Stanley's blog

Irish Taoiseach (prime minister) Brian Cowen has re-confirmed that the ‘guarantees’ promised by EU leaders at the December summit “must be legally robust in order to reassure the public about the treaty”. He said: “Whilst I respect the fact that other member states do not wish to re-ratify the Lisbon Treaty, I made it clear that for my part the legal guarantees will have to be attached to the EU treaties at the next possible opportunity.”

Denmark’s guarantees and opt-outs, promised after the first referendum on the Maastricht Treaty in 1992, were only legally implemented by the Amsterdam Treaty. But Croatia is expected to join the EU in 2010-11 and the idea is to add a protocol to the Treaty of Accession.

The question arises, however, whether it is legally possible to

add to an accession treaty an Irish-specific protocol containing such legal guarantees, or whether only issues pertaining to a state’s accession may be dealt with. “Adding this protocol to the Croatian accession treaty would leave the treaty wide open to attack in the courts.” says Andrew Duff MEP.

If an accession treaty is not used, the legal implementation would have to await a new EU treaty to be drawn up and ratified. The ‘Irish Protocol’ could, of course, be ratified by the Member States at the same time as, but separate from, the Croatian Treaty of Accession. In either case, ratification is not guaranteed.

A final decision on how to structure the guarantees is supposed to be taken at the June EU summit.

The objection raised is fascinating for lawyers but may not have any political or practical relevance. The procedure adopted by the institutions and Member States can only be invalid if (a) a case is brought before the European Court of Justice (ECJ) and (b) the Court strikes down the procedure.

There are precedents of the institutions and the Member States living with legally unsatisfactory situations. Provisions of the original treaties – notably those relating to qualified majority voting – were not implemented, not least, but not entirely, because of the Luxembourg Compromise. In some cases, they were re-enacted in the Single European Act, as if they were new. Despite extensive criticism on legal grounds the issue was never tested.

And this is what is most likely to happen with the Irish protocol. There is unlikely to be anyone with locus standi who would challenge in the ECJ non-compliance with Community law.

Even were someone with locus standi willing to bring a case, it is unlikely that the ECJ would strike the provisions down.

Author :


  1. The position of the ECJ is not quite the end of the issue – because doubts about the legality of the process might lead to reluctance to ratify the treaty in national parliaments, or challenges before national courts to the legality of ratification of the treaty, cf the 1980s case that started the whole saga of Irish referenda off in the first place. There is also the point that there are two different legal bases for accession treaties and Treaty amendments (Arts 48 and 49 TEU), which entail different decision-making processes to some extent. I think the best solution is to draw up a single treaty based on those two legal bases, which encompasses both a Treaty amendment and an accession treaty simultaneously.

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

  2. You don’t link to any external article on this, and I was wondering on what Duff bases himself to question the legality of such a protocol?

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

  3. As a lawyer, I think the statement of Mr Duff, which has inspired this post, might in itself be somewhat off the mark. Apparently he has said that “adding this protocol to the Croatian accession treaty would leave the treaty wide open to attack in the courts.” I wonder whether this is really the case. If the right honourable gentlemen is referring to a case raised before any national courts, it appears unlikely that they would be able to decide what is proper to be included in an accession treaty or not. There is no legal reason why a protocol pertaining to a different matter could not be annexed to the treatise that generally deals with the conditions of Croatia’s accession. If Mr. Duff is however thinking about a case iniated before the EU Courts in Luxembourg, there is the precedent of the Roujansky v. Council cases seems applicable (T-584/93 and C-253/94P). In these cases, the EU Courts stated that an action for annulment (ex Article 230 of the EC Treaty) cannot be staged against an EU (amendment) Treaty. The same will undoubtedly apply to challenges against accession treaties. So, if natural and legal persons have no possibility for a legal challenge, this still leaves Member States and EU institutions as possible applicants – but indeed, it would seem most strange of any of them would be attacking a treaty and protocol in court where they had an important role to play in its negotiation process. If they do not like the outcome, they can just decide not to ratify it. In sum then, as said, Mr. Duff’s remarks seem slightly erroneous from a legal point of view.

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

  4. Whatever emerges by way of agreement, there can be nothing more than an undertaking to attach it by way of protocol to a future treaty. The protocol solution seems to have been dreamed up by politicians with little awareness of the legal difficulties. The argument will have to be made that the agreement, in all likelihood, ‘a decision of the Heads of State and Government meeting within the European Council’ , is legally binding in international law, as was the case with regard to Denmark.

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

  5. I agree that no legal challenge before the EU courts against such a Protocol would be admissible, but I still think there is a possibility of a legal challenge in the national courts by some individual aggrieved by European Union integration in general – cf the challenges we have seen several times in the UK, Ireland, Germany, the Czech Republic and constitutional court proceedings also in Spain and France (I may have missed some).

    The legal point is that it would NOT be possible simply to attach a protocol dealing with anything at all to the accession treaty with Croatia, because accession treaties have a particular ‘legal base’ with a specific decision-making procedure (Article 49 TEU). But a measure amending the TEC/TEU needs the ‘legal base’ of a Treaty amendment (Article 48 TEU), which entails a different decision-making procedure in some respects from accession treaties. Also in some Member States it may be arguable that a different national ratification process would be applicable for accession treaties as compared to Treaty amendments. The only solution is to use the double legal base of Arts 48 and 49 TEU at the same time, and to combine all the features of the decision-making applicable to the two different legal bases.

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

  6. Steve & Raphael: I do not think that doubts about the legality of the process would lead to reluctance to ratify the treaty in national parliaments or challenges before national courts to the legality of ratification of the treaty, but accept that it is a possibility. The two different legal bases for accession treaties and Treaty amendments (Arts 49 and 48 TEU respectively) could be used but Andrew Duff fears that the use of Article 49 would bring into play the requirement for an intergovernmental conference (IGC), with all that entails.

    H de Waele: I agree that an attack in national courts or the ECJ is unlikely. I too cannot imagine a Member State or EU institution applying.

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

  7. With respect, these exchanges miss the point. The Danish opt-outs document, a Decision of the HOSG meeting within the EC, as you will be aware, was registered with the UN some months subsequent to its agreement in Edinburgh as an international treaty and therefore theoretically justiciable in international law. The contents were subsequently enshrined (a major political error) by way of a protocol to the Amsterdam Treaty. The texts changed the relationship that Denmark had with the EU but did not alter any existing treaty texts. No amendment was involved. The same approach will, ipso facto, have to be followed in the case of Ireland as any attempt to draft a protocol to the Lisbon Treaty would ineluctably lead to re-ratification in all Member States, exactly the situation confronting other Member States in the case of Maastricht.

    The accession treaties for Romania and Bulgaria, incidentally, contain a chapter dealing with institutional changes and refer to the now defunct Constitutional Treaty.

    One of the “legal guarantees” mentioned in the conclusions of the December 2008 European Council, that relating to taxation, is not even Ireland specific.

    The Irish situation is identical in institutional terms to that of Denmark after the first failed Maastricht vote. As the Danes were looking for real opt-outs (EMU, defence and home affairs) and, insofar as can be judged, the Irish are not, the legal problem is of considerably less significance and the instigation of a debate abot the requirement for a protocol an error, first on the part of the Irish government and second for Sarkozy in giving it credence in his subsequent press conference.

    There is no mention of it in the formal conclusions of the European Council.

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

  8. There is no need for a ‘protocol to the Lisbon Treaty’ to deal with the Irish situation – as it is perfectly possible to add a protocol to the treaties by means of a separate legal process, after the Lisbon Treaty is in force. This is obviously what is being contemplated. So the issue of re-ratification of Lisbon would not arise.

    Re the Danish deal at the 1992 summit, the mixture of public international law and EU law which this entailed was always highly problematic legally, and the Danes rightly addressed it properly in the Treaty of Amsterdam – they could have lost their referendum otherwise.

    And anyone who thinks that there is no prospect of a national legal challenge to an Irish Protocol that is at all legally questionable is underestimating Eurosceptics and overlooking their use of national courts throughout many years of Treaty revision.

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

  9. The attached report from the Irish Times may be of interest.

    The Danes had to put the Maastricht Treaty, unchanged, to their people in a second referendum on the basis of the Edinburgh agreement, irrespective of legal questions relating to the status of the latter. The Irish will have to do the same with regard to whatever emerges from the June European Council.

    In the popular mind, and as a matter of fact, a ‘protocol’ is as legally binding as the text to which it relates. This is, unfortunately, the one thing that Ireland cannot have before the October deadline for a second referendum. Neither can Ireland put to the people a text which is not yet ratified by other MS. Only what emerges from the meeting by way of agreed text can, in logic, be put to the Irish people again.

    The legal ingenuity of the initial ‘Danish solution’ was that it bridged the irreconcilable by creating a new instrument, that of a ‘Decision of the HOSG meeting within the European Council’ which it can be argued is ‘legally robust’ – to take the phrase being used by the Taoiseach – in international law. (Apart from the Edinburgh precedent, there is also the precedent that Denmark applies certain aspects of JHA under international rather than EU law).

    Assuming that the IT report was based on an official briefing, the Irish are still trying to have their cake and eat it.

    [WORDPRESS HASHCASH] The poster sent us ‘1055386704 which is not a hashcash value.

Comments are closed.