The Limits of Europe

OUP CORRECTED PROOF – FINAL, 17/10/21, SPi
The Limits of Europe

The Limits of Europe

Membership Norms and the Contestation of
Regional Integration
DA N I E L C . T HOM A S
1
1
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Daniel C. Thomas 2021
The moral rights of the author have been asserted
First Edition published in 2021
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2021911452
ISBN 978–0–19–920671–1
DOI: 10.1093/oso/9780199206711.001.0001
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
Preface and Acknowledgements
This book was conceived out of a hunch that the construction of Europe,
including the evolving community’s territorial expansion, was more deeply
contested than one would conclude from the official narrative in recent decades
about enduring commitments to democracy and human rights. The challenge was
two-fold—how to use the theories and methods of political science and the vast
resources available in the EU’s many historical archives to analyse and explain the
arc of EU constitutionalization and enlargement over time, and how to do so in a
manner that also speaks to regional integration processes in other parts of the world.
To this end, I decided to focus not on states’ decisions to seek membership nor
on the community’s readiness to admit particular applicant states, but on the
more fundamental and long-neglected question of how a regional community
decides which states are eligible for membership. This focus would reveal a great
deal, I suspected, about how political actors understand the nature and the limits
of the regional community that they are building and re-building with every
decision they make. Such a study of the conceptual and geographic limits of
Europe acquired a whole new significance as debates over cultural identity gained
salience across the community.
Of course, no such ambitions could ever be pursued, much less achieved,
without the support and cooperation of many individuals and organizations.
When the project was in its infancy, Yves Mény welcomed me back to the
European University Institute’s Robert Schuman Centre for Advanced Studies,
where I was able to explore ideas with Thomas Risse, Philippe Schmitter, and
other EUI scholars and to work in the EUI’s Historical Archives of the European
Union. Several years later, an International Affairs Fellowship from the Council
on Foreign Relations allowed me to spend a year in Brussels working at the
European Commission’s Directorate General for External Relations, which gave
me new insights into the dynamics of EU decision-making. Toward the end of
that year, the German Marshall Fund’s Transatlantic Centre in Brussels gave me a
desk and time to read through countless papers in the official archives of the
European Commission, the European Council, and the European Parliament, as
well as the archives of the Belgian, French, and German foreign ministries. In
later years, I also worked in the Barbara Sloan European Union Document
Collection at the University of Pittsburgh, and repeatedly with two first-rate
online archives—the Archive of European Integration (www.aei.pitt.edu) hosted
by the University of Pittsburgh and the Centre Virtuel de la Connaissance sur
l’Europe (www.cvce.eu) hosted by the University of Luxembourg. I will forever be
vi Preface and Acknowledgements
grateful for the help provided by these organizations, academic colleagues, and
archivist-historians.
In addition, colleagues at the universities where I have worked provided
further ideas and encouragement, including especially Alberta Sbragia and
Gregor Thum at the University of Pittsburgh, and Brigid Laffan and Ben Tonra at
University College Dublin. At Leiden University, when I decided that the book
needed a multi- method statistical chapter to complement the archive- based
process-tracing that was its core, Patrick Statsch co-authored Chapter 4 and the
Appendix while simultaneously writing his PhD dissertation at the University of
Amsterdam. I am deeply grateful for his contribution to the book. Last but
certainly not least, Frank Schimmelfennig has been an invaluable source of critical
comments and friendly encouragement over the years, even when my arguments
departed from his own ground-breaking work on similar questions.
Just as important as these many colleagues and archivists has been the undying
confidence in the project shown by Dominic Byatt, legendary editor at Oxford
University Press. Though I missed deadline after deadline due to parenting
responsibilities overlapping with a move from the USA to Ireland, then a move to
The Netherlands followed by learning Dutch and my duties as chair of a growing
department that left no time for scholarly pursuits, Dominic repeatedly assured
me of his commitment to the project. In the end, when my work was done, he and
his OUP colleagues Céline Louasli, Kim Allen and Saravanan Anandan expertly
transformed the manuscript into a book.
All epigraphs under copyright are reprinted with permission from their respective
publishers or copyright holders. The epigraph by Pierre Werner in the heading of
chapter 4 is reprinted by permission of the Pierre Werner Family archive. The
epigraph by Walter Hallstein in the heading of chapter 7 is reprinted by permission
of HarperCollins Publishers Ltd, © Walter Hallstein 1972.
Above all, I am grateful to Susanne, who supports and inspires me in ways
I could scarcely have imagined when we met years ago in the hills overlooking
Florence, and to our son Julien, who amazes me more with every passing year.
Leiden, December 2020

Contents
List of Figures ix
List of Tables xi
Note on Archival Sources xiii
PA RT O N E Q U E S T IO N S A N D A R G UM E N T S
1. The Question of Membership 3

2. Explaining Membership Eligibility 27
PA RT T WO M E M B E R SH I P OU T C OM E S
3. The Evolution of EU Membership Norms 49

4. EU Membership Eligibility in Statistical and Comparative
Perspective (with Patrick D. Statsch) 85
PA RT T H R E E M E M B E R SH I P P R O C E S SE S
5. Membership Eligibility in a Europe of Non-Communist States,

1957–1961 119
6. Membership Eligibility in a Europe of Parliamentary
Democracies, 1962–1969 137
7. Membership Eligibility in a Europe of Liberal Democracies,
1970–2005 159
8. Membership Eligibility in a Divided Europe, 2006–2020 194
PA RT F O U R C O N C LU SIO N S A N D I M P L IC AT IO N S
9. Rethinking Europe, Rethinking Regions 219
Appendix: Imputing missing Freedom House data from V-Dem data 243
(with Patrick D. Statsch)
Bibliography 247
Index 257
List of Figures
A.1. Comparison of Freedom House and V-Dem country-year scores 244

A.2. Correlations between country-year scores in Freedom House
and V-Dem datasets 244
List of Tables
2.1. Summary of explanations and hypotheses 45

4.1. EU decisions on the membership eligibility of aspirant states 87
4.2. Hypothesis 1: Membership norms 94
4.3. Hypothesis 4: Geographic location (Encyclopædia Britannica) 95
4.4. Hypothesis 5: Geographic location (United Nations) 96
4.5. Hypothesis 6: Geographic contiguity 96
4.6. Hypothesis 7, test 1: Treaty rules (change 1986) 98
4.7. Hypothesis 7, test 2: Treaty rules (change 1992) 98
4.8. Hypothesis 8, test 1: Liberal democracy (EU minimum) 99
4.9. Hypothesis 8, test 2: Liberal democracy (EU mean) 99
4.10. Hypothesis 9: Well-regulated markets 100
4.11. Hypothesis 10: Non-competitive markets 100
4.12. Hypothesis 11, test 1: Low risk states (global) 101
4.13. Hypothesis 11, test 2: Very low risk states (global) 101
4.14. Hypothesis 11, test 3: Low risk states (European) 102
4.15. Hypothesis 11, test 4: Very low risk states (European) 102
4.16. Hypothesis 12, test 1: NATO membership 103
4.17. Hypothesis 12, test 2: NATO membership or prospective membership 103
4.18. Summary of cross-tabulation analyses 105
4.19. Predicting EU membership eligibility in bivariate settings 110
4.20. Predicting EU membership eligibility in multivariate settings 111
4.21. QCA truth table 113
4.22. Analysis of causal configurations leading to EU membership
eligibility114
A.1. Imputed regime scores for country-years not covered by Freedom House data 245
Note on Archival Sources
Research for this book included extensive consultation of documents in various

official and private archives, both physical and online. Footnote references to
documents from these archives begin with the abbreviations listed below fol-
lowed by a file identifier and in most cases by a document identifier. Documents
in Dutch, French, German, Italian, and Spanish were translated to English by
Daniel Thomas or Susanne Boesch-Thomas.
Physical archives
ACCUE: Archives Centrales, Conseil de l’Union Européenne (Brussels)

(formerly: Archives Historiques du Conseil de l’Union Européenne)
ACE: Archives, Conseil de l’Europe (Strasbourg)
AHCE: Archives Historiques, Commission Européenne (Brussels)
AHPE: Archives Historiques, Parlement Européen (Luxembourg)
EUDC: Barbara Sloan European Union Document Collection, University of
Pittsburgh (Pittsburgh)
FPHS: Fondation Paul- Henri Spaak (Brussels). All contents now trans-
ferred to HAEU
HAEU: Historical Archives of the European Union (Florence)
MAEB: Archives Diplomatiques, Ministère des affaires étrangères, Royaume de
Belgique (Brussels)
MAEF: Archives Diplomatiques, Ministère des affaires étrangères, République
de France (Paris)
PAAA: Politisches Archiv, Auswärtiges Amt, Bundesrepublik Deutschland
(Berlin)
WB: Private collection of Willi Birkelbach (Frankfurt)
Online archives
AEI: Archive of European Integration: http://aei.pitt.edu/

CVCE: Centre Virtuel de la Connaissance sur l’Europe: https://www.cvce.eu/
PART ONE
QUE ST ION S A N D A RG UM E NT S
1
The Question of Membership
Any European State may apply to become a member of the Community.

—Treaty establishing the European Economic Community, 1957
The term ‘European’ has not been officially defined.
—European Commission, 19921
Europe, yes, but until where? What should be the geographic limits of
the European Union? The question has never been addressed funda-
mentally in any of the European institutions: it’s a taboo subject.
—Alain Lamassoure, Member of the European Parliament, 19992
Let’s just say for once: this is it.
—Jeroen Dijsselbloem, Eurogroup president, on EU
enlargement, 20163
By way of introduction, this chapter discusses the importance of understanding

how regional communities such as the European Union decide which states are
eligible to join and which are not. It surveys insights and gaps in existing scholar-
ship. It then introduces the argument that membership norms shape these deci-
sions by empowering certain political positions and compares its logic to
explanations derived from alternative theories of regional integration. It outlines
the empirical methods used to test the influence of membership norms against
treaty rules, geographic location, regime type, economic and security interests.
Finally, it surveys the aims and content of the chapters that follow.
1. The question and its significance
This book begins with a deceptively simple question: where does Europe begin
and end? The significance of this question reaches far beyond cartographic
debates. A recent report on the future of the European Union (EU) argues that
the European Commission should establish a new ‘Directorate General Europe’
focused on completing the unification of Europe while assigning relations with
neighbouring states without any prospect of membership to other parts of the EU
1 AEI: Europe and the challenge of enlargement (24 June 1992), p. 11.
2 Jusqu’où ? Le débat interdit, Le Monde, 9 décembre 1999.
3 Jeroen Dijsselbloem calls for end to EU expansion, Politico.eu, 10 June 2016.
The Limits of Europe: Membership Norms and the Contestation of Regional Integration. Daniel C. Thomas,
Oxford University Press. © Daniel C. Thomas 2021. DOI: 10.1093/oso/9780199206711.003.0001
4 The Limits of Europe
bureaucracy (Cvijic et al. 2019). What the report avoids, though, is the critical
issue of which states are eligible to participate in the integration process led by the
EU and which are not.
If it so decides, the EU has plenty of room to expand beyond its current 27
member states. The Council of Europe, a non-EU body designed to promote
democracy and human rights, currently has 47 member states. The Eurovision
song contest has accepted contestants from 52 states over the years. And the
European Broadcasting Union, which sponsors the song contest, has member
organizations in 56 countries. So how does the EU, which does not hesitate to
identify itself with Europe as whole, define its own limits?
This is not a new question. Struggles over the geographic limits of ‘Europe’ as a
political community have been a salient feature of European integration from the
beginning. Within months of the entry into force of the Treaty of Rome in 1958,
governments elsewhere in Europe began to talk about joining or at least preparing
for eventual membership in the new European Economic Community (EEC).4
Over the following decades, the community—later renamed the European
Union—grew from 6 to 28 member states. And despite the United Kingdom’s
recent exit from the community, neighbouring states continue to seek
membership and the European Commission continues to negotiate with them on
behalf of the Union.
Yet notwithstanding the EU’s apparently endless tendency to expand, a closer
look at the historical records reveals that over the years, some neighbouring states
were encouraged to apply, some were told to wait, and others were told definitively
that they were ineligible for membership. In some cases, the same applicant state
received positive and negative messages in relatively quick succession. This record
cries out for explanation. As Bahar Rumelili (2004: 28) asked, ‘How is it that with
respect to certain states, the EU constructs firm lines of boundary between self
and other, and with regard to others, fluid and ambiguous frontiers?’
For example, why did the EEC encourage Spain to pursue membership and
then refuse for years to offer it a path to accession? Why did the EU recognize the
Czech Republic as a membership candidate but refuse for several years to do the
same for the neighbouring Slovak Republic? Spain and the Slovak Republic are
both undeniably located on the European continent but this was not enough to
ensure their recognition as eligible to join the European Union. Similar questions
have been asked about Ukraine, whose governments have tried to place their
country on a path to EU membership since the country gained independence in
1991. When President Viktor Yanukovych seemed to abandon this goal in 2013,
4 When referring specifically to the European Economic Community, this book uses that term or
the abbreviation EEC. When referring to the entire period from the creation of the EEC through the
present, or just to the period since the entry into the effect of the Treaty on European Union, the book
uses the term European Union or the abbreviation EU.
The Question of Membership 5
demonstrators in Kiev’s Maidan Square braved sniper fire to wave EU flags and
demand closer ties to Europe. Yet despite these appeals, the EU has consistently
rebuffed Ukraine’s pursuit of what Brussels calls a ‘membership perspective’.
Meanwhile, after years of delay, the EU formally recognized Turkey as a candidate
state and began accession negotiations, but this did not quell debate among and
within the member states regarding Turkey’s suitability for membership.
These puzzles cannot be resolved simply by consulting treaties or formal
declarations. Both the 1951 Treaty of Paris and the 1957 Treaty of Rome declared
that ‘Any European state may apply to join the community’ but neither treaty
defined what makes a state ‘European’. Three and a half decades later,
the European Commission offered its own quasi-definition: ‘The term “European” . . .
combines geographical, historical and cultural elements which all contribute to
the European identity. The shared experience of proximity, ideas, values and his-
torical interaction cannot be condensed into a simple formula, and it is subject to
review by each succeeding generation.’ Perhaps deliberately, the Commission’s
open-ended definition provided no clear guidance as to whether a particular state
is actually eligible for membership: ‘The Commission believes that it is neither
possible or opportune to establish new frontiers of the European Union, whose
contours will be shaped over many years to come.’5
According to Kalypso Nicolaïdes, this vagueness is not accidental: ‘the lack of
explicit political ends to the journey has gone along with an equal indeterminacy
in terms of geographical ends. Precisely because the European project is an open-
ended process rather than a frozen structure, it has also been an open-ended
space’ (2014: 237). In fact, she argues, this open-endedness applies to both
the ultimate shape of the community’s external border, by which she means the
definition of ‘which countries should become members of the EU’, and the nature
of that border, by which she means how the border mediates between societies on
both sides. The sum of these two dimensions, says Nicolaïdes, constitutes the
fundamental ‘constitution of Europe as a political community’ (2014: 238).
How then can we begin to understand this most fundamental aspect of
European integration, or for that matter, of any regional integration project? The
formal membership rule now prevailing in the EU—that it will only consider
applications from European states with stable institutions that ensure democracy,
the rule of law, respect for human rights, and the protection of minorities, as well
as a market economy able to withstand the competitive pressures of the Single
Market—was not established even informally until many years after the EEC
Treaty. And notwithstanding the apparent clarity of this rule, the eligibility
of some applicant states—including most notably Turkey—remains deeply con-
tested. As such, the much-vaunted ‘Copenhagen criteria’ are neither an original
and immutable aspect of European integration nor a complete explanation of the

politics of eligibility for EU membership.
In fact, the definition of ‘European’ is deeply contested by the continent’s
leading politicians. For some, such as former French president Valéry Giscard
d’Estaing (2000), Europe is a space whose geographic and cultural limits are
stable and plainly seen, so eligibility for EU membership is a simple question of
location. Others, such as former EU enlargement commissioner Olli Rehn (2006: 4),
have pushed back, insisting ‘the EU is defined by its values more than by sheer
geography’ and ‘we should not draw in Indian ink some thick “fault line”
according to some notional historical borders between civilisations’. It is therefore
not surprising that Finnish Foreign Minister Erkki Tuomioja, while holding the
EU’s rotating presidency, would plead: ‘Please don’t ask me where the borders of
Europe are, that’s something we didn’t want to put on the agenda’.6
And just like the broader phenomenon of regional integration, the puzzle of
membership eligibility is not limited to Europe. Consider the Association of
Southeast Asian Nations (ASEAN), whose 1967 Bangkok Declaration announced
that ‘the Association is open for participation to all States in the South-East Asian
Region subscribing to the aforementioned aims, principles and purposes.’ But as
the organization’s former Secretary-General noted, ‘This raises the question: what
precisely is Southeast Asia?’ (Severino 2006: 41). In practice, ASEAN has been
inconsistent: it accepted Brunei’s application for membership barely a week
after that country’s independence, but Timor-Leste’s repeated assertions that
‘[g]eographically we are very much part of Southeast Asia’ (Ramos-Horta 2011)
have not convinced ASEAN that it qualifies for membership (Talesco 2016). The
same is true for neighbouring Papua-New Guinea, which has also long tried but
failed to gain recognition as being eligible for ASEAN membership.
Similarly, the Gulf Cooperation Council invited Jordan and Morocco to join,
but not Iraq, although of the three, only Iraq actually borders the gulf after which
the organization is named (Al Tamamy 2015). The construction of the Caribbean
region is equally puzzling: some states are members of both the Association of
Caribbean States and the Caribbean Community, while others in the same area
belong only to the former, and Cuba is not welcomed in either organization.
Similarly, the Shanghai Cooperation Organization recently invited India and
Pakistan to join original members China, Russia, Kazakhstan, Kyrgyzstan,
Tajikistan, and Uzbekistan, but other neighbouring states have not been given any
prospect of accession.
These European and non-European examples demonstrate that we cannot rely
on ‘geographical eyeballing or legalistic inspection’ to make sense of how regional
organizations make decisions on the eligibility of neighbouring states (Cederman
6 Les négociations d’adhésion UE-Turquie pourraient être en danger, Le Monde, 29 juin 2006.
2001: 2). Instead, we need a clear definition of membership eligibility, a theoretical

framework attuned to the complex political dynamics surrounding this issue, and
an empirical basis for assessing both general patterns and the specificities of
rhetoric and mobilization around individual cases. By examining the politics of
EU membership eligibility over seven decades in this manner, this book offers
new insights into European integration and into the general process by which
regional communities define their borders and decide which applicant states are
eligible for accession. In so doing, it contributes both theoretically and empirically
to the growing literature on the dynamics and complexity of membership in
international organizations (Droesse 2020).
This book aims to explain how regional communities, defined as groups of
states that identify with a particular geographic space and organize inter-
governmental and/or supranational institutions to promote shared interests or
values, decide which states are eligible for membership and which are not. To be
more precise, it aims to identify the factors that determine how the member states
and supranational officials of a regional community decide on the eligibility of
particular states aspiring to membership (hereafter, aspirant states).
These decisions matter: whether or not the European Union (or any other
regional community) welcomes another state that wishes to join the club can have
major implications for flows of labour, products, and investment, the life chances
of individuals, the political fortunes of governments, and the identities and
interests that shape international politics. The book’s inquiry is thus key to
understanding the nature of regional communities and the process of regional
integration and governance, both in Europe and elsewhere in the world.
To be clear, the book does not attempt to explain why non-member states apply
to join the EU: it takes their expressions of interest in membership as given. It
also does not attempt to define where on a map the ultimate limits of Europe
might lie, nor to assess the wisdom of the EU’s past decisions on enlargement, nor
to prescribe whether a particular state should be offered membership in the
future. Instead, it focuses on explaining how regional communities—in this case,
the EU—understand the potential geographic limits of their joint endeavour and
how these understandings affect their decision-making over time on which states
are welcome to join. (For further discussion of eligibility, see Section 3 below.)
In brief, I argue that the geographic limits of regional communities are
determined by norms that define the characteristics expected of member states
and thereby distinguish potential members from states that have no prospect of
accession. These norms shape the legitimacy of various membership scenarios
and thereby influence which aspirant states are considered eligible for
membership and which are not. As the norms change over time, the ultimate
limits of a regional community may expand or contract. Furthermore, the content
and effect of these ‘membership norms’ cannot be reduced or attributed to the
economic or geopolitical interests of the community’s member states, nor to their
type of government or physical location, nor to the treaty obligations that they
have assumed. To demonstrate the explanatory power of this argument, I test it
empirically across multiple phases of European integration and multiple cases of
EU decision-making with varying outcomes over seven decades. While the book
focuses empirically on Europe, I have designed the argument and analysed the
findings to speak as well to developments in other regions of the world.
2. Insights and gaps in the scholarly literature
The politics of membership eligibility deserves far more attention from scholars
of international organization, regional integration, and community building than
it has received thus far. There is extensive and insightful scholarship on why states
create regional institutions and policies (Moravcsik 1998, Mattli 1999,
Pevehouse 2003), why states apply to join an existing regional organization
(Mattli and Plümper 2002, Emmers 2005, Gray 2009), and how membership in
regional organizations influences domestic political reform, especially the
consolidation of democracy and respect for human rights (Glenn 2004,
Pevehouse 2002, Schimmelfennig 2005, Vachudova 2005, Donno 2010).
In comparison, relatively few scholars have focused on the politics that shape
the geographic limits of regional communities and in particular which applicant
states are considered eligible for membership. Some have trivialized the issue
entirely: ‘A state is coded as an eligible potential member . . . for any regional IGO
that is based in its home region (for example, Bolivia is eligible for membership in
a regional IGO based in Latin America, but not one based in sub-Saharan Africa)’
(Donno et al. 2015: 257). But the reality is far more complex than the names of
continents or the wording of treaties.
One way to conceptualize the issue is to distinguish between an inclusive
approach to building a multilateral organization that involves all potential
members from the outset and a sequential approach that begins with a core group
of states most committed to cooperation and then incorporates others as their
preferences converge with the founders. The inclusive approach is less likely to
yield deep integration because it increases the heterogeneity of integration
preferences in the decision-making process. In contrast, rather than waiting until
all potential members have adopted similar preferences, the sequential approach
allows core states to create an institution to their liking while remaining open to
any like-minded states that present themselves for membership. In this case,
expansion will be ‘a multi-step process in which the multilateral expands only
gradually as potential members or the multilateral itself attains some property not
originally possessed at the time the institution was created’ (Downs et al.
1998: 399).
The problem here is the argument regarding which states are potentially
covered by an inclusive approach, which might be temporarily excluded by a
sequential approach, and which have no prospect of participation. According to
Downs et al., these distinctions depend entirely on functional policy preferences:
all states with identical preferences regarding policy integration will be included
regardless of whether the organization opts for an inclusive or sequential
approach, while those with divergent preferences will be excluded under a
sequential approach, regardless of other considerations. This may be accurate
with respect to potentially global organizations focused on a functional issue-area
such as trade or environmental protection, but it is hard to reconcile this
expectation with the pattern evident in the course of European integration:
starting with an inclusive approach in treaty rules and practice, then becoming
more selective in practice with a considerable lag before treaty rules are adapted
to the new practice, then more recently welcoming some states (like Bosnia-
Herzegovina) while excluding others (like Ukraine) with very similar policy
preferences and administrative capacities.
In similar terms, Judith G. Kelley (2010) distinguishes between a ‘convoy’
model of membership that in principle allows all regional states to participate and
a ‘club’ model that subjects aspiring members to strict admissions criteria. The
two models have distinct implications for how a regional organization relates to
states in the region: ‘Clubs can leverage higher entry criteria to solicit behaviour
changes prior to admission for any late joiners [while] convoys, with lower entry
barriers, can take a less confrontational approach to outlier states, interacting
with them within the organization rather than erecting barriers’ (2010: 2).
Unfortunately for the purposes of this study, Kelley focuses on how the convoy
versus club choice affects the likely success of regional organizations, rather than
on how such organizations choose between a convoy and a club model or decide
on the eligibility or accession of applicant states.
In contrast, Barbara Koremenos, Charles Lipson, and Duncan Snidal (2001a,
2001b) examine the possible relationship between the enforcement difficulties
faced by an international organization and the restrictiveness of its membership
rule. Their conjecture is that membership restrictiveness increases with the
severity of enforcement problems, which one might assume would increase with
additional members (Koremenos et al. 2001a). If this were true, then one would
expect international organizations to become less open to non-member states as
the number of member states increased, and to prefer applications from individual
states over applications by groups of states. Unfortunately, these conjectures do
not fit their project’s empirical findings (Koremenos et al. 2001b) nor do they
offer a good explanation for the pattern in EU history described briefly above.
Of course, it is not necessary to conceive of regional organization simply as an
interaction of self-interested states weighing the costs and benefits of behavioural
options: the identities of states and their relationship to shared norms matter as
well. Inspired by social identity theory, Iver B. Neumann (1999) and Bahar
Rumelili (2004) argue that no group of states can claim that certain characteristics
distinguish its members without referring implicitly or explicitly to other states
that lack the same characteristics, either entirely or in sufficient measure.
Eligibility for regional membership is thus a rhetorical construction in which
certain characteristics are attributed to an Other (in this case, non-Europeans) in
order to provide meaning and coherence to the Self (in this case, Europeans). For
example, Dainotto (1997), Neumann (1998), Morozov and Rumelili (2012), and
Todorova (1997) have highlighted the reframing of Europeanness over time in
juxtaposition to the purported characteristics of various neighbours, especially
Muslims, Russians, Turks, and Mediterranean peoples. It is nonetheless difficult
to judge this approach’s contribution to explaining the politics of membership
eligibility because few of its advocates have specified causal mechanisms or scope
conditions for their arguments or tested competing hypotheses.
There is an extensive and more methodologically explicit literature on EU
enlargement in the post-Cold War period yet even the best works here suffer
from considerable shortcomings. For example, Moravcsik and Vachudova (2005)
propose a parsimonious liberal explanation of how the EU makes enlargement
decisions, but then offer only a thin empirical narrative limited to the eastern
enlargement. In a similar rationalist spirit, Skålnes (2005) proposes that
enlargement decisions are driven by geopolitical calculations, but he too pays
little attention to alternative explanations and offers little direct evidence to
support some of his key claims. Sedelmeier (2005) proposes an alternative,
constructivist explanation focused on commitments made to eastern Europe
during the Cold War, but it is not clear how his argument could apply to
other cases.
Of this genre, the most influential has been Frank Schimmelfennig’s (2001,
2003) investigation of EU and NATO enlargement. In his conceptualization, the
‘collective identity’ of a region is composed of the values of its political leaders,
the norms that they share, and the regional organization’s formal rules on
accession. He thus explains the eastern enlargement as a result of ‘rhetorical
action’: when the states of central and eastern Europe framed their applications in
terms of the Union’s multi-dimensional identification with liberal democracy, EU
member states approved their quest for full membership despite the availability of
less-costly alternatives. This argument and the evidence that Schimmelfennig
amassed to support it offer a powerful explanation of why regional communities
sometimes make membership decisions that seem inconsistent with material
incentives.
However, Schimmelfennig’s work does not associate distinctive causal
mechanisms or scope conditions with his three components of regional identity
and thus cannot separate their respective contributions to the outcome in
question. Furthermore, instead of exploring how EU values, norms, and rules

vary and maybe conflict over time, his research design treats them as unitary and
fixed: ‘European’ is equated with liberal democracy. This lack of variation in the
explanatory variable, presumably due to the project’s limited historical scope,
makes it difficult to determine whether rhetorical commitment to liberal
democracy was sufficient or even necessary to explain the outcome in question.
These shortcomings limit the study’s explanatory power in the case of Europe and
its generalizability to other cases. A less context-bound investigation of this topic
could therefore have important implications for how we understand regional
integration and the creation of supra-national political communities in Europe
and beyond.
Thomas Plümper and Christina Schneider (2007) offer an explicitly rationalist
alternative to Schimmelfennig’s constructivist explanation of the EU’s eastward
enlargement. Whereas Schimmelfennig used the costs of enlargement to frame
his puzzle, Plümper and Schneider focus their argument on the bargaining
between enlargement’s winners and losers. As they note, ‘the potential losers of
enlargement can always decide to reject the admission of further states if the
gains and costs are not redistributed to compensate them for their losses. . . . The
settlement of enlargement conflicts [thus] requires the redistribution of
the enlargement gains from the applicants and/or the relative winners among the
Union members to the relative losers’ (2007: 573). According to Plümper and
Schneider, ‘The EU’s Eastern enlargement was accomplished because after long
and heated debates, most of the accession countries agreed to be temporarily
excluded from agricultural subsidies, structural aid, and the free movement of
labor’ (2007: 569). However plausible, this argument is very difficult to test:
‘Compensation of enlargement losers through . . . discrimination of new members
is not necessarily inconsistent with Schimmelfennig’s framework’ as long the
value of compensation is less than the costs anticipated by the losers, which
Plümper and Schneider concede is ‘almost impossible’ to measure (2007: 585,
note 4). Most important here, their analysis focuses entirely on the bargaining
that occurs after a non-member state is recognized as eligible for accession, not
on how eligibility is determined in the first place.
Other studies, including Karen Smith (1998), Frank Schimmelfennig and
Ulrich Sedelmeier (2004, 2005), Gwendolyn Sasse (2008), Tina Freyburg (2013),
and Eli Gateva (2015), have examined the political conditionality that is central to
EU decision-making on applicant states. However, these works focus p redominantly
on the period between the non-member state’s formal application for member-
ship and the signature of an accession treaty. Negotiations during this period are
undeniably important, but these studies do not recognize their own selection
bias: long before submitting a formal application for membership, governments
interested in accession discuss their intentions with EU institutions and member
states, which have various ways of encouraging, discouraging, and guiding a
formal application. In many cases, the states involved have long been parties to
EU association agreements that establish strong incentives for reform. So given
that most states that submit a formal application for accession have been pre-
screened and even counselled by the community itself, the application is best
understood not as a start but as an intermediary step on the path to membership.
Thomas Plümper, Christina Schneider, and Vera Troeger’s (2006) study of the
EU’s 1993 Copenhagen summit is not framed explicitly in terms of membership
eligibility but it does address the EU decision to place ten of the formerly
Communist states ahead of others in the accession process. Their theoretical
expectation is that the EU would welcome applicant states whose policy
preferences resemble their own, as revealed by the applicants’ level of democracy
and the extent of market reforms. To test this argument and to eliminate selection
bias due to the possibility that the EU criteria may have deterred less-democratic
states from applying in the first place, they use a dynamic Heckman selection
model that distinguishes the factors that lead states to apply for membership from
the factors that shape the EU’s response to these applications. Their finding that
the EU is more open to more democratic and market-oriented states thus offers
robust insights into the politics of membership eligibility in the early post-Cold
War period. However, its generalizability is limited: the study only covers the
1993 summit and as documented later in this volume, there is strong evidence
that democracy has not always been a significant consideration in EC/EU
decision-making.
Ivan Katchanovski’s (2011) analysis of EU and NATO enlargement following
the Cold War is more explicit about membership eligibility, using an ‘EU
Accession Index’ that codes states annually from 1997 to 2010 as non-member/
not potential EU member, potential EU member, official EU candidate, and EU
member. Katchanovski’s conclusion that EU decisions on enlargement are
strongly affected by the applicants’ level of democracy and economic development
echoes that of Plümper et al. (2006), but his four-step index is hard to apply to
earlier periods because the formal designations ‘potential candidate’ and
‘candidate state’ did not exist before the 1990s. In addition, his use of country-
year data brings in country-years where there was no risk of an EU decision and
blinds the analysis to cross-temporal interdependencies: EU decisions in year 1
may affect EU decisions in year 2, etc. Finally, as Katchanovski (2011: 316)
concedes, ‘the regression analysis cannot rule out reverse causation between EU
membership and democratisation’.
In sum, the existing literature contains valuable insights into the factors that
influence how regional communities make enlargement decisions, but few
scholars have distinguished decisions on membership eligibility from decisions
on accession. Most importantly here, the literature does not offer a persuasive
explanation of how a regional community assesses the membership eligibility (as
distinct from readiness for accession) of non-member states, nor does it provide a
theoretical or methodological framework for studying this in a manner conducive

to cross-regional and cross-temporal analysis.
As a first step toward addressing this gap, the following section defines
membership eligibility and discusses empirical indicators that can be used to
study it.
3. Defining membership eligibility
To say that a non- member state is eligible for membership in a regional

community is to say that it fulfils the community’s most fundamental expectations
of all members. Membership eligibility is thus not an objective material
characteristic: it is a social status dependent upon collective recognition by the
community’s members. This consensus does not require that all member states’
governments agree on the desirability of a particular applicant—simply that they
accept a community position on its eligibility. Furthermore, recognition of
membership eligibility is distinct from assessments of whether the non-member
state’s legal code, administrative capacity, or economic competitiveness make it
ready for membership, and therefore what delays or other technical conditions
must be imposed on its accession. But only after an applicant is deemed eligible
do the potentially detailed and lengthy negotiations on the terms of accession
begin. Recognition of membership eligibility is thus the gateway through which
all states seeking membership must first pass. In social science terms, it is a
necessary but not sufficient condition for accession.7
This focus on member states reflects the predominantly inter-governmental
character of decision-making on membership in regional organizations. For
example, although supranational institutions play a key role in the drafting,
approval, and enforcement of EU legislation, the Treaty on European Union (like
its precursors) assigns principal authority for assessing and approving membership
applications to the member states. As a result, non-member states’ expressions of
interest in accession are addressed to the Council of the European Union (for-
merly the Council of Ministers), which typically comprises the foreign or
European affairs ministers of all the member states, and/or the European Council,
which comprises the heads of state or government. The intergovernmental aspect
is reinforced by the fact that the adoption of formal membership criteria, the
opening of preliminary talks or formal negotiations with an applicant state, and
the conclusion of association or accession treaties all require the unanimous sup-
port of the member states.
7 Standards of membership eligibility are also relevant in principle to current member states, but
decisions on whether the community should impose sanctions on norm-violating member states are
subject to different dynamics and thus are not theorized here.
The intergovernmental focus does not deny the important agenda-setting and
advocacy roles that the EU’s supranational bodies have always played in
enlargement decision-making (these feature prominently in the process-tracing
of EU decision-making in Chapters 5–8). Nonetheless, final decisions on the
eligibility of non-member states are made collectively by the member states. The
European Commission makes recommendations to the Council and negotiates
with non-members on its behalf, but it does not have a vote on the eligibility or
accession of any state. The original EEC treaty did not give the European
Parliamentary Assembly any role in membership decision-making, while the
1986 Single European Act and subsequent treaties require only that the European
Parliament give its assent to accession treaties.
General principles of membership eligibility are often discussed when
governments or non-state actors deliberate on the construction of a new regional
community or re-examine the foundational principles of an existing community.
However, the question of whether a particular non-member state is eligible for
accession—the ultimate focus of this book—arises most clearly when its
government communicates to a regional community its interest in joining the
club. This communication could involve a formal declaration of interest in
accession, an application to negotiate an association treaty focused explicitly on
preparation for membership (hereafter, a ‘pre-accession treaty’), or an application
to negotiate full accession.8
An applicant state is eligible for membership in a regional community when
the community’s member states conclude collectively that it possesses the
fundamental characteristics required of all members. Indicators that a regional
community considers a non-member state to be eligible for membership include
any of the following acts by the community:
• the decision to hold unconditional preliminary talks on membership with a

non-member state that has explicitly declared its interest in accession;
• the formal designation of a non-member state as a ‘pre-candidate’ or ‘candi-
date’ state;
• the decision in principle to begin formal negotiation of a pre-accession or
accession treaty;
• the start of formal negotiations on a pre-accession or accession treaty.
Regarding the first point above, a decision to hold talks with representatives of a
state that has expressed interest in closer ties to the community would not
constitute evidence of eligibility if such talks were intended to clarify whether the
8 These ‘pre-accession treaties’ are a distinct class of association treaties, which also include purely
contractual accords without any membership perspective.
non-member state seeks ultimately to gain membership or just to develop func-

tional links to the community.
As this list of indicators suggests, not all positive statements by community
institutions or officials constitute a positive position on the eligibility of a non-
member state aspiring to accession. For example, European Commission
‘opinions’ and European Parliament declarations on the ‘vocation’ of particular
non-member states may be part of a political process leading to recognition of
membership eligibility, but neither can substitute for a Council position. In
addition, a statement that is clearly intended to encourage a group of non-
member states to develop the characteristics required for accession cannot be
construed as recognition of their individual eligibility for membership. For
example, the European Council’s June 1993 statement that ‘the associated
countries of Central and Eastern Europe that so desire shall become members of
the European Union’ was accompanied by a list of standards (the ‘Copenhagen
criteria’) that aspiring states had to meet before they could be eligible for
membership. The same is true of the EU’s June 2000 statement on the Western
Balkans, which declared that ‘its objective remains the fullest possible integration
of the countries of the region into the political and economic mainstream of
Europe through the Stabilisation and Association process, political dialogue,
liberalisation of trade and cooperation in Justice and Home Affairs. All the
countries concerned are potential candidates for EU membership.’9
By the same logic, a negative position on eligibility is a regional community’s
collective conclusion that an applicant state lacks the fundamental characteristics
necessary for membership. Indicators that a regional community considers a
non-member state to be not eligible for membership include any of the following
acts by the community:
• an explicit refusal to hold preliminary talks on accession or a failure to issue

an invitation to such talks within two years of a non-member’s expression of
interest in accession;
• an explicit refusal to start negotiations on an accession treaty or pre-
accession association treaty or failure to start such negotiations within one
year of opening preliminary talks;
• a deliberate blockage of progress toward accession, including a failure to
start pre-accession or accession negotiations within one year of a declaration
of ‘candidate’ status, a freeze of all accession negotiations already underway,
or an explicit and unconditional cancellation of all accession negotiations.
9 CVCE: Conclusions of the Santa Maria da Feira European Council (19–20 June 2000).
All of these time windows (‘within two years’ etc.) are chosen to be large enough
to ensure that the indicated non-action was deliberate and not a result of the
community’s busy agenda or its need to gather information before making a
decision.
However, none of these developments would constitute evidence of non-
eligibility if it were clearly due to circumstances unrelated to the characteristics or
actions of the aspirant state, such as a diplomatic or institutional crisis that
prevents the community from responding to aspirants in a timely or normal
manner. For example, the de facto suspension of negotiations with Denmark,
Ireland, and Norway in 1963 and 1967 cannot be viewed as the result of a negative
decision on their eligibility because they were clearly due to the turmoil following
France’s veto of negotiations with the UK.
Once adopted, a regional community’s position on the eligibility of an aspirant
state is resistant but not impervious to change. In general, reputational costs and
the difficulty of achieving consensus militate against revisiting a decision on a
particular aspirant’s eligibility. However, new circumstances in the aspirant state,
within the community, or in the outside world, perhaps in conjunction with the
efforts of non-governmental or supranational policy entrepreneurs, may lead
member states to revisit an earlier conclusion. As such, the longer an aspirant
state that has been deemed eligible for membership waits before completing
accession, the more it risks having its eligibility challenged. Likewise, an aspirant
that the community has judged ineligible may be reconsidered later if its
characteristics or the community’s expectations change significantly. As a result, a
single aspirant state may be the subject of multiple eligibility decisions over time.
However, as the second list of indicators suggests, critical statements by
community institutions or officials do not necessarily constitute a negative
position on the eligibility of a particular aspirant or candidate state. For example,
European Commission ‘opinions’ and European Parliament declarations on
particular states may be part of a political process leading to a denial of membership
eligibility, or even to the retraction of a prior recognition of membership eligibility,
but neither can substitute for a collective decision by the member states in deter-
mining eligibility.
Some might argue that a community’s decision to freeze certain parts (‘chap-
ters’ in contemporary EU parlance) of accession negotiations also constitutes a
negative decision on eligibility, even if negotiations continue on other parts. After
all, accession does not have to be accomplished by any given date: negotiations
can proceed slowly if the applicant is not making the necessary reforms. As such,
a decision by the community to deny the applicant a possibility of progress in
certain areas is a political decision that calls into question the applicant state’s
eligibility for membership. If so, then eligibility would be restored when all
blockages are lifted and the progress of negotiations is limited only by the aspirant
state’s political will and administrative capacity. However, this logic does not
outweigh the fact that accession negotiations are only held with states that are
considered eligible for membership, so we consider the continuation of any such
negotiations as an indicator that the community has not revoked its prior
confirmation of membership eligibility.
Having clarified the nature of membership eligibility and ineligibility and how
they can be recognized empirically, the chapter now turns to a brief summary of
the book’s core argument and alternative explanations.
4. The argument and alternative explanations, in brief
So how can we best understand a state’s eligibility for membership in a political

community that is regional in scope and supranational in scale? This section
briefly presents a range of explanations, each informed by a different theoretical
tradition and centred on a different explanatory variable. The book’s core
argument, focused on the evolution and power of membership norms, is
summarized first, followed by five plausible alternatives focused respectively on
physical geography, treaty rules, regime type, commercial interests, and
geopolitical threats.
1. Membership norms. Membership norms express a regional community’s

collective expectations regarding the characteristics that should distinguish
community members from other states. They emerge in deliberations
among a regional community’s member state governments and supranational
officials and thereafter bolster or weaken the legitimacy of a campaign for
accession, depending on the non-member state’s fit with the prevailing
norm. A regional community will therefore recognize the membership eli-
gibility of non-member states that fit the membership norms prevailing
among its member states and regional institutions, and reject others. If no
membership norm prevails within the community, then hard bargaining
over the eligibility of non-member states would be expected. For example,
regardless of the interests or preferences of its member states, a regional
community whose membership norm limits the community to states guar-
anteeing the rule of law will not accept a non-member state that fails to
provide this guarantee within its borders.
2. Physical location. A regional community’s decisions on membership
eligibility are determined by the location of the non-member state’s terri-
tory in relation to that of the current members. This may be a matter of
territorial contiguity or location within the geographic space conventionally
associated with the region in question. Territorial contiguity increases the
likelihood of economic interdependence and thus creates a strong motive
for including the outsider in the club (Mattli 1999), and it tends to reduce
transportation costs, which have an inverse relationship with the net gains
of economic integration (Frankel et al. 1997). By this logic, a regional
community would welcome all non-member states with which a current
member state shares a land border or close proximity across water. Shifting
the focus to social purpose, which John Gerard Ruggie (1982) considers
key to understanding the costs that states are willing to bear in international
organization, casts territorial location in a new light. Quite simply, if a
regional organization aims to integrate or govern a particular geographic
space, such as a continent or portion thereof, one might expect that any
state located fully or even partly within this space would be considered
eligible for membership, and states outside this space would be ineligible.
For example, any state located in Europe would be eligible for membership
in the European Union.
3. Treaty rules. Regional communities are founded on intergovernmental
treaties that establish formal rules regarding which states may join. The
community’s decisions on the membership eligibility of non-member states
may be determined by the content of these rules. If so, then a regional
community will recognize the membership eligibility of non-member states
that fit the formal rule on memberships established by the community’s
treaty, and reject others. For example, a regional community whose
founding treaty requires that all member states must also belong to a
particular military alliance would exclude a non-member state that was not
part of the alliance.
4. Regime type. The member states of a regional community may view its
evolution as an opportunity to consolidate their own position domestically
and/or to spread their values by ensuring the prevalence of states like theirs
within the community and keeping dissimilar regimes out. A regional
community’s decision on the membership eligibility of non-member states
may therefore be shaped by variations in the member states’ form of
government. If so, then a regional community will recognize the
membership eligibility of non-member states whose regime type is similar
to that of the member states, and reject those that are dissimilar. For
example, a regional community composed of hereditary monarchies will
not consider a parliamentary democracy eligible for membership, regard-
less of the other interests of its member states.
5. Commercial interests. In regional communities including a customs union
or single market, which are likely to influence the economic profile and
competitiveness of their members, there is a strong incentive for member
states’ governments to base their preferences regarding possible
enlargement on calculations of how integrating an applicant state would
affect their own state’s economic welfare. For example, if they prefer to keep
states with inferior administrative capacities out of the customs union/

single market, then they would likely exclude applicants with a lower level
of economic development. Conversely, if they view enlargement as an
opportunity to benefit from cheaper labour, they would likely exclude
applicants with a higher level of economic development.
6. Security interests. Members of a regional community may base their
enlargement preferences on their expectations regarding how a potential
new member would affect their geopolitical security. Individual member
states would oppose any applicant states whose accession would reduce
their own security. A regional community will therefore deem a non-
member state eligible for membership if its accession would reduce all
member states’ geopolitical vulnerability or improve their capability for
resisting geopolitical threats.
It is also important to explain why the book does not give equal status to an
explanation of membership eligibility focused on socio- cultural identity.
Identities are ‘relatively stable, role-
specific understandings and expectations
about the self ’ that provide a basis for both social distinctiveness and social
solidarity (Wendt 1992: 397). Some scholars define regions as groups of states
whose populations or elites share a cultural identity (Kupchan 2010, Acharya
2011). One could therefore attempt to explain eligibility for membership in a
regional organization with reference to the socio- cultural identities of the
applicant state and the member states.
From this perspective, one might expect regional communities to welcome
membership applications from states whose cultural characteristics they share
and to reject applications from states with different cultural characteristics.
Prescriptive versions of this logic are widely heard in public debates. Former
French president Valéry Giscard d’Estaing (2000) argues that Turkey is ineligible
for EU membership because it does not share Europe’s cultural heritage, while
former Dutch politician Frits Bolkestein (2011) goes a step further, stressing the
Christian roots of European civilization. Hence Thomas Risse’s suggestion that
‘one cannot even begin to understand EU enlargement without taking identity
politics into account’ (2012: 92).
However, in order to test the identity explanation of EU decision-making over
time, it would first be necessary to determine what historical memory or set of
cultural values were shared by Europeans at various points in time since the
1950s. The problem is that apart from the claims of political entrepreneurs, ‘a
meaningful common European historical identification barely exists’ (Mayer and
Palmowski 2004: 575). As a result, scholars tend to treat the joint legal
commitments of European states as a proxy for European identity even though
the cultural sources of these commitments are ambiguous at best.
And even if it were possible to observe European cultural identity at particular

moments in time, this could not explain why regional communities distinguish
between applicant states with very similar cultural characteristics. For example,
why was the Czech Republic offered an EU membership prospect in 1994 while
the neighbouring Slovak Republic was not? Nor could this explain sudden but
consequential changes in how a regional community responds to the membership
aspirations of a single state, whose cultural characteristics presumably remained
the same. For example, why did the EEC encourage Spain to prepare for
membership in the late 1950s, reject its eligibility when it sought an association
treaty in 1962, and then reverse itself again after 1975? Why did the EEC sign a
treaty with Greece in 1961 that recognized Athens’ eligibility for eventual
membership, then close the door in 1967, and re-open it in 1975? Socio-cultural
identity is therefore not treated here as an independent explanation for the EU’s
collective decisions on membership eligibility.
5. Empirical design and methods
This part of the chapter outlines the empirical research design used to explain
how regional communities decide which states are eligible for membership. It
starts with a justification of the focus on Europe—why it makes sense to
investigate the politics of EU enlargement and why doing so should yield valid
insights with broader applicability. It then discusses how the country-cases are
identified and introduces the 48 country-cases that are the focus of the book’s
empirical investigation. It then distinguishes three distinct dependent variables
and the analytical methods used to explain them: discursive-process tracing of
EU membership norms over seven decades, statistical analysis of eligibility
outcomes in 48 country-cases, and causal-process tracing of eligibility decision-
making in three countries over time.
Regional case selection—why study Europe?
As indicated above, this book is focused empirically on regional community

building and membership decision-making in Europe. In addition to the intrinsic
importance of European integration for life within Europe and its relations with
the outside world, there are two reasons why it provides a good basis for
understanding the general dynamics of regional membership eligibility.
First of all, the strong pressures favouring alternative factors make post-1957
Europe a hard test of the membership norms argument. Given the novelty of the
European integration project, there were strong incentives for member states to
bolster their collective credibility by adhering closely to solemn commitments
they had made in the treaties. At the same time, the EU was born as an economic
community and always had economic integration as its core project, so
governments of the member states faced strong incentives to base their
membership decisions on assessments of economic fit or gain. On the other hand,
given how the rise of fascism in the 1930s had demonstrated the fragility of
democratic rule, the strength of communist parties in the post-war period created
a strong incentive for EEC member states to do whatever they could to bolster
democracy, such as limiting membership to democratic states that protect human
rights. Finally, with the Cold War raging, the Warsaw Pact’s forces never far away,
and the United States constantly pressuring its allies to reinforce NATO solidarity,
west European governments faced strong incentives to base their EC membership
decisions on geopolitical calculations. These incentives were sustained after the
Cold War by intense ethnic conflict in states bordering the EU, massive refugee
flows, and Russian military adventurism. In this complex and potentially over-
determined environment, it is far from obvious that membership norms would
have any significant impact on member states’ decisions regarding enlargement.
The second reason is purely empirical. If one aims to test competing
explanations of decision-making on membership eligibility, there have to be
enough cases to provide variation on the dependent variable and on possible
explanatory variables, as well as sufficient transparency for the researcher to
access the relevant data. The EU is ideal in both respects, given the large number
of non-member states that have sought to join the club over seven decades, some
successfully and some not, plus the relative accessibility of official records and the
intense media coverage of these developments since the beginning.
The study thus rests on a deep and varied empirical foundation. At its core lie
the most primary of primary sources in EU scholarship: preparatory studies and
preliminary drafts (travaux préparatoires) and transcripts (procès verbaux) from
meetings of the member states, the internal notes of member states and EU
institutions, as well as diplomatic correspondence and position papers found in
the historical archives of the Belgian, French, and German foreign ministries, the
Council of the European Union, the European Commission, and the European
Parliament, plus the Historical Archives of the European Communities located at
the European University Institute in Florence and the European Union Delegation
Collection held at the University of Pittsburgh. Other important documents were
found in the University of Pittsburgh’s online Archives of European Integration
(aei.pitt.edu), the University of Luxembourg’s online Centre Virtuel de la
Connaissance sur l’Europe (www.cvce.eu), and the Fondation Paul-Henri Spaak,
an independent but now-defunct institute in Brussels. These archival sources
were supplemented by interviews with senior EU officials, memoirs, speeches,
and other statements by key decision-makers, as well as print and electronic
media reporting on events as they unfolded, statistical data from the World Bank
and the Center for Systemic Peace, and some secondary historical sources.
Country-case selection—which cases and why?
As discussed above, this book aims to explain EU decision- making on

membership eligibility in situations where a non-member state has communicated
its interest in joining the club. Unfortunately, although the EU has more
experience with such situations than any other regional organization, there is no
official list of cases and outcomes that can be downloaded for analysis. The
identification of EU decisions on membership eligibility is therefore an act of
historical interpretation: the analyst must determine why a particular set of
behaviours and communication constitutes a case and what was the outcome of
each case.
Working with the aforementioned sources, I have identified 48 cases during
the period 1957–2019 in which the EU took a position on the membership
eligibility of a non-member state, including 37 cases where the state was deemed
eligible for membership and 11 cases where it was deemed ineligible. The book
uses three statistical techniques to analyse all 48 cases, and then examines in
detail EU decision- making on four states whose pursuit of community
membership extended over decades: Greece, Spain, Turkey, and Ukraine.
Dependent variables and analytical methods
In the simplest sense, the dependent variable and key unit of analysis in this study
is the EU position on the membership eligibility of an applicant state. But in order
to evaluate the various possible explanations introduced above, the study distin-
guishes between two dimensions of EU decision-making: the content of positions
on membership eligibility since 1957 and the process by which these decisions
were made in critical cases over time. Studying both dimensions yields more valid
insights into EU decision-making than studying either one alone. In addition, as
a foundation for testing the membership norms explanation, it is necessary first
to trace the evolution of EU membership norms over time. Addressing these
three empirical challenges necessitates a broad range of analytical methods whose
rationale and application are elaborated below.
Genealogy of membership norms

The first challenge is to establish the empirical reality and historical evolution of
the book’s core explanatory variable, the EU’s membership norms. To that end,
the study draws on extensive archival evidence of the language used by member
states’ representatives, European Commission officials, and members of the
European Parliament, speaking publicly or behind closed doors in official
deliberations regarding the characteristics that distinguish the Union’s member
states from others in the international system. To supplement these archival

sources, we turn to interviews, memoirs, speeches, and other statements by key
decision-makers, as well as print and electronic media accounts of the events in
question. The latter are especially important for more recent periods, where
access to primary sources is very limited.
The most promising approach to interpreting this evidence is a genealogical
method. In the social sciences, genealogy is used ‘to illuminate the contingency of
what we take for granted, to denaturalize what seems immutable, to destabilize
seemingly natural categories as constructs and confines articulated by words and
discourse and to open up new possibilities for the future’ (Crowley 2009: 341).
The particular genealogical method used here is discursive-process tracing. The
discourse analysis in this study involves reading confidential papers, speeches,
debates, and declarations from EU institutions, member states, and senior officials
in order to identify which representations of the nature and purpose of the
political community prevail in various periods of time (Neumann 2008). The
process tracing aspect involves identifying the conditions that appear to
strengthen or weaken prevailing norms, as well as the rhetorical manoeuvres that
key political actors use to maintain the salience of a dominant representation, to
promote an alternative or insurgent representation, and to frame particular
political choices in terms of their preferred representation (Blatter and Haverland
2012, Bennett and Checkel 2015). Discursive-process tracing is thus ideally suited
to investigating how political actors conceive and reconceive of eligibility for
membership in a supranational regional community.
Using this technique, we would conclude that an EU membership norm exists
when we observe that the collective deliberations and public pronouncements of
member states express a consistent definition of the constitutional, policy, or
other characteristics that current and aspiring member states are expected to
have. A change in membership norms is evident when such pronouncements
express a new definition, including when community members choose not to
contradict a novel definition by a relevant actor and then adjust their discourse
on membership eligibility to suit the new definition. However, exceptional
discourse by a minority within the community does not invalidate a norm as long
as it is criticized or at least not adopted by other community actors. Such evidence
is especially powerful when the normative discourse contradicts the material
interests or demonstrated policy preferences of the actors involved. The opening
of Chapter 3 provides more details on this analysis.
Statistical analyses of eligibility positions

The second analytical challenge is to assess the explanatory power of the
membership norms argument against that of competing arguments regarding the
membership eligibility of applicant states. This assessment is framed in terms of
the relative fit between the expectations of the various arguments and the
positions observed in the 48 country-cases (i.e. episodes where the EU took a
position on the eligibility of a particular state). Notwithstanding the nuances of
each decision, these outcomes are reduced to a binary variable (eligible–ineligible)
in order to facilitate comparison of the various arguments. Data on the outcome
in each case comes from the same sources indicated above for the genealogy of
membership norms. Variations over time in the membership norms themselves
are drawn from the preceding analysis. Data on the other arguments’ explanatory
variables comes from the EUR-Lex database (treaties), Polity IV and V-Dem
(regime type), the World Bank (commercial interests), and secondary sources
(geopolitical threats).
Once the data are assembled, we evaluate each case to determine whether the
outcome fits the expectations derived from the various hypotheses in light of the
relevant values on their respective explanatory variables. The cases are then
analysed using three complementary methodologies. First, the correlations
between explanatory variables and outcomes are presented in cross- tab or
contingency table format: this provides an initial assessment of their relative
explanatory power. Second, as a check on the robustness of the initial findings, we
analyse the same cases and outcomes using logistic regression. Third and finally,
the cases and outcomes are analysed using Qualitative Comparative Analysis,
which offers yet another robustness check. This combination of techniques is well
suited to maximizing the validity of conclusions regarding each argument’s
explanatory power. The opening of Chapter 4 provides more details on this
analysis.
Causal-process tracing of eligibility decision-making

Notwithstanding their clarity, the statistical techniques described above have
significant limitations. First of all, depending upon the alignment of explanatory
variables, competing theories may generate identical expectations about the
outcome of particular decisions, and this makes it hard to weigh their relative
explanatory power. More fundamentally, correlation is a poor indicator of
causation, especially in a relatively small population of cases. It is therefore
sensible to investigate the second face of community decision-making: the process
(rather than the outcome) by which EU member states and institutions make
decisions on the membership eligibility of applicant states.
The technique of causal- process tracing—as distinct from the discursive-
process tracing described above—focuses on the predictive accuracy of a theory’s
expectations regarding the mechanisms that link explanatory variables to key
outcomes. This technique is well established in the literature on European
integration (Schimmelfennig 2015), and it has recently been subject to extensive
methodological refinement (Blatter and Haverland 2012, Bennett and Checkel
2015). Causal- process tracing is therefore ideally suited to assessing the
explanatory power of various theoretical arguments regarding membership

eligibility.
Each of the six explanations introduced above includes certain expectations
about how a regional community (in this case, the EU) makes decisions on
membership eligibility, ranging from normative empowerment to legal
entrapment to selection by identity to rational calculation with veto option. Each
of these expectations has observable implications that are subject to empirical
verification. One can therefore weigh a theoretical argument’s explanatory power
by assessing how closely its expectations regarding causal mechanisms fit the
processes observed in a set of cases. To do so, we depend on empirical evidence of
EU decision-making processes—how agendas were crafted, who participated and
what arguments they used in negotiations, what information was available to
decision-makers, etc.—found in official archives, media reports, and participants’
memoirs. However, given the quantity of evidence necessary to conduct such an
analysis, and the limitations on access to this evidence in some cases, it is not
feasible to trace the decision-making process in all 48 of the country-cases.
This is why the process-tracing narrative is structured by the evolving norms of
membership eligibility: each of the process- tracing chapters (Chapters 5–8)
examines a single normative era and investigates how the norm prevailing in that
era shaped decision-making on several applicant states. Such a structure makes it
easier to identify the similarity in EU positions across countries within the same
normative era and to judge the historical interdependence of contemporaneous
country-cases. Of course, developments regarding these countries are not truly
independent of each other: EU decision-making on one country is likely to
influence subsequent decision- making on another country. To capture this
dynamic, the discussion of one country sometimes includes brief references to
decisions on another country. Finally, to assess whether (and if so, how) the EU’s
position on the eligibility of particular applicant states changed in response to
changes in membership norms, one has to compare decisions on that state across
historical chapters.
6. Plan of the book
Following this introduction, Chapter 2 is the theoretical heart of the book. It

elaborates the foundations, expectations, and observable implications of the
aforementioned theoretical arguments regarding how regional communities
decide on the membership eligibility of non-member states.
The next two chapters involve empirical analysis of EU membership outcomes
over time, starting with the membership norms themselves and then their relative
impact on EU decision-making. Chapter 3 uses genealogical methods to identify
four eras in the evolution of EU membership norms over seven decades. From
the signing of the Treaty of Rome in 1957 through the end of 1961, the EEC
defined itself as a community of non-Communist states and expected any state
seeking membership to fit this norm. Through the remainder of the 1960s, the
EEC defined itself as a community of non- Communist parliamentary
democracies. Starting about 1970 and continuing for another thirty-five years,
the EEC (and later EC then EU) defined itself as a community of liberal
democracies. Most recently, since about 2006, the EU has lacked a dominant
conception of itself as a political community and thus of what type of state is
eligible for membership.
Chapter 4 then uses three empirical methods—cross-tab correlation, logistic
regression, and crisp set Qualitative Comparative Analysis—to test the fit between
the six theories’ expectations regarding EU decisions on membership eligibility
and the actual decisions that were taken in the 48 cases. Both individually and
collectively, the three methods provide a strong empirical basis for the argument
that membership norms play a significant role in decision-making, arguably even
more significant than commercial or geopolitical interests. (The book’s Appendix
elaborates the techniques used to impute a small number of missing values for the
statistical analyses in this chapter.)
Building on this finding, Chapters 5–8 trace the process of EU decision-making
regarding the membership eligibility of four countries—Greece, Spain, Turkey,
and Ukraine—across four time periods. Chapter 5 examines decision-making on
Greece, Turkey, and Spain in the years 1957–1961, when the prevailing
membership norm defined Europe as a community of non-Communist states.
Chapter 6 examines decision-making on Greece, Turkey, and Spain in the years
1962–1969, when the prevailing membership norm defined Europe as a commu-
nity of parliamentary democracies. Chapter 7 examines decision-making on
Greece, Turkey, Spain, and Ukraine in the years 1970–2005, when the prevailing
membership norm defined Europe as a community of liberal democracies.
Chapter 8 examines decision-making on Turkey and Ukraine since 2006, when
no single normative definition of Europe has prevailed. Overall, these chapters
reveal a strong fit between the procedural expectations of the membership norms
argument and the actual decision-making process as it unfolded over time.
In conclusion, Chapter 9 summarizes the empirical findings and considers
their implications for the future course of EU decision-making on enlargement,
for the behaviour of regional organizations in other parts of the world, and for
future scholarship on inter-governmental decision-making and the construction
of regional communities.
2
Explaining Membership Eligibility
[A region] is not a geographic fact that has sociological consequences,

but a sociological fact that takes geographic form.
—Christopher Hemmer and Peter J. Katzenstein, 20021
Despite the broadening and deepening of regional cooperation and integration

across Africa, the Americas, Asia, and Europe over the last seven decades, scholars
differ on the meaning of ‘region’ and ‘regional community’ and on how regional
communities determine which or what kinds of states are eligible for membership.
This chapter elaborates the book’s argument on these points, with a particular
focus on the limits of regional community and thus on eligibility for membership.
After elaborating this logic and its observable implications, the chapter then
compares it to alternative explanations of membership eligibility focused on
physical geography, treaty rules, regime type, commercial interests, and geopolitical
threats.
In short, the book defines a regional community as a group of states that
identifies with a particular geographic region and acts collectively through an
inter-governmental organization. It argues that the potential geographic limits of
such communities are defined by membership norms prevailing amongst the
leaders of the member states and senior officials within the regional organization.
And like a physical border, but perhaps more so, a membership norm ‘defines
who you are and how you relate to those on the other side’ (Timmermans 2016:
8). These norms legitimate certain states’ membership in the club and distinguish
them from other states. Membership norms also shape decision-making within
regional organizations by empowering advocates of applicant states that fit the
norm and disempowering advocates of applicant states that do not fit the norm.
Yet notwithstanding their importance, the membership norms of regional
communities are often contested and subject to change over time.
1. A constructivist theory of regional community
This section of the chapter outlines a constructivist theory of regional community

and membership eligibility according to which the EU is best conceived as the
1 Hemmer and Katzenstein 2002: 587.
organizational representative of an imagined community of states whose limits

are defined by membership norms that change over time. The theory is
constructivist in the sense that it emphasizes the intersubjective interpretation of
geography as the basis for community formation and assumes that political actors
are sensitive to community norms when making behavioural (i.e. policy) choices.
However, its focus on social facts and the assumption that actors are sensitive to
norms should not be mistaken for the stronger constructivist claim (not made
here) that actors’ interests and policy preferences derive simply or largely from
their roles and interactions within the community.
The nature of regions and regional community
Many social scientists emphasize geographic proximity in their definitions of

region and regionalism. By this logic, a region is composed of states whose
borders touch or lie close to each other. However, ideas, institutions, and political
practice play a critical role in constituting the identity and limits of regions
(Mansfield and Milner 1997: 3–4; Barnett 1998; Buzan and Waever 2003: 27;
Katzenstein 2005: 6–13; Powers and Goetz 2011). The geographic extent of a
region thus cannot usefully be conceived in materialist terms, as if physical
location alone defined it. As Rumelili (2004: 39–40) explains, ‘the geographical
parameters of Europe have not only shifted throughout the centuries but also
within the short history of the European “community” as well. Therefore,
demarcating a certain geographic area as Europe has only been possible by
discourses and practices of differentiation that have also historically shifted and
changed.’
In fact, geographic proximity cannot explain some of the most puzzling aspects
of European (regional) community building. To start, states that are direct
neighbours of EU member states and not plausibly part of any continent other
than Europe have been excluded at various points from joining the process of
regional integration embodied in the EU. Europe also lacks any topographical
border that separates it indisputably from Asia, which creates intense
controversies over the Europeanness of Turkey and Ukraine, to cite just two
examples. And while the European continent may seem to be clearly distinguished
from Africa and southwest Asia by the Mediterranean and the Bosphorus, and
from the Americas by the Atlantic, this cannot explain why Greenland was once
part of the EEC or why Cyprus, the Canary Islands, and even New Caledonia
are part of the EU today.
These puzzles reflect the fact, as Edward Said (1978) recognized in his critique
of Western views of the East, that geography is meaningful and thus politically
consequential principally through the ways in which it is imagined. In turn, he
argued, the discourses that frame and reproduce these ‘imaginative geographies’
Explaining Membership Eligibility 29
are themselves sources of power for political actors with their own motives and
goals. In the same spirit, Benedict Anderson (1983: 7) defined the nation, one of
the most familiar building blocks of political geography, as ‘an imagined political
community . . . imagined as both inherently limited and sovereign’. Limits are thus
inherent to the definition and the lived experience of community: ‘The nation is
imagined as limited because even the largest of them,’ said Anderson, ‘has finite, if
elastic, boundaries, beyond which lie other nations.’
But to assert that all political communities have imagined limits is not to claim
that these limits are somehow false: they are ‘distinguished not by their falsity/
genuineness, but by the style in which they are imagined,’ wrote Anderson (1983:
6–7). Similarly, William Rogers Brubaker (1990: 380) defines the nation as ‘an
idea—and an ideal: it is a distinctive way of characterising and evaluating political
and social membership’. More recently, Roger M. Smith (2003: 20) defines a
‘political people or community [as] a potential adversary of other forms of human
association [whose] obligations legitimately trump many of the demands made
on its members in the name of other associations’. In other words, the imagined
limits of a political community may be near or far, and they may change over
time, but they shape the behaviour of its members and those who seek to join.
Anderson, Brubaker, and Smith were speaking principally about communities
composed of individuals, but their logic helps us to understand the construction
and limits of groups or communities of states. A regional community is a group of
states who identify collectively with a particular area of the world and whose
norms and rules compete with and override some of the demands on those states
deriving from other identities or institutions. These norms and rules are typically,
though not necessarily, embedded in a formal international organization designed
to govern the community and represent it to the outside world. The limits of such
a regional community are determined by the collective imagination of the
governments that comprise the community at any given time.
The nature of membership norms
So how do regional communities define which states are eligible for membership
and which are not? The governments that create a regional community often
emphasize what they have in common, such as location, cultural heritage, values,
or material interests, but building and sustaining a regional community of states
is as much about establishing difference as it is about reinforcing commonality.
After all, no group of states can claim that certain characteristics distinguish its
members from others without referring implicitly or explicitly to other states that
lack the same characteristics, either entirely or in sufficient measure. However,
this differentiation need not assume the irreconcilability of identities: the
community of states that differentiates itself from others in terms of certain
characteristics may be actively involved in overcoming the distance between itself

and the others that it considers potentially assimilable (Rumelili 2004: 37–8).
To analyse this process, we must look beyond the classical definition of
membership norms as ‘standards for including or accepting a person within a
group or social position’ (Cancian 1975: 3). The problem with the classical
definition is that the standards to which it refers could take the form of beliefs
widely held by political elites, convergent expectations that may or may not be
grounded in beliefs, or treaty rules whose negotiation was shaped neither by
beliefs nor by convergent expectations, and each of these has a different source
and different consequences for action (see Sartori 1970 on the danger of ‘concept
stretching’). Only by replacing this amalgamated concept with a more precise
definition can we possibly assess the content and effects of membership norms in
an actual community of states.
As scholars have long recognized, a key role of international organizations is to
define, communicate, implement, and enforce the shared values of the states that
they represent (Abbott and Snidal 1998: 24). The most authoritative constructions
of regional commonality and difference emerge from deliberations within the
international organization connected to the regional community in question. The
limits of a regional community—that is, which states are currently or potentially
eligible for membership—are thus defined in concrete terms by discourse and
debates within the regional organization over principles of membership eligibility
and their application to particular states.
Membership norms are best defined as the characteristics by which the
members of a regional community collectively define what they have in common
and what distinguishes them from other states in the international system. These
distinctions between ‘us’ and ‘them’ are socially convergent, and thereby qualify
as norms, as long as member states’ governments, regardless of their beliefs and/
or policy preferences, recognize them as the agreed definition of the community’s
identity and thus as the standards by which other members of the community will
judge their choices regarding membership and enlargement.
In concrete terms, a membership norm exists when the community’s
pronouncements express a common definition of the shared identity of the
community’s members and thus of the characteristics that membership applicants
are expected to have. A change in membership norms is evident when most such
pronouncements express a new definition, including when community members
choose not to contradict a novel definition by a high-profile actor and then begin
to speak about membership in similar terms. A regional community ceases to
have a membership norm when its members signal in their discursive behaviour
that they no longer recognize a shared definition of membership eligibility that
they are bound to respect.
Membership norms may be related to formal membership rules such as those
found in inter-governmental treaties, but the two are distinct logically and often
empirically, so formal rules are better treated as the basis for an alternative
explanation of decision-making. EU membership norms should therefore not be
confused with the Union’s formal rules on membership, which have been
modified several times since the 1957 Treaty of Rome. These treaty provisions
may be consistent with the membership norms prevailing at various times, but
not necessarily, and they may even be directly contradictory.
The contestation and evolution of membership norms
A regional community’s membership norms are deeply political constructions

shaped by historical experiences, ideologies, cultural legacies, the international
distribution of power, social mobilization, and the discourse of political
entrepreneurs. Like all international norms, they are subject to intense political
and ideological contestation and thus likely to change over time (Wiener 2014).
This contestation may lead to the reinforcement of existing norms or their
replacement by new norms (Sandholtz 2008), but contestation may also cause
norms to collapse without being replaced (Panke and Petersohn 2012).
Certain dynamics help to stabilize the content, applicability, and validity of
membership norms. As they emerge and are accepted by a regional community,
membership norms create incentives for community actors to accept certain
states for membership and reject others, regardless of their actual preferences.
After making a decision on whether or not to accept a particular applicant state,
policymakers may utilize a prevailing norm to justify their decision, thereby
reinforcing the salience of the norm and making it harder for later membership
choices to follow a different logic. Discursive contestation can also reinforce
existing norms by clarifying their intent and contribution to the community’s
purposes.
On the other hand, changes in a norm’s original determinants may increase
pressure for normative change, raising or lowering the salience of physical
geography, historical or cultural legacies, constitutional order, economic
structure, or state behaviour within the content of the norm. In addition, a
prevailing membership norm may not reflect the beliefs or policy preferences of
all the community’s member states’ governments and supranational actors. Given
the ideological, economic, and political interests at interests at stake, intra-
community deliberations on the eligibility of particular aspirant states are likely
to be hotly contested by the community’s supranational institutions and especially
by its member states, which could result in pressure to reform or even to abandon
prevailing norms. If they are sufficiently strong and sustained, debates over the
scope and validity of membership norms among member state governments (and
within the societies they represent) may result in the reform or breakdown of
international norms (Deitelhoff and Zimmermann 2020).
The following typology identifies four situations that may shape the creation of
regional membership norms or trigger pressure for change in existing norms, and
discusses how their dynamics are likely to differ. Although the list is not
necessarily complete nor mutually exclusive (actual cases may involve more than
one or other forces), it demonstrates the heuristic advantages of focusing on
membership norms rather than on an apolitical notion of regional identity as an
explanation of membership choices.
External challenge
External challenges to the territorial integrity, political character, economic
welfare or ideological commitments of participating states may motivate them
and other community actors to adopt particular membership norms or to contest
existing norms. Such normative changes may be part of an effort to mobilize
political support within the existing member states to resist the external challenge,
to facilitate the accession of states whose accession would help the community to
overcome the external challenge, or to exclude states whose accession would
undermine that effort. As shown in Chapter 3, this dynamic is evident around the
founding of the EEC in the late 1950s.
Internal challenge
Internal challenges to the legitimacy or constitutional principles of member
states, or to their governments’ re-electability or hold on power, may motivate
community actors to adopt particular membership norms or to contest existing
norms. In addition to current challenges, this may include memories of historical
experiences that community actors fear could be repeated. As above, these
normative changes may be designed to facilitate the accession of states with
similar constitutional values or to exclude states whose accession would
undermine the organization’s ability to resist the internal challenge it faces. As
shown in Chapter 3, this dynamic was evident several years after the EU’s
founding, when a coalition of European parliamentarians and trade unions
mobilized to insist on a reconceptualization of membership eligibility.
Rhetorical inconsistency
Gaps between a regional organization’s normative commitments and its policy
practice may motivate some community actors to contest prevailing membership
norms in order to make them congruent with recent practice on high-profile
issues. These normative changes may occur because recent practice led political
elites to rethink existing norms or simply because they wish to avoid the political
costs associated with gaps between rhetoric and practice. As shown in Chapter 3,
this dynamic was evident in the EU around 1970, when community leaders
revised the membership norm to reflect the growing salience of human rights in
international affairs.
Shift in public opinion

Changes in public attitudes toward a particular member or applicant state, or
toward people, values, or practices associated with a particular state or set of
states, may motivate some community actors to contest the validity of existing
norms in favour of an alternative definition of membership eligibility. Whether or
not this pressure is strong enough to overturn existing norms will depend on the
influence of the member state(s) where attitudes are changing and the strength of
domestic and community factions committed to retaining existing norms. As
shown in Chapter 3, this dynamic was evident in the EU around 2005, when
‘culture wars’ within the community and its member states undermined consensus
on longstanding membership norms.
Notwithstanding the variety of potential sources of normative change, it is
reasonable to expect the evolution of membership norms to fit Stephen Krasner’s
(1988) argument that social institutions and their effects exhibit a punctuated
equilibrium pattern over time, persisting for a while only to be overturned or
transformed by shocks and contestation that are difficult to foresee.
How membership norms matter
Membership norms are powerful social facts that matter a great deal in and
around regional communities of states. By defining the identity and limits of the
community and thus the socially appropriate characteristics of member states,
membership norms determine the legitimacy of various membership scenarios. If
current members of the community value their standing within the community,
and prospective members value accession, then the norms defining the criteria
according to which the eligibility of applicant states should be judged will be a
significant determinant of behaviour. This effect is not politically neutral: the
ability to deploy membership norms in political battles over the identity and
extent of a community of states represents ‘a very important category of power’
(Hurrell 2005: 40) that will have a differential impact on various actors, depending
upon how the prevailing norms fit their identities and policy preferences.
More precisely, membership norms shape the construction of a regional
community in three ways—via normative selection (norms as gates), normative
guidance (norms as roadmaps), and normative empowerment (norms as
resources and constraints). These three mechanisms are not mutually exclusive:
any or all of them may affect the politics surrounding the acceptance or rejection
of any particular application for membership.
Normative selection
By expressing the criteria that distinguish states that are eligible for community
membership from those that are not, membership norms create a gate between
current member states (who enforce the norms) and states that seek entry (who
must fit or comply with the norms if they wish to join). Membership norms thus
give members control over others who wish to be accepted as members. This
dynamic is evident in the profound adaptations of administrative structures, legal
codes, and long-standing political practices that many states made in order to
gain EU accession, although studies of this process rarely distinguish between the
effects of membership norms and the effects of formal membership rules (Glenn
2004, Vachudova 2005, Schimmelfennig and Sedelmeier 2005, Schimmelfennig
et al. 2006, Jacoby 2006, Böhmelt and Freyburg 2013).
Normative guidance
In addition to their significance for those seeking to join a regional community,
membership norms also affect the internal process by which a community of
states makes decisions on the eligibility of applicants. Policymakers committed to
the general goal of regional integration but uncertain about whether a particular
applicant should be allowed to join may rely on membership norms to guide their
choices. The process depends upon John Searle’s (1995) concept of ‘collective
intentionality,’ as reflected in John Ruggie’s (1998: 21) interpretation of the Bretton
Woods monetary negotiations, which he says ‘established intersubjective
frameworks of meaning that included a shared narrative about the conditions
that had made these regimes necessary and what they were intended to
accomplish, which in turn generated a grammar, as it were, on the basis of which
states agreed to interpret the appropriateness of future acts that they could not
possibly foresee . . .’ However, if we assume that norm compliance is cost-sensitive,
then these ‘frameworks of meaning’ would be most likely to shape behaviour
when the political or economic implications of a particular membership
application are unclear, at least to the policymaker in question.
Normative empowerment
Even if member states’ governments know what sort of enlargement outcome
they want and seek to achieve it through negotiations with other member states,
their success will depend in part upon the fit between their preferences and
membership norms prevailing within the community (Thomas 2009c, 2011). Even
those governments (presumably the majority) that do not turn to membership
norms for guidance find that prevailing norms have structured the relative
legitimacy of various membership scenarios, just as references to constitutive
legal norms are more powerful than purely political arguments (Rapp 2020).
Their instrumental action is thus shaped by a larger ‘logic of appropriateness’ that
can motivate actors to privilege social expectations over the immediate pursuit of
private preferences (March and Olson 1998). This dynamic would not affect the
balance of member states preferring to welcome or reject a particular applicant,
but it would affect how readily or effectively they pursue their preferences.
By defining the criteria by which applicant states are to be judged, membership

norms shape the distribution of authority within the community, empowering
member states and institutions whose preferences regarding community
enlargement fit the prevailing norms and constraining or disempowering those
whose preferences contradict the norms. As a result, membership norms make it
easier or harder for community actors to achieve policy outcomes consistent with
their respective preferences. Although this resource/constraint effect does not
guarantee any particular outcome, tilting the playing field for intra-community
negotiations in this manner creates a systematic bias in favour of applicants whose
candidacy can be persuasively portrayed as consistent with the prevailing
membership norm and against those whose candidacy seems inconsistent with
the norm.2
The ‘normative empowerment’ dynamic is clearly evident when one considers
how the fit between member states’ preferences and prevailing membership
norms affects intra-community discussions of an applicant state’s eligibility for
membership. In one scenario, three conditions are present: (1) applicant state X
fits the community’s membership norm; (2) member state A supports X’s
accession; and (3) member state B opposes X’s accession. In community
deliberations under these conditions, member state A is empowered and member
state B is disempowered. As a result, the community is likely to determine that
applicant state X is eligible for membership, the outcome preferred by member
state A. In a second scenario, the first of the three conditions is different: (1)
applicant state X does not fit the community’s membership norm; (2) member
state A supports X’s accession; and (3) member state B opposes X’s accession. In
community deliberations under these conditions, member state A is
disempowered and B is empowered. As a result, the community is likely to
determine that applicant state X is not eligible for membership, the outcome
preferred by member state B.
Notwithstanding their plausibility, the three mechanisms do not all play an
equal role in this study. Normative selection is very relevant to the behaviour of
applicant states, but not directly to this book’s focus on member states’ decision-
making, so it is not explored further here. Normative guidance is eminently
plausible in theory but extremely difficult to verify empirically, so it also is not
explored further in this study. In contrast, the logic of normative empowerment is
directly relevant to this study’s research question and empirical design, so it con-
stitutes the book’s core argument.
2 This ‘normative empowerment’ dynamic builds upon Schimmelfennig’s logic of ‘rhetorical

entrapment’ (2003) but nonetheless differs in three significant ways: it distinguishes the effect of
community norms from that of individual beliefs and treaty commitments; it emphasizes the role of
fellow member states rather than domestic publics; and its logic is explicitly probabilistic rather than
deterministic.
Membership norms and community decision-making
Given the previous discussion, one would expect the membership norms that
prevail within a regional community at any given time to shape its collective
decisions and actions on the membership eligibility of particular applicant states.
The core expectation here concerns which aspirant states are likely to be
recognised as eligible for membership and which are not:
Hypothesis 1: A regional community will decide that an aspirant state is eligible
for membership if it fits the community’s prevailing membership norm and will
reject states that do not, regardless of member states’ other motivations.
Of course, a regional community’s membership norms may change over time and
this could affect the community’s decision-making on a particular applicant state,
even if the preferences of the member states and the characteristics of the
applicant state do not change. As such, applicant states that are considered eligible
for membership in one period may be judged ineligible in a later period, or vice
versa, depending on changes in the fit between prevailing membership norms
and the characteristics of the applicant state.
Evidence of contrary behaviour is not necessarily evidence that membership
norms do not exist nor even that they do not matter. As John Ruggie (1998: 63)
explained, ‘we know deviation from regimes not simply by a categorical
description of acts that are undertaken, but also by the intentionality and
acceptability others attribute to those acts in the context of an intersubjective
framework of meaning’. A member state government that refuses to compromise
enlargement preferences (pro or con) that appear inconsistent with the community’s
membership norms is likely to be criticized by its peers for inappropriate behaviour.
In order to provide the criticizing states with ‘an explanation of [its] conduct in
terms of rules that they accept’ (Bull 2012: 43), the non-compliant government is
likely to attempt to justify its actions by referring to special circumstances. As
long as the violation is isolated and the ensuing condemnation is clear, this
process actually reaffirms the existence of the norm and its empowerment
(resource/constraint) effect will persist in subsequent cases. The expectation
would be as follows:
Hypothesis 2: Member states of a regional community will use prevailing mem-
bership norms to criticize apparently norm-inconsistent behaviour by a fellow
member, and the object of such criticism will reiterate its commitment to the
norm while justifying its actions in terms of extraordinary circumstances.
On the other hand, as discussed above, contestation of membership norms may

result in their deterioration and non-replacement. In such a situation, there
would be no community-level norms relevant to the membership eligibility of

non-member states. The community’s response to particular aspirant states would
therefore depend upon the distribution of preferences and bargaining power
among member states, as follows:
Hypothesis 3: When membership norms have deteriorated and not been
replaced, the member states of a regional community will engage in hard bargain-
ing over the eligibility of aspirant states.
2. Alternative explanations
It would be impossible to assess the explanatory power of the preceding argument

and hypotheses about membership norms without comparing their logic and
observable implications to alternative explanations emphasizing different
variables and processes. This exercise lays the groundwork for an empirical
assessment in later chapters of how well the expectations of the membership
norms argument fit actual EU positions on the membership eligibility of
particular applicant states (outcomes) and the decision-making dynamics by
which the EU arrived at its positions (process), and how well this fit compares to
the explanatory fit of alternative theories. To that end, this part of the chapter
introduces eleven more testable propositions based on five distinct alternative
explanations of EU decision-making on membership eligibility.
Physical geography
Perhaps the most obvious potential answer to the question of membership eligi-
bility follows the tendency in international relations to define regions in terms of
the geographic proximity of states (Buzan and Wæver 2003, Mansfield and
Milner 1997, Nye 1971, Pevehouse 2002). If a regional organization is designed to
integrate or govern a particular geographic space, such as a continent or portion
thereof, one might expect that any state located fully or even partly within this
space would be considered eligible for membership, and states located elsewhere
would be ineligible. Implicit in this argument is the assumption that a regional
community’s decisions on membership eligibility emerge through a kind of geo-
graphic automaticity rather than a rational weighing of costs and benefits or a
concession to social expectations. This is certainly a plausible interpretation of
the EU’s founding document, the 1957 Treaty of Rome, which declared that ‘Any
European state may apply to join’ without indicating any further membership
criteria.
There are several ways to test the relationship between geographic location and
membership eligibility. The first possibility is to assess the fit between decisions
on eligibility and whatever definition of the region’s limits prevails among
geographers, cartographers, and lexicographers. Assuming that such a convergent
definition can be found, the expectation would be as follows:
for membership if its territory is located physically within the geographic space
conventionally associated with the region in question, and reject others.
A second possibility would be to assess the fit between decisions on eligibility and
a definition of the region’s limits accepted by other, authoritative international
institutions. The expectation would be as follows:
that other, authoritative international institutions associate with the region in
question, and reject others.
A third version of the geographic argument would focus not on some a priori
definition of the region’s limits but on the non- member state’s geographic
proximity to the community’s current external frontier. By this logic, geography
matters but there is ultimately no final frontier of membership eligibility. The
expectation would be as follows:
for membership if its territory is contiguous to that of a current member state,
and reject others.
Treaty rules
Another possible explanation for regional organizations’ decisions on

membership eligibility focuses on the treaties that govern their operation. This
argument emphasizes actors’ responsiveness to social expectations, like the earlier
argument about membership norms, but here it is formal rules contained in the
community’s constitutive treaties that determine member states’ estimation of
what membership outcomes are socially appropriate and guide their choices
regarding applicant states. These treaties may include both procedural rules
stipulating how decisions on membership should be made and substantive rules
indicating what type of states are eligible for membership.
The significance of procedural rules on admission to membership should not

be under-estimated: different regional organizations have different rules with
distinctive implications for decision-making. For example, the Constitutive Act
of the African Union requires approval by a simple majority of member states to
admit a new member, so it is relatively difficult to block a state seeking
membership, which fits the AU’s substantive rule that ‘Any African state’ may
apply for membership. On the other hand, the EU’s founding act, the Treaty of
Rome, also stated that ‘Any European state’ may apply for membership, but it set
the procedural bar much higher, requiring that all member states must agree on
any applicant, which gave every member state a veto and thus made it far easier to
block states seeking membership. As explained by Walter Hallstein, first president
of the European Commission, this rule ‘was included in the Treaty not at the
behest of any single state, but by spontaneous and general agreement among all
the founder- members. And understandably so. For the admission of new
members . . . goes right down to the foundations on which the Treaty is built; it
touches the areas of sovereignty which the member states have not surrendered to
the Community but have retained in their own hands; it involves virtually the
conclusion of a new treaty with different members’ (Hallstein 1972: 68). The same
procedural rule applies to EU association agreements, which in some cases have
been formally presented as a stepping-stone to membership. Since then, the EU
has adopted a demanding substantive rule on membership but left the procedural
rule intact.
To complicate matters further, a treaty’s procedural and substantive rules on
membership may have conflicting implications. For example, a procedural rule
may give every member state a veto over applicant states, which invites them to
impose their preferences on community decisions, while a substantive rule may
stipulate that applicants of a certain type should be accepted or rejected, which
implies that member states should not always act on their preferences.
However, procedural rules alone cannot answer the question of which applicant
states will be considered eligible for membership. For example, a rule granting
veto power to every member state cannot explain decision- making on
membership eligibility without an explanation of why some member might
exercise its veto. For the sake of analytical clarity, it is therefore best to limit the
argument about treaty rules to the effects of substantive rules and reserve the
sources of national preferences, such as commercial interests and geo-political
vulnerability, for separate arguments presented later in this chapter.
A second possible point of confusion concerns which parts of a community
treaty count as a source of membership rules. Notwithstanding the often-lengthy
content of treaty preambles, it is important to remember that treaty drafters
choose carefully what text to place in the preamble rather than in a binding
article. If certain ideas appear only in a treaty’s preamble, this is because the
governments expecting to sign the treaty did not want those ideas to limit their
discretion in making policy choices. As such, and following standard practice

among international jurists (Yasseen 1976: 35), it is best to focus on the articles of
treaties signed and ratified by member states, and to treat preambular principles
as an interpretive aide as long as they do not contradict the apparent meaning of a
treaty article.
Finally, the argument that a regional community’s decisions on the member-
ship eligibility of applicant states are determined by substantive rules that the
community has codified in its treaties rests upon the assumption that member
states agree upon which rule is relevant to the case at hand and what it means. If
the rules are contested, as sometimes happens in international relations, then one
would have to attribute outcomes as much to the rhetorical framing and (de)
legitimation games of member states and other actors as to the content of the
treaty itself (Wiener 2004). But doing so would make this argument hard to
distinguish from the preceding argument about membership norms. For the
purposes of this study, we will assume that the formal rules of a regional
community are transparent and uncontested.
Nonetheless, the formal rules explanation of membership eligibility can
accommodate two distinct causal mechanisms: either the governments of
member states are persuaded by the prescriptive content of the formal rule or
they value the credibility of the community as a whole and the integrity of its
decision- making process above their own preferences regarding particular
membership applicants and above whatever norms may prevail within the
community. In either case, this logic suggests that a regional community will
accept applicant states that satisfy its membership rule and reject those that do
not, and that changes in treaty rules on membership will be followed by a new
pattern in membership outcomes. Put simply, this discussion yields the following
expectation:
for membership if it fits the substantive rule on membership established by the
community’s core treaty, and reject others.
Regime type
Instead of looking to the community’s membership norms or treaty rules when

deciding on the membership eligibility of an applicant state, member states’
governments may be motivated by incentives connected to their own domestic
structures. States differ in many ways and variations in ‘regime type’ may be
consequential for how the member states of a regional community make decisions
on the membership eligibility of applicant states. Membership decisions at the
community level would thus be shaped by regime characteristics at the state level.
This fits the logic of republican liberal approaches to international relations,
which emphasize the international effects of variations in social values and
domestic institutions (Moravcsik 1997: 530–3).
States contemplating the creation of a regional community or deliberating over
the accession of a new member may view the outcome as an opportunity to
consolidate their own position domestically and/or to spread their values abroad
by ensuring the prevalence of states like theirs within the community and keeping
dissimilar regimes out. If so, then we would expect that when member states all
have approximately the same regime type, the regional community will welcome
applicant states with a similar regime type and reject those with a significantly
different type. (Not all differences in regime type may matter equally: significant
differences such as democratic versus autocratic or secular versus theocratic may
matter more than minor differences such as parliamentary versus presidential
democracies, social market versus liberal market economies, or even personalistic
versus bureaucratic autocracies.) In any case, the decision-making process would
be selection-by-identity—a phenomenon well known in cognitive psychology
(Tipper et al. 1990)—without reference to social expectations or the likely costs
and benefits associated with each potential new member.
For example, a community composed of monarchies may be reluctant to admit
a democracy whose presence could call their own legitimacy into question.
Likewise, one might expect a community composed of democratic states that
protect human rights to refuse to consider membership for non-democratic or
rights-abusive regimes. If this were correct, then EU member states, which have
always been democracies, would consistently and explicitly limit membership to
other democracies and reject any non-democratic state’s expression of interest in
accession. Over time, as EU member states consolidated the protection of
individual and minority rights that characterize liberal democracy, we would
expect them to welcome membership applications only from other liberal
democracies. This discussion yields the following expectation:
for membership if its regime type is similar to that of the member states, and
reject others.
Commercial interests
Another approach to explaining decision- making in a regional community

emphasizes the impact of economic incentives on member states’ policy
preferences. This approach is well established in explanations of the EU’s founding
treaties (Milward 1992, Moravcsik 1998) but less common in studies of

enlargement. The challenge is to derive from this approach an explanation of
decision-making on membership eligibility that is generalizable to a broad range
of cases and empirically testable.
Moravcsik and Vachudova (2005) argued that the EU supported eastern
enlargement in the late 1990s because the then-15 member states expected that
they would benefit economically from including additional states in the commu-
nity. Their argument was based heavily on calculations by Richard E. Baldwin et al.
(1997: 168) that enlargement would be ‘a phenomenally good bargain for the
incumbent EU15’. More recently, Baldwin (2007) has written that geographic cri-
teria for membership are largely subjective and political criteria do nothing more
than delay the accession of states determined to join. Instead Baldwin and others
expect the commercial interests of a regional community’s member states to
determine their preferences regarding the possible accession of additional states.
As plausible as this expectation may be, it requires further specification.
In regional communities whose member states share a customs union or single
market, there would appear to be a strong incentive for member states to base
their preferences regarding possible enlargement on their calculations of how
integrating an applicant state would affect their own economic welfare. If the
likely balance of gains and losses is positive, then a member state would be
expected to welcome an applicant and support an affirmative response on its
membership eligibility by the community. This fits the conclusions of
Edward D. Mansfield and Jon C. Pevehouse’s (2013) study of the incentives that
members of a preferential trade agreement (PTA) face when deciding whether or
not to incorporate new member states: by enlarging the PTA’s internal market,
expansion creates new investment and export opportunities for firms within the
original member states, improves the members’ collective clout in multilateral
trade negotiations, and reduces their vulnerability to beggar- thy-
neighbour
policies by non-member states. And it is likely true that welcoming additional
states into a community with an existing ‘common market’ or ‘single market’
would accomplish this goal more efficiently than negotiating bilateral reductions
in trade barriers.
On the other hand, the gains and losses from an expended market would not
be distributed evenly among member states: member states with the greatest
ability to trade with and invest in an acceding state would benefit more, while
those whose leading productive sectors would face increased competition would
benefit less. Relative gains and losses would also be shaped by the effect that a
new member state would have on the existing level and distribution of the
community’s budget (in the EU case, principally agricultural and rural
development funds plus structural funds for disadvantaged regions). This too is
hard for member states to predict because they cannot easily foresee the
bargaining that would ensue once an applicant gains membership. The calculation
is also complicated by the fact that any member states whose government is
subject to elections (i.e. democracies) face a powerful incentive not to adopt
positions on enlargement whose sectoral economic effects would alienate key
domestic political constituencies.
In addition, some regional communities, including the EU since 1957, have the
option of offering aspirant states access to an integrated market but not full
membership, which allows member states to benefit from expanded markets
without alienating vulnerable constituencies or diminishing their own voting
power in regional institutions or losing access to community funds. Inviting non-
member states to form a free trade area or a customs union is thus a credible
alternative with lower costs and less commitment over time than opening the
door to community membership.
Assuming that each member state wields a veto over each applicant state,
community-level decisions on membership eligibility will therefore be positive
only if all members expect soon to benefit more economically from enlargement
than they would from offering a less costly form of association or the winners are
able to compensate the losers with gains in another policy area. With this in
mind, and given that cooperation games with multiple veto players often result in
non-agreement (Tsebelis 2002), we would expect a high percentage of membership
applicants to be rejected at the eligibility stage. However, the full argument cannot
be tested without access to detailed evidence on each member state’s economic
calculations on each applicant state, which far exceeds the scope of this study
(if they exist at all).
Alternatively, one can focus on how an applicant state’s level of economic
development compares to that of the member states. Relative economic
development has a major impact on the changes in economic competition and
investment opportunities that any community enlargement will yield. And given
the close correlation between economic development and administrative
professionalism (Evans and Rauch 1999, Mauro 1995), it also affects the applicant
state’s ability to ensure a level playing field by implementing economic and social
regulations comparable to those prevailing within the community. One possibility
is that member states welcome the expanded market offered by applicants with a
similar or higher level of development but fear the administrative incompetence
of less developed states. The contrary possibility is that member states welcome
less developed applicants because of the opportunities they offer for cheap labour
and lucrative investment, but reject similarly or more developed states out of fear
of the competition that they would bring. This logic yields the following, mutually
exclusive expectations:
for membership if its level of economic development is similar or superior to that
of member states, and reject others.
for membership if its level of economic development is below that of member
states, and reject others.
Security interests
Finally, a regional community’s decisions on membership eligibility could be

driven by security concerns. The realist argument that states must forever be
vigilant about threats to their security (Waltz 1979, Mearsheimer 2001) suggests
that decisions on regional integration will be shaped by geo-political risks. For
example, building on Pape’s (2005) concept of ‘soft balancing,’ Rosato (2011)
argues that national leaders opted for European integration in response to
uncertainty regarding the US attitude toward the expansion of Soviet power,
while Posen (2004) attributes the Europeanization of security and defence policy
to the threat posed by US unipolarity in the post-Cold War era. Similarly, recent
studies of accession to multilateral economic institutions (Davis and Milf 2017,
Davis and Pratt 2020) find that member states’ assessment of geo- political
alignment is more important than considerations of commercial functionality.
This logic suggests that a regional community’s member states would calculate
geo-political risks when making decisions on membership eligibility. The more a
member state is exposed to security risks, the more intense would be its
preferences regarding a particular applicant’s eligibility for membership
(Skålnes 2005). If every member state has a veto on enlargement decisions, the
security implications of enlargement would presumably have to be positive or at
least not strongly negative for all current members in order for an applicant state
to be deemed eligible. Regional communities would then accept as eligible for
membership a non-member state whose accession would reduce member states’
geo-political vulnerability or improve their capability for resisting geo-political
threats, and reject those that would significantly reduce the security of any cur-
rent member.
One version of this logic focuses on exposure to inter-state conflict or unman-
ageable cross-border movements. As Donno et al. (2015: 253) argue, research
suggests that ‘unstable, conflict prone, or transitional states are those that benefit
most from joining an IGO [inter-governmental organization], but from the IGO’s
perspective, these are also the states that are the most costly to integrate.’ In an
insecure world, one would therefore expect regional communities to be most
sceptical about integrating applicants that pose a security threat. In other words:
for membership if it does not pose a significant risk of involvement in inter-state
conflict or cross-border violence, and reject others.
Table 2.1 Summary of explanations and hypotheses
Explanation Hypothesis
Membership 1. A regional community will decide that an aspirant state is eligible for
norms membership if it fits the community’s prevailing membership norm
and will reject states that do not, regardless of member states’ other
motivations.
Membership 2. Member states of a regional community will use prevailing
norms membership norms to criticize apparently norm-inconsistent
behaviour by a fellow member, and the object of such criticism will
reiterate its commitment to the norm while justifying its actions in
terms of extraordinary circumstances.
Membership 3. When membership norms have deteriorated and not been replaced,
norms the member states of a regional community will engage in hard
bargaining over the membership eligibility of aspirant states.
Geography 4. A regional community will decide that an aspirant state is eligible for
membership if its territory is located physically within the geographic
space conventionally associated with the region in question, and
reject others.
membership if its territory is located physically within the geographic
space that other, authoritative international institutions associate
with the region in question, and reject others.
membership if its territory is contiguous to that of a current member
state, and reject others.
Treaty rules 7. A regional community will decide that an aspirant state is eligible for
membership if it fits the substantive rule on membership established
by the community’s core treaty, and reject others.
Regime type 8. A regional community will decide that an aspirant state is eligible for
membership if its regime type is similar to that of the member states,
and reject others.
Commercial 9. A regional community will decide that an aspirant state is eligible for
interests membership if its level of economic development is similar or
superior to that of member states, and reject others.
Commercial 10. A regional community will decide that an aspirant state is eligible for
interests membership if its level of economic development is below that of
member states, and reject others.
Geo-politics 11. A regional community will decide that an aspirant state is eligible for
membership if it does not pose a significant risk of involvement in
inter-state conflict or cross-border violence, and reject others.
Geo-politics 12. A regional community will decide that an aspirant state is eligible for
membership if it belongs to (or has a clear prospect of joining) the
security alliance to which most community members belong, and
reject others.
Alternatively, a similar realist concern might lead regional communities to focus

on how enlargement could affect their capability for resisting geo-political threats.
By this logic, enlargement would improve the security of member states if it
would strengthen the security alliance(s) to which they belong. Since the credibility
of mutual defence commitments is presumably strengthened by institutional ties
that bolster economic interdependence and mutual identification, a security
alliance would be strengthened as more of its members join a geographically
overlapping but otherwise unrelated regional community. In addition, a similar
effect could be expected for states whose path to alliance membership has been
recognized and institutionalized by the alliance.
for membership if it belongs to (or has a clear prospect for joining) the security
alliance to which most community members belong, and reject others.
To facilitate easy reference, the explanations and hypotheses outlined above are
summarized in Table 2.1.
3. Conclusions
This chapter has outlined the book’s principal argument about how membership
norms prevailing among member states of a regional community shape the com-
munity’s decision-making on the membership eligibility of other states seeking to
join the community, and then compared that logic to alternative explanations of
membership eligibility focused on very different determinants of decision-
making. Each explanation generates hypotheses that can be tested empirically.
The following chapters assess the relative explanatory power of the various logics
in light of detailed empirical evidence of EU decision-making on enlargement
over seven decades.
PART T WO
ME MBE R SHIP OU TC OM E S
3
The Evolution of EU Membership Norms
There have been—and still are—learned disputes as to the geographical

bounds of Europe. But Europe cannot possibly wait for definition, for
the end of that controversy; she does, in fact, define her own bounds by
the will of her peoples.
—Robert Schuman, French foreign minister, 19491
This chapter traces the emergence, contestation, and evolution of membership

norms within the European Union and its institutional precursors since the
Second World War.2 Its purpose is threefold: first, to demonstrate that these
norms have indeed changed over time, contrary to the assumptions of many
scholars and the claims of countless EU publications and pronouncements; sec-
ond, to illustrate the various dynamics of normative change outlined in Chapter 2;
and finally, to establish an empirical basis for investigating in later chapters how
prevailing membership norms have affected the community’s decision-making
on the eligibility of particular aspirant and candidate states.
EU decision-making on membership and enlargement has always been princi-
pally an intergovernmental competence, so the chapter gives most weight to how
the representatives of the member states speak publicly and behind closed doors
regarding the characteristics that should distinguish the community’s member
states from others in the international system. In order to reduce the risk that
member states’ preferences regarding particular membership choices give a mis-
leading picture of prevailing membership norms, the analysis discounts pro-
nouncements made with reference to a particular aspirant or applicant state. The
discourse of European Commission officials and members of the European
Parliament is considered a key potential source of normative contestation but
only as supplementary evidence of prevailing norms.
As shown below in more detail, careful tracing of the content and contestation
of normative discourse on EU membership eligibility reveals four critical
junctures and thus four normative periods of varying lengths. Despite competing
understandings of the purpose of European integration in the 1950s, the external
1 CVCE: Statement by Robert Schuman (London, 5 May 1949).

2 Parts of this chapter were published previously in Thomas (2017).
geo-political and ideological challenge of the Cold War led the governments of
the founding member states to agree that only and all non-Communist European
states were eligible for membership in the new community. This norm prevailed
in inter-governmental deliberations from the negotiation of the Treaty of Rome
through to the end of 1961. At that point, political entrepreneurship by federalists
within the European Parliamentary Assembly (precursor to today’s European
Parliament), supported by at least one of the member states, reframed the
community’s membership choices in a manner that member states with contrary
preferences could neither reject nor ignore. This resulted in a new norm that only
European parliamentary democracies based on free elections were eligible for
membership, which prevailed from 1962 through to the late 1960s.
Around 1970, deliberations within the community over how to respond to the
violent overthrow of parliamentary democracy in Greece led to a second
deepening of the preexisting norm: thereafter, only European liberal democracies
(parliamentary democracies based on rule of law and respect for human rights)
were considered eligible for membership. Finally, the breakdown of the famous
‘permissive consensus’ on European integration in the late 1990s, followed by the
politicization of immigration and national identity in the early 2000s, led a
number of member states to reject the pre- existing norm that all liberal
democracies in Europe were eligible for EU membership. As a result, no single
membership norm has prevailed in the EU since late 2005, which invites member
states to engage in hard bargaining and logrolling over the eligibility of particular
aspirant and candidate states.
1. A community of non-Communist states, 1957–1961
The landmark treaty establishing the European Economic Community (EEC)

that the governments of Belgium, France, Germany, Italy, Luxembourg, and the
Netherlands signed in Rome on 25 March 1957 called ‘upon the other peoples of
Europe who share their ideal to join in their efforts’. However, neither democracy
nor human rights were mentioned anywhere in the treaty—neither as a goal of
the new community nor as a requirement for membership or association. Instead,
the treaty’s article on membership declared simply, ‘Any European state may apply
to become a member of the Community.’ So what was the ‘ideal’ upon which the
Six had chosen to build their new community, and what additional states, if any,
were they ready to welcome as members? To answer these questions and to reveal
the forces determining this outcome, it is necessary to examine the debates and
negotiations that shaped the Treaty of Rome.
In the immediate aftermath of Germany’s defeat in 1945, the principal challenge
facing Europe was how to reconstruct the continent’s shattered infrastructure and
revive economic growth. The late 1940s and early 1950s was nonetheless a time of
The Evolution of EU Membership Norms 51
great ideological and political turmoil in western Europe. The confrontation

between liberal democratic states and authoritarian communist states that had
erupted over Berlin in 1948 then exploded into a shooting war in Korea two years
later, exacerbating fears about the intentions of local Communist parties and
about the region’s vulnerability to Soviet attack. In addition, western Europe was
divided between those who believed that the continent ought to be rebuilt on a
foundation of sovereign nation-states, those who believed that states should be
subsumed within a federal or supranational structure able to protect peace and
democracy, and those who favoured integration through ever-deeper economic
ties. All of these tensions shaped the launch of the EEC.
In the eyes of many democratic politicians and former members of the anti-
Nazi resistance, it was essential to ensure Europe’s new and restored democracies
against any revival of fascism. At a clandestine meeting in Geneva in May 1944,
one month before the allied landing at Normandy, nine national resistance groups
had called for the establishment of a federal union with supranational authority
that would be open to all European states able ‘to guarantee democratic
institutions and the free development of the human personality’.3 Advocates of
this approach called for bolstering democratic constitutions at the national level
with strong supranational institutions capable of preventing authoritarian
backsliding, including a membership norm limited to democracies that respect
human rights.In early May 1948, a congress of the nongovernmental European
Movement proposed the creation of a Council of Europe, a common declaration
on human rights, and an independent court of human rights. This new body, the
congress declared, ‘should be open to all democratic European nations which
undertake to respect fundamental human rights’ (Churchill and Spaak 1950: 48).
The following month, the European Parliamentary Union prepared its own draft
constitution for a federal United States of Europe whose ‘supreme goal is to realize
and guarantee in its States the fundamental rights of man enunciated in the
preamble of the United Nations Charter’. The same text also stipulated that
‘The Constitution of each member state must foresee the existence of a
Parliament of which at least one Chamber must be freely elected on the basis
of universal suffrage’.4
Following these proposals, the governments of Belgium, Denmark, France,
Germany, Ireland, Italy, Luxembourg, the Netherlands, Norway, Sweden, and the
United Kingdom signed the official Statute establishing the Council of Europe on
5 May 1949. Member states of this new body would be represented in a Council
of Ministers empowered to make decisions on the basis of unanimity while their
parliaments would send delegates to a Parliamentary Assembly. But membership
was limited expressly to democratic states that respect human rights and the rule
3 CVCE: Draft declaration of the European resistance movements (20 May 1944).
4 CVCE: Projet de Constitution fédérale des Etats-Unis d’Europe (Juin 1948).
of law: ‘Every member-state of the Council of Europe recognizes and accepts the
principle of the rule of law and the principle that everyone who comes under its
legal and sovereign authority is entitled to the basic human freedoms, and to
human rights . . .’ The Council’s commitment to these principles was strengthened
the following year by agreement on a European Convention on Human Rights
and Fundamental Freedoms, which stipulated the creation of a supranational
European Court of Human Rights and a Human Rights Commission to interpret
and enforce its rules.
Although separate from the institutional evolution towards and beyond the
Treaty of Rome, the Council of Europe established an important normative
benchmark against which political entrepreneurs often judged the EEC and its
successors in later years. It nonetheless bears noting that the governments
involved in drafting the European Convention on Human Rights were unable to
reach agreement on a clause guaranteeing democratic principles, institutions, or
political liberties. Despite arguments by the European Movement and others that
the Convention was designed ‘to prevent . . . the setting up in one or other of these
countries of a regime of the Fascist or Nazi type’, it proved impossible to agree on
a definition of the fundamental principles or institutions of democracy, due to
differences in the form of parliamentary democracy among the participating
governments and to British sensitivities about their overseas empire. The most
they could agree was an article in the first additional protocol to the Convention
establishing a right to free elections, and this only after the text was stripped of
the term ‘political liberty’ and other language suggesting that the people choose
their form of government (Kovács 2020: 239–42).
Many of Europe’s democratic politicians in this period were more concerned
about the challenge posed by the far-left than the far-right. Communist parties
were particularly strong in France and Italy, where they had led resistance to
fascism and occupation, and their strong ties to Moscow raised fears that the
egalitarian ideals they espoused would someday become a cover for efforts to
overthrow democracy, the rule of law and market- based economics. The
uncompromising behaviour of Communist parties in Soviet-controlled eastern
Europe, the Berlin crisis of 1948, and the outbreak of the Korean War in 1950
made this scenario all too imaginable. For politicians focused on the challenge of
Communism, both internally and externally, the surest solution lay in policies
that would boost the standard of living and deepen economic ties between the
countries of western Europe.
In May 1950, French Foreign Minister Robert Schuman declared his
government’s desire to create a High Authority to control all coal and steel
production in France and Germany. This proposal, he said, was intended ‘as a first
step in the federation of Europe’, and was to be ‘open to all countries willing to
take part’. ‘[T]he general objective assigned to the High Authority’, Jean Monnet
wrote a few weeks later, ‘can be summarized by the formula of raising the stand-
ard of living by increasing productivity’ (Stirk and Weigall 1999: 77).
On 18 April 1951, the foreign ministers of Belgium, France, Germany, Italy,
Luxembourg, and the Netherlands (the Six) signed the Treaty establishing the
European Coal and Steel Community (ECSC). From the first draft circulated by
the French in June 1950 to the final text signed ten months later in Paris, the
treaty made no mention of democracy or human rights, neither as a principle of
the community nor as a requirement for membership (United States Department
of State 1977: 727–39). Instead, the treaty declared ‘Any European state may apply
to accede.’ A treaty protocol identified close relations with the Council of Europe
as a priority for the ECSC, but membership in the Council was not a requirement
for joining the new community.
There was, however, one additional requirement expected of states seeking to
join: they had to be non-Communist. On the same day that they signed the ECSC
treaty, the six governments issued a joint statement declaring, ‘This Europe is
open to all European countries that are able to choose freely.’5 In the context of
the time, ‘able to choose freely’ (‘libre de leur choix’) referred clearly to Moscow’s
control of the foreign policy of states within its political orbit. In other words, the
new political community was not open to Communist states.
The tension between integration initiatives focused on bolstering democracy
and those focused on raising living standards remained unresolved for some
time. When French Prime Minister René Pleven proposed the creation of a
European Defence Community (EDC) to contain a rearmed Germany, Italian
federalists convinced Premier Alcide de Gasperi to insist that the EDC and the
just-launched ECSC be encompassed within a supranational European Political
Community (EPC) (Pistone 1993, Preda 1992). Responsibility for drafting an
EPC treaty was given to the members of the ECSC’s parliamentary assembly,
reconstituted for this purpose as an ‘ad hoc assembly’ under the presidency of
Paul-Henri Spaak.
The assembly’s draft treaty stated that the EPC’s first mission would be ‘to
contribute towards the protection of human rights and fundamental freedoms in
Member States’. It also incorporated the core provisions of the European
Convention for the Protection of Human Rights and Fundamental Freedoms as
‘an integral part of the present Statute’ and authorized the EPC to ‘intervene’
when a member state fails to maintain constitutional order and democratic
institutions within its territories. Most important, and in stark contrast to the
Treaty of Paris, the draft EPC treaty indicated that ‘Accession to the Community
shall be open to the Member States of the Council of Europe and to any other
5 CVCE: La déclaration commune des ministres signataires (18 avril 1951).

European State which guarantees the protection of human rights and fundamental
freedoms . . .’ and authorized the Community to ‘conclude treaties or agreements
of association . . . with such third States as guarantee the protection of the human
rights and fundamental freedoms’ (Constitutional Committee 1953a, see also
1953b). The EPC treaty would thus have excluded from the emerging European
community all states that did not respect democracy and human rights.
However, the lack of support among participating states for such a membership
norm was confirmed in 1954, after the French parliament blocked the EDC due to
concern that moves toward a European army would compromise national
sovereignty. Had the Six been committed to the EPC’s membership norms, they
could have retained them in some institutional form. Instead, they abandoned the
EPC in favour of a liberal vision of European integration focused on eliminating
barriers to trade. During the EPC negotiations in December 1952, Dutch Foreign
Minister Johan-Willem Beyen had circulated a memorandum proposing that ‘the
principal aim of European integration should be the raising of the general stand-
ard of living of the European peoples’. To this end, he favoured replacing the
ECSC’s sectoral integration plan with a general common market without national
customs duties or import quotas (Harryvan and Harst 1997: 71–4). In the same
spirit, a committee of senior German civil servants and scholars appointed by
German Economic Minister Ludwig Erhard had called in June 1953 for the creation
of a European common market characterized by the free movement of labour,
goods, and capital; a single currency; and common regulatory legislation.6
As Erhard (1963: 155–8) saw it, the failure of the EPC’s ‘mechanically calcu-
lated formula’ constituted an opportunity to pursue his vision of economic liber-
alization of the entire ‘free world,’ starting with the convertibility of currencies. So
when Beyen sat down with his Benelux counterparts Paul-Henri Spaak (Belgium)
and Joseph Bech (Luxembourg) in late 1954 to prepare a joint proposal to
relaunch European integration, there was little support in Europe’s capitals for
any formal provision limiting the process to democratic states. The influential
Benelux Memorandum of 1955 conveyed the same vision, proposing that the
ECSC member states adopt a general common market without national customs
duties or import quotas. Jean Monnet (1955: 62) described the attitude of the Six
at that point: ‘Our Community is neither a small Europe nor a limited Community.
Its limits are not set by us. They are set by those countries themselves who, for the
moment, do not join it.’
This tension between two competing visions of European integration and their
respective membership norms was soon resolved, however, by serious external
challenges. The Soviet Union’s threat to use force to repress political liberalization
in Poland in August 1956 and the Red Army’s invasion and crushing of an
6 MAEF: Europe: Généralités (1949–1955), dossier 78: Lettre de l’Ambassadeur de France en

Allemagne, 9 septembre 1953.
anti-Communist uprising in Hungary three months later, revived fears of a

Communist threat to democracy throughout western Europe. Meanwhile, the
USA’s insistence that Britain and France relinquish their military gains in the Suez
crisis, and those two countries’ inability to counter the Soviet invasion of Hungary,
clearly demonstrated the limited power of west European states in this period.
These developments bolstered support for a community membership norm that
would exclude Communist states while keeping the door open to all non-
Communist states, including some that would have been excluded by the
democracies-only norm.
In April 1956, less than a year after formal talks between the foreign ministers
of the Six ECSC states began at Messina, the Intergovernmental Conference on
European Integration proposed the creation of a European economic community
focused on the free movement of labour and capital.7 The conference’s first draft
of a treaty, prepared soon thereafter, made no mention of membership criteria.
Instead, it argued that it was ‘necessary to pursue the establishment of a Europe
united by the progressive fusion of national economies, the creation of a large
common market and the progressive harmonisation of their social legislation’.8 As
in the Treaty of Paris, no mention was made of limiting membership to states that
ensure democracy or human rights.
This was not because the negotiators were unaware of alternatives: the
conference chair, Paul-Henri Spaak, had also chaired the ad hoc assembly that
drafted the EPC statute in 1953. This awareness is clearly reflected in a 1957
working paper of the inter-governmental conference, which explicitly compared
the proposed membership clause of the new treaty to those included in the earlier
ECSC, EDC, and EPC treaties—the last of which would have limited membership
to ‘member states of the Council of Europe and all other European states that
guarantee the maintenance of human rights and fundamental liberties . . .’9 Despite
this side-by-side comparison, and the fact that adding EPC-like membership
criteria would not have prevented any of the Six from joining the common
market, the negotiators and the governments they represented chose not to limit
the potential scope of the new community in this manner. The Six simply did not
share the membership principles that had come closest to realization in the EPC
treaty and did not want their membership choices to be constrained by such
stringent criteria.
Frustrated federalist political actors outside the conference tried to force the
issue onto the agenda: ‘Modern history has seen the rise of the criminal state. It
7 AEI: Intergovernmental Committee on European Integration. The Brussels Report on the

General Common Market, June 1956.
8 HAEU: CM3-NEGO-99: Conférence intergouvernementale: projet de traité & CM3-NEGO-100:
Conférence intergouvernementale: projets d’articles.
9 HAEU: CM3-NEGO-262: Exemples de clauses d’adhésion prévues dans les traités internation-
aux, 24 fevrier 1957.
has been painfully brought home to us that the protection of our liberties
transcends national boundaries and that common measures are indispensable to
safeguard our freedom. . . . The European Community must be so constructed that
the fundamental liberties of our citizens can be effectively protected and
preserved. Not only member states as such, but also individuals and groups of
individuals, must be able to appeal when human rights are alleged to have been
violated,’ wrote Hans Nord, the Secretary General of the Netherlands Council of
the European Movement. As if hinting at developments to come, Nord added:
‘I stress the importance of this aspect of the European Community. For it may
become of the greatest value in dealing with those countries in Europe which at
present cannot participate in our joint efforts’ (Haines 1957: 223–4).
Nonetheless, the negotiators never included anything resembling the Council
of Europe or EPC criteria in any draft of the treaty to create the common market.
The Treaty of Rome’s preamble does express the signatories’ commitment ‘to
preserve and strengthen peace and liberty’, but evidence from the negotiating
record shows that the reference to liberty was understood principally in economic
rather than political terms.10 The consistent focus on economic integration and
the understanding of ‘free world’ prevailing in this period suggest that the
negotiators understood ‘liberty’ as a reference to free markets and trade within
the common market. Even 15 years later, when membership norms had clearly
changed, Hallstein (1972) continued to define the Treaty of Rome’s conception of
‘freedom’ as a commitment to the free movement of goods, services, and workers,
the free establishment of businesses, and free market dynamics regarding the
production and consumption of goods.
In the end, the treaty provision on membership (Article 237) read simply, ‘Any
European state may apply to become a member of the Community. It shall
address its application to the Council, which shall act unanimously after obtaining
the opinion of the Commission. The conditions of admission and the adjustments
to the Treaty necessitated thereby shall be the subject of an agreement between
the Member States and the applicant State. This agreement shall be submitted for
ratification by all the Contracting States in accordance with their respective
constitutional requirements.’
The treaty also included a provision (Article 238) allowing the EEC to sign
association agreements with other states or international organizations. While
Article 238 did not specify the possible purposes of association agreements nor
indicate any formal criteria for which states would be eligible, it was apparently
drafted to facilitate close ties with European states outside the EEC and with other
states (especially former colonies) with which the Six had long maintained close
relations. At the signing of the Treaty of Rome, the Six adopted Declarations of
Intent to conclude associations with Morocco, Tunisia, Libya, Italian Somalia,
Surinam, and The Netherlands Antilles, and many other states applied for EEC
10 HAEU: CM3-NEGO-182: Projet de rédaction du préambule, 14 janvier 1957, MAE 112 f/57.
association in the years that followed (Phinnemore 1999: 22). But as with regard
to membership, the treaty indicates that the Six preferred that their choices on
association not be constrained by formal criteria.
The point of these observations is not that the governments that signed the
Treaty of Rome cared little about democracy or human rights. However, the
absence of democracy and human rights from the treaty that is widely celebrated
as the source of today’s European Union was clearly no accident. Respect for these
values had already been stipulated as strict criteria for membership in the Council
of Europe and the aborted European Political Community, and they could have
been added to the Treaty of Rome without changing the function, structure,
or initial membership of the new community. The constitutions of the six
participating states were clearly based on democratic principles and all six were
committed to the Council of Europe and its European Convention of Human
Rights and Fundamental Freedoms. We are thus led inescapably to the conclusion
that the founders of the EEC made a conscious choice in 1956–57 (collectively if
not individually) not to include democracy or respect for human rights as criteria
for membership in their new community. Instead, motivated both by the liberal
economic thinking that underlay planning for the common market and by the
heightened geo-political pressures of the Cold War, the EEC would be open to all
non-Communist states in Europe.
Within months of the Treaty of Rome’s entry into force on 1 January 1958, the
EEC’s six governments and the European Commission had confirmed their
commitment to the openness of the new community with the exception of
Communist states. This first became apparent in deliberations on the question of
relations with the eleven non-EEC members of the Organisation for European
Economic Cooperation (OEEC)—Austria, Denmark, Greece, Iceland, Ireland,
Norway, Portugal, Sweden, Switzerland, Turkey, and the UK. In March 1958, soon
after taking office as first president of the European Commission, Walter Hallstein
insisted on the EEC’s willingness to consider membership for the OEEC Eleven:
‘There can only be discrimination, in other words an unwarranted differentiation
in the treatment of other European States, if the Six deny to other European States
the treatment which they accord one another, that is, if they refuse admission to a
State which is willing to pay the same price as the Six for the advantages of
membership of the custom union. Obviously that has not happened. Quite the
opposite: the Treaty embodies the principle of the open door’ (Stirk and
Weigall 1999: 149).
This openness was reconfirmed in an official memorandum submitted by the
Six to the OEEC’s Maudling Committee on 17 October 1958, in which the EEC
‘solemnly reaffirms that it is determined, both for economic and political reasons,
to arrive at an agreement which will make it possible to associate with the
Community on a multilateral basis the other Member States of the O.E.E.C.’11 For
11 CVCE: Memorandum from the European Economic Community (17 October 1958).
example, the Commission’s ‘First Memorandum’, drafted in close consultation

with the governments of the Six in early 1959, recommended that any of the
Eleven willing to pursue deeper integration should be offered association or full
membership in the EEC in accordance with Articles 238 and 237 of the Treaty of
Rome (Commission 1959).12 Spain was not mentioned in any of these documents
because it was not yet a member of the OEEC, but the casual inclusion of OEEC
member Portugal, a highly repressive dictatorship formerly allied with Nazi
Germany, confirms that the presence or absence of democracy was not an
important consideration.
The only formal condition to be imposed on OEEC states seeking membership
or association was that they accept the purposes and modalities of the Treaty of
Rome. In this spirit, the EEC’s October 1958 memorandum to the Maudling
Committee specified that any association agreement to be adopted under Article
238 ‘must not in any way prejudice either the content or the implementation of
the Treaty of Rome.’ To this end, such agreements must create more than a free
trade area: they must provide for a broad convergence of social and economic
policies and legislation akin to that applying within the EEC. The memorandum
thus reconceived association as a stepping-stone to full membership. But apart
from an applicant’s willingness to accept these administrative and legislative
requirements, it did not mention any additional constitutional or political criteria
for association.13
Looking beyond the Eleven, though, EEC officials made no secret of their
understanding that the new community would be closed to Communist states.
Robert Marjolin, Vice President of the European Commission and former head
of the French delegation in negotiations on the Treaty of Rome, made this point
unequivocally in a speech in Washington DC in April 1958. The creation of the
EEC, he said, was ‘taken in a spirit of friendly cooperation with the other
European nations in the hope that the union of the Six may in time become a
broader union of free nations. . . . We are determined to build a United Europe
because we believe in the Free World and that the Free World will be stronger
if we are stronger.’14 Fifteen years after the treaty, Hallstein recalled, ‘the Europe
that was about to unite belonged fully to the non-Communist world’ (Hallstein
1972: 19–20).
These ideas were widely shared by the governments of the Six. In a crucial
meeting of the EEC’s Council of Ministers held in early 1960 to discuss how to
respond to recent requests for association and membership, Belgian foreign
minister Pierre Wigny interpreted the treaty as requiring simply that candidates
12 AEI: First Memorandum from the Commission of the European Economic Community, 26
February 1959.
13 CVCE: Memorandum from the European Economic Community (17 October 1958).
14 AEI: The free world’s at stake in the Common Market, 17 April 1958.
for full membership must be European and must be capable of adjusting to the
common external tariff and the absence of internal tariffs. France’s Maurice
Couve de Murville introduced the possibility of association agreements with a
number of countries including Spain (then still under fascist rule), and spoke of
the ‘great value’ that he attached to ‘the idea that the Common Market must give
the impression of being an open association’. Germany’s Heinrich von Brentano
agreed that the Common Market ‘must remain open’, and warned that a failure to
welcome Greece and Turkey could increase Soviet influence over those countries
(Ministère des Affaires Étrangères 1995, document 35). Similarly, a lengthy study
of the implications of the Treaty of Rome for EEC association and membership
prepared by the Belgian Foreign Ministry in the winter of 1962 also makes no
mention of democracy or human rights as possible criteria.15
With the second Berlin crisis raging in the autumn 1961, Hallstein went to
great lengths to portray the EEC as a pillar of the ‘free world’, which he defined as
the community of states wherein economic exchange is not subject to central
control, with no mention of democracy or human rights.16 The same point is evi-
dent in the Commission’s aforementioned First Memorandum, according to
which ‘the enterprise of the Six [is] the instrument and expression of a
fundamental political idea’—namely, the ambition to create a dynamic economic
pillar within ‘Western Europe’ that would support the ‘free world’.17 And given
the evident readiness to include Portugal in the community, one cannot attribute
this anti-Communist norm to an underlying but unstated commitment to
democracy.
The openness to Portugal also cannot be explained by a strong distinction
within the EEC regarding the criteria for association and for membership:
building on the October 1958 memorandum, a May 1960 Commission document
states explicitly that association agreements with European states should be
oriented from the outset toward preparing them for full membership in the
EEC. Later that year, German economic minister Ludwig Erhard called for the Six
to accept ‘the principle of a multilateral association with regard to their economic
relations with the other free countries of Europe’, particularly in light of the
confrontation with Soviet communism. As he saw it, limiting the common market
to the six original members was ‘not sufficient to solve either the European
problem as a whole or that of the Atlantic community [and] no such limitations
were envisaged when the Rome Treaties were born’ (Erhard 1963: 317–20).
These ideas were widely shared by the governments of the Six. In a crucial
meeting of the EEC’s Council of Ministers held in early 1960 to discuss how to
15 MAEB: file 6641/1-s: Note sur les régimes d’adhésion et d’association prévus par le Traité institu-
ant la Communauté Économique Européenne, 7 février 1962.
16 AEI: The EEC and the Community of the Free World, 24 November 1961.
17 AEI: First Memorandum from the Commission of the European Economic Community, 26
February 1959.
respond to recent requests for association and membership, Belgian foreign

minister Pierre Wigny interpreted the treaty as requiring simply that candidates
for full membership must be European and must be capable of adjusting to the
common external tariff and the absence of internal tariffs. France’s Maurice
Couve de Murville introduced the possibility of association agreements with a
number of countries including Spain, and spoke of the ‘great value’ that he
attached to ‘the idea that the Common Market must give the impression of being
an open association’. Germany’s Heinrich von Brentano agreed that the Common
Market ‘must remain open’, and warned that a failure to welcome Greece and
Turkey could increase Soviet influence over those countries (Ministère des
Affaires Étrangères 1995, document 35). Similarly, a lengthy study of the implica-
tions of the Treaty of Rome for EEC association and membership prepared by the
Belgian Foreign Ministry in the winter of 1962 also makes no mention of democ-
racy or human rights as possible criteria.18
2. A community of parliamentary democracies, 1962–1969
The Treaty of Rome was not amended in the 1960s, nor did the member states
collectively issue any new authoritative criteria for EEC membership during these
years. Yet despite this continuity in formal rules, prevailing norms regarding the
political requirements for membership changed significantly early in the decade.
This normative change resulted in large part from political entrepreneurship by
members of the European Parliamentary Assembly frustrated by the absence
of democratic principles in the Treaty of Rome and disturbed by the new com-
munity’s openness to authoritarian governments in Ankara and Madrid (see
Chapter 5). Many of them had earlier belonged to the parliamentary assembly of
the Council of Europe, which was only open to democracies, or the ECSC
Common Assembly, which had drafted the statute of the European Political
Community.
Despite the failure of plans for EPC, these parliamentarians were strongly
committed to the federalist vision of strong supranational institutions requiring
and guaranteeing democracy and human rights at the national level and
repeatedly insisted on the ‘democratic character’ of the community of the Six,
which they believed was reflected most clearly in their own supranational assem-
bly of freely elected parliamentarians.19 In this sense, the struggle over member-
ship norms that re-emerged in the early 1960s was the latest phase in a larger
18 MAEB: file B-3: Note sur les régimes d’adhésion et d’association prévus par le Traité instituant la
Communauté Économique Européenne, 7 février 1962.
19 AEI: Rapport fait au nom de la Commission des Affaires Politiques et des Relations exterieures
de la Communaute—L’Assemblee Parlementaire dans l’Europe des Six. Exercice 1957–1958 Deuxieme
session extraordinaire, Document 14, fevrier 1958.
struggle over the purpose and form of regional integration that had persisted since
the early 1950s. Nonetheless, their success in changing EEC membership norms
in the early 1960s depended upon two factors: a political opening created by the
member states’ own disagreement on the terms of political union and the parlia-
mentarians’ ability to frame their demands with reference to national constitu-
tional and Council of Europe norms.
A French government initiative was questioning the Treaty of Rome from an
entirely different perspective. Unhappy with the EEC’s quasi- supranational
method of decision-making and concerned that eventual British entry would
further diminish French influence, Charles de Gaulle had proposed in 1960 that
the Six establish an inter-governmental body to coordinate political, economic,
defence, and cultural policy. Although the other five member states were sceptical
of de Gaulle’s idea, which they feared would subordinate the EEC’s new
institutions to inter-governmental control and weaken the trans-Atlantic alliance
by creating a rival to NATO, they agreed in February 1961 to explore the
possibility of a political union amongst the members of the EEC and open to
subsequent enlargement. An inter-governmental committee chaired by French
diplomat Christian Fouchet was charged with negotiating the details of such a
union, but a lack of consensus among the Six was evident as soon as the committee
began its work. The Dutch were particularly outspoken in support of the existing
EEC institutions, the priority of NATO on defence cooperation, and the
importance of remaining open to British membership (Camps 1964, Vanke 2001).
The institutional uncertainty created by the French initiative facilitated a
counter- initiative by parliamentarians concerned by de Gaulle’s well- known
aversion to supranational institutions and his government’s open support for
involving Spain in European integration. For these parliamentarians, the question
of eligibility for EEC membership was tied inextricably to their view of the EEC
as a stepping-stone to deeper political integration. French and German support
for the European ambitions of Spain since the late 1950s, as well as the
accommodative attitude of the Six toward the Turkish military coup in 1960 (see
Chapter 5), had made it clear that the Six were willing to consider any non-
Communist European state for membership in the EEC. This openness was
clearly incompatible with the federalist vision of Europe.
In response, EPA members and their colleagues in the Council of Europe’s
consultative assembly (all drawn from the same national parliaments) began to
press their own, alternative vision of European integration. Various EPA reports
and resolutions in late 1960 and early 1961 demanded that political union must
not compromise the supranational institutions of the existing EEC. But Europe’s
parliamentarians also made clear their views on which states should be associated
with the process of European integration. In January 1961, German social
democrat Annemarie Renger, a member of the Council of Europe’s assembly,
released a draft report highlighting the total absence of democracy and the
consistent violation of human rights in Franco’s Spain and warning that official
ties with NATO and the OEEC, as well as political support from the French and
German governments, were consolidating the Franco regime’s hold on power and
discouraging the opposition.20 Another early indication of their thinking emerged
around a joint conference of the European Parliamentary Assembly and the par-
liaments of various newly independent African states, held in late June 1961.21
In a document prepared before the conference, the EPA’s political commission
recommended that the forthcoming meeting declare that ‘respect for human rights
and fundamental liberties’ was a fundamental principle of any association
agreement between the EEC and its African partners.22
The subsequent evolution of proposals for political union suggests that this
pressure from the EPA had some effect. A communiqué on political union issued
by the leaders of the Six at the Bonn summit in July 1961 noted the EPA’s position,
affirmed the ‘political traditions which form their common heritage’, and
identified European integration closely with the fate of ‘free peoples’. Such
language was consistent with pre-existing EEC membership norms, as discussed
in Chapter 2, but the EPA wanted a stronger commitment. When Prime Minister
Harold Macmillan announced in late July that the UK would soon apply for EEC
membership, the newly elected president of the EPA’s socialist group, German
social democrat Willi Birkelbach, declared that British accession would ‘anchor
parliamentary and democratic principles even more firmly at the heart of
the EEC’.23
In the early autumn of 1961, the European Parliamentary Assembly’s political
commission appointed two working groups to consider how the community
should respond to the growing number of states applying for association or
membership. The first, chaired by Dutch parliamentarian Marinus Van Der Goes
Van Naters, focused on the various responsibilities of the Council, Commission,
and Parliament in decisions on membership, with an evident preference for
expanding the role of the latter. But Van Der Goes Van Naters’ report did not dis-
cuss membership criteria.24 That task was left to Birkelbach, whom the EPA
appointed as rapporteur of a second working group on political and institutional
aspects of the association and adhesion process.
20 ACE: PACECOMO19282 La situation politique in Spain (‘Rapport Renger’), Commission des
Nations Non-Représentées, Assemblée Consultative, Conseil de l’Europe, 30 novembre 1961.
21 These talks were a first step toward an association agreement signed in Yaoundé on 20 July 1963
between the EEC and a number of African states plus Madagascar.
22 AEI: Rapport fait au nom de le Commission politique et institutionnelle sur les questions poli-
tiques et institutionnelles examinées par la Conférence de l’Assemblée Parlementaire Européenne avec
les parlements d’États africains et de Madagascar. Documents de séance No.1, 23 juin 1961.
23 WB: Jean- Marie Kraft, Le Groupe Socialiste au Parlement Européen, thesis, Université de
Strasbourg, 1962, p. 149.
24 AEI: Rapport fait au nom de la commission politique sur la procédure à suivre pour la conclu-
sion des accords d’adhésion, M. van der Goes van Naters, Rapporteur. Assemblée Parlementaire
Européenne, Documents de Séance, 1961–1962, Document 75, 15 Janvier 1962.
A former political prisoner of the Nazi regime, Birkelbach believed that the
purpose of European integration was to advance democratic values and human
rights: ‘We had seen how fragile democracy really was. We had seen it destroyed
in our own country.’25 For this reason, he concluded, ‘In Europe, parliamentary
democracy, the liberties and the rights inherent in democracy must have not only
a national basis, but must also be so firmly established at the international level
that they can no longer be questioned.’26 As he later explained, ‘The parliament’s
purpose was to ensure that the political, democratic nature of the Community
was secure. It was important to show the Commission and the ministers that they
had to give priority to democratic principles and institutions.’27
To succeed, however, Birkelbach and his committee colleagues would have to
craft a definition of the community’s political identity that member states’
governments could not ignore, regardless of their particular preferences. Such a
definition would have to be radical in its conception of membership eligibility
and yet still plausibly grounded in the Treaty of Rome, whose Article 237
indicated that ‘Any European state’ could apply for membership. To be approved
by the EPA, it also had to be acceptable to the liberal, Christian democrat, and
socialist groups in the assembly. And to have any effect on the Six, it would have
to be rooted in positions they had already accepted, at least formally or
rhetorically. Fortunately for Birkelbach, De Gaulle’s proposal for political union
had created an opening for new community discourse on membership eligibility.
On 2 November 1961, after months of committee deliberation and mutual
concessions, Christian Fouchet distributed to his colleagues the first complete
draft of a treaty on political union. The draft immediately encountered stiff
resistance because of its inclusion of defence cooperation and the lack of clear
openness to British participation, among other points. However, a significant
element of the draft that does not appear to have sparked debate was its proposal
that the union would be committed, inter alia, to ‘the defense of human rights,
fundamental liberties and democracy and to justice in every sphere of social life’
and therefore that membership would be limited to states that already belonged
to the EEC and the Council of Europe, the latter of which was only open to rights-
protective democracies.
These provisions directly echoed the 1953 draft statute for a European political
community, which had been abandoned by the Six but retained as a goal by
federalist parliamentarians. Their inclusion in the draft treaty indicates that at
least by some of the member states’ governments shared the EPA’s agenda on
membership eligibility. If formally adopted in a treaty on political union, these
25 Author’s interview with Willi Birkelbach, Frankfurt, 15 July 2001. See also Birkelbach 2000.
26 WB: Jean- Marie Kraft, Le Groupe Socialiste au Parlement Européen, thesis, Université de
Strasbourg, 1962, un-numbered foreword.
27 Author’s interview with Willi Birkelbach, Frankfurt, 15 July 2001.
provisions would have constituted a major step beyond the Treaty of Rome’s
ambiguous commitment to ‘liberty’, its provision that ‘any European state may
apply to join’, and the non-Communist membership norm that prevailed in the
EEC’s early years. The problem was that such a membership norm would
contradict the demonstrated policy preferences of at least two of the member
states, France and Germany.
At the same time that the Six were debating the terms of political union,
Birkelbach and his colleagues on the EPA’s political committee were drafting
their own report. Given the high profile of the Fouchet committee and its regu-
lar contact with the EPA’s political committee, Birkelbach and colleagues would
have been fully aware that the Six were now willing to consider the possibility
that European integration should not be open to ‘all European states’ or even to
any non-Communist European state. Such ideas were not new—they had fea-
tured in the 1944 anti-fascist Ventotene Manifesto, the 1948 Statute of the
Council of Europe, and the unsuccessful 1953 draft Statute of the EPC—but
they had been rejected by the founders of the ECSC/EEC, much to the dismay
of committed Euro-federalists like Birkelbach. Now some of the Six had clearly
proposed that membership in their community should only be open to
European parliamentary democracies—a radical departure from the existing
treaty and recently prevailing norms.
The question was whether a proposal made in the midst of inconclusive inter-
governmental negotiations could be converted into a norm powerful enough to
affect collective decision-making. While recognizing that Article 237’s openness
to ‘[a]ny European state’ referred first of all to geography, Birkelbach and his EPA
colleagues believed that the creation of a common market and the treaty’s
aspiration to ‘ever closer union’ implied other economic and political characteristics
as well. Moving beyond the simple reference to ‘liberty’ in the treaty’s preamble,
the first draft of Birkelbach’s report identified a ‘democratic form of state’ as a
requirement for membership. During a meeting of the EPA political committee
on 10 November, Commission president Walter Hallstein expressed his general
approval of the draft, but a committee member voiced doubt about the definition
of democracy.28
On 19 December 1961, the political committee of the EPA unanimously
adopted Birkelbach’s report on association and adhesion, which asserted, despite
the debate swirling around the Fouchet committee, that the political character of
the Community is ‘no longer contested’. The opinion that the Community is
simply an ‘enlarged international economic treaty’ is almost never advanced, it
says, except occasionally ‘in those countries that begin to study more closely the
bases of the three treaties in order to prepare their membership or association
28 AHPE: Procès-Verbal de la réunion de la Commission Politique, Assemblée Parlementaire

Européenne, 10 novembre 1961, APE/I/PV/61–11.
with the Community’. But rather than tabling a federalist manifesto on moving
beyond the Treaty of Rome, which was unlikely to gain broad reception,
Birkelbach made sure to link the novel aspects of his text to political positions
that were already accepted by political parties and governments of both the
centre-right and centre-left.29
Building upon the pre-existing norm that Communist states were not eligible
for membership, his report asserted ‘the political structure of [an applicant state]
must not make it a foreign body in the Community’. Instead, the report proposed
that any state applying for full membership must recognize and support the
Community’s ultimate goal of political union, and must have domestic structures
and foreign policies consistent with that goal. To start, it must support the general
foreign policy commitments of the existing member states: that is, it must be
willing and able to cooperate with the western or Atlantic alliance. This principle
would clearly exclude members of the Warsaw Pact, but the report did not take a
position on whether it would also exclude neutral states such as Austria, Ireland,
Sweden, and Switzerland.
In addition, following the earlier query about defining democracy, the report
offered a simple definition of which states were eligible for membership: ‘The
guaranteed existence of a democratic form of state, in the sense of a free political
order, is a condition for membership. States whose governments do not have
democratic legitimacy and whose peoples do not participate in the decisions of
the government, neither directly nor indirectly by freely-elected representatives,
cannot expect to be admitted in the circle of peoples who form the European
Communities.’ On the other hand, the report was noticeably tentative on the
question of whether respect for human rights should also be a formal requirement
for community membership: ‘One could . . . suggest requiring of States that wish
to join the Community that they recognize the principles that the Council of
Europe has posed as a condition for those who want to be members of it.’
Explanation of this clause is even relegated to a rare footnote: ‘This involves above
all recognition of the principles of the rule of law, human rights and fundamental
freedoms (cf. article 3 of the Statute of the Council of Europe).’ Birkelbach went
further in his subsequent presentation of the report to the full EPA, expressing his
personal view that respect for human rights was integral to the political identity
of the community.30
The Birkelbach Report also addressed the question of association with the
EEC, as provided by article 238 of the Treaty of Rome. Some members of the
political committee preferred to apply the same requirements to association and
29 AHPE : Rapport fait au nom de la commission politique sur les aspects politiques et institution-
nels de l’adhésion ou de l’association à la Communauté. M. Willi Birkelbach, Rapporteur, décembre
1961, CARDOC PE0 AP RP/POLI.1961 A0-0122/61.
30 AHPE: Assemblée Parlementaire Européenne, Débats, Séance du mardi 23 janvier 1962, p. 55.
adhesion in order not to weaken the latter, but other members preferred a flexible
stance on association. The final report thus recognized that association could take
various forms, some more similar to membership and others more distant, and
insists simply that any association must be consistent with the ‘fundamental
political character of the Community’. Although it asserted that adhesion should
be the general rule or expectation, and that association should be reserved
for exceptional situations, it stopped short of establishing political criteria.
Finally, according to the report’s closing paragraph, ‘it is necessary to foresee for
each association contacts at the parliamentary level between the European
Parliamentary Assembly and the freely-elected representatives of the people in
the country in question’. The significance of this assertion would not become
apparent until the Greek crisis in 1967 (see Chapter 6).The Birkelbach Report
did not impose new legal obligations on the community’s members, nor did it
reflect shared beliefs or policy preferences regarding enlargement among all the
community’s member states and supranational actors. But its politically balanced
provisions corresponded closely to the national constitutions of the Six, to
proposals they had accepted within the Fouchet committee, and to their formal
commitments within the Council of Europe, all of which made the report harder
to refute or ignore than most emanations from the European Parliamentary
Assembly, the weakest of the EEC’s institutions. The release of the report in
December 1961 and the reception it received in early 1962 thus contributed to the
emerging norm that the EEC was a community of parliamentary democracies
within the Western alliance—a significant departure from the sparse content of
the Treaty of Rome and from the membership norm that prevailed in the first
half-decade after the treaty.
The fact that the report’s assertions were novel for the EEC and contrary to the
policy preferences of some of the member states’ governments did not stop it
from being recognized as an unassailable expression of the EEC’s membership
norm—variously praised and tolerated but never openly criticized at the
European level. When Birkelbach presented his report to the full EPA on 23
January 1962, it was met with universal acclaim from members of various political
groups within the assembly as well as representatives of the European
Commission. Socialist member Fernand Dehousse applauded the report’s
insistence that community membership was only open to states with similar
economic and political structures, defining the latter in terms of parliamentary
democracy and respect for human rights and fundamental freedoms for all.
Christian Democratic member Jean Duvieusart praised Birkelbach for looking
beyond the geographic requirement for membership and defining economic and
political criteria not articulated in the treaty: ‘This document is more than a
report . . . it will constitute a foundational text, almost a law.’ Duvieusart even
predicted that the report would have an impact on European integration
comparable to the 1955 Messina conference that led to the Treaty of Rome.
Speaking for the European Commission, both Sicco Mansholt and Jean Rey wel-
comed the report as a valuable contribution to the development of a community
‘doctrine’ on association and adhesion.31 Such was the response to the report that
Strasbourg’s leading newspaper declared the following day that it could be con-
sidered ‘the “charter” for membership and association with the European
Economic Community’.32
The strength of the emerging norm was also apparent when deliberations over
political union resumed in early 1962 following a new proposal from the French
government, reportedly drafted with personal input from Charles de Gaulle
(Camps 1964). It dropped the requirement that all members must also belong to
the Council of Europe but retained the preambular affirmation of their collective
‘attachment to the principles of democracy, of human rights, and of social justice’.
Whether or not de Gaulle was informed about the Birkelbach report, his aides
surely were, and this revised draft treaty shows the French government’s desire to
fit their proposal to the emerging membership norm. France’s five EEC partners
immediately rejected the January draft, but their counter-proposal maintained
the same preambular affirmation of democratic principles, as did a final draft
treaty adopted by the Fouchet committee on 15 March (European Parliament
1964). These negotiations on political union ultimately proved unsuccessful and
the various draft treaties were abandoned, but the fact that all had included
parliamentary democracy as a requirement for membership could only have
reinforced the new norm.
The Council secretariat distributed the Birkelbach report to the governments
of the Six on 5 February 1962, even though they would have already been well
acquainted with its contents.33 At about the same time, the Six asked the
Commission to draft a comprehensive doctrine on association for their discus-
sion.34 In an April 1962 speech, Commission vice-president for external relations
Jean Rey lamented that the member states ‘had not had the time . . . to elaborate a
doctrine regarding all these associations that rain down everywhere, to ask them-
selves what is to be done, which will be accepted, which will not be accepted, on
the basis of which criteria . . .’ Rather than focus on each application separately,
Rey said, the community should address the general question, ‘how does the
Common Market conceive of its enlargement?’35
The following month, Walter Hallstein informed the Council of Ministers
of the Commission’s full support for the Birkelbach report: ‘The Commission
is of the opinion that if it had elaborated its own document on the subject
[of association and adhesion] it would not have arrived at more valuable

31 AHPE: Assemblée Parlementaire Européenne, Débats, Séance du mardi 23 janvier 1962, p. 55.
32 Strasbourg: Passionnant Débat Politique au Parlement Européen. Les Dernières Nouvelles
d’Alsace (Strasbourg), 24 January 1962.
33 ACCUE: CM2.1.1962. 34 ACCUE: CM2.848.1965.
35 AEI: Conférence du Ministre Jean Rey, 9 avril 1962.
conclusions than those contained in the Birkelbach report’ and recommended

that the Council take it as the basis for its own discussions. Hallstein then
explained that the Commission did not believe that there existed, in the
Birkelbach Report or elsewhere, any indisputable criteria that would automatic
ally qualify an applicant state for association or adhesion; this required that the
applicant state justify its demands to the community. In short, the membership
norms proposed by Birkelbach could disqualify an applicant state, but not ensure
its acceptance.36 The following month, Hallstein gave an unreserved public wel-
come to the Birkelbach Report, declaring that its publication had ‘enhanced the
authority’ of the EPA.37
The report’s close parallel to the national constitutions and Council of Europe
commitments of the Six made it particularly difficult to rebut, even for member
states troubled by its implications for their views on community enlargement.
Neither the French nor the German government—both supportive of bringing
Spain into the EEC and thus most likely to resist the report’s recommendations—
criticized the report publicly or even behind closed doors in EEC deliberations.
Given that neither government had hesitated to criticize earlier assertions of
authority by the EPA, this silence suggests that by early 1962 they accepted the
Birkelbach Report as a legitimate (however inconvenient) expression of the
community’s new membership norm.
The persistence of the new norm alongside divergent member state preferences
was evident in 1964–65 during discussions among the Six of an Italian foreign
ministry memorandum on the future of the Community and its relations with
third countries. The memorandum itself, distributed to the Council in May 1964,
borrows the Birkelbach report’s insistence that candidates for adhesion must fulfil
geographic, economic, and political criteria, but is more demanding on the last
dimension: ‘The European construction presupposes a substantial similarity of
political systems within the member states. Membership is therefore excluded
both for countries whose international status prevents them from assuming the
obligations deriving from a future implementation of political unity and for
countries whose domestic political regime is founded on significantly different
criteria than those that inspire the six member states.’38
The memorandum was also very strict with regard to European states applying
for association: ‘For these countries, the formula of association should only be
granted on a temporary basis and as an intermediary step to arrive at the original
goal of full integration. . . . Because this involves an association oriented to
membership, it is important that such a state fulfills the political conditions
required for membership both on the domestic and international level. Otherwise,
36 ACCUE: CM2.1962.24.

37 AEI: The European Economic Community as an element in a new world order, 22 June 1962.
38 ACCUE: S/323/1964/REV1 & CM2.1965:848.
the only alternative is a commercial accord.’ Given Italy’s sense that it had paid the
highest material price for the Greek and Turkish associations, which Foreign
Minister Giuseppe Saragat had communicated to his counterparts three months
earlier, and its concern about a possible deal with Spain, it is impossible here to
disaggregate the Italian government’s economic interests from its normative
concerns.39
It is nonetheless noteworthy that in two subsequent Council discussions of the
Italian memorandum (1–3 June 1964 and 8 April 1965), Belgium, Luxembourg,
and the Netherlands endorsed the Italian position, while France and Germany
merely expressed scepticism about the practical implications of abstract yet strict
guidelines on such a political matter; France also sought to discredit the
memorandum by pointing out that it had been distributed in the midst of discus-
sions about Spain.40 Despite the clear and controversial implications of the Italian
proposals, none of the member states questioned the principle that only parlia-
mentary democracies were eligible for membership—precisely as one would
expect in the presence of an idea that had acquired normative status.
So while the EEC’s formal membership rule (‘Any European state may apply’)
remained intact throughout the first decade following the Treaty of Rome,
European parliamentarians succeeded in early 1962 in exploiting disagreements
among the member states to redefine the limits of Europe as a political commu-
nity. In the process, they replaced a membership norm that welcomed non-
Communist states with one that welcomed only Western-oriented parliamentary
democracies that accept the goal of political union.
3. A community of liberal democracies, 1970–2005
The next significant change in the EEC’s membership norms—the inclusion of

respect for human rights as a necessity for all member states—emerged out of the
community’s 1967 decision to freeze Greece’s existing association agreement
pending a reversal of the military coup and return to constitutional rule. Although
the utility of the freeze was highly contested in internal community deliberations,
its public justification by some community actors between 1967 and 1974
suggested that the military regime’s lack of respect for human rights and the rule
of law were just as important as its suspension of parliamentary democracy. This
discourse went well beyond the Birkelbach Report’s hesitant mention of human
rights and its reference to political criteria for association as ‘a very delicate
question to which one cannot respond in a summary fashion’. As a result, it took
several years before it was translated into a general membership norm.
39 ACCUE: CM2.1964:186, Saragat’s declaration to the Council, 24–25 February 1964.
40 ACCUE: CM2.1965:848.
This evolution in thinking about the European community is evident in a book

published in 1969 (and three years later in English translation) by Walter
Hallstein, former president of the European Commission. While Hallstein lists
‘freedom’ as one of the ‘basic values’ of the Community, he defines it simply
as freedom of movement of workers, capital, and services, the freedom of entre-
preneurs to make their own decisions, and the freedom of consumers to make
their own choices. On the other hand, Hallstein argued that both the member
states and the Community were bound by Article 3 of the Statute of the Council
of Europe, which declared: ‘Every member-state of the Council of Europe recog-
nizes and accepts the principle of the rule of law and the principle that everyone
who comes under its legal and sovereign authority is entitled to the basic free-
doms, and to human rights . . .’ (Hallstein 1972: 42–8). This assertion that the
member states’ individual duties under the Council of Europe extended to the
legally-separate European Community as a whole was arguable, as Hallstein
surely recognized, and he did not conclude that the Community should not admit
a state that did not respect human rights.
The first official expression of the full normative shift came in the October
1970 ‘Davignon Report’ on political cooperation by the foreign ministers of the
Six, which redefined the community’s ‘fundamental aim’: ‘A united Europe should
be based on a common heritage of respect for the liberty and rights of man and
bring together democratic States with freely elected parliaments.’41 Then, in
December 1973, following the accession of Denmark, Ireland, and the United
Kingdom, the foreign ministers of the Nine issued a joint ‘Declaration on the
European Identity’ expressing their determination ‘to defend the principles of
representative democracy, the rule of law, social justice—which is the ultimate
goal of economic progress—and of respect for human rights’. They also spelled
out the implications of this commitment for the community’s enlargement: ‘The
construction of a United Europe . . . is open to other European nations who share
the same ideals and objectives’.42 With these two declarations, the member states
identified themselves collectively for the first time as a community of liberal
democracies founded on respect for human rights.
It is important to note here that the language of human rights was not new to
Europe around 1970. However obvious, this observation relates directly to the
question of whether the 1970 and 1973 documents reflect a true normative shift
for the EEC or simply a tactical decision to formalize what had long been
understood informally. The fact that human rights language had been prominent
in Europe for at least two decades (since the signing of the European Convention
41 AEI: Report by the foreign ministers of the member states on the problems of political unifica-
tion, 27 October 1970.
42 AEI: Document on the European identity published by the nine Foreign Ministers, Copenhagen,
14 December 1973.
on Human Rights and Fundamental Freedoms in 1950 and the failed proposal for
a European Political Community in 1952) and the governments of the Six would
not have paid a domestic political price for framing the European integration
process explicitly in terms of human rights, one can only conclude that the Six did
not conceive of the community in those terms during the 1950s and 1960s. Hence
the conclusion that the 1970 and 1973 declarations heralded a real change in
membership norms.
Step by step over the next two decades, EU member states and institutions
reaffirmed and elaborated the new membership norm. In April 1977, the
European Parliament, the Council and the Commission issued an unusual joint
declaration affirming ‘the prime importance they attach to the protection of
fundamental rights, as derived in particular from the constitutions of the Member
States and the European Convention for the Protection of Human Rights and
Fundamental Freedoms’ and committing to respect these rights in ‘the exercise of
their powers and in pursuance of the aims of the European Communities’.43 In a
September 1977 conclave, members of the European Commission reaffirmed
the ‘fundamental connection’ between ‘pluralist democracy’ and Community
membership and recommended that the Council issue a ‘solemn declaration’ on
this point ‘as quickly as possible’, while negotiations with Greece, Spain, and
Portugal were still ongoing.44 At their Copenhagen summit in April 1978, the
EC’s heads of state and government adopted a ‘Declaration on Democracy’ reaf-
firming that ‘respect for and maintenance of representative democracy and
human rights in each Member State are essential elements of membership of the
European Communities’.45
In this new normative environment, it didn’t seem to matter that the EEC
treaty’s text did not mention democracy or human rights: the member states were
now retrospectively reading into the treaty certain principles on membership
eligibility that they considered binding even though they were not explicitly
contained in the text. In its 1978 submission to the European Court of Justice’s
Matheus v Doego case, the Commission expressed its new interpretation of the
EEC Treaty’s Article 237: ‘It permits the accession of the state only if that state is a
European State and its constitution guarantees, on the one hand, the existence
and continuance of a pluralistic democracy and, on the other hand, effective
protection of human rights’ (Kochenov 2004: 4). The member states made this
explicit in their December 1979 answer to a written question from the European
Parliament: ‘The Council is convinced that the States applying for accession to the
European Communities are aware of the principles underlying the Treaties and
43 CVCE: Joint Declaration by the European Parliament, Council and the Commission (5 April 1977).
44 EUDC: 441.213 ER Enlarg negoc pt.1: Commission Conclave, Europe no.2290, 19/20
September 1977.
45 CVCE: Declaration on democracy at the Copenhagen European Council (7 and 8 April 1978).
the obligations deriving therefrom, and of the content of the Joint Statement by
the European Parliament, the Council and the Commission, of 5 April 1977, on
Fundamental Rights and the Declaration on Democracy adopted by the European
Council on 7 April 1978’.46 Five years later, meeting in Stuttgart to consider the
community’s future, the Nine issued a ‘Solemn Declaration on European Union’
that explicitly reaffirmed their 1978 Copenhagen declaration on democracy.47
The community’s supranational institutions soon echoed this expectation that
only liberal democracies were eligible for membership. In 1979, the Commission’s
opinion on Greek accession declared, ‘the principles of pluralist democracy and
respect for human rights form part of the common heritage of the peoples of the
States brought together in the European Communities and constitute therefore
essential elements of membership of the said Communities’.48 In 1984, the
European Parliament adopted MEP Altiero Spinelli’s draft treaty on European
Union, which proposed that ‘Any democratic European state may apply to
become a Member of the Union’ and empowered the Union to impose penalties
‘[i]n the event of serious and persistent violation of democratic principles or fun-
damental rights by a Member State’.49 (This theme was not new for Spinelli, a co-
author of the 1941 Ventotene Manifesto calling for a European federation of
democratic states and a major influence on the 1953 draft statute for a European
Political Community.) In 1985, the Commission’s opinions on Spanish and
Portuguese applications for accession repeated the same language used six years
earlier for Greece.50
The EU then gave legal status to its new membership norm, first in the 1986
Single European Act (SEA) and then in the 1992 (Maastricht) Treaty on European
Union. Although the SEA did not modify the Treaty of Rome’s openness to ‘Any
European state’, as Spinelli had proposed, it broke new ground legally with its
expression of the signatories’ determination ‘to work together to promote
democracy on the basis of the fundamental rights recognized in the constitutions
and laws of the Member States, in the Convention for the Protection of Human
Rights and Fundamental Freedoms and the European Social Charter’. The
Maastricht Treaty also retained the ‘Any European state may apply’ formula but
its preamble reaffirmed the signatories’ ‘attachment to the principles of liberty,
democracy and respect for human rights and fundamental freedoms and of the
rule of law’ while the main text stipulated that the Member States’ ‘systems of
government are founded on the principles of democracy’.
46 Official Journal of the European Communities, C 27, Vol.23, 4 February 1980, pp. 5–6.
47 Bulletin of the European Communities 16(6), June 1983, pp. 24–9.
48 CVCE: Commission Opinion on the application for accession to the European Communities by
the Hellenic Republic (23 May 1979).
49 Official Journal of the European Communities C 77, Vol. 27, 19 March 1984.
50 Official Journal of the European Communities L 302, Vol. 28, 15 November 1985.
The Twelve elaborated the implications of their treaty commitments at the

Copenhagen summit in June 1993, declaring: ‘Membership requires that the
candidate country has achieved stability of institutions guaranteeing democracy,
the rule of law, human rights and respect for and protection of minorities, the
existence of a functioning market economy as well as the capacity to cope with
competitive pressure and market forces within the Union.’51 Although entirely
consistent with the post-1970 priority on human rights, this specific reference to
protection of minorities was connected to recognition that many of the countries
desiring accession in this period faced significant challenges in this area. At the
same time, the Copenhagen summit conclusions pointed out that enlargement
is a reciprocal deal: candidates must be able to ‘take on the obligations of
membership’ and the Union must be able to ‘absorb new members while main-
taining the momentum of European integration’. These latter clauses became
important post-2005, but they received little attention in 1993.
As such, the 1993 ‘Copenhagen criteria’—‘Membership requires that the
candidate country has achieved stability of institutions guaranteeing democracy,
the rule of law, human rights and respect for and protection of minorities, the
existence of a functioning market economy as well as the capacity to cope with
competitive pressure and market forces within the Union’—are far less ground-
breaking than is generally assumed. Similarly, the 1997 Amsterdam Treaty’s
unambiguous declaration ‘The Union is founded on the principles of liberty,
democracy, respect for human rights and fundamental freedoms and the rule of
law . . . which are common to the Member States’ is less a case of legal innovation
than a move to bring the treaty into line with a membership norm that had
acquired taken-for-granted status by the late 1970s.
At a 1999 European Council meeting in Cologne, the Fifteen authorized the
creation of a special inter-governmental and inter-parliamentary body to draft a
human rights charter for the Union. At about the same time, the possibility of
suspending the voting rights of a member state committing a serious and
persistent breach of human rights, which had been established by the Amsterdam
Treaty, suddenly became relevant. When parliamentary elections in Austria in
1999 resulted in a coalition government involving the far-right Freedom Party
(FPÖ), the EU’s other 14 governments imposed diplomatic sanctions on Vienna.
The sanctions fell short of suspending Austria’s voting rights, but they were only
lifted after an investigatory panel of ‘wise men’ appointed by the European Court
of Human Rights concluded in June 2000 that the Austrian government was
fulfilling its duties to respect human rights (Duxbury 2000). In December 2000,
51 It is interesting to note that in addition to the 1993 ‘Copenhagen criteria’, the 1973 Declaration
on the European Identity and the 1978 Declaration on Democracy were also agreed in the Danish
capital.
the European Parliament, the Council of Ministers and the European Commission
jointly approved the EU Charter of Fundamental Rights.52
The European Council was therefore restating a well-established norm when it
declared at Laeken in December 2001, ‘The European Union’s one boundary is
democracy and human rights. The Union is open only to countries which uphold
basic values such as free elections, respect for minorities and respect for the rule
of law.’53 The same can be said about the EU’s unsuccessful draft constitutional
treaty adopted in July 2003 after intensive intergovernmental and inter-
parliamentary consultations, which also declared that membership was limited to
European states committed to ‘the values of respect for human dignity, liberty,
democracy, equality, the rule of law and respect for human rights’.54 The fruits of
this consensus included the decision to offer accession to ten former Communist
states of central and eastern Europe plus Cyprus and Malta, and the decision to
grant candidate status to Turkey.
However, the norm that only European states committed to the principles of
liberal democracy were eligible for membership—and that all such states plausibly
fulfilling this standard had a right to membership—did not last much longer.
4. A divided community, 2006–2020
New conceptions of European identity and new attitudes toward EU enlargement

began to emerge in the early years of the new century, leading to a breakdown in
normative consensus at the highest levels of the European Union by late 2005.
One challenge to the pre-existing norms came from those who believed that the
‘big bang’ accession of ten (and soon to be twelve) new member states had
stretched the finances and decision-making capacity of the Union to such an
extent that it might never achieve its internal and external goals. Another
challenge came from the belief that European integration requires an underlying
cultural homogeneity that would be endangered by further enlargement. Some
people and some governments held both beliefs, although the second was rarely
expressed as clearly. On the other hand, there were still member state governments
committed to the vision of the EU as a community of liberal democracies. The
resulting normative dissensus produced a new and unprecedented politics of
membership and enlargement characterized by contention, hard bargaining, and
political deadlock on all but the easiest cases.
In order to explain this normative change, it is necessary first to examine earl
ier trends in public opinion as well as political discourse by non-governing
52 The Charter did not acquire legal status until it was appended to the 2009 Lisbon Treaty.
53 AEI: Laeken European Council, 14–15 December 2001, Presidency conclusions.
54 CVCE: Treaty establishing a Constitution for Europe, 29 October 2004.
political elites. Notwithstanding the membership norms described above, there

have always been dissenting voices including those who associated European
identity with a Christian religious heritage (e.g. Dawson 1952). Although rarely
heard from senior figures in Europe’s mainstream parties as recently as the 1990s,
such attitudes began to acquire real political salience around the start of the new
century. This change can be attributed to three overlapping developments: the
growing visibility of immigrants in towns and cities across Europe; the terrorist
attacks in New York and Washington DC in 2001 and then Madrid in 2004; and
the work of the EU’s constitutional convention between 2001 and 2003.
In the months following the terrorist attacks in New York and Washington DC,
centre-right and far-right politicians across Europe adopted a new discourse that
was unprecedented in recent times in its exclusionary view of community at both
the local and European levels. More and more politicians openly questioned
Islam’s compatibility with European values. In addition, authorities adopted more
stringent administrative practices relating to immigrants and there was a ‘very
significant rise’ in violent assaults, harassment, and vandalism directed against
Muslims, according to the EU’s own European Monitoring Centre on Racism and
Xenophobia (Allen and Nielsen 2002).55
The growing political resonance of anti-immigrant and anti-Muslim sentiment
in Europe was confirmed in April 2002, when far-right politician Jean-Marie le
Pen took second place in the first round of France’s presidential elections. Several
weeks later, rising Dutch politician Pim Fortuyn was assassinated by a (non-
Muslim) Dutch political activist angered by Fortuyn’s critique of Islam as a
‘backward’ religion. In this heated atmosphere, more established politicians saw
opportunities to gain advantage by identifying with the increasingly common
view that Europe did not extend beyond the traditional limits of Christianity. In
May 2002, for example, Bavarian premier and conservative candidate for
chancellor of Germany Edmund Stoiber insisted that the EU needed geographic
limits and these should exclude Turkey and the countries of North Africa.56
Meanwhile, the victory of the moderate Islamist but pro-EU accession Justice
and Development Party (AKP) in Turkey’s November 2002 election raised the
profile of a EU candidate country with a secular state but an overwhelmingly
Muslim population.57 The new AKP government’s repetition of its commitment
to EU accession and insistence that its view of religion’s role in public life was no
different from those of Europe’s Christian Democrat parties did not temper the
growing debate over whether a Christian religious heritage is key to European
identity. Within days of the AKP victory, the president of the EU constitutional
55 Migrants Feel Chill in a Testy Europe, The New York Times, 28 April 2002.
56 Stoiber strongly against Turkey’s EU membership, Hurriyet Daily News, 18 May 2002; EU:
Stoiber’s Remarks On ‘Limits’ To Europe Touch Sensitive Nerve, RFE/RL, 21 May 2002.
57 Religious identity enters Convention debate, EUobserver.com, 8 November 2002.
convention and former president of France, Valéry Giscard d’Estaing, declared

unequivocally ‘Turkey is a country close to Europe . . . but it is not a European
country.’58 Although inconsistent with the official position of the EU (note the
previous month’s summit recommitment that Turkish accession would be judged
‘in accordance with the same principles and criteria as are applied to other
candidate States’), Giscard’s statement reflected and legitimated a growing current
of opinion in Europe, which even its partisans recognized was contrary to norms
then prevailing in official circles (Goulard 2004).59
In early 2003, the Vatican called for an explicit reference to Christianity in the
draft Constitution’s preamble—a position that gained support from about ten of
the participating governments, as well as the European People’s Party, the largest
in the European parliament.60 In response, the new government in Turkey sug-
gested that if there were to be a reference to Christianity in the text, it should also
mention Europe’s Jewish and Muslim heritages.61 This debate within and beyond
the convention found powerful echoes across the continent.62 The Vatican was
less clear, however, on the implications of its position: its ambassador to Ankara
announced in October that the Holy See did not regard the EU as a Christian club
and would not oppose Turkish accession, but the Holy See itself did not speak out
a year later when media portrayed Cardinal Ratzinger’s vocal opposition to
Turkish accession as official Vatican policy on this controversial matter.63 In the
end, due to vigorous opposition from the governments of Belgium and France,
the final text of the constitutional treaty said only that the EU drew its ‘inspiration
from the cultural, religious and humanist inheritance of Europe’.
But rather than dissipate, the debate over Europe’s cultural identity transformed
into a row over public expression and display of religious symbols, from crosses
hanging in public schools to headscarves worn by students and public servants.64
The sense of vulnerability spurred by the ‘September 11’ attacks in the US in 2001
was exacerbated for Europeans by the terror bombings on three commuter trains
in Madrid on 11 March 2004. The cultural debate boiled over with the November
2004 murder in Amsterdam of Dutch filmmaker Theo van Gogh after his release
58 Pour ou contre l’adhésion de la Turquie à l’Union Européenne, Le Monde, 8 novembre 2002.
59 L’Europe peut-elle avoir des frontières? Le Figaro, 28 novembre 2002.
60 Streiten ȕber Gott und Geist, Sȕddeutsche Zeitung, 27 Februar 2003; Guerre de Religion à la
Convention, Le Soir, 27 fevrier 2003; Constitution—Religion in, enlightenment out, EUobserver.com,
11 June 2003; IGC to carry on Christianity debate, EUobserver.com, 13 June 2003; Seven countries
back Christianity in the Constitution, EUobserver.com, 24 October 2003.
61 Turkey calls for Islam and Judaism in Constitution, EUobserver.com. 1 December 2003.
62 Continent Wrings its Hands over Proclaiming its Faith, The New York Times, 12 November 2003;
John Paul II: Europe must recognize Christian roots, EUobserver.com, 23 December 2003.
63 Turkey gets Vatican’s support for its EU bid, EUobserver.com, 28 October 2003; US cables expose
clumsy Vatican diplomacy, EUobserver.com, 13 December 2010.
64 On Display in Italy: Classroom Crosses, and a Raw Nerve, The New York Times, 30 October
2003; High Court Rules Headscarves Okay for Teachers, Deutsche Welle, 24 September 2003; A
Francfort, continuer à vivre dans une société multiculturelle, Le Monde, 14 octobre 2003; Headscarf is
no human right, Court rules, EUobserver.com, 30 June 2004.
of a film saying that Islam discriminates against women, which sparked the
bombing of a Muslim school in Eindhoven and an even more intense clash across
Europe regarding the integration of Muslim immigrants.65
With Europe’s new culture war raging, and far-right politicians like Belgium’s Filip
DeWinter gaining support by decrying ‘the Islamisation of Europe’, centre-right
politicians across the EU—many still in opposition and preparing for elections—
became even more critical of immigration and multiculturalism. On the campaign
trail, Germany’s Angela Merkel declared ‘the notion of multiculturalism has fallen
apart’.66 Both the CDU and CSU parties made opposition to Turkish accession a
prominent plank in their manifestos for the June 2004 European parliamentary
elections. On a visit to Ankara that February, Merkl expressed her view that the
EU was ‘overburdened’ and proposed that the EU should negotiate a ‘privileged
partnership’ with Turkey instead of full membership.67
EU commissioner Frits Bolkestein, a leading member of the Dutch liberal-
conservative VVD party, was particularly outspoken in his opposition to Turkish
accession. In a book published in March 2004, Bolkestein praised eastern
enlargement as ‘an historic, a political, a moral obligation’. However, he wrote, the
European Union ‘requires an outer limit, a fixed border . . . a Europe without
frontiers, an economic ideal but a political mirage. . . . A geographical limitation of
the Union is therefore logical. In the south, the Mediterranean Sea forms a natural
divide between Europe and Africa. In the east, there is a geo-political need for a
buffer zone between the EU and Russia, which might be formed by the countries
that do not belong to either block. A similar buffer to the south-west would also
be advantageous, in order to cushion the Union against Syria, Iran and Iraq’
(Bolkestein 2004: 22–4). Another outspoken VVD member, Geert Wilders, left
the party in September 2004 to launch a new party opposed to Turkish accession
(Arts and van der Kolk 2006).
The public debate over European identity and EU enlargement did not let up.
In November, Giscard declared his belief that the ‘foundations’ of European
identity include ‘the cultural contributions of ancient Greece and Rome, the
religious heritage pervading European life, the creative enthusiasm of the
renaissance, the philosophy of the Age of the Enlightenment and the contribu-
tions of rational and scientific thought. Turkey shares none of these.’68 Such scep-
ticism about liberal values is not new in Europe: ‘anti-modernist Catholics,
conservatives, extreme rightists as well as communists’ have long held visions of
65 Muslim School in Netherlands Bombed, The New York Times, 8 November 2004; Le débat sur
l’intégration des musulmans agite l’opinion et la classe politique allemandes, Le Monde, 22 novem-
bre 2004.
66 Germans argue over integration, BBC News, 30 November 2004; Europe’s far-right parties hold
coalition talks, The Guardian, 10 December 2004.
67 Merkl’s Message to Turkey—‘No’, Deutsche Welle, 16 February 2004. See also Angela Merkl,
Türkei: Partnerschaft statt EU-Mitgliedschaft, Die Welt, 16 Oktober 2004.
68 A better European bridge to Turkey, Financial Times, 24 November 2004.
Europe and its nations ‘based on an idea of community rooted in pre-modern

religious ideas, cultural or ethnic homogeneity, or even in coercion and violence’
(Gosewinkel 2015: back cover). In this sense, the normative shift around 2005 can
be understood in part as a re-emergence of ideas hidden for several decades by a
liberal consensus among EU elites.
The EU’s Commissioner for Enlargement, Olli Rehn, tried to push back: ‘I am
often asked where Europe’s ultimate borders lie. My answer is that the map of
Europe is defined in the minds of Europeans. Geography sets the frame, but
fundamentally it is values that make the borders of Europe. Enlargement is a
matter of extending the zone of European values, the most fundamental of which
are liberty and solidarity, tolerance and human rights, democracy and the rule of
law.’69 Before long, though, it was not only opposition politicians and officials
without decision-making powers who were expressing scepticism about the EU’s
enlargement policy.
In advance of the EU’s December 2004 summit, where Turkey’s candidacy
would be on the agenda, a handful of governments made announcements that
were clearly against the spirit of the norm that all liberal democratic states in
Europe were eligible for membership. French president Jacques Chirac announced
in October that while he still supported Turkey’s quest to join the EU, he would
also support changing France’s constitution to require a referendum on the acces-
sion of any new member.70 Just hours after the December summit authorized
accession negotiations with Turkey, Wolfgang Schüssel announced that Austria
too would hold a referendum on any eventual Turkish accession.71
In the spring of 2005, French and then Dutch voters rejected the proposed EU
constitutional treaty—a rejection that was widely attributed, at least in part, to the
December decision to open accession negotiations with Turkey.72 In August 2005,
Angela Merkel and Edmund Stoiber wrote to their fellow conservatives across
Europe: ‘We are firmly convinced that Turkey’s membership would overtax the
EU politically, economically and socially and endanger the process of European
integration.’73 The following month, Merkel’s CDU party and its CSU allies won
federal elections in Germany and she took office as Chancellor on 22 November.
This was a critical turning point. During the early 2000s, high-profile politi-
cians questioning the EU’s membership norm by arguing that Turkey was ineli
gible for accession because of its religious/cultural heritage, such as Stoiber,
Merkel, Bolkestein, and Giscard, were not in government—only Austria’s
69 Values define Europe, not borders, Financial Times, 4 January 2005.
70 Chirac backs French vote on Turkey, EurActiv.com, 4 October 2004.
71 Austria says it will have referendum on Turkey, EUobserver.com, 17 December 2004.
72 A big, loud French ‘non’, The Economist, 29 May 2005; Réflexions sur la crise de l’opinion à l’égard
de l’Europe, Le Monde, 15 juin 2005. See also Arts and van der Kolk (2006) and Stefanova (2006).
73 Turkey Vote Could Mean the End of a Courtship to Join the E.U., The New York Times, 17
April 2017.
Wolfgang Schüssel sat in the EU Council in this period. But the rise to power of
Angela Merkel in 2005 (and then of Nicolas Sarkozy in 2007) signalled a growing
bloc of member states ruled by politicians who did not hide their scepticism
about multiculturalism or their readiness to include cultural similarity as require-
ment for EU accession. As such, the increasingly powerful political forces that
had been arguing since 2001 for a redefinition of the limits of Europe were now
well represented within the Council. The combination of unprecedented public
demands and powerful voices within the Council shattered the pre-existing con-
sensus on the EU as a community composed of and open only to liberal
democracies.
In this new environment, we would expect the EU to retain its treaty
commitments while emphasizing new criteria and procedures for decision-
making on enlargement that were less likely to facilitate accession by controversial
applicant and candidate states. In particular, if the new public and elite discourse
described above did indeed herald a normative change within the institutions,
and especially within the Council, we would expect to see a changed discourse on
the criteria for enlargement and the emergence of hard bargaining over particular
cases. Although the EU’s 30- year rule means that we cannot gain direct
documentary evidence of whether, and if so how, this new discourse affected
membership norms prevailing within EU institutions, we can assess whether the
empirical record is consistent with the hypothesis of normative change.
This consistency is evident in the unprecedented focus after 2005 on the EU’s
‘absorption capacity’, which was effectively a way to say no to enlargement without
openly defying the Union’s treaty commitments. The concept had first been
mentioned in the Commission’s 1976 opinion on Greek accession and then
formalized in the 1993 Copenhagen Summit’s conclusions, which stated ‘The
Union’s capacity to absorb new members, while maintaining the momentum of
European integration, is an important consideration in the general interest of
both the Union and the candidate countries.’ Despite this formalization, it had
been largely ignored after 1993 in favour of a focus on the applicant countries’
readiness and ability to meet the Union’s political, economic, and administrative
criteria. Yet almost immediately after Merkel took power, the European
Commission and European Parliament shifted their discursive focus to the
union’s ‘absorption capacity’.
Just three months after the Merkel/Stoiber letter warned that Turkish accession
would ‘overtax the EU’, and six weeks after their CDU/CSU alliance won
Germany’s elections, the Commission included an unprecedented discussion of
‘absorption capacity’ in its annual enlargement strategy report: ‘The Union’s
capacity to absorb new members, while maintaining the momentum of European
integration, is an important consideration in the general interest of both the
Union and the candidate countries. . . . The Union has to ensure it can maintain its
capacity to act and decide according to a fair balance within its institutions;
respect budgetary limits; and implement common policies that function well and
achieve their objectives.’74 But with the normative consensus on a liberal demo-
cratic membership norm broken, there was pressure to move beyond this techno-
cratic definition of the absorption capacity.
As Austria assumed the EU presidency in January 2006, Schüssel began to
push for a cultural definition of European identity linked to the EU’s ability to
absorb new member states. When presenting the Austrian presidency’s programme
to the European Parliament, Schüssel declared that discussion about the future of
Europe ‘is also about the boundaries of Europe. Translated into concrete terms,
this means the criteria that determine the absorption capacity of the Union. These
boundaries must not be drawn by surveyors or geographers. They are a political
matter. The political debate must focus on absorption capacity.’75 Just over a week
later, at a conference on the future of Europe, Schüssel insisted ‘This Europe must
not become a purely economic idea; I say that quite openly here. Europe must be
more, it must find a cultural identity, reflect on what holds us together, where its
borders lie . . .’76
But the Austrian government was not alone in pushing this theme. In late
March 2006, leading figures from conservative and Christian Democrat parties
across the EU gathered in Rome for two ‘study days on Europe’, where several
speakers urged the participants not to forget Europe’s ‘Christian roots’.77 After the
meeting, French interior minister Nicholas Sarkozy once again spoke out against
a ‘Europe without borders’.78 If translated into policy, this new focus on the cul-
tural sources of European identity would impose significant limits on the EU’s
openness to liberal democracies with clear implications for the EU’s approach to
enlargement.
At their March 2006 meeting in Salzburg, the EU’s foreign ministers reiterated
that the Union’s ‘absorption capacity has to be taken into account’ in enlargement
decision-making.79 One week later, the European Parliament asked the
Commission to ‘submit a report by 31 December 2006 setting out the principles
which underpin [the Union’s] absorption capacity’. Hinting at the answer they
wanted, the MEPs indicated, ‘defining the nature of the European Union, includ-
ing its geographical borders, is fundamental to understanding the concept of
absorption capacity’. The Parliament also declared, ‘the capacity for absorption of
the Union . . . remains one of the conditions for the accession of new countries’
74 AEI: Enlargement strategy paper. Communication from the Commission. COM (2005) 561
final, 9 November 2005.
75 AHPE: European Parliament, Official Journal, Debates, 18 January 2006.
76 CVCE: Record of the proceedings of the conference ‘The Sound of Europe’ on the future of
Europe. Salzburg, 27 and 28 January 2006.
77 Centre-right EPP revives EU constitution God debate, EUobserver.com, 29 March 2006.
78 The EU’s weary travellers, The Guardian, 4 April 2006.
79 Salzburg EU/Western Balkans Joint Press Statement, European Commission, C/06/77, 11
March 2006.
and called on the Commission to ‘factor this into the overall negotiation
timetable’.80 Austrian chancellor Wolfgang Schüssel didn’t hide his satisfaction
with these developments: ‘For the first time, we are now discussing issues like
Europe’s geographical borders and . . . the question of the EU’s capacity to absorb
new members. If you don’t see this as progress, I can’t help you.’81
At the June 2006 summit concluding the Austrian presidency, the Fifteen
instructed the Commission to prepare a report on absorption capacity, which
Schüssel commented was not simply about finances and institutional efficiency: it
also had a cultural dimension, he said. Danish Prime Minister Anders Fogh
Rasmussen insisted that the focus on absorption capacity was ‘not a new criteria’,
which was technically correct but blind to the concept’s sudden rise in political
salience and its new cultural dimension. French President Jacques Chirac
admitted that absorption capacity was partly about responding to popular
anxieties about enlargement, so there must be a means for the citizens of member
states ‘to be able to say if they accept [the candidate state] or not’.82 Such com-
ments led one journalist to conclude, ‘the west European centre right has come up
with an inelegant euphemism for Europe’s borders: “absorption capacity.” In plain
English, it means “we can’t take any more, probably” .’83
On the other hand, a number of EU governments remained committed to the
liberal democratic definition of European identity. For example, Swedish Foreign
Minister Carl Bildt criticized ‘absorption capacity’ as a ‘buzzword’ and a ‘flawed
concept’ that obscures the transformative potential of European integration. He
thus decried ‘those who want to slow down or perhaps even stop the process [of
enlargement] altogether. There is talk of defining the borders of Europe. Drawing
big lines on big maps of the east of Europe risks becoming a dangerous process.
We should know that such a process will have profound effects in those areas or
nations that fear ending up on the other side of those lines. We could easily see
forces of atavistic nationalism or the submission to other masters taking over
when the light of European integration—however vague or distant—is put out. If
that happens, the lines on the maps will certainly not protect us from the conse-
quences of what happens beyond them.’84 It is thus apparent that the Council was
deeply divided by 2006 over the nature of the EU as a political community and
thus over the definition of which neighbouring states were eligible for member-
ship, in stark contrast to earlier periods, when the member states accepted a
80 Future enlargement: ‘absorption capacity’ coming to the fore?, EurActiv.com, 23 March 2006.
81 Interview with Austrian chancellor Schüssel: Promoting the ‘European way of life’, Spiegel Online
International, 20 March 2006.
82 EU cements ‘absorption capacity’ as the new stumbling block to enlargement, EurActiv.com, 16
June 2006.
83 The EU’s weary travellers. The Guardian, 4 April 2006.
84 Open wide Europe’s doors, International Herald Tribune, 7 November 2006.
common definition of European identity regardless of their preferences regarding

particular applicant states.
The Commission’s November 2006 report on EU enlargement strategy
acknowledged broad public uncertainty about enlargement but sought to replace
discussion of absorption capacity (which sounded like an advertisement for an
infant’s undergarments) with ‘integration capacity’, conceived as a set of institutional
and financial challenges for current member states to address rather than as an
insurmountable obstacle to new member states.85 It also expressed the
Commission’s view of key treaty provisions: ‘Article 49 of the Treaty on European
Union . . . states that “Any European State which respects the principles set out in
Article 6(1) may apply to become a member of the Union.” However, this treaty
provision does not mean that all European countries must apply, or that the EU
must accept all applications. The European Union is defined first and foremost by
its values.’86 But the Commission’s arguments could not hide the growing discord
among the member states regarding the political character of their community
and thus the potential limits on its enlargement.
Where an uneasy consensus had once reigned, there was now a stark dissensus
that would make it far easier for member states to pursue their enlargement
preferences without being stigmatized by their peers. This breakdown in
normative consensus is reflected in Finnish foreign minister Erkki Tuomioja’s
cautious statement on enlargement in mid-2006, as his country was about to
assume the EU’s rotating presidency: ‘The EU has to be ready to take all those
European countries whose population wants to join Europe and that fulfil the
criteria—and please don’t ask me where the borders of Europe are, that’s some-
thing we didn’t want to put on the agenda.’87 But avoiding the issue didn’t make it
go away, as Nicolas Sarkozy made clear upon accepting his party’s nomination for
the 2007 presidential elections: ‘I want . . . to say that Europe must give itself bor-
ders, that not all the countries of the world have the vocation to join Europe,
beginning with Turkey. Enlarging without limit risks destroying the European
political union, which I do not accept.’88 Three months later, Sarkozy was elected
President of France.
By the end of the decade, the community’s formal rule on membership was
clearly divorced from the norms prevailing among the member states. The 2009
Lisbon Treaty reiterated a familiar idea, ‘The Union is founded on the values of
respect for human dignity, freedom, democracy, equality, the rule of law and
respect for human rights, including the rights of persons belonging to minorities.’
85 Nappy mirth day to EU, The Guardian, 8 November 2006; European Commission, Olli Rehn,
Enlargement package, press speaking points, 8 November 2006, SPEECH/06/663.
86 European Commission. Enlargement Strategy and Main Challenges 2006–2007. COM(2006) 649,
Brussels, 8 November 2006.
87 Les négociations d’adhésion UE–Turquie pourraient être en danger, Le Monde, 29 juin 2006.
88 Le discours d’investiture de Nicolas Sarkozy, Le Monde 15 janvier 2007.
It also repeated that any European state committed to these principles may apply
for membership. But such treaty language was no longer backed by a normative
consensus among the member states regarding the characteristics that would
make an applicant state eligible for membership. The normative consensus on this
latter point, which had undergirded similar rules in the Single European Act and
the (Maastricht) Treaty on European Union, was gone.
The resulting misfit between clear rules and normative dissensus is evident in
Italian foreign minister Franco Frattini’s 2011 plea, ‘The idea of geographic limits
based on conventional criteria is not a suitable guide for the enlargement process.
Instead, the EU should continue to offer the prospect of membership for those
countries that share our principles of democracy, freedom, and the rule of law and
are ready to assume the rights and obligations of membership, in accordance with
the European treaties and the Copenhagen criteria’ (Bini and Angelescu 2011: ix).
So while the legal definition of membership eligibility in liberal democratic terms
was not revoked, it lost normative status and thus political power.
When Jean-Claude Juncker became president of the European Commission in
2014, he made clear that further enlargement was not on the agenda of his
presidency. Although no candidate country was likely to be ready for accession
during this period, the move was clearly motivated as well by the breakdown in
consensus among the member states over which non-members were eligible for
accession and which were not. As shown in Chapter 7, this dissensus was evident
in debates on particular aspirant and candidate states throughout the Juncker
presidency.
5. Conclusions
The evolution of EU membership norms over more than six decades confirms
that the limits of regional communities are contested and redefined over time.
The contestation and redefinition described in this chapter illustrates the heuristic
value of the typology of normative change presented in the previous chapter,
including external challenge, internal challenge, rhetorical inconsistency, and
shifts in public opinion. The governments of western Europe perceived an urgent
external (Soviet communist) challenge in the mid-1950s and this motivated them
not only to create the European Economic Community but also to define its
membership eligibility in terms of non-Communism. By the early 1960s, the
external challenge had diminished but the six EEC governments were challenged
by parliamentarians and societal actors within the community whose political
mobilization and discursive tactics led the EEC to redefine itself as a community
of parliamentary democracies. Around 1970, the gap between the community’s
democracy norm and the member states’ broader constitutional and international
legal commitments to protect human rights and fundamental freedoms created
a rhetorical inconsistency that was only resolved by the broadening of the

membership norm. Finally, around 2005, shifts in public opinion across Europe
weakened many of the member states’ commitment to the definition of the EU as
a community of liberal democracies, and thus contributed to a breakdown in
normative consensus that persists to this day.
This is undeniable evidence, contrary to the rhetoric one often hears from EU
officials, that the promotion of liberal democracy has not always been central to
the member states’ shared understanding of their joint endeavour. It is therefore
mistaken to suggest, as many scholars do, that ‘The European Union (EU) from
its foundation had a commitment to democracy, bundling national democratic
institutions and European institutions in a mutually supporting manner to
solidify peace’ (Pevehouse and Russett 2006: 974). In fact, EU membership norms
were always contested and changed fundamentally at several points in the course
of European integration. In the latest, post-2005 period, the EU’s member states
and supranational institutions no longer agree on a definition of the European
Union’s political limits, and thus lack consensus on the terms by which applicant
states’ eligibility for membership should be judged.
Chapter 4 utilizes this evidence of the changing content of EU membership
norms, in addition to many other empirical indicators, in a multi-method analysis
of competing explanations of EU decision-making over six decades.
4
EU Membership Eligibility in Statistical
and Comparative Perspective
One of the causes of the trial and error and reversals of the European
idea [has been] the changing conceptions, according to political
currents, of the appropriate geographical area of a united Europe.
—Pierre Werner, Prime Minister of Luxembourg, 19661
This chapter offers various empirical tests of competing arguments regarding the
factors that shape a regional community’s decision-making on the membership
eligibility of non-member states. Simply put, it addresses the question of whether
EU decision-making on the membership eligibility of applicant states is better
explained by the fit between the membership norm prevailing within the com-
munity and the character of the applicant state, or by other factors. As discussed
in the first two chapters, the relative explanatory strength of competing argu-
ments should be evident in two forms: first, in the pattern of the community’s
decisions over time (i.e. which states are deemed eligible and which are not), and
second, in the process of community decision-making (i.e. how particular argu-
ments and political moves shaped particular decisions). This chapter assesses the
first measure by testing the fit between the six theories’ expectations regarding
EU decisions on membership eligibility and outcomes in each of the 48 cases
where the EU (including its predecessors, the EEC and the EC) made such a deci-
sion between 1957 and 2017. The second measure, requiring detailed evidence of
the EU’s decision-making process, is the focus of later chapters.
Following this introduction, Section 1 presents the parameters of the dataset of
EU decisions, including criteria for case selection, sources, and coding and
introduces the three analytical methods—cross tabulation, logistic regression
analysis, and qualitative comparative analysis—used to assess the explanatory
power of the various theoretical approaches. Section 2 examines each theoretical
argument in turn, including a brief summary of its logic, an explanation of how
its key concept(s) are converted into measurable variables. Section 3 presents the
1 AEI: Pierre Werner. Aspects de l’évolution récente de la construction européenne, 5 juin 1966.
cross-tabulation results showing the proportion of cases whose outcomes fit each
theory’s expectations. Section 4 discusses the significance and limitations of the
cross-tabulation analysis. Section 5 presents the design and results of the first
robustness check using logistic regression analysis, and Section 6 presents the
second robustness check using Qualitative Comparative Analysis. Finally, Section 7
summarizes the results of the various analyses and transitions to the chapters
that follow.
1. Data and methods
Working with a variety of primary and secondary sources, we have identified 48

cases during the period 1957–2017 in which the EU took a position on the
membership eligibility of a non-member state, including 37 cases where the state
was deemed eligible for membership and 11 cases where it was deemed ineligible.
Notwithstanding the nuances of each decision, we reduce these outcomes to a
binary variable (eligible–ineligible) in order to facilitate comparison of the
various arguments. A few cases where the EU took an ambiguous position are
coded as ineligible. Some but not all of the states deemed eligible have now
completed the accession process and joined the Union.
Most states in the dataset (Table 4.1) only appear once but some appear several
times, either because they were declared ineligible and later re-applied or because
they were declared eligible but subsequent events led the EU to revisit its position.
It is therefore possible to assess whether variation in the indicators associated
with the six explanations fit the observed pattern of decision- making on
membership eligibility.
Some of the 48 cases are well known to scholars and EU observers, while others
are virtually unknown despite their contribution to the course of European
integration. This mapping of EU decision- making differs considerably from
conventional accounts of EU enlargement, which typically present cases in terms
of regional ‘waves’ of accession: the northern enlargement in the 1970s (Denmark,
Ireland, UK), the southern enlargement in the 1980s (Greece, Portugal, and
Spain), the neutrals enlargement in the 1990s (Austria, Finland, Sweden), the
eastern enlargement in the early-mid 2000s (former Communist states plus
Cyprus and Malta), followed by Croatia in 2013. Notwithstanding their
significance in other regards, these waves reflect when accession treaties were
concluded and thus tell us little about decision-making on the membership
eligibility of particular applicants.
Several situations where the government of a non-member state declared its
interest in accession are not included in the dataset, either because the government
soon reversed its position (Switzerland 1992) or the community’s member states
were distracted by unrelated events from making a collective decision on
Membership Eligibility in Statistical and Comparative 87
Table 4.1 EU decisions on the membership eligibility

of aspirant states
Aspirant state Year Eligibility decision
Greece 1959 Eligible

Turkey 1959 Eligible
Spain 1960 Eligible
United Kingdom 1961 Eligible
Denmark 1961 Eligible
Norway 1962 Eligible
Spain 1962 Not eligible
Ireland 1962 Eligible
United Kingdom 1963 Not eligible
Greece 1967 Not eligible
United Kingdom 1967 Not eligible
United Kingdom 1970 Eligible
Greece 1976 Eligible
Spain 1977 Eligible
Portugal 1978 Eligible
Morocco 1987 Not eligible
Czechoslovakia 1990 Eligible
Poland 1990 Eligible
Hungary 1990 Eligible
Bulgaria 1991 Eligible
Austria 1992 Eligible
Finland 1992 Eligible
Sweden 1992 Eligible
Romania 1992 Eligible
Slovak Republic 1993 Eligible
Czech Republic 1993 Eligible
Cyprus 1993 Eligible
Estonia 1994 Eligible
Latvia 1994 Eligible
Lithuania 1994 Eligible
Slovenia 1995 Eligible
Malta 1995 Eligible
Moldova 1997 Not eligible
Slovak Republic 1997 Not eligible
Slovak Republic 1999 Eligible
Ukraine 1999 Not eligible
FYR Macedonia 2000 Eligible
Croatia 2000 Eligible
Albania 2003 Eligible
Serbia & Montenegro 2005 Eligible
Bosnia-Herzegovina 2005 Eligible
Ukraine 2005 Not eligible
Serbia & Montenegro 2006 Not eligible
Montenegro 2006 Eligible
Serbia 2007 Eligible
Iceland 2010 Eligible
Kosovo 2013 Eligible
Georgia 2013 Not eligible
eligibility and the government subsequently clarified that it was only seeking to
develop commercial ties (Portugal 1962 and 1969). The most ambiguous of the
cases not included in the dataset is that of Cape Verde, whose Prime Minister José
Maria Neves spoke openly starting in early 2005 about his interest in pursuing EU
membership for his country (Vieira and Ferreira-Pereira 2009). However, the
Cape Verdean government never approached the EU formally about accession.2
Cape Verde did inquire in Brussels in late 2005 about possible participation in the
European Neighbourhood Policy (ENP). Despite a supportive resolution by the
European Parliament in January 2006, which recognized Cape Verde’s ‘geograph
ical proximity, cultural and historical affinity’ to Europe, the country’s inclusion in
ENP was later denied by the EU on geographic grounds. One could imagine, given
Neves’ previous rhetoric, that the EU may have viewed the ENP application as a
first step in a quest for accession, in which case the community’s negative reply
would be de facto a negative decision on membership eligibility. However, there is
no evidence in Council or Coreper records that consideration of Cape Verde’s ENP
application was in any way linked to the possibility of accession.3 For this reason,
the 2006 decision on ENP is not included in the dataset.
It is also important to note that Table 4.1 does not include situations where the
EU has quietly sustained (i.e. not reversed) its earlier acceptance or denial of an
applicant state’s eligibility. For example, after the EEC accepted Turkey’s eligibility in
1959, it could have but did not reverse that decision after the military takeover and
political repression in Ankara the following year. More recently, after the EU
rebuffed Ukraine’s pursuit of a membership perspective in 2011, it could have but
did not reverse that decision following the ‘Euromaidan’ revolt of 2013–14. Such
non-events (like Sherlock Holmes’ famous dog that didn’t bark) are notoriously dif-
ficult to categorize, but they can play an important role in counter-factual hypoth
esis testing when viewed in light of theories that suggest a different outcome
(Fearon 1991). Although they do not appear in Table 4.1, some situations of this
sort are clearly relevant to testing competing explanations of how the EU decides on
membership eligibility and are referenced as such in the empirical analysis.
Data on the outcome in each case comes from a variety of official EU docu-
ments and pronouncements, supplemented by contemporary media accounts,
2 During a visit to Lisbon in May 2005, Neves declared his plan to ‘go as far as possible’ in
cooperation with the EU and indicated that he would not be satisfied with ‘associate status’. The
following month, Cape Verde’s parliament debated the possibility of EU membership. On the other
hand, according to two leading Portuguese media sources, Cape Verde was only seeking ‘a special
status from the European Union’ and ‘the hypothesis of membership was never officially assumed by
the Portuguese and Cape Verdean governments, who always favoured the more consensual option of a
special strategic partnership’. Sources: Cabo Verde poderá apresentar proposta de adesão à UE ainda
este ano—PM, www.rtp.pt, 6 May 2005; Cabo Verde vai pedir adesão à UE, Correio da Manhã, 7 May
2005; Parlamento de Cabo Verde debatirá ‘adhesión’ a la Unión Europea, Afrol News, 10 June 2005;
Cabo Verde ganha estatuto especial com União Europeia. Publico.pt, 26 October 2007; author’s
correspondence with Jim Cloos, EU Council official, July 2017.
3 Author’s correspondence with Jim Cloos, EU Council official, July 2017.
scholarly sources, and interviews. Variations over time in the membership norms
themselves are drawn from the historical reconstruction in Chapter 3. Data on
each aspirant state’s status with regard to the various explanatory variables are
explained below in Section 3.
To test the explanatory power of the various theoretical arguments, we analyse
the dataset using three distinct empirical methodologies: cross tabulation, logistic
regression, and crisp set qualitative comparative analysis (csQCA). First, we evalu-
ate each case to determine whether the outcome fits the expectations derived from
the various hypotheses in light of the relevant values on their respective explanatory
variables. Then, for each argument, the correlations between explanatory variable
and outcomes are presented in cross-tab or contingency table format. Although
correlation is not equivalent to causation, a strong correlation between variables
with a plausible relationship in theory would indicate the possible presence of a
causal or constitutive relationship that merits further investigation, while a low cor-
relation indicates that a significant relationship between the variables is unlikely.
A strong correlation is thus necessary but not sufficient evidence of a causal or con-
stitutive relationship and a valuable step toward explaining EU decisions over time
in this important and under-studied area. Comparing the empirical fit of the five
arguments thus provides an initial assessment of their relative explanatory power.
We then ran two types of robustness check on this initial finding—one using
logistic regression and one using QCA. These three methods rely on different
logics of causation, but we have chosen to use them together to increase our
confidence that whatever pattern, if any, we observe in the data is indeed
indicative of a causal relationship. And if we observe no pattern in the data, the
triple methodology will increase our confidence that we have not overlooked
something important. Cross-tabulations and logistic regression define causation
in terms of the degree of correlation between explanatory variables and the
outcome variables; they are typically used to assess the relative power of
competing explanations. In contrast, QCA relies on a conjunctural logic better
suited to capturing independent multivariate causal pathways to the same
outcome. The point here is not to debate ontological or epistemological claims
regarding causation but to increase our confidence in the project’s empirical
findings by subjecting the data to multiple analytical methods.
2. Explanations and hypotheses
As previewed in Chapter 1 and elaborated in Chapter 2, there are at least six

distinctive explanations for variation in EU decision-making on membership
eligibility: membership norms, geography, treaty rules, regime type, commercial
interests, and geo-political interests. This section of the present chapter briefly
reviews each of the six explanations.
Membership norms
The principal argument proposed and tested in this book is that EU decisions on
the membership eligibility of states seeking accession are shaped by the outsider’s
conformity with membership norms prevailing within the Union—namely,
among member states’ governments and senior supranational officials. In
empirical terms, the expectation is that non-member states seeking accession that
fit the norm will be recognized as eligible for membership, while those that do
not fit the norm will be considered ineligible. As such, applicant states that are
considered eligible for membership in one period may be judged ineligible in a
later period, or vice versa, depending on changes in the fit between prevailing
membership norms and the characteristics of the applicant state.
for membership if it fits the community’s prevailing membership norm and will
reject states that do not, regardless of member states’ other motivations.
The other two hypotheses on membership norms presented in Chapter 2 are

tested through process-tracing in Chapters 5–8.
Geography
Geographic explanations of membership eligibility suggest that if a regional com-

munity aims to integrate or govern a particular geographic space, then any state
located fully or even partly within this space would be considered eligible for
membership, and states located elsewhere would be ineligible. Two slightly
different relationships are suggested by this logic—one focused on an applicant
state’s location with a certain limited area and the other on physical closeness to
others. To test this argument, we consider here three distinct propositions, focused
respectively on conventional geography, institutionally recognized geography,
and geographic contiguity.
conventionally associated with the region in question, and reject others.
that other, authoritative international institutions associate with the region in
question, and reject others.
for membership if its territory is contiguous to that of a current member state,
and reject others.
Treaty rules
The argument that rules in a regional community’s governing treaty determine its
decision-making on applicant states is consistent with two possible causal
mechanisms: either the governments of member states are persuaded by the
prescriptive content of the treaty rule or they value the credibility of the
community as a whole and the integrity of its decision-making process above
their own preferences regarding particular membership applicants and above
whatever norms may prevail within the community. In either case, this argument
yields the following testable expectation: a regional community will recognize the
membership eligibility of non-member states that fit the substantive rule on
memberships established by the community’s treaty, and reject others.
for membership if it fits the substantive rule on membership established by the
community’s core treaty, and reject others.
Regime type
States vary widely in their domestic structures and governance norms and this
variation may determine whether or not a regional community considers them
eligible for membership. In particular, we hypothesize that a regional community
will accord membership eligibility to aspirant states whose regime types are
similar to that of member states and treat others as ineligible. In the EC/EU
context, both during and after the Cold War, such variation has most often been
conceived in terms of a state’s degree of liberal democracy.
for membership if its regime type is similar to that of the member states, and
reject others.
There are at least two distinct ways that a regional community’s decisions on
membership eligibility may be shaped by the commercial interests of its member
states. The first proposes that the economic opportunities created by integrating
new states with prosperous populations and strong regulatory capacities would
motivate the community to accord membership eligibility to aspirants whose
level of economic development is similar or superior to that of the current
member states. Another possibility is that the members of a regional community
will see states with lower levels of economic development as attractive markets for
their goods and investment capital and thus welcome their eventual integration
into the community’s common market area while rejecting more developed states
whose own exports could threaten domestic producers and workers.
for membership if its level of economic development is similar or superior to that
of member states, and reject others.
Hypothesis 10: A regional community will decide that an aspirant state is eli
gible for membership if its level of economic development is below that of mem-
ber states, and reject others.
Security interests
In contrast to the geographical explanations focused on the physical location of

an applicant state, these explanations focus on the likely implications of
enlargement for the security of the community. Put simply, they suggest that a
regional community will welcome a non-member state whose accession would
reduce member states’ exposure to violent threats or improve their capability for
resisting such threats, and reject those whose accession would have the opposite
effect. There are at least two ways to conceptualize how the members of a regional
community understand the link between enlargement and security—one
suggesting that decisions on membership eligibility would be driven by the risk of
conflict associated with the aspirant state in question, and the other suggesting
that decisions would be driven by a desire to boost the coherence and credibility
of security alliances.
gible for membership if it does not pose a significant risk of involvement in inter-
state conflict or cross-border violence, and reject others.
gible for membership if it belongs to (or has a clear prospect for joining) the
security alliance to which most community members belong, and reject others.
3. Cross-tabulation analysis
As a first test of the various expectations, this section presents the results of cross-
tabulation analyses of the correlation between the potential explanatory variables
and the EU’s positions in the 48 cases. The six explanations are subjected to a total
of 16 tests using distinctive indicators of the various explanatory factors. The
higher the correlation or per cent fit, the stronger the predictive power of the
explanation and hypothesis in question; a 50 per cent fit indicates that the
expectation is incorrect as often as it is correct, like a coin toss. Hypotheses with
an observed correlation around or below 50 per cent are thus considered weak.
Membership norms
The explanatory variable here is the fit between the prevailing EU membership
norm and the character of the state seeking membership in the year that the latter
expresses its interest in accession. As discussed in the previous chapter, the EU’s
membership norms have evolved through four identifiable normative periods.
During the first three periods, the norms applicable to membership eligibility
indicated that states holding or seeking membership should be a non-communist
European state (1957–1961), a European parliamentary democracy (1962–1969),
and a European liberal democracy (1970–2005). The liberal democracy norm
broke down in late 2005 but the governments of the member states have not
converged on an alternative definition of membership eligibility, so the norms
hypothesis becomes indeterminate at that point. As a result, this hypothesis can
only be tested with 42 of the 48 cases; six post-2005 cases are excluded.
In order to assess various states’ conformity with the non-Communist norm
during the first period, we relied on membership in the leading military and
economic organizations of the Soviet bloc—the Warsaw Pact and the Council on
Mutual Economic Assistance. Both organizations had identical memberships in
this period: Albania, Bulgaria, Czechoslovakia, the German Democratic Republic,
Hungary, Poland, Romania, and the Soviet Union. To this list, we added
Yugoslavia, which was non-aligned but explicitly socialist in ideology. Given the
non-Communist norm prevailing in 1957–1961, we would expect any of these
states would be considered ineligible for membership in the European
Community, while any other European state would be considered eligible.
In order to assess a country’s status as a parliamentary democracy (the second
normative period) or liberal democracy (the third period), we rely principally on
the categories and data provided by Freedom House (FH). The FH dataset
includes a clearly defined ‘electoral democracy’ category that corresponds closely
to the ‘parliamentary democracy’ norm and a more stringent ‘Free’ category
limited to states that ‘can be considered both electoral and liberal democracies’,
meaning that they also ensure respect for civil liberties, including freedoms of
expression and belief, associational and organizational rights, rule of law, and
personal autonomy (Puddington et al. 2015). These are preferable for two reasons
to leading alternative indexes. The Varieties of Democracy dataset (V-Dem)
places states on various 1–10 scales without over-arching categories (Lindberg
et al. 2014) while Polity IV uses a 20-point scale divided into three categories
Table 4.2 Hypothesis 1: Membership norms
Fits membership Does not fit

norm membership norm
Eligible 27/42 = 64% 6/42 = 14%

Not eligible 4/42 = 10% 5/42 = 12%
(autocracy −10 to −6, anocracy −5 to +5, and democracy +6 to +10) that obscure
the distinction between parliamentary and liberal democracies.
However, in order to measure a non-member state’s conformity with the ‘par-
liamentary democracy’ norm using FH’s ‘electoral democracy’ category or its
conformity with the ‘liberal democracy’ norm using FH’s ‘Free’ category, we had
to overcome the FH dataset’s limited temporal coverage, which extends back only
to 1989 for the electoral democracy indicator and to 1972 for the Free/Not Free
distinction. As a result of this limited range, none of the cases of countries want-
ing to start membership negotiations during the period of the ‘parliamentary
democracy’ norm is covered by FH’s electoral democracy indicator, while one
case (the UK in 1970) covered by the ‘liberal democracy’ norm precedes FH’s
Free/Not Free categorization. To solve this problem, we use V-Dem data that
covers the full post-1945 period to impute missing country scores on the FH
variables. For discussion of this technique, see the book’s Appendix.
As shown in Table 4.2, of the 42 cases of EU decision-making on the member-
ship eligibility of a non-member state that occurred while the EU had determinate
membership norms, 32 or 76 per cent of the cases resulted in decisions that fit the
prevailing norm, including cases of eligibility when the non-member state fit the
norm and cases of ineligibility when it did not fit the norm; only 10 cases or 24 per
cent fell outside the hypothesized pattern. The membership norms argument does
not predict a particular outcome for the six cases that occurred when the EU
lacked a determinate norm; instead, it simply expects that such decisions will be
shaped by interest-based bargaining rather than by normative constraints.
This evidence of a strong relationship between the content of EU membership
norms and the outcome of EU decisions on membership eligibility suggests that
the former may play an important role in regional community decision-making.
However, we cannot really assess its significance before comparing it to evidence
of how other possible explanatory variables relate to the same outcomes.
Geography
The geography explanation of membership eligibility can be tested in various

ways. The first of these (Hypothesis 4) focuses on a conventional geographic
definition of Europe: the European Union will accord membership eligibility to
non-member states whose territory is located physically within the geographic

space conventionally identified as Europe, and reject others. To test this expect
ation we first use Encyclopædia Britannica’s definition of the European continent:
‘Geographically, Europe is the Western peninsula of Eurasia, conventionally
delimited by the Mediterranean Sea, the Black Sea, and the Caucasus Mountains
in the South, the Atlantic Ocean in the West, the Arctic Ocean in the North, and
the Urals and the Caspian Sea in the East’ (cited in Schimmelfennig 2016: 178–9).
By this definition, small parts of Kazakhstan and Turkey and a large part of Russia
lie within the furthest eastern and southeastern reaches of Europe, but most of
Turkey, all of Cyprus, and North Africa lie beyond Europe (Table 4.3).
This test suggests a strong relationship between conventional understandings
of Europe and EU decisions on membership eligibility. If we assume that the
approximately 5 per cent of Turkish territory that lies north of the Bosporus
qualifies the entire country as lying within the conventional geographic limits of
Europe, then 37 of 48 cases lie within conventional Europe and were judged
eligible for EU membership while one case lies outside Europe and was considered
ineligible, meaning that 37 of 48 or 77 per cent of the cases fit the hypothesis. On
the other hand, 10 of 48 lie within conventional Europe but were judged ineligible
while one lies outside conventional Europe but was considered eligible for
membership, meaning that 23 per cent defy the hypothesis.
It is also possible to test this hypothesis using a definition of Europe’s geo-
graphic limits accepted by another authoritative international institution. To this
end, we considered two different regional classifications used by the United
Nations. One, the UN General Assembly’s regional voting bloc ‘West European
and Others Group’, is too clearly political to be useful here: Australia, Canada,
Israel, and New Zealand are permanent members and the USA has Observer sta-
tus. Instead, we used the regional codes developed ‘for statistical processing pur-
poses’ by the United Nations’ Statistics Division, whose ‘Europe’ category includes
53 countries and areas including micro-states like Andorra and Monaco and
autonomous sub-state units like the Åland Islands and Gibraltar.4
When tested against the 48 cases of EU membership (in)eligibility, this test
supported the conventional geography hypothesis just as strongly overall as the
Encyclopædia Britannica test: 35 of 48 cases lie within this UN definition of
Table 4.3 Hypothesis 4: Geographic location

(Encyclopædia Britannica)
Within Europe Beyond Europe
Eligible 36/48 = 75% 1/48 = 2%

Not eligible 10/48 = 21% 1/48 = 2%
4 United Nations, Statistics Division. Standard Country or Area Codes for Statistical Use (M49).
http://unstats.un.org/unsd/methods/m49/m49regin.htm.
Europe and were judged eligible for EU membership while two cases lie outside
the UN definition and were considered ineligible, meaning that 37 of 48 or 77 per
cent of the cases fit the hypothesis. On the other hand, nine of 48 lie within the
UN definition of Europe but were judged ineligible while two lie outside the UN’s
‘Europe’ but were considered eligible for membership, meaning that 23 per cent
defy the hypothesis (Table 4.4).
An alternative version of the geographic argument focuses not on an a priori
definition of the region’s limits but on the non-member state’s geographic proxim-
ity to the community’s current external frontier. The expectation (Hypothesis 6)
would be as follows: The European Union will accord membership eligibility to
non-member states whose territory is contiguous to that of a current member
state, and reject others. To test this expectation, we define ‘contiguous’ to mean
sharing a land border with an EU member state or separated from the territory of
a member state by no more than 50 miles (80 km) of sea. Some studies of member-
ship in international organizations use a higher measure of separation by sea, but
given our focus on regional organizations, the 50-mile measure (also used by
Mansfield and Pevehouse 2013) seemed most appropriate. By this logic, geography
determines the zone of potential membership eligibility but there is no ultimate
frontier of eligibility as long as the community’s external border keeps expanding.
When tested against the 48 cases of EU membership (in)eligibility, this hypoth
esis proved far weaker than the preceding conventional geography hypotheses: 21
of 48 cases were contiguous to a member state and were judged eligible for EU
membership while four cases were non-contiguous and were considered ineli
gible, meaning that only 25 or 52 per cent of the cases fit the hypothesis. On the
other hand, seven of 48 were contiguous but not accorded eligibility while 16 were
accorded eligibility despite being non-contiguous, meaning that 48 per cent defy
the hypothesis (Table 4.5).
Table 4.4 Hypothesis 5: Geographic location

(United Nations)
Within Europe Beyond Europe
Eligible 35/48 = 73% 2/48 = 4%

Not eligible 9/48 = 19% 2/48 = 4%
Table 4.5 Hypothesis 6: Geographic contiguity
Contiguous Not contiguous
Eligible 21/48 = 44% 16/48 = 33%

Not eligible 7/48 = 15% 4/48 = 8%
Treaty rules
According to Article 237 of the 1957 Treaty of Rome, ‘Any European State
may apply to become a member of the Community. . . . The conditions of
admission and the adjustments to this Treaty necessitated thereby shall be
the subject of an agreement between the Member States and the applicant
State.’ A literal reading of this rule indicates that only ‘European’ states may
be considered for membership but not necessarily that all ‘European’ states
would be deemed eligible. However, the founding member states could have
included a more stringent definition of eligibility in the treaty, and the treaty’s
travaux préparatoires reveal that this option was indeed considered (see
Chapter 3). It is therefore reasonable to conclude that the Treaty of Rome
deemed all ‘European’ states minimally eligible for membership. We use
Encyclopædia Britannica’s version of the conventional geographic definition
of ‘Europe’ (see above) to test this requirement, which remained in effect for
decades.
In fact, there are two ways to interpret when the EU treaty rules on member-
ship eligibility changed, so we test them both. By one reading, the 1986 Single
European Act (SEA) did not modify the Treaty of Rome’s indication that ‘All
European states may apply . . .’ but the multiple and explicit statements of the com-
munity’s commitment to democracy, fundamental human rights, and the rule of
law contained in its Preamble removed any doubt that these factors had become
treaty requirements. By this reading, the treaty rule hypothesis would expect the
community to accord membership eligibility to all ‘European’ applicant states
(measured by the Encyclopædia Britannica definition) in the years prior to 1986,
and only to European liberal democracies (measured by the Freedom House ‘free’
category) in the years since 1986. As shown in Table 4.6, this test yields an
explanatory fit of 69 per cent.
Alternatively, a reading that focuses on the treaties’ articles rather than their
preambular text would conclude that EU treaty rules on membership eligibil-
ity did not change until the 1992 (Maastricht) Treaty on European Union,
which retained the ‘All European states may apply’ formula but added that the
Member States’ governments are ‘founded on the principles of democracy’ and
constitutionally committed to respect fundamental rights. By this reading
(Hypothesis 4, test 2), the treaty rule hypothesis would expect all ‘European’
applicant states (measured by the Encyclopædia Britannica definition) to be
accorded membership eligibility prior to 1992, and only European liberal
democracies (measured by the Freedom House ‘free’ category) in the years
since 1992. This test yields an explanatory fit of 68 per cent, just like the first
test (Table 4.7).
Table 4.6 Hypothesis 7, test 1: Treaty rules (change 1986)
Fits treaty rules Does not fit treaty rules
Eligible 28/48 = 58% 9/48 = 19%

Not eligible 6/48 = 13% 5/48 = 10%
Table 4.7 Hypothesis 7, test 2: Treaty rules (change 1992)
Fits treaty rules Does not fit treaty rules
Eligible 28/48 = 58% 9/48 = 19%

Not eligible 6/48 = 13% 5/48 = 10%
Regime type
States vary widely in their domestic structures and governance norms and this
variation may determine whether or not a regional community considers them
eligible for membership. In particular, we hypothesize that a regional community
will accord membership eligibility to aspirant states whose regime types are simi-
lar to that of member states and treat others as ineligible. In the EU content, both
during and after the Cold War, such variation has most often been conceived in
terms of a state’s degree of liberal democracy. To test this hypothesis, we test the
effect of regime similarity in two ways: first we test whether the EU only accords
membership eligibility to aspirant states whose liberal democracy score is equal
to or greater than that of the lowest-scoring member state (Table 4.8), and then we
test whether it limits eligibility to aspirant states whose score is equal to or greater
than the mean score of all member states (Table 4.9). For both tests, we use the
V-Dem scoring of liberal democracy since 1900, which includes all of our 48
cases. As shown below, the community-minimum test yields an explanatory fit of
60 per cent and the community-mean test yields a fit of just 44 per cent.
Regime type is thus a poor predictor of membership eligibility. The ‘liberal
democracy’ score of aspirant states, whether tested in comparison to the member
states’ mean or to the lowest-scoring member state, does not correlate consistently
with EU decisions. In short, the EU has not required that aspirant states exhibit a
level of liberal democracy equal or greater than of states already belonging to the
community. It is also interesting to note that the two correlations associated with
this variable (44 per cent and 60 per cent) are inconsistent with the opposite
hypothesis, that the EU prioritizes aspirant states with a poor score on liberal
democracy. This does not mean that regime type is irrelevant to EU decision-
making—it clearly plays a role in the membership norms and treaty rules
explanations (see results above and discussion below)—but it is remarkably weak
as a stand-alone variable.
Table 4.8 Hypothesis 8, test 1: Liberal democracy (EU minimum)
V-Dem Score >= EU minimum V-Dem Score <EU minimum
Eligible 20/48 = 42% 17/48 = 35%

Not eligible 2/48 = 4% 9/48 = 19%
Table 4.9 Hypothesis 8, test 2: Liberal democracy (EU mean)
V-Dem Score >= EU mean V-Dem Score <EU mean
Eligible 12/48 = 25% 25/48 = 52%

Not eligible 2/48 = 4% 9/48 = 19%
We tested two hypotheses related to commercial interests. The first is that the eco-
nomic opportunities created by integrating new states with strong regulatory
capacities into a common market area would motivate the community to accord
membership eligibility to all aspirants whose regulatory capacity is similar or
superior to that of the current member states. Given that any member state may
block consensus on membership eligibility, testing this hypothesis requires evi-
dence of the regulatory capacity of each aspirant state and that of the member
states in the same year. Unfortunately, the time scopes of the two standard sources
of such data—the World Bank’s Worldwide Governance Indicators5 and its Doing
Business Data6—are far too limited in historical scope to be useful here: the first
begins in 1996 and the second in 2003. We therefore rely on GDP per capita as a
proxy for regulatory capacity, using data in current US$ available from the World
Bank and OECD.7 The hypothesis thus leads us to expect that the EU will accord
membership eligibility to aspirant states whose GDP per capita (proxy for regula-
tory capacity) is equal to or higher than that of the worst performing member
state, and deny it to others (Table 4.10).
Another possibility (Hypothesis 10) posits the opposite relationship between
GDP and membership eligibility—namely, that the members of a regional com-
munity will see states with lower levels of economic development as attractive
markets for their goods and investment capital and thus welcome their eventual
5 World Bank, Worldwide Governance Indicators. http://info.worldbank.org/governance/wgi/#home

6 World Bank, Doing Business Data. http://www.doingbusiness.org/data
7 GDP data is unavailable for Greece 1959 and Turkey 1959 because both precede World Bank and
OECD coverage. In order not to lose observations, we coded both countries based on their 1960
scores. Given the large difference in terms of economic performance between the two cases and the
economically weakest EC member state in 1960 (Italy), we are confident in coding their performance
as inferior.
Table 4.10 Hypothesis 9: Well-regulated markets
Equal or higher GDP per capita Lower GDP per capita
Eligible 11/48 = 23% 26/48 = 54%

Not eligible 2/48 = 4% 9/48 = 19%
Table 4.11 Hypothesis 10: Non-competitive markets
Lower GDP per capita Equal or higher GDP per capita
Eligible 26/48 = 54% 11/48 = 23%

Not eligible 9/48 = 19% 2/48 = 4%
integration into the community’s common market area while rejecting wealthier
states whose own exports could threaten domestic producers and workers. This
hypothesis leads us to expect that the EU will accord membership eligibility to
aspirant states whose GDP per capita is lower than that of the worst performing
member state, and deny it to others, which we test using the aforementioned
World Bank and OECD data (Table 4.11).
Overall, these commercial or economic explanations are strikingly weak. The
hypothesis that the EU will favour more prosperous aspirants because their well-
regulated markets are more attractive for investment correlates with actual EU
decisions in only 42 per cent of the cases. The alternative hypothesis that the com-
munity will favour less prosperous aspirants because they offer more lucrative
investment opportunities and less risk of disruptive competition fares better at
58 per cent, but this still leaves a substantial share of cases unexplained. In sum,
commercial motives do not appear to have had a major influence on EU decision-
making on the membership eligibility of aspirant states.
Security interests
This explanation focuses on the likely implications of enlargement for the geopol
itical security of the community. Put simply, they suggest that a regional commu-
nity will welcome a non-member state whose accession would reduce member
states’ exposure to violent threats or improve their capability for resisting such
threats, and reject those whose accession would have the opposite effect. There
are at least two ways to conceptualize how the members of a regional community
understand the link between enlargement and security—one focused on the risk
of conflict and the other on alliance credibility—and each can be tested empiric
ally in several ways.
In the risk-of-conflict conceptualization, decisions on membership eligibility

are based upon the apparent probability of an applicant state’s ‘involvement in an
international militarized conflict . . . based on key elements of [its] internal char-
acteristics and its external security environment’ (Donno et al. 2015: 253, 255). By
this logic, a regional community will accord membership eligibility to non-
member states that pose a low risk of involvement in inter-state conflict, and deny
it to others. To test this possible determinant, we rely on Nordhaus et al. (2012)
global ranking of states’ risk of involvement in international conflict, whose time
boundaries exclude 10 of the 48 cases in our universe of EU decisions on mem-
bership eligibility and the case of Malta in 1995. We subject the remaining 37
cases to four distinct tests.
Our first test defines ‘low risk’ as states falling into the bottom 50 per cent of
the global ranking, so a low-risk state is one that in a given year is less at risk than
the median country of that year. As shown in Table 4.12, the results of this first
test support the hypothesis in just 49 per cent of the cases.
This finding could give rise to several objections. To start, being among the
50 per cent of states less at risk than the other half might not satisfy the security
demands of a regional community such as the EU. In other words, a country just
slightly less risky than the median country in a given year might still be perceived
as an unacceptable security risk, so the community might be more demanding on
conflict risk when evaluating the membership eligibility of aspirant states. In a
second test, presented in Table 4.13, we therefore redefine ‘very low risk’ as states
falling into the bottom 25 per cent of the yearly global ranking, which lowers the
hypothesis’ empirical fit to 22 per cent.
On the other hand, one might object that the global ranking is not the most
relevant benchmark for regional communities to assess potential members’ risk
of conflict. Instead, they might base their membership eligibility decisions on a
Table 4.12 Hypothesis 11, test 1: Low risk

states (global)
Low risk High risk
Eligible 14/37 = 38% 15/37 = 41%

Not eligible 4/37 = 11% 4/37 = 11%
Table 4.13 Hypothesis 11, test 2: Very low

risk states (global)
Very low risk Higher risk
Eligible 0/37 = 0% 29/37 = 78%

Not eligible 0/37 = 0% 8/37 = 22%
Table 4.14 Hypothesis 11, test 3: Low risk

states (European)
Low risk Higher risk
Eligible 12/37 = 32% 15/37 = 41%

Not eligible 3/37 = 8% 7/37 = 19%
Table 4.15 Hypothesis 11, test 4: Very low

risk states (European)
Very low risk Higher risk
Eligible 2/37 = 5% 27/37 = 73%

Not eligible 0/37 = 0% 8/37 = 22%
comparison to other states in their geographic region. In other words, when com-
pared to other states in one’s own geographic region, does a potential member
pose a low or high risk of conflict? To limit our analysis to states in the EU’s
region, we rely in the third and fourth tests (Tables 4.14 and 4.15) on the UN
Statistics Division’s measure of Europe and define low risk and very low risk as
states falling into the bottom 50 per cent and 25 per cent of the regional yearly
ranking. Neither of these tests indicates a strong relationship between apparent
conflict risk and membership eligibility.
A second version of the security interests explanation posits that a regional
community’s decisions on an applicant state’s eligibility for membership are
linked to the members’ interest in reinforcing the credibility of a security alliance
to which they also belong. By this logic, a regional community will accord mem-
bership eligibility to aspirant states that are or soon will be members of the secur
ity alliance to which most of the community’s members already belong, and deny
it to others. In the case of the EU, this concerns membership in NATO, the secur
ity alliance to which most members of the community also belong.
In particular, we consider the impact of two concentric indicators of closeness
to NATO: states that are full members of the alliance, and states with a clear path
to achieving NATO membership. The first test measures the fit between NATO
membership and EU membership eligibility at the time of the latter’s decision.
The results are quite weak: as shown in Table 4.16, only 38 per cent of the cases fit
the expectation that NATO members would be granted EU membership eligibility
while non-members would be considered ineligible.
It is also possible that the EU would grant eligibility to applicant states that
have a clear prospect of NATO membership in addition to those that already
belong. The best indicators of such a prospect are formal recognition as a future
Table 4.16 Hypothesis 12, test 1: NATO membership
NATO member Not NATO member
Eligible 9/48 = 19% 27/48 = 56%

Not eligible 3/48 = 6% 9/48 = 19%
Table 4.17 Hypothesis 12, test 2: NATO membership or

prospective membership
Current or prospective Not current or prospective

NATO member NATO member
Eligible 13/48 = 27% 24/48 = 50%

Not eligible 3/48 = 6% 8/48 = 16%
member by the NATO Council or after April 1999 by signature of a NATO

Membership Action Plan (MAP). According to NATO, agreement on a MAP
‘does not guarantee membership, but is a key preparation mechanism’.8 (In con-
trast, NATO’s Partnership for Peace (PfP) is not designed as a stepping-stone to
NATO accession and does not strengthen the alliance’s defensive capabilities, so
PfP participation does not indicate a membership prospect.) As shown in
Table 4.17, the results here are also rather weak: only 44 per cent of the cases fit
the hypothesized relationship.
Overall, despite multiple empirical tests, security interests seem to offer a weak
explanation of EU decisions on membership eligibility. The two tests of a possible
alliance credibility motive only fit that hypothesis in 38 per cent and 44 per cent
of the cases, while the two tests of conflict risk only fit the hypothesis in 22 per
cent and 49 per cent of the cases compared to all states, and 27 per cent and 51
per cent of the cases compared to European states depending on the risk thresh-
old.9 Interestingly, on conflict risk, there is almost no difference between the tests
using global and European benchmarks but lowering the risk threshold from 50
per cent to 25 per cent does produce a major reduction in explanatory fit in both
global and European tests. In fact, the very weak correlations (22 per cent global
and 27 per cent European) with the lower risk threshold seem to indicate that the
EU is biased against very low- risk states when deciding on membership
eligibility—the opposite of the hypothesis!
8 NATO Enlargement & Open Door. North Atlantic Treaty Organization, Fact Sheet, July 2016.
9 To ensure that these low correlations are not an artefact of our 25 per cent threshold for very-low
risk, we re-analysed the data with a 33 per cent threshold and found that this has only a minor effect
on the explanatory fit, which rises from 22 per cent (global comparison) and 27 per cent (Europe
comparison) to just 26 per cent and 29 per cent.
It is certainly plausible that the community might try to use accession as a

means to stabilize risky states in its neighbourhood, but this conclusion is belied
by the fact that all European states during the Cold War (plus states in south-
eastern Europe after the Cold War) were exposed to a relatively high risk of conflict,
so there were almost no very low-risk states seeking accession. It is nonetheless
apparent that aspirant states whose conflict risk makes them ‘the most costly to
integrate’ (Donno et al. 2015: 253) are frequently deemed eligible for integration.
In the end, it appears that EU decision-making on membership eligibility since
1957 has not been shaped very strongly by geo-political pressures.
To facilitate comparison, Table 4.18 summarizes the correlations found in the
preceding cross-tabulation analyses of various possible determinants of EU deci-
sions on the eligibility of non-member states seeking accession and the outcomes
observed in all identifiable cases of such decisions since the birth of the commu-
nity in 1957. Figures in the fit columns are the sum of cases where the hypothesis
correctly predicted a positive decision and cases where it correctly predicted a
negative decision. To facilitate comparison of the results, the column on the far
right presents the figures only for the pre-2006 cases.
4. Discussion of cross-tabulations
The preceding 16 tests of various explanations of membership eligibility revealed

considerable variation in fit between the empirical expectations and the actual
positive and negative outcomes in the 48 cases. Explanations focused on mem-
bership norms and geographic location correlate with the outcomes in 76–77 per
cent of the cases, while the treaty rules hypothesis correlates with outcomes in 69
per cent of the cases. These strong findings are suggestive of a significant causal or
constitutive effect on EU decisions.
In contrast, well-established explanations of EU decision-making exhibit a far
weaker fit with the pattern of EU decisions on membership eligibility: territorial
contiguity (52 per cent), regime similarity (44 per cent, 60 per cent), commercial
interests (42 per cent, 58 per cent), conflict risk (49 per cent, 22 per cent, 51 per
cent, 27 per cent), and alliance strength (38 per cent, 44 per cent). These poor
correlations—similar to and sometimes far worse than the results of a coin toss—
suggest that these factors had little causal or constitutive effect across the universe
of cases. And as is evident in the marginal difference between the two fit columns
in Table 4.18, limiting the analysis to pre-2006 cases, when all the hypotheses can
be tested equally, does not significantly affect their predictive or apparent explana
tory power.10
10 These findings relate only to the pattern of decisions on membership eligibility. The weak impact
of regime type and commercial and security interests on eligibility decisions reveals nothing about
Table 4.18 Summary of cross-tabulation analyses
Explanation Hypothesis Variable(s) Fit (full scope) Fit (pre-2006)
Membership 1. A regional community will decide that an aspirant state is Membership norms 32/42 = 76% 32/42 = 76%
norms eligible for membership if it fits the community’s prevailing 1957–1961, 1962–1969,
membership norm and will reject states that do not, regardless 1970–2005, (2006–2015)
of member states’ other motivations.
Geography 4. A regional community will decide that an aspirant state is ‘Europe’—Encyclopædia 37/48 = 77% 33/42 = 79%
eligible for membership if its territory is located physically Britannica
within the geographic space conventionally associated with the
region in question, and reject others.

Geography 5. A regional community will decide that an aspirant state is ‘Europe’—United Nations 37/48 = 77% 32/42 = 76%
eligible for membership if its territory is located physically
within the geographic space that other, authoritative
international institutions associate with the region in question,
and reject others.
Geography 6. A regional community will decide that an aspirant state is Territorial contiguity 25/48 = 52% 23/42 = 55%
eligible for membership if its territory is contiguous to that of a
current member state, and reject others.
Treaty rules 7. A regional community will decide that an aspirant state is Treaty rules -1957–1986, 33/48 = 69% 29/42 = 69%
eligible for membership if it fits the substantive rule on 1987–2015 33/48 = 69% 29/42 = 69%
membership established by the community’s core treaty, and Treaty rules—1957–1992,
reject others. 1993–2015
Regime type 8. A regional community will decide that an aspirant state is V-Dem ‘liberal’ aspirant vs. 21/48 = 44% 18/42 = 43%
eligible for membership if its regime type is similar to that of mean member state 29/48 = 60% 25/42 = 60%
the member states, and reject others. V-Dem ‘liberal’ aspirant vs.
least ‘liberal’ member state
Continued
Table 4.18 Continued
Explanation Hypothesis Variable(s) Fit (full scope) Fit (pre-2006)

Commercial 9. A regional community will decide that an aspirant state is Aspirant’s GDP per capita 20/48 = 42% 17/42 = 40%
interests eligible for membership if its level of economic development >= GDP per capita of the
is similar or superior to that of member states, and reject poorest member state
others.
Commercial 10. A regional community will decide that an aspirant Aspirant’s GDP per capita < 28/48 = 58% 25/42 = 60%
interests state is eligible for membership if its level of economic member states’ GDP per
development is below that of member states, and reject others. capita
Security 11. A regional community will decide that an aspirant 50% least risky, global 18/37 = 49% 18/37 = 49%
interests state is eligible for membership if it does not pose a significant 25% least risky, global 8/37 = 22% 8/37 = 22%
risk of involvement in inter-state conflict or cross-border 50% least risky, Europe 19/37 = 51% 19/37 = 51%
violence, and reject others. 25% least risky, Europe 10/37 = 27% 10/37 = 27%
Security 12. A regional community will decide that an aspirant NATO member 18/48 = 38% 15/42 = 36%
interests state is eligible for membership if it belongs to (or has a clear NATO member or candidate 21/48 = 44% 17/42 = 41%
prospect for joining) the security alliance to which most
community members belong, and reject others.
As shown above, the cross- tab analysis reveals striking support for the
hypothesis that EU decisions on membership eligibility are shaped by member-
ship norms prevailing amongst the community’s member states and supranational
institutions. In fact, the correlation for membership norms (76 per cent) is signifi-
cantly higher than that found for other, more traditional explanations of EU
decision-making focused on regime type, treaty rules, commercial incentives,
and security interests. This suggests strongly that the EU’s decisions on its geo-
graphic limits as a political community cannot be understood without serious
attention to the evolution and effects of membership norms in the period
1957–2005 and to their breakdown in the post-2006 period.
The scope of the membership norms argument is clearly limited by its inability
to explain decisions made when the community’s member states and supranational
institutions do not converge on a shared definition of eligibility. For example, EU
member states seem to agree in the post-2005 period that all states seeking mem-
bership must be ‘European’ but they did not agree on what criteria are sufficient
for membership eligibility. As a result, membership norms cannot explain the
eight decisions on eligibility that the EU has made since 2005. However, this limi-
tation is consistent with the theory’s explicit scope condition—namely, that deci-
sions on membership eligibility will only be shaped by membership norms when
such norms exist, meaning when member states share an understanding of what
types of states qualify for membership. When this condition is not satisfied, as in
the EU since 2006, the theory expects that eligibility decisions will reflect hard
bargaining among member states.
The analysis also shows strong support for geographic explanations. Although
the territorial contiguity hypothesis has very limited explanatory power (52 per
cent fit), the two tests of aspirant states’ location within ‘Europe’ (both 77 per cent
fit) suggest that understandings of geography do matter. This is not surprising,
given that the European Economic Community was created explicitly to ensure
peace and freedom in Europe and the Treaty of Rome invited all ‘European states’
(and implicitly, only them) to apply for membership. In that context, the commu-
nity’s 1987 rejection of Morocco’s quest for accession was rather predictable, since
it lies clearly beyond the conventional geographic limits of Europe.11
However, looking beyond the statistical tests, it is clear that conceiving of
‘European’ simply as a reference to physical location can be seriously misleading.
For example, the EU never questioned the eligibility of Cyprus, whose territory is
far closer to Southwest Asia than to the European mainland, but it did in earlier
decades declare Spain, Greece, and the United Kingdom ineligible. Or, consider
their possible influence on subsequent decision-making regarding whether a particular candidate

state is ready for accession or the conditions that the community may set for its accession: these deci-
sions occur at a different stage in the pre-accession process and have different consequences, so they
could well be shaped by different considerations.
11 W. Europe Bloc Bars Morocco as a Member, Los Angeles Times, 21 July 1987. See also Flory 1986.
the Urals and the Bosporus, which are often cited as the eastern frontiers of
Europe. The size and proportion of Kazakhstan’s territory that lies west of the
Ural Mountains and Ural River is far larger than the size or proportion of Turkish
territory that lies north of the Bosporus, yet Turkey is recognized as a candidate
for membership while Kazakhstan is covered by the EU’s Regional Strategy for
Central Asia and never mentioned as a potential member.
Conceiving of regional geography simply in terms of physical location thus
cannot provide a satisfactory explanation of a state’s eligibility to join a regional
organization. In short, the expectation that the EU would consider all European
states eligible for membership has very limited explanatory power unless one
includes variables that cannot be reduced to physical location. As the European
Commission conceded in 1992, ‘The term “European” combines geographical,
historical, and cultural elements which all contribute to European identity [and]
is subject to review by each succeeding generation.’12 But if political actors in
Europe hold similar conceptions of the physical limits of their region at particular
moments in time, and make membership decisions accordingly, their actions are
being shaped by shared ideas and not physical location.
Finally, the cross-tab analysis also shows fairly high predictive power for the
hypothesis that decisions on membership eligibility depend on the aspirant state’s
conformity with membership rules in the community’s core treaty. In fact, the
results (69 per cent) do not change depending on whether the change in EU treaty
rules is dated to 1986 or 1992. Although slightly weaker than the results for
physical location, this finding is also not surprising, given that the community
was created by a treaty and the members states know that its future depends on
the credibility of their treaty commitments. As such, it would have been surpris-
ing if this hypothesis had exhibited a lower correlation with actual outcomes.
In sum, membership norms, treaty rules and an aspirant state’s physical
location all seem to play a significant role in community decision-making on
membership eligibility. If we only had cross-tab results, it might be difficult to
separate the effects of these three factors. For example, while treaty rules and
membership norms changed over time, both consistently limited membership to
‘European’ states and none of the various membership norms over time ever
favoured accepting a state that did not comply with the community’s treaty rules.
We therefore re-assessed the explanatory power of membership norms using
two alternative empirical methods. Both robustness checks yielded positive
results, as shown below in Parts V and VI. First, logistic regression analysis shows
that our findings on membership norms hold even when we control for the
impact of the other potential explanatory factors. Second, crisp set Qualitative
Comparative Analysis (csQCA) shows that membership norms remain important
even when we adopt a conjunctural (rather than correlational) logic of causation

open to the possibility of independent multivariate causal pathways to the same
outcome. The fact that both of these checks confirmed the importance of
membership norms for explaining decisions on EU membership eligibility makes
us confident with regard to the internal validity of the claims we have made.
5. Logistic regression analysis
As a first check on our findings from the cross-tab analysis, we ran a range of
logistic regression models predicting EU membership eligibility. We did so
despite two reasons for being sceptical about the utility of this method: first, the
number of observations (48) is lower than normally recommended for statistical
regression; and second, the assumption of independent observations is violated
within our data—some countries appear multiple times and the individual out-
comes on these cases are probably related to one another, while decisions on some
countries might well have been influenced by other cases. However, since the 48
decisions on EU eligibility in our dataset constitute the whole population of cases,
and the clustering of observations is limited (i.e. most countries appear only
once), we concluded that checking our previous findings by means of logistic
regression analysis is an appropriate way to increase the confidence in our earlier
conclusions.
To that end, Tables 4.19 and 4.20 report a range of (multivariate) logistic
regression models. The models confirm our earlier findings: membership norms
turn out to be a meaningful predictor of EU membership eligibility. In Table 4.19
we regress decisions on EU eligibility in bivariate models on the same set of inde-
pendent variables that were used in the previous sections. We include one factor
for each of the theoretical arguments developed above, and used the one with the
greatest explanatory power in the main analysis, when multiple operationaliza-
tions were available. We see that conformity with the membership norms prevail-
ing within the Union (Model 1), is the only factor that significantly influences
decisions on EU eligibility in a bivariate setting. The lack of evidence for any of
the remaining, more traditional explanations of EU membership eligibility—even
in a simple bivariate setting—is striking.
The great advantage of regression analysis over the cross-tabulation method
lies in the possibility to estimate the effects of variables while controlling for the
influence of other factors. To employ this strength, Table 4.20 displays three
multivariate models that particularly aim to disentangle the effects of member-
ship norms and regime similarity, given that regime similarity is also reflected in
conformity with the prevailing membership norms. They unequivocally show
that applicants that meet the normative requirements held by the EU member
Table 4.19 Predicting EU membership eligibility in bivariate settings
Dependent variable:
EU eligibility (1 = eligible)
(1) (2) (3) (4) (5) (6)
Membership norms 1.73**

(1= norms are met) (.81)
Regional location 1.39
(1= within Europe according to (1.47)
Encyclopædia Britannica)
Treaty rules .62
(1= treaty rules with change in 1986 (.83)
are met)
Regime similarity 1.44
(1= country scores higher than min. (.87)
V-Dem score in EU in given year)
Market regulation .42
(1= country has well-regulated market) (.89)
Conflict risk .16
(1= Country scores lower than mean (.82)
risk in EU in given year)
Intercept .18 −0.00 .85 .76* 1.19*** 1.22**
(.61) (1.41) (.69) (.46) (.43) (.51)
Observations 42 42 42 42 42 37
Log Likelihood −19.5 −21.4 −21.6 −20.3 −21.7 −19.3
AIC 43.0 46.8 47.1 44.5 47.4 42.6
Note: * p < 0.1; ** p < 0.05; *** p < 0.01. Entries are logit coefficients; standard errors in parentheses.
states have a higher likelihood of being granted membership eligibility, even

when controlling for other factors.13
In Model 7, which includes all theoretically relevant variables except for the
regime similarity variable, we estimate a significant effect of the membership
norms variable (p <.05), while no other coefficient reaches acceptable levels of
significance. In Model 8 we observe a similar finding with regard to the effect of
regime similarity (p <.10). However, when including both of these variables in the
same model, only the effect of membership norms remains significant (p <.10)
and in the expected direction. In summary, these regression analyses provide
strong support for the findings of our cross-tab analyses: conformity with the
membership norms held within the EU play an important role in decisions on EU
membership eligibility—even when controlling for other theoretical explanations.
13 The treaty rules variable has been omitted from the multivariate tests, since it is too highly cor-
related with both membership norms and regime similarity and renders the estimation of models
impossible.
Table 4.20 Predicting EU membership eligibility in multivariate settings
Dependent variable:
EU eligibility (1 = eligible)
(7) (8) (9)
Membership norms 3.13** 2.63*

(1= norms are met) (1.36) (1.49)
Regional location .73 .19 .22
(1= within Europe according to Encyclopædia (1.62) (1.72) (1.74)
Britannica)
Regime similarity 2.35* 1.10
(1= country scores higher than min. V-Dem score (1.43) (1.79)
in EU in given year)
Market regulation −.65 .16 −.73
(1= country has well-regulated market) (1.40) (1.21) (1.45)
Conflict risk −.92 −1.34 −1.32
(1= Country scores lower than mean risk in EU (1.29) (1.39) (1.56)
in given year)
Intercept −.78 .59 −.29
(1.62) (1.67) (1.72)
Observations 37 37 37
Log Likelihood −14.95 6.89 −14.76
AIC 39.89 43.78 41.51
Note: * p < 0.1; ** p < 0.05; *** p < 0.01. Entries are logit coefficients; standard errors in parentheses.
6. Qualitative Comparative Analysis
Qualitative Comparative Analysis (QCA) relies upon a conjunctural understand-

ing of causation whereby multiple configurations of variables (‘conditions’ in
QCA parlance) may lead to a given outcome, so it is a useful way to check for
causal patterns that could be missed by cross-tab correlations or logistic regres-
sion analysis (Schneider and Wagemann 2012). Although we have not formulated
our theoretical expectations in terms of causal heterogeneity, we have recognized
multiple plausible explanations of membership eligibility and we can imagine a
number of plausible configurations leading to membership (in)eligibility.
Moreover, our dataset is very well suited to the requirements of QCA: it includes
a rather small number of observations (n = 48), the independent and dependent
variables are dichotomized, and each observation is placed in a mutually exclu-
sive subset of cases. For instance, cases belong either to the group of countries
that were deemed eligible or to the group of those that were not; they either lie
within Europe or not, have a competitive market or not, and are at a high risk of
inter-state conflict or not. This permits us to use QCA inductively to re-check the
robustness of our initial finding that EU membership norms play an important
role in the community’s decision-making.
To do so, we start by excluding all post-2005 cases from the dataset, as we have
missing information for a range of variables for these cases. To capture the condi-
tions that potentially form part of the path leading to membership eligibility, we
include one factor for each of the theoretical arguments developed above. When
multiple operationalizations are available for the same test, we used the one with the
greatest explanatory power in the main analysis. We then constructed a truth table
displaying the configurations of conditions and outcomes observed in our data
(Table 4.21). For this, we removed all rows of configurations of conditions and out-
comes that were not empirically observed at least once. In addition, as we have a
large number of observations (by QCA standards), quite precise and strong theor
etical expectations, and a high confidence in our coding of the data, we set the con-
sistency cut-off value to be larger than .75. This means that rows in our truth table
for which more than 75 per cent of the cases contained exhibit the expected out-
come are identified as sufficient configurations for membership eligibility.
Finally, we use logical minimization to compare the various truth table rows
with the sufficient conditions. As shown in Table 4.22, this exercise identifies four
parsimonious configurations of conditions that lead to EU eligibility. The consist-
ency of each of the individual configurations is high (82–89 per cent), meaning
that almost all cases sharing a particular configuration of conditions also share
the expected outcome. The consistency and coverage of the solution as a whole
are also high: 90 per cent of the cases covered exhibit the expected outcome, with
82 per cent of the cases being covered.
Importantly, membership norms feature prominently in the solution, appear-
ing in three of the four identified configurations: meeting the membership norms
and not being among the bottom half of the European risk ranking; meeting
the membership norms and not having a well-regulated market; and meeting the
norms and not being more democratic than the least democratic member state all
lead to EU membership eligibility. Interestingly, the additional conditions identi-
fied enter the configurations in their logically negated form. While this is in line
with one of our theoretical arguments concerning applicant states’ level of market
regulation, the same is not true with regard to the conflict risk, or regime similar-
ity argument.
This combination of evidence in favour of the membership norms argument,
and against two more traditional explanations of EU decision-making, bolsters
our claim that membership norms need to be taken into account as potential
explanations of EU membership eligibility. The one configuration of conditions
leading to membership eligibility that does not include membership norms does
not invalidate this claim, although the actual configuration—being located on the
European continent, not meeting the formal treaty requirements, and not being
among the bottom half of the European risk ranking—is indeed puzzling. The
QCA analysis thus bolsters our conclusion that membership norms need to be
taken seriously in explanations of EU membership eligibility.
Table 4.21 QCA truth table
Membership Part of Europe Fit with Regime similarity Well-regulated Low conflict risk Number of cases Consistency
norms (Encyclopædia treaty rules (comparison to market (GDP per (Country scores per (proportion of
Britannica) (1986 least democratic capita is higher lower than configuration cases with

change) member state) than EU minimum median European eligibility =1)
in given year) ranking)
1 0 0 0 1 1 1 1
1 1 1 0 1 0 1 1
1 1 1 1 1 0 2 1
1 1 1 1 0 1 5 1
1 1 1 1 0 0 5 1
0 1 0 0 0 0 7 0.857143
1 1 1 0 0 0 9 0.777778
1 1 1 1 1 1 8 0.75
0 1 0 0 0 1 1 0
0 0 0 0 0 0 1 0
0 1 1 0 0 0 2 0
Table 4.22 Analysis of causal configurations leading to EU membership eligibility
Country meets membership Country meets membership norms AND Country meets Country is part of
norms AND is not among the left does not have a well-regulated market membership norms AND Europe AND does not
half of European risk ranking is not more democratic fit with treaty rules
than least democratic AND is not among the
member state left half of European
risk ranking
Raw 0.45 0.52 0.27 0.18

coverage
Unique 0.06 0.15 0.03 0.18
coverage
Consistency 0.88 0.89 0.82 0.85
Cases Greece 1959, Greece 1959, Turkey 1959, Spain 1960, Greece 1959, Turkey 1959, Romania 1992, Slovak
covered by Turkey 1959, Spain 1960, Ireland 1962, Portugal 1978, Spain 1960, Spain 1977, Republic 1993, Slovak
solution Denmark 1961, Czechoslovakia 1990, Hungary 1990, Bulgaria 1991, Cyprus 1993, Republic 1999, Ukraine
Greece 1976, Spain 1977, Poland 1990, Bulgaria 1991, Czech Malta 1995, Slovak 1999, Macedonia FYR
Czechoslovakia 1990, Hungary Republic 1993, Estonia 1994, Latvia 1994, Republic 1997, Croatia 2000, Albania 2003,
1990, Bulgaria 1991, Estonia 1994, Lithuania 1994, Malta 1995, Slovenia 1995, 2000, Serbia and Bosnia and Herzegovina
Latvia 1994, Lithuania 1994, Slovak Republic 1997, Croatia 2000, Serbia Montenegro 2005, Ukraine 2005
Malta 1995, Slovak Republic and Montenegro 2005, Ukraine 2005 2005
1997, Croatia 2000, Serbia and
Montenegro 2005, Ukraine 2005
Consistency cutoff: 0.77
Solution consistency: 0.9
Solution coverage: 0.82
Uncovered cases: AUT 1992, FIN 1992, GRE 1967, MOL 1997, MOR 1987, NOR 1962, ESP 1962, SWE 1992, UK 1961, 1963, 1967, 1970
Cases in bold are inconsistent: they exhibit the configuration of conditions but do not share the expected outcome.
7. Conclusions
In this chapter, we used three distinct empirical methods—cross-tab correlations,

logistic regression analysis, and qualitative comparative analysis—to assess the
impact of EU membership norms on the community’s decisions regarding the
eligibility of states seeking membership. All three methods support the conclu-
sion that membership norms are a powerful (though not all-determining) factor
in EU decision-making. This finding is particularly notable given the widespread
tendency to assume that EU decision-making is dominated by economic and/or
security concerns, the novelty of the membership norms argument, and the effort
made here to distinguish these norms conceptually and empirically from treaty
rules and various indicators of state interests. Moving beyond statistics, Chapters
5–8 provide rich evidence of membership norm dynamics from detailed process
tracing of EU decision-making on the eligibility of Greece, Spain, Turkey, and
Ukraine over seven decades.
PART THREE
M E MBE R SHIP PRO CE S SE S

5
Membership Eligibility in a Europe of
Non-Communist States, 1957–1961
The Common Market is not only an economic instrument. It also has a

political mission . . . Let’s draw Greece and Turkey toward the Common
Market to consolidate their engagements in the Western world.
—Heinrich von Brentano, German foreign minister, 19601
The ink was barely dry on the 1957 Treaty of Rome when the new European
Economic Community received its first enquiries from other states, including
Greece, Turkey, Spain, and others, seeking membership in the near or more dis
tant future. The original member states were pleased with this apparent af fi rm
ation of their new community, but that alone did not answer the question of how
they should reply to the membership ambitions of such a diverse group of neigh
bours. In addition, this was hardly an easy time for the Six to contemplate expand
ing their new club: the EEC was just launched and had yet to implement its plans
to complement the Common Market with a Common Agricultural Policy; the
unabashedly-eurosceptic Charles de Gaulle had just returned to power amidst
political turmoil in France; and a Cold War was raging across the continent. How
then did the EEC respond to states seeking membership in the years immediately
following the Treaty of Rome?
As shown in Chapter 3, the new community was neither as open in its early
years as the 1957 Treaty of Rome seemed to suggest, nor as limited as the
democracy-focused discourse of later decades would lead us to believe. Instead,
there was an informal understanding among the governments and representa
tives of the Six that the Community would be open to all of the continent’s non-
Communist states and closed to Communist states. Other membership criteria
that had once been salient in the immediate postwar era disappeared from official
EEC discussions, even behind closed doors.
This chapter demonstrates that these membership norms exerted a powerful
effect on enlargement decision-making from the earliest days of the EEC. It thus
provides a first set of tests of the book’s argument and establishes an empirical
1 Documents diplomatiques français 1960. Tome 1: 1er janvier-30 juin. Paris : Ministère des Affaires
Étrangeres, Commission de Publication des Documents Diplomatiques Français, 1995, document 35 :
‘Consultation des Ministres des Affaires Étrangères des Six pays membres des Communautés
Européennes, Rome, 25–26 janvier 1960.’
baseline against which later cases can be compared. Sections 1, 2, and 3 respectively,
trace official decision-making regarding Greece, Turkey, and Spain’s efforts to join
the process of European integration in the first half-decade after the Treaty of
Rome, highlighting the various ways in which these membership norms shaped
the language that political actors used and the policy outcomes that emerged
from their deliberations and confrontations. Finally, Section 4 compares the
explanatory power of the book’s alternative hypotheses in all three cases and con
siders the implications of these cases for subsequent analyses.
1. Opening to Greece
Greece’s pursuit of EEC membership began on 8 June 1959, when the government
submitted a formal request for association with the community, making it the
first country to do so after the Treaty of Rome’s entry into force. As justification
for its initiative, the government cited Greece’s commercial, economic, spiritual,
social, and political ties to Europe, as well as its ‘geographic position at the
periphery of the defence of free Europe’. Just as important, the letter made clear
that Athens intended this association to lead over time ‘to the full participation of
Greece in the Treaty of Rome’—that is, to full membership. The Greek application
thus engaged two articles of the Treaty of Rome: Article 237 on membership and
Article 238 on association. As shown below, the timing and conditions of the
community’s openness to Greece were the subject of detailed negotiations but the
country’s fundamental eligibility for membership was never questioned. Instead,
as one community diplomat recalled, it was considered ‘more or less natural’.2
The first sign of the EEC’s openness was the Council of Ministers’ decision to
authorize the Commission to undertake exploratory talks with the Greek
government. When the latter restated its ultimate desire for full membership
during the opening of these talks, the Commission delegation was not alarmed:
association would provide a critical step toward this goal, it replied.3 An ECSC
High Authority report drafted shortly thereafter indicates that association would
transform the Greek economy and thus facilitate full EEC membership ‘in the
long term’.4 The first page of a Commission internal document, apparently also
drafted in September, stated without qualification that ‘The association of
Greece . . . will take the form of a customs union in the GATT sense whose final
objective will be the complete membership of Greece in the Community.’5
2 CVCE: Interview with Charles Rutten: The Community’s First Association Agreements, 29
November 2006.
3 AHCE: BAC/1/1971/27: Compte rendu des conversations exploratoires, 10–12 septembre 1959.
4 AHCE: CEAB/5/792: Note pour Mr. le Président du Groupe de Travail Relations Extérieures, 21
septembre 1959.
5 AHCE: BAC/1/1971/27: Notes sur quelques aspects fondamentaux de l’association de la Grèce à
la Communauté, undated.
A EUROPE OF NON-COMMUNIST STATES, 1957–1961 121
The EEC’s unquestioning attitude toward Greece’s eligibility for full membership
is particularly puzzling because Greek accession would impose significant
economic costs on the Six. Greece’s per capita income was well below that of Italy
(the poorest of the Six) and less than half that of France. A high proportion of the
Greek labour force worked in agriculture, many of them relying on outdated
techniques and inefficient landholding patterns but nonetheless producing crops
that would compete directly with Italian and French producers. As such, allowing
Greece to join the community would require a high tolerance in the short–
medium term for imports that undercut domestic producers, and over time, con
siderable investments to bring the country closer to the level of development
enjoyed by the Six. With or without such investments, Greece would be a major
competitor for funds from the community’s soon-to-be launched common agri
cultural policy.6 In the meantime, Greek consumption of exports from the Six
would be marginal. Yet when considering Greece’s eligibility for membership, the
Six repeatedly subordinated these commercial considerations.
Instead, decision-making on Greece’s application seems to have been shaped
from the very beginning by the prevailing EEC norm that all non-Communist
European states were eligible for membership. In a decisive discussion of Athens’
request in a July 1959 meeting of the EEC’s Committee of Permanent Representatives
(typically know as ‘Coreper’), all six delegations agreed that Greece’s position at
the periphery of ‘free Europe’, its membership in NATO, and its ‘internal political
situation’ (i.e. the strength of the local Communist Party) must take priority over
questions of the country’s readiness for the EEC or the likely implications of
Greek association for the economies of the Six.7 This attitude was not limited to
the governments of the Six: a European Parliamentary Assembly committee
report from this period cites Greece and Turkey’s membership in the Organisation
for European Economic Cooperation (OEEC) as evidence that both countries are
‘traditionally tied to us’ and thus endorses their pursuit of EEC association.8
Over the following year and a half, the norm that the EEC was a club of non-
Communist states was a constant point of reference in EEC deliberations on
Greece, and the Commission used it repeatedly to gain concessions on the
negotiation mandate from member states otherwise anxious to protect their
commercial interests. For example, as the Six prepared their negotiation mandate
in February–March 1960, the Commission commented that the result was ‘more
an accumulation of the restrictive positions of different delegations than a
reasonable basis for the opening of negotiations’ and warned the Six that a failure
of the negotiations would make it harder for the Greek government to resist
6 AEI: The European Economic Community and Greece. Newsletter on the Common Agricultural
Policy, no. 36, European Commission Directorate General for Agriculture, June 1965.
7 AHCE: BAC/1/1971/27: Extrait du procès-verbal, Coreper, 16 juillet 1959.
8 AHCE: CEAB/5/792: various documents.
Soviet economic overtures. In response, the Six conceded that the mandate could
be reconsidered in light of the progress of the negotiations.9
Keeping the pressure on, Hallstein described Greece and Turkey as ‘frontline
posts of the free world [that] merit all our sympathy and our cooperation: one
cannot judge the situation of these countries and our relations with them only
from a commercial perspective’ and highlighted Europe’s ‘responsibility to the
economic stability and cohesion of the Atlantic Community [which applies] not
only to our Community of the Six but equally to the other countries of Europe’.10
Two months later, the Commission informed the Council that its negotiations
with Athens would fail without a loosening of the commercial and financial terms
of the mandate. When the Six failed to reach agreement on further concessions,
the Commission reminded them that a failure of the talks could have ‘grave
consequences’ for Greece’s place among the ‘Western nations’.11
In response to this repeated warning, the Council then approved a radical step:
as long as the Commission consulted closely and regularly with the governments
of the member states, it was authorized to continue negotiations without a spe
cific mandate.12 And when these consultations subsequently failed to yield the
concessions that the Commission deemed necessary, it did not hesitate to remind
the Six of the ‘major political interest’ represented by Greece’s quest for association:
‘As the general situation of the free world . . . worsens in the Far East, Africa, and
Cuba, it does not seem that we can allow a defeat in Europe.’13 In late September,
the Six agreed to proceed in accordance with NATO’s call for increased economic
and financial aid to Greece and Turkey.14
The following spring, some of the member states remained unhappy with the
Commission’s prominent role in the negotiations with Greece, and this controversy
threatened to derail the talks. But even in the face of such obstacles, Greece’s
European vocation (the French term routinely used to indicate both eligibility
and destiny to join the community) was never questioned by the Six, and their
thinking on this matter was shaped significantly by Greece’s identity as a non-
Communist state. For example, Belgian Foreign Minister Paul-Henri Spaak wrote
in May 1961, ‘For several years, within NATO, Greece has attracted our attention
because of the Communist countries’ growing share of its external trade. Greece
is the object of constant pressure and tempting offers by the USSR. It is essential
to allow it to resist; its adhesion to the Common Market is therefore an essential
political act.’15
9 AHCE: CEAB/5/792: Extrait du procès-verbal, Conseil de Ministres, 10 mars 1960.

10 La Communauté Économique Européenne—Son Rôle, Ses Responsabilités, Son Rayonnement
dans le Monde. L’Opinion en 24 heures, 18 mars 1960. Reprinted in Hallstein 1979, 185.
11 AHCE: CAEB/5/792: Extrait du procès-verbal, Conseil de Ministres, 10–12 mai 1960.
12 AHCE: CAEB/5/792: Extrait du procès-verbal, Conseil de Ministres, 10–12 mai 1960.
13 AHCE: CAEB/5/792: Communication de la Commission au Conseil, 12 juillet 1960.
14 AHCE: BAC 26/1969/262: Note de la Délégation Néerlandaise, 23 septembre 1960.
15 AHEU: Fond Spaak PHS-638: Lettre à Premier Ministre Amintore Fanfani, 11 May 1961.
Greece’s eligibility for EEC membership was formally confirmed on 9 July 1961,
when representatives of the Greek government and the EEC signed the ‘Athens
Agreement’—an association treaty providing for a range of commercial and
financial measures designed to achieve full customs union between Greece and
the EEC within 12 years. Just as importantly, the treaty formally acknowledged
that the purpose of the association was to prepare Greece for full membership
of the EEC. Its preamble declared, ‘the support given by the [EEC] to the efforts of
the Greek people to improve their standard of living will facilitate the Accession
of Greece to the Community at a later date’, while Article 72 specified, ‘As soon as
the operation of this Agreement has advanced far enough to justify envisaging
full acceptance by Greece of the obligations arising out of the Treaty establishing
the [EEC], the Contracting parties shall examine the possibility of the Accession
of Greece to the Community.’
The treaty also established that a joint commission of EEC and Greek
government officials would supervise the progress of the association. At the ini
tiative of the European Parliamentary Assembly, and against the wishes of the six
member states, this official commission was soon supplemented by a mixed par
liamentary commission including members of the EPA and the Greek parliament.
Alongside the treaty, the parties also agreed a financial protocol providing for a
European Investment Bank loan to Greece of 125 million dollars. The significance
of both the joint parliamentary commission and the EIB loan would not be fully
apparent until four years later.
2. Opening to Turkey
The government of Prime Minister Adnan Menderes was quick to recognize that
Turkey would be affected by economic integration in western Europe and it
wasted no time in exploring how Turkey might be associated with the new
community. What makes this case so interesting in retrospect is that neither the
poor state of the Turkish economy, nor the Muslim faith of most Turks, nor a
brutal military takeover in 1960–61 prevented the Six from welcoming Ankara’s
aspirations. In fact, the content and motivation of the community’s ultimate
position was remarkably similar to that of its nearly simultaneous decision-
making on Greece.
Menderes’ government first tested the Six’s receptivity to an association
agreement in the late winter of 1959 by submitting a formal aide-memoire
alleging EEC discrimination against Turkish exports and arguing that Turkey
should receive privileged access to the Common Market. A subsequent European
Commission analysis described the stabilization of Turkey as a ‘clear interest for
the Community’, but observed pointedly that the ‘bilateral association of Turkey
to the Community offers no economic advantage’. In fact, notwithstanding the
Menderes government’s success in spurring growth through economic liberalization,

the Commission highlighted Turkey’s high inflation, stagnating exports, and
over-dependence on agriculture. Given these facts, the Commission concluded,
an association agreement with Turkey ‘would bring little to the Community, at
least in economic terms, while making it responsible for the financial difficulties
of a country whose economy would not easily integrate into the Common
Market’.16 But Ankara did not give up.
In late July 1959, just seven weeks after the Greek request for EEC association,
the Turkish government formally requested ‘admission to the Community as an
Associate Member’.17 In separate letters to the president of the European
Commission and to the acting president of the Council of Ministers, Menderes
proposed that Turkey belongs geographically to Europe, that it is economically,
politically, and militarily inseparable from Europe, and that its destiny is
irrevocably tied to that of the Western world. The letters also state that exclusion
from the Common Market would weaken the Turkish economy, which could
have ‘depressing consequences for free Europe’. Subsequent commentary by
Turkish officials made it clear that Ankara was seeking an association agreement
that would prepare the country for full EEC membership (Çaliş 2004: 79). As
such, the Turkish application, like the Greek application just before it, had
implications for two distinct provisions in the Treaty of Rome—Article 238 on
association and also Article 237 on membership.
From an economic perspective, opening the door to Turkey did not make
obvious sense at the time. According to the European Commission, the Turkish
economy was very weak and not ready to join the EEC’s customs union, as was
then under discussion with the Greeks. Compared to Greece, whose development
lagged well behind that of the Six, Turkey was deemed to have a weaker financial
sector, a less coordinated pattern of industrialization, fewer competitive industries,
more anti-competitive state intervention in the economy, and a shortage of foreign
exchange. In addition, its population was three times larger than that of Greece,
which magnified all its problems. An association agreement designed to achieve
customs union with the EEC would thus pose a far greater challenge for Turkey
than it would for Greece, and may not be realistic, the Commission concluded.18
As shown below, the six member states all shared this dour assessment of the
Turkish economy.
On the other hand, there were strong political pressures in favour of a positive
response to Turkey’s aspirations. To start with, the fact that Turkey belonged to
the Organization for European Economic Cooperation, the Council of Europe,
16 AHCE: BAC/3/1978/285: Notes pour Mr. Seeliger, 6 mars 1959 & 10 mars 1959.
17 AEI: The Third General Report on the Activities of the Community (21 March 1959–15 May 1960),
pp. 245–6.
18 AHCE: BAC 3/1978/285: Note sur l’Association de la Turquie, 23 septembre 1959.
and the North Atlantic Treaty Organization brought it clearly within the scope of
the EEC’s prevailing norm that all non-Communist European states were eligible
for membership. One of the few positive notes in a separate ‘Turkey dossier’
prepared by the Commission in September 1959 was the fact that ‘Turkey has
clearly pronounced itself against the USSR and for the West’ and that it belonged
to the aforementioned organizations of ‘free Europe’.19 In addition, the Six were
aware that for geo-political reasons the USA was strongly supportive of Turkey’s
application to the EEC. At the time, says former Dutch diplomat Charles Rutten,
‘the most important political considerations were of a strategic nature. It had to
do with the Cold War.’ A memo from the German Foreign Office emphasized that
Turkey’s economic weakness could lead to political instability and thus perhaps to
closer ties to the Soviet Union, Rutten recalled.20 Despite its concerns about the
Turkish economy, the Commission issued a positive recommendation.21
The EEC Council of Ministers met on 11 September to discuss Turkey’s request.
Hallstein opened the discussion by referencing his recent arguments in favour of
opening the door to Greece, which he said also applied to this case despite the
‘more complex’ economic implications of Turkey’s potential association. Turkey
was, he argued, ‘a State connected to the Community by close political and
military relations that the Community would have an interest in reinforcing by
strengthened relations in the economic domain’. As such, Hallstein urged the Six to
adopt the same favourable position that it had adopted for Greece. France’s State
Secretary of Finances, Valéry Giscard d’Estaing, then expressed his government’s
support for opening negotiations with Turkey, explaining that the EEC ‘is not a
closed Community, preoccupied only with the material interests of its members,
but intends to be open to third countries that understand its deep significance
and declare themselves ready to accept its rules’. German representative Hilger
van Sherpenberg agreed, stressing the importance of managing the Greek and
Turkish applications in parallel. The Luxembourg, Belgian, and Italian ministers
also indicated their governments’ support for opening negotiations with Turkey.
In conclusion, ‘the Council, while recognising that the association of Turkey
poses a certain number of problems that will have to be resolved, decide[d]
unanimously to welcome favourably the request of the Turkish government’
and charged the Commission with opening negotiations over the form of the
association.22
Given Ankara’s explicit indication of its interest in eventual accession, the
community’s decision on 11 September 1959 to begin negotiations without
excluding the possibility of accession can thus be interpreted as its first
19 AHCE: BAC 3/1978/285: Dossier Turquie, 17 septembre 1959.

20 ‘Turkije en Europa,’ VPRO broadcast, 14 December 2004, transcript at: https://www.andereti
jden.nl/aflevering/442/Turkije-en-Europa.
21 AHCE: BAC 3/1978/285: Dossier Turquie, 17 septembre 1959.
22 HAEU: CM 2/1959–14: Procès-verbal, 23e Session du Conseil, 11 septembre 1959.
recognition of Turkey’s eligibility for membership. Subsequent discussions among

the member states over the form and speed of Turkey’s integration with the
community do not detract from the significance of their fundamental openness.
In addition, the content of the discussion on 11 September indicates that the
decision to recognize Turkey’s membership eligibility was motivated in large part
by a shared conception of the EEC as a community of non-Communist states.
In fact, subsequent discussions among the member states and within community
institutions provide further evidence of the community’s motivation. A November
1959 report by the European Parliamentary Assembly describes Turkey as
‘traditionally tied to us’ and embraces its pursuit of EEC association.23 When the
EEC foreign ministers met in late January 1960, Germany’s Heinrich Van Bretano
commented, ‘Let’s draw Greece and Turkey toward the Common Market to
consolidate their engagements in the Western world.’ France’s Maurice Couve de
Murville agreed, ‘The association of Greece and Turkey presents many economic
inconveniences and great political advantages. It provokes strong reactions
in French agricultural circles. But we have an interest in not refusing these two
countries because of their location in the East, their proximity to Russia, and
the solidarity that we must show them’ (Ministère des Affaires Étrangères 1995,
document 35).
Nonetheless, the Council of Ministers’ openness in principle to the Turkish
request for association did not yield easy agreement among the Six over what
form the Community’s relationship with Turkey should take. The exploratory
talks concluded in December 1959, but the way forward was not clear and the
Commission was busy launching its negotiations with Greece, so Turkey’s
application stalled for several months. In the meantime, the Commission seems to
have had second thoughts about its earlier recommendation. In March 1960, the
Commission recommended to the Council that the EEC offer Turkey an ‘accord
of cooperation and assistance’, rather than the association agreement it had
recommended for Greece. Although both agreements would culminate in a cus
toms union (after 12 years for Greece and 22 years for Turkey), the Commission
proposal would have denied Turkey’s request for an agreement explicitly oriented
toward eventual membership.24
When Coreper took up the Commission’s proposal on 22 March, Germany
conveyed Turkey’s concern that its negotiations were lagging behind those with
Greece. Belgium sympathized with the Commission’s proposal, but France,
Germany, and Italy preferred that the same terminology be used for Greece and
Turkey even if the content of the two agreements had to differ for economic
reasons. The Netherlands added that Greece and Turkey were both of great
23 AHCE: CEAB/5/792: Assemblée Parlementaire Européen, Rapport au nom de la commission

des affaires politiques, doc.68, Novembre 1959.
24 MAEB: 6641/1-t Turquie 1959–63: Note sur l’Association de la Turquie, 23 mars 1960.
importance for NATO, so negotiations with Ankara should not be delayed. With
agreement still elusive, the issue was deferred until the next meeting of Coreper. It
is nonetheless significant that both Italy and France, the two member states most
threatened by Turkish exports, and thus presumably the most motivated to block
an opening to Turkey, opposed the Commission’s proposal of 22 March.25 Unable
to reach agreement, Coreper referred the issue to a group of national experts who
proposed several weeks later that two association options be presented to the
Council of Ministers—a customs union analogous in time-span and orientation
to that being negotiated with Greece, and a customs union to be achieved
gradually over a long period of time.26
When the Council of Ministers finally discussed the issue in May, almost all
delegations began their interventions by stressing the ‘political importance’ of the
decision to be made. (As explained by Rutten above, this was understood as a
reference to the anti-Communist purpose of the EEC.) Germany and the Benelux
preferred to open negotiations with Turkey as fast as possible on an association
agreement that would be similar to the Greek model while taking into account
the economic and commercial differences between the two countries, while
France and Italy preferred a two-stage scheme to be implemented over 22 years.
In the end, the Council agreed that the Commission should begin negotiations
with the Turkish government on the form that an EEC–Turkey association
agreement might take, and that it should seek to conclude such an agreement as
soon as possible, but it did not specify the form of association or implementation
time-span the Commission should aim to achieve.27
Meanwhile, within Turkey, strong tensions between the ruling Democratic
Party and the opposition Republican People’s Party were producing massive street
demonstrations and destabilizing a political system unaccustomed to inter-party
competition (Harris 1970). Formal negotiations between the EEC and Turkey
were set to begin in June 1960, but after the military staged a coup d’état in Ankara
on 27 May, the new Turkish government informed the Commission that it was
not ready to proceed.
Military rule continued through October 1961, during which time the Turkish
parliament was suspended, a new Constitution institutionalizing the military’s
influence over politics was approved, and senior opposition politicians (including
the deposed prime minister Adnan Menderes) were executed. As expected by
both the membership norms hypothesis and the security threats hypothesis, these
violations of democratic and human rights principles are never mentioned in
internal EEC deliberations. Rather than express their ‘shock’ at the execution of
25 AHCE: CEAB 5/794: Procès-verbal, réunion Coreper, 22 mars 1960 and 7 avril 1960.
26 AHCE: CEAB 5/794: Aide-Mémoire sur la réunion du Groupe d’Experts, 26 avril 1960.
27 AHCE: CEAB 5/794: Procès-verbal, réunion Conseil, 10–12 mai 1960.
Menderes, Charles Rutten later recalled, the Six agreed to overlook it.28 A French
foreign ministry report from February 1961 discussed the status of the association
in great detail without mentioning the Turkish military’s seizure of power or
subsequent violations of human rights. Instead, the focus was on the implications
of the pending Greek association for the negotiations with Ankara (Ministère des
Affaires Étrangères 1997). In short, as long as Turkey remained allied with the
West, neither the non-democratic nature nor the repressive behaviour of its
regime was considered relevant to its eligibility for European integration.
Nor did any of the Six move to expel Turkey from the Council of Europe for its
blatant failure to respect human rights. Instead, the EEC Council of Ministers
resolved in September 1960 that the Community would coordinate its policies
toward Greece and Turkey with NATO’s call for economic assistance to these
countries, and agreed to inform NATO that the Community’s association talks
with Ankara had been ‘temporarily suspended on the demand of the Turkish
government’.29 When the new military- dominated government in Ankara
signalled the following month that it was ready to resume negotiations on EEC
association, the Six agreed.
When negotiations resumed in mid-October 1960, the Turkish government
tabled an agenda quite different from that of its pre-coup predecessor. Instead of a
22-year programme of assistance and gradual progress toward customs union,
which the previous government had accepted, the new government informed
the Commission that it preferred an accord on the Greek model, involving EEC
loans at market rates, radical reforms, and rapid progress toward customs union.
When Jean Rey, member of the Commission responsible for external relations,
presented this new Turkish agenda to the Council of Ministers, he highlighted
that notwithstanding these new ambitions, the country’s financial situation had
degraded significantly since the previous spring.30 Rather than be alarmed, the
Six were committed to doing whatever they could to quickly stabilize the Turkish
economy and thus anchor its place in the western alliance.31
And although Turkey’s new insistence on a Greek-style accord indicated that it
viewed customs union as a step toward full membership, this point did not spark
debate in the Council. Instead, Germany’s representative expressed concern that
instability in Turkey could create an opening for Soviet influence and called for
an acceleration of the association negotiations. Belgium’s representative asked
Rey about the extent of Turkey’s foreign trade with the EEC relative to that with
28 ‘Turkije en Europa’, VPRO broadcast, 14 December 2004, transcript at: https://www.andereti
jden.nl/aflevering/442/Turkije-en-Europa.
29 MAEB: 6641/1-t Turquie 1959–63 & AHCE: BAC 26 1969 262.
30 ACCUE: CM2/1960/19: Procès- verbal réunion Conseil 17–19 octobre 1960; AHCE:
CEAB/5/794: Association de la Turquie, 27 octobre 1960.
31 ACCUE: CM2/1960/19: Procès-verbal réunion Conseil 17–19 octobre 1960; ‘Turkije en Europa’,
VPRO broadcast, 14 December 2004: https://www.anderetijden.nl/aflevering/442/Turkije-en-Europa.
the Eastern bloc. No member state used the opportunity to express doubt about
moving forward with a military-dominated government or even to argue that the
Greek model was inappropriate for such a weak economy. Instead, the Council
instructed the Commission to continue talks with Ankara while seeking current
economic data on Turkey from the OEEC and IMF.32
Notwithstanding the Council’s October 1960 instructions to reconvene
negotiations, the Commission deferred doing so for months, apparently due to
doubts that were emerging within both the Commission and Coreper about the
wisdom of proceeding on the basis approved by the Council the previous year.33
According to a Council document, these doubts were based on the experience
gained in negotiations with Greece and on the political, economic, and financial
situation in Turkey.34 In March 1961, the Commission proposed that the Council
consider three options: (1) an accord providing for rapid progress toward customs
union, like that about to be signed with Greece, (2) an accord designed to achieve
customs union in two stages, with progress to the second stage dependent on
unanimous agreement that Turkey had fulfilled its obligations in the first stage, or
(3) an even more limited accord including a statement of common objectives, a
permanent consultative mechanism, and EEC financial aid to Turkey.35 When the
Council met on 2–3 May 1961, it deferred any decision on the Commission’s
recommendations on Turkey until after the expected completion of the Greek
association in early July.36
By the summer of 1961, the views of the six member states were well established.
France and Italy favoured a limited accord covering technical, financial, and
commercial assistance to stabilize the Turkish economy. Their representatives
frequently cited doubts about Turkey’s ability to implement the demands of a
customs union, but they were undoubtedly also motivated by fear of commercial
competition. In contrast, Germany favoured an association agreement for Turkey
resembling that offered to Greece but with provisions that conditioned progress
towards a customs union on the state of the Turkish economy. This position was
based on Bonn’s determination that a preferential trade agreement would violate
GATT rules, its view that it was ‘politically impossible’ to consider any solution
for Turkey that would amount to a refusal to offer Ankara an association on the
same level as Greece, and on the ‘particular strategic importance of Turkey’. The
three Benelux states were sympathetic to the French-Italian position but inclined
by ‘political imperatives’ to support the German position.37
32 ACCUE: CM2/1960/19: Procès-verbal réunion Conseil 17–19 octobre 1960.

33 AHCE: CEAB/5/916: Note: Accord d’Association avec la Turquie, 15 mars 1961.
34 AHCE: CEAB/5/916: Procès-verbal réunion ministériel Conseil, 2–3 mai 1961.
35 AHCE: CEAB/5/916: Note: Accord d’Association avec la Turquie, 15 mars 1961.
36 AHCE: CEAB/5/916: Procès-verbal réunion ministériel Conseil 2–3 mai 1961.
37 AHCE: CEAB/5/916: Problème de l’association de la Turquie, 20 juillet 1961.
The surprise construction of the Berlin Wall in August 1961 raised the salience
of these political considerations. An analysis written the following month by the
Belgian ministry of foreign affairs noted that the Turkish economy was extremely
weak and would require many years of structural reform and foreign assistance
before it could meet EEC requirements. However, it declared, ‘overwhelming
political factors argue for an accord as ambitious as possible. . . . It is therefore
necessary to find a formula that reconciles the political necessity of passing an
accord with the imperatives of an economic association that otherwise would
hardly be justifiable because of the fragility and weaknesses of the Turkish
economy.’38
At a Council meeting in late September, there was consensus on the German
representative’s view that ‘Turkey is an essential element in western defense and
that it constitutes a bridge with the Middle East and Africa [and that] these
elements must be taken into consideration as long as Turkey’s internal situation is
not stabilized.’ Speaking for the Commission, Jean Rey advised that ‘the general
political reasons that led the Community to envisage an association agreement
with Turkey have not changed and that, if certain recent internal political
developments cannot have a very favorable effect on the position that the
Community must take, these circumstances should not give the impression that
the Turkish negotiations could be called into question.’ None of the Six challenged
the description of military rule as a simple lack of internal stability, nor did any
suggest that it justified a suspension of negotiations. The only discord concerned
the technical form of Turkish association, which the Council resolved to separate
from its political commitment to pursue an association agreement.39
When a special ‘associations committee’ met in September to bridge the gaps
between the member states, the representatives of the Six agreed that Turkey
would be unable to fulfil the obligations accepted by the Greeks and expressed
tentative support for a new Dutch proposal that any agreement with Turkey should
divide its association into two phases—a preparatory phase in which Turkey
would have limited obligations but the Community would make concessions to
facilitate Turkey’s progress, and a second phase involving greater obligations on
both sides.40 However, a subsequent Coreper meeting revealed considerable
divergences regarding the content of the second phase, with Germany and the
Benelux favouring full customs union if certain criteria were met, France favour
ing an open-ended package of commercial, economic, and financial measures,
and Italy reserving its position.41
38 MAEB: 6641/1-t Turquie 1959–63: Association de la Turquie, 8 septembre 1961.

39 AHCE: CEAB/5/916: Procès-verbal du Conseil de la CEE, 25–27 septembre 1961.
40 AHCE: CEAB/5/916: Modalités possible d’une association de la Turquie, 14 septembre 1961.
41 AHCE: CEAB/5/916: Note pour M. Wehrer, 22 septembre 1961.
Yet another meeting of the associations committee failed to bridge the gap,
with France and Italy supporting a Commission proposal that the content of the
second phase could include customs union but should not be defined in advance,
while Germany and the Benelux insisted that the goal of customs union should be
formally retained.42 When the issue was discussed by the Council of Ministers in
mid-November, neither France nor Italy was swayed by the Benelux-German
argument that the deterioration of the Turkish economy made it politically
imperative to reinforce Turkey’s links to the West. Instead, the French and Italian
ministers expressed concern about competition with Turkish agricultural
products.43
As 1961 drew to a close, some close observers of the EEC deliberations (and
surely some in Turkey) began to wonder whether the Six really had the political
will to conclude an association agreement of any kind. A senior Commission
official conveyed to Jean Rey his impression that certain member states’ refusal
of ‘extremely modest’ concessions in Turkey’s favour indicated that they are
fundamentally opposed to any association but wish to avoid saying so openly.
This was different from the situation with Greece, he added, where none of the
member states was opposed.44 This impression is consistent with the reflection by
retired Dutch diplomat Charles Rutten, who recalled decades later (without
further explanation) that contemplating eventual full membership for Turkey was
‘not natural at all’ for the Six in this period but that ‘most considered it so far off
that it was not necessary to have a deep discussion of something that would
probably never happen’. In any case, Rutten added, there was ‘no discussion of the
problem of Islam or the influence of Islam in Europe . . . ’45
However, the evidence contradicting Rutten’s interpretation is far stronger. In
response to his colleague’s memo, Jean Rey insisted that even at this difficult stage
in the negotiations, the differences ‘relate more to tactics’ than to ‘a divergence of
views on the fundamentals’.46 In other words, he was saying, the Six agreed on the
ultimate goal but some were delaying formal agreement in order to influence the
terms of agreement and perhaps to secure some advantage for themselves. And it
was Rey, not the sceptical colleague, who had actually participated in the relevant
Council meetings.
Rey’s interpretation is also supported by the response of the Six to the two
memoranda that the Turkish government had submitted in late August 1961
restating its motivations for seeking closer ties to the EEC and its vision of what
any prospective accord should contain. Unlike previous Turkish memoranda,
42 AHCE: CEAB/5/916: Comite Associations Pays Tiers, 13 octobre 1961.

43 MAEB: 6641/1-t Turquie 1959–63: Extrait du Projet de Procès-Verbal de la réunion restreinte à
l’occasion de la 55e session du Conseil, 13 et 14 novembre 1961.
44 AHCE: BAC/3/1978/288: Note pour M. Rey, 17 novembre 1961.
45 CVCE: Interview with Charles Rutten (29 November 2006).
46 AHCE: BAC/3/1978/288: Note pour M. Seeliger, 27 novembre 1961.
these documents state repeatedly and without qualification that Ankara’s goal was
‘membership’ and ‘complete integration’ with the EEC.47 In fact, Ankara called on
the Six to ‘provide the Commission a mandate that would permit Turkey’s
membership in the EEC as fast as possible’.
If anything would have provoked fundamental opponents of Turkish
membership to show their cards, it was far-reaching statements such as these. Yet
when the associations committee met two weeks later, all six representatives
reaffirmed the ‘political interest’ that their governments attached to Turkey’s
association to the EEC. The only mention of the recent memoranda was a
statement that it would be difficult to realize the goals outlined in the memoranda
within the first five years of an association agreement.48 On balance, the evidence
suggests that the debate at this stage among the Six, and between them and
Turkey, concerned how and at what speed Turkey would progress toward EEC
membership, but not whether Turkey was eligible to join the community.
While commercial interests clearly played a role in EEC deliberations at this
point, they do not seem to have been decisive in the community’s willingness,
despite deep concerns about the Turkish economy’s suitability to the Common
Market, to open negotiations on the first steps toward Turkey’s eventual accession.
France and Italy were clearly less willing than Germany and the Benelux to
guarantee Turkey’s entry to the EEC’s customs union, which is likely explained by
the fact that Turkish exports were more likely to compete with French and Italian
products. On the other hand, key decision-makers repeatedly discounted the
Commission’s argument that the Turkish economy was ill suited to the competitive
pressures or administrative requirements of the Common Market, and in the end,
neither Rome nor Paris used its veto to ensure that talks with Turkey were limited
to a form of association that would be less threatening to their commercial
interests.
In contrast, this puzzling behaviour is consistent with three hypotheses.
Throughout the EEC’s deliberations, Turkey’s credentials as a European state were
often cited and never questioned, so the Six’s openness to Turkey could be
explained, in line with the formal rules hypothesis, by the Treaty of Rome’s
provision that ‘all European states may apply to join the community’. But while
Article 237 clearly establishes a minimum for eligibility, it does not stipulate
which states the community should accept.
Thus we turn to the membership norms hypothesis. At various stages of the
negotiations, including the interruption occasioned by the military takeover
in Ankara, potential economic and political obstacles to Turkish membership
were overcome by agreement on the importance of solidarity with Turkey as a
47 AHCE: CEAB/5/916: Note de transmission: Aide-mémoire du Gouvernement turc, 30 août

1961. (The French terms were ‘adhésion’ and ‘intégration complète’.)
48 AHCE: CEAB/5/916: Modalités possible d’une association de la Turquie, 14 septembre 1961.
non-Communist European state. Of course, if the member states believed that

anchoring Turkey in the Western military alliance required welcoming it into the
EEC, as seems to have been the case, then the same pattern could also be
explained by the security interests hypothesis.
3. Opening to Spain
Although Spain did not formally apply for EEC association until 1962, and is thus
not directly comparable to Greece or Turkey in that regard, the government in
Madrid made no secret after 1957 of its desire for closer ties to the new community.
It is therefore interesting to examine how EEC officials and the governments of
the Six responded to Madrid’s quest for acceptance in Europe during the years
immediately following the Treaty of Rome. Interestingly, as in the Turkish case,
the backwardness of the Spanish economy and the authoritarian character of its
political regime do not seem to have prevented the EEC from responding posi
tively to Madrid’s overtures in this period.
Despite the military defeat and political demise of its German and Italian allies
in 1945, the fascist regime of Francisco Franco, which had controlled Spain since
the 1930s, remained in control after the Second World War. By the late 1950s,
Spain’s ruling elites were divided over questions of reform, with economic liberals
around the Opus Dei movement favouring closer ties to European and international
institutions, and conservative veterans of the civil war period seeking to preserve
the status quo. The implications for Spain of the creation of the EEC were a central
point in these debates (Guirao 1997, 1999). Even though Franco’s government
had not shown any interest in political liberalization, both Germany and France
welcomed Madrid’s economic reform programme and actively supported Spain’s
effort to join the process of European integration.49
In 1956, with the support of Bonn and Paris, Spain applied to become an
associate member of the OEEC. A decision on this application was delayed for
two years by Belgian foreign minister Paul-Henri Spaak’s reluctance to welcome
Franco into any pan-European organization and by a general assessment that the
Spanish economy was unprepared for foreign competition.50 But with economic
reform continuing in Spain, Spaak eventually bowed to French and German
pressure, which allowed Spain to gain associate membership in the OEEC in 1958
and full membership the following year, when it also joined the International
Monetary Fund and the World Bank (Griffiths 1997).
49 ACE: PACE:COMO19006: La situation politique en Espagne (Rapporteur: Anne-Marie Renger),

Conseil de l’Europe, Assemblée Consultative, AS/NR(13)14, 13 janvier 1961.
50 FPHS: Documents 4628 & 4629: Entretien du Ministre des Affaires Étrangères avec l’Ambassadeur
d’Espagne, 12 janvier 1955, plus annexe.
Membership in these bodies was a first step away from the autarchic trade
practices that had drained Spain’s currency reserves and isolated its economy
from the rest of Europe, but the Spanish reformers’ real goal remained stronger
ties to the EEC. This became an urgent priority as London began to rethink its
sceptical position toward the EEC, leading the Spanish government to fear that its
agricultural exporters would lose out to their Italian competitors if the UK were
to join the common market.51 Spanish officials interested in moving closer to the
EEC were heartened by the encouragement they received from within the
community (Trouvé 2008: 47–8).
When asked in a June 1959 press conference about the possibility of Spain’s
accession to the EEC, Walter Hallstein replied ‘it’s up to the government of Spain
to decide this’ once Madrid’s accession to the OEEC was completed.52 In other
words, Hallstein did not foresee the EEC imposing any insurmountable obstacles
to Spanish accession. At a crucial meeting of the EEC’s Council of Ministers held
in early 1960 to discuss association and membership, French foreign minister
Maurice Couve de Murville introduced the possibility of association agreements
with a number of countries including Spain, although he conceded that its ‘entry
into the Common Market seems technically impossible’ (Ministère des Affaires
Étrangères 1995, document 35). Given the context and word choice, this concession
seems to indicate his judgement that the Spanish economy was not capable of
withstanding the tariff reductions and removal of other trade barriers required by
the Treaty of Rome—the same issue that had delayed Spain’s accession to the
OEEC until 1959. However, the fact that his positive mention of Spanish associ
ation was not contradicted by any of the other five foreign ministers is consistent
with the prevailing norm that the EEC was open to all non-Communist states
in Europe.
The principal critics of the new rapprochement between Franco’s regime and
the other states of western Europe were found in socialist circles. In February
1960, Germany’s Social Democratic Party condemned Bonn’s discussion of
military cooperation with Madrid. The Bureau of the Socialist International
(including Willi Birkelbach) protested the German government’s policy of
‘seeking sites for military depots, hospitals and administrative units in Spain
through agreement with the regime which came into being with the Nazi’s help’.
The language of the SDP’s condemnation suggests, however, that the socialists
were preparing for an eventual discussion of Spanish relations with the EEC:
‘Spain, it is true, is a country with a great European cultural tradition, but under a
fascist dictatorship. . . . We desire trade and communications with Spain, but our
political dealings with the present regime in Spain must be confined to correct
51 AHCE: BAC/3/1978/235: ‘L’Espagne et la CEE’, translated article from Nieuwe Rotterdamsche
Courant, 24 June 1961.
52 AEI: Interview of Three European Community President, 11 June 1959.
diplomatic dealings.’53 But such criticisms were rare and never voiced within
the institutions of the new community of Six, as one would expect given the
membership norm.
Another sign of official west European sympathy for Spain’s EEC ambitions
came in December 1960, in response to a colloquium on Spain’s integration into
the European economy being planned by a coalition of liberal, socialist, and
Euro-federalist groups. To symbolize the organizers’ critical attitude toward the
Franco regime, the event was to be held in a building in Strasbourg belonging to
the Council of Europe, the institution responsible for the European Convention on
Human Rights and Fundamental Freedoms. However, after Franco’s government
sent strong protests to the European Commission and the governments of the Six,
one of them intervened and blocked the use of the Council of Europe building.54
Throughout the winter and spring of 1961, Spanish officials were increasingly
outspoken about their government’s intentions to apply to the EEC. In a February
1961 letter to Commission president Walter Hallstein, Spain’s ambassador to the
EEC Comte de Casa Miranda spoke of ‘the desire of my government that Spain
participate in the European integration movement.’55 In June, Franco himself
spoke about the advantages of access to European markets.56 Later that summer,
presumably with the consent of the Six, Hallstein led a Commission delegation to
Madrid, where they held three meetings with members of Franco’s government.57
In September, Hallstein gave a speech that Spanish officials interpreted as an
endorsement of the French government’s view that European countries seeking
deeper ties to the EEC should apply for full membership, rather than association.
In fact, the Spanish embassy in Brussels highlighted Hallstein’s indication that the
Athens Treaty had demonstrated the EEC’s readiness to sign association agreements
with European states unable to fulfil all the requirements of membership.58 So by
the end of 1961 the government in Madrid has received multiple indications of
various sorts that Spain would be welcome in the EEC and no powerful messages
to the contrary.
In sum, the EEC’s openness to Spain in this period is consistent with several
but not all of the hypotheses considered in this volume. First, the EEC’s posture
clearly fits the expectations of the membership norm hypothesis: the anti-
Communist credentials of Franco’s government were beyond dispute. Similarly,
53 Socialist International Information (London) X:10, 5 March 1960 and X:12, 19 March 1960.
54 AHCE: BAC/3/1978/235.
55 WB: Letter to the Spanish Foreign Ministry from Conde de Casa Miranda, chief of Spain’s
Mission to the European Economic Community, Brussels, 21 September 1961.
56 AHCE: BAC/3/1978/235: Note à Monsieur le Président, C.E.E. Porte-Parole de la Commission,
Bruxelles, le 3 juillet 1961.
57 WB: Jean-Marie Kraft, ‘Le Groupe Socialiste au Parlement Européen,’ thèse, Université de
Strasbourg, 1962.
Mission to the European Economic Community, Brussels, September 21, 1961.
the EEC’s welcoming attitude toward Spain may have been driven by fear of a
Soviet attack on western Europe, as expected by the security interests hypothesis.
After all, Spain had allowed the USA to operate naval and air bases on its territory
since 1953 and cooperated with France on intelligence matters in North Africa.
By 1960, the German government was talking with Madrid about access to mili
tary bases, supply depots, and hospitals in Spain that could be used by NATO
forces in case of a Soviet attack.59 And finally, given that Spain is indisputably
located on the European continent, the clear minimum for membership eligibility
under the Treaty of Rome, the EEC’s openness to Madrid also fits the formal rules
hypothesis.
In contrast, it is difficult to explain the EEC’s openness in terms of regime type
or commercial interests. The EEC’s welcoming attitude toward Spain clearly
contradicts the regime type hypothesis that the Six would exclude authoritarian
states in order to reinforce their own, democratic constitutional order. It is
also inconsistent with apparent short- or medium-term commercial calculations,
since Spain’s principal exports in this period (grains, fruits, olive oil) would
compete directly with local production in France and Italy—the former being one
of Madrid’s strongest supporters in Europe.
4. Conclusions
From the outset of the European Economic Community, a norm that non-
Communist states were eligible for membership prevailed among the six original
member states and shaped their decision-making on the efforts of Greece, Spain,
and Turkey to associate with and eventually to join the community. As shown
above, member states followed the norm even when they had strong economic
interests in taking a different position. The effects of the non-Communist norm
thus explains policy choices that might appear puzzling from a contemporary
perspective, such as the community’s openness to these three states despite the
economic weakness of Greece and Turkey, military rule in Turkey in 1960–61,
and the persistence of fascist rule in Spain. Chapter 6 traces how the convergence
around a new membership norm in late 1961 affected subsequent decision-
making on the same three states through the 1960s, including military rule in
Greece starting in 1967.
59 The Allies: Room of One’s Own. Time Magazine, 7 March 1960.
6
Parliamentary Democracies, 1962–1969
The guaranteed existence of a democratic form of state, in the sense of a

free political order, is a condition for membership. . . . States whose
governments do not have democratic legitimacy and whose peoples do
not participate in the decisions of the government, neither directly nor
indirectly by freely-elected representatives, cannot expect to be admitted
in the circle of peoples who form the European Communities.
—‘Birkelbach Report’, European Parliamentary Assembly,
January 19621,2
Compared to the initial years after the Treaty of Rome, when the EEC signalled in
various ways that it would someday be willing to offer membership to three states
in its southern neighbourhood—Greece, Turkey, and Spain—the Community
then changed positions and closed the door to Spain and Greece while continuing
to treat Turkey as eligible for membership. Given how ready the new community
had been to welcome these applicants just a few years earlier, how can we explain
its sudden reversal on two of the three states? There was a change of regime in
Greece (following the 1967 military coup) but not in Spain and yet both were
rejected while Turkey’s eligibility status remained unchanged.
The key to this puzzle is found in the transformation of EEC membership
norms less than five years after the signing of the Treaty of Rome, which
fundamentally shifted the terms by which a state’s eligibility to join the community
were assessed. Starting in early 1962, the norm that all non-Communist European
states were eligible to join the Community was replaced by an understanding that
only parliamentary democracies were eligible. As shown in Chapter 3, this
normative change resulted from the intersection of political entrepreneurship by
members of the European Parliamentary Assembly able to leverage the domestic
1 AEI: Rapport fait au nom de la commission politique sur les aspects politiques et institutionnels
de l’adhesion ou de l’association a la Communaute, 16 janvier 1962.
2 ACCUE: CM2.848.1964.
and international legal commitments of the member states and opportunities

created by the French government’s unsuccessful effort to change the form of
European integration set by the treaty. The resulting membership norms
influenced decision-making on association and adhesion in the decade that
followed, producing outcomes that differed systematically from those of the
immediate post-Treaty of Rome period and that cannot be explained by treaty
rules or the member states’ commercial or strategic interests.
This chapter examines the impact of these new norms on EEC decisions
regarding the three states’ pursuit of membership in this period, showing how the
norms facilitated certain applicant states and impeded others. To start, the
community’s new norms were critical to the EEC’s 1962 decision, despite Paris
and Bonn’s strong support for Madrid, to reject Spain’s quest for association as a
first step to membership, and to the 1963 conclusion of an association agreement
with Ankara despite deep reservations in the Commission and the member states
regarding Turkey’s economic under-development. The new norms were also
critical to the 1967 decision to freeze Greece’s progress from association toward
full membership until democracy was restored in that country, despite the weak
legal basis for such a move and significant concerns about the possibility of a
Communist take-over in Athens.
1. Rejecting Spain
After years of preparation, Spain finally applied for EEC association on 9 February
1962. The letter that Spanish Foreign Minister Fernando M. Castiella to sent to
Maurice Couve de Murville, France’s foreign minister and EEC president-in-
office, used language very similar to that used by Greece and Turkey three years
earlier. In particular, he requested ‘the opening of negotiations to examine the
possibility of establishing an association to the Community capable of leading in
time to a complete integration after the completion of the necessary steps that will
permit the Spanish economy to align itself with the conditions of the Common
Market’. Castiella also cited Spain’s ‘European vocation’, its geographical position
and territorial contiguity with the Community, and its programme of economic
reform.3 As one historian noted, it was ‘less than an application for membership
but more than a simple application for association’ (Trouvé 2008: 72).
As implied by the terms of Castiella’s letter, Spain’s economy was not ready for
full EEC membership. One month before Madrid’s application, Belgium’s leading
3 CVCE: Letter from the Foreign Minister, Fernando Castiella, 9 February 1962.
A EUROPE OF PARLIAMENTARY DEMOCRACIES, 1962–1969 139
business daily had made a similar observation: ‘the engagement of Spain in the
European union does not appear desirable as soon as one considers its economic
structure’.4 However, given that both France and Germany had embraced Spain’s
desire to integrate itself into the new community, and the fact that the Six had not
treated military rule in Turkey in 1960–61 as an obstacle to that country’s association
or eventual membership, the Spanish government had good reasons to expect
that the EEC would respond positively to its request. Even the Italian government,
which was most threatened by Spanish exports, appears initially to have favoured
responding to Castiella’s letter with an offer to hold ‘exploratory conversations’,
the same process that had been used for Greece and Turkey.5
The unpredictable variable in this process was the EPA’s effort to transform the
community’s membership norms so as to exclude authoritarian states, including
the uncertain political impact of the Birkelbach Report that was the keystone of
this effort. In an early indication of what was to come, conservative French
newspaper Le Figaro, which supported Spain’s application for EEC association,
predicted that the Birkelbach Report could ‘complicate’ EEC decision-making,
but it refrained from criticizing the contents of the report, whose publication had
been so well received.6 Likewise, the chief of Spain’s mission to the Community
advised his superiors in Madrid that the Birkelbach Report was significant
precisely because the articles on association and membership in the Treaty of
Rome were ‘so flexible and fluid’ and indicated that it would matter ‘even though
the Parliamentary Assembly does not have to be consulted for the adhesion of a
new member’.7
The French and German governments were sympathetic to Spain’s request but
fully aware that it would be controversial. France’s ambassador in Madrid sent the
Quai d’Orsay a list of responses to opponents of the Spanish application. Since the
EEC was principally economic in character, he argued, there was ‘no reason why
the nature of the Franco regime should provide a valid pretext to exclude it’.
Furthermore, he suggested, there was little opposition to Franco’s rule in Spain,
where the majority of the citizens are ill suited to ‘the forms of our democracy’.8
For its part, following the pattern established with Greece and Turkey, the
German government’s preference was that the EEC should inform Madrid that
4 L’Echo de la Bourse, 13 January 1962, cited in ‘Réactions de la presse espagnole et internationale’,

Revue du Marché Commun 69 (mai 1964), 150.
5 CVCE: Note from the Italian Government on relations between Spain and the Common Market,
9 February 1962.
6 Débats Difficiles à Prévoir sur l’Entrée de l’Espagne. Le Figaro, 13 janvier 1962.
Mission to the European Economic Community, Brussels, 25 January 1962.
8 MAEF: EUROPE: Espagne (1961–1970) Dossier 329: Télégramme de l’Ambassadeur Margerie,
19 février 1962.
exploratory talks would begin in the second half of 1962.9 Spain’s foreign minister
later remarked publicly, ‘I do not believe that in all of Europe a government has
welcomed our integration request as has the German government. . . . Germany
also helps us, in words and in deeds, to achieve our objectives: we begin with
association with the Six; then we envisage a possible integration into the European
Community.’10
On the other hand, Spain’s application provoked an immediate negative
response from EPA members and trade unions. The EPA’s socialist group (also
headed by Birkelbach) adopted a motion on 19 February declaring that it was
‘categorically opposed to entertaining any request coming from a dictatorial
government. . . . The merger of economic interests has no meaning unless it
strengthens democracy. In other words, no European country whose government
lacks democratic credentials and which does not guarantee the basic freedoms
and rights of man can become a member of the Community or become associated
with it.’11 Although this wording was more categorical than that of the Birkelbach
Report (which had been drafted to satisfy various parties in the EPA), its reference
to ‘freedom’ and ‘rights’ clearly referred to that document and thus aimed to
disempower Spain’s supporters within the Council of Ministers.
The European Free Trade Union Confederation (CISL) used similar language
to urge the Council of Ministers and the European Commission to reject Spain’s
request: ‘We believe that the association of a country in which the forces that, in
our countries, have contributed to the creation of our Community, are pursued
and oppressed by a ruthless dictator, is something unimaginable and contrary to
the interests of our countries. If the free West wants to remain true to the
democratic tradition and if its governments wants to continue to count . . . on the
support of those forces represented by our union movement, a categorical refusal
must be given to a country that has the audacity to offend us by its advances.’12
Soon thereafter, the secretary of the European branch of the International
Confederation of Christian Trade Unions (CISC) declared that the organization
was ‘strongly opposed’ to all negotiations between the EEC and Spain regarding
association or membership. His letter explained, ‘The membership or even the
association of a non-democratic state to the Community appears to us contrary
to the conception of the latter and to the objectives that it pursues. It would be in
particular contrary to the principles affirmed in the Preamble of the Treaty.’ In
particular, the letter said, the government of Spain does not respect the rights
recognized in the Universal Declaration of Human Rights or the European
Convention on Human Rights and Fundamental Freedoms.13
9 PAAA: B-26/167: Spanischer Antrag auf Assoziierung an die EWG, 15 Februar 1962.
10 L’Espagne se félicite de l’ampleur et des formes de l’aide allemande, Le Monde, 28 mars 1962.
11 Socialist International Information XII:10, 10 March 1962, 133.
12 AHCE: BAC/26/1969/667. 13 ACCUE: CM2.1970.851.
The CISL and CISC protests attracted further media attention to the debate.14
Press in the Benelux was particularly critical of Spain’s application. An editorial in
an influential Dutch newspaper asked why ‘in the free world it is not understood
that Spain has no place there’.15 Belgian socialist daily Le Peuple commented,
‘From a political point of view, Spanish association would be inappropriate,
unacceptable.’16 A liberal member of the EPA argued in the Lëtzebuerger Journal
that Spain should be excluded from the EEC despite strong economic and strategic
reasons to the contrary: ‘Europe is not only oranges and tomatoes, coal and
steel, cars and furniture. To build it is not only to pursue the abolition of customs
barriers or the growth of cultural exchanges. It is also and above all to strive to
create a political community.’ As such, he argued, the EPA could not incorporate
members ‘selected by a dictatorship’. A copy of the article was forwarded to Paris
by the French ambassador in Luxembourg, along with reference to a socialist
paper that had reprinted the CISL’s letter to the Council of Ministers.17 A group of
prominent Spanish political exiles also petitioned the EPA to oppose any opening
of negotiations with Spain.18
The real question, however, was how Spain’s application would be received
within the EEC’s institutions, most importantly the Council of Ministers. When
Birkelbach addressed the EPA on February 20, he posed a formal question to the
EEC bodies with legal authority over adhesion and association: ‘The Spanish
government has recently asked to open negotiations with the European Economic
Community regarding the association and eventual membership of Spain in the
Common Market. Do the Council of Ministers and the Commission believe that
it is necessary to consider such a request coming from a regime whose political
philosophy and economic practices are in complete opposition to the conceptions
and structures of the European communities?’ Birkelbach’s question elicited great
attention in the Council, which had never before been faced with a direct oral
question from a member of the parliamentary assembly.19 The Commission had
more experience with parliamentary questions, but its directorate on external
relations remained unsure how to respond to Spain, and thus to Birkelbach.20
The Six were now confronted with unexpected rhetorical pressure from parlia-
mentary and trade union circles mobilized in opposition to Spain’s request. Just as
important, the Belgian government was divided between the free-market tenden-
cies of Prime Minister Van Zeeland and the federalist tendencies of Foreign
14 Viva opposizione nel MEC all’ingresso della Spagna, La Giustizia, 20 febbraio 1962.
15 Nieuwe Rotterdamse Courant, 10 februari 1962.
16 L’Espagne aux portes du Marché Commun, Le Monde 11–12 fevrier 1962.
17 MAEF: Europe: Espagne (1961–1970) file 329: Lettre de Mr. Guyon, 20 février 1962.
18 CVCE: Information document submitted by the Socialist Group in the European Parliamentary
Assembly (February 1962).
19 ACCUE: CM2.1970.851: Note d’Information: Question orale no.2, Conseils des Communautés
Européennes, Secrétariat Général, 26 février 1962.
20 AHCCE: BAC/3/1978/235: ‘Projet: Note pour le Ministre Rey’, 9 mars 1962.
Minister Paul-Henri Spaak, a long-standing critic of Franco’s regime. The q

uestion
was which tendency would prevail within the Belgian government and whether
the resulting policy would be sufficiently firm to resist the EEC’s heavyweights.
Just three years earlier, Belgium had acquiesced to French and German support
for Spain’s entry into the OEEC that was no less strong than their support for
Spain’s association with the new EEC.
The first opportunity for member states to deliberate formally on Spain was the
Coreper meeting on 20–21 February, but that proved inconclusive when the three
Benelux delegations insisted that the matter be referred to the Council of
Ministers’ session scheduled for 5–6 March.21 Lest there be any doubt, Germany’s
Vice Chancellor and Economy Minister Ludwig Erhard then announced publicly
that Bonn was ‘ready to be favorable’ to Spain’s request to join the EEC.22 Two
days later, Erhard reaffirmed his position in a private conversation with the
Belgian ambassador to Bonn.23 To confirm its support, the German government
delivered a loan of 200 million DM to the government in Madrid.24 In a private
meeting, though, Erhard alerted Spain’s finance minister to the possible difficulty
of mustering support for Spain’s request from all of the Six.25
During the two weeks between Birkelbach’s question and the Council of
Ministers meeting, socialist groups and trade unions increased their focus on the
Spanish issue. On 23 February, the Belgian Metalworkers Union expressed its
total opposition to Spanish association. Four days later, the Dutch Socialist Party
issued an appeal to the Commission and to the Six against Spanish adhesion or
association to the EEC. On 1 March, the leadership of the German Trade Union
Federation wrote to the chancellor: ‘The association request of Spain is not
acceptable as long as the fundamental democratic rights are not re-established in
this country and the free exercise of human rights . . . are not assured.’26 In
response to such pressure, Belgium’s Foreign Trade Office, which lobbied within
the Ministry of Foreign Affairs for a decision in favour of an association with
Spain, devoted nearly half of its position paper to rebutting the Birkelbach Report
and pressure from the trade unions, which it qualified as ‘passionate’ rather than
‘objective’.27 Notwithstanding such rebuttals, support for Madrid’s request was
becoming politically untenable.
21 ACCUE: CM2.1970.851: Extrait du Compte Rendu Sommaire de la Réunion Restreinte Tenue à
l’Occasion de la 205ème Réunion du Comité de Représentants Permanents à Bruxelles, le 20 et 21
février 1962.
22 L’Espagne Séduite à Son Tour par le Marché Commun, Les Dernières Nouvelles d’Alsace, 10
février 1962; Per la Spagna nel MEC ‘benevoli’ i tedeschi, Il Giorno (Milan), 22 febbraio 1962.
23 MAEB: 6641/1-t (Espagne 1962–1963): Lettre de Mr. Baert à Mr. Spaak, 23 février 1962.
24 L’Espagne se félicite de l’ampleur et des formes de l’aide allemande, Le Monde, 28 mars 1962.
25 PAAA: B-26/167: ‘Besuch des spanischen Finanzministers Navarro Rubio’, 26 Februar 1962.
26 Pas de Franco au Marché Commun, Le Peuple (Brussels), 27 février 1962; Contre l’adhésion de
l’Espagne à la CEE, Journal de Genève, 27 février 1962; Contre l’association de l’Espagne au Marché
Commun, Le Populaire de Paris, 1 mars 1962.
27 MAEB: 6641/1-t (Espagne 1962–1963): Note pour l’Administration, 19 mars 1962.
When the Council of Ministers convened on 5 March, France’s Couve de

Murville opened the discussion by proposing that the EEC reply to Spain as it had
replied to earlier association requests from Austria, Sweden, and Switzerland: this
would have meant indicating that the EEC had received Madrid’s request ‘with
satisfaction’ or ‘with interest’. Germany’s Rolf Lahr then offered a more moderate
proposal: given the political problems posed by Spain’s application, the Council’s
reply should refer to the importance of future relations between Spain and the
Community. Italy’s Antonio Segni proposed that the Council respond by asking
Madrid to specify exactly what sort of relationship it wished to establish. All three
proposals would have kept the door open to initiating negotiations with Spain at a
future date, perhaps when public and parliamentary attention had diminished.28
Unlike three years earlier, though, when Spaak had reluctantly conceded
Spanish entry into the OEEC, his views were now reinforced by a vigorous public
debate and mobilization of societal forces. He thus argued for responding to
Spain in the ‘most neutral terms possible’. In support of Spaak, the Netherlands’
Jan Willem De Pous expressed concern that Lahr’s text would raise Spain’s hopes
that the Council actually intended to pursue negotiations, and Luxembourg’s
Eugéne Schaus proposed a simple acknowledgement that the request had been
received. At this point, the French and German ministers backed down.
Apparently unwilling to pressure his Benelux colleagues on an issue where
France’s position had attracted so much negative public attention, Couve de
Murville responded only that Spain’s geographical and historic links to Europe
made it impossible to exclude it indefinitely from the Community. Lahr then
stressed that the response must be courteous and not too short. In the end, the
Council approved a non-committal accusé de reception and agreed to delay
further consideration of Spain’s request. In addition, the Council agreed that the
Coreper should coordinate with the Commission its response to Birkelbach’s
question.29 Shortly thereafter, the Council replied to Birkelbach that it had
acknowledged receipt of Spain’s request but had not yet discussed the problems
raised by the request and was thus not yet in a position to respond fully to the
question that he had posed.30
When the Commission’s Jean Rey appeared before the parliamentary assembly
on 29 March, Birkelbach declared that the EEC ‘would cease to be worthy of
confidence if it envisaged establishing with the Madrid regime a tight connection
in the form of an association or even a complete membership’. After insisting that
the Commission had not yet developed a position on Spain’s request, Rey
conceded that ‘[i]n the context of our discussions to define a policy of association,
28 ACCUE: CM2.1970.851: Procès-verbal de la réunion restreinte tenue à l’occasion de la 63ème

session du Conseil de la Communauté Économique Européenne, Bruxelles, 5–7 mars 1962.
29 ACCUE: CM2.1970.851: Note: Réponse à la question orale no.2, Secrétariat Général, Conseils
des Communautés Européennes, 12 mars 1962.
30 ACCUE: CM2.1970.851: Letter from Mr. Couve de Murville to Mr. Fuller, 26 mars 1962.
we have read and discussed the report by Mr. Birkelbach’. Speaking personally, he
added, ‘European politics is not made only of interests, but also of feelings and
ideas and . . . important movements of opinion like those that we learned about
through the large trades union organizations . . . constitute, it seems to me, one of
the elements that must be taken into consideration by all the responsible
European authorities . . .’31
The Council of Ministers’ formal response and the Commission’s refusal to
articulate a public position were seen in Madrid as an indication that the door
might still be open. The Spanish Ministry of Foreign Affairs thus set out to reverse
the delegitimation of its position by the Birkelbach Report and the EPA’s socialist
group. In a memorandum delivered to the German Foreign Ministry on 27
March, the Spanish ambassador asserted, ‘There do not exist any arguments
flowing from the text of the Treaties of Rome that pose legal obstacles to Spain’s
association to the EEC.’ He also highlighted the legislative power of the Cortes,
and argued that ‘all essential fundamental rights are guaranteed and protected’ in
Spain by constitutional law.32 The same day, the Spanish ambassador in Rome
delivered two messages to the Italian Foreign Ministry: the first argued that ‘no
legal clause appears in the Treaty of Rome . . . justifying the opposition; to the
contrary, the terms of the Treaty are clearly and explicitly favorable to the Spanish
proposal’, and the second claimed that the Spanish government ‘accepts all the
consequences and knows all the conditions that govern the question—including
those in political matters’. That said, he concluded, ‘The question of knowing in
what manner it will be proper to apply these principles to the politics of Spain is
an internal problem that will be resolved gradually as we approach the goal . . . but
it is fair that [Spain] be given in this respect a reasonable margin of confidence.’33
On 30 March, the Spanish consul in Strasbourg submitted to the Spanish
embassy in Paris an optimistic analysis of the recent debate in the parliamentary
assembly. As he saw it, members of the Christian Democratic and Liberal groups
had failed to applaud Birkelbach’s speech and pointed questions, but had
applauded Jean Rey’s reserved remarks, which he took to indicate openness to a
‘realistic’ presentation of Spain’s position. In addition, he wrote, ‘our friends’ in
the assembly report that ‘the socialists represent a political force with much more
power to attack than to defend’. In conclusion, he recommended an intensification
of Madrid’s efforts to convince the Commission and the governments of the Six.
31 Parlement Européen, Débats, 29 mars 1962. See also: Rey zum Assoziationsgesuch Spaniens,
Neue Zürcher Zeitung, 31 März 1962.
32 ACCUE: CM2.1970.851: Note: Démarches espagnols auprès des Gouvernements allemand et
italien. Le Conseil, Communauté Économique Européenne, Bruxelles, 9 mai 1962, Annexe I.
italien. Le Conseil, Communauté Économique Européenne, Bruxelles, 9 mai 1962, Annexe II & III.
At that point, he estimated, Spain could count on the support of the governments
of Germany and France.34
One week later, the Spanish ambassador to Rome wrote to the Italian Foreign
Ministry that Spain’s request to open negotiations with the EEC ‘is not a question
of this or that government; it simply concerns Spain as a nation’. He also linked
the socialist group’s opposition to views expressed in the Soviet newspaper Pravda
and accused the parliamentarians of ‘collaborating’ with ‘Communist’ forces that
‘have always opposed’ the idea of European integration.35 The Spanish foreign
ministry also delivered to the Council of Europe a detailed rebuttal of the earlier
Renger report, which had argued strongly that the Spanish regime was ineligible
to participate in European integration.36
The anti-Franco forces nonetheless sustained their rhetorical attack on the
possibility of Spanish association or accession. On 2 April, the chairman of the
Labor group in the Dutch Parliament published a commentary on Spain’s rela-
tions with the EEC: ‘Can anybody seriously believe that the Socialists of Europe
can or will co-operate with these mortal enemies of freedom? The admission to
the EEC of a Spain ruled by a Fascist dictator would be intolerable. . . . We cannot
allow Franco, whose power rests on terror and bloodshed, this man who was the
third of the fascist tyrants who held Europe in subjection, to pose as a believer in
the ideals of peace and freedom as professed by Western Europeans.’37 When the
Political Committee of the Council of Europe’s Consultative Assembly issued its
recommendations on EEC enlargement and association policy in May 1962, the
report recognized the economic and geo-political arguments for integrating
Spain into European institutions; but echoing Birkelbach’s argument, it concluded
that the same political considerations that would exclude Spain from the Council
of Europe would also exclude it from membership in the EEC. Instead, the report
concluded, ‘however impossible may be any close form of political co-operation,
some form of association between Spain and the EEC is desirable’.38 But to
emphasize how Spain’s lack of democracy differentiated it from the other eight
applicant states, the committee’s final recommendation relegated it to a separate
document urging ‘the Members of the EEC to examine the possibility of some
form of economic agreement between Spain and the EEC, bearing in mind the
constitutional changes that will be necessary before any form of political associ
ation can be contemplated’.39
34 WB: Letter from Spain’s Consul in Strasbourg to Spain’s Ambassador to France, 30 March 1962.
italien. Le Conseil, Communauté Économique Européenne, Bruxelles, 9 mai 1962, Annexe V.
36 CVCE: Comments on the ‘Renger Report’ and the Situation in Spain (April 1962).
37 CVCE: Démocratie et désintégration: L‘Espagne membre de la CEE? Courier Socialiste Européen
(2 avril 1962).
38 AHPE: PACEDOC3191: Report on the general policy of the Council of Europe. Consultative
Assembly of the Council of Europe, 9 May 1962, Doc.1420, pp. 11–14.
39 ACE: Council of Europe, Parliamentary Assembly, Recommendation 314 (17 mai 1962).
Meeting in Munich on 7–8 June, over a hundred representatives of Spanish

opposition groups from within the country and abroad declared, ‘the integration
of any country with Europe, whether in the form of full membership or of
association, requires democratic institutions on the part of that country’. Their
resolution then stipulated a long list of political conditions that Spain must fulfil,
including the establishment of authentically representative and democratic
institutions, effective guarantees for all personal human rights, the exercise of
trade union freedom on a democratic basis, the freedom to organize social
movements and political parties, and respect for the right of opposition.40 Although
the EEC had never formally adopted these as conditions for membership, they
clearly echoed Council of Europe norms and the language of the Birkelbach Report.
Two weeks later, the European Parliament’s socialist group endorsed the Munich
declaration and reaffirmed its ‘complete opposition to any and all participation of
Franco’s Spain in the institutions of free Europe’. In a cable to Bonn, the German
embassy in Madrid admitted, to its apparent dismay, that the Munich congress had
increased the political sensitivity of Spain’s quest for EEC association.41
The European federalists’ rhetorical offensive stretching from the Birkelbach
Report to the Munich congress had thus made it impossible for the Spanish
government and its supporters in Paris, Bonn, and elsewhere to control the terms
of debate. On 16 June, a congress of European liberal politicians resolved that
only democratic states were eligible to join the community, followed three days
later by an identical statement from a congress of Christian Democratic parties
(Trouvé 2008: 82). The principle that the EEC should only consider membership
for parliamentary democracies had acquired full normative status in official
community discussions by this point. An editorial in one of Belgium’s leading
newspapers prefaced its endorsement of Spain’s application with the legally
dubious assertion that ‘the Treaty of Rome, which is the Constitution of the future
Europe, excludes non-democratic states’. Having conceded this point, it could
then only argue that perhaps the Spanish people had the government they
deserved or wanted, and in any case that the Spanish government’s application to
the EEC demonstrated its intention to democratize.42
The fate of Spain’s application had thus already been settled by the time the
Commission decided in October 1962 that association and membership
negotiations with all other countries would be put on hold until those with the
United Kingdom were complete. In short, Madrid’s first overture to the EEC was
rejected because European parliamentarians, federalist militants, and sympathetic
trade unions had transformed the prevailing normative definition of the EEC as a
40 Keesing’s Contemporary Archives, Vol. 13, 1961–1962: ‘September 1962 – Spain’. London:
Keesing’s Ltd.
41 PAAA: B26/162: Die Bundesrepublik und Spanien, 23 Juni 1962.
42 Faire l’Europe avec ou sans les socialistes, La Libre Belgique, 2 mai 1962.
political community, replacing anti-Communism with a focus on parliamentary

democracy, and then insisted that this precluded any opening to Franco’s Spain.
In February 1964, two years after its initial overture to the EEC, the Spanish
government reiterated its interest in opening negotiations with the Community.
This time, though, Madrid emphasized practical measures rather than the longer-
term goal of full integration, effectively delinking its overture from any aspiration
to eventual membership and thus from the community’s new membership norms.
Shortly thereafter, the Council of Ministers agreed to propose to Madrid
‘conversations whose goal would be to examine the economic problems that the
development of the EEC poses for Spain and to search for appropriate solutions’.43
These talks were still ongoing in 1969 when the Commission recommended to
the Council that the Community offer preferential trade agreements to less
developed Mediterranean states like Spain. In so doing, the Commission noted
that only ‘countries that enjoy institutions and regimes comparable to those of the
founding States’ are eligible for association under the Treaty of Rome’s Art. 238,
which had already been defined as a first step to full membership for European
states.44 The significance of this latter move should not be under-estimated: the
Commission was in effect using the community’s new membership norm to
interpret a pre- existing treaty provision that actually said nothing about
regime type.
2. Confirming the opening to Turkey
As discussed in Chapter 5, the Turkish military’s brutal rule after its seizure of
power in 1960 did not cause the EEC to question the association negotiations it
had begun earlier that year with the democratically elected government in
Ankara. By early 1962, though, the membership norms prevailing in the EEC had
changed fundamentally, causing the community to reject Spain’s quest for an
association agreement because of the nature of its political regime. Had the
Turkish military not allowed elections in October 1961 and then returned to the
barracks, it seems probable that the EEC would soon have sent the same message
to Ankara: no negotiations until the restoration of parliamentary democracy. But
by the time the EEC’s new membership norm was established in early 1962,
Turkey was back under democratic rule.
Turkey’s military rulers, who had suspended negotiations with the EEC in early
1961, requested a return to talks in early 1962. When the Council of Ministers
met in May 1962 to discuss the terms of renewed talks, all six member states were
43 ACCUE: CM2.1970:852, includes Spain’s letter (14 February 1964) and the Council’s reply
(2 June 1964).
44 EUDC: 441.212 EFTA states 1959–77: Réponse à la question écrite n. 410/70.
reluctant to offer a rapid path to customs union that they believed Turkey was
incapable of accomplishing. Beyond that consensus, a rift was apparent: on one
side were France and Italy, both concerned about competition with their own
agricultural sectors and thus inclined to favour talks focused on financial aid or a
limited commercial accord; on the other side was Germany, which favoured
speedy progress toward a comprehensive association accord in order to ensure
political stability in a critical NATO member state.45
The first sign of a breakthrough came in July of that year, several months after
negotiations resumed, when the Italian ambassador to Turkey signalled that
Rome’s foot-dragging on Turkish association was tactical and would soon be
lifted: ‘We would not thwart a political operation for hazelnuts,’ he told his Belgian
counterpart in Ankara.46 The negotiations were interrupted in October of that
year by the Commission’s decision to focus on the British application, but were
then helped along by a Franco-German deal: Paris would drop its objections to
the association with Turkey if Bonn would drop its objections to a proposed
association agreement with the African states and Madagascar, which was finally
signed in Yaoundé on 20 July 1963.47
Finally, on 12 September 1963, an ‘Agreement Creating an Association between
the Republic of Turkey and the European Economic Community’ was signed in
Ankara. The agreement accorded Turkey the same formal status as that which
had been accorded two years earlier to Greece, but its economic content was quite
different: a preparatory phase (5–9 years) and transition phase (12 years) would
precede the gradual implementation of a customs union in industrial and
agricultural products, freedom of movement and establishment for workers,
freedom of movement for services and the application of the Community’s rules
on competition. Nonetheless, the accord anticipated the possibility of Turkey’s
accession to full membership in the Community over the long term. Copying
verbatim from the Athens agreement, the preamble to the Ankara Agreement
declared, ‘the support given by the [EEC] to the efforts of the Turkish people to
improve their standard of living will facilitate the Accession of Greece to the
Community at a later date’, while Article 28 specified, ‘As soon as the operation of
this Agreement has advanced far enough to justify envisaging full acceptance by
Turkey of the obligations arising out of the Treaty establishing the [EEC], the
Contracting parties shall examine the possibility of the Accession of Turkey to the
Community.’
EEC officials went out of their way to confirm the country’s European vocation.
During his speech at the signing ceremony of the Ankara Agreement, Commission
45 ACCUE: CM2.1962:24: Réunion du Conseil des Ministres, 14–15 mai 1962.
46 MAEB: 6641/1-t Turquie 1959–63: Lettre de Mr. Fenaux, 6 juillet 1962.
47 CVCE: The Case of Turkey (8 July 2016).
President Walter Hallstein repeated twice ‘Turkey is part of Europe.’48 Also at the
ceremony, when Dutch Foreign Minister Joseph Luns, who signed the agreement
as President of the EEC Council, was asked if Turkey was a European country
geographically, he answered, ‘Yes, here is Asia, but political and economic
agreements transcend and even redraw geographical frontiers’ (Çaliș 2004: 93).
Two months later, in a speech to the European Parliament, Luns said again that
the Ankara Agreement had confirmed Turkey’s European character and its des-
tiny to accede to full EEC membership at a later date.49
At the same session, other members of the European Parliament stressed that
democratic rule was key to eligibility for Community membership. Belgian
socialist August De Block commented, ‘The new [Ankara] treaty is, like the
others, about the economy and social questions. Everybody knows and admits
this new Europe must be a democratic Europe. We almost never speak about it. In
my opinion, this is a mistake.’ The democracy norm is critical, De Block
concluded, because a government can raise economic living standards without
protecting fundamental liberties.50
Finally, Belgian socialist MEP Fernand Dehousse highlighted the fact that
some states seeking closer ties to the EEC satisfied community norms while
others were radically inconsistent with those norms. Given the detail of his
observations on the countries examined here, they are worth quoting at length:
‘Another problem emerges every time we conclude association agreements, that
of the democratic character of the countries with whom we have such relations. I
know that this problem will be raised regarding Turkey, as it was regarding
Greece, and I would like to give very sincerely my personal sentiment. I make a
clear and absolute distinction between the case of Greece and Turkey on the one
hand, and possibly those of Spain or Portugal on the other. These latter countries
are based upon an antidemocratic public law. The principle of democracy is not
only rejected by the constitution of these countries or by what occurs, it is
formally condemned there. The current legal philosophy of Franco’s Spain and
Salazar’s Portugal is directed against the democratic spirit and the parliamentary
system. There’s nothing of the same in Greece and Turkey. The constitutions of
these two countries, which have subscribed to the European Convention on
Human Rights of 1950, are pointed in another direction. We have been able to
observe in recent events that some of the reasons for complaints addressed to
them in this regard are disappearing. My dear colleagues, the elections have been
free in these two countries. The legislative elections have been free in Greece. The
48 AHCE: CEAB/5/1152: Allocution du Professeur Hallstein, 12 septembre 1963.

49 ‘Turkije en Europa,’ VPRO broadcast, 14 December 2004, at: https://www.anderetijden.nl/pro-
gramma/1/Andere-Tijden/aflevering/442/Turkije-en-Europa.
50 Parlément Européen, Séance du 28 novembre 1963, p. 153.
same is true for recent municipal elections in Turkey. The presentation of several
lists was permitted. I am still waiting to see several in Spain and Portugal.’51
What Dehousse could not foresee, of course, was how dramatically conditions
in Greece would change nor how the norms that he endorsed would shape the
community’s reaction.
3. Suspending the opening to Greece
Although the EEC’s membership norm became significantly more restrictive just
six months after the signing of the 1961 Athens Agreement, this had no immediate
impact on the status of Greece, which remained a parliamentary democracy at
the time. On 21 April 1967, though, a group of senior military officers in Greece,
citing an imminent threat of Communist take-over, overthrew the country’s
democratically elected government and replaced it with a military regime. The
new regime immediately suspended the Greek parliament and arrested many of
its members, as well as large number of opposition activists, journalists, and
others. Despite the military’s claims that its rule would be of short duration, these
developments provoked protests across Europe, especially among socialists and
trade unionists. Before long, though, European institutions began to react.
The first reaction from a European institution came not from the EEC but from
the Council of Europe, whose explicit purpose was the protection of democracy
and human rights. Within days of the coup, the parliamentarians belonging to the
Council of Europe’s Consultative Assembly passed a resolution expressing
‘sadness’ over the imposition of martial law and calling on Greece to ‘replace the
military regime with a parliamentary democracy’ (Papandreou 1998). On 25 May,
the assembly’s Legal Committee recommended that Greece be referred to the
Human Rights Commission for failing to justify its assertion of a right to derogate
from the European Convention on Human Rights.52
In contrast, the governments of the EEC’s member states were slow to take a
position on the Greek coup.53 After all, neither the coup itself nor the subsequent
actions of the military junta violated the explicit terms of the Athens agreement
or the Treaty of Rome. When the French foreign ministry sought advice on the
coup’s implications for the association agreement, its legal service replied that
France could either assert that the agreement had been invalidated by changes to
the fundamental conditions prevailing at the time of its signature or it could
utilize the new political conditions in Greece to reopen the case for an association
51 Parlement Européen, Séance du 28 novembre 1963, p. 150.

52 AHCE: BAC/173/1995/1430: Conseil de l’Europe: La Grèce et les Droits de l’Homme, 26 mai
1967. See also Buergenthal (1968).
53 In contrast, the United Kingdom (not yet an EEC member) was less circumspect. See
Nafpliotis (2012).
agreement with Spain (see Chapter 5 for details on this case).54 Advisors to
Belgian foreign minister Pierre Harmel urged him to proceed cautiously because
the Greek military could soon return to the barracks (‘as had happened in Turkey
in 1960’), political isolation encourages extremism, and in any case, recognition
of states is not the same as approval of particular governments.55
Within the EEC’s institutions, the overthrow of democratic rule in Greece was
first raised officially during a 2 May meeting of the European Parliamentary
Assembly’s Committee on the Association of Greece. The meeting began with
testimony from leaders of two major trans-European trade union networks, both
of which called upon the EPA to advise the European Investment Bank (EIB) to
suspend the loans accorded to Greece under the association agreement. The
European Commission’s representative stated that although the Commission had
not yet deliberated on the recent developments, it ‘could not remain indifferent to
the constitutional evolution of a country’ destined for full EEC membership. In
particular, he expressed his expectation that the mixed EEC-Greece Parliamentary
Commission, which had been established by the 1961 Athens Agreement, could
not function until the Greek parliament had been re-established.56
Although seemingly a factual statement, this proposition that the disruption of
the joint parliamentary commission might call the entire association agreement
into question was actually a controversial claim. Less than five years earlier,
within months of the signing of the Athens Agreement, members of the European
Parliamentary Assembly indicated that they interpreted Article 71 of the
agreement as authorization for the creation of a joint commission with the Greek
parliament. Acting through Coreper, the member states quickly informed the
EPA’s president Hans Furler that in their opinion, the Association Agreement did
not authorize creation of such a commission and advised that he block its
creation. In response, Furler expressed his strong disagreement with Coreper’s
view on Article 71 of the agreement, asserted the sovereignty of the APE and
indicated that the parliament would proceed with the joint commission.57 By
portraying the role of the joint commission as legally self-evident in the aftermath
of the coup, the Commission official was thus staking a risky position in a
sensitive battle between the member states and the APE. It is difficult to imagine
that he would have made such comments without the support of the post-1962
normative environment.
54 MAEF: EUROPE: Grèce (1961–1970) Dossier 233: Incidence des événements d’Athènes sur
l’association de la Grèce a la CEE, 28 avril 1967.
55 MAEB: 6641/1-s Grèce 1961-mars 1968: Note pour Mr. Le Ministre, Bruxelles, 5 mai 1967; Note:
Reconnaissance d’États [et] de Gouvernements, 5 mai 1967.
56 AHCE: BAC/6/1968/13: Commission de l’Association avec la Grèce, compte rendu, session 2
mai 1967.
57 MAEF: EUROPE: Grèce (1961–1970) Dossier 233: Télégramme, Paris, 17 février 1962; Note
pour le ministre, Paris, 10 mars 1962; Télégramme, Bonn, 15 mars 1962.
In the end, the meeting resolved that recent developments in Greece threatened
the ‘foundations’ of the 1961 agreement, which it described as a treaty ‘founded on
a shared conception of democracy of the contracting states’, and that no further
actions should be taken on the association, including by the EIB, ‘until the situ
ation in Greece is normalised and all the institutions established by the agreement
could resume their functions’. Finally, the committee resolved that the EPA would
formally question the European Commission on three points—whether it has a
position on events in Greece, whether it believes that the current situation pre-
vents the functioning of the association agreement whose purpose is to facilitate
the eventual full membership of Greece, and whether it has further information
on the fates of Greek parliamentary members of the mixed commission estab-
lished by the association agreement.58
The European Commission was thus caught between member states inclined
to avoid condemning the Greek junta and parliamentarians pushing for a robust
reply. Just one day after the meeting with the EPA, the college of the European
Commission decided that it did not yet have sufficient information to express a
formal position on the events in Greece, but that it would instruct the EIB to
defer its decisions on two loans to Greece pending further study by the
Commission.59 On 5 May, the president of the Coreper advised the Commission
to proceed cautiously in its response to the EPA’s questions, mentioning that some
member states were particularly insistent on this point.60
When the Parliament Assembly convened on 8 May, the presidents of its
political committee and its committee on the Greek association were outspoken
in their condemnation of the coup and subsequent developments as a violation of
the democratic principles underlying the process of European integration and the
1961 treaty linking Greece to the EEC. In response, the Commission’s vice-
president expressed his institution’s concern about developments in Greece but
avoided any specific comment on their implications for the association. He also
observed that the member states had maintained a similar reserve. In the debate
that followed, spokesmen for the parliament’s socialist, Christian democrat and
liberal groups also issued unambiguous condemnations of military rule as
incompatible with progress toward EEC membership. The session concluded with
a parliamentarian’s assertion that if military rule and recent repressions in Greece
did not soon end, then the Commission ‘would have the duty to question the
validity of the association agreement based on the recognition of the democratic
regime and civil and political liberties’ that existed in Greece when it was signed.61
58 AHCE: BAC/6/1968/13: Commission de l’Association avec la Grèce, compte rendu, session 2
mai 1967.
59 AHCE: BAC/209/1980/401: Procès-verbal de la réunion de la Commission, 3 mai 1967.
60 AHCE: BAC/209/1980/402: Procès-verbal de la réunion de la Commission, 8 mai 1967; MAEF:
EUROPE: Grèce (1961–1970) Dossier 233: Télégramme, Bruxelles, 8 mai 1967.
61 Parlement Européen, Séance du lundi 8 mai 1967.
No speaker mentioned commercial considerations and only Lando Ferretti, a for-

mer close associate of Benito Mussolini, discussed the Greek crisis in terms of the
West’s confrontation with Communism, which had been the dominant normative
frame before 1962. Given that neither Greece’s geographic location nor the nature
of its economy had changed since 1961, and the threat of an internal Communist
takeover had arguably increased, this can only be explained by a change in the
normative environment that rendered such discourses inappropriate and thus
ineffective in the context of a threat to democracy.
Despite the Commission’s reserve and the Council’s silence, the Parliament did
not let the issue drop. On 8 May and 10 May, members of the Parliament posed
formal questions to the Commission and Council asking whether Greece was still
eligible for EEC membership, whether the association accord should be
suspended, whether the normal execution of the accord could continue, and
whether any further information was available on the fate of the missing members
of the joint EEC-Greece parliamentary commission. On 11 May, the Parliament
approved a resolution stating that the suspension of the Greek parliament had
prevented an essential element of the association from functioning and calling for
a speedy return to parliamentary democracy in Greece. During discussion of the
resolution, all speakers emphasized the priority of respect for democracy and
human rights; one specifically insisted that the Community should only sign
commercial agreements with democratic regimes.62
The governments of the Six were fully informed of the Parliaments’
deliberations and formal questions, and the Commission worked closely with
Coreper to draft replies to each question.63 This process, like the larger debate
over how to respond to developments in Greece, was shaped by the diversity of
views within the Commission and especially among the member states. Both the
French and German governments preferred a return to constitutional rule in
Athens but worried that strong pressure could push Greece out of the Western
camp; the latter also worried that any rupture with Greece could harm its
commercial ties to the country (Pelt 2006, Plassmann 2012). However, the
membership norm that had consolidated since 1962—that the EEC is a
community of democratic states—ensured that the Council could not ignore the
parliamentarians’ concerns. In the face of this pressure, the Commission and
Council moved slowly toward tougher positions on the Greek junta.
On 29 May, European Commissioner for External Relations Jean Rey submitted
to his fellow Commissioners a detailed report on the state of affairs in Greece, in
which he noted that the European Parliament’s various resolutions were less
demanding than the outcome preferred by those who favoured a full suspension
62 Parlement Européen, Séance du jeudi 11 mai 1967.

63 ACCUE: CM2.1967.1101: Lettre à Monsieur le Président du Conseil, 12 mai 1967; Note
d’information, 16 mai 1967.
of the association. In conclusion, Rey recommended that the Community

distinguish between the current administration (‘gestion courante’) of the
association, which he said should be maintained, and the further development of
the association, which he said should be refused.64 He then qualified this recom-
mendation by proposing that the current administration should be loosely inter-
preted to permit the continued dispersal of funds by the EIB, contrary to the
demands of various parliamentarians and trade unions.65 Three days later, the
college of Commissioners adopted the essence of Rey’s proposal as its
recommendation to the Council, including the proposal that the Council
communicate its position to the Greek government. There was no mention of
geo-political or commercial factors in Rey’s report or in the ensuing discussions
among the Commissioners.66
In its own effort to shape the discussion, the Greek government then submitted
two documents to the Council and the Commission—a memorandum designed
to rebut calls for a suspension of the association agreement, and a declaration by
the new prime minister committing his government to the re-establishment
of parliamentary rule under a new Constitution.67 The fact that the Greek govern-
ment chose to frame its case in terms of constitutional legitimacy, rather than by
stressing its anti-Communist credentials, is yet another indicator of how much
EEC norms had changed in less than 10 years. Nonetheless, when the Council
finally met on 5 June, France and Germany opposed any declaration directly
critical of the Greek regime, which they feared could weaken a key NATO ally.
Instead, the six foreign ministers endorsed Rey’s call that the association be
limited to current administrations.68 But this was not the end of the struggle to
shape the community’s position on military rule in Greece.
Now that the Commission and Council had determined their positions, it was
finally possible for the two institutions to reply officially to questions posed earlier
by members of the Parliament. With regard to the question concerning Greece’s
continued eligibility for membership, the Commission responded on 4 July that it
‘could not remain indifferent to the constitutional regime of a country, like
Greece, that was destined to become a member of the EEC [and that the] events
that have transpired since 21 April raise concerns that could compromise the
development of the Association’. With regard to the question of whether the
‘normal execution’ of the agreement could continue after the military takeover,
the Council and Commission responded jointly on 10 July: ‘The Community is
64 Here and elsewhere, I translate ‘gestion courante’ and ‘application courante’ as ‘normal manage-
ment’ and ‘normal application’, with ‘courante’ understood as normal in the sense of ordinary, every-
day, or standard.
65 AHCE: BAC/173/1995/1430: Note pour MM. les Membres de la Commission, 29 mai 1967.
66 AHCE: BAC/209/1980/405: Procès-verbal de la réunion de la Commission, 1 juin 1967.
67 ACCUE: CM2/1967/1115: Letter and attachments to Mr. P. Harmel, 2 juin 1967.
68 ACCUE: CM2/1967/39: Procès-verbal de la réunion du Conseil, 5–6 juin 1967.
examining with the greatest attention the repercussions the evolution of the
situation in Greece could have on the application of the Athens Accord. At pre-
sent, the normal application of the Association is assured.’69 The reference to ‘nor-
mal application’ was critical because it referred implicitly to the many calls for a
suspension of any further development in the association.
A Member of Parliament posed two follow- up questions, asking the
Commission to clarify which aspects of the Association were still being applied
and which had been suspended.70 On 13 September, the president of the
Parliament’s Commission on the Association with Greece informed the European
Commission that because of the lack of progress toward any return to
parliamentary democracy in Greece, the Parliament would soon revisit the status
of the association agreement.71 The Commission replied two weeks later that the
association agreement’s legal provisions on tariffs and trade were being
implemented, but that the mixed parliamentary commission was not functioning,
the harmonization of agricultural policies had been suspended, the negotiation of
a new financial protocol had been postponed indefinitely, and 125 million dollars
in investment funds allocated to Greece would likely expire on 31 October with-
out being dispersed.72
Behind the scenes, though, the status of Greece’s EIB loans remained
controversial, with all sides recognizing the symbolic importance of this otherwise
technical issue. The Council thus worked hard to balance its commitment to
avoid any political decision that could be seen as advancing the association with
its preference to avoid taking a clear stance. On 11 September, the EIB’s president
formally requested the Council’s advice on what to do with the remaining $125
million, including a specific $10 million loan request from Athens for a road pro-
ject in Crete.73 On 26 September, the EIB requested the Commission’s advice on
the $10 million loan request. The Commission discussed the matter the following
day and decided to advise against disbursement of the loan. In addition, the
Commission invalidated the EIB administrative council’s earlier vote in favour of
the loan and asked the Council to address the larger issue of the Greek associ
ation as soon as possible.74 Two days later, the APE’s commission on the Greek
association issued its own negative opinion on the loan. The Greek government
responded with a strong criticism of the Commission for over- stepping its
authority and called on the Six to consider the implications of this move.75
69 AHCE: BAC/173/1995/1430: Note d’Information, 4 juillet 1967 and 10 juillet 1967.
70 AHCE: BAC/26/1969/696: Note d’Information, 17 juillet 1967.
71 AHPE: Commission de l’Association avec la Grèce: Procès-verbal de la réunion 29 septembre
1967, Annexe II.
72 AHCE: BAC/173/1995/1450: Note d’Information, 28 septembre 1967.
73 ACCUE: CM2.1967/1115.
74 AHCE: BAC/259/1980/8.
75 MAEB: 6641/1-t Grèce 1967: Télégramme Athènes, 2 octobre 1967.
But the Six seem to have shared a preference (consistent with their earlier
rejection of a proposed formal decision and communiqué to Athens) for allowing
the 31 October deadline to pass without action. The EIB’s request for instructions
was discussed at multiple Coreper meetings in September and October, but no
clear answer was given. Instead, they sought legal advice from the Council’s legal
service, which advised that Greece had no legal right to the $125 million, which
had been intended only as a limit on possible lending. When Coreper met in late
September, they rejected the Commission’s proposal that the Council re-examine
the association but failed to reach agreement on how to advise the EIB regarding
the loan. Soon thereafter, the Council secretariat advised the member states that
the $125 million figure was simply a maximum allocation, not a legal commit-
ment to Greece.76 Keeping up the pressure for a decision, the EIB’s president
wrote again to the Council on 9 September, reminding the member states that the
deadline for disbursing the funds would soon expire and asking for advice.77
Civil society pressure for a rupture with Athens was also mounting. Since June,
various unions, euro-federalist associations and student groups had called on the
EEC to break all contacts with Athens until democracy and respect for human
rights were restored.78 One of the largest trade unions networks in Europe,
the International Confederation of Free Trade Unions, added its weight to the
campaign in October, but this mobilization does not seem to have swayed the
member states.79
When Coreper reconvened on 13 October, Belgium and Germany insisted that
both the $10 million loan and the larger question of the association were matters
of member state competence and all agreed that they should be decided at the
forthcoming Luxembourg Council.80 In the meantime, the Belgian foreign minis-
try’s legal service advised that the Commission should be over-ruled on several
grounds: it had exceeded its competence under the treaty; its opinion had been
submitted two days after the EIB’s deadline for consultation; and its argument was
based on a misinterpretation of Article 71 as requiring a functioning joint parlia-
mentary commission.81
When the six foreign ministers discussed the issue at the Luxembourg Council
on 24 October, five were inclined to overrule the Commission while Italy insisted
that the release of funds must await the release of the Greek MPs. Italy’s
commercial interests may have been a factor here, but it is unlikely that Rome
would have been willing to reverse its earlier support for the Greek association
76 ACCUE: CM2.1967/1115. 77 ACCUE: CM2.1967/1115.

78 AHCE: BAC/173/1995/1430: Lettre à M. Walter Hallstein, 22 juin 1967; AHPE: Commission de
l’Association avec la Grèce, Procès-verbal de la réunion 29 septembre 1967, Annexe I.
79 AHCE: BAC/173/1995/1430: Lettre à Monsieur Rey, 25 octobre 1967.
80 ACCUE: CM2.1967:1115 & CM2.1967:1101; MAEB: 6641/1- t Grèce 1967: Compte-
rendu,
Réunion du Coreper, 13 octobre 1967.
81 MAEB: 6641/1-t Grèce 1967: Lettre à Mr. Harmel, Bruxelles, 19 octobre 1967.
without the political cover provided by the new normative environment. But in
the absence of Italy’s agreement, the only option was to leave the matter in the
hands of the EIB.82 Coreper nonetheless allowed another week to pass before
informing the EIB on 31 October that no new loans should be made after the
deadline the following day.83 After failing to reply to repeated inquiries, Coreper’s
president finally informed the Greek ambassador on 14 November that the funds
were no longer available because the deadline for accessing them had expired; no
further explanation was provided.84
In the meantime, however, both the European Commission and the European
Parliament had informed the Greek ambassador that the association agreement
could not function normally without a full restoration of democracy in Greece.85
In late November, responding to an earlier question from the EPA, the Council
reaffirmed its concern about the political situation in Greece and its commitment
to blocking any further development of the association agreement until the
restoration of democracy. Just as important, the Council declared that the
association agreement had ‘political scope’ because it foresees Greece’s eventual
adhesion to the Community. And for the first time, the Council linked the
freezing of the association to the non-functioning of the joint parliamentary
commission, which it described as an interesting and useful element of the
association—a marked departure from the member states’ earlier refusal to attribute
any legal or political significance to the parliamentary commission.86
As this episode shows, the normative definition of the EEC as a community of
parliamentary democracies had acquired a fully taken-for-granted status by this
point, regardless of what preferences were actually held by the governments of the
member states. And as such, European states without ‘free institutions’, like
Greece under military rule, were now indisputably ineligible for EEC membership,
regardless of their geographic location or position in Cold War politics. Greece’s
association agreement with the EEC was thus ‘frozen’ in late 1967 and remained
so until the fall of the military regime in 1974. The Community continued to
implement its obligations under the EEC-Greece customs union to reduce tariffs
on agricultural and industrial imports from Greece but the Greece- EEC
association council no longer met at ministerial level, no new loans were granted
to Athens, and no further steps were taken to liberalize the movement of labour,
services, and capital or to harmonize legislation or agricultural policies—all
measures that were critical to preparing Greece for accession.87
82 MAEF: EUROPE: Grèce (1961–1970) Dossier 233: Fonctionnement de l’association de la Grèce”
10 octobre 1967; Télégramme, Bruxelles, 25 octobre 1967; MAEB: 6641/1-t Grèce 1967: Europe
Bulletin Quotidien no. 2785, 24 octobre 1967.
83 ACCUE: CM2.1967:1115. 84 AHCE: BAC/17/1969/38.
85 AHCE: BAC/17/1969/38. 86 AHCE: BAC/173/1995/1430.
87 MAEF: Europe 1944—Sous-Série Grèce 1971–1976, carton 3314: Note sur l’Accord d’Association
de la Grèce à la CEE. Athènes, 23 août 1974.
4. Conclusions
Whereas the previous chapter illustrated how EEC membership norms facilitated
closer ties between the community and Greece, Spain, and Turkey immediately
after the Treaty of Rome, this chapter’s account of the EEC’s encounters with the
same three states in the following half-decade supports the book’s argument that
membership norms can also be a powerful obstacle to states seeking to join a
regional community. Not long after the newly-created EEC had pursued closer
ties with Spain despite its dictatorial regime, a radical change in EEC membership
norms in 1962 raised the political costs for member states favouring this goal to
such an extent that they ended up refusing an association agreement as long as
Spain remained under non-democratic rule. A similar outcome is evident in the
Turkish case, although there it was a case of both international and domestic
change: the restoration of democracy in the early 1960s kept the country in line
with the EEC’s evolving membership norms, so the norms continued to support
Ankara’s quest for closer ties to the community. Greece also underwent a regime
change in this period but the post-1962 democratic norm could not be reconciled
with the military junta that seized power in Greece in 1967, so the community
chose to freeze progress toward eventual membership.
As shown in Chapter 7, the community’s membership norms became even
more stringent after 1970, with significant consequences for the membership
aspirations of all three countries, and after 1991 for Ukraine.
7
Liberal Democracies, 1970–2005
It is unthinkable even to imagine that a state which did not respect

human rights would ever be allowed to join the Community. These
rights are an expression of all that is civilized in the constitution of the
states, of all that is democratic; they are essential, irrevocable conditions
for membership of the Community.
—Walter Hallstein, former President of the European
Commission, 19721
The principles of pluralist democracy and of respect for human rights
form part of the common heritage of the peoples of the States brought
together in the European Communities and constitute therefore essen-
tial elements of membership of the said Communities . . .
—European Commission, 19852
By the start of the 1970s, the European community had redefined its membership
norm to require that aspirant states must ensure not only parliamentary democ-
racy, but also the rule of law and respect for human rights—a point that the
Birkelbach report had considered controversial less than a decade earlier. As dis-
cussed in Chapter 3, this normative expansion was driven above all by the com-
munity’s deliberations in the 1960s over membership overtures by military
dictatorships in southern Europe. Before long, though, it had acquired full nor-
mative (i.e. taken-for-granted) status in the community and its neighbourhood.
The best-known examples of EU decision-making entrapped by the commu-
nity’s membership norm in the post-1970 period are the eastern enlargements of
2004 and 2007, but this process has been so well documented and analysed else-
where (e.g. Schimmelfennig 2003, Sedelmeier 2005) that it need not be redone
here. Instead, this chapter focuses first on the consequences of the EU’s expanded
membership norm for Greece’s, Spain’s, and Turkey’s continued pursuit of acces-
sion from the 1970s through the early 2000s. It then explores a new case, Ukraine,
showing that the EU’s membership norm impeded Kiev’s pursuit of closer ties to
1 Hallstein 1972: 48.

2 CVCE: Commission opinion on the applications for accession to the European Communities by
the Kingdom of Spain and the Portuguese Republic (31 May 1985).
Europe in the first decade and a half after interdependence in 1991 because of
widespread concern within the Union regarding Ukraine’s actual commitment to
liberal democracy and the rule of law.
1. Reopening to Greece
Civilian rule was restored quickly in Greece following the collapse of the military
junta in July 1974, but the new government led by Prime Minister Konstantinos
Karamanlis faced significant challenges. After years of living under a right-wing
dictatorship that had never been fully rejected by Bonn, Paris, or Washington, the
Greek population was restive and suspicious of NATO. Greece had weathered the
1973–74 recession better than many other countries, but its sheltered and heavily
nationalized economy was ill prepared for international competition. Finally, the
Cyprus crisis and military standoff with Turkey in the Aegean created real
strategic uncertainty. For Karamanlis, reestablishing Greece on the path to EEC
membership that had been blocked in 1967 was critical to addressing all of these
challenges (Karamouzi 2015).
But doing so would not be easy. First, the 1962 Athens Agreement had
envisioned a 22-year transitional period before completion of an EEC–Greece
customs union, as well as other measures to prepare Greece for eventual accession
to full membership, but most of these had been interrupted in 1967. In addition,
the EEC had barely absorbed the accession of Denmark, Ireland, and the UK in
1973 when the unexpected collapse of the Greek junta thrust the question of
enlargement back onto the community’s agenda. And it was no secret that the
Greek economy was less developed than that of Ireland, the poorest of the Nine,
so the member states knew that Greek accession would require a major investment
of community resources. Moreover, in the agricultural sector where Greece was
actually competitive, its exports (fruit, wine, olives, and olive oil) posed a direct
threat to French and Italian producers, so Greek accession posed economic risks
for two of the largest member states. Finally, the longstanding community
practice of taking a balanced approach to Greece and Turkey’s separate quests for
closer integration with Europe would make it difficult to put Athens on a fast
track to accession. Yet this is exactly what Karamanlis wanted.
The key to this diplomatic game, as Karamanlis understood better than his
EEC counterparts, was getting the community to accept in principle that Greece
was eligible for membership, thereby reversing its 1967 decision to block the
country’s path to accession. As a Belgian foreign ministry official recalls, ‘In the
period after the Colonels’ eviction, everyone agreed that something had to be
done for Greece, and everyone agreed that they should not enter
immediately. . . . And Karamanlis, very intelligently, immediately went around the
capitals. And even though everyone had agreed that they were going to be very
A EUROPE OF LIBERAL DEMOCRACIES, 1970–2005 161
careful, all prime ministers and heads of state promised Karamanlis

membership. . . . And then, it was necessary to live with this situation.’3 This brief
recollection begs the question of why EEC heads of state and government agreed
to put Greece on a fast track to accession rather than pursue other options, such
as a significant pre-accession transition period, that better fit their objective
assessment of what was best for the Community and even for Greece itself.
Fortunately for Karamanlis, by the time the Greek junta collapsed in 1974, the
EEC’s membership norm had been revised again. Starting around 1970, the EEC
had deepened its self-definition as a community of parliamentary democracies by
adding an explicit commitment to human rights and fundamental freedoms. In
fact, as demonstrated in Chapter 3, the public and institutional contestation over
how to respond to the 1967 coup in Athens had contributed significantly to this
normative change from parliamentary democracy to liberal democracy. This
liberal democratic norm enabled Karamanlis in 1975–76 to overcome significant
resistance within the community to his government’s desire to move Greece
quickly toward EEC membership. As shown below, the norm delegitimated
opposition to Karamanlis’ insistence that the Community, including those
member states with deep reservations, had no choice but to reward his
government’s commitment to institutionalizing democracy, the rule of law, and
respect for human rights.
The day after Karamanlis’ return to power in Athens on 24 July 1974, President
of the European Commission Francois- Xavier Ortoli sent him a letter of
congratulations indicating ‘the progress of democracy in Greece can only have
beneficial effects on the development of our association’.4 Within days, the French
government (then holding the EEC presidency) began to urge its community
partners to make a more visible and concrete gesture of support for the new
Greek government.5 On 2 August, the French delegation in Brussels informed the
Commission that it intended to initiate a discussion in Coreper or the Council on
defreezing the Greek association; Ortoli’s initial reaction was cautious but his
cabinet began to prepare papers on what this might involve.6
On 22 August, less than a month after the collapse of military rule, Athens
informed the EEC (via an aide-mémoire to the French presidency-in-office) of its
desire to tighten its ties to Europe, including reestablishing the association
agreement and reaffirming Greece’s vocation for accession.7 To justify this goal,
the document focused on Greece’s democratic credentials: ‘the support that the
3 HAEU: INT614: Etienne Davignon. Interview transcript, 9 November 1998–18 January

1999, p. 35.
4 AHCCE: BAC 50/1982–18: Ortoli letter to Karamanlis, 25 July 1974.
5 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3314: Télégramme 3260–68,
Paris à DELFRA BRUXELLES, 9 septembre 1974.
6 AHCCE: BAC 50/1982–18: Note: The Association with Greece, 2 August 1974.
7 AHCCE: BAC 50/1982–18: Letter by Francis Lott plus attachment, 27 August 1974.
Greek people as a whole give [to the new government] derives from its re-
establishment of respect for democratic principles and individual and human
rights’. It also framed its request in terms of the Community’s past commitments:
‘the immediate “unfreezing” of the association would constitute for the Greek
people an act of faith to the ideals that the European Community demonstrated
during the country’s submission to the dictatorial regime’. Even more audaciously,
‘The current exercise of power by a representative government and the full respect
of democratic principles and fundamental liberties fulfill all the conditions
necessary for the full reestablishment of the functioning of the association
between Greece and the European community.’
These arguments found immediate sympathy in the European Parliament,
especially the parliament’s president, Cornelis Berkhouwer, who considered his
institution the guarantor of the Community’s democratic spirit.8 Following
meetings with the government in Athens, Berkhouwer told a press conference
that he expected the Council of Ministers would soon consider full membership
for Greece and that this goal should be achievable within less time than the
Greece–EEC association had been frozen (i.e. seven years). But within the
European Commission, Berkhouwer’s pronouncement was not appreciated: ‘It
does not seem to correspond with what we have discussed up to now.’9
The Commission’s official Communication to the Council, dated 30 August,
summarized the state of the association agreement and identified a number of
steps that could be taken to revive it and thereby demonstrate that ‘the
Communities’ aim is to intensify relations with a democratic Greece’. However,
the same document insisted, ‘A full reactivation of the association will only be
possible when a fully democratic regime is established in Greece. That is to say,
after free elections have been held and democratic institutions re-established.’10
Within the Ministry of Foreign Affairs in Paris, the assessment was also measured:
‘It is advisable . . . to be cautious on the chances of success and the calendar of
restarting the association relationship as foreseen by the Athens accord. First,
certain member states will tie their attitude to the progress of the reestablishment
of democracy. Second, the complexity of economic and political problems raised
by the Greek demands calls for a distinction between the medium and the short
term.’ As for the Greek government’s interest in rapid accession to full
membership: ‘It is permissible to question the consequences [of accession] on the
general balance of the Community and on the attitude of certain interested
European countries such as Spain or Turkey.’11
8 CVCE: Interview de Cornelis Berkhouwer accordée à 30 Jours d’Europe (septembre 1974).

9 AHCCE: BAC 50/1982–18: Note to Mr. Missir, 23 August 1974, plus attachment.
10 AHCCE: BAC 50/1982–18: Communication de la Commission, 30 August 1974, p. 10.
11 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3314: Note: La Grèce et la CEE,
Paris, 3 septembre 1974.
When Greek Foreign Minister Georgios Mavros visited France, Germany, and
EEC headquarters in Brussels in September, he stressed his government’s desire
not only to reactivate the association but also to move quickly toward full
membership.12 In response, the French government instructed its Brussels
delegation that the Council should be persuaded to return to the normal
application of the association agreement and ‘to restate, if the Greek party insists
on it, the final objective formulated in 1961, that is to say the eventual accession
of Greece to the communities’.13 During Mavros’ visit to Bonn, the German
government announced its support for a relaunch of Greece’s association to the
EEC, its accession to full membership of the community, and its return to the
Council of Europe.14 But such rhetoric was no guarantee that accession would
happen soon. When Finance Minister Yiangos Pesmazoglou visited Brussels later
the same month, EEC officials stressed that the harmonization of agricultural
policies, which was already deadlocked before the 1967 coup, could take years to
implement (Karamouzi 2015: 7). As the French delegation in Brussels reminded
Paris, Greece must understand that ‘their country’s history may have stopped in
1967 but the same was not true for the community, which since then has regulated
its agricultural markets and welcomed three new member states’.15
The Nine were thus quick to celebrate Greece’s return to democracy but not
ready to embrace the new government’s quest for a short timetable to EEC
membership, as evident in this joint statement from 16 September: ‘The Nine
applaud the restoration in Greece of personal and political freedom and
most heartily welcome the efforts in this direction by the Government of
Mr. Karamanlis. The clearly affirmed resolve by the Greek Government to lose no
time in completing the process of democratization, the initial action taken to this
end and especially the restoration of the democratic Constitution of 1952 are seen
by the Nine as enabling Greece to regain her place in democratic Europe and
particularly within the Council of Europe. As far as the Nine themselves are
concerned, they ask the appropriate authorities of the Council of Europe to take
the action required to attain this objective.’16
The following day, the Council of Ministers (not to be confused with the
Council of Europe) announced very limited concessions in response to the
ambitions expressed by the Greek government in August: they agreed to reactivate
12 Greek Foreign Secretary Meets President of the EEC Francois-Xavier Ortoli, 11 September 1974,
Associated Press, https://www.youtube.com/watch?v=POYADnI1GOU.
Paris à DELFRA BRUXELLES, 9 septembre 1974.
14 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3314: Télégramme 3644/50,
Bonn à Paris, 11 septembre 1974.
DELFRA BRUXELLES à Paris, 13 septembre 1974.
16 Commission of the European Communities. Bulletin of the European Communities, no.9,
1974, p. 87.
Greece’s association with the community, to unblock the $55 million loan from
the EIB that had been frozen in 1967, and to raise the annual quota on imports of
Greek wine intended for the production of vermouth. But Athens principal
aim—a renewed commitment to full accession, preferably without major delay—
was not mentioned and no other steps were taken to move in this direction.17 By
this point, it was clear to Athens that its European campaign was not succeeding.
Several weeks later, Karamanlis announced that free elections would be held
on 17 November to replace the transitional government of national unity that he
had created after the fall of the junta in July (Clogg 1987: 208). These elections
produced a huge parliamentary majority for Karamanlis’ conservative New
Democracy party. In the lead-up to a subsequent national referendum on whether
the Greek state should be a constitutional monarchy or a parliamentary republic,
Karamanlis then disappointed some in his party by refusing to endorse a return
of the king, and voters opted decisively for a republic. These major steps toward
the re-democratization of Greek institutions were completed with the indirect
election by parliament of a new president on 18 December.
Karamanlis did not waste time in exploiting these victories at home for gains in
Europe. On 26 November, during preliminary Greece–EEC talks on reviving the
association treaty, the Greek government tabled another aide-mémoire stating its
intention to join the community as a full member in the very near future and
requesting a transitional period like that of the three newest member states,
which had been granted special derogations for a period of time after accession.18
Two days later, the Council of Europe invited Greece to rejoin the organization,
from which the junta had resigned in 1969 to avoid expulsion due to its human
rights violations.19 The resulting boost to Greece’s democratic credentials would
make it easier for Karamanlis to leverage the liberal democratic norm with which
the EEC had identified itself in recent years.
On 2 December, at the first meeting of the EEC-Greece Association Council
following the elections, Commission Vice- President Christopher Soames
congratulated Greece on its ‘determined return to democratic traditions [that]
had made it possible for the two partners to rapidly reestablish confidence’, while
French Foreign Minister Jean Sauvagnargues, speaking on behalf of the Nine,
expressed their desire for ‘a growing relationship between the Community and its
associated partner . . . that should lead . . . to Greece’s full membership in the
Community’. But such rhetoric did not mean that all scepticism about Athens’
goals and timeline had evaporated. For example, frustrated by a discussion of
17 AHCCE: BAC 50/1982–18: Press release and Council statement, 17 September 1974 & Letter
from Mr. J. Sauvagnargues, 3 October 1974.
18 Council of Europe (1974). Annuaire Européen/European Yearbook, Vol. XXII, p. 565. La Haye:
Martinus Nijhoff.
19 CVCE: Resolution (74) 34 of the Committee of Ministers of the Council of Europe
(28 November 1974).
fruits and vegetables, the Greek head of delegation asked pointedly ‘Does the
Community expect to place Greece in the framework of its global Mediterranean
policy or is it understood that Greece is part of an Association agreed with an aim
of complete membership?’20
When Sauvagnargues visited Athens in late February 1975, Karamanlis openly
requested full EEC membership at the earliest possible date (Karamouzi 2015: 9).
In return, Sauvagnargues reassured him that France remained favourable to
Greece’s joining the community but explained ‘it is nonetheless necessary to
observe a certain prudence as an accession negotiation will pose numerous
problems . . . ’21 The difficulty of satisfying Karamanlis’ request were apparent
when the Nine discussed Greek accession in early May. Within a meeting of
Coreper, Belgium called for a lengthy pre-accession transition period to ensure
that Greece would be ready to join the common market, Italy proposed some sort
of ‘special regime’ to bring Greece closer to the community without the conse-
quences of full membership, and Germany indicated that it did not expect Greece
to be ready for accession before 1985.22 A Council of Ministers discussion of the
same issue was summarized as follows by the French foreign ministry: ‘All our
partners intend to address this question with a prudence equivalent in some cases
to strong reservations. Certainly, no member state challenges the principle of
Greece’s accession to the EEC. But most of them consider the project premature.’23
In public, EEC governments continued to support the principle of Greek
accession. When Karamanlis visited Bonn in mid- May, Chancellor Helmut
Schmidt declared his government’s support for Greece’s candidacy as a full
member and indicated that he would advocate it within community discussions.24
It appears, however, that such statements did not reflect actual preferences or
intentions. In early June, the French embassy in Bonn informed Paris: ‘For various
reasons, the German government remains committed to delaying the process of
accession.’25 This combination of principled support in public and deep
reservation in private seems to have motivated Karamanlis to escalate his
campaign by insisting that Greece was a liberal democracy and thus entitled to
full membership as soon as possible.
20 AHCCE: BAC 50/1982–18: Procès- Verbal de la 37ème session du Conseil d’Association,

18 décembre 1974.
21 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3315: Note: Grèce/CEE, 4
avril 1975.
Delfra Bruxelles à Paris, 6 mai 1975.
23 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3315: Note: CEE/Grèce, 6
juin 1975.
24 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3315: Note: La visite officielle à
Bonn du Premier Ministre Grec M. Caramanlis 15–17 mai 1975, undated.
Bonn à Paris, 9 juin 1975.
The first step to this end came on 12 June 1975, when the Greek government
submitted a one-sentence letter to Irish foreign minister and EEC President-in-
Office Garret Fitzgerald: ‘On behalf of the Government of the Hellenic Republic, I
have the honour to inform Your Excellency that Greece hereby applies to become
a member of the European Economic Community in accordance with the
provisions of Article 237 of the Treaty establishing that Community.’26 The brevity
of this letter belied its significance: by referencing Article 237 (about accession)
rather than the 1961 Athens Agreement, which was based upon Article 238
(about association), Karamanlis was signalling that he considered Greece to be
eligible for membership without significant delay.
That same day, Karamanlis summoned the EEC’s ambassadors to his office in
Athens to explain his move. As reported by the French ambassador, the prime
minister ‘especially insisted on the political aspect of his request. Not only is
Greece’s destiny to belong fully to the Europe in formation, but its accession will
help to fortify the democratic regime that has just been reestablished.’
Furthermore, in contrast to the Athens Agreement’s indication that accession
could happen in 1984, and some voices suggesting that even this might be too
soon given the rupture during military rule, Karamanlis mentioned a transition
period of just five years.27 Karamanlis was thus forcing the member states to act
in accordance with the community’s prevailing membership norm, to which they
had recently and repeatedly paid rhetorical homage, or to violate the norm
publicly by following their preferences for a lengthy and perhaps undefined pre-
accession transition period.
Initial communications among member states regarding the Greek application
confirmed earlier indications of divergent preferences. Denmark declared it was
‘favorable to the enlargement of the community framework provided that it is
compatible with respect for the treaties in effect’.28 Several days later, the Italian
government declared that it agreed in principle that Greece should become the
tenth member of EEC but it was concerned about modalities and the timing of
accession. In concrete terms, Italy did not want to pay economically for a ‘political
decision’. Instead, Italy expected compensation from other member states for
consequences that would affect it first and most heavily, such as the import of
Greek wine and oil that competed directly with Italian products.29 The German
government welcomed Greece’s application as proof of the Community’s
attractiveness but stated that it would consider ‘the complex problems in detail
26 CVCE: Greece’s application for accession to the EEC (Athens, 12 June 1975).
Athènes à Paris, 12 juin 1975.
Copenhague à Paris, 13 juin 1975.
29 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3315: Télégramme 1.519/1.522,
Rome à Paris, 16 juin 1975.
and in a spirit of open-mindedness’.30 The UK informed its partners that it

favoured a negative reply to the Greek request (Soriano 2016: 149).
Given this range of preferences among the member states and the significant
consequences that Greek accession would have for some of them, one might
expect that the only possible consensus would be a return to the institutions and
calendar of the Athens agreement. And yet, when the Council of Ministers met
on 24 June, discussions ‘revealed the cleavage between the delegations favorable
in principle to Greece’s accession and those who resign themselves to it with some
difficulty’—an indication that no member state was resolutely opposing it. In the
end, the Council agreed to recognize ‘with satisfaction Greece’s decision to
request, following its return to democracy, its accession to the Community’.31
This outcome shows that Karamanlis’ ability to link his agenda to the community’s
outspoken identification with liberal democracy as the basis for membership
eligibility was beginning to entrap sceptical member states and move the
Community towards Greece’s position. This momentum was reinforced three
days later when the President of the Council reminded the EEC-Greece Joint
Parliamentary Commission that the Community had ‘welcomed with great
satisfaction the restoration of parliamentary democracy in Greece’ and declared
that the ‘[t]he time has come to give a new stimulus to our Association, the
ultimate objective of which is Greece’s accession to the Community as a full
member’.32 In turn, the Joint Commission recognized the Greek people’s
‘attachment to democratic principles’ and recommended that the Community
‘accelerate the procedures foreseen for Greek membership’.33
The next critical phase came in the autumn, as the European Commission
prepared its official opinion (‘Avis’) on Greece’s quest for accession. In July,
Karamanlis had urged the Commission not to ‘mix up the Greek application for
membership with too many other questions [and] not to give too much attention
to purely material issues’.34 Nonetheless, at a Council meeting in mid-September,
Commission President François-Xavier Ortoli asked for guidance on how to
handle overlapping political and economic considerations in the Greek case
including conflicts of interest between member states, especially in the
agricultural sector which he said ‘should not be under-estimated’, and the risk
that further enlargement could impede consensus in the Council. Subsequent
30 CVCE: Athen beantragt offiziell den Beitritt zur Europäischen Gemeinschaft. Frankfurter
Allgemeine Zeitung 13 juni 1975.
31 MAEF Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3315: Télégramme 2225–30,
Delfra-Bruxelles à Paris, 26 juin 1975.
32 HAEU: CM5/ADH-00002/001: Basis for a speech to be given by the President of the Council, 20
June 1975.
33 AHCCE: BAC 50/1982–19: Commission Parlementaire Mixte C.E.E.-Grèce, IXe Session, 25–7
juin 1975.
34 AHCCE: BAC 50/1982–30: Meeting between the Prime Minister of Greece and Sir Christopher
Soames, 30 July 1975.
discussion revealed that a number of member states remained sceptical about

Greek accession in the near future. As the French delegation reported to Paris,
‘Fearing to appear to reopen the agreement in principle already given on the
Greek request, all the delegations agreed to refuse to open the policy debate
requested by the Commission.’35
In the months that followed, continued dissensus behind closed doors among
the member states and within the community’s executive arm did not derail the
community’s welcoming response to Greece’s request. For example, the economic
and commercial counsellors of the EEC delegations in Athens undertook a
detailed study of the Greek economy and its readiness for accession. The
conclusion to their joint report, marked ‘confidential’ when it was distributed,
differed significantly from the scenario preferred by the Greek government:
‘Greek accession to the EEC would doubtless require a long transitional
period . . . so that the harmonisation of agricultural policies has the possibility to
be fully realised and Greek industry can resist competition from European
industry.’36 The Commission too was divided internally on the content of its Avis.
A dispassionate Commission review of political and economic instability in
Greece over recent decades concluded in November in favour of ‘a transition
period focused on a final clarification of Greece’s place in the free world’.37 Among
the Commissioners themselves, Ortoli managed to gain consensus on accession
in principle but they continued to disagree on ‘concrete procedures’ including the
desirability and form of a pre-accession transition period.38 Key elements of the
text were debated and variously adopted or rejected at the last moment.39
The final content of the Commission’s Avis, issued on 29 January 1976, reflected
these internal divergences as well as the normative framing of the issue that
Karamanlis had imposed on the Community. On the one hand, it echoed the
Council’s official position in a manner pleasing to Athens: ‘Given the avowed
aims of the Community in establishing the Association and Greece’s return to a
democratic form of government, there can be no doubt, in the view of the
Commission, that the Community must now give a clear, positive answer to the
Greek request’. Elsewhere in the document, the Commission expressed its
reasoning in more detail: ‘It is clear that the consolidation of Greece’s democracy,
Delfra-Bruxelles à Paris, 16 septembre 1975.
36 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3315: Rapport Commun sur la
Situation Économique de la Grèce dans ses Relations avec la Communauté Européenne, Athènes, 24
novembre 1975.
37 AHCE: BAC 15/1993–56: L’évolution politique et économique en Grèce, Novembre 1975.
Delfra-Bruxelles à Paris, 27 janvier 1976 & Télégramme 287/297, Delfra-Bruxelles to Paris, 30 jan-
vier 1976.
39 AHCE: BAC 259/1980–177: Procès-Verbal Spécial: Réunion de la Commission, 28 & 29 jan-
vier 1976.
which is a fundamental concern not only of the Greek people but also of the
Community and its Member States, is intimately related to the evolution of
Greece’s relationship with the Community. It is in light of these considerations
that the Commission recommends a clear affirmative reply be given to the Greek
request and that negotiations for Greek accession be opened.’40
However, the Avis did not recommend that negotiations be concluded
promptly or that accession should be completed in the near future, as Athens
wanted. Instead, by recommending an open-ended pre-membership transitional
period to address structural problems in the Greek economy, the Avis cast doubt
on when, if ever, Greece would actually be able to join the community. In so
doing, the Avis seemed to contradict the Council’s earlier recognition of Greece’s
eligibility for membership. This apparent contradiction was accentuated by the
fact that just a few years earlier, Ireland and the UK had been granted a special
period to adjust to membership after their accession. Finally, the Avis urged the
Community to ensure the efficiency of its internal decision- making before
undertaking any further enlargements.41 By highlighting economic and technical
difficulties in this manner, the Commission was effectively rejecting Karamanlis’
effort to frame accession as a political choice dependent only upon Greece’s fit
with the EEC’s liberal democratic membership norm.
The Greek government was surely aware that the content of the Avis did not
differ significantly from the actual range of member state policy preferences, but
it did not want this more conditional reframing of Greece’s European vocation to
stand. It thus launched a strong rhetorical offensive against the Avis, which it
considered inconsistent with the more positive attitude of certain member states
since July 1974 and of the Council of Ministers since June 1975. Two days after
the Avis was released, Karamanlis again summoned all EEC ambassadors to his
office to deliver a blistering critique. After describing the text as ‘incoherent’ and
asserting that it ‘runs contrary to the expressed position of the European
governments’, Karamanlis rejected any linkage between Greece’s EEC eligibility
and relations with Turkey and asked pointedly, ‘If the Greeks are convinced that
they cannot find a future in democratic Europe because of Turkey what are they
going to do with their future?’ In the end, he urged the Community to encourage
his country’s further democratic development by ‘by approving Greece’s request
for immediate integration’.42 Given that the Avis had stated explicitly that the start
of negotiations should not depend on the Turkey issue and had instead justified
40 CVCE: Commission Opinion on Greek application for membership of the European

Communities (29 January 1976).
41 In addition, the Avis discussed difficulties that Greek accession could create for the EEC’s
approach to the Cyprus crisis and for the past practice of treating Greece and Turkey in a balanced
manner, though it did state explicitly ‘it would . . . be inappropriate for the decision on Greek member-
ship to be dependent on it.’
42 MAEF: Série Europe 1944—Sous-Série Communauté Européenne, carton 3803: Télégramme
Lux Coreau à All Coreu, 1 février 1976.
its proposal for a pre-accession period on economic and structural grounds,

which Karamanlis did not mention, his address to the ambassadors was a clear
effort to tie the Community’s decision- making to its liberal democratic
commitments.
And once again, this tactic worked. In private discussions, the German foreign
ministry described Karamanlis’ reaction as ‘excessive’ and ‘based on internal
political considerations’ but nobody dared to say so publicly. Instead, at a press
conference in Bonn on 3 February, German Foreign Minister Hans-Dietrich
Genscher and Dutch Foreign Minister Max van der Stoel condemned the
Commission’s ‘mistaken interpretation’, declaring that the EEC must begin
accession negotiations immediately and must not impose on Greece a transitional
stage before membership.43 When the Council of Ministers discussed the issue six
days later, the member states most reticent about Greek accession—Belgium,
Denmark, Ireland, and the UK—were swept along by ‘the current of unanimity’
supporting a clear and positive response.44 In the end, the Council acknowledged
the Commission’s opinion but announced that it supported Greece’s request for
accession and that preparatory talks necessary for actual negotiations should
begin without conditions and ‘as soon as possible in a positive spirit.’45 As Le
Monde summarized the outcome, ‘The Commission gets its ears boxed.’46
The extent to which EEC member states had been entrapped into recognizing
Greece’s eligibility for full membership without delay, rather than motivated to do
so by their own assessment of the situation, is evident in the Community’s
inconsistent behaviour. Soon after the Council of Ministers welcomed Greece’s
request for accession in June 1975, the process was stalled by member states’ and
the Commission’s concerns about the country’s readiness. After strong pressure
from Athens, the Council agreed in February 1976 to begin negotiations with
Greece but preparatory talks had progressed very little by 1977 and the country’s
accession became bound up in intra- Community discussions about how to
respond to Portugal and Spain, which had recently also applied to join
(Karamouzi 2014: ch. 4).
Nonetheless, Athens’ sustained rhetorical focus on its democratic credentials
eventually broke the logjam. In April 1978, the Commission advised the Council,
‘The three countries have entrusted the Community with a political responsibility
which it cannot refuse, except at the price of denying the principles in which it is
itself grounded.’47 As French president Valéry Giscard d’Estaing, an advocate of
43 MAEF: Série Europe 1944—Sous-Série Grèce 1971–1976, carton 3315: Télégramme 298/301
Bonn à Paris, 4 février 1976.
Delfra-Bruxelles à Paris, 10 février 1976.
45 CVCE: Council press release on Greece’s application for accession to the European Communities
(9 February 1976).
46 CVCE: Ins Wasser, Der Spiegel (16 Februar 1976).
47 AEI: General Considerations on the Problems of Enlargement, 20 April 1978.
Greek accession, later recalled, ‘It was impossible to exclude Greece, the mother
of all democracies, from Europe’ (Karamouzi 2010: 265). The accession of Greece
was confirmed by treaty in May 1979 and took effect on 1 January 1981, far earlier
than envisioned by the 1961 Athens Agreement. It is therefore little exaggeration
to conclude, as veteran Belgian diplomat Etienne Davignon did decades later,
‘The negotiation of Greece’s accession somehow was something that nobody
wanted and that happened.’48
2. Reopening to Spain
Formal relations between Spain and the EEC during the early–mid 1970s were
limited to the implementation of a preferential trade agreement that the two sides
signed in June 1970. During this period, the community’s liberal democratic
membership norm was both an obstacle to Spanish accession and an indication of
how it might someday become eligible.
Behind the scenes, however, the community was divided on the question of
Spain’s relationship to the community.
Ever since the 1962 rejection of Spain’s quest for an association agreement
oriented toward full membership, the community had indicated repeatedly that
Spain would be welcome once it democratized its political system. By the 1970s,
respect for human rights had become part of the requirement, effectively raising
the bar on membership eligibility. For example, in October 1975, the nine EEC
Foreign Ministers condemned the Spanish government’s execution of five regime
opponents as a violation of human rights and ‘the common heritage of the peoples
of Europe’ and expressed their joint ‘desire that a democratic Spain might take its
place in the assembly of European nations’.49 The death of Franco in November
1975, after 35 years in power, thus removed an important obstacle to Spain’s
On the other hand, Franco’s death did not open an easy path to accession
because some of the Nine were in no rush to see Spain actually join the
community. These reservations were due in part to suspicion that Spanish
industry was not ready to compete with more developed economies and in part to
fears about the influx of cheap agricultural produce into the Common Market. In
particular, French and Italian farmers felt threatened by the prospect of competing
with Spanish grain, wine, citrus, olives, and olive oil.50
48 HAEU: INT614: Etienne Davignon. Interview transcript, 9 November 1998–18 January

1999, p. 35.
49 CVCE: Statement by the Foreign Ministers on the Situation in Spain (Luxembourg, 6
October 1975).
50 CVCE: Espagne: demande d’adhésion à la C.E.E. le 28 juillet. Le Figaro, 20 juillet 1977.
Within three months of Franco’s death, the foreign minister of Spain’s transi-
tional government travelled to Brussels to inform the EEC that the country would
submit its application for rapid accession as soon as its transition to democracy
was complete.51 As shown in this section, the community’s liberal democratic
norm enabled Spain to progress much faster in this quest than certain member
states apparently preferred. As in the Greek case, unfolding at about the same
time, the Spanish government’s framing of their country’s European vocation
made it politically impossible for Community decision-makers to resist, notwith-
standing their deep concerns about the economic consequences of Spanish acces-
sion for the EEC’s current member states.
Soon after Franco’s death, European parliamentarian Pierre Lagorce asked the
Council of Ministers whether the community was now open to Spanish accession.
In discussions among the Nine regarding how to answer Lagorce’s question, one
can see unmistakably how respect for democracy and human rights had become
central to membership eligibility in this era. On 5 February 1976, Luxembourg,
then holding the EEC presidency, proposed that Lagorce should be told that if
and when Spain applies for membership, ‘The question of respect for fundamental
values of democracy and liberty on which the community rests will be one of the
important elements of the community’s deliberations.’ Five days later, Denmark
proposed that respect for these values should be ‘a decisive element’ of the
community’s deliberations. Several weeks later, the Nine adopted a French
proposal that these values would be ‘one of the most important elements’ in the
community’s deliberations.52 The Council’s final answer to Lagorce noted that
Spain had not yet applied but that such an application would be assessed ‘in light
of all the principles and legal rules on which the Community is founded’.53
After a seven-month transition government, Adolfo Suárez was appointed
prime minister of Spain in July 1976. Despite his prior experience in the Francoist
regime, Suárez moved quickly to uproot the old system and replace it with a
pluralist democracy. He pushed a major law on political reform through the
Cortes (Spain’s parliament) and then called a national referendum in December
1976 in which 94 per cent of voters approved the changes. He also freed political
prisoners, legalized some once-banned organizations and political parties, and
ensured the adoption of a new electoral law to prepare for the country’s first free
election in four decades. After the parliamentary election in June 1977, which was
won by Suarez’s Union of the Democratic Centre, the new Cortes began drafting a
new constitution.
51 CVCE: La longue marche du douzième État membre de la C.E.E. La Libre Belgique, 21 décem-
bre 1978.
52 MAEF: Série Europe 1944—Sous-Série Communauté Européenne, carton 3803.
53 AHPE: PEO AP QP/QE E-0738/75, pièce 0030: Réponse à la question écrite n. 738/75.
At the same time, Suarez was also active on the European front. Within days of
taking office, his government reached out to the Council of Europe, the continent’s
foremost guardian of democracy and human rights. Its Parliamentary Assembly
replied by sending a delegation to Madrid to meet with the new government as
well as opposition politicians, journalists, and others. Unsatisfied by the partial
nature of Suárez’s reforms, they insisted upon free elections and the full protection
of civil and political rights.54 Following the elections in June 1977, the Council of
Europe congratulated Suárez’s government for its progress on democratization.55
From this point onwards, the credibility of Spain’s commitment to liberal
democracy allowed Madrid to frame its EC aspirations in terms of membership
eligibility rather than commercial concerns.
On 26 July 1977, Suárez submitted a letter to Belgian foreign minister and EC
president- in-
office Henri Simonet asking to open negotiations for full
membership. In that letter, Suárez declared his wish ‘to put on record the Spanish
Government’s identification with the ideals underlying the Treaties establishing
the European Communities and to express our hope that these negotiations will
lead us within a reasonable space of time to satisfactory results for the European
Economic Community and for Spain’.56 Yet even if Spain now held the normative
high ground, its candidacy still had to be approved by the community’s nine
member states.
Support from Germany seemed assured: Genscher had recently told the
Bundestag, ‘We wish to leave no one in doubt that we want a Europe of democracy,
a pluralistic Europe, a Europe of liberty . . . [W]e are determined to pave the way
into the Community for those states which have overcome nondemocratic forms
of government in recent years and which now wish to become members of the
Community. We must not disappoint the democracies of Greece, Portugal, and
Spain. We see the economic problems clearly, but such problems do not exist in
these three countries alone. We must respond to the political decision of these
states in favour of Europe with an equally responsible political answer.’57
In contrast, the French government was less inclined to support Spain’s
ambitions. As Le Monde described it, ‘the French position is clear: Spain, as well
as Greece and Portugal, will not be full members of the Common Market until
Mediterranean agricultural production is first reorganised’.58 The problem for
President Giscard was that Spain was rapidly consolidating its identity as a liberal
democratic state and thus fulfilled the community’s membership norm. As a
54 CVCE: Resolution of the Council of Europe on the situation in Spain (22 September 1976).
55 CVCE: Spain’s accession to the Council of Europe.
56 CVCE: Spain’s application for accession to the European Economic Community (26 July 1977).
57 EUDC: 441.213 ER Enlarg negoc pt.: Bundestag Supports EC Membership for Greece, Spain and
Portugal. The Bulletin, 8 June 1977.
58 Espagne: demande d’adhésion à la C.E.E. le 28 juillet, Le Monde, 20 juillet 1977; Marché com-
mun: la question espagnole, L’Express, 8 aout 1977.
result, when Suarez visited Paris in August, the reference to commercial

considerations in Giscard’s communique—‘France is committed not to sacrifice
its Mediterranean agriculture for Spanish membership’—could only be presented
as an issue affecting the terms of Spanish accession, not the country’s fundamental
eligibility for membership. Instead, Giscard declared somewhat vaguely, ‘France
has every reason to hope for the accession of Spain, a neighbouring country,
Latin, democratic, and friend, to the European Community.’59
At the same time that Spain was pursuing accession to the European
Community, it continued its outreach to the Council of Europe, which was
certain to boost his country’s acceptance by the EC. In response to concern within
the former organization’s Parliamentary Assembly that Spain had not yet adopted
a new constitution ensuring democracy and human rights, the heads of the
leading Spanish parliamentary parties adopted a joint declaration on 8 October
1977 confirming their ‘firm decision to provide constitutional guarantees of the
rule of law, adherence to the ideals enshrined in the Statute of the Council of
Europe and, in particular, the human rights and fundamental freedoms laid down
in the European Convention signed in Rome on 4 November 1950’. Four days
later, the Parliamentary Assembly welcomed the declaration and endorsed Spain’s
accession to the Council of Europe. The following day, the Spanish government
submitted its formal application for membership. On 18 October 1977, the
Council of Europe’s Council of Ministers noted the Spanish government’s
intention to sign the European Convention on Human Rights, and offered
membership to Spain. On 24 November 1977, Spain joined the Council of Europe
as a full member.
Meanwhile, the European Economic Community was following its own
internal procedures to decide on Spain’s application to start accession negotiations.
The fact that this process unfolded strictly in accordance with Article 237 of the
Rome Treaty, without any delays due to commercial or other concerns, is strong
evidence of how Spain benefited from fulfilling the community’s membership
norm. At a meeting on 19 September 1977, the EEC Council of Ministers
welcomed Spain’s application and asked the European Commission to prepare its
Opinion. The following February, the Commission’s communication to the
Council proposed a transitional period to facilitate Greece’s, Portugal’s, and
Spain’s adaptation to the acquis communautaire, which would start during
negotiation of the accession treaty but would not delay accession.60 The
Commission issued another communiqué in April offering a detailed discussion
59 CVCE: Communiqué de l’Élysée à l’issue de la visite à Paris du premier ministre espagnol

M. Adolfo Suárez (31 août 1977). Also: Poorer Nations’ Bid Puts Common Market on the Spot. The
New York Times, 5 September 1977.
60 CVCE: The transitional period and the institutional implications of enlargement.
Communication from the Commission of the European Communities to the Council. Bulletin of the
European Communities. February 1978, Supplement 2/1978.
of economic and institutional problems this triple enlargement would cause for
the community while simultaneously explaining why it had to be accepted: ‘The
three countries have entrusted the Community with a political responsibility
which it cannot refuse, except at the price of denying the principles in which it is
itself grounded.’61 These conflicting priorities are even reflected in the
Commission’s formal Opinion on Spain, issued in November 1978, which lists
weaknesses in the EEC and Spanish economies and discusses problems that
Spanish accession would create for industry and especially agriculture in the
Community before concluding with positive advice on Spain’s accession!62
On 20 December 1978, the Council decided to open accession negotiations
with Spain. One month later, the European Parliament noted that ‘that the
observance and defence of democratic principles form one of the essential
cornerstones of the Community’, welcomed ‘the fact that Greece, Portugal and
Spain have evolved from dictatorships into pluralist, parliamentary democracies’
and therefore expressed the Parliament’s ‘political will’ to see them join the
Community. To support their claim that pluralist democracies are eligible for
accession, the parliamentarians cited not the Treaty of Rome but the membership
norm that had become clear in recent years: ‘the statement made in December
1973 by the Heads of State or Government of the Community Member States on
the European Identity, the common declaration of April 1977 by the European
Parliament, the Council and the Commission of the European Communities on
the protection of fundamental rights and the declaration made by the European
Council in April 1978 on democracy’.63
The actual Spain–EC negotiations began on 5 February 1979. At the opening
ceremony, EC president-in-office Jean François-Poncet made the same point as
the Parliament while summarizing the changes that had made Spain eligible for
full membership instead of the limited commercial accord signed in 1970: ‘Spain,
without which one cannot speak properly of Europe, has in turn experienced
fundamental changes: the democratization process has led to the reestablishment
of pluralist democracy and fundamental rights enshrined in the new Constitution
that the Spanish people approved by referendum on 6 December 1978.’64
Commercial concerns reappeared on 5 June 1980, just one day before a planned
meeting of the Nine and Spain and one week before an EC Summit, when Giscard
called for a halt to further enlargement in an address to the Assembly of French
Chambers of Agriculture.65 His government explained that this was needed to
61 AEI: General Considerations on the Problems of Enlargement, 20 April 1978.

62 CVCE: Commission Opinion on Spain’s Application for Membership (29 November 1978).
Bulletin of the European Communities, Supplement 9/78.
63 CVCE: European Parliament resolution on the prospects of enlargement of the Community
(18 January 1979).
64 CVCE: Statement by Jean François-Poncet at the opening of the accession negotiations with
Spain (Brussels, 5 February 1979).
65 CVCE: The Member States’ reservations towards Spain’s accession (8 July 2016).
save the Community from excessive financial obligations and to prevent the
Community from degenerating into a free trade zone. The response from France’s
EEC partners was swift and critical. The British Foreign Office announced that it
wanted Spain and Portugal to join quickly, the German government declared that
it was still in favour of Spanish and Portuguese accession despite the difficulty of
the issues under negotiation, and the Belgian Foreign Ministry reiterated its belief
that the EEC is open to any democratic European country.66 In the end, despite
Giscard’s call and the concerns underlying it, the French government did not
interrupt the meeting the following day, and the Community confirmed the
continuation of negotiations at its Venice summit the following week. According
to the Financial Times, Giscard bowed to the fact that EC norms gave Spain a
‘right to membership, a right which the EEC has recognised’.67 The accession
treaty was finally signed in Madrid on 12 June 1985 and Spain joined the
community as a full member on 1 January 1986.
3. Reaffirming the opening to Turkey
Ankara’s quest to join the community was controversial among political elites and
voters in the member states during the period 1970–2005, partly because of
developments within the community and partly because of developments in
Turkey. However, the community’s membership norm and its earlier commitment
to Turkey’s membership eligibility both played a role in keeping the relationship
on track despite this controversy. At the same time, as in the Greek and Spanish
cases, those who were sceptical of Turkey’s European vocation in this period
attempted to use the community’s membership norm to legitimate their
preferences and proposals.
In 1970, the EEC and Turkey signed an Additional Protocol to govern progress
toward the customs union envisioned by the Ankara Treaty, which had recognized
Turkey’s membership eligibility seven years earlier. Relations between the parties
were damaged, however, by Turkey’s occupation of northern Cyprus following a
Greek-sponsored coup there in 1974 and then by the EEC’s start of accession
negotiations with Greece in 1976. The EEC was also deeply concerned by the
weakness of the Turkish economy in 1978 and sent several senior delegations to
discuss assistance. Significant assistance was eventually provided but not enough
to satisfy Ankara or to stabilize the Turkish economy.68
In September 1980, the Turkish army staged a coup d’état, the third of its kind
in two decades. In his first press conference, the chair of the new National Security
66 EUDC: 441.213 ER Enlarg negoc pt.1: EEC Enlargement: Europe no.2923, 7 June 1980.
67 A challenge to the EEC. Financial Times, 14 October 1980.
68 HAEU: EN.16 Association de la Turquie, fonds EN-1637, EN-1698, EN-2753.
Council declared that the coup aimed ‘to establish a civilian government in a
reasonable time [that would be] liberal, democratic, secular based on the rule of
law [and] would respect human rights and freedoms’ (Dagi 1996: 125). This
statement was likely crafted to hold off pressure from the EEC, whose liberal
democratic membership norm was clearly recognized in Turkey by this time. As
Ecevit himself had noted in 1976, ‘The European Economic Community attaches
great importance to modes of government. It is known that free democratic regimes
are a clear necessity within the EEC. There is no place in the EEC for those that do
not obey this condition’ (Döşemeci 2014: 195). But rhetoric aside, the community’s
membership norm does not seem to have constrained Turkey’s new regime: during
the following three years of military rule, according to Amnesty International,
‘650,000 people were detained and questioned, 250,000 were arrested, and count-
less others tortured, executed or disappeared’ (Döşemeci 2014: 210).
Given the EEC’s increasingly explicit commitment to liberal democracy and
human rights in this period, this brutal record would likely have been enough to
lead the community to suspend its association agreement with Turkey, as it had
done with Greece’s association agreement following the Greek coup d’état in 1967.
However, the community did not immediately have to consider such an action
because Ecevit had imposed a five-year freeze on Turkey’s implementation of the
association agreement in October 1978 out of frustration with the community’s
economic assistance at that time (Çalış 2017: 140). Instead, with the EEC–Turkey
relationship officially frozen, EEC and Council of Europe officials worked behind
the scenes to foster a return to democratic rule in Turkey (Dagi 1996). With
Ecevit’s five-year suspension of the EEC–Turkey association agreement due to
expire in 1983, officials in Ankara were surely aware that the community might
take stronger measures, including even a definitive cancellation of Turkey’s
The military government finally dissolved itself in December 1983 and
constitutional rule and civilian authority were reestablished in Turkey. Less than
four years later, on 14 April 1987, Turkey’s government submitted a formal
application for full membership in the European Community. It based this
application not on Article 28 of the Ankara Treaty, but on Article 237 of the
Treaty of Rome, which indicates that ‘Any European State may apply’, presumably
because the customs union foreseen by the Ankara Treaty had not yet come into
effect. Two weeks later, the Council took note of Turkey’s application and asked
the Commission for its opinion, as required by the treaty. The Commission was
unusually slow to issue its recommendation, which suggests that the Council was
not anxious to act on this dossier. But given Turkey’s status as a liberal democracy
whose membership eligibility the community had recognized long ago, it could
not wait forever.
In December 1989, the Commission re-confirmed Turkey’s eligibility for mem-
bership ‘in accordance with the political will shown at the time of the signing of
the Ankara Treaty’, but recommended against opening negotiations on accession,

for three inter-related reasons. First, the EC was working to implement the Single
European Act and to pursue economic, monetary, and eventually political union.
This agenda imposed far greater burdens on states seeking membership while
challenging the Community ‘to adapt its institutional operation in such a way as
to enable it to welcome within its fold new members without running the risk of
weakening its management and decision- making capacity’. As such, the
Commission advised that it would be ‘unwise’ for the Community to launch
accession negotiations with any applicant before 1993 ‘except in exceptional
circumstances’.
Furthermore, said the Commission, ‘any decision to open negotiations with a
particular country must be based on a strong conviction that a positive conclusion
is possible, indeed probable, within a reasonable period. This presupposes first
that the candidate country is considered capable, at the end of a transitional
period, of bearing all the constraints and disciplines now applying to Member
States . . .’ In this regard, the Commission was very sceptical about Turkey, given
its size and relative underdevelopment. Despite impressive growth and infra-
structure improvements, the Commission argued that the Turkish economy was
unlikely soon to overcome the major adjustment problems it faced, including
high levels of industrial protection, macro-economic imbalances, and low levels
of social protection. These conditions would make it difficult for Turkey to assume
the obligations of membership, impose a heavy burden on Community finances,
and risk large flows of labour that would exacerbate unemployment problems in
the current member states. Finally, but no less important, while recognizing that
recent political reforms in Turkey had ‘resulted in a parliamentary democracy
closer to Community models’, the Commission argued that the country’s human
rights situation and respect for minorities ‘have not yet reached the level required
in a democracy’.69
For all these reasons, ‘without casting doubt in any way on Turkey’s eligibility
to accede, [the Commission] proposed a set of measures aimed at helping the
country to modernize politically and economically’.70 On 3 February 1990, the
Council of Ministers approved the Commission’s Opinion, including both its
reaffirmation of Turkey’s membership eligibility and its conclusion that it was not
yet appropriate to open accession negotiations, and called upon the Commission
to make more proposals for deepening Turkey’s ties to Europe.71
The question of Turkey’s readiness for accession re-emerged in the late 1990s,
as the EU-Turkey customs union was completed and five post-Communist states
69 CVCE: Commission of the European Communities. Opinion on Turkey’s request for accession to the
Community. SEC (89) 2290 final/2, 20 December 1989.
70 AEI: Bulletin of the European Communities No 12 1989 Volume 22, p. 88.
71 European Commission. EU–Turkey Relations, Press Corner Memo/96/42, 25 April 1996.
of central and eastern Europe (Estonia, Poland, the Czech Republic, Hungary,
and Slovenia) moved closer to EU membership. Despite the Council’s
reaffirmation of Turkey’s membership eligibility in 1990, the issue was increasingly
controversial. In early 1997, Dutch Foreign Minister Hans van Mierlo told the
European Parliament, it was ‘time for us in Europe to be honest’ about Turkey:
‘There is a problem of a large Muslim state. Do we want that in Europe? It is an
unspoken question.’72 Wilfried Martens, head of the Christian Democrats in the
European Parliament, commented, ‘The EU is in the process of building a
civilization in which Turkey has no place.’73
The fact that van Mierlo considered this Europe’s ‘unspoken question’ is strong
evidence that the liberal democratic membership norm remained firmly
entrenched within the Council in this period. Explicitly cultural arguments were
simply not acceptable in official deliberations: regardless of their actual
motivations, opponents of Turkey would not contradict the prevailing
membership norm. When the foreign ministers of the five largest member states
(France, Germany, Italy, Spain, the UK) gathered in Rome for a meeting with
their Turkish counterpart, they reassured her that Turkey has ‘every right’ to
aspire to full membership and that there are ‘no obstacles in principle’ to its
accession. However, they added, Turkey needed to make progress on human
rights, the treatment of Kurds in the country, and on relations with Greece.74
This entrapment was clear after Christian Democrats in the European
parliament (the EPP group) resolved that Turkey’s religious and cultural traits
were ‘barriers’ to European integration and that Turkey had no prospect of
membership in the short, medium, or long term. Despite his comment to the
European Parliament just a few weeks earlier, van Mierlo, then holding the EU’s
rotating presidency, had to distance himself from the EPP position: ‘I do not
agree with the statement of the Christian Democrats.’75 At an informal meeting a
couple of weeks later, all EU foreign ministers backed van Mierlo, saying that
Turkey’s Muslim identity was not an obstacle to membership and that its progress
would be judged by the same criteria used for other European states. Even
German Foreign Minister Klaus Kinkel, whose government was the most
sceptical of Turkish accession in the Council at the time, could only add, ‘Turkey
should be kept on track to the EU [but] this is a long track.’76
At the Luxembourg summit in December 1997, the EU once again confirmed
Turkey’s membership eligibility while refusing to start the accession negotiations
that Ankara wanted. In particular, the Fifteen included Turkey (along with Cyprus
and ten states of central and eastern Europe) on a list of ‘European States’
72 Turkey Finds European Door Slow to Open. The New York Times, 23 February 1997.
73 Turkey and Europe: Just not our sort. The Economist, 13 March 1997.
74 Europe: The Truth about Turkey? It’s Not Wanted by the EU. RFERL.org, 9 February 1997.
75 EU term president supports Turkey’s bid. Hurriyet Daily News, 10 March 1997.
76 EU supports Turkey against its own Christian Democrats. Hurriyet Daily News, 21 March 1997.
committed to ‘the principles on which the European Union is founded’ and thus
recognized ‘Turkey’s eligibility for accession to the European Union . . . on the
basis of the same criteria as the other applicant States’. But the summit conclusions
made clear that the EU would henceforth be very attentive to applicant states’
implementation of their normative commitments. In this regard, the continued
use of torture in Turkish prisons and the Turkish military’s behind the scenes
manoeuvring to force the resignation of Prime Minister Necmettin Erbakan in
February 1997 could not have helped.77 As a result, the Fifteen decided at
Luxembourg not to launch accession negotiations with Turkey: ‘While the
political and economic conditions allowing accession negotiations to be envisaged
are not satisfied, the European Council considers that it is nevertheless important
for a strategy to be drawn up to prepare Turkey for accession by bringing it closer
to the European Union in every field.’
The controversy surrounding the prospect of Turkish accession persisted
nonetheless, as did the entrapment effect of the community’s liberal democratic
membership norm. In December 1999, for example, French MEP and former
minister Alain Lamassoure proposed a way to block Turkey’s accession without
making the cultural argument, calling on the upcoming European Council
meeting in Helsinki to declare unambiguously that the ‘European’ character of a
country, and thus its potential eligibility for EU membership, was determined by
geography, not religion or culture, and that such a geographic limit would
necessarily exclude Turkey.78 But such a strict geographic limit on membership
eligibility was inconsistent with prevailing membership norms.
Instead, the EU’s leaders reiterated at Helsinki that ‘Turkey is a candidate State
destined to join the Union on the basis of the same criteria as applied to the other
candidate States’ and welcomed ‘recent positive developments in Turkey as noted
in the Commission’s progress report, as well as its intention to continue its
reforms towards complying with the Copenhagen criteria’.79 In other words, as
long as Turkey remained a liberal democracy committed formally to respect
human rights and ensure the rule of law, its eligibility for membership would not
be questioned. Instead, Turkey’s progress toward membership would depend
upon its implementation of the Copenhagen criteria.
Over the next two years, the Turkish government adopted a series of political
reforms designed to satisfy the Copenhagen criteria, such as removing military
judges from civilian courts, lengthening prison terms for those convicted of
torture, making it harder for prosecutors to close down political parties, and
77 Briefing: The Continued Use of Torture in Turkey, Washington, DC: US Commission on Security
and Cooperation in Europe, 8 April 1997; Turkish Generals Raise Pressure on Premier, The New York
Times, 13 June 1997.
78 Jusqu’où?: le débat interdit, Le Monde, 9 décembre 1999.
79 AEI: Helsinki European Council, 10–11 December 1999, Presidency Conclusions.
putting the execution of Kurdish rebel leader Abdullah Ocalan on hold.80 These
reforms were not extensive or deep enough to convince the Commission to rec-
ommend the start of accession negotiations, but they helped convince the EU
foreign ministers in December 2000 to approve an Accession Partnership
outlining the conditions for accession negotiations with Turkey.81 Council
conclusions from Feira in June 2000, Nice in December 2000, and Göteberg in
June 2001 showed that the Fifteen were noticing Ankara’s efforts. At the Laeken
summit in December 2001, they confirmed that Turkey’s progress towards
complying with the political criteria had ‘brought forward the prospect of the
opening of negotiations with Turkey’.82
By the autumn of 2002, Turkey had moved even closer to meeting EU
membership norms. Since Bülent Ecevit’s return as prime minister in 1999, his
government had ‘rewritten one-third of the Turkish constitution by adopting
international human rights laws, ending capital punishment, expanding women’s
rights, discouraging torture, and improving prison conditions. New laws curtailed
existing restrictions on freedom of expression, civil society, and the media, as well
as diminishing the Turkish military’s long- standing dominance of politics’
(Pope 2012: 46). These developments, combined with the strong support that
Germans of Turkish origin gave to the SPD in the Bundestag elections of
September 2002, made Gerhard Schröder’s government more sympathetic to
Turkey’s desire to start accession negotiations. In October, Sweden’s government
signalled that it too had changed position and was inclined to support the opening
of talks with Turkey (Ludlow 2002).
Also important, however, was the overwhelming victory of the Islamist but
strongly pro-reform Justice and Development Party (AKP) in Turkey’s November
2002 elections, followed by AKP leader Recep Tayyip Erdoğan’s goodwill tour of
European capitals, where he reiterated his commitment to further political reform
and EU accession.83 In this context, the EU’s refusal to negotiate accession with
Turkey contrasted sharply with its embrace of central and eastern European states
including Romania, whose performance on political rights and civil liberties was
only marginally better than that of Turkey.84 Austria and the Netherlands, the two
member states most opposed to Turkish accession, still retained a veto but were
finding it increasingly difficult to resist the opening of negotiations. As the Danish
presidency of the EU prepared for the December 2002 summit, the question was
therefore not whether the Fifteen would make a concession on negotiations but
what its terms would be (Ludlow 2002).
80 Why are we waiting? Economist.com, 8 June 2000.

81 EU adopts Accession Partnership with Turkey, Euractiv.com, 5 December 2000.
82 AEI: Laeken European Council. 14–15 December 2001. Presidency conclusions and annexes.
83 Turkey all out efforts, Hurriyet Daily News, 28 November 2002.
84 According to Freedom House’s Freedom in the World index for 2002, Turkey rated a 3 for political
rights and a 4 for civil liberties while Romania rated a slightly better 2 for both.
In the end, the Copenhagen summit concluded with the EU’s first, albeit
conditional, mention of a date to start negotiations: ‘The Union encourages
Turkey to pursue energetically its reform process. If the European Council in
December 2004, on the basis of a report from the Commission, decides that
Turkey fulfils the Copenhagen political criteria, the European Union will open
accession negotiations with Turkey without delay.’ In other words, the EU gave
Turkey what Erdoğan had just recently refused: ‘a date for a date’.85 This was
another major step in Turkey’s decades-long pursuit of EU membership.
Although Erdoğan quickly reversed himself and declared the Copenhagen
result ‘an historic victory’, the AKP government continued to pursue political
reforms.86 Just before the Thessaloniki summit in June 2003, the government
(headed since March by Erdoğan as prime minister) announced an ambitious
package of further constitutional reforms to secure freedoms of expression and
religion, allow radio and TV broadcasts in Turkish, and expand women’s rights.87
The package was well received in the EU but questions persisted about the
implementation of the reforms, which was clearly central to any evaluation of
Turkey’s position.
A wave of terrorist bombings of British and Jewish targets in Istanbul in
November 2003 revived suspicions about Turkey across the EU and threatened to
derail Ankara’s campaign, but here too references to membership norms kept the
debate within bounds. In Germany, vice-chairman of the CDU Wolfgang Bosbach
declared that if Turkey were to join the EU, ‘it would import the terror problem to
Europe’, a view quickly echoed by Bavaria’s justice minister Gunther Beckstein. In
response, fellow CDU heavyweight Elmer Brok, chair of the European
Parliament’s foreign affairs committee, criticized his colleagues for linking
Turkish membership to the attacks, adding ‘I am against Turkish membership
because the political criteria are not fulfilled.’ Yet after Germany’s justice minister
Otto Schily replied that the bombings had shown the need to give Ankara ‘a clear
perspective’ on membership, the terrorism issue faded from public discussion.88
The question of future enlargement, and particularly the status of Turkey, was
very high on the political agenda in 2004, with many conservatives and other
opposition politicians outspoken in their opposition to Turkish accession. In
Germany, both the CDU and CSU parties made opposition to Turkish accession a
prominent plank in their manifestos for the June 2004 European parliamentary
elections. On a visit to Ankara that February, Merkl expressed her view that the
EU was ‘overburdened’ and proposed that the EU should negotiate a ‘privileged
85 Turkey demands EU talks date, BBC News Online, 5 December 2002.
86 Turkey and the EU: Well, thank you, The Economist, 3 January 2003.
87 Avant le sommet de Thessalonique, la Turquie accélère les réformes, Le Monde, 13 juin 2003.
88 Bombings bring debate on Turkey’s EU future, Financial Times, 21 November 2003; German row
over Turkish EU membership, EUobserver.com, 25 November 2003.
partnership’ with Turkey instead of full membership.89 Positioning himself for the
upcoming presidential election in France, Nicholas Sarkozy departed from the
position of the government in which he was economy minister, declaring his
belief that Turkey is not a European state, his opposition to Turkish accession,
and his support for a referendum on the matter.90
In March 2004, EU commissioner Frits Bolkestein published a collection of
interviews entitled The Limits of Europe whose cover showed him standing in
front of a map of Europe with his body blocking the countries east of Croatia and
the Czech Republic. In the book, Bolkestein praised eastern enlargement as ‘an
historic, a political, a moral obligation’ but warned against ‘a Europe without
frontiers, an economic ideal but a political mirage’. As he saw it, ‘A geographical
limitation of the Union is therefore logical. In the south, the Mediterranean Sea
forms a natural divide between Europe and Africa. In the east, there is a geo-
political need for a buffer zone between the EU and Russia, which might be
formed by the countries that do not belong to either block. A similar buffer to the
south-west would also be advantageous, in order to cushion the Union against
Syria, Iran and Iraq’ (Bolkestein 2004: 22–4). In a speech at Leiden University,
Bolkestein warned that Turkey’s accession would cause the ‘implosion’ of
Europe.91
As the December 2004 summit approached, opponents of Turkish accession
began to make cultural arguments. The same week as Bolkestein’s Leiden speech,
a letter on Turkey written by Austrian commissioner Franz Fischler to his
Commission colleagues was leaked to the press. In addition to questioning the
financial consequences of Turkish accession, Fischler described Turkey as ‘more
oriental than European’ with only a superficial commitment to secularism.92
Cardinal Josef Ratzinger, a very senior figure in the Vatican (and later pope),
declared, ‘Europe is a cultural and not a geographic continent. Its culture gives it a
common identity. In this sense, Turkey always represented another continent
through history, in permanent contrast with Europe.’93 Just before the summit,
the leader of Belgium’s parliamentary opposition Herman van Rompuy told his
colleagues, ‘Turkey is not part of Europe and will never be part of Europe.’
Europe’s fundamental values are also Christian values, he argued, and thus would
89 Merkl’s Message to Turkey—‘No’, Deutsche Welle, 16 February 2004; see also Angela Merkl,
Türkei: Partnerschaft statt EU-Mitgliedschaft, Die Welt, 16 oktober 2004.
90 Sarkozy chatouille Chirac sur l’adhésion de la Turquie, Libération, 28 septembre 2004; Turquie:
Sarkozy contredit Chirac, Le Nouvel Observateur, 20 décembre 2004.
91 L’accession de la Turquie pourrait provoquer ‘l’implosion’ de l’Europe, selon Bolkestein, EurActiv.
fr, 7 septembre 2004; EU Commissioner says Turkey’s entry will end European integration, AsiaNews.
it, 8 September 2004.
92 Fischler challenges EU plans for Turkey, EurActiv.com, 10 September 2004; European Union
struggles to contain Turkey accession debate, RFE/RL, 11 September 2004.
93 L’accession de la Turquie pourait provoquer ‘l’implosion’ de l’Europe, selon Bolkestein, EurActiv.
fr, 7 septembre 2004; EU Commissioner says Turkey’s entry will end European integration, AsiaNews.
it, 8 September 2004.
be threatened by the accession of a large, predominantly Muslim country.94 At

about the same time, Valéry Giscard d’Estaing rejoined the debate, declaring that
European identity is founded on ‘the cultural contributions of ancient Greece and
Rome, the religious heritage pervading European life, the creative enthusiasm of
the renaissance, the philosophy of the Age of the Enlightenment and the contri-
butions of rational and scientific thought. Turkey shares none of these.’95
But Merkl, Sarkozy, van Rompuy, and the others were all in opposition during
this period, not in government, so they were not involved in the Council’s
deliberations and their outspoken views on the limits of Europe cannot be taken
as indicators of a change in EU membership norms. Bolkestein and Giscard did
hold senior positions in the Commission and Convention, respectively, but there
is little evidence that their views were widely shared in those institutions. As one
Commission official said in response to Fischler’s letter, ‘Religion is not a
condition for membership. The important thing is whether you share the values
of democracy and tolerance.’96
Two weeks later, after completing his review of Turkey’s compliance with the
Copenhagen criteria, EU enlargement commissioner Günter Verheugen praised
Ankara’s reform efforts and commented, ‘No remaining outstanding obstacles
remained on the table.’97 In other words, Turkey had satisfied the EU’s membership
norms and should therefore be allowed to begin negotiations. On the eve of the
December 2004 summit, Bolkestein admitted that the EU was entrapped: ‘If a
secret vote was held, in the commission as well as among government leaders,
then only a small minority would be in favor of Turkish accession. We are in a
harness and don’t know how to get out of it.’98
This does not mean that it would be easy to reach consensus among the
member states. While Belgium, Germany, Greece, Italy, Portugal, Spain, Sweden,
and the UK supported negotiations, a number of other member states remained
deeply sceptical about Turkish accession. The Austrian, Danish, Dutch, and
French publics were strongly opposed to Turkish accession and Austrian
chancellor Wolfgang Schüssel had long favoured some sort of privileged
partnership. In the background, a Eurobarometer poll conducted in October–
November 2004 had revealed that over a third of citizens in the 25 member states
were opposed to any further enlargement, with the numbers rising to 51 per cent
in France, 57 per cent in Germany, and 62 per cent in Austria.99
94 Van Rompuy: ‘Turkey will never be part of Europe’, EUobserver.com, 18 November 2009; Pers
boos om ‘anti-Turkse benoeming’, De Standaard, 21 november 2009.
95 A better European bridge to Turkey, Financial Times, 24 November 2004.
96 Fischler urges EU to rethink Turkish entry, Financial Times, 9 September 2004.
97 Turkey moves closer to EU after retreat on adultery law, The Guardian, 24 September 2004.
98 Bolkestein: Turkey could complete talks in five years, Hurriyet Daily News, 7 December 2004.
99 Over a third of EU citizens against further enlargement, EUobserver.com, 10 December 2004.
The first hurdle to opening negotiations with Turkey, then, was the Dutch
presidency of the EU, which would be responsible for managing the December
summit. As discussed above, recent events in the Netherlands had clearly
demonstrated the Dutch public’s deep divide on issues related to multiculturalism
and immigration, and the question of Turkish accession touched on both of these.
When the Dutch cabinet first discussed the issue in February 2004, only two
senior ministers were strongly supportive of an opening to Turkey, while others
were deeply opposed (Ludlow 2005: 10). To complicate matters further, the Dutch
government had recently become increasingly pragmatic about using the EU to
achieve national aims and was thus divided internally over whether its presidency
should be focused on bridge-building, as it had been in 1997, or on agenda-
setting, as it had been in 1991 (Keulen 2004). But given divisions among the
member states, the odds of agreeing to start talks with Turkey would be greatly
diminished if the Dutch presidency were not supportive.
In this context, the strength of the prevailing membership norm appears to
have been a critical determinant of the Dutch presidency’s agenda. By the time
that Prime Minister Jan Peter Balkenende addressed the European Parliament in
July, his government’s approach to negotiations with Turkey was clear: ‘The
decision must be arrived at honestly, under the ground rules to which we
previously, in 2002, firmly committed ourselves. That means strict application of
the criteria laid down, but without inventing any new criteria.’ To resist Turkey
because of its predominant religion, said Balkenende, would be ‘not consistent
with Europe’s shared values.’ After the Commission recommended on 6 November
that the EU open negotiations with Turkey, the Dutch presidency set out to
overcome opposition in the Council (Ludlow 2005: 11).
The principle obstacle was Austrian Chancellor Wolfgang Schüssel’s strong
preference that the EU offer Turkey only a privileged partnership and, barring
that, his insistence that the conclusions must incorporate language agreed by the
European People’s Party the previous month: ‘Maintaining the Union’s capacity to
act as well as acting in accordance with its capacity to absorb new members is as
important as the fulfillment of the accession criteria by Turkey’ (Ludlow 2005:
28). With nearly two-thirds of Austrians opposed to Turkish accession and only
one-fifth in favour, according to Eurobarometer, Schüssel had good reason to
refuse any compromise.100 It is thus not surprising that during the summit delib-
erations, Schüssel ‘tested his colleagues’ patience to the limits’. As the BBC
described it, ‘Austria was a sticking point’ (Ludlow 2005: 28).
In the end, the European Council concluded in December 2004, ‘Turkey
sufficiently fulfills the Copenhagen criteria to open accession negotiations’ and
agreed to do so on 3 October 2005. On the other hand, the Council imposed a
100 In 1683 Turkey was the invader. In 2004 much of Europe still sees it that way, The Guardian, 22
September 2004.
number of important conditions, including a provision that ‘accession

negotiations are an open-ended process; the outcome cannot be guaranteed
beforehand’ and the possibility that Turkey’s accession might involve ‘[l]ong
transitional periods, derogations, specific arrangements or permanent
safeguards . . . for areas such as freedom of movement of persons, structural
policies or agriculture’ (Ludlow 2005: 11). In the meantime, Ankara had to accept
a protocol extending the EU-Turkey customs union to the ten new member states
including Cyprus, which Turkey had theretofore refused to recognize.101
Notwithstanding these conditions, it is noteworthy that Schüssel achieved
neither of his core objectives prior to the December summit, but still did not
exercise Austria’s right to veto the conclusions. If he had insisted on his position,
the presidency would have had to compromise. Instead, the fit between Turkey’s
candidacy and EU membership norms appears to have forced Schüssel’s hand:
even though several other member states were deeply sceptical about Turkey, he
simply could not muster support for a policy that so clearly contradicted the
membership norms prevailing within the community. Ten months later, Schüssel
made another last-minute attempt to convince his partners, many of whom
remained sceptical about Turkish accession, to reverse the 2004 commitment and
offer an alternative to full membership.102
In the end, though, Schüssel failed again to gain support for his position and
the EU agreed on 3 October 2005 to open accession talks with Ankara. Austria’s
concession was softened by an EU decision to open negotiations with Croatia,
which Vienna had long favoured, but media reports of a simple quid pro quo are
seriously misleading: the green light for Croatia was based principally on a
declaration by the international war crimes tribunal in The Hague that Croatia
was cooperating fully with its work, as required by the EU.103 In any case, given
the EU’s 2004 conclusion that Turkey’s reforms had satisfied the liberal democratic
membership norm, and the absence of new evidence to the contrary, the 2005
decision to open talks was nearly a pre-ordained conclusion.
4. ‘Neither open nor closed’ to Ukraine
Ukraine became independent upon the break-up of the Soviet Union in 1991,
and the government in Kiev was quick to express interest in joining the
EU. However, the liberal democratic membership norm prevailing in the
community at that time proved to be a major obstacle to the EU’s recognition of
101 CVCE: Conclusions of the Brussels European Council (16 and 17 December 2004).
102 Austria repeats Turkey objections, BBC News Online, 1 October 2005.
103 Austria repeats Turkey objections, BBC News Online, 1 October 2005; Europe embraces Turkey
as diplomatic deadlock is broken, Guardian, 4 October 2005.
Ukraine as a state eligible for membership. Other factors also shaped EU decision-
making, including the state of Ukraine’s economy and the risk of antagonizing
Russia, but without convincing liberal democratic credentials, Kiev could not
take advantage of the normative entrapment dynamics that had helped other
aspirant and applicant states to overcome the preferences of reluctant
member states.
Despite Ukraine’s rampant political corruption and economic decline in the
years following independence, the EU signed a Partnership and Cooperation
Agreement (PCA) with President Leonid Kravchuk in 1994. The pact was
designed to promote closer political relations, expanded trade and technical
cooperation between the parties and thus to help Ukraine consolidate its new
democracy and complete the transition to a market economy. The PCA was
intended to run for an initial period of 10 years and could be renewed thereafter
but unlike some of the community’s earlier association agreements, it did not
mention accession as an ultimate possibility.
Ukraine’s new president Leonid Kuchma, who took office one month after the
signing of the PCA, had more ambitious goals. In 1996 Kuchma declared EU
membership as a strategic objective for his country (Wolczuk 2003). Two years
later, he issued a formal presidential decree that the ‘national interests of Ukraine
require . . . full-fledged EU membership’. Crucially, however, the decree stated that
the ‘involvement of Ukraine into the European political, economic and legal
space and acquiring on this basis a status of the EU associated member’ was the
country’s ‘main foreign policy priority in the middle-term perspective’.104 So
while Kuchma was looking ahead to full membership, he was not proposing to
start accession negotiations any time soon. On the other hand, he was clearly
hoping to gain EU confirmation of Ukraine’s eligibility for EU membership.
The problem was that regardless of Kuchma’s ambitions, the country was still
plagued by rampant corruption, electoral fraud, political interference in the
judiciary, and significant limits on freedom of the press and freedom of
association, among other abuses of power. In its 1999 survey, Freedom House
rated Ukraine ‘partly free’ and warned of a ‘downward trend arrow due to
increased government pressure on the independent media, presidential elections
in October–November which were not free and fair, and attempts by the executive
after the elections to increase presidential powers at the expense of parliament’.105
As such, Ukraine was not heading in the right direction on the EU’s ‘Copenhagen
criteria’ principles of democracy, the rule of law, and respect for human rights. In
addition, despite a macro-economic turnaround under Kuchma, the absence of
104 Decree by the President of Ukraine: On the Approval of the Strategy of Ukraine’s Integration to
the European Union. No. 615/98, 11 June 1998. http://www.sdfm.gov.ua/content/file/site_
docs/2007/16.05.06/decree.htm
105 Freedom in the World 1999. Washington, DC: Freedom House, 1999.
structural reforms meant that Ukraine was still far from having a functioning
market economy. As one observer put it, Ukraine’s political leaders ‘acted as if
they could achieve integration by declaration, or simply by joining and
participating in international organizational and political clubs rather than by
undertaking concrete structural changes’ (Wolczuk 2003: 2).
EU statements on Ukraine in this period highlighted this disconnect between
Kiev’s stated ambitions and the reality on the ground. For example, in their joint
statement following the EU–Ukraine summit in July 1999, Ukraine’s President
Kuchma, President of the European Council Paavo Lipponen, and European
Commissioner Hans van den Broek announced that they had ‘discussed further
steps to use to the full the considerable potential for further rapprochement
between the European Union and Ukraine [and] exchanged information on
Ukraine’s strategy of integration with the EU’, but they made no mention of
moving beyond the previous year’s PCA, much less of possible accession.106
Five months later, at the European Council’s December 1999 Helsinki summit,
rather than endorsing Ukraine’s ultimate eligibility for membership, as Kiev had
hoped, the EU said only that it ‘underlines the importance it attaches to the
emergence of a democratic, stable, open, and economically successful Ukraine as
a prominent actor in the new Europe [and] takes account of Ukraine’s European
aspirations and pro- European choice’. The contrast to the same summit’s
declaration that ‘Turkey is a candidate State destined to join the Union on the
basis of the same criteria as applied to the other candidate States’ could not have
been clearer.107 By recognizing that Ukraine is ‘in Europe’ but refusing to give it
the same membership perspective that it was giving to other neighbouring states
also in need of reform, the European Council made clear that it did not consider
Ukraine eligible for EU membership.
Given the aforementioned conditions in Ukraine at the time, this decision was
fully consistent with the community’s prevailing norm that limited membership
eligibility to liberal democracies. There were other motives at play, of course.
Most notably, the French and German governments were concerned that offering
Ukraine a membership perspective would antagonize Russia, with whom they
wanted to maintain good relations. This combination of factors may explain
European Commission President Romano Prodi’s 2002 comment that Ukraine
had as much chance of EU membership as New Zealand.108 It was not long,
though, before developments in Ukraine put the country’s European orientation
back on the EU’s agenda.
106 European Commission. Joint Press Communique, EU-Ukraine Summit Kiev, 23 July 1999,
C/99/244.
107 CVCE: Conclusions of the Helsinki European Council (10 and 11 December 1999).
108 Russia, ex-Soviet republics will not join EU, says Prodi, The Irish Times, 28 November 2002.
In Ukraine’s 2004 presidential election, the principal candidates were incumbent

Prime Minister Viktor Yanukovych, who was supported by Russia, and opposition
leader Viktor Yuschenko, who favoured a westward orientation for the country.
Yanukovych was officially declared the winner after the second, run-off stage of
the election, but strong evidence of electoral fraud confirmed by international
observers caused thousands of Ukrainians to take to the streets in protest. Behind
the scenes, senior EU officials including the presidents of Lithuania and Poland
and the EU’s high representative for foreign and security policy participated in
roundtable talks with Kuchma, Yanukovych, and Yushchenko. The Ukrainian
Supreme Court then annulled the run-off results and ordered a re-run of the
second round, which resulted in a convincing win for Yushchenko on 26
December. This ‘Orange Revolution’ seemed to indicate that Ukraine was now
firmly committed to EU values (Åslund and McFaul 2006).
These dramatic developments reframed the Ukrainian eligibility question for
EU member states. Before the election at least 10 member states favoured
recognizing Ukraine’s membership eligibility, but the ambiguities of Ukrainian
politics meant that more sceptical governments (including France, Germany, and
The Netherlands, among others) faced little pressure to follow their line
(Gromadzki el al. 2006: 13–14). The incoming government promised to eliminate
the ambiguity: ‘For Mr Kuchma “European ambitions” were a tactical device with
which he regulated his multi-vectoral policy of balancing between Russia and the
West, lacking the political will to fulfil criteria of EU membership. Nothing could
be further from the truth regarding the government of Viktor Yushchenko, which
regards relations with Russia as strategically important but is adamant about the
European destination and future of Ukraine’ (Gromadzki et al. 2006: 24–5).
The effects of this new reality were quickly evident in the position of the
European Parliament. On the eve of the first round of elections in late October,
the EP had approved a resolution recognizing Ukraine’s European aspirations,
praising its people’s struggle for democracy, the rule of law and human rights, and
pledging support for ‘their country’s assumption of its rightful place in the
European community of democratic nations’ while avoiding any direct reference
to membership.109 Following the Orange Revolution, MEPs adopted a new
resolution celebrating recent events in Ukraine and alluding to the country’s
legibility for EU membership under ‘the provisions of Article 49 of the Treaty on
European Union, which state that EU membership is an option for all European
countries that satisfy the relevant conditions and obligations . . . ’110 Although still
cautious, the change in wording showed MEPs’ increased willingness to challenge
member states resisting any opening to the eventuality of Ukrainian accession.
109 European Parliament, 28 October 2004, resolution RC-B6-0095/2004.

110 European Parliament, 13 January 2005, resolution RC-B6-0038/2005.
In the brief interval between Viktor Yuschenko’s inauguration and his first visit
to Brussels as Ukraine’s president, politicians in EU member states scrambled to
adjust their positions, at least rhetorically. In Germany, opposition Christian
Democratic foreign policy spokesman Wolfgang Schäuble declared, ‘The
European Union is now called upon to put its promises into practice—promises it
has made in recent weeks.’ Fellow CDU member Friedbert Pflueger went further,
saying that ‘Ukraine is a European country by any standards’ and deserved an EU
perspective, at least comparable to that given to Turkey.111
In contrast, the leaders of France and Germany remained reluctant. Jacques
Chirac’s congratulatory letter to Yushchenko treaded carefully on the issue of EU-
eligibility: ‘Rest assured that France has weighed the importance of the December
26 election, both for Ukraine and for Europe. I hope this election will continue to
reinforce the ties binding all the people of our continent.’112 Gerhard Schroeder
told a meeting with the Bundestag’s European Affairs Committee, ‘At present it’s
about a partnership—not accession.’113
The European Commission struggled to encapsulate, much less lead, the EU’s
response to the new reality emerging in Ukraine. Commissioner for external
relations Benita Ferrero-Waldner declared that the community’s ‘door is neither
open nor closed’. However ambiguous, the plausibility of this statement was
bolstered when the Commission member responsible for communication,
Margot Wallström, admitted, ‘We discussed this a lot in the Commission and
agree that eventually it is a realistic vision for the future that Ukraine should join
[the EU] without today going out and saying we have a concrete date or an offer.’
The Commission’s spokesperson responded with a blanket denial: ‘There has
been no discussion on whether Ukraine is a European country or not.’114
Around the same time, the new government in Kiev sent senior diplomat Oleh
Rybachuk to Brussels to press for EU recognition of Ukraine’s membership
eligibility. Much to his frustration, he was ‘shuffled from meeting to meeting with
MEPs, officials in the European Commission and the EU Council, and with
diplomats in EU countries’ embassies’. They were willing to discuss EU
‘benchmarks’ and ‘criteria’, but were not yet ready to give a positive signal on the
question of membership eligibility.115
Nonetheless, given the EU’s prevailing norm that liberal democracies in Europe
were eligible for membership, it was becoming increasingly difficult for the
Commission and sceptical member states to resist any opening to the eventuality
of Ukrainian accession. After Yushchenko’s January visit to Brussels, Commission
president Jose Manuel Barroso suggested that the issue of Ukraine’s membership
111 German Politicians Demand Closer Links to Ukraine, Deutsche Welle, 27 December 2004.
112 Chirac applauds Yushchenko’s win in Ukraine, Expatica.com, 7 January 2005.
113 Schroeder cool on EU membership for Ukraine, Expatica.com, 27 January 2005.
114 Don’t know much about geography, Financial Times, 25 January 2005.
115 EU and Ukraine: What went wrong? EUobserver.com, 25 November 2013.
eligibility was more about when than whether: ‘I believe that the future of Ukraine
is in Europe, but now the journey is not for European Union membership.’
Instead, he suggested that EU relations with Ukraine were covered adequately for
the time being by the Union’s Neighborhood policy.116 Eight months later, after a
meeting in Brussels with Kiev’s new prime minister Yuri Yekhanurov, Barroso
took another rhetorical step in Ukraine’s direction: ‘Our door remains open. The
future of Ukraine is in Europe.’117
On the other hand, and despite Barroso’s statements, the Commission could
not stray far beyond the position of the member states. As the year wore on, long-
standing French and German desires not to antagonize Russia, plus a sense that
the Union needed to digest the ‘big bang’ inclusion of 10 new member states,
were complemented by new, domestic reasons to be suspicious of further
enlargement. In the late spring, voters in French and then Dutch referenda
rejected the EU Constitutional Treaty. Opinion polls showed that anti-
enlargement sentiment, which many voters linked to an anti-immigration stance,
had been a powerful motive in both countries. Other EU governments took note.
Meanwhile, Angela Merkel, known for her outspoken opposition to Turkish
accession, was a rising star in German politics and seemed likely to win the next
election. In short, consensus on the EU’s liberal democratic membership norm
was fast deteriorating (see details in Chapter 3). Notwithstanding the new
opportunity created by Ukraine’s Orange Revolution, this was simply not a
propitious time to press for inclusion in the EU’s enlargement queue.
On 9 November 2005, the European Commission released its annual strategy
paper on enlargement, which would have been prepared with a careful eye on
member states’ preferences. As commissioner Olli Rehn summarized it, the
strategy was very conservative: ‘We need to consolidate our enlargement agenda
but be cautious with new commitments.’ In effect, this meant that achieving the
accession of Croatia and promoting the progress of Turkey and candidate states
in the Western Balkans would occupy the EU’s enlargement agenda for the
foreseeable future, leaving little room for Ukraine and others waiting at the
door.118 Once again, the EU had opted neither to offer Ukraine a clear membership
perspective nor to rule out any such possibility.
5. Conclusions
In May 2001, the Economist magazine published a collection of articles on

European Union enlargement headlined ‘The limits of Europe’ and asked ‘Where
116 European Commission Press Release, SPEECH/05/84, 9 February 2005.

117 Ukraine gets enlargement wink, EUobserver.com, 6 October 2005.
118 Blow to Kiev as Brussels closes door to further enlargement, EUobserver.com, 9 November 2005.
will it all end?’119 In that issue, a German scholar argued that enlargement ‘has
proved a really excellent way of persuading countries to behave in the ways that
you want. But sooner or later the EU has to be able to draw a line and work out
new forms of relationships with countries that are not going to join the Union.’
Yet as this chapter has demonstrated, the EU’s collective commitment to a liberal
democratic membership norm between 1970 and 2005 shaped deliberations
among the member states and the conclusions they reached as community,
preventing them from drawing such a definitive line.
The expansion of EU membership norms around 1970, which added liberal
democratic principles of human rights and the rule of law to the pre-existing
focus on parliamentary democracy, exerted a powerful positive effect on Greece’s,
Spain’s, and Turkey’s pursuit of accession in the two and a half decades that
followed. In the cases of Greece and Spain, the community’s self-definition as a
community of liberal democracies trapped the member states into accepting
Athens’ and Madrid’s quest for rapid accession following the collapse of
dictatorship in the mid- 1970s, despite the Commission’s deep reservations
regarding both countries’ readiness.
The power of membership norms is even more evident in the case of Turkey, a
country whose potential accession was anathema to several EU governments.
Multiple pressures shaped EU deliberations on Turkey in this period, but the
critical decision to open accession negotiations is best explained by the strength
of the EU norm that liberal democratic European states are entitled to
membership, which Erdoğan could plausibly claim applied to Turkey and
Schüssel felt obligated to accept regardless of his preferences. The irony, as noted
many years later by Hungary’s Prime Minister Viktor Orban, was that the EU
effectively ‘told a candidate country that we want to negotiate while significant
countries said that we will never admit that country into the European Union’.120
In contrast, EU decision-making on Ukraine following the Orange Revolution
of 2004 exhibited a weak form of normative entrapment—just strong enough to
get EU institutions to acknowledge Ukraine as a European state (which Article 49
sets as a precondition for membership eligibility) but not strong enough to gain
explicit recognition that Ukraine was on track for membership. The difference
between the EU’s official, albeit reluctant openness to Turkey and its refusal to
acknowledge Ukraine’s membership eligibility cannot be explained by geography:
Ukrainian territory lies entirely in the zone that geographers call Europe while
more than 90 per cent of Turkey lies in Asia. Nor can it be explained by public
opinion or domestic politics: polls in numerous member states showed far more
public sympathy for Ukraine than for Turkey.
119 The Economist, 17 May 2001.

120 Hungary’s Orban says EU ‘insincere’ with Turkey on accession, APnews.com, 8 October 2018.
Instead, the difference seems to be due to two factors: the relative inconsist-
ency of the two states’ efforts to establish and consolidate practices that fit the
EU’s prevailing membership norm, and the degree of prior commitment by the
EU itself. In the case of Turkey, its governments worked consistently starting in
the 1990s to adopt and implement reforms promoting democracy, the rule of law,
and human rights, and the EU itself had recognized as early as 1959 (see
Chapter 5) that the country was eligible for accession. The Ukrainian govern-
ment’s post-1991 pursuit of an explicit membership perspective unfolded in an
identical EU membership norms framework but was undermined by the govern-
ment’s failure to rid itself of Soviet-era tendencies to politicize the judiciary,
intimidate political opponents, and rig elections, and the fact that the EU was not
yet on record regarding the country’s membership perspective. As a result,
Turkey’s campaign for EU-eligibility was far more successful in this period than
that of Ukraine.
In sum, the liberal democratic membership norm prevailed among the
member states for three and a half decades, and during this time the community
avoided drawing any clear geographic limits to its potential enlargement. As EU
commissioner for enlargement Olli Rehn explained in 2005, ‘Enlargement is a
matter of extending the zone of European values, the most fundamental of which
are liberty and solidarity, tolerance and human rights, democracy and the rule of
law.’121 Notwithstanding the clarity of Rehn’s assertion, the community’s
consensus on membership eligibility was about to break, leading to a very
different pattern of negotiation and decision-making in the following decades.
8
Membership Eligibility in a Divided
Europe, 2006—2020
The EU has to be ready to take all those European countries whose

population wants to join Europe and that fulfil the criteria—and please
don’t ask me where the borders of Europe are, that’s something we didn’t
want to put on the agenda.
—Erkki Tuomioja, Foreign Minister of Finland, June 2006, just
before Finland took over the EU’s rotating presidency1
This is not a European Union negotiation. This is an intergovernmental
negotiation where every member state can veto whatever it wants to
veto. . . . Every member state remains master of the negotiating process.
—Bernard Bot, Foreign Minister of The Netherlands, December
2006, describing decision-making on Turkey’s candidacy2
As shown in the previous three chapters, EU decision-making on states seeking

accession was strongly shaped for decades by membership norms prevailing
among the community’s member states. This culminated in the post-1970 norm
that only European states committed to the principles of liberal democracy were
eligible for EU membership and that all such states fulfilling this standard had a
right to membership. The most high-profile fruits of this norm included the
accession of ten former Communist states of central and eastern Europe plus
Cyprus and Malta. As shown in Chapter 7, it also contributed directly to the deci-
sion to grant candidate status to Turkey and, due to Ukraine’s inconsistent com-
mitment to liberal democracy, to a reluctance to recognize that country’s
However, as shown in Chapter 3, new conceptions of European identity and
new attitudes toward EU enlargement began to emerge in the early years of the
new century, leading to a breakdown in normative consensus among the member
states by late 2005. As a result of this dissensus, and pressure from particular
member states, the EU’s manager of enlargement, the European Commission,
adopted a new approach to enlargement, starting with a subtle re-interpretation
of the Treaty on European Union’s Article 49, which had established that ‘Any
1 Les négociations d’adhésion UE-Turquie pourraient être en danger, Le Monde, 29 June 2006.
2 Cyprus veto looms over EU-Turkey talks, EUobserver.com, 12 December 2006.
a Divided Europe, 2006–2020 195
European State which respects the principles set out in Article 6(1) may apply to
become a member of the Union.’ In the 2006 edition of its enlargement strategy,
the Commission asserted that Article 49 ‘does not mean that all European
countries must apply, or that the EU must accept all applications. The European
Union is defined first and foremost by its values.’3 This reference to values was
neither new nor controversial, but highlighting the fact that the Union was not
obligated to accept all European applicants was clearly an attempt to limit the
dynamics of normative entrapment.
At the same time, the Commission document stipulated that enlargement
decision-making would henceforth take into account the ability of the Union to
integrate new members without compromising its policies, budget, institutional
functioning, or democratic legitimacy. This new stance stopped short of creating
a new formal criterion for accession, and it increased pressure on member states
to prepare the Union’s institutions for enlargement. Nonetheless, by highlighting
such a complex new condition (references to ‘absorption capacity’ since 1993 had
never received such attention), it was clearly intended to slow the momentum
towards an ever-larger community. According to one report, a French diplomat
stated that the reframing of the treaty and new decision- making condition
‘reflects the current mood on enlargement’ while a Dutch contact added ‘[T]he
Netherlands has pushed hard for this.’4 But not all the member states embraced
this development: Swedish Foreign Minister Carl Bildt cautioned his colleagues,
‘There is talk of defining the borders of Europe. Drawing big lines on big maps of
the east of Europe risks becoming a dangerous process.’5
This unprecedented dissensus among the member states and new approach to
collective decision- making produced a new politics of membership and
enlargement characterized by contention, hard bargaining and political deadlock
on all but the easiest cases. This pattern fits the expectation, presented in
Chapter 2, that the decision-making dynamics associated with the presence of
membership norms degrade in the absence of such norms and are replaced by
hard bargaining over the policy preferences of the member states. This final
empirical chapter illustrates the consequences of this normative dissensus by
tracing EU discourse and decision-making on the membership eligibility of
Turkey and Ukraine since 2006. For practical reasons, it focuses on key moments
in each case over this period.
Before proceeding to the empirical analysis, however, it is important to note
that the regulations governing access to official documents make it far more
difficult to gain reliable evidence of norms prevailing (or not) among EU
4 Brussels to take more self-protecting approach to enlargement, EUobserver.com, 8 November 2006.
5 Open wide Europe’s doors, International Herald Tribune, 7 November 2006.
governments in this period than it was for earlier periods falling on the open side
of the 30-year ‘moving wall’ of secrecy. The same is true for evidence of how
particular decisions on enlargement were made. So instead of using private
discourse (travaux préparatoires and procés verbaux from EU meetings and the
internal notes of member states) to identify norms and trace decision-making
processes, we have to rely more heavily on public speeches, debates, policy
declarations, and press accounts. This raises issues of reliability, given the
incentive to please public audiences by concealing one’s private views, but there is
no alternative for the present study. We can only trust that the tight linkage
between membership norms and policy choices demonstrated in earlier chapters
will boost the credibility of this chapter’s similar findings in the latest period.
1. Bargaining over the opening to Turkey
Although the EU had first recognized Turkey’s membership eligibility in 1959,

and reconfirmed it more explicitly in 1963, 1997, 2004, and 2005, the issue re-
emerged at the highest levels of decision-making after 2005 and remains unsettled
today. This is not because the member states’ preferences changed suddenly and
radically against Turkey in 2005, apart perhaps from Angela Merkel’s victory in
Germany’s elections. Frits Bolkestein was likely correct when he suggested in
December 2004, ‘If a strictly secret vote was held, in the commission as well as
among government leaders, then only a small minority would be in favor of
Turkish accession.’6 The key change at the end of 2005, as discussed in Chapter 3,
was the breakdown of the liberal democratic membership norm that previously
had limited the willingness of member states’ governments to speak or act on
their preferences.
In the new, more contentious normative environment after 2005, EU member
states were unusually direct in their comments on Turkish accession. In addition
to expressing concern about Ankara’s implementation of the acquis, and thus the
progress of accession negotiations, Turkey’s critics re-opened debate on its basic
eligibility for membership. The negotiations were thus plagued by an obstacle that
could not be resolved by a diplomatic horse-trading: ‘The will is not there at the
moment,’ Gaston Thorn, former prime minister of Luxembourg and president of
the European Commission, observed in early 2006. ‘If you listen to the statements
of the Heads of State or Government, everyone is trying to find arguments in
order not to do it, but not in order to do it,’ said Thorn.7 This had major
consequences for the progress of negotiations on Turkish accession. This part of
the chapter provides detailed evidence of this change in 2006, the first year of
6 Turkey could complete EU bid in five years, Times of Malta, 7 December 2004.
7 CVCE: Interview with Gaston Thorn: Turkey and the European Union (6 February 2006).
accession negotiations, and then various illustrations of its consequences in later

years. In so doing, it draws attention to an under-recognized dimension of the
breakdown of EU–Turkey relations that persists to this day.
Following the EU’s formal decision in 2005 to open accession negotiations
with Turkey, the first ‘chapter’ to be raised for negotiations in 2006 dealt with
science and research. There was very little EU legislation in this area, so it should
have been easy. However, Cyprus seized the opportunity to pressure Ankara to
make good on its obligation under the EU-Turkey customs union to open its
ports and airspace to Cypriot traffic. Austrian chancellor Wolfgang Schüssel,
holder of the EU’s rotating presidency, set ‘the end of 2006’ as a deadline without
specifying the consequences of non-compliance.8 Although the EU ‘closed’ the
first chapter several months later, Austria’s foreign minister warned, ‘If there are
no concrete steps here, there will be serious difficulties sooner or later. There’s a
risk of running into a dead end.’9
A similar reluctance was evident in April 2006 around the opening of the
second negotiation chapter concerning education and culture policy, which is
also typically not difficult in accession talks. The Austrian presidency, supported
strongly by Cyprus and France, tried to get all member states to agree that
Turkey’s still-controversial record on human rights and democracy could be
considered in talks on the education and culture policy chapter. The UK, Finland,
and Spain blocked this move, arguing that the linkage was unprecedented, but the
proposal itself signalled that certain member states were still not reconciled to
Turkey’s membership eligibility and were prepared to attempt unusual procedures
to block it.10 And given that every member state has a potential veto over all 35
negotiating chapters, this threat could not be ignored.
Within Germany’s ruling coalition, social democrats generally supported
Turkish accession but Merkel’s own CDU party and its Bavarian CSU allies were
steadfastly opposed. According to Bavarian premier Edmund Stoiber, ‘Turkey is
not European and does not belong in Europe.’11 CDU general secretary Ronald
Pofalla declared that Turkish accession is ‘not defendable and wrong’. When
Merkel herself visited Ankara in early October, she made the startling announce-
ment that only after the EU–Turkey negotiations are completed will it be decided
‘what the result is’.12 In other words, even if Turkey fully satisfied all EU require-
ments and closed all 35 negotiation chapters, the EU might still decide to deny it
membership. This threat went well beyond the European Council’s December
8 Austrian presidency says Turkey must recognize Cyprus this year, EUobserver.com,
22 February 2006.
9 Turkey warned as EU opens the door, Euractiv.com, 13 June 2006.
10 EU states clash on political criteria in Turkey talks, EUobserver.com, 6 April 2006.
11 German premier Stoiber implies no Muslim country can be EU member, TurkishWeekly.net,
25 September 2006.
12 Verheugen warns EU against ‘dangerous spiral’ on Turkey, EUobserver.com, 9 October 2006.
2004 decision (see Chapter 7) that a future Turkish accession might involve long
transitional periods or even permanent safeguards.
European commissioner (and German social democrat) Guenter Verheugen
replied to Merkel by blasting the European tendency to send Turkey ‘almost
exclusively negative signals’, which he said were undermining support for reforms
in the candidate country. Indeed, popular support within Turkey for joining the
EU was dropping rapidly. By September 2006, only 32 per cent of Turks believed
that their country ‘must certainly enter the EU’, down from 57 per cent in 2005
and 67 per cent in 2004. Meanwhile, more than 25 per cent opposed EU accession,
compared to just 10 per cent in 2005.13 But in the absence of a strong membership
norm helping to keep member states in line behind the 2005 decision, such
criticisms were ineffective.14
By the autumn, it had become apparent to the European Commission that
political reforms in Turkey had stalled.15 In addition, the EU was increasingly
concerned that Ankara had failed to open its air and sea ports to Cypriot traffic,
as it had committed to do two years earlier. An attempt by the Finnish EU
presidency to broker a settlement on the Cyprus issue was cancelled just days
before the Commission was due to release its annual enlargement strategy,
reportedly because of the Turkish government’s anger that Greece would not
participate.16 Merkel then warned Turkey that its refusal to allow free movement
of goods with Cyprus, a member state, was creating ‘a very, very serious
situation . . . regarding the continuation of the accession talks’.17 Cyprus threatened
to veto opening of negotiations on the education and culture chapter if the free
movement issue was not settled.18
The Turkey section in the Commission’s 2006 enlargement strategy, released in
early November, reported on a lack of progress on human rights and the rule of
law, as well as the slow pace of legislative alignment on agriculture, environment,
and internal market issues. On human rights and rule of law issues, it reported
ongoing problems related to free expression, the use of torture, civilian control of
the military, corruption, judicial independence, and the rights of Kurds, women,
religious minorities, and trade unions.19 Enlargement commissioner Olli Rehn
nonetheless tried to relativize the situation: ‘In the public debate, one may get the
impression that Turkey is backtracking on the reforms. This is not the case.
Turkey has continued political reforms, even though their pace has slowed down
13 Less than a third of Turks keen to join the EU, EUobserver.com, 25 October 2006.
14 Verheugen warns EU against ‘dangerous spiral’ on Turkey, EUobserver.com, 9 October 2006.
15 EU alarmed by Turkey reform woes, BBC News, 3 November 2006.
16 EU cancels emergency Turkey talks, EUobserver.com, 3 November 2006.
17 Merkel warns Turkey on ‘very, very serious situation’, EUobserver.com, 6 November 2006.
18 EU report contains stinging criticism of Turkey, EUobsever.com, 7 November 2006.
19 European Commission. Enlargement Strategy and Main Challenges 2006–2007. COM (2006)
649, Brussels, 8 November 2006.
during the last year.’20 Nonetheless, the document announced, the Commission
would intensify monitoring of Turkey’s compliance with the political criteria for
membership and continued violations could lead to a suspension of accession
negotiations.21
As expected, the document also highlighted Turkey’s continued denial of
free movement of traffic from Cyprus, which violated Turkey’s obligations under
the customs union agreement. There were, however, differences of opinion within
the Commission and between member states regarding how the EU should respond,
including the possibility of suspending some or all of the accession negotiation
chapters. Within the Commission, the Cypriot, Greek, and French commissioners
argued that Ankara’s non-compliance merited a firm response, while others,
including Olli Rehn and President Barroso, favoured giving the Finnish EU presi-
dency more time to broker a solution. Rehn criticized European politicians who,
he said, used the Cyprus issue to promote their opposition to Turkish member-
ship: ‘Those who continue to question Turkey’s EU membership goal create a
vicious circle . . . [I]t is essential that the train remains on track.’22
Yet once again, in the absence of a strong membership norm, Commission
advocacy could not overcome differences among the member states on Turkey’s
candidacy. The Cypriot, French, and German governments favoured some
suspension of accession negotiations while Italian premier Romano Prodi, among
others, were against stopping the talks. ‘There is an idea in some parts of Europe
to stop it all,’ Prodi said, ‘[b]ut I think this is a great historical challenge.’
Recognizing these differences, the Commission delayed until December any
recommendation on whether or not the EU should suspend accession talks with
Turkey based on the Cyprus issue.23
Talks between the EU and Turkey continued for several weeks but broke down
definitively in late November, with Turkey insisting that the normalization of its
relations with Cyprus required that the EU honour its promise to end the
economic isolation of the Turkish- controlled northern part of the island.
Knowing the preferences of the member states on this issue, including Austria’s
call for a pause in the entire negotiations, the Finnish foreign minister announced,
‘There will be consequences . . . business as usual cannot continue.’24 Two days
later, after a ‘serious and earnest debate’ in the college of commissioners, Olli
Rehn announced the Commission’s recommendation: suspending the opening of
eight chapters until Ankara allowed trade from Cyprus. As he described it, the
20 EU treads carefully and gives Turkey talks one more chance, EUobserver.com, 8 November 2006.
22 EU treads carefully and gives Turkey talks one more chance, EUobserver.com, 8 November 2006.
23 Brussels set to give Turkey more time, EUobserver.com, 7 November 2006; EU treads carefully
and gives Turkey talks one more chance, EUobserver.com, 8 November 2006.
24 EU–Turkey talks in crisis after Finnish plan fails, EUobserver.com, 27 November 2006.
recommendation was based on two points—first, as a community based on law, the

EU could not tolerate Turkey’s failure to meet its legal obligations, and second, both
Turkey and the EU had a strong interest in keeping the negotiations alive. ‘There is
no train crash, no freeze, no hibernation, but yes, a slowing down,’ said Rehn.25
Beyond Rehn’s attempt to spin the issue, though, it was clear that adopting the
Commission’s recommendation would kill any momentum remaining in the EU–
Turkey relationship. And if they so desired, there was ample precedent for the
member states to ignore Commission recommendations on enlargement. For
example, in 1975, the Greek government convinced the member states to over-
ride the Commission by linking its desire for a speedy accession to a membership
norm that liberal democracies were eligible for accession (see Chapter 7).
However, the member states were deeply divided on how to handle the Cyprus
issue and even more so on the broader question of Turkish accession. And unlike
in the case of Greece in 1975, there was no longer a clear membership norm to
guide decision-making. In informal statements, Italy endorsed the Commission’s
recommendation but Spain, Sweden, and the UK found it too hard on Turkey
while Austria and Cyprus favoured a freeze of the entire negotiations.26 France and
Germany then consulted separately and agreed that the EU should impose a new
deadline on Turkey to normalize relations with Cyprus, only for Merkel to abandon
this joint position two days later and endorse the Commission plan following direct
contacts with Olli Rehn and Turkish premier Recep Tayyip Erdoğan.27
On 11 December, the EU’s foreign ministers agreed a position designed to bal-
ance the preferences of all the member states. At first, they adopted the
Commission’s recommendation to freeze eight of 35 chapters until the ports issue
was resolved. They did not establish a deadline by which Turkey had to comply
but they did instruct the Commission to review the situation on an annual basis.
In addition, the ministers agreed that ‘chapters for which technical preparations
have been completed’, such as economic and monetary issues, industry, financial
control, and education and culture, ‘will be opened in accordance with established
procedures’. Most importantly, they raised the bar for any future enlargements by
formally establishing the union’s ‘capacity to absorb new members’ as a
consideration in EU decision-making. In the end, it was nether membership
norms nor formal EU rules that determined this outcome—it was a simple
horse-trade. ‘There were delegations which may have thought that this was too
harsh, but there were also those who said this was too lenient,’ said the Finnish
foreign minister.28
25 Brussels proposes ‘slowing down’ of Turkish EU talks, EUobserver.com, 29 November 2006.
26 Cyprus threatens to block EU deal on Turkey talks, EUobserver.com, 1 December 2006.
27 Merkel and Chirac to urge new EU deadline on Turkey, EUobserver.com, 4 December 2006; No
ultimatum for Turkey, says Germany, EUobserver.com, 6 December 2006.
28 EU division marks key Turkey talks meeting, EUobserver.com, 11 December 2006; EU clinches
deal on Turkey talks, EUobserver.com, 11 December 2006; Cyprus vote still looms over EU-Turkey
talks, EUobserver.com, 12 December 2006.
The something-for-everyone nature of the deal became apparent when the

European Council endorsed it two days later. The content of the Council’s
decision allowed the Swedish prime minister to urge that the summit be
remembered for ‘keeping the doors open and sticking to our obligations’, and the
Danish prime minister to say ‘I don’t think we could or should draw a line across
Europe once and for all. But we have to take into careful consideration the
capacity of the EU to include new member states.’ At the same time, the Austrian
chancellor could take satisfaction from having suspended part of the Turkey
negotiations and reinforcing the principle that closing any chapter required
unanimous support from the member states. Finland’s prime minister, who had
helped to broker the deal, proudly declared, ‘We found the way forward. We did
not set any new criteria. The door is still open.’ Nevertheless, the compromise did
not prevent the Irish prime minister from predicting accurately that future
enlargement will be ‘much more difficult’.29
These developments, culminating in the freezing of eight chapters in Turkey’s
accession negotiations in December 2006, set a pattern in EU–Turkey relations
that persisted for years to come. The EU froze three more chapters in 2007 and six
more in 2009, effectively blocking progress on nearly half of the areas linked to
accession. Two of the blocked chapters—chapter 23 (judiciary and fundamental
rights) and chapter 24 (justice, freedom, and security)—covered core issues of
‘EU values’, leading a senior Turkish official to complain ‘You can’t ask us to
compete benchmarks which were never given to us and then bash Turkey on
lacking political reforms—it simply makes no sense.’30 Meanwhile, chapter 19 on
social policy and employment was not frozen but the EU refused to open
negotiations until Turkey commit to international legal texts that multiple
member states (Germany, Spain, and others) have either not signed or not
ratified.31 So notwithstanding real problems in Turkey, it began to appear that the
EU was using ‘EU values’ as a cover for other motives.
Three of the 17 blocked chapters were later re-opened, but several member
states insisted that they would only consider chapters that promoted functional
cooperation between the parties, not chapters tied specifically to accession.32 In
2017, the EU responded to worsening problems with human rights, democracy,
and the rule of law in Turkey by cutting funding designed to help the country
meet the requirements for accession. Although the EU did not revoke Turkey’s
candidacy and the Commission continued to press Turkey for reforms, leading
EU politicians called more and more openly for accession negotiations to be
replaced by a ‘privileged partnership’. In short, the member states were no longer
29 EU leaders prepare for delicate enlargement dinner, EUobserver.com, 14 December 2006; New
EU ‘consensus’ over enlargement highly fragile, EUobserver.com, 15 December 2006.
30 Turkey: EU political benchmarks ‘were never given to us’, Euractiv.com, 7 February 2013.
31 The EU’s non-negotiations with Turkey, Euractiv.com, 21 March 2013.
32 Turkish EU talks partially blocked by France, EUobserver.com, 26 June 2007.
unified behind their earlier recognition of Turkey as a candidate state eligible for
membership and bound for accession as soon as it demonstrated its compliance
with EU requirements. The EU–Turkey relationship thus entered a largely
transactional phase in which the country’s membership perspective was reduced
to empty rhetoric and bureaucratic procedures.
The EU and Turkey launched a ‘Positive Agenda’ in May 2012 whose purpose,
said EU enlargement commissioner Štefan Füle, was ‘to keep the accession
process alive and put it properly back on track after a period of stagnation’. It
would involve working groups composed of EU and Turkish officials, as well as
contacts with civil society, on issues of legislative alignment, political reforms and
fundamental rights, visa, mobility and migration, trade, energy, counter-
terrorism, and foreign policy.33 Yet regardless of its official purpose, the highly
technical ‘positive agenda’ could not overcome the fundamental obstacle to
progress on Turkish accession after 2005—namely, the fact that member states
opposed to Turkish accession, whether in principle or because of conditions
prevailing there, did not hesitate to block the process.
The situation changed after the massive influx of migrants to Europe in 2015,
many of them refugees fleeing dictatorship and civil wars in Syria and Afghanistan,
destabilized politics and fuelled the rise of far-right parties in multiple EU member
states. Germany initially adopted an open-door policy for refugees but as the
political limits of this policy became apparent, Merkel and other EU leaders looked
for alternative solutions.34 Notwithstanding their growing concern about rule of
law problems in Turkey and about Turkey’s increasingly aggressive posture toward
the Kurds in northern Syria, top EU officials and then Angela Merkel travelled to
Ankara hoping that Turkey would help solve the EU’s refugee crisis by limiting the
flow of migrants through southeastern Europe.
With the EU suddenly in the demandeur position, Ankara had new bargaining
leverage.35 To start with, publication of the European Commission’s annual
progress report on Turkey, which contained harsh reviews of rule of law problems
in the country, was delayed from mid-October until after the Turkish elections on
1 November.36 Meanwhile, despite just recently reiterating that she was ‘against
EU membership’ for Turkey, Merkel agreed to support a revival of accession
negotiations: ‘Germany is ready this year to open Chapter 17 [on economic and
monetary policy] and to make preparations for 23 and 24. We can talk about the
details,’ she said.37
33 European Commission, Positive EU-Turkey agenda launched, Press Release MEMO/12/359, 17
May 2012.
34 How Merkel and Europe came to embrace Erdoğan, Politico.eu, 16 October 2015.
35 Turkey’s Erdogan calls the shots at EU summit, Euractiv.com, 21 October 2015.
36 Commission accused of playing ‘hide and seek’ with Ankara, Euractiv.com, 28 October 2015.
37 Merkel against Turkey joining the EU, Euractiv.com, 8 October 2015; Merkel, in bind on
migrants, ready to back Turkish EU bid, Reuters.com, 18 October 2015; Merkel says ready to support
Turkey EU accession process, Euractiv.com, 19 October 2015.
The EU and Turkey adopted a Joint Action Plan in November 2015 to promote
cooperation in stopping irregular migrant flows and to support Syrian refugees
within Turkey, followed in March 2016 by a more comprehensive agreement by
which all irregular migrants crossing from Turkey to Greece would be returned to
Turkey. In exchange, the EU would significantly increase its financial support to
Turkey, and the EU would accept one Syrian refugee for settlement in Europe for
every one stopped in Turkey. In addition, the EU agreed to accelerate
implementation of a visa liberalization scheme allowing Turks visa-free entry into
the EU, to upgrade the existing Customs Union that had previously been blocked
by the Cyprus dispute, and to re-energize Turkey’s accession process starting with
negotiations on chapter 33, which had been frozen since 2006.
These moves, especially the EU’s various concessions to long-standing Turkish
demands, were controversial. According to EU foreign policy chief Federica
Mogherini, the new EU–Turkey relationship was pragmatic: ‘The sense of
direction is this: rather than having an ideological debate about enlargement
tomorrow or no enlargement, take concrete steps to bring us closer.’38 Others
were less charitable, describing it as a ‘transactional relationship dressed up as an
accession nobody believes in’.39 One EU diplomat conceded, ‘Bowing in front of
Erdoğan is certainly not ideal, but we do not have much leverage.’40 Yet regardless
of the controversy, the migrant deal ensured that Turkey’s EU candidacy stayed
alive, at least formally, through the stormy events of the next few years.
Although the migrant deal temporarily calmed official questioning of Turkey’s
eligibility for EU membership, the issue was re-opened dramatically by the
attempted coup d’état on 15 July 2016 and especially by the massive crackdown
that Erdoğan’s government launched in its aftermath. Tens of thousands of
soldiers, politicians, civil servants, judges, prosecutors, journalists, and teachers
were jailed or dismissed from employment. Over the following weeks, with the
rule of law seemingly collapsing in Turkey, the EU’s enlargement commissioner
said it was ‘legitimate’ to question Turkey’s membership eligibility, Austria’s
Chancellor Christian Kern called Turkey’s membership prospects a ‘diplomatic
fiction’, and Luxembourg’s Foreign Minister Jean Asselborn said the purges were
reminiscent of Nazi Germany.41 Yet most senior officials rejected the call for an
end to accession negotiations, generally on utilitarian grounds. Commission
President Jean-Claude Juncker said he didn’t think such a move would be ‘helpful’
while German Finance Minister Wolfgang Schäuble insisted the process must
38 EU warms to Turkey, Serbia membership bids, Euractiv.com, 15 December 2015.

39 The EU’s hypocrisy on Turkey, Politico.eu, 9 December 2015.
40 EU countries still bickering over Turkey, Politico.eu, 26 November 2015.
41 Johannes Hahn: It is ‘legitimate’ to question Turkey’s accession to the EU, Poltico.eu, 29 July
2016; Austrian chancellor: Turkey’s EU accession talks a ‘diplomatic fiction’, Politico.edu, 4 August
2016; Turkey’s EU candidacy hangs in the balance, top official says, Reuters.com, 9 November 2016.
continue: ‘I do not like what Erdoğan is doing at all, but I don’t think that we
should stop cooperating with him,’ said Schäuble.42
When the EU’s foreign ministers met on 14 November, there was a general
consensus that any reintroduction of the death penalty, which Erdoğan had
recently proposed, would be a red line for the EU. But apart from Austria and
Luxembourg, there was little support for ending the accession negotiations, which
all recognized would put an end to the migrant deal. ‘We need a rational and
cool-headed response,’ said Hungarian foreign minister Peter Szajarto.43 Yet still
the issue did not go away. In a non-binding vote on 24 November, a large majority
in the European Parliament supported a temporary suspension of all accession
talks with Turkey in order to protest the post- coup crackdown.44 Erdoğan
immediately threatened to cancel the migrant deal and flood Europe with
refugees. A Germany foreign ministry spokesperson then urged calm, ‘It is
important that we do not freeze the accession negotiations because that would
only further damage the relationship between Turkey and Europe, and that would
not be in the interests of Turkey or of Europe.’45
In early December 2016, the Dutch representative in Brussels informed the
other member states that The Hague favoured a suspension of all membership
talks with Turkey until the arrest of political opponents, the dismissal of civil
servants and plans to establish a more centralized, constitutional presidency were
dropped.46 Austria’s foreign minister Sebastian Kurz then pushed back against
other member states’ calls for pragmatism, saying ‘there is a big difference
between keeping up the dialogue on issues of common interests and pretending
that a country will eventually be part of the European Union’. When the foreign
ministers met again, most favoured an unstated but de facto suspension of the
accession talks, which they believed would avoid a rupture of relations with
Ankara and thus a collapse of the migrant deal. Kurz was in a distinct minority,
but he opted to use ‘the leverage I have as a foreign minister’ and blocked draft
recommendations for the upcoming European Council summit. This tactic was
effective but not appreciated: ‘Austria burnt a lot of credibility and capital today,
despite everyone’s effort to keep them on board,’ said one diplomat. Instead of
issuing Council Conclusions including a position on Turkey, which require
unanimity among the member states, the Council would issue ‘presidency con-
clusions’ without a position on Turkey.47
42 Austrian chancellor: Turkey’s EU accession talks a ‘diplomatic fiction’, Politico.edu, 4 August
2016; Juncker dismisses calls to halt Turkey membership talks, Politico.edu, 4 August 2016; Wolfgang
Schäuble: EU needs Turkey for migration crisis, Politico.edu, 17 August 2016.
43 EU must not ditch Turkey, ministers warn, EUobserver.com, 15 November 2016.
44 Parliament backs freeze of Turkish membership talks, Euractiv.com, 24 November 2016.
45 EU should not freeze accession talks with Turkey, Germany says, Reuters.com, 25 November 2016.
46 Dutch join push for EU-Turkey accession talks freeze, Euractiv.com, 2 December 2016.
47 Austria to the EU: We need to talk about Turkey, Politico.eu, 12 December 2016; Austria calls for
freeze of Turkey’s membership talks, FT.com, 13 December 2016.
The pattern in 2017 was basically like that in 2016. The EU’s institutions and
member states were informed about a worsening human rights situation within
Turkey, including repeated violations of media freedom, 40,000 people arrested
and more than 100,000 people dismissed from their jobs during the seven months
since the attempted coup.48 When an April referendum marred by significant
irregularities and condemned by the OSCE and Council of Europe endorsed
Erdoğan’s proposal for a more centralized, presidential republic, enlargement
commissioner Johannes Hahn urged the member states to consider a ‘different
type of relations’ with Turkey.49 Gianni Pitella, head of the socialist group in the
European Parliament, joined in, saying ‘We would have liked to keep Europe’s
doors to Turkey open. Unfortunately, the referendum . . . has forced us to request
the suspension of the accession negotiations.’50
As before, the member states were divided: Austrian foreign minister Sebastian
Kurz repeated his familiar calls to end accession negotiations, while Germany’s
foreign minister Sigmar Gabriel said this would be ‘the wrong reaction’. French
foreign minister Jean-Marc Ayrault was clear about how his government saw the
stakes: ‘How can we ignore Turkey? We have to fight terrorism, we want Turkey to
respect the migration deal . . .’51 According to Kati Piri, the European Parliament’s
rapporteur on Turkey, the EU’s inability to take a strong position in defence of
‘EU values’ in Turkey was easy to explain: ‘28 member states can’t say anything.
You need unanimity.’52 Piri’s diagnosis is undeniable and clearly captures part of
the reason for the EU’s paralysis. However, as shown in previous chapters, similar
divisions in earlier eras were often resolved by reframing the choice in terms of a
membership norm that no member state dared to disavow. In this period, there
was no such robust normative consensus.
As long as Turkey did not reintroduce the death penalty, the accession process
would continue, formally alive but effectively frozen.53 ‘Turkey is and stays a
candidate country,’ said EU foreign policy chief Federica Mogherini in the face of
continuing calls for a total end to talks. After yet another ‘high level political
dialogue’ with Turkey failed to produce progress, Hahn called again for a new
approach: ‘I believe it is time for the member states to discuss the strategic
implications of this behavior. . . . Shrugging alone is not a political strategy in the
long run.’54 However, there would be no significant change as long as Turkey
continued to deliver on other issues valued by member states. On the eve of an
EU summit in October, Bulgarian prime minister Boyko Borissov made clear that
48 Human rights group urges Turkey to ‘urgently change course’, Politico.eu, 15 February 2017.
49 Commission urges ‘different type of relations’ with Turkey, Euractiv.com, 25 April 2017.
50 Gianni Pitella: ‘Authoritarian’ Turkey cannot join EU, Politico.eu, 10 May 2017.
51 Germany rejects calls for EU to end Turkey accession talks, DW.com, 28 April 2017.
52 Turkey rapporteur: ‘We are sending a very bad signal to the Balkans’, Euractiv.com, 5 July 2017.
53 Juncker: Death penalty in Turkey would mean end to EU accession talks, Euractiv.com, 1 June 2017.
54 EU and Turkey fail to defuse tensions, EUobserver.com, 25 July 2017; EU enlargement commis-
sioner calls for a new Turkey policy, Politico.eu, 22 August 2017.
the migration deal was key to his calculation: ‘Up to now, Turkey strictly abides
by the agreement. President Erdoğan stands by his word in this respect, and we
have zero migration. So my position will be to respect the EU-Turkey agreement,
not to cut [accession] negotiations.’55 Another priority for some member states
was Turkey’s cooperation in stopping European ‘jihadis’ who tried to return home
after fighting in Syria.56 The one action that member states could agree was to
reduce some of the pre-accession funding that the EU had been giving Turkey.57
The same pattern prevailed in 2018. In April, Johannes Hahn issued the
strongest rebuke of Turkey’s record on human rights and the rule of law that
the Commission had ever produced, concluding that the country ‘continues
to take huge strides away from the EU’.58 In Council Conclusions from their
June 2018 summit, the member states seemed to agree, saying they were ‘especially
concerned about the continuing and deeply worrying backsliding on the rule of
law and on fundamental rights including the freedom of expression’ and noting
that ‘Turkey has been moving further away from the European Union’.
Nonetheless, they concluded, ‘Turkey remains a candidate country and a key
partner in many areas.’59 It was left to Hungarian prime minister Viktor Orban to
admit that the EU’s intergovernmental bargaining was sustaining an ‘insincere’
position: ‘I don’t remember a time when we told a candidate country that we want
to negotiate while significant countries said that we will never admit that country
into the European Union.’60
And again in 2019. In March, the European Parliament passed another
resolution calling for all accession negotiations with Turkey to be formally
suspended, with MEP and Turkey rapporteur Kati Piri explaining, ‘If the EU takes
its own values seriously, no other conclusion is possible than to formally suspend
the talks on EU integration.’61 At a meeting several days later of the EU-Turkey
Association Council, the first one held after the lifting of the State of Emergency
imposed in Turkey the previous year, the EU reiterated its recognition of Turkey
as a candidate country, despite presenting a long list of serious problems related
to democracy, human rights, and the rule of law in the country.62 In June, the
Commission’s 2019 Enlargement Package warned again that Turkey had
continued to move further away from the EU, ‘with serious backsliding not only
55 Borissov gives preview of Turkey debate before EU summit dinner, Euractiv.com, 19 October 2017.
56 Turkey funding cuts signal EU mood shift, EUobserver.com, 20 October 2017.
57 Merkel: EU to cut Turkey pre-accession funds, EUobserver.com, 20 October 2017.
58 Hahn: Turkey taking ‘huge strides’ away from the EU, Euractiv.com, 18 April 2018.
59 Council of the European Union. Council Conclusions on Enlargement and Stabilisation and
Association Process, 10555/18, 26 June 2018.
60 Hungary’s Orban says EU ‘insincere’ with Turkey on accession, APnews.com, 8 October 2018.
61 Parliament wants to suspend EU accession negotiations with Turkey, European Parliament
News, 13 March 2019.
62 Council of the European Union, Press statement following the 54th meeting of the Association
Council between the European Union and Turkey, Brussels, 15 March 2019.
in the areas of the rule of law and fundamental rights but also through the weakening
of effective checks and balances in the political system.’63 Yet once again, in the
absence of a strong membership norm, such appeals were unable to sway ‘the few
European countries that still believe’ Turkey’s future lies in the EU, as Greece’s
foreign minister described his country’s unchanging position.64
Just as consensus among the member states regarding the political character of
their community made the EU more receptive to Turkey in earlier decades, the
significance of normative dissensus today should not be under-estimated. In the
absence of consensus among member states on any membership norm after 2005,
opponents of Turkish accession found a way to frustrate the process while others
blocked them from reversing Turkey’s formal candidacy for membership.
Meanwhile, the Turkish government found that it was able to use functional
cooperation with the EU, most recently the 2016 migrant deal, to keep its
membership prospect alive while steadily moving away from the EU’s formal
expectations on democracy, human rights, and the rule of law, among other
topics. The same stalemate continued in 2020, with no end in sight.65
This account demonstrates the naivety of the European Commission’s claim,
‘The accession negotiations are based on Turkey’s own merits and progress in the
negotiations, measured against requirements, among which the respect for the rule
of law, human rights and fundamental freedoms, respect for international law and
good neighbourly relations.’66 It is far more accurate to observe, as another senior
Commission official did in off-the-record remarks, ‘The enlargement process is
technically driven but at the end of the day, the quantity of water in the glass is a
political judgment.’67 In the post-2005 phase of EU decision-making on Turkey’s
quest for EU accession, this political judgement has essentially been an intergov-
ernmental bargain among the member states, with very little role played by the
membership norms that had repeatedly been so decisive in earlier phases.
2. Bargaining over Ukraine
Altough it does not have the same historical legacy or political salience as
the Turkish case, Ukraine’s quest for EU membership eligibility was also caught
up in the normative dissensus, enlargement scepticism and intergovernmental
bargaining of the post-2005 era. The EU remained in close contact with a series of
governments in Kiev following its de facto decision in 2005 not to offer Ukraine a
clear membership perspective (see Chapter 7), but it did not change its position.
63 Commission: Turkey has continued to move away from the EU, Euractiv.com, 29 May 2019.
64 Greek foreign minister: EU should keep door open for Turkey, Politico.eu, 16 May 2019.
65 Commission seeks to revive enlargement with new package, Euractiv.com, 6 October 2020.
66 Turkey and EU shadow-box over illusive accession process, Euractiv.com, 24 January 2018.
67 Author interview with a senior European Commission official, 11 May 2012.
At the 2006 EU–Ukraine summit in Helsinki, EU leaders praised political reforms

that the Yushchenko government had adopted in the year and a half since the
Orange Revolution: ‘The elections showed that the consolidation of democracy
and the freedom of speech had been key achievements of the past two years.’ Yet
the EU still refused to grant a clear membership perspective: ‘Ukraine is not ready
and we are not ready,’ said José Manuel Barroso.68
Over the following eleven years, as the EU and Ukraine negotiated and then
finally adopted an association agreement and a 2000-page partnership and free
trade agreement designed to set the agenda for cooperation between Ukraine
and the EU for many years to come, the EU refused again and again to make
any explicit statement on Ukraine’s eligibility for eventual accession. The EU
persisted in this position even after the country went through another revolu-
tion in which citizens died to overthrow their government in favour of a more
democratic and EU-oriented future for their country. To illustrate the pattern,
this narrative focuses on key moments involving various member states as lead
veto player.
The EU and Ukraine began talks on a free trade agreement when President
Kuchma was still in power in 1999 and launched formal negotiations in 2008,
several years after the Orange Revolution. The parties also began talks on a broad
association agreement during the same period. Throughout these negotiations,
Ukraine pressed repeatedly but unsuccessfully for a perambulatory clause
recognizing Ukraine’s EU membership eligibility, as the EU had agreed in
association agreements with other countries.69 The EU even refused to refer to
Ukraine as a ‘European state’ because this sounded too much like the EU treaty’s
Article 49, which says that all ‘European states’ may apply for accession.70 Yet
despite these refusals, the EU continued to press Kiev for reforms in the country’s
judicial system and electoral laws before it would agree to sign the texts.71
This happened because EU member states were divided in their preferences
and unrestrained by any strong membership norm. On one side were those that
wanted to accommodate Ukraine’s desire for a membership perspective and
accelerate the country’s association to the EU, including Sweden, Poland,
Lithuania, and Estonia. On the other side were those that insisted on further
reforms, including France and Germany.72 These calculations were connected to
competing visions of Ukraine’s future: ‘For some member states, and many
Ukrainians, the country’s association accord was to be the first step to
68 Is the EU ready for Ukrainian membership? EUobserver.com, 25 February 2014.

69 Ukraine takes EU to task for weak words on new treaty, EUobserver.com, 22 January 2007; EU
initials Ukraine agreement ‘to keep momentum’, Euractiv.com, 30 March 2012.
70 EU and Ukraine: What went wrong? EUobserver.com, 25 November 2013.
71 Füle warns Ukraine not to ruin its EU hopes, Euractiv.com, 21 September 2011.
72 Eastern countries add Ukraine to foreign ministers’ agenda, Euractiv.com, 27 August 2013;
Analysts slam Germany’s Ukraine policy shortcomings, Euractiv.com, 19 November 2013; Germany
brings Ukraine relations to EU summit table, Euractiv.com, 18 December 2013.
membership. For others it was precisely a means of holding that off, though helping
to create a stable buffer to the east of the EU.’73 Although the EU had included
recognition of membership eligibility in earlier association agreements with other
countries, France and Germany refused to compromise on Ukraine.
Meanwhile, Moscow pressed Kiev to join a new Russia-sponsored customs
union with Belarus and Kazakhstan. To make its point clear, the Russian
government began to limit imports from Ukraine, including measures targeted at
a giant Ukrainian confectionary producer whose owner Petro Poroshenko was
openly pro-European.74 EU foreign policy chief Catherine Ashton warned that
the Union ‘cannot lose Ukraine’ and Russian Foreign Minister Sergei Lavrov
replied that his country also ‘cannot lose Ukraine’.75 Despite their growing
awareness of the geo-political stakes, EU member states still did not agree to
grant Ukraine’s wish for a clear membership perspective.
Then, less than two weeks before the association agreement was due to be
signed at a summit in Vilnius, the Ukrainian government revealed that it would
prioritize relations with Russia and other members of the Commonwealth of
Independent States over integration with Europe. President Yanukovich informed
EU enlargement commissioner Štefan Füle that he could not accept the deal
because it would cost his country more than $600 billion in lost trade with Russia
and measures to comply with EU standards. As if to remove any doubt about the
country’s new direction, the Ukrainian parliament failed to pass laws that would
allow jailed opposition leader Yulia Tymoshenko to leave for medical treatment
in Germany, which the EU had said must happen before the signing of the
agreement.76
Thousands of people gathered within hours on Kiev’s Maidan square and
elsewhere in the country to protest Yanukovich’s decision, many of them carrying
EU flags. These ‘Euromaidan’ protests, as they came to be known, continued for
weeks despite efforts by the government to quell them through intimidation of
journalists, targeted abductions and beatings, violent charges by riot police, and
even sniper fire from rooftops. Dozens of people were killed and hundreds injured
but the protests continued. Yanukovych resigned and fled to Russia in late
February and a new government led by Arseniy Yatsenyuk took power in Kiev.
Within days, Russian troops with unmarked uniforms began to take up positions
in the Crimean Peninsula.
Like the earlier Orange Revolution, the Euromaidan protests forced the EU to
reconsider its position on Ukraine. Yet despite the size and persistence of the
protests and the violence exhibited in response by Ukraine’s authorities, EU
73 Setting the limits of Europe, Irishtimes.com, 26 November 2013.

74 Moscow warns Ukraine over EU pact signature, Euractiv.com, 19 August 2013.
75 Eastern countries add Ukraine to foreign ministers’ agenda, Euractiv.com, 27 August 2013;
Lavrov: Russia cannot lose Ukraine, TASS.com, 19 October 2014.
76 Ukraine stuns EU by putting association deal on ice, Euractiv.com, 25 November 2013.
officials were careful to avoid any new commitment on the membership eligibility
issue. At the EU’s December 2013 summit in Brussels, Commission president
Barroso confessed, ‘When we see those European flags in the streets of Ukraine in
this very cold temperature, we cannot resist to say they are indeed part of the
European family.’ Lithuanian President Dalia Grybauskaite added, ‘The European
Union is open to Ukrainian people but not necessarily the current Ukrainian
government—that’s the message.’77 Additional statements by senior EU officials
over the next few months were slightly more accommodating on the issue of
Ukraine’s future relationship to the Union, but it was the distribution of
preferences among the member states, rather than normative entrapment or
empowerment, that defined their content.
As protests in Kiev and other cities escalated in early 2014, the EU’s position on
Ukraine’s membership perspective began to soften. At the Munich security
conference on 1 February, EU enlargement commissioner Štefan Füle said that
only the promise of accession could transform former Soviet states: ‘If we are
serious about helping this part of Europe to transform, the association agreement
is only the first step. The next one should be the light at the end of the tunnel. You
can’t transform this part of Europe without using this most powerful instrument.’
Romanian President Traian Basescu agreed: ‘You can’t just invite them to associate
with the EU but say: “Yes. But this does not guarantee you will ever get into
the EU” .’78
Such arguments may have been backed by historical precedent and the
dramatic events still unfolding in Kiev, but they could not shift the distribution of
preferences among the member states when the EU foreign ministers gathered in
Brussels on 10 February. Polish Foreign Minister Radek Sikorski argued for a
clear commitment to Ukraine, while his French counterpart opposed any mention
of Ukraine’s membership prospects in the communique and even attempted to
insert language making clear that Ukraine would never become a member. The
resulting statement was hopeful but still non-committal: ‘The Council expresses
its conviction that this [association] agreement does not constitute the final goal
in EU-Ukraine co-operation.’79
After the meeting, Lithuania’s Foreign Minister Linas Linkevicius, whose
government had long favoured an EU perspective for Ukraine, acknowledged
‘deliberate ambiguity’ in the communique. ‘It’s as far as we can go jointly at this
point’, given the views of the various member states, he said. Sikorski gave an
optimistic spin to the Brussels statement, saying it should be understood in the
context of Article 49, which allows ‘any European state’ to join if it meets the
77 EU leaders close the door on Yanukovych, EUobserver.com, 20 December 2013.

78 EU commissioner calls for Ukraine accession promise, EUobserver.com, 1 February 2014.
79 EU gives Ukraine enlargement hint, EUobserver.com, 10 February 2014; Ukraine accession
promise would help EU-Russia relations, EUobserver.com, 13 February 2014.
c riteria. ‘We have opened the door and given hope to the Ukrainian nation that, if
Ukraine embarks on the course of reform, then it has the chance to take full
advantage of European integration and treaty provisions,’ he added.80 But the fact
remained that the EU had once again refused to clearly recognize Ukraine as
eligible for membership.
The issue emerged again in early March at a meeting in Dublin of the European
alliance of Christian Democrats, the European People’s Party. (By this time, pro-
Russian demonstrators had mobilized in Ukraine’s eastern province of Donbass
and a full-blown war with Russia suddenly seemed possible.) Fortunately for the
advocates for Ukraine, the centre-left government of France was not represented
at the Dublin gathering. As a result, the EPP resolved that ‘Article 49 of the EU
treaty refers to all European states, including Ukraine, which has a European
perspective and may apply to become a member of the Union, provided that it
adheres to the principles of democracy, respects fundamental freedoms and
human and minority rights and ensures the rule of law.’81 But even as the largest
party group in the European Parliament, the EPP could not speak for the EU.
On such an issue of high politics, that responsibility was borne by the EU’s
heads of state and government, who met in Brussels on 6 March for an emergency
summit of the European Council. After an opening session with Ukraine’s new
prime minister, Council members exchanged their views. The Polish prime
minister, Donald Tusk, with support from Germany and others, wanted the EU to
finally offer an EU membership perspective to Ukraine. But France, represented
by President Francois Hollande, would not consider this, not even if the offer
were linked to specific conditions. Other member states were focused on the need
to maintain unity in this critical moment, and wary of escalating tensions between
Russia and Ukraine.82 In the end, their final communique restated support for the
still-unsigned association agreement and endorsed a comprehensive financial
assistance package for Ukraine, but made no mention of Ukraine’s ultimate rela-
tionship to the Union.83
As a gesture of support for the government in Kiev, the EU agreed to sign the
political chapters of the association agreement on 21 March, but without the
explicit preambulatory confirmation of Ukraine’s membership eligibility that
the government had sought. The main text of the agreement, containing a Deep and
Comprehensive Free Trade Agreement, was left for signature after the presiden-
tial elections that were held on 25 May. In the meantime, Kiev continued to seek
support for an explicit EU statement on its membership eligibility. For example,
80 EU gives Ukraine enlargement hint, EUobserver.com, 10 February 2014; Ukraine accession
promise would help EU-Russia relations, EUobserver.com, 13 February 2014.
81 Centre-right leaders give Ukraine hope of EU membership, EUobserver.com, 7 March 2014.
82 France denies EU membership perspective to Kyiv, Euractiv.com, 12 March 2014.
83 European Council. Statement of the Heads of State or Government on Ukraine. Brussels, 6
March 2014.
at a conference in Brussels organized by the European Policy Centre, Kiev’s

ambassador to the EU, Kostyantin Yelisieiev, argued that a membership
perspective would be ‘the light at the end of the tunnel’ that Ukraine so badly
needed. Former EU Ambassador to Russia Marc Franco agreed, explaining that
this commitment was necessary to counter-balance the social and economic
disruption (‘hara-kiri of the previous regime’) that the association agreement
would cause.84 Yet once again, such appeals proved unsuccessful.
At the signing of the final association agreement on 27 June, the newly-elected
president of Ukraine, Petro Poroshenko, could only repeat his country’s long-
standing request.85 Commission president Jose Manuel Barroso responded with
the same non-committal phrase that the foreign ministers had agreed in February,
saying that ‘signing the agreements should not be seen as the end of the road, but
the beginning of a journey’.86 Herman van Rompuy, President of the European
Council and thus de facto spokesperson of the member states, had to be even less
specific: ‘In Kiev and elsewhere, people gave their lives for this closer link to the
European Union. We will not forget them.’87
The member states’ collective reluctance to offer a clear membership
perspective to Ukraine was not shared by the European Parliament. Two weeks
after the signing ceremony, the parliament passed a resolution echoing the EPP’s
statement from early March: ‘Pursuant to Article 49 of the Treaty on European
Union, Georgia, Moldova and Ukraine—like any other European state—have a
European perspective and may apply to become members of the Union provided
that they adhere to the principles of democracy, respect fundamental freedoms
and human and minority rights and ensure the rule of law.’88 This resolution went
well beyond the recent position of the European Council, but the parliament
could not speak on behalf of the EU so the resolution did not represent the com-
mitment that Kiev had been seeking.
The same pattern was repeated in the buildup to the Eastern Partnership
summit in Riga in May 2015. Krzysztof Bobinski, Polish journalist and co-chair
of the Eastern Partnership Civil Society Forum, urged the member states to grant
Ukraine a clear membership perspective. Not doing so, he argued, would
discourage further reforms.89 The day before the summit, Ukrainian Foreign
Minister Pavlo Klimkin restated his government’s position: ‘Now we want con-
crete assurances in Riga that Ukraine is eligible for future membership of the
European Union and has the chance to become an accession candidate in the
84 Diplomats polemicise over Ukraine’s membership perspective, Euractiv.com, 3 June 2014.
85 Ukraine, Georgia, Moldova sign EU deals, Politico.com, 27 June 2014.
86 Borroso: Supporting Moldova, Georgia, Ukraine is a solemn commitment for the EU, Euractiv.
com, 27 June 2014.
87 Ukraine signs landmark agreement with the E.U., Washington Post, 27 June 2014.
88 European Parliament resolution RC-B8-25/2014, 17 July 2014.
89 Eastern countries deserve an enlargement perspective at the Riga summit, Euractiv.com,
13 April 2015.
future.’90 Yet once again, the EU refused. ‘The Eastern Partnership is not an
instrument for enlargement of the European Union, but it is an instrument of
rapprochement with the European Union,’ said Angela Merkel when she arrived
at Riga. The summit’s final declaration said that the participants ‘acknowledge the
European aspirations and European choice of the partners concerned’, but did not
mention their eligibility for membership.91
The issue returned acutely in the context of a 2016 referendum in The
Netherlands on approval of the EU–Ukraine association agreement. Despite the
Dutch government’s insistence that the association agreement was not an
accession treaty, and its consistent opposition in previous years to any recognition
of Ukraine’s eligibility for eventual membership, Euro-sceptics framed the issue
as a vote on Ukrainian accession and European enlargement more generally.92
The referendum result was 61 per cent negative. Rather than acknowledge the
outcome as advisory but non-binding on his government, Prime Minister Mark
Rutte attempted to gain approval from other member states for a legally-binding
guarantee that the association agreement did not constitute a step toward
membership for Ukraine, among other assurances.93 Although many EU
governments were unhappy with the Dutch proposal, they accepted it as the
necessary price to complete ratification of the association agreement. ‘It wasn’t
easy, it wasn’t pleasant, but it is necessary, because it ensures that the EU can
continue to form a united front against destabilising Russian foreign policy,’ Rutte
said.94 Once again, an EU decision on the membership eligibility of an aspirant
state was determined by hard bargaining among member states unrestrained by
community norms.
But this would not be the last time. Barely six months later, on the eve of an
EU–Ukraine summit in Kiev to cap off recent agreement allowing Ukrainians
visa-free travel to the EU and the final ratification of the long-awaited free trade
agreement, The Netherlands vetoed a draft summit communique that ‘the
European Union acknowledges Ukraine’s European aspirations and welcomes its
European choice’. The fact that the language of the communique had been in
negotiation for four months, and that identical language had been included in the
2015 Riga declaration, did not stop the Dutch from blocking agreement and
forcing the Kiev summit to end without any final statement.95
90 Ukraine wants roadmap to EU accession at Riga summit, Euractiv.com, 21 May 2015.
91 Ex-Soviet states accept limited EU perspective, EUobserver.com, 22 May 2015.
92 Farage to back ‘no’ vote in Dutch referendum on Ukraine’s EU association, Euractiv.com,
29 February 2016; Dutch referendum on Ukraine treaty will test anti-EU sentiment, Euractiv.com,
6 April 2016.
93 Netherland’s Rutte still mired in Ukraine referendum aftermath, Euractiv.com, 21 October 2016.
94 ‘Many not happy’ with Dutch limits of Ukraine pact, Euractiv.com, 15 December 2016; Ukraine
urges Dutch to sign EU deal, Euractiv.com, 16 December 2016.
95 Dutch block aspirational communique at EU-Ukraine summit, Euractiv.com, 12 July 2017; EU
and Ukraine—close but not that close, Politico.com, 13 July 2017.
This reconstruction of decision-making after 2006 refutes the tendency to

attribute the EU’s negative stance on Ukraine’s membership eligibility to
‘enlargement fatigue’. While it is undeniable that public opinion in some member
states became less accepting of enlargement over time, and member states’
governments grew increasingly concerned about the EU apparatus’ ability to
function with more and more member states, the record shows that the negative
response to Ukraine is not easily explained by these amorphous and presumably
universal pressures. As shown above, the member states negotiated repeatedly
among themselves about Ukraine’s membership eligibility during the five years of
negotiations with Ukraine on the association agreement. There is strong evidence
of deep differences among them, and little evidence that they adjusted their
positions over time. Britain, Sweden, Poland, Latvia, Lithuania, and Estonia
referred repeatedly to formal rules and expectations related to EU membership,
and to developments on the ground in Ukraine that they considered a good
reason for granting Kiev a membership perspective, especially the liberal
democratic tendencies of the Euromaidan revolution and the Russian annexation
of Crimea, while Germany, France, Austria, Belgium, and The Netherlands
remained steadfastly opposed.96 In short, the outcome was shaped by hard
bargaining among member states.
3. Conclusions
The Ukrainian case reveals, like in the Turkish case before it, that the post-2005
breakdown of normative consensus among EU member states had the effect of
replacing the dynamics of normative entrapment and empowerment with inter-
governmental bargaining in which a sub-set of member states impose their
preferences on the Union regardless of past commitments or expectations. This
sort of barely restrained inter-governmentalism, whether motivated by long-term
geo-political calculations or short-term domestic political needs, was markedly
different from the pre-2006 pattern evident in earlier chapters, where member states
tended to compromise preferences that contradicted prevailing membership norms.
The clearest expression of this development is Dutch Foreign Minister Bernard
Bot’s remark in the aftermath of the December 2006 EU foreign ministers’
meeting on Turkey’s non-recognition of Cyprus, ‘This is not a European Union
negotiation. This is an intergovernmental negotiation where every member state
can veto whatever it wants to veto. . . . Every member state remains master of the
negotiating process.’97 Although made in a specific context, Bot’s remark
accurately forecast the centrality of the member states in EU decision-making on
96 Vijf EU-landen blokkeerden lidmaatschap Oekraïne, Volkskrant.nl, 1 April 2016.

97 Cyprus veto looms over EU-Turkey talks, EUobserver.com, 12 December 2006.
enlargement for many years to come. Hence Christophe Hillion’s observation

four years later of a ‘creeping nationalisation’ of EU enlargement policy led by
Member States ‘showing less scruple in instrumentalising enlargement for
domestic political gains’ (Hillion 2010: 40).
This development has real-world consequences. In the Ukrainian case, the
EU’s refusal to grant a clear membership perspective leaves the government in
Kiev having to make costly reforms to demonstrate its bona fides to Brussels
without knowing or being able to tell the public what its efforts will yield over
time.98 In the Turkish case, the lack of a clear membership perspective has
contributed, in addition to developments within the country itself, to a situation
where Turkey remains a candidate for membership but nobody in Ankara or
Brussels seems to believe that accession will ever actually happen. Something
similar is happening with regard to the countries of the western Balkans, all of
which have been recognized officially as eligible for EU membership: ‘The
candidate countries pretend they want to reform and we pretend we want them to
join the EU.’99
The concluding chapter of this book explores further some of the theoretical
and practical implications of these developments in the discourse and direction
of European integration.
98 Poroshenko says Ukrainians at risk of losing faith in EU path, Euractiv.com, 17 January 2017.
99 Macron and the EU enlargement make-believe, EUobserver.com, 24 May 2019.
PART FOUR
C ONC LU SIONS A ND
IMPL IC AT IONS
9
Rethinking Europe, Rethinking Regions
I am often asked where Europe’s ultimate borders lie. My answer is that

the map of Europe is defined in the minds of Europeans. Geography sets
the frame, but fundamentally it is values that make the borders
of Europe.
—Olli Rehn, EU commissioner for enlargement, 20051
The big question . . . is how the EU will react to neighbours who share its
values and are keen to join the club.
—Fraser Cameron, former European Commission official, 20182
The question of the limits of Europe as a political community has long been one
of the most persistent questions in European debates and all signs indicate that it
will remain salient for years to come. As demonstrated in the preceding chapters,
this question cannot be answered with simple references to Europe’s physical
geography, nor to its cultural values and practices, nor to the formal rules of the
European Union, nor even to the commercial or security interests of its member
states. As Kalypso Nicolaïdes observes, ‘There is no specific tipping point, no
point on the map where we can say: EUness stops’ (Nicolaïdes 2014: 244). A more
reliable answer is found in the community’s membership norms—that is, the
community’s prevailing definition of which type of state is eligible for member-
ship. These norms have been repeatedly contested since the 1950s and evolved
significantly during this time, with observable consequences for the community’s
decision-making on its territorial enlargement and for how the community pre-
sents and justifies its very existence.
Importantly, the historical record shows that these norms cannot be attributed
simply to member state preferences. In some cases, changes in the norms were
clearly provoked by developments beyond the control of powerful member states
and moved in a direction that contradicted their preferences. In multiple cases,
one can observe member states with clear preferences on the eligibility of a par-
ticular aspirant state struggling to translate those preferences into a community

2 Georgia’s European Economic Area ambitions, Euractiv.com, 22 July 2019.
decision in a political environment shaped by the prevailing membership norm.

As such, the effect of EU membership norms can be separated from member
states’ views on enlargement in general and on the eligibility of particular states.
This chapter’s concluding observations are presented in three parts. Section 1
summarizes the book’s quantitative and qualitative findings regarding the
evolution of EU membership norms since the late 1950s and their contribution to
EU decision-making on the enlargement of the community during this period.
Section 2 considers the study’s general lessons for our understanding of regions
and regional integration. Finally, Section 3 combines normative critique and
historically-informed speculation in a discussion of the study’s implications for
salient issues in the future of European governance.
1. The construction of Europe
As shown in this book, the changing norms of the EU as a political community

have shaped EU decisions on enlargement, which have in turn reshaped the scope
and character of the organization most closely identified with ‘Europe’. This
finding confirms Christophe Hillion’s suggestion that ‘studying the “nitty-gritty”
of the enlargement process of the European Union is a fruitful exercise to expose
vital components of the EU constitutional order’ (Hillion 2007: 269)—or better
yet, to expose vital components of the EU’s contested and evolving constitu-
tional order.
This part of the conclusion summarizes the book’s empirical findings in four
steps. First, it reviews the patterns evident in various quantitative analyses of
every EU decision on the membership eligibility of aspiring and applicant states.
Second, it reviews how the membership norm prevailing in each of four normative
eras shaped community decision-making on various cases in that era. Third, to
illustrate the process, it reviews the quest of one country (Greece) for community
membership across the first three of the four normative eras. And finally, it
reviews the apparent explanatory power of each of the principal alternative
explanatory factors.
Quantitative analyses
As outlined in Chapter 4’s exploration of the 48 cases evident in the historical

record since 1957, there is extensive statistical support for the argument that
membership norms play a critical role in EU decisions on the membership
eligibility of aspirant states. In a cross-tabulation analysis, the explanatory fit of
the membership norms argument was nearly identical to a hypothesis that the EU
Rethinking Europe, Rethinking Regions 221
is open to any state seeking membership and to two tests of the geography
argument. Just as important, the membership norms argument performed
better—and sometimes far better—than various tests of the treaty rules, regime
similarity, territorial contiguity, commercial interests, and security interests
arguments outlined in Chapter 2. Two robustness checks—one using logistic
regression and one using crisp set Qualitative Comparative Analysis—also
supported the membership norms argument. Given how well established the
alternative logics are in the literature on EU integration and decision-making,
these multiple statistical confirmations of the power of membership norms beg
further exploration.
Normative eras and decision-making effects
Evidence for the impact of membership norms on EU decision-making is not

limited to statistical results. As shown in Chapters 5–8, careful tracing of the EU
decision-making process regarding four states seeking to join the community also
supports the argument that decisions on membership eligibility were shaped
significantly by membership norms and by their absence post-2006. In the first
half-decade after the Treaty of Rome, when prevailing membership norms
defined the EEC as a community of non-Communist European states, the Six
welcomed applications from Greece and Turkey and encouraged Spain to apply,
despite a military takeover in Turkey and persistent authoritarian rule in Spain.
In the second half- decade, when prevailing norms redefined the EEC as a
community of parliamentary democracies, the Six remained open to Turkey,
which had since returned to democratic rule, but rejected Spain and Greece
because of their authoritarian regimes. Around 1970, the community again
redefined itself around a norm of liberal democracy including respect for human
rights, which enabled democratizing governments in Greece, Spain, and even
Turkey to overcome resistance among certain member states and to gain
recognition as eligible for membership.
The normative consensus broke down in late 2005 and has not been replaced
since then by a new membership norm. As member states competed in the
post-2005 era over the membership eligibility of various states, they often framed
their preferences with arguments about which types of states should be eligible
for membership and how the ultimate geographic limits of the community should
be defined. But in the absence of any single prevailing membership norm, these
references could do little more than maintain a shared sense of purpose among
like-minded member states and perhaps reassure like-minded voters at home.
Within EU-level decision-making, they could not constrain the behaviour of
member states with different preferences. This normative dissensus yielded
incoherent responses to Ukraine’s quest for membership and a deadlock on

Turkish accession, well before the start of Ukraine’s civil war or Ankara’s turn
away from reforms designed to promote democracy, the rule of law, and respect
for human rights.
An illustration: Greece and Europe over time
Case studies of EU decision-making on Greece across multiple normative eras

confirm the norms–policy relationship hypothesized in Chapter 2. Looking back
at Greece’s quest for community membership from the initial application for
association in 1959 through its successful pursuit of membership eligibility in the
mid–late 1970s and its accession in the 1980s, it is undeniable that various factors
influenced decision-making by individual member states and the Community as
a whole. However, detailed tracing of community decision-making on Greece
identified three key turning points that are best explained by the presence of
particular membership norms at each point in time and the opportunities and
constraints they established for governments and other decision-makers. In fact,
each of the three sub-cases exemplifies a distinct way in which membership
norms contributed to membership outcomes.
The EEC’s initial openness to the Greek government’s quest in 1959 to join the
process of European integration was less a calculated policy choice than an
instinctive response to a taken-for-granted understanding that Greece belonged
in a community of Europe’s non-Communist states. By what means and how
soon Greece should be integrated was the subject of much discussion, but the
country’s fundamental right to join the new community was never questioned.
This pattern is exactly what one would expect if EEC decisions on Greece were
being shaped by a non-Communist membership norm.
The same outcome might also be explained as an effort to bolster a NATO ally
vulnerable to internal subversion and external attack, but if such geopolitical
factors had been dominant, we would expect to find much clearer evidence
of decision-makers weighing geo-political gains against economic costs and
prioritizing the former over the latter. Alternatively, if EEC decision-making
had been dominated by economic considerations, the Six would probably have
opted for a much looser association agreement oriented at promoting trade and
investment rather than preparing for membership. After all, the Commission
repeatedly pointed out that Greece was too small and underdeveloped to be a
major market for industrial products from the Six, while domestic producers in
France and especially Italy would compete directly with Greek agricultural
exports.In contrast, the EEC’s decision to freeze Greece’s membership eligibility
following the military coup in 1967 was deeply contested. Yet here too the impact
of the Community’s membership norm was clearly evident in the language used
by those seeking to put Greece’s relationship to Europe on hold: the ability to link
this goal to the redefined sense of the EEC as a community of parliamentary
democracies enabled otherwise-weak actors such as European parliamentarians
and trade union leaders to stop the clock on Greece’s move toward Europe and to
condition any new movement on a return to democratic rule. This outcome is
hardly explicable in terms of geo-political or Cold War necessity.
Following the collapse of the junta in 1974, economic incentives were mostly
negative and geo-political incentives—namely, the priority of stability on NATO’s
south-eastern flank—did not dictate that the Community must accept the Greek
government’s insistence that it deserved a fast track to full membership
unconditioned by the Greek economy’s actual readiness for the competitive and
regulatory challenges of the common market. Even if the Community wanted to
ensure the stability of Greece’s newly democratic government, it could have done
so by reaffirming the accession commitment and timetable made in the 1961
Athens Agreement while helping Greece to institutionalize civilian control of its
armed forces and the independence of its judiciary and by promoting the
structural reforms and policy harmonization that had been delayed by the
post-1967 freeze. The fact that the EEC did not pursue this more modest course,
and instead placed Greece on a path to rapid accession, is explained by
Karamanlis’ ability to silence member states with contrary preferences by
highlighting his country’s newly- restored fit with the Community’s liberal
democratic membership norm.
Alternative explanatory factors
Notwithstanding the explanatory power of membership norms, the book’s

statistical and process-tracing analyses provide considerable evidence that EU
enlargement has also been shaped by other factors. To truly assess the member-
ship norms explanation, we must therefore compare its explanatory power to
that of potential alternatives. So what can we conclude from this study about
the relative explanatory power of alternative approaches to EU decision-making
on membership and enlargement? In other words, how does the power of
membership norms compare to the power of other, more familiar factors such as
formal rules, physical geography, commercial, and geo-political incentives?
While patterns of EU enlargement have certainly been shaped by the Union’s
formal membership rules, this book’s cross- tabulation analysis reveals a
correlation between rules and outcomes that is less strong than that between
membership norms and outcomes, while the process-tracing indicates that formal
rules were no more than a permissive factor. For example, the fact that Article 237
of the Treaty of Rome authorized ‘any European state’ to apply for membership
helps explain why so many states quickly expressed interest in accession, but it
cannot explain why some applicants were offered association agreements designed
to facilitate eventual membership while others were refused. As one observer put
it, ‘The hardest rule is the geographical criteria but this has turned into one of
those “I know it when I see it” rules. . . . As is often the case in the EU, it is not the
formal rules that matter most’ (Baldwin 2007).
The case studies also show the significance and the limitations of commercial
and security calculations. On the one hand, the process-tracing demonstrates that
member states were sensitive to the commercial and security consequences of
various enlargement choices. In fact, such calculations may have been the
dominant determinant of particular member states’ positions on the desirability
of particular applicants. However, the general pattern of EU collective decision-
making shows that when commercial or security interests conflict with the
community’s prevailing membership norm, member states have repeatedly
compromised these interest calculations in order to ensure an outcome consistent
with the norm.
In fact, the case studies offer little support for the hypothesis that the decision-
making behaviour of member states is determined by the short-term interests of
sectors within their national economy threatened by competition from applicant
states seeking to join the Common (and later Single) Market. For example, France
and Italy had strong commercial incentives to block Greece and Turkey’s access to
the Common Market, and the case studies show that French and Italian diplomats
often acted in accordance with these incentives. But the short-term commercial
interests hypothesis cannot explain why France and Italy ultimately chose not to
veto association agreements for Greece and Turkey, nor why France strongly
supported Spain’s initial application despite the comparable threat posed by that
country’s agricultural producers.
This raises the question of whether the member states were perhaps prioritizing
long-term over short-term commercial interests. As discussed earlier, it is not
easy for the governments of democratic states to ignore the interests of key
commercial sectors. Nonetheless, the six case studies provide mixed evidence
regarding the hypothesis that the prospect of long-term gains from trade drives
member states’ decision-making on enlargement. A September 1959 report by
the ECSC High Authority recognized that full EEC membership for Turkey
would challenge Italian and French agricultural exports but indicated that these
effects would be outweighed over time by the large and growing Turkish popula-
tion’s demand for industrial goods.3 Likewise, a 1964 French foreign ministry
document estimated that the agricultural competition produced by Spain’s entry
into the Common Market would be relatively limited and largely compensated by
the opening of the Spanish market to French exports (Ministère des Affaires
3 AHCE: CEAB/5/792: Note pour Monsieur le Président, 21 septembre 1959.

Étrangères 2002, document 180). However, this putative long-term commercial

rationality cannot explain why the EEC refused Spain’s application in 1962 or why
the EEC froze Greece’s progress toward membership in 1967, just six years after the
Athens Agreement recognized that country’s eligibility for eventual membership.
Finally, the statistical analyses reveal an explanatory fit of only 22–49 per cent
for the hypothesis that the EU would accept applicants whose membership
seemed likely to reduce the current member states’ exposure to war and/or
strengthen the security alliance to which member states belong, depending on the
specific hypothesis and indicator. The process-tracing evidence shows that geo-
political considerations often figured into decision-making, but some crucial
cases remain puzzling. All six of the EEC member states during this period
belonged to NATO, whose principal purpose was to deter or defeat a military
attack by the Warsaw Pact. As expected by this hypothesis, there was considerable
anxiety among the Six regarding the strategic reliability of Greece and Turkey,
which also belonged to NATO. For example, the Council of Ministers resolved in
September 1960 to inform NATO that, in accordance with the North Atlantic
Council’s call for Alliance-wide efforts to address Greece’s and Turkey’s economic
difficulties, it was pursuing association talks with Athens and open to talks with
Ankara.4 However, this hypothesis cannot explain why the Six would reject the
application from Spain, which was critical to NATO plans as a location for air
bases and strategic depth in case of a Soviet attack, nor why they would freeze
Greece’s association in response to the coming-to-power in 1967 of a strongly
pro-NATO government.
Another explanation of EU enlargement decision-making may lie in analysing
the bargains that member states make when considering particular applicants.
By this logic, derived from public choice theories of parliamentary vote-trading
or ‘log-rolling’ (Miller 1977), member states would trade support for one mem-
bership applicant in exchange for support on another applicant or even on an
unrelated issue. If enlargement decision-making proceeded in this manner,
the EU’s response to states seeking membership would be explicable in terms of the
dynamics of specific and diffuse reciprocity (Keohane 1986) but not the simple
preferences of the various member states. Instead, when some member state
(or states) opposes a particular applicant, we would expect the EU nonetheless to
offer membership if the applicant’s supporters were willing and able to offer its
opponent(s) valuable concessions on other issues. Alternatively, following diffuse
reciprocity, opponents of a given applicant may yield to its supporters without a
specific quid pro quo if they are confident that they will be rewarded later
with concessions that they will value as much as or more than the current
membership choice.
4 AHCE: BAC/26/1969/262: Note de la Délégation Néerlandaise, 23 septembre 1960.

Testing this log- rolling argument is difficult because such bargains and
expectations are rarely mentioned in official speeches or recorded in the minutes
of EU meetings, so decisive empirical evidence is not easily found. Nonetheless,
there are multiple incidents in the book’s case studies where a member state
government had powerful reasons to deny (or award) a membership perspective
to an aspirant state but nonetheless adjusted its position to fit the prevailing norm.
For example, according to Eurobarometer, the Austrian population in 2004–05
was deeply sceptical of Turkish accession, more so than the population of any
other member state. Despite this powerful incentive, Chancellor Schüssel
repeatedly backed down from his demands that the EU offer Turkey a partnership
agreement instead of membership.5 Some observers attributed Schüssel’s final
reversal on Turkey to a last-minute backroom deal in which he secured a positive
decision on Croatia, but this interpretation is undermined by strong evidence
that other factors shaped the Croatia decision, as discussed in Chapter 7. The
analytical focus on membership norms thus appears to offer a superior
explanation of EU decision-making on enlargement than a range of alternative
approaches, including those focused on formal rules, geo-political calculations,
commercial interests, and even inter-governmental bargaining and reciprocity.
2. Understanding regions and regional integration
This book demonstrates the explanatory value of an ideational approach to the

composition of regions and the dynamics of regional integration. More
specifically, it shows that the limits of regional communities are not given by
geography or even by culture—they are defined within regional organizations by
member states’ governments plus supranational actors deliberating over a
common definition of the characteristics that members and potential members
are expected to share. The membership norms that emerge in these deliberations
create powerful incentives for those seeking to promote or block an applicant
state, so focusing on the content and evolution of these norms offers valuable
insights into how regional communities decide which states are eligible to join
the club and where the club’s ultimate geographic limits lie.
In so doing, the book demonstrates the political importance of two purely
intersubjective and potentially recurring moments in any regional integration
venture—the moment(s) when members of a regional community, whether
existing or in-formation, define what type of state is eligible or ineligible for
membership, and the moment(s) when they announce whether a particular state
is eligible or ineligible for membership. These acts of regional construction have
5 Why Austria was a sticking point, BBC News Online, 3 October 2005.
no tangible dimension, and sometimes not even legal status. Their political
significance is nonetheless clearly evident in various actors’ efforts to shape and
re-shape their community’s standards of eligibility, and to obtain or block a
declaration of eligibility for a non-member. As such, they deserve far more
attention from scholars of regional integration and governance, who tend to focus
their analytical energies, if they study membership politics at all, on an applicant
state’s decision to apply for accession, on community decisions regarding an
applicant’s readiness for accession (distinct from its eligibility for membership),
and/or on how the quest for and experience of membership affects the structure
and policies of the candidate (or later member) state.
This theoretical conclusion builds on ideational or norm-focused understandings
of international institutions and space. Contrary to the focus on institutional
design that dominates traditional analyses of international institutions, John
Ruggie argued that we know them ‘not simply by some descriptive inventory of
their concrete elements, but also by their generative grammar, the underlying
principles of order and meaning that shape the manner of their formation and
transformation’ (Ruggie 1998: 63). Likewise, argued Ruggie, ‘space is not given in
nature. It is a social construct that people, somehow, invent’ (Ruggie 1999: 235).
As such, regions should not be understood ‘simply as abstractions or as a priori
spatial givens, but instead as the results of social processes that reflect and
shape particular ideas about how the world is or should be organized’ (Murphy
1991: 24). This book’s findings thus reinforce prior evidence of the heuristic
value of this approach in international relations, such as Ruggie’s (1982, 1991)
explanations of the stability of the post-Second World War international
order and Hemmer and Katzenstein’s (2002) explanation of why there is ‘no
NATO in Asia’.
The conventional norm-focused approach is limited, however, by its tendency
to over-state the taken-for-granted nature of normative structures and thus to
under-estimate the potential for contestation that is inherent in any normative
order. After all, the politics of international norms involves not only the manage-
ment of compliance with social expectations—it also involves the contestation of
ideas and their link to structures of power and resistance (Acharya 2004, 2009,
Wiener 2004, 2008, 2014). The challenge for social science is how to incorporate
norm contestation into a theoretical framework that can explain important
political decisions and institutional developments—in other words, how to
examine the consequences of normative structures at particular historical
moments while recognizing the dynamics of normative contention and avoid-
ing the positivist temptation to reduce norms to causes that uniformly produce
particular outcomes.
In this regard, it is likely futile to attempt to theorize and document the
‘identity’ of regional organizations and communities (Thomas 2017). If an identity
is composed of ‘relatively stable, role-specific understandings and expectations

about the self ’ (Wendt 1992: 397), then the concept is best applied to individuals
and perhaps to groups of individuals but it cannot encompass the internal variety
and multi-level, dynamic complexity of a regional community of states. For exam-
ple, Acharya’s (1997) early definition of an emerging Asia-Pacific regional identity,
which he linked to four principles extrapolated from the ‘ASEAN way’ of inter-
governmentalism, overlooks the inexorable contestation of international norms
that is the focus of Acharya’s later work.
A similar point can be made about Schimmelfennig’s (2001) definition of
European identity as an ensemble of leaders’ values, the norms they share, and
the formal rules that they must obey. Although perhaps useful for examining
periods of time where contestation is relatively low, the general explanatory value
of this definition is seriously undermined by the fact that each of Schimmelfennig’s
three dimensions is potentially connected to different causal mechanisms and
scope conditions. This did not weaken his analysis because he focused on a region
where and a time when contestation was indeed low—Europe in the years
immediately after the fall of Communism and the end of the Cold War—but it
does limit the argument’s generalizability to other regions and across time.
More recently, Acharya offered a revised definition that merits more extensive
consideration: ‘The regional identity of Southeast Asia, one that yields the notion
of Southeast Asia as a distinctive region and sets it apart from neighbouring
regions such as South Asia or Northeast Asia, is not a given, and is not
preordained. Nor is it based merely on the facts of geography, or shared historical,
political, and cultural features and experiences. These are important but not
sufficient conditions for regional identity. Rather, Southeast Asia’s identity, which
is the basis of the identity of the Association of Southeast Asian Nations (ASEAN)
as a regional organisation, is socially and political constructed, through
interactions amongst its governments and societies. To the extent that it is a
contrived but meaningful notion, ASEAN identity is also subject to challenge and
change due to changing political, strategic, and economic currents in the region
and beyond’ (Acharya 2017: 25). Interestingly, here Acharya has reversed his
earlier explanatory claim: Southeast Asia’s identity is now the basis of ASEAN’s
identity, rather than vice-versa. More important, compared to his definition of
two decades earlier, this one is far more complex and cognizant of the reality of
norm contestation and evolution.
In these regards, Acharya’s new definition resembles the European
Commission’s (2006) official definition of ‘European’—a key but undefined term in
EU treaties: ‘The term “European” combines geographical, historical and cultural
elements which all contribute to European identity. The shared experience of
ideas, values, and historical interaction cannot be condensed into a simple
timeless formula and is subject to review by each succeeding generation.’6

Although clearly convenient for a regional organization faced with competing
political pressures, this approach is so inclusive that it cannot answer common-
sensical questions about the limits of a region (in this case, Europe) and cannot
generate strong testable propositions for studying regional (European) integra-
tion. The basic problem is that ‘identity’ is just not sufficiently precise or differen-
tiated to yield valuable insights into the dynamics of regional communities.
Instead, this book incorporated both the enabling/constraining power of
norms and the contestation/evolution of norms into a more precise and coherent
theoretical framework. The resulting concept of membership norms can accom-
modate the fact that regions are more social than material in nature and yet sub-
ject to discursive contestation and change over time, while remaining distinct
from the beliefs and values of individuals and the formal rules of institutions.
To assess the concept’s explanatory power, the book’s research design paired a
punctuated equilibrium model of normative change with a concept of normative
effects including both constraints and empowerment, followed by both statistical
and process-tracing explorations of important outcomes in European integration
and enlargement. Just as important, it compared the relative explanatory power of
membership norms concept to other explanations of regional dynamics focused
on physical geography, material interests, and formal rules.
More work of this type can deepen our understanding of the truly dynamic
nature of regional integration including how the ideational construction of a
region shapes policy choices and how those policy choices in turn reshape the
construction of the region itself. In the process, future scholars will undoubtedly
develop more precise conceptualizations of the sources and dynamics of
normative contestation and change and better ways to leverage multiple empirical
methods to assess the explanatory power of their theoretical expectations.
3. The future of Europe
The arguments and empirical findings presented above have significant implications
for salient topics in contemporary European governance, such as the future of EU
enlargement, relations with Turkey, the dynamics of ‘differentiated’ integration
and dis-integration, the EU’s role as a bastion of liberal democracy, and the orien-
tation of EU foreign policy—all themes that are likely to feature prominently in
the EU’s upcoming multi-year Conference on the Future of Europe, which aims
6 European Commission. Europe and the challenge of enlargement. Bulletin of the European
Communities, Supplement 3/92, Prepared for the European Council, Lisbon, 26–7 June 1992, p. 11.
to address, in the words of French president Emanuel Macron, ‘all the necessary
changes to our political project, without any taboos, not even treaty revision’.7
The future of enlargement
The processes of enlargement decision-making examined in this book are very

relevant to Europe’s future. In addition to Turkey, whose accession negotiations
have been effectively frozen for years, there are states in the Balkans whose
membership eligibility the EU has already confirmed, as well as Georgia,
Moldova, Ukraine, and perhaps others that hope someday to be recognized
formally as eligible for membership. These states are unlikely to stop knocking on
the EU’s door. As Northern Macedonia’s Foreign Minister Nikola Dimitrov points
out, ‘For those of you who are inside it’s easy to forget how cold it is outside’
(Munich Security Conference Foundation 2019). And while the characteristics of
these potential candidate states differ considerably, the EU has not been engaged
in a nuanced discussion of their individual merits. Instead, it has been wracked in
recent years by fundamental debates over whether or not to remain open to new
members, and if so, how future decisions should be made.
These debates are sometimes linked, at least implicitly, to regrets over the
prominent role played by membership norms in past decisions. For example,
Valéry Giscard d’Estaing argued in 2012 that Greece is ‘basically an Oriental
country’ and therefore should never have been offered community membership,
regardless of its democratic credentials.8 (This argument is particularly striking
given that Giscard was a strong advocate of Greek accession while President of
France in the mid 1970s.) And Giscard’s scepticism was not focused solely on
Greece: along with many others, he was also an outspoken critic of accession
negotiations with Turkey, which he insisted is not EU-eligible for cultural and
geographic reasons.
Yet it was not by chance that the EU judged these two countries eligible for
membership. As shown in earlier chapters, EU leaders made deliberate choices on
multiple occasions in both the Greek and Turkish cases based in part on what was
normatively appropriate at the time. The community welcomed Greece when it fit
the ‘non-Communist’ norm in 1959–61, suspended Greece when it violated the
‘parliamentary democracy’ norm following the military coup in 1967, and then
reluctantly welcomed Greece again following the collapse of military rule in 1974
when the country’s new government linked its accession ambitions to the
community’s ‘liberal democracy’ norm. A similar pattern is evident in the Turkish
case since 1959, where calculations of economic or political gain were repeatedly
subordinated to the demands of membership norms. Whether or not a more
7 Conference on EU future edges closer—but with clear limits, Politico.eu, 24 June 2020.
8 European Luminaries Reflect on Euro, Spiegel Online International, 11 September 2012.
interest-focused process would have yielded preferable outcomes in particular

cases is difficult to assess and anyway beyond the scope of this project.
However controversial particular enlargement moves may be in hindsight, the
finding that they were shaped significantly by membership norms does not imply
that such norms are best avoided in future decision-making. As discussed in the
first chapter, regional communities face no more important choices than those
they make regarding which non-member states to welcome into the club and
which to reject. And there is no reason to expect that the log-rolling that typically
dominates inter-governmental bargaining would necessarily yield results that are
superior to a process that conditions accession on agreements among member
states regarding the fundamental purpose of their community. Consider the
uneven quality of another series of critical decisions made through inter-
governmental bargaining in the presence of far more complete information—the
selection of presidents of the European Commission.
Most important, the decisions on eligibility for membership analysed in this
book are distinct from subsequent decisions on various states’ readiness for
accession, and it is the latter decisions that have engendered such controversy. It is
undeniably true, as shown in some of the case studies, that decisions on eligibility
sometimes placed the community’s member states and institutions in a rhetorical
trap that made it harder for them to resist calls for quick accession, even when the
objective analyses suggested that some delay was needed. Yet however strong
rhetorical or normative entrapment may be, the real problem does not lie there.
The first and over-riding problem is the member states’ unreadiness to insist
that candidate states must institutionalize their commitment to the norms that
had helped them to be recognized as EU-eligible in the first place. As a result,
compliance with EU expectations often dropped off soon after accession. The
second problem is the tendency of member states’ governments to blame ‘the EU’
for decisions that they helped to make, rather than taking responsibility for their
decisions and explaining why they considered a particular state’s accession to be
appropriate or advantageous. This tendency is obviously explicable in political
terms, but its result is a widespread public disenchantment with the entire idea
and practice of enlargement.
Long after the ‘big bang’ enlargements of the early 2000s, there is much talk of
‘enlargement fatigue’ among voters—a phenomenon that some of today’s member
state governments seem to welcome and others seem unwilling or unable to
challenge. This sentiment may be linked to the fact that enlargements have
generally worsened economic inequality between the EU’s regions and member
states, and a fear that the only way to forestall mass migration from Turkey, the
western Balkans, or Ukraine would be to institute equally-unwelcome transfers of
wealth from richer to new, poorer member states.9 There is thus widespread
9 Branko Milanovic, Where are the limits of Europe? 14 May 2019, https://www.socialeurope.eu/
where-are-the-limits-of-europe.
c oncern that proceeding further with enlargement will bolster support for populist
and nationalist parties on the far-right and far-left whose agendas are inimical to
European integration. For this reason, Jean-Claude Juncker famously declared in
2014 that there would be no further enlargement during his term as Commission
president and there are no accessions on the immediate horizon under the current
Commission. On the other hand, the EU has not declared that enlargement is
finished, nor has it revoked the membership eligibility granted earlier to a number
of states, so formal negotiations and preparations continue.
Officially, the EU has reformed its enlargement practices to avoid certain
deficiencies seen in earlier rounds, including a tendency to treat candidate states
in groups that allowed reform laggards to hide behind reform leaders, and the
tendency to announce target dates for accession well before negotiations and
reform were completed, which made it difficult to stop the process when reforms
were lagging.10 For example, at its Thessaloniki summit in 2003, the EU promised
‘unequivocal support for the European perspective’ of states in the Western
Balkans while clarifying that ‘the speed of movement ahead lies in the hands of
the countries of the region’.
In practice, however, due to the breakdown of normative consensus after 2005,
individual member states no longer face pressure to avoid conditioning the
further unification of Europe on their particular calculations of self-interest, so
inter-governmentalism dominates collective decision-making. This has made it
easier for individual member states to block applicant states’ progress toward
accession, regardless of the latter’s actual commitment and record in meeting EU
expectations. And there are plenty of opportunities for these vetoes to be exercised
behind closed doors: ‘each member state has at least 70 occasions to pause or
block accession negotiations (such as during the opening and closing of each of
the 35 negotiating chapters . . .) Multiply this by 28 member states and it turns out
that both in theory and in practice, member states have about 2,000 different
opportunities to block the negotiation process for EU accession.’11
This kind of unrestrained intergovernmentalism was evident in the struggle
during 2019–20 over whether to open accession negotiations with Albania and
North Macedonia, the two Balkans states that had progressed furthest in their
reforms. Among others, Austria, Germany, and the Visegrad four (the Czech
Republic, Hungary, Poland, and Slovakia) backed the European Commission’s
judgement that talks should start, while France blocked North Macedonia and
Denmark and The Netherlands blocked Albania. France was especially outspoken
in its call to rethink the enlargement decision-making process, but critics charged
10 Romania and Bulgaria were not ready for accession, EU auditors confess, Euractiv.com, 13
September 2016; Let’s set things straight: Accession talks do not equate EU membership promise,
Euractiv.com, 24 June 2019.
11 Let’s set things straight: Accession talks do not equate EU membership promise, Euractiv.com, 24
June 2019.
that Macron’s government was motivated first by domestic concerns. European

Commission President Jean-Claude Juncker described the French veto as ‘a grave
historic mistake’ and then repeated himself in French to make his point, while
European Council President Donald Tusk declared, ‘Let me be very clear: North
Macedonia and Albania are not to blame for this.’12 Counter-proposals by the
European Commission and a number of member states eventually persuaded
France to lift its veto, but the message emerging from this episode was clear: the
EU’s attitude toward aspirant and candidate states is increasingly delinked from
their actual governance records.13
In the EU’s accession queue, states that have gone the furthest with reforms
expected by the EU are actually lagging behind states that are struggling to reform
their political and economic systems (Emerson and Noutcheva 2018). Due to
lobbying from several member states, the final statement from a May 2020
summit of leaders from the EU and Western Balkans states referred to the region’s
‘European perspective’ but omitted any reference to enlargement.14 However
convenient this may be for member state governments concerned about how to
sell enlargement to a sceptical public or parliament, it casts doubt on the
community’s prior commitments and future credibility. In the absence of evidence
that the EU will deliver a membership perspective, much less a final deal on
accession, there is little incentive for the governments of aspirant or candidate
states to make costly reforms. Even worse, giving each member state an easy veto
invites each of them to hold membership candidacies hostage to the narrowest of
national interests, or to national moods expressed via referenda, with little regard
to broader community interests.
Consider for example how inter-governmentalism has created a link between
the EU- eligibility (for now purely hypothetical) of Catalonia and Scotland.
Several years before the UK’s referendum on Brexit, Scottish nationalists began to
argue that an independent Scotland would immediately seek EU membership,
but not by the conventional means of applying under Article 49 of the Treaty on
European Union. Instead, they said, Edinburgh would use Article 48, which
allows the treaties to be amended by unanimous consent, to seek immediate
accession as a direct continuation of Scotland’s membership as part of the UK.15
This idea was immediately rejected by Spanish Prime Minister Mariano Rajoy:
‘I know for sure that a region that would separate from a member state of the EU
would remain outside the EU.’16
12 France halts EU enlargement, Euractiv.com, 16 October 2019; EU’s Balkan breakdown reveals
split among leaders, Politico.eu, 21 October 2019.
13 9 EU countries push back on French enlargement revamp, Politico.eu, 13 December 2019; EU set
to modify member accession method to placate France, Euractiv.com, 5 February 2020.
14 Western Balkans summit encourages ‘European perspective’ but omits enlargement, Euractiv.
com, 5 May 2020.
15 Scotland maps out own path to EU membership, European Voice, 28 November 2013.
16 Scottish leader hits back at Spain on independence, EUobserver.com, 28 November 2013.
By signalling that Madrid would never allow Scotland to rejoin, Rajoy was
sending a message to separatists in his country that independence could come at
the cost of exclusion from Europe. This move led a Catalan politician to accuse
the Spanish government of ‘projecting its own internal phantoms in third
countries and different situations’.17 In any case, nowhere did the Spanish
government appear to consider how its preferences would affect the EU as a
community of democratic states committed to respect human rights. Then,
several years later, apparently hoping to gain EU support for its aims on Gibraltar’s
status after Brexit, the Spanish government suggested after all that it might not
block a Scottish application.18 Yet here too the message was clear: Spain would
threaten or withhold its veto over an applicant state’s EU-eligibility based not on
community interests or values but on its own territorial concerns.
This incoherence may be explicable in terms of inter-governmental bargaining
and issue-trading, but it leaves outsiders wondering ‘how the EU will react to
neighbours who share its values and are keen . . . to become a fully integrated
members of the club of European democracies’.19 For this reason, a recent report
by the Open Society Foundations proposes that decision- making on the
intermediary stages of accession negotiations, such as monitoring reforms by
applicant states, should be based on qualified majority voting (55 per cent of
member states representing 65 per cent of the EU population), as is already done
for decisions on visa liberalization (Cvijic et al. 2019). This would leave member
state governments and national parliaments responsible for approving accession
treaties on a unanimous basis while reducing the risk that national preoccupations
undercut negotiations that are advantageous to the community as a whole. After
all, the breakdown in normative consensus post-2005 has not been total: the
majority of member states remain committed to liberal democratic values.
The reform proposed by the Open Society report would thus make it more
likely that the unification of Europe proceeds on the basis of enduring principles
rather than on the basis of ad hoc bargains with unpredictable effects. What it
would not address, of course, is the problems that arise when the EU fails to invest
sufficient resources and political capital in ensuring that current member states
also remain committed over time to the community’s official values, with credible
remedies available in case they do not. As one of the same authors warned
separately, ‘Balkan nations aspiring to join the EU are watching Orbán and
friends establish one-party states inside the EU’s borders with interest. Should the
EU let Hungary slide further into authoritarianism, Brussels can be sure the
Balkans will follow.’20
17 Spain to Scotland: You’re not special, Politico.eu, 3 February 2017.

18 Spain eases opposition to an independent Scotland in EU, Reuters.com, 2 April 2017.
19 Georgia’s European Economic Area Ambitions, Euractiv.com, 22 July 2019.
20 Macron and the EU enlargement make-believe, Euractiv.com, 24 May 2019.
On the other hand, even if EU member states were to rebuild the previous
consensus on liberal democratic norms and invest the resources needed to back
them up, they would still have to find a solution to economic inequality within
the community that convinces their citizens to support further enlargement.
Without this, a new normative consensus at the intergovernmental level could
quickly succumb to resistance from below.
The Turkey question
The preceding discussion raises the perpetually vexing issue of Turkey. As

discussed in Chapter 8, Turkey gained explicit recognition of its EU eligibility in
1999 and accession negotiations opened in 2005. Progress towards accession has
been blocked by the subsequent breakdown of the EU’s membership norm and by
the Turkish government’s turn away from liberal democracy, the rule of law, and
respect for human rights, most clearly since the aborted coup attempt of 2016.
Turkey thus appears to be stuck between candidacy and membership, with little
hope of accession regardless of how the current crisis is resolved, what reforms
Ankara adopts, or how many times it mentions the EU’s earlier promises.
We thus have good reason to doubt Alexander Bürgin’s (2010) claim that
entrapment dynamics have been too strong for opponents of Turkish accession to
overcome, as well as Frank Schimmelfennig’s (2009) expectation that if Turkey
resumes its earlier path of liberal reform, the EU will be ‘entrapped again’ by its
rhetorical commitments. This nonetheless begs the question of what developments,
if any, could end the current deadlock and thereby clarify whether or not the EU
still considers Turkey eligible for membership.
The simplest scenario would involve the member states re-converging around
a new membership norm, a new definition of their community that encompasses
rather than excludes Turkey. For example, although there is little indication now
of this happening any time soon, the EU could interpret its official motto—
‘United in diversity’—in a way that emphasizes cultural and religious pluralism.
Or, the EU could adopt an updated version of the membership norm that
prevailed in the late 1950s and thereby open the doors to all European states that
resist Russia’s territorial ambitions and political subterfuge. Yet here too the
member states are deeply divided and Turkey’s position vis- à-
vis Russia is
increasingly ambiguous. As such, a normative re- embrace of Turkey seems
unlikely.
Another possibility would be that EU member states re-converge on a strong
liberal democratic membership norm and agree that decisions related to
membership must be determined by that norm rather than by other geo-political
considerations such as the unity of NATO or the continuation of bilateral deals
with Ankara. If this were to happen, and assuming no other significant
reorientation in Turkish domestic politics, then the EU would likely decide that
Turkey no longer qualifies for a membership perspective, as it did on Greece fol-
lowing the military takeover in 1967. This could result in a formal freeze on all
discussions of accession, or even a definitive decision to replace accession negoti-
ations with an alternative partnership deal, as some member states have been
proposing for years. This scenario is unlikely to happen soon given the continued
strength of illiberal governments within the EU that reject any EU role in policing
the domestic political behaviour and constitutional choices of member states,
even though many of them would actually welcome a decision to end all possibil-
ity of Turkish accession.
A third possibility would be that EU member states re-converge on a clear
membership norm, but instead of the liberal democratic norm, it would be the
kind of conservative EU-as-defender-of-Christian-civilization norm advocated in
recent years by governments in Hungary, Italy, and Poland. If this were to happen,
then a definitive end to accession negotiations with Turkey and denial of any
future membership prospect would be virtually guaranteed. Given the once-
unthinkable move away from liberal values in various member states, and an
international environment increasingly conducive to nativist appeals, this
possibility cannot be excluded. However, the continued commitment to liberal
values in many member states makes this scenario also rather unlikely.
In a fourth scenario, the current deadlock on Turkey’s membership prospects
would end without any revision of EU membership norms. This would involve a
domestic political reorientation in Turkey and the emergence of a new
government fully committed to putting the country back on the liberal reformist
path of the late 1990s and early 2000s. Such a government might link its ambitions
to the EU’s formal commitments to Turkey, lobby the EU to reconfirm its
membership eligibility and to re-start accession negotiations. Recent developments
in Turkey suggest that the AKP’s hold on power may be slipping, so political
change in the coming years cannot be excluded. However, such a strategy would
only succeed if EU governments can be entrapped by their predecessors’ formal
commitments to Turkey in the absence of any normative consensus among them
that EU eligibility depends upon liberal governance. This is possible (Schimmelfennig
2009) but unlikely unless the new government manages simultaneously to reverse
the anti-Turkey sentiment that now seems widespread and deeply-rooted within
many member states.
There is, however, one more scenario in which the current deadlock is broken
without any new EU membership norm or change of government in Turkey. As
seen in Chapter 8, member states have depended since 2016 on the EU’s migrant
readmission deal with Turkey to shield them from domestic backlash over
uncontrolled immigration. Opponents of Turkish accession take comfort in the
fact that negotiations are frozen, supporters of the principle of Turkish accession
take comfort in the fact that Turkey’s EU-eligibility has not been revoked, and
governments on both sides benefit domestically from limiting the inward flow of
migrants. This win–win situation underlies the tacit agreement among member
states not to push hard for either advancing or cancelling Turkey’s EU membership
perspective.
However, Turkey’s foreign minister announced in 2019 that Ankara would
suspend the readmission deal out of frustration with the EU’s limited acceptance
of migrants screened within Turkey.21 This has not yet happened but member
states’ acceptance of the status quo in relations with Turkey is likely to evaporate if
the deal does collapse and massive inflows of migrants resume through the
Balkans and across the Aegean Sea. In this case, barring other changes, it is easy
to imagine the EU deciding once and for all to reverse its prior membership
commitments to Turkey and replace them with some sort of partnership
agreement based on overlapping and complementary interests. On the other
hand, if the migrant deal survives, normative dissensus continues to prevail
within the EU, and the political status quo within Turkey persists, the deadlock in
EU–Turkey relations could continue for years.
Differentiated integration and disintegration
For the sake of theoretical clarity, this book’s argument has assumed that states are
either in or out of a regional community, either members or non-members. In
reality, though, when states consider participation in regional integration, they face
more complex choices than simply whether or not to join a regional community.
For example, association agreements of various types grant non-members
access to some aspects of the community without the privileges or obligations of
full membership. However, as seen clearly in several of the EU cases outlined in
this book, association agreements are sometimes a potential stepping stone to
accession, so member states or community institutions may insist that states
seeking association also comply with the community’s membership norms. In
such situations, the reach of regional membership norms extends beyond the circle
of states that already belong to or aspire soon to join the community in question.
An even more complex exception to the in-or-out assumption lies in the pos-
sibility of ‘differentiated integration’, whereby states choose to join the regional
integration of certain policy areas while leaving other areas under national con-
trol, or accept supranational authority where others prefer inter-governmental
decision-making (Leuffen, Rittberger and Schimmelfennig 2013). For example,
the ‘Eurozone’—the group of states that use the euro currency—has its own mem-
bership norms and decision-making. Most EU member states chose in the 1990s to
21 Turkey suspends deal with the EU on migrant readmission, Euractiv.com, 24 July 2019.
adopt the euro but a few chose to retain their national currency; other states that
joined the EU in 2004 or later are required to join the Eurozone but most have
been told they are not ready. The dynamics of membership norms discussed in
Chapter 1 apply as well to such ‘differentiated integration’ initiatives that involve a
subset of a larger community.
Applying the same logic in reverse yields the concept of ‘opting out’ or
‘differentiated disintegration’ whereby certain member states choose to separate
themselves from particular integration initiatives without leaving the community
as a whole (Adler-Nissen 2014; Leruth et al. 2019). Membership norms are
relevant and potentially problematic here as well. Imagine a regional community
that requires all members to behave in a particular manner or to integrate
particular policy domains, and then a member state reverses its previous
acceptance of these requirements, changing its behaviour or re-nationalizing the
domain in question while insisting on its right and desire to remain within the
larger community. In such situations, other member states and community
institutions would be forced either to accept the violation of a norm that they
once deemed essential, with the attendant risk of norm decay, or to impose
sanctions on the non-compliant state that could have the unintended effect of
pushing it to exit from the community.
Membership norms may also be related to the extreme situation where a
member state chooses to leave a regional organization entirely, as in the case of
Brexit. If this occurs because the government of a member state concludes that its
policy preferences or the country’s economic welfare or cultural values are
incompatible with the membership norms of a regional organization to which it
belongs, then the norms would have been the proximate cause of the choice to
leave. Many such situations are possible. A government committed to limiting the
ability of judges to oversee the constitutionality of legislation may clash with
community institutions enforcing a membership norm that requires all member
states to ensure an independent judiciary and the rule of law. A government may
prefer to privilege a particular form of religious practice while the membership
norm of a regional community to which it belongs insists on freedom of religion.
In such situations, the member state is likely to challenge the norms, either
through discursive contestation or non-compliant behaviour. If this contestation
fails or if non-compliance provokes the community to impose sanctions, then the
member state in question has either to make concessions or leave the community.
If they choose the latter, then it would be fair to attribute the member state’s exit
to the community’s membership norms.
It is thus clear that membership norms can play a role not only in decision-
making on accession to a region’s meta-community, as seen throughout this
book, but also in a range of other scenarios related to regional integration and
disintegration.
The EU as a bastion of liberal democracy
This book’s findings speak directly to one of the most salient issues in
contemporary European affairs—the EU’s commitment to liberal democratic
values. By exposing the contestation and evolution of EU membership norms
over time, and tracing their contribution to decision-making on EU enlargement,
the book challenges one of the core assumptions voiced by generations of EU
officials and European integration scholars, namely that the EU has always been
committed to liberal democracy and respect for human rights as the foundation
of peace. Regardless of their normative preferences, social scientists would be well
advised to join their historian colleagues (Wilson and van der Dussen 1993:
106–43; Delanty 1995: 111–14; Heffernan 1998: 131–78; Gosewinkel 2015) in
recognizing the historical falsity of this claim.
The point is not that the EU’s member states have never shared a conception of
themselves as a community of democracies or even liberal democracies: they
certainly have, in various ways, at particular points in time. But as this book’s
genealogy of EU norms and its analysis of EU decision-making on membership
and enlargement since the 1950s have shown, the historical reality has been far
more dynamic and complex than the conventional wisdom would suggest.
The constitutive norms of the European political community have changed con-
siderably over time, and in the process, exerted a significant influence over the
course of EU enlargement, which has in turn transformed the community itself.
Over the last decade and a half, these normative currents have brought anti-
liberal ideas with deep historical roots back into the mainstream of EU debates.
Those concerned about the future of liberal democracy today cannot afford to
whitewash this reality.
In fact, the breakdown of the EU’s liberal membership norm around 2005 has
accelerated over time. Whereas the EU’s strong commitment to democracy,
respect for human rights and the rule of law contributed significantly to the
liberalization of applicant states in the 1990s (Schimmelfennig and Sedelmeier
2005), there is today good reason to question the depth of the EU’s collective
commitment to these values. Many member states remain committed to liberal
democracy and the rule of law but others—most notably Hungary and Poland—
have pursued ‘a political and constitutional agenda that is increasingly at odds
with the principles underlying the EU legal order’ (Hillion 2020).
Starting in 2014, the Hungarian premier Viktor Orbán has repeatedly pro-
fessed his commitment to ‘illiberal democracy’.22 And he has not hesitated to
translate this rhetoric into practice. ‘After taking power in 2010 elections, Prime
22 Orbán wants to build ‘illiberal state’, EUobserver.com, 28 July 2014; Viktor Orbán: Era of ‘liberal
democracy’ is over, DW.com, 10 May 2018.
Minister Viktor Orbán’s Alliance of Young Democrats–Hungarian Civic Union

(Fidesz) party pushed through constitutional and legal changes that have allowed it
to consolidate control over the country’s independent institutions. More recently, the
Fidesz-led government has moved to institute policies that hamper the operations of
opposition groups, journalists, universities, and nongovernmental organizations
(NGOs) whose perspectives it finds unfavorable’ (Freedom House 2020).
The Orbán government’s pursuit of this agenda has not been deterred by
negative publicity. In its 2015 ‘Nations in Transit’ report, Freedom House
downgraded Hungary from Consolidated Democracy to Semi- Consolidated
Democracy (Habdank-Kołaczkowska 2015). In a separate 2019 ranking, Freedom
House downgraded Hungary’s rating from Free to Partly Free—a first for any EU
member state—‘due to sustained attacks on the country’s democratic institutions
by Prime Minister Viktor Orbán’s Fidesz party, which has used its parliamentary
supermajority to impose restrictions on or assert control over the opposition, the
media, religious groups, academia, NGOs, the courts, asylum seekers, and the
private sector since 2010’ (Freedom House 2019). In its 2020 ‘Nations in Transit’
report, Freedom House noted ‘a stunning democratic breakdown’ in Hungary,
removing it from the ranks of democracies and re-classifying it as a Transitional/
Hybrid Regime (Csaky 2020).
As similar trend, albeit less advanced, is evident in Poland. A 2018 paper by
Freedom House described the Polish government’s assault on judicial
independence as ‘incompatible with the conditions of membership of the
European Union’ (Davies 2018). Two years later, Freedom House observed, ‘Since
taking power in late 2015, [the Polish government] has enacted numerous
measures that increase political influence over state institutions and threaten to
reverse Poland’s democratic progress’ (Freedom House 2020). In its 2020 ‘Nations
in Transit’ report, Freedom House downgraded Poland from Consolidated
Democracy to Semi-Consolidated Democracy (Csaky 2020).
The European Parliament has criticized these developments and the European
Commission has threatened sanctions on Hungary and Poland, but this pressure
has not reversed the breakdown of normative consensus. More worrisome, the
EU may actually be helping to consolidate authoritarian regimes within its
borders through the member states’ reluctance to criticize each other’s domestic
actions, through funding and investment from the community budget without
political conditions, and the ability of dissatisfied citizens to move elsewhere in
the bloc, leaving the Union ‘stuck in an autocracy trap’ (Kelemen 2020; see also
Closa 2020). Such developments threaten to transform political life within the EU
as well as the Union’s international presence at a time when the liberal
international order is being challenged on multiple fronts.
If the EU fails to re-converge on a liberal democratic membership norm, then
member states and community institutions still committed to these values will
find it harder to resist the spread of ‘illiberal democracy’ within the Union.
Non-member states in eastern and south-eastern Europe that still aspire to EU

accession will have little external incentive to ensure democratic institutions, the
rule of law and respect for human rights within their borders. Instead, they will
recognize that they can gain entry to the Union simply by demonstrating their
geo-political value-added or their readiness for inclusion in the Single Market.
The warning signs are clear: as noted by Davor Stier, former foreign minister of
Croatia, ‘The EU’s enlargement policy still formally exists . . . but it is more of a
leftover bureaucratic procedure than a real strategic policy for stabilization and
reform. That needs to change if the EU is to prevent further fragmentation and
the erosion of its democratic values.’23
If this trend continues, then EU enlargement policy, once recognized as one
of the EU’s most successful policies, will be reduced to a technocratic procedure
and the EU will lose influence as a defender of democracy and human rights
abroad. Given the EU’s limited reserves of hard power, such a self-inflicted loss of
soft power is likely to exacerbate the Union’s tendency to ‘punch below its weight’ in
international affairs (Thomas 2012). And if the EU ceases to champion democracy,
human rights, and the rule of law around the world, it is no longer clear which
other state or group of states, if any, would lead this cause.
The orientation and effectiveness of EU foreign policy
The EU’s recent lack of consensus on membership norms also has real implications
for the future of EU foreign policy. By definition, the practice of foreign policy
involves relations with an Other, which presumes some prior understanding of
the Self. A regional community’s membership norms reflect this understanding
of the Self and thereby inform its relations with the Other. A regional community
is thus likely to relate differently to states that it considers eligible for membership
(i.e. part of the potential Self) and those that will forever be foreign (i.e.
permanently part of the Other). This was clear when the then- European
Community deliberated in the early 1970s over its approach to the multilateral
initiative that became the Conference on Security and Cooperation in Europe:
the EC’s emerging liberal democratic membership norm contributed directly to
its collective insistence, much to the displeasure of US Secretary of State Henry
Kissinger, that it would not consider normalizing relations with its central and
east European neighbours unless respect for human rights were part of the dis-
cussion (Thomas 2001).
A regional community’s membership norms may also affect relations with
states that stand no chance of accession. By communicating what the community’s
23 Europe unwhole and unfree, Politico.eu, 2 November 2018.

members consider makes them distinctive, membership norms shape the agenda
for relations between the community and outside powers. For example, ASEAN’s
repeated assertions that it takes no position on the constitution or internal
behaviour of member states is likely to have undermined those that might have
favoured a more outspoken position on human rights abuses by neighbouring
states. Similarly, after defining itself repeatedly as a community of liberal
democracies committed to human rights and the rule of law, the EU antagonized
the USA by giving strong support to the International Criminal Court even
though some member states were sympathetic to the Clinton Administration’s
concern about the scope of the new court and then the Bush Administration’s
campaign to exempt Americans from the court’s jurisdiction (Thomas 2009a).
The breakdown of consensus on membership norms since 2005 thus denies the
EU a potential source of foreign policy unity. Given the difficulty of gaining agree-
ment on common positions among the EU’s many member states (Schimmelfennig
and Thomas 2009, Thomas 2009b, 2009c), this normative d issensus will weaken
the EU’s ability to respond collectively to the diverse array of foreign policy chal-
lenges it is likely to face in coming years, including Russia’s assertiveness in
Europe’s backyard, the USA’s retreat from multilateralism, China’s effort to remake
the international order, and the complexities of climate change mitigation and
adaptation. Past experience has shown that unity is not sufficient for foreign policy
effectiveness (Thomas 2012), but it is likely necessary for ensuring that the EU’s
influence on a changing world is commensurate with its size and ambitions.
APPENDIX
Imputing missing Freedom House

data from V-Dem data
The test of the membership norms argument in Chapter 4 relies principally on Freedom
House’s (FH) ‘electoral democracy’ and ‘freedom’ scores, which provide clear criteria for
assessing a country’s fit with the membership norm prevailing at various periods of time.
However, FH has limited temporal coverage and thus does not include all our cases. This
appendix describes why and how we used data from the University of Gothenburg’s
V-Dem Institute to impute scores missing in the FH data.
First, we established that V-Dem data are highly predictive of FH’s ‘electoral democracy’
indicator. As shown in Figure A.1, FH electoral democracies form clusters of high-scoring
countries on both relevant V-Dem variables (‘liberal democracy’ and ‘polyarchy’) and have
significantly higher mean scores as compared to countries without electoral democratic
institutions and practices (t = −75.8, p < .01 for the V-Dem polyarchy Index; t = −77.3,
p < .01 for the V-Dem Liberal Democracy index).
The correlations related to liberal democracy are similarly high. Teorell and colleagues
(2016: 29) have shown that V-Dem’s ‘Polyarchy’ index is highly correlated with FH’s ‘civil
liberties’ and ‘political rights’ scores, and as shown in Figure A.2; we find strikingly similar
patterns with regard to the relationship between V-Dem’s ‘liberal democracy’ index and
FH scores.
We then used simple (logistic) regression models to predict a country’s FH scores based
on all country-year observations available in both FH and V-Dem. Employing V-Dem’s
‘polyarchy’ and ‘liberal democracy’ indices as dependent variables, as well as including
country dummies (fixed effects), we predict a country’s freedom rating, that is, its mean
value on civil liberties and political rights in a given year, using OLS. Countries are coded
as liberal democracies when their predicted value is no higher than 2.5. For a country’s
classification as electoral democracy, we estimate a logistic regression model including the
same predictor variables, and count countries as electoral democracies when their pre-
dicted probability to belong to this category was higher than 0.5. The predicted scores were
calculated as follows:
Freedom =∝+ VDemLib. Dem. * b1 + VDemPolyachy * b 2 + Country fixed effect ;
and
∝+VDemLib . Dem. b + VDemPolyachy b + Country fixed effect
e * 1 * 2
P ( Electoral Democracy = 1) = ∝+VDemLib . Dem.* b1 + VDemPolyachy *b2 + Country fixed effect
;
1+ e
where α and β stand for the estimated coefficients.
Both models fit the data reasonably well with 94 per cent of variance being explained by
the OLS model, and 90 per cent of correct classifications and a reduction of prediction
244 Appendix
0.8
0.6
0.4
0.2
0.0
No Yes
FH electoral democracy
V-Dem Liberal democracy

V-Dem Polyarchy
Figure A.1 Comparison of Freedom House and V-Dem country-year scores
0.8 0.8
V-Dem Liberal Democracy
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
1 2 3 4 5 6 7 1 2 3 4 5 6 7
FH Civil Liberties FH Political Rights
Spearman rank correlation coefficient = –.91 Spearman rank correlation coefficient = –.93
Figure A.2 Correlations between country-year scores in Freedom House and V-Dem

datasets
error by 74 per cent by the logistic regression model. Nonetheless, in order to increase our
confidence in the predicted scores we ran a number of robustness checks, including slightly
different model specifications, comparable models based on Polity IV data, models includ-
ing Polity IV and V-Dem predictors, and models for European countries only. All these
robustness checks provided substantially similar results. In sum, the advantage of having a
Appendix 245
Table A.1 Imputed regime scores for country-years not covered by Freedom House data
Country-Year Parliamentary Liberal Democracy

Democracy (scores 1–7, with 1.0–2.5
(1 = yes; 0 = no) signifying liberal democracy)
Ireland (1962) 1 Not applicable

Norway (1962) 1 Not applicable
Spain (1962) 0 Not applicable
United Kingdom (1963) 1 Not applicable
Greece (1967) 0 Not applicable
United Kingdom (1967) 1 Not applicable
United Kingdom (1970) 1 1.5
Serbia and Montenegro (2006) 1 2
Kosovo (2013) 1 4
clear and discriminating classification scheme for different regime types, as provided by
FH data, far outweighs the minor uncertainty introduced by this imputation procedure.
Finally, using the estimated regression equations, we imputed scores for regime charac-
teristics in country-years not included in the FH dataset, as shown in Table A.1.
Bibliography
Abbott, Kenneth W. and Duncan Snidal (1998). Why States Act through Formal
International Organizations. Journal of Conflict Resolution 42(1), 3–32
Acharya, Amitav (1997). Ideas, identity, and institution-building: From the ‘ASEAN way’ to
the ‘Asia-Pacific way’? Pacific Review 10(3), 319–346
Acharya, Amitav (2004). How Ideas Spread: Whose Norms Matter? Norm Localization and
Institutional Change in Asian Regionalism. International Organization 58(2), 239–75
Acharya, Amitav (2009). Whose Ideas Matter: Agency and Power in Asian Regionalism.
Ithaca, NY: Cornell University Press
Acharya, Amitav (2011). The Making of Southeast Asia: International Relations of a Region.
Ithaca, NY: Cornell University Press
Acharya, Amitav (2017). The Evolution and Limitations of ASEAN Identity. In: Aileen
Baviera and Larry Maramis, eds., Building ASEAN Community: Political-Security and
Socio-cultural Reflections. ASEAN@50, Volume 4. Jakarta: Economic Research Institute
for ASEAN and East Asia, 25–38. http://www.eria.org/ASEAN_at_50_Vol_4_Full_
Report.pdf
Adler-Nissen, Rebecca (2014). Opting Out of the European Union: Diplomacy, Sovereignty,
and European Integration. Cambridge: Cambridge University Press
Allen, Christopher and Jørgen S. Nielsen (2002). Summary Report on Islamophobia in the
EU after 11 September 2001. Vienna: European Monitoring Centre on Racism and
Xenophobia
Anderson, Benedict (1983). Imagined Communities: Reflections on the Origin and Spread of
Nationalism. London: Verso
Arts, Kees and Henk van der Kolk (2006). Understanding the Dutch ‘No’: The Euro, the
East and the Elite. PS: Political Science & Politics 39(2), 243–6
Åslund, Anders and Michael McFaul (2006). Revolution in Orange: The Origins of Ukraine’s
Democratic Breakthrough. Washington, DC: Carnegie Endowment for International Peace
Baldwin, Richard E. (2007). Is enlargement unlimited? 17 May. Blog post at http://voxeu.
org/article/enlargement-unlimited
Baldwin, Richard E., Joseph F. Francois, Richard Portes (1997). ‘The Costs and Benefits of
Eastern Enlargement: The Impact on the EU and Central Europe.’ Economic Policy
12(24), 125–76
Barnett, Michael N. (1998). Dialogues in Arab Politics: Negotiations in Regional Order. New
York: Columbia University Press
Bennett, Andrew and Jeffrey T. Checkel (2015). Process Tracing: From Metaphor to Analytic
Tool. Cambridge: Cambridge University Press
Bini, Federiga and Irina Angelescu, eds (2011). The Frontiers of Europe. A Transatlantic
Problem? Washington, DC: Brookings Institution Press
Birkelbach, Willi (2000). Fazit: gelebt-bewegt. Marburg: Schüren Verlag
Blatter, Joachim and Markus Haverland (2012). Designing Case Studies: Explanatory
Approaches in Small-N Research. Basingstoke: Palgrave Macmillan
Böhmelt, Tobias and Tina Freyburg (2013). The temporal dimension of the credibility of
EU conditionality and candidate states’ compliance with the acquis communautaire,
1998–2009. European Union Politics 14(2), 250–72
Bolkestein, Frits (2004). The Limits of Europe. Tielt: Uitgeverij Lannoo
248 Bibliography
Bolkestein, Frits (2011). The Roots of Europe’s Cultural Masochism. Wall Street Journal, 3 June.
Brubaker, William Rogers (1990). Immigration, Citizenship, and the Nation-State in France
and Germany: A Comparative Historical Analysis, International Sociology 5(4), 379–407
Buergenthal, Thomas (1968). Proceedings against Greece under the European Convention
of Human Rights. American Journal of International Law 62(2), 441–50
Bull, Hedley (2012). Anarchical Society: A Study of Order in World Politics, 4th ed. Oxford:
Oxford University Press
Bürgin, Alexander (2010). Cosmopolitan Entrapment: The Failed Strategies to Reverse
Turkey’s EU Membership Eligibility. Perspectives 18(2), 33–56
Buzan, Barry and Ole Wæver (2003). Regions and Powers: The Structure of International
Security. Cambridge: Cambridge University Press
Çalış, Şaban H. (2004). Formative Years: A Key for Understanding Turkey’s Membership
Policy towards the EU. Perceptions: Journal of International Affairs 9(3) (Autumn), 73–96
Çalış, Şaban H. (2017). Turkey’s Cold War: Foreign Policy and Western Alignment in the
Modern Republic. London: I.B. Taurus
Camps, Miriam (1964). The Six and Political Union. The World Today 20(11), 473–80
Cancian, Francesca M. (1975). What are Norms? A Study of Beliefs and Action in a Maya
Community. Cambridge: Cambridge University Press
Cederman, Lars-Erik, ed. (2001). Constructing Europe’s Identity: The External Dimension.
Boulder, CO: Lynne Reinner Publishers
Churchill, Winston Spencer and Paul-Henri Spaak Churchill, Winston Spencer and Paul-
Henri Spaak (1950). European Movement and the Council of Europe. London: Hutchinson
Clogg, Richard (1987). Parties and Elections in Greece: The Search for Legitimacy. London:
C. Hurst & Company
Closa, Carlos (2020). Institutional logics and the EU’s limited sanctioning capacity under
Article 7 TEU. International Political Science Review. Online First.
Crowley, Úna (2009). Genealogy Method. In: Rob Kitchin and Nigel Thrift, eds,
International Encyclopedia of Human Geography, Vol. 4. Oxford: Elsevier, 341–4
Csaky, Zselyke (2020). Nations in Transit 2020: Dropping the Democratic Facade. New
York: Freedom House. https://freedomhouse.org/report/nations-transit/2020/dropping-
democratic-facade#Facade
Cvijic, Srdjan et al. (2019). From Enlargement to the Unification of Europe: Why the
European Union needs a Directorate General Europe for Future Members and Association
Countries. Open Society Foundations, European Policy Institute
Dagi, Ihsan D. (1996). Democratic Transition in Turkey, 1980–83: The Impact of European
Diplomacy. Middle Eastern Studies 32(2), 124–41
Dainotto, Roberto M. (1997). Europe (in Theory). Durham, NC: Duke University Press.
Davies, Christian (2018). Hostile Takeover: How Law and Justice Captured Poland’s
Courts. Freedom House, Nations in Transit, May. https://freedomhouse.org/sites/
default/files/poland%20brief%20final.pdf
Davis, Christina L. and Tyler Pratt (2020). The Forces of Attraction: How Security Interests
Shape Membership in Economic Institutions. Review of International Organizations.
Online 15 September
Davis, Christina L. and Meredith Wilf (2017). Joining the Club: Accession to the GATT/
WTO. Journal of Politics 79(3), 964–78
Dawson, Christopher (1952). Understanding Europe. Garden City, NJ: Image Books
Deitelhoff, Nicole and Lisbeth Zimmermann (2020). Things We Lost in the Fire: How
Different Types of Contestation Affect the Robustness of International Norms.
International Studies Review 22(1), 51–76
Delanty, Gerard (1995). Inventing Europe. Idea, Identity, Reality. Basingstoke: Macmillan
Bibliography 249
Donno, Daniela (2010). Who Is Punished? Regional Intergovernmental Organizations and

the Enforcement of Democratic Norms. International Organization 64(4), 593–625
Donno, Daniela, Shawna K. Metzger, and Bruce Russett (2015). Screening Out Risk: IGOs,
Member State Selection, and Interstate Conflict, 1951–2000. International Studies
Quarterly 59(2), 251–63
Döşemeci, Mehmet (2014). Debating Turkish Modernity: Civilization, Nationalism, and the
EEC. Cambridge: Cambridge University Press
Downs, George W., David M. Rocke, and Peter N. Barsoom (1998). Managing the
Evolution of Multilateralism. International Organization 52(2), 397–419
Droesse, Gerd (2020). Membership in International Organizations: Paradigms of Membership
Structures, Legal Implications of Membership and the Concept of International Organizations.
The Hague: T.M.C. Asser Press.
Duxbury, Alison (2000). Austria and the European Union—The Report of the ‘Three Wise
Men’. Melbourne Journal of International Law 1(1), 169–74
Emerson, Michael and Gergana Noutcheva (2018). Political and Economic Governance in
the Balkans and Eastern Europe Compared. Brussels: Centre for European Policy Studies,
CEPS Policy Brief, July
Emmers, Ralf (2005). The Indochinese enlargement of ASEAN: Security expectations and
outcomes. Australian Journal of International Affairs 59(1), 71–88
Erhard, Ludwig (1963). The Economics of Success. Princeton, NJ: Van Nostrand
Evans, Peter, and James E. Rauch (1999). Bureaucracy and Growth: A Cross-National
Analysis of the Effects of ‘Weberian’ State Structures on Economic Growth. American
Sociological Review 64(5), 748–65
Fearon, James D. (1991). Counterfactuals and Hypothesis Testing in Political Science.
World Politics 43(2), 169–95
Flory, Maurice (1986). Note sur la demande d’adhésion du Maroc à la Communauté
économique européenne. Annuaire de l’Afrique du Nord 23, 705–9. Paris: Éditions du CNRS
Frankel, Jeffrey A., Ernesto Stein, and Shang-Jin Wei (1997). Regional Trading Blocs in the
World Economic System. Washington, DC: Institute for International Economics
Freedom House (2019). Freedom in the World 2019: Democracy in Retreat. New York.
https://freedomhouse.org/report/freedom-world/freedom-world-2019
Freedom House (2020). Freedom in the World 2020: A Leaderless Struggle for Democracy.
New York. https://freedomhouse.org/report/freedom-world/2020/leaderless-struggle-
democracy
Freyburg, Tina (2013). The Temporal Dimension of the Credibility of EU Conditionality
and Candidate States’ Compliance with the acquis communautaire, 1998–2009.
European Union Politics 14(2), 250–72
Gateva, Eli (2015). EU Enlargement Conditionality in the Context of Eastern and South-
Eastern Enlargement of the Union. Basingstoke: Palgrave Macmillan
Giscard d’Estaing, Valéry (2000). A Better European Bridge to Turkey. Financial Times, 24
November 2004
Glenn, John K. (2004). From Nation-states to Member States: Accession Negotiations as an
Instrument of Europeanization. Comparative European Politics 2(1), 3–28
Gosewinkel, Dieter (2015). Anti-liberal Europe: A Neglected Story of Europeanization. New
York: Berghahn Books
Goulard, Sylvie (2004). Le Grand Turc et la République de Venise. Paris: Fayard
Gray, Julia (2009). International Organization as a Seal of Approval: European Union
Accession and Investor Risk. American Journal of Political Science 53(4), 931–49
Griffiths, Richard T. ed. (1997). Explorations in OEEC History. Paris: Organization for
Economic Co-operation and Development
250 Bibliography
Gromadzki, Grzegorz et al. (2006). Will the Orange Revolution Bear Fruit? EU-Ukraine
Relations in 2005 and the Beginning of 2006. Warsaw: Stefan Batory Foundation
Guirao, Fernando (1997). Association or Trade Agreement? Spain and the EEC,
1957–1964. Journal of European Integration History 3(1), 103–20
Guirao, Fernando (1999). Solvitur Ambulando: The Place of the EEC in Spain’s Foreign
Economic Policy, 1957–1962. In: Anne Deighton and Alan S. Milward, eds, Widening,
Deepening and Acceleration: The European Economic Community 1957–1963. Baden-
Baden: Nomos, 347–58
Habdank-Kołaczkowska, Sylvana (2015). Nations in Transit 2015: Democracy on the
Defensive in Europe and Eurasia. New York: Freedom House. https://freedomhouse.org/
sites/default/files/2020–02/FH_NIT2015_06.06.15_FINAL.pdf
Haines, C. Grove, ed. (1957). European Integration. Baltimore, MD: Johns Hopkins
University Press
Hallstein, Walter (1972). Europe in the Making. Trans. Charles Roetter. London:
Allen & Unwin
Hallstein, Walter (1979). Europäische Reden. Stuttgart: Deutsche Verlags-Anstaalt
Harris, George S. (1970). The Causes of the 1960 Revolution in Turkey. Middle East Journal
24(4), 438–54
Harryvan, A.G. and J. van der Harst, eds (1997). Documents on European Union. New
York: St. Martin’s Press
Heffernan, Michael J. (1998). The Meaning of Europe: Geography and Geopolitics.
London: Arnold
Hemmer, Christopher and Peter J. Katzenstein (2002). Why Is there no NATO in Asia?
International Security 56(3), 575–607
Hillion, Christophe (2007). Negotiating Turkey’s Membership to the European Union. Can
the Member States Do as They Please? European Constitutional Law Review 3(2), 269–84
Hillion, Christophe (2010). The Creeping Nationalisation of the EU’s Enlargement Policy.
Stockholm: Swedish Institute for European Policy Studies, Report No.6
Hillion, Christopher (2020). Poland and Hungary are withdrawing from the EU. 27 April.
Blog post. https://verfassungsblog.de/poland-and-hungary-are-withdrawing-from-the-eu/
Hurrell, Andrew (2005). Power, Institutions and Global Governance. In: Michael Barnett
and Rayond Duvall, eds, Power and Global Governance. Cambridge: Cambridge
University Press, 33–58
Jacoby, Wade (2006). The Enlargement of the European Union and NATO. Cambridge:
Cambridge University Press
Karamouzi, Eirini (2014). Greece, the EEC and the Cold War 1974–1979: The Second
Enlargement. London: Palgrave Macmillan
Karamouzi, Eirini (2015). A Strategy for Greece: Democratization and European
Integration, 1974–1975. Cahiers de la Méditerranée 90, 1–15.
Katchanovski, Ivan (2011). Puzzles of EU and NATO Accession of Post-Communist
Countries. Perspectives on European Politics and Society 12(3): 304–19
Katzenstein, Peter J. (2005). A World of Regions: Asia and Europe in the American
Imperium. Ithaca, NY: Cornell University Press
Kelemen, R. Daniel (2020). The European Union’s Authoritarian Equilibrium, Journal of
European Public Policy 27(3), 481–99
Kelley, Judith G. (2010). The Role of Membership Rules in Regional Organizations. ADB
Working Paper Series on Regional Economic Integration. No. 53 (June)
Keohane, Robert O. (1986). Reciprocity in International Relations. International
Organization 40(1), 1–27
Keulen, Mendeltje van (2004). The Netherlands 2004 EU Council Presidency. Swedish
Institute for European Policy Studies, SIEPS Paper 2004:1op (October)
Bibliography 251
Kochenov, Dimitry (2004). Behind the Copenhagen Façade: The Meaning and Structure of
the Copenhagen Political Criterion of Democracy and the Rule of Law. European
Integration Online Papers 8(10), 1–24.
Koremenos, Barbara, Charles Lipson, Duncan Snidal (2001a). The Rational Design of
International Institutions. International Organization 55(4), 761–99
Koremenos, Barbara, Charles Lipson, and Duncan Snidal (2001b). Rational Design:
Looking Back to Move Forward. International Organization 55(4), 1051–82
Kovács, Kriszta (2020). Parliamentary Democracy by Default: Applying the European
Convention on Human Rights to Presidential Elections and Referendums. Jus Cogens
2(3), 237–58.
Krasner, Stephen D. (1988). Sovereignty: An Institutionalist Perspective. Comparative
Political Studies 21(1), 66–94
Kupchan, Charles A. (2010). How Enemies Become Friends: The Sources of Stable Peace.
Princeton, NJ: Princeton University Press
Leuffen, Dirk, Berthold Rittberger and Frank Schimmelfennig (2013). Differentiated
Integration: Explaining Variation in the European Union. Basingstoke: Palgrave Macmillan
Lindberg, Staffan I. et al. (2014). V-Dem: A New Way to Measure Democracy. Journal of
Democracy 25(3), 159–169
Ludlow, Peter (2002). The Copenhagen European Council: December 12 and 13, 2002.
EuroComment Briefing Note No. 10
Ludlow, Peter (2005). Dealing with Turkey: The European Council of 16–17 December 2004.
EuroComment Briefing Note 3:7
Mansfield, Edward D. and Helen V. Milner, eds (1997). The Political Economy of
Regionalism. New York: Columbia University Press
Mansfield, Edward and Jon Pevehouse (2013). The Expansion of Preferential Trading
Arrangements. International Studies Quarterly 57(3), 592–604
March, James G. and Johan P. Olsen (1998). The Institutional Dynamics of International
Political Orders. International Organization 52(4), 943–69
Mattli, Walter (1999). The Logic of Regional Integration. Cambridge: Cambridge University
Press
Mattli, Walter and Thomas Plümper (2002). The Demand-Side Politics of EU Enlargement:
Democracy and the Application for EU Membership. Journal of European Public Public
Policy 9(4), 550–74
Mauro, Paolo (1995). Corruption and Growth. Quarterly Journal of Economics 110(3), 681–712
Mayer, Franz C. and Jan Palmowski (2004). European Identities and the EU: The Ties that
Bind the Peoples of Europe. JCMS: Journal of Common Market Studies 42(3), 573–98
Mearsheimer, John J. (2001). The Tragedy of Great Power Politics. New York: Norton
Miller, Nicholas R. (1977). Logrolling, Vote Trading and the Paradox of Voting: A Game-
Theoretical Overview. Public Choice 30(1), 51–75
Milward, Alan S. (1992). The European Rescue of the Nation-State. London: Routledge Press
Ministère des Affaires Étrangères (1995). Documents diplomatiques français, 1960—Tome I
(1er janvier–30 juin). Paris: Ministère des Affaires Étrangeres, Commission de Publication
des Documents Diplomatiques Français,
Ministère des Affaires Étrangères (1997). Documents diplomatiques français, 1961—Tome I
(1er janvier–30 juin). Paris: Ministère des Affaires Étrangeres, Commission de Publication
des Documents Diplomatiques Français,
Ministère des Affaires Étrangères (2002). Documents diplomatiques français, 1964—Tome
II (1er juillet–31 décembre). Bruxelles: Peter Lang
Monnet, Jean (1955). Les Etats-Unis d’Europe ont Commencé. Paris: Robert Laffont
Moravcsik, Andrew (1997). Taking Preferences Seriously: A Liberal Theory of International
Politics. International Organization 51(4), 513–53
252 Bibliography
Moravcsik, Andrew (1998). The Choice for Europe: Social Purpose and State Power from
Messina to Maastricht. Ithaca, NY: Cornell University Press
Moravcsik, Andrew and Milada Anna Vachudova (2005). Preferences, Power, and
Equilibrium: The Causes and Consequences of EU Enlargement. In: Frank Schimmelfennig
and Ulrich Sedelmeier, eds, The Politics of European Union Enlargement: Theoretical
Approaches. London: Routledge, 198–212
Morozov, Vlatcheslav and Bahar Rumelili (2012). The external constitution of European
identity: Russia and Turkey as Europe-makers. Cooperation and Conflict 47(1), 28–48
Munich Security Conference Foundation (2019). Munich Security Report 2019. https://
www.securityconference.de/en/publications/munich-security-report
Murphy, Alexander B. (1991). Regions as social constructs: the gap between theory and
practice. Progress in Human Geography 15(1), 22–35
Nafpliotis, Alexandros (2012). Britain and the Greek Colonels: Accommodating the Junta in
the Cold War. London: I. B. Tauris
Neumann, Iver B. (1998). European Identity, EU Expansion, and the Integration/Exclusion
Nexus. Alternatives 23(3), 397–416
Neumann, Iver B. (1999). The Uses of the Other: ‘The East’ in European Identity Formation.
Minneapolis: University of Minnesota Press
Neumann, Iver B. (2008). Discourse Analysis. In: Audie Klotz and Deepa Prakash, eds,
Qualitative Methods in International Relations: A Pluralist Guide. Houndmills: Palgrave
Macmillan, 61–77
Nicolaïdes, Kalypso (2014). Europe’s Ends. In: Claudia Wiesner and Meike Schmidt-
Gleim, eds, The Meanings of Europe: Changes and Exchanges of a Contested Concept.
New York: Routledge, 236–57
Nordhaus, William, John R. Oneal, and Bruce Russett (2012). The Effects of the
International Security Environment on National Military Expenditures: A Multicountry
Study. International Organization 66(3), 491–513
Nye, Joseph S. (1971). Peace in parts: Integration and conflict in regional organization.
Boston: Little, Brown
Panke, Diana and Ulrich Petersohn (2012). Why international norms disappear some-
times. European Journal of International Relations 18(4), 719–742
Papandreou, Nikos (1998). The Council of Europe Fights for Democracy in Greece,
1967–1969. Athens: Andreas G. Papandreou Foundation
Pape, Robert A. (2005). Soft Balancing against the United States. International Security
30(1), 7–45
Pelt, Mogens (2006). Tying Greece to the West: US–West German-Greek Relations,
1949–1974. Copenhagen: Museum Tusculanum Press
Pevehouse, Jon C. (2002). With a Little Help from my Friends? Regional Organizations
and the Consolidation of Democracy. American Journal of Political Science 46(3), 611–26
Pevehouse, Jon C. (2003). Democratization, Credible Commitments, and Joining
International Organizations. In: Daniel W. Drezner, ed., Locating the Proper Authorities:
The Interaction of Domestic and International Institutions. Ann Arbor: University of
Michigan Press, 25–48
Pevehouse, Jon C. and Bruce Russett (2006). Democratic International Governmental
Organizations Promote Peace. International Organization 60(4), 969–1000
Phinnemore, David (1999). Association: Stepping-Stone or Alternative to EU Membership?
Sheffield: Sheffield Academic Press
Pistone, Sergio (1993). Il ruolo di Altiero Spinelli nella genesi dell’art.38 della Comunità
Europea di Difesa e del progetto di Comunità Politica Europea. In: Gilbert Trausch, ed.,
Bibliography 253
The European Integration from the Schuman-Plan to the Treaties of Rome. Baden-Baden:
Nomos Verlag, 393–413
Plassmann, Lorenz (2012). Comme dans une nuit de Pâques: Les relations franco-greques,
1944–1981. Brussels: Peter Lang
Plümper, Thomas and Christina J. Schneider (2007). Discriminatory European Union
Membership and the Redistribution of Enlargement Gains. Journal of Conflict Resolution
51(4), 568–87
Plümper, Thomas, Christina J. Schneider, and Vera E. Troeger (2006). The Politics of EU
Eastern Enlargement: Evidence from a Heckman Selection Model. British Journal of
Political Science 36(1), 17–38.
Pope, Hugh (2012). Erdoğan’s Decade. Cairo Review of Global Affairs 4 (Winter), 43–57
Posen, Barry R. (2004). ESDP and the Structure of World Power. The International
Spectator 39(1), 5–17
Powers, Kathy and Gary Goertz (2011). The Economic-Institutional Construction of
Regions: Conceptualization and Operationalization. Review of International Studies
37(5), 2387–415
Preda, Daniela (1992). From a Defence Community to a Political Community: The Role of
de Gasperi and Spinelli. In: Andrea Bosco, ed., The Federal Idea, Vol. II: The History of
Federalism since 1945. London: Lothian Foundation Press, 189–206
Puddington, Arch et al. (2015). Freedom in the World 2014: The Annual Survey of Political
Rights and Civil Liberties. Washington, DC: Freedom House
Ramos-Horta, José (2011). Why Timor-Leste Should Join ASEAN now. East Asia Forum,
16 May http://www.eastasiaforum.org/2011/05/16/why-timor-leste-should-join-asean-
now/
Rapp, Kyle (2020). Law and Contestation in International Negotiations. Review of
International Studies 46(5), 672–90
Rehn, Olli (2006). Europe’s Next Frontiers, Speech 06/654, Finnish Institute of
International Affairs, Helsinki, 27 October. http://www.fiia.fi/assets/doc/upi_next_
frontiers_27-10–2006.pdf
Risse, Thomas (2012). Identity Matters: Exploring the Ambivalence of EU Foreign Policy.
Global Policy 3(1s), 87–95
Rosato, Sebastian (2011). Europe United: Power Politics and the Making of the European
Community. Ithaca, NY: Cornell University Press
Ruggie, John Gerard (1982). International Regimes, Transactions, and Change: Embedded
Liberalism in the Postwar Economic Order. International Organization 36(2), 379–415
Ruggie, John Gerard (1991). Embedded Liberalism Revisited: Institutions and Progress in
International Economic Relations. In: Emanuel Adler and Beverly Crawford, eds,
Progress in Postwar International Relations. New York: Columbia University Press
Ruggie, John Gerard (1998). Constructing the World Polity: Essays on International
Institutionalisation. London: Routledge Press
Ruggie, John Gerard (1999). What Makes the World Hang Together? Neo-Utilitarianism
and the Social Constructivist Challenge. In: Peter J. Katzenstein, Robert O. Keohane,
and Stephen D. Krasner, eds, Exploration and Contestation in the Study of World Politics.
Cambridge, MA: MIT Press, 215–45
Rumelili, Bahar (2004). Constructing Identity and Relating to Difference: Understanding
the EU’s Mode of Differentiation. Review of International Studies 30(1), 27–47
Said, Edward (1978). Orientalism. New York: Pantheon Books
Sandholtz, Wayne (2008). Dynamics of international norm change: Rules against wartime
plunder. European Journal of International Relations 14(1), 101–131
254 Bibliography
Sartori, Giovanni (1970). Concept Misformation in Comparative Politics. American

Political Science Review 64(4), 1033–53
Sasse, Gwendolyn (2008). The Politics of EU Conditionality: The Norm of Minority Protection
during and beyond EU Accession. Journal of European Public Policy 15(6), 842–60
Schimmelfennig, Frank (2001). The Community Trap: Liberal Norms, Rhetorical Action,
and the Eastern Enlargement. International Organization 55(1), 47–80
Schimmelfennig, Frank (2003). The EU, NATO and the Integration of Europe. Cambridge:
Cambridge University Press
Schimmelfennig, Frank (2005). Strategic Calculation and International Socialization:
Membership Incentives, Party Constellations, and Sustained Compliance in Central and
Eastern Europe. International Organization 59(4), 827–60
Schimmelfennig, Frank (2009). Entrapped Again: The Way to EU Membership
Negotiations with Turkey. International Politics 46(4), 413–31
Schimmelfennig, Frank (2015). Efficient Process Tracing: Analyzing the Causal Mechanisms
of European Integration. In: Andrew Bennett and Jeffrey T. Checkel, eds, Process Tracing.
From Metaphor to Analytical Tool. Cambridge: Cambridge University Press, 98–125
Schimmelfennig, Frank (2016). Europe. In: Tanja E. Boerzel and Thomas Risse, eds, The
Oxford Handbook of Comparative Regionalism. Oxford: Oxford University Press, 178–201
Schimmelfennig, Frank and Ulrich Sedelmeier (2004). Governance by Conditionality: EU
Rule Transfer to the Candidate Countries of Central and Eastern Europe. Journal of
European Public Policy 11(4), 661–79
Schimmelfennig, Frank and Ulrich Sedelmeier (2005). The Europeanization of Central and
Eastern Europe. Ithaca, NY: Cornell University Press
Schimmelfennig, Frank and Daniel C. Thomas (2009). Normative Institutionalism and EU
Foreign Policy in Comparative Perspective (with Frank Schimmelfennig). International
Politics 46(4), 491–504
Schimmelfennig, Frank, Stefan Engert, and Heiko Knobel (2003). Costs, Commitment and
Compliance: The Impact of EU Democratic Conditionality on Latvia, Slovakia and
Turkey. JCMS: Journal of Common Market Studies 41(3), 495–518
Frank Schimmelfennig, Stefan Engert, Heiko Knobel (2006). International Socialization in
Europe: European Organizations, Political Conditionality and Democratic Change.
Basingstoke: Palgrave Macmillan
Schneider, Carsten Q. and Claudius Wagemann (2012). Set-Theoretic Methods for the Social
Sciences: A Guide to Qualitative Comparative Analysis. Cambridge: Cambridge
University Press
Searle, John (1995). The Construction of Social Reality. New York: Free Press
Sedelmeier, Ulrich (2005). Constructing the Path to Eastern Enlargement: The Uneven
Policy Impact of EU Identity. Manchester: Manchester University Press
Severino, Rudolfo C. (2006). Southeast Asia in Search of an ASEAN Community: Insights
from the Former ASEAN Secretary-General. Singapore: ISAES Publishing
Skålnes, Lars S. (2005). Geopolitics and the Eastern Enlargement of the European Union.
In: Frank Schimmelfenning and Ulrich Sedelmeier, eds, The Politics of European Union
Enlargement: Theoretical Approaches. London: Routledge, 213–33
Smith, Karen (1998). The Use of Political Conditionality in the EU’s Relations with Third
Countries: How Effective? European Foreign Affairs Review 3(2), 253–74
Smith, Rogers M. (2003). Stories of Peoplehood: The Politics and Morals of Political
Membership. Cambridge: Cambridge University Press
Soriano, Victor Fernández (2016). ‘Quel pays plus que la Grèce?’ La place de la Grèce
dans la construction de l’Europe: une mise en perspective historique. Histoire@Politique
29(2), 141–57
Bibliography 255
Stefanova, Boyka (2006). The ‘No’ Vote in the French and Dutch Referenda on the EU
Constitution: A Spillover of Consequences for the Wider Europe, PS: Political Science &
Politics 39(2), 251–5
Stirk, Peter M.R. and David Weigall, eds (1999). The Origins and Development of European
Integration. London: Pinter
Talesco, Cristian (2016). How East Timor’s Democracy is Making it an Outcast. Foreign Policy,
10 May. http://foreignpolicy.com/2016/05/10/how-east-timors-democracy-is-making-
it-an-outcast-asean-southeast-asia/
Al Tamamy, Saud (2015). GCC Membership Expansion: Possibilities and Obstacles.
Al-Jazeera Dossiers, 31 March. http://studies.aljazeera.net/en/dossiers/2015/03/201533
171547520486.html
Thomas, Daniel C. (2001). The Helsinki Effect: International Norms, Human Rights and the
Demise of Communism. Princeton: Princeton University Press
Thomas, Daniel C. (2009a). Rejecting the US Challenge to the International Criminal
Court: Normative Entrapment and Compromise in EU Policy-Making. International
Politics 46(4), 376–94
Thomas, Daniel C. (2009b). The Challenge of EU Policy-Making: Moving from Divergent
Preferences to Common Policies. International Politics 46(4), 335–8
Thomas, Daniel C. (2009c). Explaining the Negotiation of EU Foreign Policy: Normative
Institutionalism and Alternative Approaches. International Politics 46(4), 339–57
Thomas, Daniel C. (2011). Making EU Foreign Policy: National Preferences, European
Norms and Common Policies. Basingstoke and New York: Palgrave Macmillan
Thomas, Daniel C. (2012). Still Punching Below its Weight? Coherence and Effectiveness
in European Union Foreign Policy. Journal of Common Market Studies 50(3), 457–74
Thomas, Daniel C. (2017). Beyond Identity: Membership Norms and Regional Organization.
European Journal of International Relations 23(1), 217–40
Timmermans, Frans (2016). Community: Discovering Ties that Bind. London: Policy Network
Tipper, Steven P., Brehaut, Jamie C., and Driver, Jon (1990). Selection of Moving and Static
Objects for the Control of Spatially Directed Action. Journal of Experimental Psychology:
Human Perception and Performance 16(3), 492–504
Todorova, Maria (1997). Imagining the Balkans. Oxford: Oxford University Press
Tsebelis, George (2002). Veto Players: How Political Institutions Work. Princeton: Princeton
University Press
Trouvé, Mathieu (2008). L’Espagne et l’Europe: de la dictature de Franco à l’Union européenne.
Bruxelles: Peter Lang
United States Department of State (1977). Foreign Relations of the United States, 1950,
Volume III: Western Europe. Washington, DC: United States Government
Printing Office
Vachudova, Milada Anna (2005). Europe Undivided: Democracy, Leverage and Integration
after Communism. Oxford: Oxford University Press
Vanke, Jeffrey W. (2001). An Impossible Union: Dutch Objections to the Fouchet Plan,
1959–62. Cold War History 2(1), 95–112
Vieira, Alena Vysotskaya Guedes and Laura C. Ferreira-Pereira (2009). The European
Union-Cape Verde Special Relationship: The Role of Portugal. Portuguese Journal of
International Affairs 1 (Spring), 42–50
Waltz, Kenneth N. (1979). Theory of International Politics. Reading: Addison-Wesley
Wendt, Alexander (1992). Anarchy is What States Make of It: The Social Construction of
Power Politics. International Organization 46(2), 391–425
Wiener, Antje (2004). Contested Compliance: Interventions on the Normative Structure of
World Politics. European Journal of International Relations 10(2), 189–234
256 Bibliography
Wiener, Antje (2008). The Invisible Constitution of Politics: Contested Norms and
International Encounters. Cambridge: Cambridge University Press
Wiener, Antje (2014). A Theory of Contestation. Heidelberg: Springer
Wilson, Kevin and Jan van der Dussen, eds (1993). The History of the Idea of Europe.
London: Routledge
Wolczuk, Kataryna (2003). Ukraine’s Policy towards the European Union: A Case of
‘Declarative Europeanization.’ Online paper. http://www.batory.org.pl/ftp/program/
forum/eu_ukraine/ukraine_eu_policy.pdf
Yasseen, Mustafa Kamil (1976). L’interprétation des traités d’après la Convention de Vienne
sur le droit des traités. Académie de droit international. Recueil des cours, Volume III.
Leiden: Martinus Nijhoff
Index
Note: Figures and tables are indicated by an italic ‘f ’, ‘t’ and notes are indicated by ‘n’ following the
page numbers.
‘absorption capacity’ 79–82, 195, 200 Barroso, José Manuel 190–1, 199, 208, 210, 212
‘integration capacity’ 82 Basescu, Traian 210
accession Bech, Joseph 54
accession negotiations 15, 16–17, 234 Beckstein, Gunther 182
accession treaty 11, 14, 15, 174, 176, 213 Berkhouwer, Cornelis 162
membership eligibility as condition for Beyen, Johan-Willem 54
accession 13 Bildt, Carl 81, 195
membership eligibility/readiness for accession Birkelbach, Willi 62–4, 134
distinction 12, 13, 227, 231 Spain 140, 141, 143, 144
non-member states’ expressions of Birkelbach Report 64–7, 69, 137, 159
interest in 13 Spain and 139, 142, 144, 146
pre-accession treaty 14, 15 Bobinski, Krzysztof 212
qualified majority voting and Bolkestein, Frits 19, 77, 78, 184, 196
decision-making 234 The Limits of Europe 183
readiness for accession 168, 178 Borissov, Boyko 205–6
reform and 34, 233, 234 Bosbach, Wolfgang 182
Acharya, Amitav 228 Bosnia-Herzegovina 9
African states 62n.20, 148 Bot, Bernard 194, 214
African Union 39 Brentano, Heinrich von 59, 60, 119, 126
Albania 232–3 Brok, Elmer 182
Amsterdam Treaty 73 Brubaker, William Rogers 29
Anderson, Benedict 29 Bulgaria 232n.10
applicant state 14 Bull, Hedley 36
formal application 11–12 Bürgin, Alexander 235
membership norms and 27
see also candidate state Cancian, Francesca M. 30
ASEAN (Association of Southeast Asian candidate state 14, 234
Nations) 6, 228, 242 Czech Republic 4
aspirant state 233 potential candidate 12, 15, 230
membership eligibility 7, 16 ‘pre-candidate’ state 14
regional community and 7, 14–16 Turkey 5, 74, 88, 108, 180, 194, 201, 203, 205,
see also candidate state 206, 207, 215, 235
Ashton, Catherine 209 see also applicant state; aspirant state
Asselborn, Jean 203 Cape Verde 88
Association of Caribbean States 6 Caribbean Community 6
Austria 57–8, 73, 78 Castiella, Fernando M. 138–9, 140, 144
authoritarian regimes 149, 221, 240 Catalonia 233–4
Ayrault, Jean-Marc 205 Central Europe: EU membership 10, 15, 74, 179,
181, 194
Baldwin, Richard E. 42, 224 China 242
the Balkans 230, 232, 234, 237 Chirac, Jacques 78, 81, 190
Western Balkans 15, 191, 215, 231–2, 233 Christian Democrats (European parties) 63, 75,
Balkenende, Jan Peter 185 80, 144, 146, 152, 179, 190, 211
258 Index
CISC (International Confederation of Christian Council of Europe 61

Trade Unions) 140–1 1948 Statute 64, 65, 70
CISL (European Free Trade Union 1962–1969 membership norms
Confederation) 140–1 evolution 61–2, 65, 66, 67, 68
climate change 242 creation of 51
Cold War 21, 50, 57, 91, 104 democracy 4, 51–2, 57
post-Cold War period 10, 12, 44, 91, 104 Greece 150, 163, 164
Turkey 125 human rights 4, 51–2, 57
commercial interests membership 4, 51–2
Common Market 92 rule of law 51–2
customs union 18–19, 42, 43 Spain 145, 146, 173, 174
EU Eastern enlargement 42 Turkey 124, 128, 205
EU’s founding treaties 41–2 Council of the European Union 13
Greece 121, 160, 168, 222 Turkey 185, 206
hypotheses on community decision- Ukraine 210
making 43–4, 45t, 91–2, 99–100, 100t see also Council of Ministers
membership eligibility and 17, 18–19, 41–4, Council of Ministers (EEC) 13
224, 225 Greece 120, 153, 154–7, 162, 163–4, 165,
membership eligibility in statistical and 167–8, 170
comparative perspective 91–2, 99–100, Spain 134, 140, 141, 143, 144, 147, 172,
100t, 104, 106t 174, 175
OECD data 99, 100 Turkey 125, 127, 128–9, 131, 147–8, 178
relative gains and losses 42 see also Council of the European Union; the
Spain 136, 138, 147, 172, 175–6 Nine; the Six
Turkey 123–4, 127, 129, 130, 131, 132, 138, Council on Mutual Economic Assistance 93
148, 176, 178 Couve de Murville, Maurice 59, 60, 126, 134,
weakness of commercial explanations for 138, 143
membership eligibility decisions 100, 104 Croatia 186, 191, 226
World Bank data 24, 99, 100 Crowley, Úna 23
see also membership eligibility customs union
Common Market 64, 67, 224 commercial interests 18–19, 42, 43
commercial interests 92 Greece 123, 127, 157
ECSC 54 Turkey 127, 128, 129, 130, 132, 148, 177, 178,
EEC 59, 60, 119 186, 197, 199, 203
Spain 171, 224 currency 54
Turkey 132, 224 euro 237–8
see also Single Market Cyprus 74, 107, 160, 179, 194
Communism 52, 54–5 Turkey and 176, 186, 197, 198, 199, 200, 203
Conference on Security and Cooperation in Czech Republic 4, 20
Europe 241
conservatives (European parties) 77, 78, 80, 182 Dainotto, Roberto M. 10
constructivism 10–11 Davignon, Etienne 160–1, 171
constructivist theory of regional Davignon Report 70
community 27–37 De Block, August 149
‘Copenhagen criteria’ 5–6, 15, 73, 83, 187 De Gasperi, Alcide 53
Turkey 180, 182, 184, 185 De Gaulle, Charles 61, 63, 67, 119
Copenhagen summit Dehousse, Fernand 66, 149–50
1978 71, 72 democracy
1993 12, 73, 79 1957–1961 membership norms evolution
2002 181–2 51–2, 53, 54, 55, 57, 59
Coreper (Committee of Permanent 1962–1969 membership norms evolution 60,
Representatives, EEC) 88 61–2, 63, 65, 67
Greece 121, 151, 156–7, 161, 165 1970–2005 membership norms evolution
Spain 142 72–3, 74
Turkey 126–7, 129, 130 Council of Europe 4, 51–2, 57
Index 259
EEC and 21, 51, 59 EIB (European Investment Bank): loan to

EU membership norm 239, 241 Greece 123, 151, 152, 154, 155–7, 164
Greece 150–5, 161–2, 221, 223 ENP (European Neighborhood Policy) 88
membership norms evolution 26, 84 EPC (European Political Community) 53–4, 55,
membership rule 5, 12 56, 57, 60, 64, 71, 72
representative democracy 70, 71 Erbakan, Necmettin 180
Spain 171–3, 174, 175, 221 Erdoğan, Recep Tayyip 181, 182, 192, 200,
Treaty on European Union 72 203–4, 205, 206
Treaty of Rome 57, 59, 60, 97 Erhard, Ludwig 54, 59, 142
Turkey 127–8, 178, 180, 181, 193, 197, 201, EU (European Union)
206–7, 235 liberal democracy 26, 91, 239–41
Ukraine 187, 188, 193, 208 membership 4
see also liberal democracy; parliamentary membership decision-making 13, 14, 16
democracy membership eligibility, EU decisions on 21,
Denmark 16, 57–8, 160 22, 25, 26, 85, 94, 95, 103, 104, 107
De Pous, Jan Willem 143 member states 13, 16
DeWinter, Filip 77 EU, future of 229–42
Dijsselbloem, Jeroen 3 differentiated integration and
Dimitrov, Nikola 230 disintegration 237–8
Donno, Daniela 44, 101, 104 EU as bastion of liberal democracy 239–41
Downs, George W. 8–9 EU foreign policy, orientation and
Duvieusart, Jean 66 effectiveness of 241–2
future of enlargement 230–5
Eastern Europe 52, 104 Turkey question 235–7
commercial interests 42 EU: geographic limits 3, 7, 85, 219
EU Eastern enlargement 10, 11, 42, 77, 86, ‘absorption capacity’ and 80–1
181, 183 ambiguous borders 4, 5, 28, 219
EU membership 10, 15, 74, 179, 194, 241 avoiding the issue of 3, 6, 82, 194
membership norm and Eastern clear/fixed borders 4, 6, 77, 81, 195
enlargement 159 debate on 6, 49
Eastern Partnership 212–13 Encyclopædia Britannica 95, 97
Eastern Partnership Civil Society Forum 212 EU as open-ended space 5, 80, 183
Ecevit, Bülent 177, 181 European Commission 5, 78
economic integration 18, 21, 56, 123 European integration and 4
ECSC (European Coal and Steel European Parliament 80
Community) 53–5, 60, 64, 120, 224 European values and 219
1955 Benelux Memorandum 54 geographic contiguity 96
European common market 54 need of 75, 77, 80, 82, 183
EDC (European Defence Community) shifting nature of 28
53–4, 55 topographical borders 28, 77, 95, 108, 183
EEC (European Economic Community) 21 Turkey 95, 108, 149, 180, 192, 230
association as stepping-stone to full UN regional classifications 95
membership 58 EU Accession Index 12
Common Market 59, 60, 119 EU Charter of Fundamental Rights 74
democracy 21, 51, 59 EU Constitutional Treaty 78, 191
economic character of 139 EU enlargement 220
founding of 32, 50–1, 58 2006–present membership norms
human rights 21, 59 evolution 74, 79–83
liberal democracy 26 ‘absorption capacity’ 79–82, 195, 200
membership eligibility 4, 20 anti-enlargement sentiment 191
membership norms 26 Eastern enlargement 10, 11, 42, 77, 86,
OEEC and EEC membership 57–8, 121, 124, 181, 183
129, 142 economic inequality within the EU 231, 235
as pillar of the ‘free world’ 59 endless tendency to expand 4
EEC Treaty see Treaty of Rome ‘enlargement fatigue’ 214, 231
260 Index
EU enlargement (cont.) European Council 13

EU enlargement policy 215, 241 1970–2005 membership norms evolution
European Commission 82, 83, 191, 194–5, 71–2, 73–4
232, 233 2006–present membership norms evolution
European values and 193, 195, 219, 234 79, 81
future of 230–5 membership eligibility 15
membership norms 230–1, 235, 239 Turkey 179, 182, 185–6, 201, 204
membership rules 223 Ukraine 188, 190, 211, 212
political judgment and 207 European Court of Human Rights 52, 73
potential limits of 82, 191–2, 193 European integration 4, 7, 20, 229, 232
public opinion 214 1957–1961 membership norms evolution 54
reform 232 1962–1969 membership norms
scepticism about 78, 79, 233 evolution 61, 62, 63
‘waves’ of accession 86 1970–2005 membership norms evolution 73
see also membership eligibility 2006–present membership norms
Eurobarometer 184, 185, 226 evolution 74
‘European’ (term) ‘Copenhagen criteria’ 5–6
contestation of the term 6 EU’s geographic limits and 4
definition 5–6 inclusive/sequential approach 9
European Commission 3, 5, 108, 228–9 see also regional integration
European identity 5 European Movement 51, 52, 56
liberal democracy and 10, 11 Europeanness 10, 28
vagueness of the term 5 European Parliament
European Broadcasting Union 4 1970–2005 membership norms
European Commission evolution 71, 72, 74
1957–1961 membership norms 2006–present membership norms
evolution 57, 58, 59 evolution 76, 80
1970–2005 membership norms EU: geographic limits 80
evolution 71, 72, 74 membership decision-making 14
2006–present membership norms membership eligibility 15, 16
evolution 82, 83 membership norms evolution 49
‘absorption capacity’ 81, 82 Turkey 204, 206
‘Avis’ on Greece’s quest for accession 167–70 Ukraine 189, 212
Birkelbach Report 137 European Parliamentary Assembly (EEC)
democracy 159 1957–1961 membership norms evolution 51
‘Directorate General Europe’ 3 1962–1969 membership norms
EU: geographic limits 5, 78 evolution 60–3, 66, 68, 138
EU enlargement 82, 83, 191, 194–5, 232, 233 draft constitution 51
EU membership 4, 14 Greece 123, 151–5, 157, 162, 175
‘European’ (term) 3, 5, 108, 228–9 membership decision-making 14
Greece 120, 121–2, 151–7, 162, 167–70, membership norm 50
174–5, 192 Portugal 175
human rights 71 Spain 139, 140, 141, 143, 145–7, 172,
membership eligibility 5, 15, 16 174, 175
membership norms evolution 49 Turkey 126, 149
Portugal 174–5 European People’s Party 76, 185, 211, 212
Spain 135, 140, 144, 146, 147, 174–5, 192 European political union 5, 61–5, 67, 69, 82, 178
Treaty of Rome 71 European Social Charter 72
Turkey 123–4, 125, 126, 127, 128–9, 148, Eurovision (song contest) 4
177–8, 181, 182, 184, 185, 198–200, 203, Eurozone 237–8
205, 206, 207
Ukraine 188, 190–1, 209–10 fascism 21, 51, 52
European Convention on Human Rights and federalism 55, 60, 61, 63, 64, 65, 135, 141–2,
Fundamental Freedoms 52, 53, 70–1, 72, 146–7, 156
135, 149, 150, 174 Ferrero-Waldner, Benita 190
Index 261
Ferretti, Lando 153 1962–1969 membership eligibility 137, 138,

Fischler, Franz 183, 184 150–7, 158, 221
Fitzgerald, Garret 166 1967 freezing of association agreement 69,
Fortuyn, Pim 75 138, 157, 158, 222–3, 225
Fouchet, Christian 61, 63 1967 military coup/crisis 66, 69, 137, 150–7,
Fouchet committee 63, 64, 66, 67 158, 161, 222, 230
France 75, 78 1970–2005 membership eligibility 160–71,
veto by 16, 232–3 192, 221
Franco, Francisco 1970–2005 membership norms
death of 171–2 evolution 69, 72
fascist regime 61–2, 133, 134–5, 136, 139–47, 1974 collapse of the military junta 160, 161, 223
149, 158 applications for accession 72, 120, 221
Franco, Marc 212 civil society 156
François-Poncet, Jean 175 commercial interests and economy 121, 160,
Frattini, Franco 83 168, 222
Freedom House 93, 187, 240, 243–5, 244f, 245t Coreper 121, 151, 156–7, 161, 165
2015 ‘Nations in Transit’ report 240 Council of Europe 150, 163, 164
Freyburg, Tina 11 customs union 123, 127, 157
Füle, Štefan 202, 209, 210 democracy 150–5, 161–2, 221, 223
EEC 57–8, 59, 60
Gabriel, Sigmar 205 EEC association 120, 122, 123, 126–7, 138,
Gateva, Eli 11 148, 161–2
GATT (General Agreement on Tariffs and EEC Council of Ministers 120, 153, 154–7,
Trade) 120, 129 162, 163–4, 165, 167–8, 170
Genscher, Hans-Dietrich 170, 173 EEC membership 120–1, 123, 163, 165,
geography 28 170, 222
‘imaginative geographies’ 28–9 EIB loans 123, 151, 152, 154, 155–7, 164
region: geographic proximity 28, 37, 38 European Commission 120, 121–2, 151–7,
regional community: geographic limits 7, 162, 167–70, 174–5, 192
8, 27, 226 European Parliamentary Assembly 123,
see also EU: geographic limits; physical 151–5, 157, 162, 175
location human rights 156, 161
geopolitical threats see security interests ‘liberal democracies’ norm 161, 164, 165, 167,
Georgia 230 169–70, 175, 192, 200, 221, 223, 230
Germany membership eligibility 20, 79, 123, 157, 167,
Berlin crises 51, 52, 59 170, 230
Berlin Wall 130 membership norms hypothesis 121, 122, 222–3
CDU Party (Christian Democratic Union of NATO 121, 122, 126–7, 128, 154, 160, 222,
Germany) 77, 78, 79, 182, 190, 197 223, 225
CSU Party (Christian Social Union in the Nine 163, 164–5, 166–7, 170
Bavaria) 77, 78, 79, 182, 197 ‘non-Communist states’ norm 121, 122, 221,
EU’s refugee crisis 202 222, 230
Gibraltar 234 overthrow of parliamentary democracy
Giscard d’Estaing, Valéry 79 in 50, 67, 150
EU: geographic limits 6 ‘parliamentary democracies’ norm 157, 221,
European identity 77 223, 230
Greece 170–1, 230 rejection of Greek accession 137, 138, 150–7,
Spain 173–4, 175–6 221, 230
Turkey 19, 76, 125, 184 reopening to 160–71
Gosewinkel, Dieter 78 the Six 121–2, 123, 150–1, 153, 154, 155–7,
Greece 22 221, 222
1957–1961 membership eligibility 120–3, trade unions 151, 154, 156, 223
126, 136, 221, 222 Treaty of Rome 120, 150
1961 Athens Agreement 123, 148, 150–2, Grybauskaite, Dalia 210
154–5, 160, 166, 167, 171, 223, 225 Gulf Cooperation Council 6
262 Index
Hahn, Johannes 205, 206 irreconcilability of identities 29–30

Hallstein, Walter liberal democratic definition of European
basic values of the Community 70, 159 identity 81
Birkelbach Report 64, 67–8, 137 membership norms and 30, 33
EEC 58, 59 regional identity 10, 19–20, 28, 32, 227–8
Greece 122 socio-cultural identity 19–20
OEEC Eleven: EU membership for 57 Southeast Asian regional identity 228
Spain 134, 135 IMF (International Monetary Fund) 129, 133
Treaty of Rome 39, 56 immigration 75, 231, 237
Turkey 122, 125, 149 anti-immigrant sentiment 75, 77, 185, 191
Harmel, Pierre 151 integration of Muslim immigrants 77
Heckman Selection Model 12 Turkey, migration deal and EU’s refugee
Hemmer, Christopher 27, 227 crisis 202–6, 207, 236–7
Hillion, Christophe 215, 220, 239 Intergovernmental Conference on European
Hollande, Francois 211 Integration (1956) 55
human rights 70–1 inter-governmentalism 214–15, 228, 232–3
1957–1961 membership norms evolution 52, inter-governmental bargaining 194, 206, 207,
53–4, 55–6, 57 214, 226, 231, 234
1962–1969 membership norms evolution 60, membership eligibility, intergovernmental
61–2, 63, 65, 67 aspect 13–14, 49
1970–2005 membership eligibility 159, International Confederation of Free Trade
192, 221 Unions 156
1970–2005 membership norms evolution 69, International Criminal Court 242
70–1, 72–4, 221 Ireland 16, 57–8, 160, 169
Council of Europe 4, 51–2, 57 Islam 75, 76–7, 131
EEC 21, 59 Turkey, Islamic culture/religion 76–7, 131,
EU membership norm 32, 239, 241 179, 183–4, 230
European Commission 71 Italy 68–9, 236
Greece 156, 161
membership rule 5 Juncker, Jean-Claude 83, 203, 232, 233
the Six 71
Spain 61–2, 140, 144, 146, 171, 172, 174 Karamanlis, Konstantinos 160–1, 163–70, 223
Treaty on European Union 72 Katchanovski, Ivan 12
Treaty of Rome 57, 59, 60, 71, 97 Katzenstein, Peter J. 27, 227
Turkey 127–8, 177, 180–1, 182, 193, 197, 198, Kazakhstan 95, 108, 209
201, 203–5, 206–7, 235 Kelley, Judith G. 9
Ukraine 187, 208 Kern, Christian 203
Human Rights Commission 52, 150 Kinkel, Klaus 179
Hungary 55, 236, 239–40 Kissinger, Henry 241
Fidesz Party (Alliance of Young Democrats– Klimkin, Pavlo 212–13
Hungarian Civic Union) 240 Korean War (1950) 51, 52
Hurrell, Andrew 33 Koremenos, Barbara 9
Krasner, Stephen 33
Iceland 57–8 Kravchuk, Leonid 187
identity Kuchma, Leonid 187–8, 189, 208
‘Declaration on the European Identity’ Kurz, Sebastian 204, 205
70, 73n.50
European cultural identity 19–20, Lagorce, Pierre 172
76, 80 Lahr, Rolf 143
European identity 5, 19–20, 77–8, 80–2, 108, Lamassoure, Alain 3, 180, 182
175, 184, 194, 228–9 Lavrov, Sergei 209
European identity and Christianity 74–6, left-wing politics 52, 65, 211, 232
80, 183–4 Le Pen, Jean-Marie 75
identity politics 19 ‘liberal democracies’ norm
Index 263
Greece 161, 164, 165, 167, 169–70, 175, 192, Greece 20, 79, 123, 157, 167, 170, 230
200, 221, 223, 230 indicators of eligibility 14–15
Portugal 175 indicators of non-eligibility 15–16
Spain 171, 173, 192, 221 intergovernmental aspect 13–14, 49
Turkey 177, 179, 180, 185, 186, 192, 194, 221 member states, importance of 13, 16
Ukraine 160, 186–7, 188, 190, 191, 194 non-European regional communities 6
see also membership eligibility: 1970–2005; politics of 8, 10, 12
membership norms evolution: 1970–2005 readiness for accession/membership eligibility
liberal democracy distinction 12, 13, 227, 231
anti-liberal ideas 239 regional community and 3, 7, 12–13,
EC 26, 91 14–16, 17–20
EEC 26 scholarship on 8–13
EU 26, 91 Spain 4, 20, 171
EU as bastion of liberal democracy 239–41 Turkey 5, 19, 75–6, 123, 126, 132, 176, 177–8,
‘European’ (term) and 10, 11 179–80, 193, 196, 215, 235
European identity 81 Ukraine 4–5, 187, 188, 190, 193, 194, 208–10,
‘illiberal democracy’ 239–40 211, 213–14, 230
membership norms evolution 84 see also commercial interests; log-rolling;
regime type and membership eligibility 41, 98 membership norms; physical location;
see also democracy regime type; treaty rules
liberals (European parties) 63, 144, 152 membership eligibility: 1957–1961 (Europe of
limits of Europe 219 non-Communist states) 26, 119–20,
avoiding the issue of 3, 6, 82, 194 136, 221
see also EU: geographic limits Cold War 125
Linkevicius, Linas 210 Greece 120–3, 126, 136, 221, 222
Lipponen, Paavo 188 NATO 121, 122, 125, 126–7, 128, 136
Lipson, Charles 9 ‘non-Communist states’ norm 119, 121, 122,
Lisbon Treaty 74n.51, 82–3 125, 126, 127, 128–9, 130, 132–3, 134–5,
location see physical location 136, 221
log-rolling 225–6, 231 Spain 133–6, 221
see also membership eligibility Treaty of Rome 120, 124, 132, 134
Luns, Joseph 149 Turkey 123–33, 136, 147, 221
see also Greece; Spain; Turkey
Maastricht Treaty see Treaty on European Union membership eligibility: 1962–1969 (Europe of
Macmillan, Harold 62 parliamentary democracies) 26, 137–8,
Macron, Emanuel 230, 233 158, 221
Madagascar 62n.20, 148 Greece 137, 138, 150–7, 158, 221
Malta 74, 101, 194 ‘parliamentary democracies’ norm 137–8,
Mansfield, Edward D. 42, 96 146–7, 149, 157, 221
Mansholt, Sicco 67 Spain 137, 138–47, 158, 221, 225
Marjolin, Robert 58 Turkey 137, 147–50, 158, 221
Mavros, Georgios 163 see also Greece; Spain; Turkey
Mayer, Franz C. 19 membership eligibility: 1970–2005 (Europe of
Mediterranean Sea 77, 95, 183 liberal democracies) 26, 159–60, 191–3, 221
membership eligibility Greece 160–71, 192, 221
argument and alternative explanations 17–20, human rights 159, 192, 221
25, 36–46, 221–6 ‘liberal democracies’ norm 159–60, 161,
avoiding the issue of 4 164, 165, 167, 169–70, 171, 173, 175, 177,
definition 7, 13–17 179, 180, 185, 186–7, 188, 190, 191, 192,
EU decisions on 21, 22, 25, 26, 85, 94, 95, 103, 193, 221
104, 107 rule of law 159–60, 192
European Commission 5, 15, 16 Spain 171–6, 192, 221
European Council 15 Turkey 176–86, 192, 221
European Parliament 15, 16 Ukraine 159–60, 186–91
264 Index
membership eligibility: 2006–present (divided membership eligibility and 36–7, 194

Europe) 26, 195–6, 214–15, 221–2 membership eligibility in statistical and
intergovernmental bargaining 194, 206, comparative perspective 90, 93–4, 94t, 104,
207, 214 105t, 107, 108–9, 110, 112, 115, 220–1
normative dissensus 196, 198, 199, 200, 205, membership norm/membership rule
207, 208, 213, 221–2 distinction 30–1
Turkey 196–207, 214, 215, 222 nature of 29–31
Ukraine 207–14, 215, 222 regional community and 7, 27–8, 29, 30, 33,
membership eligibility in statistical and 226, 229, 237, 241–2
comparative perspective 26, 85–6, Spain 135
115, 220–1 strength of membership norms explanation
commercial interests 91–2, 99–100, 100t, for membership eligibility decisions 94,
104, 106t 107, 108–9, 110, 112, 115
cross-tab analysis 26, 89, 92–109, 105t, 110, Turkey 132–3
111, 115, 220–1 membership norms: change in 23, 25, 27, 30,
data and methods 86–9 84, 219–20
EU decisions on membership eligibility of external challenge 32, 83
aspirant states 86, 87t, 88 internal challenge 32, 83
explanations and hypotheses 89–92 normative changes 31–3, 49, 60, 74–5, 78, 79,
FH data (Freedom House) 93, 243–5, 83–4, 138, 146–7, 154, 159, 161, 229
244f, 245t rhetorical inconsistency 32, 83–4
geography 90, 94–6, 95t, 96t, 104, 105t, shift in public opinion 33, 83
107–8, 221 see also membership norms evolution
logistic regression 26, 85, 89, 108, 109–10, membership norms: effect of 7–8, 27, 33–5
110t, 111t, 115, 221 membership eligibility 7, 33, 219–20
membership norms 90, 93–4, 94t, 104, 105t, normative empowerment 33, 34–5, 36
107, 108–9, 110, 112, 115, 220–1 normative guidance 34, 35
Polity IV 93–4, 244 normative selection 33–4, 35
Qualitative Comparative Analysis 26, 85, 89, membership norms evolution 25–6, 31–3, 49,
108–9, 111–12, 113t, 114t, 115, 221 83–4, 219, 221–2
regime type 91, 98, 99t, 104, 105t, 109–10 democracy 26, 84
security interests 92, 100–104, 101t, 103t, European Commission 49
104, 106t European Parliament 49
treaty rules 91, 97, 98t, 104, 105t, 108 genealogical methods 25
V-Dem (Varieties of Democracy dataset) 93, liberal democracy 84
94, 98, 99t, 243–5, 244f member states 49
membership norms 3, 17, 20, 229 normative dissensus 74, 80, 81–3, 84, 195,
community’s identity 30, 33 207, 221, 237, 242
constructivist theory of regional variations over time 24, 36, 49, 60, 69, 84, 89,
community 27–37 137–8, 146–7, 154, 159, 161, 239
content of 7–8, 31 see also membership norms: change in;
contestation of 27, 31–3, 36, 49, 74, 84, membership norms evolution: 1957–1961;
219, 239 membership norms evolution: 1962–1969;
definition 30 membership norms evolution: 1970–2005;
differentiated integration and membership norms evolution:
disintegration 238 2006–present
EU enlargement 230–1, 235, 239 membership norms evolution: 1957–1961
EU membership norm 23 (‘non-Communist states’ norm)
genealogy of 22–3 49–60, 221
Greece 121, 122, 222–3 Cold War 57
hypotheses on community decision-making Communism 52, 54–5
and 36–7, 45t, 90, 94t, 222 Council of Europe 51–2, 57
identity and 30, 33 democracy 51–2, 53, 54, 55, 57, 59
legitimacy of membership scenarios and 7 ECSC 53–5, 60
limits of Europe 219 European Commission 57, 58, 59
Index 265
European integration 54 EU enlargement 74, 79–83

European Parliamentary Assembly 51 European Commission 82, 83
human rights 52, 53–4, 55–6, 57 European Council 79, 81
Intergovernmental Conference on European European identity 74–8, 80–2
Integration 55 European integration 74
‘non-Communist states’ norm 53, 55, 57, 58, European Parliament 76, 80
59, 83, 93, 119, 121, 122, 125, 126, 127, hard/intergovernmental bargaining 17,
128–9, 130, 132–3, 134–5, 136, 221 45t, 50, 74, 79, 107, 194, 195, 206, 207,
parliamentary democracy 52 213, 214
the Six 50, 53, 54, 56, 57–60, 221 immigration 75, 77
Soviet threat 51, 52, 54–5, 59, 60, 83 liberal democratic values 234
Treaty of Rome 50–1, 56, 57–60 Lisbon Treaty 82–3
see also EEC; EPC normative dissensus 74, 80, 81–3, 84, 93,
membership norms evolution: 1962–1969 194–5, 196, 198, 199, 200, 205, 207, 221–2,
(‘parliamentary democracies’ norm) 50, 232, 239
60–9, 221 normative dissensus and lack of foreign policy
association with the EEC 65–6, 67–8 unity 242
Birkelbach Report 64–7 political discourse by non-governing political
British membership 61, 62, 63 elites 74–5, 78–9
Common Market 64, 67 public opinion 74, 79
Council of Europe 61–2, 65, 66, 67, 68 Turkey 75–6, 77–8
democracy 60, 61–2, 63, 65, 67 membership question, significance of 3–8
European integration 61, 62, 63 membership and regional integration
European Parliamentary Assembly 60–3, theories 3, 8–11
66, 68, 138 constructivism 10–11
European political union 61–5, 67, 69 ‘convoy’/‘club’ model of membership 9
federalist vision of Europe 60, 61 inclusive/sequential approach 8–9
human rights 60, 61–2, 63, 65, 67 membership restrictiveness/enforcement
Italian foreign ministry memorandum 68–9 problems relationship 9
‘parliamentary democracy’ norm 62, 63, 64, rationalism 11
66, 67, 69, 83, 93, 137–8, 146–7, 157, 221 social identity theory 10
rule of law 65 membership rule 5, 9, 30, 40, 70
the Six 61, 63–4, 67, 68, 69, 83, 221 democracy 5, 12
Treaty of Rome 60 EU enlargement 223
membership norms evolution: 1970–2005 (‘liberal human rights 5
democracies’ norm) 50, 69–74, 221 market economy 5, 12
Davignon Report 70 membership norm/membership rule
democracy 72–3, 74 distinction 30–1
European Commission 71, 72, 74 protection of minorities 5
European Council 71–2, 73–4 rule of law 5
European integration 73 member states 13–14, 16
European Parliament 71, 72, 74 membership norms evolution 49
Greece 69, 72 see also inter-governmentalism
human rights 69, 70–1, 72–4, 221 Menderes, Adnan 123–4, 127–8
‘liberal democracies’ norm 70, 72, 74, 93, Merkel, Angela 79, 213
159–60, 161, 164, 165, 167, 169–70, 171, EU’s refugee crisis 202
173, 175, 177, 179, 180, 185, 186–7, 188, Turkey 77, 78, 79, 182–3, 184, 191, 196,
190, 191, 192, 193, 194 197–8, 200, 202
minorities protection 73, 74 Messina conference 55, 66
rule of law 69, 72–3, 74 methodology 20–5
SEA 72 causal-process tracing of eligibility decision-
membership norms evolution: 2006–present making 24–5, 223, 224, 225, 229
(normative dissensus) 50, 74–83, 194, 221–2 country-case selection 22
‘absorption capacity’ 79–82 cross-tab analysis 24, 26, 89, 92–109, 105t,
culture wars 33, 77 110, 111, 115, 220–1, 223
266 Index
methodology (cont.) ‘non-Communist states’ norm

dependent variables and analytical Greece 121, 122, 221, 222, 230
methods 22 Spain 134–5, 221
discourse analysis 23 Turkey 125, 126, 127, 128–9, 130, 132–3, 221
discursive-process tracing 20, 23, 24 see also membership eligibility: 1957–1961;
empirical research design 20–5, 37 membership norms evolution: 1957–1961
EUR-Lex database 24 Nord, Hans 56
genealogy of membership norms 22–5 Nordhaus, William 101
hypotheses 36–7, 38, 40, 41, 43–4, 45t, 46, 90–2 normative dissensus
logistic regression 24, 26, 85, 89, 108, 109–10, membership norms evolution 74, 80, 81–3,
110t, 111t, 115, 221, 243, 244 84, 195, 207, 221, 237, 242
Polity IV 24, 93–4, 244 Turkey 196, 198, 199, 200, 205, 207,
Qualitative Comparative Analysis 24, 26, 85, 221–2, 235
89, 108–9, 111–12, 113t, 114t, 115, 221 Ukraine 207, 208, 213, 221–2
regional case selection 20–1 see also membership eligibility: 2006–present;
sources and data 21, 22–3, 24, 195–6 membership norms evolution:
statistical analyses of eligibility positions 22, 2006–present
23–4, 223, 229 norms/normative approach 226–9
V-Dem (Varieties of Democracy dataset) 24, norm contestation 227, 228, 238
243–5, 244f norm decay 238
see also membership eligibility in statistical see also membership norms; membership
and comparative perspective; rhetoric norms: change in; membership norms:
minorities protection 5, 41, 82, 178, 198 effect of; membership norms evolution
1970–2005 membership norms North Macedonia 232–3
evolution 73, 74 Norway 16, 57–8
Mogherini, Federica 203, 205
Moldova 230 OECD (Organisation for Economic
Monnet, Jean 52–3, 54 Co-operation and Development) 99, 100
Moravcsik, Andrew 10, 42 OEEC (Organisation for European Economic
Morocco 107 Cooperation) 62, 133, 134, 143
Morozov, Vlatcheslav 10 EEC membership and 57–8, 121, 124, 129, 142
multiculturalism 77, 79, 185 Open Society Foundations 234
Orbán, Viktor 192, 206, 234, 239–40
nation 29 Ortoli, Francois-Xavier 161, 167, 168
nationalism 81, 232, 233
NATO (North Atlantic Treaty Palmowski, Jan 19
Organization) 61, 62, 227 parliamentary democracy 52
1957–1961 membership eligibility 121, 122, Greece 50, 67, 150
125, 126–7, 128, 136 Greece, ‘parliamentary democracies’
enlargement 10, 12 norm 157, 221, 223, 230
Greece 121, 122, 126–7, 128, 154, 160, 222, Spain, ‘parliamentary democracies’
223, 225 norm 146–7, 221
MAP (Membership Action Plan) 103 Turkey, ‘parliamentary democracies’
PfP (Partnership for Peace) 103 norm 147, 149
security interests and NATO see also membership eligibility: 1962–1969;
membership 102–3, 103t, 225 membership norms evolution: 1962–1969
Spain 136, 225 Pesmazoglou, Yiangos 163
Turkey 122, 125, 126–7, 128, 148, 225 Pevehouse, Jon C. 42, 84, 96
see also security interests Pflueger, Friedbert 190
Neumann, Iver B. 10 physical location 6, 8
Neves, José Maria 88 Encyclopædia Britannica 95, 95t
Nicolaides, Kalypso 5, 219, 222 geographic contiguity 96, 96t
the Nine (EEC) 70, 72, 160 hypotheses on community decision-
Greece 163, 164–5, 166–7, 170 making 38, 45t, 90, 95t, 96, 96t
Spain 171, 172, 173, 175–6 membership eligibility and 6, 8, 17–18, 37–8
Index 267
membership eligibility in statistical and incentives for 12

comparative perspective 90, 94–6, 95t, 96t, Spain 133
104, 105t, 107–8, 221 Turkey 130, 180–1, 182, 184, 186, 193, 198–9,
region 28 201, 222
strength/weakness of geography explanations Ukraine 187–8, 208, 215
for membership eligibility decisions 96, regime type
104, 107–8 hypothesis on community decision-
territorial contiguity and economic making 41, 45t, 91, 98, 99t
interdependence 17–18 liberal democracy 41, 98
Treaty of Rome 37 membership eligibility and 17, 18, 40–1
UN regional classifications 95, 96t membership eligibility in statistical and
see also EU: geographic limits; membership comparative perspective 91, 98, 99t, 104,
eligibility 105t, 109–10
Piri, Kati 205, 206 selection-by-identity 41
Pitella, Gianni 205 Spain 136
Pleven, René 53 weakness of regime type explanations for
Plümper, Thomas 11, 12 membership eligibility decisions 98, 104
Pofalla, Ronald 197 see also membership eligibility
Poland 54, 236, 239, 240 region 27, 28, 226–9
policy geographic proximity 28, 37, 38
EU enlargement policy 215, 241 regional identity 10, 19–20, 28, 32, 227–8
EU foreign policy, orientation and regional community
effectiveness of 241–2 aspirant states 7, 14–16
policy integration 9 constitution of Europe as political
policy outcomes 35, 120 community 5
policy preferences 9, 12, 23, 28, 30, 31, 33, constructivist theory of 27–37
34–5, 41, 64, 66, 169, 195, 238 definition 7, 27, 29
populism 232 geographic limits of 7, 8, 27, 226
Poroshenko, Petro 209, 212 international organizations and 30, 226
Portugal membership eligibility and 3, 7, 12–13,
applications for accession 72, 88 14–16, 17–20
EEC 57–8, 59, 175–6 membership norms 7, 27–8, 29, 30, 33, 226,
European Commission 174–5 229, 237, 241–2
European Parliamentary Assembly 175 nature of regions and regional community 28–9
‘liberal democracies’ norm 175 regional construction 226–7
Salazar’s authoritarianism 149 regional governance 7
Posen, Barry R. 44 regional integration 7, 226–9, 237
Prodi, Romano 188, 199 differentiated integration and
PTA (preferential trade agreement) 42 disintegration 237–8
public opinion see also European integration; membership
2006–present membership norms and regional integration theories
evolution 74, 79 regionalism 28
change in membership norms and 33, 83 Rehn, Olli 6, 78, 191, 193, 198–200, 219
EU enlargement 214 Rey, Jean 67, 128–9, 130, 131, 143–4, 153–4
Turkey 184–5, 192, 198, 226 rhetoric
Ukraine 192, 213 discursive-process tracing 23
membership eligibility 10
Rajoy, Mariano 233–4 membership norms: rhetorical
Rasmussen, Anders Fogh 81 inconsistency 32, 83–4
Ratzinger, Josef (Cardinal) 76, 183 rhetorical entrapment 35n.2, 231
realism 44, 46 right-wing politics
reform Austria 73
accession and 34 centre-right 65, 75, 77, 81, 211n.81
EU accession and 12, 233, 234 European identity and Christianity 75
EU enlargement 232 far-right 52, 73, 75, 77, 202, 232
268 Index
Risse, Thomas 19 membership eligibility and 17, 19, 24, 27,

Romania 181, 232n.10 44, 46, 224
Rosato, Sebastian 44 membership eligibility in statistical and
Ruggie, John Gerard 18, 34, 36, 227 comparative perspective 92, 100–104,
rule of law 5 101t, 103t, 104, 106t
1962–1969 membership norms evolution 65 NATO membership 102–3, 103t, 225
1970–2005 membership ‘soft balancing’ 44
eligibility 159–60, 192 Spain 136
1970–2005 membership norms evolution 69, Turkey 133
72–3, 74 US 44
Council of Europe 51–2 weakness of security explanations for
EU membership norm 239, 241 membership eligibility decisions 103, 104
Spain 174 see also membership eligibility; NATO
Treaty on European Union 72 Sedelmeier, Ulrich 10, 11
Turkey 180, 193, 198, 201, 202, 203, Segni, Antonio 143
206–7, 235 Shanghai Cooperation Organization 6
Ukraine 160 Sikorski, Radek 210–11
Rumelili, Bahar 4, 10, 28, 30 Simonet, Henri 173
Russett, Bruce 84 single market 18–19, 42, 54
Russia 183, 242 EU Single Market 5, 224, 241
annexation of Crimea by 209, 214 see also Common Market
Ukraine and 187, 188, 189, 191, 209, 211, 213 the Six (Belgium, France, Germany, Italy,
see also Soviet Union Luxembourg, the Netherlands)
Rutte, Mark 213 1957–1961 membership norms evolution 50,
Rutten, Charles 125, 127, 128, 131 53, 54, 56, 57–60, 221
Rybachuk, Oleh 190 1962–1969 membership norms evolution 61,
63–4, 67, 68, 69, 83, 221
Said, Edward 28–9 ‘democratic character’ of 60
Salazar, António de Oliveira 149 ECSC 53, 54
sanctions 13n.7, 73, 238, 240 EEC 50, 61, 119
Sarkozy, Nicolas 79, 80, 82, 183, 184 Greece 121–2, 123, 150–1, 153, 154, 155–7,
Sasse, Gwendolyn 11 221, 222
Sauvagnargues, Jean 164–5 human rights 71
Schäuble, Wolfgang 190, 203–4 OEEC 57–8
Schaus, Eugéne 143 Spain 135, 139–43, 144
Schily, Otto 182 Treaty of Rome 56, 57–60
Schimmelfennig, Frank 10, 11, 35n.2, 228, 235 Turkey 123, 125, 126–7, 128–9, 130–1,
Schmidt, Helmut 165 147–8, 221
Schneider, Christina 11, 12 Skålnes, Lars S. 10
Schroeder, Gerhard 190 Slovak Republic 4, 20
Schuman, Robert 49, 52 Smith, Karen 11
Schüssel, Wolfgang 78–9, 80, 81, 184, 185–6, Smith, Roger M. 29
192, 197, 226 Snidal, Duncan 9
Scotland 233–4 Soames, Christopher 164
SEA (Single European Act, 1986) 14, 72, 83, 178 socialists (European parties) 63, 134, 152, 205
changes of EU treaty rules on membership Spain 140, 142, 145, 146
eligibility 97 Soviet Union 93
Searle, John 34 Communist threat posed by 51, 52, 54–5,
security interests (geopolitical threats) 59, 60, 83
alliance credibility 100, 103, 104, 225 invasion of Hungary by 54–5
conflict risk 100–1, 103–4, 225 see also Russia
cross-border movements 44 Spaak, Paul-Henri 53, 54, 55, 122, 133,
hypotheses on community decision- 142, 143
making 45t, 46, 92, 101t, 103t Spain
inter-state conflict 44 1957–1961 membership eligibility 133–6, 221
Index 269
1962–1969 membership eligibility 137, Switzerland 57–8, 86

138–47, 158, 221, 225 Szajarto, Peter 204
1970–2005 membership eligibility 171–6,
192, 221 terrorism 75, 76–7, 182
applications for accession 72, 133, 138–9 Thorn, Gaston 196
Birkelbach Report 139, 142, 144, 146 Timmermans, Frans 27
commercial interests and economy 136, 138, Todorova, Maria 10
147, 172, 175–6 trade unions
Common Market 171, 224 Greece 151, 154, 156, 223
Coreper 142 Spain 140–1, 142, 146–7
Cortes 144, 172 Treaty on European Union (Maastricht
Council of Europe 145, 146, 173, 174 Treaty) 4n.4, 72, 97
democracy 171–3, 174, 175, 221 Article 48 234
EEC 58, 59, 60, 61, 68 Article 49 82, 189, 192, 194–5, 208,
EEC association 133, 138 210–11, 212
EEC Council of Ministers 134, 140, 141, 143, changes of EU treaty rules on membership
144, 147, 172, 174, 175 eligibility 97
EEC membership 135, 174, 176 democracy 72
European Commission 135, 140, 144, 146, human rights 72
147, 174–5, 192 membership norm 72, 83
European Parliamentary Assembly 139, 140, member states’ role in assessing and
141, 143, 145–7, 172, 174, 175 approving membership applications 13
Franco’s fascist regime 61–2, 133, 134–5, 136, reframing of 194–5
139–47, 149, 158 religious heritage 76
geographical and historic links to rule of law 72
Europe 143, 175 Ukraine 189, 208, 210–11, 212
human rights 61–2, 140, 144, 146, 171, Treaty of Paris 5, 53, 55
172, 174 Treaty of Rome (EEC Treaty) 3, 4n.4, 14, 31
‘liberal democracies’ norm 171, 173, 192, 221 1957–1961 membership eligibility 120, 124,
membership eligibility 4, 20, 171 132, 134
membership norm hypothesis 135 1957–1961 membership norms evolution
Munich declaration 146 50–1, 56, 57–60
NATO 136, 225 1962–1969 membership norms evolution 60
the Nine 171, 172, 173, 175–6 absence of democracy and human rights as
‘non-Communist states’ norm 134–5, 221 criteria for membership 57, 59, 60, 71, 97
‘parliamentary democracies’ norm Article 237 56, 58, 63, 64, 71, 97, 120, 124,
146–7, 221 166, 174, 223–4
reform 133 Article 238 56, 58, 65, 120, 124, 147, 166
regime type hypothesis 136 association agreements 56–8, 59–60, 224
rejection of Spanish accession 137, 138–47, contestation of 60–1
158, 221, 225 debates and negotiations on 50–1, 58
reopening to 171–6 European Commission 71
rule of law 174 ‘freedom’ 56
security interests hypothesis 136 Greece 120, 150
the Six 135, 139–43, 144 on membership 5, 37, 39, 50, 56, 57,
trade unions 140–1, 142, 146–7 58–60, 63, 107
Treaty of Rome 144, 146, 147 physical location 37
treaty rules hypothesis 136 preamble 56, 97
Spinelli, Altiero 72 the Six 56, 57–60
Statsch, Patrick D. 85–115, 243–5 Spain 144, 146, 147
Stier, Davor 241 Turkey 124, 132
Stirk, Peter M.R. 53, 57 treaty rules
Stoiber, Edmund 75, 78, 79, 197 changes of EU treaty rules on membership
Suárez, Adolfo 172–4 eligibility 97
Sweden 57–8 EUR-Lex database 24
270 Index
treaty rules (cont.) customs union 127, 128, 129, 130, 132, 148,
hypothesis on community decision-making 177, 178, 186, 197, 199, 203
40, 45t, 91, 97, 98t Cyprus and 176, 186, 197, 198, 199, 200, 203
membership eligibility and 17, 18, 38–40, deadlock on Turkish accession 222, 235–7
223–4 death penalty 204, 205
membership eligibility in statistical and democracy 127–8, 178, 180, 181, 193, 197,
comparative perspective 91, 97, 98t, 104, 201, 206–7, 235
105t, 108 EEC 57–8, 59, 60, 61, 88
procedural rules 38–9 EEC association 121, 123–4, 125–7, 130–2,
Spain 136 138, 148–9
strength/weakness of treaty rules explanations EEC Council of Ministers 125, 127, 128–9,
for membership eligibility decisions 97, 131, 147–8, 178
104, 108 EEC/EU membership 124, 128, 132, 148–9,
substantive rules 38, 39, 40 177, 182
treaty articles 39–40 European Commission 123–4, 125, 126, 127,
treaty preambles 39–40 128–9, 148, 177–8, 181, 182, 184, 185,
Treaty of Rome 39 198–200, 203, 205, 206, 207
Turkey 132 European Council 179, 182, 185–6, 201, 204
see also membership eligibility European Parliament 204, 206
Troeger, Vera 12 European Parliamentary Assembly 126, 149
Tuomioja, Erkki 6, 82, 194 the Fifteen 180, 181
Turkey future of Turkey’s accession 235–7
1957–1961 membership eligibility 123–33, geographic limits of Europe and 95, 108, 149,
136, 147, 221 180, 192, 230
1960 military coup/rule 61, 88, 123, 127–8, human rights 127–8, 177, 180–1, 182, 193,
130, 136, 139 197, 198, 201, 203–5, 206–7, 235
1962–1969 membership eligibility 137, Islamic culture/religion 76–7, 131, 179,
147–50, 158, 221 183–4, 230
1963 Ankara Agreement 148–9, 176, 177–8 ‘jihadis’ returning home from Syria 206
1970 Additional Protocol 176 ‘liberal democracies’ norm 177, 179, 180, 185,
1970–2005 membership eligibility 176–86, 186, 192, 194, 221
192, 221 membership eligibility 5, 19, 75–6, 123, 126,
1980 military coup/rule 176–7 132, 176, 177–8, 179–80, 193, 196, 215, 235
2006–present membership membership norms hypothesis 132–3
eligibility 196–207, 214, 215, 222 migration deal and EU’s refugee crisis 202–6,
2006–present membership norms 207, 236–7
evolution 75–6, 77–8 NATO 122, 125, 126–7, 128, 148, 225
2016 attempted coup d’état 203, 205, 235 ‘non-Communist states’ norm 125, 126, 127,
accession negotiations 197–206, 235 128–9, 130, 132–3, 221
AKP (Justice and Development Party) 75–6, normative dissensus 196, 198, 199, 200, 205,
181–2, 236 207, 221–2, 235
breakdown of EU–Turkey relations opening to Turkey 147–50, 158
197–206, 222 opposition to Turkish accession 75–6, 77–8,
candidate status 5, 74, 88, 108, 180, 194, 201, 82, 131, 179, 181, 182–5, 196, 197, 199, 202,
203, 205, 206, 207, 215, 235 207, 230, 236
Cold War 125 ‘parliamentary democracies’ norm 147, 149
commercial interests and economy 123–4, ‘Positive Agenda’ 202
127, 129, 130, 131, 132, 138, 148, 176, 178 ‘privileged partnership’ 77, 184, 185, 201,
Common Market 132, 224 226, 236
controversy on Turkey’s membership 176, public opinion 184–5, 192, 198, 226
179, 180, 182–5, 196, 203 reaffirming the opening to 176–86
‘Copenhagen criteria’ 180, 182, 184, 185 reform 130, 180–1, 182, 184, 186, 193, 198–9,
Coreper 126–7, 129, 130 201, 222
Council of Europe 124, 128, 205 rule of law 180, 193, 198, 201, 202, 203,
Council of the European Union 185, 206 206–7, 235
Index 271
security interests hypothesis 133 Russia and 187, 188, 189, 191, 209,
the Six 123, 125, 126–7, 128–9, 130–1, 211, 213
147–8, 221 Treaty on European Union 189, 208,
socio-cultural identity 19 210–11, 212
terrorism 182 Turkey/Ukraine comparison 192–3
Treaty of Rome 124, 132 veto on 213
treaty rules hypothesis 132 UN (United Nations)
Ukraine/Turkey comparison 192–3 UN Charter 51
US and 125 UN regional classifications 95, 96t
veto on 132, 181, 186, 194, 197, 198, 214 US (United States) 242
Tusk, Donald 211, 233 security interests and 44
Tymoshenko, Yulia 209 Suez crisis 55
Turkey and 125
UK (United Kingdom) 57–8
Brexit 4, 233, 234, 238, 241 Vachudova, Milada Anna 10, 42
EEC membership 61, 62, 63, 134, 146, 148, Van den Broek, Hans 188
160, 169 Van der Goes Van Naters, Marinus 62
France’s veto on negotiations with 16 Van der Stoel, Max 170
Ukraine Van Gogh, Theo 76–7
1970–2005 membership eligibility Van Mierlo, Hans 179
159–60, 186–91 Van Rompuy, Herman 183–4, 212
2006–present membership eligibility 207–14, Van Sherpenberg, Hilger 125
215, 222 Vatican 76, 183
corruption 187, 193 Ventotene Manifesto 64, 72
Council of the European Union 210 Verheugen, Günter 184, 197, 198
Deep and Comprehensive Free Trade veto
Agreement 211 decision-making on membership 39, 43, 44,
democracy 187, 188, 193, 208 181, 194, 197, 198, 232–3, 234
economy 187, 188 France’s veto 16, 232–3
EU association 208–9, 210, 211–14 treaty rules 39
EU membership exclusion 9, 88, 188, 207, Turkey 132, 181, 186, 194, 197, 198, 214
209, 211–14, 215 Ukraine 213
‘Euromaidan’ protests 5, 88, 209–10, 214 visa liberalization 203, 234
European Commission 188, 190–1, 209–10
European Council 188, 190, 211, 212 Wallström, Margot 190
European Parliament 189, 212 Warsaw Pact 21, 65, 93, 225
geographic limits of Europe and 192 Weigall, David 53, 57
human rights 187, 208 Wendt, Alexander 19, 228
‘liberal democracies’ norm 160, 186–7, 188, Werner, Pierre 85
190, 191, 194 Wigny, Pierre 58–9, 60
membership eligibility 4–5, 187, 188, 190, 193, Wilders, Geert 77
194, 208–10, 211, 213–14, 230 World Bank 24, 99, 100, 133
membership perspective 5
normative dissensus 207, 208, 213, 221–2 xenophobia 75
‘Orange Revolution’ 189, 191, 192, 208, 209
PCA (Partnership and Cooperation Yanukovych, Viktor 4, 189, 191, 209
Agreement) 187, 188 Yatsenyuk, Arseniy 209
public opinion 192, 213 Yekhanurov, Yuri 191
reform 187–8, 208, 215 Yelisieiev, Kostyantin 212
rule of law 160 Yushchenko, Viktor 189, 190, 208

The Limits of Europe - Membership Norms and The Contestation of Regional Integration

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Limits of Europe - Membership Norms and The Contestation of Regional Integration

Uploaded by

Copyright:

Available Formats

OUP CORRECTED PROOF – FINAL, 17/10/21, SPi

The Limits of Europe

Preface and Acknowledgements

vi Preface and Acknowledgements

Leiden, December 2020

1. The Question of Membership 3

3. The Evolution of EU Membership Norms 49

5. Membership Eligibility in a Europe of Non-­Communist States,

9. Rethinking Europe, Rethinking Regions 219

A.1. Comparison of Freedom House and V-­Dem country-­year scores 244

2.1. Summary of explanations and hypotheses 45

Note on Archival Sources

Research for this book included extensive consultation of documents in various

ACCUE: Archives Centrales, Conseil de l’Union Européenne (Brussels)

AEI: Archive of European Integration: http://aei.pitt.edu/

Any European State may apply to become a member of the Community.

By way of introduction, this chapter discusses the importance of understanding

1. The question and its significance

4 The Limits of Europe

The Question of Membership 5

6 The Limits of Europe

and immutable aspect of European integration nor a complete explanation of the

The Question of Membership 7

2001: 2). Instead, we need a clear definition of membership eligibility, a theoretical

8 The Limits of Europe

2. Insights and gaps in the scholarly literature

The Question of Membership 9

10 The Limits of Europe

The Question of Membership 11

question. Furthermore, instead of exploring how EU values, norms, and rules

12 The Limits of Europe

The Question of Membership 13

theoretical or methodological framework for studying this in a manner conducive

3. Defining membership eligibility

To say that a non-­ member state is eligible for membership in a regional

14 The Limits of Europe

• the decision to hold unconditional preliminary talks on membership with a

The Question of Membership 15

non-­member state seeks ultimately to gain membership or just to develop func-

• an explicit refusal to hold preliminary talks on accession or a failure to issue

16 The Limits of Europe

The Question of Membership 17

4. The argument and alternative explanations, in brief

So how can we best understand a state’s eligibility for membership in a political

1. Membership norms. Membership norms express a regional community’s

18 The Limits of Europe

The Question of Membership 19

states with inferior administrative capacities out of the customs union/

20 The Limits of Europe

And even if it were possible to observe European cultural identity at particular

5. Empirical design and methods

Regional case selection—why study Europe?

As indicated above, this book is focused empirically on regional community

The Question of Membership 21

22 The Limits of Europe

Country-­case selection—which cases and why?

As discussed above, this book aims to explain EU decision-­ making on

Dependent variables and analytical methods

Genealogy of membership norms

The Question of Membership 23

states from others in the international system. To supplement these archival

Statistical analyses of eligibility positions

1. The Question of Membership 3

5. Membership Eligibility in a Europe of Non-Communist States,

A.1. Comparison of Freedom House and V-Dem country-year scores 244

2.1. Summary of explanations and hypotheses 45

To say that a non- member state is eligible for membership in a regional

non-member state seeks ultimately to gain membership or just to develop func-

Country-case selection—which cases and why?

As discussed above, this book aims to explain EU decision- making on

Causal-process tracing of eligibility decision-making

Membership norms and community decision-making

would be no community-level norms relevant to the membership eligibility of

Another approach to explaining decision- making in a regional community

1. A community of non-Communist states, 1957–1961