A Proposal Towards a European Defence Union

In the context of profound (geo-)political changes, and following the Conference on the Future of Europe, the European Parliament (EP) has called for wide-ranging Treaty changes. In November 2023, it adopted proposals for Treaty reform, which would have far-reaching implications for the area of defence within the European Union. This blog post analyses the proposed formation of the European Defence Union (EDU) and the introduction of qualified majority voting (QMV), among other major propositions, while concluding that the new framework would likely create contradictory outcomes and undesirably challenge the current constitutional balance. Hence, building on the current Treaty framework would be a preferable solution.

‘European Defence Union’ and extraordinary powers of the Commission concerning the emergency measures

Although a general discussion of the proposal has been included on Verfassungsblog, it has received relatively little attention, let alone a legal analysis of its implications for the area of defence. Such an analysis is relevant considering the security challenges for Europe, leading to calls for far more integrated EU defence and even nuclear weapons. At the same time, for the EU as a value-based community, the democratic legitimacy of decision-making is an important checkpoint against which these reforms have to be judged. Furthermore, rule of law considerations require a review of legal necessity and consistency of proposed modifications.

The ‘flagship’ of this proposal is the creation of an EDU. Before the proposal was adopted by the EP, the EDU seemed mainly to be used as a political term without a fixed or legal meaning. In 2017, Juncker, as then President of the European Commission, wanted to achieve a ‘fully-fledged’ EDU by 2025. The current President agrees with that objective but seems to have let go of the target of delivery in 2025 and specifically includes candidate-Member States in that ambition.

It is noteworthy that the proposal of the EP uses the EDU to label the newly-envisaged military dimension of the EU. That ignores the current Treaty provisions that already provide the concept of a ‘common defence’. It is not entirely clear whether this was a conscious decision of the ‘framers’. Article 42(2) TEU provides the basis for (the establishment of) a ‘common defence’. The procedure of the Article prescribes that the European Council can decide to this end unanimously, after which the Member States must approve that decision and test it against their own constitutional provisions. This procedure is comparable to the simplified treaty amendment.

The proposed EDU is intended to actively provide security for the Union. The proposal, cosmetically, attempts to bring Article 42(7) TEU more in line with the famous Article 5 of NATO. The 2023 Treaty Reform Proposal explicitly states that ‘an armed attack on one Member State shall be considered to be an attack on all Member States’. However, Article 42(7) TEU is to a large extent already comparable to Article 5 of the NATO Treaty and is arguably formulated more compellingly and broadly for EU Member States: it contains ‘the duty’ to provide assistance ‘by all means’ at their disposal. And for the record, ‘all’ also includes military resources. A novelty is that in the new text the EDU itself is mentioned as an entity (besides Member States) upon which the obligation of military solidarity rests.

Closely related to military solidarity, especially in a whole-of-society-approach and in relation to hybrid threats, are emergency measures. Article 222 TFEU currently contains the ‘solidarity clause’, reserved for terrorist attacks and natural or man-made disasters. Institutionally, the Council has a central role in that regard. The proposed amendment of Article 222 TFEU aims to change this considerably. The Council, together with Parliament, may in case of an ‘emergency’ grant the Commission ‘extraordinary powers’, which includes ‘those to enable it to mobilise all necessary instruments’.

Remarkably in light of these considerations, the proposal leaves Article 4(2) TEU untouched. Currently, this Article figures as a cornerstone in limiting the scope of EU defence, as it stipulates that the EU respects the essential state tasks of (among other things) territorial defence and national security. Therefore, there seems to be a tension between the proposal and this provision. A complete transfer of military units and powers, as for example was intended with the European Defence Community (EDC) in the 1950s, is in any case not possible without amending Article 4(2) TEU. However, in the eyes of the EP, apparently, putting military units under ‘operational command’ of the EU (more about that later), is compatible with the current Treaty. That explanation seems defensible. After all, respecting national tasks does not preclude the EU from executing additional functions which do not conflict with the national prerogatives. Further debate is necessary to decide where the borderline of this provision lies exactly.

QMV in CFSP and CSDP

Besides the creation of an EDU, the proposal also procedurally reforms defence cooperation. Currently decisions within CFSP and CSDP are generally taken by unanimity. This decision-making has been the result of the development of the CFSP and CSDP, and their initial architecture as a special intergovernmental pillar in the Maastricht Treaty. It reflects the assumption that (sovereign) Member States, through the (European) Council, provide the democratic legitimation of decisions in these areas. The consensus-seeking attitude within the Council is still highly valued. However, it is being assailed by the phenomenon of ‘politicisation’ of decision-making resulting in strategic vetoes. Following these developments, the Commission, Parliament, Council and Member States to various extents all recognised the unanimity-requirement often holding the EU back in defending its interests and pursuing its targets.

In that light, the introduction of the QMV in the entire CFSP and CSDP domains envisaged in this proposal is reasonable. Not only would it incorporate the domains more into ‘regular’ EU policies, but it would also increase effectiveness and (combined with other measures) could even strengthen democratic legitimacy. Within the proposal however, some existing guarantees also disappear. Although in the proposed amendment a Member State still ‘may request’ that a matter be referred to the European Council, the ‘emergency-brake’ is stripped of its effectiveness. The European Council also only adopts decisions by QMV. Referring it may contribute to better substantive deliberation with a Member State concerned, but within the European Council a Member State will also be ‘out-voted’. The instrument of ‘constructive abstentions’, necessary to uphold the unity of Union-policy, is abolished. Taken together with the reduced role of the ‘emergency brake’ these changes contain a danger that the consensus-seeking attitude in the (European) Council will eventually fade. This is probably even intended by the authors, as the proposal aims to change the role of the Council into a role comparable to a senate.

However, these are not the only concerns. The unanimity requirement has strong roots in national law and ultimately safeguards national sovereignty. German constitutional law is crucial, partly because the Federal Constitutional Court in the (in)famous Lisbon-Urteil set strict limits on EU integration, including in the field of defence. According to the Federal Constitutional Court, the principle of unanimity (of Article 43 TEU) implies that none of the Member States can be forced to participate in missions against their will – not unthinkable in an EDU with QMV. A Treaty amendment would be needed to change that. However, the Federal Constitutional Court also determined that if Germany would cooperate in such a treaty change, that would be contrary to its own constitution – specifically the constitutive requirement of parliamentary approval. Germany would therefore have to amend its constitution, but because it concerns articles under the protection of the eternity clause, Germany would have to adopt a completely new constitution. Although this is not technically impossible, it seems highly unlikely from a political perspective. Hence, this appears to be an insurmountable obstacle to QMV.

Intricacies concerning the institutional balance

Inconsistencies further emerge when considering the (new) role the EU-institutions have to play in the day-to-day functioning of the EDU. The role of the Parliament increases considerably: it will have to consent to the way CFSP and CSDP are defined and implemented (Article 24(1) TEU). Moreover, the new Article 42(4a) TEU states that Parliament will have to consent to civilian or military EU missions. Notwithstanding the introduction of the QMV, the role of the Council in this new constellation is unclear. This could endanger the goals of these proposals: whereas the increased role of the EP increases democratic legitimacy, it could at the same time undo the envisaged advantage of the QMV (efficient decision-making), as voting within Parliament might present its own inefficiencies.

The newly introduced ‘military units including a permanent rapid deployment capacity’ (RDC) in the revised Article 42(3) TEU provide a good example. The RDC is placed under the ‘operational command of the Union’. This formulation not only implies radical changes, but also creates uncertainty. First, the current RDC which is being set up consists solely of the Member States’ forces and is based on modified EU Battlegroups. According to the new proposal, the RDC will be part of the EDU whereby its forces are part of the EU, whereas Member States may provide additional forces.

Second, it is unclear what type of deployment this RDC is to be tasked with. The new Article 42(4a) TEU requires approval by both the Council and Parliament for missions. These missions refer to Article 42(1) TEU, and have to be deployed ‘outside the Union’. The catalogue of missions in Article 43 TEU is updated in the proposal, but also only refers to missions to be undertaken outside of the Union. The updated collective self-defence clause of Article 42(7) TEU however seems to suggest that the RDC would also be used in cases of self-defence. In cases of aggression, not only the Member States, but also the Defence Union itself shall have the obligation of aid and assistance. In fact, this showcases another contradiction since collective self-defence mentioned in Article 42(7) TEU (and Article 5 NATO Treaty) is derived from Article 51 UN Charter. As only Member States can invoke this Article, the EDU would be unable to activate Article 42(7) TEU.

The third uncertainty created by the proposed Article 42(3) TEU is the terminology of ‘operational command’. Generally, ‘operational command’ refers to the authority exercised over certain (allocated) units and is limited to the accomplishment of mandated tasks. Differently, ‘full command’ refers to the sovereign power to decide on the deployment or retrieval of armed forces altogether. ‘Operational command’ therefore seems to suggest that not the Union, but another entity, retains full command of the RDC – the most logical candidates being the Member States. In that sense, the ‘armed forces’ would not be part of the EDU but rather be forming a European Defence Alliance. Although this would from a legal standpoint and in light of preceding considerations be more correct, it contradicts the new element of provision 42(7) TEU, which creates an obligation for the EDU to provide aid and assistance in case of aggression.

Conclusion

The proposed changes seem to flow from a general and abstract vision of the functioning of the EU. Accordingly, the relationship between the executive and the representative branch would be more conventional than in the current EU, and the national government’s role in the decision making would be reduced to something similar to a senate.

Without judging this vision of the EU an sich, in our view it would not necessarily lead to a better defence union. Furthermore, due to its different and uncertain constitutional nature, its creation would be in the end a shot in the dark. It is not clear how the newly proposed framework would solve the current problems and why it would be preferable over the options under the current Treaties.

Analysis of the proposed EDU, introduction of QMV and consequences for institutional balance shows that amending provisions often fail to solve existing problems. The principles underlying the current Treaties (most prominently Article 4(2) TEU) are left intact. Yet, the CFSP and CSDP provisions are amended to such an extent that they contradict the respective principles. This creates inconsistency as well as legal uncertainty, which is problematic from a democratic perspective.

In our view, implementing an EDU and accompanying procedures should correspond to the current functioning of the CDSP and its legal basis. The current Treaties, already designed to guard democratic legitimacy, provide sufficient procedural possibilities to achieve desired aims. Building upon these foundations not only strengthens the EU externally, but also internally.