See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/374695986

Collaborating on the Edge of Failure: Frame Alignment Across Multiple

Interaction Arenas in Multi-Stakeholder Partnerships

Article in Academy of Management Journal · September 2023

CITATIONS READS

0 152

2 authors:

Julia Grimm Juliane Reinecke

Stockholm University University of Oxford
18 PUBLICATIONS 28 CITATIONS 102 PUBLICATIONS 3,256 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Juliane Reinecke on 13 October 2023.

The user has requested enhancement of the downloaded file.

 1

Collaborating on the Edge of Failure:

Frame Alignment Across Multiple Interaction Arenas in Multi-Stakeholder Partnerships

Julia Grimm
Stockholm Business School
Stockholm University
[email protected]

Juliane Reinecke
Saïd Business School
Oxford University
[email protected]

Forthcoming in the Academy of Management Journal

Citation suggestion: Grimm, J. & Reinecke, J. Collaborating on the Edge of Failure: Frame
Alignment Across Multiple Interaction Arenas in Multi-Stakeholder Partnerships. Forthcoming
in the Academy of Management Journal.

Acknowledgements

We would like to thank our editor Tammar Zilber for her constructive guidance throughout the
review process, as well as the three anonymous reviewers who have helped us shape this paper.
We would also like to thank Florian Überbacher, Shaz Ansari, Paolo Aversa, Dennis
Schöneborn, Christine Moser, Jack Fraser, Jette Steen Knudsen, Jimmy Donaghey and Norbert
Steigenberger, as well as participants of AOM, EGOS, the OSSW, the Edinburgh AMJ PDW,
OTREG, and LOST for their feedback on earlier versions of this paper. We also want to thank
colleagues at the University of Cambridge, Bayes Business School as well as the University of
Bath for their helpful feedback during seminars there. Finally, we thank our anonymous
respondents for supporting us with their insights, and Matt Jones for his help proofreading this
paper.
 2

Collaborating on the Edge of Failure:

Frame Alignment Across Multiple Interaction Arenas in Multi-Stakeholder Partnerships

ABSTRACT
One of the greatest challenges of multi-stakeholder partnerships lies in forging a shared
understanding and obtaining and sustaining commitment among parties representing different
interests and goals. While previous studies have emphasised the importance of developing shared
frames for enabling collaboration and collective action through frame alignment, scant attention
has been paid to how stakeholder representatives can attain commitment from their constituents
“back home” to the frames negotiated on their behalf. Our longitudinal process study explores
how participants in the German Partnership for Sustainable Textiles successfully confronted the
challenge of aligning frames across multiple interaction arenas, highlighting how failing to tackle
this “two-table-problem” can risk partnership collapse. Our process model captures how back-
and-forth interactions enabled the stretching of shared frames across interaction arenas, thereby
propelling the partnership from near-collapse to deepened commitments. While stretching frames
heightens the risk of frame break, our analysis shows how such iterative ongoing efforts are
essential for deepening commitments and advancing collaboration. We thus contribute to
framing theory by highlighting how frame alignment can be achieved across multiple interaction
arenas by “collaborating on the edge of failure”. We further contribute to scholarship on multi-
party collaboration by unpacking the multi-table negotiation dynamics that help explain
collaborative outcomes.

INTRODUCTION
Multi-stakeholder partnerships (MSPs1) are forms of inter-organisational collaboration that

enable companies to engage voluntarily with competitors and stakeholders in collectively

addressing challenges too complex and wide-ranging for any single entity to tackle alone (Doh,

Tashman, & Benischke, 2019; Gray & Purdy, 2018). MSPs are inherently difficult to constitute

and sustain, however, on account of the often conflicting agendas of the parties involved (Selsky

& Parker, 2005). Indeed, companies often only participate in MSPs and engage in collective

action as a result of external pressures, including pressures arising in the aftermath of reputation-

harming events such as the 2013 Rana Plaza catastrophe in Bangladesh (Reinecke & Donaghey,

1
MSPs are defined as “private governance mechanisms involving corporations, civil society organizations, and
sometimes other actors, such as governments, academia, or unions” (Mena & Palazzo, 2012: 528). Other labels used
to describe inter-organisational collaboration include “cross-sector partnerships” (XSPs), “multi-stakeholder
initiatives” (MSIs), and “public-private partnerships” (PPPs) (see Koschmann et al. (2012)).
 3

2015; Schuessler, Lohmeyer, & Ashwin, 2023). Key barriers to successful collaboration

identified in previous studies (Margerum & Robinson, 2016) include collaborative inertia

(Agranoff, 2012), ambiguity of aims (Turcotte & Pasquero, 2001; Zuzul, 2019), and power

disparities among participating organisations (Gray, Purdy, & Ansari, 2022; Purdy, 2012). These

factors frequently lead companies to water down their initial commitments, to engage only

symbolically with partnerships, or even to disengage from them completely once public pressure

has waned.

In view of these barriers and pitfalls, a pressing task for researchers and practitioners is to

identify ways in which diverse parties can negotiate and maintain a shared understanding of

which joint commitments are feasible. Scholars taking up this task have found interactional

framing theory especially useful for gaining insights into “how parties negotiate meaning in

interactions” (Dewulf et al., 2009: 156; Gray & Purdy, 2018; Gray, Purdy, & Ansari, 2015;

Reinecke & Ansari, 2021). By co-constructing collective interpretations through repeated

interactions, it is argued, parties can attain the common ground needed for effective collaboration

(Alvarez & Sachs, 2021; Ferraro & Beunza, 2018). Put simply, this is because interactions over

time can foster “frame alignment” by affording opportunities for diverse parties to converge

sufficiently to achieve “at least a minimal level of agreement about what action to take” (Ansari,

Wijen, & Gray, 2013: 1032; Donnellon, Gray, & Bougon, 1986).

While scholarship has proffered valuable explanations of how frame alignment can be

attained (Helms, Oliver, & Webb, 2012; Rao & Kenney, 2008), these insights have mostly been

related to alignment as the basis for agreeing on the creation of MSPs. Even when successfully

established, however, MSPs all too often fail to generate substantive governance and collective

action (Alamgir & Banerjee, 2019; Barnett & King, 2008), because commitments tend to become
 4

watered down and underlying conflicts among members are prone to reignite over time (Wijen,

2014; Zuzul, 2019).

One explanation for the difficulty of maintaining shared frames and commitment that has

often been overlooked relates to the sheer complexity of inter-organisational collaboration

(Bryson, Crosby, & Stone, 2015). This complexity arises in part because negotiators not only

need to develop shared frames amongst each other but must also gain acceptance for any

negotiated frame from their constituents “back home” (Gray & Purdy, 2018). Referred to by

scholars as the “two-table problem”, borrowing a metaphor used in the negotiation and

diplomacy literature, this challenge entails aligning frames across multiple different interaction

arenas by engaging in “level-two negotiations” (Colosi, 1983, 1985; Putnam, 1988; Sebenius,

2013). Achieving frame alignment across multiple interaction arenas is again inherently

challenging, however, since moving from one table or level to the next can lead to frame

misalignment, reignited framing contests, and possibly frame breakage.

The intrinsic difficulty of aligning frames across multiple interaction arenas is readily

apprehensible if we accept that frames are not fixed cognitive representations but interactional

co-constructions (Dewulf et al., 2009; Lewicki, Gray, & Elliott, 2003) that are thus contingent on

the social situations in which framing takes place (Reinecke & Ansari, 2021). From this it clearly

follows that a frame that emerges in one interaction arena cannot simply be ‘transported’ into

another. To date, however, the question of how a shared frame can be expanded or ‘stretched’

beyond the interactions from which it emerges has received little attention (Gray et al., 2015).

To study interactional frame alignment across multiple interaction arenas, we draw on a six-

year-long study of the German Partnership for Sustainable Textiles (‘the Textiles Partnership’),

an MSP initiated by the German government in response to the 2013 Rana Plaza disaster. Having
 5

originally selected this case as an example of the near-failure of an MSP in spite of considerable

reputational and governmental pressure for it to succeed, we were surprised when the companies

involved later agreed to commitments they had initially rejected, especially as this turnabout

transpired after public pressure had significantly decreased. In revealing how this realignment

was achieved through stretching frames sufficiently to pull parties towards agreement on the

highest common denominator, our analysis highlights the extent to which preserving and making

progress in partnerships entails collaborating ‘on the edge of failure’. This is because the process

of stretching frames inevitably involves the risk of frame break if at-home constituents reject the

compromises made on their behalf. We further elaborate how this risk ensues from the relational

dynamics between negotiators and their constituents; for while we show how cumulative

laminations of relational, issue and process framing can enable frame alignment among

stakeholder representatives, the accretion of these meanings through interactions further

complicates the transferral of frames to constituents. Our process model captures these framing

dynamics, highlighting how effective navigation across multiple interaction arenas can deepen

and renew commitments among collaborating parties.

We contribute to framing theory by shedding light on the dynamics involved in navigating

frame alignment across multiple interaction arenas. By unpacking the role of relational, issue and

process framing that laminate new shared meanings onto frames in frame alignment processes,

we draw attention to the extent to which frames are grounded in the relationships of the

interactants who “do” the framing (Dewulf et al., 2009; Gray et al., 2015; Reinecke & Ansari,

2021). We thus provide a far more dynamic model and account of how shared frames are

continuously (re-)negotiated and stretched across overlapping interaction arenas.

By exploring the “two-table problem” in MSPs, our study also provides an explanation of
 6

how MSPs can transform into more robust institutions of collaborative governance (Ferraro,

Etzion, & Gehman, 2015). Unlike studies that have emphasised the role of external incentives

and political pressure in explaining why companies cede to societal demands (Bartley & Child,

2014), we foreground the role of negotiation dynamics in sustaining such collaboration.

THEORETICAL MOTIVATION
Interest in multi-party collaboration for collective action has proliferated in organisation theory

of late, specifically in the context of grand challenges (George, Fewer, Lazzarini, McGahan, &

Puranam, 2023). While this interest springs from the promise of such collaboration to address

problems too complex for any organisation to tackle alone (Gray & Purdy, 2018), there is broad

consensus that collaborating amongst multiple parties is a complex challenge in itself (George et

al., 2023). Numerous studies have shown how failure to align parties on joint goals can lead to

conflicts that can ultimately derail collaboration (Zuzul, 2019), including research on inter-

organisational collaboration (Hardy, Lawrence, & Grant, 2005; Ring & van de Ven, 1994),

multilateral networks (Human & Provan, 2000), and self-regulatory industry coalitions (Barnett

& King, 2008). This challenge is exacerbated in MSPs (George et al., 2023), where different

interest groups must “integrate their diverse perspectives and goals, which are frequently

misaligned or competing” (Gray & Purdy, 2018: 68).

Scholars exploring how collaboration and collective action can be established in the first

place among multiple stakeholders have focused especially on the communicative processes

whereby parties can attain a common understanding (Koschmann, Kuhn, & Pfarrer, 2012;

Reinecke & Ansari, 2016). The application of framing theory in particular has yielded useful

insights into how diverse actors seeking to work together can reach settlement on potentially

contentious issues by negotiating a shared frame (Helms et al., 2012; Lee, Ramus, & Vaccaro,

2018). Defined by Goffman (1974: 21) as “schemata of interpretation”, frames “promote a

 7

particular problem definition, causal interpretation, moral evaluation, and/or treatment

recommendation” (Entman, 1993: 52). Again, however, establishing shared frames is an

inherently challenging task, not least because different interest groups tend to “strategically try to

win over others” (Cornelissen & Werner, 2014: 197), engaging in “framing contests” (Guérard,

Bode, & Gustafsson, 2013; Kaplan, 2008) in which they vie for their respective constructions of

meaning to prevail (Lee et al., 2018). Only by aligning their respective frames, it is argued, can

parties move beyond such contests (Ansari et al., 2013; Helms et al., 2012; Rao & Kenney,

2008). As Goffman (1974: 127) proposed, this is because frame alignment renders the issue at

stake “inter-subjectively meaningful and understandable” to all parties involved, thereby

enabling a shared frame to be “constructed and agreed upon”. Such alignment can be achieved

either by interacting parties incorporating elements of each other’s frames (Cornelissen &

Werner, 2014; Donnellon & Gray, 1990) or by one or more parties transforming their own

frames (Lee et al., 2018), as a result of “frame shifts” (Ansari et al., 2013; Reinecke & Ansari,

2016). In social movement theory the concept of “frame alignment” serves for explaining how

social movement organisations can strategically frame their activities to create linkages in

“interpretive orientations” that render diverse individual interests, values and beliefs “congruent

and complementary” (Benford & Snow, 2000; Snow, Worden, Rochford, & Benford, 1986: 464).

Most early studies of frame alignment examined this phenomenon at field level (Weber,

Heinze, & DeSoucey, 2008; Zietsma, Groenewegen, Logue, & Hinings, 2017) with a focus on

documenting “framing strategies” that facilitate consensus around field frames (Granqvist &

Laurila, 2011; Meyer & Hoellerer, 2010). For example, Helms et al. (2012: 1124) showed how

parties applied framing strategies in multi-stakeholder negotiations of the global CSR standard

ISO 26000, adopting various negotiation frames as discursive tools to shape the views and
 8

actions of others. In a study of how a shared “climate change logic” evolved among diverse

actors over decades of interaction, Ansari et al. (2013) traced a series of field level frame shifts

whereby different interpretations of the meaning of climate change converged sufficiently to

enable agreement on joint action, further showing how securing commitment on low-priority

issues through “issue linkage” succeeded in breaking deadlock and paving the way for parties to

cooperate on high-priority issues. In another field level study, Lee et al. (2018) highlighted the

role of “strategic frame brokerage” in reaching frame alignment, showing how a Sicilian anti-

racket organisation strategically reconciled the competing meaning systems of anti-racketing

activists and industry actors. Through such strategic brokerage, frames were transformed by

“keying” and “laminating” (Goffman, 1974: 45) additional meanings onto activities to

encompass the concerns of both sides. In this process, any frames that could not be transformed

were partially retained or selectively referred to maintain ideological consistency with the

movement while avoiding framing clashes with industry actors (Lee et al., 2018).

Frame Alignment from an Interactional Perspective

Organisational scholars dissatisfied with the tendency in previous studies to “focus more on the

content of frames than on their use in social interactions” (Leibel, Hallett, & Bechky, 2018: 165)

have since turned to an interactional perspective on framing (Gray et al., 2015; Reinecke &

Ansari, 2021). Conceptualising framing as the “dynamic enactment and shaping of meaning in

ongoing interactions”, these scholars focus on “how parties negotiate meaning in interactions”

(Dewulf et al., 2009: 156). This perspective is especially useful for explaining how competing

frames can be aligned in multi-party collaborations, showing how parties can co-construct

collective interpretations through repeated interactions to negotiate “at least a minimal level of

agreement about what action to take” (Ansari et al., 2013: 1032; Donnellon et al., 1986). Studies

applying this perspective draw attention to the people who “do the work of framing” (Leibel et
 9

al., 2018: 165). For example, Dewulf et al. (2009) highlight how parties frame not only the issue

at stake but also their ongoing interactions, who they are and how they relate to one another.

From this perspective, Ferraro and Beunza (2018) studied how a faith-based investor coalition

first needed to establish a common understanding of their interactions with two major U.S. car

manufacturers as ‘dialogue’ rather than confrontation before being able to establish a shared

language and vocabulary. Ferraro and Beunza (2018) refer to this latter process as “creating

common ground”, whereby interacting parties build up a “sum of mutual, common, or joint

knowledge, beliefs, and [communicative] suppositions” (Clark, 1996: 93; Cornelissen & Werner,

2014). In building common ground, interactants also co-create a common worldview (Alvarez &

Sachs, 2021) that can lead them to greater shared commitment to joint action.

Other scholars have drawn attention to the role of emotions in interactional framing. For

example Gray et al. (2015: 122) have proposed that the emotional energy “generated in an

interaction often heightens identification with the framing of participants in that interaction”. In

their study of collective action frames in the Occupy movement, Reinecke and Ansari (2021)

found that emotion-laden interactions were central to intensifying activists’ attachment to a new

collective action frame.

Numerous studies have shown that frame alignment is hard to maintain even when parties do

manage to align divergent frames, showing that alignment can easily fall apart if framing

contests reignite (Turcotte & Pasquero, 2001; Zuzul, 2019). This can lead to merely symbolic or

otherwise inadequate action that can undermine the capacity of collaborating parties to achieve

joint objectives (Alamgir & Banerjee, 2019; Ansari et al., 2013). In a study of two large-scale

“smart city” projects, Zuzul (2019) documented the failure to maintain a shared understanding

among diverse experts as to what smart cities are and how they should be developed, showing
 10

how this led to intractable conflicts and the eventual collapse of collaboration despite significant

resource commitment and initial enthusiasm. While highlighting the fragility of frames, however,

the focus of these studies has not extended to how frames can be maintained and realigned.

Scholars of interactional framing certainly recognise that meaning is always provisional and

that frames thus continually emerge and evolve over time (Litrico & David, 2017), subject

always to unanticipated shifts, updates and replacements. Indeed this is implicit in their

conceptualisation of the construction of meaning as a dynamic and evolving process that is

“continuously being constituted, contested, reproduced, transformed, and/or replaced” (Benford

& Snow, 2000: 628). Nevertheless, we note that studies in this stream have tended to focus on

how frames are stabilised or solidified (Reinecke & Ansari, 2016) and how commitment to a

particular frame escalates at the expense of alternative framings (Cornelissen, Mantere, & Vaara,

2014; Reinecke & Ansari, 2021). As a result, much more is known about how interactional

frames are established than about the ongoing process of maintaining and (re-)aligning them.

The Challenge of Frame Alignment in the Context of Multi-Party Collaboration

A key complication that contributes to reigniting framing contests is that multi-party negotiations

unfold across multiple interaction arenas, with negotiating parties needing not only to reach

agreement with each other but also to “ratify their agreement with back-home constituents”

(Gray & Purdy, 2018: 91) and defend them towards their inner group (i.e., constituents back

home) (Rao & Kenney, 2008). A concomitant of this “two-table problem” (Gray & Purdy, 2018)

is that shared frames can break in “level-two negotiations” (Sebenius, 2013) if constituents reject

frames developed at the main negotiation table. That this dynamic has been under-explored in

the literature to date constitutes a significant gap in our understanding of frame alignment in the

context of multi-party collaboration. In particular, it remains unclear how “potential partners may
 11

frame issues at different levels” or in different interaction arenas to “prevent eleventh hour

rejection of potential partnership agreements” (Gray & Purdy, 2018: 220; 172).

The two-table problem and the threat it poses to partnerships calls for serious attention to the

question of how interactionally negotiated frames can be extended beyond the local contexts

from which they emerge. Gray et al. (2015) have proposed that frames generated in microlevel

interactions can in theory move across levels and escalate at the field level through a process of

“frame amplification” whereby frames are first expanded through their “adoption by a broader

group of people” until a particular frame gains traction through repetition “and/or emotional

intensification” (Gray et al., 2015: 120). Precisely how this process unfolds in practice still

remains unclear, however, and scholars concede that frame amplification is far from

straightforward given that interactional frames are contingent on the social situations in which

they emerge through interactions among specific actors (Lewicki et al., 2003).

Aligning frames across multiple interaction arenas involves at least two interrelated

challenges. First, any notion of frames as “free-floating packages of meaning to be deployed at

will and transplanted from one context into another” (Reinecke & Ansari, 2021: 381; Steinberg,

1999; Tannen, 1985) must clearly be jettisoned if we understand frames as interactional co-

constructions that emerge from the “dynamic enactment and shaping of meaning in ongoing

interactions” (Dewulf et al., 2009: 162). From an interactional perspective, it is clear that frames

co-constructed in one interaction arena are unlikely to be perfectly replicated in another and that

negotiators are likely to struggle upholding frames across arenas. Expanding frames across

arenas thus “may trigger a frame break and generate conflict between [interaction] networks,

which can then amplify further” (Gray et al., 2015: 120).

 12

Second, one reason why the nuances of multi-arena framing have been overlooked in field

level studies may be that, rather than focusing on the interactions of individual negotiators as the

interacting participants in framing processes, such studies have – with only a few exceptions

(e.g., Ferraro & Beunza, 2018) – focused on collectives of actors such as firms, social

movements, NGOs, and nation-states, treating these as aggregated groups or generic “social

actors” (Lee et al., 2018; Rao & Kenney, 2008; Reinecke & Ansari, 2016). Whilst reflecting a

common simplification in language usage, treating entities such as the UN and the EU as if these

collective entities were doing the framing in multi-party negotiations obscures the complex

challenge faced by individual negotiators acting on behalf of constituent organisations of needing

to negotiate meanings both within and between collective actor groups. A focus on individual

negotiating representatives rather than generic social “actors” is necessary because the concept of

an actor is already in itself an aggregate construct and yet it is not aggregate actor groups but

specific interactants who negotiate frames on behalf of their constituencies.

Focusing on the people who actually “do” the framing in multi-party collaboration (Leibel et

al., 2018: 164) brings to light the relational dynamics among stakeholder representatives

themselves and between them and their constituents. Scholars applying this focus in studies of

inter-organisational collaboration have highlighted the role of the personal bonds developed by

individuals through repeated interactions in facilitating cooperation (Berends & Sydow, 2019;

Ring & de Ven, 2019). For example, Hardy et al. (2005) have emphasised the importance of

personal ties between negotiators in achieving effective collaboration between diverging groups

and developing a collective identity. Other scholars have explored the role of “negotiating group

representatives” as “boundary spanning persons”, highlighting the pressure these interactants

face from their constituents (Aaldering & De Dreu, 2012; Jackson & King, 1983). While such
 13

studies have revealed that the relationships between representatives and their constituents are

crucial for upholding negotiated outcomes, however, what remains unanswered is how multiple

interaction arenas influence framing dynamics and how those who “do” the framing can navigate

complex multi-arena settings. To understand frame alignment across interaction arenas,

therefore, we address the research question: “How do multiple interaction arenas affect frame

alignment, and how can frame alignment be navigated across multiple interaction arenas?”

METHODS
Applying a process perspective and an inductive, open-ended and longitudinal research design

(Langley, 1999), we studied the role of stakeholder negotiations in steering the German Textiles

Partnership away from the brink of collapse towards deepened commitments.

This paper forms part of a larger research agenda initiated in 2013 to investigate national

and transnational governance responses to the Rana Plaza catastrophe. Through extensive

fieldwork in Bangladesh and internationally, Reinecke and Donaghey studied the

implementation of various governance initiatives to address the issues exposed by the

catastrophe, notably the Bangladesh Accord for Fire and Building Safety (Reinecke &

Donaghey, 2015, 2021, 2022; Donaghey & Reinecke, 2018). Whereas the Accord proved

successful in establishing multi-party collaboration, we initially selected the Textiles Partnership

as a contrasting case of failure affording insights into the factors explaining the success or

demise of collaborative efforts. From this starting point, Grimm (2019) conducted a pilot study

to explore what at first appeared to be a lack of substantive institutional formation that led to the

near-collapse of the partnership, focusing on the incentives that led companies to join

partnerships. To our surprise, this preliminary research uncovered an unexpected strengthening

of commitments in the collaboration. These early findings also alerted us to the complex

challenges involved in sustaining agreements voluntarily reached by diverse parties.

 14

Accordingly, we resolved to further follow the Textiles Partnership with a new research focus on

the negotiation of multi-party agreements and how they can be sustained.

As we continued to track the development of the Textiles Partnership, we significantly

expanded our data collection and further refined our research design to explore negotiation

dynamics and the dynamics of frame alignment in greater depth. Our additional data and analysis

revealed the significant extent to which multi-stakeholder negotiation processes are shaped by

the relationships between stakeholder representatives and their constituents. In particular, we

identified a central challenge to such negotiation processes in the “two-table problem” faced by

stakeholder representatives who need to negotiate shared frames not only through interactions at

one ‘table’ but also ‘back home’ with their constituents. The Textiles Partnership thus constitutes

a revealing case for yielding insights into frame alignment across multiple negotiating arenas.

Research Setting
As the world’s second-largest importer of textiles, Germany established the Textiles Partnership

in early 2014 as a national response to the 2013 Rana Plaza garment factory collapse in

Bangladesh (Schuessler, Frenkel, & Wright, 2019). Initiated by the then Federal Minister for

Economic Cooperation and Development, Gerd Mueller, the Textiles Partnership was intended

to demonstrate the willingness of Germany’s textiles industry to take responsibility for

improving the conditions of workers in their global textiles supply chains. Bringing together five

different stakeholder groups with “equal rights in the discussions and decision-making

processes” (Doc_B_62), the partnership sought to seek to develop collective solutions through

collaborative governance, defined as “a collective decision-making process that is formal,

consensus-oriented, and deliberative” (Ansell & Gash, 2007: 544). As of 2021, the Textiles

Partnership had 124 members in total, including 15 trade associations and 70 companies, 20

NGOs, seven standards organisations, two trade unions, the German government, and nine
 15

consulting members. Stakeholders are represented in the Steering Committee (short ‘SC’) as the

partnership’s central decision-making body “responsible for its strategic management and

ongoing development”, including “amendments to the Plan of Action” (Doc_B_62).

Our study focuses on the perspectives of the 12 members of the SC, as regular partnership

members were initially bystanders in the negotiation process. These individuals were nominated

or elected by their stakeholder groups, including companies (represented by our informants

Wilhelm and Florian), trade associations (represented by Thorsten and Klaus), NGOs

(represented by Silvia, Simone, and Fabian), the German government (represented by Patricia,

Jonas, and Alexander), trade unions (represented by Erich), and standards organisations

(represented by Sabrina). Figure 1 depicts the SC negotiation table with the crucial protagonists

in our case grouped as ‘objectors’, ‘neutral’, or ‘collaborators’. (All names are pseudonymised.2)

------------------------------------
Insert Figure 1 about here
------------------------------------
As outlined below and visualised in Figure 2, the Textile Partnership’s development process

can be subdivided into six negotiation phases based on critical negotiation outcomes. Each phase

was characterised by interlinking “topic-framing cycles” that tested the “red lines” previously

agreed upon by the stakeholder groups represented in the partnership’s SC.

------------------------------------
Insert Figure 2 about here
------------------------------------
Phase I saw the near-collapse of the partnership in 2014 as a consequence of the stakeholder

representatives having failed to address the two-table problem. Although these representatives

managed to negotiate a shared frame among themselves over the course of recurrent roundtable

meetings held from April to October 2014, propelled at least in part by political pressures and

2
References to the names of respondents indicate their group and negotiation table, e.g., “Silvia (NGO-SC)” refers
to the respondent Silvia, who is a member of the NGO group and a representative in the steering committee.
 16

bolstered by the personal involvement of the high-profile minister Gerd Mueller, the industry

representatives subsequently failed to recruit their constituents to the highly ambitious

commitments enshrined in Action Plan 1.0., including fixed goals and timelines.3 Against the

insistence of NGOs on these fixed goals being “non-negotiable” and legally binding for all

participants, the industry group declared that uniform and legally binding goals crossed their “red

lines”, insisting they would only commit to more flexible and “company-specific” goals. Just one

day before the official inauguration of the Textiles Partnership, the majority of companies

publicly announced their refusal to join the agreement.

Phase II, from late 2014 to 2015, began with Gerd Mueller stepping down as the public face

of the partnership after the refusal of the industry group to ratify Action Plan 1.0. Despite this

major setback, the partnership’s organising committee reconvened negotiations from December

2014. In these revived negotiations the industry group succeeded in upholding their red lines on

roadmaps needing to contain company-specific goals. While this encouraged numerous

companies, together representing some fifty percent of Germany’s textiles industry, to join the

partnership, the NGO representatives now struggled to recruit their constituents to the revised

agreement, called ‘Action Plan 2.0’, which crossed their initial red line that goals must be

uniform and legally binding. In another major blow to the partnership, Greenpeace decided to

leave the Textiles Partnership, dismissing Action Plan 2.0 as a “light version” that “gave in to

pressure from the textiles lobby” (IND16). Thus, also Phase II failed to establish an action plan

that all involved parties could agree with.

3
For example, Action Plan 1.0 stipulated that the involved organisations would increase their use of cotton produced
according to the Textiles Partnership’s standards by 20% by 2016, 50% by 2020, and 100% by 2024 (Doc_B_9).
 17

Phases III, IV and V,4 from 2015 to 2019, saw three decisive shifts in the positions of the

interactants over the course of prolonged negotiations, with both groups backing down from

several of their initial red lines. First, in October 2016 the industry group agreed to the

reintroduction of roadmaps with yearly fixed time and quantity goals that they had hitherto

rejected while the NGO group agreed to these goals being company-specific rather than uniform.

Second, in November 2018 both groups agreed to commit to making annual roadmaps public in

line with the NGOs’ insistence on this measure for ensuring transparency. In doing so, both

groups thus stepped back from their red lines; while the industry group dropped its demand that

roadmaps remain confidential and accessible only to partnership members, the NGO group

compromised on agreeing for the roadmaps to be made available to the public only after a “trial

year”. Third, in November 2019, the industry group agreed to a stricter review process entailing

corporate due diligence, thereby giving up their previous demands that roadmaps be reviewed

only by the partnership committee and that there should be no punishment for organisations with

unambitious roadmaps. These concessions were again partly matched by the NGO group, who

now dropped their demands for third-party verification. These mutual concessions came as a

positive surprise to many observers, especially since government attention and public pressure in

the wake of the Rana Plaza catastrophe had waned significantly over the years.

Phase VI, from 2019 to 2021, was marked by renewed framing contests that eventually led

26 companies and two NGOs to leave the partnership. In this phase, which included an

interruption of face-to-face negotiations due to the COVID-19 pandemic, the NGO side brought

forward new demands, with their red lines now including the stricter implementation of rules and

4
We have split the negotiation process here into three different phases because different aspects were negotiated
within each time-bracket and topic-framing cycle. We describe these phases in the same single sub-section,
however, as the framing dynamics in these phases were similar.
 18

goals to address the worsening situation of supply chain workers. Meanwhile, the industry side

urged for ‘softening’ the agreement on account of the financial burdens of the pandemic as

companies struggled with backorders and lockdown requirements. While the question of what to

do next remained unresolved and less advancements were made during the pandemic, later on

new concessions were made on both ends of the negotiation table (Doc_B_64).

Data Sources
To capture how interactional dynamics unfolded in the Textiles Partnership, we adopted a

longitudinal research design (Langley, 1999), conducting multi-phase data collection from

November 2015 to November 2021. Applying a process perspective (Langley & Tsoukas, 2010),

we “traced backward” the Textiles Partnership’s evolution since its establishment in early 2014

and “followed forward” its development from late 2015 onwards. Table 1 provides an overview

of all data sources we collected throughout this period.

------------------------------------
Insert Table 1 about here
------------------------------------
Formal & informal interviews. Our primary data source comprises in-depth interviews with

members of the Textiles Partnership, with a focus on the 12 stakeholder representatives in the

SC. Over a period of six years, we conducted 92 interviews: 81 semi-structured formal

interviews and 11 informal interviews. We spoke with a total of 48 individuals up to five times

each. Most of these interviews were audio-recorded and lasted between 18 and 102 minutes. Our

respondents included representatives of NGOs (n=22), the federal government (n=11), trade

associations (n=9), standards organisations (n=5) and trade unions (n=3), as well as

representatives from the German textile industry in positions ranging from project management

to C-suite executives in MNCs (n=24) and SMEs (n=9). We also spoke to advisory members,

including respondents from academia (n=7) and the media (n=2).

 19

Our selection of interviewees was guided by our aim of capturing multi-party and multi-

arena interactions when following negotiation rounds over time. We thus interviewed most SC

members several times over the course of our study (n=47). We then identified other key players

from each stakeholder group and interviewed these regular members at stakeholder level (n=32).

In drawing on interviews to capture our respondents’ recollections of interaction

experiences, our approach is similar to that taken by Balogun and Johnson (2004) and

Soderstrom and Weber (2020) in seeking “traces of interactions”. Identifying interaction traces

in our interview data helped us reconstruct key events such as SC meetings and excursions, as

well as to understand how singular interactions connected over time. To capture our

interviewees’ “knowledge of events” and “tap into [their] beliefs, perceptions and experiences”

(Langley & Meziani, 2020: 372), we focused our semi-structured interview guide on the types of

interactions that took place, the individuals involved, the content that was being negotiated, and

the general dynamics of these interactions. Importantly, we were able to cross-check different

accounts of the same events and interactions from the perspectives of multiple respondents

(Eisenhardt & Graebner, 2007), particularly since the 12 steering committee members remained

the same individuals throughout the period of our study. As we built up growing rapport and

trust with respondents over time (Dundon & Ryan, 2010), they increasingly shared more detailed

insights, including their personal perceptions of internal dynamics, political tensions and identity

issues. Since a core theme that emerged early on was the challenge faced by SC representatives

of navigating different negotiation arenas or “tables”, we adapted our interviews questions to

focus specifically on these dynamics.

We interviewed respondents directly or as soon as possible after SC meetings and other

events, inviting them to revisit their experiences of events and interactions in detail by
 20

privileging “how” over “what/why” questions in line with Langley and Meziani’s (2020: 378)

interpretive interviewing techniques for “capturing the meaning of lived experience”. We

“avoided naming the topic”, for example, by wording our questions based on the interviewee’s

own terms, e.g., asking about “homecoming” and their “perceptions” rather than referring to ‘the

two-table problem’ or ‘frames’. Focusing on respondents’ experiences of the social situation

rather than the content of the negotiations yield insights into unexpected aspects of our case,

including the role of interpersonal relations developed between the interactants over time.

Observations. Although SC meetings were strictly limited to signatories, we were invited to

present our work at two key stakeholder group meetings. This bolstered our position as

trustworthy observers and afforded a valuable opportunity to observe group dynamics first-hand,

with our notes from these meetings further informing our revisions of the semi-structured

questionnaire to deepen our understanding of the two-table problem. We also observed selected

public events, including a full-day workshop entitled “Rana Plaza – never again!” in November

2015 in Dortmund. Besides affording a better understanding of conflicts and tensions between

stakeholders, observing these meetings provided us with useful prompts in follow-up interviews.

Documents & archival records. To complement our primary data, we collected publicly

available documents and archival records related to the partnership. Published between August

2010 and November 2021, these documents included press releases, media reports, newsletters,

website information, comments, agendas, and presentations, while the archival records included

sustainability and annual reports, brochures, handbooks, guidelines, mission statements, and

research reports, as well as internal position papers prepared ahead of SC meetings. All these

written data were ordered chronologically and divided according to stakeholder groups,

amounting to a total of 3,120 pages. In addition to using these written data sources to construct
 21

our event chronology and as background information for interviews, we also used them to cross-

check interview accounts, including accounts of negotiated outcomes.

Data Analysis
We applied a combination of inductive process analysis strategies (Langley, 1999) to unpack the

evolution of the Textiles Partnership and capture the dynamics behind the shift from near-

collapse to a strengthening of the parties’ commitments. After documenting the overall process,

we mapped the interactants and interaction arenas before outlining topic-framing cycles and

tracing frame alignment within and across interaction arenas, thus building our theoretical

process model “outside-in”.

Step 1: Constructing a process chronology & visual map. To gain an overview of the

Textiles Partnership’s evolution, we ordered our transcribed data and documents in an event

chronology in line with the narrative strategy proposed by Langley (1999). Based on an initial

round of open coding, we first identified key phases and bracketed them in relation to milestones

in negotiation outcomes. In seeking to make sense of the shift in our respondents’ interpretations

of commitments, we initially assumed that external factors were the main reason for their having

agreed to fixed time- and quantity-goals similar to those they had rejected before. However, our

process analysis soon alerted us to the key role played in this change by the ongoing

communicative process between the specific individuals participating in the negotiations, as was

evident in many of our informants’ responses, e.g., “our perspective has changed through

ongoing debate, through working together and by getting to know one another” (IND29-SC), and

“It was a process of finding one another at first … Everyone has their own perception of how

things should be” (IND12, emphasis added). Drawing on interactional framing theory (Dewulf et

al., 2009; Gray et al., 2015), we noticed that these interpretive “perceptions” were shaped by

specific interactions that our respondents described as “finding one another”.

 22

To gain insights into the framing dynamics that facilitated the changes we observed in our

respondents’ interpretations of what they could or could not do about the issues at stake, we

extended our process chronology to visual mapping (Langley, 1999). This helped us identify

core topics of contention. Our mapping followed the process through which “parties interactively

construct the meaning of issues in the conflict situation” (Dewulf et al., 2009: 166). We

identified six interconnected “topic-framing cycles” over six phases (see Figure 2 above).

Step 2: Mapping interaction arenas & interactants. In trying to understand why ever-new

framing contests kept on emerging and why aligned frames were at risk of breaking, we noticed

that topic-framing cycles were not influenced solely by negotiations between SC members.

Instead, our process chronology revealed frequent references by respondents to what they termed

“homecoming”, i.e., the times when SC representatives returned to their stakeholder groups and

presented the frames they had negotiated to these constituents. They also spoke repeatedly of

“red lines”, referring to the boundaries that stakeholder representatives were not meant to

overstep when negotiating on their constituents’ behalf.

Having identified the need for SC members to navigate back and forth between the SC and

their respective group interaction arenas as a core dynamic, we related this challenge to the “two-

table problem” in the negotiations literature (Sebenius, 2013). Informed by SC members that

“the real exchange of blows takes place between companies and NGOs” (NGO18-SC), we turned

our attention to the “homecoming” of NGO and industry representatives, focusing our visual

mapping on the challenges involved in navigating frame alignment across multiple arenas. This

mapping confirmed that a shared frame co-constructed in one interaction arena could

subsequently be destabilised in another, further revealing that such destabilisation invariably

depended on the interactions that took place within each arena.

 23

Step 3: Tracing frame alignment within and across interaction arenas. In a third step we

turned to the literature on framing contests and settlements (Helms et al., 2012; Rao & Kenney,

2008) and the various frame alignment processes whereby interactants develop mutually

resonant frames (Snow et al., 1986). However, we soon realised that extant scholarship has

mostly treated negotiated meanings as relatively stable (Reinecke & Ansari, 2016), with most

studies focused on how frames are amplified or escalated at field level (Gray et al., 2015;

Cornelissen et al., 2014; Reinecke & Ansari, 2021). Accordingly, we returned to our data to

decipher how negotiated meanings were affected when stakeholder representatives navigated

frame alignment across interaction arenas. Following Dewulf et al.’s (2009: 163)

recommendation to “focus on specific points or episodes in conflict interaction”, we coded for

our respondents’ recollections of key interactions, taking the “interaction traces” (Soderstrom &

Weber, 2020) between representatives as our “unit of data” (Zilber & Meyer, 2022).

We further followed Dewulf et al.’s (2009: 165) recommendation to unpack “what is it that

gets framed”. In doing so we noted that parties not only negotiated the issue at stake but also

engaged in framing their identities and relationships and mutual interactions in the partnership.

This was evident across our interview data in our informants’ reflections on e.g., “What is my

counterpart like?” and “How are we working here?”. By iterating between our data and the

literature, we discovered that these accounts corresponded to what Dewulf et al. (2009: 166)

define as “relational-, issue- and process framing”. To capture this, we identified first-order

codes such as “getting to know one another” and “approaching each other on a personal level” as

indications of relational framing, identifying any changes described in relationships as

indications of shifts in relational framing (e.g., from referring to discussions with “enemies” to

“talking to a person like me”). We then reengaged with our data to gain deeper insights into how
 24

relational, issue and process framing shape frame alignment within interaction arenas (see Figure

3 later on). In doing so we were able to identify the details of how relational framing comes

about, including via modes of inquiring (e.g., “What are you doing?”), re-evaluating (e.g., “they

don’t mean it that way”), and identifying (e.g., “this person is also a person just like me”).

To theorise on the processual dimension of our longitudinal data and gain a sense of patterns

in the processes captured in these data, we coded for relational, issue and process framing in each

topic-framing cycle. This yielded insights into the framing processes that occurred during both

“outgoing” and “homecoming” interactions. What struck us as particularly important was how

any misalignment of frames was followed by navigation between interaction arenas and

reoccurring frame (re-)alignment (see Figure 4 later on). By developing a more specific set of

codes for the various moves involved in within- and across-table frame alignment, we identified

three emergent “patterns in the process” (Langley, 1999) that recurred in the topic-framing

cycles with regards to the back-and-forth navigation between interaction arenas. We found that

these “cross-arena framing processes” were key for “stretching” frames not only within one

interaction arena but also across the boundaries of this arena to other arenas (see Figure 5 later

on). We labelled these as processes of “reciprocal, integrative and bilateral frame stretching”.

Step 4: Building theory. In the final step of our data analysis, we developed our theoretical

process model by interrelating insights gained from our event chronology and visual map (step 1)

with our interaction mapping (step 2) and our tracing of frame alignment (step 3). In this way we

were able to abstract the details of the overall process and to identify the framing dynamics and

interrelationships between topic-framing cycles. Constructing our process model (see Figure 6

later on) thus enabled us to understand the how and why of the ongoing frame alignment process

that facilitated the maintenance of a shared frame.

 25

We use “composite vignettes” as a way of vividly portraying “a critical event or moment in

the field … to illuminate a theoretical concept … alongside more interpretative explanatory text”

(Jarzabkowski, Bednarek, & Lê, 2014: 280). These vignettes portray critical moments of

interaction, while our accompanying analysis shows how we theorised the processual dimension

of our longitudinal data. In composing our vignettes, we merged our informants’ recollections of

specific interaction episodes from multiple interviews and event observations into single

narratives to highlight dynamics illustrative of those we identified throughout our data.

FINDINGS
Our findings document how the Textiles Partnership was maintained after near-collapse and how

the parties eventually agreed on a shared frame, committing to goals they had previously rejected

and that went far “beyond their own limited vision of what is possible” (Gray, 1989: 5). In

tracing the negotiation process over time, we elucidate how this turnaround unfolded in four

parts. First, we explain how the partnership’s near-collapse was due to the difficulty of aligning

interactional frames across multiple interaction arenas, evidencing how SC members failed to

recruit their constituents to the shared frame negotiated on their behalf. Second, we explain why

and how different interaction arenas affect frame alignment, highlighting how frame alignment

entails laminations not only of issue framing but also of relational and process framing. Third,

we document how frame break can ensue from a failure to build up these laminations among

constituents. Fourth, we show how frame alignment can be navigated across interaction arenas to

maintain shared frames by ‘collaborating on the edge of failure’.

Multiple Interaction Arenas Pose a Challenge to Interactional Frame Alignment

In October 2014, the German Minister for Economic Cooperation and Development, Gerd

Mueller, proudly unveiled Action Plan 1.0 of the German Textiles Partnership in Berlin: “For the

first time,” Mueller announced, “it was possible to reach a content-based consensus among key
 26

actors about their common goals” (Doc_A_3). This announcement proved premature, however,

as most German textiles companies refused to sign up to the action plan, including Adidas, C&A,

the Otto Group, and Tchibo, with the exception of the sustainability champions Hess Natur and

Vaude. This failed inauguration was a major blow to the German government and Mueller.

Having spent the previous six months in a series of roundtable negotiations with representatives

of the German textiles industry, trade unions and civil society, the “state secretaries and the

minister were shocked” (IND2-SC) when the textile companies now claimed that the action plan

contained overly ambitious commitments, including “unrealistic” fixed quantity goals for

sustainable sourcing and annual progress reviews with third-party verification.

Vignette 1 highlights the lead-up to the near-collapse of the partnership, capturing how

industry representative Thorsten failed to recruit his constituents to join the partnership and his

attempts to warn the government about this impending failure.

Vignette 1
In late September 2014, industry representative Thorsten is meeting with senior managers from the ten largest
German textiles companies in his office in Berlin’s Mitte district. Prior to the meeting, Thorsten sent out the final
draft of Action Plan 1.0, which the companies had yet to sign to join the Textiles Partnership.
Against expectations, all the managers present, including the managers of Adidas, Otto, and H&M, now
unanimously announce that the binding fixed goals and timelines are major obstacles to ratification. One manager
starts by declaring “This is a no-go!”, followed by another who states “We will not sign this!”. A third manager
explains their common objection by citing the opinion of their legal department: “The liability issue remains
unresolved. And that’s a disaster. We just can’t do this.” By the end of the meeting it is clear that the main players in
the industry group will not commit to join the partnership.
As soon as the meeting ends, Thorsten worriedly phones Franz, who serves on behalf of the German government as
assistant to minister Mueller and moderator in the partnership negotiations. Thorsten knows that he must meet Franz
as soon as possible: “We need to talk about how to deal with this! Any chance you could come by?”
Franz comes by Thorsten’s office the following evening. In their one-on-one conversation, Thorsten relays the
industry group’s objections: “We have to talk about this now! If we stick to the fixed time and quantity goals in the
action plan, I can guarantee we [the industry group] will not sign the partnership agreement.” Outraged by this
objection, Franz asks “Why?” Thorsten replies that “There’s this one question – the liability question. I guarantee
you, we [the industry representatives] are not going to get this through [in the industry group].”
Franz is not willing to budge. He thinks Thorsten is bluffing in order to water down the agreement. Thorsten
suggests extending the negotiations to develop a revised version: “If we adjust the liability issue, if we add two or
three more points for clarification in the document, then I think we will be able to make companies sign the
agreement.” But Franz insists on going ahead with the action plan as is. He does not understand why Thorsten is
bringing up this issue only now after having discussed the action plan together for months.
In a major blow to the government, Thorsten and Klaus (both SC-industry) release a collective press statement
announcing that the industry group will not sign the partnership agreement: “It is necessary to state realistic and,
 27

more importantly, feasible goals that are acceptable not only to specific companies but also to the majority of textile-
trading companies.” (Doc_G_68)

Vignette 1 reveals that it was above all the misalignment between the SC and stakeholder

group frames that led to frame break and the near-collapse of the Textiles Partnership. Notably,

this crisis led companies in the otherwise extremely heterogenous German textiles industry to

“line up in agreement with each other” and collectively draw hard boundaries as a precondition

for any continuing negotiations: “We, the industry, had set several ‘red lines’ which we didn’t

ever want to cross.” (IND16)

Laminations of Frame Alignment: Relational, Issue and Process Framing

Despite the near-collapse of the partnership in October 2014 and the watered-down Action Plan

2.0 agreed to in April 2015, the parties were able to strengthen Action Plan 2.0 significantly

through negotiations from mid-2016 onwards. Over the course of numerous lengthy SC

meetings, the representatives managed to agree on re-integrating many of the commitments

formerly rejected, with both the industry and NGO groups repeatedly crossing their previously

declared red lines. Our findings unpack how this was achieved through a process of repeated

iterations between frame alignment in the SC and stakeholder group arenas. In the sections

below, we trace the three types of framing laminations that underpinned interactional frame

alignment within the SC interaction arena. Later on, we highlight how interactional frame

alignment was navigated across interaction arenas.

Relational framing. Initially, representatives of the NGO and industry group perceived

themselves as sitting around a table with “enemies” (NGO17-SC). Over the course of prolonged

negotiations, however, and after experiencing frame break early in the process, they came to “get

closer, get to know each other”. This led to a shift in attributional orientation, with interactants

no longer seeing each other as “enemies” to whom they assigned blame but as allies with whom
 28

they shared responsibility: “We now have the spirit to collaborate. We are all here in different

roles, with different interests in the same thing, but what we now all have in common is that we

want to improve the textiles supply chain.” (GOV2-SC). Central to this development was a shift

in relational framing that occurred in beyond-the-table negotiations, including informal meetings

over meals or drinks. Industry representative Florian told us how he came to reconstruct the

meaning of “self” and “other” in these “backstage” interaction contexts, especially through

informal meetings with NGO representative Silvia:

We had an evening in Bielefeld. And that was nice ... when I chatted privately with Silvia about ‘What are you
doing?’, ‘Where were you on holiday?’. Then you get a completely different bonding level with each other,
which is perhaps more stable, because you know ‘Ah, this person is also just a person like me’, right? ‘The two
of us represent completely different interests at this table, but she is also a person like you and me, who also
shares her viewpoints.’ … and when you tell each other a few personal things … that’s what you do in the
evening when you talk to each other in a relaxed atmosphere, I think you get to a different level of relationship.
(IND35-SC)

According to Florian, getting to know his counterparts significantly helped to improve

formal negotiations:

We in the SC organised events where we had ‘time outs’ for two days somewhere, where we met outside [the
negotiation table] with an overnight stay to have an evening with a glass of wine. Here we developed and
discussed strategic things once again and on a different level … this is a social process of getting to know each
other better, of learning to appreciate each other. Pleasant relationships developed in these settings … And that
certainly contributed to it [finding consensus] … when you can meet informally and perhaps exchange a few
personal words. That simply helps … There were different temperaments. So that’s what you get to know.
‘What is the temperament of one or the other?’, ‘How do I actually judge them when they rant like that?’ You
start to understand ‘Okay, they don’t mean it that way. They just have a funny way of expressing themselves.’
And that’s also quite a relief when you get to know each other and it becomes a bit easier ... [whereas] before
you might have found these outbursts off-putting, which at times hindered the progress of negotiations.
(IND40-SC)

Florian’s description of his interactions with counterparts yields several important insights,

including how his judgement changed through his interactions with Silvia and how he

constructed a new meaning of their relationship. Florian notes how agreement might previously

have been hindered by his and others’ reactions to people “ranting” at the negotiation table but

that getting to know counterpart had enabled him to develop an understanding and tolerance of

the different ways people articulated themselves and their “temperaments”. This helped him re-
 29

evaluate and “appreciate” them as individuals “like you and me”. Instead of being “put off” by

“outbursts” and allowing this distaste to colour his judgment, he now saw beyond the form to the

content of their contributions, attributing “rants” to the fact that people “just have a funny way to

express themselves”. Through repeated interactions in formal and informal settings, Florian and

Silvia’s relationship gradually became socially embedded in informal interpersonal ties.

In assigning new meaning to their counterparts, SC members also assigned new meaning to

their relationships with these individuals, no longer encountering each other solely in their

formal roles as representatives of opposing interest groups but as individuals with whom they

had a personal relationship. This helped them move beyond their antagonistic perceptions of

each other as “enemies” to see each other at least as “frenemies” with whom they could share a

meal, ultimately enabling the development of a collective “we” identity.

Lamination of relational framing onto issue framing. Relational framing proved crucial in

shaping how the stakeholder representatives subsequently constructed their understandings of the

issues at stake. This was explicitly acknowledged by industry representative Florian as follows:

On this other relationship level it becomes, I would say, more stable, yes? And on this more stable level it is
also more resilient when you have problems where you know: ‘Aha, she now has a different opinion in terms
of content. I have a different one too. Okay, so we have to discuss this, since nevertheless I respect her and
nevertheless I think she’s somehow super nice as a person.’ (IND35-SC)

These new relational framing dynamics in turn led to a change in the nature and frequency

of framing contests between group representatives as their attributional orientation shifted from

mutual dismissal and blame to a shared willingness to focus on the underlying content-related

reasons for their disagreements, recognising that others “have a different opinion in terms of

content” (IND35-SC).

Issue framing. The second frame lamination we observed was that of (re-)framing the issue

at stake. This came about as ongoing interactions led to a shift from a subjective to an
 30

intersubjective understanding. Reflecting on the early phases of the partnership, our informants

commonly identified a lack of mutual understanding as having been a key barrier to progress.

Although the SC members all spoke the same national language in negotiations, they later

recalled how they had not been “speaking the same language … like having two different, two

foreign languages coming together” (UNI1-SC). As the moderator observed of this period:

Their words are different to a degree that even certain words [in the same national language] have a different
meaning for each group. That means they use words at times that cause a reflex in the other group so that
people in the other group aren’t able to really listen any longer to what else is being said. (GOV8-SC)

At this stage, industry representatives perceived their NGO counterparts as ignorant of

commercial realities and supply chain complexities and thus lost in a “theoretical nirvana” or

“make-a-wish-world” (IND39-SC). For their part, the NGO representatives claimed that the

“multinational corporations didn’t even make an effort” (NGO4-SC) to understand the plight of

workers in their supply chains.

Through repeated interactions, however, SC members gained a better understanding of the

specific challenges each other faced. As Silvia explained: “We, the industry and us [the NGOs],

have of course gotten to know each other better … you get to know one another and you also

start to see the problems on the other side of the table” (NGO9-SC). The more the interactants

immersed themselves in the complexity of the issues at stake, the more they grew to appreciate

the difficulties faced by the other group, as evident from Thorsten’s account:

Silvia and the others [NGO representatives] learned a lot about the logic of business firms in this process –
how businesses work and what they can and cannot do. We [industry representatives] on the other side also
learned a lot from them, we developed quite a lot of understanding for how they work and what they do.
(IND19-SC)

Vignette 2 outlines an interaction between SC members Silvia and industry representative

Wilhelm, highlighting how they aligned their framing on the issue of adjusting Action Plan 2.0

(“roadmap transparency” in topic-framing cycle 4).

Vignette 2
 31

In August 2017 the first company-specific roadmaps are being compiled and made accessible in a secure databank
for all partnership members. Industry representative Wilhelm calls NGO representative Silvia to discuss whether the
compromise made on company-specific instead of uniform roadmaps is sufficient. Wilhelm says “I’ve read through
a lot of these roadmaps and in my honest opinion we have stronger and weaker ones. We also have big differences
in the level of ambition.” Silvia replies “From what I’ve seen, this isn’t enough. I’m worried the companies won’t
move.” Wilhelm counters “But it’s not only the industry side that’s delivered insufficient roadmaps. Some of the
NGO roadmaps are essentially an impertinence as well!” However, he acknowledges that “Still, it’s true roadmaps
on the industry side are just as bad. I won’t name names, but some are really weak – full of spelling mistakes and,
yes, very vaguely worded.”
Having reviewed the roadmaps at an internal workshop prior to this phone call, Silvia knows that many regular
members in her group are dissatisfied with the outcome. She tells Wilhelm “They [the roadmaps] don’t seem to be
what we thought they would be.” Wilhelm agrees that they have to “recognise the weaknesses” but also defends
those organisations that have handed in inadequate roadmaps. “The specifications we made were too vague,” he
says, referring to the reporting tool formulated by the SC members. “The questions weren’t always good … There
was room for a lot of interpretation.”
Wilhelm knows from a recent workshop with his own stakeholder group that many companies who filled in the
roadmaps went about it somewhat lackadaisically, devising their own interpretations of what the reporting tool was
all about. However, he now explains to Silvia that it was only after his closer engagement with companies on how to
fill in the reporting tool that many of them said “Ah! Now I finally have clarity. Now I know what you want ...
Because in the past, this was somehow very nebulous.” Even companies that were regarded as frontrunners told
Wilhelm that “this whole topic has been made overly complex … the degree of academisation of this roadmap
activity has nothing at all to do with things on the ground … [The reporting tool] cannot be implemented at all.”
Wilhelm seeks to reassure Silvia that “you will see in 2018 the roadmaps will already be much more specific than
they are now.” Silvia says that she is willing to “wait and see whether they [companies] work with the roadmaps or
not.” Wilhelm proposes to engage more with the regular members of his own group to ensure “the partnership is
continuously moving forward.”
The call ends with Wilhelm and Silvia being on the same page regarding the reasons for why roadmaps are not as
promising as they’d hoped. They agree that Silvia will explain to her group why the roadmaps were lacking in
ambition, while Wilhelm will communicate the need for much stronger roadmaps to his group.

Vignette 2 yields several important insights. Firstly, Silvia and Wilhelm’s engagement in

dialogue meant they were willing to provide each other with a sense of the reasoning within their

respective groups. Although Silvia started off bluntly (“this isn’t enough”), Wilhelm was aware

by now that she was a seasoned and “hard-hitting” negotiator (“not known for her pure

gentleness” (IND21-SC)) and thus quickly went on from defending his group (“it’s not only the

industry side”) to conceding that “it’s true” the industry could do more.

Secondly, the vignette highlights how both representatives gained intersubjective

understanding of the issue. Building on their relational framing, Wilhelm was willing to share

what he himself had learned at the workshop with his group, and Silvia later admitted that

Wilhelm’s explanations of why the roadmaps were not as ambitious as she had hoped had helped
 32

her “to see the problems on the other side of the table” (NGO9-SC). This extended her

interpretative scope, as new meanings were added to how she had previously interpreted the

reason for insufficient roadmaps (from “companies are not moving” to “companies do not

understand”). Such lamination of new meaning and transition from a subjective to an

intersubjective understanding led to a further alignment of the shared frame within the SC.

Lamination of issue framing onto process framing. Relational and issue framing not only

helped shift framing contests from person-related disagreements to content-related debates and

from a subjective to an intersubjective understanding of the issues at stake but also prompted SC

representatives to reframe the meaning and purpose of their ongoing interactions. This further

facilitated the shift from an antagonistic to a collaborative interaction process in which the

groups now moved forward together to reach shared goals rather than continuing to pursue their

own respective agendas. As a result, the industry side shifted its position towards blurring its

previously uncrossable red lines:

Through ongoing debate and by getting to know one another and understanding their [the NGOs] point of view,
through working together, our [the industry’s] perspective has changed from ‘we can’t’ towards ‘well maybe
we can’. (IND29-SC)

Process framing. The third frame lamination we observed was tied to how interactants (re-

)framed the meaning of their ongoing interactions. Conflict and “shouting matches” (NGO17-

SC) had initially characterised interactions in the SC, with NGO representatives accusing

industry representatives of being unwilling to make any substantive commitments (“They’re

completely against it [investing in improved conditions] – they are very obstructive and do not

want it under any circumstances” (NGO3-SC)) and with industry representatives accusing NGOs

of pursuing a ‘naming and shaming’ agenda. Repeated interactions over the course of ongoing

negotiations fostered a willingness to compromise, however, with interactions increasingly

framed as collaborative rather than antagonistic. According to industry representative Wilhelm

 33

(IND10-SC): “What made the partnership continue nevertheless, was the strong will to make it

happen – on both sides. In other words, to be ready to compromise, to really seek compromise.”

Reflecting on past conflicts, Wilhelm admitted that the greatest compromises had always

been reached when negotiations were “close to failing”:

I also remember this SC meeting … at the end of the day we weren’t able to reach compromise. And there were
many discussions after the meeting. The whole thing was really close to failing. And yet the following meeting
was the one that brought us the furthest – that achieved the most progress, because it was clear that no one
wanted to drive the thing against the wall ... And that led to the fact that the next meeting brought so much
progress, because each side knew ‘We now have to clear our position to a certain extent and approach the other
side so that this can work’. (IND10-SC)

As this account suggests, it was above all the interactants’ mutual recognition that their

collaboration was “really close to failing” which increased their willingness to compromise:

“You can’t just keep to your position in such a partnership and say ‘No, I won’t do it!’. Then it’s

a failure.” (IND26-SC). The threat of imminent collapse made the interactants realise they had to

“clear their positions” and cross their red lines to keep the partnership alive. This shift to a more

compromising position was described by Florian’s assistant Susanne (who joined SC meetings

on a regular basis):

I don’t want to make it sound as if it was so easy to move forward … Because of course I’d say the bar set by
the NGOs for time and quantity targets of roadmaps was much higher … Getting to the point where we met in
the middle and said ‘Yes, good. I’ll say that we can go along with the formulation of the time and quantity
targets’ was a long process ... And let me put it this way, the NGOs had to get off their, shall I say, high horse.
Just like the industry had to. (IND26-SC)

The parties’ collective recognition of the need to compromise by crossing their red lines

changed the nature of framing contests in the SC, with all groups now seeking primarily to ensure

the partnership’s survival rather than solely pursuing their respective interests.

Over the course of interrelated topic-framing cycles, further laminations of meaning were

added to the shared frame (see Figure 3). These laminations of relational, issue and process

framing facilitated further frame alignment and gradually increased the willingness of

representatives to cross the red lines they had set internally within their groups.
 34

------------------------------------
Insert Figure 3 about here
------------------------------------
Misalignment of Laminated Interaction Frames Across Multiple Interaction Arenas
As a key finding from our data analysis, we discovered that the frame alignment attained within

the SC arena was by no means automatically ‘translated’ to the constituent groups on whose

behalf this frame had been negotiated. Instead, SC representatives struggled to uphold the shared

frame when “coming home” to their own groups. Vignette 3 highlights how the challenge of

maintaining frame alignment across arenas again brought the partnership to the brink of collapse.

This vignette describes an interaction within the NGO group in Phase IV when representative

Silvia was confronted with criticism for agreeing to roadmaps not being accessible to the public

in a first “trial year” – something her group had internally agreed to be a “non-negotiable”.

Vignette 3
The NGO group’s annual meeting takes place from 27 to 28 January 20185 in a remote country hostel on the
outskirts of Jena, Germany. A group of 40–50 motivated and enthusiastic NGO members and activists have gathered
to discuss the activities and campaigns for the upcoming year. Silvia is attending as one of the group’s three SC
representatives in the Textiles Partnership.
Many of those here today have seen the company-specific roadmaps so far submitted by the textile companies and
found them either lacking in ambition or implausible. NGO members share their frustration. They see the lack of
progress as evidence of companies’ unwillingness to improve: “We’ve seen now that there are companies that have
put together implausible roadmaps. For now, they got the chance to amend their roadmaps. But now we’ve reached
the point where we need to ask ‘Is this process still trustworthy?’” Another member insists that “if it becomes clear
that it [the engagement of companies] doesn’t go beyond what is currently mainstream in the textiles industry, then
participating in the partnership is of no use for us. And then the NGO group has to leave.”
This issue is one of the sticking points of today’s meeting: the question of whether to leave the partnership or not.
My6 own presentation on the progress made in the partnership is first on the agenda. I highlight the positive
advancement of the partnership, as well as the NGOs role in it. I present some direct quotes from industry
representatives, including “I’ve talked a lot bilaterally with Silvia”, “discussions on an equal footing with Silvia”
and “Silvia is someone who is willing to make compromises.”
Immediately a loud and disorganised debate erupts among the crowd. One of the more confrontational group
members, Lars, says in outrage “This presentation perfectly illustrates why we should immediately leave the
partnership.” He is referring to the perceived role of NGOs, insisting “Our goal is not to become allies of the
industry.” Another objector declares “We have lost our purpose by being so closely engaged in the day-to-day
business of the partnership. Structural changes are not in sight. We invest a lot of time – time that’s now not

5
Representatives have faced the challenge of coming home already earlier in the process, in 2014, as Vignette 1
highlights. However, we have chosen to outline here data that we collected later in the process, in 2018, as we only
then gained access to meetings and got to observe the homecoming challenge first-hand. Whilst before, we had only
heard of this challenge from our respondents in their retrospective sense-making efforts, this meeting allowed us to
make notes of how the interaction unfolded in real-time.
6
“I” here refers to the first author, who conducted extensive fieldwork for the study and presented intermediate
results at the described NGO group’s annual meeting in January 2018.
 35

available for campaigns. We legitimise the partnership. What is our original task? First of all, it’s campaigning –
exposing company scandals like labour rights violations.”
Despite being described by her constituents as having previously been “confrontational for years”, Silvia holds back
throughout most of this discussion. She gives the impression of being rather irritated. Later, she reflects that “we [all
representatives in the SC] got to know each other much better. That makes the difference [to how the objectors
within the NGO group see things]. Or at least that’s how I explain it to myself … I think the industry has perhaps
seen us a little differently before. They always see us as the ones who make a fuss.” Silvia makes clear that “that’s
what we continue to do [make a fuss]. That’s what we want to maintain.” She reflects that “if they [the industry]
think they can buy me in, then of course I don’t think that’s great”. Silvia is taken aback “When I think back to it
now, I think ‘Man! We were a bit more radical back in the days.’ I have the feeling that we’ve also become
somehow (laughs) tamer.” However, she also concedes that dialogue in the SC has enabled representatives to solve
important conflicts and create a stronger partnership.
By a narrow margin, the meeting ends with a decision in favour of the NGO group staying in the partnership for the
time being, depending on whether the group’s new demands are met for making the roadmaps more transparent.

This vignette yields three key insights. First, it highlights how Silvia could not simply

“transport” the new meanings she had co-constructed with other representatives in the SC to her

constituents “back home” but rather was immediately confronted with the misalignment of this

SC frame with the frame of her constituents. Since these regular NGO group members were not

part of SC interactions and had not built up the same frame laminations as their representative,

Silvia’s compromises in the SC arena were not “inter-subjectively meaningful and

understandable” (Goffman, 1974: 127), in particular her agreement to company-specific

roadmaps and a “trial year” of not publishing. Unlike Silvia, these members’ relational framing

had not shifted, hence they still saw themselves as enemies or certainly “not allies of the [textile]

industry”. In terms of issue framing, regular NGO members felt that companies submitting weak

roadmaps were being treated “too gentle”, whereas Silvia had developed an understanding of

some of the genuine challenges faced by companies. In terms of process framing, meanwhile,

these members questioned whether the “process [was] still trustworthy” or if it made more sense

to leave, whereas Silvia had come to feel, along with her counterparts in the SC arena, a

commitment to the partnership’s survival.

Second, this vignette highlights how Silvia engaged in renewed relational, issue and process

framing through the vociferous efforts of her NGO constituents to realign her frame with theirs.
 36

In terms of relational framing, interactions with her own group led Silvia to reflect on how she

was now seen by them as “tamed” and “as not making a fuss anymore”, prompting her to

question the meaning of “self”. While she saw herself in SC interactions as “someone who talks

to everyone” (NGO9-SC), these interactions “back home” reminded Silvia that she had been

more radical before and made her determined to “make more of a fuss” again. Similarly,

interactions in this arena added and induced her to attach new meanings to how she saw the issue

at stake. While she had previously accepted Wilhelm’s explanations for why roadmaps were

lacking in ambition (see Vignette 2), renewed interactions with her own group made her question

whether the industry was “trying to buy her in”. In terms of process framing, Silvia was

reminded of the importance of continuing to confront rather than act “as allies of the industry”.

Third, this vignette highlights how Silvia was influenced by her exposure to different frame

laminations through different interaction arenas. By Silvia’s own account – as she ‘explains to

herself’ in this vignette – what “makes the difference” between her present and previous attitudes

towards the industry is that, unlike her constituents, she has gotten to know industry

representatives in the SC. Being able to view an issue from the perspective of a broader

repertoire of framings can help negotiators navigate multiple interaction arenas. However, on

returning to the SC negotiations Silvia ultimately decided against overstepping her group’s red

lines too far by not “waving through” unambitious roadmaps. Instead, following the meeting

with her own group she demanded greater transparency for roadmaps even though this risked

breaking the shared frame negotiated in the SC.

As highlighted in this vignette and Vignette 1 (and as shown in Figure 4), moving across

interaction arenas risks breaking the shared frame established within an interaction arena. In

Phases I and II this ‘frame stretching’ led to frame break, first in the industry group and then in
 37

the NGO group, risking the near-collapse of the partnership. Indeed, our data reveal the

partnership was close to failure on multiple occasions, as evidenced by the number of individual

companies who refused to overstep their red lines along the way and eventually left the

partnership (see ‘exits’ in Figure 2). While frame misalignment continued to emerge in each

topic-framing cycle, however, over time the SC representatives improved their skills in

navigating frame alignment across interaction arenas.

------------------------------------
Insert Figure 4 about here
------------------------------------
Navigating Frame Alignment Across Multiple Interaction Arenas on the Edge of Failure
Exposure to the risk of frame break during “homecoming” moments also alerted SC

representatives to the fact that it was not enough only to achieve compromise among themselves

but that they also needed to ‘recruit’ their constituents to the frames negotiated in the SC arena.

As the moderator of the Textiles Partnership observed of both group representatives’

experiences: “Silvia and Thorsten got hit on the head by their own troop, because for those who

were not involved in the process the concessions that one side or the other made were partly too

far-reaching” (GOV8-SC). In short, as one industry representative concluded, the representatives

learnt that “you need to be able to take your own people with you” (IND19-SC). Moreover, they

also grew conscious of the ever-present risk of “the other side’s” constituents breaking away:

At several points in time there was a great danger that the demands were not ambitious enough and that the
NGOs would say ‘Now we’re leaving!’ or that the commitments would become too concrete and then
companies would leave by the score. It was always a tough struggle [on both sides], but we were determined to
make this work. (IND10-SC)

We identified three key processes of cross-arena framing by which stakeholder

representatives navigated frame alignment across interaction arenas and managed to steer their

groups beyond their respective red lines from Phase III onwards. The common basis for all these

processes was the realisation of the SC representatives that their constituents were not passive
 38

“frame takers” willing to adopt whatever shared frame was achieved within the SC but active

“frame makers” who needed to be included in frame alignment processes. For this reason, each

of the cross-arena framing processes of ‘stretching’ we describe below – reciprocal, integrative

and bilateral – created an interlocking of frame laminations across interaction arenas (see Table 2

for illustrative examples). The frame shared by SC representatives was thus no longer tied to a

single interaction arena but stretched across arenas (see Figure 5). Although this did not

eliminate frame misalignment altogether, it helped avoid frame break and collaboration failure.

------------------------------------
Insert Figure 5 and Table 2 about here
------------------------------------
Reciprocal frame stretching. The first cross-arena framing process we observed was that of

stretching the frame within the SC arena as representatives recognised the need to ease each

other’s “homecoming” by helping each other recruit their constituents to the shared frame they

had co-constructed in the SC. In this process, representatives made reciprocal concessions that

expanded the frame they shared in ways that encompassed the concerns of the constituents in

their counterparts’ arenas. Such reciprocal frame stretching was enabled by laminations of

previous relational, issue and process framing that had emerged through recurrent interactions.

First, reciprocal frame stretching was enabled by cumulative laminations of relational

framing through which the parties had previously reconstructed their attributional orientation so

that they now saw each other as trustworthy partners. These laminations increased parties’

confidence that they could rely on the reciprocal support of their counterparts to make equal

concessions and defend compromises vis-à-vis constituents. When negotiating the thorny

question of roadmap transparency, for example, industry representative Thorsten described how

he was able to negotiate reciprocal concessions with NGO representative Silvia by building on
 39

their existing personal relationship, meaning he was able to “rely on her” to reciprocate and

uphold a mutually agreed compromise that crossed his group’s red lines:

In these personal four-eye meetings with [Silvia], whom I now also address on a first-name basis, that’s just
something personal. We like each other ... And there you can make agreements and say ‘Listen! Let’s do it this
way.’. And then you have to be able to rely on the other person to do exactly the same and defend it in your
own group. And that’s what made it successful, that you could say to each other ‘Listen! There and there,
that’s our concession.’ So I’m ready in the negotiation to go beyond my current hard point. ‘I offer you now, in
private: I would still go along with [that]. But beyond that there’s no way.’ (IND38-SC)

Second, reciprocal frame stretching built on previous laminations of issue framing and

intersubjective understanding among representatives. In topic-framing cycle 4 on roadmap

transparency, both sides recognised that while transparency was “a stomach and soul issue”

(IND25-SC) for the NGO group it was also a “topic of resistance” (NGO8-SC) for the industry

group. Silvia thus agreed to “pursue the topic not too strongly”. In turn, the industry

representatives agreed to make it clear to their group that a compromise had to be reached.

Third, by building on laminations of process framing that had fostered the interactants’

ability and willingness to compromise, representatives now recognised that reciprocal

concessions were crucial for representatives from both sides to present the other side’s

concession as a “win” back home. Having understood they would need to take risks to progress

the partnership, group representatives eventually crossed their own groups’ red lines to facilitate

each other’s homecoming. By accepting a concession from the other side, moreover, “a personal

liability arises” (IND24-SC), and this compelled SC members to defend these compromises

when coming home to their constituents: “Then we must stand by it and say ‘It just doesn’t help,

we couldn’t get our way’.” (IND25-SC).

Integrative frame stretching. The second cross-arena framing process we observed is what

we term “integrative frame stretching” whereby the SC frame was stretched to the interaction

arena of constituents. In integrative frame stretching, interaction arenas became interconnected

as regular members were integrated in the negotiation process through the creation of multiple
 40

intermediate interaction arenas that included constituents from different stakeholder groups.

Specifically, from Phase III onwards the partnership’s organising committee initiated several

expert working groups in which regular partnership members could participate. These working

groups were each tasked with negotiating agreement around a particularly contentious topic such

as the debate around a living wage and the contentious review process for making roadmaps

more transparent and accountable. These groups were structured in a way that primed

participants towards relational, issue and process framing.

First, in terms of relational framing, allowing only experts with a deeper understanding of

the topic to join meetings helped overcome previous obstacles to progress because these

interactants could now relate to one another as legitimate experts. Previously, industry group

members had often accused NGOs of “laziness” or “simple incompetence”, with one respondent

from the industry group complaining, for instance, that “NGOs really do not know what they are

talking about most of the time” (IND18). As had happened in the case of within-arena framing,

however, the shift in attributional orientation associated with the establishment of expert groups

changed the framing contest from being person-related to being content-related, which thus now

applied also across interaction arenas.

Second, the creation of expert working groups further facilitated alignment on issue framing

amongst participants because issue-based expertise was made a prerequisite for participation.

Whereas seats in working groups had initially been allocated randomly, it quickly became clear

that relevant expertise was necessary to focus discussions on problem-solving activities:

We made great progress by moving away from open working groups where everyone was able to hop in and
out and you basically had to start all over again. So we set up this idea of expert groups and you had to prove
beforehand why you thought you had expert knowledge in this field. So that was very effective. So not just
anyone with idealistic ideas from the civil society could sit there but ideally someone who also deals with the
core issue in their NGO work, right? So that led to the fact that, for example, [NGO X] was kicked out because
they didn’t have the expertise for fibre compositions and technology processes and all that. And the space was
freed up for someone who used to be with [NGO Y] and is now with a standard-setting organisation and could
bring in a completely different expertise from that point of view. (IND37)
 41

Third, in terms of process framing the more focused and technical orientation of expert

working groups framed interactions as processes of pragmatic problem-solving. Working in

smaller groups focused the interactants on identifying achievable steps that eventually helped

them move across their red lines:

Cooperation happens in the expert group … When you’ve identified certain common lines in one area [laughs],
you dare to approach the companies. And in the expert groups this may work more easily because, let’s say,
people think more constructively about these issues. (NGO12-SC)

Moreover, expert groups not only facilitated the participation of regular members in the

partnership but enabled them to integrate experts from their own organisations to stretch frames

even further across several interaction arenas. This was crucial to build up laminations of process

framing beyond the steering committee and stakeholder group to include the organisational

arena. One respondent who had experienced difficulties in aligning working group proposals

with their own organisation explained that “we send experts from our company to these

committees now” to ensure that agreements can no longer be rejected based on non-feasibility:

In the beginning I was in all the working groups ... I came home with something that I thought was totally
purposeful and made sense ... but the responsible departments that had to implement it later said ‘No, we can’t
or don’t want to do it’. And then we said ‘OK, then it would be more effective if staff from the departments
that are supposed to implement it later took part in the discussions in the working groups.’ For example, in the
chemicals group these were our chemists or in the fibre working groups our quality managers ... Because the
work is in vain if decisions from the working group cannot be implemented in the organisation afterwards, i.e.,
if you must go back to the working group to say ‘What we’ve discussed here cannot be implemented by my
organisation’. (IND37)

Expert working groups as intermediate interaction arenas had a dual effect. Not only did

they facilitate agreement among participants on specific topics but also made it much more

difficult for both the SC members and stakeholder groups to reject solutions that had been

negotiated, since unless critics had the requisite expertise to participate in working groups, their

objections could easily be discounted. Through these groups the SC frame was thus stretched

towards the solutions negotiated in intermediate interaction arenas.

 42

Bilateral frame stretching. The third cross-arena framing process we observed was

“bilateral frame stretching”. Here, the frame of regular members was stretched to the extent that

it overlapped with the frame shared within the SC. Again, this involved integrating regular

stakeholder group members as active “frame makers” by connecting interaction arenas through

what our respondents referred to as “scanning”. Such scanning involved “beyond the table”

negotiations in which representatives reached out to their own constituents but also “sound[ed]

out the other side” (IND38-SC).

First, in terms of relational framing, scanning was undertaken by SC members and aimed at

making regular members of their own and counterparts’ group re-evaluate themselves and

identify as front-runners. For example, one company that was known to be more progressive

recalled being repeatedly approached by NGO representative Silvia in her efforts to recruit them:

Of course, the [NGOs] tried to convince me, and others perhaps also. There were informal conversations on the
side, and Silvia said ‘Look, you could go further, why are you holding back? Can’t you do more?’. And then
she sprinkled a bit of sugar and was quite nice and acted a bit friendly and suggested ‘Come on, step a bit more
on the gas’. And then I always thought ‘Yes, man, actually you need to step on the gas a bit more’ … so you
start thinking again, you take on a different perspective, you think about it again. (IND34)

These interactions with Silvia thus made the manager reconsider his identity and his

company’s purpose and position within the industry group. Ultimately, bilateral scanning

leveraged divisions between sustainability ‘leaders’ and ‘laggards’ within the industry group.

These divisions had been overcome in Phase I when the industry’s opposition had served as a

unifying force in rejecting Action Plan 1.0. Cross-arena scanning from Phase III onwards thus

aimed at breaking up groups and recruiting individual members as allies to cross their red lines.

Second, in terms of issue framing, bilateral frame stretching was aimed at working out

where meaning was already aligned (“we work in a similar way” (IND36)) and identifying

misalignments (“how can we dissuade them?” (IND36)). Through informal bilateral

conversations, interactants developed an understanding of possible compromise lines: “There

 43

have been telephone calls ... to talk bilaterally to question topics again: ‘Why is the red line here,

where would a possible consensus be?’” (IND36). Aligning positions across interaction arenas in

advance was crucial to support negotiations in the SC: “We try to work out a line of compromise

in meetings with our key members before an SC meeting but also in the follow-up.” (IND22-

SC). Again this was achieved by breaking up groups:

If you address things in advance in smaller groups, if you already make phone calls before the meetings, partly
to talk to people individually before the meetings, to carry offers from one side to the other and say that if the
others would suggest ‘We make the roadmap look like this’ then what would you say to that? Would that be a
no-go for you? Where is the red line for you?’ As a rule, before we went into the big meetings, you could say I
tried to negotiate everything beforehand. (GOV8-SC)

By going back and forth between interaction arenas, representatives could “check the

temperature” in their own stakeholder group:

Everyone tries to scan their own group or the different fractions in their group. And if you have someone on
each side who can judge their own people well enough or knows ‘Well, I won’t get away with that – whatever
I do they’ll tear my head off!’, then you have to discuss something like that again in a small circle and try to
bring about a solution. (IND38-SC)

Scanning members helped representatives avoid clashes and escalations of conflict that

would have broken out if they took decisions that went too far beyond the group’s red lines.

Third, bilateral frame stretching through cross-table scanning shifted the interactants’

process framing towards identifying allies among the constituents both of their own groups and

of the other parties. Bilateral talks helped representatives break up groups to identify those who

were willing to overstep their group’s red lines and could help persuade their peers:

One then got allies and said: ‘Hey, we work in a similar way, let’s see how we can get this to work? How can
we dissuade the others from their refusal attitude or red line? How can we work on moving forward?’ (IND36)

Through various “beyond-the-table” interactions, representatives thus sought to sound out

how far their own constituents were prepared to compromise on the issues at stake and thus how

far their initial group frames could be stretched. Such cross-table scanning proved vital for

averting renewed framing contests and the risk of frame break by easing the homecoming.

Breaking up groups into potential allies and collaborators versus objectors also led the frames
 44

shared within arenas to become more “elastic”. While this led to more progress, it also ran the

risk of more confrontational members deciding to leave (as occurred in Phases II, IV, and VI).

DISCUSSION & CONCLUSION

Whereas a strategic frame alignment perspective focuses on how the content of frames is

negotiated by incorporating elements of different frames or transforming frames, we adopt an

interactional perspective which understands frame alignment as the outcome of interactions. In

doing so, we raise and address several critical but unexplored questions regarding how

interactional frames can be extended beyond the local interactional contexts from which they

emerge. We highlight how multiple interaction arenas influence the process of frame alignment

and explain why and when misalignment occurs. Focusing on the people who “do” the framing

helps to understand how the challenge of framing across multiple interaction arenas can be

navigated. We identify three cross-arena framing processes that are crucial for interactants to

access frame laminations of relational, issue and process framing of different interaction arenas

that can help them interlock misaligned interactional frames and avoid frame break and failure.

We find that collaborating on the edge of failure through such processes, while not eliminating

misalignment, can propel negotiations forward between groups with competing interests as the

threat of collapse makes interactants realise they must transcend their frame boundaries to keep

inter-organisational collaboration alive. We derive a theoretical process model that unveils the

dynamics of interactional frame alignment across multiple interaction arenas.

A Process Model of Navigating Frame Alignment Across Multiple Interaction Arenas

In Figure 6 we depict the theoretical process model we abstracted from our case, schematising

the process of frame alignment across multiple interaction arenas. Our model responds to our

research question in two parts.

------------------------------------
Insert Figure 6 about here
 45

------------------------------------
The left part of our model addresses the first part of our research question: “How do multiple

interaction arenas affect frame alignment?” Illustrating the two interaction arenas and processes

of frame alignment, Figure 6 depicts how frame laminations gradually lead to the emergence of a

shared frame within an interaction arena. However, these laminations are not available across

interaction arenas, leading to misalignment between arenas and potential frame break. Our model

highlights that frame alignment is not only about constructing shared meaning of the issue at

stake but that interactants also engage in assigning new meanings to their relationships and

ongoing interactions (Dewulf et al., 2009), with each new meaning laminated on the other.

Through relational framing, “parties interactively construct the meaning of self, other and

relationships” (Dewulf et al., 2009: 166). As our in-depth process study discloses, the industry

and NGO representatives in our case reframed how they saw each other and their relationships

with counterparts through “beyond-the-table” interactions. Such “backstage” settings (Goffman,

1978) can be crucial for interactants to re-evaluate and connect with one another (Watzlawick,

Beavin, & Jackson, 1967). Shifting from an antagonistic “us versus them” relationship towards a

collective identity (Gamson, 1992) is also likely to lead to a shift in “attributional orientation”

away from mutual accusations of blame or attributions of responsibility to certain causes or

actors (Snow & Benford, 1988: 474). When blame is no longer attributed to the other side, the

nature of the framing contest can move from being person-related to being content-related.

Such relational framing laminates onto issue framing, i.e., “how parties interactively

construct the meaning of issues” (Dewulf et al., 2009: 166). Shifting relational framing away

from blame towards identifying what the actual problem is (Benford & Snow, 2000; Snow &

Rochford, 1983) increases the willingness of parties to see the problem from the other’s

perspective. Through this “extension of interpretive scope” (Benford & Snow, 2000), interactants
 46

can move from a subjective to an intersubjective understanding of the issue at stake (Goffman,

1974). In our case the interactants’ relational framing of each other as trustworthy interlocutors

prevented the NGO and industry representatives from attributing insufficient progress on core

issues to each other’s lack of willingness. Instead, they were now more open to jointly exploring

the reasons for insufficient progress.

Issue framing and the shift from subjective to intersubjective understanding further

laminates on process framing, i.e., the way interactants construct “the meaning of the ongoing

interaction process between them” (Dewulf et al., 2009: 166). By framing their interaction as

collaborative, the participants in our case became more willing to compromise and move forward

with each other rather than solely pursuing their own goals (Donohue & Roberto, 1993). Frame

alignment is thus “constituted of an innumerable number of elements, amalgamated during the

ongoing process of interaction” (Gonos, 1977: 860), including cumulative laminations of

relational, issue and process framing. As such, frame alignment can be compared to a process of

sedimentation whereby shared frames emerge from an accumulation of laminations.

Ultimately, however, frame alignment always remains inherently fragile. This is because

relational, issue and process laminations are not readily accessible across different interaction

arenas, meaning frames cannot simply be “transported” and that efforts by interactants to

introduce frames into another interaction arena may lead to “slippage and misfirings” (Gray et

al., 2015: 119). Failing to take account of the need for frames to be aligned across interaction

networks can thus renew framing contests and jeopardise frame alignment.

This brings us to the second part of our research question: “how can frame alignment be

navigated across multiple interaction arenas?” We derive three interrelated cross-arena framing

processes that connect multiple interaction arenas to create interlocking frame alignment. These
 47

are depicted in the right-hand side of our process model (Figure 6), which provides a zoomed-in

view of how interactants navigate across multiple interaction arenas.

Through the process of “reciprocal frame stretching”, interactants stretch their shared

interactional frame within an interaction arena. In our case, after experiencing frame break in

Phases I and II the representatives began to discursively integrate not only their own constituents

but also the constituents of their counterparts in the negotiation process from Phase III onwards.

By offering each other reciprocal concessions, the representatives “mutually facilitated”

(Sebenius, 2013) each other’s crossing of red lines. This proved crucial for helping each other

avoid frame break in their respective “homecomings” by enabling the representatives to present

concessions made by their counterparts as a “win” to their constituents.

Our model highlights how a shared frame created in one interaction context can be stretched

to overlap with another frame through both integrative and bilateral frame stretching. In our case,

the frame shared in the SC was stretched through the integration of regular members into the

negotiation process. By participating in expert working groups, these constituents then came to

be recognised as active “frame makers” who now also “did” the framing. The representatives

also scanned their own and their counterparts’ groups for potential compromise lines, stretching

their constituents’ frames to enable overlap with the SC frame. Through scanning,

representatives tried to identify where frames already overlapped and where meanings still

needed to be matched. Matching meanings (Kellogg, 2014) and explicating a shared frame to

include interactants from other interaction arenas is crucial for enabling meanings to be

amplified beyond the interaction context from which they emerge (Gray et al., 2015; Snow,

2013; Snow, Vliegenthart, & Ketelaars, 2019).

 48

Through “reciprocal”, “integrative” and “bilateral” frame stretching, interactional frames

become more “elastic”. In our case, such stretching means that interactants no longer needed to

move away from their initial frames entirely when engaging in a new interaction context.

Instead, interactional frames became “interwoven and embedded within one another” (Diehl &

McFarland, 2010: 1716) across multiple interaction arenas. Although frame stretching ran the

risk of frame break and the collapse of collaborative partnership, this process was necessary to

push negotiating parties forward towards agreement on the highest common denominator.

Indeed, when negotiating parties collectively realised that they were collaborating “on the edge

of failure” this recognition of risk helped them move away from rejecting comprises towards

agreeing on ambitious commitments in order to maintain the collaboration.

Theoretical implications for framing theory. Our model contributes to framing theory by

highlighting the iterative nature of interlocked framing processes across multiple interaction

arenas. This contribution highlights an underexplored consequence of conceptualising frames as

interactional co-constructions rather than “free-floating packages of meaning” (Reinecke &

Ansari, 2021: 381). While framing scholarship has increasingly focused on the micro-processes

of framing and the negotiation of meaning through ongoing interaction (Cornelissen & Werner,

2014; Dewulf et al., 2009; Gray et al., 2015; Klein & Amis, 2021; Reinecke & Ansari, 2021),

studies in this stream have thus far favoured theoretical explanations of how frames solidify as

agreement emerges (Helms et al., 2012), how frames become dominant (Klein & Amis, 2021),

escalate (Cornelissen et al., 2014; Reinecke & Ansari, 2021), and eventually become

institutionalised (Gray et al., 2015). However, a closer look at multi-arena framing dynamics

reveals that frame alignment is an ongoing process in which frames are challenged, rejected or

revised as they are extended into a new interaction context. Our model shows that while frames
 49

cannot move beyond the interactions in which they emerge, they can be stretched and made more

“elastic” to incorporate a broader repertoire of laminated meanings (Gray et al., 2015).

In conceptualising frame alignment as an ongoing endeavour, we emphasise the agency of

audiences in ways that go beyond the notion of frame “resonance’ (Giorgi, 2017; Snow et al.,

1986), showing that frames do not simply resonate with a targeted audience (Snow & Benford,

1988) on account of their inherent properties or fit but gain resonance through interactional

dynamics in social situations. Our findings in this respect both align with and extend Dewulf et

al.’s (2009) suggestion that it is not only issues that get framed but also identities, relationships,

and the interaction process itself, revealing how resultant frames are laminated upon one another

and shape framing trajectories. Specifically, our model shows that interactants do not passively

adopt or “synchronise” frames (Cornelissen et al., 2014; Zilber, 2002) but are themselves “frame

makers”. This leads to constantly evolving laminations of interactional frame alignment. Our

processual account calls for and facilitates a shift away from outcome-oriented studies of

framing towards research aimed at understanding the emergent, processual, and recursive

character of interactional frame alignment.

Our notion of multi-arena framing dynamics also has important implications for

understanding framing as a source of meaning at the field level (Ansari et al., 2013). In

particular, by unveiling the iterative and interlocked process of framing across interaction arenas,

we respond to recent calls to examine framing as a way to “provide a fully dynamic, layered

account of meaning making in fields” (Leibel et al., 2018: 165). Studies of framing contests to

date have focused on such contests between specific groups, such as social movements versus

corporations (Vaccaro & Palazzo, 2015). However, conceptualising fields as constituted by a

myriad of overlapping interaction arenas entails a shift in focus onto iterative and ongoing frame
 50

alignment processes across arenas. Whereas prior studies have focused on the construction of

“field settlements” as a precursor to the institutionalisation of more widely shared meaning

systems such as new cultural codes (Weber et al., 2008) or organisational forms (Rao & Kenney,

2008), our study shows that fields rarely converge around a fully unified consensus through

frame shifts at the field level (Ansari et al., 2013; Lounsbury, Ventresca, & Hirsch, 2003;

Reinecke & Ansari, 2016). According to our model, frame alignment rarely results in the

definitive resolution of issues between rival groups but instead is a work-in-progress that

requires ongoing and back-and-forth alignment between and across arenas. This answers to

Cornelissen and Werner’s (2014) call for researchers to move beyond the dichotomy of

cooperation versus competition in framing struggles by emphasising the complexity of iterative

multi-arena framing in which collaborative and competitive framing processes co-exist.

Our study also has implications for understanding the process of frame dissemination.

Specifically, we add nuance to conceptualisations of “frame amplification”, i.e., “the process by

which frames generated in microlevel interactions move to the meso and macro levels” (Gray et

al., 2015: 120). Such amplification, as theorised by Gray et al. (2015: 120), relies on the

diffusion of frames via the “transfer of meaning through broadening or overlapping networks of

interactants”. However, we find that efforts at such transferral can result in conflict if others in

overlapping networks contest the frame. Our findings thus highlight the importance of focusing

on the iterative dynamics of framing, since framing processes in overlapping networks can be

mutually reinforcing or undermining, serving either to accelerate frame amplification or to

destabilise shared frames. Our research provides the basis for a much more dynamic and multi-

pronged approach to the study of framing processes that can help explain how and why
 51

interactional frames are successfully amplified in some cases but merely “fizzle out” in others.

Iterative framing contests can thus shape field dynamics in unexpected ways.

We further add to framing theory by drawing attention to the undertheorized relational

dynamics that are crucial for constructing shared frames. Our study thereby answers to calls from

framing scholars to seek insights into the relational dynamics between those who “do” the

framing (Cornelissen & Werner, 2014; Reinecke & Ansari, 2021). By studying frame alignment

processes across interaction arenas, we highlight the consequences of frames being relationally-

bound to interactants in specific interaction arenas (Lewicki et al., 2003). In contrast to how

frames have been conceptualised in the social movement literature, i.e., as cognitive constructs

that are upheld through words (Snow et al., 1986) and which can be emotionally charged through

e.g., photographs (Klein & Amis, 2021), our case demonstrates that frames are grounded in

relationships between those interacting. This explains why a new frame that emerges from

relational framing in one arena may be prone to break in another arena due to the lack of

relational frame laminations in the latter. This insight encourages scholars to not only study the

properties and content of frames but also the relationships in which frames are grounded.

Building on these insights, future research could zoom in even more closely on the

emotional and identity challenges faced by the individuals tasked with the complex role of

navigating between different interaction arenas. Such challenges to social identity are inevitable

when navigating multiple interaction arenas, we argue, since “parties work out definitions of

their identities and relationships by negotiating them in social interaction” (Dewulf et al., 2009:

171), which explains why interactants can have “multiple, fragmented, processual and

situational, rather than coherent, fixed and stable” identities (Karreman & Alvesson, 2001: 63).

Scholars could thus study how representatives overcome the difficulties they face when they
 52

experience a shift of frame at a personal level yet still feel compelled in their roles as appointed

representatives to stay “true” to their initial group’s frame. How do such individuals cope with

possible identity conflicts when interacting across multiple interaction arenas over time? If it is

accepted that not only frames can be emotionally charged but also interactions themselves

(Collins, 2004; Reinecke & Ansari, 2021), then another important question for future research

concerns the emotional dynamics between those who co-construct a frame.

Theoretical implications for inter-organisational collaboration and MSPs. Our study also

contributes to the literature on MSPs and inter-organisational collaboration more generally.

Previous research has focused on the substantive question of why multi-party collaboration

succeeds or fails (Alamgir & Banerjee, 2019; Barnett & King, 2008; Zuzul, 2019). Such research

has repeatedly shown how collaboration often only induces symbolic commitment (Wijen,

2014). However, our case presents a rare example of multi-party collaboration that progressed

despite coming close to collapse in its early stages. In the Textiles Partnership, corporate

participants eventually agreed to ambitious commitments with timelines and goals for

sustainable sourcing they had initially rejected as unfeasible for them to implement. In contrast

to studies that have largely focused on the role of societal and political pressures as key drivers

for companies to make concessions to societal demands and engage in MSPs (Bartley & Child,

2014), our study suggests that such external pressures alone are not sufficient to sustain

commitment. In our case, public pressure in the wake of the 2013 Rana Plaza disaster had largely

subsided by the time the initially rejected commitments were re-introduced. By highlighting an

especially challenging but under-explored complication in negotiating agreement in multi-party

settings, i.e., the “two-table problem” (Colosi, 1983; Gray & Purdy, 2018; Sebenius, 2013), our

analysis foregrounds what happens inside collaborative processes as we unpack the two-table
 53

problem and identify the processes whereby different negotiation tables can be navigated. A key

insight of our study is that successful collaboration is likely to happen “on the edge of failure”

because representatives must push each other to cross their respective group’s red lines to reach

the highest rather than the lowest common denominator as a basis for substantial collaboration.

And while this propels collaboration forward and deepens commitments, it also risks conflict and

failure, as evidenced by several “exit waves” of companies who refused to undertake the more

ambitious commitments their representatives had negotiated on their behalf.

Our close investigation into the “two-table problem” reveals how the complexities of

collaborative relationships have been concealed by the tendency of research on settlements

between stakeholder groups to treat these groups as if they were homogenous actors (Bartley,

2007; Helms et al., 2012). To understand processes of conflict and settlement in collaboration,

we argue, researchers need to “drop down” to a lower level of analysis that not only considers

the relationships between stakeholders at one level but further examines these dynamics as they

play out between representatives and their constituent groups. While we agree that successful

collaboration “depends upon clear communication between representatives and their

organizations” (Margerum, 2008: 495-496), our in-depth study unpacks the cross-arena framing

processes through which negotiators continuously realign locally negotiated frames across

negotiation tables. In exploring the role of multi-arena framing dynamics among participants

with divergent goals and interests, our study sheds light on important but hitherto under-studied

aspects of multi-party negotiations and other inter-organisational forms of collaboration.

Our analysis of the “two-table problem” also highlights the need for conceptualising

collaboration as a dynamic multilevel system (Bryson et al., 2015). This contrasts with

scholarship that has tended to focus on the creation of MSPs as an outcome in itself (Helms et
 54

al., 2012; Koschmann et al., 2012). Addressing the paucity of empirical research on the iterative

processes whereby frames are renewed and realigned over time and across negotiation tables, our

six-year study of the Textiles Partnership unpacks the “nonlinear and emergent nature of

collaboration”, showing how “collaboration evolves as parties interact over time” (Thomson &

Perry, 2006: 22). Specifically, our account of interconnected topic-framing cycles evidences and

elucidates the dynamic nature of “interaction goals” (Wilson & Putnam, 1990) and the ways in

which parties continuously redefine issues as they navigate multiple negotiation tables. This

demonstrates that MSPs are never ‘complete’ just because initial goals have been agreed upon

but instead require an ongoing “balancing act” to keep parties aligned. As such, our process

model points to the importance of ongoing and within-MSP negotiations that can renew and even

increase commitments after the initial establishment of collaboration.

Finally, our process model has implications for the governance of voluntary collaboration

more broadly, contributing both to the wider scholarship on inter-organisational collaboration

(Hardy et al., 2005; Human & Provan, 2000; Ring & van de Ven, 1994; Zuzul, 2019) and to

voluntary governance arrangements (Barnett & King, 2008; Levy & Prakash, 2003; Ostrom,

1990; Prakash & Potoski, 2007). With our model we provide insights into how commitment to

collective goals can be negotiated and sustained amongst a large number of autonomous

organisations. Although frame alignment is likely more difficult among groups with competing

interests, maintaining commitment to joint goals has been shown to be difficult even in shared

interest coalitions such as multi-firm networks, including industry coalitions, membership

associations, and other inter-organisational relations (Barnett & King, 2008; Human & Provan,

2000). In all these constellations, representatives from different groups negotiate on behalf of

their groups with external parties and are therefore likely to encounter the two-table problem.
 55

Such negotiations can rapidly unravel if negotiated frames are subsequently revoked by

participants. Unlike situations such as trade union bargaining in which representatives have clear

mandates, representatives in partnerships are typically only loosely mandated by their

constituents to negotiate on their behalf. This inevitably places them in a precarious position,

since they can lose their legitimacy, risking the collapse of the collaboration. Future research

could focus on how different degrees of authorisation vested in representatives and different

levels of accountability affect the strategies adopted for managing framing across multiple

interaction arenas and associated outcomes (Jackson & King, 1983).

Boundary Conditions
Our study has several important boundary conditions. First, due to the confidential nature of

negotiations in the Textiles Partnership we were not able to directly observe interactions during

SC meetings. While scholars have proposed that interpretive interviewing is capable of capturing

the construction of meaning (Langley & Meziani, 2020), interviews are nevertheless subject to

potential hindsight bias. They are also likely to miss out on interactional dynamics observed

through ethnographic methods. Future research could thus apply more “close-up” observational

methods or conversation analysis to identify additional interactional dynamics that shape cross-

arena frame alignment processes.

Second, in this study we have focused only on two main interaction arenas, i.e., partnership

and stakeholder group arenas. Future research could expand our insights to the intra-

organisational arena wherein framing contests often erupt between departments with different

sub-cultures, identities, and routines (Howard-Grenville, 2006; Kaplan, 2008; Kellogg, 2014).

Our study has hinted at a third negotiation table in participants’ home organisations, specifically

in the context of conflicts between CSR and procurement. CSR managers often function as

“internal activists” who need to “sell” CSR issues by framing them internally (Wickert & De
 56

Bakker, 2018). Future research could build on this existing knowledge and usefully focus on how

negotiators can build up engagement tactics to help win over “at-home” groups.

The third boundary condition concerns the role of government as a convener in the Textiles

Partnership. Even though the role of the government became much less prominent as public

attention subsided, it still played an important role in bringing parties to the negotiating table.

Although much stricter private agreements have been achieved that were not initiated by any

public agency, as in the case of the Bangladesh Accord on Fire and Building Safety (Reinecke &

Donaghey, 2015), the role of governments nonetheless remains an important question in

collaborative governance (Kourula, Moon, Salles-Djelic, & Wickert, 2019). Future research

could usefully compare negotiation dynamics in MSPs with and without government

sponsorship, as well as the influence of external influences more broadly, such as media

attention, activists’ campaigning, local legislation, and international supervision.

 57

REFERENCES
Aaldering, H., & De Dreu, C. K. 2012. Why hawks fly higher than doves: Intragroup conflict in
representative negotiation. Group Processes & Intergroup Relations, 15: 713–724.
Agranoff, R. 2012. Collaborating to manage: A primer for the public sector. Washington, DC:
Georgetown University Press.
Alamgir, F., & Banerjee, S. B. 2019. Contested compliance regimes in global production
networks: Insights from the Bangladesh garment industry. Human Relations, 72: 272–297.
Alvarez, S., & Sachs, S. 2021. Where do stakeholders come from? Academy of Management
Review, 48. http://dx.doi.org/10.5465/amr.2019.0077
Ansari, S., Wijen, F., & Gray, B. 2013. Constructing a climate change logic: An institutional
perspective on the “Tragedy of the Commons”. Organization Science, 24: 1014–1040.
Ansell, C., & Gash, A. 2007. Collaborative governance in theory and practice. Journal of Public
Administration Research and Theory, 18(4): 543–571.
Balogun, J., & Johnson, G. 2004. Organizational restructuring and middle manager sensemaking.
Academy of Management Journal, 47: 523–549.
Barnett, M. L., & King, A. A. 2008. Good fences make good neighbors: A longitudinal analysis
of an industry self-regulatory institution. Academy of Management Journal, 51: 1150–
1170.
Bartley, T. 2007. Institutional emergence in an era of globalization: The rise of transnational
private regulation of labor and environmental conditions. American Journal of Sociology,
113: 297–351.
Bartley, T., & Child, C. 2014. Shaming the corporation: The social production of targets and the
anti-sweatshop movement. American Sociological Review, 79: 653–679.
Benford, R. D., & Snow, D. A. 2000. Framing processes and social movements: An overview
and assessment. Annual Review of Sociology, 26: 611–639.
Berends, H., & Sydow, J. 2019. Introduction: Process views on inter-organizational
collaborations. In J. Sydow & H. Berends (Eds.), Managing inter-organizational
collaborations: Process views, vol. 64: 1–10. Bingley, UK: Emerald Publishing.
Bryson, J. M., Crosby, B. C., & Stone, M. M. 2015. Designing and implementing cross‐sector
collaborations: Needed and challenging. Public Administration Review, 75: 647–663.
Clark, H. H. 1996. Using language. New York: Cambridge University Press.
Collins, R. 2004. Interaction ritual chains. Princeton, NJ: Princeton University Press.
Colosi, T. 1983. Negotiation in the public and private sectors: A core model. American
Behavioral Scientist, 27: 229–253.
Colosi, T. 1985. A core model of negotiation. In R. J. Lewicki, & J. A. Litterer (Eds.),
Negotiation: Readings, exercises and cases: 313–319. Homewood, IL: Irwin.
Cornelissen, J. P., Mantere, S., & Vaara, E. 2014. The contraction of meaning: The combined
effect of communication, emotions, and materiality on sensemaking in the Stockwell
shooting. Journal of Management Studies, 51: 699–736.
Cornelissen, J. P., & Werner, M. D. 2014. Putting framing in perspective: A review of framing
and frame analysis across the management and organizational literature. Academy of
Management Annals, 8: 181–235.
Dewulf, A., Gray, B., Putnam, L., Lewicki, R. J., Aarts, N., Bouwen, R., & Van Woerkum, C.
2009. Disentangling approaches to framing in conflict and negotiation research: A meta-
paradigmatic perspective. Human Relations, 62: 155–193.
 58

Diehl, D., & McFarland, D. 2010. Toward a historical sociology of social situations. American
Journal of Sociology, 115: 1713–1752.
Doh, J. P., Tashman, P., & Benischke, M. H. 2019. Adapting to grand environmental challenges
through collective entrepreneurship. Academy of Management Perspectives, 33: 450–468.
Donaghey, J., & Reinecke, J. 2018. When industrial democracy meets corporate social
responsibility – A comparison of the Bangladesh accord and alliance as responses to the
Rana Plaza disaster. British Journal of Industrial Relations, 56: 14–42.
Donnellon, A., & Gray, B. 1990. An interactive theory of reframing in negotiation.
Unpublished manuscript, Pennsylvania State University Center for Research in Conflict and
Negotiation: State College, PA.
Donnellon, A., Gray, B., & Bougon, M. G. 1986. Communication, meaning, and organized
action. Administrative Science Quarterly, 31: 43–55.
Donohue, W. A., & Roberto, A. J. 1993. Relational development as negotiated order in hostage
negotiation. Human Communication Research, 20: 175–198.
Dundon, T., & Ryan, P. 2010. Interviewing reluctant respondents: Strikes, henchmen, and Gaelic
games. Organizational Research Methods, 13: 562–581.
Eisenhardt, K. M., & Graebner, M. E. 2007. Theory building from cases: Opportunities and
challenges. Academy of Management Journal, 50: 25–32.
Entman, R. M. 1993. Framing: Toward clarification of a fractured paradigm. Journal of
Communication, 43: 51–58.
Ferraro, F., & Beunza, D. 2018. Creating common ground: A communicative action model of
dialogue in shareholder engagement. Organization Science, 29: 1187–1207.
Ferraro, F., Etzion, D., & Gehman, J. 2015. Tackling grand challenges pragmatically: Robust
action revisited. Organization Studies, 36: 363–390.
Gamson, W. A. 1992. The social psychology of collective action. Frontiers in Social Movement
Theory, 1: 53–76.
George, G., Fewer, T. J., Lazzarini, S., McGahan, A. M., & Puranam, P. 2023. Partnering for
grand challenges: A review of organizational design considerations in public-private
collaborations. Journal of Management. http://dx.doi.org/10.1177/01492063221148992
Giorgi, S. 2017. The mind and heart of resonance: The role of cognition and emotions in frame
effectiveness. Journal of Management Studies, 54: 711–738.
Goffman, E. 1974. Frame analysis: An essay on the organization of experience. Cambridge,
MA: Harvard University Press.
Goffman, E. 1978. The presentation of self in everyday life. Harmondsworth, UK: Penguin.
Gonos, G. 1977. “Situation” versus “frame”: The “interactionist” and the “structuralist” analyses
of everyday life. American Sociological Review, 42: 854–867.
Granqvist, N., & Laurila, J. 2011. Rage against self-replicating machines: Framing science and
fiction in the US nanotechnology field. Organization Studies, 32: 253–280.
Gray, B. 1989. Collaborating: Finding common ground for multiparty problems. San
Francisco, CA: Jossey-Bass.
Gray, B., & Purdy, J. M. 2018. Collaborating for our future: Multistakeholder partnerships for
solving complex problems. Oxford, UK: Oxford University Press.
Gray, B., Purdy, J. M., & Ansari, S. 2015. From interactions to institutions: Microprocesses of
framing and mechanisms for the structuring of institutional fields. Academy of Management
Review, 40: 115–143.
 59

Gray, B., Purdy, J. M., & Ansari, S. 2022. Confronting power asymmetries in partnerships to
address grand challenges. Organization Theory, 3.
https://doi.org/10.1177/26317877221098765
Grimm, J. 2019. Private governance as an institutional response to wicked problems. Nomos
Verlagsgesellschaft mbH & Co. KG.
Guérard, S., Bode, C., & Gustafsson, R. 2013. Turning point mechanisms in a dualistic process
model of institutional emergence: The case of the diesel particulate filter in Germany.
Organization Studies, 34: 781–822.
Hardy, C., Lawrence, T. B., & Grant, D. 2005. Discourse and collaboration: The role of
conversations and collective identity. Academy of Management Review, 30: 58–77.
Helms, W. S., Oliver, C., & Webb, K. 2012. Antecedents of settlement on a new institutional
practice: Negotiation of the ISO 26000 standard on social responsibility. Academy of
Management Journal, 55: 1120–1145.
Howard-Grenville, J. A. 2006. Inside the “black box”: How organizational culture and
subcultures inform interpretations and actions on environmental issues. Organization &
Environment, 19: 46–73.
Human, S. E., & Provan, K. G. 2000. Legitimacy building in the evolution of small-firm
multilateral networks: A comparative study of success and demise. Administrative Science
Quarterly, 45: 327–365.
Jackson, C. N., & King, D. C. 1983. The effects of representatives' power within their own
organizations on the outcome of a negotiation. Academy of Management Journal, 26: 178–
184.
Jarzabkowski, P., Bednarek, R., & Lê, J. K. 2014. Producing persuasive findings: Demystifying
ethnographic textwork in strategy and organization research. Strategic Organization, 12:
274–287.
Kaplan, S. 2008. Framing contests: Strategy making under uncertainty. Organization Science,
19: 729–752.
Karreman, D., & Alvesson, M. 2001. Making newsmakers: Conversational identity at work.
Organization Studies, 22: 59–89.
Kellogg, K. C. 2014. Brokerage professions and implementing reform in an age of experts.
American Sociological Review, 79: 912–941.
Klein, J., & Amis, J. M. 2021. The dynamics of framing: Image, emotion and the European
migration crisis. Academy of Management Journal, 64.
https://doi.org/10.5465/amj.2017.0510
Koschmann, M. A., Kuhn, T. R., & Pfarrer, M. D. 2012. A communicative framework of value
in cross-sector partnerships. Academy of Management Review, 37: 332–354.
Kourula, A., Moon, J., Salles-Djelic, M.-L., & Wickert, C. 2019. New roles of government in the
governance of business conduct: Implications for management and organizational research.
Organization Studies, 40: 1101–1123.
Langley, A. 1999. Strategies for theorizing from process data. Academy of Management
Review, 24: 691–710.
Langley, A., & Meziani, N. 2020. Making interviews meaningful. The Journal of Applied
Behavioral Science, 56: 370–391.
Langley, A., & Tsoukas, H. 2010. Introducing perspectives on process organization studies.
Process, Sensemaking, and Organizing, 1: 1–27.
 60

Lee, M., Ramus, T., & Vaccaro, A. 2018. From protest to product: Strategic frame brokerage in a
commercial social movement organization. Academy of Management Journal, 61: 2130–
2158.
Leibel, E., Hallett, T., & Bechky, B. A. 2018. Meaning at the source: The dynamics of field
formation in institutional research. Academy of Management Annals, 12: 154–177.
Levy, D. L., & Prakash, A. 2003. Bargains old and new: Multinational corporations in global
governance. Business and Politics, 5: 131–150.
Lewicki, R. J., Gray, B., & Elliott, M. 2003. Making sense of intractable environmental
disputes. Washington, DC: Island Press.
Litrico, J.-B., & David, R. J. 2017. The evolution of issue interpretation within organizational
fields: Actor positions, framing trajectories, and field settlement. Academy of Management
Journal, 60: 986–1015.
Lounsbury, M., Ventresca, M., & Hirsch, P. M. 2003. Social movements, field frames and
industry emergence: A cultural-political perspective on US recycling. Socio-Economic
Review, 1: 71–104.
Margerum, R. D. 2008. A typology of collaboration efforts in environmental management.
Environmental Management, 41: 487–500.
Margerum, R. D., & Robinson, C. J. 2016. The challenges of collaboration in environmental
governance: Barriers and responses. Cheltenham, UK: Edward Elgar Publishing.
Mena, S., & Palazzo, G. 2012. Input and output legitimacy of multi-stakeholder initiatives.
Business Ethics Quarterly, 22: 527–556.
Meyer, R. E., & Hoellerer, M. A. 2010. Meaning structures in a contested issue field: A
topographic map of shareholder value in Austria. Academy of Management Journal, 53:
1241–1262.
Ostrom, E. 1990. Governing the commons: The evolution of institutions for collective action.
Cambridge, UK: Cambridge University Press.
Prakash, A., & Potoski, M. 2007. Collective action through voluntary environmental programs:
A club theory perspective. Policy Studies Journal, 35: 773–792.
Purdy, J. M. 2012. A framework for assessing power in collaborative governance processes.
Public Administration Review, 72: 409–417.
Putnam, R. D. 1988. Diplomacy and domestic politics: The logic of two-level games.
International Organization, 42: 427–460.
Rao, H., & Kenney, M. 2008. New forms as settlements. In R. Greenwood, C. Oliver, K. Sahlin,
& R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism: 352–371.
Los Angeles, CA: SAGE.
Reinecke, J., & Ansari, S. 2016. Taming wicked problems: The role of framing in the
construction of corporate social responsibility. Journal of Management Studies, 53: 299–
329.
Reinecke, J., & Ansari, S. 2021. Microfoundations of framing: The interactional production of
collective action frames in the Occupy Movement. Academy of Management Journal, 64:
378–408.
Reinecke, J., & Donaghey, J. 2015. After Rana Plaza: Building coalitional power for labour
rights between unions and (consumption-based) social movement organisations.
Organization, 22: 720–740.
 61

Reinecke, J., & Donaghey, J. 2021. Political CSR at the coalface – The roles and contradictions
of multinational corporations in developing workplace dialogue. Journal of Management
Studies, 58: 457–486.
Reinecke, J., & Donaghey, J. 2022. Transnational representation in global labour governance and
the politics of input legitimacy. Business Ethics Quarterly, 32: 438–474.
Ring, P. S., & van de Ven, A. H. 1994. Developmental processes of cooperative
interorganizational relationships. Academy of Management Review, 19: 90–118.
Ring, P. S., & van de Ven, A. H. 2019. Relational bonds underlying cooperative inter-
organizational relations in different societal contexts. In J. Sydow & H. Berends (Eds.),
Managing inter-organizational collaborations: Process views, vol. 64: 13–37. Bingley,
UK: Emerald Publishing.
Schuessler, E., Frenkel, S. J., & Wright, C. F. 2019. Governance of labor standards in Australian
and German garment supply chains: The impact of Rana Plaza. ILR Review, 72: 552–579.
Schuessler, E., Lohmeyer, N., & Ashwin, S. 2023. “We can’t compete on human rights”:
Creating market-protected spaces to institutionalize the emerging logic of responsible
management. Academy of Management Journal, 66: 1071–1101.
Sebenius, J. K. 2013. Level two negotiations: Helping the other side meet its “behind‐the‐table”
challenges. Negotiation Journal, 29: 7–21.
Selsky, J. W., & Parker, B. 2005. Cross-sector partnerships to address social issues: Challenges
to theory and practice. Journal of Management, 31: 849–873.
Snow, D. A. 2013. Social movements. In D. A. Snow, D. della Porta, B. Klandermans, & D.
McAdam (Eds.), The Wiley‐Blackwell encyclopedia of social and political movements:
1201–1205. Malden, MA: Wiley Blackwell.
Snow, D. A., & Benford, R. D. 1988. Ideology, frame resonance, and participant mobilization.
International Social Movement Research, 1: 197–217.
Snow, D. A., & Rochford, E. B. 1983. Structural availability, the alignment process and
movement recruitment. Paper presented at the Annual Meeting of the American
Sociological Association, Detroit, MI.
Snow, D. A., Vliegenthart, R., & Ketelaars, P. 2019. The framing perspective on social
movements: Its conceptual roots and architecture. In D. A. Snow, S. A. Soule, H. Kriesi, &
H. J. McCammon (Eds.), The Wiley Blackwell companion to social movements (2nd ed.):
392–410. Hoboken, NJ: Wiley Blackwell.
Snow, D. A., Worden, S. K., Rochford, E. B., & Benford, R. D. 1986. Frame alignment
processes, micromobilization, and movement participation. American Sociological Review,
51: 464–481.
Soderstrom, S. B., & Weber, K. 2020. Organizational structure from interaction: Evidence from
corporate sustainability efforts. Administrative Science Quarterly, 65: 226–271.
Steinberg, M. W. 1999. The talk and back talk of collective action: A dialogic analysis of
repertoires of discourse among nineteenth-century English cotton spinners. American
Journal of Sociology, 105: 736–780.
Tannen, D. 1985. Frames and schemas in interaction. Quaderni di Semantica, 6: 326–335.
Thomson, A. M., & Perry, J. L. 2006. Collaboration processes: Inside the black box. Public
Administration Review, 66: 20–32.
Turcotte, M. F., & Pasquero, J. 2001. The paradox of multistakeholder collaborative roundtables.
The Journal of Applied Behavioral Science, 37: 447–464.
 62

Vaccaro, A., & Palazzo, G. 2015. Values against violence: Institutional change in societies
dominated by organized crime. Academy of Management Journal, 58: 1075–1101.
Watzlawick, P., Beavin, J. H., & Jackson, D. D. 1967. Pragmatics of human communication.
New York: W.W. Norton.
Weber, K., Heinze, K. L., & DeSoucey, M. 2008. Forage for thought: Mobilizing codes in the
movement for grass-fed meat and dairy products. Administrative Science Quarterly, 53:
529–567.
Wickert, C., & De Bakker, F. G. 2018. Pitching for social change: Toward a relational approach
to selling and buying social issues. Academy of Management Discoveries, 4: 50–73.
Wijen, F. 2014. Means versus ends in opaque institutional fields: Trading off compliance and
achievement in sustainability standard adoption. Academy of Management Review, 39:
302–323.
Wilson, S. R., & Putnam, L. L. 1990. Interaction goals in negotiation. Annals of the
International Communication Association, 13: 374–406.
Zietsma, C., Groenewegen, P., Logue, D. M., & Hinings, C. R. 2017. Field or fields? Building
the scaffolding for cumulation of research on institutional fields. Academy of Management
Annals, 11: 391–450.
Zilber, T. B. 2002. Institutionalization as an interplay between actions, meanings, and actors:
The case of a rape crisis center in Israel. Academy of Management Journal, 45: 234–254.
Zilber, T. B., & Meyer, R. E. 2022. Positioning and fit in designing and executing qualitative
research. The Journal of Applied Behavioral Science, 58: 377–392.
Zuzul, T. W. 2019. “Matter battles”: Cognitive representations, boundary objects, and the failure
of collaboration in two smart cities. Academy of Management Journal, 62: 739–764.
 63

Figure 1: The Steering Committee Negotiation Table in the Textiles Partnership

Figure 2: Process Chronology of the Textiles Partnership

 64

Figure 3: Laminations of Interactional Frame Alignment Within Interaction Arena

Figure 4: Misalignment of Laminated Frames Across Multiple Interaction Arenas

 65

Figure 5: Reciprocal, Integrative and Bilateral Cross-Arena Framing Processes

 66

Figure 6: A Process Model of Interlocking Multiple Interaction Arenas

 67

Table 1: Multi-Phase Data Collection Process

COLLECTION 2014 2015 2016 2017 2018 2019 2020 2021 n.n. Sum
PERIOD (and (and
before) after)

Formal Interviews
Steering Committee
Industry reps 0 2 4 5 1 0 6 18
NGO reps 0 2 2 3 4 1 4 16
Standards rep 0 0 1 0 1 0 1 3
Union rep 0 0 1 0 0 1 1 3
Government reps 0 0 2 1 1 2 1 7
Regular members
Industry members 0 3 9 3 4 0 4 23
NGO members 0 0 3 0 2 0 0 5
Standards members 0 0 2 0 0 0 0 2
Government members 0 1 1 0 0 0 0 2
Non-members
Media 0 1 1 0 0 0 0 2
Total number 0 9 26 12 13 4 17 81
Total hours 00:00:00 02:37:05 19:49:39 05:06:29 05:22:46 02:38:05 14:52:04 50:26:08

Informal Interviews*
Total number 0 2 3 4 0 0 2 11
*including members of NGOs, government, academia and industry

Observations
Total hours 08:00:00 00:00:00 08:00:00 17:00:00 00:00:00 00:00:00 00:00:00 33:00:00
 68

Documents & Archives

Industry 25 505 280 237 248 230 316 2 110 1953
NGOs 1 3 29 8 43 447 0 7 24 562
Standards 14 0 63 3 0 0 0 0 7 87
Unions 1 1 2 3 0 0 0 0 0 7
Government 49 37 125 82 35 38 8 57 80 511
Total number (in pages) 90 546 499 333 326 715 324 66 221 3120
 69

Table 2: Data Table with Empirical Examples of Cross-Arena Framing Processes

Reciprocal frame stretching Integrative frame stretching Bilateral frame stretching

through reciprocal concessions through integrating experts through cross-table scanning

Relational Relating to each other as Relating to each other as Relating to each other as
framing trustworthy partners legitimate experts progressive frontrunners
“how parties “[Peoples] personalities were very “Partnership members can participate in “Of course, the [NGOs] tried to convince
interactively construct important because ‘the tone makes the working groups if they have the necessary me, and others perhaps also. There were
the meaning of self, music’. A lot depends on trust. I really expertise.” (IND26-SC) informal conversations on the side, and
other and relationships” have to believe my counterpart. I have – Silvia said ‘Look, you could go further,
(Dewulf et al., 2009: to a certain extent – also to trust that what “Respect has to be earned. And basically it why are you holding back? Can’t you do
166) he says about himself will actually has to be earned in working groups where more?’. And then she sprinkled a bit of
happen and can be implemented … And we have gotten to know each other more.” sugar and was quite nice and acted a bit
trust has grown over time.” (GOV3-SC) (UNI2-SC) friendly and suggested ‘Come on, step a
bit more on the gas’. And then I always
thought ‘Yes, man, actually you need to
step on the gas a bit more’ … so you start
thinking again, you take on a different
perspective, you think about it again.”
(IND34)

“[P]ersonal relationships were built up

through this work in the Textiles
Partnership – relationships that never
existed before. By engaging with one
another face-to-face, by trying to have a
conversation, whether over coffee during
the break or something else, by coming
into contact with people and also
clarifying things with each other over
coffee or at a meeting and suddenly
realising ‘Well, they have a point after
all!’ It was important that our
relationships with one another grew.”
(IND36)
 70

Issue framing Creating shared understanding Creating a shared understanding Creating an understanding of
of why issues require of issues based on expertise lines of compromise on issues
“how parties concessions
interactively construct “[In expert groups] topics were dealt with “We [steering committee members]
the meaning of issues” “What has ultimately caused this [to much more intensively. We put all check out beforehand ‘Where are the red
(Dewulf et al., 2009: collaborate through crossing red lines information on the table, to surface and lines, where do we have consensus,
166) was] working with each other. An talk openly about what obstacles there are where do we have no consensus?’.”
increase in trust also between the [for each group], what reasons there are (GOV7-SC)
different groups of actors and ultimately [that hinder progress] … It became clear
also the evolving understanding that this that it wasn’t about just defeating one
partnership can actually achieve things another per se. Instead, we really started to
and that we have to make concessions if work on establishing consensus-oriented
we don’t want the partnership to fail.” solutions together, so we said ‘Okay, this
(IND22-SC) is the big picture. This might not be
possible yet, but what could we do as a
first step?’ And I think this was key to
moving from our red lines … There are
still red lines we can’t move away from,
but in areas where we started to speak with
others, and through speaking with each
other, we moved closer together and
started to implement concrete actions.”
(IND36)

Process framing Framing interaction as taking Framing interaction as problem- Framing interaction as finding
risks together solving activity allies in bilateral negotiations
“how parties
interactively construct “[We realised] we have to try together, to “Cooperation happens in the expert group “There have also been telephone calls in
the meaning of the see how we can possibly bring this topic … When you’ve identified certain which people have simply talked on the
ongoing interaction closer to its goal ... It’s difficult to really common lines in one area (laughs), you phone, bilaterally, or three people have
process between them” calculate [what participating in such an dare to approach the companies. And in spoken on the phone ... to talk rather than
(Dewulf et al., 2009: agreement will cost you]. We realised the expert groups this may work more in the big group, to question topics again
166) that the risk is actually that you either fail easily because, let’s say, people think – ‘Why are you on this side, why is the
alone or you fail with everyone else more constructively about these issues.” red line here, where would a possible
(laughs).” (IND15-SC) (NGO12-SC) consensus be?’ So then various
possibilities arose. This wasn’t organised
“In these working groups, solutions are by the Textiles Partnership’s organising
prepared based on consensus among committee, but I think it came out of the
working group participants. These group … we looked at what the fears
 71

solutions are then submitted to the steering were [others had] and where the
committee. So actually these proposals possibilities were to move forward ...
[for solutions] are adopted twice, so to Also then we got allies and said ‘Hey, we
speak. Once by the members who’ve think alike, let’s see how we can get this
worked on them together, and then again to work? How can we dissuade the others
by the steering committee where all the from their, yes, refusal attitude or red
stakeholder groups are mirrored and meant line? How can we work on moving
to just approve it.” (IND26-SC) forward on the issues?’.” (IND36)
 72

Biographical Sketch

Julia Grimm ([email protected]) is an Assistant Professor of Management, Organisation

and Society at Stockholm Business School, Stockholm University. Drawing on qualitative
research methods, Julia studies multi-stakeholder partnerships, transnational governance, and
social movements, all in the context of global supply chains and social and environmental
sustainability issues. Julia’s research takes a process perspective and is grounded in framing and
paradox theory.

Juliane Reinecke ([email protected]) is Professor of Management Studies at Saïd

Business School, University of Oxford. Juliane uses qualitative methods to study transnational
commons governance, multistakeholder collaborations, business sustainability, social
movements, and time and temporality. Juliane received her Ph.D. from the University of
Cambridge. She serves as a trustee of the Society for the Advancement of Management Studies
(SAMS) and as Associate Editor of the Academy of Management Journal.

View publication stats