The "no case to answer" standard and the process of fact-finding in the Gbagbo case

Trial Chamber I issued yesterday its reasons for the oral decision of 15 January 2019 in the Gbagbo case. The decision of the majority is quite short, instead one needs to read the reasons of Judge Geoffrey Henderson and the opinion of Judge Cuno Tarfusser to understand the majority.

The introduction of the "no case to answer" procedure at the ICC
I have had some concern about the introduction of the "no case to answer" procedure at the ICC since such a procedure is neither provided for in the Rome Statute nor in the Rules of Procedure and Evidence. I have previously held the same view as Burrow (2010 in Khan et al., p. 691) that the confirmation of charges proceedings could fulfil the same purpose, namely to ensure that only cases in which the prosecution’s case is sufficiently strong to commit the person to trial, are confirmed (Klamberg, 2013, p. 147). The "no case to answer" procedure does not fit well with the other potential features of the ICC procedural framework as adopted in some of the cases, as illustrated by the Gbagbo case. Judge Henderson comments in the following way about the fact that the Trial Chamber has not made any admissibility rulings.

this Trial Chamber did not make any admissibility rulings. This means that there has been absolutely no filter on what the parties were able to submit into the case record. This has resulted in the case record being flooded with documents of doubtful authenticity as well as documents containing significant anonymous hearsay. This sits uncomfortably with the traditional Adversarial/Common Law no-case to answer test, which instructs Chambers to consider the Prosecutor’s evidence at its highest. (para. 4)

However, I will not devote the rest of this blog post arguing that it was wrong to introduce the "no case to answer" procedure at the ICC. The Appeals Chamber has in Ntaganda confirmed, that trial chambers may decide to conduct a "no case to answer" procedure on the basis of their power to rule on relevant matters pursuant to article 64(6)(f) of the Statute and rule 134(3) of the Rules (Judgement 5 september 2017, para. 44). Thus, it is procedure available at the ICC and none of the parties - not even the prosecution - appear to argue that the availability of the procedure is wrong as such. However, the mere availability of the procedure may have consequences for other parts of the ICC procedural framework, including for the confirmation of charges proceedings and admission of evidence. The different components of the systems are interrelated and there are some limits on the suitability of different procedural combinations.

The "no case to answer" standard and the process of fact-finding
Instead I will focus on the "no case to answer" standard in the Gbagbo case. It would however  restrict this discussion to be only about a "standard" since my view is that evidentiary standards also have consequences for the method of evaluating evidence, it would therefore also be suitable to describe as a process of fact-finding. In other words, how did Judge Henderson go about when deciding to end the case against Mr Laurent Gbagbo and Mr Charles Blé Goudé?

Inventory of potential components or steps to take when evaluating evidence
Before we assess how Judge Henderson reasoned we could make an inventory of potential components or steps to take when evaluating evidence. I have previously (Klamberg, 2013 and Klamberg, 2015) discussed different components or steps to take in relation to evaluating evidence. I have also argued that this depends on the stage of the proceedings. This begs the question, which steps has Judge Henderson taken? In order to make the final determination of guilt all steps need to be taken at some stage of the proceedings, including:

1. The judge must ascertain what elements and facts in issue need to be established
2. The evidence has to be structured with, applying the following components and questions.
i) Is the evidence relevant in order to prove the charge? If the proposition is deconstructed into parts, one may exclude irrelevant circumstances.
ii) What is the bearing of the evidence, is it a part of the evidence for the prosecution or for the defence?
iii) What kind of evidence has been submitted? Is it “direct” or “circumstantial” evidence?
Yvonne McDermott has similarly pointed to the utility of Wigmorean Analysis (McDermott, 2015).
3. The probative value and weight relevant for each fact in issue has to be determined. Reliability and credibility are relevant factors when assessing probabative value.
4. If several evidential facts, i.e. several pieces of evidence (for example witness testimony, documents, etc), are relevant to a fact in issue, the Court has to weigh the pieces against each other. I try to avoid the phrase "holistic evaluation of evidence", but if the phrase is ever to be used it is in this context when weighing evidence pieces of evidence against each other in order to asses a specific fact in issue. At this stage of evaluating evidence, statistical methods could be used with caution. Frequency-type probability is useful for assessing DNA evidence and similar types of evidence, but arguably not for assessing the entire body of evidence in a case. Other probability methods could be used with caution - as thumb rules only - when assessing corroborative and circumstantial evidence.
5. When the probative value of the evidence has been determined the court must be satisfied that it is strong enough to meet the standard of proof. Three factors have to be considered before a final determination is made as to whether the standard has been satisfied.
i) The court must first weigh the evidence in favour or against the proposition.
ii) Second, the requirement on the standard for an adequate investigation must be satisfied. If the standard for an adequate investigation has not been satisfied and the evidence is not robust, the judges have the options to a) do nothing during the trial and wait until the final deliberations to acquit the accused; b) dismiss the case through a summary judgment or c) order the production of more evidence.
iii) Third, as part of evaluating whether the weight of the evidence meet the standard of proof, the judges should test and eliminate alternative hypotheses which could give a different reasonable explanation to the charges made by the prosecution.

I am going to read Judge Henderson's reasons with these steps in mind.

The "no case to answer" standard and the process of fact-finding at the ad hoc tribunals
Let us first consider the procedure at the ad hoc tribunals. The original ICTY rules of procedure and evidence had no specific rule concerning motions to dismiss the prosecution case through a summary judgment. In both the Tadić case and the Delalić et al. case motions of no-case-to-answer were filed. In both cases the Trial Chambers examined the motions, which shows the Chambers’ concern for the rights of the accused (Tadić, ICTY T. Ch., 13 September 1996 and Delalić et al.  ICTY T. Ch., 18 March 1998). The Trial Chamber rejected the motions on their merits. However, they could also have dismissed the motions on the ground that, at the time, there was no rule of no-case-to-answer and the accused was entitled to be presumed innocent. As a consequence of these decisions Rule 98 bis was adopted allowing for filing of motions of acquittal. The rule went even further, conferring on the Trial Chamber at the close of the prosecutor’s case the proprio motu power to enter a judgment of acquittal on any count if there is no evidence capable of supporting a conviction. In Jelisić the Appeals Chamber set the standard with regard to the test to be applied under rule 98 bis, namely whether the “evidence, if believed, is insufficient for any reasonable trier of fact to find that guilt has been proved beyond reasonable doubt.” The test is not whether the trier would in fact arrive at a conviction beyond reasonable doubt on the prosecution evidence (if accepted) but whether it could (Jelisić, ICTY A. Ch., 5 July 2001, para. 37).

The "no case to answer" standard and the process of fact-finding in Ruto and Sang
ICC Trial Chamber V(A) adopted a similar approach in Ruto and Sang stating that "the test to be applied in determining a ‘no case to answer’ motion […] is whether there is evidence on which a reasonable Trial Chamber could convict", adding that "the test to be applied in determining a ‘no case to answer’ motion […] is whether there is evidence on which a reasonable Trial Chamber could convict" (Ruto and Sang, Decision No. 5 on the Conduct of Trial Proceedings (Principles and Procedure on ‘No Case to Answer’ Motions), 3 June 2014, ICC-01/09-01/11-1334, para. 32.).

The "no case to answer" standard and the process of fact-finding in Gbabgo 
Kevin Jon Heller hoped in January 2019 to see the standard of proof expressed in the phrase “evidence (if accepted) upon which a reasonable tribunal of fact could be satisfied beyond reasonable doubt of the guilt of the accused” in the forthcoming decision.

The majority decided not to follow Ruto and Sang (Henderson, paras. 1-8; Cuno Tarfusser, para. 67)

Considering the approach to admissibility taken in the Gbabgo case Judge Henderson stated that "[i]t cannot be assumed ... that all the evidence on the record has at least some minimal probative value" (para. 5) and decided to adopt a different approach by engaging in "a full review of the evidence submitted and relied upon by the Prosecutor" (para. 8). What standard of evidence does that entail? Judge Henderson explains that "a decision that there is no case to answer is not a formal judgment of acquittal on the basis of the application of the beyond reasonable doubt standard in accordance with article 74 of the Statute" (para. 17) and does not use that phrase in any other instance of his reasons. Does that mean that Kevin Jon Heller should be disappointed? What is the standard used by Judge Henderson?

My reading is that of Judge Henderson reasons is that he takes all the five steps that I have described above in evaluating the evidence. I would like to highlight some of the steps taken.

1. Judge Henderson explains in para. 91 and parts IV-VI which elements and facts in issue that need to be established. The evidence submitted by the Prosecution is structured in the manner understood by the Judges and an assessment is made of the probative value and weight relevant for each fact.

2. Above I use the phrase "standard for an adequate investigation" and robustness which relates to question whether the prosecutor has gathered evidence that covered all the facts in issue, a matter of quantity. Judge Henderson does not use these word but appears to question the investigation, as illustrated in the following phrase.

As will be noted several times throughout this opinion, although the available evidence is voluminous, a lot of essential information is still missing. Whether this is due to the information being unavailable to the Prosecutor or to the fact that the Prosecutor did not look (hard enough) for it is not for me to say." (para. 5)

He also finds the available evidence far from complete (para. 67).

3. In part IX of the reasons Judge Henderson entertains as an option to ending the case, the idea of recharacterising the charged facts and circumstances under regulation 55. This would under subregulation 3(b) allow the parties to call a new witness or to present other evidence. This option is dismissed since none of the parties has made a serious effort to raise the question of the existence of an armed conflict during the period relevant to the charges.

4. Judge Henderson describes in part III the prosecutor's case theory as a "narrative", he criticises the Prosecutor's "one-sided version of the situation in Côte d’Ivoire" (para. 66), presents an alternative to the Prosecutor's case theory of the President being in control (para. 67) and points to "alternative reading" of the evidence (para. 1171).  Judge Henderson describes how the Prosecutor has failed to prove that "the combined effect of all her evidence is greater than the sum of the individual (non-incriminating) parts" (para. 88).

Judge Cuno Tarfusser appears to have a similar approach when he states the following.

the evidence on the record not only fails to convince me that any of the charged incidents did indeed occur pursuant to the Prosecutor’s narrative, but is rather suitable to point to one or more alternative readings which are equally, if not more, plausible. (para. 75)

In other words, the Prosecutor has failed in eliminating all reasonable, alternative explanations to the Prosecutor's case theory. Judge Cuno Tarfusser's statement suggests that the Prosecutor even has failed to reach the balance of probabilities standard.

In conclusion, the Judge Henderson has taken steps in relation to "no case to answer" motion which are very close or almost identical to a finding under the "beyond reasonable doubt" standard. Is an appeal by the Prosecution warranted, has the trial chamber used the wrong evidentiary standard? My reading of the Judge Henderson's reasons is that it is not only a blow against the Prosecution, is also a rebuffal of the confirmation of charges decision. Evidence may have been missing already with a lower standard, as suggested by the dissenting opinion of Pre-Trial Judge Christine Van den Wyngaert.

Making the procedural framework coherent and efficient
As argued above, the mere availability of the "no case to answer "procedure may have consequences for other parts of the ICC procedural framework, including for the confirmation of charges proceedings and admission of evidence. I am not sure if the ICC has got their approach to admissibility of evidence right, not even if one has a civil law approach. In para. 20 of Judge Henderson's reasons one may read that "the Chamber by majority decided that decisions on the admissibility and relevance of the evidence submitted by a party ... will be deferred to the final judgment". This should be compared with Judge Cuno Tarfusser who argues that the evidence should have been seriously filtered in  order to shape the trial (para. 20). On the other hand he argues that "to the so-called admissibility ruling system, and [his] preference for a system whereby the evidence as a whole is considered within the context of the final determinations of the trial" (para. 69). I would argue that it is fine to asses credibility and reliability in the final judgement, but even in continental European systems one may find mechanisms whereby judges consider relevance before the evidence is allowed to be presented and be part of the submitted evidence.