Between military deployment and democracy: use of force under the German constitution

ABSTRACT The German regime on the use of military force provides an important reference point for legal comparison. In a seminal judgment of 1994, the Constitutional Court identified a constitution-based requirement for each military deployment to have parliamentary approval. The formalities of the involvement of the Bundestag were, in 2005, codified in a statute. Recent German participation in coalitions of the willing have raised the question whether such operations are still covered by the constitutional bases, and participation in anti-Islamic State action in Syria is currently under review by the Constitutional Court. The article concludes that the tension between the need to effectively integrate military forces into multinational operations, democratic accountability, and judicial oversight has been uniquely resolved in the German constitution and statutory and case law. It illustrates the feasibility of upholding standards of democracy and the rule of law in foreign and military affairs.


Historical context and current practice
German history has produced a unique and complicated constitutional regime on the deployment of the German military abroad. In acknowledgment of its responsibility for the wars of aggression the nation waged since 1939, and in line with the initial complete demilitarisation of the country after defeat in 1945, the state is constitutionally committed to world peace and to the renunciation of aggression.
Germany became a member of NATO in 1955. In 1956, it amended its constitution, the Basic Law of 1949, in order to allow the state to contribute to the integrated forces of NATO for defending Western Europe against a possible communist aggression (which never materialised). In 1968, the constitutional provisions on the use of the military both inside the country and abroad were again revised in the context of drawing up a so-called 'constitution on the state of emergency'. The text of the German constitution regarding military matters and armed forces is frozen in the status of those two historical moments. The resulting incoherence mirrors the intense political controversies of the time.
The end of the global East-West split went hand in hand with German reunification and the formal international recognition of the state's national sovereignty in 1990, 1 enabling the UN Security Council to overcome its prior paralysis and to authorise military action against threats to the peace. This contributed to a 'fundamental reorientation' 2 : both from the side of Germany's allies and in some quarters from inside the country, the expectation arose that Germany would actively engage its military towards achieving world peace and security. But the nation remained reluctant to participate in collective security effortsa reluctance rooted in the deeply pacifist, and some might say irresponsible, attitude of the population and the classe politique. Refusal of any military involvement had until the reunification of 1990 been masked with the somewhat legalist argument that the Basic Law prohibited it. 3 The Basic Law served 'as a shield and pretext' for Germany to avoid participation in UN peace missions. 4 After the national and international political and legal context had changed profoundly in the 1990s, sections of the German population gradually accepted this new responsibility, and scholars argued that military engagement within the UN would be constitutionally admissible. 5 Nevertheless, it remained impossible to amend formally the constitutional text so as to explicitly allow for and regulate such military action.
In this political deadlock, the German constitutional court issued in 1994 a truly law-making judgment. 6 By holding that the constitutional text did allow German military participation both in NATO and in UN peacekeeping missions (a thesis that had been deeply controversial up to that moment), while at the same time giving (without much textual basis in the Grundgesetz) the Bundestag (Parliament's first chamber) the last word on military operations abroad, the Constitutional Court opened the way for German participation in multilateral peace and security missions and in collective defence efforts. Since then, the piecemeal-like constitutional text on military matters, which is stuck in the status of 1956 and 1968, has been complemented by a judge-made regime 7 the principles of which were partly codified in a parliamentary statute of 2005. 8 Since the start of German foreign deployments in 1992 more than 130 mandates for foreign deployments have been issued and prolonged by the Bundestag, amounting to more than 60 different operations. 9 By 2015, the additional costs amounted to €19 billion. 10 In December 2017 alone, Parliament issued seven mandates. One of them, by way of example, is participation in UNIFIL Lebanon (involving up to 300 German troops; with 131 currently deployed). 11 Examples for operations in the framework of the EU common security and defence policy (on the basis of European Council decisions adopted under Article 42(4) and 43(2) TEU) include the anti-piracy operation ATALANTA 12 off the Somali coast (to which Germany agreed to contribute up to 950 troops; with 197 currently deployed 13 ) and the anti-human trafficking operation SOPHIA 14 in the Mediterranean Sea (German participation of up to 950 troops; with 66 currently deployed). 15 The most controversial operations were those that led to proceedings before the Federal Constitutional Court and some other deployments, namely German participation in the monitoring of compliance with the UN-authorised embargoes against Serbia and the monitoring of a no-fly zone over Bosnia andHercegovina (1992-1995), 16 7 See section 2. 8 See section 4. 9 Since 1990, eight parliaments have been elected, currently (since the elections of 24 September 2017, the 19th parliament is sitting, and has so far authorised 15 operations). See for a complete list of all mandates until 2012 with exact documentation Wiefelspütz (n 2) 357-75, later data based on the author's own research. are Article 26 (the ban on a war of aggression) 25 and Article 115a (declaration of state of defence). 26 Article 24(2) GG runs: With a view to maintaining peace, the Federation may enter into a system of mutual collective security; in doing so it shall consent to such limitations upon its sovereign powers as will bring about and secure a lasting peace in Europe and among the nations of the world. 27 The relevant sections of Article 87a GG read: '(1) The Federation shall establish Armed Forces for purposes of defense. […] (2) Apart from defense [Verteidigung], the Armed Forces may be employed [eingesetzt] only to the extent expressly permitted by this Basic Law.' 28 Article 87a(2) GG has so far always been considered as the second-best legal ground, as a residual constitutional basis for military operations abroad that cannot be called 'collective' in the sense of Article 24(2) GG. In fact, except for two evacuation operations (in Albania in 1997 and in Libya in 2011), Germany has, as yet, never acted outside a multilateral schemevery broadly conceived as acting alongside the UN and/or NATO, or as a minimum in a coalition of the willing. The government regularly cites Article 24(2) GG as the constitutional basis for German participation in these operations.
In the lead judgment of 1994 (which will be discussed in section 2.2), the Court defined a system of 'mutual collective security' 29 as one whichthrough a peace-securing system of rules and the establishment of a specially dedicated organisationcreates for each member a status of being bound by international law, by which all of them mutually commit themselves to secure peace and to provide security. 30 Put differently, in order to qualify as a system of mutual collective security in the sense of Article 24(2) GG, an arrangement must possess three characteristics: (1) its rules must be directed at securing peace; (2) it must be institutionalised, and (3) it must impose legally binding rules. 25 Article 26(1) GG runs: 'Acts tending to and undertaken with intent to disturb the peaceful relations between nations, especially to prepare for a war of aggression, shall be unconstitutional. They shall be made a criminal offense.' 26 This provision was inserted into the Basic Law by amendment of 24 June 1968. Section 1 reads: Any determination that the federal territory is under attack by armed force or imminently threatened with such an attack (state of defense) shall be made by the Bundestag with the consent of the Bundesrat. Such determination shall be made on application of the Federal Government and shall require a two-thirds majority of the votes cast, which shall include at least a majority of the Members of the Bundestag.
The precursor norm was Article 59a GG of 19 March 1959 which foresaw the declaration of the state of defence by the Bundestag and which was repealed in 1968. 27 Emphasis added. 28 Emphasis added. This contribution uses the official English translation by Christian Tomuschat, David P Currie, Donald P Kommers, in cooperation with the Language Service of the German Bundestag. 29 Article 24(2) GG. 30 BVerfGE 90, 286 (n 6) paras 231 and 236, and holding no. 5a); author's translation.
The Court settled that the constitutional concept covers not only the UN but NATO also. 31 It is notably irrelevant whether the intention of the treaty scheme is to guarantee peace among members (as with the UN) or whether it obliges members to furnish support in case of attack from the outside (as with NATO). 32 The Court also found that the WEU qualifies, 33 whereas the post-Lisbon EU 'does not yet take the step towards a system of mutual collective security', despite the TEU's collective defence obligation exactly mirroring NATO's. 34 A mere coalition of the willing without any link to the UN or NATO would not qualify as a system in terms of Article 24(2) GG. 35 Lacking a treaty basis and possessing no institutions, it could not be called a 'system'. A bilateral treaty, for example between Germany and Iraq, would not qualify either.
Next, Article 24(2) GG forms not only the basis of German membership to the system but also constitutionally covers the individual deployments 'insofar as these occur within and pursuant to the rules of such a system'. 36 This condition is easily satisfied by UN peacekeeping operations. 37 Article 24(2) GG also covers peace-enforcement actions, namely military operations authorised by the Security Council under Chapter VII of the UN Charter. 38 The constitutional clause further covers operations conducted within NATO (Article 5 and non-Article 5 operations 39 ) and finally NATO actions implementing UN enforcement decisions. 40 Before the 1994 Constitutional Court's seminal judgment, scholarship struggled to bring into line the facially restrictive constitutionnotably the seemingly clear wording of Article 87a GGwith the emerging practice of military activity abroad (such as airspace monitoring and humanitarian aid), and attempted to declare constitutionally admissible German participation in peace and collective defence missions.
Interpretative strategies ranged from either construing the notion of 'defence' in Article 87a GG very broadly so as to include, for example, rescue operations for German nationals abroad; reading the notion of BVerfGE 90, 286 (n 6) para 231 and holding no. 5a). 33 Ibid, para 232. Scholarship further counts the OSCE as a system in the sense of Article 24(2) GG. 34 BVerfGE 123, 361 -Lisbon (2009), para 390. See Article 42(7) TEU, subsection 1 sentence 1 TEU. Arguably, practice since the Lisbon-judgment has been treating the EU as a system of collective security in the constitutional sense. 35 See section 3 for the problematic operation against IS which has some link to the UN through UN SC res 2249. 36 See n 49. 37 Peacekeeping was the object of the lead judgment BVerfGE 90, 286 (n 6) para 240. 38 This follows from BVerfGE 90, 286 (n 6) para 245 which mentions the UN Security Council resolutions 713, 757, 781, and 816 (all adopted under Chapter VII). 39 As long as these can be counted as practice developing the NATO Treaty as opposed to tacitly amending it. See n 245-n 246 and accompanying text. 40 See in scholarship for the various types of NATO operations Christian Hillgruber 'Article 24' in Hans Hofmann and Hans-Günter Henneke (eds), GG Kommentar zum Grundgesetz (Carl Heymanns, 13th edn 2014) paras 45, 51. employment very narrowly, 41 for example by distinguishing military from police operations; or by qualifying small-scale activities as a mere 'usage' (Verwendung) of the army as opposed to employment (Einsatz).
It was also pointed out that the original intent of Article 87a GG 42 was chiefly or even exclusively concerned with the deployment of the Bundeswehr domestically, and therefore no obstacle to military non-defensive engagement abroad. However, the plain wording of Article 87a(2) GG does not distinguish between domestic and foreign deployments and thus stands against limiting the applicability of the constitutional rule to purely internal deployments. 43 To conclude, the facial constitutional prohibition of any non-defensive military action until the early 1990s furnished a convenient explanation for not participating in UN peace missions, as these were not expressly permitted by the Basic Law.

The lead judgment of 1994: the invention of the parliamentary prerogative
In the eminently important judgment of 1994, 44 the Constitutional Court found a way out of the impasse. It clarified the key parameters both for the admissibility of military deployment (and the substantive requirements flowing from the choice of the legal basis) and the respective powers of government and Parliament when deciding on deployments.
This judgment was rendered in a dispute between constitutional organs, 45 upon requests by two parliamentary factions and their members, and joined together four proceedings concerning the German government's decisions to contribute to various military operations. 46 All operations were authorised by the UN Security Council (partly mediated by implementation decisions of NATO and WEU) and were located in two different settings: Yugoslavia and Somalia. Germany contributed planes, ships, and troops. For example, German soldiers participated in an integrated NATO unit in AWACS planes (Airborne Early Warning and Control System) to monitor the no-fly zone over Bosnia and Hercegovina, German ships patrolled the 41 In this article, the author uses the terms employment, deployment, dispatch, and operation as synonymous translations of the German word Einsatz.

42
In 1956, when the 'rearmament' provision was adopted, the text figured in then Article 143 GG. 43 The  45 'Organstreitverfahren', see section 6 for detail on this type of proceeding. 46 See n 6. A. PETERS Mediterranean Sea to enforce an embargo against the then Federation of Serbia and Montenegro, and German supply and transport battalions distributed humanitarian aid under UNOSOM II. Hence, the judgment dealt exclusively with military operations in the framework of international organisations proper, as opposed to unilateral action or coalitions of the willing.
As regards the constitutional admissibility of extraterritorial military activity and their legal basis, the Court made four points: first, that NATO is a 'system of mutual collective security' in the sense of Article 24(2) GG, although the drafters had the UN in mind; second, the parliamentary statute under Article 59(2) GG, which approved of German accession to NATO in 1955, 47 also covers the incorporation of military staff into integrated units and their participation in military action conducted by NATO and under NATO command to the extent that such integration or participation has already been pitched in the founding treaty; 48 third, in a kind of implied powers argumentation, the Court held that Article 24(2) GG, which expressly allows the German Federation to 'enter' (einordnen) into such a system of mutual collective security, must likewise implicitly allow the country to assume those tasks that are typically connected to membership importantly, this means that Article 24(2) GG is also the constitutional basis for the use of German armed forces in operations 'insofar as these occur within and pursuant to the rules of such a system'; 49 and fourth, Article 87a GG does not bar such operations. 50 Having clarified that the German constitution allows extraterritorial operations, the next set of findings concerns the constitutional powers of the executive and legislative branch in this matter: first, the out-of-area action under NATO's new strategy of 1991 did not require a parliamentary statute under Article 59(2) GG because this new strategy did not constitute a treaty amendment. 51 However, the bench (Senat) was split on this point. It has already been the case that in the 1994 judgment, four of the eight judges opined that the powers of Parliament were undermined by the purely executivist development 47 Article 59(2) GG demands a parliamentary statute for the ratification of certain types of important international treaties, namely those 'that regulate the political relations of the Federation or relate to subjects of federal legislation'. 48 BVerfGE 90, 286 (n 6) para 238. 49 This English translation is taken from BVerfGE 121, 135 (n 59) para 62, which uses the identical German phrase as BVerfGE 90, 286 (n 6) para 255: 'im Rahmen und nach den Regeln' (no official translation available of the 1994 judgment). See section 3 for detail on this requirement. 50 Ibid, paras 253-56. The Court left open the doctrinal relationship between the two constitutional provisions which can be construed in two ways so as to allow foreign deployments: Either Article 24(2) GG might be qualified as an 'express permission' as required by Article 87a GG (although Article 24 does not say anything expressly on deployments), or Article 87a GG (of 1956/1968) might be read as regulating only the use of the armed forces inside the country, and therefore as leaving completely untouched the clause of Article 24(2) GG, which is older (it dates from 1949). of NATO. 52 In order to accommodate this concern, and to avoid that the executive branch could have decided alone on military deployment, the Court ruled that 'The Basic Law requires a constitutive parliamentary approval for any military deployment of armed forces.' 53 This requirement corresponded to constitutional tradition since 1918, the Court said. 54 Secondly, the Court described in detail which types of military action demanded parliamentary assent in each concrete case. Thirdly, it asked the legislature to draw up a statute, which was finally adopted a decade later in 2005. 55 Any assessment of the judgment must note that the parliamentary prerogative 56 was unknown in the constitutional landscape until its invention by the Constitutional Court. 57 It was a stretch to extrapolate from Parliament's 'classic' right to declare (defensive) war 58 the requirement of parliamentary approval for every single military action. But this extrapolation facilitates military engagement to protect peace and security, which was in principle favoured already by the original constitution even though the founders did not anticipate when and how Germany would become part of this collective endeavour. Hence, the creative reading adapted the constitution to totally new circumstances after 1990, something that had not been foreseen in 1949. The requirement of parliamentary approval was not the only possible answer to the global change, but it was an admissible progressive interpretation of the Basic Law (probably transgressing the blurry boundary with judicial lawmaking) in line with the constitutional spirit.
2.3. The AWACS II judgment of 2008: when is parliamentary approval required?
The second key judgment was rendered in 2008, 59 four years after the adoption of the Statute on Participation. 60 It was, again, issued in a proceeding of a 52 The four defeated judges found the Constitutional provision on parliamentary ratification of treaties to be applicable in an analogous fashion to the strategic documents expanding the NATO mandate independent of their qualification as a treaty under international law or not (BVerfGE 90, 286 (n 6) paras 296-310). 53 Ibid, para 324 (author's translation). See on the parliamentary power ibid, paras 319-50. 'Constitutive' means that the approval by the Bundestag is a necessary basis of the constitutional legality of the deployment. If it is missing, deployment is unconstitutional. 54 Ibid, paras 325, 329. dispute between constitutional organs. It related, again, to a German contribution to airspace monitoring by AWACS, this time over Turkey in the spring of 2003. The context was the Iraq war waged by the US and its allies. Turkey feared that Iraq might target its territory and had requested consultations under Article 4 of NATO. Airspace surveillance with German planes was thus seen (including by the Court) as preparing for possibly needed collective self-defence under Article 5 of NATO. 61 In March 2003 the German Chancellor, Gerhard Schröder, explained to Parliament that the German military's exclusive duty was strictly defensive aerial surveillance of Turkish airspace and did not extend to providing any support for deployment in or against Iraq. There was 'a strict dividing line' between the NATOled AWACS aircraft, under the command of NATO's Supreme Allied Commander Europe (SACEUR), and the US command. The geographical separation and the difference in mandates was, the Chancellor said, the reason why there was no need for an approval by the Bundestag. 62 The applicant, a faction of the liberal Opposition party Freie Demokratische Partei, (FDP), successfully complained against the government for bypassing Parliament. 63 The Court found that German involvement in the aerial surveillance of Turkish airspace was a deployment of armed forces. Even if no actual combat took place, German forces were involved in armed operations and this would have required parliamentary approval. 64 The first and major contribution of this judgment was to flesh out in more detail when exactly a parliamentary approval was needed. Doctrinally, this hinges on the question of when a 'deployment' is present. 'Deployment' is the key term of the 2005 statute, but it is a constitutional (and not merely a statutory) concept which goes back to the 1994 judgment. 65 Second, the judgment offered a new rationale for the involvement of Parliament. It moved away from the bookish constitutional tradition on armed forces and placed the approval squarely into the context of democracy and the separation of powers. 66 This judgment had the consequence of further strengthening the Bundestag.

61
The Court which framed the case as explained probably got the international legal aspects of the case wrong: on the premise that the attack by the United States on Iraq was unlawful, Iraq had the right to self-defence. This might have lawfully included some action spilling over to Turkey. Then, self-defence would not have been permitted against such action.  The respondent was the government composed by the Social Democratic Party and the Green Party. It should be noted that the Social Democratic faction had, a decade earlier, when it was in the opposition, successfully sued the conservative (CDU/CSU) government and obtained the 1994 judgment. In the AWACS II-proceedings, the Court issued an order denying the requested provisional measures on the merits. Plaintiffs had sought an injunction to compel the government to seek approval of the lower house or otherwise immediately discontinue the operation. But the Court found the negative consequences of an injunction stopping the operation to prevail (BVerfGE 108 (n 59) 34-52). 64 BVerfGE 121, 135 (n 59) para 83.  The judgment on the Treaty of Lisbon examined the newly designed European scheme for a Common Security and Defence Policy (Article 42-46 TEU) and whether a German ratification of the EU reform treaties of 2007 would be compatible with the Basic Law. 67 Article 42(1) TEU allows for 'missions outside the Union for peacekeeping, conflict prevention and strengthening international security in accordance with the principles of the United Nations Charter'. The undertakings are further detailed in Article 43(1) TEU which, inter alia, mentions the 'tasks of combat forces'. The concrete missions and measures must be decided by the Council (Article 43(2) in conjunction with Article 42(2) sentence 1 TEU). The actual performance then falls on the member states. 68 Article 42(7) TEU contains an 'obligation of aid and assistance', which explicitly refers both to Article 51 of the UN Charter and to 'commitments under NATO'. The principle of voluntariness governs: only willing and able member states will be entrusted and must agree among themselves and in association with the High Representative of the Union for Foreign Affairs and Security Policy on the management of the tasks. 69 The Court examineddetached from any concrete casethe treaty's compatibility with the Basic Law's provision 70 on collective security. 71 The scheme manifests, according to the Court, the EU member states' 'intention to retain the sovereign decision on the deployment of their armed forces which is rooted in the last instance in their constitutions'. 72 Under the TEU, any common defence operation would have to be unanimously decided by the European Council. 73 This decision would only have the legal force of a recommendation, and would need to be implemented by the EU member states 'in accordance with their respective constitutional requirements'. 74 Should some member states propose an amendment of the EU Treaty to abolish the requirement of unanimity for a Council decision on this matter, then '[t]he Federal Republic of Germany would be constitutionally prohibited to take part in such a treaty amendment', the Court said. 75 In other words, majority voting in the European Council on a 67 BVerfGE 123, 267 of 30 June 2009 (-2 BvE 2/08 -, -2 BvE 5/08 -, -2 BvR 1010/08 -, -2 BvR 1022/08 -, -2 BvR 1259/08 -, -2 BvR 182/09 -), Organstreit proceeding. 68 Article 42(1) sentence 4 TEU: 'The performance of these tasks shall be undertaken using capabilities provided by the Member States.' 69 Article 44(1) TEU. 70 Article 24(2) GG. 71 BVerfGE 123, 267 (n 67) paras 254-55 and 381-91. It found the EU not to form a 'system of collective security' in terms of Article 24(2) GG (ibid, para 390). 72 Ibid, para 384. The Court here pointed to wording and legislative history of the new TEU. 73 Article 42(2) sentence 2 TEU. 74 Article 42(2) sentence 3 TEU. 75 BVerfGE 123, 267 (n 67) para 391. common defence operation could only be introduced after a German constitutional amendment. 76 The Court notably interpreted Article 42(7) sentence 2 TEU ('This shall not prejudice the specific character of the security and defence policy of certain Member States') as catering for the German requirement of parliamentary approval to deployment decisions. 77 Moreover, 'the constitutive requirement of parliamentary approval for the deployment of the Bundeswehr abroad is not open to integration'. 78 This means that any concrete deployment decisionalso within an EU operationwill always require a specific approval by the Bundestag. This principle cannot be overturned by any (potential) supranationalisation of collective security and by the primacy of EU law over German law as allowed by Article 23 GG. 79 The requirement of parliamentary approval may not be circumvented by secondary EU law either. 80 The judgment stated obiter dictum (possibly by mistake) that deployments abroad, if outside a system of mutual collective security, are constitutionally permitted only 'in case of national [territorial] defense [Verteidigungsfall]'. 81 This seems to preclude other types of operations that may be permitted by international law. 82 The best view is that, rather than outlawing these forms of military force, the Court simply left them aside as negligible for the purposes of this judgment, which concerned the EU and European integration. 83 In sum, it is doubtful whether the judgment really drew a 'red line' 84 for military supranationalisation; in any case it created legal confusion.  BVerfGE 123, 267 (n 67) para 386. 78 Ibid, para 255 ('integrationsfest'; official English translation). 79 Ibid, para 255. 80 Ibid, para 387. 81 Ibid, para 254. The phrase is misleading, because it is the technical term used in Article 115a GG (territorial defence of Germany), and does not cover the broader scenario permitted by Article 87a GG ('Verteidigung' ['defence'] tout court). whether the approval of the Bundestag is constitutionally required for urgent military actions once they are completed. Under the statute of 2005, the government is empowered to decide alone on a deployment in the case of 'imminent danger' and 'for the rescue of humans from specific danger when their lives would be endangered by the public involvement of the Bundestag'. 86 The statute says the Bundestag must be informed before and in the course of action. It also says that the government must 'immediately raise the request for approval. If the Bundestag rejects the request, deployment must be terminated'. 87 Most commentary had opined that the obligation to ask for approval would also govern deployments that are completed. 88 But the Court followed the opposing scholarly view 89 and held that in this scenario the Constitution does not oblige the government to seek an ex post approval, but only to furnish timely, complete, detailed, and written information to the Bundestag as a whole. In this judgment, the Court for the first time defined the legal effects of a parliamentary approval and clarified, and maybe slightly modified, the rationale of the parliamentary prerogative. It also sharpened the concept of 'deployment', which triggers the need for parliamentary approval, denied a political or military margin of appreciation of the government, and for the first time spelt out the parameters of the constitutional obligation to inform.
The doctrinal explanation was that the government's lawful use of its urgency powers has the same legal effect as the exercise of the combined governmental and parliamentary power 90 in normal situations. It is an 'auxiliary' stand-alone power which 'modifies' the principle of parliamentary codecision in situations wherefor purely factual reasons -Parliament cannot exercise its competence. 91 If an operation is ongoing, parliamentary approval or disapproval has only an ex nunc effect. In the words of the official translation of the judgment: 'The required immediate involvement  of Parliament after a deployment has begun […] does not have the legal effects of a retrospective decision, namely that if such retrospective approval were denied, the deployment would have been illegal from the beginning.' 92 The rationale and effect of parliamentary involvement as identified by the Court points against the need for an ex post approval of operations that are already terminated. Normally, the purpose of the parliamentary prerogative is that Parliament decides jointly with the executive and influences the concrete shape of the operation. Once the operation is over, this function can no longer be fulfilled. 93 By contrast, the purpose of parliamentary approval is not to assess authoritatively (verbindlich) the legality of an operation; this is incumbent upon the Court. 94 Political control by Parliament can be exercised through other types of instruments, for example parliamentary resolutions or even a motion of censure against the government. 95 Importantly, the Court for the first time ruled that parliamentary approval is needed for all deployments of German armed military forces abroad ('warlike' or not), and for deployments outside a system of collective security; 96 indeed approval is needed for 'every unilateral deployment' 97irrespective of its military or political relevanceincluding, notably, purely humanitarian operations. 98 The legal threshold, namely at what point a 'deployment' calls for parliamentary involvement, is identical in all cases; it is a 'uniform threshold'. 99 The executive branch does not enjoy a margin of purely military and political appreciation that would be beyond the reach of Parliament and of the Constitutional Court. Such leeway does not exist, not even in situations of emergency. 100 The key concept of 'imminent danger' is a legal concept, too; it is fully reviewable by the Court and does not leave space for political assessment by the government. 101 Applying these principles as a benchmark, the Court found that the rescue operation in Nafurah was indeed a 'deployment' in the sense of the Basic Law and the relevant statute of 2005, not least because 'at the time of the executive's decision on the deployment the likelihood of having to use armed force was particularly high'. 102 The non-occurrence of actual combat activity 92 Ibid, para 87. 93 Ibid, para 99. 94 Ibid, para 99. 95 Ibid, para 101. 96 Ibid, headnote 1; see also para 66. 97 Ibid, para 69. A parliamentary deliberation is even more needed in operations that have not been discussed with allies (ibid). The question whether there was an enabling provision in the constitution for the unilateral operation (which is controversial) was not submitted for decision in this proceedings (para 69); see section 3 on this question. 98 Ibid, paras 71 and 80. 99 Ibid, para 77. 100 Ibid, para 70: The Constitution 'does not […] grant any interpretative leeway'. 101 Ibid, paras 91-94. 102 Ibid, para 115. did not alter the fact that there had been a concrete expectation that German soldiers would be involved in combat during the evacuation. 103 The core contribution of the judgment is the new constitution-based obligation to inform. In order to allow Parliament to exercise the various forms of political control, it is the duty of the government to inform it in detail about military operations. Parliament enjoys an 'entitlement to information' 104 : the government 'must inform the Bundestag promptly and in a qualified manner about the deployment of armed forces'. 105 Furthermore, it must present the factual and legal considerations on which the government based its decision, and details regarding the operation (with the level of detail depending on its political and military importance) and its outcome. This must happen as soon as possible, to the Bundestag as a whole, and normally in writing. The information must be 'clear, complete' and such that it 'can be easily reproduced'. 106 The 2015 Pegasus judgment has been criticised on the grounds that the Court without sufficient justification cut back both the parliamentary prerogative 107 and also failed to do justice to the democratic and rule of law rationale of the requirement. 108 The critique is that even after completion of a deployment, a parliamentary approval could still deploy the legally relevant function of assuming 'democratic' co-responsibility for the entire operation. 109 The government would factor into its urgent decisionmaking the necessity to ask Parliament afterwards, and this prospect aloneforeseeable deliberation in the Bundestagwould influence government decisions. The control of legality by the Constitutional Court cannot fully replace scrutiny by Parliament because the scope of review is limited. 110 Against this criticism, this author would argue that a backward-looking political responsibility seems quite empty, and that a more extended legal control beyond the review of the Constitutional Court is not necessary. Intensified information to the Bundestag, as required by the judgment, is feasible 103 Ibid, paras 105-17. 104 Ibid, para 104, author's translation of 'der parlamentarische Informationsanspruch'. The government has an 'obligation of formal information' (para 103; official translation). 105 Ibid, headnote 4. Most likely, the information provided by the government in this case had not satisfied these conditions. However, because the plaintiffs had not even claimed insufficient information, this question was outside the subject matter of the proceedings (ibid, para 105). 106 Ibid, paras 103-04, quotes from para 104. 107 See for the critique the authors mentioned in n 109-n 110. In 2008, the Court had found that '[t]he German Bundestag must approve every deployment of armed forces without exception [ausnahmslos]' (BVerfGE 121, 135 (n 59) para 89). 108 See section 5 on the rationales of the requirement. and satisfies democratic and the rule of law principles. 111 With its clear statements against any executive leeway, the judgment does not mark a turn towards de-parliamentarisation.
2.6. The operations against the Islamic State in Syria and Iraq since 2015 and the stretch of 'collective security' Since 2015 112 the German military has participated in OIR against Islamic State (IS) in Iraq and Syria, deploying not only around 500 troops but a number of Tornado and aerial refuelling aircrafts, as well as one frigate. Action includes the training of Peshmerga combatants in Northern Iraq, aerial refuelling and surveillance, reconnaissance, the exchange of information, and evacuation. The Bundestag was properly involved and provided a mandate for up to 1200 troops, which has so far been extended until 31 October 2018. 113 In its formal request to the Bundestag, the German government relied on both traditional justifications in the law of nations for using military force. It stated that the international legal basis for the deployment decision was 'Art. 51 of the UN Charter in conjunction with Art. 42(7) TEU as well as resolutions 2170 (2014), 2199 (2015), 2249 (2015) of the Security Council'. 114 From the perspective of international law, neither justification works easily for the military action in and over Syria, 115 and this has an impact on their constitutionality, as will be shown in section 3.
Members and factions of the parliamentary Opposition filed a complaint before the Constitutional Court against the deployment decisions in the context of OIR. 116 It is unlikely that the Constitutional Court will reach the merits of the case. 117 Should this happen, the Court might scrutinise the constitutional basis of the deployment and then would have to pronounce itself incidentally on its international lawfulness as well. 3. The alignment between the constitutional and international law on the use of force

Overview
The German constitution specifically hooks onto and refers to the international law of peace. It contains a 'constitutional command of peace' 118 and a constitutional principle of 'friendliness towards international law'. 119 Moreover, Article 25 GG says that the 'general rules of international law shall be an integral part of federal law'. This clause thus transports the international law-based prohibition on the use of force into domestic law. 120 A breach of the ban on force as a 'general rule' of international law will, however, not automatically lead to a violation of the constitution, because the 'general rules', while taking precedence over federal laws, still rank below the Basic Law.
Finally, the constitutional prohibition of a war of aggression 121 is 'tied to the concept in international law'. 122 The constitutional concept follows the definition of aggression as spelt out in the UN GA resolution of 1974 and in Article 8bis ICC-Statute after the amendments of Kampala. Any activity that evidently constitutes an aggression in the sense of international law is at the same time unconstitutional. 123 The result of the combination of these constitutional provisions is that any manifest violation of the relevant rules of international law in the area of the use of force 124 will normally also violate the German constitution. This means that German constitutional law cannot, as a rule, be more permissive towards the use of force than international law.
The more difficult question is whether the constitution sets a higher bar and is more pacifist than international law. As a historical matter, this assumption was and is widespread as a lesson to learn from Germany's past as an aggressor state. Because the Basic Law mentions only defence and 118 BVerfGE 104, 151 (n 51) para 159 (para 30 of the English shortened translation). This 'precept' is read out of numerous provisions that mention peace (notably the prohibition of aggression in Article 26, the preamble, Article 24(2), and Article 1(2) GG collective security, there is no obvious constitutional basis for other types of military action. 125 Also, the relationship between the two explicit constitutional bases among each other is not fully clear. Under international law, self-defence (Article 51 of the UN Charter) and Chapter VII of the Charter are two distinct tracks. Under German constitutional law, these two modes are not mutually exclusive; there is 'no strict antagonism between collective security and collective defense', as the Constitutional Court put it (referring to Article 24(2) and Article 87a GG). 126 Finally, the constitutional bases are not fully congruent to international law. Article 24(2) GG is broader than Chapter VII, while Article 87a GG has been understood by some constitutional lawyers to be narrower than Article 51 of the UN Charter. Because of a traditionally narrow reading of 'defence' under Article 87a GG, the alternative constitutional basiscollective security 127has during the history of German military deployments been continuously expanded. In all instances, the German government argued thatfrom a constitutional perspectivethe German deployments pursuant to these resolutions do 'occur within and pursuant to the rules' of a system of collective security and are therefore safely covered by Article 24(2) GG. 129 But this constitutional coverage has been doubted by critics. Qualms relate to the international legality of the operation and to the constitutionality of German participation, and both aspects are linked. Herein the present author will concentrate on OIR, the ongoing anti-IS operation in and over Syria and Iraq. 130 Resolution 2249 (2015) is itself not sufficient as an authorisation for the use of force against IS under international law. 131 The resolution is not based on Chapter VII and therefore had to use a milder wording ('calls upon' as opposed to 'authorises'). Thus, the Security Council did not exactly 'authorise' the operations, as the Constitutional Court had formulated in its lead judgment of 1994. 132 Moreover, the resolution specifically asked the member states to act 'in compliance with international law', which can be understood as referring to the parameters of self-defence (as opposed to enforcement action). Another special feature is that OIR is directed against a terror group (IS), and not against an attacker state, and therefore resembles collective defence as opposed to 'collective security', which is the object of the constitutional provision of Article 24(2) GG.
The non-activation of Chapter VII of the UN Charter might taint the constitutionality of the deployment. Does it kick the Security Council resolution out of the 'framework' of a 'system of mutual collective security', as a decision exactly not taken 'within and pursuant to the rules' of the 'system'? This is what the critics of German contributions to the anti-IS operation had argued in Parliament. They opined that the German deployment decision is not covered by the state's ratification of the UN Charter in 1973 and therefore lacks a constitutional basis (not covered by Article 24(2) GG). 133 Against this critique, it is submitted that the German contribution to the anti-IS operations is linked to the UN Security Council exercising a competence laid out in the Charter, and thus fulfils the Federal Constitutional Court's legal requirements for coverage by Article 24(2) GG, even short of 'authorisation'. 134 In its lead judgment, the Constitutional Court had said that UN membership will be the constitutional basis for German participation in operations 'when the competent organs of the UN assume tasks, competences, and powers that are laid out in the Charter'. 135 With resolution 2249 (2015) the Security Council issued a binding decision, albeit outside of Chapter VII. 136 The resolution was adopted unanimously and qualified IS as 'a global and unprecedented threat to international peace and security' (in the preamble). It used typical Chapter VII language, namely to 'take all necessary measures'. 137 The operation against IS, undertaken pursuant to this resolution, is multilateral. Germany never launched a unilateral strike. This satisfies the rationale of the constitutional provision of Article 24(2) GG, which is intended to make Germany a reliable ally, and which in turn requires that Germany does not shy away from military contributions. 138 The German contribution has therefore been givenit is submitted -'within and pursuant to the rules' of the UN and is thus covered by Article 24(2) GG. However, it is a borderline case that strains the constitution, notably because the operation against IS is 'collective' only in a minimal sense (with two UN members, the US and Russia, basically fighting against each other).
It notably does not and should not follow that Security Council resolutions simply condemning terrorism without suggesting any member state action would suffice to bring military deployments under the umbrella of Article 24(2) GG. But such an extrapolation can hardly be avoided once the slippery slope of reliance on a rather soft Security Council resolution has been taken.

The training mission in Northern Iraq of 2014
Article 24(2) GG had been stretched further with the ongoing training mission in Northern Iraq, to which Germany contributed from the summer of 2014 to the spring of 2018, 139 and which was in 2018 transformed into a mandate on capacity building in Iraq. 140 144 The interventions by MPs on international law in the Bundestag debate were muddled. The academic legal service of the Bundestag opined that the international legal basis was both intervention by invitation and Article 51 of the UN Charter (in the form of collective self-defence), and that no additional Security Council authorisation was needed. 145 Again, the constitutional basis of the operation is to some extent linked to the international legal basis. The government relied only on Article 24(2) GG. 146 But the government's view that the Security Council's presidential statement suffices to bring a military operation 'within and pursuant to the rules' of a system of collective security is far-fetched. Such presidential statements are not legally binding and are issued as political pronouncements exactly because no agreement could be reached in the Security Council, and because the Council President is not empowered to determine a 'threat to the peace' in terms of Chapter VII. It is better therefore to admit that the training mission in Northern Iraq during its initial eighteen months (before the adoption of UN Security Council resolution 2249) was not covered by Article 24(2) GG. 147 In contrast, it should and could have been based on Article 87a (2) GG, which would have meant entering 'constitutional virgin territory', as the Parliament's academic service put it. 148 It has rightly been criticised that the 'key' of Article 24(2) GG risks becoming a passe-partout. 149 The breaking point of Article 24(2) GG might have been reached with OIR. To avoid a further dilution of the constitutional concept of a 'system of mutual collective security', and instead of using Article 24(2) GG as a mere facade, it might be better to 'activate' Article 87a GGwhere possibleas a legal basis. However, Article 87a, to which we now turn, is also fraught with problems and notably does not offer a clear solution for anti-terror operations such as OIR.

'Defence' under Article 87a GG
Article 87a GG only covers operations 'for purposes of defence'. 150 The interpretative question is what 'defence' means here. The constitutional notion of 'defence' is a concept of domestic law and as such autonomous from international law. On its face, the constitutional principle of friendliness towards international law would seem to require that the interpretation of the constitutional term should follow the evolution of international law. Indeed, both the German Federal Administrative Court 151 and prevailing commentary assumes that the constitutional term 'defence' in Article 87a GG is fully congruent with the international law of self-defence, 152 and therefore would seem to obey the same yardstick of legality. This discussion has involved two questions. Who may be defended? And against whom? Both questions have emerged simultaneously and have been merged, because all controversial German contributions to collective action outside the framework of Article 5 of NATO or Chapter VII of the UN Charter (after 2011 in Afghanistan and after 2014/2015 in Syria) took place far away from German soil and were directed against terrorist attacks.

Defence of Germany and other states
The first question is whether defence in terms of Article 87a GG means only defence of German statehood. A conventional understanding had been that the constitution only allows military action needed to defend against attacks that threaten Germany as a stateeven if remotely. 153 This assumption led a German minister of defence in 2002 to proclaim, with regard to German engagement in Afghanistan, that the security of Germany is also defended at the Hindukush. 154 Today, basically all scholarly commentary acknowledges that Article 87a GG covers actions to defend Germany as a state, to defend German allies (Bündnisverteidigung, notably within NATO 155 ), and that Article 87a(2) GG also covers the collective defence of non-allied states even if this requires operations outside and far away from the country. This would mean that collective self-defence, for example in favour of France (as an ally) and Iraq (asking for military support) in 2015, is permitted under Article 87a GG as long as it is covered by the international law of self-defence. 156 3.3.2. Defence against non-state actors, notably in the operation against IS 'Defence' as a constitutional concept might be read as including defence against non-state actors. Along that line, the constitutional concept would follow a putative evolution of the international legal concept of self-defence moving towards allowing self-defence against large-scale armed attacks by terrorists (with a tenuous link to a state, or even without any attribution to the territorial state in which defensive military action unfolds).
Proponents of relying on Article 87a GG as the basis for defence against terrorist threats argue that this constitutional construction would avoid the overuse of the constitutional concept of 'collective security', bring the constitution more in line with the two distinct tracks (collective security or selfdefence) available under international law, and open the way for more military engagement. The preference of German political actors always to rely on Article 24(2) GG had the historical motive of signalling to its citizens and to the world that Germany was firmly integrated into multilateral structures and would not go it alone. Such a signal and reassurance is arguably no longer needed.
On the other hand, important arguments speak in favour of sticking to a narrow reading of the constitutional concept of 'defence'. The chief consideration is that the international law of self-defence is currently unclear and in flux. This counsels against schematically linking the constitutional concept to the (blurry) international one. 155 A source of confusion was that NATO has, after the disappearance of the communist threat, expanded its range of action with help of a row of 'new strategies'. The member states have thereby mandated NATO to go 'out of area' both in a geographic and a material sense, notably responding to non-state threats such as terrorists and pirates, far beyond the traditional armed attack by states which Article 5 NATO in conjunction with Article 51 UN Charter originally addressed. A proper Article 5 operation has actually never been conducted. Article 5 NATO was invoked by the North Atlantic Council so far once in history (on 12 September and 4 October 2001, after the terrorist attacks of 9/11). Since then, the casus foederis ('case for the alliance') has not been formally lifted. The German Left and Green parties occasionally but in vain tried to bring the German government to work towards the termination of the casus foederis within NATO bodies. See Zur Dauer des NATO-Bündnisfalles, BT-WD 2-3000-142/08 (13 November 2008). 156 See the scholarly voices in n 152.
The problem shows up in the anti-IS OIR. The international legal basis conjured by the German government, namely collective self-defence for Iraq (and initially also for France), 157 is problematic. In its letter to the Security Council, Germany argued: ISIL has occupied a certain part of Syrian territory over which the Government of the Syrian Arab Republic does not at this time exercise effective control. States that have been subjected to armed attack by ISIL originating in this part of Syrian territory, are therefore justified under Article 51 of the Charter of the United Nations to take necessary measures of self-defense.
The letter claimed that 'support [to] the military measures of those States that have been subjected to attacks by ISIL' was a lawful exercise of the right of collective self-defence by Germany. 158 In the Bundestag's debate some days earlier, the German government and the parliamentary majority heavily relied on a legal opinion by the parliamentary academic service in which that service opined that 'obviously, last but not least against the background of the recent Paris attacksan evolution of customary law is ongoing' in the direction of admitting self-defence against non-state actors. 159 But this evolution is less obvious than the academic service claims. The international lawfulness of self-defence against armed attacks by non-state actors without any imputation to the state on the territory of which the defensive action strikes back is disputed. 160 This uncertainty is mirrored in the controversial interpretation of Article 87a(2) GG. The literature is actually split on this point. Some authors allow defence against terrorist attacks without further conditions. 161 Others opine that military reactions against terrorist attacks count as 'defence' in the sense of Article 87a GG only if the threat resembles an inter-state armed attack. 162 Finally, a sizeable number of commentators insists on the need of attributing the attack to a state. 163 However, they do not explain which criteria of attribution must be satisfied so as to trigger 'defence' in terms of Article 87a GG.
Presuming that an evolution of the international law of self-defence is ongoing (which is controversial) and that a novel, broader reading of Article 51 of the UN Charter is gaining ground, 164 it is not clear that the constitutional concept of 'defence' automatically follows. The parliamentary oppositional faction of Die Linke (the Left Party), in its pending complaint to the Constitutional Court, argues that, should the Constitutional Court follow the broad (and fairly novel) interpretation of Article 51 of the UN Charter in the direction of admitting self-defence against armed attacks by non-state actors, this would be an inadmissible judge-driven evolution of the UN Charter. Then, so the argument runs, this judicial law-making would no longer be covered by the initial approval of Germany's accession to the UN Charter by the federal statute of 1973, but would require a fresh approval by the legislative branch or at least by the Bundestag. 165 Against the fluent and uncertain state of international law, in order to be safely in conformity with international law on the use of force (as the Basic Law prescribes 166 ), German constitutional law needs to remain restrictive as long as the presumed evolution of international law towards extension is not yet firm. This author would therefore argue that, for the time being, 167 Article 87a GG only covers operations that respond to ongoing or imminent armed attacks by a state or are attributable to a state under the 'classic' uncontroversial Nicaragua criteria. 168 164 See the reference to scholarship in n 160. The author leaves aside the additional problem of a putative evolution of underlying customary law and its interplay with the interpretation of the UN Charter. She also leaves aside the view that Art. 51 of the Charter has always covered self-defensive actions against non-state actors, so that the current practice does not constitute any expansion. The assessment of past (pre-9/11) incidents is mixed, and the ICJ leans towards limiting self-defence to actions against states. 165 Organstreit proceeding, launched by the parliamentary faction Die Linke against the deployment decision (complaint lodged on 31 May 2016; proceeding is pending). The complaint is directed both against the Federal government and against the Bundestag as respondents. The applicants seek the declaration that the respondents, with their joint deployment decision (of 1 and 4 December 2015), violated competences of the Bundestag flowing from Article 24(2) in conjunction with Article 59(2) GG. Applicants' memo of 31 May 2016, at 110-20. 166 See n 118-n 124 and accompanying text. 167 As long as the state of international law has not shifted towards a clear espousal of the lawfulness of self-defence against non-state actors, Art. 87a of the Basic Law could not easily be amended and complemented by a phrase which clarifies that 'defence' in the constitutional sense includes defence against terrorist attacks. Besides being in political terms highly unlikely, such an amendment would risk of infringing international law. 168  and at the same time avoids constitutional practice violating international law (and by extension the constitutional principle of friendliness towards international law). In contrast, anti-terrorist and anti-piracy operations need to be based on Article 24(2) GG and can thus be conducted only within a collective, as a multilateral action broadly conceived. Thereby, because the scope of the constitutional authorisation under Article 24(2) GG is much broader than Chapter VII of the UN Charter, Germany is able to shape a practice that might contribute to the crystallisation of the permission of self-defence against non-state actors.

Constitutional law is more restrictive than the international law on the use of force
Other types of military action that are under some conditions allowed under international law, notably rescue operations of own nationals, intervention by invitation, humanitarian interventions in the framework of the responsibility to protect, and anti-terror and anti-piracy operations, are more difficult to justify under the Basic Law. Subsuming all these actions under Article 24(2) GG or, as recently suggested, under Article 87a GG is a stretch. Firstly, and contrary to a scholarly suggestion, Article 87a GG should not be seen to cover the 'defence of nationals' 169 (twinning territorial defence), that is, operations to evacuate and rescue nationals abroad. So far Germany has twice conducted such operations, but the Constitutional Court itself did not pronounce on their constitutional basis. 170 Unilateral 'Blitz-type' rescue operations, especially non-combat evacuation operations, such as the aforementioned Operation Pegasus in Libya, are mostly (and rightly) considered to be permitted under the law of nations, although opinion is divided as to the legal explanation. Many scholars allow such action under the heading of self-defence. This qualification opens the door to Article 87a GG. If, however, the international lawfulness is sought elsewhere (notably in an independent rule of customary law), Article 87a GG would not be available. Then, a constitutional basis found either in Article 32 GG (conduct of foreign affairs), 171 in the right to life (Article 2(2) GG), 172 or in Article 25 GG (general rules of international law as part of the law of the land) would seem to work only on the premise (not shared in this article) that unilateral non-defensive action is allowed by the Basic Law, and not barred by Article 87a GG. 173 The issue illustrates 169 In favour of personal defence notably Depenheuer (n 153) paras 39-40. 170 The evacuation from Libya led to BVerfG Pegasus, but the proceeding was only about the division of competences between government and parliament and not about the constitutionality of the operation in the first place. See section 2.5. 171 Wiefelspütz (n 2) 183-85. 172 Fassbender (n 149) para 77. 173 Cf. on the doctrine n 50. the tensions arising from the mismatch between the international legal bases and the constitutional grounds for military action abroad.
Secondly, a military strike upon invitation (such as the request by Iraq in 2014 174 ) may only be conducted if covered by either of the two constitutional provisions as discussed previously. Thirdly, current international law does not recognise an independent allowance to undertake humanitarian intervention. Any military delivery of the responsibility to protect may only happen in the framework of Chapter VII of the UN Charter, 175 and this would then also be covered by Article 24(2) GG.
Overall, the constitutional situation for Germany is that the state may not undertake all forms of military operations that are allowed under international law, except if multilaterally embedded. This is fully in line with the constitution's wording and intent, based on Germany's historic legacy. In terms of legal policy, Germany should avoid hiding behind a sham 'collective security' that is no more than several states acting in parallel and should likewise avoid pursuing a 'defence interventionism'. 176

Procedures and modalities under the statute on parliamentary participation of 2005
The case-law has interpreted the Basic Law as creating a 'combined power' 177 of the government and Parliament's first chamber to decide on any military deployment abroad. 178 The 2005 statute on parliamentary participation (Parlamentsbeteiligungsgesetz) 179 says that any 'deployment of German armed forces outside the scope of the Basic Law requires approval of the Bundestag', 180 a requirement that flows 'directly from the constitution'. 181 The statute only regulates the 'form and extent of the Bundestag's participation', as its opening paragraph puts it. 182

The trigger concept of 'deployment'
The term around which the parliamentary approval turns is 'deployment'. The provision of § 2(1) of the statute defines this as: 'A deployment of armed forces 174  is present when soldiers of the German army are involved in armed undertakings or when the involvement in an armed undertaking is to be expected.' 183 The provision of § 2(2) defines which activities are not to be considered as a 'deployment'. These are, firstly, 'preparatory measures and planning' and, secondly, 'humanitarian aid service and auxiliary action [humanitäre Hilfsdienste und Hilfsleistungen]' with arms carried only for self-defence 'when it is not to be expected that soldiers will be involved in armed undertakings'. 184 In 2008, the Federal Court fleshed out in more detail the concept of deployment. It also established a legal presumption for the requirement of parliamentary approval. 185 'Deployment' can be distinguished from law enforcement activities by looking at the 'genuinely military' character of an operation and at the belligerence of the context. 186 The qualification as a 'deployment' is independent of the constitutional and international legal basis of the military operation. This also means that putative illegality under national or international law is irrelevant to the requirement of the Bundestag's approval.
The need for secrecy is no reason in itself to forego Parliament 187 but might justify the delegation to parliamentary committees whose sessions are not public and which can be committed to strict confidentiality, for example the committee on defence. 188

Concrete expectation of combat is sufficient
Most importantly, the 2008 judgment of the Constitutional Court clarified that it is irrelevant whether an armed conflict or combat was already happening. The 'concrete expectation that German soldiers will be involved in armed conflicts [bewaffnete Auseinandersetzungen] suffices', says the Court. 189 The statute uses the term armed undertakings (bewaffnete Unternehmungen). 190 What is meant here is combat activity, not the existence of an armed conflict in terms of IHL. The Court spelt out the statute and cut back the parliamentary prerogative slightly: if combat is only a remote and abstract 'mere possibility', then the action does not yet require approval by the Bundestag. 191 The expectation of combat must be 'concrete' and 'well-founded', with 'sufficient tangible actual evidence that a deployment […] may lead to the use of armed force' and a 'particular proximity to the use of armed force'. 192 Involvement in armed conflicts, in the sense of actually using arms, 'must be expected immediately'. 193 The assessment of probability must examine the 'aggregation of factual circumstances'. 194 In integrated NATO operations, the participation of German soldiers, for example in the aerial surveillance of an adjacent country, can quasi 'automatically' lead to their use of armed force, because once the soldiers are dispatched, Germany can no longer influence the course of events. 195 It is irrelevant whether combat actually takes place; what counts is the ex-ante assessment of factual and temporal proximity. 196 It is neither necessary nor sufficient that the deployed personnel carry arms themselves. The threshold to deployment can be reached even if members of the military only supply combat-relevant information, perform reconnaissance, or give orders for the use of arms. 197 In order to assess when Parliament needs to be involved, the most difficult line to draw is probably between the 'preparatory measures' and the point where deployment begins. Borderline activities are the movement of troops to the territory of an ally, the delivery of arms, training, surveillance, and the furnishing of information. Training missions, which are in practical terms highly relevant, will normally qualify as 'deployment' only if they take place in a dangerous environment (such as those in Afghanistan 198 ). Another example of a borderline case is Germany's participation in the EU 'Sea Guardian' operation in the Mediterranean. The 'multifunctional mandate' comprises surveillance, capacity building, and combatting terrorism and weapons smuggling. Government and Parliament treated this as a 'deployment' not least because it was in political terms controversial (due to its scope). 199 In the aftermath of 9/11, logistical activitiessuch as air transports of other nations' peacekeeping soldiers and of materials, as well as the dislocation of AWACS airplanes in October 2001 to the USwere considered to be below the threshold. In contrast, the dispatch of German soldiers and the transfer of AWACS aircrafts from their home base in Germany to a Turkish base with a view to monitoring the airspace over Turkey in 2003 was determined to be a deployment by the Court because of the fear of a spill-over from the Iraq war. 200 An important question, linked to the definition of the threshold, is the timing of parliamentary approval. If approval is sought late it risks ratifying a fait accompli. 201 But when exactly does an operation reach the point that the Bundestag must approve it? Under the statute, the request by government must be 'in time [rechtzeitig] before the beginning of the deployment'. 202 An example is the training of Kurdish Peshmerga forces. The German government decided on 31 August 2014 to support training in Northern Iraq by furnishing military equipment, weapons and ammunition to the Peshmerga. Earlier in August, six German soldiers had been dispatched in Northern Iraq. In September 2014, further 'non-lethal' equipment was delivered and more soldiers stationed. In parallel, Iraqi Peshmerga combatants were trained in Germany. 203 The government requested approval by the Bundestag only in December 2014, specifically for the 'training mission in the Kurdistan-Iraq region'. Approval was then granted by the Bundestag in January 2015. 204 At the opposite end of the spectrum, approval sought and given very early would risk becoming a blanket authorisation. So only the right timing will allow the Bundestag to embrace responsibility in a meaningful way.

Formalities
Normally, deployments are authorised for one year. The government starts to prepare the decision for prolongation several months before the lapse of this time period. The governmental request must contain information on the mandate, the territory, the international and constitutional legal basis, the maximum number of troops, the capacities, the planned duration, and expected costs and financing. The government diligently complies with this and even goes beyond it, for example by explaining the political context. 205 The governmental requests are treated like parliamentary bills except that the second and third reading in the plenary are joined. 206 Four committees (foreign affairs, defence, budget, and legal affairs) are usually involved in the preparation.
Practice shows that lengthiness in Parliament is no real problem. Normally, the processing of a request by the government runs over several weeks. But the rules of procedure of the Bundestag allow for accelerated treatment under certain conditions. 207 For example, the approval of OIR took only four days. 208 The decision in the Bundestag must be taken by a simple majority. 209 The statute foresees a simplified procedure for 'deployments with low intensity and moderate consequences', 210 which requires involvement of two parliamentary committees, the foreign affairs committee and the defence committee. This has very rarely been used for prolongations of peace missions 211 and has not become particularly practical. 212 The Bundestag can make an explicit reservation or formulate a condition, and thus tie its approval to specific factual or legal circumstances. Indeed, the government has occasionally asked for approval explicitly conditioned on the persistence or renewal of a Security Council mandate, for example for participation in peace missions in Sudan and Lebanon. 213 This means that the approval would expire should the Security Council not renew its authorisation. Otherwise, the approval will expire eo ipso only when circumstances have manifestly changed: an easy eo ipso expiration would create legal insecurity and thus compel the government to constantly seek renewal. 214 Also, in less obvious situations of change, expiration is unnecessary because Parliament can always make use of its right to recall; 215 the Bundestag remains the 'master of its assent'. 216 Recall is a matter of political discretion (within the limits of the law) and may become relevant after elections. If Parliament recalls, then the government is obliged to withdraw as quickly as possible. The remote risk of incurring state responsibility, for example for failure to honour commitments under international law, is inherent in the constitutional division of competences.
The clause of § 5 of the Parlamentsbeteiligungsgesetz allows deployment without prior approval of the Bundestag in two scenarios: imminent danger 217 or rescue operations, when public involvement of the Bundestag would endanger the lives of those to be rescued. 218 In these two cases, government must inform the Bundestag, 219 and must promptly ask for subsequent approval for the operation. If the Bundestag refuses, the operation must be terminated. 220 This statutory scheme implies that the urgent or rescue operation is still ongoing and not yet completed when the approval is sought. The Constitutional Court's Pegasus judgment of 2015 clarified the legal situation, notably the obligation to inform, for completed rescue operations. 221 To conclude, the practice of parliamentary approval leans towards the extensive involvement of Parliament. Government reckons with the fact that the parliamentary Opposition will reclaim the prerogative, and therefore readily ask for approval in order to avoid a complaint before the Federal Constitutional Court. Practice also shows serious and engaged debates which have not become ritualised. The only routine are mandates on the basis of Security Council resolutions (Chapter VII-authorisations or peacekeeping). 222

The aborted statutory reform
Against the background of the intensification of military engagement, the need for rapid action (for example within the NATO response force), and the prospects of pooling military capacities (such as foreseen by the Permanent Structured Cooperation (PeSCo) of the EU), 223 and constant worries about Parliament obstructing the reliability of Germany as a military ally, a reform of the 2005 statute was ventilated but ultimately abandoned. The reform bill of 2016, 224 based on a parliamentary committee report of 2015, 225 sought to define operations below the threshold of deployment that would not need parliamentary approval. The bill also sought to codify the Pegasus judgment by prescribing the ex post information of Parliament on urgent rescue operations. Finally, it foresaw an ex post evaluation of deployments, reporting about secret operations, and annual reporting about 'multilateral military alliance capacities'. The reform agenda had been from the outset criticised by Die Linke and Die Grünen (the Green Party) as a manoeuvre to undermine parliamentary power. Parliament could not and should not surrender its political responsibility for each and every single deployment by adopting a statute which would make government more of a gatekeeper. 226 The reform petered out in 2017, not least because constitutional law experts doubted its compatibility with the nuanced constitutional case-law. The possible gain in legal security would have been lost through the risk of judicial proceedings (Organstreit) before the Constitutional Court.

The requirement of parliamentary approval
The requirement of parliamentary approval to military deployments was in 1994 created through judicial law-making as an informal constitutional amendment. 227 What are its rationales?

Separation of powers
In Germany, as in other democratic states, foreign affairs have traditionally been viewed as the domain of the executive. The perceived needs for secrecy, swiftness, uniformity (speaking with one voice), and the demands of military alliances (Bündnisfähigkeit) 228 stand in tension with parliamentary debate, which is open, slow, and a cacophony of many voices.
Still in 1984, the Constitutional Court admitted a 'tendency of parliamentarisation in the formation of foreign affairs objectives' but nevertheless strictly limited the involvement of Parliament to the approval of specific treaties, as explicitly foreseen in Article 59(2) GG. 229 It denied any further powers of Parliament in the field of foreign relations and rejected the idea of an 'erroneous all-encompassing parliamentary prerogative' by pointing to the separation of powers, and it postulated a 'core field of executivist responsibility'. 230 Ten years later (in 1994), parliamentary involvement was still seen by the Court as an exception to the allocation of foreign affairs to the executive. 231 Today, this view has been overcome: both foreign and military affairs are generally regarded as a mixed power. 232 The German political system is a parliamentary democracy in which the government emerges out of Parliament and mirrors the result of the elections to Parliament. In such a system, the separation between the legislature and the executive does not have great political significance. The parliamentary majority does not need an additional instrument because its political viewpoint is catered for in the government itself. So, politically speaking, parliamentary approval is a tool of the Opposition. This tool enjoys judicial protection because the Opposition possesses standing to instigate an Organklage proceeding claiming the violation of parliamentary power. 233

Shifting rationale: from military to democracy
In the course of the almost twenty-five years since the invention of the requirement of parliamentary approval, its justification and rationale has evolved. In 1994, the Constitutional Court originally anchored the requirement in an ostensible German constitutional tradition and framed the requirement as a feature of the constitution of the armed forces (Wehrverfassung 234 ). The conceptualisation, which prevailed from 1994 to 2008, placed the power to deploy the military outside the area of foreign affairs and thus avoided contradicting the aforementioned traditional view that foreign affairs was the domain of the executive. 235 While this legal pedigree rendered the requirement of approval acceptable to traditionalists, it was more an academic postulate than living constitutional law. The German Parliament had the constitutional power to declare war only during a brief period of the Weimar Republic (1918)(1919)(1920)(1921)(1922)(1923)(1924)(1925)(1926)(1927)(1928)(1929)(1930)(1931)(1932)(1933), though in practice never used it. German constitutional reality was such that the Reichs President held the dominant power over the army. By inventing the 'parliamentary army' in 1994, 236 the Federal Constitutional Court belatedly satisfied the nineteenthcentury quest for a constitutionalised citizens' army and elevated a liberal aspiration to constitutional law. 237 In later judgments and decisions, this special doctrinal construct was abandoned. Parliamentary approval was fitted into the pattern of the rule of law and democracy and thus 'normalised'. 238 In our times of increasing global interdependence and the blurring of the spheres of internal and external political affairs, the need for a thick democratic basis of 'foreign' relations is generally acknowledged. The involvement of the Bundestag in military dispatches is no longer an anomaly but in hindsight turns out to be a forerunner 239 of an ongoing parliamentarisation and democratisation of foreign affairs, which is visible, for example, in the publication of traditionally secret treaty negotiations that have come under intense public and parliamentary scrutiny.

Making good for executivist NATO transformations
The requirement of parliamentary approval is also a corollary and a counterpoint to the government's monopoly on developing the relevant treaty regimes, notably NATO. Parliament has been kicked out of participation in making the strategic decisions, that is, the successive NATO strategies by which the range of NATO activity has been much expanded. Inversely, it has gained the ultimate say on each single deployment. During the epochal change in the decade following the end of the Cold War, the executive branch's power in the military sphere was amplified through constitutional interpretation in two ways: firstly by facilitating foreign deployments through a broad reading of Article 24(2) GG, 240 and secondly by acknowledging the executive's power to 'develop' international treaties, notably NATO, without the involvement of Parliament on the basis of a narrow reading of Article 59(2) sentence 1 GG. In this context, the function of the prior approval of Parliament on the domestic level is to counterbalance the executive's power.
Under the Basic Law, important international treaties, namely those 'that regulate the political relations of the Federation or relate to subjects of 236 BVerfGE 90, 286 (n 6) para 321. 237 Fassbender (n 149) paras 90-91, 94-98. 238 Kleinlein (n 110). 239 Ibid, 60. 240 See section 2. federal legislation', need to be approved by federal statute (Article 59(2) GG). The function of this statute is to make the treaty part of the domestic legal order and to secure the democratic basis of the treaty which was negotiated and concluded by the executive. 241 While NATO had regularly adopted strategic concepts since its foundation in 1949, it was only after 1989 that 'new' strategic concepts led to a pronounced substantive and geographic expansion of the organisation's activity. 242 Thereby, the organisation reacted to the melting away of the military menace posed by the Socialist Block, which had been its raison d'être, by refocusing on the emergence of new threats, notably global terrorism. NATO here gave allowance to 'non-Article 5 operations'. Because these operations are beyond the substantive and geographical scope of the collective defence against an armed attack under Article 5 and 6 of NATO, they have come to be called 'out-of-area' operations. 243 From a democratic perspective, the question is whether these strategic concepts are still covered by the original treaty and thus merely 'developed' the treaty-based system of collective security. If not, they would have actually changed the system and thus amounted to (informal) treaty revisions (amending treaties), which, given their political importance, fall under the rule of Article 59(2) sentence 1 GG and would thus have required a consenting statute by the German Parliament. The Constitutional Court had to deal repeatedly with this question. 244 It stated that even important strategic documents that are not treaties will not require a parliamentary statute (or any other form of participation by the Bundestag). A statute is needed only for the conclusion of treaties (including amending them). 245 241 In the tradition of considering foreign affairs to be a prerogative of the government, parliamentary participation in treaty-making was seen as an exception, as an extraordinary interference of parliament in the executive's domain (BVerfGE 1, 372, judgment of 29 July 1952 -2 BvE 2/51 -Deutsch-Französisches Wirtschaftsabkommen, paras 96-97; BVerfGE 1, 351, judgment of 29 July 1952 -2 BvE 3/51 -Petersberger Abkommen, para 83). 242 After 1989, new strategic concepts and similar documents were adopted in 1991, 1999, 2006, 2009, and, most recently, 2010. 243 The term 'out-of-area' confusingly referred to an extended geographic area of action, to an extension of tasks, and to the extensions of members of NATO. It was in the German debate mostly used to denote military action beyond collective defence as foreseen in Article 5 and 6 NATO Treaty and notably referred to the German participation in UN peace missions or sanctions, partly implemented by NATO. 244 The leading case is BVerfGE 104, 151 (n 51), official English translation shortened. See already BVerfGE 90, 286 (n 6) paras 257-95 (no official English translation). 245 The Court opined that the application of the clause on treaty-making (Article 59(2) GG) in an analogous fashion to the informal evolution of international treaties (and thus requiring a formal parliamentary statute of assent) would give rise to legal uncertainty, stymie the executive, and would run against the separation of powers (BVerfGE 104, 151 (n 51) para 149 in the German text; para 20 in the English shortened official translation). The follow-up question then is when a treaty (as opposed to a political or soft law document) is present. The Court here looked at the 'intention to be legally bound' as the decisive criterion (ibid, German text paras 133-38) (which it found to be absent). This is almost completely circular because the task is exactly to find out whether and when the parties intended to conclude an amending treaty to the NATO Treaty.
From the perspective of the Basic Law, every German military contribution to each specific operation that 'grows out of the seeds planted in the founding treaty' 246 enjoys a democratic basis. The reason is that German participation is covered by the initial parliamentary statute (under Article 59(2) GG) assenting to the founding treaty of the relevant organisation (such as the UN, NATO, and the EU). This also means that an ultra vires operation would not only be outside the particular organisation's founding treaty but at the same time outside the parliamentary statute that brought that treaty into the domestic legal order and would thus lack constitutional and democratic legitimacy. 247 The gist of the newer Constitutional Court decisions on military deployments is that the parliamentary decision on each single military deployment is seen as compensating for the overall strategic decisions lacking parliamentary foundation. 248 The AWACS II judgment of 2008 exposes this: German participation in the overall strategic direction of NATO and in decision-making as to specific deployments of the alliance is quite predominantly in the hands of the Federal Government […] But the freedom of the Federal Government to structure its alliance policy does not include the decision as to who, on the domestic level, is to determine whether soldiers of the Bundeswehr will take part in a specific deployment that is decided in the alliance. By reason of the political dynamics of an alliance system, it is all the more important that the increased responsibility for the deployment of armed forces should lie in the hand of the body that represents the people. 249 The Lisbon treaty judgment of 2009 holds (in the awkward official English translation) that '[p]articularly sensitive for the ability of a constitutional state to democratically shape itself are decisions […] on the disposition of the monopoly on the use of force by […] the military towards the exterior'. 250 Curiously, the division of labour between the executive and Parliament has been reversed. Normally the legislature sets the general rules and the executive implements them. With regard to foreign military engagement, Parliament has been kept at bay from setting the rules in the form of developing NATO, and instead now decides on concrete cases of deployment. 251 To sum up, the question of democratic legitimacy has arisen not only with regard to concrete operations but with regard to strategic changes in relating to NATO. The involvement of Parliament has been designed as 'an essential corrective', as the Court put it 252 : as a counterweight to the executivist grand design. Today, this compensatory function has moved to the backgrounduntil another transformation of NATO happens to occur.

Non-delegation à l'allemande: 'Wesentlichkeitstheorie'
The German conception of democracy requires important, or 'essential' (wesentlich), political decisions to have a parliamentary basis. The jurisprudential 'rule of essential matters' (Wesentlichkeitstheorie) is the functional equivalent to non-delegation doctrines. It was developed with a view to restrictions of fundamental rights and has more recently been applied to the projection of military force abroad. The Lisbon treaty judgment of 2009 uses the term 'wesentlich' by stating that '[t]he deployment of armed forces is of paramount importance for the individual legal standing of soldiers and of others affected by military action and involves danger of far-reaching implication'. 253 It is not difficult to see that a military deployment is indeed an important decision, for foreign politics, for the lives of soldiers and their licence to kill in armed conflict, and their ensuing obligations under international humanitarian law. Now neither the German constitution nor the 2005 statute offer much guidance for deployments. This indeterminacy of the legal basis is normally problematic under Wesentlichkeitstheorie. The requirement of parliamentary approval is apt to compensate for the weak substantive legal limits on foreign deployments by furnishing fine-tuned deliberative and public procedures for taking decisions. 254 By endorsing the government's request, the Bundestag does not give a one-off assent but 'assumes an ongoing co-responsibility'. 255 Political (democratic) accountability is thus supplied. The essentiality rationale notably explains why parliamentary approval is needed for all deployments independent of NATO (purely within the UN framework), including potential defensive action under Article 87a(2) GG. In those cases, executivist treaty development requiring a parliamentary 'corrective' is no issue. But because deployment is always essential, Parliament must be involved. from a comparative law perspectivea far reaching involvement of the judicial branch in foreign and military affairs. 256

Abstract control of norms?
It has been suggested 257 that the Bundestag's approval (not the government's request) of a deployment should be challengeable in a judicial proceeding called '(abstract) control of norms'. 258 The chamber's approval decisions belong to the body of federal law, but they are not 'norms' in a technical sense. 259 The argument in favour of subjecting these acts to judicial proceedings is that they function like parliamentary acts: as 'quasi-statutes'. 260 Opening this type of proceeding could fill a gap in judicial protection because the benchmark here would be the entire constitution (and not only the competences of the Bundestag). Finally, the standing of the governments of the Länder in this judicial proceeding might compensate for the lack of involvement of the Parliament's second chamber (which represents the Länder) in the deployment decision.
On the other hand, the qualification of the Bundestag's decisions as statutelike is unpersuasive not least because the chamber cannot modify the government's request and can only take it or leave it. 261 This does not have a lawmaking quality but rather amounts to 'military micromanagement' by Parliament. 262 The norm control proceeding does not fit and has so far never been tried. This situation is widely perceived to constitute a gap in the rule of law.
Against this background, and in response to the Bundestag's approval of the first anti-IS mandate in December 2015, Die Grünen suggested introducing (through amendment of the statute on the Constitutional Court (Bundesverfassungsgerichtsgesetz)) a new type of procedure before the Constitutional Court, specifically to review the conformity of deployment decisions with the constitution (and indirectly with international law), with standing for three quarters of the Opposition in the Bundestag. 263 But the proposal was defeated, and instead the Left faction tried an Organstreit against the anti-IS mandate, which will be explained in the next section.

Organstreit proceedings
The Organstreitverfahren ('dispute between constitutional organs') 264 is a weapon in the hands of the Opposition. It looms over every government and motivates careful legal assessment both in the formal written request for authorisation of a military deployment by the executive and in parliamentary debates. The double legal scrutiny performed by both branches of government is then often complemented or rectified by the third layer of limited legal scrutiny by the Court. This has so far happened to more than ten cases on foreign deployments since the early 1990s, sometimes accompanied by provisional measures.

Procedural requirements, standing, and limitation of judicial review
The Organstreit proceeding, inter alia, allows a political group (faction) in the Bundestag to file a complaint. 265 So far, every complaint against a military deployment has been instituted by one or several factions. The purpose of the Organstreit proceeding is to protect the competences of constitutional organs. This means that plaintiffs cannot directly attack the substantive decision of the government to participate in military action. 266 The only admissible grievance is that a constitution-based power ('right') of the Bundestag has been breached by the government. 267 The Court may thus only examine whether the constitutional competences of the organs have been respected and whether the procedure of parliamentary involvement was correct.
Historically, the first two cases arose because the government had omitted to seek the parliamentary chamber's approval or had not sought approval as a . Also, the Organstreit proceeding may neither be lodged by individual members of Parliament nor by a minority which is not a faction, and only against the government as a whole, not against the minister of defence (see for all this BVerfGE 90, 286 (n 6) paras 199, 204, 210-17). 266 The 'wrong' claim normally leads the Court to reject the application as inadmissible and is not treated in the merits. 267 § 64(1) BVerfGG: The application shall only be admissible if the applicant asserts that an act or omission on the part of the respondent violated or directly threatened to violate the rights and obligations conferred on the applicant or on the applicant's organ by the Basic Law. (Official translation) matter of constitutional obligation. 268 In both instances, one or several political factions complained that the Bundestag's right to participation was violated by those decisions. 269 Since then, governments have been careful to seek the approval of the Bundestag. The more difficult question now is under what conditions an Organstreit proceeding might be admissible and meritorious even after the Bundestag had approved a deployment.
6.2.2. Overstepping the mandate of the Bundestag It would seem admissible for a parliamentary faction to claim that the Bundestag's right to participation is violated when a concrete operation oversteps the confines of the Bundestag's prior approval in geographic, temporal, or substantive terms. So far, no judicial decision on a putative ultra vires operation has been rendered. The proceeding coming closest to this scenario relates to Kosovo. The ongoing German deployment is based on Security Council resolution 1244 (1999) and Article 24(2) GG. The Security Council authorisation itself is not time-limited. It is a constitutional practice to annually renew the mandate in the Bundestag. 270 When Kosovo declared its independence in 2008, the parliamentary faction of Die Linke instituted an Organstreit proceeding against the government, arguing that government should have sought a new approval of the Bundestag for German participation in KFOR because the new circumstances (Kosovan independence) had led to an expiry of the Bundestag's mandate. The Court ruled that only a manifest change of circumstances could lead to an expiration of parliamentary assent, and that the declaration of independence of Kosovo was not such a case. The Court was not allowed to examine the overall constitutionality or the international lawfulness of the deployment and only declared that the competences of the Bundestag had not been violated. 271

No overall constitutional control
The more difficult situation is that the faction claims that the deployment violated international law, and by extension also the parallel constitutional law. In 1998, the German CDU/FDP government decided to contribute to the air operations of the intervention troops formed by NATO members and under NATO leadership against Serbia and which were undertaken without a UN Security Council authorisation. The Bundestag approved it. 272 The PDS faction (Partei Deutscher Sozialisten, Party of German Socialists), having been outvoted in Parliament then instituted a complaint on the grounds that the 'humanitarian' intervention in Kosovo breached international law and thus also the constitutional ban on aggressive war. 273 The Federal Constitutional Court, however, found this complaint inadmissible because the Organstreit proceeding does not permit an 'overall constitutional review'. 274 Constitutional law as a whole is no benchmark in the Organstreit proceeding because the Bundestag was not installed to make sure that measures taken by the executive are in conformity with the Basic Law. 275 6.2.4. The logic and review of 'excess' In contrast, the 2007 Tornados judgment 276 introduced a specific type of constitutional scrutiny that even considers clear breaches of international law in individual deployments. The background to the Tornados judgment was that the UN Security Council had in 2003 geographically extended the ISAF mission to Afghanistan as a whole, and NATO had agreed to continue carrying out the extended mission. This led to a territorial overlap with the US Operation Enduring Freedom (OEF). In parallel, NATO had successively revised its overall strategy in the direction of responding to global security and novel threats, notably at the 2006 Riga summit. 277  of data gathered by German surveillance to OEF 'only if necessary for the successful realisation of the ISAF-operation or for the security of ISAF-forces'. 278 The grievance of the suing factions PDS/Die Linke (as interpreted and partly reformulated by the Constitutional Court) was that the government's decision to participate in the extended ISAF mission overstepped the statutory approval of Germany's original accession to NATO (in 1955) and thus violated the rights of the Bundestag.
The Court examined both the regional extension of NATO activity beyond the Euro-Atlantic area and its substantive orientation towards peace. An excess in either dimension would not be covered by the original NATO treaty and thus not covered by the parliamentary statute approving Germany's accession to the organisation and would thus escape parliamentary co-responsibility. The Court first performed a 'constitutional review as to whether essential structural decisions of the NATO Treatyin this case a connection to NATO's regional purposehave been exceeded'. 279 Importantly, the Court did not review the actions of NATO in themselves but limited itself to a 'review of the connection between the NATO actions and the regional framework'. 280 On this point, the Court found that NATO had not 'departed in a general sense from its connection to a specific region'. 281 Notably: [t]hose responsible in connection with NATO were and are entitled to assume that the securing of the rebuilding of Afghanistan's civil society also contributes directly to the Euro-Atlantic area's own security; in view of present-day threats from globally acting terrorist networks, as 11 September 2001 showed, threats to the security of the NATO area cannot any longer be territorially restricted'. 282 Second, NATO's (putative) turning away from peace would constitute a substantive excess. 283 Such an excess would not only mean that NATO and its operations were no longer covered by the NATO Treaty (and the complementary federal statute) but also would violate the constitutional provision of Article 24(2) GG, which allows Germany to participate only in systems of collective security that are directed at maintaining peace. This would mean that Germany would have to leave NATO. If not, the powers of the Bundestag would be violated because it [the excess] removes the treaty basis of the alliance from the responsibility of the German Bundestag and in doing so infringes the German Bundestag's right under Article 59(2) sentence 1 of the Basic Law in conjunction with Article 24(2) of the Basic Law. 284 Importantly, this review allowed for and required an incidental assessment of international law. The reason is that a breach of international law through individual NATO military operations, 'in particular the violation of the prohibition of the use of military force, may be an indicator' that NATO has transformed itself and that it 'is structurally departing from its constitutionally mandatory orientation towards the maintenance of peace'. 285 On this point, the Court did not see any tendency of NATO to turn away from its objective to secure peace. Notably, the cooperation of ISAF with OEF did not change the character of the NATO Treaty because the two operations remained distinct in legal and factual terms. 286 The Court therefore rejected the application as unfounded on the merits. However, the judgment marks some extension of judicial review, the exact scope and impact of which is not yet fully clear.
This extension will be tested in a pending proceeding launched by the Die Linke faction against the anti-IS OIR, which had been approved by the Bundestag. The complaint filed in 2016 is directed both against the Federal government and against the Bundestag. 287 The applicant seeks the declaration that the respondents, with their joint deployment decision (of 1 and 4 December 2015), violated competences of the Bundestag flowing from Article 24(2) in conjunction with Article 59(2) GG. The faction argues that the deployment occurred outside a system of collective security 288 and would therefore have required a new statute 289 and that the deployment constituted an extensive reading of Article 51 of the UN Charter, which oversteps the parliamentary statute approving Germany's accession to the United Nations in 1973.
It remains to be seen whether the logic of 'excess' and 'departure' from the original federal statutes approving membership in the two organisations (NATO and UN)which is the only window to constitutional reviewwill suffice to confer standing on the faction. The related question is whether this logic will, on the merits, allow for an assessment, however summarial, of the lawfulness of self-defence against non-state actors (under the heading of an 'excess' beyond the original meaning of Article 51 of the UN Charter and Article 5 of NATO).
Besides the issue of 'overstepping' the federal statutes of 1955 (NATO accession) and 1973 (UN accession), the constitutional law claim in the IS case is that OIR finds no basis in either Article 24(2) GG, because it is not conducted inside a system of collective security, or in Article 87a (2) GG, because it is directed against non-state actors. However, the lack of a constitutional basis as such (beyond the logic of 'excess') is not reviewable by the Court in the Organstreit proceeding. 290

No constitutional complaints of soldiers
It has occasionally been argued that an individual soldier could file a constitutional complaint (Verfassungsbeschwerde) 291 with the argument that a military command ordering his participation in a specific operation violates his fundamental rights. 292 On the level of statutory law, the Statute on Soldiers regulates the military duties of obedience. A soldier may refuse to obey only when an order runs against human dignity, and he must refuse to obey when the order asks for the commission of a crime. 293 Orders of this kind are nonbinding. So far, no one has claimed that these statutory provisions are not in conformity with the constitution. The statute as such lawfully restrains the soldiers' fundamental right to personal liberty. 294 A war of aggression is a felony under the German statute on international crimes, 295 so that an order to participate in such a war would be nonbinding. Beyond this, most commentary opines that orders that request manifest grave violations of international law are nonbinding as well.
Against this legal background, it is unlikely that a constitutional complaint could be admissible let alone successful on the merits because the soldiers would (and could) first of all have to attack the commands by reference to the statutory law. In contrast, in a constitutional complaint, the Federal Constitutional Court only examines violations of fundamental rights, and the scrutiny is limited specifically to constitutional law. International law cannot become a benchmark for the Court simply via Article 25 or 26 GG. It is conceivable that a soldier's freedom of conscience 296 is affected by a also against the Chancellor) would have empowered the latter to request the dissolution of Parliamentbut confidence was expressed with only a two-vote majority. 303 Roughly speaking, the conservative CDU/CSU faction seems to construe the scope of parliamentary power a bit narrower than the Left factions. Finally, most Organstreit proceedings against deployments have been launched by Liberal, Left, and Green factions.
From a comparative law perspective, the German constitutional regime on the use of military force abroad is heavily legalised through case-law and codification. The dominant role of the Constitutional Court (which created the regime) stands out but corresponds to the overall strong position and eminent political function of this body in the German constitutional set-up. It is the Court that insists on the strong role of Parliament, with the 'joint decision-making power' 304 of the government and the first parliamentary chamber (Bundestag) on each single deployment. The judicial concept of a 'requirement of constitutive parliamentary approval' 305 on military operations abroad was in 2005 codified in a statute. 306 But the prerogative of the Court remains and has so far prevented a statutory amendment. The main reason for abandoning a well-prepared reform of the statute in 2017 was the consideration that the Constitutional Court demands a parliamentary decision in each individual deployment, and that generalisations enshrined in the statute would not suffice for the securing of political accountability.
In the German parliamentary system of government, in which the government is elected by Parliament and accountable to it, the case for an additional parliamentary decision on the use of military force might seem less obvious than in a Presidential system where the executive branch enjoys legitimacy independent from Parliament. But it finds its explanation in the idea of the 'parliamentary army'. Parliamentary involvement is not only superimposed onto a genuinely executive power but coconstitutes every deployment decision. The consequence for timing is that the approval of the Bundestag must be sought at the outset and not only after a certain lapse of time, such as after sixty days in the US, or after four months in France.
The tension between the potentially growing necessity to integrate a national army effectively into multinational forces on the one hand, and democratic accountability on the other hand has in Germany been resolved with an accent on parliamentary control. 307 This is compatible with the  304 See n 90 and accompanying text. 305 See n 53 and accompanying text. 306 See n 55. 307 See Miller (n 257) 205-06 for praise for the Basic Law's contribution to curbing Germany's militarist tradition. The 'reconciliation of military and democracy will have to count equally as one of the great successes of the German Basic Law'. extant treaties (NATO and TEU), which leave room for member states' specific constitutional requirements. 308 In a complicated and confusing legal web of international law, UN law, the law of NATO or EU, specific international agreements with host states, and domestic constitutional law, it is parliamentary approval which guarantees that every single military operation will be double-checked and allocates the ultimate political responsibility to the body that most represents the German people. From a cosmopolitan or transnational perspective, the strong role of Parliament is welcome. Although parliamentary involvement does not necessarily lead to greater consideration for foreign and global interests, it does lead to public debate. Together with the need for official governmental explanations in their formal request to Parliament, these procedures contribute to the expression of an opinio iuris of the state on matters of the use of force.
The historical reintegration of Germany into collective military action has been achieved. It seems that we are now entering a new political era on the use of force, characterised by the renewed blockade of the Security Council, and an increased acceptance of military action against terror groups under the guise of self-defence. This calls for reflection on how to adapt constitutional law to the new demands. The constitutional text's commitment to defence and to collective security and the historically motivated firm renunciation on unilateral or non-defensive military action makes it difficult to accommodate other types of engagement, ranging from operations to rescue German citizens abroad to anti-terrorist action and combatting piracy. The interpretation of the German constitution to tally every operation under the heading of either defence or collective security has become somewhat strained. Reminiscent of doctrinal suggestions before 1994, it is again proposed to construe 'defence' broadly so as to include defence against non-state (terrorist) attacks, and to use Article 87a GG as the constitutional basis for such action. 309 The German experience is relevant for the debate on the exceptionalism of foreign and, notably, military affairs and their law. 310 The normative claim of exceptionalism is that governance decisions in this sphere are and need only be subjected to less demanding constitutional standards: perceived or assumed requirements of speed, uniformity, and flexibility would, according to the exceptionalist thesis, suffer less democracy, less transparency, less checks and balances, less rule of law, less judicial control, and less respect 308 Article 11 NATO 'This Treaty shall be ratified and its provisions carried out by the Parties in accordance with their respective constitutional processes' (emphasis added). See for Article 42(7) TEU the text accompanying n 77. 309 BT-WD 2-3000-025/16 (n 152). 310 Cf. for human rights than other areas of law and governance. But this claim is weakened by the successful German practice, which shows that foreign affairs law need not follow a completely distinct logic of legitimacy than other constitutional law. This ties back to global constitutionalism, which holds that in a world characterised by global flows of information and digital hyper-connectivity, global supply chains, intense global trade, foreign investment, and migration, the starting point of any analysis should be that foreign affairs is not a categorically distinct type of politics but, rather, resembles 'world internal politics'. Therefore, prima facie, foreign relations measures and the law that governs them are and should be subject to the usual constitutional standards. The burden of explanation rests on those who advocate the exception and who seek to water down the constitutional standards governing foreign relations law. German constitutional law and practice in deciding on military deployments abroad demonstrates that it is sufficient to lower the standards only slightly.

Acknowledgements
The author thanks Jochen Frowein, Wiebke Rückert, and an anonymous reviewer for helpful suggestions on a prior version of this article, and her student assistants, notably Leander Beinlich, for their extensive research. A very short version of this article appears in Curtis A. Bradley (ed), The Oxford Handbook of Comparative Foreign Relations Law (Oxford University Press, 2019). The author thanks the handbook editor for the permission to use material from the handbook chapter for this article.

Disclosure statement
No potential conflict of interest was reported by the author.