II. EVIDENTIARY STANDARDS IN INTERNATIONAL LAW

In any adjudicative process, the application of an evidentiary requirement to a legal question has two aspects. The first of these is the quantum of evidence necessary to substantiate the factual claims made by the parties.Footnote ¹⁴ This is what may be labelled the required ‘standard of proof’. The second question in any evidentiary assessment is: upon which of the parties does the burden of meeting this standard fall?Footnote ¹⁵ This is best termed the ‘burden of proof’. The approach taken in international adjudication to the burden of proof issue is relatively easy to identify. International courts and tribunals essentially employ the concept of actori incumbit probatio: the party who relies upon a contention of fact is obliged to establish it.Footnote ¹⁶ This article deals instead with the former question: what evidentiary standard is necessary to ‘establish’ the validity of factual claims, upon which legal claims are premised?

In general, international law does not have a clear benchmark against which the persuasiveness or reliability of evidence may be gauged for the purposes of attributing responsibility or assessing legal claims.Footnote ¹⁷ In other words, there is no consistent standard of proof with regard to international obligations. International courts, tribunals and arbitrators essentially determine their own evidentiary standards.Footnote ¹⁸ Moreover, in most cases, there is no requirement that the standard employed remains the same within a tribunal across its decisions.Footnote ¹⁹

This is not to say that international arbitral or judicial decisions are made without any application of a standard of proof.Footnote ²⁰ Rather, the level of proof required to meet the evidentiary burden in any given dispute is often opaque, and the standard employed is far from consistent across decisions. Thus, as with other international arbitral bodies, the ICJ has essentially limitless power to determine what evidentiary standard it will require in any given case.Footnote ²¹ As such, it has avoided explicitly articulating a general standard with regard to its decisions. Instead, the Court has employed different standards, depending upon the dispute before it. These variable standards are often merely implicit in the decisions, if indeed they can be ascertained at all.Footnote ²² In disputes before the ICJ, then, ‘we generally know which party carries the burden, but we do not know with certainty what the burden is.’Footnote ²³

However, irrespective of this lack of clear evidentiary standards in international adjudication, it is possible, by examining the decisions of international courts and tribunals, to distinguish various evidentiary standards that have been employed with regard to specific international legal disputes. Four distinct evidentiary standards may be identified as appearing, ad hoc, throughout the jurisprudence of a variety of international courts and tribunals.Footnote ²⁴ These standards sometimes are applied explicitly; however, in the majority of instances the evidentiary standard employed is implicit in the manner in which the court or tribunal examined the evidence in the given dispute. The evidentiary standards that can be identified in international adjudication are similar to those that may be found in the domestic law of many jurisdictions, and so are likely to be rather familiar to most lawyers. Nonetheless, it is worth briefly reiterating them here.

The first of the standards that may be identified in the jurisprudence of various international arbitral bodies is what may be termed a ‘prima facie’ standard. This represents a test of very low degree with regard to the assessment of evidence: it simply requires that the evidence produced is indicative of the proposition claimed. The second identifiable test is that of a ‘preponderance’—or, alternatively, a ‘balance of probabilities’—standard. This refers to evidence that is more convincing than the evidence that is offered in opposition to it, or evidence that establishes that the factual proposition of the relevant party was more likely than not. It can be seen that this test is similar to, but is somewhat more onerous than, requiring mere prima facie evidence.

These two standards—prima facie and preponderance—may be contrasted with the ‘beyond a reasonable doubt’ standard. Somewhat self-evidently, the latter is a strict standard of proof, requiring that the proposition being presented is supported with evidence of a nature that there can be no ‘reasonable doubt’ as to the factual validity of the proposition. Under this standard, then, a proposition must be virtually indisputable, given the evidence.

Falling in between the first two ‘low level’ standards and the strict ‘beyond a reasonable doubt’ approach is what is often termed the ‘clear and convincing’ evidentiary standard.Footnote ²⁵ To prove something by a ‘clear and convincing’ standard, the party with the burden of proof must convince the arbiter in question that it is substantially more likely than not that the factual claims that have been made are true. This is obviously a more onerous test than a mere prima facie or preponderance standard, but does not require the virtual certainty of the ‘beyond a reasonable doubt’ test.

It has been stated that ‘the preponderance of evidence is predominantly applicable in international procedure.’Footnote ²⁶ This is possible: an evidentiary standard of ‘preponderance’ may well be the informal default position in international arbitration. However, this is a difficult claim to make with any degree of certainty. As we have already noted, there is little consistency in the application of evidentiary standards in international law: this varies greatly across tribunals and across decisions.

Indeed, it is arguable that a specific evidentiary standard for determining international legal questions in an adjudicatory context would in fact be undesirable. The inherent flexibility of the current approach suits the variety of challenges that face the international legal system.Footnote ²⁷ It may certainly be argued that the context or ‘type’ of dispute in question should be relevant to the evidentiary standard employed in establishing relevant facts.Footnote ²⁸ The ICJ has repeatedly demonstrated that, with regards to evidentiary standards, the context of each dispute will be an important factor. This pragmatic approach makes conceptual sense, due to the fact that the standard of evidence that it is desirable for the relevant party to meet to establish a factual claim may be higher in the context of certain disputes than in others. Inevitably, the gravity or consequence of certain breaches of international law will differ from others.Footnote ²⁹ As such, the evidentiary standard for establishing such a breach should differ.Footnote ³⁰

Therefore, it is not herein argued that a single standard for assessing evidence across all international law disputes is desirable. Nonetheless, it is contended in the next section that in regard to the specific question of whether an armed attack has occurred for the purposes of assessing the lawfulness of a self-defence claim, it is necessary that a consistent standard of proof can be identified. In other words, the evidentiary standard for establishing the factual basis for one claim of self-defence should be the same as for any other self-defence claim, and, if possible, that standard should be explicit.

III. IS THE LACK OF AN EXPLICIT STANDARD FOR SELF-DEFENCE DESIRABLE?

It is perhaps unsurprising, given our previous discussion, that the practice of the ICJ in relation to self-defence mirrors that of its decisions in other contexts. The Court has not yet articulated an explicit standard of proof for the establishment or attribution of an armed attack as a precursor to assessing the lawfulness of claims regarding the right of self-defence.Footnote ³¹ In the 1986 Nicaragua merits judgment, for example, the Court on a number of occasions rejected evidence as being ‘insufficient’ to establish the facts necessary to determine the lawfulness or unlawfulness of the self-defence claim made by the United States.Footnote ³² Thus, the Court stated at paragraph 159 of the Nicaragua judgment that it was ‘unable to give weight to alleged statements to the effect of which there is insufficient evidence.’Footnote ³³ However, in each instance, the ICJ determined that evidence was ‘insufficient’ without indicating where the line between sufficiency and insufficiency actually lay.

Similarly, in the 2003 Oil Platforms case, the ICJ held that ‘if at the end of the day the evidence available is insufficient to establish that the missile was fired by Iran, then the necessary burden of proof [to establish that the missile attack constituted an ‘armed attack’] has not been discharged by the United States.'Footnote ³⁴ As in Nicaragua, the Court in Oil Platforms did not then go on to explicitly indicate what the ‘necessary burden of proof’ was. DRC v Uganda followed its predecessors, in that no explicit evidentiary standard for self-defence was articulated in that decision either.Footnote ³⁵

Of course, it may be questioned as to whether an explicit and consistent evidentiary standard to be applied to all cases involving self-defence is desirable. One could take the view that there are good reasons for the Court to avoid indicating a generic standard for such cases. We have already seen that, in general, international courts and tribunals employ various evidentiary standards, and that there are sound reasons for this flexibility.Footnote ³⁶ Why should self-defence be any different? As with disputes in general international law, each dispute involving a claim of self-defence will necessarily have context specific factors. Nicaragua, for example, involved a conflict of a relatively large scale between contra forces and the Sandinista government. In contrast, Oil Platforms concerned far more limited uses of force, both in terms of the alleged attacks and the responses taken. Perhaps the severity or scale of the dispute involving self-defence could be seen as requiring a different evidentiary standard to be applied to each given dispute?

It is argued here that a consistent evidentiary standard is desirable for self-defence. Or, to clarify this: it is argued that an explicit standard should be adopted for the question of whether an armed attack has occurred and if so who perpetrated it (the sine qua non legal trigger for self-defence).Footnote ³⁷ Self-defence is a rather special claim within international law, in that it constitutes the only clearly recognised legal justification for the unilateral use of military force. Outside of UN enforcement, self-defence alone legally validates institutionalised violence (on a potentially large scale). It is also, at least in the first instance, a legal determination that is self-assessed. States decide whether an armed attack has occurred, and whether it is necessary to respond with force. As such, the potential for abuse of the right of self-defence (and the degree of human suffering that can be caused by such abuse) is obvious.Footnote ³⁸ On occasion, States make disingenuous claims of ‘self-defence’, and support such claims with spurious ‘evidence’. It is therefore crucial that international law requires that claims of self-defence are supported by evidence, even if this evidence is only made available ex post facto.

To ensure that such evidence is actually present in cases where States avowedly act in self-defence and that it can be produced if necessary, comparatively strict rules of evidence are needed. The very nature of the use of military force heightens the need for strict evidentiary requirements in respect of legal claims justifying such actions. As part of this need for tighter evidentiary rules in this context, it is important—when one turns to the question of the occurrence of an armed attack—that the standard of evidence necessary to establish this as a factual matter is both consistent and explicit across cases.

This is because a transparent consistent standard would have the effect of producing a level of certainty with regard to the use of military force.Footnote ³⁹ It would be clear what quality of evidence a State may be expected to produce, and should expect to have to produce, in support of a claim of self-defence. This will inevitably contribute to limiting the abuse of the right. The current position—or lack of a position—of the ICJ on the question of an evidentiary standard for self-defence means, ultimately, that a State party before the Court that is defending itself against a military attack must attempt to establish the factual existence of that attack without knowing what evidentiary threshold will be required by the Court before it will accept the State's claims. Such uncertainty is surely undesirable when it comes to such an important operational decision.

We have seen that differences in the ‘gravity’ of the subject-matter relating to a claim under international law may mean that differing evidentiary standards are desirable: as a general rule evidentiary standards are and should be context specific.Footnote ⁴⁰ However, this does not apply to the question of the occurrence of an armed attack, because that question is itself based upon the gravity of the use of force against the responding State. In this respect, the question being asked in Nicaragua was the same as that asked in Oil Platforms: did an armed attack occur? Despite the factual differences between the two cases, the legal ‘gravity threshold’ for this question was the same. That is, could the attack or attacks against the responding State be considered to be of ‘the most grave form of the use of force’?Footnote ⁴¹ This is not a context specific question, despite some contentions to the contrary.Footnote ⁴² It is a standard of scale that must be met in all instances of self-defence,Footnote ⁴³ and as such, the evidence required to establish this can also be set at the same level.

To put this differently, the question of whether an armed attack has occurred is an objective one, in both a factual and legal sense. We are not here talking about the difference between a legal determination of genocide and the breach of a trade agreement, but instead about whether a particular legal criterion has been met—something for which the evidentiary standard should be the same in any instance, especially when that criterion is the primary trigger for initiating military force. The perfectly sensible assertion made by Judge Higgins that ‘the graver the charge the more confidence there must be upon the evidence relied on’Footnote ⁴⁴ is surely not relevant when the ‘charge’ is the same.

As the principal judicial organ of the UN, the ICJ would seem well placed, when suitable cases come before it, to elucidate upon such a standard, irrespective of the fact that this is not its usual practice. Although not binding beyond the case in dispute, any such articulation of a standard for this question would possess a good deal of authority coming from the ICJ. Moreover, if the Court were explicit about the evidentiary standard employed in such cases, it would enable to judges themselves to fully engage with questions of evidence, and be clear in their own minds what evidence will suffice, something that is likely to strengthen the quality of the adjudicative process.Footnote ⁴⁵

Given all of the above, the Court has received criticism for not setting out an evidentiary standard in relation to self-defence.Footnote ⁴⁶ Most notably, such criticism has emerged from within; in their separate opinions to the Oil Platforms merits decision, Judges Higgins, Owada and Buergenthal all raised concerns with the fact that the Court refused to set out any clear evidentiary standard with regard to self-defence in that case.Footnote ⁴⁷ Judge Buergenthal, for example, asked: ‘What is meant by ‘insufficient’ evidence? Does the evidence have to be ‘convincing’, ‘preponderant’, ‘overwhelming’, or ‘beyond a reasonable doubt’ to be sufficient?'Footnote ⁴⁸

Judge Higgins extended her criticism to the Nicaragua decision, and thus implied that Oil Platforms should be viewed as compounding the mistakes made in this regard in 1986.Footnote ⁴⁹

Moreover, even if one takes the contrary view to that put forward here—that due to the inherent contextual nature of each and every dispute, it may be desirable to employ different evidentiary standards in relation to different disputes—Nicaragua and Oil Platforms may nonetheless be criticised in regard to their approach to evidentiary standards. In neither decision did the Court at any point explicitly state what the standard required was in the case before it:

As Wolfrum has argued, the general principle of freedom of evidence in international arbitration should not ‘absolve’ international courts and tribunals of the need to be explicit about the evidentiary standard being employed in any given decision, or with regard to any particular factual question, even if standards vary across cases.Footnote ⁵¹

IV. THE IMPLICIT ‘CLEAR AND CONVINCING’ STANDARD

In the previous section, we noted the lack of explicit elucidation by the ICJ of an evidentiary standard for assessing self-defence claims. Whilst this is certainly true, if one undertakes a close textual examination of the three merits decisions of the Court that directly relate to the application of the right of self-defence, it is possible to identify implicit standards of evidence that appear to have been employed by the Court. Despite a lack of explicit invocation of evidentiary standards, implicit approaches may be identified in each of the judgments. The standards employed can be inferred from cumulative indications as to the approach of the Court.

It is important to note that this process must necessarily be a tentative one. When examining what the Court appears to have done, rather than what it has actually explicitly held, there is always a danger of misinterpretation, or of reading too much into the text of the judgment. In the Nicaragua case, the Court stressed that it does not possess ‘authority to ascribe to States legal views which they do not themselves advance.’Footnote ⁵² When determining the conclusions reached by the ICJ, one must be careful not to ascribe conclusions to the Court that it did not itself advance. As such, the following section is accompanied with the caveat that it constitutes an interpretation of the evidentiary standards employed by the ICJ.

Leaving such concerns to one side, upon examination of both the Nicaragua and the Oil Platforms merits decisions, it is striking that a consistent approach to evidentiary issues may be identified throughout. The inference that may be drawn from both judgments is that the standard employed was one of ‘clear and convincing evidence’.Footnote ⁵³

For example, in the Nicaragua decision the Court stated that it ‘must attain some degree of certainty [regarding the claim of the United States that El Salvador et al. had suffered an armed attack] and … [ensure] that the facts on which it is based are supported by convincing evidence.’Footnote ⁵⁴ Elsewhere, the Court asserted that there was ‘no clear evidence of the United States having exercised such a degree of control [of contras]’.Footnote ⁵⁵ This suggests a need for something beyond a mere preponderance of evidence. Similarly, prima facie evidence supporting the existence of ‘interventions’ conducted by Nicaragua into Honduras and Costa Rica was rejected on the basis that such evidence was insufficient to establish that these constituted armed attacks.Footnote ⁵⁶ In addition, there are a number of other points in Nicaragua that equally suggest a reliance on a ‘clear and convincing’ standard.Footnote ⁵⁷

A similar pattern emerges when one turns to the Oil Platforms decision. Thus, at paragraph 71, the Court took the view that the available evidence regarding the Iranian responsibility for the mine that struck the USS Samuel B Roberts—being that it was surrounded by other moored mines bearing serial numbers attributable to Iran—was ‘highly suggestive, but not conclusive’.Footnote ⁵⁸ This evidence was clearly indicative, prima facie, of Iranian responsibility for an armed attack on the United States. In rejecting that evidence, the Court implicitly rejected a balance of probabilities approach, and indicated the need for a higher evidentiary standard. Elsewhere the Court held that ‘the evidence indicative of Iranian responsibility for the attack on the Sea Isle City is not sufficient to support the contentions of the United States.’Footnote ⁵⁹ Thus ‘indicative evidence’—evidence that meets a prima facie standard—was viewed as being ‘insufficient’ to establish factually the attribution of an armed attack. Equally, the assessment of evidence in the decision clearly did not approach a strict ‘beyond a reasonable doubt’ standard.

Statements of this kind in the jurisprudence do not in themselves establish with any clarity the evidentiary standard employed by the ICJ in this context. However, the cumulative weight of these suggestions as to the Court's approach to evidence is certainly indicative of reliance, in both Nicaragua and Oil Platforms, upon a ‘clear and convincing’ standard, rather than a prima facie or preponderance standard, or the converse ‘beyond a reasonable doubt’ approach.Footnote ⁶⁰ Indeed, aside from one anomalous exception,Footnote ⁶¹ this approach appears to have been consistent throughout each decision and, also, between them.

The fact that there exists no explicit standard in the Court's jurisprudence remains, in the view of the present author, unfortunate. However, it is arguable that—following Oil Platforms—a reasonably consistent ‘clear and convincing’ standard was tentatively emerging from the case law, albeit without concrete expression. This process constituted a desirable second best to having an explicit standard. Indeed, a ‘clear and convincing’ standard appears particularly appropriate in the context of self-defence. A standard of either prima facie evidence or of preponderance would not be strict enough given the fact that such disputes involve the use of force. If an armed attack is established by the evidence, this constitutes a green light for a State to initiate, legally, military force against another. This determination surely should not hinge on a mere ‘balance of probabilities’ approach. Equally, the right of self-defence relates directly to a State's national security. A strict ‘beyond a reasonable doubt’ standard seems too onerous when it is considered that States making a genuine claim of self-defence will be faced with a defensive necessity for a military response. Certainly, expecting a State faced with such a necessity to ensure that it can meet a ‘beyond a reasonable doubt’ standard of proof before responding is wholly unrealistic.Footnote ⁶²

Moreover, the ‘clear and convincing’ standard may concord with the standard of evidence increasingly employed in practice by States—notably the United States—with regard to self-defence claims made outside of the adjudicative context. This is not the place to set out this practice, but it is worth noting the work done by Mary Ellen O'Connell in identifying a number of instances where the United States has claimed that it has ‘compelling’ or ‘convincing’ evidence establishing a right to self-defence.Footnote ⁶³ This evidentiary standard was claimed to have been met by the United States with regard to the bombings of Libya in 1986, the raid against Iraq in 1993 in response to the plot to assassinate former President Bush, and the attacks on Afghanistan and Sudan in 1998 following the bombings of United States embassies.Footnote ⁶⁴ Whilst the practice of one State is far from legally constituting (in the sense of the formation of a customary standard), it is nonetheless telling that there has been an apparently consistent approach to evidence in the context of self-defence from the world's sole superpower.Footnote ⁶⁵ Thus, it is possible to argue that the ‘clear and convincing’ standard seemingly employed in Nicaragua and Oil Platforms may not only be desirable, it may also be an accurate reflection of an embryonic formalist approach to evidence with regard to self-defence claims more generally.

V. DRC v UGANDA: FLUCTUATING STANDARDS

In DRC v Uganda, the third contentious self-defence decision of the ICJ, the Court offered a comparatively high degree of commentary on the weight it ascribed to various pieces of evidence presented before it. This more overt approach to evidentiary issues is commendably transparent, and was therefore an improvement upon the Oil Platforms case in this regard.Footnote ⁶⁶ Nonetheless, the case in no way introduced an explicit standard of evidence for self-defence claims. The Court assessed whether individual pieces of evidence were credible, but rarely stated what ‘level’ of credibility the parties were required to attain (particularly as a cumulative whole) to support assertions as to the occurrence of one or more armed attack. For example, the Court stated that it considered that ‘Uganda had not produced sufficient evidence to show that the Zairean authorities were [responsible for] attacks against Ugandan territory.’Footnote ⁶⁷ As in its previous decisions, the Court did not then explicitly indicate what would constitute ‘sufficient evidence’.

Again, as with Nicaragua and Oil Platforms (in spite of a lack of explicit articulation), the DRC v Uganda judgment does contain a number of implicit indications regarding the evidentiary standard applied. However, unlike in preceding cases, it appears that the Court in DRC v Uganda referenced different standards at different points of the judgment with regard to the same question (that of the existence of an armed attack, be it an armed attack against Uganda, or, in the context of the counter-claim, against the DRC). In other words, the Court not only appeared to refer to a ‘clear and convincing’ standard throughout the DRC v Uganda case, it also appeared, in some instances, to assess evidence relating to the self-defence claims of the parties based upon other standards.

At points in its decision, the ICJ clearly rejected evidence that would prima facie establish that an armed attack had occurred, either against the DRC or against Uganda, and instead indicated the need for evidence to meet a weightier ‘clear and convincing’ standard. Thus, in examining whether Ugandan attacks had in fact occurred in the eastern part of the DRC based upon the available evidence, the Court rejected a ‘sketch map’ provided by the DRC as being inadequate to establish such attacks.Footnote ⁶⁸ The map indicated the presence of Ugandan troops (or, perhaps, Ugandan directed troops) at various positions within the eastern part of the DRC. The map, in conjunction with other available evidence, would meet a low level prima facie test (and perhaps even a preponderance test) for establishing the presence of these troops. Instead, the ICJ held that Ugandan action in the eastern part of the DRC needed to have been ‘convincingly established by the evidence’.Footnote ⁶⁹ Further, the Court echoed this terminology later in the judgment when reiterating that it had not been established that attacks of the character of armed attacks had occurred in the eastern part of the DRC, holding that the evidence that had been produced that did indicate this was insufficient because it was not clear and convincing.Footnote ⁷⁰ Therefore, in the context of the alleged armed attacks in the eastern areas of the DRC, it seems relatively apparent that the Court was not satisfied with indicative evidence, but instead required that it meet a clear and convincing test.

Similarly, when the Court examined whether the DRC itself had committed an armed attack (or armed attacks) against Uganda—potentially justifying Uganda's resort to force—it concluded that alleged attacks by the Allied Democratic Forces (an anti-Ugandan militant group) and support for that group by the DRC, was not established on the evidence. Here, Uganda presented a variety of pieces of evidence that prima facie supported this factual claim. However, this evidence—media reports and other secondary accounts, including witness statementsFootnote ⁷¹—was not seen as satisfying the standard of evidence applied by the Court. It held that this evidence did not establish the attacks by the Allied Democratic Forces, or the DRC's involvement with that group, due to the fact that the available evidence was not ‘weighty and convincing’.Footnote ⁷² The Court rejected this evidentiary material, at least to the extent that it established the occurrence of an armed attack, irrespective of the fact that it was relatively independent of the parties, because it was not of sufficient clarity, indicating that these materials did not have the necessary ‘quality or character’ to support the contentions of Uganda in this regard.Footnote ⁷³ This again suggests that the Court rejected prima facie evidence in favour of a clear and convincing standard.

It would seem then that with regard to the respective claims of self-defence made by both parties, that the ICJ assessed the question of whether one or other party had perpetrated an armed attack against the other with reference to a clear and convincing evidentiary standard. This mirrors the approach taken by the Court in its earlier decisions.

However, at other key points in the DRC v Uganda decision, the Court appeared to accept evidence (with regard to the establishment of an armed attack) based upon a notably lower standard: either that of preponderance or of prima facie evidence. This can be shown, for example, by reference to the Court's examination of whether Uganda was responsible for alleged attacks against the DRC by the Mouvement de Libération du Congo (a rebel group comprised predominantly of Banyamulenge Tutsis), and therefore whether it could be said that Uganda had perpetrated one or more armed attacks against the DRC. With regard to this specific question, the Court found that the evidence provided by the DRC was not indicative of Ugandan responsibility in this context.Footnote ⁷⁴ However, it would appear that indicative evidence was all that the Court would have required to make such a finding. The Court held that there was not enough evidence to ‘suggest’ (to use the Court's word) that Uganda was responsible for any armed attacks perpetrated by the Mouvement de Libération du Congo, but that there was enough prima facie evidence to establish a more general breach of international law.Footnote ⁷⁵

When assessing whether an alleged aerial operation by Uganda at Kitona in the DRC constituted, factually, an armed attack by Uganda, the Court examined evidence provided by the DRC to support this contention. In doing so, it appeared to assess the evidence provided by the DRC on the basis that such evidence should meet a balance of probabilities test. There was no indication that the Court sought clear and convincing evidence to establish Ugandan responsibility for the Kitona incident and the required gravity of that incident, only that this needed to be shown to have been more likely to have been the responsibility of Uganda than not.Footnote ⁷⁶

It therefore seems apparent that the ICJ applied a ‘clear and convincing’ standard at some points of the judgment, but then elsewhere referred to a notably lower test, with regard to the same issue: the factual basis of an armed attack.Footnote ⁷⁷ Apparently, the Court adopted a clear and convincing standard with regard to whether the actions in the eastern part of the DRC, and the attacks against Uganda by the Allied Democratic Forces constituted armed attacks, yet adopted a lower standard with regard to whether the actions of the Mouvement de Libération du Congo and the aerial attack at Kitona were similarly armed attacks.

As such, evidence set out by both parties to support directly comparable factual claims were assessed by the Court based upon differing evidentiary standards. It has already been argued that there are good reasons for employing the same evidentiary standard with regard to all assertions that an armed attack has occurred.Footnote ⁷⁸ This is particularly true when such assertions are made within the context of the same overall self-defence claim, as was the case in DRC v Uganda. There is no contextual difference between the empirical occurrence of an attack in the eastern DRC as opposed, for example, to an alleged aerial attack at Kitona: both are instances of the same character—an unlawful use of force that must be assessed against the ‘gravity test’ to determine whether they constitute an armed attack has been met, and whether the State against which self-defence measures are being taken is responsible. There is no justification for applying different evidentiary standards to these separate uses of force. Indeed, to do so is illogical.

Ultimately, then, not only must the standard applied by the Court be teased from the DRC v Uganda judgment (as with its predecessors), but it appears that the ICJ applied different standards at different stages of the same judgment. The seeming ‘dual standard’ aspect of the case undermines any conclusion that it clarifies the evidentiary standard that is likely to be adopted in future cases. Despite a commendable commentary on the process of the examination of much of the evidence relied upon in the case, in terms of a standard of evidence, the decision actually further confuses the already somewhat murky waters of Nicaragua and Oil Platforms. The practice of applying differing evidentiary standards in the same judgment to the same factual question inevitably opens the ICJ up to accusations of bias.Footnote ⁷⁹ Having said this, it is not as though the Court applied a ‘clear and convincing’ standard to the evidence produced by the DRC in support of its claim that it had suffered an armed attack, and then switched to a ‘prima facie’ or ‘preponderance’ standard for the corresponding counter-claim of Uganda. It applied fluctuating standards across the claims of the parties. For this reason an accusation that the Court acted in a biased manner towards either party cannot be sustained. A charge of arbitrariness, however, could certainly be levelled at the ICJ. As such, this aspect of the DRC v Uganda decision cannot be good for the long term credibility of the Court.

Interestingly, this aspect of the DRC v Uganda case worryingly echoes the approach to evidentiary standards employed in the Corfu Channel decision of 1949. In that judgment, as with DRC v Uganda, a comparatively open approach to evidentiary issues belied the fact that the Court employed a fluctuating standard of proof. Of course, the Corfu Channel decision did not specifically examine the lawfulness of a self-defence claim. Nonetheless, it dealt directly with the use of force, and the claim advanced by the United Kingdom amounted to an analogous one of ‘self-help’.Footnote ⁸⁰ It is notable that in Corfu Channel, the Court at one point referred to the need for ‘conclusive evidence [and] … a degree of certainty’.Footnote ⁸¹ The terminology employed here is indicative of an approach similar to the ‘clear and convincing’ standard, or, as some have argued, to the application of a ‘beyond a reasonable doubt’ standard.Footnote ⁸² Such a conclusion is strengthened by a passage on the next page of the decision, where the Court noted that the evidence must ‘leave no room for reasonable doubt.’Footnote ⁸³

However, in between these two allusions to a relatively high evidentiary standard, the ICJ referred to the establishment of ‘prima facie responsibility’ on the part of Albania.Footnote ⁸⁴ It then went on to argue that in certain circumstances, a State which has been the victim of a breach of international law but ‘is unable to furnish direct proofs of facts … should be allowed more liberal recourse to inferences of fact and circumstantial evidence.’Footnote ⁸⁵ This passage demonstrates a notable shift from a higher standard—the requirement of ‘conclusive evidence’—to a much lower standard of evidence in the context establishing responsibility for the use of force, within the space of a page.

Despite some suggestions that the Corfu Channel constituted a rare success with regard to the assessment of evidence in the ICJ,Footnote ⁸⁶ it is submitted that this kind of flexibility with regards to evidentiary standards is at least somewhat detrimental to the credibility of the decision. The ad hoc approach employed in Corfu Channel was not apparent in Nicaragua or Oil Platforms, but it is something that was a notable aspect of DRC v Uganda. A return to the methodology employed in Corfu Channel in this respect is wholly undesirable.