I have started to review the Katanga judgment and the dissenting opinion of Judge Van den Wyngaert. Considering the field of my research (evidence) I specifically sought for information revealing how the Judges understand the standard beyond reasonable doubt.
Due to my poor French I am not sure on the view of the majority. I belive the relevant passage is paras. 68-70.
I cannot see that the majority is talking about alternative narratives and the need for the Prosecutor to eliminate such alternatives. The Majority appear to state in para. 70 that even if a person is declared non-culpable that does mean that he/she is innocent, interesting.
Fardeau de la preuve
68. Aux termes de l’article 66 du Statut, l’accusé est présumé innocent jusqu’à ce que le Procureur ait prouvé sa culpabilité. Pour qu’il soit déclaré coupable, il faut que chacun des éléments de l’infraction reprochée ait été établi « au - delà de tout doute raisonnable » .
69. La Chambre précise que le principe d’établissement de la preuve « au - delà de tout doute raisonnable » doit être appliqué s’il s’agit d’établir l’existence d’un élément du crime ou du mode de responsabilité retenu à l’encontre de l’accusé, ou encore, s ’il s’agit d’établir l’existence d’un fait indispensable pour entrer en voie de condamnation.
70. Pour la Chambre, le fait qu’une allégation ne soit, selon elle, pas prouvée au - delà de tout doute raisonnable n’implique pas pour autant qu’elle mette en cause l’ existence même du fait allégué. Cela signifie seulement qu’elle estime, au vu du standard de preuve, ne pas disposer de suffisamment de preuves fiables pour se prononcer sur la véracité du fait ainsi allégué. Dès lors, déclarer qu’un accusé n’est pas coupa ble ne veut pas nécessairement dire que la Chambre constate son innocence. Une telle décision démontre simplement que les preuves présentées au soutien de la culpabilité ne lui ont pas permis de se forger une conviction « au - delà de tout doute raisonnable ».
I found this in Judge Van den Wyngaert's dissenting opinion (in English).
32. ... charges must represent a coherent description of how certain individuals are linked to certain e vents, defining what role they played in them and how they related to, and were influenced by, a particular context. Charges therefore constitute a narrative in which each fact belonging to the ‘facts and circumstances’ has a particular place.
33. ... It is crucial to note that it is insufficient to simply compare ‘stories’ in order to see to what e xtent they contain some of the same elements. It is equally important to analyse the legal significance of each fact within the framework of each narrative, because this determines how an accused would defend him or herself against the charges as formulated. It matters a great deal, in this respect, how important certain parts of the story are within each narrative. A similar fact may be a mere detail in one narrative, but constitute the linchpin of another. Accordingly, a defendant may have chosen not to devote scarce resources to such a fact because it could not be expected to have any tangible effect on the outcome of the case, whereas he or she would in all likelihood concentrate all his or her investigative efforts on that same fact if that fact wer e to perform a different function in an alternative narrative.
It is interesting that Judge Van den Wyngaert uses the expression "alternative narrative". I write about narrative evidence theory on pp. 166-167 in my book Evidence in International Criminal Trials. Van den Wyngaert continues under the same philosophy.
145. In particular, there are countless points where I think it is beyond dispute that reasonable alternative explanations can be given to the evidence. It is uncontroversial that the Chamber can only rely on the incriminating version of events if all alternative explanations can be rejected for being unreasonable. However, the Majority only engages in this exercise selectively and often simply states that it is not convinced by the explanations offered by the Defence. With all due respect, this is not the appropriate approach. Instead, the Chamber must convincingly explain why the alternative explanation is considered to be unreasonable.
146. The Defence does not shoulder any burden of proof in this regard. Yet, this is very often the attitude taken by the Majority. Rather, unless the Defence raises an explanation that is patently absurd, it is the Prosecutor’s task to disprove it.
Again, she uses expressions such as "alternative explanations" and that it is the the Prosecutor’s task to disprove it. This what I write in my book in relation to theories on the matter (pp. 183-186):
In essence, Cohen proposes an inductive method whereby alternative hypotheses may be graded ordinally in accordance with the factfinders conclusions as to how well each hypothesis is accounted for by the evidence; how well each hypothesis withstands attempts to eliminate it based on the evidence; and the extent to which any given hypothesis shows itself evidentially superior to the others in terms of completeness and quality of the evidence. Diesen suggests that alternative hypotheses should be tested when they are reasonable in the sense that they are not pure scepticism or speculation. More specifically, a reasonable hypothesis is 1) compatible with at least some of the facts in the case, and 2) commensurable with the indictment, i.e. the alternative hypothesis should be possible to formulate as a legal and evidential theme.... A hypothesis that raises sufficient reasonable doubt to acquit the defendant should be compatible with the defendant’s own version or with an explanation of why the defendant has not presented such an alternative in his or her testimony. ... The fact-finding process is according to Ferrajoli always guided by a working hypothesis; the formulation of the hypothesis is always done in order to reach a possible or desired validation. The Court’s evaluation of evidence requires that the process fulfil three criteria: 1) The prosecution, which carries the burden of proof, has the duty to present the facts confirming the indictment and related circumstances; 2) the defence has the right to defy the hypothesis of the indictment by using counterevidence which are compatible with an alternative hypothesis which in its turn the prosecutor must eliminate; 3) The Court may only accept the hypothesis of an indictment if it is corresponds with all the evidence and may withstand all possible counterevidence
I also make the following conclusion on the case law of the ad hoc tribunals (p. 195).
the ad hoc tribunals have adopted “the exclusion of every reasonable hypothesis of innocence method” whereby alternative hypothesis are eliminated.
One may find a similar statement by the ICC Appeals Chamber, see Al-Bashir (3 February 2010, para. 33)
requiring that the existence of genocidal intent must be the only reasonable conclusion amounts to requiring the Prosecutor to disprove any other reasonable conclusions and to eliminate any reasonable doubt.
My conclusion on p. 196:
the statements in al-Bashir and Lubanga may be analysed together with previous case law from the historical tribunals and the ad hoc tribunals. The conclusion is that when judges in international criminal proceedings evaluate whether evidence meets the “beyond a reasonable doubt” standard, they are acting according to the approach whereby alternative hypothesis are eliminated.
Note that Judge Van den Wyngaert has an entire section in her opinion with the title "Another reasonable reading of the evidence is possible" covering 42 pages (pp. 100-142).
Judge Van den Wyngaert also has a very interesting arguement about the scope of investigations in para. 92.
even if the investigations yield no useful new evidence whatsoever, this does not mean – even with hindsight - that they were not necessary. ... defence investigations that yield no significant result play a very important role in confirming the validity of the conviction. However, if no investigation takes place at all, there always remains the reasonable possibility that evidence might have been found that could contradict the available incriminating evidence.
Here is what I write on the matter (pp. 153-156):
the standard of an adequate investigation is a quantitative issue while the standard of proof is a qualitative issue. To elucidate what is an adequate investigation, the concept “robustness” and its theoretical underpinnings are introduced. ... Keynes distinguished between the probability of an argument and the weight of an argument. ... Strandberg has elaborated and clarified Keynes’s theory: The degree of probability depends on the relation between the pieces of evidence in favour of a proposition and the pieces of evidence against it, while the degree of weight depends on the sum of the pieces of evidence in favour of a proposition and the pieces of evidence against. Hence if new evidence is added that is unfavourable to the proposition the probability will go down but the weight will go up. For example, assume that probability is 0.85 for a given piece of evidence (h1). Subsequently evidence h2 is added and the probability of h1 and h2 combined is now 0.45. Then a third piece of evidence h3 is added which makes the probability return to 0.85. Even if the probability is the same (0.85) before and after h2 and h3 are added, the matter has become more elucidated. Keynes would state that the probability is the same before and after the two new pieces of evidence but the weight is higher. ... Keynes’s theory is interesting but his terminology may not be transposed directly to international criminal procedure because the concept “weight” has a different meaning in case law and legal scholarship. ... Strandberg finds Keynes’s theory and terminology ambiguous and incomplete. Instead he suggests the use of other concepts such as massiveness, completeness and robustness, concepts with closely similar meaning. Other alternative concepts to Keynes’s concept of “weight of evidence” include the “resilience” or “comprehensiveness” of evidence. For the present purpose, the concept “robustness” is used when the quantity of evidence is discussed. By robust evidence is meant that additional evidence will not change the probative value. ... the standard of an adequate investigation, and thus robustness, relates to the amount of evidence that should be collected before an evaluation of evidence is possible.
Following this terminology Judge Van den Wyngaer appears to argue that the standard of an adequate investigation has not been meet, the evidence is not robust.