Knappe (2020)

Journal of International Relations and Development
https://doi.org/10.1057/s41268-020-00197-6
ORIGINAL ARTICLE
Representation as practice: agency and relationality

in transnational civil society
Henrike Knappe1
© Springer Nature Limited 2020
Abstract
This paper argues that political representation in transnational civil society networks
needs to be investigated as practice with regard to its flexibility, relationality and
dialogical agency. Analysis of transnational representation from a practice-theo-
retical perspective can facilitate a better understanding of the actual representation
practices of civil society actors in a transnational setting. The question raised is this:
how do representation practices ensue, change or shift within the broader structures
in which they are embedded? By focusing on flexibility, relationality and dialogical
agency as the key theoretical concepts for representation practice, this study delves
deeper into those aspects through empirical analysis of qualitative interviews with
activists from two major transnational civil society networks: the Clean Clothes
Campaign and Friends of the Earth. The study finds that transnational representation
in civil society networks evolves in a non-linear fashion, is characterised by shifting
agency, discursive claims and a disembodiment of representation. The paper con-
cludes with a discussion of how future research can pursue such critical engagement
without falling back into standard, static notions of political representation.
Keywords Political representation · Practice theory · Representative claims ·

Transnational civil society
Introduction
Transnational civil society actors claim to represent constituencies—for example,

garment workers in Bangladesh or farmers in Nigeria—without having been for-
mally elected by these groups to act on their behalf. This kind of non-electoral rep-
resentation takes place inside and outside international organisations and marks a
representation practice which is not bound institutionally to the authorisation and
* Henrike Knappe
henrike.knappe@iass‑potsdam.de
1
Institute for Advanced Sustainability Studies e.V. (IASS), Berliner Strasse 130, 14467 Potsdam,
Germany
Vol.:(0123456789)
H. Knappe
accountability mechanisms of electoral politics. This growing space of transnational

representation among civil society actors leaves International Relations (IR) schol-
ars with the question of how to observe and analyse such new forms of transnational
representation. It cannot easily be analysed with standard categories of (electoral)
representation. This paper argues that political representation in transnational civil
society networks needs to be investigated as practice with regard to its flexibility,
relationality and dialogical agency. Analysis of transnational representation from
a practice-theoretical perspective can facilitate a better understanding of the actual
representation practices of civil society actors in a transnational setting. The cen-
tral research question raised is: how do representation practices in transnational
civil society ensue, change or shift within the broader structures in which they are
embedded? By focusing on flexibility, relationality and dialogical agency as the key
theoretical concepts for representation practice, this study delves deeper into those
aspects through empirical analysis of qualitative interviews with activists from two
major transnational civil society networks: the Clean Clothes Campaign and Friends
of the Earth. The study finds that transnational representation in civil society net-
works evolves in a non-linear fashion, is characterised by shifting agency, discursive
claims and a disembodiment of representation.
How actors create and shape the make-up of constituencies, the role of the audi-
ence or the tasks of representatives through the very act of representing becomes
much more discernible in non-electoral representation. Non-electoral representation
thus accentuates the constitutive aspect of representation to a greater degree than
formalised electoral representation. Furthermore, the agency and efficacy of the rep-
resentative, that is his or her ability to ‘render others politically present’ (Montanaro
2018: 7) becomes even more relevant than in electoral representation.
Applying a practice-based constructivist lens to representation this paper recon-
structs how transnational representation unfolds as flexible and relational practice
which relies heavily on forms of dialogical agency. In concrete terms, this means,
first, that representation practices do not develop in a linear fashion. They emerge
from reciprocal or complementary interaction, a ‘back-and-forth’ between repre-
sentatives and those they represent; representation practices are thus experienced as
‘organic’ processes. This also involves the exchange and modification of represen-
tation roles as well as a strong emphasis on trust in managing representation rela-
tionships. Second, instead of claiming to represent ‘real people’, actors in the civil
society networks tend to represent concepts or positions with respect to public dis-
courses. In particular, when it comes to long-distance representation, civil society
actors shy away from representing constituent groups directly, preferring instead to
speak out for a particular stance within a broader debate. Third, the spread of digi-
tal communication technology as a vehicle for representation leads to disembodied
communication. This is problematic for accountability in representation and shows
the need for real-world encounter, namely having material space and face-to-face
contact between actors in representation dialogues.
This paper addresses two strands of literature. First, IR literature has paid only
scarce attention to theory-building of non-formalistic representation in transnational
settings from a constructivist perspective (see Mulieri 2013; Montanaro 2018). Also
the way civil society actors actually practise representation and what this means for
Representation as practice: agency and relationality in…
their operation (Hahn and Holzscheiter 2013; Holzscheiter 2016) is only seldomly
an explicit research focus in IR. Second, representation theorists who have explic-
itly defined representation as performance have only seldomly been able to breach
the muddy waters of non-electoral transnational politics (Näsström 2015; Castigli-
one and Warren 2019; Saward 2010; Montanaro 2018) and instead rather focused
on representation in the formalised politics of nation-states (Mansbridge 1999;
Rehfeld 2009; Severs 2010). For this reason, this study mobilises international prac-
tice theory which defines practices as competent, patterned performances that rest
on background knowledge and weave together ‘the discursive and material worlds’
(Adler and Pouliot 2011: 7‒8). It provides an account of how representation is per-
formed through representative claims (Saward 2010) in transnational civil society
networks. This paper is intended to contribute to and exceed the existing scholarship
operating with fixed, formal categories in order to better assess whether and how
civil society actors succeed or fail as democratic representatives in global govern-
ance (Beisheim and Mariann 2001; Bendell 2006; Scholte 2007; Bexell et al. 2010;
Beauzamy 2010).
The study presented here investigates political representation in two transnational
civil society networks, namely the Clean Clothes Campaign (CCC) and Friends of
the Earth (FoE). Both are typical transnational civil society networks; at the same
time, each covers a different issue area, with its own distinct campaign strategy and
lobbying repertoires of its civil society actors. As a case study of typical cases, this
study aims at generating new hypotheses and theory-building (see Gerring 2007: 67)
in a barely investigated field: representation practice in transnational politics.
In the first part of this paper, I develop the argument that the emphasis of inter-
national practice theory and constructivist representation theory on flexibility, rela-
tionality and agency can facilitate a better understanding of transnational represen-
tation. In the second part, I present empirical findings from interviews with civil
society actors in each of the two networks, CCC and FoE. I reconstruct how these
actors describe their own representation practices within their respective networks.
This in turn gives us insight into the dynamics, interactions and actor positions that
evolve in representation practice in transnational civil society. Finally, I discuss the
empirical findings in the light of practice-theoretical assumptions about relational-
ity, flexibility and dialogical agency, and conclude with remarks on the potential of
practice-theoretical research for transnational representation studies.
Flexibility, relationality and dialogical agency
In recent decades civil society networks and coalitions in particular have become
research objects as visible and powerful actors in the transnational realm (Keck
and Sikkink 1998; Khagram et al. 2002) exercising different forms of influence and
using different mobilisation strategies (Tarrow 2005). IR research on transnational
civil society focuses mainly on the effects of representation, namely on the inclusion
and acceptance of non-governmental organisations (NGOs) in international organi-
sations such as the UN or WTO (Bäckstrand 2006; Esguerra et al. 2017; Jönsson
and Tallberg 2010; Tallberg et al. 2013). The representative role of NGOs is seen as
H. Knappe
the basis for legitimising the participation of civil society in international organisa-
tions. Inclusion of civil society is discussed even more so as a potential for democ-
ratising international politics (Koenig-Archibugi and MacDonald 2013). However,
as Kuyper and Bäckstrand (2016) note, representation is often not linked to matters
of actual accountability, therefore the concrete relationship dynamics between repre-
sentatives and those represented has not been a particular focus of transnational civil
society research. In their role as representatives, mainly in international organisa-
tions or vis-à-vis a transnational public, civil society actors act as spokespersons on
behalf of particular groups—garment workers, for instance—or they claim to stand
for and represent supporters of broader concepts such as animal rights or biodiver-
sity. The characteristic structures of these actors’ respective transnational civil soci-
ety networks ‘blur the clearly defined roles of accountability holders and holdees
in favour of a situation in which each actor is equally an accountability holder and
holdee’ (Esmark 2007: 282); therefore, the precise nature of representation practice
can hardly be captured by standard representation theory. Standard representation
theory coined the notion of representation as a mere mirroring of already exist-
ing, stable interests (Disch 2011). Political representation is commonly understood
as ‘acting in the interest of the represented, in a manner responsive to them’ (Pit-
kin 1967: 209). Conventional views of representation refer mainly to the idea of an
unmediated, authentic political reality (the citizens’ real preferences), which then
provides ‘an origin and point of reference for assessing the accuracy and faithfulness
of any attempt to represent it’ (Disch 2011: 104). Nevertheless, the new forms of
representation in transnational civil society still tend to be analysed and evaluated on
the basis of such standard accounts of representation, which makes them, by way of
comparison, seem to be ill-functioning. The analytical lens of standard representa-
tion theory is mismatched to the empirical object it seeks to investigate.
I argue that this conventional notion of political representation which is borrowed
from the nation-state context of representative government does not do justice to the
transnational context of shifting boundaries, open coalitions and temporal arenas.
Political representation makes constituencies (Montanaro 2018) and mobilises iden-
tities and interests (Disch 2011). Representation, be it in formal or informal contexts,
cannot be understood fully by merely investigating its formal rules and institutions.
The analysis of representation should focus more on the process of representation
than merely on its effects. It is not the transnational reality that needs to be adjusted
to representation theory; rather it is the standard understanding of political represen-
tation that needs to be reconsidered. The reality of transnational relations is that of
plural and fluid (power) relations and processes (MacDonald 2008; MacDonald and
MacDonald 2010).
In order to fruitfully investigate representation practices in transnational civil
society, I build on two different bodies of literature: one encompassing the ongoing
debate in international practice theory (IPT) (Adler and Pouliot 2011; Bueger and
Gadinger 2014) and the other comprising the strand of constructivist representation
theory (Saward 2010; Disch 2011). Both bodies of literature suggest that practices
bring about the world we live in. Social phenomena are no longer seen to be ‘prod-
ucts’ of fixed structures, institutions, norms or power dynamics and interests; instead,
social phenomena are constructed (see Ankersmit 1996) or in the making (Schatzki
et al. 2005). Practice theorists understand the social as ‘a field of embodied, materi-
ally interwoven practices, centrally organised around shared practical understand-
ings’ (ibid.: 3). Analysing representation from a practice-theoretical angle has an
enormous advantage: it allows us to examine political representation in transnational
politics beyond electoral politics and practice as we know them from nation-state
governance and it does so in three important ways. First, practice theory informs
us about the social aspect to flexibility and thus sheds light on the uncertainty and
fragility of human life in the ‘open spaces between routines (performing estab-
lished practices) and crises (creating new practices)’ (Gadinger 2016: 5). Practice
theory also assumes relationality in the socio-political context—that is how trans-
national practices are embedded in multiple structural contexts and how relations
between transnational, national or local practices and actors can be reconstructed
from a practice-theoretical perspective. Finally, practice theory takes agency into
account and thus opens the analysis up to including actors and how they perform.
The ontological status fluctuating between flexibility and stability (Gadinger 2016)
is an overarching assumption of practice theory; the concepts of relationality and
dialogical agency can be seen as contributing to this ontological status. The embed-
dedness of practices in relations means that practices constantly change, depending
on the particular relations between various actors at a given time, in order that these
actors be able better to accommodate or adapt to newly formed relationships. When
actors are credited with real agency in performing certain practices, this means that
those practices rely on individual choices and interactions. It also means that prac-
tices thus generated need to remain flexible.
Flexibility
Practices can be relatively stable patterns and they can have stabilising effects; how-
ever, they are still ‘moving, shifting and changing entities’ (Bueger and Gadinger
2014: 78). This ontological status of practices, fluctuating between stability and flex-
ibility, is mainly stressed by poststructuralists. Butler for example states that the reit-
eration of practices can stabilise norms that are historically contingent and, at the
same time, that such reiterations are open for shifts and subversion (Butler 2004).
Thus, in view of the actors performing such practices, ‘practices stabilize agents, but
never do so completely, leaving identities and the practices that underpin them open
for appropriation and change’ (Braun et al. 2018: 11).
The use of a practice lens in the study of representation speaks to a debate in
representation theory that culminated in the ‘constructivist turn’. Here representa-
tion is seen as performance that develops in the course of different practices, which
is not systematically laid out a priori (Disch 2011). Representation is therefore an
open-ended process. More specifically, this view focuses on the claims-making of
a person or group who wants to represent a certain constituency (Saward 2010).
Saward coined the term ‘representative claims-making’ to describe a practice rather
than a static, institutionalised relationship. He argues that representation starts when
persons claim to be representing someone or something (ibid.). Those construc-
tivist and performative understandings of representation stand in stark contrast to
H. Knappe
the standard assumption of representation as a fixed relationship which is clearly

sequenced through elections. According to standard representation theory, represen-
tation is time-limited through and institutionally bound to regularly held elections.
Once a representative has been elected there are two main representation formats
that standard theory defines for him or her: either the representative may act as a
trustee of his or her respective constituency, carrying out tasks and making deci-
sions with a view to the common good; or the representative may act as a delegate
of his or her respective constituency and try to mirror the interests of that constitu-
ency (Pitkin 1967; Mansbridge 2003; Rehfeld 2011). While trusteeship is captured
as a deliberative and relational account of representation (Dryzek and Niemeyer
2008), delegation is seen as an aggregative and substantive form of representation
(Rehfeld 2011).1 Many have argued that this ideal standard of democratic represen-
tation is nothing more than an illusion, because the representative will necessarily
betray parts of her or his constituency in the act of representation (e.g. Latour 2003)
and that she or he can never be fully responsive to all of the different values and
perspectives within the constituency. Representation therefore will always be partial
and selective (Saward 2009).
Relationality
Practice theory emphasises relationality—that is actors’ behaviour interwoven with

relevant material and immaterial entities including physical bodies, space, time or
affects (Solomon and Steele 2017), as well as the relationship between practices:
‘the world is made up of different “constellations of practices”, that is assemblages
of communities and their practices that interact, overlap and evolve. […] It is the
permanent state of connectivity and tension inside a constellation of practices that
fuels transformation’ (Adler and Pouliot 2011: 20). Emirbayer (1997) points out the
primacy of contextuality and relations in analysing the social world by citing the
analogies of the game and the conversation. A game cannot be understood by merely
analysing the individual players as isolated units, nor can a conversation be under-
stood by analysing the individual interlocutors. To understand a game or a conversa-
tion, one must take into account the dynamics and processes of those interactions
(Emirbayer 1997: 289‒290). Relationality is also tightly linked to a processual
ontology: ‘Everything is potentially moving, all is in the making. In other words,
continuity is not to be confused with stability, but with ongoing processes that (re)
produce that stability’ (Guzzini 2017: 371). This processual ontology is also coined
in the central practice theoretical term of ‘performativity, that is the claim that phe-
nomena such as power, facts, political orders, nation states or regimes are effects of
relations and practices’ (Bueger and Christian 2017: 330).
1
Aggregative accounts of representation clearly adhere to the realist assumption of an authentic and
static political constituency with fixed interests and preferences, but the trusteeship model leaves room
for a more dynamic way of doing representation through deliberation and persuasion. The trusteeship
model does, however, still hold to the idea of unity of the constituency and a coherent and unified com-
mon good that can be achieved in the process of representation.
The idea of relationality is taken up in representation theory by scholars who

problematise the aggregative notion of standard representation theory and instead
propose a deliberative form of representation. Discursive representation2 empha-
sises the process of constructing discourses while representing, instead of assum-
ing, an institutional schematic relationship. The focus is here on the articulation, the
making visible and the very constitution of multiple ideas, values and preferences
through discursive representation (Dryzek and Niemeyer 2008: 8‒10). Similarly,
in their study on participatory practices, Chilvers and Kearnes (2015: 16) empha-
sise the ‘multiple, diverse, entangled and interrelating collectives of public involve-
ment within particular political constitutions, systems or issue spaces’. From this it
follows that the practice dynamics ‘can never be fully understood in isolation, but
should always be read in terms of their relational interdependencies and connected-
ness’ (ibid.).
Applying a practice lens to representation therefore means that transnational rep-
resentation cannot be analysed by neatly cutting it into single separate units such as
different roles for actors (representative, constituency) or different sequences (cam-
paigning, election, holding office, re-election), which remain stable and discrete
units of analysis.
Dialogical agency
Practice theory ‘reconsiders agency in a more substantial manner […] [and] submits
that actors’ dispositions are replaced by actors who are “active, not passive”’ (Bol-
tanski 2011: 26, cited in Gadinger 2016: 5). Foucauldian-inspired practice theorists
argue that ‘[a]gents’ properties, such as preferences and interests but also beliefs,
are not to be assumed before the analysis, but treated as emerging in the process.
And if properties are emerging, the process needs to be included in the analysis, i.e.
endogenised, and not simply taken for granted’ (Guzzini 2017: 371). These emerg-
ing preferences and interests are not to be seen as straightforward and determined
processes, but rather as an ‘unsystematic interplay of discourses that converge as
well as conflict with one another’ (Brown and Scott cited in Guzzini 2017: 371).
In applying the practice-theoretical concept of agency, I want to further
emphasise its dialogical character, that is, the observations of representation
necessitate taking into full account the dialogue between at least two agents—the
representative and the represented. This is not the standard assumption of repre-
sentation theory; the standard intuition has been that only representatives ought
to be responsive to their electorates. As Disch (2011: 100) notes this inclination
creates the image of a unidirectional relationship. However, as recent theorising
and empirical research indicates, representation practices are not only about the
agency of representatives, but also about the object of representation, the constit-
uency represented (Saward 2010; Holzscheiter 2016). This makes representation
2
Discursive representation operates under the assumption that people have multiple interests and ideas
which change dynamically. Such ideas and interests are then subsumed in certain discourses (Dryzek and
Niemeyer 2008: 8‒10).
H. Knappe
a political process, in which representations are ‘act[s] of power’ (Berger and

Esguerra 2018: 219). Representation is not an unambiguous mirroring of pre-
existing interests, but ‘[s]peaking for others involves choosing between alter-
natives, not only between good and bad alternatives but rather between various
interpretations that could be represented otherwise’ (ibid.).
For example, in her study on transnational advocacy in the field of child labour,
Holzscheiter (2016: 223‒225) shows how representative claims-making can be
understood as an antagonistic practice and not just a ‘one-way street’. She points
out that former constituents represented in the child-labour politics and policy
arena—the working children themselves—have voiced their demand to represent
themselves and contested the dominant representation claims heretofore made on
their behalf:
They have […] voiced their resistance to the discursive construction of their
lives and identities in the performative practices of those advocacy organisa-
tions involved in the formulation of international child labour standards. They
have rejected the vocabularies of victimhood, vulnerability, and distress that
are omnipresent in global civil society activism on behalf of working children,
claiming their right to political citizenship, that is, to participate in communi-
cative spaces where they are given the opportunity to express their own opin-
ions and perceptions. (Holzscheiter 2016: 223)
Holzscheiter’s study highlights the contested nature of representative claims. A
power struggle evolved in this case over representation claims in an (asymmetri-
cal) interaction between disproportionately powerful civil society organisations,
intergovernmental bodies and relatively powerless working children’s organisations.
The Holzscheiter study shows how representation practices develop from interac-
tions or dialogue between representatives and constituencies (ibid.). This back-and-
forth between different actors results in different kinds of interpretations and role
ascriptions. When actors claim to represent a certain constituency, they construct the
constituency through the act of representing. At the same time, constituencies can
actively participate in representation practices: they can be forceful respondents and
discussants of policies, demonstrators for or against particular policies or ‘counter-
representatives’ for alternative visions. This active performance and the construction
of constituencies through the act of representing is somewhat under-conceptualised
in standard accounts of political representation (Peruzotti 2010: 156); for example,
constituencies are reduced to passive groups like an electorate represented by mem-
bers of parliament. The same limitation applies to the role of representatives in con-
ventional theory: representatives are confined to acting on behalf or mirroring the
interests of their constituency. However recent debates in representation theory have
highlighted the multiplicity of types of political representation and have empha-
sised its creative potential (Mansbridge 2003; Severs 2010; Rehfeld 2011). Creativ-
ity here refers to the different ways in which the roles of representatives and those
represented can be filled, in which new roles can be established or where dominant,
conventional frames can be circumvented. Constructivist representation theory and
practice theory both shed light on the role of actors and their agency—their abil-
ity to shift, change and invent representation practices. Actors can modify practices
(Butler 2004) or re-interpret and get around them in positioning themselves vis-à-vis
other actors.
Furthermore, this new way to conceptualise ontologically and analyse empirically
representation as practice can shed light on the agency—and identity-constituting
effects of representation (Braun et al. 2018) from the standpoint of those represent-
ing and those being represented (Saward 2010; Disch 2015). For example, group
identity can be stabilised through representation practices (Castiglione and Warren
2019) or by mobilising new representation roles. Particularly on the transnational
level representative efficacy depends on actors’ ability to create and mould or adapt
representation (see Disch 2016: 793).
Transnational civil society’s representation practices
What do we see when we examine representation practices? The Clean Clothes

Campaign and Friends of the Earth are typical examples of transnational civil soci-
ety networks; they reflect the diversity of transnational civil society particularly as
regards overall goals, internal relations, more narrowly targeted aims, strategy and
capacity. Representation occurs over great geographical distance among various
communities and actors, often mediated through digital communication technology.
Civil society actors face the unique challenge of transnational representation. They
must cross geographical boundaries and overcome language barriers—that is they
must translate messages and interpret issues from a plethora of languages across
great territorial expanses. Transnational civil society organisations are not formally
elected by their constituents and consequently do not operate within a formalised
context of legislative periods, demarcated electoral districts or defined mandates.
The Clean Clothes Campaign (CCC) was founded in Amsterdam (Netherlands)
as the ‘Schone Kleren Campagne’ in 1989. The campaign is ‘one of Europe’s most
influential multi-stakeholder initiatives for pressuring companies to assume respon-
sibilities for workers’ rights at their suppliers’ factories’ (Egels-Zandén 2011: 259).
The CCC is highly institutionalised and campaigns on a continual basis. The organi-
sation consists of seventeen national platforms in sixteen countries. The CCC began
as a protest movement against clothing retailer, C&A, in the Netherlands; today it
concentrates its efforts on ‘improving working conditions in the global garment
industry’.3
Friends of the Earth (FoE) is an international grassroots environmental network,
founded in 1971 by organisations from France, Sweden, England, and the United
States. FoE belongs to the three biggest environmental non-governmental organi-
sations, but in contrast to the other two, Greenpeace and the World Wildlife Fund
(WWF), FoE addresses environmental issues in reference to social and political
inequality, and voices explicit critique of neoliberalism on a broader ideological
agenda than does either Greenpeace or the WWF (Doherty 2006: 862). Seventy-six
3
See https://cleanclothes.org/about/mission (last accessed on 24 October, 2019).
H. Knappe
member organisations overall and two million members4 are coordinated in a federal
structure.
The empirical insights presented in this part of the paper are based on a recon-
structive analysis of interviews conducted with 26 members, activists and coordi-
nators from both transnational civil society networks.5 Through the interviews and
interview transcripts, I have reconstructed the stances of actors towards political
representation practices.6 From the interviewees’ narrations and interpretations of
coordination and communication practices in the network, I was able to reconstruct
an evolution of practices, the roles of representation and the actors’ access to repre-
sentation.7 For the interview analysis I adopt the concept of discursive practices in
order to ‘refer to both the practice component of discourses and to the net of mean-
ings necessary to see practices as intersubjectively shared’ (Guzzini 2017: 371). In
order to replicate the nets of meanings that surround and entrench practices, I rely
on linguistic expressions as a main source for the data analysis. The inseparability of
practice from language is a core methodological argument for applying reconstruc-
tive analysis to interview material. The choice of interview data and the reconstruc-
tive method of analysis is further justified by the specific characteristics of transna-
tional civil society networks. In qualitative interviews the subjects discuss practices
and in so doing they discursively construct meaning for what they think about how
such practices occur and how they ought to be classified and judged. Therefore
assuming that people can talk about practices (Hitchings 2012), qualitative inter-
views are valuable tools for reconstructing the justifications for and detailing the
long-term processes of certain decision-making practices or strategies that are not
actually observable.8
The representation practices that I have reconstructed for both networks can be
grouped into four categories: (1) representation as an organic process, (2) trust and
shifting agency, (3) representative claims at a distance and (4) disembodiment in
representation. I will present the findings of the empirical analysis in the following
sections before I go on to discuss the relevance of flexibility, relationality and dia-
logical agency in the empirical findings.
4
See http://www.foei.org/en/who-we-are (last accessed on 24 October, 2019).
5
Interviews are numbered consecutively—for the FoE: F1, F2 …; for the CCC: C1, C2 …
6
This interview analysis was part of my dissertation. The empirical data and parts of the analysis are
taken from that work (see Knappe 2017). The interviews were conducted in 2012 and 2013.
7
Because the interviews were transcribed with intonations and accentuations, excerpts from them read
differently compared to standard English-language quotations. The quoted material here tries to repli-
cate spoken language to the degree that it is still readable and understandable. Accentuations are marked
using upper-case letters; all other non-accentuated words, including nouns and pronouns such as ‘I’, are
written exclusively in lower case.
8
Interview data are of course limited in analysing the very micro-level material or bodily practices of
everyday interaction in meetings, discussions, etc. However, since this paper is interested in the broader
ensembles of practice in transnational civil society networks, there is some justification for sticking to
this data.
Representation as an organic process
Transnational meetings of the CCC and FoE are central sites of representation.
Local member organisations of both networks gather at such meetings and discuss
campaign progress, goals and strategies. From meeting to meeting representation
practices add up to a set of activities, routines and norms. If new persons, groups
or even methods enter into the practice, then preparation routines or accountabil-
ity mechanisms are newly interpreted and challenged. Representation practices
therefore evolve in a dynamic back-and-forth exchange between different actors,
developing non-linearly and organically. These practices are mainly constructed
through the positional agency and knowledge of actors and their interpersonal
relations. Organically evolving representation practices therefore enable a diverse
set of actors to take on various roles. As one interviewee noted (F4), giving rep-
resentatives a mandate or holding them accountable is an ‘organic’ process that
develops over time but not in a pre-determined manner. The organic-process
metaphor indicates a non-linear, relational way of doing representation. Bits and
pieces of representation practice come together and find an adaptive fit; they may
relate to one another but do not necessarily point in the same direction. They can
change their form and their function over time and in specific situations and con-
texts. However this way of doing representation can also reinforce unequal power
dynamics, when representation practices become arbitrary and non-transparent.
Representatives in both networks under investigation experienced representa-
tion as a multi-layered dialogue necessitating much discussion at international as
well as national and local level. One interviewee described the role of representa-
tives as being like a ‘hinge’ between the national platform and the international
campaign network (C9).
Representatives position themselves as mediators between the two levels and
often seek active engagement by their constituency ‘at home’, as one interviewee
described a typical decision-making situation.
maybe two-thirds of the steering committee who thought it was not a good
idea so they dropped [it] – i dropped the idea fine. i mean i need the backup
… i will try to if i’m really convinced[;] i will try to convince them but if
they disagree i will drop the idea. (C4: 77)
At least for contested issues the different views and proposals need to be
broadly and openly discussed in order that representatives actually have a man-
date to decide on behalf of the constituency—in this case the national group of
volunteers.
i think it is about … DISCUSS[ING] all things like that [–] this is not some-
thing that’s happening often … but if it DOES happen it is important to have a
meeting WITHIN the organization where you discuss PROs and CONs and if
it is something we [stand] beHIND or NOT[;] and if it is NOT we then just go
back and say [‘]sorry this is not something that we can WORK on[‘]. so it is
VERY important to have this implemented within the organizations since a lot
of the work is done by volunTEERS. (F9: 41)
H. Knappe
However, the dialogue between representative and represented can look very dif-
ferent from the perspective of a member of the local constituency; it can create an
impression of opacity.
i think the decision making processes in other organizations are always big
mysteries for their colleagues and then sometimes EVEN for the organizations
themselves because it’s always a big struggle … because organizations are part
of international networks … and often it’s the director that goes there and then
there is a big discussion and then there is a result and there were so many steps
involved. (F1: 98)
So dialogues between representatives and constituencies can be interpreted differ-
ently by representatives and constituency members. Perceptions range from see-
ing these dialogues as being highly interactive to their being very opaque. But the
implicit assumption is that constituencies actively participate in representation prac-
tices and therefore need to be kept informed and on track about what is going on
when their representatives go to international meetings.
Trust and shifting agency
The dialogical relationship between representatives and represented in both of the

studied transnational civil society networks is based on trust and knowledge. In one
interviewee’s words: ‘I know in the spirit of whom I have to act’ (F4: 54). There is
a sense of caring involved on the part of representatives who are sent to the higher-
level transnational meetings, replicated in the stances they take. Especially when
representatives have been in charge for a longer time, they will have established a
greater rapport with their respective constituencies—that is to other local organisa-
tions and to individual members of the national organisation. Representatives who
enjoy such rapport with their constituencies generally consider themselves knowl-
edgeable and feel confident to represent these constituencies accordingly, without
further need of extensive and in their view unnecessary interaction.
i have been the coordinator for sixteen […] or have worked with ccc for six-
teen years. i know what issues are delicate. so i know when i have to get back
to my platform [how] to be able to express our position at the euro-meetings.
(C4: 70)
When asked on which grounds an international coordinator is given a mandate
to represent other groups, one interviewee responded thus: ‘good will [and] good
understanding of how the organisation operates’ (F5: 72). Thus the representative’s
knowledge plays a significant role in this kind of trusteeship-representation practice.
Some even see the trust others have in them as a great opportunity to keep the work
of the organisation going.
i can say [–] have the chance [–] that they [the national organisations in
the platform] really trust me for the international level because i think [–]
i mean [–] there are different kind[s] of involvement for the national plat-
forms in the international network[:] some are really involved …, [but] in
[own country] it’s not really the case[.] so it’s not that people are … [not]
interested or do not feel it’s important, but they feel the work i do as a
national coordinator participating in the international network is sufficient
or is enough and that we have to deal with other [–] with a LOT of other [–]
topics at the national level. (C3: 77)
This rapport between representatives and fellow members of their own organisa-
tion enables them to do the job of representing more effectively. Placing one’s
trust in the person who is sent to transnational meetings is necessary if one wants
to avoid time-consuming control mechanisms (see Castiglione and Warren 2019).
Nevertheless the failure of more experienced and trusted representatives to
mobilise their constituencies was also problematised by interviewees.
there is not TOO much interest in really preparing those meetings and the
person going there knows generally what’s important for [own organisation]
so there is no NEED to prepare it in a better way[;] but … i think it would
be good to give more attention [to the constituency] so that people feel more
connected to the process. (F1: 45)
‘Feeling more connected to the process’ was a desire often voiced in both net-
works. Trusted relationships and time constraints can limit dialogue and mobi-
lisation between representatives and constituencies. At an extreme the caring
and trusting aspects of representation can result in the exclusion and silencing
of constituent groups (C3) and in so doing constructs a constituency of passive
receivers of information. However, as one representative noted, the constant dia-
logue and struggle with disagreement is deeply inherent in their interpretation of
representation.
because i feel it’s my role both to deal with agreement and disagreement.
sometimes i even know that beforehand, that some organisation or some
person in [the] organisation will disagree[;] so i will look for that disagree-
ment just to make sure that they have done a proper consultation. (C4: 73)
As representatives position themselves, they reflect on the disagreements they
have had in the past and try to foresee potential disagreements they might be fac-
ing in the future.
One must see this kind of rapport between representatives and members of
their constituencies—characterised by caring and knowledgeability—as the
result of a longer process of learning. Newer members of both networks in my
study tended to pursue a more rigorous strategy of preparing representatives for
international meetings and having them report back after such meetings to the
local and national membership, the constituency. These newer network members
describe dialogues between representatives and constituencies as an ongoing
learning process in which representatives and constituencies acquire informa-
tion and knowledge about the international meetings—the sites of representation
activity—but also about one another and the evolving interests and ideas within
their own respective organisations (F7: 24).
H. Knappe
i have also learned a lot because at the beginning i was really inexperienced
but now i can really work with the projects and everything so it’s really good
for me[;] i understood the logic of EC projects. (C8: 75)
However, these learning processes are sometimes evaluated critically as struggling
to adapt.
sometimes like i said before we have to adopt different ways of working here
and there is no space for such [a] big discussion in [the] euro[-level] meeting
(…) so that’s why it’s so difficult[;] i think it takes us more time to follow the
processes in a way that it [all] makes sense for us. (C7: 12)
Here we can observe how organically grown representation practices are difficult
for newcomers to understand. Whereas older members of the network witnessed the
back-and-forth of developing representation practices, newcomers struggled with
hidden rules and insider jargon.
Political representation in both transnational civil society networks in this study
is also characterised by shifting peer-to-peer relationships which constantly undergo
negotiation between actors. Since networks usually do not have very strict, formal
hierarchical structures, representation more often takes place between network
member organisations. The practice of representing other peer organisations at
formal meetings is quite common. The use of proxy voting in the FoE is just one
example of this kind of representation practice. If one national organisation can-
not attend an annual general meeting, then it can ask another organisation to vote
in its stead, as its proxy. This kind of horizontal representation must be formally
expressed through a written declaration of authorisation. It is often practised among
organisations which are similar or share similar languages like the Scandinavian or
Central-Eastern European organisations (F4: 48). Thus likeness or regional kinship
is a main factor in horizontal peer-to-peer representation. Beyond a formal declara-
tion the mandate for representation often remains vague. Some of the represented
peer organisations explain their mandate for the representing organisations based on
the former’s impression that the latter will decide in the best interests of those it
represents, because both organisations are so similar in their worldviews and share
the same aims. Actors assume that shared languages or other characteristics eases
understanding between the representing organisation and the one represented (C11).
This horizontal representation among peers is by definition mutual and interactive.
The roles of representative and represented are interchangeable; they can be modi-
fied and renegotiated from meeting to meeting. At the same time, peer representa-
tion is a variation of descriptive representation: peer organisations are positioned as
standing for another similar organisation.
Another kind of practice in peer-to-peer representation gives us a slightly dif-
ferent perspective on it. For example, some of the bigger and wealthier organisa-
tions in a network may give their ‘seat’ in transnational meetings to smaller, not-so-
well-resourced organisations and fund the latter’s travel expenses (F11). Although
both organisations are peers, they differ in their capacities to attend such meetings.
Thus peer representation in this case enables under-resourced and potentially mar-
ginalised groups the access to representation positions. In this case the meaning of
representation has shifted. It is strongly connected to the notions of inclusion and

participatory equality among members in the network, such that some of the mem-
ber organisations may feel the responsibility to care for others by providing them
with needed resources (like travel funds). This form of inclusion is not formalised
in any universally applicable rules. Such opportunities for representation are given
on a personal and relational basis. Informal practice like this can enable better repre-
sentation and access to levels in the network hierarchy that may otherwise be out of
reach for some members. It can also lend formerly passive constituents empowered
agency but, at the same time, it can also reinforce already existing inequalities and
foster unwanted dependencies.
Representative claims at a distance
The voicing of representative claims towards sometimes distant constituencies

shows how practices are embedded in multiple relationships between different actors
and discourses. It also reveals how practices can shape contexts and shift in relation-
ship to them and to specific assumptions of the actors involved. In the transnational
civil society networks investigated here, for someone or some group to claim to rep-
resent some other particular and distant constituency was something done only with
a measure of caution or reticence. Nobody in the CCC, for example, made a general
claim to represent garment workers in Asia.
i do not really know how to formulate this, because i think it would be too
BOLD to say that we are actually representing the workers? but they are the
ones that it in the end it’s all about. (C10: 39)
While this was expressed by a Western European organisation, similar statements
were also made by NGOs in the Global South which were and are in close contact to
the workers.
i will think that as campaigner we … only play this supportive role and the
garment workers[,] they have to stand up to the fight for their own rights.
(C12: 26)
From the two interview excerpts above, C10 and C12, we can surmise that the role
of the representative is limited to that of an assistant who supports the workers in
their fight. Both of these CCC interviewees rejected the idea of taking an active rep-
resentation role; instead agency was ascribed to the workers themselves. Here we
can observe a complex constellation of representation practices. Whereas the local
NGOs claimed to support workers’ demands vis-à-vis the international-level or
European-level network organisations; the European-level organisations, unlike the
local NGOs, claimed to be the amplifier of workers’ demands vis-à-vis companies
and national governments.
When asked whom a local NGO represents, an NGO activist referred to the work-
ers’ rights and to corporate social responsibility (CSR), not to the workers them-
selves. Representatives claim to represent certain normative concepts and refer to
H. Knappe
broader discourses rather than a particular constituency per se: ‘we represent the
workers’ rights and CSR interested to call corporates accountable and to uphold
workers’ rights’ (C14: 55‒56).
Similarly the only bold and emotional representative claims by the FoE were
made with regard to a constituency only vaguely described as ‘the environment’;
the Western European activist in question claimed to ‘give effect’ to the ‘voice of
the environment’9 (F10: 2). Even here the claims-maker assumed the passive role of
someone who just carries out an already defined task, as if ‘the voice of the environ-
ment’ were clearly speaking and as if that claims-maker were merely carrying out
instructions or fulfilling demands.
So many representative claims are cautious, vague and abstract. CCC-activists in
Western Europe and Asia emphasise the autonomy of garment workers and the mere
instrumental role of the representative serving the constituency. If direct claims are
made, then this is done in relation to norms such as workers’ rights, standards for
socially responsible entrepreneurship or broader concepts like the environment.
Cautious representation practices shift the addressees of representation from con-
crete persons to concepts or discourses. This was interpreted by the interviewees as
a highly sensitive response resulting from heightened awareness of unjustified and
illegitimate claims to representation. On the other hand, this shift away from con-
crete individuals to representing abstract ideas or discursive positions also encour-
aged conflict avoidance and pre-emptive exclusion of critical voices from among
local activists. In any case a re-interpretation of representation as a claim to stand for
broader normative concepts points to forms of discursive representation that make
concepts and discourses their main reference points rather than peoples’ voices.
Disembodiment in representation
Embodiment, the intertwined character of activity and physical presence in practice

(Schatzki et al. 2005: 2) and the lack thereof, disembodiment, can be experienced
intuitively by everyone who has ever had a skype meeting with more than three
persons.
there is a difference. …when you have a skype meeting …sometime[s] it’s not
easy to express exactly what you have to let people know about[.] but physi-
cally … i do not know[,] but it’s really GOOD to have people physically [pre-
sent], to talk to people to meet people[.] we are used to that and we really like
that. (F12: 38)
The relationality of practices becomes apparent when tangible relationships cannot
be experienced by actors in the same way they were used to in the past. Because
of the necessity for online communication between network members dispersed in
different countries and across continents, representation practices become disen-
gaged from face-to-face communication and personal contact. This disembodied
9
Author’s translation.
communication is understood as communication mediated through technical

devices—computers or telephones—in which physical presence is excluded from
talk. Disembodied communication has the advantage of gaining a broader scope and
reaching broader circles of persons and organisations. Such online communication
as a daily practice in transnational networks can work well in terms of providing and
distributing information, keeping in touch and updating involved persons on the cur-
rent situation of a campaign.
in fact i report in between euro-meetings and in between steering meetings by
e-mail … [with] let’s say these ten people [from] my network. they are very
busy so if i send an e-mail i do not disturb them during their work and they
can read [it] if they like[.] and if they don’t read it, [well] it’s a pity but after a
while i can [remind] them, [and then] they start reading like five e-mails one
after the other and okay they catch up with it. so the e-mail is like a sort of
NICE way to stay in touch. if i REALLY need input i will call them. (C4: 73)
Nevertheless the expansion and bridging of very long distances can diminish the
binding nature of representation in actual practice. In transnational digitally medi-
ated representation, bodies, or rather the absence of bodies, affects representation
responsibility.
we also have skype talks which is …i think … [a] major feature of our work,
and personally i find it a bit even … exaggerated[–] the amount of time that we
spend on [these] discussions [–] because they tend to be quite slow especially
if you have a group of people and then people might not be always focused on
the call if you are just sitting on your earphones for one and half an hour[s].
(F3: 14)
Furthermore, decision making on skype is seen to be possible only if the partici-
pants have met at least once in person (C11). The interviewee from the CCC who
brought up this point argued that regular decision making usually ends with non-
verbal signs of agreement or disagreement.
when you’re with a group of colleagues and you have to make a decision, you
always look around [at the faces of] people … [–] whether they oppose or
whether they consent [to] what is going on [–] and you can see much more
than they actually express. (C11: 25)
Because the conference-call participants cannot see each other on skype, such non-
verbal communication is not possible. Consequently skype talks are very time-con-
suming if all the participants express their opinions verbally; participants must to
some extent assume how one or the other person is likely to decide an issue, based
on how well they already know each other. Not only does digitally mediated com-
munication and representation affect how positions are expressed and how focused
the conference participants are, it can also frustrate individuals and prevent them
from participating in representation practices.
chaos because conference call[s] with five [or] six people on phone [–] you know
each other by person but on phone you do not know [who’s who], you do not
H. Knappe
see the gesture[s] of the people … someone is speaking maybe … too long, you
want to say SOMEthing then it’s right to interrupt but then the communication is
slightly postponed [there is a slight delay in the transmission] … but in the very
end it works. (F6: 30)
Furthermore, disruptions in communication can occur especially when, for example,
skype talks become inaccessible for persons in some of the countries of the global
South because of technological problems making it a struggle for them to participate in
this type of internet conference.
in africa internet is not good, the bandwidth of internet is not good …. so most of
the time it will be difficult to have a voice call[.] you will just type it and then you
wait for the reply and you type another time … it’s not really usual to have inter-
net with voice with discussions like we are DOing … because in [home country]
i have a café[, a] cyber café, where i can have such discussions [with] live voice
and [messaging –] you can type the message[. in the café] i can have VOICE and
discussion, but IN the office it’s not usual[,] it’s not common to have people on
skype with voice…. and in togo, the situation is WORSE there. so most of the
time they can just call them through the telephones and they can [get] through
[with the] telephone eas[il]y, just to have an information[.] yeah they can try to
call them like that but most of the time for MEETing for discussions[,] for long
discussions, we use skype, conference calls. (F12: 49)
This representative (F12) from an African FoE member organisation points out clearly
how difficult it is to have those skype meetings, which have in the meantime become
the main way to communicate within the network. In order to handle the situation,
adaption practices have become necessary. Because the internet is too slow for voice
call, some participants type their messages in skype at the same time that others are
talking on skype. For those participants who communicate over the keyboard, typing
in messages prevents them from hearing what is said during discussion, leaving them
in a passive and disconnected position during the discussion. A similar experience is
described by an interviewee from Latin America who concludes that it was impossible
for them to communicate via skype. Their campaign group which consisted of many
organisations in Asia, Latin America, Africa and Europe took the decision to commu-
nicate via e-mail instead of skype (F13: 70). Even though e-mail communication was
perhaps a less satisfactory choice because it was seen to hamper the campaign’s coordi-
nation process, at the same time it did not produce inequalities between campaigners in
different countries.
In sum, then, disembodied representation practices play a crucial role in long-dis-
tance transnational civil society network communication. However the lack of physical
presence and the spatial divide between representatives and their constituents creates
communication gaps and unclear communication situations.
Flexibility, relationality and dialogical agency in empirical practice
In the following, I discuss the results of my study in light of the practice-theoretical

assumptions of flexibility, relationality and dialogical agency, putting the empirical
findings in dialogue with the broader theoretical frame developed in the first part of
this paper. The empirical insights as exemplary cases will thus help us to contribute
to theory-building in the field of transnational representation. They let us think fur-
ther about the workings, characteristics and dynamics of transnational representa-
tion practices.
Representation practices in both networks evolve in non-linear, organic ways
(F4). Rather than pursuing clearly sequenced, linear forms of representation,
actors in the network go back and forth between the international level and local or
national groups, accumulating knowledge and gaining trust in this process of learn-
ing with and from one another. Representation in both networks needs to remain
flexible and thus seems to rely more on deliberation and constant negotiation than
on counting votes or a clearly timed electoral process. While standard representa-
tion theory usually assumes authorisation via election of a representative before the
actual representing occurs, Montanaro (2018) argues however that there is a form of
non-electoral authorisation which is granted by a constituency after a claim has been
made by a group or an individual to represent or advocate on behalf of that constitu-
ency (ibid.). The constant process of negotiation and discussion on representation
practice within the CCC and the FoE can be seen as such a form of processual non-
electoral authorisation. This negotiation about representation can also evolve contro-
versially, as we could see from the interview material. These practices are embedded
in relations that can create tensions and conflict (see Guzzini 2017: 371), for exam-
ple over the way representatives report back to their constituency. Furthermore, and
most crucially, they can create new, or reinforce existing inequalities, as we could
observe in the case of newer members in the networks and the proxy-representation
practices for resource poor organisations. Representatives and those they represent
need to constantly negotiate and renegotiate their roles and representation practice.
In a positive light this results in openness to new forms of representation and pos-
sible adaptation, but in a more negative one it also results in arbitrary practices that
are not easily accessible for newer or resource-poor organisations. Transnational
representation practices in both investigated civil society networks were character-
ised accordingly by constant tension.
The relationality aspect of representation becomes most apparent when network
members are physically separated by large geographical expanses so that represen-
tation must be practised long-distance. As we saw in the examples of disembodied
communication via skype and e-mail, actors may struggle with loosened relation-
ships and distant, disembodied communications partners, which can even go so far
as actors’ declining to take part in representation practices. From this we can see
clearly that representation practices are relational and embedded in specific contexts
and that these practices need to be continually actualised and frequently repeated.
The interpretation of disembodied communication via skype and e-mail as prob-
lematic also highlights the significance of materiality in representation practices.
H. Knappe
Communication cannot be just transferred to another space, cybernetic space: actors

feel the need for the physical presence of others as well as the physical spaces of
encounter. Body language needs to be seen, experienced and interpreted in order for
those communicating to effectively make decisions. In practice-theoretical accounts
the argument that space as well as materiality and physical bodies be included in the
analysis of social phenomena is an important one (see Solomon and Steele 2017).
Knowledge—for example, about one’s conversation partner or about the matter
awaiting disposition—is then incorporated and it materialises in practice (Reckwitz
2003: 290‒291).
Another important finding is the discursive claims-making. CCC actors prefer to
refer to workers’ rights rather than to workers themselves, and FoE actors prefer to
refer to the environment rather than to the people being affected by environmental
destruction. This changes representation practice from a standard form of aggrega-
tive representation to a form of discursive representation in which representatives
stand and act for discourses and represent a fluid group of supporters and subscrib-
ers to those discourses rather than a fixed group of individuals with specific or
vested interests (Dryzek and Niemeyer 2008). When CCC activists refer to workers’
rights and corporate social responsibility, they connect sub-sets of different ideas
into newly assembled discourses which in turn they present to the public in order
to engage with subscribers of these discourses. However, practising representation
in this way can also make the real people on whose behalf others claim to advocate
invisible, thereby rendering them unable to contest those representative claims. But
the visibility of constituents is crucial for the legitimacy of representative claims.
This was indicative as well for how dialogical agency in representation shifts con-
stantly. Agency can be acquired by actors but it can also be denied them. Actors may
negotiate over perceived and ascribed agency, they may criticise a lack of agency or
they may empower others in order to increase the latter’s agency.
Conclusions
This study investigated political representation in two transnational civil society net-
works, the CCC and the FoE, as a practice carried out in a system of various actors,
differing and changing situations and diverse backgrounds. Transnational civil
society actors remain within the confines of established practice and at the same
time actors modify representation practice whenever they position themselves with
regard to particular aspects of representing or whenever they attempt to circumvent
or re-interpret established practices (see Butler 2004). Such dynamic practices break
with conventional theoretical accounts of representation, which prescribe static rela-
tions between representatives and constituencies and a clearly timed representation
sequence.
I have argued in this paper that political representation in the transnational sphere
should be analysed from a practice-theoretical and constructivist angle in order
to open up the potential repertoire of representation practices and to do justice to
the flexible and dynamic character of transnational representation. I proposed that
representation practices in transnational civil society networks be approached from

the perspectives of flexibility, relationality and dialogical agency.
The particular role of flexibility in transnational representation is spelled out as
non-linearity and organically evolving representation practices. This supports the
argument that representation practices are pieced together in the making instead of
following an externally given plan. Relationality also becomes crucial in the way
actors practice representation via technical devices over large distances. The physi-
cal presence of actors in decision-making processes is decisive for the quality of
representative relations. Similarly, cautious claims to represent distant constituen-
cies often diffuse into claims about abstract concepts and discourses; this shows the
embeddedness of representation practices in broader societal discourses (see Dry-
zek and Niemeyer 2008). Dialogical agency in representation was reconstructed as a
means to ascribe particular roles to actors, as a way to shift roles among actors and
as a means to empower them. The observation that many representation practices
are matters of negotiation shows how crucial the agency of representatives as well
as that of the represented is in developing efficacious representation practices. This
flexibility points to an equalising function of representation, whereby representation
roles are relatively open to being filled by a number of different individuals (Saward
2014). Political representation in the transnational civil society networks I investi-
gated is characterised by attempts to create egalitarian horizontal forms of repre-
sentation as well as trusting and caring relationships between representatives and
constituencies. Face-to-face meetings and close relationships further trusting envi-
ronments, promote effective discussions and enhance the empowerment of diverse
actors. However, geographical expanse and digital communication styles can cre-
ate distance between and a lack of accountability among representatives and repre-
sented. Furthermore, the exclusion of newer—especially resource-poor—members
of a network or the stabilisation of dependencies is also a common observation in
both networks. The tensions that grow out of proximity to or distance from one’s
constituents are often matters of struggle and negotiation among actors in the course
of practising representation.
What all of this shows is that practice-theoretical reconstruction of political
representation in transnational civil society networks enriches and expands our
account of political representation by providing us with new insights on repre-
sentation dynamics. Applying a practice-theoretical lens to representation shows
us in greater detail how representation actually works. Representation practices
enable scholars of representation theory to better understand how representation
unfolds in the space between formal electoral procedures and official appointment
of representatives. This broadens the scope of representation beyond the narrower
conventional perspective which sees it as occurring only in light of formal pro-
cedures or specific institutional arrangements. Practice-theoretical analysis does
greater justice to events transpiring in the transnational sphere, where institution-
alisation is not very extensive. At the same time, it should also complement the
analytical lens used in more formal, conventional settings where representation
is similarly shaped by its actors and undergoes shifts in actual practice. This case
study contributed to theory-building of transnational representation practice. Its
empirical insights, namely the non-linearity of representation, the shifting agency
H. Knappe
within representation practice, the discursive claims and the disembodiment in

transnational representation should be tested by future empirical research.
Acknowledgements Earlier versions of this paper were presented at the 2017 ECPR panel on ‘Representa-
tion Studies Beyond the Constructivist Turn’ and the 2017 EISA Panel on ‘Translating International Prac-
tices’—I want to thank the organisers and participants of these panels and other smaller fora, in particular,
Henrik Enroth, Alejandro Esguerra, Frank Gadinger, Anna Holzscheiter, Frank Nullmeier, Vincent Pouliot
and Daniel Schade as well as two anonymous reviewers and the editors of JIRD for their valuable com-
ments. I would also like to thank Mary Elizabeth Kelley-Bibra for her excellent language editing.
Appendix
Clustered list of interviewees
Clean Clothes Campaign:
Large Western European organisations C1

C2
C6
C9
C10
Smaller Western European organisations C4
C5
Southern European organisations C3
Northern European organisations C11
Central-and Eastern European organisations C7
C8
Asian organisations C12
C14
Friends of the Earth:
Large Western European organisations F1

F2
F4
F5
F10
Southern European organisations F6
F8
Northern European organisations F9
Central-and Eastern European organisations F3
F7
F11
South American organisations F13
African organisations F12
References
Adler, Emanuel and Vincent Pouliot (2011) International practices, New York: Cambridge University
Press.
Ankersmit, Frank R. (1996) Aesthetic politics: Political philosophy beyond fact and value, Stanford:
Stanford University Press.
Bäckstrand, Karin (2006) ‘Democratizing global environmental governance? Stakeholder democracy
after the world summit on sustainable development’, European Journal of International Rela-
tions 12(4): 467‒98.
Beisheim, Mariann (2001) ‘Demokratisierung einer klimapolitischen Global Governance durch
NGOs?’ [Democratization of Global Climate Politics through NGOs?], in Achim Brunnengräber,
Ansgar Klein and Heike Walk, eds, NGOs als Legitimationsressource. Zivilgesellschaftliche Par-
tizipationsformen im Globalisierungsprozess, 115‒36, Leverkusen: Leske + Budrich.
Beauzamy, Brigitte (2010) ‘Transnational Social Movements and Democratic Legitimacy’, in Eva
Erman and Anders Uhlin, eds, Legitimacy Beyond the State? Re-examining the Democratic Cre-
dentials of Transnational Actors, 110‒29, Basingstoke: Palgrave.
Bendell, Jem (2006) Debating NGO accountability, NewYork, Geneva: United Nations.
Berger, Tobias and Esguerra, Alejandro (2018) ‘Conclusion: power, relationality and difference’, in
Tobias Berger and Alejandro Esguerra, eds, World politics in translation: Power, relationality
and difference in global cooperation, 217‒31, London, New York: Routledge.
Bexell, Magdalena, Jonas Tallberg and Anders Uhlin (2010) ‘Democracy in global governance: The
promises and pitfalls of transnational actors’, Global Governance: A Review of Multilateralism
and International Organizations 16(1): 81‒101.
Boltanski, Luc (2011) On critique: A sociology of emancipation, Cambridge: Polity Press.
Braun, Benjamin, Sebastian Schindler and Tobias Wille (2018) ‘Rethinking agency in International
Relations: performativity, performances and actornetworks’, Journal of International Relations
and Development 21(2): 1‒21.
Bueger, Christian (2017) ‘Practice’, in Xavier Guillaume and Pinar Bilgin, eds, Routledge Handbook
of International Political Sociology, 328‒37, London, New York: Routledge Taylor & Francis
Group.
Bueger, Christian and Frank Gadinger (2014) International practice theory: New perspectives, Basing-
stoke: Palgrave Macmillan.
Butler, Judith (2004) Undoing gender, New York: Routledge.
Castiglione, Dario and Mark E. Warren (2019) ‘Rethinking democratic representation: Eight theoretical
issues and a postscript‘. in Lisa Disch, Mathijs van de Sande, and Nadia Urbinati, eds, The Con-
structivist Turn in Political Representation, 21–4, Edinburgh: Edinburgh University Press.
Chilvers, Jason and Matthew Kearnes (2015) Remaking participation: Science, environment and emer-
gent publics, London: Routledge.
Disch, Lisa (2016) ‘Representation’, in Lisa Disch and M. E. Hawkesworth, eds, The Oxford Handbook of
Feminist Theory, 781‒802, New York: Oxford University Press.
Disch, Lisa (2015) ‘The “constructivist turn” in democratic representation: A normative dead‐end?’, Con-
stellations 22(4): 487‒99.
Disch, Lisa (2011) ‘Toward a mobilization conception of democratic representation’, American Political
Science Review 105(1): 100‒14.
Doherty, Brian (2006)’Friends of the earth international: Negotiating a transnational identity’, Environ-
mental Politics 15(5): 860‒80.
Dryzek, John S. and Simon Niemeyer (2008) ‘Discursive representation’, American Political Science
Review 102 (4): 481‒93.
Egels-Zandén, Niklas (2011) ‘Clean clothes campaign’, in Thomas Hale and David Held, eds, Handbook
of transnational governance: Institutions and innovation, 259‒65, Cambridge: Polity Press.
Emirbayer, Mustafa (1997) ‘Manifesto for a Relational Sociology’, American Journal of Sociology
103(2): 281–317.
Esguerra, Alejandro, S. Beck and R. Lidskog (2017) ‘Stakeholder Engagement in the Making: IPBES
Legitimization Politics’, Global Environmental Politics 17(1): 59–76.
Esmark, Anders (2007) ‘Democratic accountability and network governance—problems and potentials’,
in Eva Sørensen and Jacob Torfing, eds, Theories of democratic network governance, 274‒96, Bas-
ingstoke: Palgrave Macmillan.
H. Knappe
Gadinger, Frank (2016) ‘On justification and critique: Luc Boltanski’s pragmatic sociology and interna-
tional relations’, International Political Sociology 10(3): 187‒205.
Gerring, John (2007) Case Study Research, Cambridge: Cambridge University Press.
Guzzini, Stefano (2017) ‘International Political Sociology, or: the Social Ontology and Power Politics of
Process’. In: Xavier Guillaume and Pinar Bilgin (eds) Routledge Handbook of International Politi-
cal Sociology, pp. 368‒77. London, New York: Routledge Taylor & Francis Group.
Hahn, Kristina and Anna Holzscheiter (2013) ‘The ambivalence of advocacy: Representation and contes-
tation in global NGO advocacy for child workers and sex workers’, Global Society 27(4): 497‒520.
Hitchings, Russell (2012) ‘People can talk about their practices’, Area 44(1): 61‒67.
Holzscheiter, Anna (2016) ‘Representation as power and performative practice: Global civil society
advocacy for working children’, Review of International Studies 42(2): 205‒26.
Jönsson, Christer and Jonas Tallberg (2010) Transnational actors in global governance: Patterns, expla-
nations and implications, Basingstoke: Palgrave Macmillan.
Keck, Magaret E. and Kathryn Sikkink (1998) Activist beyond borders. Advocacy networks in interna-
tional politics, Ithaca: Cornell University Press.
Khagram, Sanjeev, James V. Riker and Kathryn Sikkink (2002) Restructuring world politics: Transna-
tional social movements, networks, and norms, Minneapolis: University of Minnesota Press.
Knappe, Henrike (2017) Doing democracy differently: Political practices and transnational civil society,
Leverkusen: Budrich UniPress.
Koenig-Archibugi, Mathias and Kate Macdonald (2013) ‘Accountability-by-proxy in transnational non-
state governance’, Governance 26(3): 499‒522.
Kuyper, Jonathan W. and Karin Bäckstrand (2016) ‘Accountability and representation: Nonstate actors in
UN climate diplomacy’, Global Environmental Politics 16(2): 61‒81.
Latour, Bruno (2003) ‘What if we talked politics a little?’, Contemp Political Theory 2(2): 143–64.
Macdonald, Kate and Terry Macdonald (2010) ‘Democracy in a pluralist global order: Corporate power
and stakeholder representation’, Ethics & international affairs 24(1): 19‒43.
Macdonald, Terry (2008) Global stakeholder democracy: Power and representation beyond liberal
states, Oxford: Oxford University Press.
Mansbridge, Jane (2003) ‘Rethinking Representation’, American Political Science Review 97(4): 515‒28.
Mansbridge, Jane (1999) ‘Should Blacks Represent Blacks and Women Represent Women? A Contingent
“Yes”’, Journal of Politics 61(3): 628‒57.
Montanaro, Laura (2018) Who Elected Oxfam? A Democratic Defence of Self-Appointed Representatives,
Cambridge: Cambridge University Press.
Mulieri, Alessandro (2013) ‘Beyond Electoral Democracy? A Critical Assessment of Constructivist Rep-
resentation in the Global Arena’, Representation 49(4): 515–27.
Näsström, Sofia (2015) ‘Democratic Representation Beyond Election’, Constellations 22(1): 1‒10.
Peruzotti, Enrique (2010) ‘Democratic credentials or bridging mechanisms? Constituents, representa-
tives, and the dual politics of democratic representation’, in Eva Erman and Anders Uhlin, eds,
Legitimacy Beyond the State, 153‒72, Basingstoke: Palgrave Macmillan.
Pitkin, Hanna F. (1967) The concept of representation, Berkeley/Los Angeles: University of California
Press.
Reckwitz, A. (2003) ‘Grundelemente einer Theorie sozialer Praktiken. Eine sozialtheoretische Perspek-
tive’ [Basic Elements of a Theory of Social Practices. A Perspective in Social Theory], Zeitschrift
für Soziologie 32(4): 282–301.
Rehfeld, Andrew (2011) ‘The concepts of representation’, American Political Science Review 105(3):
631‒41.
Rehfeld, Andrew (2009) ‘Representation Rethought: On Trustees, Delegates, and Gyroscopes in the
Study of Political Representation and Democracy’, American Political Science Review 103(2):
214–30.
Saward, Michael (2014) ‘Shape-shifting representation’, American Political Science Review 108(4):
723‒36.
Saward, Michael (2010) The representative claim, New York: Oxford University Press.
Saward, Michael (2009) ‘Authorisation and authenticity: Representation and the unelected’, Journal of
Political Philosophy 17(1): 1‒22.
Schatzki, Theodore, Karin Knorr-Cetina and Eike von Savigny (2005) The practice turn in contemporary
theory, London, New York: Routledge Taylor & Francis Group.
Scholte, Jan A. (2007) ‘Global civil society–opportunity or obstacle for democracy’, Development Dia-
logue 49(3): 15‒28.
Severs, Eline (2010) ‘Representation as claims‐making. Quid responsiveness?’, Representation 46(4):

411‒23.
Solomon, Ty and Brent J. Steele (2017) ‘Micro-moves in international relations theory’, European Jour-
nal of International Relations 23(2): 267‒91.
Tallberg, Jonas, Thomas Sommerer and Theresa Squatrito (2013) The opening up of international organi-
zations, New York: Cambridge University Press.
Tarrow, Sidney (2005) The New Transnational Activism, Cambridge: Cambridge University Press.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.
Henrike Knappe is Scientific Project Leader of the ‘Politicizing the Future’ project at the Institute for
Advanced Sustainability Studies, Potsdam. Her research interests include practices of democratic repre-
sentation, intersectional future studies and sustainability politics. She obtained her Ph.D. at Lueneburg
University. She was visiting researcher at Washington University Seattle and Stockholm University.

Knappe (2020)

Uploaded by

Copyright:

Available Formats

Knappe (2020)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Knappe (2020)

Uploaded by

Copyright:

Available Formats

Journal of International Relations and Development

Representation as practice: agency and relationality

© Springer Nature Limited 2020

Keywords Political representation · Practice theory · Representative claims ·

Transnational civil society actors claim to represent constituencies—for example,

accountability mechanisms of electoral politics. This growing space of transnational

Flexibility, relationality and dialogical agency

the standard assumption of representation as a fixed relationship which is clearly

Practice theory emphasises relationality—that is actors’ behaviour interwoven with

The idea of relationality is taken up in representation theory by scholars who

a political process, in which representations are ‘act[s] of power’ (Berger and

Transnational civil society’s representation practices

What do we see when we examine representation practices? The Clean Clothes

Representation as an organic process

Trust and shifting agency

The dialogical relationship between representatives and represented in both of the

representation has shifted. It is strongly connected to the notions of inclusion and

Representative claims at a distance

The voicing of representative claims towards sometimes distant constituencies

Embodiment, the intertwined character of activity and physical presence in practice

communication is understood as communication mediated through technical

Flexibility, relationality and dialogical agency in empirical practice

In the following, I discuss the results of my study in light of the practice-theoretical

Communication cannot be just transferred to another space, cybernetic space: actors

representation practices in transnational civil society networks be approached from

within representation practice, the discursive claims and the disembodiment in

Clustered list of interviewees

Clean Clothes Campaign:

Large Western European organisations C1

Friends of the Earth:

Large Western European organisations F1

Severs, Eline (2010) ‘Representation as claims‐making. Quid responsiveness?’, Representation 46(4):

You might also like