a "create a secure, efficient, flexible and accessible workspace for now and future proofed," "recognize the value of the building and its heritage collections by safeguarding, conserving and enhancing," or "optimize the environmental impacts". There was a considerable discussion, excitement, and some disagreement amongst participants in various workshops. Participants had long, in-depth discussions about whether the new proposed design should remain faithful to the traditional mid-19th century structure and what parts should be transformed to meet Parliamentary requirements in the 21st century. Participants concentrated on developing a bold and innovative vision of the future, whilst protecting a rich and symbolic heritage of the building. However inspirational and engaging, the abstract and general nature of the statements hindered the development of clear, concrete, and practical guidance for those preparing for an outcome that should endure for at least 150 years without requiring another large-scale refurbishment.

Challenge#2: How to deliver the job? The risk of strategic misalignment
The second challenge in organizing the front-end of the R&R program was the difficulty of aligning the strategic objectives of the program with the delivery functions and capabilities that need to be in place before commencing with the actual program of work. The shadow sponsor and delivery bodies were directed by the Program Leadership Team (PLT). The sense of urgency that created a momentum for program initiation in 2018 now manifested in various exercises to create deadlines and milestones to reach the Substantive state. For example, 60-Day plan (completed in December 2018) and a 240-Day plan were initiated to develop the first baseline for the program. Since there was limited time before the shadow bodies would become substantive, the PLT was under intense pressure to define the R&R program structure and objectives. Drawing on the previous studies (Pre-feasibility and IOA), various documents such as Program Execution Plan (PEP), workbreakdown structures, program risk assessments, reporting policies and other policies and processes were produced to prepare for the substantive state in 2020. In practice, however, this work often simply replicated what had taken place on previous programs, rather than making the adjustments required to address the specific challenges and context of the R&R program.
Although PLT members had extensive knowledge and experience gained on other major UK programs, they struggled initially when preparing for the R&R program because the outcome and arrangements for achieving it were so unclear. Weekly Governance Panels and, monthly PLT Strategy meetings were put in place to drive the instigation of program policies and strategies.
Following one of the monthly strategy meetings organized by the PLT, participants engaged in a "future perfect strategizing" exercise. Referring to the strategy that worked well in the London Olympics 2012 project, participants were invited to draw on the strategic themes of the program, think beyond the current situation, and use future-perfect thinking to imagine a successful end-state when the refurbished palace would be "fully open for business and used by stakeholders". Working backward from various end states, participants were identified the particular course of action needed to achieve them.
The exercise showed that participants held surprisingly divergent views, varying assumptions and interpretations of what should be done.
Whereas one participant wanted to narrow down options identified by the groups, others wanted more open and expansive discussions about the difference between "outputs and outcomes" or "transition and transitioning". Ideas about how to organize the front-end activities differed in terms of the nature of the work (e.g. classifying assumptions based on restoring the old and constructing the new builds), function of the various spaces in the buildings (e.g. education facilities and meeting places), and governance required to deliver a desirable future (e.g. the clarity of the governance structure and capacity of the supply chain).
Groups in the workshop were also asked to narrow down their various interpretations and convert them into a few manageable categories to address scope and requirements, business case and funding, and program design and delivery. Categories of activities were placed in chronological order and various strategic models were developed to organize and coordinate the work. For example, one model depicted a linear progression of activities while another one represented an iterative cycle, which many believed was the most accurate representation of their real-time experience. While one senior manager found the whole process "too abstract", others wanted greater clarity to assist in the planning and definition of program tasks, with clearer start and end dates. After a few rounds of discussion, there was little agreement on the "method" use to organize tasks. Although many hoped that this exercise would clarifying the end state for the R&R program, participants could not reach an agreement because their interpretations were so subjective, fragmented and insufficiently concrete to guide what needed to be done to achieve it.

How can we know what the task is? Moving forward through constructing 'spaces of experimentation' and 'reconfiguring boundaries'
From April to June 2019, members of the PLT and shadow bodies shifted from efforts to articulate program outcomes and roadmaps to reach those outcomes to a new focus on developing a shared understanding and reaching a consensus on how to move forward in the face of uncertainty. There was an emerging realization among the leaders about the "fuzzy", messy and ambiguous character of the situation, where the conventional approach of "let's just get on with it" no longer applied. Senior managers wanted to know "what is the task and how can we deliver it?", whereas now they faced the more fundamental challenge of cognitively reframing the problem and figuring out "how can we know what the task is?". According to one director, the conceptual outcome of the program was ambitious and based on whole life value of the project and long-term benefits. But there was still no shared view or agreement about the meaning of those conceptual ideas and no plan to achieve it.
The PLT organized workshops to develop a strategic framework comprised of program workpackages and schedules so that they were able to progress the work. Participants were encouraged to listen to each other's perceptions, transcend divergent opinions, achieve a common understanding and build the consensus on how to move to the substantive state. One director reflected on the new approach: "This [referring to uncertainty and ambiguity] is the nature of the project at this stage. It is all about starting point and then further definition, and once you get to enough further definitions you can actually build something. The further definitions evolve, it is like a design process. Now the same principles apply to design of a building as to design of an organization: to be able to say what does good look like, you've got to go and do some work first. So, they have their thinking: what does a good organization look like? What are the systems and process I think I might need? Let's write those down, OK! You compare that against the actual work, and you go through the cycle of maturity and eventually you get to a product, hopefully, that says: that's what we want; that's what we need; that's what the world needs; let's do that. But that creative side, you know, some get stuck, they dream stuff and then it falls down because they don't put that dream into action." In a half-day PLT workshop held in May 2019, program leaders were encouraged to think of both the desired future outcome and current circumstances and priorities. To achieve this, they had to start thinking in terms of the interplay between the present in terms of "the definition and validation of the work" (including categories such as next generation building services, restoration work and new construction) and the future in terms of "conditioning of the work" (including program's strategic themes). A two-dimensional figure was created to visualize the problem. One dimension addressed what had to be accomplished to achieve the program baseline and minimum requirements to restore and maintain the Palace. The other dimension identified the program's aspirations to achieve more transformational outcomes. A conceptual "space" was constituted at the intersection of the two dimensions, representing the "actual projects to get the work done". Nine workpackages were developed by challenging existing assumptions about the attributes, uses and functions of the building (e.g. building services, heritage, and education facilities) and sources of failure and improvements. This work helped participants understand how aspirations to achieve the program's strategic themes at a future date had to be grounded in present circumstances, such as the capabilities required for "defining and validating" and "exploring and conditioning" the work. Projecting the present understanding of the work into the future allowed actors to engage with the abstract interpretations and themes to envision future outcomes, while moving forward with actionable and viable strategic frameworks grounded in current concerns, practical priorities and expertise.
The importance of the new approach was underlined when participants said that it provided a "vehicle", "platform" or "machinery" to deliver the program in a highly complex, contested and uncertain environment. It provided leaders with a safe space to repeatedly reinterpret the meaning of the program strategic themes and experiment with varied operational practices associated with different meanings. Whereas in most projects managers seek to narrow options, achieve predictability and freeze the plan as early as possible, having a space for experimentation during the front end encouraged participants to create a strategic framework that was open, adaptable and responsive to unexpected circumstances and changes in future conditions. During this period, the attention of the PLT shifted from trying to define the program outcomes to focus more on progressing to a substantive state. Constructing a shared understanding of what was required to become substantive became increasingly apparent as participants with diverse and often conflicting perspectives began to interact, voicing their concerns and preferences. For example, one member to know what the difference was between "the plan to get to the substantive state versus the plan to get to what the outcome is." In the same vein, members engaged in various discussions about "what the actual job [end state] is?" versus "the organizational capabilities required to deliver the job". In this process of reflection and collective sensemaking, the PLT gradually integrated, refined and established the strategic framework required to achieve a substantive state. A senior manager considered the new approach was quite nuanced in the way in which it accommodated differences of opinion and tensions. Members agreed on the following definition (noted in an internal document) of the substantive state: "the levels of organizational maturity required to discharge all obligations placed on the respective entity".
The research findings revealed another major shift in this period. The PLT had originally assumed that the Sponsor Body was solely responsible for envisioning the intended outcome, specifying the scope and identifying the objectives of the R&R program, whereas the Delivery Authority would organize and execute the program as planned. But given the ambiguity surrounding the two-tier structure of the program, the PLT recognized that a more collaborative approach was required and informally agreed to dissolve the pre-defined boundaries and merge the two bodies into a single, co-located management team for the remaining period to forge the collaboration necessary to achieve the shared understanding, clarity of purpose and focus needed to transition the substantive state and prepare for program delivery. One senior director describe it as follows: "We don't have a program that is fully defined, so we are not yet at the stage to operate separately because there are many common functions and tasks that require joint accountability and endeavor. We join together [at this stage], but when we get to substantive stage we separate, beautifully coordinated, and we can actually join together again if we need tocall it 'the Velcro model'".
By the end of the third period, the joint shadow team constructed a shared understanding of how to prepare for the substantive state. It developed strategic accounts that were concrete enough to be practically feasible in present circumstances, but sufficiently abstract to pursue ambitious goals in ways that challenge and transform current arrangements.

Discussion
This study examined how actors cope with seemingly contradictory imperatives in the front-end of projects and construct a path of action forward. The case illustrates how a project's inter-generational time horizons, fundamental uncertainty, and high diversity limit, if not foil, attempts to predict the future and to understand various alternatives and their consequences. The findings show that to mobilize collective action and to initiate a project, key actors may create a sense of urgency around a shared issue or cause. Creating a shared sense of urgency fosters progress by allowing participants with diverging interests and priorities to converge around a common goalsomething needs to be done. However, to maintain engagement over the long period of time, participants may engage in imaginative practices and future possibilities to inspire a variety of stakeholders to commit to the project. Yet, the dormant tensions surface over time when different forces contradict rather than complement one another, such as attempts to seek consensus when participants have fundamentally opposed interests and conflicting agendas, and efforts to form a bold vision of the future and the need for realistic and achievable path of action.
Extant literature on front-end management recognizes these tensions as the major challenges facing large complex projects (Edkins et al., 2013;Williams & Samset, 2010). However, studies often emphasize one side of this dichotomies. For instance, the assertion of "failure" (with the benefit of hindsight) implies some pre-determined benchmark of success against which the strategizing efforts are evaluated. As the result, much of the research portrays ambitious and often unachievable goals as the result of human biases, shortcoming in methodologies and deliberate strategic misrepresentations (e.g., Flyvbjerg, 2009;Williams et al., 2019). Actors are therefore advised to avoid or seek to correct "fallacies" and dysfunctional behavior which lead to lack of clarity and highly ambiguous project objectives.Whereas the consequences of such imaginative and aspirational practices for engaging and orienting collective action and initiating a project, as illustrated in the R&R case, have been largely overlooked (Engwall, 2002).
Instead, a pragmatist approach directs attention towards unexpected rather than pre-determined outcomes, and emergent rather than rational decision-making and organizing processes (Simpson, 2009). Pragmatism's antidualistic stance encourages scholars to avoid a one-sided focus on one category over another. Pragmatists argue that this theoretical distinction (or implicit dualism) creates a barrier to understanding project formation and development because it misses how various strategies are recursively linked and mutually constitutive of one another (Farjoun et al., 2015). As Simpson (2009) points out, this is not to suggest that pragmatist view of practice is unconcerned with outcomes and the results of strategizing; but it challenges the dominant view of the front-end as a stage-gated, goal-oriented, rational process where the focus is purely on designed or intended outcomes. The pragmatist approach is also concerned with emergent outcomes (see Table 2).
In this vein, the pragmatist perspective presented here complements the project-as-practice literature by offering a way of engaging with "how" practice emerges and changes, tensions surface and are experiences by participants and the consequences for organizing the front-end, rather than "what" practices are in use (Sergi et al., 2020). Our findings illustrate that to tackle the challenges, leaders alternate between the forward-looking projection of the outcomes and practical evaluation of the situation to evoke action. Over time, however, as project leaders emphasize one strategy over another, latent tensions surface and demand navigation. As explained in the sections that follow, ignoring the contradictory nature of front-end organizing can result in undesirable outcomes including entrapment and inaction and strategic misalignment. The findings illustrate that embracing these complexities, being open to alternative perspective, and ability to cross establish boundaries to navigate this territory will be a key to unlocking the innovative potential of project organization. Over time, leaders develop reflective capabilities when they realize that established habits, assumptions and old ways of thinking and acting are no longer efficacious, which in turn enables the participants to apply iterative approaches to coordinate front-end activities. Applying this pragmatist lens, the following sections further discuss the main challenges that participants faced and how they managed to engage with tensions and overcome those challenges.

Duality of mobilization
The first tension that exists in organizing the front-end stems from efforts to mobilize a diverse group of actors, each with significant influence on the strategic direction of the project, to engage with and commit to the cause over prolonged timespans. Given the long-term horizon of large projects, initiating, and maintaining collective action in such contexts may enhance a duality of mobilization. The strategies that help to converge and mobilize diverse participants in the project's early stages generate a misalignment of interests and trigger divergence in later stages that may eventually impede collective action.
Our findings illustrate how key actors strategically created a shared sense of urgency to evoke collective action. This was achieved by focusing attention on preventing a detrimental outcome and generating a pressure to act swiftly. Reporting, lobbying and other evidence-gathering activities were instrumental in amplifying the urgency of the situation and drawing attention to the building's visible disrepair. By associating failing to act with detrimental outcomes (e.g., risk of fire), actors could overcome inertia and stimulates calls for action (Slawinski & Bansal, 2015). This was a convergent process, consolidating diverse group of people and organizations (Gil, 2017).
Despite the underlying divergence of opinions, formal approvals (i. e., Parliamentary vote in 2018) and efforts to amplify the importance of the project in public discourse (e.g., emphasizing the heritage, prestige, and symbolic value of the Palace) contributed to a situation of escalating commitments (Cantarelli et al., 2010) and irreversible decisions. While these strategies created a strong impetus for action, however, they also obscured the fact that there many different views and ideas about what actually needed to be done (Challenge #1 in the case). Such situations create a fertile ground for the "escalating indecision", identified by Denis et al. (2011) as one of the main decision-making pathologies in public projects. The tendency to render decisions acceptable to participants may actually mask a growing divergence of opinions where several of the most challenging points remained vague and unsettled. In such situations, where the goal of the project is not yet explicitly defined or articulated, participants continually "make, unmake, and remake strategic decisions", resulting in prolonged efforts with little concrete strategic action, "the constant possibility of reversal or reorientation, and potential widening scope of decision activity" (Denis et al., 2011, p.225, 237). Thus, leading to "decisions" that embed the need for more decision making. Escalating indecision is, therefore, a key challenge during front-end project organizing.

Duality of transformation
The second issue stems from the tension between the transformative objectives of the project and feasibility of achieving those objectives. The research findings show that the front end, where the future objectives are uncertain and highly contested, provides a fertile ground for previously unappreciated duality of transformation. The imaginative practice of envisioning transformational future outcomes based on abstract and hypothetical ideas (which provides inspiration for a diverse group of stakeholders to engage with the project) may also become an obstacle to the creation of a strategic framework and concrete action.
The R&R case illustrates how by engaging with possible project outcomes, actors were able to transcend the here and now, and move beyond current discourses and experiences. This is in line with Zerjav et al. (2021) who highlight that to find a common ground between different stakeholders, it is important to focus on the value of the transformational potential of the project, rather than execution efficiency. In the series of imaginative exercises, participants in the R&R program were encouraged to challenge existing assumptions and creatively consider how Parliament might successfully operate, function and be used by future generations. The forward-looking thinking underpins projective efforts to engage with future end states and creates the prospect of achieving a transformative change, rather than efforts to preserve the current situation (Alimadadi et al., 2021;Gil & Fu, 2021). Such activities encourage participants to look beyond the status quo, think creatively, and imagine hypothetical futures (Mische, 2014). Forces of aspiration and hope for improvement encourage actors to take on challenging endeavours of delivering a large project that might otherwise seem costly, risky, slow, frustrating, and often painful (Alimadadi et al., 2021).
Images of desired outcomes, especially in a distant future, however, are often abstract and general, tied to broader ideologies and values, rather than concrete and detailed instances (Augustine et al., 2019). The case shows how participants frequently resorted to vague and general terms when trying to imagine the future, such as the heritage and symbolic value of the Palace as the "emblem of parliamentary democracy". The increasing degree of abstraction used to imagine aspirational outcomes created wide-ranging appeal among actors as they envisaged various possibilities, critically evaluated current approaches, and created the conditions for transformative change. Abstract features and uncertainty associated with project outcomes may, however, prevent immediate coordinated action and hamper efforts to create a feasible path of action (Challenge #2 in the case). This, in turn, increased the probability of misalignment between envisioned project outcomes (e.g. strategic themes) and the tactical concerns and current priorities of the project (e.g. managing program baseline).

Overcoming the tensions
To avoid escalating indecision and misalignment between the program's objectives and ongoing activities, participants developed an appreciation of the underlying tensions and applied novel approaches to mobilize action and achieve the strategic objectives. The research provides a dynamic understanding of field-level strategic actions, taking into account how participants' understandings evolve as they engage with various organizing practices and dynamically try to find a way forward in the face of uncertainty. This was achieved in the R&R case by efforts to strike a balance between these opposing forces, as actors created spaces of experimentation and constructed flexible boundaries.
The R&R case provides new way of understanding of front-end project organizing. Participants may benefit from mutually enabling qualities (Smith & Lewis, 2011) if they are able to manage the tensions and conflicting demands of urgency for action (manifested in program baseline by various deadlines and milestones) and ambitious visions for the outcomes. As tensions arising from dualities of mobilizations and transformation surfaced and were experienced by participants, actors gradually increased their reflective capabilities to understand, reframe and address the (Farjoun et al., 2015). This interpretive shiftor reflectivitydemonstrates an ability to question taken-for-granted "habits of thought" (Emirbayer & Mische, 1998) and conventional approaches -"let's just get on with it" -to program management. It is important to note that the shift occurred as an emergent outcome of the interplay between urgency for the work to progress and inefficiency of fallible habits, established routines and predispositions in the face of uncertainty. However, it was only when the puzzling situation created doubt among participants about the (effectiveness of) existing arrangements and routines that they started to engage in a process of pragmatic inquiry and "abductive" reasoning towards their work (Mantere & Ketokivi, 2013;Peirce, 1878). This process is driven by experimentation and collective learning (Dewey, 1938), enabling actors to reflect on where they are going, where they want to go, and how they can move from the present to the chosen destination.
Participants created "spaces of experimentation" that allowed them to explore and address strategic tensions without drifting toward one at the expense of the other. Two dimensions of this conceptual frame created metaphorical "guardrails" consisting of themes and categories, each associated with one side of the duality (Smith & Besharov, 2019). Guardrails prevented participants from losing sight of the visions for the future and what had to be accomplished to progress, while meeting the minimum requirements of restoring and maintaining the Palace. Within this bounded space, participants constructed workpackages addressing these imperatives and involving experts (both individual and organizational) such as project managers, heritage experts, architects, historians, Members of Parliament.
This approach to front-end organizing created "workable certainties" for participants and enabled them to move forward and explore the scope of the program despite the lack of clarity about the program objectives (Lüscher & Lewis, 2008;Weick, 1995). Instead of making long-term commitments to a certain course of action or conception of the project's end state, leaders adopted provisional approaches that shifted and evolved over time (Andriopoulos & Lewis, 2009). The framework facilitated experimentation with practices and supported creative thinking (Smith & Besharov, 2019).
The approach presented here is inherently different from the dominant view of the front-end wherein actors are concerned with the choice of concept and path dependence (e.g. Samset & Volden, 2016;Williams et al., 2009), rather than sensemaking and path creation (Garud & Karnøe, 2001). This approach offers an alternative perspective grounded in the interaction between stability and flexibility, and between needs and ambitions. Instead of premature attempts to fix the plan or to freeze the design too early, through "small wins" actors could collectively legitimize change in previously institutionalized ways of working (Reay et al., 2006;Weick, 1984). Small wins are defined as "a series of controllable opportunities of modest size that produce visible results" (Weick, 2001: 427).
The case illustrates how over time actors questioned their habits and shifted their attention towards on what pragmatists term "abductive" approach in which plausible explanations are inferred from specific observations and general principles (Bartel & Garud, 2003;Mantere & Ketokivi, 2013). Experimentation to generate small wins is an iterative and evolutionary process: "When a solution is put in place, the next solvable problem often becomes more visible" (Weick, 1984, p. 43). As leaders become more aware of the benefits of spaces for experiments, they can proactively build the formal organizational structure to create the guardrails to manage the tensions (Smith & Besharov, 2019).
Furthermore, our findings revealed significant efforts that were required to overcome the inertia and constraints imposed by prevailing structural and practical dispositions (Gilbert, 2005;Leonard-Barton, 1992). At the outset of the R&R program, adopting the governance arrangement used successfully for other large infrastructure projects in London created a suitable structure for the participants to initiate the work. Collective efforts were required for actors to overcome established boundaries and reflect on connections between the different parts of the project organization, whilst focusing on establishing the capabilities required to achieve a large, complex and strategic endeavour. For example, the hierarchical relationship between the shadow bodiesthe sponsor and the delivery organizationrestricted efforts to mobilize actors and attract the resources required to adhere to their responsibilities as they perceived necessary (McCarthy & Zald, 1977). The space for experimentation resulted in a new, and more collaborative arrangements to facilitate the process. To support the new approach, new structural arrangements such as cross-boundary groups are essential in order to bring together participants with different expertise and roles to work on specific problems.
New strategies were created as the aspirational and functional aspects the program became mutually constituted through interaction. By alternating between planning and imaginative orientations, participants in the program eventually managed to create a hybrid frame for experimentation that was sufficiently flexible to facilitate efforts to think abstractly, open up possibilities and create room to move beyond current constraints and exigencies, whilst being firmly grounded in practical concerns and priorities needed to create momentum for concrete action. These processes are inherently relational, dynamic and recursive. Fig. 1 illustrates how the findings were analyzed and connected to conceptual constructs.

Conclusion
This research aimed to study how actors creatively resolve conflicting imperatives in the front-end of projects and construct a path of action to move forwards. The study responds to the call by project scholars (e.g. Buchan & Simpson, 2020;Cicmil, 2006;Tywoniak et al., 2021) to highlight the opportunities that the pragmatist perspective could offer for revising the dominant execution view of front-end project organizing. The main theoretical contributions are as follow.
First, this study contributes to the literature on the project's frontend (e.g Edkins et al., 2013;Williams & Samset, 2010;Williams et al., 2019) addressing some of the main challenges that project participants face during this period. The pragmatist approach applied here seeks to transcend the dichotomous assumptions between means versus ends, abstract goals and feasible, and challenge established theoretical and empirical assumptions. Rather than favouring one side of these dichotomies at the expense of the other, it values both as complementary aspects of what we have called the dualities of mobilization and transformation. The present study illustrates how both aspects of these dualities can advance our understanding of the often overlook dynamics and consequent challenges of engaging a diverse group of stakeholders to both mobilize and maintain collective action in the face of considerable uncertainty and urgency.
In line with Sergi et al. (2020) and Brunet et al. (2021), the findings contribute to the front-end management literature by showing how "performativity" (Gond et al., 2016) of various strategies and organizing practiceshow project participants utilised a sense of urgency as a convergence mechanism, and later applied imaginative practices to expand the alternative outcomesgives rise to and sustains collective action at different stages of the front-end. However, pragmatism reminds us that it is only through reflection and proactive engagement with the tensions that leaders could overcome challenges. By tapping into the potential synergistic power of contradictory goals, project leaders can create a strategic framework to engage participants with diverse views to explore flexible future outcomes, while ensuring actionability. In this vein, the tensions stemming from dualities of mobilization and transformation can be perceived to be the driving force during the front-end of projects that leads to a change and transformation of the practices, the fuel for the dynamics of organizing the front-end. The conceptual model illustrates the complexity of organizing the front-end from an alternative perspective by explaining how renewal and transformation occur amidst structural constraints, and how tensions arising from dualities allow for experimentation and novelty which can enable and generate impetus for change of existing processes and practices, not merely undermine them (Dewey, 1922).
Second, the present study contributes to the growing body of literature exploring projects from a pragmatist perspective (e.g. Lalonde et al., 2010;Sergi, 2012;Simpson et al., 2018). Building on the existing work on actuality of projects , this study shows how pragmatist perspective can enhance our understanding of the challenges of organizing front end as experience by the participants. By situating various practices within a broader timeline and by showing how certain practices may come to be interconnected over time, the findings complement and extend our understanding of how various organizing practices and processes change and evolve as project participants grapple with constructing a viable path of action. The research also adds to studies that highlight the role of pragmatic inquiry in projects (e.g. Buchan & Simpson, 2020;Lurino, 2018) by highlighting the dynamics of collective action under plurality of interests and uncertainty about the future outcomes, situations that characterizes the front-end of projects. Figure 1. Conceptual structure of the findings While we concur with the pragmatist studies that the breakdown of established habits and the disequilibrium of a situation constitute the starting point for reflection (Buchan & Simpson, 2020) the result of this study adds to these insights by identifying the underlying tensions that arises as the result of plural identities, diverse (and often opposing) interests, and diffuse power (Denis et al., 2007) of various participants in the front-end. Under such circumstances, it becomes more challenging for actors, as a collective, to notice a situation that could be characterized as abnormal and puzzling (Weick, 1995), before (if at all) being able to collectively make sense of it and develop reflectivity. The conceptual model presented here translates the pragmatist ideas into the context of organizing front-end of a project by placing the dynamic view of collective action at the heart of understanding how project participants engage in abductive processes to create consensus and restore unity.
The paper has important implications for practice. Applying a pragmatist perspective may encourage practitioners to see the recursiveness in their actions and how decisions change the context for their own actions. Through reflective practices and engaging with tensions, rather than abandoning them, practitioners may realize new possibilities. Pragmatists encourage managers to face complexity and proactively explore tensions by tapping into the potential synergistic power of seemingly opposing strategies (Langley & Tsoukas, 2012;Weick, 1979).
A limitation of this study is that it primarily focused on capturing the processes of front-end organizing as they unfolded. However, the openendedness of this approach, in contrast to retrospective studies, can be problematic as the focus of action may shift and different outcomes might emerge as the project evolves gradually. Therefore, in future work, scholars might examine the effectiveness of the strategies discussed here in achieving the project objectives. That is, how constructing a sense of urgency or using imaginative exercises may be applied to enable collective learning, trigger, and develop reflective capabilities and apply iterative approaches to coordinate front-end activities. Furthermore, while recognizing the plurality of project participants implies that importance of their positions and the distribution of power among the leadership team, the present study did not explore what role the power structure plays in the process of front-end development, and when and how leaders values and preferences shape the project. Therefore, in future work, scholars might also examine the framing and meaning-making activities (e.g. Kaplan, 2008;Spee and Jarzabkowski 2017) and the role of different actors in creating an agreed-upon action frame during front-end project organizing, paying particular attention to power dynamics in facilitating or prohibiting the emergence of the changes in hierarchical and relational arrangements among participants. Furthermore, scholars might further examine.

Funding
The data cannot be made public due to reasons of confidentiality. The usual disclaimers apply.