25 Jan International Law’s Invisible Frames Symposium: Inter-Temporality, Legal Fictions and Wartime Emotions in the Investigation of International Humanitarian Law Violations
A Comment on Shiri Krebs’ chapter “The Invisible Frames Affecting Wartime Investigation: Legal Epistemology, Metaphors, and Cognitive Biases”
[Emiliano J. Buis is a Professor of Public International Law at the University of Buenos Aires (UBA) and at the Central University of the Province of Buenos Aires (UNICEN), and researcher at the National Research Council for Science and Technology (CONICET)]
Shiri Krebs’ insightful contribution to the book International Law’s Invisible Frames. Social Cognition and Knowledge Production in International Legal Processes, edited by Andrea Bianchi and Moshe Hirsch, is a thought-provoking piece. The idea defended by the author is that there are normative, cognitive and legal frames that operate in ex post wartime investigations, which have serious consequences in their outcomes. According to Krebs, the existence of future-focused epistemological, cognitive and motivational biases affect the way in which these investigations are deployed, since they can help legitimise actions that seem to be otherwise illegal under international law.
One of the most important assets of the chapter is the belief that the practice of international law heavily relies on extra-legal considerations, and that the role that different actors play in the making and in the interpretation of rules depends on perceptions, which are mostly related to subjective experiences. Krebs acknowledges that the identification of the various biases affecting the way in which international legal rules are understood and applied can be explained through a “behavioural approach”: the actual existence of international law is corroborated through an empirical perspective that can provide a more realistic method capable of overcoming the deficits of theoretical frameworks that can only account for an idealized normativity.
My intention here is not to summarise Krebs’ arguments, with which I mostly agree. I will rather rely on her line of argument in order to draw some conclusions and suggest some questions on the “inter-temporal” fictional mechanism deployed by fact-finding investigators who are tasked with the appraisal of wartime operations. This will allow me to move beyond Krebs’ text in order to discuss the “emotional” elements present at the core of the controversial invention of a “rational” solution (the ‘reasonable commander’ test) to deal with the particular challenges posed by the military experience in situ belli.
Foggy times: the fiction of inter-temporality and the limits of retrospection
In the complex context of war, it has been stated that commanders take decisions based on the (sometimes very limited) knowledge available at hand. The decision to act (or to refrain from acting) is the result of an assessment and evaluation of the unclear circumstances surrounding military operations. In this sense it has been said that it cannot be reasonably expected that commanders possess all the data that should be theoretically required when targeting or launching an attack, and so their conduct has to be examined in accordance with these factual limitations. The “fog of war” —an expression coined by von Clausewitz to refer to this natural vagueness— is the main metaphor used by Krebs to illustrate her argument: since gaps, ambiguities, and doubts are inherent aspects of real-time choices during armed conflicts, the element of uncertainty has been incorporated a posteriori in order to conclude on the legality of military actions that a thorough and unbiased investigation should have rather found unlawful. In spite of the problematic transplant of this “incomplete factual framework” from the context of hostilities to the desk of fact-finding investigators, the inevitability of the “fog of war” has been frequently implemented to assess how commanders behave when deciding on a course of action.
International humanitarian law (IHL) deals with this uncertainty in several provisions, where it is explicitly stated that the inevitable ex ante ignorance should be compensated by an appropriate prevision of possible outcomes. Article 51(5)(b) of Additional Protocol I (AP-I), for instance, prohibits those attacks “which may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated” (italics added, here and elsewhere). A similar language related to future expectations is present in Article 3(3)(c) of the 1980 Protocol II to the Convention on Certain Conventional Weapons, as well as in Article 3(8)(c) of its 1996 amended text.
In some cases, a reference to “knowledge” is also included. According to Article 85(3)(b) of AP-I, “launching an indiscriminate attack affecting the civilian population or civilian objects in the knowledge that such attack will cause excessive loss of life, injury to civilians or damage to civilian objects…” is defined as a grave breach. In the same vein, under the 1998 Statute of the International Criminal Court “intentionally launching an attack in the knowledge that such attack will cause incidental loss of life or injury to civilians or damage to civilian objects … which would be clearly excessive in relation to the concrete and direct overall military advantage anticipated” amounts to a war crime in the context of an international armed conflict (Article 8(2)(b)(iv)).
The evaluation of the future effects of an attack included in these articles is also required to comply with the principle of proportionality enshrined in customary IHL (rule 14 and its practice) according to which the reasonable mental construction of the scenario, founded on the information available at the time, determines whether a commander has acted appropriately or not.
With the aim of examining the circumstances in a way that is compatible with the reality of wartime behaviour, a well-grounded standard —the so-called Rendulic rule—has been developed in ius in bello when discussing the responsibilities arising from the violation of IHL during the conduct of hostilities. When Colonel General Lothar Rendulic was tried by a US military tribunal in the context of the so-called Hostage case (1948) because of the actions ordered in the context of the German operations in northern Norway in 1944, the judges acquitted him of the charge of the devastation of houses, transport and stores in Finnmark on the basis of an examination of the facts in retrospect:
…we are obliged to judge the situation as it appeared to the defendant at the time. If the facts were such as would justify the action by the exercise of judgment, after giving consideration to all the factors and existing possibilities, even though the conclusion reached may be faulty, it cannot be said to be criminal.
The standard, as established by the military court, involved a series of relevant questions both at the subjective and the objective levels. According to B. J. Bill (2009: 136):
What the commander subjectively knew at the time is liable to the difficulties of proof. The commander may testify to what he knew at the time; for example, Rendulic testified that he expected the Russians to attack in force. Of course, commanders in such a situation have a real incentive to minimize the knowledge they possessed. Accordingly, the objective standard of reasonableness is applied, which can take several forms. What information was actually before the commander, and could a reasonable commander be expected to have appreciated the importance of the information? What information was reasonably available to, but not directly before, the commander, and would a reasonable commander have sought it out? Would a reasonable commander have acted in the absence of information?
In its judgment in the Prosecutor v. Galić case in 2003, the Trial Chamber of the International Criminal Tribunal for the Former Yugoslavia (ICTY) insisted on the importance of this Rendulic rule, stating (§58) that:
In determining whether an attack was proportionate it is necessary to examine whether a reasonably well-informed person in the circumstances of the actual perpetrator, making reasonable use of the information available to him or her, could have expected excessive civilian casualties to result from the attack.
The explicit mention of the “circumstances” in which an armed attack takes place shows the importance of time as a key factor in military strategy (as recently discussed by O. Schmitt). In chronological terms, the “reasonable commander” standard, as an ex post consideration, amounts to a fictional return to the past. In her chapter, Krebs identifies this as a “law-fulfilling prophecy”, in the sense that investigators examining the situation and the decisions taken would not consider the reality of what actually happened after the attack, but rather concentrate on the predictions made at the very time of the event. Krebs attributes this dynamic (which she explores in some case studies, mainly the Israeli targeted killing of Hamas leader Salah Shehadeh in Gaza in 2002) to a series of multiple biases that need to be exposed and eradicated, because they perpetuate the justification of unlawful actions though an act of imagination (a fiction, I would say) according to which every operation in the field is always the result of the inevitable gaps present in the (always insufficient) available information. By focusing on estimated forecasts and not collecting information of the actual attack, the reasoning behind these investigations is flawed, in the sense that it disregards the concrete and tangible consequences of military operations.
In my opinion, behind the cognitive and motivational biases identified by Krebs there are other relevant aspects which cannot be overlooked when analysing the attitudes of investigators. I contend that in their reasoning there is a conscious appeal to a fictional “inter-temporality”, borrowing the expression from a doctrine well known in international law according to which —as judge Max Huber expressed in the 1928 decision on the Island of Palmas arbitration case (p. 845)—, “a juridical fact must be appreciated in the light of the law contemporary with it, and not of the law in force at the time when a dispute in regard to it arises or falls to be settled” (on the debates around the concept see Elias 1980). In spite of the difference between its original sense and the dilemma explained by Krebs, I believe that the concept of “inter-temporality” becomes useful in order to describe the artificial displacement of fact-finding investigators or court judges from the present time to a specific moment in the past where the legal situation under review took place.
Inter-temporality constitutes a legal fiction that helps investigators move back in time and overcome the gap between the “external” legal evaluation of responsibility and the “internal” experience of military commanders. By placing themselves in the shoes of the decision-takers (“as if they were” the commander in loco and in tempore), investigators or magistrates develop a sort of “fictional uncertainty”. To conclude about the legality or illegality of a specific action and account for the “truth” of what happened, it seems enough to rely on the scarce knowledge possessed by the commanders at the time of his decision, instead of drawing conclusions from the (well-known a posteriori) objective result of their actions.
Imagining a “reasonable commander”: from rational fiction to emotional reality
An additional point can be made about the “emotional” fabrication of reasonableness in the context of inter-temporality. The traditional “rational” justification of IHL responds to the presence of “objective” rules that strike a balance between antithetical principles: when considering that the conduct of hostilities should be performed on the fragile equilibrium between military necessity and humanity, there seems to be a well-thought decision to assist combatants in dealing with the specific challenges they face during war. A similar explanation can be offered to explain the principle of proportionality and the resulting notion of collateral damage, which are the result of logical assessments of risk and contextual elements that give little room to emotional considerations. Nevertheless, as Krebs demonstrates, the “rational” dimension of normativity does not fully explain the attitudes that different subjects and agents show when discussing the legality of wartime decisions. The imagined position of the investigators and reviewers in the role of the military commander responds to an affective scheme that “humanizes” decision-making.
The contribution of Krebs’ chapter in this regard is also noteworthy. There is still today a wide opinion in favour of excluding personal emotions when endorsing the applicability of IHL. The problem with these traditional positions that insist on leaving emotions aside is that, for the sake of generalizing and consolidating an “abstract” normative system, they lose track of the reality of how IHL is exercised on the ground. As R. Sutton (2021) has recently shown, when studying the efficacy of IHL it is of the utmost importance to pay attention to the human component. Through a careful sociological analysis, she claims that the relevance of emotions in IHL is related to the role of specific actors in the field, who express specific feelings and perceptions in their social interaction.
Thinking outside the conventional legal framework and relying on the practical importance of emotions require considering new perspectives that can be implemented to better assess the nature and scope of ius in bello, both in theory and practice (cf. N. K. Modirzadeh, 2020). In its study on The Roots of Restraint in War (2018), the International Committee of the Red Cross (ICRC) has acknowledged the limits of a “legalistic” approach when dealing with the applicability of IHL and (contrary to the content and spirit of its 2004 previous study on the matter) promoted the need for interdisciplinary tools and strategies to understand the reasons behind the respect of rules regulating armed conflicts. As D. Stephens (2014) has affirmed when examining the place of law, morality and self-identity in wartime behaviour, the role of emotions and collective values in IHL should not be underestimated.
A clear example of one of the areas in which this debate has grown stronger in recent times is provided by artificial intelligence and lethal autonomous weapons systems. Many argue that, by removing human emotions such as anger or fear, killer robots seem to be more ready to engage in hostilities without the possibility of making mistakes. The absence of any emotional experience would allow, according to this position, for an unbiased participation in hostilities. But this, of course, is not clear, since the absence of combatant decision-making ends up de-humanising the conduct of hostilities and reducing war to a cold and inexpressive algorithm.
Likewise, the standard of the “reasonable commander” —which could be employed to show which emotions should be adequately felt and expressed by military leaders— has also given rise to a falsely “rational” belief that the best soldiers are those who can avoid becoming too attached to their surrounding circumstances. However, as Saab (2021) has correctly pointed out,
International humanitarian law uses the concept of ‘reasonable commanders’ especially in assessing the proportionality principle. Reasonable commanders are reasonable not because of a lack of emotions, but precisely because of their ability to experience human emotions, their ability to feel empathy, show compassion, and understand the importance of dignity.
To consider the honest and reasonable estimate of the facts which are available to a commander seems to be an objective way of weighing the proportionality of an attack in retrospect. But, at the same time, this standard relies on an empathic real-time approach to the commander in times of difficult choices. As stated by R. Bartels (2013: 279) quoting N. Hayashi, “while judges need not have been soldiers themselves, some degree of familiarity with, and sensitivity to, the nature of the military profession is essential for realistic judicial decisions”. Paradoxically, then, the objectiveness of the “reasonable commander” standard is based upon the subjective need to “feel” the circumstances as they existed in the theatre of operations. Although the creation of a legal standard to identify reliable decision-takers during a military operation seems a convenient fiction, investigators cannot evaluate any compliance with IHL if they ignore the context in which a measure was really decided, an order was really given or an action really took place.
As Krebs correctly points out, it seems adequate to expect a fact-finding commission to investigate and collect ex post data in order to understand the nature and availability of the information considered by a commander when launching an assault; this critical exercise helps overcome the reproduction of an incomplete factual history and avoids creating legalities where, in fact, illegal acts were committed. However, at the same time, it should not be ignored that the use of figures, evidence and material only gathered after the facts can become problematic when addressing the individual responsibility of a military superior acting in the heat of an armed conflict.
Although criticized by Krebs as generating “faulty processes and human insecurity”, the method of limiting the examination to the knowledge accessible when a course of action was adopted pays attention to the emotional context of decision-making and brings reassurance in times of considerable uncertainty. Therefore, when evaluating responsibility, a behavioural approach such as the one observed by Krebs cannot ignore the affective elements involved in the concrete measures taken to ensure legal security in the foggy times of armed violence. As an invisible frame, a realistic, practical and human perspective of IHL has to guarantee also that no upright “reasonable commander”, performing his/her duty in good faith, should fear that the actions undertaken would be judged in the future on the basis of elements which are not yet at his/her disposal.
I very much appreciate this thoughtful engagement with my chapter. I am also grateful for the opportunity to clarify an important point that was only mentioned in the chapter due to the limited space available. While the chapter does suggest a move toward outcome-based evaluation of military operations, identifying gaps between ex ante fact-finding efforts and the actual conditions on the ground, it does so in a particular context: improving military fact-finding and risk assessments processes. This argument does not extend to criminal investigations or individual criminal responsibility, where the knowledge perception of the individual decision-maker should remain central. As I note in the chapter, individual decision-makers should continue to be judged based on their imperfect perception of the available facts in real time. However, the limited and biased nature of real-time fact-finding processes should not deter fact-finders from reshaping this knowledge ex post, lifting the ‘fog of war’ and identifying cognitive processes affecting ex ante knowledge production.