Violation and confirmation of the law: the intricate effects of the invocation of the law in armed conflict

ABSTRACT Within the jus contra bellum there is an apparent contradiction between states’ verbal commitments to the law and the prevalence of armed conflicts. Taking this contradiction as a starting point, this article aims to provide empirical insights into how states invoke international law to justify their participation in armed conflicts. It develops a typology of how law can be confirmed by its invocation, taking an inductive approach based on case analysis. Do recent military interventions indicate a decline of international law? This article argues that there are three dimensions of confirmation. Firstly, law can be confirmed as an instrument of communication between states. Secondly, in a set of uncontroversial cases, the specific substantive rules of international law are confirmed through what is described as coherent practice. Thirdly, the article explains why even in controversial cases substantive rules may be confirmed through their invocation, even where the action is in fact illegal.


Introduction
What states do and what states say they do are often two different things. At the verbal level commitments to international law are numerous. At a practical level, however, military conflicts appear to be as frequent as ever, despite the existence of a prohibition of the use of force and despite states' general and repeated verbal commitments to that rule. Obviously we often face a significant contradiction between words and deeds. 1 It has therefore been a common theme in international legal scholarship to reflect on the possible death of the compliance of states with international law since the 1990s have led to numerous insights into whether and why states comply with various international legal regimes, 7 the jus contra bellum has largely been exempt. This is, however, certainly not because of a lack of thorough empirical studies, but because the assessment of the effects of invocations of the law appearsmore than in relation to other areasto be guided by pre-existing normative beliefs. Whereas liberals aim to uphold the law and therefore value states' verbal commitments, 8 positions with a (neo)realist drive put an emphasis on the practice and tend to view state declarations as mere 'cheap talk '. 9 Against this backdrop, this article serves a twofold purpose. It aims to provide empirical insights into how states invoke international law to justify their participation in armed conflicts over recent decades, examining a specific sample of conflicts that occurred between 1990 and 2014. 10 This empirical foundation will provide the debate with some numerical data in order to help assess the gravity and prevalence of the phenomena at issue. Secondly, the paper aims to make a conceptual contribution in developing a typology of how law can be confirmed by its invocationtaking an inductive approach based on the analysed cases. It will be argued that there are three dimensions to this typology.
First of all, and as the most general type of confirmation, we may understand the fact that states justify their political actions in terms of international law to be a confirmation of international law as the language of states, i.e. as an instrument of communication between states. States use the law to express and explain their actions and thereby confirm the status of international law as a general reference system, without, however, confirming specific substantive rules. Secondly, the analysis reveals that in a number of cases, states' participation in armed conflicts is uncontroversial and largely regarded to be in conformity with international law. In these cases, substantive rules of international law are confirmed through what may be described as coherent practice. The focus of this article will rest on the third and most challenging type of confirmation. It will explain why, even in controversial cases, the law may be confirmed through its invocation, even where the action is in fact illegal. However, it also will be shown that there are significant limitations to this confirmation hypothesis so that its actual scope is much more limited than implied by the ICJ or argued for by some scholars. A particular problem here are cases that may be described as cynical uses of the law, meaning that states verbally commit to the law, while in fact undermining it with their actions. This article will suggest how to deal with that apparent tension.
Section 2 begins by explaining the empirical basis and method of the analysis. Based on the empirical findings, sections 3-5 then explore the three different wayspreviously notedin which the invocation of norms can lead to a confirmation of international law. Section 6 then provides a brief overall assessment.

Empirical scope and method of the analysis
This article's method combines a quantitative approach (aiming to provide a glimpse into the prevalence of the approaches of states to international law in armed conflict) with a qualitative assessment (providing a theoretical categorisation of the invocations). Both aspects are pursued simultaneously in that the empirical findings are presented within the typology that has been induced from the analysed conflicts.
The difficulty in selecting a set of conflicts for the analysis is that one needs to be, on the one hand, sufficiently inclusive so as to include not only the obvious examples that have extensively been discussed in scholarship. On the other hand, in view of the countless small-scale violations of the prohibition on the use of force, the selection needs to be sufficiently discriminatory in order to allow the author to identify and work with a manageable number of cases, for which a qualitative assessment is possible because further information on the conflict and the legal justifications put forward by states is available. A number of organisations collect conflict-related data and provide databases or regular reports containing various conflict-related information. 11 The data on which the present analysis is built has been drawn from the Uppsala Conflict Data Program (UCDP), 12 using the 11 The most well-known and reputable compilations are provided by The Correlates of War Project (www. correlatesofwar.org/), the Heidelberg Institute for International Conflict Research (http://hiik.de/en/ index.php), the Center for Systemic Peace (www.systemicpeace.org/) and the Uppsala Conflict Data Program (http://ucdp.uu.se/). 12 The UCDP has been chosen for two reasons. Firstly, it allows for the identification of international conflicts, i.e. those that have a trans-border dimension and, therefore, are relevant for the state-centred framework of international law. In contrast, the reports of the Heidelberg Institute for International Conflict Research have a different emphasis, strongly focus on internal conflicts and do not clearly mark cases with an international dimension. Secondly, the UCDP allows one to identify not only major escalations, but also contains information on smaller confrontations (described as 'minor' confrontations, with at least 25 battle-related deaths per year). Thus, it also provides information on less intense confrontations, which nevertheless are relevant for the general prohibition on the use of force. online database, 13 as well as journal publications. 14 Based on these sources, conflicts were selected that cumulatively fulfil the following three criteria. First, the conflicts took place or were ongoing between 1990 and 2014. 15 This timeframe has been chosen as it marks a significantly changed situation of world politics, especially in regard to the end of the Cold War confrontation and corresponding major power shifts from a bipolar to, as it seemed then, a unipolar world. 16 It was also selected because of the corresponding emerging capacity of the Security Council to take action during this period (as demonstrated, e.g. in regard to Iraq's intervention in Kuwait). 17 Secondly, conflicts were selected where the confrontation resulted in at least 25 battle-related deaths (in at least one year of the conflict). This includes the categories of 'war' and 'minor conflict' as used by the Uppsala Conflict Data Program. This excludes, obviously, a number of interventions and conflicts that are relevant for international politics as well as for the international legal debate but resulted in fewer casualtiesfor example interventions carried out for the rescue of nationals abroad. However, this limitation was necessary for reasons of feasibility.
Thirdly, the conflicts selected had an international dimension, meaning that in addition to the state upon whose territory the conflict took place, at least one other state participated in the hostilities. The data available through the Uppsala Conflict Data Program provides an annual overview of the locations of conflict, the parties involved, the number of deaths and some further aspects. In order to get a view of the 'the big picture', these annual sets of data have been grouped into conflict situations, paying attention to a nexus in territory, actors and disputed issue. For example, the 13 The data set has been created using the online 'customized reports' tool of the Uppsala Conflict Data Program, drawing the data on 8 February 2016. This online tool allowed the selection of cases inter alia based on the intensity of the conflict as indicated by the number of casualties per year (minor con-flict=>25; war=>1000), the degree of international involvement (pure intrastate conflict, intrastate with international involvement, interstate), and the years in which the conflict took place. However, after a recent update of UCDP's website this tool is no longer available online: only the complete data sets can be downloaded. entire conflict between Kuwait and Iraq is treated as one 'case', including the initial invasion by Iraq in 1990, and the later international response of 'Operation Desert Storm' in 1991 aiming to repel Iraqi forces. 18 Based on these sources and criteria, 45 conflict situations were identified. 19 For these conflicts the current author then investigated the legal claims (if any) that the main parties involved have put forward, paying particular attention to the justifications that the major intervening state(s), i.e. the prima facie violators of the prohibition of the use of force have proclaimed. The sources for this investigation have been (in order of importance): (a) official statements and submissions to the UN (above all to the Security Council); (b) official state publications, such as press releases; (c) where such sources were unavailable, reputable newspapers containing declarations of state officials have been considered. In cases where multiple justifications were offered by 18 It is important at this point also to mention the challenges and limits in working with UCDP and other databases as a means for case selection. The notion of 'conflict' employed is often not in line with what is relevant from a legal point of view. The data is usually actor-based, not sufficiently taking into account the territory from which the actors operate, although this is of utmost concern for an analysis from a jus contra bellum point of view. For example, the UCDP lists the United States' operations against Al Qaida as one conflict within the category of war [JPR (2002) 635]. However, from a legal standpoint this is not the relevant perspective on the conflict, since the prohibition on the use of force does not protect Al Qaida as a non-state actor. From the perspective of the prohibition on the use of force, the relevant dimensions are the interventions in specific states in the fight against Al Qaida, particularly the intervention in Afghanistan (which UCDP lists as a separate conflict). Thus, for the scope of this article, the US conflict with Al Qaida could not be taken into account as a separate conflict. Another problem is that a complex set of data, such as the UCDP, almost necessarily has shortcomings in relation to the accuracy of the collected data. For example, the UCDP lists the conflict in Libya breaking out in 2011 as a merely internal conflict (i.e. not acknowledging an international dimension), although foreign states played a significant role in overthrowing the Gaddafi government and took active part in the hostilities by means of bombing campaigns. The same is true for the 2007/8 interventions of Turkey in Northern Iraq, which the UCDP does not list as a separate conflict of international dimension, but only refers to the general conflict between Turkey and the PKK [JPR (2009) 583]. For the present framework these, for the current author, obvious mistakes have been corrected. It must, therefore, be underlined that the identification of cases iseven when using databases like the UCDPnot exact science but also includes an evaluative component. one state or by various participating states, only the major justification has been ascribed to the case. For example, the 1999 Kosovo intervention was counted as a case in which states put forward a legal justificatory claim (namely the doctrine of humanitarian intervention) even though the majority of the NATO states only provided political justifications and did not claim legality under international law. Individual classifications could therefore give room for diverging views, but breaking down the conflicts into their major dimension proves necessary as any other approach would end up treating one conflict as multiple conflicts, depending on the number of participants and their respective justifications. Such complexity, however, would render the current aim of establishing an empirically grounded overarching understanding of the effects of invocations impossible.
The identified cases have been analysed for the following aspects. First, did the state(s) in question rely on or refer to international law in justifying their participation in an armed conflict? 20 The cases in which a reference to international law could be found were then confronted with further questions: secondly, was the law used in controversial ways (i.e. were states disagreeing about the application of the law?); and, thirdly, if states were disagreeing consideration was given to whether they relied on established norms and interpretations of the law or whether they invoked (as in the Kosovo case) norms that did not reflect the state of the law at the time.
This third question isas will be shownvery important for determining whether a rule is confirmed under controversial circumstances. A rule can only be confirmed when it is invoked in its established understanding. It is necessary to distinguish conflicts in which states dispute the abstract general content of a rule (especially in regard to the existence of a rule, to the inclusivity of a rule, to the priority among several rules, and to the interpretation of a rule) from disputes of a rule's application to the concrete case, including the establishment of the relevant facts. 21 These distinctions may sometimes be difficult to apply, but as a matter of principle they can be upheld. 22 For example, a dispute about the legality of humanitarian interventions takes place at the level of the abstract general content of rules. The dispute is about the existence of such a rule as a justification for the use of force, and is also about the priority of rules (namely of human rights protection vis-à-vis the protection of territorial integrity), as well as being about the interpretation of rules (e.g. concerning whether existing practice already speaks in favour of a reinterpretation of rules in light of subsequent practice). Territorial disputes, on the other hand, usually are about 20 Limiting the perspective to state actors was necessary as non-state actors are too numerous to fully take into account and because their actions do not have direct effects on the law. 21 See, for a discussion of these criteria, Francis Anthony Boyle, World Politics and International Law (Duke University Press, 1985) 108-12; Joel H Westra, International Law and the Use of Armed Force (Routledge, 2007) 22. 22 See, e.g. Christine Gray, International Law and the Use of Force (Oxford University Press, 3rd edn 2008) 10-1.
the application of the rule to the concrete case. States do not dispute the rules in general, but, rather, the application to the (potentially contested) facts or about the necessary concretisation of the abstract general provisions to the case at hand (e.g. regarding the question whether a certain response action was proportionate or not). We will subsequently see that this has significant effects in regard to the confirmation of the law.
Chart 1 provides a summary of the findings of this study in relation to these questions, in reference to the 45 conflicts under investigation herein. The data will be discussed and interpreted in detail in the following sections, but it is worth summarising some main results here.
The small black field indicates the cases in which states did not make reference to international law (3 cases = 7%). The transparent field on the left of Chart 1 marks cases in which states' participation in armed conflicts was uncontroversial, i.e. where other states did not object (or did not object significantly) to the legality of the intervention (19 cases = 42%). The two grey blocks to the right contain cases in which the invocation of the law was controversial among states. The dark grey field represents 16 cases (35%) in which states referenced established norms of international law, but contested their application to the case at hand. The light grey field represents seven cases (16%) in which states relied on new interpretations of the law that departed from the established state of the law.
Based on this empirical framework, the following three sections will discuss the different dimensions in which we may conceptualise how international law can be confirmed through its invocation.

International law as the language of states
International law is widely seen as providing for a language of states orin other wordsfor a 'grammar' in which states may articulate their political Chart 1. Total number of cases = 45; black field: no reference to international law (3 cases = 7%); transparent field: uncontroversial interventions (19 cases = 42%); light grey field: invocation of unestablished norms (7 cases = 16%); dark grey field: reference to established norms (16 cases = 35%).
views and justify their course of action in a form that is understandable to other states. Understood like this, international law works as an instrument of communication between states in which political and social claims are transformed into categories that are universally understandable. In particular, proponents of Critical Legal Studies view this as the crucial dimension of international law 23 but it is seen as an important function of international law way beyond work following that analytical approach. 24 The first question for this article therefore is: do states confirm international law, and more specifically the jus contra bellum, as an instrument of communication between states? Do they make use of international law and confirm its status as a 'language', or do they refer to other ways of justifying their actions (for example to political arguments)? The potential implications being considered here do not refer to the level of the substance of specific rules, but, rather, to the function of international law in more general terms. So how does such a confirmation work?
When states refer to international law in justifying their actions they declare different things. The most obvious and explicitly raised content is the claim for the legality of the state's action in a concrete case. A state might say, for example, that 'an attack against another state was lawful as it was carried out in self-defence'. However, the explicit content of such a declaration does not exhaust the communicative content that a state conveys. Language philosophy uses the term 'performativity' to uncover actions that can be performed while speaking and which can create social meaning or perform social practices. 25 In that sense, the explicit proposition comes with implicit claims that are inherent in the state's justification.
Consider, for example, the ongoing conflict between India and Pakistan over the Kashmir region. India claims to defend its sovereignty over Kashmir, which it regards as part of its sovereign territory as the result of a (controversial) accession in 1947. 26 Pakistan, by contrast, holds that the situation is still unresolved under international law and demands that the people of Kashmir be able to exert their right to self-determination. 27 One performative content that the mere use of legal categories implies in a situation like this is that international law is acknowledged as a relevant international framework for articulating and justifying a state's political course. International law is used as an instrument here, as a reference system that can be helpful in expressing and potentially explaining political actions to other actorsin the case of Pakistan, for example, the interest to allow the Kashmiris to exert a right to self-determination and to hold a referendum on the status of Kashmir. India finds different international legal principles relevant for the safeguarding of its interests, namely the principles of sovereignty and territorial protection. The law serves as the language that allows both parties to articulate their interests, to structure political conflicts, and potentially negotiate political solutions. Just by its invocation, the law is generally confirmed as a relevant international system. If it were not relevant, states would refrain from any 'law talk' and simply provide, for example, political justifications for their actionsan alternative states have and are aware of. 28 The results of the empirical analysis undertaken for this article are very much in line with a common sentiment, namely that international law is omnipresent in the declarations of states and plays a crucial role. As indicated, 45 conflict situations have been identified for this study, of which 19 were of undisputed legality. These uncontroversial conflicts will be discussed in more detail in section 4. Here it suffices to point out that the intervention wasin all of these 19 conflictsbased on the invitation of the territorial state. Such invitations were often expressed in general terms, for example in mutual defence treaties or on an ad hoc basis in order to allow for the provision of immediate military support. In one way or another, the governments involved relied on the expression of consent, which constitutes a legal act, and, therefore, relied on international law as a means of organising their relationships. In this sense, international law and the rules of the jus contra bellum, which acknowledge consent as a justification for the use of force on the territory of another state, were confirmed as the language of states.
It is also the case that in regard to the remaining conflicts in which the legality of the intervention was disputed international law mostly played an important role for the justification of a state's intervention. In fact, only three conflicts lacked at least some form of invocation of the law by the intervening states (7% of the overall 45 cases). Two of these cases were early 1990s conflicts for which material sources are scarce: one concerning the occupation and the related use of force by Syria on Lebanese territory taking place from the mid-1970s to the early 1990s, 29 and one related to support provided by 28 See also the approach of the ICJ, which carefully differentiated between legal and political justifications when analysing the justificatory claims of the US. See, e.g. Nicaragua (merits) (n 5) para 208. 29 Naomi Weinberger, Syrian Intervention in Lebanon: The 1975-76 Civil War (Oxford University Press, 1986) 209 (quoting Syria's Prime Minister, who said that Syria's intervention was 'motivated by nationalist and humanist sentiments, in response to the request of a group of citizens who were in a state of despair and fear, prompting them to appeal for assistance to sister Syria').
Burkina Faso to Charles Taylor's National Patriotic Forces of Liberia (NPFL) and their campaign to overthrow the government in 1990. For both cases no legal justificatory statements were traceable. The third case was the Operation 'Desert Strike' against Iraq, conducted by the US in 1996 in order to make Saddam Hussein 'pay a price' 30 for attacks on Kurdish-controlled territory and to prevent further Iraqi action. In this case the US provided political justifications, arguing with interests, not at all with the law. 31 The Kosovo interventiona further candidate for a lack of invocation of the lawdoes not fall into this group of cases. Though many states invoked mere political justifications aiming to legitimise the intervention, 32 arguments for the legality of humanitarian interventions overall were present and so we find legal justifications for the intervention. 33 In all the remaining cases, states justified their actions under reference to the law in one way or another, and thereby contributed to international law's general relevanceeven if the invocation was often not very precise and, in some cases, not especially proficient from a technical legal standpoint. Even in cases in which the intervention was almost unanimously considered to be illegal, states advanced legal arguments to justify their actionfor example Iraq for its invasion of Kuwait, 34 the US for its intervention in Iraq in 2003, 35 or Russia for its annexation of Crimea. 36 The question whether international law is invoked at all in armed conflicts points to a general danger that, for example, Operation 'Desert Strike' highlights. States can decide to act outside the framework of the law and without even relating their actions to the existing norms. This is the ultimate threat to international law, namely that the law loses its relevance for states and that political decisions are taken without reference to a legal framework. One could argue that on the plus side of this situation lies the fact that states act honestly and do not cover up their actions with a questionable appeal to international law. 37 However, such lack of reference to the law is the result of undisguised political power that does not consider a general normative framework to be the benchmark for its actions. Such actions, therefore, communicate an appeal to unilateralismwhich marks the end of the cooperative efforts that international law is based on. In any case, if the primacy of political justifications were to become general practice, law would ultimately fade into insignificance and another 'grammar', of prudent statesmanship for example, would take over. As of now, however, we see that some sort of invocation of the rules of the jus contra bellum remains the dominant practice for justifying the participation in armed conflicts.

Coherent practice as a confirmation of the law
The discussion in section 3 concerned the role of the jus contra bellum as a means of communication between states and, thus, it remained at a very general or even formal level. The more demanding question is whether and under which circumstances a confirmation of specific rules of international law takes place. This section therefore shifts from analysing international law as an instrument to analysing it as a system containing specific prescriptive rules that oblige states to behave in a certain way.
Consideration of the empirical dataset underpinning this article shows that in a significant number of cases, the rules of the jus contra bellum function as an uncontroversial normative framework for armed conflicts. States participate in military conflicts, and rely on an accepted legal justification for the use of force, and their legal claim is not (or only to an insignificant extent) contested, especially among the parties directly involved in the conflict. Through the congruence of a state's actions with the law and the explicit or (in part) implicit invocation of substantive legal norms the substance of the prohibition of the use of force is strengthened.
The analysis of the conflicts of the past 25 years shows that 19 out of 45 conflicts (42%) fall within this category. In all of the 19 cases states relied on the doctrine of intervention by invitation. Although not spelt out in the Charter, it is universally accepted that states may in principle consent to other states using force on their territory. 38 Many scholars, however, consider significant limitations to apply concerning the legality of interventions where these interventions aim at settling internal strife. 39 The rationale is that under these circumstances the intervention would violate the right to self-  40 This limitation does not, however, apply where the intervention aims at combating terrorism, as under these circumstances no violation of the right to self-determination takes place. 41 While the lines between internal struggle and terrorism may be thinespecially in view of the unsettled concept of terrorism 42the practice under analysis here shows that numerous cases of interventions did not spark controversies. In the 19 cases at issue now, the inviting state was (in almost all cases) targeting internal terrorist groups and sought the help of a foreign government. The vast majority of these interventions (16 cases) took place in African countries, 43 two in Asia 44 and one in the Middle East. 45 In these conflicts the legality was uncontroversial, not in the least because of the fact that states targeted non-state actors, which do not have a (formal) voice in assessing the legality of interventions. Moreover, the acting (mostly) Islamist terrorist groups do not have any open advocates in the community of states which makes a consensus on targeting them easy to achieve.
To name just a few cases, France was involved in a number of military conflicts, for example in the Central African Republic in 2006 (supporting the government against an internal rebellion), 46  . The intervention in Yemen is certainly also a borderline case. Some criticism was put forward (e.g. by Iran), but overall international reaction (by states) was still very limited so this case was counted as an uncontroversial intervention.  In all of these conflicts, practice of uncontroversial legality confirmed the rules of the Charter in regard to the prohibition of the use of force.

Violation and confirmation
Confirmation of specific rules by means of coherent practice only applies where the legality of an intervention is uncontroversial. But what happens in conflicts in which states do not agree about the legality of an intervention but, rather, express their diverging views in the 'language of the jus contra bellum'. Can we assume, under such circumstances, a confirmation of the substance of specific rules? The empirical basis for this section includes the conflicts in which states' invocation of international law was controversial, i.e. 23 conflict situations. 52 This section starts by presenting the basic rationale of the ICJ's confirmation hypothesis (5.1). In a second step, it is necessary to take a brief doctrinal look and determine how the invocation of the law in controversial cases may strengthen norms of international law (sections 5.2 and 5.3). Subsection 5.4 then develops the requirements and limits of the confirmation hypothesis and presents the empirical data on which this research is based. Subsection 5.5 provides a conclusion on the issue of confirmation and violation.

The ICJ's confirmation hypothesis
The most prominent articulation of the view that the invocation of the law leads to its confirmation, even if the actual conduct of a state is eventually found to be illegal, has come from the ICJ. The starting point for this hypothesis is that in the vast majority of cases, military interventions are not mere facts, but are accompanied by declared norms. States accompany their actions with a specific legal evaluation, according to which the intervention in question is not in violation of the law. The ICJ held in a much-quoted passage from its Nicaragua decision: If a State acts in a way prima facie incompatible with a recognized rule, but defends its conduct by appealing to exceptions or justifications contained within the rule itself, then whether or not the State's conduct is in fact justifiable on that basis, the significance of that attitude is to confirm rather than to weaken the rule. 53 In Nicaragua, the Court needed to establish whether a general principle of non-intervention and a prohibition of the use of force existed under customary international law (to which its jurisdiction was limited). The challenge here was, in fact, that interventions into a state's internal affairs did not appear to be the rare exception to the rule, but, rather, a widespread phenomenon. In view of that fact, the confirmation hypothesis fulfilled an important argumentative function in the endeavour to determine the state of the law. Established doctrine requires custom to be based on a practice of states that has been accepted as law. 54 In applying this approach to determining the existence of custom, it is common to start with identifying state practice and, then, in a second and separate step, to check whether this practice is also supported by an opinio juris. The confirmation hypothesis proves valuable at this point. It not only allows one to disregard conflicting practice as long as it is justified with references to the law, but even provides an argument to allow one to draw law-confirming conclusions from violations of the law. In doing so, the hypothesis helps to attribute more relative weight to confirming state practice. This, in the end, eases the argumentative burden in establishing a certain norm or of defending a norm against its dissolution.
The confirmation hypothesis has triggered significant controversy. D'Amato generally attested the ICJ judges to be 'collectively naive about the nature of custom' 55 and saw an 'empty theory that any practice inconsistent with [… an original rule] does not count'. 56 Glennon holds that 'the assumption that the state's intent is necessarily to "confirm" the rule is arbitrary'. 57 In view of this controversy, it is necessary now to take a closer look at the requirements of the confirmation hypothesis, which will enable an assessment of its merits and limits in more differentiated terms.

The invocation of the law as practice and opinio juris
In order to precisely assess the effects of the invocation of the law, an analysis of the exact doctrinal categories in which such effects have to be measured is needed. Attention must thus be turned to the sources of the jus contra bellum, which include the provisions of the UN Charter, especially Article 2(4) and Article 51, and, secondly, a distinct and parallel set of norms existent under customary international law. 58 Actions and declarations of states are potentially relevant under both sources. Customary rules require 'a general practice accepted as law'; 59 rules of the UN Charter are shaped and substantiated by interpretation, which gives significant weight to states' 'subsequent practice in the application of the treaty which establishes the agreement of the parties'. 60 While the UN Charter comes with the gravity of a foundational treaty, its substantive meaning can only be established when taking into account later interpretive documents (the most prominent example being the Friendly Relations Declaration 61 ) but also the general practice of states. 62 Rules of both sources therefore depend on an objective component of practice and on a subjective componentan opinio juris according to which the practice is accepted as a reflection of legal obligation. 63 But how does the invocation of norms affect the law in doctrinal terms? First of all, state declarations are, in principle, suitable ways to express opinio juris. The requirement of opinio juris is a subjective assessment expressing a belief that a certain practice reflects an obligation under international law, in contrast to one that might be motivated by other factors 64 (so thatin the latter caseit would be a usus out of comity at best, but not law). International legal scholarship largely agrees that the evidence for 'acceptance as law' includes a wide range of material and depends on the concrete circumstances and on the rule in question. 65 We can refer to states' actions, but also to their declarations, in determining the beliefs of states and, therefore, we may reference states' invocation of the law in relation to the use of force in order to draw conclusions as to the extent/existence of opinio juris.
A more controversial question regards the suitability of verbal declarations as also counting as practice. The notion of practice traditionally aimed to capture that states were 'doing certain actions', 66 i.e. it referred to the material conduct 'on the ground', not to reflexions about it. 67 This traditional approach lives on in the positions advanced by a significant number of writers who are hesitant to attribute too much weight to the mere declarations of states. D'Amato, for example, argues that 'a state has not done anything when it makes a claim; until it takes enforcement action, the claim has little value as a prediction of what the state will actually do'. 68 States might deliberately seek to mislead others through their declaration 69 and we would therefore naively fall for the empty promises of those states, so these authors fear. 70 However, a number of factors point against such a narrow understanding of practice. The ways in which states express themselves are considerably more complex and havewith the proliferation of international forums, the development of media and increased publicnesssubstantially developed since the nineteenth century. 71 States' legal positions do not only manifest themselves in material actions, but also in verbal declarations, accessible diplomatic correspondence, opinions of advisors and so on. Disregarding these verbal acts would mean substantially limiting the potentially relevant acts because statements are, in fact, the more common form of the articulation of state positions than physical conduct. 72 'Actual practice' is very rare in some fields, meaning that establishing custom would become even more difficult or impossible. 73 Moreover, restricting relevant practice could in fact encourage confrontation (e.g. regarding sovereign rights over disputed territory) because states would need to make a practical point in order to be taken seriously. 74 This would ultimately also marginalise less powerful states within the process of the formation and change of customary rules as they have less capability to act, but depend on their verbal declarations to count. 75 In addition to that, the attitudes of states engaged in aggression, genocide and other similar acts would count more vis-à-vis the mere declarations of law-abiding states, which, for example, condemn such actions. 76 The ICJ has considered various forms of verbal declarations as practice 77 and the ILC also argues that practice 'may take a wide range of forms. It includes both physical and verbal actions.' 78 It is therefore sensible to accept verbal declarations that states put forward to justify their actions not only as expressions of opinio juris, but also as practice. 79 The consequence of this conclusion is that verbal declarations can fulfil a double role, as they can be seen to indicate both practice and opinio juris. Whether we take them into account as one or the other depends on the circumstances and is a question of judgment and overall assessment. 80 It is important, though, not to count them twice and to take, for example, a single statement as an indication of practice and opinio juris in regard to the very same rule. 81

Rereading violation into confirmation
A focus merely on confirmation of practice and opinio juris, however, still leads to a mixed overall picture. For example, in the territorial dispute between Cambodia and Thailand, ultimately only one state can be right in relying on the law and, therefore, one side must be in violation (at least concerning a precise escalation, not necessarily in relation to the entire conflict). Would this mean that the invocation is invalidated by the ultimately illegal action? In other words, does not direct military action on the ground 'speak' much louder than the invocation of the law, thereby disqualifying the 'weaker practice' constituted by the mere invocation of the law?
In fact, state practice should not be considered in isolation, but in relation to other acts by the state in question. 82 One must not engage in cherry picking and only pay attention to certain parts of the practice; there must be engagement in an overall assessment of a state's behaviour. This means that when a state acts inconsistently, when its practice points first in one direction and then in another, the practice of this state may eventually not be of much worth for confirming a rule, because the direction of a state's practice and opinio juris remains inconclusive. 83 In the context of the current study, this would mean that where the action on the ground apparently violates the law, whereas the 'verbal action' confirms it, there exists a contradiction.
However, the invocation of the law does more than just add another layer of opinio juris and practice into the overall assessment. Rather, verbal declarations are capable of totally absorbing the law-negating dimension of states' physical action and letting even an illegal action shine in a totally different light. 84 The reason is that at the time of action, it simply is not known which legal assessment will prevail, as this is in practice an often difficult and controversial exercise. Based on the openness of factual and legal questions, it is often not possible clearly to condemn the actions of one or other side of the conflict, as it was factually and legally unclear which side is in fact right. In a number of circumstances the ambiguity of the application of the law to a concrete case accounts for why states violate the law. 85 The consequence is that both parties to a conflict can invoke a common abstract and general understanding of the very same rule, but draw quite different conclusions in regard to the application of the law to the facts of the concrete case. As is often the case in regard to the law, there is room for different perspectives and, before an authoritative decision has been rendered, no ultimate clarity exists on whether an act is lawful or not. In the multipolar process of international law, such a determination is usuallyif at allrendered in a likewise multipolar process of states and international organisations expressing condemnation or support for an intervention. In this process the existing collective organs, such as the Security Council or the General Assembly, are very important. However, the declarations of these organs often remain ambiguous. It is notable, for example, that the Security Council usually does not take a clear position, as its mandate is above all the preservation of peace and not a quasi-judicial function of deciding on the legality of an action. 86 Ultimate clarity and an authoritative determination concerning a dispute could be achieved by a court decision, but, of course, most controversies in international law do not fall under the jurisdiction of a court.
As a consequence, in most cases an authoritative decision on a controversial legal issue is never reached. Accordingly, disputes often remain in a state where both parties can legitimately claim to have the law on their side. This is due to the imperfect nature of international law, the underdeveloped institutional system of which does not provide ultimate legal clarity in the majority of cases. Nevertheless, the parties to such a dispute can, by means of their declared legal opinions, relate to the law and confirm the content of the abstract legal provisions, even though these provisions may not, due to a lack of clear international reactions or a lack of jurisdiction, provide legal clarity in the case at hand. Without an authoritative decision, all that remains in such a case is the common reference of two states to the same substantive rule of international law. Moreover, this does not change if an authoritative decision is rendered at a later time.
The consequence is that the invocation of jus contra bellum norms not only constitutes evidence of practice and of opinio juris, but in fact may change the legal meaning of actual conduct, dismantling it of its seemingly law-negating character.

Requirements, limits and empirics of the confirmation hypothesis
Having established that invocation may act as legal confirmation in principle, this section now explores the specific requirements as well as limits of the confirmation hypothesis, taking into account the empirical data employed in this article.

Invocation of established norms and interpretations
The core requirement for the confirmation hypothesis to take effect is that a state invokes an established norm of international law. This reference does not, of course, have to use the specific legal terminology, but it has to be clear that the state is referring to an established norm, because only an existing norm can be confirmed. A rule is established if it isamong states and, in a subsidiary sense, by scholarshipregarded to reflect the current state of international law. This means that a confirmation cannot occur in the cases in which states invoke international law, but where they rely on justifications for the use of force that are not (yet) established under the lex lata. The ICJ has made that clear: ' exception to the principle might, if shared in principle by other States, tend towards a modification of customary international law.' 87 Such claims aim to change the law, not to confirm it. Interventions aiming to protect democracy or human rights without Security Council authorisation are examples of the reliance on new justificatory rules.
A related requirement is that states must refer to an established interpretation of the law. In international law, a 'novel right' does not always have to come in the form of a completely newly proclaimed legal foundation: it may also come as a previously unestablished interpretation of an existing right (e.g. of self-defence). States make reference to all sorts of (proclaimed) legal norms: from firmly established ones to controversial ones, and to allegedly emerging (i.e. not yet established) ones. For obvious reasons they do not differentiate their claims in this way in the language that they use, but particularly try to let actions relying on shaky legal foundations appear to be justified by invoking the language of established rules. In the recent Syrian crisis, for example, many states have claimed to lawfully act in self-defence against Islamist terrorists on Syrian territory, above all against the group Islamic State (IS). A number of these statesnotably the United States, the United Kingdom, Turkey and Francehave claimed to act in individual self-defence against IS or Khorasan. 88 Some of the states that made this individual self-defence claim, and a number of others involved in the conflict that did not, alternatively or additionally relied on the collective self-defence of Iraq, 89 thereby 87 Nicaragua (merits) (n 5) para 207. presupposing that Iraq was suffering an armed attack in the sense of Article 51 UNC. However, the possibility of invoking self-defence against non-state actors operating from the territory of another state in fact relates to an interpretation of self-defence that cannot be understood to reflect 'established law', in the sense that it represents a largely uncontroversial interpretation of the current state of the law. It is established law that self-defence actions may be exercised against nonstate actors and on the territory of a foreign state, where the actions can be attributed to that state. 90 Of course, an extended framework of self-defence has been invoked by a number of states and has been supported by many scholars. Particularly since the attacks of 11 September 2001 and the subsequent US-led Afghanistan intervention and, even more so in view of the interventions in Syria, many have claimed that the law has changed and now allows for interventions against non-state actors on the territory of a non-consenting state. 91 However, this extended interpretation is not found in the largely uncontroversial/agreed framework of self-defence. 92 By contrast, the conditions of self-defence are far from agreed on and the participating states have invoked quite differing rationales for their interventions, ranging from the 'unwilling or unable' standard (proclaimed by the United States, Australia, Canada and Turkey) 93 to the criterion of effective territorial control (emphasised by Belgium and Germany). 94 The interpretation of international reactions to these claims does not lead to unambiguous results and, therefore, the claims are not sufficiently established in order to apply the confirmation hypothesis. Rather, by claiming self-defence, states take a position within the process of the development of the law and aim to (re-)shape the law on self-defence and to support an interpretation that has undoubtedly gained significant international support in the last two decades. For the purposes of this article it is not necessary to further elaborate on what the current state of the law actually is. It suffices to state that the law is not sufficiently settled in order to regard the possibility of self-defence against non-state actors as an uncontroversial interpretation. As the law isin that sensestill unsettled, we face a situation which resembles more what the ICJ, in the previously quoted passage, has described as the '[r]eliance by a State on a novel right' 95 that tends to modify the current legal framework.
This 'self-defence against non-state actors' example shows that while the acting states affirm the right to self-defence in their declarations, they may nevertheless be presenting a general interpretation of the norm that does not confirm an established reading. Thus, it is crucial in any given instance to examine which norm a state is referring to in substancenot only in the sense of the broad and uncontroversial categories (self-defence, intervention by invitation), but in relation to the specific interpretation of a norm that states invoke, to decide whether this interpretation is in line with the established reading of the law.
The decision on whether a rule invoked by a state is established or not has to be based on a comparison between the understanding of a rule that states have so far generally practised and accepted with the one that is proposed in the respective case. This is obviously not a question to be answered by exact science. Rather, it poses a theoretical challenge on how to determine the scope of what may count as the established reading of the law, which is all too often in itself a challenging question (as the controversial issue of self-defence against non-state actors illustrates). It is important to see, however, that in practice it largely is possible to differentiate new from established interpretations. This determination has to be made based on the reactions of states towards a claim, but also scholarly works on the current state of the law, which provide an important foundation for this assessment. The confirmation hypothesis applies where the state of the law is sufficiently clear, as states and (subsidiary) scholarship agree on an established reading on the norm which the acting state then invokes. Where we find significant dispute about the state of the law, the confirmation hypothesis may not lead to a confirmation. Under these circumstances, the invocation of a norm does not confirm what is established, but rather contributes to a specific legal view within an unsettled field of the law.
Consideration of the empirical cases will be illustrative at this point. We have seen that in three of the 26 controversial cases no legal norms were invoked at all and, therefore, the confirmation hypothesis does not apply from the outset. 96 The analysis of states' justificatory claims shows, moreover, that in seven cases states invoked a new norm or a controversial/non-established interpretation of an existing norm for justifying their intervention. The vast majority of these conflicts concerned the legality of self-defence against non-state actors (six cases). 97 States took military actions on the territory of another sovereign state without its consent because non-state controlled military groups were operating from the third state's territory. The remaining casethe Kosovo interventionwas justified by some states with the non-established concept of humanitarian intervention. 98 In these seven cases, states challenged the law. However, such challenges cannot be read to reflect a general opposition to international law as such: rather, they often also have a positive and productive function that allows international law to develop and to adapt to new empirical challenges. Recent scholarship has pointed out the productive effects of what has been described as 'operational noncompliance' 99 unfriendly unilateralism 100 or efficient breach of international law. 101 Challenging the established law through practical opposition is a widely employed way to push for the development of the law and to initiate a reinterpretation of existing provisions. As states cannot appeal to a centralised law-making institution, shaping law through practice is often the only de facto option that states possess in order to push for amending the law.
Leaving aside these seven cases we are now left with 16 remaining cases (around 35% of the 45 analysed conflicts) in which states justified their actions by reference to (a common understanding of) the established law, but drew opposite conclusions as to the legality of specific state actions. These are the cases in which the ICJ's confirmation hypothesis could generally lead to a confirmation of the law. The conflicts of this category predominantly have a traditional inter-state character. Twelve of them concerned disputes regarding territory, including those in which aspects of self-determination 96 See section 3. 97 of parts of the population are at issue. 102 In the remaining cases the interventions were directly or indirectly aimed at initiating regime change. 103

No invocation of untenable interpretations of the law
It has been shown that states may invoke an interpretation of the law that actually expresses their desire of how the law should be, with the aim of finding supporters of their understanding of the law. In contrast to this 'legislative' approach, there are also cases in which states provide a legal justification that they do not actually advance as a generally tenable and desirable legal view, but as a sort of pretext for their action, thereby taking a cynical approach to the law. In other words, they advance a legal justification, but this justification is in fact only a disguised political justification as it does not fulfil even the minimum standards of a tenable legal claim.
The category of untenability is a difficult one, because the quality of legal arguments obviously varies and is difficult to assess. However, there is a tipping point at which invoking the law clearly can be viewed as cynical, because it is so badly done. In its most abstract terms, this tipping point is reached when the legal argumentation is obviously not sustainable in view of the standards of the profession. Of course, determining exactly where this tipping point is reached may turn out to be a difficult exercise and remains a question of judgment. The standards of the profession are controversial themselves and the realist government lawyer takes a fundamentally different approach to the constitutionalist. Thus, there will always be lawyers (by qualification) who will justify a specific political course by providing legal arguments, but there is a point at which those attempts can be seen as being entirely absorbed by politics and, thus, as losing their connection to the normative system of law. The following criteria may help to determine whether a legal justification is in fact untenable: . If the majority of states, and/or international organisations, reject the state's legal claim, this suggests the justification is untenable. Also, the condemnation of an intervention by a state with usually close relations to the acting state provides strong evidence of untenability. 104 . The strength of other states' condemnation may be expressed in sanctions, and, in many cases more severe sanctions will indicate a stronger rejection of the legal claim.
. The lack of any substantial international support for a state's legal viewpoint also indicates untenability. . Ultimately, the degree of scholarly consensus is also an important indicator. Where scholars agree on an international level, a strong argument exists towards the (un)tenability. Where scholars engage in an extensive and nuanced legal debate, this may indicate that the claim has a degree of tenability and does not reach the level of cynicism.
There is not the space in this article for an exploration of all 16 cases and for a determination of whether the untenability threshold has been reached. A clear case of an untenable claim is, for example, Russia's justification for its 2014 intervention in Crimea. Putin justified the intervention by reference to a letter of Yanukovych, who requested that Russia 'use the armed forces of the Russian Federation to restore law and order' 105 and then initiated the annexation of a part of Ukraine's territory based on that very formulation. 106 It is obvious that the Russian action is not covered by the alleged legal foundation as it goes far beyond the scope of the authorisation (notwithstanding the question of whether Yanukovych was still in the position to invite intervention at all). 107 The US justification for the 2003 Iraq war provides another example. The US justified the intervention in view of Iraq's material breaches of its disarmament obligations, and claimed that the old Chapter VII authorisation issued by the Security Council in 1991 was reactivated. 108 However, the Council had made clear in its Resolution 687 (1991) that it regarded enforcement actions to be the prerogative of the Council when it decided 'to be seized of the matter and to take such further steps as may be required for the implementation of the present resolution'. 109 While numbers do not necessarily account for truth, it is nevertheless an indication for the unsustainability of the US approach that notwithstanding some apologetic writers, 110 the vast majority of legal scholars and states have clearly rejected the US justification for the invasion. 111

No propagandistic use of the law
Implicit disrespect for the law can also be expressed when states invoke legal justifications in blatant denial of facts. Generally speaking, a state that denies being engaged in any illegal acts or aims to obscure its activities performatively recognises the established law as the relevant normative benchmark. 112 However, apparent verbal commitments to the law can be undercut if the gap between the declarations and actual facts becomes too wide. This is the rationale that underlies positions that are generally critical towards attributing too much significance to verbal declarations of states.
Certainly, there are cases in which the declared commitment to the law is nothing more than political propaganda. Consider a very drastic historical example that makes the point clear. In 1939, Hitler justified the German war of aggression against Poland as a response measure, i.e. essentially invoking the rationale of self-defence, against the alleged use of weapons by Polish soldiers on German territory. 113 The Nazi regime even faked the Polish attack against which it claimed to be responding. The Nazis clearly employed a cynical attitude towards the law. Hitler had made that explicit when stating before Wehrmacht officers: 'The conflict will be initiated by a suitable act of propaganda. Its credibility doesn't matter, the law lies in victory.' 114 Everybody would agree that it was absurd to regard Hitler's propagandistic act and the accompanying declaration as confirming the law: rather, it signifies utmost cynicism and its most extreme weakening.
Turning to the present and not drawing any historical parallels, there nevertheless exist cases in which the obvious denial of facts prevails and in which the necessary overall assessment does not warrant the assumption of a law-confirming effect of the invocation anymore. It is not possible to engage in an in-depth discussion of these cases here, as the empirical material would overburden the current scope. However, recent decades have witnessed a number of cases in which a blatant denial of facts was crucial. Iraq's invasion of Kuwait, which Iraq justified inter alia with the aim that it was coming to the 'rescue of the Provisional Free Government of Kuwait' 115 falls in that category. Similarly, the US' 'major intelligence failure' 116 concerning the possession of weapons of mass destruction by Iraq that played a role in explaining why Iraq was in material breach of its disarmament obligations and paved the way for the 2003 Iraq invasion raises serious doubts. Moreover, especially in border disputes (such as the one between Ethiopia and Eritrea or between Thailand and Cambodia), both sides usually justify their actions in an escalation with self-defence and it is almost impossible to discern from the outside whether the escalation results from an unfortunate chain of events or whether one side simply lies about the facts of a current escalation. As a consequence, it is empirically very difficult to determine whether the cynical disrespect towards international law which such political lies communicate is in fact present in a particular case or not.
There is no absolute certainty here, but the assessment of whether states are in denial of the facts has to be based on the accessible material. This may include official sources, such as related judgments, or reports by factfinding missions, declarations of states, and media reports. A sensible indicator may also be found based on a critical analysis of the narratives presented by states. Do the parties to a conflict offer a sound version of what has happened and does it correspond to other sources? Does a party soundly address the allegations brought up by its opponent or does it merely provide a simple and implausible denial? Evidently, this requires an engagement with the material concerning specific cases.
A rather clear recent case can be found in the Russian declarations and actions in the 2014 Ukrainian crisis. In the beginning of the conflict, Russia repeatedly declared to be law abiding and that no Russian soldiers were actively involved in Crimea. 117 In that sense, the Russian government affirmed the provisions of international law, performatively expressing the viewpoint that not sending soldiers to Ukraine was the right thing to do. Of course, after the incorporation of Crimea into the Russian Federation, 115  Putin acknowledged that Russian soldiers played a key role. 118 In view of that, prior commitments to the law are invalidated and cannot be seen to change the meaning of Russia's actions on the ground. Empirically it can be very challenging to determine in a specific case whether states are in blatant denial of the facts. In the example of Ukraine such assessment was made possible because Putin ultimately admitted the involvement and this made further investigation of the facts superfluous. However, if states remain in denial, only a thorough judicial investigation or an authoritative, independent fact-finding commission could bring clarity, which isof courseusually not available. These empirical challenges also make it impossible to draw any concrete conclusions as to when the law has been used in a propagandistic way in the cases under consideration here.

What role for the confirmation hypothesis?
As a result, we are left with a differentiated view on the confirmation hypothesis. The invocation of the law contributes to practice and opinio juris and has the power to shift the meaning of practice and, hence, the ICJ's assumptionaccording to which the invocation of the law may confirm its norms even where the actual conduct is eventually found to be in violation of the lawproves generally correct. As has been discussed in this article, however, this comes with significant limitations as a confirmation does not occur where states invoke unestablished norms (or interpretations thereof) or where they employ a cynical attitude towards the law.
An important result from the prior discussion is that the confirmation may not be regarded as an issue that may be settled once and for all. A formalistic assessment that a state somehow invoked an established norm is not enough, but sometimes the factors that limit the confirmation hypothesis only become evident after the passage of time. Experience tells that the international legal assessment by states and scholars may also require some time in order for a clear position to be reached. The assessment of the 2003 Iraq war, for example, developed over years. While initially more diverse and numerous states supported the 'coalition of the willing', eventually even some of the states that participated in the campaign have acknowledged the illegality of the intervention. 119 The picture is now relatively clear, but it took many years to really reach a (near) consensus here.
In a similar vein, the facts of a conflict may appear in a different light after some time has passed. Russia's admission of the presence of its troops in Ukraineinvalidating its commitment to international lawwas sudden, following a period of uncertainty and denial of the facts. This means that the evaluation of an action as cynical may change over time and accordingly also the confirmation hypothesis should not be understood as a tool allowing a final conclusion on an issue; it requires an assessment in view of the latest developments.
The determination of the limits of the confirmation hypothesis leads to an undeniable epistemological problem. The declarations of states are documented rather reliably and can be traced: therefore, there is usually a significant amount of accessible 'verbal practice' that can be read as a confirmation of the law. The indicators that limit the value of these declarations, especially as evidence of states' propagandistic use of international law, by nature do not lie in the open and are considerably blurry and difficult to assess. The consequence is that the law-confirming effect of verbal declarations comes with the suspicion that a state's declaration might just be the ideological surface of an actually cynical attitude towards international law.
This constellation highlights the danger of an ever-widening gap between international law's normative demands and its effect on practice. The worstcase scenario is one in which states permanently commit to the law in their declarations, but violate it in their actual conduct whileat the same time covering their tracks. This, however, seems to be the general problem of a decentralised legal system that is based on an underdeveloped institutional framework and lacks compulsory binding judicial means of dispute settlement.
Notwithstanding this danger, the confirmation hypothesis can serve important functions. First, it takes seriously the performative commitment that the violator expresses towards the law when trying to hide a violation. As La Rochefoucauld expressed in one of his maxims, '[l]'hypocrisie est un hommage que le vice rend à la vertu.' 120 Second, as Kant has noted, law is about the external behaviour of subjects, not about their secret motivations or hidden intentions. Also, for international law, one can only take into account what can be assessed based on public sources. Such assessment isas this article has shownnecessary, but where a prudent assessment of the legal claims and facts of the case does not give any indication for a cynical approach towards the law, we can confidently ignore the abstract possibility that this might nevertheless be the case. Any other approach would lead into a field of mere speculation.
Third, the confirmation hypothesis provides us with an important argumentative device to deal with the complexity of empirical situations and significantly helps to assess, structure and interpret state practice.

Conclusion
The analysis presented in this article shows that the continuing announcements of the death of international law's prohibition of the use of force remain, as Henkin pointed out a long time ago, 'greatly exaggerated'. 121 The rules may be in permanent danger of becoming the mere façade for states' actions, but they can hardly be seen to have fallen into desuetude.
Although the empirical sample of conflicts occurring between 1990 and 2014 under investigation here does not include 'smaller' interventions (with less than 25 battle-related deaths), the analysis nevertheless demonstrates that the situation is not as bad as some commentators suggest. Of the 45 conflicts under analysis, 'only' 26 interventions were controversial among states. 122 The analysis of states' legal justifications for their participation in international conflicts of the past two and a half decades has shown that the rules of the jus contra bellum play a significant role in the practice of states and that their invocation can have three different types of law-confirming effects.
The first and weakest type of confirmation results from the mere reference to international law in states' justifications of their actions. Here international law is confirmed as the language of states. Through the invocation of legal arguments and the use of legal practices, states acknowledge the capacity of international law to serve as a communicative tool that allows for articulating their state interest. Some sort of invocation of jus ad bellum rules is present in the vast majority of cases, which demonstrates that international law forms a highly relevant system for states to articulate and explain their political course.
The law may, secondly, be confirmed by what has been described herein as coherent practice. This type of confirmation refers to the substantive provisions of the prohibition of the use of force. In a significant number of cases (namely in 19 out of 45 analysed armed conflicts) international law functioned as an unproblematic framework for international participation in armed conflicts. Here the law was confirmed as interventions took place within the boundaries of the law.
The third and most complex type concerns the confirmation of the substantive rules of international law in cases where the action is legally disputed and might even violate the law. This author has, in this article, defended the ICJ's confirmation hypothesis and has shown how the invocation of the law contributes to practice and opinio juris and may let the actual practice shine in a different light. The confirmation hypothesis nevertheless comes