Has the majority in the Bemba case treated circumstantial evidence with the logics that apply to hearsay evidence?

The Appeal Judgement in the Bemba case has resulted in a debate on several issues, including evaluation of evidence. I would like to highlight two paragraphs in the separate opinion of Judge Christine Van den Wyngaert and Judge Howard Morrison.

11. We are also concerned about the fact that the Trial Chamber relied on a large amount of circumstantial evidence in relation to a number of key findings. Again, the Trial Chamber stated the correct principle that circumstantial evidence can only lead to findings beyond a reasonable doubt when the proposed inference is the only plausible one, but has often failed to adhere to this principle in its actual analysis.  

12. For example, in paragraphs 676 to 684, the Conviction Decision lists eight circumstantial factors that it considered cumulatively proved the existence of a policy to attack a civilian population. We are far from persuaded that there was sufficient evidence to support the eight ‘factors’ that were relied upon. In this regard, it is sometimes argued that only the material facts must be established beyond a reasonable doubt and that it is unnecessary to establish subsidiary facts to the same standard.7 While this is legally correct, it does not mean that the quality of the evidence for subsidiary facts is irrelevant from an evidentiary point of view. This is especially true in relation to circumstantial evidence. By definition, drawing inferences from circumstantial evidence only adds uncertainty. Therefore, if the factual basis of the circumstantial evidence is weak, the inferences drawn from it will be even weaker.

I have a problem especially with the last sentence which is arguably wrong: "if the factual basis of the circumstantial evidence is weak, the inferences drawn from it will be even weaker."

To the contrary, even if separate pieces of evidence are too weak by themselves to prove guilt, the combined evidentiary value will be stronger than any of the individual pieces if the pieces of evidence are independent and seek to prove the same fact in issue. These logics apply to circumstantial evidence which is a form of concurrent (corroborative) evidence. The opposite logics apply to successive (chain) evidence, all pieces of evidence in a chain that are less than certain will have the result that the combined inference will be even weaker. Hearsay is an example of this. From the paragraph above it appears as the judges in the majority erroneously have applied the logics of successive evidence when evaluating circumstantial evidence (concurrent evidence). In other words, has the majority in the Bemba case treated circumstantial evidence with the logics that apply to hearsay evidence?

I have written about this in my book Evidence in International Criminal Trials (2013), pages 177-179 and more recently in the article The Alternative Hypothesis Approach, Robustness and International Criminal Justice (2015). Let us compare successive (chain) evidence and concurrent (corroborative) evidence while leaving counterevidence aside.

First, successive (chain) evidence concerns evidentiary facts that each are links in a chain. The links all have a probative value that is less than certain, which will have the result that the combined evidentiary weight can never be higher than the evidentiary value of the weakest link. The rule of thumb, or cautionary advice, is that successive evidence is often overestimated because the focus of the evaluation of evidence tends to be towards the last link of the chain neglecting the previous links. This is the reason why the evidentiary value of hearsay evidence (which is a form of successive evidence) tends to be weaker than direct evidence. International judges appear to be aware of this danger. The International Criminal Tribunal for the former Yugoslavia (ICTY)Appeals Chamber has in Aleksovski and Kordic and Cerkez listed indicia of reliability for hearsay evidence where of one indicia is whether it is ‘first-hand or removed’. This approach has been repeated at the ICTY and at the International Criminal Tribunal for Rwanda (ICTR), Special Court for Sierra Leone (SCSL) and International Criminal Court (ICC); case law suggests that hearsay evidence normally should be afforded less probative value or weight.

Secondly, counterevidence…  

Thirdly, concurrent (corroborative) evidence concerns two or more independent pieces of evidence that concurs and separately has a probative value supporting the fact in issue. An example would be two witnesses that observe the same event independently of each other. The probative value in such cases may be higher than the highest probative value of any of the separate pieces of evidence, which [at a first glance] may appear illogical. The rule of thumb is thus that in cases of concurrent evidence the weight is, according to Ekelöf, often higher than what is normally expected. With a similar logic, judges at the ICTR and ICC have declared that they attach higher probative value to those parts of a testimony which may be corroborated;  

Klamberg, The Alternative Hypothesis Approach, Robustness and International Criminal Justice (2015), pages 540-541

I dont have a problem with what  Judge Christine Van den Wyngaert and Judge Howard Morrison write about [alternative] plausible explanations in para. 14, it fits very well with my take on evaluating evidence. It is just another part in the process.