Working Paper n.º 8
The Europeanization of Ratification Procedures: Towards a EU-Wide Constitutional Convention
01 | 09 | 2005
Carlos Closa, Centre for Political and Constitutional Studies, Madrid

Abstract

The word “Europeanisation” has acquired so many meanings that it is difficult to disagree with Olsen that it is not very useful in theoretical terms (Olsen; 2001). In this context, it results unavoidable a previous clarification of the meaning in which it is used here, even though the object of this paper is not a theoretical inquiry on the concept, but rather an assessment on whether the process has happened.

Europeanization means the impact of the EU on the domestic structures or, in other words, the political system. Since the object of this inquiry refers to the ratification of the EU constitution, the inquiry can be framed along the following question: what is the influence of the EU norms on domestic processes of ratification? The response is immediate. EU ratification rules do not seem prima facie to impose constraints on national ones and, hence, impact could not be easily identified. If Europeanization is defined as the transformation brought about by the impact of EU level on the national one, then, it has not happened. However, a different model of “europeanisation” has happened: member states show signs of a trend towards converging in certain models of ratification. Naturally, this does not mean that all they use the same procedures and in the same form. It means, rather, that they watch closely each other and they adopt and adapt their own ratification procedures according to the learning of what other member states have done.

The paper reviews this argument in the following sections. Firstly, it examines briefly the changes that the draft constitution introduced and compares it with the American model of ratification. It then examines the three main procedures for ratification to asses whether some convergence can be appreciated. This paper discusses the three procedures used for ratification (constitutional reform, parliamentary ratification and referendums) as well as the actors involved (constitutional courts and political parties) with the aim of mapping out the eventual emergence of a “constitutional convention” on the ratification of the Constitution. The underlying inquiry seeks to establish whether ratification of the EU Constitution produces an Europeanization of procedures. It finally ponders the extent to which we can refer to “Europeanisation” of ratification procedures understood as an increased convergence among Member States.

1. EU rules for ratification and their comparison with the US constitution

In contrast with the vivid debates of the American constitution making exercise, the Convention on the Future of Europe discussed ratification very scarcely. Political realism made even the most ardent federalist to accept that ratification would proceed along the traditional international law mechanisms. Article IV-447 uses the typical normalised language of international law treaties and it is, following de Witte’s opinion, the clearest formal confirmation that the constitutional treaty is, from the point of view of its drafters, a truly international treaty (de Witte; 2005: 194).

The constitution maintains the ratification acquis consolidated in former treaties and adopts the same procedures and requirements: as for the threshold requirement, Constitution drafters did not modify the former requirement of unanimity. This bounds together the fate of all and every Member State: an eventual failure to ratify in one of them will spill over automatically in all the EU. However, it is not this common fate what national provisions take as the criteria for ratification but rather, specific national circumstances. National definitions and understandings of membership both in constitutional and political terms remains the basic backbone for national ratification and, in this form, ratification responds to the coincidence of different (and even antagonistic) national projects.

The Constitutional procedure follows the path of earlier treaties: ratification will proceed in accordance with national constitutional requirements (art. IV-447). This has been interpreted as a mild constraint or a “constitutionally sensible” international law mechanism, meaning the need of parliamentary approval in all signatory estates (de Witte; 2005). Even though parliamentary ratification is common to all Member States, national constitutions pose a vast group of additional procedures for ratification (constitutional reform, previous intervention of courts, parliamentary ratification with or without reinforced majorities, referendums –either consultative or binding).

This limited constraint of EU rules compares with the strength of the rules for ratification included in the US Constitution that directed towards strong “Americanization or “nationalization” of States and states constitutions. This happened by means of two rules; being the first the number of ratifications from contracting parties required for the Constitution to be valid. The US Constitution (article VII) established that the ratification of the Conventions of nine states shall be sufficient for the establishment of this constitution between the states so ratifying the same. The effects of this threshold shorter than unanimity are known: The nine state rule (…) gave each state an incentive to consider its long run as well as its short-term interests (Beer, 1993: 332). Despite overwhelming opposition against, the New York convention ratified the Constitution when other 9 states had effectively done so.

This threshold does not only provide a frame for the rational calculation of the options of a single state. It has also theoretical significance. In words of Madison, the unanimous assent of the several parties that are part to it [transforms] the act establishing the constitution into an act of the people, as forming so many independent states, not as forming an aggregate nation.[1] In contrary sense, Calhoun argued that ratification was the act of the several states acting in their separate capability and, consequently, final unanimity did not transform the basic fact that the act of ratification (…) established it as the constitution between the states ratifying it, and only between them (Calhoun; 1849: 92 and 87).

The second issue of the American experience with paradigmatic value for the EU refers to the procedures for ratifying the Constitution. In the USA case, the Philadelphia Convention hotly debated whether state conventions or legislative assemblies should be the organs in charge of ratification. The option for either of these has strong implications on the identification of the locus of sovereignty and the construction of a constituent power: ratification through state parliaments implies recognising the subjection of the Constitution to normal state legislatures. The contrary option implies creating an independent and unbounded source of constitutional consent. Again, Calhoun argued that the states retained after ratification of the constitution, the distinct, independent, and sovereign character in which they formed and ratified it, is certain, unless they divested themselves of it by the act of ratification, or by some provision of the constitution (Calhoun; 1849: 87).

These decisions taken by the US constitutional Convention defined the nature of the new document and conditioned its fate during the subsequent ratification stage. They, moreover, influenced deeply the nature of the new polity. Thus, Calhoun reduced the issue of whether the government is federal or national to a simple question: whether the act of ratification, of itself, or the constitution, by some one, or all of its provisions, did or did not, divest the several states of their character of separate, independent and sovereign communities, and merge them all in one great community or nation, called the American people (Calhoun; 1849: 88).

As it has been argued, EU norms for ratification impose scarce pressures on national constitutions. They just establish a timetable constraint and the referent to national constitutional requirements (that, at best, for some, imply an obligation to observe and adhere to parliamentary ratification). There is not great pressure for change nor need for institutional adaptation for the top down level. Europeanization, thus, has not been imposed from the top constitutional norms

However, Europeanization has happened in a different way, as an increasing convergence of procedures. Europeanization may result from the bottom-up convergence of national ratification models. Even lacking a common framework based on the EU constitution and the application of 25 different procedures, the current process may reveal some traits of an emerging constitutional convention on ratification of the EU constitution whose importance will be only evident ad futurum. If there is a degree of convergence on procedures, then the emergence of a constitutional convention that goes further than treaty rules can be assessed. This constitutional convention may become essential for determining the nature of the polity in the long run. The following three sections review the procedures for ratification of the constitution in the 25 Member States.

2. The EU constitution and national constitutions’ reforms: reasons and actors

The EU Constitution maintains the ratification procedures of former treaties and, accordingly, its legal validity does not derive from other sources than national constitutional requirements. National constitutions remain still the source of legal validity of the EU constitution being constitutional clauses enabling membership the essential provisions. These clauses reflect a wide variety of circumstances even though a basic typology would distinguish between general clauses (these designed for incorporation to international organisations in general) and specific EU clauses. An additional distinction within the first group would be between the absence of any reference whatever to the EU or the introduction of reforms to integrate specific aspects of the EU and EU policies.[2] With great cautions, it could be argued a progressive mode towards the “Europeanization” of national constitutions since only six member states do still have not adapted at all their respective texts to the requirements of membership and/or created a specific basis to it. What remains, though, is wide heterogeneity in the way national constitutions grant validity to the EU law.

Typology of Constitutional clauses for EU membership |Ver PDF|

The EU Constitution has not modified greatly the former panorama. A priori, it might be assumed that a document bearing the name of “constitution” may raise questions on its articulation with the national ones. Constitutional reform could be expected on different grounds. This may happen, firstly, because there were incompatibilities appreciated. Thus, the current round of ratification has put on the agenda a large number of issues for constitutional interpretation in very different domestic constitutional contexts. Concerns reach many of the new provisions but among these, the Charter of Fundamental Rights and the primacy of EU law (art. I-6) are the most important ones. Else, the wish and/or necessity of grounding in more solid constitutional basis what in symbolic terms may be appreciated as a substantive leap forward may also be the driving force behind reform of national constitutions. In this line, Cartabia argues in favor of ratification through constitutional reform or, at least, qualified majorities. In her opinion, ratifying through law implies accepting that national constitutional values could be modeled by decisions taken by concurrent governmental majorities in EU institutions (Cartabia, 2005: 284). The following two subsections examine the perceptions of national governments and national constitutional courts and other advisory bodies on this issue.

2.1 National governments perceptions

Basically, the prevailing (even though not only one) interpretation considers that the EU constitution does not mean a substantial change in the conditions of membership, a view largely shared by governments, constitutional courts and other advisory bodies. Among governments, the view of the Spanish Foreign Affairs Minister was that the Treaty fit perfectly within the current constitutional provisions. In Finland, the interpretation that the Constitution does not mean any significant transfer of powers to the Union prevailed backed by a strong pro-constitutionalist consensus: nearly all of the major parties represented in the Riksdagen, the Finnish Parliament accept the European Constitution. The Swedish Prime Minister argued that the parliament is the right body to settle the issue, especially since the constitution does not (according to the Prime Minister) change Sweden’s relationship to the EU. For the British government, the draft constitution raised not particularly difficult constitutional issues and the changes do not alter the fundamental relationship with Member States.

The prevalent view among Eastern European member states is that the EU constitution is more of an international treaty than a Constitution and hence, there is no need of constitutional reform and/or referenda since the Constitution did not imply a change in the conditions for membership established in national constitutions and ratified through referendum.. Most of the Members acceding in 2004 share this view. In Lithuania, the prevalent interpretation was that the accession referendum and the Constitution already covered the changes introduced by the EU Constitution. The forthcoming EU constitution was anticipated in Latvia and Estonia, that introduced protective clauses in their respective constitutions.[3] In Latvia, any substantive change in the conditions of membership had to be introduced by means of a referendum (art. 68). And in Estonia, the constitutional duty to preserve independence was interpreted as meaning a prohibition to participate in a federal entity (Albi, 2005: 415). The official rationale behind the position emphasized the fact that by the time the Estonian accession referendum was held (September 14, 2003), the end result of the Convention and the prospect of an IGC were already known and voters could take this into account when casting votes on accession. The Slovak government insisted that the Slovak people already agreed to the country's entry to the EU in the 2003 referendum. The opposition argued that the EU Constitution is "such an important document that we should give citizens the chance to express their opinions about it".

Thus, the prevalent interpretation among these member States was the referendums on accession and the constitutional reforms implemented for membership covered already the EU Constitution which, implicitly, was not considered a significant transformation of the EU. There is one exception to the prevailing interpretation among governments. In Denmark, the Ministry of Justice issued a Declaration that argued that the EU Constitution would compromise the country's sovereignty in privacy protection laws, diplomatic immunity, the transfer of capital, intellectual property rights, health issues and free movement of citizens. In particular, the "flexibility clause," providing for the expansion of EU's jurisdiction without parliamentary ratification or approval by referendum, conflicts with the Danish Constitution. Since the EU Constitution would limit Danish sovereignty, five-sixths of Denmark's Parliament would have to vote for it or a referendum would need to be won for the EU Constitution to be approved.

2.2 The view from advisory bodies and constitutional courts

In the cases in which a previous judgment on the constitutionality of the EU constitution is required, this process offers a window of opportunity for certain actors with the capability to shape the debate and define the sense and path of European integration: the Constitutional and/or Supreme Courts and/or other advisory bodies. Advisory bodies (Councils of State) coincided also in their interpretation was akin. In the Netherlands, the Council of State (Raad van Sate) asked by the government on the expected implications of the European Constitution for the Dutch legal order argued that the draft constitution is only a codification and continuation of EU legal order with no conflict with Dutch Constitution. Curiously, the distinctiveness of the EU constitution was argued as the main reason for convening a referendum. In its general comment, the Dutch government explains that special powers are granted regarding criminal law and external policies as well as the inclusion of the Charter of Fundamental Rights. It is not just another Treaty amendment, and therefore more involvement of citizens is necessary. The government wanted to strengthen and improve the legitimacy of decision-making through the consultative referendum. The Dutch Council of State agreed that the consultative referendum was a good form to consult the citizen.

In Spain, the law required the government to request a preliminary ruling (Dictamen) from the Council of State (an advisory body) for every Treaty that needs the authorisation of the Cortes. The Council declared that the EU Constitution is a “supranational integration treaty” and it argued that there was not contradiction as regards the new system of competences of the Union and the Charter of Fundamental Rights and the Spanish constitutional order of values, rights and freedoms (even though punctual conflicts might appear). However, the Council argued that an eventual conflict may exist between the clause of primacy of EU law (art. I-6) and the Spanish clause of constitutional supremacy (article 9.1).[4] Because of this, the Council opined that the Constitutional Court should declare whether there exist contradiction between the Spanish constitution and the EU constitution. In this way, the Council bypassed government’s earlier intentions and it empowered indirectly the Constitutional Court. Spanish Constitution (art. 95.2) establishes that the government or either chamber of the Parliament may ask to the Constitutional Court to declare whether a Treaty is coherent with Spanish Constitution (although the government is not obliged to so automatically).

The Spanish Council of State argued that if there exists antinomy and/or contradiction between both, then constitutional reform is required. The Council suggested immediately the possible reforms: the introduction of a “clause of integration” that allows a general opening of the Spanish legal order to EU law in such a way that the consistency of EU law with the Spanish constitution is presumed a priori. Further, the Council suggested the idea to “Europeanise” the Spanish constitution by means of an explicit reference to the EU.

Finally, constitutional courts coincided in the same evaluations in their preliminary rulings. The French Court ruled[5] that in two of the cases examined (primacy of EU law and Charter) there was not need for a previous reform. However, certain dispositions on EU functioning and policies, as well as the new powers recognised to national parliaments required a previous constitutional reform. The French constitutional Court designed a full road map for constitutional reform that was successfully completed on 28 February 2005, when the French Congress (National Assembly and Senate seating in joint session) approved the constitutional reform by 730 votes in favour, 66 against and 36 abstentions.[6]

In Spain, a preliminary ruling was not compulsory but the opinion of the Council of State and pressure from opposition parties forced the government to seek it. The consult was framed along four questions:

1.                                                   The existence or not of contradiction between the Spanish Constitution and article I.6 of the Treaty

2.                                                   The existence or not of contradiction between the SC and articles II-III and II-112 of the Treaty

3.                                                   The sufficiency or not of article 93 of the SC for the ratification of the Treaty

4.                                                   If constitutional reform was required for ratification, the procedure to be followed.

The Declaration of the Constitutional Court gave a negative response to questions 1 and 2, a positive one to question 3 and it declared improcedent a response to question 4. Largely shared interpretations coincide in arguing that the Constitutional Court has tried to avoid the huge legal problem that different solutions could have created on a issue that is not politically divisive in Spain (vis-à-vis questions such as reform of statutes of Autonomy for regions, etc.)

The last case of intervention of Constitutional Courts is the Czech Republic, in which a preliminary constitutional ruling is not compulsory but the Constitutional Court Act envisages the possibility of a review of compatibility of international treaties (under which regime the EU Constitution still falls) with the Czech constitution. After some initial hesitation, Czech President Vaclav Klaus asked in February 2005 whether the European Constitution is in line with the Czech Constitution and whether changes need to be made to the country's constitution in order to pave the way for the new European treaty.

Taken together, there is a certain move towards sanctioning EU membership by means of an explicit constitutional clause (what certain authors have called Europeanization of national constitutions). It seems also that national constitutional courts have moved towards a tacit acceptance of systemic compatibility of the two legal orders. This grounds the claims for a kind of “multilevel constitutionalism” or European “block of constitutionality”.

3. Parliamentary Ratification

The political nature of the EU depends (among other things), on the role of parliaments in ratification. The US constitutional debate shows a fierce dispute on the nature of the federation depending on this issue. Two (theoretical and practical) models contra posed each other: ratification through a vote of parliamentary assemblies or ratification by means of state conventions. Disputes on the models and their implications appeared when the Committee of the Whole appointed to consider the state of the American Union discussed the 19^th resolution requiring the ratification of the Constitution by state conventions. Ellsworth, tabled a proposal to refer the plan to state legislatures. Against this proposal, Mason argued that legislatures are mere creatures of constitutions and cannot be greater than its creator. Succeeding legislatures would have equal power to repel the Constitution as the current one to adopt it. This view was supported among others by Randolph, Gorham and Madison and opposed by Gerry and Ellsworth. Madison argued that legislatures were incompetent: it would be a novel and dangerous doctrine that a legislature could change the constitution under which it held its existence. He considered the differences between a system founded on the legislatures only and one founded on the people, to be the true difference between a league or treaty and a constitution (Elliot; 1907: 356). Madison refined his arguments in Federalist no. 43: should state legislatures act as the ratifying bodies, this would place the Constitution on a similar standing to any other laws approved by these bodies. Hence, a similar instrument could override it or its provisions (Federalist 43). Ellsworth’s motion to transfer the plan to state legislatures’ was rejected (3 yes and 7 no) and the 19^th Resolution was approved by 9 yes and 1 nay. Hence, ratification happened through state Conventions.

But even after ratification was completed, the issue was not settled and disputes persisted on the interpretation of the ratification procedures. For Calhoun, the leading theorist of nullification and the supremacy of states over the federation, the process preparatory to ratification, and the acts by which it was done, prove beyond the possibility of a doubt, that it was ratified by the several states, through conventions of delegates, chosen in each state by the people thereof; and acting, each in the name and by the authority of its state; and, as all the states ratified it –we the people of the United States” – means We the people of the several states of the Union (acting as free, independent and sovereign states) (Calhoun; 1849: 93). In other words, Calhoun interprets that the fact of avoiding state legislatures as the ratifying organs was insufficient to create a new and independent source of validity for the US constitution that was still subject to the States: the authority which ordains and establishes is higher than that which is ordained and established (Calhoun; 1849: 94).

Of course, no similar debate on the role of parliaments occurred on the EU constitution: all Member States treat the EU Constitution as an international treaty similarly to former EU Treaties. And in almost all of them, the procedure by default for ratification is by means of a vote in parliament. But this is by no means, a classical feature and, in fact, the involvement of Parliament in ratification processes became a regular constitutional feature during the 20th century (de Witte, 2004a). Classical political theorists had either neglected or rejected the role of parliament. Locke defined the external policy as “federative power” containing, among other things, the capability to carry through all necessary negotiations with alien persons and communities. Differently to the executive power, the “federative one” must be trusted to the prudence and wisdom of those in charge of it (the government) and not regulated beforehand by positive law. Locke´s anthropological cum organicist vision of the body politic explains his opinion: being the state a single person and the international society under the state of nature, the norm to be followed when foreigners are concerned depends very much on the form of behaving of these and the changes in their purposes and interests.[7]

Rousseau coincided also from different anthropological views and with a more democratic bias. L’exercice extérieur de la puissance ne convient pas au Peuple, les grandes maximes d’Etat ne son pas à as portée; il doit s’en rapporter là-dessus à ses chefs qui, toujours plus éclairs qui lui sur ce point, n’ont guère intérêt à faire au-dehors des traits désavantageux à la patrie.[8]

These restrictive views had progressively been modified into an almost universal acceptance of the role of parliaments in ratification. Even in these countries in which ratification may be considered to be the sanction of an international treaty delegated to the executive, ad hoc rules have modified this perception. Thus, in the UK, ratification of an international treaty requires merely an executive act of on the part of the Foreign Secretary, acting on behalf of the Crown in the exercise of the Royal Prerogative. However, the so-called Ponsonby Rule since the 1920s has effectively required that a treaty to be subject to ratification to be laid before Parliament for 21 sitting days before ratification, for information and to give Parliament the opportunity to debate such a Treaty. In practice, ratification of treaties such as the Constitutional Treaty requires an Act of Parliament because of the domestic and budgetary effects of such amending treaties.

Moving within this general regime, EU treaties enjoy a reinforced integrity in comparison to other international treaties since national parliaments cannot enter reservations (De Witte; 2004a). Prima facie, this may seem as a foreclosing of any form of “conditional membership”. However, Member States have used protocols and opt-out clauses to negotiate specific application of theacquis.

Table 1: Parliamentary Ratification Procedures in the Member States of the EU |Ver PDF|

Parliamentary ratification is the main (and single) process in most countries and even when a referendum has been convened, parliament ratification is also formally required. Formally, a distinction can be drawn between these cases in which the competent authority for ratification is the parliament itself (authorization) and those in which it is not and only a consultation is formally required. The later category comprises UK and Ireland, being formally facultative in the first and compulsory in Ireland. For all other Member States, parliamentary authorization is required. In these cases, the intervention of Parliaments is constructed as a mechanism of control on the authority with the effective capability to sign the Treaty (normally, the government). In this situation, rejection seems a highly unlikely outcome (and, in fact, the only rejection from a national parliament happened in 1954 with the EDC by the French National Assembly: since most EU Member States are parliamentarian systems, it is supposed that the majority that supports a government will also ratify the Treaty negotiated by the same government. Some authors have precisely argued that in the European parliamentary democracies, a parliament’s law cannot be considered the expression of the general will but an expression of the political direction of the governmental majority (Cartabia; 2005: 275).

Starting from this assumption, parliamentary ratification may diverge from it in three situations. Firstly, ratification may require qualified majorities what implies building broad coalitions involving also opposition parties (Austria if constitutional reform is involved, the Czech Republic, Denmark, Finland if constitutional reform is involved, Germany, Greece, Hungary, Luxemburg, The Netherlands, Slovakia, Slovenia and Sweden). What results striking is that, so far, large majorities have endorsed ratification in parliamentary votes. Secondly, parliamentary ratification may change its function as a mere sanctioning of governments whishes if elections change the parliamentary majority that supported a government in the negotiation of a Treaty (as it may be the case in Poland and/or the Czech Republic). Thirdly, in certain cases, more than one Chamber may intervene and modify the relationship with the government (the approval of the two Chambers is required in Austria, the Czech Republic, Hungary and Italy; whilst in other cases, the Lower Chamber usually suffices. In Belgium, the two Chambers, plus the three Regions (if their competences are affected) plus the two communities must ratify the Constitution, what requires the participation of 7 Chambers). The clearest example of this situation happens in Germany, where the German Bundesrat has the power to veto ratification, what renders it a powerful actor in negotiations. In fact, facing a threat to vote against the Constitution coming from some Länder, German Chancellor Schröder promised the representatives from four regions that their competences at the EU level would be widened and that the upper house would be involved in the choosing of judges for the European Court of Justice.

In summary, parliamentary ratification has so far backed government stances with strong majorities, in all cases well above the threshold required what reveals that parliamentary acceptance of the Treaty is, so far, very high in contrast with the trends in public opinion.

4. Referendums on the Constitution

The referendums convened to ratify the EU constitution are something of a novelty in one respect: never before so many countries held a referendum on the same issue mobilising such an enormous amount of citizens to back a decision and a specific text. Between February 2005 and June 2006, more than 250 million people in 10 countries will be asked whether they accept or not the EU constitution.

From the purely legal point of view and strictly speaking, referendums are necessary in Ireland and semi-compulsory in Denmark (if not majority of 5/6 is obtained in the Folketing). Technically, these two are different cases: in Denmark is a direct technique for ratification of the Treaties whilst in Ireland is an incidental technique deriving from the necessity of reforming the Constitution (Lepka and Terebus; 2003: 79-80). What are the explanations for the wave of referendums for ratifying the Constitution?

Table 1: Referendums on the EU Constitution |Ver PDF|

4.1 Ratification through referendums as result of domestic tactical considerations

One of the explanations is the coincidence of domestic factors. In several of the referendums on the EU constitution, partisan factors (i.e. strategic calculus of the effects of referendums on the domestic electoral struggle) may account for the recourse to this specific institute of direct democracy. Parties calculate the electoral advantage that may derive from using the referendum on a specific party setting. Whether or not their calculations are accurate, politicians seem to base their behaviour on a plain calculus of the hypothetical advantage. In many cases, the referendums reflect not so much an overdue aspiration to explain Europe to the citizen as a manoeuvre to avoid short-term domestic political problems. Several of the cases taken in this paper confirm that the decision for convening of the referendum resulted also from a calculus of its electoral and partisan impact (UK, France, Germany, the Czech Republic, Poland). These decisions are constructed in four situations (Closa; 2004).

1.                  Governments may use referendums as tactical weapon in strengthening its power (Bogdanor; 1994: 31). Governments use non-required votes it in order to strengthen its own position either by attempting to gloss over internal divisions or creating divisions in the opposition. Some of the historical experiences of EU reform-related referendums confirm this tactical use: the 1992 Mitterrand decision of convening a referendum on the Maastricht Treaty was based (apart from other considerations) on the thought that the vote would undermine his political opponents. The 2004 situation mirrored the former one. Chirac decision was perceived as an attempt to undermine potential rivals within the socialist party (in which divergences on the lack of “social” provisions in the new Constitution existed). The bet pay off since the convening spilt over the internal party struggle between the socialist leader François Hollande and the former Prime Minister Laurent Fabius. Both would like to be the socialist presidential candidate in 2007, and Hollande supported the constitution but Fabius opposed it. The issue was settled through an internal referendum clearly won by the yes camp.

The British case shows traits of the same pattern even though other reasons apply. Tony Blair, the British prime minister, announced a referendum at a moment of great political weakness, when he faced accusations of dishonesty over the reasons for his going to war in Iraq. The constitutional referendum was supposed to dispel accusations of arrogance and show he would listen to public opinion. It aimed also to neutralise opposition arguments on the referendum.

2.      Referendums are a mechanism for securing ratification in the absence of parliamentary majorities. This happens either when there is a split between government and opposition on the issue and the governmental majority can not secure the legal threshold required or when the majority itself is divided on the issue. These are the cases of the Czech Republic and Poland. The Czech government did not initially support the idea of holding a referendum (that had a secondary position in the Czech constitution). But the two Czech opposition parties (the Civil Democrats –ODS- and the Communists) opposed the Constitution and the Czech President Vaclav Klaus is an outspoken critic of the Constitution. Facing the risk of an eventual parliamentary defeat, the government yielded to the idea of holding a referendum. Since opinion polls showed large support from public opinion on the Constitution, ratification seemed easier by means of a referendum. To complicate more the situation, the future bill on referendum requires a significant majority in parliament but government and opposition have very different views. The centre-left dominated government wants referendums a common instrument of decision-making, while the main opposition centre-right Civic democrats (ODS) proposed a bill applying only to the vote on the EU Constitution.

In Poland the Sejm rejected a motion in favour of a referendum in September 2003 and the two majority parties (the Democratic Alliance of the Left and the Civic Platform) opposed to it. However, the ability of a weakened prime minister to forge a parliamentary majority in favour of the Constitution raised serious doubts. The parties opposed to it, Justice and Law and the League of Polish Families, campaigned in favour of a referendum. Again, referendum resulted convenient in this situation since public opinion is favourable to the Constitution.

The reverse situation also happens: quite often, opposition parties (even though they may agree with government on ratification) argue in favour of a referendum as a means for weakening governments. In fact, opposition parties called for referendums during the ratification of the Maastricht Treaty in almost all Member States but Italy and the Netherlands. In cases in which referendums have not been convened and there exist a strong underlying parliamentary consensus on the Constitution, they are invoked as tactical instruments for partisan struggle. Thus, in Germany, the government speculated with a referendum (despite early rejection). In Autumn, the leadership of the SDP declared itself in favour of holding a referendum if reforms of the Fundamental Law could be completed before December 2004. Since the agreement of the CDU was necessary for this reform and the CDU opposed holding a referendum, the proposal has to be interpreted as a tactical one that sought to undermine the CDU, also because its sister party CSU supported convening it. In July, the coalition government rejected holding a referendum arguing the unexpected consequences of a popular consult: whilst a positive vote was sure in Parliament, a referendum may trigger unexpected consequences.

Similar tactical considerations (coated with the logic of justification referred to below) appear in calls in Greece by PASOK (despite the traumatic experience of the 1974 referendum on the abolition of monarchy) or the Latvian main opposition party, the Popular Party (20 seats out of 100).

3.            A third situation occurs when a strong majority in parliament favours ratification and it does not coincide with the prevalent trend in public opinion. It means, in all cases, that the parliament is keener on the EU constitution than the citizenry at large. This factor explains the non-convening of a referendum, since in these cases, governments who do not want to risk a rejection have good cases to avoid popular votes. Sweden is perhaps a clear case of a country prima facie committed to the principle of a popular referendum that is not holding one because of these reasons. In Sweden (where a referendum recently rejected adoption of the euro as the national currency against all odds and against the recommendation of the political establishment), the EU enjoys broad cross-party support in parliament, but very questionable support in public opinion. Both the governmental Social Democrats and the right-wing opposition agreed that ratifying the Constitution via the parliament would be enough.

4.      Referendums may also serve to resolve situations in which the parties of the majority are internally divided within themselves along anti-integrationist/pro-integrationist lines. This new cleavage threatens to overtake the classic left/right divide on which the party structure in most European countries is built. This is a powerful element for explaining British Labour Party policy on the issue (even though it is not the only one).

During the year 2003, Prime Minister Tony Blair expressed repeatedly his unwillingness to hold a referendum on the EU Constitution. Blair suddenly changed his mind in April 2004. The decision appears as a tactical (Leonard and Gowen; 2004: 58) and highly partisan move with a view to the performance of his party in both the EP elections of June 2004 and the anticipated General Elections of 2005. In this way, Blair impeded that an issue that threatened the internal cohesion of the Labour Party polluted the campaign (apparently, pressure from Minister of Treasury, Gordon Brown, Secretary of State, Jack Straw and Deputy Prime Minister John Prescott resulted decisive in the U-turn). He removed a tactical weapon regarding the role of plebiscitary democracy from his political opponents and he created the possibility of a cross-party consensus of pro-European elements that could be capable of leading a successful referendum campaign. The opposition was left in a very uncomfortable situation; in the one hand, they wanted a referendum to undermine the government but, on the other, they were left without a strong alibi for criticizing the government. The Conservative leader, Michael Howard considered that the pool was not a good idea. Whilst the Maastricht Treaty that transferred powers to the Union was not ratified by referendum, the Constitution (that did not) would be ratified by referendum.

Whilst the British case results a very obvious one, other countries display similar situations. Thus, in Italy, calling a popular vote would have probably highlighted the internal divisions over the Constitutional Treaty of both the government and the opposition coalition.

4.2. The emerging of a European convention on ratification through referendums

Apart from these domestic factors, decisions to convening ratification referendums have been shaped by arguments that portray the constitution as a fundamental change in the nature of the EU that requires citizens’ involvement to guarantee its legitimacy. The outcome negotiated (i.e. a Constitution) (as well as the process of negotiation itself) created an interpretative framework that induced (either directly or indirectly) a logic justifying a revision of the domestic arrangements for ratification. Additionally, Member States imitated policy solutions (i.e. ratification procedures).

Among the countries holding a referendum, governments appealed commonly to the fact that the EU Constitution represents a qualitative change in the process of European integration that requires an input of direct citizens’ legitimacy. These arguments were voiced in Spain, France, Portugal and Poland. In Spain, the Conservative government initially opposed holding a referendum. Former Foreign Affairs Minister and member of the Convention Ana Palacio rejected this proposal in 2002 and said that since it is a Treaty, it should not be subject to a popular consultation, “nobody has ever seen a treaty being ratified by a referendum. It is not the usual way, since a referendum always refers to internal matters of the member states”.[9] In a similar U-turn to the British case, the Conservative government changed mind and proposed holding the first consult on the EU to be held in Spain. The main supporting argument became that the Constitution was not another Treaty but a re-foundational act whose importance should be backed by citizens’ endorsement. And in June 2003, Palacio announced the possibility to hold a national referendum on the European Constitution. Once the official position of the former governing Partido Popular, Peoples Party (PP), turned in favour of a referendum, no other opposition against holding a referendum was left, because all parties, with different motivations, and the civil society had already supported and argued in favour of the convocation of a referendum. Some commentators expressed puzzlement for this unnecessary (in terms of public demand and domestic constitutional need)[10] referendum.

Similar arguments were voiced by the Dutch government (in its general comment of the Treaty) for whom it was not merely another Treaty amendment and, hence, it required more involvement from the citizens (hence, a referendum). In doing so, the Dutch government disregarded the opinion of the Council of State for whom the EU Constitution did not affect Dutch Constitution and it did not imply new transfer of powers. On this, the government followed the claims of the opposition parties that argued that the referendum would increase the role of citizens in EU politics and their awareness on EU issues in general (Hussain, Maitland and Whitman; 2005). Also in Portugal, Prime Minister Durao Barroso invoked in Autumn 2003 the necessity that the (Portuguese) people would legitimise the Constitution.

In countries not holding a referendum, opposition parties borrowed arguments developed in other Member States. They emphasised that decisions concerning reform of the EU had a cross-partisan character and affected the nation in a different way to normal politics. This happened in Germany and Greece. In Germany, the FDP submitted two Proposals of Law[11] for reforming the Fundamental Law with the objective of holding a referendum on the EU constitution. The justification argued that the EU constitution entails a fundamental change of the EU that conditions the future development both in its very nature and its competencies. Hence, citizens should have the possibility to decide on it. The Green Party argued also that the EU Constitution is something qualitatively different from former treaties and it enhances the efficiency, transparency and democracy in the Union. A referendum will increase its legitimacy and it would create a European space for debate. In Germany, the EU referendum issue became intermingled with earlier and ongoing discussions on the necessity to reform the Constitution in order to introduce elements of direct democracy. In Greece, the Socialist PASOK party expressed its support for the text, calling it a "key step for a democratic, politically strong and fair Europe" in a resolution approved in its Congress on March 2005. However, the Greek socialists argued that the citizens should debate the Constitution because of its importance. The party said a referendum would "satisfy the demand of citizens to be fully informed and to decide the future of Greece in the EU."

In summary, domestic political and electoral reasons explain why convening referendums but the EU Constitution raised also a strong doctrinal case on the necessity of ratification through referendum. The coincidence of a growing number of member states on referendum as the essential ratification mechanism (even in countries that normally do not use it) marks an emerging “europeanisation” of the procedure.

Concluding remarks: How much Europeanisation?

Despite the reference to national constitutional procedures, the three procedures used show a certain degree of convergence in their application by Member States: constitutional reform and constitutional courts’ ruling have become increasingly coincident in their interpretation of EU law. Parliamentary ratification follows also similar pattern of coincidence and the EU constitution has provided the basis for the emergence of a Pan-European cuasi constitutional convention on the utilisation of referendums for ratification. Europeanization (understood as a convergence in ratification mechanisms) has happened as an adaptation of national constitutions to EU rules, as well as the recourse to referendums as the essential ratification mechanism