Solidarity and deterrence in the Eastern Mediterranean An analysis of the delicate question of collective defence between EU member states vis-à-vis Turkey Recherches & Documents N°13/2021 Elie Perot Brussels School of Governance (BSoG), Vrije Universiteit Brussel (VUB) June 2021 www.frstrategie.org List of acronyms AWACS CSDP EEZ EIB ESDP EU EUGS HR/VP ICJ MBT NATO TEU TFEU UN UNCLOS WEU Airborne Warning And Control System Common Security and Defence Policy Exclusive Economic Zone European Investment Bank European Security and Defence Policy European Union European Union Global Strategy High Representative of the Union for Foreign Affairs and Security Policy/Vice-President of the European Commission International Court of Justice Modified Brussels Treaty North Atlantic Treaty Organization Treaty on European Union Treaty on the Functioning of the European Union United Nations United Nations Convention on the Law of the Sea Western European Union F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E TABLE OF CONTENTS Executive summary............................................................................................................. 1 Introduction: the collective defence dimension of the dispute with Turkey in the Eastern Mediterranean ................................................................................................................... 2 1. Reacting to the crisis in the Eastern Mediterranean: collective defence through the EU as an option of last resort ................................................................................................... 5 1.1. At the root of NATO’s embarrassment: institutional limitations and the hesitant leadership from the United States ............................................................................................... 5 1.2. The European Union’s prudent strategy: treading a thin line between dialogue opening and limited sanctions .................................................................................................................. 7 1.3. Greece’s strategic rationale for raising the possibility of triggering article 42.7 TEU: alarming friends and foes alike ................................................................................................. 11 2. Activating article 42.7 TEU in the Eastern Mediterranean: under which conditions? . 13 2.1. Admissibility: would it be formally possible to invoke the EU collective defence clause against a NATO ally? ................................................................................................................. 13 2.1.1. Historical background: the WEU’s Petersberg Declaration and the European Council’s 2002 assurances on EU-NATO relations ............................................................................................................... 13 2.1.2. Assessing the consequences (or lack thereof) of the Petersberg Declaration and the European Council’s 2002 assurances with regard to article 42.7 TEU ......................................................................... 15 2.1.3. Interpreting the “NATO caveat” directly enshrined in the EU collective defence clause............... 16 2.2. Casus foederis: how could the terms of article 42.7 TEU come into play in the Eastern Mediterranean? ........................................................................................................................ 17 2.2.1. Reaching a sufficient degree of gravity: the threshold of “armed aggression” ............................ 17 2.2.2. The geographical scope of article 42.7 TEU and the Eastern Mediterranean: possible interplays and issues .................................................................................................................................................... 19 3. Looking forward: scenarios and implications............................................................. 24 3.1. What if the EU collective defence clause were to be activated in the Eastern Mediterranean? ........................................................................................................................ 24 3.1.1. Flexibility: adapting EU member states’ responses to circumstances ........................................... 24 3.1.2. Differentiation: the possibility for EU member states to contribute in distinct manners to implementing article 42.7 TEU .................................................................................................................... 27 3.2. The question of the operationalisation of article 42.7 TEU and the Eastern Mediterranean .......................................................................................................................... 29 3.2.1. A growing ambition: to clarify and strengthen the EU collective defence clause ......................... 29 3.2.2. The EU collective defence clause and the recent tensions in the Eastern Mediterranean: what impact on the Strategic Compass and the envisaged declaration on article 42.7 TEU?.............................. 31 Conclusion: a scenario bound to remain taboo? ............................................................... 35 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Executive summary Paper completed on June 3, 2021 During the crisis with Turkey in the Eastern Mediterranean in 2020, the Greek government explicitly raised the possibility of activating article 42.7 of the Treaty on European Union (TEU) – the EU collective defence clause which notably provides that “[i]f a Member State is the victim of armed aggression on its territory, the other Member States shall have towards it an obligation of aid and assistance by all the means in their power, in accordance with Article 51 of the United Nations Charter”. The objective of the present report is to analyse this complex and particularly sensitive issue. To this end, the report first places the discussion in context by reviewing the Western reactions to the recent Greek-Turkish crisis. In so doing, the report shows that it was largely due to both the internal difficulties encountered within NATO in dealing with Turkey’s actions in the Eastern Mediterranean and the perceived softness of the EU in this regard that Greece found it necessary to sound the alarm by raising the possibility of triggering the collective defence mechanism provided for in the EU treaties. The report then examines, from a legal perspective, the conditions under which article 42.7 TEU could be activated in the Eastern Mediterranean. As a preliminary step, the report verifies that it would be formally admissible to invoke the EU collective defence clause against a member of NATO. The report then looks at the two criteria that would have to be fulfilled to activate article 42.7 TEU – namely that a given EU member state should have been the victim of “armed aggression” and that such armed aggression should have taken place on the “territory” of that member state – and examines in particular how this could prove problematic in the specific context of the Eastern Mediterranean. Turning finally to the way forward, the report first outlines how the EU collective defence clause could be implemented if necessary in the Eastern Mediterranean. In this regard, the report stresses that the intensity of the actions undertaken according to this provision would have to be commensurate to the specific circumstances in which the latter would be invoked and that EU member states would most likely contribute to such actions in different ways. The report also considers to which extent the latest tensions between Greece and Turkey should lead to a reassessment of the ambition to operationalise article 42.7 TEU that has gradually emerged in recent years at the EU level. The report concludes, however, that the possible link between these two issues is likely to remain indirect and ambiguous, including in the context of the elaboration of the Strategic Compass and a possible political declaration on article 42.7 TEU, on account of the political sensitivities involved. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 1 Solidarity and deterrence in the Eastern Mediterranean Introduction: the collective defence dimension of the dispute with Turkey in the Eastern Mediterranean After the diplomatic crisis which erupted in the summer of 2020 in the Eastern Mediterranean, tensions between Greece and Turkey, both nominal allies within the North Atlantic Treaty Organization (NATO), kept ebbing and flowing. These fluctuations closely matched the activity cycles of the Oruç Reis, the Turkish seismic research vessel which was exploring the Eastern Mediterranean under a military escort, looking for hydrocarbon reserves in areas viewed by Greece and Cyprus as part of their respective exclusive economic zones (EEZ) or continental shelves1. Thus, when the Turkish government decided to recall the Oruç Reis to the port of Antalya around mid-September, this move was logically welcomed by Greece as a first sign of de-escalation2. But when the same Turkish ship and its military escort resumed their contested maritime activities one month later, regional tensions were reignited, prompting Athens to intensify its diplomatic efforts to get the European Union (EU) and its member states to finally adopt a tougher stance towards Ankara. On 19 October 2020 in particular, the Greek minister for Foreign affairs, Nikos Dendias, sent a series of letters across European chancelleries. In a first letter, he asked some of his European counterparts to stop exporting military equipment to Turkey3. In a second letter, he enjoined the European Commissioner for Neighbourhood and Enlargement, Olivér Várhelyi, to consider suspending the EU-Turkey customs union4. Finally, in a third letter addressed to the High Representative of the Union for Foreign Affairs and Security Policy (HR/VP), Josep Borrell, the Greek minister for Foreign affairs denounced what his government regarded as Turkey’s “illegal and aggressive behaviour” in the Eastern Mediterranean. This third letter strikingly concluded that “[i]n the short term, the only way forward for the EU [was] to stick to its principles, – especially on internal solidarity and mutual assistance between its Members, as enshrined in EU Treaties, including article 42.7 TEU [Treaty on European Union] – while unambiguously resisting any attempt by Turkey to bully its way forward”5. The purpose of the present report is to analyse the most sensitive point raised by the Greek foreign minister during the dispute with Turkey in the Eastern Mediterranean last year, namely the possibility of invoking article 42.7 TEU – the EU collective defence clause which 1 For an overview of the different dimensions of the Eastern Mediterranean dispute, see for instance: Branislav Stanicek, “Turkey: Remodelling the Eastern Mediterranean. Conflicting Exploration of Natural Gas Reserves”, Briefing, European Parliamentary Research Service (EPRS), September 2020. 2 Turkish officials sent contradictory messages, however, regarding the withdrawal of the Oruç Reis. President Erdoğan in particular stated that this move was made to “give a chance to diplomacy”, but also that the Oruç Reis was docked “for restocking and maintenance” and that it would “return to mission in the same fashion after its maintenance [would be] over” (Recep Tayyip Erdoğan, “The Oruç Reis will Return to Mission after its Maintenance is Over”, Turkish Presidency, 18 September 2020). 3 Sotiris Nikas, “Greece Seeks Arms Embargo, Halt to EU-Turkey Customs Union”, Bloomberg, 20 October 2020. 4 Ibid. 5 Emphasis added. The letter is available via: Aurélie Pugnet, “Face à la Turquie, la Grèce invoque une ‘vraie’ solidarité… et la clause d’assistance mutuelle de l’article 42.7. La lettre de Dendias à Borrell”, Bruxelles2, 20 October 2020. A copy of the letter has also been posted on Twitter by Nikos Chrysoloras, Bloomberg’s Brussels bureau chief. 2 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean notably provides that “[i]f a Member State is the victim of armed aggression on its territory, the other Member States shall have towards it an obligation of aid and assistance by all the means in their power, in accordance with Article 51 of the United Nations Charter”6. To be clear, it is not the intention of this report to suggest that article 42.7 TEU should be activated, or that this will probably be the case in the future. Indeed, such an eventuality should be contemplated only in the hypothesis of a dramatic re-escalation on the part of Ankara and only in reaction to it. It should also be stressed from the outset that activating the EU collective defence clause against Turkey would force extremely difficult choices on European countries, since Ankara remains a major NATO ally, a key partner on migration and trade, and, nominally, a candidate to EU membership. It goes without saying, furthermore, that resorting to article 42.7 would not address, as such, the political root causes of the dispute in the Eastern Mediterranean and that it would be infinitely preferable for Turkey and its EU neighbours to resolve their differences by peaceful and legal means, via direct diplomatic negotiations or international arbitration. That being clarified, it is nonetheless striking that the possibility of resorting to article 42.7 TEU in the Eastern Mediterranean has elicited very few reactions thus far, even though this option was explicitly put on the table by the Greek government a few months ago7. It can be assumed that the political volatility of the subject, its technical complexity, as well as sheer incredulity may have all contributed to this lack of discussion8. However, even now that 6 In full, article 42.7 TEU reads as follows: If a Member State is the victim of armed aggression on its territory, the other Member States shall have towards it an obligation of aid and assistance by all the means in their power, in accordance with Article 51 of the United Nations Charter. This shall not prejudice the specific character of the security and defence policy of certain Member States. Commitments and cooperation in this area shall be consistent with commitments under the North Atlantic Treaty Organisation, which, for those States which are members of it, remains the foundation of their collective defence and the forum for its implementation. Since article 51 of the UN Charter recognizes for its part the inherent right of states to “individual or collective self-defence”, this paper uses the term “collective defence” rather than more ambiguous or euphemistic expressions like “mutual assistance” when referring to article 42.7 TEU. 7 Needless to say, references to article 42.7 TEU have not been a personal fixation of the Greek minister for Foreign affairs Nikos Dendias. Greek Prime Minister Kyriakos Mitsotakis, for instance, also explicitly mentioned the possibility to invoke the EU collective defence clause in autumn 2020 during a press conference at the Thessaloniki Helexpo Forum. Responding to a question about whether the security agreement that was under negotiation between Greece and France would include a bilateral mutual defence pledge, the Greek Prime Minister said: “You asked me about defence cooperation, the possible strategic relationship we can sign with France. Indeed, we are at an advanced stage to sign such an agreement which will reaffirm the very close ties that Greece has with France. I must say, however, that on the issue of the mutual defence clause, I would like to remind you that such a clause exists at the European Union’s level. It is article 42 (7) of the Treaty on European Union, which stipulates that member states have the obligation to assist, and assist with all the means at their disposal, a state affected by an armed attack on its territory. […] I hope that [Greece] will never need to activate this clause” (Kyriakos Mitsotakis, Press conference of the Prime Minister at the Thessaloniki Helexpo Forum, Greek Prime Minister’s Office, 13 September 2020). It is also worth noting that Syriza, the main opposition party in Greece, led by former Prime Minister Alexis Tsipras, explicitly called in October 2020 for “activating EU mutual defence clauses [sic]” to protect Greece’s sovereign rights in the Eastern Mediterranean (see: “SYRIZA: National problems cannot be resolved using public relations stunts and the media”, Athens-Macedonian News Agency, 20 October 2020). 8 See for instance the evasive answer that NATO Secretary General Jens Stoltenberg gave (for obvious diplomatic reasons) when a journalist from a Turkish news channel asked him about the signifi- F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 3 Solidarity and deterrence in the Eastern Mediterranean Greece and Turkey have resumed their “exploratory talks” over their bilateral disagreements9 and that Turkish authorities have toned down their confrontational rhetoric10, it would still be imprudent to ignore the possibility of an invocation of the EU collective defence clause in the Eastern Mediterranean. Indeed, unless Ankara clearly renounces to assert its territorial claims through a strategy of faits accomplis, it is likely that the threat of a military conflict will persist in the region in the months and years to come11. Accordingly, without denying the acute political sensitivity of a potential recourse to the EU collective defence clause against Turkey, the low probability of such an event occurring or its undesirable consequences, it remains important to examine this worst-case scenario carefully and think through the possibilities and complexities that it could involve. To this end, the rest of this report is divided into three main sections. Its first section reviews the Western reactions thus far to the recent crisis in the Eastern Mediterranean, explaining in particular why the possibility of invoking article 42.7 TEU against Turkey came to be explicitly highlighted by the Greek government in this context. Its second section considers under which conditions this provision could be activated from a legal viewpoint. More specifically, this second section examines whether it would be formally admissible to invoke the EU security guarantee against a NATO ally and under which circumstances its terms could effectively come into play in the Eastern Mediterranean. The third section of this report finally looks at the way forward. It does so in two ways: first, by discussing the broad contours of how the EU collective defence clause could be implemented if necessary in the Eastern Mediterranean; and, second, by examining the extent to which the recent crisis between Greece and Turkey may be factored into the ongoing work at the EU level to make this mechanism more operational, particularly in the context of the elaboration of the Strategic Compass and a possible political declaration on article 42.7 TEU. cance, from a NATO viewpoint, of the explicit reference by Greece to article 42.7 TEU during its dispute with Turkey (NATO, “Online pre-ministerial press conference by NATO Secretary General Jens Stoltenberg ahead of the meetings of NATO Defence Ministers”, 21 October 2020). Similarly, when asked about a possible activation of article 42.7 TEU in the Eastern Mediterranean, the EU spokesperson for foreign and security affairs largely dodged the question (Sarantis Michalopoulos, “An ‘accident’ in the Mediterranean is in no one’s interest, EU official says”, Euractiv.com, 3 September 2020). 9 The reason why these talks are called “exploratory” is that they are meant to identify the points on which future substantive negotiations could be held. Between 2002 and 2016, Athens and Ankara thus engaged in 60 different rounds of exploratory talks, making little progress overall. While Greece would accept to negotiate over the boundaries of its EEZ and continental shelf with Turkey, the latter would like to bring a broader range of issues to the table, like the delimitation of Greece’s airspace and territorial waters or the demilitarised status of certain Greek islands in the Aegean Sea. 10 Simon Von Dorpe, “Erdoğan: Turkey wants to ‘turn new page’ in EU relations”, Politico, 10 January 2021. 11 See: Sinem Adar and Ilke Toygür, Turkey, the EU and the Eastern Mediterranean Crisis. Militarization of Foreign Policy and Power Rivalry, Comment n° 62, German Institute for International and Security Affairs (SWP), December 2020. On Turkish territorial claims in the Eastern Mediterranean, as encapsulated in the so-called “Blue Homeland” (Mavi Vatan) doctrine, see for example: Ryan Gingeras, “Blue Homeland: the heated politics behind Turkey’s new maritime strategy”, War on the Rocks, 2 June 2020; Murat Sofuoglu, “Turks still debate whether Treaty of Lausanne was fair to Turkey”, TRT World, 26 January 2018. 4 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean 1. Reacting to the crisis in the Eastern Mediterranean: collective defence through the EU as an option of last resort In response to the recent crisis in the Eastern Mediterranean, Western countries have followed what can be described as a dual strategy. They have strived, on the one hand, to smooth rifts with Turkey through a strategy of engagement and de-escalation, both at the NATO and EU levels. On the other hand, a growing emphasis has been put on deterrence, in particular through the EU with the threat of economic sanctions. Thus, it is noticeable that only the Greek government openly alluded to the possibility of resorting to article 42.7 TEU as part of this containment effort. Yet, it is important to understand how these different developments are related to each other. It was indeed due to both the internal difficulties encountered within the NATO framework to deal with Turkey’s actions in the Eastern Mediterranean (1.1) and the EU’s perceived softness in this regard (1.2) that Greece felt necessary to sound the alarm by raising the possibility of triggering the collective defence mechanism provided for under the EU treaties (1.3). 1.1. At the root of NATO’s embarrassment: institutional limitations and the hesitant leadership from the United States At first sight, the Atlantic Alliance would seem to offer a natural platform for resolving the conflict between Athens and Ankara. But NATO is in reality poorly equipped, at least institutionally, to settle quarrels between its members. The formal goal of the North Atlantic Treaty, signed in Washington in 1949, has indeed never been to create a system of collective security, aimed at resolving tensions between its parties, but to establish a system of collective defence, designed to deal with threats posed by external actors (initially the Soviet Union)12. Thus, at most, the first article of the Washington Treaty reminds NATO allies of their obligation under the Charter of the United Nations (UN) “to settle by peaceful means any international disputes in which they may be involved” and “to refrain in their international relations from the threat or use of force in any manner inconsistent with the purposes of the United Nations”. But the North Atlantic Treaty does not provide for any mechanism to adjudicate internal disagreements, nor does it foresee the possibility of sanctioning, suspending or even excluding a member of the organisation that would have caused significant harm to its fellow allies13 . 12 This is not to say that systems for collective defence and collective security are necessarily incompatible. For instance, the Inter-American Treaty of Reciprocal Assistance of 1947, also known as the Rio Pact, is meant to serve both purposes. Its collective defence clause (article 3) was thus invoked in favour of the United States after the September 11 terrorist attacks, while its collective security provisions have been put to use most recently (and quite controversially) in September 2019 to sanction the de facto government of Nicolás Maduro in Venezuela (see e.g.: Peter Meyer, “The Inter-American Treaty of Reciprocal Assistance and the Crisis in Venezuela”, Congressional Research Service, 11 December 2019). 13 In theory, the Washington Treaty could be revised to include provisions allowing for the sanction, suspension or expulsion of a country from the Atlantic Alliance. In practice, however, amendments to this treaty have been extremely rare and much more modest (so far, all the amendments to the North Atlantic Treaty have been marginal adjustments to its article 6, which delimits the geographical scope of NATO’s collective defence clause, itself enshrined in article 5). A revision of the Washington Treaty F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 5 Solidarity and deterrence in the Eastern Mediterranean To be sure, this does not mean that the structures of the Atlantic Alliance cannot play in practice a useful role in reducing frictions between its members. Last October, Secretary General Jens Stoltenberg and his team managed to make Greece and Turkey agree on establishing a bilateral de-confliction mechanism, including a hotline based on NATO networks, and on cancelling two “provocative” military exercises that were supposed to take place on the dates of their respective national holidays14. The hope that was expressed at the time at NATO was that these measures, designed to limit the risk of incidents at sea and in the air between Turkish and Greek armed forces, would help to rebuild trust between the two neighbours, thus opening the diplomatic space needed for holding talks about the substantive causes of their dispute in the Eastern Mediterranean15. Nonetheless, if NATO has often effectively contributed to pacify the relations between its members throughout history, this stabilising role has never really been predicated on technical arrangements such as the one concluded last October between Greece and Turkey. Instead, NATO’s latent geopolitical ability to foster peace between its members has always ultimately rested on the political leadership exercised by the United States over its European partners, whether inside or outside the formal structures of the Atlantic Alliance. Thus, in 1995-1996, it was thanks to the direct intervention of President Clinton and his closest advisors with officials in Athens and Ankara, not to NATO structures per se, that an armed conflict between Greece and Turkey was narrowly averted over the Imia/Kardak islets in the Aegean Sea16. Yet, what was striking in 2020 in the Eastern Mediterranean was, precisely, the absence of resolute American leadership. The latest crisis between Greece and Turkey intensified during the final months of Donald Trump’s turbulent presidency, when the latter was mostly focused on his re-election campaign. Thus, throughout the summer and autumn of 2020, President Trump only made half-hearted appeals to dialogue between Turkey and its neighbours, demonstrating in reality very limited interest in playing the role of policeman in the region (as in the rest of the world for that matter). It would be unfair, of course, to say that US diplomacy under President Trump tried nothing to appease tensions in the Eastern Mediterranean, as Secretary of State Michael Pompeo repeatedly called on Turkey and Greece to settle their dispute “diplomatically and peacefully”17. But it has to be recognised that, overall, these diplomatic efforts produced only very limited results. The other factor that has complicated US efforts to contain tensions within NATO in recent times is that the United States has found itself directly at odds with Turkey on an increasing would require, in any event, the unanimous consent of NATO allies and could thus be blocked by a veto from any one of them – including Turkey. 14 Talks on a de-confliction mechanism were initially refused by Athens, as Turkish ships were still operating in disputed maritime zones in the Eastern Mediterranean (Micheal Peel, Kerin Hope, Laura Pitel, “Greece pulls back from Nato eastern Mediterranean talks”, Financial Times, 4 September 2020). 15 NATO, “Military de-confliction mechanism between Greece and Turkey established at NATO”, 1 October 2020. 16 Michael Robert Hickok, “The Imia/Kardak affair, 1995-96: A case of inadvertent conflict”, European Security, 7(4), 1998, pp. 118-136. 17 See for instance: Michael Pompeo, Remarks at the Cyprus Centre for Land, Open-seas, and Port Security – Memorandum of Understanding Signing Ceremony, 12 September 2020. 6 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean number of issues – a growing antagonism which has complicated Washington’s ability to play its traditional role of honest broker between Ankara and Athens. US-Turkish dissensions flared up in particular in December 2020 during the NATO Foreign affairs ministers’ summit, when a sharp exchange took place between Michael Pompeo and his Turkish counterpart, Mevlüt Çavuşoğlu. On this occasion, the US Secretary of State reportedly accused Ankara of undermining the cohesion of the Atlantic Alliance through its unilateral actions in the Eastern Mediterranean or its acquisition of the Russian S-400 ground-to-air missile defence system – a decision for which the Trump administration (under pressure from the US Congress) eventually imposed sanctions against the Turkish defence procurement agency and some of its senior officials18. Given Joseph Biden’s particularly harsh words towards Recep Tayyip Erdoğan during the Democratic presidential primaries (the former portraying the latter as an “autocrat”19), it has been generally assumed that the new US presidential administration would adopt an even tougher stance towards Turkey than the one initiated in the last days of the Trump administration20. However, it is quite possible that the Biden administration will be much less unequivocal in practice, forced to strike a balance between confronting Turkey and preserving the cohesion of the Atlantic Alliance – as already evidenced by the more conciliatory tone adopted by the new Secretary of State, Anthony Blinken, towards Ankara so as not to irrevocably push the latter into the arms of Moscow21. 1.2. The European Union’s prudent strategy: treading a thin line between dialogue opening and limited sanctions Because of the crucial role played over the years by the United States in resolving disputes between Greece and Turkey, the Europeans have long been able to comfortably avoid becoming embroiled in the crises that have periodically erupted between the Aegean neighbours. In 2020, however, the difficulties encountered within the transatlantic framework to resolve the spat between Athens and Ankara forced the EU and its member states to step up their efforts and take on greater diplomatic responsibility. In that perspective, the EU’s strategy has been to try to strike a balance between engaging Ankara and defending the interests of its member states which seemed threatened by Turkish actions22. This twofold strategy, trying to combine conciliation and firmness, was illustrated by the opposite (but perhaps ultimately complementary) roles that Germany and 18 David Herszenhorn, Nektaria Stamouli, Rym Momtaz, “US and Turkey target each other in NATO meeting”, Politico, 2 December 2020. On US sanctions over the procurement of the S-400 air defence system, see: US Department of State, Factsheet: CAATSA Section 231 “Imposition of Sanctions on Turkish Presidency of Defense Industries”, 14 December 2020. 19 Editorial Board, “Interview with Joe Biden, former vice president of the United States”, New York Times, 17 January 2020. 20 Reuters Staff, “U.S. secretary of state nominee Blinken says Turkey not acting like an ally”, Reuters, 19 January 2021. 21 Alexandra Brzozowski, “US, NATO have strong interest in keeping Turkey close, says Blinken”, Euractiv.com, 24 March 2021. 22 European External Action Service, “Video conference of Foreign Affairs Ministers: Main outcomes”, 14 August 2020; President of the European Council, “Conclusions by the President of the European Council following the video conference of the members of the European Council”, 19 August 2020. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 7 Solidarity and deterrence in the Eastern Mediterranean France respectively played at the bilateral level in the summer 202023. Berlin, which was then holding the rotating presidency of the Council of the European Union, redoubled its efforts to calm down tensions and bring the parties back to the negotiations table, while Paris, by contrast, focused on deterrence, strengthening from mid-August its air and maritime presence in the region and participating in military exercises alongside Greek and Cypriot forces – a show of force that Italy later joined as well24. Officially, the EU has given (and still gives) priority to the first part of this twofold strategy, that is, to de-escalation and dialogue. Since November 2019, however, a legal framework exists to impose sanctions on Turkish businessmen (personal travel ban and asset freeze) over their participation in unauthorised drillings in Cypriot waters25, a framework which the EU chose to apply in February 2020, initially against a vice-president and a vice-director of the Turkish Petroleum Corporation (TPAO)26. Yet, despite the strong insistence of Greece or Cyprus, most EU countries have been particularly loath to go much further than these limited measures, for fear of burning bridges with Ankara. Thus, during the summit of the European Council that took place in early October 2020, EU heads of state and government tried hard to maintain the door open for diplomacy. Instead of wielding the “stick” of additional sanctions, they offered a “carrot” to Turkey, consisting in “a positive political agenda” including “the modernisation of the Customs Union and trade facilitation, people to people contacts, High level dialogues, continued cooperation on migration issues”. EU leaders also called for the organisation of a “Multilateral Conference on the Eastern Mediterranean” to discuss issues such as “maritime delimitation, security, energy, migration and economic cooperation”27. At this occasion, the President of the Commission, Ursula von der Leyen, nonetheless emphasised that EU leaders “expect[ed] that Turkey from now on [would abstain] from unilateral actions” and that “[i]n case of such renewed actions by Ankara the EU [would] use all its instruments and options available”28. EU leaders 23 Joseph de Weck, ‘Pariscope: Hegelian Diplomacy in the Eastern Mediterranean’, Internationale Politik, 2 September 2020, https://internationalepolitik.de/en/pariscope-hegelian-diplomacy-easternmediterranean. 24 Note, however, that the Italian destroyer Luigi Durand de La Penne, which participated in the QUAD-EUNOMIA 2020 deployment alongside Greek, Cypriot and French forces, had just returned from a training with Turkish frigates south of Cyprus, as reported on the Twitter account of the Marina Militare (see in particular: https://twitter.com/ItalianNavy/status/1298613863072567296?s=20). The political message sent by Italy on that occasion was thus not as straightforward as some might have interpreted it. 25 Council of the European Union, Council Regulation (EU) 2019/1890 of 11 November 2019 concerning restrictive measures in view of Turkey’s unauthorised drilling activities in the Eastern Mediterranean. In November 2020, the applicability of this framework was prolonged for another year (see: Council of the European Union, Council Decision (CFSP) 2020/1657 of 6 November 2020 amending Decision (CFSP) 2019/1894 concerning restrictive measures in view of Turkey’s unauthorised drilling activities in the Eastern Mediterranean). 26 Council of the European Union, Council decision (CFSP) 2020/275 of 27 February 2020 amending Decision (CFSP) 2019/1894 concerning restrictive measures in view of Turkey’s unauthorised drilling activities in the Eastern Mediterranean. 27 European Council, Special meeting of the European Council (1 and 2 October 2020) – Conclusions, EUCO 13/20, 2 October 2020. 28 Opening remarks by President von der Leyen at the joint press conference with President Michel following the Special European Council meeting, Brussels, 1 October 2020. 8 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean also agreed to review their position at their next meeting, making it clear that they would not hesitate to resort to sanctions “in case of renewed unilateral actions or provocations in breach of international law” by Turkey29. But relations with Ankara did not immediately improve – quite the contrary. As noted in the introduction, Turkey resumed its prospecting activities from mid-October onwards in the Eastern Mediterranean, thereby ostensibly thumbing its nose at the red line drawn by the European Council. In fact, the whole relationship between the EU and Turkey was pushed to a low point during the autumn of 2020, due in particular to repeated provocations by Turkish officials, such as President Erdoğan’s insults about French President Emmanuel Macron’s mental sanity30 or his call to boycott French products due to the French government’s allegedly discriminatory attitude towards Muslims31. Despite these negative developments, the European Council nonetheless decided to adopt only very limited additional sanctions when it reconvened two months later in December 2020. Indeed, due to the persistent opposition of some EU member states (Germany and Bulgaria in particular), the Union only agreed to add a few names to the existing list of Turkish individuals sanctioned for their key role in the unauthorised drilling operations off the coast of Cyprus32. The Union did not agree, however, to draw up fresh or broader sanctions on a new legal basis. Rather, the EU reiterated its offer of “a positive EU-Turkey agenda” and indicated its intention to continue to support Ankara on migration issues, including through financial assistance33. In truth, EU leaders considered it more prudent on this occasion to delay any major decisions regarding Turkey in order to be able to coordinate their approach regarding Turkey with that of the Biden administration from January 2021 onwards. Contrary to what happened throughout the autumn of 2020, however, tensions with Ankara finally decreased in the first months of 2021 with the halt of Turkish gas exploration in the Eastern Mediterranean and the resumption of exploratory talks with Greece. This positive evolution enabled the European Council to assert in March 2021 that the EU was ready to “enhance cooperation [with Ankara] in a number of areas of common interest”, in particular by improving economic cooperation (modernisation of the customs union), launching “high level dialogues with Turkey on issues of mutual concern, such as public health, climate and counter-terrorism as well as regional issues”, and exploring “how to strengthen cooperation with Turkey on people to people contacts and mobility”34. European leaders cautioned, however, that the EU would re-engage with Turkey “in a phased, proportionate and reversi- 29 European Council, Special meeting of the European Council (1 and 2 October 2020) – Conclusions, EUCO 13/20, 2 October 2020. 30 Patrick Wintour, “France recalls ambassador to Turkey after Erdoğan questions Macron’s mental state”, The Guardian, 25 October 2020. 31 William Adkins, “Erdoğan calls on Turks to boycott French goods”, Politico, 26 October 2020. 32 Sarantis Michalopoulos, “Merkel and Borissov blocked EU sanctions against Turkey at summit: sources”, Euractiv.com, 11 December 2020. 33 European Council, European Council meeting (10 and 11 December 2020) – Conclusions, EUCO 22/20, 11 December 2020. 34 European Council, Statement of the Members of the European Council, SN 18/21, 25 March 2021, p. 6. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 9 Solidarity and deterrence in the Eastern Mediterranean ble manner”, calling once again the country “to abstain from renewed provocations or unilateral actions in breach of international law”35. In this regard, there is a wide range of tools with which the EU could impose political and economic costs on Ankara in the event of renewed tensions. In a joint report “on the state of play concerning the EU-Turkey political, economic and trade relations and on instruments and options on how to proceed” presented in preparation of the European Council last March, the HR/VP and the European Commission listed, for instance, five measures that the EU could implement in crescendo. These are: 1) adopting the additional listings already agreed in December 2020; 2) “enhancing” the economic sanctions undertaken within the current framework for sanctions, including with the possibility of applying sanctions not only to individuals (as has been the case thus far) but also to legal persons (i.e. Turkish companies); 3) restricting EU-Turkey economic cooperation, such as lending from the European Investment Bank (EIB)36; 4) targeting other sectors of the Turkish economy like tourism (which is particularly dependent on European travellers) through administrative measures such as a ban on providing tourism services or the publication of negative travel advice; 5) extending EU sanctions to “the energy and related sectors energy and related sectors, such as import/export bans on certain goods and technologies”37. Beyond these measures, which could be put into effect rather quickly, the EU could also (as mentioned in the introduction) impose a comprehensive and legally binding arms embargo on Ankara38, or suspend the EUTurkey customs union (or, at the very least, suspend its planned modernisation)39. Another option, reportedly suggested by member states like Austria and France, would be to terminate Turkey’s EU accession negotiations40 – a process that has been effectively frozen for 35 Ibid. 36 Turkey’s access to financing from the EIB (which used to be Ankara’s largest single lender) has, in fact, already been severely restricted since 2019 due to its drillings off the coast of Cyprus (see: Marc Jones, “EIB to halt Turkey government-linked lending till year-end”, Reuters, 24 July 2019; Marc Jones, “EU lending arm EIB set to keep tight Turkey restrictions in place”, Reuters, 21 January 2020). Thus, “no loans to Turkey were signed in 2020” by the EIB, as the HR/VP and the Commission observed in their joint report of March 2021 (see: European Commission & High Representative, Joint communication to the European Council: State of play of EU-Turkey political, economic and trade relations, JOIN(2021) 8 final/2, 22 March 2021, p.12). If maintained, these restrictions could affect in particular the ability of the Turkish Petroleum Pipeline Corporation (BOTAS) to exploit the TransAnatolian Natural Gas Pipeline (which carries gas from Azerbaijan through Turkey and on to Europe via the Trans-Adriatic Pipeline), thus complicating the realisation of Turkey’s geopolitical ambition of becoming a regional energy hub (see: David O’Byrne, “EU applauds Turkey’s more ‘constructive’ attitude to East Med relations”, S&P Global Platts, 23 March 2021). 37 State of play of EU-Turkey political, economic and trade relations, Ibid., pp.14-15. 38 Robin Emmott, “EU to discuss arms exports to Turkey with NATO and U.S., Merkel says”, Reuters, 11 December 2020. EU member states already committed in October 2019 to adopt “strong national positions regarding their arms export policy to Turkey” over its offensive in northern Syria. But this political agreement did not amount to a complete, EU-wide embargo on weapons sales (see: Council of the European Union, Outcome of the Council meeting – 3720th Council meeting Foreign Affairs, 13066/19, 14 October 2019, p. 4. For an analysis of this decision, see: Aurélie Pugnet, “Les 28 tombent d’accord pour sanctionner la Turquie pour deux motifs. Le Royaume-Uni résiste », Bruxelles2, 14 October 2020. 39 Branislev Stanicek, “EU-Turkey customs union: Modernisation or suspension? », Briefing, European Parliamentary Research Service, December 2020. 40 Sarantis Michalopoulos, “EU gives Turkey one month as pressure for sanctions mounts”, Euractiv.com, 14 July 2020. 10 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean several years anyway, especially since the 2016 coup attempt and its extensive repression by the Turkish authorities41. 1.3. Greece’s strategic rationale for raising the possibility of triggering article 42.7 TEU: alarming friends and foes alike None of the measures just mentioned would represent, however, the last “rung” on the EU’s escalation ladder if tensions were to peak again with Turkey42. In the event of a particularly serious escalation, the possibility of activating the EU collective defence clause, as enshrined in article 42.7 TEU, could indeed eventually arise. This provision was introduced by the Lisbon Treaty in 2009 and has so far been invoked only once, by France, in response to terrorist attacks in and around Paris in November 2015. The prospect that Greece may have to resort as well to article 42.7 TEU surfaced as early as mid-July 2020, as tensions started to brew in the Eastern Mediterranean. Indeed, according to press reports, this clause was mentioned by Greek Foreign minister Nikos Dendias during a session of the Council of the EU on 13 July 202043. A day after this meeting, the minister confirmed on Greek television that he had stressed in front of his European counterparts that his country would not hesitate to trigger the EU collective defence clause against Turkey if push came to shove44. Nikos Dendias went on to explain that, in his eyes, article 42.7 TEU represented “the European equivalent of article 5 of the NATO Treaty [the collective defence clause of the Washington Treaty]. An article that, through its existence, is actual proof of European solidarity”45. Therefore, when the Greek minister for Foreign affairs explicitly referred to article 42.7 TEU in his letter to the EU’s High Representative in October 2020, he was in substance repeating himself, although this time in writing and, therefore, with greater solemnity. Given the wording of the letter addressed to Josep Borrell, which, to repeat, asked the EU “to stick to its principles – especially on internal solidarity and mutual assistance between its Members, as enshrined in EU Treaties, including article 42.7 TEU –”, it seems that the Greek government’s intention was not to immediately call for the activation of the EU collective defence clause against Turkey46. It is more likely, instead, that the Greek authorities, disappointed by a perceived lack of support from the rest of the Union, have wished to recall to their EU partners that they all have an obligation of mutual solidarity vis-à-vis the outside 41 In response to a question from a member of the European Parliament, the European Commission indicated in March 2020 that Turkey would not receive any pre-accession financial assistance under the EU’s Multiannual Financial Framework for 2021-2027 (see: European Commission, Answer given by Mr Várhelyi on behalf of the European Commission, E-000168/2020, 17 March 2020). 42 See: Herman Kahn, On Escalation. Metaphors and Scenarios, New York: Praeger, 1965. 43 “EU gives Turkey one month as pressure for sanctions mounts”, op. cit. 44 Nikos Dendias, Interview on STAR TV’s evening news with journalist Mara Zacharea, Greek Ministry of Foreign Affairs, 14 July 2020. 45 Ibid. 46 See “Face à la Turquie, la Grèce invoque une ‘vraie’ solidarité…”, op. cit. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 11 Solidarity and deterrence in the Eastern Mediterranean world, both on a permanent basis47 as well as upon express request in exceptional circumstances – as per article 42.7 TEU indeed48. From a strategic perspective, the letter of the Greek Foreign affairs minister can thus be viewed as a twofold warning, addressed to friends and potential foes alike49. On the one hand, the explicit mention of article 42.7 TEU was used by Greece to try to garner the diplomatic support of its EU partners by making them face up to their eventual responsibilities in case of escalation in the Eastern Mediterranean. It is likely, in other terms, that the Greek government tacitly played on the fear of entanglement of its EU partners, believing that if the latter were informed that a confrontation in the region would potentially lead to an activation of the EU collective defence clause, they would be much less inclined to stand aside and let the situation deteriorate in the first place. On the other hand, this letter – which was conveniently revealed to the press and, therefore, to the general public as well – was also a message meant to be indirectly heard in Ankara, in order to dissuade the Turkish authorities from pushing the confrontation too far. The unambiguous reference to article 42.7 TEU indeed implied that Athens expected (and, in all likelihood, still expects) that the rest of the EU would come to its aid in the event of armed aggression on the part of Turkey and, therefore, that the balance of power – and military power in particular – was in reality not as favourable to Ankara as it might appear on a purely bilateral level50. 47 Articles 24.2, 24.3 and 32 (TEU). 48 Another provision, article 222 of the Treaty on the Functioning of the European Union (TFEU), provides for European solidarity in case of terrorist attack or of natural or man-made disaster. 49 This intent was especially transparent in the July 2020 interview in which Nikos Dendias declared: “To be absolutely clear with our friends, our family, the European Union, and with anyone else – Turkey, in this case – I warned that Greece will request measures based on article 42 if it is attacked by anyone” (see: Nikos Dendias, Interview on STAR TV’s evening news, op. cit.; emphasis added). 50 According to the latest SIPRI estimates, Turkey spent $20.4 billion on its military in 2019, four times more than Greece ($5.5 billion) over the same period. As a whole, however, the EU-27 spent that year $219.6 billion on defence, almost eleven times more than Ankara alone (see: “Military expenditure by country, in constant (2018) US$ m., 1988-2019”, Stockholm International Peace Research Institute, 2020). The amount of defence spending is, to be sure, only one part of the story but it may still give a rough estimate of the balance existing in that regard between countries or groups of countries. 12 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean 2. Activating article 42.7 TEU in the Eastern Mediterranean: under which conditions? There is little doubt that Greece’s repeated references to article 42.7 TEU should be taken seriously by other EU member states, especially if tensions were to run high again with Turkey in the months or years to come. At the same time, it must be recalled here that the EU security guarantee does not constitute a blank cheque and functions, to the contrary, within well-defined legal limits. To some, this may sound like an overly legalistic point to make. After all, one might argue, political considerations, not legal ones, would surely be decisive if the EU collective defence clause were to be invoked (and all the more so if this were to happen against a NATO ally such as Turkey!). Yet, it is also fair to assume that law and politics could not operate entirely at cross-purposes: even if ultimately driven by political motives, a request (or, for that matter, a refusal) to invoke article 42.7 TEU would need in other words to be backed up by solid legal arguments at the very least. It remains important, in consequence, to understand what the legal requirements would be for triggering the EU collective defence mechanism in the Eastern Mediterranean. In that perspective, two questions are successively examined in the following section: first, would it be possible, at least as a matter of principle, to invoke the EU collective defence clause against a NATO ally (2.1)? And, second, how could the terms of article 42.7 TEU come into play in the specific context of the Eastern Mediterranean (2.2)? 2.1. Admissibility: would it be formally possible to invoke the EU collective defence clause against a NATO ally? The question of whether an intra-EU security mechanism could be used against a NATO ally, while uncomfortable from a political standpoint, is however not entirely new. Indeed, an analogous question arose regarding WEU-NATO relations during the early 1990s as well as when the EU security and defence policy was launched in the early 2000s (2.1.1). Yet, it is unlikely that the diplomatic solutions found at the time (2.1.2) or the “NATO caveat” directly enshrined in article 42.7 TEU (2.1.3) could formally prevent the invocation of this provision against a NATO ally if such a scenario were to effectively arise. 2.1.1. Historical background: the WEU’s Petersberg Declaration and the European Council’s 2002 assurances on EU-NATO relations The preparations for Greece’s accession as a full member of the Western European Union (WEU) in the early 1990s posed a problem similar, mutatis mutandis, to the one discussed in this report51. This process implied indeed that Athens would benefit by definition from the date of its accession of the protection conferred by the WEU’s collective defence guarantee, 51 During the Cold War, the WEU largely remained a dormant military organisation, in the shadow of NATO. After the Maastricht Treaty (1992), which established a Common Foreign and Security Policy (CFSP) and envisaged the eventual framing of a common defence policy, the WEU was repurposed to serve as the “defence arm” of the EU, while working as well in close cooperation with the Atlantic Alliance. It is in this context that the EU member states that did not already belong to the WEU were encouraged to participate in its activities, either as full or as associate members. Only full members were formally bound, however, by the WEU’s collective defence commitment. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 13 Solidarity and deterrence in the Eastern Mediterranean as it used to be encapsulated under article V of the Modified Brussels Treaty (MBT)52. Already back then, however, the tumultuous relationship between the Aegean neighbours proved to be a stumbling block in that regard, as WEU members feared that Greece would invoke the WEU’s security guarantee in the event of a future confrontation with Turkey53. To prevent such a situation from ever arising, WEU members hence explicitly agreed in the Petersberg Declaration of June 1992 that: the security guarantees and defence commitments in the Treaties which bind the member States within Western European Union and which bind them within the Atlantic Alliance are mutually reinforcing and will not be invoked […] in disputes between member States of either of the two organizations54. Furthermore, in order to leave absolutely no room for ambiguity, this section of the Petersberg Declaration was specifically referred to in the accession protocol that Greece had to sign five months later to join the WEU55. Thus, at the cost of “slightly undermining the substance” of article V of the MBT (as one observer put it), the principle that the respective collective defence obligations of the WEU and NATO would not be used against each other was clearly established56. When the European Security and Defence Policy (ESDP) was set in motion in the early 2000s, Turkey sought to replicate the existing WEU-NATO guarantees, or at least to ensure that the new defence tools then being developed under the EU framework (often, in fact, as a result of direct transfers from the WEU) would not threaten its security either. A political agreement was brokered in this perspective in December 2001 by the United Kingdom and the United States with the government of Turkey and couched in a text known as the “Ankara Document”57. The gist of this deal regarding EU-NATO cooperation was then approved in the presidency conclusions of the European Council at the Brussels summit of October 200258. 52 Article V of the Modified Brussels Treaty (1954) provided that: If any of the High Contracting Parties should be the object of an armed attack in Europe, the other High Contracting Parties will, in accordance with the provisions of Article 51 of the Charter of the United Nations, afford the Party so attacked all the military and other aid and assistance in their power. 53 Franklin Dehousse, Benoît Galer, “De Saint-Malo à Feira : les enjeux de la renaissance du projet de défense européenne”, Studia Diplomatica 52(4), p. 40. 54 WEU, Petersberg Declaration made by the Council of Ministers, Bonn, 19 June 1992, Part III, Section A. 55 WEU, Protocol for the accession of Greece, Rome, 20 November 1992. 56 See: Cédric Sangaletti, Interview of Charles Goerens, 28 September 2009, p. 9. Charles Goerens, a Luxembourgish politician, served as president of the WEU’s Parliamentary Assembly from 1987 to 1990. 57 On this point, see: Antonio Missiroli, “EU-NATO Cooperation in Crisis Management: No Turkish Delight for ESDP”, Security Dialogue, 33(1), 2002, pp. 9-26. 58 At the insistence of Greece, these conclusions were made to refer to all non-EU European allies and not just to Turkey (Ibid., p. 21). 14 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean These conclusions provided in particular that: 1. The Treaty on European Union states (Article 17(1)) [in the current version of the TEU, as last amended in 2009 by the Lisbon Treaty, this corresponds to the second paragraph of article 42.2 TEU] : “The policy of the Union in accordance with this Article shall not prejudice the specific character of the security and defence policy of certain Member States and shall respect the obligations of certain Member States, which see their common defence realised in the North Atlantic Treaty Organisation (NATO), under the North Atlantic Treaty and be compatible with the common security and defence policy established within that framework.” 2. For the Member States concerned, this means that the actions and decisions they undertake within the framework of EU military crisis management will respect at all times all their Treaty obligations as NATO allies. This also means that under no circumstances, nor in any crisis, will ESDP be used against an Ally, on the understanding, reciprocally, that NATO military crisis management will not undertake any action against the EU or its Member States. It is also understood that no action will be undertaken that would violate the principles of the Charter of the United Nations59. 2.1.2. Assessing the consequences (or lack thereof) of the Petersberg Declaration and the European Council’s 2002 assurances with regard to article 42.7 TEU Now, could either of the two texts mentioned above apply to the case of article 42.7 TEU and prevent it from being invoked against a NATO ally? Let us consider each case in turn. Article 42.7 TEU is often presented as the “successor” to the WEU’s collective defence clause. The reason for this is straightforward: the members of the WEU explicitly referred to the entry into force with the Lisbon Treaty to justify the dissolution of the organisation, indicating that the creation of an EU collective defence clause signified that they remained “strongly committed to the principle of mutual defence” that was enshrined in article V MBT60. A question could therefore arise as to whether the principle of non-invocation against a NATO ally that was established with the Petersberg Declaration of 1992 could have, in a way, been “transferred” to article 42.7 TEU. The answer to this question must be, however, clearly negative. Indeed, the link between the WEU’s and the EU’s respective security guarantees, while historically and politically indisputable, does not entail as such any legal consequence. As the Legal Service of the Council of the EU noted, article 42.7 TEU constitutes “a distinct and autonomous legal provision, applicable to all EU Member States, which must be interpreted and applied in its own context within the overall framework of the EU Treaties”61. In other words, the EU collective defence clause cannot be considered as being bound by interpretations pertaining to a different legal order, such as those that used to be attached to the WEU’s security guarantee. 59 European Council, Presidency Conclusions of the European Council on 24 and 25 October 2002 in Brussels, 26 November 2002, Annex II: ‘ESDP: Implementation of the Nice provisions on the involvement of the non-EU European allies’. Italics in the original. 60 WEU, Statement by the Presidency of the WEU Permanent Council on the termination of the Brussels Treaty, Brussels, 31 March 2010. 61 Council of the European Union, Opinion of the Legal Service: Article 42(7) TEU, 11176/16, 12 July 2016, p. 5. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 15 Solidarity and deterrence in the Eastern Mediterranean Let us thus turn to the European Council in 2002 on EU-NATO relations. It is not altogether certain, first of all, that these could directly apply to article 42.7 TEU. It could be argued indeed that these assurances apply only to the “crisis management” dimension of the EU defence policy and not, strictly speaking, to “collective defence”, if one refers both to the exact wording of these assurances or to the fact that the latter preceded the introduction of a collective defence clause in the EU treaties by several years62 . However, even if we reject these arguments, which may be more or less convincing, and accept that the 2002 assurances can indeed apply to the case of article 42.7 TEU, these assurances could not really prevent in any case the invocation of the EU collective defence clause against a NATO country. The precondition that must be fulfilled to activate the EU collective defence clause – its casus foederis in legal parlance – is, indeed, that an EU member state must have suffered an “armed aggression on its territory” (more on this in 2.2). Yet, if such armed aggression were to be perpetrated by a NATO ally against an EU member state, this would undoubtedly constitute a violation of the principles of the UN Charter, in particular of the general prohibition on the use or threat of force in relations between states63. But this, in turn, would violate one of the conditions on which the European Council’s assurances were given in 2002, as can be deduced from the last sentence in the excerpt cited above. Therefore, because activating the EU collective defence clause necessarily presupposes a violation of the UN Charter, there exists by definition no situation in which such activation could be refused on the grounds of the 2002 assurances: either there is not a sufficient reason to activate article 42.7 TEU and the assurances that were given by the European Council remain valid, or it is possible to activate this provision because an armed aggression has effectively been committed against an EU member state by a NATO ally, in which case these assurances would no longer apply. 2.1.3. Interpreting the “NATO caveat” directly enshrined in the EU collective defence clause In consequence, we need to turn to another and, in fact, more obvious question: can it be inferred directly from the text of the EU collective defence clause that it would be impossible to invoke this clause against a NATO ally? The second paragraph of article 42.7 TEU includes indeed an important caveat regarding the Atlantic Alliance, providing that: Commitments and cooperation in this area [i.e. collective defence] shall be consistent with commitments under the North Atlantic Treaty Organisation, which, for those States which are members of it, remains the foundation of their collective defence and the forum for its implementation. Although this caveat is quite similar to ex-article 17.1 of the TEU (now article 42.2 TEU, second paragraph), which, as we have just seen, was the basis on which the assurances of the 62 For a remark in this sense, see: Willem van Eekelen, From Words to Deeds. The Continuing Debate on European Security, Centre for European Policy Studies, 2006, p. 67. 63 As provided under article 2.4 of the UN Charter. Of course, not all uses of force are necessarily unlawful under international law, nor do they all automatically reach the level of armed aggression (see 2.2.1). However, all armed aggressions constitute by definition unlawful uses of force under the UN Charter. 16 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean European Council were given in 2002, its exact implications have never been so clearly drawn. The purpose of the second paragraph of article 42.7 TEU seems to be mainly to reco-gnise the primacy of NATO for those EU countries that are also parties to the Washington Treaty to avoid unnecessary duplication in terms of military planning in cases where the mandates of the two organisations overlap in the domain of collective defence (which would be the case in many, but not all, situations)64. It is important to understand, however, that this remains an issue separate from whether it would be permissible (or not) to trigger the EU collective defence clause in the eventuality of armed aggression committed by a member of NATO against an EU member state. It has to be recognised in this respect that nothing in the second paragraph of article 42.7 TEU explicitly implies that it would be impossible to resort to this clause in such a scenario. It should be noted, moreover, that no official interpretation to that effect has been produced to date concerning this particular paragraph (at least to the author’s knowledge). Therefore, in the absence of any express and official statement to the contrary, it remains more prudent to consider that article 42.7 TEU could indeed be invoked in the case of armed aggression against an EU member state, even if the latter were to be perpetrated by a NATO country. 2.2. Casus foederis: how could the terms of article 42.7 TEU come into play in the Eastern Mediterranean? If one accepts the above conclusion, this still does not mean, however, that the EU collective defence clause could be triggered at will in the Eastern Mediterranean. Indeed, according to the wording of article 42.7 TEU, two conditions would still need to be satisfied to activate this provision: firstly, an EU member state would need to have been the victim of an “armed aggression” (2.2.1) and, secondly, this armed aggression would need to have taken place on the “territory” of that member state (2.2.2). In what follows, we will examine how these two criteria could prove problematic in the specific context of the Eastern Mediterranean. 2.2.1. Reaching a sufficient degree of gravity: the threshold of “armed aggression” To determine, first of all, whether a given EU member state has been the victim of an “armed aggression”, two elements must be considered: firstly, whether the attacks directed against this member state originated from abroad and, secondly, whether these attacks reached a sufficient level of gravity. The first of these two criteria does not really represent an issue in the case at hand, given that the dispute in the Eastern Mediterranean involves 64 For an analysis of the overlaps, or lack thereof, between articles V of the Washington Treaty and 42.7 TEU (as well as 222 TFEU), see for instance: Elie Perot, “The art of commitments: NATO, the EU, and the interplay between law and politics within Europe’s collective defence architecture”, European Security, 28(1), 2019, pp. 40-65). For a similar discussion, with a special focus on “hybrid threats”, see: Aurel Sari, “The Mutual Assistance Clauses of the North Atlantic and EU Treaties: The Challenge of Hybrid Threats”, Harvard National Security Journal, 10(2), 2019, pp. 405-460. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 17 Solidarity and deterrence in the Eastern Mediterranean well-identified state actors and their armed forces65. We will therefore set aside here this aspect of the discussion and focus exclusively on the second criterion – the question of the threshold for armed aggression. Because the EU collective defence clause derives from article 51 of the UN Charter66, international law may provide some guidance on the type of actions that would be sufficiently serious to trigger this provision67.Thus, in its famous Nicaragua case in 1986, the International Court of Justice (ICJ) found it important to distinguish “the most grave forms of the use of force” – which constitute an armed aggression – from “other less grave forms”68. But the ICJ did not provide on that occasion a very sharp criterion for how to make this distinction in other cases, indicating only by way of example that a “mere frontier incident” may qualify as a use of force under the UN Charter but not necessarily as an armed aggression69. Doubts, in consequence, remained regarding the precise extent of the gap that may exist in international law between a minimal use of force and an armed aggression. In its Oil Platforms judgment in 2003, the ICJ suggested, however, that this gap was in reality quite narrow, as the Court refused in that instance to “exclude the possibility that the mining of a single military vessel might be sufficient to bring into play the ‘inherent right of selfdefence’” 70. Of course, in the specific case of article 42.7 TEU, one may point out that it would be up, in any case, to the Council of the EU to accept a request to activate this provision and, in consequence, that the threshold for armed aggression will ultimately be set in reality whe-rever EU member states themselves will deem it fit71. This realist argument, while valid, should not 65 This point was however of special importance when the decision was taken to invoke article V of the Washington Treaty in 2001. The activation of this provision was indeed fully confirmed only once the fact had been clearly established that the 9/11 terrorist attacks had been directed from abroad (see: NATO, “Statement by NATO Secretary General, Lord Robertson”, 2 October 2001). 66 Article 51 of the UN Charter provides in full that: Nothing in the present Charter shall impair the inherent right of individual or collective self-defence if an armed attack occurs against a Member of the United Nations, until the Security Council has taken measures necessary to maintain international peace and security. Measures taken by Members in the exercise of this right of self-defence shall be immediately reported to the Security Council and shall not in any way affect the authority and responsibility of the Security Council under the present Charter to take at any time such action as it deems necessary in order to maintain or restore international peace and security. 67 The UN Charter uses the term, more exactly, of “armed attack”. But for the sake of simplicity this paper uniformly uses the expression of “armed aggression” employed in article 42.7 TEU. Both expressions should be considered nonetheless as functional equivalents from the standpoint of international law (see the author’s explanation on this point in “The art of commitments…”, op. cit., p. 45). 68 ICJ, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), 1986, §191. 69 Ibid., §195. 70 ICJ, Oil Platforms (Islamic Republic of Iran v. United States of America), 2003, §72. 71 As happened in November 2015, when France invoked for the first time article 42.7 of theTEU (see: Council of the European Union, Outcome of the Council Meeting, 14120/15, 16-17 November 2015, p. 6; EEAS, Remarques introductives de la Haute Représentante et Vice-Présidente Federica Mogherini lors de la conférence de presse avec Jean-Yves Le Drian, Ministre de la Défense français, 17 November 2015). Because of the momentous implications of the decision to activate the EU collective defence clause, it is extremely likely, however, that this decision would need to be preliminarily vetted in practice at the highest political level, that is, by EU heads of state and government (on this point, see: Alexander Mattelaer, Audition in front of the Sub-Committee on Security and Defence of the 18 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean be exaggerated, however. It is very unlikely, indeed, that EU countries would agree to view as an armed aggression some acts that would not involve any use or threat of force at all or, on the contrary, that they would refuse to recognise an armed aggression where one would have manifestly taken place against a fellow member state. Hence, it is mostly for “borderline” cases that EU member states would enjoy indeed real discretion in deciding whether or not the threshold of armed aggression has been crossed. In any event, such a question has not really arisen yet in the case at hand. It would be very hard indeed to argue at present that Turkey’s actions in the Eastern Mediterranean have been severe enough to be considered as approaching the threshold for armed aggression, let alone exceeding it, under international law. True, a relatively serious accident occurred in mid-August 2020 when a Greek and a Turkish frigates were involved in a collision (without causing any casualty on either side)72. Also, the fact that a military escort consisting of several warships has been constantly protecting the Oruç Reis during its contested seismic surveys may be viewed as a form of intimidation on the part of Turkey. Nonetheless, one could not reasonably qualify these events, individually or even collectively, as an “armed aggression” without stretching this notion to the implausible73. 2.2.2. The geographical scope of article 42.7 TEU and the Eastern Mediterranean: possible interplays and issues Now, if an armed aggression were to be effectively perpetrated against an EU member state, this fact, on its own, would still not be sufficient to trigger article 42.7 TEU. As stated above, the armed aggression in question would also need to have taken place on the “territory” of this member state. This second criterion could seem less problematic than the previous one since the territorial boundaries of EU member states are in general well-delineated and hence much less susceptible to interpretation. Nevertheless, given the special geographical features of the Eastern Mediterranean theatre, it is important to underscore here the two following points: first, that article 42.7 covers in reality only a fraction of the different maritime zones that can fall under the jurisdiction of EU member states; and, second, that the complex and contested nature of Greece’s boundaries in the Aegean Sea could raise some issues regarding the territorial applicability of the EU collective defence clause in case of conflict in this particular region. Let us start with the general question of the limited applicability of the EU collective defence clause in the maritime domain. By disassembling the territory of a state into its constituent elements under international law, one can first deduce that article 42.7 TEU covers more specifically EU member states’ land areas, internal and territorial waters as well as the na- European Parliament, 1 December 2015, quoted in: Suzana Elena Anghel, Cirlig Carmen-Cristina, Activation of Article 42(7) TEU. France's request for assistance and Member States’ responses, European Parliamentary Research Service, July 2016, p. 3. 72 Michele Kambas, Tuvan Gumrukcu, “Greek, Turkish warships in ‘mini collision’ Ankara calls provocative”, Reuters, 14 August 2020. 73 This is not to say, of course, that Ankara’s neighbours have no good reasons to regard Turkish gas and oil drilling operations as “unfriendly”, or even “hostile”, and to denounce them as incompatible with international law. Nor does it mean that one can definitively rule out the possibility that further Turkish actions in the Eastern Mediterranean might rise to the level of armed aggression in the future. But the fact is, quite simply, that such an assessment does not seem warranted at the moment of this writing. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 19 Solidarity and deterrence in the Eastern Mediterranean tional airspace over each of them74. However, if we then refer to the different maritime zones defined under the United Nations Convention on the Law of the Sea (UNCLOS)75, this implies, by contrast, that article 42.7 TEU would not apply if an armed aggression against an EU member state were to take place on or over its EEZ/continental shelf or on or over the high seas76. Accordingly, in the event of such a contingency, the other member states would be legally correct to stress the geographical limits of article 42.7 TEU and refuse its activation77. Needless to say that this could represent a major issue in the case at hand since the maritime confrontation with Turkey has unfolded so far mostly in areas that are considered by Greece or Cyprus as belonging to their respective EEZs/continental shelves, but not in their territorial waters78. The second set of difficulties concerning the territorial applicability of the EU collective defence clause is more specific to the case at hand, as it arises from the existing disagreements between Greece and Turkey over the delimitation of Athens’ territorial waters and national airspace in the region of the Aegean Sea79. 74 See: Robert Jennings, Arthur Watts, “Part 2 The objects of international law, Ch.5 State territory, The Different Parts of State Territory”, in Oppenheim's International Law: Volume 1 Peace (9th edition), Oxford University Press, 2008. In theory, one could add archipelagic waters to this list but no EU member state has claimed the status of archipelagic state under international law. 75 One of the main features of the UNCLOS, which entered into force in 1994 and has been ratified by the vast majority of countries in the world, is the distinction it makes between different maritime zones, notably the territorial sea, the EEZ, the continental shelf and the high seas. Thus, according to the UNCLOS, a coastal state has the right to exercise full sovereignty over its territorial waters up to a maximum of 12 nautical miles from its shores, to the exception that foreign vessels must be granted the right of “innocent passage” through these waters. Beyond the territorial sea lies the EEZ, which may extend up to 200 nautical miles. The coastal state has sovereign rights in terms of exploration and economic exploitation in its EEZ but must allow the free navigation in and overflight of this area by foreign ships and aircraft. The continental shelf, for its part, corresponds to the seabed beyond a coastal state’s territorial waters. The continental shelf may stretch up to 200 nautical miles or more under specific circumstances and grants exclusive rights to the exploitation of the resources it contains to the coastal state. Finally, the high seas designate the remaining part of the seas and oceans that extends beyond the different areas just described. 76 The author is not aware of any publication that has yet explicitly stressed this point. However, by reading between the lines, this point can be inferred from the following report: Tiia Lohela, Valentin Schatz (eds.), Handbook on Maritime Hybrid Threats — 10 Scenarios and Legal Scans, European Centre of Excellence for Countering Hybrid Threats (Hybrid CoE), November 2019, pp. 23-24. 77 In that regard, it may be interesting to note that an interaction of this kind happened during the recent conflict between Azerbaijan and Armenia over Nagorno-Karabakh (fall 2020). When pressed by Yerevan to intervene on its side, Russia underscored the territorial limitations of the security guarantee provided under the Collective Security Treaty, declaring that it would actively support Armenia under the terms of this treaty if and only if hostilities were to directly affect the latter’s internationallyrecognised territory (see: “Russia has security obligations to Armenia, not Nagorno-Karabakh – Kremlin”, TASS, 7 October 2020). 78 In recent months, Turkish ships came very close to Greece’s territorial waters but have also clearly refrained from stepping into them (see e.g.: Vassilis Nedos, “Oruc Reis reaches up to 7 miles off Kastellorizo”, Ekathimerini, 27 November 2020). 79 For an overview of the different dimensions of the “Aegean dispute”, of which the disagreements over Greece's territorial waters and airspace are only one part, see for example: Alexis Heraclides, The Greek-Turkish Conflict in the Aegean. Imagined Enemies, Palgrave Macmillan, 2010. The respective Turkish and Greek positions regarding the Aegean dispute are also succinctly exposed on the websites of the foreign affairs ministries of both countries. See the following webpages and related 20 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean In that perspective, let us consider first the dispute regarding Greek territorial waters. This dispute concerns not the current demarcation of these waters but rather the right that Greece theoretically enjoys under the UNCLOS to unilaterally extend their width, set for the most part at 6 nautical miles (nm) from the Greek coastline, up to a maximum limit of 12nm80. Yet, unlike most countries that have ratified the UNCLOS, Greece has refrained for decades from exercising this right81. The reason for this restraint is that Turkey has repeatedly stressed that it would consider such an extension in the region of the Aegean Sea as a casus belli82. Ankara, which is not a party to the UNCLOS and therefore does not consider itself bound by its provisions83, argues that expanding Greek territorial waters up to 12 nm would unduly close the Aegean Sea to Turkish military, economic or scientific activities, constrain coastal navigation around Anatolia and reduce the free access of Turkish ships to the open sea, especially for vessels coming from the Turkish Straits84. Greece’s traditionally cautious stance regarding the possibility to extend its territorial waters has nonetheless recently started to evolve. Faced with mounting tensions on its eastern border with Turkey, the current Greek government led by Prime Minister Kyriakos Mitsotakis has aimed over recent months to quickly resolve the outstanding maritime disputes that Greece had with its other neighbours. Bilateral agreements over the demarcation of bordering EEZs were thus reached with Italy in June 2020 and with Egypt in August, while in October the Greek and Albanian governments agreed to refer their maritime dispute to the ICJ. These developments have, in turn, emboldened Greece to finally assert its rights under the UNCLOS regarding the width of its territorial waters. As a result, Prime Minister Mitsotakis announced in August 2020 the intention of his government to expand the country’s territorial waters from 6 to 12 nm along its western coast, in the Ionian Sea85. This measure, which represents Greece’s first territorial change since the incorporation of the Dodecanese islands hyperlinks: Turkish Ministry of Foreign Affairs, “Turkish-Greek Relations/Aegean Problems”; Greek Ministry of Foreign Affairs, “Issues of Greek-Turkish Relations”. 80 Article 3 UNCLOS. 81 Greece ratified the UNCLOS in 1995. 82 This was made clear in particular in a resolution voted by the Turkish Grand National Assembly in June 1995. Greece has complained, in turn, that this declaration constitutes a threat of force and is, therefore, contrary to the principles of the UN Charter. 83 Greece retorts to this argument based on the res inter alios acta principle that the delimitation of territorial waters is not just a matter of treaty law but also of customary international law (which the UNCLOS is supposed to reflect), and that Turkey itself has already extended the breadth of its territorial sea up to 12nm in the Black Sea and in the easternmost part of its Mediterranean coastline. 84 Turkish Ministry of Foreign Affairs, “The Breadth of Territorial Waters”. Regarding the free access of Turkish ships to the open sea when coming from the Turkish Straits, it is worth noting here that Ankara strategically exploited in 2020 the two maritime “corridors” in the Aegean Sea that remain outside of Greek territorial waters (one corridor closely follows the coastline of Anatolia, the other passes through the middle of the North Aegean and then between the island of Levitha on one side and the islands of Patmos, Leros and Kalymnos on the other side) in order to ship weapon systems from the Haydarpasa port near Istanbul to Libya, in violation of the UN embargo (see: UN Panel of Experts established pursuant to resolution 1973 (2011), Final Report, S/2021/229, 8 March 2021, pp. 168-173). It is on the occasion of one of these shipments that an incident occurred between French and Turkish warships in June 2020, leading France to suspend its role in NATO’s Operation Sea Guardian (see: Ibid., pp. 172173; Robin Emmott, John Irish, “NATO to investigate Mediterranean incident between French, Turkish warships”, Reuters, 18 June 2020). 85 Reuters Staff, “Greece plans to extend its western territorial waters”, Reuters, 26 August 2020. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 21 Solidarity and deterrence in the Eastern Mediterranean in 1947 and increases its national territory by roughly 10 per cent, was unanimously approved by the Greek Parliament in January 2021. On that occasion, the Greek government nonetheless made clear that it considered that Greece retained the “right to extend its territorial waters to 12 nautical miles wherever and whenever it wants to”86. In this respect, the issue that could come quite soon to the top of the agenda would be to similarly increase Greek territorial waters in the area south of Crete87. Such a decision, however, could bring the outer limit of Greece’s “extended” territorial waters much closer to Turkey’s maritime claims in the region – especially around the south-eastern part of Crete – and would carry, therefore, a much greater risk of confrontation with Ankara88. Now, how could the changing delimitation of Greek territorial waters concretely interact with the question of the territorial applicability of article 42.7 TEU? Since the EU collective defence clause can be activated in case of armed aggression happening on the territory of an EU member state, this implies that the geographical scope of the EU collective defence clause depends by construction on how each member state defines the limits of its national territory, whether on land, at sea or in the air. In the case at hand, this means more specifically that article 42.7 TEU covers for the time being only a maritime belt of 6 nm in the Aegean Sea, since this is what Athens officially deems to be its territorial waters at present in that region. Yet, if Greece were to decide to expand the breadth of its territorial waters up to 12 nm in the region of the Aegean Sea, just like it recently did in the Ionian Sea, this would mechanically expand by the same token the geographical scope of article 42.7 TEU – at least at the margin89. It is nevertheless likely that the Turkish government would consider the expansion of Athens’ territorial waters in the Aegean as a challenge from which it cannot easily turn away, as we have seen. Pressed to respond, Turkey could for example dispatch warships within the corridor situated between 6 and 12 nm from the Greek coastline. The dispute between Greece and Turkey regarding the extent of the latter’s territorial waters could trigger, in other terms, a chain of events in which the rest of the EU could find itself entangled via article 42.7 TEU should things take a turn for the worse. 86 Nikos Dendias, Speech during the parliamentary debate on the draft law “Determining the breadth of territorial waters in the maritime zone of the Ionian and the Ionian Islands up to Cape Tenaro in the Peloponnese”, Greek Ministry of Foreign Affairs, 19 January 2021. 87 Idem. This corresponds in particular to the area that felt within the scope of Greece’s recent agreement with Egypt (see: Agreement between the Government of the Hellenic Republic and the Government of the Arab Republic of Egypt on the delimitation of the exclusive economic zone between the two countries, 6 August 2020). 88 Turkey’s overlapping claims in the area are notably reflected in the memorandum that Ankara signed in 2019 with the internationally-recognised government in Libya (see: Memorandum of Understanding Between the Government of the Republic of Turkey and the Government of National AccordState of Libya on Delimitation of the Maritime Jurisdiction Areas in the Mediterranean, 27 November 2019). Note that a French research vessel, L’Atalante, escorted by the Greek frigate Elli¸ was reportedly pushed back by two Turkish frigates in April 2021 after entering this area for scientific exploration (see: Yorgo Kırbaki, “Turkey pushes back French, Greek vessels out of its continental shelf”, Hürriyet Daily News, 19 April 2021). 89 In that perspective, it should be noted that the EU (which is itself a party to the UNCLOS alongside all EU member states) has been supportive of the principle that Greece has the right under international law to extend its territorial waters up to 12 nm (Sarantis Michalopoulos, “Commission asks Turkey to respect international agreements after new spat with Greece”, Euractiv.com, 24 October 2018). 22 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean The problem could become even more delicate, however, regarding the contested boundaries of Greece’s national airspace over the Aegean Sea. Let us recall first that, in international law, the horizontal boundaries of the airspace of a given country are supposed to precisely coincide with its land and maritime borders – the national airspace being quite simply the column of air situated right over one’s territory90. The delimitation that Greece has historically given to its national airspace constitutes, however, an exception to this rule. While, as we have just seen, the limits of Greek territorial waters have been fixed for decades at 6 nm from the shore, Athens considers that the breadth of its national airspace goes up instead to 10 nm, creating, as a result, an unusual – even unique – discrepancy between the maritime and air domains that Greece considers as part of its territory91. Turkey’s contestation of the “unconventional” delimitation of Greece’s national airspace has become more forceful over past decades, both in words and deeds, with numerous military overflights being carried out in the region by Turkish fighter jets92. Athens, in return, regularly denounces what it considers as violations by Turkey of its national airspace, indicating that Turkish overflights have taken place not only within the contentious corridor of Greek national airspace situated between 6 and 10 nm but also, at times, directly over Greek inhabited areas93. This tense situation regularly leads to intercepts, provocations and incidents between Greece’s and Turkey’s respective air forces. The EU, for its part, has repeatedly condemned Turkish incursions into Greek airspace94. But the Union has always been careful to avoid taking an explicit position on the exact delimitation of Greek national airspace over the sea – probably because, like the United States, most EU member states would have in reality some serious reservations on this point95. As a result, for reasons similar to that evoked earlier regarding Greek territorial waters, the definition of the Greek national airspace in the Aegean Sea could equally become the source of a spiral of escalation, potentially involving the rest of the EU via article 42.7 TEU. What 90 See article 2 UNCLOS as well as articles 1 and 2 of the Chicago Convention on International Civil Aviation. 91 In theory, Athens could resolve this incongruity simply by extending the breadth of its territorial waters up to 10 nm or further – which is what precisely happened in the Ionian Sea, where the boundaries of Greek territorial sea and national airspace now perfectly coincide at 12 nm from the Greek shore. But a similar move would be much more delicate to perform in the region of the Aegean Sea due to the threat of casus belli emanating, as we have seen, from Turkey. 92 Although Greece extended its airspace up to 10 nm in 1931, Turkey started to protest against this decision only four decades later, after the Cyprus crisis of 1974. 93 Greek claims regarding the violations of its national airspace by Turkish aircraft are registered on the website of the Hellenic National Defence General Staff (see: “Violations of National Airspace – Infringements of Air Traffic Regulations ICAO”, Hellenic National Defence General Staff). 94 European Commission, Commission Staff Working Document – Turkey 2020 Report, Accompanying the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, 2020 Communication on EU Enlargement Policy, SWD(2020) 355 final, 6 October 2020, pp. 65-66. See also: Council of the European Union, Statement of the EU Foreign Ministers on the situation in the Eastern Mediterranean, 15 May 2020. 95 The US Department of State recently refused to provide the US Congress with a full list of confirmed violations of the Greek airspace by Turkish aircraft on the grounds that the United States and Greece officially disagree over the extent of Greek national airspace (see: “‘No consensus on extent of Greek airspace, State Department report says’”, Ekathimerini, 11 November 2020; see also: “Greece: Summary of Claims”, United States Navy Judge Advocate General’s Corps, 2014). F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 23 Solidarity and deterrence in the Eastern Mediterranean would be different in this case, however, is that there would exist a legitimate doubt about whether the EU collective defence clause could apply within the outer rim of Greece’s national airspace over the Aegean Sea if an armed aggression were to occur precisely there – an uncertainty which could lead, in turn, to an extremely divisive situation among EU member states. 3. Looking forward: scenarios and implications Activating article 42.7 TEU in the Eastern Mediterranean would not represent, of course, an end in itself. It is important therefore to consider not only how this provision could be triggered but also how it could be implemented, in broad terms at least, in this eventuality. A key point to understand in this regard is that the exercise of collective defence between EU member states could come in different shapes and sizes, so to speak, as this section will explain first (3.1). That being said, it is clear that the most likely prospect remains that the EU collective defence clause will ultimately not be activated in the Eastern Mediterranean, especially now that tensions seem, hopefully, to be easing with Turkey. This does not mean, however, that no implication should be drawn for EU security and defence policy from the recent episode of tensions in the Eastern Mediterranean. This is why this section will then focus on whether this experience may be factored into the discussions currently taking place at the EU level to make article 42.7 TEU more operational (3.2). 3.1. What if the EU collective defence clause were to be activated in the Eastern Mediterranean? Implementing the EU collective defence clause in the Eastern Mediterranean would undoubtedly require flexibility and differentiation, meaning that the intensity of the actions undertaken according to article 42.7 TEU would have to be, on the one hand, commensurate to the specific circumstances in which this clause would be invoked (3.1.1) and, on the other hand, that EU member states would most likely contribute to such actions in different ways (3.1.2). 3.1.1. Flexibility: adapting EU member states’ responses to circumstances In Europe, debates about collective defence tend to be fixated on the hypothesis of a largescale military confrontation, in particular with Russia96. Such a scenario, while obviously important, does not cover the full range of situations that could fall within the scope of collective defence. In this perspective, it is worth recalling that the only time that both NATO’s and 96 See for instance the (otherwise fascinating) debate which recently took place around the question of whether Europe can or cannot “defend itself” and which characteristically centered on whether Europeans could defeat a major conventional offensive from Russia in the Baltics (see Barry Posen, “Europe Can Defend Itself”, Survival, 62(6), 2020, pp. 7-34; also in Survival, 63(1), 2021: Douglas Barrie, Ben Barry, Henry Boyd, Nick Childs, Bastian Giegerich, “Europe’s Defence Requires Offence”, pp.19-24; François Heisbourg, “Europe Can Afford the Cost of Autonomy”, pp. 25-32; Stephen G. Brooks, Hugo Meijer, “Europe Cannot Defend Itself: The Challenge of Pooling Military Power”, pp. 33-40; Barry R. Posen, “In Reply: To Repeat, Europe Can Defend Itself”, pp. 41-49; see also: Hugo Meijer, Stephen Brooks, “Illusions of Autonomy: Why Europe Cannot Provide for Its Security If the United States Pulls Back”, International Security, 5(4), 2021, pp. 7-43). 24 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean the EU’s security guarantees have been effectively activated was in response to terrorist attacks, those of 11 September 2001 in the United States and of 13 November 2015 in France respectively – both events which, although unquestionably tragic, caused far less damage than what a major conflict between modern states could do. The responses implemented in both cases were thus by the same token rather modest, at least by comparison to the type of responses that are usually contemplated when talking about collective defence. Thus, when article 5 of the Washington Treaty was invoked in 2001, NATO allies implemented only a package of limited measures, the most notable of which were the deployment of NATO’s Standing Naval Force Mediterranean to the Eastern Mediterranean and the dispatch of NATO AWACS aircraft to help monitor the national airspace of the United States97. But the ensuing military intervention that took place in Afghanistan to destroy the Al-Qaida terrorist network and defeat the Taliban regime (Operation Enduring Freedom) was not, at least not initially, carried out under the auspices of NATO but directly by the United States and the United Kingdom. Likewise, the direct results of the first activation of article 42.7 TEU were not particularly spectacular98. One obvious reason is that France and other EU member states were already actively engaged in the international coalition led by the United States to defeat the Islamic State when the terrorist attacks took place in and around Paris in November 2015. The activation of the EU security guarantee in 2015 had thus essentially two results. It led, firstly, some EU countries – the United Kingdom in particular – to perform additional air strikes in Syria and Iraq and/or to provide greater logistical and training support for operations in this theatre99. Secondly, it pushed certain member states to provide greater political and intelligence support to France and/or to indirectly assist Paris by increasing their contributions (albeit often modestly) in missions under EU, UN or French command in other regions of the world so as to relieve the general pressure under which French military forces were put at the time100. If it is useful to keep these precedents in mind when considering how article 42.7 TEU could be implemented in the Eastern Mediterranean, this not because effects would necessarily be the same in the case at hand (most likely, they would not). What these two examples highlight, however, is that the activation of the right of collective defence can occur in response 97 The other measures adopted by NATO allies in response to the first invocation of article 5 were to: enhance their intelligence sharing and co-operation; prepare for mutual assistance in case of terrorist counter-attacks; increase security for US and other allied forces present on their respective territories; backfill allied assets diverted to counter-terrorist operations; provide blanket overflight clearances; and provide access to allied ports and airfields, including for refueling (see: Christopher Bennett, “Aiding America”, NATO Review, 1 December 2001). 98 See for example : Nicolas Gros-Verheyde, “La demande française de solidarité : un semi flop”, Bruxelles2, 4 February 2016. 99 For an overview of the bilateral assistance provided by EU member states to France, see: Activation of Article 42(7) TEU. France's request for assistance and Member States’ responses, op. cit.; Jana Puglierin, “‘Aid and Assistance by All the Means in their Power’. The EU Mutual Assistance Clause as a Test Case for European Defense”, Kompakt, n° 22, German Council on Foreign Relations, December 2016. 100 As stressed in particular by: Niklas Nováky, “The Invocation of the European Union’s Mutual Assistance Clause: A Call for Enforced Solidarity”, European Foreign Affairs Review, 22(3), 2017, pp. 357375. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 25 Solidarity and deterrence in the Eastern Mediterranean to a much wider range of circumstances than is generally appreciated in the European strategic debate and that its exercise does not necessarily lead to “extreme” measures such as the conduct of major combat operations. Indeed, for reasons that have to do as much with strategy as with the need for political legitimacy and respect of international legality, any action undertaken in the name of collective defence would need to be both proportionate to the threat actually faced and necessary to address this threat101. A more concrete discussion may help to better grasp this point in the case at hand. Let us return to the scenario evoked in the previous section about the tense standoff that would most certainly ensue between Greek and Turkish armed forces should Greece decide to extend the breadth of its territorial waters from 6 to 12 nm in the region of the Aegean Sea (as Athens is entitled to do, let us recall, under the UNCLOS). If an incident causing significant damage or injury – say, the sinking of a Greek warship by Turkey – were to occur in those circumstances, an invocation of the EU collective defence clause could be warranted102. Although undoubtedly serious, such a situation would not be, however, of the same level of gravity as a full-scale military conflict – or at least not yet. Thus, in this event, the primary objective of the EU member states would likely be to avoid the deterioration of an already bad situation. The invocation of article 42.7 TEU could be used in that context first and foremost to signal the solidarity of EU member states and their resolve to defend each other. Such a “warning shot” could be immediately accompanied by demonstrations of political and diplomatic support by EU countries at the international level, notably at the UN Security Council103. Performing military manoeuvres of a purely demonstrative nature, such as joint patrols in Greek territorial waters and airspace, could finally be envisaged to deter further incidents on the part of Turkey. It cannot be excluded, however, that such immediate measures would prove insufficient or too slow to avoid an aggravation of the situation, making the effective use of military force inevitable. Yet, even in this case, distinctions could and, indeed, should be made in the exercise of collective defence by EU member states vis-à-vis Turkey. One way to keep the intensity of a military confrontation in check could be, generally speaking, to restrict the type or number of weapons and platforms used in combat for example. The geographical scope of hostilities could also be limited, for instance by trying to confine military operations to the 101 The requirements of proportionality and necessity in the exercise of self-defence have been most clearly underlined by the ICJ in: ICJ, Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, 1996, §41. In addition, the condition of immediacy, in the sense that “action taken in selfdefence must be taken within a reasonable timeframe in relation to the occurrence of the attack”, should also be respected under customary international law (Terry Gill, “Legal Basis of the Right of Self-Defence under the UN Charter and Under Customary International Law” (Chapter 8), in Terry Gill, Dieter Fleck (eds.), The Handbook of the International Law of Military Operations (2nd edition), Oxford University Press, 2015, p. 223). 102 Recall, as we have seen earlier, that in the Oil Platforms case, the ICJ did not exclude the possibility that the mining of a single ship may give rise to the right of self-defence (ICJ, Oil Platforms (Islamic Republic of Iran v. United States of America), op. cit., §72). 103 As article 42.7 TEU derives from article 51 of the UN Charter, this means that all the measures taken as a result of the activation of the EU collective defence clause would need (at least in theory) to be immediately reported to the UN Security Council until the latter adopts appropriate measures to maintain or restore international peace and security. 26 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean maritime domain (or even just a designated part of it) while avoiding the Turkish mainland104. Of course, it remains very difficult to assess in the abstract what the dynamics of the process of escalation would be if a military conflict were to break out in the Eastern Mediterranean. For one thing, frictions, uncertainty, failures of command would hamper, as always, how finely measured and controlled the use of military force would be. For another, the dynamics of such a conflict would crucially depend not just on EU member states’ decisions but also on the actions and counter-actions by Turkey, as well as on the involvement or noninvolvement of many other parties – most notably the United States, but also Russia and all the regional powers with which Greece has started to develop close bilateral ties in recent months, such as Egypt, Israel, Saudi Arabia or the United Arab Emirates105. All this does not invalidate, however, the key point here which is that activating article 42.7 TEU would not – at least not in and of itself – automatically lead to an all-out military confrontation between EU member states and Turkey. 3.1.2. Differentiation: the possibility for EU member states to contribute in distinct manners to implementing article 42.7 TEU If article 42.7 TEU were to be invoked in the Eastern Mediterranean, all EU countries would be legally bound to provide “aid and assistance by all the means in their power” to the member state that would have been the victim of an armed aggression on its territory. Yet the means employed to fulfil this commitment could considerably vary from one member state to another. This can be inferred from the text of the EU collective defence clause which specifies that the obligation of aid and assistance “shall not prejudice the specific character of the security and defence policy of certain Member States”. This reservation is often understood as referring to those EU member states that have a tradition of neutrality or non-alignment in foreign policy, namely Austria, Cyprus, Finland, Ireland, Malta and Sweden. Nonetheless, it is possible to consider that this caveat could apply more broadly to all EU member states, with the effect of modulating their obligation of aid and assistance under article 42.7 TEU according to their respective national situations106. The protocol “on the concerns of the Irish peo- 104 During the Falklands war, for instance, the United Kingdom unilaterally limited the exercise of its right of self-defence by defining an exclusion zone around the islands invaded by Argentina, while explicitly refraining from attacking Argentine forces or installations on the South American continent (see: Lawrence Freedman, Virginia Gamba-Stonehouse, Signals of War. The Falklands Conflict of 1982, London: Faber and Faber, 1990, pp. 247-252). The Eastern Mediterranean is, of course, a much narrower theatre of operations than the South Atlantic, which would make it difficult, but not necessarily impossible, to respect some sort of geographical limitation of hostilities. 105 The United Arab Emirates in particular appears to have become one of Greece’s closest partners in facing off Turkey over recent months (see: Paul Iddon, “UAE Dispatches Fighter Jets To Support Its Allies Against Turkey”, Forbes, 26 August 2020). Thus, in November 2020, Athens and Abu Dhabi formalised their alliance by signing an agreement which included a bilateral mutual defence clause. At this occasion, Greek Foreign affairs minister Nikos Dendias underlined that this represented “the first time since the end of World War II that such a clause is included in an agreement signed by [Greece]” (see: Nikos Dendias, Statement of the Minister of Foreign Affairs, Nikos Dendias, following his briefing of parliamentary group representatives, Greek Ministry of Foreign Affairs, 26 November 2020). 106 In its assessment of the Lisbon Treaty in 2009, the German Federal Constitutional Court considered for instance that this reservation could also encompass the constitutional requirement under F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 27 Solidarity and deterrence in the Eastern Mediterranean ple on the Treaty of Lisbon”, in particular, seems to support such an interpretation. Indeed, this protocol provides that “[i]t will be for Member States – including Ireland, acting in a spirit of solidarity and without prejudice to its traditional policy of military neutrality – to determine the nature of aid or assistance to be provided to a Member State which is the object of a terrorist attack or the victim of armed aggression on its territory”107. Such a language suggests that all EU member states, and not just traditionally “neutral” member states such as Ireland, would retain a very large degree of discretion in determining the scope and nature of the help they would provide under article 42.7 TEU. Regardless of legal obligations, it is clear that the implementation of the EU collective defence commitment would depend in any case on the assets and resources that would be available in practice to each member state at the time of the invocation of this provision. It is likely for instance that some EU countries would only be able to provide help of a civilian nature such as humanitarian108 or financial assistance109 if article 42.7 TEU were to be invoked in the Eastern Mediterranean. It is also possible that some member states would be ready to offer some military assistance, but only indirectly – for example by providing logistical support, overflight clearances, or medical assistance, or by deploying troops and equipment to other theatres so as to backfill the armed forces of the (probably few) EU countries that would be prepared, by contrast, to directly participate in military operations in the region. Also, it should be noted that, from an organisational viewpoint, the EU collective defence clause does not foresee any formal role in its implementation for EU institutions. This means that any assistance undertaken under article 42.7 TEU could be provided either bila-terally, as was done when Paris managed itself the various contributions it received from other member states after the attacks of November 2015, or through an ad hoc group of EU countries working closely together110. Nonetheless, if member states were to agree, it could still be envisaged to involve relevant institutions in Brussels (like the European External Action Service) to implement the EU collective defence clause, since its text does not expressly exclude this possibility either. This could be done for instance to exchange information between national capitals or even to coordinate the aid and assistance provided under article 42.7 TEU, or at least part of it (e.g. its civilian component). German law to receive parliamentary approval before deploying the Bundeswehr abroad (see: Bundesverfassungsgericht, Judgment of the Second Senate of 30 June 2009 – 2 BvE 2/08, 2009, §386). 107 See: Protocol on the concerns of the Irish people on the Treaty of Lisbon, Official Journal of the European Union (OJEU), 2 March 2013, L.60, pp. 131-139. Note that the explicit mention of a “terrorist attack” is a reference not to article 42.7 TEU but to article 222 TFEU. 108 Humanitarian assistance could be useful for example to manage flows of refugees if the migration issue were to be instrumentalised by Ankara (like in February-March 2020) as a destabilising measure in the midst of a broader conflict. 109 Economic and financial help could be provided in particular under article 122.2 TFEU, which specifies that such assistance may be granted to a member state if the latter “is in difficulties or is seriously threatened with severe difficulties caused by natural disasters or exceptional occurrences beyond its control”. 110 In the Eastern Mediterranean, the core of such a group would most likely include Cyprus, Greece and France. 28 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean 3.2. The question of the operationalisation of article 42.7 TEU and the Eastern Mediterranean It is worth considering, finally, the wider implications of the recent crisis between Greece and Turkey for EU security and defence policy. This is true especially in the light of the ambition that has progressively emerged about making article 42.7 TEU more operational (3.2.1). As this ambition is expected to be clarified in the upcoming months through the development of the EU’s so-called “Strategic Compass” and a specific declaration on article 42.7 TEU, it is nevertheless unlikely, as this report will show, that a direct connection will be made in this context between the operationalisation of the EU collective defence clause and the latest tensions in the Eastern Mediterranean (3.2.2). 3.2.1. A growing ambition: to clarify and strengthen the EU collective defence clause The ambition to make the EU collective defence clause more operational has grown first and foremost out of the somewhat frustrating experience of its first invocation in 2015. When France first activated this provision, it appeared that the other member states had little idea on how to respond, since, in the absence of an integrated military command structure at the EU level, virtually nothing had been put in place to ensure the effective and coordinated implementation of this commitment in case of need111. As a result, in a resolution adopted two months later, the members of the European Parliament regretted that “no analysis and no guidelines were available when the mutual defence clause was activated for the first time”, leading to a “situation requiring ad hoc measures, ad hoc management and ad hoc cooperation”112. The European Union Global Strategy (EUGS) that was presented in June 2016 by HR/VP Federica Mogherini seemed to take this criticism into account. While reminding that NATO remains the primary framework for collective defence for those EU member states that belong to it, the EUGS called on Europeans to “live up” to their security commitments such as article 42.7 TEU and to “translate [them] into action”113. Yet, the Implementation Plan on Security and Defence which followed in November 2016 to give concrete expression to the EUGS in the field of security and defence remained in the end very evasive in this regard, merely “highlighting” the importance of the EU collective defence clause in the perspective of the protection of the Union and its citizens114. Since then, one country – France – has nonetheless kept pushing for a reinforcement of the EU security guarantee. In August 2018, President Macron announced that he intended “to spearhead a project to strengthen European solidarity in security matters”, arguing that 111 See: Sven Biscop, “The European Union and mutual assistance: more than defence”, The International Spectator, 51(2), 2016, pp.119-125. 112 European Parliament, Resolution of 21 January 2016 on the mutual defence clause (Article 42(7) TEU) (2015/3034(RSP)), §7. 113 High Representative of the Union for Foreign Affairs and Security Policy, Shared Vision, Common Action: A Stronger Europe : A Global Strategy for the European Union’s Foreign And Security Policy, June 2016, p. 14, p. 19. 114 Council of the European Union, Implementation Plan on Security and Defence, 14392/16, 14 November 2016, p. 4. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 29 Solidarity and deterrence in the Eastern Mediterranean Europeans “should give more substance to Article 42(7) of the Treaty on European Union” and signalling that France was “ready to enter into concrete discussions with European States on the nature of reciprocal solidarity and mutual defence relations under our Treaty commitments”115. Three days later, during a trip to Finland (which, significantly, is a nonNATO EU member state), the French head of state specified that he was proposing to reform the EU collective defence clause to create “a kind of reinforced article 5 [of the North Atlantic Treaty] within the European Union, which would mark a very strong solidarity between the countries of the European Union in the area of defence”116. Emmanuel Macron’s call for greater European solidarity was surely motivated by the doubts that emerged during Donald Trump’s presidency about the reliability of the US commitment to defend Europe117. But it seems that the French president saw (and in all likelihood still sees) the strengthening of defence mechanisms such as article 42.7 TEU not just as a pragmatic means to hedge against a potential weakening of the American security umbrella, but also as an objective in its own right, representing in a way the logical culmination of the “European project” as a political community. When in Finland, President Macron justified his proposal to strengthen the EU collective defence clause by saying that this was in his opinion “what we owe to each other and what we owe to Europe to make it a reality for each member state and each of our citizens”118. On a more recent trip to the Baltic states and Poland in February 2020, the French leader used a similar language (though, without making an explicit reference to article 42.7 TEU), saying that he would “be happy the day when Poles say to themselves: ‘the day I am attacked, I know that Europe will protect me’, because then the attachment to Europe [le sentiment européen] will be indestructible”119. The fact that France has been at the forefront of the efforts to make the EU collective defence clause more operational is, of course, hardly surprising. Paris is after all the only EU capital thus far to have invoked this provision, and it tends, in general, to be supportive of a more robust role for the EU in security and defence matters. But France has not been the only one to push in this direction. Thus, in May 2020, the Defence ministers of France, Germany, Italy and Spain sent a joint letter to their EU counterparts and the HR/VP in which 115 Emmanuel Macron, Speech by the President of the Republic at the Ambassadors’ Conference, French Ministry for European and Foreign Affairs, 27 August 2018. 116 Emmanuel Macron, Joint press conference with Sauli Niinistö, President of the Republic of Finland, Élysée, 30 August 2018. 117 “Europe can no longer entrust its security to the United States alone. It is up to us to assume our responsibilities and to guarantee European security and thereby sovereignty” (Speech by the President of the Republic at the Ambassadors’ Conference, French Ministry for European and Foreign Affairs, op. cit.). 118 Joint press conference with Sauli Niinistö, President of the Republic of Finland, op. cit. A few weeks later, Finnish President Sauli Niinistö described article 42.7 TEU as “the true core of European defence”, saying that he had been “delighted that France’s President Emmanuel Macron [had] been willing to discuss the content of that article” (see: Sauli Niinistö, Speech at the opening of the 227th National Defence Course, 5 November 2018). 119 Emmanuel Macron, Joint press with Andrzej Duda, President of the Republic of Poland, Élysée, 3 February 2020. Even as France’s minister of Economy and Industry under François Hollande, Emmanuel Macron defended similar ideas, once declaring that European solidarity implied that “we may have to accept to die for others, because it is our project, because we are Europeans, because we have chosen it, because we will choose it tomorrow” (Emmanuel Macron, Speech at the College of Europe, “Quelle Europe pour la nouvelle génération ?”, 18 April 2016). 30 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean they notably underlined “the importance of European solidarity to act and react to crises”, indicating that “[a] key work, in that regard, will be the operationalisation of the Article 42(7) of the TEU”. The four ministers argued, furthermore, that “[r]egular scenario-based discussions, wargames and exercises could help reach a common perspective on the possible threats and bolster the political interactions among our capitals”, specifying that “[s]uch tabletop exercises should cover all possible worst-case scenarios of crisis”120. As a result, the ambition to better operationalise the EU security guarantee has recently started to reappear in EU documents. In its conclusions on security and defence of June 2019, the Council invited member states to (finally) “discuss the lessons identified following the first activation of Article 42.7 TEU” in 2015121. Taking up the joint recommendation from France, Germany, Italy and Spain (but without insisting on the need to cover “all possible worst-case scenarios of crisis”), the Council agreed, furthermore, in June 2020 to hold “table top exercises and scenario-based policy discussions” in the future, focusing in particular on the practical modalities for implementing the EU collective defence clause to reach a better common understanding of what fulfilling this commitment might entail in practice122. 3.2.2. The EU collective defence clause and the recent tensions in the Eastern Mediterranean: what impact on the Strategic Compass and the envisaged declaration on article 42.7 TEU? The ambition to strengthen article 42.7 TEU, as we can see, emerged long before the crisis which occurred between Greece and Turkey in the summer of 2020 and thus independently from any perspective of a possible confrontation with Turkey. One may wonder, however, to which extent this ambition should now be reassessed in view of the recent tensions in the Eastern Mediterranean123. This question is particularly relevant today as the debate around the final scope of the operationalisation of the EU collective defence clause should soon be settled with the adoption in the coming months of two key documents, namely the EU’s Strategic Compass and, possibly, a specific declaration on article 42.7 TEU. It may be useful to recall here that the Strategic Compass, first of all, is a document that is currently being developed to update the EU’s level of military ambition by identifying the 120 Florence Parly, Annegret Kramp-Karrenbauer, Margarita Robles Fernández, Lorenzo Guerini, Letter from the defence ministers of France, Germany, Italy and Spain to their European counterparts and the High Representative of the Union for Foreign Affairs and Security Policy, 29 May 2020, p. 2. Emphasis added. 121 Council of the European Union, Council Conclusions on Security and Defence in the context of the EU Global Strategy, 10048/19, 17 June 2019, p. 13. 122 Council of the European Union, Council Conclusions on Security and Defence, 8910/2, 17 June 2020, p. 4. Thus, in early 2021, two scenario-based discussions related to article 42.7 TEU have been on the agenda of the Political and Security Committee (see: Council of the European Union, Notice of Meeting and Provisional Agenda: Political and Security Committee, CM 1143/1/21 REV 1, 13 January 2021; Council of the European Union, Notice of Meeting and Provisional Agenda: Political and Security Committee, CM 2380/21, 23 March 2021). A document “consolidating the lessons identified from these exercises and scenario-based policy discussions” should be presented to EU member states by the end of 2021 (see: Council of the European Union, Council Conclusions on Security and Defence, 8396/21, 10 May 2021, p. 4). 123 A point raised (but not discussed further) in: Niklas Nováky, The Strategic Compass. Charting a New Course for the EU’s Security and Defence Policy, Wilfried Martens Centre for European Studies, December 2020, p. 7. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 31 Solidarity and deterrence in the Eastern Mediterranean threats and challenges its member states will face in the coming years as well as the means to address them124. The elaboration of this new politico-military strategy was initiated in June 2020 in the run-up to the German presidency of the Council and should be completed during the French presidency in the first semester of 2022125. The first phase of this process involved a threat analysis, which was concluded in November 2020126. In this regard, it is worth noting that Turkey was explicitly pointed out in this analysis127. This does not mean that a real consensus exists across the Union to explicitly view Ankara as a threat since the EU’s threat analysis was, concretely speaking, “only” an unsorted compilation of all the threats and challenges that the intelligence services from each member state were willing to identify for the next 5 to 10 years128. But this mention remains quite remarkable for a country that is both a NATO member and still nominally a candidate for EU membership129. At the moment of this writing, the second phase of the development of the Strategic Compass is ongoing. The latter involves a dialogue between member states to determine the future priorities and objectives for the Common Security and Defence Policy (CSDP), with issues to be discussed grouped into four “baskets”: crisis management, resilience, capability development, and partnerships130. It is in particular within the second basket about “resilience” that the scope and implications of a reinforcement of the EU collective defence clause are supposed to be examined131. In parallel to the Strategic Compass process, discussions have been ongoing on adopting a separate declaration that would be specifically dedicated to article 42.7 TEU and could be approved, similarly, in early 2022132. The exact content of this declaration, its articulation with the forthcoming Strategic Compass as well as its concrete implications remain, however, rather vague to date, since almost nothing has been officially disclosed on these points 124 Council of the European Union, Council Conclusions on Security and Defence, 8910/2, 17 June 2020, p. 3. For a comprehensive overview of the EU’s Strategic Compass, see for example: Daniel Fiott, Uncharted Territory? Towards a common threat analysis and a Strategic Compass for EU security and defence, Brief n° 16, European Union Institute for Security Studies, July 2020. 125 European External Action Service, Factsheet: Towards A Strategic Compass, February 2021. 126 Josep Borrell, “Defence cooperation: How strategic perseverance will pay off”, European External Action Service, 20 November 2020. 127 See: Nicolas Gros-Verheyde, “La boussole stratégique : un premier effet visible pour les services de renseignement”, Bruxelles2, 15 February 2021. 128 European External Action Service, Memo: Questions and answers: Threat Analysis – a background for the Strategic Compass, 20 November 2020. 129 “La boussole stratégique : un premier effet visible pour les services de renseignement”, op. cit. 130 Council of the European Union, Council Conclusions on Security and Defence, 8910/2, 17 June 2020, p. 4. 131 Nicole Koenig, The EU’s strategic compass for security and defence: Just another paper?, Hertie School, Jacques Delors Centre, 10 July 2020, p. 6; Claudia Major, Christian Mölling, Europe, Germany and defense: priorities and challenges of the German EU presidency and the way ahead for European Defense, Note n° 63/20, Fondation pour la recherche stratégique, 13 October 2020, p. 9 ; Nicolas Gros-Verheyde, “Boussole stratégique. L’UE veut devenir un acteur maritime, cyber et spatial à part entière”, Bruxelles2, 15 February 2021. 132 Martin Quencez, Sophia Besch, The Challenges Ahead for EU Defense Cooperation, Policy Brief n° 2, The German Marshall Fund of the United States, January 2020, p. 4. 32 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean thus far – even on the part of France, which is said to be at the origin of this initiative133. One can nonetheless suppose that if France wishes to see all EU member states adopt a political declaration on the EU collective defence clause, it is because of the symbolic significance that the country attaches, or would like to see attached, to this provision134. In so doing, the broader objective of the French government may be to try to raise awareness among its European partners about the necessity to ensure their own security, not just as NATO allies but also as Europeans – a goal that would be congruent with the attitude Paris adopted in summer 2020 when it strongly backed Greece and Cyprus against Turkey, precisely in the name of European solidarity135. Thus, given what is publicly known to date about the Strategic Compass and the possible declaration on article 42.7 TEU, one may expect that the practical dimension of the operationalisation of the EU collective defence clause will be mainly addressed in the first of these two documents, while its symbolic and political dimension could be more strongly emphasised in the second. At this stage, however, nothing officially indicates whether and how the specific issue of the Eastern Mediterranean will be handled in this context and in particular whether a connection will be explicitly acknowledged with article 42.7 TEU in this respect. Discussions among experts in the context of the Strategic Compass process have not directly addressed this connection either, focusing instead on the question of the desirable division of labour to be achieved with NATO. The consensus that seems to have emerged among defence experts in this perspective is that a strengthening article 42.7 TEU could be useful, but only to the extent that this effort would focus on countering cyber, terrorist or hybrid threats while leaving the conventional (let alone nuclear) defence of the European continent entirely in the hands of the Atlantic Alliance136. In support of this position, experts have underlined in particular that any clear foray by the EU into the domain of collective defence would likely be viewed with some apprehension, if not suspicion, by many member states – 133 The so-called “Strategic Update” which was published in January 2021 by the French Ministry of Defence could have provided some indications about the intentions of the French government in that regard. The document remained instead rather evasive, simply stating that: Europe’s convergence around common defence and security interests will be one of the issues in efforts to define a “strategic compass”, launched by Germany and due to continue until the French Presidency of the EU in the first half of 2022. The operationalisation of Article 42.7 of the Treaty on European Union should also make it possible to strengthen the effectiveness of solidarity mechanisms between Europeans. See: French Ministry of Defence, Strategic Update 2021, January 2021, p. 34. 134 It is possible indeed that, from a French perspective, dealing with the subject of article 42.7 TEU only in the context of the Strategic Compass would be perceived as downplaying this issue, since it would turn it into just another point of discussion about the future of CSDP. Conversely, a declaration specifically devoted to the EU collective defence clause and endorsed by all member states would arguably confer a much stronger visibility to this provision – which would be more in line with President Macron’s conceptions in this regard, as we have seen earlier. 135 See for instance: “Macron slams Turkey’s ‘violation’ of Greek, Cypriot sovereignty”, Al Jazeera, 23 July 2020. 136 See: Christian Mölling, Torben Schütz (eds.), The EU’s Strategic Compass and Its Four Baskets. Recommendations to Make the Most of It, Report n° 13, German Council on Foreign Relations, November 2020, pp. 8-9; Thierry Tardy, For a New NATO-EU Bargain, Security Policy Brief n° 138, Egmont Institute, February 2021; Sven Biscop, EU and NATO Strategy: A Compass, a Concept, and a Concordat, Security Policy Brief n° 141, Egmont Institute, March 2021. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 33 Solidarity and deterrence in the Eastern Mediterranean especially, but not only, Central and Eastern European countries which remain particularly wary of any initiative which might weaken NATO and antagonise the United States137. It has been emphasised, furthermore, that expecting a noticeable contribution from the EU to the collective defence of its member states could further undermine its credibility as a security actor, given in particular the serious difficulties that CSDP already faces in the “less demanding” area of crisis management138. Since most EU member states are also members of NATO, the distribution of tasks envisaged by experts, based on the comparative advantages of both organisations, seems obviously sensible – at least as a broad outline. There is no dispute indeed that the Union disposes of a set of civilian and military tools that makes it well-positioned to address “peripheral” aspects of security like disinformation, border management, terrorism, critical infrastructures, or economic and energy coercion, while the Atlantic Alliance remains, by comparison, much better equipped to deal with “traditional” military threats139. And as we have already seen, EU treaties recognise in any event NATO’s pre-eminence when it comes to collective defence for those EU member states that belong to it. However, there is an important, yet less often mentioned, obstacle to such a clear division of labour between the EU and NATO, namely the imperfect overlap between the mandates of both organisations in the field of collective defence140. Indeed, in “situations involving nonNATO EU Member States or EU Member States’ territories that are outside the North Atlantic area and are therefore not covered by the Washington Treaty, or situations where no agreement on collective action has been reached within NATO”, the EU and its member states would not be able to rely on NATO to ensure their collective defence, as the European Parliament pointed out as early as 2012141. One would tend to think in this regard about the possibility of an attack against Finland or Sweden, or against EU overseas territories. But an armed aggression by Turkey in the Eastern Mediterranean against Greece or Cyprus would also fall into the category of situations where NATO would be unable by definition to act as a forum for collective defence, but where the EU might. As this scenario is, in addition, the only one for which the possibility of resorting to the EU collective defence clause has been explicitly raised in recent times (see section 1.3), it could seem logical – and, in fact, eminently prudent – to place this possibility at the centre of the current efforts to make article 42.7 TEU more operational. It must be immediately added, however, that such a conclusion – which would make sense from a strictly EU-centric perspective – is unlikely to be acted upon in practice, given the broader geopolitical context in which EU member states conduct their foreign and defence 137 Uncharted Territory?..., op. cit., p. 7; Katarina Engberg, A European Defence Union by 2025? Work in progress, Swedish Institute for European Policy Studies, January 2021, p. 11. 138 See e.g.: Daniel Fiott, “As You Were? The EU as an Evolving Military Actor” (Chapter 8) in The CSDP in 2020. The EU’s legacy and ambition in security and defence, European Union Institute for Security Studies, 2020. 139 On this see for example: Gustav Lindstrom, Thierry Tardy (eds.), The EU and NATO. The essential partners, European Union Institute for Security Studies, 2019. 140 “The art of commitments…”, op. cit. 141 European Parliament, Resolution of 22 November 2012 on the EU’s mutual defence and solidarity clauses: political and operational dimensions (2012/2223(INI)), §12. 34 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean policies. Indeed, as said above, many EU countries remain concerned about the negative impact that the ambition to further operationalise article 42.7 TEU might have on NATO and transatlantic relations in general. Thus, if these efforts were to be developed in direct link with the hypothesis of a confrontation with Turkey, it is likely that any progress on this front would become extremely divisive, if not downright impossible – not to mention the many diplomatic and political complications that would result from such an explicit lin-kage, and in particular the harm that could be done to the very fruitful cooperation that has developed in recent years between the EU and NATO142. Given these foreseeable political difficulties, it could thus be tempting to ask whether it would not be preferable, in fact, to try to handle the problem the other way round, so to speak. The development of the Strategic Compass and the planned declaration on article 42.7 TEU could indeed appear as an opportunity to establish the explicit principle whereby the EU and NATO security guarantees should never be invoked in disputes between members of either organisation, along the lines of what was done for example in 1992 with the Petersberg Declaration (see 2.1.1). In this way, any possibility that the EU collective defence clause could be invoked one day against Turkey would simply be made impossible. Yet, a diplomatic impasse should be expected in this case as well. In the current strategic context, it is very likely that those EU member states that have recently felt threatened by Turkey and have stressed in response the need for European solidarity – that is, Greece and Cyprus, as well as France – would regard such a proposal as simply unacceptable and, in consequence, veto it. It would be surprising, in brief, if the Strategic Compass and/or the envisaged declaration on the EU collective defence clause could open the door to any real planning at the EU level to respond to the prospect of an armed aggression from Turkey. But, conversely, neither document is likely to be used to definitively exclude this possibility either. In these circumstances, the link between the ambition to make article 42.7 TEU more operational and the recent tensions in the Eastern Mediterranean is thus likely to remain at most indirect and ambiguous. Conclusion: a scenario bound to remain taboo? It is unlikely that the EU collective defence clause will ultimately be invoked against Turkey. It is indeed more reasonable to expect that Ankara will come to see it as not being in its interest to try to impose its claims by force in the Eastern Mediterranean and, failing that, that it will be deterred from pursuing such a dangerous course of action by the various “carrots and sticks” that EU member states and others could wield to that effect. That said, the fact that the possibility of triggering article 42.7 TEU was explicitly raised in 2020 in connection with the situation in the Eastern Mediterranean remains worthy of note. This was, first of all, the first time that this provision has been employed in a preventive manner. Indeed, although France, unlike Greece, did invoke the EU collective defence clause in November 2015, this was after a disaster – a series of terrorist attacks – had already occurred. In the case at hand, however, article 42.7 TEU was signalled in advance by the Greek government, both to mobilise the attention of its EU partners on an unfolding crisis and to 142 European External Action Service, Factsheet: EU-NATO cooperation, June 2020. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 35 Solidarity and deterrence in the Eastern Mediterranean forestall a possible escalation on the part of Turkey. Of course, for students of alliance politics, there is nothing particularly surprising about a collective defence commitment being explicitly flagged so as to send a warning to friends and potential foes alike. Such a move is, however, unprecedented in the history of the EU as a security and defence actor and thus deserves to be underscored. This episode also highlighted the close attention that needs to be paid to the legal foundations of collective defence in Europe. As we have seen in this report, relatively technical points of law could become major sources of division between EU member states in case of crisis (e.g. in the event of an armed aggression in the maritime domain) and thus have important political and strategic implications. Europeans need therefore to think in detail about the full range of collective defence scenarios they might face in the future and prepare themselves in particular for the legal debates that these scenarios might generate, as well as for the fact that potential adversaries may try to leverage the legal vulnerabilities of their collective defence framework to their advantage. Thinking about collective defence in broader terms is, of course, even more important on a practical level. Indeed, although the strategic debate in Europe has tended to focus in recent years on the prospect of a major conventional offensive, particularly from Russia, the recent tensions with Turkey have underlined that other collective defence scenarios are possible, in terms of both the actors involved and potential military intensity. In other words, the exercise of collective defence in Europe should not be seen as a task that is always equal but, on the contrary, as closely dependent on the specific political and military circumstances in which it may occur. And it is precisely political circumstances that make the case at hand in this report especially challenging to deal with. As a long-standing member of NATO, Turkey is indeed supposed to be part of the transatlantic security community within which the use of force has become, if not unthinkable, at least extremely unlikely143. Actively preparing for, or even openly thinking about, the possibility of a conflict with Ankara thus implies going against deeply ingrained inhibitions among European policy-makers – something that is unlikely to happen, as this report has concluded, in the context of the elaboration of the EU’s Strategic Compass and a possible declaration on article 42.7 TEU. And yet the question remains: is the general reluctance to consider the possibility of a serious confrontation between Ankara and EU member states tenable? To give a categorical answer, one way or the other, is far from easy. One does not need to be perfectly cynical to agree that it may be politically necessary, at least sometimes, to avoid mentioning certain “taboo scenarios” so as to preserve harmony within a group or an organisation. The obvious risk with this attitude, however, is that it can lead to blindness and therefore unpreparedness in the face of a looming disaster144. It is beyond the scope of this paper to try to evaluate whether this risk has become too great or, on the contrary, remains acceptable regar- 143 Karl Deutsch, Political Community and the North Atlantic Area: International Organization in the Light of Historical Experience, Princeton: Princeton University Press, 1957; Emanuel Adler, Michael Barnett, Security Communities, Cambridge: Cambridge University Press, 1998. 144 Paul Schoemaker, Philip Tetlock, “Taboo scenarios: How to think about the unthinkable”, California Management Review, 54(2), 2012, pp. 5-24. 36 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E Solidarity and deterrence in the Eastern Mediterranean ding the delicate question of collective defence among EU member states vis-à-vis Turkey; it is nevertheless the author’s hope that this report will have helped the reader to form his or her own opinion in that regard. F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E 37 Solidarity and deterrence in the Eastern Mediterranean Les opinions exprimées ici n’engagent que la responsabilité de leur auteur. 38 F O N D A T I O N pour la R E C H E R C H E S T R A T É G I Q U E
关注微信
