Spirituality and Duress: Local Culture Beliefs at the International Criminal Court

[Dr. Sigurd D’hondt is an Associate Professor at the Department of Language and Communication Studies, University of Jyväskylä (Finland). Dr. Juan-Pablo Peréz-León-Acevedo and Dr. Fabio Ferraz-de-Almeida are researchers associated with the project reported on here, and Elena Barrett is a PhD candidate at the department.]

Introduction

On 4 February 2021, Trial Chamber IX (TC-IX) of the International Criminal Court (ICC) convicted Dominic Ongwen of 61 charges corresponding to crimes against humanity and war crimes, by far the largest number in the ICC’s history. On 6 May 2021, he was sentenced to 25 years in prison. The conviction and sentencing have been appealed. Ongwen’s trial presented several unprecedented complexities for the ICC and, possibly, international criminal justice at large. The most obvious relates to the defendant’s past: Ongwen was recruited in the Lord’s Resistance Army (LRA) as a child soldier but eventually rose in rank to the level of commander, which presented pressing questions about the dividing line between perpetrator and victim. 

However, there is also a subtler, more intricate legal dilemma at play here, and that has to do with testimony about local spirituality and belief. The defence placed spirituality at the centre of its case, arguing that it provided grounds for excluding Ongwen’s criminal responsibility under the defence of duress, enshrined in article 31(1)(d) of the ICC Statute. Kony’s invocation of Christian dogma intertwined with local spirituality, the defence maintained, was extremely effective in instilling belief (coupled with a healthy dose of fear) in LRA members. These attempts to exculpate Ongwen turned out to be fruitless, and the defence failed to convince the ICC Judges that the widespread belief in the spiritual powers of Joseph Kony fulfilled the requirement of ‘a threat of imminent death or of continuing or imminent serious bodily harm against that person or another person’, which is a key element of duress under article 31(1)(d) of the Statute. In its assessment of the evidence, TC-IX developed an argument to support this rejection that appears sound and straightforward, as shown in figure 1: 

TC-IX acknowledged the centrality of Acholi spiritualism within the LRA but concluded that ‘for many persons who stayed in the LRA longer’ these beliefs also evolved over time. This evolution allegedly corresponded to a fixed pattern: newly abducted children believed in Kony’s spiritual powers, but this belief weakened and dissipated as they grew older and remained in the LRA long enough to see and experience evidence to the contrary.

From here on, TC-IX’s assessment of the evidence about spirituality continued with a series of references to witnesses’ testimonies that illustrate this pattern. Figure 2, the next paragraph from the judgment, is an example of this:

In this blog post, we problematize the apparent clarity that TC-IX found in the evidence submitted by the parties. To make our point, we will draw on insights from linguistic anthropology on the role of language in the legal process before moving on to potential legal-normative implications. 

A Linguistic-Anthropological Reading of the Ongwen Judgment

Linguistic anthropologists have amply documented how texts travel through the different stages of the legal process and how these chains of continuous de- and recontextualizations may eventually result in a transformation in the meaning of the evidence. Often, such transformations are premised on a reading practice, a particular mode of working with and interpreting written texts, that only takes into consideration literal content but loses sight over how meaning is negotiated in interaction. In what follows, we draw on these insights to chart how TC-IX arrived at its conclusion. We reconstruct the ‘trajectory’ of the evidence to bring out the implicit reasoning behind TC-IX’s rejection of the duress argument. To make our claim, we will engage in an analytic exercise with which (some) readers of this blog may be less familiar: we will extensively quote from trial transcripts, to confront the clarity of the decision with the nitty-gritty of questions and answers through which evidence is delivered in the courtroom. Looking in more detail at what exactly the witnesses said, as well as when and in response to what, will help us to better understand the testimony and, perhaps, assess the consistency of the judgement.  

We have done this exercise for all nine testimonies quoted in the judgment as substantiating the pattern that we could retrieve from the ICC website. In this blog post, however, we specifically focus on the testimony of witness P-0231 referenced in paragraph 2646 above, the first ‘explanation’ of the pattern quoted by the Judges. In the footnotes, the judgment refers to two passages from the hearing transcript, shown here as figures 3 and 4:

It immediately becomes clear that the evidence substantiating the pattern was produced in response to different questions, posed by different participants for different purposes. The footnotes also refer to very short stretches of talk, leaving aside the interactional context in which the statement was made. The ‘explanation’ of the pattern cited by TC-IX is thus not referencing a monological statement by P-0231 delivered in a continuous stretch of talk, as an isolated reading of the judgment might suggest. Instead, the pattern is constructed out of two separate question-answer pairs: 1) an affirmative response to a question by the defence about whether the witness believed in Kony’s spiritual power at a specific moment (1995) and specific location (Palutaka), and 2) an account of how he grew sceptical of Kony’s powers while still in the LRA, in response to a question by the judge.

Interpreting, assessing and referencing evidence always requires the strategic inclusion of some elements to the detriment of others. In this case, however, the selectivity with which the Judges quote short bits of testimony effectively alters the meaning of that testimony, even if the passages in question are requoted verbatim in the verdict. This becomes increasingly tangible once we take a step further back and consider the surrounding talk before and after the excerpts referenced in the judgment. Figure 5 below is a more comprehensive rendition of figure 3. It illustrates the extent to which meaning produced in testimony can be altered through ignoring its wider interactional context:

Once the Judge obtains the answer that later reappears as part of ‘the pattern’ (lines 12-14), he returns the floor to the defence counsel. The latter in turn invites the witness to recontextualize his previous answers, by asking two targeted questions about the time frame of his belief (lines 17-18 and 20-21). In the answer (only partially reproduced here), the witness responds with a biographical account that refers to personal and political rifts within the LRA and is set in ‘historical’ time. This specificity of time and place not only refutes the abstract and decontextualized temporality that characterizes ‘the pattern’ (as a ‘psychological’ regularity), but also situates this senior LRA figure’s loss of faith after the facts for which Ongwen was charged (e.g. the attacks on the Abok and Lukodi IDP camps took place in the spring of 2004). His answers furthermore suggest that his growing scepticism had as much to do with a loss of ‘trust’ in the LRA’s political project as with spirituality per se. Again, this nuance is entirely lost in the Trial Judgment. 

In such testimony on spirituality, one also frequently encounters alternations between passages where the witness ‘exhibits’ beliefs (e.g. when he describes the changes in Kony’s appearance while he was allegedly possessed by Juma Oris) and episodes where he ‘claims’ beliefs (e.g. his responses to the Judge’s question in lines 12-14). Only the latter are referenced in the Trial Judgment. Sometimes open conflicts arise between these ‘claimed’ and ‘exhibited’ beliefs. Another witness whose testimony was quoted by the Judges as exemplifying the pattern (witness P-0205, cited in paragraph 2650) explicitly disavows believing in Kony’s spiritual powers, but at the same time ‘exhibited’ his acceptance of the LRA commander’s ability to predict the future – an apparent inconsistency that the defence kept hammering on in cross-examination (Trial transcript 8 March 2017, starting at line 20 of page 22).  Here as well, only the witness’s explicit claiming of disbelief made its way into the verdict. This further exemplifies how the practice of selectively quoting contributes to projecting a ‘binary’ image of the LRA as a belief-environment in which all grey areas are removed. The judgment leaves no room for the potentially ambiguous ways in which individual LRA members internalized, appropriated and/or resisted the group’s ‘master narrative’ about Kony’s spiritual powers. It is questionable whether such binarism can do justice to the complexity of navigating a civil war-like situation in which multiple normative orders compete for legitimacy. 

What is the significance of this? Nistor, Merrylees and Holá identified four modes in which ICC trial actors interact with local understandings of spirituality, arguing that the ‘description’ of spirit beliefs alternates with their ‘translation’ into the categories of international law, which may eventually result in the ‘mutilation’ and ‘rationalization’ of these beliefs (where trial actors assess spirit beliefs based on their presumed truth-value). By analyzing the courtroom talk through which evidence about spirituality is produced and how such testimony is selectively recontextualized in the judgment, we elucidated the discursive mechanisms through which translation and mutilation operate. 

This linguistic-anthropological analysis has significant normative implications. First, it suggests that TC-IX failed to properly examine the substance of the duress elements and their context as applied to this case, in particular its subjective, cultural and spiritual aspects. Instead, the Chamber dodged the issue by primarily invoking evidentiary issues (see also Ambos, p. 471). The spiritual elements that were so constitutive of this case necessitated greater consideration of culture-related elements, both in the trial proceedings and in the interpretation of the evidence as it came to light in the judgment. This more nuanced mode of interpretation is justified under article 31(2): ‘The Court shall determine the applicability of the grounds for excluding criminal responsibility provided for in this Statute to the case before it’ (emphasis added). 

In the end, the Court’s resistance to a more context-sensitive analysis of the cultural elements related to duress could be perceived as privileging ‘western’ views (see Nistor, Merrylees and Holá and Titeca), and hence reinforcing the African backlash against the ICC. It also raises further questions as to whether the Court is able to guarantee the accused person the right to a fair trial (see Higgins). To circumvent these criticisms, the ICC should analyse the cultural aspects of the duress defence more thoroughly – not just in black and white but also allowing for shades of grey, if not full colour. 

TC-IX’s unwillingness to analyse cultural elements more comprehensively also prevented them from tackling a notable indeterminacy within the ICC Statute’s article on duress. Article 31(1)(d) partially merges ‘necessity’ and ‘duress’, which increases the burden inherent in successfully invoking a duress defence. As Ambos (p. 461) has explained, ‘necessity’ is a justification (unlawfulness eliminated by the choice of the lesser evil) while ‘duress’ is an excuse (unlawfulness is not eliminated but only tolerated, as the accused could not fairly resist the threat due to human frailty). This distinction requires further jurisprudential clarification, especially in cases where local cultural elements play such a central role. From this angle, the Chamber missed an opportunity to distinguish more clearly, and in a more culturally informed manner, between these two defences. 

Conclusion

In light of this analysis, it is questionable whether, and to what extent, the ICC is the most appropriate transitional justice forum to handle complex cases like this one. Situations that require very specific, expert culture-based analyses also require a context-informed discussion and context-sensitive application of key legal categories. Other transitional justice alternatives (or supplements), including national criminal courts, truth and reconciliation commissions and ‘non-western’ local justice mechanisms, could and should also be considered. In any event, more attention should be drawn to approaches guided by restorative justice, so that victims, defendants, and local communities are granted a more central role in dealing with atrocities committed in complex cultural contexts.