Journal of European Public Policy 13:5 August 2006: 607 – 626

The common sense of European

integration
Giandomenico Majone

ABSTRACT The rejection of the Constitutional Treaty and the various events
following the negative referenda provide an excellent occasion for reconsidering
the real meaning of European integration. Paradoxically, the integration process is
often praised for its clumsy and ultimately unsuccessful attempts to mimic the
nation state, while its truly important contribution to European civilization – the
establishment of a supranational constitutional order – is belittled or even
ignored. An example of this distorted vision is the debate on the so-called democratic
deficit – a condition which could be easily corrected if a majority of Europeans sup­
ported a supranational federal state. Since it is obvious that no such majority exists,
now or in the foreseeable future, the ‘democratic deficit’, however defined, is the
price we pay for wishing to integrate our national economies while preserving
the core of national sovereignty. The current crisis is methodological rather
than systemic: it amounts to a rejection of the stealthy approach to European
integration – cryptofederalism – which has entailed the triumph of process over
outcome. The legitimacy problem of the EU can be solved by limiting, rather
than continuously expanding, the competences of the supranational institutions.The
institutional system established by the founding fathers was not designed for effective
policy-making, but largely to pursue objectives of negative integration.
KEY WORDS Community method; constitutional referenda; cryptofederalism;
democratic deficit; harmonization and race to the bottom; legitimacy standards;
positive and negative integration; supranational constitutionalism.

Il fallait une révolution pour ramener les hommes au sens commun

(J-J. Rousseau, Discours sur les sciences et les arts)
If not a revolution in the sense intended by Rousseau in the above quotation –
where he was referring to such history-making events as the fall of Constantino­
ple in 1453 and the birth of Renaissance civilization – the events of the second
half of 2005 certainly represent a turning point in the less heroic history of the
European Union (EU). As such, they provide an excellent occasion to reconsider
the common sense of European integration: what are the really significant and
unique features of this process rather than its clumsy and failed attempts to
Journal of European Public Policy
ISSN 1350-1763 print; 1466-4429 online # 2006 Taylor & Francis

http:==www.tandf.co.uk=journals

DOI: 10.1080=13501760600808212

608 Journal of European Public Policy

mimic the nation state. The ambitious, but highly ambiguous, goal of ‘ever
closer union’ has been a constant source of confusion and misunderstandings.
Consider, for example, the much debated question of the ‘democratic deficit’
of the EU – the absence or underdevelopment of the institutions and processes
of parliamentary democracy at the European level. Now, there would be a simple
way of eliminating such a deficit: if the majority of European voters desired to be
governed by a full-fledged European federal state, they could use the electoral
process to force their national leaders to transfer sufficient powers and resources
to a democratic and federal union. To deny this is to assert that the democratic
process in the member states is seriously flawed – which is not only patently
false, but contrary to the basic assumption of the EU that all its member
states are democratic. The simple truth is that after half a century of integration
there is no majority, or even a sizeable minority, advocating a federal solution.
On the other hand, the majority of Europeans are in favour of some form of inte­
gration: even among those who voted against ratification of the Constitutional
Treaty in summer 2005, 83 per cent in France and 73 per cent in the Nether­
lands expressed support of continued EU membership. We may conclude that
the so-called democratic deficit is, so to speak, democratically legitimated,
being the price voters are willing to pay in order to preserve the essential
elements of national sovereignty. However, and this is the other important
lesson of the recent events, this price remains acceptable only as long as the com­
petences of the EU remain limited. In the following pages I try to present what I
take to be the ‘common sense’ of European integration – simple truths that have
been obscured by an overly optimistic and/or naive literature on European inte­
gration. I do this by revisiting some of the key arguments presented in Dilemmas
of European Integration, a book published in March 2005, in light of the
constitutional débâcle and of the events that followed it in the second half of
2005 – the true annus horribilis of the European Union.

1. THE REVOLT OF THE MASSES AND THE DESERTION

OF THE ELITES
The most striking aspect of the French and Dutch referenda on the Consti­
tutional Treaty is the huge gap between elite and popular opinion. All the
major political parties – on the left and on the right, in government and in
opposition – the vast majority of members of the national parliaments, econ­
omic leaders (in the Netherlands, also trade union leaders), major newspapers
and other media, the European Parliament and Commission – all supported
ratification, at least officially. In the Netherlands, two-thirds of Members of
Parliament voted for the new constitution, while almost two-thirds of the citi­
zens voted against. The rejection of the new constitution did not come as a sur­
prise, either in France or in the Netherlands, having been predicted by virtually
all opinion polls. What was surprising in both cases were the high voter partici­
pation (compared to all previous European elections) and the size of the negative
vote: almost 55 per cent against ratification and 45 per cent in favour in France;
 G. Majone: The common sense of European integration 609
61.5 and 38.1 per cent, respectively, in the Netherlands. In an extraordinary
meeting in Brussels in early June 2005, the presidents of the Commission, of
the European Parliament and of the EU Council at first tried to minimize
what had happened. They insisted that the ratification process must continue
so that at the end of 2006, when the process was supposed to be completed,
a general reassessment of the situation would be made. The hopes of the
European leaders were dashed by the British decision of 6 June to postpone
indefinitely the referendum originally scheduled for the first half of 2006.
Denmark, the Czech Republic and Poland soon followed the British example,
reinforcing the impression of many commentators that the Constitutional
Treaty – which was supposed to provide the institutional framework for the
enlarged Union – was effectively dead. The last popular referendum was held
in Luxembourg on 10 July 2005. As was to be expected in a country which
benefits enormously, both politically and economically, from EU membership,
the ‘yes’ prevailed, but the shock came from the 40 per cent who voted against
ratification in spite of massive support for the Treaty by all political parties,
trade unions and economic interests, and despite the threat by Prime Minister
Juncker that he would resign in case of rejection.
Never before had the distance separating the common people from the poli­
tical, economic and intellectual elites appeared so great. The title of Ortega y
Gasset’s bestseller of the 1930s – The Revolt of the Masses – fits perfectly the situ­
ation currently prevailing in Europe. Most analysts agree that the causes of the
French and Dutch rejection of the constitution were socio-economic rather
than political or ideological: fear of unemployment caused by eastern enlarge­
ment and the possible admission of Turkey; concerns about the low rate of econ­
omic growth in the Eurozone; and general disappointment about the failure of
monetary union to produce all the wonderful results, including price transpar­
ency and price stability, that had been promised by the political leaders. The
common currency, which should have provided tangible proof that the inte­
gration process had reached the point of no return, was cited by one-third of
Dutch voters as one significant reason for their rejection of the new constitution.
Even before the result of the Dutch referendum was known, an opinion poll
published by the popular German magazine Stern indicated that 56 per cent of
Germans would have liked to return to the Deutschmark, 48 per cent thought
that the problems of the economy were also due to the euro, and 90 per cent
made the common currency responsible for the price increases of the last
years. According to the same magazine, the Minister of Finance, the president
of the Bundesbank and a select group of economists had met secretely to
discuss the possibility and consequences of a collapse of monetary union. In a
similar vein, in August the London Times discussed the hypothesis that Italy
might leave Economic and Monetary Union (EMU) if the European Central
Bank (ECB) did not loosen its restrictive monetary policy. In both cases, the
revelations, true or alleged, were promptly and officially denied, but one
thing was at any rate clear: the debate about the wisdom of an economically
premature and politically driven monetary union had moved beyond a
610 Journal of European Public Policy
narrow circle of specialists to become public. The rhetorical strategy of the ECB
changed accordingly: while previously the monetary authority had dismissed
doubts about EMU simply by repeating that the euro – like a De Beers
diamond – ‘is forever’, now it started to argue more analitically about the
costs of leaving EMU. These, then, were some of the indications of the level
of popular dissatisfaction with the elitist methods followed in fifty years of
European integration. How did the elites react?
In the book already referred to, Ortega y Gasset suggests that the revolt of the
masses is closely correlated to the desertion of the elites, by which term
the Spanish philosopher meant the lack of plans, objectives and ideals of the
leaders. This terminology seems again a fair characterization of the current
crisis of leadership in the EU. Even more than the absence of contingency
plans so apparent after the constitutional débâcle, and of clear perspectives for
the future, the return to brutal methods of defence of the national interest
revealed the depth of the crisis. Of course, national interests always become
highly visible in moments of crisis. Nevertheless a behavioural rule had devel­
oped over the years, according to which the defence of the national interest
should not exclude concessions to the other member states and at least the
formal acknowledgement of the existence of a higher common interest. De
Gaulle, Chirac and Mrs Thatcher are the exceptions that confirm the general
rule. At the Brussels summit of 16 –17 June 2005, where the financial frame­
work for the period 2007 to 2013 was to be defined, the style of interaction
among the national leaders reached levels of stubborness seldom, if ever, experi­
enced before. Thus, in its defence of a narrowly construed national interest, the
Dutch Prime Minister Jan Balkenende was as irremovable as the British,
Spanish, Swedish and Finnish prime ministers, or as the French president.
His country was tired, Balkenende effectively said, of being the main paymaster
of the Union (on a per capita basis). Unless the net contribution of the
Netherlands was significantly reduced, there would be no long-term financial
framework. The position of the leader of a traditionally ‘federalist’ country
illustrates a key point, developed in a later section of this article, about the
impossibility for a would-be European federal state to pursue significant re-
distributive policies across countries and regions, and as already suggested,
without a full-fledged federal state there can be no true parliamentary democracy
at the European level. Before developing these themes further, however, it is
useful to examine critically the kind of roundabout, or cryptofederalism that
has been practised for the last half century.

2. CRYPTOFEDERALISM
After the collapse, in 1954, of the European Defence Community (EDC),
and consequently of the European Political Community (EPC) which was sup­
posed to provide a pre-federal, democratic framework for the EDC and for the
Coal and Steel Community, most federalists chose to continue to fight,
so to speak, underground: political integration by stealth. I use the label
 G. Majone: The common sense of European integration 611
‘cryptofederalism’ to characterize the new approach to integration, so different
from that of the early, ‘Hamiltonian’, federalists who openly worked for a con­
stitution dividing the powers of government between a federal Europe and its
member states, with democracy at each level. In the decade following the end
of the Second World War, Hamiltonian federalism had been a significant
factor in European politics, especially in Germany, Italy and the Netherlands.
Even realpolitiker like Konrad Adenauer and Alcide de Gasperi had been
attracted, albeit for a short time, by the federalist vision. By the mid-1950s,
however, this vision had already lost its credibility and whatever popular
support it had enjoyed in the immediate post-war period. The main reason:
all the analyses and predictions it had inspired had been clearly refuted by
experience. Federalists like Altiero Spinelli had repeated ad nauseam that
none of the dramatic problems of post-war Europe could be solved without
first building a powerful European federal state on the smoking ruins of the
nation state: not the economic, social and democratic reconstruction of the
continent, nor the solution of the German problem, or an effective defence of
the western part of the continent against the Soviet threat. Soon it became
clear, however, that all these problems were being solved, more or less well,
without any radical political innovation. Indeed, perceptive analysts could
already see that the nation state, far from withering away, was in the process
of becoming the omnicompetent European welfare state. This loss of credibility
of the federalist vision explains why by the mid-1950s the demise of the EDC
and EPC was received with a sigh of relief not only in Paris, but also in Bonn
and Rome, not to mention London.
The rise of cryptofederalism can only be understood against this background
of (to use Popperian language), conjectures and refutations. Paul-Henry Spaak
and Jean Monnet are the best-known leaders of federalist revisionism, a method
which, like the Marxist brand, is characterized by its concentration on means
and its indifference to distant ends. True, in June 1955 Monnet had left the
High Authority of the European Coal and Steel Community to set up the
Action Committee for the United States of Europe, and thus he is generally con­
sidered to have been an open, rather than a crypto- federalist. But as Max Kohn­
stamm, one of his closest associates, later explained, Monnet used the expression
‘United States of Europe’ more as a tribute to the USA, a country he knew well
and loved, than as a clear ideological commitment. In fact, the French statesman
and ‘father of Europe’ used several expressions more or less interchangeably: in
addition to United States of Europe also ‘European entity’, ‘union’, ‘federation’
and ‘confederation’ (Kohnstamm 1989).
Also for Spaak ‘everything which tends toward European organizations’
was good, independently of what such organizations might be or do. Indeed,
in 1949 he commended a proposal for common European postage stamps as
having equal value with any other proposal and concluded that: ‘It is when
we will have worked in directions which will sometimes appear perhaps as a
little opposed, even contradictory, that it will even so be possible after all to
make a synthesis’ (cited in Milward 1992: 324). In sum, for both founders of
612 Journal of European Public Policy
criptofederalism as for Eduard Bernstein, the founder of nineteenth-century
German revisionism, the final aim was nothing, the movement everything (Gay
1962: 74). Under what may be called the ‘Monnet method’ – related to, but
distinct from, both the classic Community method (see the next section), and
also, I would argue, from neofunctionalism – it is never clear whether European
policies are initiated in order to solve some concrete problem which cannot be
tackled at the national level, or whether they are to serve some unstated (insti­
tutional or political) objectives. Economists know that attempts to pursue
several objectives with the same policy instrument usually produce sub­
optimal outcomes. What is perhaps less appreciated is that when several objec­
tives are pursued simultaneously, accountability is reduced to vanishing point.
In Dilemmas of European Integration (see especially pp. 108– 9) I identify this
ambiguity about goals and instruments as the major methodological flaw of
the Monnet method. When objectives are multiple and ill-defined it becomes
easy, and often convenient, to confuse means and ends, process and outcome.
This ambiguity worked for a while, but could not be sustained indefinitely.
After all, voters, like consumers, are interested in results, not in process.
When they see that monetary union, far from producing the promised economic
benefits, seems to impede growth and job creation, they tend to view the euro,
not as the visible symbol of closer political integration, but as an arbitrary con­
straint imposed on them by the powers that be. Hence the widespread dissatis­
faction with the European institutions – rather than with European integration
as such – that pollsters have detected since the French and Dutch votes. Dis­
satisfaction with the European institutions implies dissatisfaction with their
methods: the current crisis is methodological rather than systemic.

3. RISE AND DECLINE OF THE CLASSIC COMMUNITY

METHOD
As codified by the Commission in its White Paper on European Governance
(Commission 2001: 12), the Community method, which defines the role of
the various European institutions and the modes of their interactions, rests on
three principles:
(1) The Commission is independent of the other European institutions; it alone
makes legislative and policy proposals. Its independence is meant to
strengthen the ability to execute policy, act as the guardian of the Treaty,
and represent the Community in international negotiations.
(2) Legislative and budgetary acts are adopted by the Council of Ministers and
the European Parliament, always on a proposal made by the Commission.
(3) The European Court of Justice (ECJ) guarantees the maintenance of the
balance among European institutions, and respect for the rule of law.
Several distinctive features of this scheme should be noted. First, the
Commission and the ECJ are major players in areas where the Community
method applies. Their importance is enhanced by the fact that most
 G. Majone: The common sense of European integration 613
European policies are regulatory (Majone 1996). Second, these major
players are non-elected, politically independent bodies – non-majoritarian
institutions – operating within poorly specified accountability structures. For
instance, even though the Commission may be submitted to parliamentary
censure, in practice the threat of censure is often a strategic weapon in the per­
manent inter-institutional struggle which characterizes EU politics, rather than
an instrument of control of the Commission by the directly elected Parliament.
Even if the EP took seriously the task of enforcing public accountability, it
would run against two major obstacles. First, the collegial nature of the Com­
mission: Parliament is understandably reluctant to dismiss the entire college
in order to sanction a single commissioner – in spite of this possibility
having been offered by the Amsterdam Treaty. Second, because the Commis­
sion has been assigned a large number of separate tasks, it is extremely costly
to dismiss it even when there is intense dissatisfaction with how it carries out
one such task. Thus, control of the policy agenda and absence of clear account­
ability standards allow the Commission to pursue objectives of political inte­
gration and self-aggrandizement while pretending to solve specific policy
problems: integration by stealth.
The most striking feature of the Community method, however, is its rejec­
tion both of the model of parliamentary democracy and of the principle of sep­
aration of powers. The Commission’s monopoly of legislative and policy
initiative – its agenda-setting power – has no analogue either in parliamen­
tary or in presidential democracies. In parliamentary systems, legislators intro­
duce relatively few bills; most legislative proposals are instead presented by
bureaucrats to the Cabinet, which then introduces them as draft legislation to
Parliament. Once legislators receive such proposals, however, they are free to
change or reject them. This is not the case under the Community method
where as a rule the Council may modify Commission proposals only under
the stringent requirement of unanimity. In parliamentary systems, moreover,
neither civil servants nor their political masters can pre-empt the right of initiat­
ive of parliamentary parties and individual members of the legislature. Besides,
national executives are the expression of the party or coalition which won the
elections, and this is certainly not the case of the Commission. In a separ­
ation-of-powers system like the United States, on the other hand, only legis­
lators can introduce bills. In the course of a typical congressional term,
members of Congress will introduce several hundred bills on behalf of the Pre­
sident or of executive-branch agencies, but during the same period they will
introduce on their own behalf as many as 15,000 or 20,000 bills.
It is important to understand clearly what is implied by the Commission’s
monopoly of agenda-setting. First, other European institutions cannot legislate
in the absence of a prior proposal from the Commission. It is up to this insti­
tution to decide whether the Community should act and, if so, in what legal
form, and what content and implementing procedures should be followed.
Second, the Commission can amend its proposal at any time while it is under
discussion in the Committee of Permanent Representatives of the member
614 Journal of European Public Policy
states, or in the Council of Ministers, while the Council can amend the proposal
only by unanimity. If the Council unanimously wishes to adopt a measure
which differs from the Commission’s proposal, the latter can deprive the
main Community legislator of its power of decision by withdrawing its own
proposal. Finally, neither the Council nor the EP nor a member state can
compel the Commission to submit a proposal, except in those few cases
where the Treaty imposes an obligation to legislate. In addition to its monopoly
of legislative initiative, the Commission, as guardian of the Treaty, can take
autonomous decisions in order to determine whether member states have com­
plied with their treaty obligations, or to permit them in appropriate cases to
deviate from their obligations. In some cases the Commission can also take
general measures (directives) without Council approval, for example, to
ensure that state-owned, as well as private, firms satisfy European rules on com­
petition (Article 86 of the EC Treaty). The member states have repeatedly, but
always unsuccessfully, challenged the powers of the Commission under this
article.
The delegation of such extensive agenda-setting powers to the Commission,
under the Community method, is clearly a means used by the member states to
enhance the credibility of their commitment to European integration (Majone
1996; Moravcsik 1998). However, this particular ‘technology of commitment’
presupposes that the delegated powers will not be abused. The decline of the
Community method begins when the member states become convinced that
the Commission is incapable or unwilling to exercise self-restraint. September
1991 is a convenient point in time from which to observe the progressive
decline of the method. This is the date when the Dutch presidency proposed
that both the common foreign and security policy (CFSP) and justice and
home affairs (JHA) were to be brought within the sphere of application of Com­
munity law. What was proposed, in other words, was the realization of that con­
stant aspiration of the European Commission: the generalization of the
Community method. As so often in the history of the EC/EU, this abortive
attempt to establish an explicit commitment to a pre-federal Europe produced
a backlash and sharpened conflicts among the (then) twelve member states. The
proposal of the Dutch presidency was supported by the Commission and by the
EP, as one would expect, but was rejected by all the other countries, except
Belgium. As a result, the negotiations leading to the Treaty on European
Union (TEU) were pursued on the basis of the Luxembourg scheme of a Euro­
pean Union founded on three separate pillars. In the institutional architecture
designed by the TEU the Community method is largely confined to the first
pillar. The other two pillars (CFSP and JHA) were smaller, more specific in
scope, and only partly adjusted to that method. Here the European Council –
the only distinct organ of the EU as a whole – can act without a prior proposal
from the Commission, and is not subject to any control by the EP, or to the
scrutiny of the Court of Justice, except indirectly. The Commission is supposed
to be ‘fully associated’ with the tasks in CFSP and JHA, but its power to monitor
the fulfilment of treaty obligations by the member states was explicitly omitted
 G. Majone: The common sense of European integration 615
and its power of agenda-setting severely restricted. Under the TEU, the EP is
simply informed about decisions taken in CFSP and JHA, and may express
its opinion, while the general jurisdiction of the ECJ is explicitly excluded.
Article 34 of the Treaty stipulates that the legislative decisions which may be
adopted in the area of police and judicial co-operation in criminal matters
shall not entail direct effect: they do not confer rights on individuals which
may be enforced before national courts.
Even in the first pillar, the Maastricht Treaty introduced new competences in
a way that limits the exercise of Community powers, by explicitly ruling out har­
monization of national laws and regulations. Since harmonization is the most
powerful legal instrument available to the Commission to pursue positive inte­
gration (see section 7 below), the limits imposed on its use are a clear indication
of the progressive loss of confidence in this institution’s capacity for self-
restraint. The framers of the Amsterdam and Nice Treaties followed the same
strategy of excluding harmonization when assigning new competences to the
Community level. As Joseph Weiler (1999) observed, the member states,
rather than relying on implicit competences, whose limits (under Article 235
of the Treaty of Rome) seemed out of control, opted for an explicit grant
that delimits the modes of action and the reach of the new policies. Even the
grant of co-decision powers to the EP by the TEU, and the extension of
those powers by the Amsterdam and Nice Treaties, entailed a non-negligible
cost to the Commission. Under the procedure of Article 251 EC, if the
Council accepts all the amendments contained in the EP’s opinion, it may
adopt the proposed act, as amended, by a qualified majority. This means that
the Commission cannot oppose an act on which the Council and EP are in
agreement, whereas, as we know, in general the Council can modify a Commis­
sion’s proposal only by a unanimous vote.
Other signs of erosion of the Commission’s monopoly of policy initiation are
visible elsewhere. For instance, the system of social dialogue established by the
Maastricht Treaty allows officially recognized representatives of employers and
workers to enter into voluntary agreements that are subsequently enacted as
directives by the Council. The social dialogue, when successful, conforms
with the Community method on the output side, since it leads to a legally
binding directive, but not on the input side, since the initiative for proposing
legislation rests with the social partners, not with the Commission (Scott and
Trubek 2002). Arguably the most ominous challenge to the Commission’s
exclusive power of agenda-setting comes from the growing tendency of the
European Council to deliberate on specific policy issues at thematic summits,
such as the Lisbon Summit on economic and social policy, the Luxembourg
Summit on employment, or the Tampere Summit of October 1999, a true land­
mark for policy and institutional developments in the area of Justice and Home
Affairs.
One of the key functions of the European Council is, of course, to define the
general policy guidelines of the Union, but it can also deal with whatever pro­
blems it chooses, in the manner it judges most appropriate. Not being (yet) a
616 Journal of European Public Policy
European institution, the powers and decision-making procedures of the
Council are not determined by the Treaty. This means, in particular, that it
is not legally bound to follow the Community method or to respect the
treaty-based rights and privileges of the Commission. At the June 2002
Summit in Seville, for instance, the heads of state or government, in order to
improve decision-making in Brussels, agreed to extend the time frame in
which their key decisions are taken by drawing up annual and three-year pro-
grammes. According to press reports, the Commission’s president was seriously
concerned that if the Council decided on such strict planning without previous
agreement with the Commission, there would inevitably be serious conse­
quences for his institution’s monopoly of legislative initiative. The Consti­
tutional Treaty, if ratified, would have further weakened the position of the
Commission and the role of the Community method, especially by the pro­
visions (Article I-21) concerning the President of the European Council,
elected by a qualified majority of the Council for a term of two and a half
years, renewable once.
In conclusion, there is clear evidence that the Commission, and arguably also
the Court of Justice, on many occasions have used the Community method well
beyond the limits envisaged by the drafters of the Treaty of Rome, and that the
member states have reacted to this lack of self-restraint by limiting the scope of
delegation to the supranational institutions. Concerning those limits, textual
analysis shows that the Treaty put more emphasis on negative than on positive
integration, whereas the Commission has consistently tried to reverse this
ranking. As I tried to show in Dilemmas of European Integration, the results
could only be disappointing since the key concept of the Community
method is the preservation of institutional balance, not policy activism.
However, it would be a serious mistake to discard the method altogether.
Rather, the method should be preserved, and even strengthened, but its appli­
cation should be mainly restricted to the enforcement of negative integration
(see section 7 below).

4. DEMOCRACY AND INTEGRATION: THE BIG TRADE-OFF

As was seen in the preceding section, the democracy vs. integration dilemma –
which the early federalists had either ignored or denied, but the inescapable
reality of which became obvious in the mid-1950s – has been resolved by the
Community method in favour of integration. Later attempts to reduce this so-
called ‘democratic deficit’ by such devices as the direct election and steady expan­
sion of the powers of the European Parliament, and various ‘social’ programmes
– from the agricultural, regional and cohesion policies to pro-labour regulations
– have failed entirely to alter the nature of the basic dilemma. It could not have
been otherwise, given the lack of popular support for the federal vision. Euro­
pean integration was always, and as the recent events have shown, continues
to be, an elitist project. It is striking that even in countries like Germany
where, for historical and geopolitical reasons, support for European integration
 G. Majone: The common sense of European integration 617
was very strong, popular consensus never went beyond passive acceptance of the
agreements reached in Brussels by the national government. Perhaps even more
than the Germans, the Dutch used to have the reputation of being great
champions of European integration. ‘Nonetheless, Dutch enthusiasm for
Europe has been the result of a pragmatic assessment of national interests
rather than an unmitigated commitment to federalism. Indeed, the federal
route is quickly dismissed when it may conflict with Dutch interests’ (Klabber
1998: 355). Recent events prove the correctness of this analysis.
Of course, all key concepts of modern history, from popular sovereignty to
the idea of nation and the principle of nationality, were originally advanced
by intellectual and political elites – were ‘invented’, to use the language of
Edmund Morgan’s well-known book, Inventing the People (1988). But these
ideas proved their vitality by their capacity to mobilize the people and push
them into political action. In addition, the idea of a politically and economically
united continent has been propagated by some of the most distinguished intel­
lectual and political leaders of twentieth-century Europe: names like Couden­
hove-Kalergi, Aristide Briand, Denis de Rougemont, the Spanish philosopher
Ortega y Gasset, and two future Nobel Prize-winning economists – Maurice
Allais and Lionel Robbins – come to mind. Determined to realize the ideas
of these men, after the Second World War a new generation of federalists
became convinced that only by appealing to the peoples of Europe could a
federal state be built against the resistance of the national governments and
parliaments. Unfortunately, federalist ideas proved unable to mobilize the
masses and push them into political action. In the course of half a century a
certain Europeanization of intellectual, economic and political elites has taken
place, but this process has hardly touched the large majority of citizens. No
‘Europeanization of the masses’ has taken place even remotely comparable to
that ‘nationalization of the masses’ analysed so perceptively by George Mosse
in the case of Germany, but which occurred in all countries of Western
Europe at the end of the Napoleonic wars (Mosse 1975).
It is important to understand that the decision to proceed with economic
integration, postponing political integration to an indefinite future, entailed
the acceptance of the old liberal principle of the separation of politics and econ­
omics. This principle, which at the national level seemed to have been repealed
since the Great Depression of the 1930s, if not sooner, was recognized by the
founding fathers of the European Economic Community as an essential prequi­
site for integrating the national economies of a group of states traditionally
wedded to the principles of state interventionism and protectionism, and extre­
mely jealous of their political sovereignty. Only by separating, as far as possible,
politics from economics was it feasible to minimize the effect on the integrating
European market of different legal orders, national economic structures and
systems of administration and citizenship. It should be noted that acceptance
of the necessity of separating (national) politics from (supranational) economics,
and hence the willingness to sacrifice democracy in order to protect the process
of market integration, has nothing to do with an ideological choice in favour of
618 Journal of European Public Policy
‘neoliberalism’. Indeed, it is hard to imagine such a choice being made by
statesmen raised in the tradition of dirigisme, and still practising it at home
within the limits set by European law, and often beyond such limits. The
simple truth is that, absent popular support for a federalist solution, the only strat­
egy open to cryptofederalists à la Monnet – sectoral or general economic inte­
gration, on the road to ‘ever closer union’ – could not be pursued using
democratic means. This, in a nutshell, is the common sense of European inte­
gration, and I find it puzzling that those who keep complaining about the ‘demo­
cratic deficit’ seem to be unable to understand it. If they are really concerned about
the insufficient development of democracy at European level they should support
a drastic reduction of delegated tasks under the Community method. Instead,
they grasp at pseudo-solutions, such as giving ever increasing powers to the EP.

5. DIFFERENT STANDARDS OF LEGITIMACY

In Chapter 2 of Dilemmas of European Integration (Majone 2005) I argue that to
speak of a democratic deficit of the current EU is to commit a ‘category
mistake’, which consists in discussing certain facts as if they belonged to one
logical type or category, when they actually belong to another. Philosophers
point out that concepts or standards that may be legitimately applied to one cat­
egory become meaningless or seriously misleading when applied to a different
category (Ryle 1949). To exemplify, it is a category mistake to discuss an insti­
tution such as a university in the same terms in which one discusses its depart­
ments, professional schools or administrative offices: the university is not
identical with any one of these units; rather, it is the way the units are organized.
The pitfall consists in thinking of the university as if it stood for an extra
member of the class of which department, schools and offices are members.
A similar category mistake consists in discussing the EU with the same con­
cepts we use for its component units – as if the Union were a state rather than
an organization of corporate bodies: the member states and the European insti­
tutions. The principle which Alexander Hamilton considered ‘the great and
radical vice’ of the American Confederation – namely, ‘the principle of legis­
lation for states or governments in their corporate or collective capacities, and as con­
tradistinguished from the individuals of which they consist’ (cited in Rakove
1997: 191, emphasis in original) – still prevails at the European level. Judicial
doctrines of direct effect and supremacy, and the fact that European law can
create rights and duties for citizens of the member states – rights which can
be enforced only by national courts – do not change this basic fact, which
has again been acknowledged by the ill-fated Constitutional Treaty. True,
there is also a citizenship of the Union, but only for persons already holding
the nationality of a member state: ‘Citizenship of the Union shall complement
and not replace national citizenship’ (Article 17 EC). In other words, citizenship
is not an autonomous concept of European law, but is defined exclusively by the
legislation of the member states. Again, since 1979 we have a directly elected
European Parliament consisting of ‘representatives of the peoples of the States
 G. Majone: The common sense of European integration 619
brought together in the Community’ (Article 189 EC), but its competences do
not include those essential powers of a true legislature: the power to tax and
spend, to initiate legislation, and (in parliamentary systems) to form a govern­
ment. Moreover, the EP is seriously deficient as a system of representation of
individual interests. These interests are still largely rooted at the national level
and hence find their natural expression in national parliaments and political
parties. In sum, the EP cannot represent a (non-existent) European people in
the same sense in which national democratic institutions represent an histori­
cally defined demos.
To deny that the notion of ‘democratic deficit’ is applicable to the current EU
is not to deny the existence of a legitimacy problem. Rather, it is to say that
democratic legitimacy is only one member of a class of normative standards
used in assessing public institutions. In a democracy, all non-majoritarian insti­
tutions – including constitutional or supreme courts, independent central
banks and independent regulatory commissions – have a legitimacy problem,
but this does not mean that they suffer from a ‘democratic deficit’: by design,
such institutions are not accountable to the voters or to their elective represen­
tatives. Rather, they have to acquire their legitimacy by other means: they must
prove their ‘distinctive institutional competence’ by generating and maintaining
the belief of being, of all feasible institutional arrangements, the most appropri­
ate one for a given range of problems (Choper 1980). In the case of the EU, for
example, the role of the European Court of Justice could not be played as effec­
tively by any other European institution. But also a body such as the European
Central Bank can reasonably claim to enjoy a distinctive institutional compe­
tence to manage the common currency – given the conditions under which
monetary union was introduced. The Bank’s very high level of independence
was meant to compensate for the serious shortcomings of macroeconomic gov­
ernance at European level. For example, some economists have argued that the
socially optimal delegation of monetary policy is not to a completely indepen­
dent central bank. Rather, in order to maximize social welfare, governments
should have the option of overriding the central bank’s decisions under some
particular circumstances. Thus, the ‘optimal’ central bank should follow a
non-linear decision rule: in the case of small output shocks it determines the
acceptable inflation level independently, while in the case of large output dis­
turbances it follows the government’s preferences. At European level,
however, there is no economic government, no generally accepted political
counterweight to the central bank; hence it is not clear how appropriate
procedures for overriding ECB decisions could be designed and enforced.
Incidentally, to acknowledge the ECB’s distinctive institutional competence
is not to overlook the risks entailed by the political decision to move to monetary
union. Given this political decision, and the peculiar governance structure of the
EU, however, something like the extremely independent ECB was arguably the
best arrangement that could be devised – a second-best which takes into
account the governance deficit as well as the artificial quality of the monetary
union: the EU is far from being an optimal currency area, in the sense of
620 Journal of European Public Policy
being able to make easy domestic adjustments to external shocks. Precisely
because of the difficulty of reaching agreement on matters of economic govern­
ance, the Maastricht Treaty, although extremely detailed on procedural matters,
left a number of fundamental policy questions unanswered. As the vicissitudes
of the Stability and Growth Pact suggest, it may have been a mistake to assign
price stability as the sole objective of the central bank, but this was the decision
of the member states, not of the ECBs. From the point of view of accountability,
at least, this explicit, treaty-based commitment has the advantage that the per­
formance of the Bank can be measured unambiguously against the rod provided
by the regular statistical measurements of inflation in the Eurozone. In this
sense, at least, the ECB is an accountable institution, which is more than can
be said about the European Commission.
It is, in fact, much more difficult to identify the distinctive institutional com­
petence of the Commission. Most EU policies are regulatory in nature, and in
this respect the Commission may be considered a sort of super-agency.
However, it has been assigned a variety of other functions: executive, legislative
and quasi-judicial. This multiplicity of functions and objectives expands the
scope of the Commission’s discretionary choices, thus greatly complicating
the task of evaluating the overall quality of the institution’s performance. As
was already noted, the collegial nature of the institution, and the multiplicity
of different tasks assigned to the Commission, further complicate matters. As
a result, both political accountability and accountability by results are reduced
to vanishing point. Indeed, accountability, as well as legitimacy in the sense
of distinctive institutional competence, are only possible where objectives are
clear and limited. This is probably the reason why the issue of insufficient demo­
cratic legitimacy was never raised before the European Single Act of 1986 greatly
increased Community competences.

6. EUROPA: GEHT ES NICHT EINE NUMMER KLEINER?

‘Europe: couldn’t it be one size smaller?’ The title of an article which appeared in
the influential German weekly Die Zeit a couple of years ago expresses well the
feelings of many Europeans about the apparently unstoppable growth of EU
competences. Just as people no longer see any natural limit to the geographical
expansion of the Union – regardless of geopolitical and other risks – so they
realize that the distant goal of ‘ever closer union among the peoples of Europe’
may be used to justify almost any expansion of the powers of Brussels, even in
areas unrelated to economic integration. Hence the impression that there is no
effective barrier to the continuous, if incremental, expansion of EU compe­
tences. In 1990 a distinguished jurist could write that ‘there is no nucleus of
sovereignty that the Member States can invoke, as such, against the Commu­
nity’ (Lenaerts 1990). When these words were written, few people seemed to
object. The results of the recent referenda show how much the public mood
has changed since then.
 G. Majone: The common sense of European integration 621
According to some scholars such worries are unjustified because the compe­
tences of the EU remain quite limited by comparison with the powers of the
national governments. This is a misleading argument. The appropriate standard
of comparison is not the omnicompetence of the national welfare state, but the
level of the resources available to the EU. When normative, administrative and
financial resource limitations are taken into account, it is easy to see that in very
many areas the Union is doing too much rather than too little. And because it
tries to do too much, pursuing several objectives simultaneously, it too often
produces suboptimal outcomes. In Dilemmas of European Integration (Majone
2005: 111– 14) I discuss the instructive example of the conservation part of
the Common Fisheries Policy (CFP). Over its more than twenty-year history,
this policy has largely failed in its aim of conserving fishery resources, despite
its seeming institutional advantages over other international fisheries regimes.
The reason, according to Symes and Crean (1995), is that the underlying prin­
ciples of the CFP ‘have more to do with reinforcing the concept of European
unity than with the effective management of a seriously depleted resource’.
Because under the CFP all EU fishermen can fish in the offshore waters of
any member state, the international-commons problem (which the UN Law
of the Seas attempted to solve with the introduction of the 200-mile Exclusive
Economic Zone) has been perpetuated. In turn, the perception that equal access
leads to overfishing has led to the loss of the CFP’s legitimacy, especially among
British and Irish fishermen. They are inclined to see the CFP as a policy aimed at
redistribution rather than conservation, and fear that their compliance with
quotas will simply result in Spanish and other foreign fishermen getting the
fish. The situation would be different if the EU had the means to monitor
implementation of European rules and quotas. The Brussels authorities have
no such means, but at the same time they make it impossible for the British
Navy, say, to prevent overfishing in British waters: from the point of view of
conservation, a no-win situation. In spite of such an obvious policy failure,
the drafters of the Constitutional Treaty included the conservation part of the
CFP among the few exclusive competences of the EU. Many more similar
examples could be mentioned – some are discussed in Dilemmas of European
Integration – to support the widespread impression that the EU is constantly
trying to bite off more than it can chew. The inability to produce satisfactory
results undermines the legitimacy of European policies and explains, in part,
the revolt of French and Dutch voters.
In sum, the real problem – in the sense of a condition about which some­
thing could conceivably be done under present circumstances – is the ‘account­
ability deficit’, not an alleged ‘democratic deficit’ of the EU: accountability,
rather than an unattainable democratic legitimacy, should be the main
concern. I have already argued that democracy and integration could be made
compatible, rather than conflicting, values, only if a majority of Europeans
supported a federalist outcome of the integration process. Even then it is
to be feared that after the first moments of enthusiasm, a European federal
state – a United States of Europe satisfying all the criteria of parliamentary
622 Journal of European Public Policy
democracy – would be unable to attract and retain the loyalty of its citizens.
This is because, absent a European demos – in the sense of ‘a collectivity in
which the identification of members with the group is sufficiently strong to
override the divisive interests of subgroups in cases of conflict’ (Scharpf 2001:
364) – the federal government could not legitimately provide many of the
public goods (including large-scale income redistribution) which Europeans
have come to expect from their government. Unable to compete with the
national welfare states, the federation would always be considered a virtual,
rather than a real, government, so that the nation state would continue to be
the real arena of democratic politics (Majone 2005). Paradoxically, only a with­
ering away of the European welfare state, perhaps as a consequence of globaliza­
tion, could facilitate the popular acceptance of a European federal state by
drastically reducing the difference between what can legitimately be done at
the national and at the European level.

7. POSITIVE AND NEGATIVE INTEGRATION

The cryptofederalist bias of many students of European integration is particularly
evident in the asymmetric treatment of positive and negative integration. The
Rome Treaty did not attach any normative connotation to the distinction
between these two approaches. The Common Market was to be achieved by
both methods, but in fact by greater reliance on negative law – witness the sig­
nificance of the articles dealing with: the elimination of customs duties (Articles
12-17), and of barriers to intra-Community trade (Articles 30-7); free movement
of persons, services and capital (Articles 48-73); distortion of competition
(Articles 85-94); and, not least, with equal pay for male and female workers
(Article 119). Indeed, repeated policy failures suggest that the institutional
system designed by the Treaty is more suitable to the promotion of negative inte­
gration, which limits the discretionary powers of the member states, than to the
development and implementation of effective supranational policies. Against all
evidence, however, many scholars continue to associate positive integration
with positive values such as social protection and the correction of market failures,
negative integration with deregulation and the narrow interests of traders. In fact,
economic and other special interests often find it convenient to support measures
of positive integration, while fundamental rights and the diffuse interests of con­
sumers are generally better protected by measures of negative integration.
The evidence presented in Chapter 6 of Dilemmas of European Integration
supports the view that EU policies tend to be the by-product of processes
largely unrelated to the stated policy objectives. Even though its specialized
services may be sincerely devoted to problem-solving, the Commission as a col­
legial body will never propose a policy which does not contribute, directly or
indirectly, to an expansion of EU, and its own, powers – the ‘common interest’
as perceived in Brussels. Likewise, neither the member states nor the European
Parliament will support a proposed solution, however effective, which is
unfavourable to their national or institutional interests. Since the organizing
 G. Majone: The common sense of European integration 623
principle of the Community is the representation and balancing of interests
(Majone 2005: ch. 3), each European institution sees itself primarily as the
bearer of a particular interest, national or supranational, which it strives to
protect and promote. Hence, the main theme of political conflict in the EU
is less the formulation and implementation of policy than the defence of the
respective prerogatives of Council, Parliament and Commission – hence the
importance attached to the question of the legal basis of any proposed measure.
It is true that also at the national level public policy is often made less to solve
concrete problems than to serve party-political or other special interests. Most
political scientists assume that the main goal of elected politicians is to maximize
the probability of being re-elected. Hence, politicians tend to care little for
actual policy results, and in any case prefer to support distributive rather than
efficient policies. Favouring special interest groups or geographical districts
appears to be a more promising strategy for achieving their re-election objective
than pursuing the public interest, however defined. However, recent research
has shown that citizens’ preferences can constrain legislators’ actions, provided
that issues are framed in a way which allows citizens to reward or punish their
representatives for specific policy decisions. If legislators are forced to take public
positions on specific programmes, voters can hold their legislators accountable
for the positions they take and for the effects they produce. Too wide a gap
between stated objectives and actual results invites punishment at the polls
(Arnold 1990). No equivalent controls exist in the EU, where sub-optimal pol­
icies can persist, unscrutinized and unchallenged, for decades – as was already
mentioned, Article I-12 of the Constitutional Treaty listed the conservation part
of the Common Fisheries Policy (a clear case of policy failure if there ever was
one) among the exclusive Union competences, together with monetary policy
the common commercial policy and customs union!
Many, perhaps most, measures of positive integration in the areas of health,
safety and environmental regulation, are justified by the argument that without
EU-level harmonization member states would engage in a socially undesirable
‘race to the bottom’. Indeed, not only in the EU but also in most federal
states, a standard argument for centralizing social regulation is that centraliza­
tion prevents member states from competing for industry by offering environ­
mental or other standards that are too lax relative to the preferences of their
citizens. Such competition is said to lower the level of social protection that
states would pursue if they did not face international or inter-jurisdictional com­
petition to attract industry and the additional wages and taxes produced by
industry migration. It is assumed that, other things being equal, firms will try
to reduce costs by moving to the jurisdiction that imposes the least stringent
regulatory requirements. Superficially, such a race to the bottom appears to
be an example of the prisoners’ dilemma. If the two states could enter into a
co-operative agreement to adopt the optimally stringent standard they could
maximize social welfare without engaging in ‘unfair’ competition for industry.
This presupposes, of course, that the agreement is enforceable and that prefer­
ences for environmental quality are exactly the same in the two jurisdictions.
624 Journal of European Public Policy
As long as the jurisdictions are independent states, any co-operative agreement
would lack credibility, but the situation is different if they are part of a federal or
quasi-federal system. In such a case the sub-optimal outcome could be avoided if
the federal standards were harmonized at the higher level, provided that the
harmonized standards were equal to the standards the two states would find
optimal if they were still independent. The proviso about the harmonized stan­
dards corresponding to the actual preferences of the member states is crucial: if
states have different preferences for environmental protection then standards
that maximize social welfare will be different. Hence, a uniform European stan­
dard, even one that sets only a minimal level of pollution control and allows the
member states to adopt more stringent national standards, will not be optimal,
unless the minimum standard is so low that it is exceeded by all the national
standards. It is thus quite possible that even if there is a race to the bottom
(an assumption which in fact lacks empirical support), a European standard
might still reduce aggregate social welfare.
Morover, as Revesz (1992) has pointed out, race-to-the-bottom arguments
are incomplete because they fail to consider that there are more direct means
of attracting foreign direct investments than lowering social standards. The
advocates of harmonization assume implicity that states compete over only
one variable, such as environmental quality. Given the assumption of a ‘race’,
however, it is more reasonable to suppose that if harmonization prevents com­
petition on the environmental dimension, states would try to compete over
other variables, such as worker safety, minimum wages or taxation of corporate
profits. To avoid these alternative races, the central regulators would have to har­
monize national rules so as to eliminate the possibility of any form of interstate
competition altogether. This would amount to eliminating any trace of national
autonomy, so that the race-to-the-bottom argument is, in the end, an argument
against subsidiarity.
Naturally, the fear of a ‘race’ is not the only rationale for harmonization of
social standards. A more plausible argument for EU-wide harmonization
of such standards is the need to dismantle non-tariff barriers to trade within
the single market. Even in this respect, however, ex-ante harmonization may
have been pushed too far. Today it is recognized that an initial difference in
health, safety or environmental standards need not distort international trade;
rather, it is trade itself that leads to their eventual convergence. The reason is
that such standards are positively correlated with the standard of living.
Hence, as wealth grows as a result of free trade, the endogenous demand for
higher social standards grows as well. It is worthwhile keeping in mind that
the Treaty of Rome rejected the view, fairly common even then, that differences
in social conditions between the member states can represent a form of ‘unfair’
competition, so that harmonization is needed in order to prevent ‘social
dumping’. Apart from the condition of equal pay for male and female
workers, introduced under French pressure, the Treaty never prescribes that
social policies be harmonized prior to, or even concurrently with, trade liberal­
ization within the common market. In line with the conclusions of the 1956
 G. Majone: The common sense of European integration 625
Spaak Report, the Treaty maintains that harmonization should in general be
regarded as a corollary of, rather than a requirement for, market integration.
This sequence – free trade first, followed by a more or less spontaneous har­
monization of social standards – was generally accepted by European policy­
makers until the mid-1970s. After a decade of hesitation, a clear shift in the
direction of social harmonization occurred in the mid-1980s. Several factors
contributed to this shift: the enlargement of the Community, creating wide
differences in labour costs between member states; high unemployment and
stagnating real wages; and, not least, the social activism of the Delors Commis­
sion (Sapir 1996). Today the wisdom of ex-ante, top-down harmonization –
positive integration – is again being questioned.
If the necessity of positive integration has too often been overrated, the role of
negative integration in legitimating the process of European integration has not
been sufficiently appreciated. Negative integration is not only about removing
national restrictions to trade and to the free movement of the factors of pro­
duction. It is also about limiting monopoly power and market dominance, pro­
tecting the diffuse interests of consumers, and about fighting discrimination on
grounds of gender, nationality, age and other factors. The best-known example
of negative law in the area of individual rights is Article 119 of the Treaty of
Rome (now Article 141 EC), which requires application of the principle of
equal pay for male and female workers, for equal work or work of equal
value. The article itself conferred no positive regulatory power until its amend­
ment by the Treaty of Amsterdam. The new paragraph (3) inserted by that
Treaty extends the scope of the article to positive measures ensuring equality
of opportunity and is thus not restricted to measures simply outlawing discrimi­
nation. So far, however, the most dramatic results have been achieved by Article
119 in its original, ‘negative’ formulation. In sum, the economic constitution of
Europe finds expression in the rules of negative integration, while attempts to
pursue positive integration with inadequate material and normative resources
too often undermine the legitimacy of the constitutional order.

Biographical note: Giandomenico Majone is Professor of Public Policy, Emer­

itus, European University Institute.

Address for correspondence: Giandomenico Majone, Department of Public

Policy, European University Institute, Via Lappeggi, 26, 50015 Grassina, FI,
Italy. email: giand.majone@tin.it

REFERENCES
Arnold, R.D. (1990) The Logic of Congressional Action, New Haven, CT: Yale Univer­
sity Press.
Choper, J.H. (1980) Judicial Review and the National Political Process, Chicago,
IL: University of Chicago Press.
Commission of the European Communities (2001) European Governance,
Luxembourg: Office for Official Publications of the European Communities.
626 Journal of European Public Policy
Gay, P. (1962) The Dilemma of Democratic Socialism, New York: Collier Books.
Klabber, J. (1998) ‘The Netherlands’, in D. Dinan (ed.), Encyclopedia of the European
Union, Boulder, CO: Lynne Rienner, pp. 355 –7.
Kohnstamm, M. (1989) ‘Jean Monnet face à l’Union Européenne’, in G. Majone,
E. Noel and P. Van den Bossche (eds), Jean Monnet et l’Europe d’Aujourd’hui,
Baden-Baden: Nomos, pp. 39 – 44.
Lenaerts, K. (1990) ‘Constitutionalism and the many faces of federalism’, American
Journal of Comparative Law 38: 205– 63.
Majone, G. (1996) Regulating Europe, London: Routledge.
Majone, G. (2005) Dilemmas of European Integration: The Ambiguities and Pitfalls of
Integration by Stealth, Oxford: Oxford University Press.
Milward, A.S. (1992) The European Rescue of the Nation State, London: Routledge.
Moravcsik, A. (1998) The Choice for Europe, Ithaca, NY: Cornell University Press.
Morgan, E.S. (1988) Inventing the People: The Rise of Popular Sovereignty in England and
America, New York: W. W. Norton.
Mosse, G.L. (1975) The Nationalization of the Masses, New York: New American
Library.
Rakove, J.N. (1997) Original Meanings, New York: Vintage Books.
Revesz, R.L. (1992) ‘Rehabilitating interstate competition: rethinking the “race-to-the­
bottom” rationale for federal environmental regulation’, New York University Law
Review 67: 1210– 54.
Ryle, G. (1949) The Concept of Mind, New York: Barnes & Noble.
Sapir, A. (1996) ‘Trade liberalization and the harmonization of social policies: lessons
from European integration’, in J.N. Bhagwati and R.E. Hudec (eds), Fair Trade and
Harmonization, Vol. 1, Cambridge, MA: The MIT Press, pp. 543– 70.
Scharpf, F.W. (2001) ‘Democratic legitimacy under conditions of regulatory compe­
tition: why Europe differs from the United States’, in K. Nicolaidis and R. Howse
(eds), The Federal Vision: Legitimacy and Levels of Governance in the United States
and the European Union, Oxford: Oxford University Press, pp. 355– 74.
Scott, J. and Trubek, D.M. (2002) ‘Mind the gap: law and new approaches to govern­
ance in the European Union’, European Law Journal 8(1): 1– 18.
Symes, D. and Crean, K. (1995) ‘Historic prejudice and invisible boundaries: dilemmas
for the development of the Common Fisheries Policy’, in G. Blake, W. Hildesley,
R. Pratt, R. Ridley and C. Schofield (eds), The Peaceful Management of Transbound­
ary Resources, London: Graham and Trotman.
Weiler, J.H.H. (1999) The Constitution of Europe, Cambridge: Cambridge University
Press.

Final version accepted for publication 16/03/06