Content uploaded by Sebastian Haunss
Author content
All content in this area was uploaded by Sebastian Haunss on Oct 15, 2017
Content may be subject to copyright.
Electronic copy available at: http://ssrn.com/abstract=1162209
Conflicts About Intellectual Property Claims
The Role and Function of Collective Action Networks
Sebastian Haunss & Lars Kohlmorgen
Author Posting. © Taylor & Francis, 2010. This is the author's version of the work. It is posted here by
permission of Taylor & Francis for personal use, not for redistribution. The definitive version was published in
Journal of European Public Policy, Volume 17 Issue 2, March 2010.
doi:10.1080/13501760903561534 (http://dx.doi.org/10.1080/13501760903561534)
Abstract:
Decision making processes in Europe involve complex networks of actors who are trying to
influence them on the various levels of the European multi-level governance system. Interest
group research often assumes that the ability of an actor to exert influence depends mainly on
its financial and personal resourcefulness, on its ability to provide expert knowledge and on
its economic and/or political power. Recent conflicts in which “weak” actors were able to
persist have challenged this assumption. We claim that a careful analysis of the actor networks
is able to complement the traditional actor-resource-centered perspective, and that paying
attention to the structure of collective action networks is necessary to fully grasp the dynamics
of decision-making processes in Europe in which the power of networks sometime outweighs
the power of resources.
Keywords: Europe, interest groups, intellectual property, network analysis, protest
1. Introduction
In the Fall of 2003, two proposals were pending in the European Parliament that would affect
the rules governing intellectual property (IP) in Europe. Tw o years later the deliberation on
both proposals had come to an end – with opposing outcomes:
In March 2004, the European Parliament adopted the “Directive on the Enforcement of
Intellectual Property Rights” (IPRED 1, generally referred to as simply the Enforcement
Directive) by a vote of 330 to 150. The directive was intended to strengthen and harmonize
the enforcement of intellectual property rights, including copy, trademark, and patent rights,
in the EU member states (COM 2003). It gives rights holders more possibilities to bring civil
suit against counterfeiters and other violators.
The second proposal, the “Directive on the Patentability of Computer Implemented
Inventions”, known as the Software Patent Directive, was soundly rejected by the European
Parliament in July of 2005, by a margin of 648 to 14. This directive was drafted by the
1
Electronic copy available at: http://ssrn.com/abstract=1162209
European Commission to introduce patent rights for inventions “implemented on a computer
or similar apparatus which is realized by a computer program” (COM 2002, 13). This
surprising decision marked the end of nearly four years of discussion, lobbying, campaigning
and bargaining between EU institutions, business associations, civil society groups, national
parliaments, and the media, in which the opponents of the Directive would normally have
been considered much weaker than its proponents..
These two conflicts were largely concurrent and had many similar characteristics. Both took
place in the same policy field (intellectual property rights), in an initially similar institutional
and procedural setting, and involved a significant number of the same kinds of actors. Yet one
passed by a 2 to 1 margin and the other was overwhelmingly defeated. Why did this happen?
Why would interest groups traditionally regarded as strong succeed in the first case but fail in
the other? Why were “weak interests” able to mobilize in a way no one had anticipated to
prevent the adoption of the Software Patents Directive? So far neither the literature on
European policy making, on interest groups in Europe (Eising 2004; Greenwood 2003;
Richardson 2000; Bouwen 2004; Bennett 1999) nor the literature on policy networks (Kenis
and Schneider 1991; Kohler-Koch 2002; Kriesi, Adam, and Jochum 2006; Rhodes 1997) can
give satisfactory answers to these questions. The empirical puzzle leads to a reconsideration
of the existing literature in light of recent conflicts in the field of politics of intellectual
property (IP), suggesting that additional factors must play a role in determing the success or
failure of attempts by weak actors to influence the policy decisions.
In this article we suggest that research on social movements and social networks provides
tools that allow us to complement existing approaches by identifying additional mechanisms
that help explain the unexpected trajectories of these two conflicts.
After a discussion of the relevant literature, we present results of an analysis of the actor
networks involved in both conflicts, identifying two network-related mechanisms that help
explain these outcomes, which effectively contradict central assumptions about interest group
politics in Europe. We argue that relational aspects are crucial to understanding the structure
of the conflicts and their outcomes. We will show that the structures of the action networks
and of the coalitions built among individual actors strongly affected the decision-making
process and the actors’ chances of influencing it.
2. Interest Groups in the European Polity
Both conflicts were characterized by intense lobbying and political mobilization by various
interest groups. In the sizable literature on interest groups in Europe there is broad agreement
2
that in the complex multi-level system of European governance the impact of interest groups
differs decidedly from existing national and international settings (Eising 2004; Hooghe and
Marks 2001; Kohler-Koch 1997; Scharpf 2002). As an action and governance system sui
generis, the European political arena is characterized by a multiplication of negotiation
arenas.
Following Olson's (1968) classic theorem that the incentive to invest resources for the
creation of collective goods diminishes with group size, studies of interest representation in
Europe usually assume that small groups with specific interests or large individual firms have
the best chance of influencing policies (Eising 2004). This general asymmetry should be even
more pronounced at the European level, as interest representation there requires actors to be
active on multiple levels of the governance system simultaneously (Bouwen 2004; Buholzer
1998; Eising and Kohler-Koch 1994; Grande 1996; Kohler-Koch 1996; Bennett 1999).
The mainstream of research on interest groups in Europe clearly places public and social
interests as well as small and medium-sized enterprises (SMEs) at a disadvantage relative to
transnational corporations and large business associations. The former set of groups usually
either lack the necessary resources to establish a continuous presence in Brussels or they are
unable to satisfy the specific information and knowledge needs of the two most powerful
European institutions, the Council of the European Union and the European Commission
(Burns 2004; Hayes-Renshaw and Wa lla ce 1997). Nevertheless, from this perspective,
consumers, workers or civil society groups might be able to compensate for their structurally
weak position by politicizing contentious issues (Kohler-Koch 1997; Beyers 2004). But until
now there has been no systematic empirical research that would tell us the conditions under
which such a strategy would succeed.
Given the constellation of actors involved in the two IP conflicts, the interest groups literature
would predict relatively clear outcomes. In the software patent case, where all the “strong”
actors supported the directive and those opposing the directive would generally be seen as
weak, we would expect an easy win for those supporting the directive. In the case of the
Enforcement Directive, the situation was more complicated. While most European business
associations supported the directive, a significant number of large firms and business
associations (mainly from the telecommunications, generic medicine, and auto parts
industries) opposed it. Here it is much less clear what outcome the interest group literature
would predict. But in both cases the de facto conflict trajectories were quite different: Despite
some minor protest and with very limited modifications of the original proposal (the
withdrawal of the originally proposed additional criminal sanctions), the Enforcement
Directive was adopted without a hitch, while the Software Patent Directive failed. These
3
results suggest that factors other than those elaborated in the interest group literature may
have actually been decisive in these two conflicts.
The strength of interest group research is that it shows how the resourcefulness of an actor
usually corresponds with its ability to have its interests heard, or more precisely, how different
key resources matter at different levels of the European governance system. Research in this
area, however, has primarily focused on strong actors and elite interaction (Imig and Tarrow
2001) and is not well suited to explain the occasional success of actors it regards as weak.
Moreover, actors are classified as strong or weak mainly on the basis of their access to
resources. We believe that this view is too static and ignores interactional variables such as
the structure of interactional networks among actors, rather than the attributes of the actors
themselves might better explain success or failure.
The literature on policy networks analyzes European policy making from just such a
relational perspective (Kenis and Schneider 1991; Kriesi, Adam, and Jochum 2006;
Richardson 2000; Rhodes 1997), though it tends to focus on new forms of governance
characterized by “informal, decentralized and horizontal relations” (Kenis and Schneider
1991, 32) or on relatively stable interactions between established actors. There are also a few
studies of temporary, issue-oriented networks, as well as networks of NGOs and other civil
society actors, but there is no agreement if and/or under which conditions these issue
networks or policy coalitions would be more effective at interest representation than stable
policy networks. The dominant assumption is that “policy communities”, the most stable and
integrated type of policy network, will generally be more likely to accomplish their goals than
less integrated networks (Rhodes 1997). Other studies, however, contend that, at least in the
field of environmental policies, short-term, issue-specific coalitions have been more effective
than broader long-term networks (Warleigh 2000).
2.1. Social Movements and Networks
In contrast to the literatures on both EU interest groups and policy networks, social movement
research has traditionally and systematically focused on weak actors and paid closer attention
to networks of interaction. Studies from the political opportunity structure perspective (Kriesi
1995; McAdam 1996; Tarrow 1994; Tilly 1978), for example, underline that social
movements act within a social environment that structures their chances to mobilize adherents
and to influence policy-making processes. A recurring claim is that factors such as the relative
openness or closure of the political system, the stability of political alignments, the
availability of elite allies, and the state’s capacity and propensity for repression critically
4
structure the outcomes of movement campaigns. This perspective suggests that in European
policy conflicts it would therefore be advisable to look for alliance structures, especially with
strong allies inside and outside the institutions, and to look for fissures and conflicts between
the various European institutions.
Other authors stress that discursive factors must also be taken into account to explain the
success or failure of social movements (Snow et al. 1986; Snow 2004), especially the creation
of a collective identity, which they argue is a precondition for collective action (Gamson
1992; Haunss 2004; Melucci 1996). Following this line of thought, rather than trying to act as
a (loose) coalition of individuals with a common interest, weak actors should work to
construct a coherent collective action frame and some sort of collective identity that allows
them to identify the field of opportunities and constraints of their action and holds them
together as a collective actor. The network perspective on social movements also draws
attention to the multiplicity of linkages that connect people, organizations, events, and frames
(Diani and McAdam 2003) and to the social structures that facilitate or hinder mobilization
along existing interpersonal ties (McAdam 2003).
Social network analysis in general developed in explicit opposition to the methodological
individualism of other approaches that focus on attributes of discrete social units or groups. It
interprets behavior as the result of patterned interaction between social actors embedded in a
social structure that is itself a network of networks (Wellman 1988, 20). Its central assumption
is that it is often not the discrete characteristics of an actor that determine its role and
influence but its embeddedness in a larger network of interaction and its ties to other actors in
that network. Depending on the specific situation strong or weak ties, direct or indirect ties
can be more important (Granovetter 1973). Generally speaking, the number of links
connecting an actor to the network (referred to as degree centrality) is a relevant measure of
its importance. But there are also times when actors that bridge otherwise unconnected
subnetworks may be more powerful than those that act as a hub within a network (Burt 1992).
Thus, the literatures on social movements and social networks would both point to the
topography of the collective action networks as a critical factor in the two European IP
conflicts of concern here. For that reason, and because social movement research explicitly
focuses on precisely those actors that interest group research usually classifies as “weak”, our
analysis draws mainly on these two perspective – without neglecting the importance of
resources and political context.
5
3. The networks of interaction in two IP conflicts
The existing literature does not provide a solid basis to formulate strong hypotheses regarding
the nature of successful actor networks. But in the absence of an explanatory framework
exploratory network analysis can be used to explain success or failure by identifying
important relational characteristics of the network.
Drawing on the existing research we suggest four network characteristics which may
significantly affect the outcomes of such policy struggles:
1) Density of cooperative links: Actors in influential policy networks as well as social
movement actors tend to form strong cooperative links with other actors (Rhodes 1997).
Strong cooperation is generally interpreted as an expression of social cohesion that facilitates
consensus building within a group (Wasserman and Faust 1994, 250 f.). Groups of actors
should have a greater chance of exerting influence if they appear as a cohesive collective
actor with a clear profile and a persuasive collective action frame, rather than a loose alliance
of disparate special interests. They should be able to formulate a common position and
develop a unified strategy. In short, we would expect successful coalition networks in political
conflicts to be characterized by a high density of cooperative links.
2) Location of actors in the network: Not all actors in large networks are equally important.
Some are better connected than others, some act as hubs that are connected to many other
actors, others may act as brokers connecting otherwise unconnected or only weakly connected
subnetworks. On that basis, we might expect formal coalitions and established policy
networks to have a different distribution of central actors than would conflict networks that
develop more like a social movement mobilization. Similarly, different types of networks
should exhibit differing degrees of network centralization, i.e. we would see variation in the
centrality scores of different kinds of networks (Wasserman and Faust 1994, 169 ff.).
3) Duration of cooperation: Interest group research often assumes that having regular and
established contact with EU institutions, i.e. insider status, is crucial for influencing decision-
making processes in the EU (Broscheid and Coen 2003). This view is largely supported by the
policy networks literature. An analysis of the actor networks should therefore pay attention to
the extent to which they are built on pre-existing structures of formal or informal cooperation.
4) Size: Obviously network size is likely to be an important factor. All else being equal, a
large network of actively cooperating actors should be able to mobilize significantly more
resources than a smaller network, even if each actor's contribution is relatively small. At the
same time, a large network may be less flexible and react more slowly to changing conditions.
6
4. Data and Methods
For an exploratory network analysis that focuses on the above mentioned factors, the relevant
network is the collective action network comprising all actors actively involved in a particular
policy conflict. We define collective action networks as those networks that include all
interacting actors involved in the conflicts, ranging from civil society organizations and firms
to public institutions (such as parliaments and the European Commission). To identify these
networks we used a triangulation method, combining data from different sources. News
coverage in major national newspapers was the first source. Using the political claims
analysis framework developed by Koopmans and Statham (1999) we conducted a content
analysis of all newspaper articles published in selected quality newspapers of four countries
(Süddeutsche Zeitung, die tageszeitung, Frankfurter Rundschau, Die We lt, Stuttgarter Zeitung
for Germany; Daily Mail, The Times, The Guardian, Financial Times, Western Mail, Morning
Star, The Daily Telegraph, The Business, The Independent, The Observer for the UK; Le
Figaro, Liberation, Les Echos, Le Monde for France; Gazeta Wyborcza, Polityka,
Rzeczpospolita, Wprost for Poland) that mentioned either or both of the conflicts.1 From this
analysis we generated a list of actors that had been mentioned in the press. A second source
was semi-structured interviews conducted with 25 key actors about their perception of the
conflict, their role in it, and their cooperation networks. We further expanded the list of actors
and relations by doing a content analysis of documents published on the web, and finally we
sent a questionnaire to all the actors we had identified up to that point.
The number of actors involved in the two conflicts was very large. More than 90.000
individuals and firms had signed the EuroLinux petition2 by the end of 2001, all 785 MEPs
had received emails and other lobbying material regarding one or both of the conflicts, and
the European Information & Communications Technology Industry Association (EICTA), one
of the major proponents of software patents, claims on its website3 to represent more than
10,000 European firms. Obviously not all of these actors played significant roles in either of
the conflicts. Many of the information technology firms represented by EICTA, for example,
may not have even known what the disputed directives were about, and most of the signers of
the EuroLinux petition did nothing beyond signing the petition. The actors we included in the
networks, therefore, included only those mentioned in the press, those from whom we
received completed questionnaires (including those we interviewed), and members of the
1Overall a total number of 170 articles (G: 75, UK: 37, F: 45, PL: 31) were coded. For more inform-
ation on the coding and a detailed list of the codes used see (Haunss and Kohlmorgen 2008).
2http://petition.eurolinux.org/
3http://www.eicta.org/index.php?id=10
7
business and civil society associations, NGOs, and ad-hoc coalitions who showed significant
commitment beyond signing a petition.
In the resulting networks, nodes represent actors and edges represent cooperative relationships
as indicated by any kind of joint activity, such as membership in a formal coalition,
organizing a hearing together, or signing a petition or letter together. Because we assume that
cooperation is reciprocal, relationships in our graphs are undirected. To mitigate this relatively
strong assumption, we base our main argument only on the analysis of the network cores (k-
core ≥ 2) and ties where we have reports of cooperation from at least two independent sources
(as indicated by a line value ≥ 2).
5. Conflicts about the EU directives on software patents and IP enforcement
The two directives that we have chosen have played a central role in shaping the regulatory
framework for intellectual property rights in the EU during the last decade. The “Directive on
the Enforcement of Intellectual Property Rights” (IPRED 1) aims to strengthen and harmonize
the enforcement of intellectual property rights in the EU member states, including copy,
trademark, and patent rights. It requires all member states to apply “penalties which must be
effective, proportionate and deterrent” (COM 2003, 19) against counterfeiting and piracy.
The “Directive on the Patentability of Computer Implemented Inventions” – the Software
Patents Directive (SWPat) – was intended to introduce patents on inventions “implemented on
a computer or similar apparatus which is realised by a computer program” (COM 2002, 13).
Whether this definition would include “software as such”, which is explicitly exempted from
patentability in the European Patent Convention, was highly disputed. Certainly the
opponents of the directive succeeded in framing it as the “Software Patents Directive”, and
only the core supporters called it the “Computer Implemented Inventions” (CII) Directive.4
The scope and impact of these two directives differ significantly. The Enforcement Directive
touches on several issues, including intellectual property rights in music, movies,
pharmaceuticals, luxury goods, automotive parts, and initially also software. It also covers
different forms of intellectual property rights, such as patents, copyrights and trademarks. The
Software Patents Directive, on the other hand, had a much narrower scope. This difference in
scope might have significant implications for mobilization, since an issue that affects a
broader constituency would start off with a larger potential mobilization pool. At the same
time, the breadth of the issue could also hinder mobilization by making it more difficult to
construct a collective action frame broad enough to convince and mobilize all affected parties.
4According to a former commission employee even the Commission circulated its the preparatory
documents with filenames containing “swpat”.
8
Thus, the narrow focus of the Software Patents Directive could make it easier to construct a
collective action frame, while simultaneously making frame bridging – building coalitions
with actors who are not directly affected by the directive – more complicated (a dilemma
confronting many single issue movements).
Aside from the difference in scope, the similarities between the two directives are striking.
The time frame and the institutional setting of both directives were nearly identical. The
respective Green Papers were published within a year and a half of each other (SWPat in June
1997 and IPRED 1 in October 1998), consultations for both were held in 1999, and the
proposals were published in February 2002 (SWPat) and January 2002 (IPRED 1). Both
legislative processes were carried out using the “co-decision procedure.” The Commission
backed both proposals, arguing that they were necessary to harmonize the internal European
market and to comply with international treaties. It furthermore argued that both directives
would strengthen the global competitiveness of European industries.
There were also strong similarities in the constellations of actors involved in these conflicts.
In both cases the Commission received strong support from industry lobbying groups that
represented powerful players in the respective fields (see Table 1). It was also true, however,
that business interests did not unanimously support either of the Commission’s proposals.
Major firms from the European telecommunications industry opposed the Enforcement
Directive, and a large number of mostly SMEs opposed the Software Patents Directive.
Lastly, consumer interest and other civil society groups mobilized against both directives.
Table 1: Actors supporting and opposing the Software Patents Directive and the
Enforcement Directive
Software Patents Directive Enforcement Directive
Pro • Commission
• Large firms
• Business associations
• Most national governments
• Commission
• Business associations
• Most national governments
Contra •SMEs
•NGOs
• Civil society organizations
• Some national parliaments
• Large firms
•NGOs
Despite the structural similarities between these two decision-making processes and the
similarities in the kinds of actors involved, there were also significant differences in their
trajectories and the intensity of the conflicts. There was a heated debate about the desirability
of software patents – an issue that initially had seemed relatively uncontroversial. In contrast,
9
the legislative process in the case of the Enforcement Directive was relatively smooth and
uneventful, though one would have expected much more conflict here, since the directive
touches upon issues like file-sharing, which have received much more public attention than
the arcane issue of software patents.
Aside from the above mentioned difference in scope, one reason for the differing levels of
conflict was that there was a significant difference in the de-facto decision-making process: In
the case of the Enforcement Directive, the process was considerably sped up, through the
introduction of a so-called “trialogue,” i.e. informal negotiations between the European
Parliament, the European Commission, and the Council of the European Union. As a
consequence of the intense conflict over the Software Patents Directive, the main actors
involved wanted to avoid a similar conflict over the Enforcement Directive. Additionally
there were concerns that the new EU member states (which in some cases suffered from
widespread infringement of intellectual property rights) might complicate and slow down the
decision-making process. Consequently, they aimed for an adoption after the first reading in
the European Parliament, in an attempt to pass the directive before the EU enlargement took
place May 2004. This specific form of decision making closed off certain avenues of
influence for extraparliamentary opponents of this directive, who were already weaker and
less formally organized than those supporting the directive. At the same time, this unusual
procedure would not have been available but for the weakness of the directive's opponents,
since a stronger oppositional mobilization would most likely have made the trialogue
infeasible.
Thus, we have two political conflicts in the same broader policy field with some differences
in the scope of the proposed directives but with strong similarities in the institutional settings
and actor constellations at the beginning of each conflict.
5.1 The Software Patents Directive
In the case of the Software Patents Directive the central actor network comprised about 800
actors (see figure 1). This included six large membership and support networks. On the side of
the proponents these were the formal membership network of the European Information &
Communications Technology Industry Association (EICTA) that included 37 different
national associations and 50 individual corporations, and the Business Software Alliance
(BSA), which presents itself on its website as “the voice of the world's commercial software
industry and its hardware partners”. Even though a number of relevant companies (e.g. SAP,
Intel, Adobe Systems, Apple and Symantec) belong to both EICTA and BSA , the network
10
data shows minimal cooperation between the two associations. This may be due to the latent
rivalry between the two associations about leadership in representing the major high-tech
industries in Europe.
On the opponents' side, there was one formal membership network, the Foundation for a Free
Information Infrastructure (FFII), and three large informal support networks, the EuroLinux
Alliance, Economic-Majority.com, and patentfrei.de.
Figure 1: Network Involved in the Conflict Over the Software Patents Directive
Edges represent cooperation, vertex size represents betweenness-centrality, color indicates affiliation to the
respective organizations/networks.
The most important difference between the proponents’ and opponents’ networks is that the
latter did not exist prior to the start of the conflict. Preexisting networks such as the two SME
associations, Confédération Européene des Associations de Petites et Moyenne Entreprises
(CEA-PME) and Union Européenne de l'Artisanat et des Petites et Moyennes Entreprises
(UEAPME), played only a minor role and did not actively contribute to the mobilization or
the framing of the conflict, though they did provide some infrastructural support. CEA-PME
cooperated closely with FFII and used their established contacts to MEPs. Because most of
the opponents’ network emerged and was actively constructed during the mobilization, we
define it as a situational network.
11
FFII
EPP-ED
ALDE
Greens/EFA
PSE
EP
UEN
Independent
UEAPME
Ericsson
Deutsche Telekom
Microsoft
Nokia
Sun Microsystems
BSA
Florian Müller
AIPPI
European Commission
EICTA
SAP
Siemens
PROSA
EDRi FSFE
ICC BEUC
CompTIA
EPC
Red Hat
MySQL
Intellect
Alcatel
Philips
UNICE
BVDW
BDI
ZVEI
IBM
Hewlett-Packard
BMJ
ANIE
Hartmut Pilch
ISOC Polska
CEA-PME
Barrapunto.com
EFF
SIUG
FFS
EPO
FeONIC
Aleph One Ltd
Datalab
Akin Gump
Sonologic
Adobe
Apple
Cisco Systems
Symantec
Dell Computers
economic-majority.com
GrowthPlus
BITKOM
SEDISI patentfrei.de
BVMW
ACEA
CECED
COCIR
EDMA
ProTon Europe
ITEA
MEDEA+
Polish Government
AB Strakt
Nexedi
Council
Eurolinux
Pro
Contra
MEPs
Commission
Council
Other
And because its ad-hoc nature was combined with a thematic focus on just one issue
(software patents), we further classify it as a single issue situational network. Since those
participating in the network joined with the clear objective of preventing the directive's
passage, many of them showed a high degree of commitment and dedication. The majority of
those opposing the directive were individual software programmers or small software
companies.
In contrast, the preexisting large membership networks of the proponents were neither
situational nor focused on a single issue. Rather, they were ongoing associations concerned
with a range of issues, one of which was the Software Patents Directive. Thus, the proponents'
network had a more continuous character and its members brought in different interests and
different motivations for participating in the network.
Figure 2: Close Cooperation in the SWPat-Network Core
(line values ≥ 2, k-core ≥ 2, 42 vertices)
Edges represent cooperation, node size represents betweenness-centrality, color indicates affiliation to the
respective organizations/networks.
A closer look at the core (figure 2) of the Software Patents Directive network illustrates the
uneven access the two camps had to the European institutions. Only those in the proponents'
camp were able to establish stable cooperative relationships with the European Commission.
This fact supports the contention in the EU interest groups literature that European
associations and single large firms would have the best access to the Commission. Neither
12
FFII
Greens/EFA
EICTA
BSA
Florian Müller
Pro
Contra
Eurolinux
economic-majority.com
EICTA SMEs
patentfrei.de
Council
EPP-ED
ALDE
PSE
EP
UEAPME
Microsoft
Nokia
Sun Microsystems
AIPPI
European Commission
Siemens
FSFE
BEUC
EPC
IVSZ
Red Hat
ECIS
Philips
UNICE
ANIE Attac
Hartmut Pilch
ISOC Polska
CEA-PME
Barrapunto.com
FFS
EPO
Aleph One Ltd
NightLabs GmbH
Polish Government
Pro
Contra
MEPs
Commission
Council
Other
camp was able to establish direct cooperative links with the Council. This, too, accords with
the research on interest representation in the EU, which sees the Council as the institution
most open to the lobbying efforts of national interest groups via the national lobbying route.
We see this strategy when, for example, FFII activists tried to lobby national parliamentarians
and government members, asking them to reject the directive in the Council meetings. Thus,
the network had only indirect links to the Council via national institutions. Networks on both
sides of the conflict established close cooperative relationships with MEPs from all the
relevant groups, but the directive's opponents were clearly more successful in this regard.
Moreover, although it is not visible in the graphs, which present only a static illustration of
the conflict, the opponents established their links with the MEPs earlier in the conflict. This is
because, unlike the other side, they immediately understood the importance of the Parliament
in the co-decision procedure, whereas the business associations relied for a long time on their
established connections to the Commission.
Figure 3: Core of SWPat Oppositional Network
(left: k-core ≥ 2, 26 nodes; right: without 3 central nodes)
The oppositional network as shown in figure 3 is relatively dense (0.25) has a moderately
high degree centralization (0.58) and a relatively low betweenness centralization (0.24). The
moderately high degree centralization, which measures the variance of degree centrality in the
network, is an indicator for the existence of central actors. The relatively low betweenness
centralization combined with the high density indicates that the network is nevertheless fairly
nonhierarchical and remains strongly connected even if the central actors are taken out (as on
the right side of Figure 3). The network exhibits the main features of the decentralized,
13
1&1 Internet AG
Sun Microsystems
PROSA
EDRi
FSFE
BEUC
Red Hat
ECIS
MySQL
Attac
Eurolinux
Hartmut Pilch
CEA-PME
Barrapunto.com
SIUG
FFS
Aleph One Ltd
Datalab
LinuxHaus Stuttgart
Sonologic
economic-majority.com
patentfrei.de
NightLabs GmbH
FFII
Greens/EFA
1&1 Internet AG
Sun Microsystems
Florian Müller
PROSA
EDRi
FSFE
BEUC
Red Hat
ECIS
MySQL
Attac
Eurolinux
Hartmut Pilch
CEA-PME
Barrapunto.com
SIUG
FFS
Aleph One Ltd
Datalab
LinuxHaus Stuttgart
Sonologic
economic-majority.com
patentfrei.de
NightLabs GmbH
polycentric collective action networks that have been identified as typical for civil society
networks by Baldassarri and Diani (2007).
Moving now to the proponents' side, their network (figure 4) differs significantly from that of
the opposition. The proponents' network has a density of 0.3, a degree centralization of 0.74,
and a betweenness centralization of 0.55, making it slightly more dense, more centralized, and
more hierarchical than the opponents' network. If we take out the three central nodes, EICTA,
UNICE, and Siemens, the network decomposes into three sub-networks consisting of
European and national business associations and large IT firms that are BSA members. As
Figure 4 illustrates, EICTA mainly tried to mobilize other business associations to support the
directive. BSA and EICTA both mobilized essentially the same large IT firms, and Siemens,
with its strong patent department, was the only non-associational actor that tried to mobilize
independently.
Figure 4: Core of SWPat Proponents' Network
(left: k-core ≥ 2, 31 vertices; right: without 3 central nodes)
The network analysis shows that the opponents managed to build a broad and diversified, yet
flexible network. The proponents' network was much more institutionalized and had only few
important nodes. For example, only a few lobbyists contacted the MEPs, whereas many actors
from the opponents’ network contacted them. These manifold avenues of influence certainly
contributed to the success of the “No Software Patents” campaign.
The successful mobilization against the Software Patents Directive had many characteristics
of a grassroots mobilization. Many committed actors who would have been directly affected
by the directive's passage actively took part in the campaign by writing papers, uploading
websites, organizing demonstrations, and lobbying MEPs. The network was very open so that
all interested actors would be able to participate. This kind of grassroots mobilization also had
an effect at the discursive level. As committed individuals, the directive's opponents had a
high level of credibility among many MEPs. The decentralized structure of the oppositional
14
Ericsson
Microsoft
Nokia
BSA
European Commission
EICTA
SAP
Siemens
ICC
Intellect
Alcatel
Philips
UNICE
IBM
Hewlett-Packard
EPO
LTU Technologies
Intel
Open Forum Europe
GrowthPlus
BITKOM
Agoria
SEDISI
Svenskt N?ringsliv ACEA
CECED
COCIR
EDMA
ProTon Europe
ITEA
MEDEA+
Ericsson
Microsoft
Nokia
BSA
European Commission
SAP
ICC
Intellect
Alcatel
Philips
IBM
Hewlett-Packard
EPO
LTU Technologies
Intel
Open Forum Europe
GrowthPlus
BITKOM
Agoria
SEDISI
Svenskt N?ringsliv ACEA
CECED
COCIR
EDMA
ProTon Europe
ITEA
MEDEA+
network was broadly transnational, with bases in almost all EU member countries. The
opponents also utilized the multilevel structure of the EU by being active at the European
level but also at the national level where they lobbied national governments, parliamentarians,
and parties. Thus, the diversified, transnational character of the network gave the campaign
momentum and was clearly enhanced the opponents' ability to influence the decision-making
process.
Within the diversity of the network as a whole, the FFII was not only a critically important
node in terms of connecting different actors and providing an infrastructure for the campaign;
it also provided the opponents' network with expertise and played a central role in their
collective action framing, especially with regard to interpretation and argumentation. In their
framing, the FFII and other opponent actors combined the notions of the competitiveness of
SMEs with civil rights arguments about freedom of speech, open access, and democratic
procedures. This set of frames was convincing to many of the MEPs as well as the general
public.5
On the other side, the proponents’ network was characterized by a small number of central
actors and clear unanimously supported framing, but it nevertheless did not succeed in
constructing an effective collective actor. This can be explained by the relatively low level of
commitment on the part of individual network actors and by the fact that, for the most part,
the campaign was run by professional lobbyists. The proponents did not manage to build a
mobilization of the type that creates and in turn is fueled by a strong collective identity.
However, once they realized what they were up against, they did try to mimic the methods
and grassroots style of mobilization used by their opponents. EICTA gathered several SMEs
to sign a petition arguing in favor of patents, and the Campaign for Creativity tried to stage an
“astroturf campaign”. But in the end, EICTA and BSA were not able to overcome their rivalry
and put little effort into building a strong common network.
3.2 The Enforcement Directive
The network of relevant actors involved in the conflict on the Enforcement Directive was
much smaller than the Software Patent network (incorporating approximately 300 nodes).
Among the proponents the main actor was the International Federation of the Phonographic
Industry (IFPI), made up of 50 national record industry associations and about 1400
companies in over 70 countries. Together with 12 other business associations (including BSA,
the Motion Picture Association, the International Video Federation, and the European
5We have analyzed the collective action framing of the actors involved in the conflict in a separate
paper (Haunss and Kohlmorgen 2009).
15
Newspaper Publishers' Association), IFPI formed the informal Anti-Piracy Coalition. This
single issue situational network was formed to fight product piracy in Europe and to lobby for
EU legislation against the infringement of intellectual property rights.
Under the leadership of IFPI, this Anti-Piracy Coalition was centrally involved in drafting the
proposed directive and thereby exerted great influence on the whole debate from the start. The
BSA was another important actor in the proponent camp. In contrast to the Software Patents
Directive, where individual MEPs played important but not central roles, one MEP was
central in the network supporting the Enforcement Directive: Janelly Fourtou (formerly of
EEP, now with ALDE), who was the rapporteur in the legislative process and who had close
contacts to both the BSA and the IFPI. She was also already involved in drafting the directive
and actively campaigned for it. The fact that Janelly Fourtou is married to the then CEO of the
French media conglomerate Vivendi Universal, was seen by some organizations and MEPs as
a conflict of interest.
Figure 5: The Enforcement Directive Network
Edges represent cooperation, vertex size represents betweenness-centrality, color indicates affiliation to the
respective organizations/networks.
Other important actors in the network opposing the Enforcement Directive included as main
actors/single networks the European Digital Rights Initiative (EDRI) and the Campaign for an
Open Digital Environment (CODE), two civil society, digital rights initiatives that were
established explicitly for this mobilization. EDRI comprises 25 member organizations, and
16
ALDE
European Commission
EPP-ED
PSE Greens/EFA
EP
IFPI
IPR-Alliance
Deutsche Telekom
Telecom Italia
Yahoo
France Telecom
ETNO
British Telecom
Belgacom
Telenor
TeliaSonera
MCI
Anti-Piracy Coalition
BSA
MPA
IVF
EPC
CODE
BEUC
EDRi
FFII
FSFE
Netzwerk Neue Medien
EGA
FFS
ICC
EICTA
ANIE
IFPM
UNICE
BVDW
Microsoft
EFF
Red Hat
SCONUL
Sun Microsystems
Nokia
ECIS
Irish Government
European Council
Adobe
Corel
IBM
Oracle
Vodafone Group
Verizon
Soros Foundation
Auto Replacement Part Manufacturers
ISPs
Independent Music Producers
EIF
Pro
Contra
MEPs
Commission
Council
CODE, which was mainly organized by the US civil society organization, IP Justice, had 53
members. EDRI had a designated organizer for the campaign against the Enforcement
Directive, who tried to bring together a range of civil society groups, scientists, and small
software developers. They faced an immediate difficulty, however, because at that point,
many software developers and FFII were still involved in the campaign against the Software
Patents Directive, and consequently ended up not being very active in the campaign against
the Enforcement Directive – a fact that obviously weakened the opponents’ network. There
were also attempts to forge cooperative ties between civil society actors and private industry,
but these attempts largely failed. The economic actors that actively opposed this directive
were the telecommunications firms and internet service providers, who did not want to be
held accountable for their customers' infringements of intellectual property rights. They relied
on the European Telecommunications Network Operators' Association (ETNO) and the
informal European Net Alliance to represent their interests. Other industries who saw their
interests negatively affected by the directive were manufacturers of generic medicines and
generic automotive parts. In the end, however, EDRI/CODE, representing civil society and
civil rights groups, and ETNO/European Net Alliance, representing telecommunications
companies and the interests of other private industries, were too different to forge a stable
common network.
Figure 6: Network Cores For and Against the Enforcement Directive
(k-core ≥ 2, line values ≥ 2; left, pro-network; right, contra-network)
Figure 6 shows the two respective network cores. There are a number of noteworthy
differences in their structures. First of all, the proponents’ network (density: 0.38, degree
centralization: 0.63, betweenness centralization 0.76) is highly centralized with the IFPI led
Anti-Piracy Coalition at its center, where the oppositional network (density: 0.13, degree
centralization: 0.28, betweenness centralization 0.48) is much more sparse. On the opponents'
17
Deutsche Telekom
Telecom Italia
NetAlliance
ETNO
British Telecom
CODE
EDRi
FFII
FSFE
Netzwerk Neue Medien
Contra
FIPR
EGA
FFS
Iuridicum Remedium
ECIS
MCI
BEUC
Sun Microsystems
IFPI
Anti-Piracy Coalition
BSA
MPA
IVF
EPC
UNICE
side, one potentially powerful actor, the generic pharmaceuticals industry, is not strongly
connected to the rest of the network, while other industry and civil society interests are
connected only through FFII, which invested only minimal energy in the conflict, and were
organized in a very informal alliance in any case.
This points to a second difference. While both networks included actors with significant
resources, the proponents network coordinated its efforts more effectively for a number of
reasons. Because the Anti-Piracy Coalition was a situational network focused exclusively on
intellectual property rights and the Enforcement Directive, this small but relatively dense
alliance functioned as a kind of relay station for interaction among the various parts of the
larger proponents' network. The involvement of Janelly Fourtou was also an important avenue
for the business associations to exert influence on the European Parliament, and IFPI’s crucial
role in drafting the directive was a great advantage for its proponents. Besides good contacts
with the European Parliament, IFPI also cooperated extensively with the Commission. In
sum, although the proponents' network was not very big, it proved to be very effective and
assertive.
While the opponents' network was also partially made up of situational networks, such as
EDRI, CODE, and the informal European Net Alliance, that were specifically created to lobby
against the Enforcement Directive, this network was too small and developed too late to exert
significant influence on the decision-making process. In addition to its relative low density,
the two main civil society initiatives, EDRI and CODE, had largely overlapping
memberships, making for a smaller base from which to mobilize. Consequently, EDRI and
CODE were not able to initiate a significant political mobilization or any real grassroots
support.
4. Conclusion
Our analysis has shown that relational characteristics of the actor networks can, indeed, help
explain the contrasting outcomes of the two European conflicts over intellectual property
rights legislation. In analyzing the core networks on either side of these two conflicts, two
network-related factors appear to have been especially important in determining the
outcomes.
1. Network size, structure, and intensity of cooperation: The difference in size between the
two networks was not a result of the objective scope of the directives' impact. On the contrary,
the Software Patents Directive objectively affected a much smaller constituency than the
Enforcement Directive, and yet the SWPat network was much larger. This suggests that
18
network size had more to do with the relative effectiveness of mobilization strategies than
with the number of people affected by the decision. In the software patents case the central
mobilizing actors were able to create a snowball effect. The campaign had a relatively open
structure and developed the characteristics of a grassroots mobilization. This created a broad
and diversified network of organizations and individuals and leant the campaign against the
Software Patents Directive for a momentum that largely explains its success.
In both conflicts the successful networks were single issue situational and focused collective
action networks that did not rely solely on preexisting membership. These collective action
networks were able successfully to mobilize support for their position even against
established, resource-rich actors that are usually considered to be more powerful. In other
words, the dynamic mobilization structures of the situational networks were able to counter
the static power of resources.
In the case of the Enforcement Directive, the supporters’ and the challengers’ networks were
both to some degree situational networks centered around the Anti-Piracy Coalition and
around EDRI/CODE, respectively. But the oppositional network did not develop a grassroots
dynamic and lacked a stable connection between the civil society actors and the economic
actors. The supporters of the directive were successful because they combined traditional
forms and avenues of lobbying with engagement in an informal and flexible coalition –
traditional resource-based power with the power of the focused situational network. This
finding suggests that policy networks or, to use our terminology, situational networks may be
more important at the EU level than in national political conflicts, where advocacy coalitions
are more likely to prevail (Warleigh 2000).
2. Commitment: Directly related to these characteristics of the networks is their ability to
mobilize not just support but highly committed participants. While the resource-rich players
relied mostly on traditional lobbying tactics, the anti-software patents network largely
compensated for its lack of financial resources by persuading many individuals to invest a lot
of time in the campaign. While lobbyists in the European institutions are generally accepted
as competent and informed, they are nevertheless also regarded with some skepticism.
Conversely, the highly committed individuals, who were mobilized to lobby against the
Software Patents Directive earned significant credibility through the persuasive presentation
of their own interests. This is where the differences between institutions became most visible.
The Commission only cooperated with established lobbyists and associations, whereas the
Parliament was much more open and responsive to the arguments of concerned individuals.
19
At a broader level, our analysis suggests that for weak actors to prevail in policy conflicts
over established resource-rich opponents, they must undertake a broad mobilization, forge a
dense web of network ties, and construct a convincing master frame. Put in more abstract
terms, in order to be successful, weak actors have to build situational coalitions that fulfill the
conditions of a collective actor with a recognizable collective identity. This implies the
formulation of aims and strategies as well as a shared interpretation of the problem and its
solution.
More generally, these findings show that relational aspects of collective action networks must
be taken into account in assessing the capacity of particular actors to influence policy making.
Obviously actor attributes like economic resources and political power play an important role,
but they do not completely determine an actor’s potential to exert influence. Equally
important is the structure of the collection action networks in which the actors participate. In
the software patents conflict the power of network mobilization was effectively able to
counter the power of resources. Similarly, in the conflict over the Enforcement Directive, the
network structure helps to explain why one coalition of resource-rich actors was more
successful than the other. While we cannot generalize to all policy struggles on the basis of
our two cases, our findings nonetheless suggest that the structure and shape of networks must
be included in any model that seeks to explain the influence of interest groups on policy
decisions. Further research strategically comparing cases across policy fields may be able to
construct a typology of networks and conflicts relating network structures and types of policy
conflicts to the success or failure of interest representation.
Biographical note: Sebastian Haunss is lecturer in political science at the University of
Konstanz. Lars Kohlmorgen is researcher at the University of Hamburg. Correspondence
should be addressed to the first author (sebastian.haunss [at] uni-konstanz.de).
Address: Sebastian Haunss, Universität Konstanz, Fach 90, 78457 Konstanz
Acknowledgements: This research was supported by grants from the Fritz-Thyssen-
Foundation and the Hans-Böckler-Foundation. We would like to thank the anonymous
reviewers for their feedback on an earlier version of this article.
20
References
Baldassarri, D. and Diani, M. (2007), 'The Integrative Power of Civic Networks', American
Journal of Sociology 113(3): 735-780.
Bennett, R.J. (1999), 'Business Routes of Influence in Brussels: Exploring the Choice of
Direct Representation', Politcal Studies 47: 240-257.
Bouwen, P. (2004), 'Exchanging Access Goods for Access: A Comparative Study of Business
Lobbying in the European Union Institutions', European Journal of Political Research
43: 337-369.
Broscheid, A. and Coen, D. (2003), 'Insider and Outsider Lobbying of the European
Commission', European Union Politics 4(2): 165-189.
Buholzer, R.P. (1998), Legislatives Lobbying in der Europäischen Union. Ein Konzept für
Interessengruppen, St. Galler Studien zur Politikwissenschaft ; 21, Bern u.a.: Haupt.
Burt, R. (1992), Structural Holes: The Social Structure of Competition, Cambridge: Harvard
University Press.
COM (2002), Proposal for a Directive of the European Parliament and of the Council on the
patentability of computer-implemented inventions, Brussels: Commission of the
European Communities.
--- (2003), Proposal for a Directive of the European Parliament and of The Council on
measures and procedures to ensure the enforcement of intellectual property rights,
Brussels: Commission of the European Communities.
Diani, M., and D. McAdam, eds (2003), Social Movements and Networks. Relational
Approaches to Collective Action, Oxford: Oxford Univ. Press.
Eising, R. (2004), 'Multilevel Governance and Business Interests in the European Union',
Governance: An International Journal of Policy, Administration, and Institutions 17(2): 211-
245.
Eising, R. and Kohler-Koch, B. (1994), 'Inflation und Zerfaserung: Trends der
Interessenvermittlung in der Europäischen Gemeinschaft', in W. Streeck (ed.), Staat und
Verbände, Sonderheft 25 der Politischen Vierteljahresschrift, Opladen: Westdeutscher
Verlag, pp. 176-206.
Gamson, W.A. (1992), 'The Social Psychology of Collective Action', in A.D. Morris and C.M.
Mueller (eds.), Frontiers in social movement theory, New Haven, CT: Yal e University,
pp. 53-76.
Grande, E. (1996), 'The State and Interest Groups in a Framework of Multi-Level Decision-
Making: the Case of the European Union', Journal of European Public Policy 3(3): 318-
358.
Granovetter, M.S. (1973), 'The Strength of Weak Ties', The American Journal of Sociology
78(6): 1360-1380.
Greenwood, J. (2003), Interest Representation in the European Union, Houndmills: Palgrave
Macmillan.
Haunss, S. (2004), Identität in Bewegung. Prozesse kollektiver Identität bei den Autonomen
und in der Schwulenbewegung, Wiesbaden: VS Verlag für Sozialwissenschaften.
Haunss, S. and Kohlmorgen, L. (2008), 'Codebook for the analysis of political claims in
conflicts on intellectual property rights in Europe', http://www.ipgovernance.eu/
publications/IPGovCodebook.pdf.
--- (2009), 'Lobbying or Politics? Political claims-making in IP conflicts', in S. Haunss and
K.C. Shadlen (eds.), The Politics of Intellectual Property, Cheltenham, Glos: Edward
Elgar Publishing, pp. 107-128.
21
Hooghe, L. and Marks, G. (2001), Multi-level governance and European integration,
Governance in Europe, Lanham, Md. [u.a.]: Rowman & Littlefield.
Kenis, P. and Schneider, V. (1991), 'Policy Networks and Policy Analysis: Scutinizing a New
Analytical Toolbox', in B. Marin and R. Mayntz (eds.), Policy Networks - Empirical
Evidence and Theoretical Considerations, Frankfurt, pp. 25-59.
Kohler-Koch, B. (1996), 'Catching Up with Change: the Transformatione of Governance in
the European Union', Journal of European Public Policy 3(3): 359-380.
--- (1997), 'Organized Interests in the EC and the European Parliament', European Integration
Online Papers 1(9), http://ssrn.com/abstract=302669.
--- (2002), 'European Networks and Ideas: Changing National Policies?', European
Integration online Papers (EIoP) 6(6).
Koopmans, R. and Statham, P. (1999), 'Political Claims Analysis: Integrating Protest Event
And Political Discourse Approaches', Mobilization 4(2): 203-22.
Kriesi, H. (1995), 'The Political Opportunity Structure of New Social Movements: Its Impact
on Their Mobilization', in J.C. Jenkins and B. Klandermans (eds.), The Politics of
Social Protest, Minneapolis, MN: University of Minnesota Press, pp. 167-198.
Kriesi, H., Adam, S. and Jochum, M. (2006), 'Comparative analysis of policy networks in
Western Europe', Journal of European Public Policy 13(3): 341-361.
McAdam, D. (1996), 'Conceptual Origins, Current Problems, Future Directions', in D.
McAdam, J.D. McCarthy, and M.N. Zald (eds.), Comparative Perspectives on Social
Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings,
Cambridge: Cambridge U Press, pp. 23-40.
--- (2003), 'Beyond Structural Analysis: Toward a More Dynamic Understanding of Social
Movements', in M. Diani and D. McAdam (eds.), Social Movements and Networks,
Oxford: Oxford Univ. Press, pp. 281-298.
Melucci, A. (1996), Challenging Codes. Collective Action in the Information Age, Cambridge:
Cambridge Univ. Press.
Olson, M. (1968), The Logic of Collective Action: Public Goods and the Theory of Groups,
New York: Schocken Books.
Rhodes, R.A. (1997), Understanding Governance: Policy Networks, Governance, Reflexivity
and Accountability, Buckingham: Open University Press.
Richardson, J. (2000), 'Government, Interest Groups and Policy Change', Political Studies
48(5): 1006-1025.
Scharpf, F.W. (2002), 'Regieren im europäischen Mehrebenensystem. Ansätze zu einer
Theorie', Leviathan 30(1): 65-92.
Snow, D.A. (2004), 'Framing Processes, Ideology, and Discursive Fields', in D.A. Snow, S.A.
Soule, and H. Kriesi (eds.), The Blackwell companion to social movements, Blackwell
companions to sociology, Oxford: Blackwell, pp. 380-412.
Snow, D.A., Rochford Jr., B.E., Worden, S.K. and Benford, R.D. (1986), 'Frame Alignment
Processes, Micromobilization and Movement Participation', American Sociological
Revue 51(4): 464-481.
Tarrow, S. (1994), Power in Movement. Social Movements, Collective Action and Politics,
New York: Cambridge University Press.
Tilly, C. (1978), From Mobilization to Revolution, Reading, Mass.: Addison-Wesley.
Warleigh, A. (2000), 'The hustle: citizenship practice, NGOs and 'policy coalitions' in the
European Union - the cases of Auto Oil, drinking water and unit pricing', Journal of
European Public Policy 7(2): 229-243.
22
Wasserman, S. and Faust, K.L. (1994), Social Network Analysis: Methods and Applications,
New York: Cambridge University Press.
Wellman, B. (1988), 'Structural Analysis: From Method and Metaphor to Theory and
Substance', in B. We llm an and S.D. Berkowitz (eds.), Social Structures: A Network
Approach, Cambridge: Cambridge University Press, pp. 9-61.
23
