Journal of European Public Policy

ISSN: 1350-1763 (Print) 1466-4429 (Online) Journal homepage: https://www.tandfonline.com/loi/rjpp20

Governance vs. politics: the European Union's

constitutive ‘democratic deficit’

Myrto Tsakatika

To cite this article: Myrto Tsakatika (2007) Governance vs. politics: the European Union's
constitutive ‘democratic deficit’, Journal of European Public Policy, 14:6, 867-885, DOI:
10.1080/13501760701497840

To link to this article: https://doi.org/10.1080/13501760701497840

Published online: 10 Sep 2007.

Submit your article to this journal

Article views: 2313

View related articles

Citing articles: 18 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rjpp20
Journal of European Public Policy 14:6 September 2007: 867 – 885

Governance vs. politics: the

European Union’s constitutive
‘democratic deficit’
Myrto Tsakatika

ABSTRACT Governance practices fail key democratic tests coming from interest
aggregation, deliberative and agonistic theories of democracy. It is argued that the
European Union is best described as a complex web of governance practices and,
therefore, that its ‘democratic deficit’ is constitutive. In contrast to what is often
thought, it is not by strengthening ‘good governance’ practices but by treading par-
allel paths of constitutionalization and politicization that the European Union could
eventually close the democratic gap.
KEY WORDS Agonism; democratic deficit; European Union; governance;
politicization.

INTRODUCTION
The early 1990s witnessed the transformation of the citizens’ ‘permissive
consensus’ that sustained élite decisions in the direction of European integration
into a ‘constraining dissensus’ (Hooghe and Marks 2005). Its latest and to date
gravest manifestation has been the rejection of the European Constitution in
popular referenda held in two of the Union’s founding member states: France
and the Netherlands. A persistent phantom, long familiar to observers of Euro-
pean Union (EU) politics, is once again casting its shadow over the continent:
the ‘democratic deficit’.
It is often thought that addressing the ‘democratic deficit’ is a question of
appropriate institutional engineering. If we agree on a democratic constitution,
if we further strengthen the European Parliament, if the Commission is better
held under check, if the Council is more transparent and national parliaments
more involved in the political process at EU level, the problem is bound to dis-
appear. On the other hand, there is the view that the EU cannot be democratic
because it lacks essential preconditions of democracy. There is no European
‘demos’ (Weiler 1995). The Union is not a sovereign state, but a union of sover-
eign states whose component parts are the only entities with legitimacy to make
and enforce democratic decisions. In the absence of a pan-European demos,
Journal of European Public Policy
ISSN 1350-1763 print; 1466-4429 online # 2007 Taylor & Francis
http:==www.tandf.co.uk/journals
DOI: 10.1080/13501760701497840
868 Journal of European Public Policy
institutional engineering will be pointless and the ‘democratic deficit’ a peren-
nial feature of European public life. A third cut on the ‘democratic deficit’ is
that if the EU is to be evaluated from the democratic point of view, such assess-
ment must involve non-statist democratic canons (Banchoff and Smith 1999;
Eriksen and Fossum 2000: 6– 9). The EU is, notoriously, neither a state nor
an international organization; it is a sui generis political entity. Consequently,
we should revise our democratic theories in order to set up novel democratic
tests appropriate for the evaluation of the kind of entity that the EU is.
In what follows the claim will be that the Union is not one of a kind. Rather,
it is best described as a complex web of governance practices. If we must revise
our theories of democracy in order to be able to capture a changing political
reality, this is the reality of a world populated with governance practices. It
will be argued that the EU is faced with a constitutive ‘democratic deficit’,
not by virtue of the fact that it is badly constructed in institutional terms or lin-
guistically and culturally diverse, but by virtue of the fact that it is a system of
governance. From the perspective of our revised theories of democracy, govern-
ance as such is argued to fail the democratic test. Therefore, if the ‘democratic
deficit’ is to be addressed, what needs to change is the EU’s governance
texture. Institutional engineering will be necessary but can only be part of the
solution. Another equally significant part will be the gradual development of
practices of political contestation, the sine qua non for the emergence of a
pan-European demos.

THE EUROPEAN UNION AS A WEB OF GOVERNANCE

PRACTICES
The fascinating question about the EU is not whether it is sui generis or not,
since all political entities, regardless of their size, the homogeneity of their popu-
lations or the extent to which they can claim sovereignty are in a strict sense
unique; the question is what the EU can reasonably be compared to. In what
follows the EU will be presented as a web of diverse governance practices,
comparable to other practices or constellations of practices of governance.
The term ‘governance’ is increasingly being used to denote a specific type of
organization that is on the ascendant in response to a historical trend. The state,
as a historical form of political organization, seems to be losing its claim to the
monopoly of ‘governing’ (Hirst and Thompson 1999: 269). This is the result of
the twin processes of internationalization (or globalization) and functional
differentiation that are ‘eroding’ the nineteenth-century state as to its sover-
eignty, as to its exclusive control over territory and population, usurping func-
tions that it once exclusively performed.
One key characteristic of ‘governance’ is complexity: it does not involve a
clearly delineated top-down exercise of the functions of governing by the
state, but power-sharing in the context of complex patterns of multi-directional
interaction between diverse actors, public and private, operating at subnational,
 M. Tsakatika: Governance vs. politics 869
national, transnational and supranational level, among which the state is but
one player.
It is a second characteristic of governance, which is due to the complexity and
functional specialization of the issues it is called to deal with, that it requires
expertise rather than statesmanship or the layperson’s political judgement
alone. The type of participant involved in governance processes will therefore
not be the average citizen or politician, but the expert: an independent expert
or a qualified official or representative of a specialized governmental or corporate
actor, group, or committee.
Third, being a joint undertaking of actors that operate in the absence of hier-
archy, governance is directed towards the achievement of consensus on solutions
that can be acceptable to all or most participants and thus inevitably involves
locking out disagreements on issues that all or most participants would rather
resolve on their own. The stress, in other words, is on reaching consensus,
which requires that only certain types of issue be included in the range of
common action.
Finally, insofar as the state participates as one actor among others, responsi-
bility for governance outcomes is shared with other actors whose legitimacy to
exercise governance functions is dependent upon the results they are able to
achieve, rather than upon the appropriateness of the processes of authorization
that they were subject to in order to exercise these functions. Efficiency thus
becomes the primary concern in the context of ‘governance’.
Despite it becoming a ‘catchword’ only in the 1990s, ‘governance’, under-
stood as a complex, expert, consensus and efficiency oriented practice, has
been part and parcel of the EU from the outset. What was the Community
Method if not an ingenious innovation in governance, involving regular pat-
terns of interaction at multiple levels and horizontal interactions between differ-
ent public and private specialized actors, whose aim was to consensually exercise
governing functions efficiently?
In the course of European integration, governance practices multiplied and
diversified, in response to the different needs of policy areas, the growing
body of work, the proliferation of actors and successive enlargements. In
2002, Scott and Trubek identified no less than six different modes of governance
apart from the classical community method itself, three of which they con-
sidered variations of the Community Method: Community Method with
greater use of framework directives, extended use of comitology, or increasing
civil society involvement in the policy-making process, and the other three
radical departures from it: partnership, social dialogue and the open method
of co-ordination (Scott and Trubek 2002). The EU is in this sense best
thought of as a web of densely intertwined governance practices.
This is not to say that there are no ‘government’ aspects to it. Pointing to the
EU’s ‘multi-level’ governance character, Jachtenfuchs has convincingly argued
that ‘the EU includes its member states in an encompassing system while at
the same time, national political, economic, or legal systems continue to
exist’ (Jachtenfuchs 1997: 40). Yet, on the one hand, the governance mode is
870 Journal of European Public Policy
increasingly gaining ground within member states through the proliferation of
administrative practices such as new public management (Pollitt and Bouckaert
2000). On the other hand, ‘(t)he parallel existence of “governance” and “govern-
ment” gives rise to conflicts between different logics of action’ (Benz and Papa-
dopoulos 2006: 15). Such conflicts may well lead to a gradual strengthening of
the ‘governance’ logic with respect to ‘government’ in the EU.
When it comes to examining the EU from the democratic point of view, its
make-up as a web of governance practices is of the utmost importance. Can gov-
ernance practices be democratic? Philippe Schmitter, among others, answers the
question in the affirmative: governance can be democratized. To this end he has
elaborated detailed proposals (Schmitter 2001). The thesis set out in what
follows directly challenges this view: governance cannot be democratic, therefore
the ‘democratic deficit’ will not be remedied as long as the EU remains primarily
a system of governance.
Before turning to directly address the question, it is important to clarify what
‘democratic’ is to refer to here. As much as we need to revise our democratic
theories in order to take into account the emergence and proliferation of govern-
ance practices such as those that compose the EU and to construct appropriate
democratic benchmarks in order to assess such practices, there is one qualifica-
tion that must be put forward: our revised theories must still be theories of
democracy. The question here is the following: what is the bottom line of demo-
cratic rule? What can we safely subtract from any theory of democracy before it
can no longer be considered a theory of democracy? What, on the contrary, is it
essential to hold on to?
We can envisage a democracy without a people, in the sense of persons
sharing a common ethnic, cultural, linguistic or religious identity (Habermas
1996). We can conceive of the possibility of a democracy without a community
of rights-bearers, in the sense of persons bearing constitutionally entrenched
civil and political rights (Taylor 2000). Yet, we cannot conceptualize democracy
in the absence of a demos, that is, persons that decide on equal terms about those
affairs they have come to consider of common concern. Further, we can envisage
a democracy without a state, unitary, federal or confederal; or a democracy
without authority to decide on the full range of common concerns. However,
we cannot conceive of a democracy that does not involve a mechanism
through which the demos can arrive at and enforce joint decisions on matters
of common concern.
What we end up with is quite familiar: in the context of the nation-state it
was called popular sovereignty. In the context of governance practices such as
those that make up the EU, the equivalent of popular sovereignty is collective
self-rule. ‘Collective’ refers to the demos, understood in the minimal sense
described above. ‘Self-rule’ need not be as pervasive or watertight as sovereignty,
but it does require that there is a binding process of decision-making by a
demos. The bottom line of democratic rule then, what distinguishes democracy
from other forms of political regime, is collective self-rule, even in the case of
governance practices.
 M. Tsakatika: Governance vs. politics 871
GOVERNANCE AND DEMOCRACY
Contemporary theories of democracy attribute the utmost importance to collec-
tive self-rule, despite the fact that they divide radically on what exactly collective
self-rule entails. In this section we will focus on three leading contemporary the-
ories of democracy, attempting to spell out the conception of collective self-rule
that they incorporate. Governance will then be examined from the point of view
of the three theories in order to draw conclusions about the extent to which it
lives up to their standards of collective self-rule.
A pertinent distinction has been drawn by Chantal Mouffe between (a) the-
ories that conceptualize democracy as a mechanism of interest aggregation; (b)
theories of ‘deliberative’ democracy according to which democratic decision is
about adopting solutions that all can reasonably accept, emerging from a
process of fair and equal public deliberation; and (c) theories of ‘agonistic’
democracy where democratic practice is considered an open contest between
different groups of active citizens who assert different values and express differ-
ent political identities (Mouffe 2000).
According to the aggregative conception, democracy is no longer about the
mobilization of citizens towards consensus on the common good, but about
aggregating individual preferences in what is now a pluralistic, morally
diverse civic universe. In democracies, such aggregation of interests principally
takes place through voting in competitive elections (Dahl 1972), with political
parties being the central actors in the process. Citizens, moved by the rational
pursuit of their own interest, choose the party that, in their judgement, best rep-
resents that interest in the context of the electoral process. Competition for
office-holding is in this view a fundamental condition of democracy (Dahl
1972) as office-seeking politicians compete for the people’s vote claiming to
be the best representatives of their interest (Downs 1957). Clear and under-
standable accountability chains are necessary if voters are to identify and evalu-
ate the performance of political parties in government and to be in a position to
‘punish’ those whose performance they judge inadequate (Przeworski et al
1999). Between elections citizens are to stay away from the political process,
as ‘back seat driving’ can be ineffective and dangerous (Schumpeter 1947).
Collective self-rule is thus principally manifested in competitive elections, and
consists in the power of the people to ‘throw the rascals out’.
For deliberative democrats, democracy is about more than aggregating inter-
ests: the moral standing of political procedures and outcomes should be a top
democratic concern. Consensus on substantive moral grounds can be reached,
despite the fact that pluralism is indeed a fundamental sociological condition of
contemporary societies. According to Jurgen Habermas, attempts to reach rational
consensus will only bear fruit if conducted in the context of a process of equal and
fair public deliberation, through argument and reason-giving (Habermas 1992).
A democratic constitution guaranteeing rights to private and public autonomy,
a political culture embodying democratic values, an autonomous public sphere
and a process whereby deliberation in the public sphere feeds into the
872 Journal of European Public Policy
deliberations of representative institutions (parliament, administration, judiciary)
are the context in which democratic deliberation is to operate (Habermas 1996).
While deliberative theorists have provided different versions of what fair and equal
public deliberation entails (Bohman 2000: 34–7; Dryzek 1994: 15; Cohen and
Sabel 1997; Habermas 1996: 307–8; Eriksen 2000: 52), three conditions may
be singled out as common to most accounts:
1. that all affected parties can participate effectively and contribute to the
agenda;
2. that weight is given to the merit of the argument rather than to the relative
power of participants;
3. that deliberation is public, meaning public-regarding (not appealing to
partial interests), addressed to the public (to all rational agents rather than
only to those physically present) and open (not conducted in secret).
The deliberative conception of collective self-rule is ‘de-substantialized’ and
‘intersubjective’. It is understood in terms of the ‘communicatively generated
power’ emerging through deliberation in independent public spheres, which
is filtered through the complex of representative democratic institutions, to ulti-
mately exert influence upon public policy decisions (Habermas 1996: 301– 2).
Such decisions are taken with several other considerations in mind, efficiency
being a significant one among them.
While agreeing with deliberative democrats that democracy should be con-
sidered as something more than a mechanism of interest aggregation, agonists
consider that the emphasis of deliberative theories on reaching substantive con-
sensus means not facing up to the realities of value pluralism (Mouffe 2000). It
is certainly true that deliberative democrats consider conflict as one of the very
reasons to deliberate in the first place. It is also true that their account of the
topics that can be discussed in public deliberations does not exclude ‘private’
or ‘pre-political’ issues ex ante, insofar as they are of common concern
(Habermas 1996: 308). The problem, from the agonists’ point of view, is
deliberative democrats’ belief that conflict over such issues can be dealt with
in a (communicatively) rational and impartial way.
Agonists consider this approach not only misguided, but ultimately also
dangerous for democracy itself. Misguided because citizens are constituted by,
and cannot be abstracted from, their position in social and political power
relations, language, cultural or other group belongings, which, given value plur-
alism, means that conflict cannot be eliminated through rational deliberation.
To claim that conflict has been eliminated would be to overlook or conceal
the particular power distribution or established hegemonic project that has
temporarily crystallized (Wolin 1996). Dangerous because the emphasis on
consensus may discourage citizen participation (Honig 1993) and in so doing
direct ‘collective passions’ away from contestation around different projects
for the common good, i.e. liberal-conservative, social-democratic, green,
radical democratic, etc., and towards forms of contestation that cannot be
managed through democratic practice, i.e. identity politics (Mouffe 2000).
 M. Tsakatika: Governance vs. politics 873
Democracy must be seen not as an attempt to settle conflict but as an ongoing
practice meant to ‘domesticate’ conflict by transforming antagonism (conflict
between enemies which is the result of irreconcilable pluralism) into agonism
(civil contestation between political adversaries) (Mouffe 2000). Agonistic
democratic politics are in this sense directed towards disagreement, as
opposed to both aggregative and deliberative democratic politics, which are
‘irenic’, that is, directed towards agreement (Roberts-Miller 2004).1 Democracy
requires the availability of ‘institutions, discourses and forms of life’ that allow
for agonistic, active citizen participation in democratic politics, which is the
means by which citizens grow attached to the democratic process and demo-
cratic values (Mouffe 2000). Nancy Fraser has argued that apart from the ‘domi-
nant publics’ such spaces are to be found in ‘subaltern counterpublics’, where
disadvantaged or excluded groups can express identities and articulate claims,
as well as demand that their claims are taken into account (Fraser 1992). For
agonists, collective self-rule is citizen identification with, and ownership of,
the democratic process.

Competition for office and accountability

Interest aggregation democrats will be concerned with the fact that the complex-
ity of governance enhances the obscurity of decision-making and implemen-
tation, obstructs the identification of responsibility and renders accountability
difficult. Advocates of governance may reply that despite their complexity, or
rather, because of it, governance practices will ensure greater accountability
than traditional ‘government’- type accountability mechanisms, albeit by differ-
ent means. Governance processes involve multiple actors, whose constant
horizontal exchanges result in greater dissemination of information and
transparency. Information and transparency are essential to the effective exercise
of accountability, on behalf of those meant to exercise it. However, the aggrega-
tive stress on accountability is due to the opportunity it provides for the people
to sanction those in power. Governance practices will still tend not to provide
citizens at large with the opportunity to sanction the élite that makes
common decisions and thus it presents that élite with no credible threat that
it may somehow be ‘punished’ for its choices.
Interest aggregation democrats will rejoice in the fact that it is specialists
rather than citizens at large who call the shots in governance practices. This is
because they are highly sceptical when it comes to ordinary citizens’ capacities
and judgement for the governing of common affairs.
On the other hand, being consensus-oriented, governance practices do not
involve élite competition for office: at first blush, there is no ‘office’ to compete
for. Rather, the ‘office’ is shared between actors concerned with arriving at a
consensus. Advocates of governance may counter-argue that governance practices
incorporate better checks to power exercised horizontally among participants than
those offered through traditional party competition for office. This is because the
check is exercised continuously rather than exceptionally, as in the case of periodic
874 Journal of European Public Policy
elections, which means that those exercising it have the opportunity to develop
greater thoughtfulness and skill in carrying it out; it avoids the rigidity of the
process of setting out competing electoral manifestos and programmes, but is flex-
ible and therefore more adaptable to changing circumstances; it is not centralized
and winner-takes-all, but decentralized and mutual, which means that power will
tend to be dispersed rather than exercised in the process. These counter-arguments
will not impress interest aggregation democrats. From the aggregative point of
view, in the absence of élite competition for office there is but one élite that
runs governance processes co-operatively. In other words, party competition is
necessary for democracy because it alone ensures that the interests of all rather
than those of one élite (plural but ultimately united in a co-operative enterprise)
will be served.
Efficiency, as the search for best solutions and a key aim of governance, will sit
in well with aggregative concerns because democracy is thought of exactly as a
mechanism for arriving at optimal outcomes through the aggregation of
interests.
It seems that two of the principal conditions of collective self-rule for interest
aggregation democrats, that is, élite competition for office and the opportunity
for the people to vote élites out of office in elections, are simply not a part of
governance processes.

Fair and equal public deliberation

Deliberative theorists may prima facie welcome the complexity of the governance
form: it entails involving a greater range of actors. It is therefore likely to be more
inclusive of diverse stakeholders and their concerns. It also entails non-hierarchy,
which is bound to render governance a set-up more conducive to argument and
deliberation rather than power and bargaining. Yet, there are at least two ways in
which the power of the actor, rather than the strength of the argument, is bound
to matter in governance practices. Deliberation presupposes equal rights of
participation. But, the actors that participate in governance practices are more
likely than not to be institutionally endowed with different weights as to their
say in policy-making. Let us not forget that institutions will normally matter;
‘where you stand depends on where you sit’ (Allison 1971: 176) and where
you sit is important when it comes to the influence you are able to exercise.
Moreover, effective participation in governance practices requires expertise,
which means that the layman will find himself at a disadvantage with respect
to the expert, both to participate and to add items to the agenda.
Expertise is not a feature of governance that deliberative democrats are likely to
prize: participants in deliberation are not meant to be experts, but persons who are
able to put forward public-regarding arguments. Expertise will therefore put unwel-
come limits on who can be included. Moreover, expertise is bound to violate delib-
erativists’ publicity condition. Governance practices may indeed provide a space for
public deliberation, in the sense of it appealing to public-regarding reasons and
addressing itself to all rational members of the public. However, this is a space
 M. Tsakatika: Governance vs. politics 875
that is by and large isolated from the public eye; it mainly consists in deliberation
between experts, part of whose proceedings and the full range of whose results
may be made public, but which are not accessible to the public at large.
The consensus-orientation of governance practices will sit in well with delib-
erative democrats. Less so the priority governance places on efficiency. For delib-
erative democrats efficiency is important, but only to be balanced with
considerations of collective self-rule.
Insofar as governance practices are not inclusive, do not ensure that power
will not matter and shun the publicity condition, ‘communicatively generated
power’ is unlikely to result. What is more, deliberation is meant to take place
both in the autonomous public sphere, whose locus is civil society, and in the
context of representative institutions. Deliberative democrats insist on the sep-
aration of opinion-formation in the public sphere and will-formation in the
representative decision-making institutions. It is free flowing communication
between the two that makes for collective self-rule. Governance practices are
bad news for the autonomy of the public sphere. In effect they involve a
fusion of civil society and government, the risk being that organized civil
society will be co-opted by the representative institutions. All things considered,
governance does not fully live up to the deliberative democratic challenge either.

Political contestation and participation

Agonists will welcome neither the complexity nor the expert character of gov-
ernance, as they both discourage direct citizen participation. Complexity
makes the process difficult to follow for the non-specialist, non-insider. Agonists
will not welcome the type of participation that ‘governance’ requires for the
exact same reasons that interest aggregation democrats will: the expertise
required of governance participants makes demanding claims on participants’
knowledge and skills and thus relies upon participation of the enlightened few.
Classical objections from élitism can be raised against agonists at this point:
left to their own devices, the masses are not only inefficient and irresponsible
decision-makers, but also dangerous for the survival prospects of democratic
rule. Demagogy, scapegoating and short-termism are known to characterize
the politics of mass democratic publics, often to the detriment of individual
and minority rights and ultimately of democracy itself. Governance processes
allow for sober consideration of complex and delicate policy choices, peaceful
and inclusive mediation between diverse interests and ensure that the rights of
individuals and the concerns of minorities are protected in the process.
Agonists can respond with the equally old argument going back to Rousseau
and J.S. Mill: the people learn about the issues at stake, at least what they need
to know in order to articulate and express their preferences and learn to exercise
their judgement through participation in democratic politics. Excluded from the
possibility of exercising political judgement, they will remain inept and inexperi-
enced judges of the common good and never develop their citizenship potential
(Pateman 1970). Governance practices effectively deprive citizens of the
876 Journal of European Public Policy
possibility of developing their citizenship potential by eliminating the possibility
for them to participate actively in democratic decision-making.
Agonists will object above all to the consensus orientation of governance,
which strikes at the core of agonistic democratic politics, given that it obscures
differences between political alternatives, de-politicizes choice and thus
discourages political contestation. An objection that may be raised against
democratic agonistic conflict, particularly in culturally and politically diverse
political settings, is that political contestation has the tendency to polarize
and ‘freeze’ identities and thus create centrifugal tendencies in the democratic
realm (Dryzek 2005). Governance practices defuse disaggregating clashes such
as those linked to national identities. Agonists will reply that ethno-nationalistic
conflicts find space to emerge in the first place when political contestation
around values and political identities is muted and power relations concealed
under the efficiency cloak. Governance practices are likely to lead to unmanage-
able polarization between culturally diverse groups in the long run, whereas
overt politicization could work as a centripetal demos-creating unifying force
for the common democratic space.
The efficiency orientation towards ‘best’ solutions will also be criticized, given
that for agonists there can be no ‘best’ solution in politics: any solution is bound
to reflect the temporary dominance of particular groups and will only be ‘best’
for them. Given that governance practices discourage both political partici-
pation and political contestation, they undermine popular ownership of the
political process and are therefore democratically unsound from the agonistic
point of view. Table 1 summarizes the argument set out in this section.

Table 1 Governance and theories of democracy

Deliberative
Governance Interest aggregation democracy Agonism

Complexity Negative, lack of Positive, more Negative,

accountability inclusive, discourages
conducive to participation
deliberation
Expertise Positive, on grounds Negative, Negative,
of élitism deliberation disallows
not generally effective
accessible participation
Consensus- Negative, no Positive, the aim Negative,
orientation competition for being ‘rational de-politicization
office consensus’
Efficiency- Positive as the aim Negative, not Negative insofar
orientation of democracy is balanced with as it conceals
optimal outcomes democratic choices on
considerations values
 M. Tsakatika: Governance vs. politics 877
THREE PATHS TO DEMOCRATIZATION
If the EU can be conceptualized as a complex web of governance practices and
if governance fails to live up to all three tests of collective self-rule, the EU
suffers a constitutive ‘democratic deficit’. Aggregative, deliberative and agon-
istic assumptions can in fact be distinguished in the three most widespread
versions of the EU’s ‘democratic deficit’: citizens cannot effectively remove
EU-level representatives from office through elections, given lack of party com-
petition and given the inadequacy of accountability mechanisms; the EU is not
endowed with an independent public sphere where civil society can engage in
free and equal public deliberation which will then in turn inform political
debate in EU institutions and ultimately the policy-making process; the EU
is ‘far from its citizens’, in the sense that citizens do not understand or identify
with and therefore feel that they have no power over European democratic
politics. Constitutionalization, ‘good governance’ and politicization, three
distinct strategies that have been put forward as appropriate to address the
‘democratic deficit’, can be viewed as responses to the three different under-
standings of the ‘democratic deficit’, that is, lack of accountability, delibera-
tion, or direct participation.
The first strategy for addressing the ‘democratic deficit’, whose prospects,
however, seem quite bleak at present, was recently adopted by pro-federalist
forces in the European Parliament, Germany, and several of the Union’s
smaller member states, when advocating the idea of calling an EU-wide Consti-
tutional Convention and establishing a Constitution for the EU.
The Constitutional Treaty was meant to address the question of simplifica-
tion and clarification of roles in the structures of EU governance (della Sala
2001). One of its main purposes is to systematize and delineate the tasks of
the European institutions; to proceed towards a clearer allocation of competen-
cies of the European and national levels of government in the different policy
areas; to simplify decision-making procedures and legislative instruments. In
so doing, the aim was to render policy-making more transparent and under-
standable and the system more accountable to European voters. Clearly, a
focus on simplifying accountability chains is central to the concerns of interest
aggregation democrats whose main requirement in terms of collective self-rule is
the possibility for voters to oust their representatives from office through the
vote. The Constitutional Treaty also formalized steps that had gradually been
taken in the direction of rendering the Commission Presidency an office that
could be the identifiable prize of party competition: by linking the choice of
Commission President to the outcome of European elections and by strength-
ening the accountability relationship of the Commission to the European
Parliament.
The Constitutional Treaty also incorporated the EU Charter of Basic Rights.
The entrenchment of basic rights in the EU’s constitutional regime can be
thought of as a step in the direction of an EU-wide constitutional democracy,
of the kind that is a precondition for the emergence of an EU-wide independent
878 Journal of European Public Policy
public sphere. Moreover, the Treaty was arrived at through a new method that
opened up a deliberative space at the constitutional level. The convention
method was inclusive, as it was open not only to the designated representatives
of governments, parliaments and EU institutions, but also to civil society organ-
izations; it involved deliberation where power differences were set aside to some
extent (Bellamy and Schonlau 2004), while the procedure was indeed public,
that is, substantively public-regarding, accessible to the non-specialist and
open. Clearly these are positive developments for the deliberative democrat.
A second strategy to address the ‘democratic deficit’ consists in gradually but
steadily strengthening ‘good governance’ practices in the EU. The European
Commission, seconded by ‘third way’ political forces, academics and civil
society organizations, advanced ‘good governance’ arguments to that purpose,
the apex of which was the European Commission’s ‘White Paper on Govern-
ance’, published in 2001 (COM WP 2001). Good governance in Commission
jargon is said to aim at enhancing the legitimate exercise of EU governance
through the institutionalized engagement of policy-makers with civil society
organizations such as the social partners, non-governmental organizations,
human rights organizations, churches, and other groups. Civil society is here
conceived ‘non-politically, as spontaneous, private and prior to politics’
(Hirst 1996: 99). It is not by chance that political parties were not included
among the privileged civil society associations that the Commission expresses
the desire to consult in its White Paper (Steinberg 2001).
The ‘good governance’ strategy clearly maps on to deliberative democratic
concerns with the quality of democratic procedures and the availability of
spaces, institutions and practices of deliberation, the inclusion of a multitude
of diverse views representing those with a stake in the process, and the possibility
of arriving at consensual agreements in the context of fair and equal public
deliberative procedures.
A third response to the ‘democratic deficit’ seems to be on the rise in the EU.
National political parties and social movements, coalitions of political forces at
European and national levels, and lately, groups of citizens mobilized around
opposite camps in referenda for the ratification of the European Constitution,
may be driving forward an EU politicization trend that has only recently
begun to gradually develop (Fossum and Trenz 2005: 3). ‘Agonistic’ politics
have so far been a property of national constituencies. At the EU level, politics
have mostly been ‘irenic’, consensual and efficiency oriented, at least up until the
1990s. The novel trend seems to be developing at the European level around the
two main political cleavages that characterize EU party politics: Left– Right and
GAL/TAN as Hooge and Marks have recently dubbed the distinction between
political forces that are Green/Alternative/Libertarian and Traditionalist/
Authoritarian/Nationalist (Hooge and Marks 2005).
The solution to the ‘democratic deficit’ is here more politics that involve an
active citizenry (Magnette 2003). Assumptions are agonistic: political contesta-
tion and, through it, more direct identification of citizens with distinct EU-level
political alternatives addressing European-level concerns; therefore, more
 M. Tsakatika: Governance vs. politics 879
ownership of the political process by citizens. Claims for building up more
agonistic political confrontation at the EU level are articulated in terms of citi-
zens’ being able to express values, promote projects, assert political identities and
take part in decisions over issues like the basic principles underlying economic
and social policy at EU level, the degree of homogenization in policy areas where
it is felt that traditional values are threatened, the openness or closure of
European societies to immigration flows. Pan-European referenda, petitions,
citizen legislative initiatives, but also campaigns, strikes and protests organized
at EU level (Imig and Tarrow 2000), are the types of instrument that the
politicization strategy adopts.
At this point we can bring our previous discussion to bear upon our assess-
ment of alternative democratization strategies. Taking the argument set out in
the previous sections into account, the ‘good governance’ strategy must be
rejected. Advocates of ‘good governance’ build on existing governance practices
and argue for their strengthening and improvement in the direction of more
deliberation, more inclusiveness and more transparency. Yet it has been
argued that governance practices constitutively work against public deliberation
of the kind that generates ‘communicative power’. ‘Good governance’ stands on
unsuitable premisses and will be as undesirable from the deliberative democratic
point of view as any form of governance. This is certainly not to say that fair and
equal public deliberation is not to be considered a valuable part of democratic
decision-making; but it is to say that ‘good governance’ cannot involve fair and
equal public deliberation and therefore that the strategy of strengthening ‘good
governance’ practices in the EU is not the solution to the ‘democratic deficit’.
Attention must rather be turned towards the other two ways of addressing the
‘democratic deficit’: constitutionalism and politicization. It has recently been
argued that politicization and constitutionalization cannot be pursued at the
same time, because politicization is emerging as an obstacle to European inte-
gration and therefore to Europe’s spill-over laden path towards polity-hood
(Hooghe and Marks 2005). Yet, there is no good reason to think of
constitutionalization as necessarily leading to more European integration; nor
is it fair to equate politicization with Euro-scepticism. Constitutionalization
may lead to less rather than more integration in the sense of transfer of compe-
tencies from the national to the European level; it may involve transfer of com-
petencies to the EU level, but it may also involve re-nationalization of
competencies. Politicization means exactly the articulation of concrete political
alternatives that citizens can identify with; Euro-scepticism is likely to be only
one of these alternatives. Agonistic dynamics are developing in more dimensions
than one and will evolve in ways that we cannot predict, just as we cannot
predict the outcomes of national agonistic dynamics. These may favour
giving more, less or different responsibilities to the European level, according
to the occasional European-level interpretation of the common good and the
hegemonic project that prevails.
There seems to be a stronger case in favour of the view that the appropriate way
to go about resolving the ‘democratic deficit’ would be to acknowledge that
880 Journal of European Public Policy
constitutionalism and politicization go hand in hand in the EU (Føllesdal and Hix
2005; Fossum and Trenz 2005). As long as the ground rules of EU decision-
making discourage the identification of citizens with political divisions that they
can understand, identify with, and mobilize around, they cannot effectively
serve the cause of collective self-rule in interest aggregation terms, which
demands the possibility of removing élites from office by popular vote. Citizens
cannot ‘throw the rascals out’ because they know neither who the rascals really
are, nor why they should be removed or returned to office, and may even have
doubts about going to vote in the first place. Constitutional rules that discourage
contestation cannot serve collective self-rule in deliberative terms either: the gen-
eration of ‘communicative power’ goes through public deliberation in the public
sphere on disagreements. When disagreements are muted in the context of the repre-
sentative institutions, they may also fail to emerge in the context of the public
sphere. Deliberation and the construction of a public sphere itself are set back.
On the other hand, without giving citizens the possibility of knowing which
office-holders are to be held accountable and/or of knowing what the controver-
sial issues at stake really are, EU democracy cannot make the most of its emer-
ging agonistic potential: citizens might be alienated from political participation,
perhaps shifting, as Mouffe warns, to ‘ethno-nationalist’ confrontation, or they
might direct their democratic pathos to the wrong target, at the wrong time, for
the wrong reasons.

CONCLUSIONS
We are led to the conclusion that any future European Constitution should be
so articulated as to encourage popular participation and contestation and there-
fore greater popular ownership of the democratic process; at the same time, it
may be the case that EU citizens will be ready to accept a constitution only
when EU-level political agonism has fired up, triggering a process of identifi-
cation of EU citizens with a common EU democratic arena.
Certainly the proponents of aggregative, deliberative and agonistic con-
ceptions of democracy would opt for different versions of a common consti-
tutional settlement. One issue of contention would be the extent to which
direct or representative democracy ought to be privileged in constitutional
design.
In conceptual terms, the distinction between direct and representative
democracy cuts across all three conceptions of democracy discussed here.
Cohen and Sabel argue that direct participatory democracy is conducted by citi-
zens on the substance of policy, in aggregative or deliberative fashion; I would
here add agonistic. Likewise for representative democracy: citizens choose repre-
sentatives who make decisions on substance of policy in legislative assemblies
that operate in aggregative or deliberative fashion (Cohen and Sabel 1997).
Again I would add agonistic, involving groups of representatives engaging in
open political contestation. Nonetheless, it would be fair to argue that interest
aggregation democrats, deliberative democrats and agonists would differ as to
 M. Tsakatika: Governance vs. politics 881
their approach towards institutionalizing forms of respectively direct or repre-
sentative democracy in the European Constitution.
From the point of view of interest aggregation democrats, it would be
important to limit provisions for direct popular participation through pan-
European referenda, petitions and citizen initiatives as much as possible,
given their élitist distrust of citizen involvement and the dangers they see
in ‘back seat driving’. Agonists would oppose, requiring that such measures
be provided for as extensively as possible, given that they would not only
empower citizens at large, but also give expression to and the possibility of
intervening in ‘subaltern counter-publics’. Mid-way between the two, delib-
erative theorists would call for an increased role for the representative insti-
tutions, particularly the European Parliament and the Court and spaces
where deliberation can freely develop along with the coming of age of a
European civil society (Habermas 2001). Among the three there would be
room to discuss which types of decision and which policy areas would be
more appropriate for direct popular involvement and where, on the contrary,
caution would be warranted and more representative forms of democracy are
to be preferred.
A second issue of disagreement might be the extent to which the constitution
should be rigid or flexible. For aggregative democrats it would be important to
fix clear and stable constitutional arrangements in order to ensure accountabil-
ity. From an agonistic point of view, on the other hand, it would be consistent to
require that any constitutional solution for the EU be flexible. For agonists, any
institutional arrangement reflects the dominance of particular interests and
power inequalities; it is therefore best to render the constitution susceptible to
change so as to make it more responsive to shifts in the dialectics of agonistic
contestation (Tully 1999). For deliberative democrats, a constitution should
entrench basic rights and provide necessary safeguards for cultural and linguistic
diversity, preconditions of free and equal deliberation. Yet, they would call for
not too definite a competence catalogue which would be open for revision at
fixed dates, this being the realization of ‘a system of basic rights under changing
historical circumstances’ (Habermas 2001: 23). A discussion would here need to
take place regarding possible ways to reconcile the requirements of accountabil-
ity, the protection of rights and the need to embrace change in line with the per-
ennial open agonistic search that is the political.
With appropriate constitutional arrangements that would provide the insti-
tutional incentives put to one side, which key actors could best provide the cat-
alyst for the articulation and strengthening of agonistic dynamics? Political
parties have traditionally played a crucial role in this respect when it came to
the national context. Yet, as Peter Mair has argued, parties are failing to fulfil
their basic functions: to engage citizens and to serve as the privileged political
space for political leaders (Mair 2005: 8).
Citizen participation and identification with political parties is declining,
which is detrimental to agonistic politics. It is giving up its place to apathy
for the many and specialized, partial and discontinuous engagement for the
882 Journal of European Public Policy
few through associations of ‘civil society’. At the same time, politicians are
increasingly engaging with the techniques of power and policy-making, insulat-
ing themselves from ideological party competition for office and popular
control. In other words, they are increasingly engaging in governance practices
rather than in party politics.
Civil society is an insufficient substitute for party political participation. ‘The
networks of global civil society may be useful in manifold ways, but as a substi-
tute for political communities they tend to eliminate the specific qualities of
politics in favour of its convergence with functionally determined social
relations’ (Thaa 2001: 520). However, it cannot be argued that two types of pol-
itical engagement are necessarily direct opponents in a zero-sum game. As Weale
has argued, party competition simplifies and simplification brings exclusion.
Civil society organizations in many cases bring forth the claims of excluded
groups (Weale 2006). They can and have been complementary components
of political contestation at the national level. There are signs that the develop-
ment of a similar relationship at the European level may be in progress (Wessels
2004). Were civil society organizations to invest a bit more of their efforts and
resources in pushing their claims and concerns up EU-level party agendas and in
public deliberation over how their concerns relate to party political concerns,
and a bit less on lobbying European institutions and participating in governance
practices, their actions would enhance agonistic democratic practice. And were
political parties to keep their doors more open to the concerns, challenges and
expertise of civil society organizations, social movements and ‘subaltern counter-
publics’ there might well be renewal and renewed interest in their activities by
the citizenry at large.
At the EU level the role of transnational political parties is determining the
development of agonistic democratic politics, because they are the only actors
that would be in a position to articulate general European-level political alterna-
tives, thus capturing the imagination of, and engaging, European citizens at
large. This is not only because political parties as types of collective actors are
meant to generate comprehensive political alternatives. It is also because they
would be in a privileged position to creatively reconcile differences in national
political cultures, overcome organizational difficulties, and provide unifying fra-
meworks for national political parties that share common political traditions, by
appealing to common value commitments and articulating principled, clear and
distinct responses to key European and global issues.
Were European-level political parties to assume the responsibility of such
tasks, they would not only become attractive public loci for deliberation intern-
ally and in relation to civil society, but they could also succeed where national
political parties have failed in Mair’s terms, rekindling the kind of agonistic
citizen engagement that is fading in national politics and capturing once
again political leaders’ undivided attention.

Biographical note: Myrto Tsakatika is currently Adjunct Lecturer at the Athens

University of Economics and Business, Greece.
 M. Tsakatika: Governance vs. politics 883
Address for correspondence: Myrto Tsakatika, Department of International
and European Economic Studies (DIEES), Athens University of Economics
and Business, Patision 76, Athens 10434, Greece. email: [email protected]

ACKNOWLEDGEMENTS
The author would like to thank Dimitri A. Sotiropoulos, Euclid Tsakalotos,
Albert Weale and two anonymous referees for detailed comments on previous
drafts.

NOTE
1 Etymologically, ‘agonistic’, agon ¼ contest and ‘irenic’, irene ¼ peace, in Greek.

REFERENCES
Allison, G.T. (1971) Essence of Decision: Explaining the Cuban Missile Crisis, Boston:
Little, Brown & Company.
Banchoff, T. and Smith, M.P. (1999) ‘Introduction: Conceptualizing legitimacy in a
contested polity’, in T. Banchoff and M.P. Smith (eds), Legitimacy and the European
Union, London: Routledge.
Bellamy, R. and Schonlau, J. (2004) ‘The good, the bad and the ugly: the need for con-
stitutional compromise and the drafting of the EU constitution’, in L. Dobson and
A. Follesdal (eds), Political Theory and the European Constitution, London: Routledge.
Benz, A. and Papadopoulos, Y. (eds) (2006) Governance and Democracy. Comparing
National, European and International Experiences, London: Routledge.
Bohman, J. (2000) Public Deliberation: Pluralism, Complexity and Democracy, Cambridge,
MA: MIT Press.
Cohen, J. and Sabel, C. (1997) ‘Directly deliberative polyarchy’, European Law Journal
3(4): 313 – 43.
Commission of the European Communities (2001) European Governance: A White
Paper, COM (2001), 428 final, Brussels, 25 July.
Dahl, R.A. (1972) Polyarchy: Participation and Opposition, New Haven, CT: Yale
University Press.
Della Sala, V. (2001) ‘Constitutionalising governance: democratic dead end or dead on
democracy?’, Constitutionalism Web-Papers, ConWEB, No. 06/2001.
Downs, A. (1957) An Economic Theory of Democracy, New York: Harper.
Dryzek, J.S. (1994) Discursive Democracy: Politics, Policy and Political Science, Cambridge:
Cambridge University Press.
Dryzek, J.S. (2005) ‘Deliberative democracy in divided societies: alternatives to
agonism and analgesia’, Political Theory 33(2): 218 –42.
Eriksen, E.O. (2000) ‘Deliberative supranationalism in the EU’, in E.O. Eriksen and
J.E. Fossum (eds), Democracy in the European Union. Integration Through Delibera-
tion?, London: Routledge.
Eriksen, E.O. and Fossum, J.E. (2000) ‘Post-national integration’, in E.O. Eriksen and
J.E Fossum (eds), Democracy in the European Union. Integration Through Delibera-
tion?, London: Routledge.
Føllesdal, A. and Hix, S. (2005) ‘Why there is a democratic deficit in the EU: a response
to Majone and Moravcsik’, European Governance Papers (EUROGOV) No. C-05-
02. http//:www.connex-network.org/eurogov/pdf/egp-connex-C-05-02.pdf
884 Journal of European Public Policy
Fossum, J.E. and Trenz, H. (2005) ‘The EU’s fledgling society: from deafening silence
to critical voice in European constitution making’, ARENA Working Paper, 19/05.
Fraser, N. (1992) ‘Rethinking the public sphere: a contribution to the critique of actu-
ally existing democracy’, in C. Calhoun (ed.), Habermas and the Public Sphere,
Cambridge,: MA MIT Press.
Habermas, J. (1992) ‘further reflections on the public sphere’, in C. Calhoun (ed.),
Habermas and the Public Sphere, Cambridge, MA: MIT Press.
Habermas, J. (1996) Between Facts and Norms, Cambridge: Polity Press.
Habermas, J. (2001) Why Europe needs a constitution’, New Left Review 11: 5 –26.
Hirst, P. (1996) ‘Democracy and civil society’, in P. Hirst and S. Khilnani (eds), Rein-
venting Democracy, Oxford: Blackwell.
Hirst, P. and Thompson, G. (1999) Globalisation in Question; The International
Economy and the Possibilities of Governance, Cambridge: Polity Press.
Honig, B. (1993) Political Theory and the Displacement of Politics, Ithaca, NY: Cornell
University Press.
Hooghe, L. and Marks, G. (2005) ‘The neofunctionalists were (almost) right: politiciza-
tion and European integration’, Constitutionalism Web-Papers, ConWEB No. 5/2005.
Imig, D. and Tarrow, S. (2000) ‘Political contention in a Europeanizing polity’, West
European Politics 23(4): 73– 93.
Jachtenfuchs, M. (1997) ‘Democracy and governance in the European Union’, in
A. Føllesdal and P. Koslowski (eds), Democracy and the European Union, Berlin:
Springer Verlag.
Magnette, P. (2003) ‘European governance and civic participation: beyond elitist citi-
zenship?’, Political Studies 51(1): 144 –60.
Mair, P. (2005) ‘Democracy beyond parties’, Center for the Study of Democracy,
University of California, Irvine, Working Paper 05/06.
Mouffe, C. (2000) ‘Deliberative democracy or agonistic pluralism’, Institute for
Advanced Studies, Vienna, Political Science Series.
Pateman, C. (1970) Participation and Democratic Theory, Cambridge: Cambridge
University Press.
Pollitt, C. and Bouckaert, G. (2000) Public Management Reform: A Comparative
Analysis, Oxford: Oxford University Press.
Przeworski, A., Stokes, S. and Manin, B. (eds) (1999) Democracy, Accountability and
Representation, Cambridge: Cambridge University Press.
Roberts-Miller, P. (2004) Deliberate Conflict: Argument, Political Theory, and Compo-
sition Classes, Carbondale, IL: Southern Illinois University Press.
Schmitter, P. (2001) How to Democratize the EU. . . and Why Bother? Lanham, MD:
Rowman & Littlefield.
Schumpeter, J. (1947) Capitalism, Socialism and Democracy, New York: Harper &
Brothers.
Scott, J. and Trubek, D.M. (2002) ‘Mind the gap: law and new approaches to govern-
ance in the European Union’, European Law Journal 8(1): 1 –18.
Steinberg, P. (2001) ‘Agencies, co-regulation and comitology – and what about poli-
tics?’. A critical appraisal of the Commission’s White Paper on Governance’. In
C. Joerges, Y. Mény, and J.H.H. Weiler (eds), Mountain or Molehill? A Critical
Appraisal of the Commission White Paper on Governance, http://www.iue.it/RSC
Taylor, R. (2000) ‘Liberalism and democracy in Habermas, Rawls, and Constant’,
http://www.gongfa.com/haeimasiluoersiziyouminzhu.htm
Thaa, W. (2001) ‘“Lean citizenship” The fading away of the political in transnational
democracy’, European Journal of Political Research 7(4): 503– 24.
Tully, J. (1999) ‘The agonic freedom of citizens’, Economy and Society 28(2): 161– 82.
Weale, A. (2006) ‘Party competition and deliberative democracy’, in J. Bara and
A. Weale (eds), Political Democracy and Party Competition, London: Routledge.
 M. Tsakatika: Governance vs. politics 885
Weiler, J. (1995) ‘Does Europe need a constitution? Demos, telos and the German
Maastricht decision’, European Law Journal 1(3), 219 – 58.
Wessels, B. (2004) ‘Contestation potential of interest groups in the EU: emergence,
structure and political alliances’, in G. Marks and M. Steenbergen (eds), European
Integration and Political Conflict, Cambridge: Cambridge University Press.
Wolin, S. (1996) ‘Fugitive democracy’, in S. Benhabib (ed.), Democracy and Difference:
Contesting the Boundaries of the Political, Princeton, NJ: Princeton University Press.

Final version accepted for publication 07/03/07