Excuse and justification: What’s explanation and understanding got to do with it?

Abstract

A well-worn French proverb pronounces ‘tout comprendre c’est tout pardonner’ (‘to understand all is to forgive all’). Is forgiveness the inevitable consequence of social scientific understanding of the actions and lives of perpetrators of serious wrongdoing? Do social scientific explanations provide excuses or justifications for the perpetrators of the actions that the explanations purport to explain? In this essay, I seek clarification of these intertwined explanatory and moral questions.

Introduction

A well-worn French proverb pronounces ‘tout comprendre c’est tout pardonner’ (‘to understand all is to forgive all’). ¹ Is forgiveness the inevitable consequence of social scientific understanding of the actions and lives of perpetrators of serious wrongdoing? Do social scientific explanations provide excuses or justifications for the perpetrators of the actions that the explanations purport to explain? Many think that social science explanation and understanding does indeed have these effects, implications or meanings, and for this reason they are hostile towards, or suspicious of, it.

The issues arise most dramatically, starkly and clearly with social scientific studies of events of extreme violence such as the Holocaust, genocide, terrorism and crimes against the person. Some have taken the supposed excusing or justifying effects of explanation and understanding to be decisive reason for refusing even to attempt the explanation and understanding of heinous events. Thus, for example, the celebrated survivor of Auschwitz and writer Primo Levi (1995, p. 227) maintained that ‘one cannot, what is more one must not, understand what happened, because to understand is almost to justify’. Likewise, the psychologist and concentration camp survivor Bruno Bettelheim (1986) reports: ‘I restricted myself to trying to understand the psychology of the prisoners and I shied away from trying to understand the psychology of the SS – because of the ever-present danger that understanding fully may come close to forgiving’. These sentiments were echoed by John Major, UK Prime Minister in 1993, in the aftermath of the brutal murder of 2-year-old James Bulger by two 10-year-old boys, when he pronounced that ‘society needs to condemn a little more and understand a little less’ (McNutt, 2010).

Social scientists are often keenly aware that their explanatory and interpretive endeavours are likely to be seen by other people as having excusing or justifying consequences or implications. Social psychologist James Waller (2007, p. 17), for example, concedes that there is a legitimate ‘fear that to explain the behaviors of perpetrators of extraordinary evil is to justify those behaviors’. However, most summarily dismiss the very idea of explanation and understanding leading or amounting to excuse or justification. The eminent historian of Nazi Germany and the Holocaust, Christopher Browning (2001, p. xviii), proclaims: ‘I do not accept the old clichés that to explain is to excuse, to understand is to forgive. Explaining is not excusing; understanding is not forgiving'. Likewise, Philip Zimbardo (2007, p. xi), the social psychologist famed for his Stanford Prison experiment, insists that ‘attempting to understand the situational and systemic contributions to any individual’s behaviour does not excuse the person or absolve him or her from responsibility in engaging in immoral, illegal, or evil deeds’. And psychologist Susan Fiske (2013, p. 605), in an article on ‘extremism’, warns: ‘make no mistake: understanding is not condoning’.

A very few brave social scientists come close to accepting, or at least not disowning, a definite relation between explaining and understanding on the one hand, and excusing, justifying or forgiving on the other. Social psychologist Arthur Miller (2016, p. 207) is one such, averring that ‘the partial exoneration of perpetrators’ is ‘a direct implication of situationism’. ²

The above survey of stances might seem to cover all possible bases, that is, that explanation/understanding: (i) should be foresworn because it leads to excuse, justification or forgiveness; (ii) that it has no such effects or implications and is therefore safe to pursue; (iii) that it does issue in some degree of excuse, justification or forgiveness but this is not untoward or unwarranted, and perhaps even to be welcomed. However, I do not think the key questions can be asked so bluntly and answered so unequivocally. Each of the principled stances reported above contain confusions and conflations – so much so that no clear answer can emerge when expressed in these simplistic terms. I hasten to add that I do not want to imply that the scholars quoted above are theoretically obtuse. Rather, the stances they take call for philosophical reflection, analysis and clarification, and this is what I undertake in this essay. The key issues to be explored are: (1) the relevant characteristics of social scientific explanation and understanding; (2) the nature of, and distinction between, excuse and justification, and the conditions for forgiveness; (3) the modality of the relation between explanation/understanding and excuse/justification; and (4) what it is about social science explanation and understanding that has bearing on the conditions for excuse and justification.

Explanation and understanding

I don’t want to venture into the essential form of explanation or understanding, nor be unduly formal or prescriptive. The main characteristic I attribute to social scientific explanation is that it (purports to) tell us something about a social phenomenon of interest that neither the actors involved nor the explanation’s intended audience knew prior to the explanation. That is, explanation goes beyond the self-understanding of the actors involved and beyond common-sense observation and beliefs about what the actors are doing and what motivates them. In a word, explanation aims to be revelatory. What makes an explanation a social, or social scientific, explanation is that it explains the actions, beliefs and motivations of people through the identification of social causes and conditions that impact upon their perception, beliefs, opportunities and actions.

Understanding can be just that in which a good explanation issues, whereby the explanation facilitates the grasp of something illuminating or insightful about the actors or their actions, beliefs, perceptions or attitudes. An explanation that doesn’t enhance our understanding is not much of an explanation! But also, understanding might be the aim of a particular, non-explanatory, non-generalising, kind of inquiry that issues in a distinctive kind of insight into and appreciation of the experiential lifeworld of a particular social group. This is the practice and delivery of Verstehen, where the desideratum is to gain an understanding and appreciation of the way of life of a particular group which is rooted in its members’ self-understanding. It is, famously, to attempt to see and experience how things are from the point of view of those that one wants to understand, in all their depth and particularity.

Because explanation and understanding are so intertwined and complementary, henceforth I will often just refer to explanation, but this should be read expansively, to include both the understanding that explanation might yield and specifically ‘interpretive’ aspects and forms of inquiry. However, the final section of this essay finds a quite distinctive role for Verstehen that transcends propositional belief and understanding.

Justification, excuse, forgiveness

In the above quotations (and in plenty more that I could have quoted) of social scientists on the mitigatory effects or implications of explanation, the operative words are ‘excuse’, ‘justify’, ‘forgive’, ‘exonerate’ and ‘condone’. I want to begin by clarifying the nature of, and distinction between, excuse and justification. These concepts are frequently conflated or misapplied, both by social scientists and non-social scientists. ³ In clarifying these concepts, I am not simply imposing philosophical legislation. The nature of excuse and justification, and the distinction between them, is embedded in the criminal law, which is itself a reflection of the use of the concepts in everyday life (Baron, 2007, p. 22 n1).

Excuse: An agent may be excused for performing a morally wrongful action if there was something about their capacities or circumstances that made it difficult for them to avoid or refrain from acting thus, or to know that in doing so they were acting wrongly. The common dismissal ‘that’s just an excuse!’ should be noted. The phrase is linguistically odd because it actually means the opposite of what is explicitly asserted, namely, that what was proffered is a counterfeit, not a genuine, excuse, thus not an excuse at all. A genuine excuse can, and in some cases should, issue in the agent not being blamed, or their blame being diminished, for their wrongful action.

Justification: Justification comes from the moral status of the agent’s action, signifying that it is permissible, not morally wrong (Baron, 2007, p. 22), or is morally right, and the agent is therefore justified in performing it. The question of justification only arises when there is a prima facie case to be answered (it is otiose to call transparently morally unproblematic acts ‘justified’, even though they are justified). An action that causes harm or suffering may appear to be wrongful, but ascertainment of all the relevant facts, or deeper or wider reflection, can show it to be permissible or right, hence justified. Justification is complete exoneration of the actor: if one is excused from blame it remains the case that one has acted wrongly (if one hadn’t acted wrongly there would be no basis for excuse); whereas if one acted with justification one has not acted wrongly, so there is nothing to excuse.

Here are examples of the two concepts in use, for illustration:

The German Order Police carried out massacres of hundreds of thousands of defenceless Jewish civilians in Poland, 1942–1943. The actions of these men are morally appalling. But suppose (counterfactually) that they had complied with orders to kill out of fear for their own lives if they refused. ⁴ In that case, it would probably be appropriate to excuse them from blame or substantially reduce it. Now suppose that there was no credible threat to the policemen’s lives but there were powerful social-psychological pressures to comply with the orders (see Browning, 2001). In this case, it might be appropriate to excuse them from some blame (this will be discussed further below).

Allied aircrew massacred hundreds of thousands of German civilians in bombing raids in the later stages of World War II. Most people think that their actions were morally justified in virtue of the strategic aims of the bombing in pursuit of a just cause (see Schwartz & Comer, 2018). If so, the aircrew need no excuse for their actions because the actions were justified, that is, not morally wrong, and many would say they were morally right. But if one thinks that the bombing was morally wrong, then one must think the aircrew’s actions unjustified – though one might think that they are eligible for some degree of excuse, that is, exoneration from, or diminishment of, blame for the wrongful actions.

The other mitigatory concepts that feature in the concerns that social scientists express about the effects or implications of explanation, as quoted above, are condone, exonerate and forgive. Little need be said about condonation and exoneration because they are close cognates of excuse (though ‘exonerate’ can also be used to mean ‘found not to have done anything wrong’, which makes it a close cognate of ‘justified’; but it is clear that Miller above uses it in the sense of ‘excuse’).

Forgiveness, though, is unlikely to feature among the mitigatory consequences and implications of social scientific explanation. Only people who have been wronged have the standing to grant forgiveness to the wrongdoer. A murderer might be forgiven by their victim’s parent, but the parent has standing only to forgive the murderer for what he has done to them by killing their child, not to forgive him for what he has done to the child. One could say that murder, in distinction from the loss that it inflicts on the bereaved, is literally an unforgiveable crime because only the victim can forgive the perpetrator, and with their murder the victim has been rendered unable to forgive.

In the light of the foregoing conceptual clarification, we can see that although some of the scholars quoted above refer to justification and forgiveness, what they are all concerned with is actually excuse, not justification or forgiveness. There can be no question of justifying the kinds of actions they are talking about (even were the perpetrators to be granted some degree of excuse, the wrongness of their actions remains unequivocal and undiminished). Forgiveness has little to do with the possible effects or implications of social scientific explanation for two reasons. First, as just noted, it is only those that have been wronged by perpetrators that have the standing to issue forgiveness. Second, whereas explanation can yield reasons why perpetrators should be excused, there is not even a weak requirement to forgive; nobody is required to forgive wrongdoers, least of all by the deliverances of social scientific explanation. Conversely, if the forgiver so chooses they may forgive the perpetrator for actions that cannot be excused. Forgiveness is a gift which stands in no need of good reasons for its bestowal.

Social science explanation can have little if any relevance for justifying actions because whether an action is justified depends on whether it belongs to a kind of action that is justified. Justification is achieved by showing that the apparently wrongful act is not of the kind that it first seemed to be but actually of a kind consisting of actions that are permissible. So, for example, a woman stabs a man to death. Her action seems to be of the kind murder. But if we discover that she was being attacked by the man she killed we see now that her action is of the kind self-defence, which is morally permissible (justified). Justification, then, is a process of re-classifying actions that at first seemed to be of an impermissible kind to a permissible kind. In disputed cases, the dispute turns on which kind an action should be classified under. For example, does Allied bombing of German citizens belong to the kind war crime massacre or to just war self-defensive killing? Can social science explanation feed into this process of justification via re-classification or disputed classification? It could, were it able to reveal some pertinent features of an action-situation that are not apparent to ordinary observation, but it is hard to see how explanation as such might do this, and no clear examples come to mind. Take the case of Allied bombing again: which kind these actions should be classified under seems not to be resoluble by the discovery of some previously unknown facts, or anything else that social science explanation might yield (it’s a moral, not an explanatory question).

Justificatory re-classification also occurs not just with individuals’ actions but with the re-classification of whole kinds of action, where the kind itself changes from being impermissible to permissible (and sometimes vice versa). For example, consensual sexual acts between individuals of the same sex used to be classified as ‘unnatural’, ‘perverted’ and criminal but have now been re-classified as morally and legally unproblematic ordinary sexual activity. Differential employment and service to Black and White people was re-classified from being legitimate discretionary acts to unlawful and unjust discrimination. Social science theory and explanation might have played some part in these processes of re-classification, but the major roles are played by moral and political protest, debate and deliberation, and legal reform.

There is another domain in which social science theory and explanation is much more actively involved: the justification of institutions. A prime historical example of this is the Marxist historical-materialist theory according to which Ancient slavery in Athens and Rome was not, and could not have been, unjust, because it was intrinsic to the ‘mode of production’ that enabled those societies to function. ⁵ A more contemporary example is the classic ‘functionalist’ theory of economic inequality advanced by Kingsley Davis and Wilbert Moore (1945). According to this theory, the ‘universality’ (as Davis and Moore claimed it to be) of economic inequality is explained by its necessity: without its central incentivising role no developed society could function sustainably.

Necessity is widely taken to justify a social practice or institution. Most people think that money is a necessary institution – without it there would be no civilised society. Likewise meat consumption – many people think (erroneously according to contemporary scientific and medical consensus) it necessary for health and nutrition. If a practice or institution really is necessary – necessary for a tolerably decent way of life – then it is thereby justified. Davis (1953) sought to deny a critic’s charge that he and Moore were ‘justifying’ economic inequality, by pointing out that they were only explaining its universal existence (by showing it to be necessary), not justifying it (i.e. claiming it to be justified). But he was being naive or disingenuous; to claim that an explanation shows an institution to be necessary ipso facto is to claim it to be justified.

With these caveats on the relevance of social science explanation for justification in place, the reminder of this essay concentrates on the consequences or implications of social scientific explanation for excuse. Invariably, where there is some concern over ‘letting perpetrators off the hook’, what is at issue is excuse, even where the concern is expressed using the words ‘justification’ or ‘justify’.

The relation between explanation and excuse

Primo Levi and Bruno Bettelheim (quoted above) depict the relation between explanation and excuse as a constitutive one: to explain and understand is to excuse (forgive or justify, as they put it). If the relation between explanation and excuse is a constitutive one, then every (adequate?) explanation would thereby, ipso facto, excuse the perpetrators. Further, it would not be possible, conceptually, to deny that perpetrators whose behaviour has been explained and understood have simultaneously been excused by that explanation.

But it is quite clear upon just a little reflection that the relation between explanation and excuse is not a constitutive one (whereas there is a constitutive relation between explanation and justification where an explanation discloses the necessity of the institution or practice explained). Why would anyone think otherwise? One possible reason for Levi and Bettelheim conceiving the relation as a constitutive one is that it makes their stance seem incontrovertible. If to explain the actions of Nazi perpetrators just is thereby to excuse them, then the reasonable thing to do is to eschew explanation, with no need for further debate.

Another possible reason is that if social scientific explanation shows that a group of perpetrators literally could not have acted other than they did, then they would thereby be excused. It might be that many people, including some social scientists, do think that social science explanation can, or purports to, show just this. But it is a misconception. Only metaphysical analysis of the nature and conditions of human agency and causation can claim to show (inconclusively, by the very nature of metaphysics) that people could not have acted other than they did (see Pleasants (2019) on ‘determinism’). Moreover, such analysis can only claim to show that no one can ever act other than as they did; it cannot claim to show merely that some particular group of perpetrators could not have acted other than they did. What social science explanation can show is that a group of perpetrators acted under substantial social pressure, force and influence, making it difficult, to some greater or lesser extent, for them to have acted otherwise. The relation between excusability and difficulty will be explored below.

Likewise, counter-asserting, as do Browning, Zimbardo and Fiske (quoted above), that explaining perpetrators’ actions doesn’t thereby excuse the perpetrators also seems to remove the need for any further consideration. The implicit message is: ‘there’s no need to worry – explanation doesn’t excuse the perpetrators; I (the social scientist) no more think these people deserve to be excused than you (the reader) do’. It might also be the case that social scientists who deny that there is a constitutive relation between explanation and excuse do so on the basis that social scientific explanation does not show that perpetrators could not have acted otherwise, therefore explanation does not excuse. Fiske (2013, p. 605) is probably right to say that ‘the public assumes that the understanding excuses the doing’. Thus, those social scientists who deny a constitutive relation could simply be denying what they take to be commonly assumed. But as I will go on to argue, to observe that explanation does not constitute excuse is not to show that explanation has no effect on, or implications for, excuse.

By issuing forthright denial of there being a constitutive relation between explanation and excuse, it seems quite easy to dismiss any worry on the part of the audience that there is a strong or substantial relation between explanation and excuse. But if I were worried about the possibility of excuse issuing from explanations of perpetrator behaviour I would retort: ‘it’s not the possibility of a constitutive relation that I’m worried about’! The denial of a constitutive relation between explanation and excuse does nothing to address the possibility of there being other kinds of relation, kinds which, moreover, are not so easily dismissed. The most likely and plausible other kinds of relation are what I call ‘normative’ and ‘causal’.

Let’s consider first what a normative relation between explanation and excuse amounts to. Conceived normatively, the question is: Do social science explanations of perpetrator behaviour provide evidence or reasons in the light of which one should excuse the perpetrators? I will defer answering this question fully to later sections of this essay, but for now will just assert that how one responds to the question in particular cases requires moral reflection, reasoning and judgement. How or whether one should excuse a group of perpetrators on the basis of social scientific explanation of their actions is not itself a social scientific question, it’s a moral one.

The normative question, then, is whether or how people should excuse perpetrators on the basis of reasons furnished by social science explanation. But the question is: Does explanation make people judge the culpability of perpetrators more leniently than they would in the absence of explanatory mediation? Explanation could have this effect regardless of whether one (and those subject to the effect) thinks it should or shouldn’t. Unlike the normative relation, the causal relation is amenable to social scientific investigation. And there has been some empirical inquiry, yielding some interesting findings. The leading study, conducted by social psychologists Miller et al. (1999), found the following:

Experimental subjects who were tasked with producing an explanation of the behaviour of perpetrators of harmful acts before judging their culpability judged the perpetrators more leniently than subjects who gave their judgement of culpability before generating an explanation of the behaviour.

Subjects who read a ‘situational’ (social) explanation of some wrongful behaviour judged ‘the explanation’ and ‘the researcher’ (i.e. the author of the explanation, the psychologist) to be more ‘exonerating’ of the perpetrator than subjects who read a ‘dispositional’ explanation (one based on perpetrators’ personal characteristics) of the same behaviour.

Subjects who judged the exponent of a ‘situational’ explanation to have an exonerating/excusing stance towards perpetrators nevertheless were not themselves moved to adopt a more exonerating stance towards the perpetrators.

In sum, while Miller et al. found evidence that the work of producing explanations had a mitigatory effect on subjects’ judgement of culpability, merely entertaining social scientists’ situational explanations had no mitigatory impact on their judgement. Subjects did, according to Miller et al. (1999, p. 265), see what they describe as ‘the exonerating implications of social-psychological explanations’, ⁶ even though they rejected those implications themselves. There is some suggestion that Miller et al. (1999, p. 266) regard the ‘clear reluctance on the part of our participants…to endorse the exonerating implications of social-psychological explanations’ as perversely recalcitrant.

There are some puzzling phenomena which this study throws up but which receive little explicit attention from the authors. First, why is it that when subjects produce their own explanation they adopt a more excusing stance, whereas when they just read a psychologist’s (situational) explanation they don’t? The likely answer is implicit in Miller et al.’s (1999, pp. 255–260) theoretical discussion of the different cognitive processes involved in producing an explanation (it takes the deployment of considerable cognitive resources) and making a judgement of culpability (a more or less instant reaction that deploys little cognitive effort). That is, what induces people to be more excusing is the activity of engaging in explanatory work. But when they just passively ‘consume’ the kind of explanation that Miller et al. (and many other psychologists) believe to have intrinsic ‘exonerating implications’, there is no mitigatory effect. It would be good to know whether subjects could be induced by situational explanations to adopt a more excusing stance in an experiment designed to get them to engage explicitly and actively, over an extended period of time, with such explanations. Bearing in mind that coming to understand and appreciate an explanatory theory usually requires sizeable amounts of study time, experiments in which subjects have only a few minutes to grasp and assimilate a theory are ill-equipped to assess its potential influence on their moral judgement.

Second, Miller et al. (2002, p. 321) assert that ‘social-psychological explanations…will be resisted by those construing them as absolving perpetrators’ but give very little detail on the evidential basis for this assertion. The suggestion is that the credence that people accord to different types of explanation (social-situational versus individualistic-intentional) is a consequence of their prior moral attitude towards excusability. Thus, Miller (2016, p. 207), reporting on another study (by Newman & Bakina, 2009), says that subjects who were opposed to the exoneration of perpetrators ‘were most favourable to the interactionist and least to the situational explanation’ (interactionist explanations are those that combine situational and ‘dispositional’ (agential) factors). I don’t doubt that in life outside the psychologist’s laboratory it often happens this way too, but I suspect it is far from always so.

I have been delivering an undergraduate course on ‘the Holocaust, genocide and society’ for many years. It attracts quite large numbers of students. The course includes examination of a range of different explanatory theories, including situationism and its critics, and concludes with explicit reflection on moral responsibility and possible grounds for diminished culpability for various categories of perpetrator. I would say that roughly half of the students find social/situational explanations the most plausible, and roughly half reject them in favour of individualist/’agential’ explanations and interpretations. Those that endorse social/situational explanations do indeed exhibit a relatively excusing attitude towards perpetrators, encapsulated by the reflective thought that ‘there but for the grace of odds go I’ (a poignant phrase that I lift from Alford (1997, p. 733)). But I would say, impressionistically, that some of these adopt a more excusing attitude as a consequence of learning about social/situational explanations, not vice versa as Miller seems to insist upon. Indeed, it is just this effect of explanation on moral judgement that worries those social scientists who believe that their explanations do not, and should not be taken to, provide grounds for exoneration.

In sum, the empirical data that we have on the ‘excusing’ effects of social scientific explanations of perpetrators’ behaviour are, I think, suggestive but limited. Note, because what we are considering here is a causal relation between explanation and judgement of culpability, the subject who undergoes it does not – cannot – know that it has occurred. Or rather, they can only know a posteriori. As soon as one becomes aware of an explanation inducing one to judge more excusingly the relation ceases to be causal and becomes normative (unless explanation can make one knowingly judge more excusingly against one’s will).

If there is a significant causal relation between explanation and judgement of culpability it poses a difficulty for social scientists such as those quoted above who vehemently deny that their explanations excuse. Although they don’t actually state that they personally deny any kind or degree of excuse to those whose actions they seek to explain, it is safe to infer that this is their stance. The problem is that while they themselves resolutely refuse to excuse, and they recommend by unmistakable implicature that the reader should not excuse either, nevertheless at least some readers will be induced to take a more excusing stance through engagement with the explanations proffered. Social scientists simply dismissing the very possibility that their explanations can or should induce readers to adopt a more excusing stance is not helpful, either for themselves or for their readers.

What the foregoing analysis has shown is that the assertion that explanation constitutively excuses, and the counter-assertion that it has no effect on, or implications for, excuse are both untenable. We have also seen that in spite of social scientists’ denunciations of the very idea that their explanations have any effect on, or implications for, the excusability of perpetrators, the explanations might exercise such effects anyway. I turn now to a detailed analysis of the ways in which explanations might, normatively, give us reasons for judging perpetrators excusingly to some degree.

What can social scientific explanation bring to judgement of perpetrator responsibility?

Excuse may be granted to agents whose actions were affected by two different types of excusing condition, namely, volitional and cognitive conditions. Excuse, under the volitional condition, can be granted where an agent’s ability to have acted differently to their wrongful act was substantially constrained, hampered or impaired by adverse or unpropitious circumstances. In short, whether, and to what degree of severity, blame (and punishment) is deserved, and appropriate, for acting wrongfully depends on how easy or difficult it was for the agent to have avoided acting thus. This assumes that they acted as they did partly because of coercive or difficult circumstances Take, for example, the compliant Order Policemen referred to above, as studied by Browning. Browning presents evidence that most of these men complied with their orders to kill with reluctance and regret, knowing that doing so was morally wrong (though believing it not to be legally wrong). He argues that social and situational pressures made it very hard for them to choose to avoid participation in killing operations, even though they knew that they could do so with official impunity. As the quote above exhibits, Browning himself insists that this explanation of the men’s behaviour gives absolutely no reason for them to be excused. But I think that one could (I’m not saying one should) disagree with him on this. I will discuss this further below.

Excuse under the cognitive condition may apply when agents are found not to have been aware that in acting as they did they acted wrongly. This comes about when the agents in question have false beliefs on the permissibility of their action. Daniel Goldhagen (1997) studied exactly the same group of perpetrators – the German Order Police – as Browning. But he maintains that the Policemen willingly and enthusiastically (not reluctantly and regretfully as Browning claims) participated in killing operations, believing – falsely of course – that it was morally necessary, right and justified to do so. Along with Browning, Goldhagen seems vehemently to take the view that although the men’s actions, according to his explanation, were based on false beliefs that they blamelessly held, this gives no reason to grant any kind of excuse. ⁷ Many of his critics, though, insist that were his explanation correct the men would thereby be excused (for they could hardly have believed otherwise, according to his explanation) (see Pleasants, 2018, p. 25 and 27, 28).

These two conditions, the volitional and the cognitive, mark out the boundaries of agent responsibility and culpability. At the limit, if an agent could not reasonably be expected to have acted otherwise (due to overwhelming forces, pressures and constraints, or forbiddingly harsh circumstances), then no reasonable person will blame them for their wrongful act. Likewise, if an agent didn’t know, and could not reasonably be expected to have known, that acting as they did was wrong, then no reasonable person will blame them for doing so. At these outer limits, we hardly need social scientific explanation or interpretation to reveal excusing conditions. Once we know, for example, that military personnel who killed civilians under threat of themselves being killed for refusal to do so, there’s nothing more we need to know or understand in order to judge their moral responsibility to be at least significantly diminished. Likewise, when we discover that A didn’t know that someone else had put poison in the sugar that he spooned into B’s coffee, no further explanation or interpretation is needed to decide that A should be excused for (inadvertently) poisoning B. It is with regard to cases that fall some way short of exhibiting such obvious excusing conditions that social science explanation and interpretation is pertinent to judging excusability.

Considering that the so-called ‘problem’ of structure and agency is often reckoned to dramatise the core feature of human existence with which the social sciences must grapple, the relevance of social scientific explanation to the volitional condition of excusability should be apparent. On this view, the central task of the social sciences is to discover and reveal the nature and effects of social-structural forces and conditions on people’s capacity to manifest their personal agency. The reason that the relation between social-structural causation and individual agency is depicted as a problem is that, unless cognitively impaired, individuals always retain their agency (their capacity to have acted other than they did). But yet they also always have to act in conditions in which their ability to do as they would prefer is shaped, influenced and impacted by social-structural and situational constraints, forces, pressures, inducements and affordances (see Pleasants, 2019). So any good social scientific explanation will contain information that is pertinent to assessing the opportunities and options available to agents and the kinds and degrees of difficulty they may have encountered in the course of acting as they did.

Social science explanation and interpretation can also deliver information relevant to assessing the cognitive condition for excuse, namely, what it was that actors knew or believed. Broadly speaking, the sociology and psychology of knowledge is well placed to assess not just what particular groups did in fact believe, but what, if anything, they could and should have done to check the validity of their beliefs, seek and assess alternative sources of evidence, engage in particular lines of inquiry and so on. In short, what people believe is very largely a function of the communities of belief accessible to them, and what they’re socially expected to believe. Thomas Kuhn (1996, p. 210) famously insisted that ‘scientific knowledge, like language, is intrinsically the common property of a group’, and this applies equally to moral knowledge.

Verstehen: Judging oneself in the other ⁸

The central question that emerges from the foregoing analysis is: What stance, in general, should one adopt on the excusability of perpetrators in the light of social scientific explanation of their actions? The basic point that needs to be made is that one should see that there is indeed a normative relation between explanation and excuse. There is, then, a moral obligation to recognise that one’s active moral reasoning and judgement is required to work out, in particular cases, whether perpetrators deserve some degree of excuse subsequent to what explanation of the circumstances of their actions reveals about them as agents.

To get a sense of the space in which moral reasoning and judgement is required, and when it isn’t, consider the following examples. Let’s return to the case of the German Order Police under the counterfactual supposition that the men complied with orders to kill because they feared their own lives were under threat if they didn’t comply. Many people, I think, would say that under these conditions, the men simply must be excused from blame. If pushed on why, the response is likely to be ‘because they couldn’t have done any differently’, or ‘anyone would do the same in that situation’. Neither claim is strictly true: the men could have refused and taken the likely consequence; some people, albeit no doubt very few, wouldn’t kill even under these conditions. Nevertheless, I agree that it would indeed be entirely reasonable to excuse under such conditions. However, it would also not be unreasonable, though it would be harsh, to refuse excuse. For one might take the view that even in such dire circumstances, the men are culpable because morality requires that one does what is right even when it is very hard or costly to do so. A relative of the policemen’s victims might well indignantly maintain that it was cowardly of the men to choose to kill such weak, vulnerable, innocent people (most of whom were babies, children, women, the sick and the old) in order to save themselves. I am not arguing that the Order Policemen should be blamed even if they killed only because they feared being killed themselves; just that the difficulty and costliness of acting in such circumstances does not automatically dictate how they must be judged – moral judgement is still required.

Consider now the case of the Order Policemen as it actually was, where there was in fact no threat to the policemen’s lives. According to Browning’s explanation, although the men knew that they would not face any formal punishment for not carrying out their orders (indeed, some were given the option of not doing so by their commanding officer (Browning, 2001, p. 2)), there were powerful social-psychological and situational pressures to comply. Many people would probably think, as Browning himself does, that social-psychological and situational pressure is not a coercive enough force to warrant excusing the men from blame for killing weak and defenceless civilians. ⁹ But one could disagree with him on this, and judge that his refusal to consider any degree of excuse underplays just how powerful and forceful are the situationally generated pressures and influences to which he adverts. Browning himself presents reasons that might be taken to support such an excusing stance. Reflecting on the circumstances of the Order Police as revealed by his explanation, he tellingly concedes ‘I must recognize that in the same situation, I could have been either a killer or an evader’ (2001, p. 188). Browning also thinks that what goes for him goes for everyone else too: ‘If the men of Reserve Police Battalion 101 could become killers under such circumstances, what group of men cannot?’ (2001, p. 189). Doesn’t the fact that Browning thinks that the social-psychological and situational pressures that bore upon the policemen were of such a kind that he cannot say with confidence that he himself would have been able to resist them give reason to think that maybe the men should qualify for some degree of excuse?

My aim has been neither to argue that the threat to a perpetrator’s life does not count as legitimate grounds for excuse, nor that forceful social-psychological pressures do. What I have sought to do is to show how even in such seemingly clear-cut cases, the facts of perpetrators’ circumstances as revealed by explanatory theory do not determine how they should be judged with regard to excusability. These kinds of case can be contrasted with ones in which there is no need, and no room, for moral judgement, such as the above case of inadvertent poisoning. Moral judgement is not called for here because no reasonable person would think of attaching any degree of blame under such circumstances. What I have been trying to show is that explanation does not dictate how excusability is to be judged, but it can yield vital information to inform the making of that moral judgement.

All of the scholars quoted above (with the exception of Miller et al.) take for granted that if explanation and understanding lead to excuse for perpetrators that would be a bad thing. They evidently think that wrongdoers do not deserve to be excused. When it comes to moral judgement, social scientists are ordinary folk (as are philosophers), and many, perhaps most, ordinary folk do not like to consider the possibility that perpetrators of serious wrongdoing deserve some kind or degree of excuse. Why might one think otherwise?

In the course of our everyday lives, we are thoroughly versed in making moral assessments and judgements of people in our familiar milieus, and in deploying excuse, mitigation, forgiveness, justification and so on (as well as condemnation, criticism, censure, etc.). But we lack knowledge and experience of people different from us and conditions of action different from our own. Social science explanation and understanding can give us insight into the difficulties, challenges and constraints faced by other people in circumstances different to our own.

When we learn from social science about the wide and long prevalence of serious wrongdoing in different places and times, this learning might occasion recognition of the strong likelihood that we too are implicated in various kinds of socially structured wrongdoing. Some of these we may be quite unaware of, such as participation in an unjustly exploitative international economic order, and some, such as life-threatening poverty in faraway places and the effects of drastic climate change on future people, we may be aware of but think that we personally are powerless to redress. We have first-hand experience of, and insight into, the difficulties, forces, pressures, constraints and epistemic limitations that make it hard for us to avoid (unwitting) participation in socially structured wrongdoing. Our knowledge and understanding of the constraints, limitations, difficulties and challenges faced by others outside of our social milieus, in sharp contrast, is second-hand, mediated by social scientific theory, explanation and understanding. ¹⁰

Let’s revisit the circumstances of the German Order Police, as Browning presents them. If we had only his explanatory theory to go by, we might well be incredulous at the idea that that such seemingly mundane phenomena as social-psychological forces, pressures and influences were instrumental in people committing mass murder against weak and defenceless civilians. Although Browning candidly admits that he doesn’t know if he would have been a ‘killer or an evader’, I strongly suspect that most of us would be quite sure that we wouldn’t have killed in the policemen’s situation (cf. beliefs on how people would behave in ‘obedience’ experiments, note 10).

Is this difference in how we think we would behave due to Browning fearing he might lack moral fortitude, or is he just more honest or less self-deceived than us? I think it’s mainly for another reason. What distinguishes Browning from his readers, I think, is his profound adoption of Verstehen vis-à-vis the policemen and their circumstances of action. Verstehen, I suggest, can bring to life the mundane phenomena of social-psychological forces, pressures and influences that explanatory theory discloses. Browning achieves this by striving to imagine himself into the phenomenology of the policemen’s lifeworld, which he reconstructed from their testimonies. This process enabled him to attain a deep sense of what it would actually be like to live, think, perceive, experience, judge and act under the thrall of these social-psychological forces, in those circumstances.

Verstehen in the social sciences is usually seen as a methodological device for attaining richly authentic understanding of a social group in its own terms, from their point of view. Browning attained just this insightful understanding and communicates it graphically and grippingly to the reader. This understanding is the fruit of Browning’s adoption of a Verstehende stance in relation to the policemen, their actions and the circumstances in which they acted. But there is something more to Verstehen, as Browning enacts it, which cannot be conveyed propositionally to the reader. It consists in him taking to heart the task of seeking vicariously to put himself into the policemen’s situation, in order to see and experience what they saw and did – to the extent that anyone can. This psychological manoeuvre goes beyond conceptual, interpretive, propositional thought; borrowing a pithy phrase from Wittgenstein (1980, p. 17) (with Gramscian echoes), it has ‘to do with the will, rather than with the intellect’. This aspect of Verstehen is not methodological, theoretical, conceptual, interpretive or explanatory; it is deeply psychological, and I think it is the basis for Browning’s poignant observation that he doesn’t know if he would have been a ‘killer or an evader’.

I would not say that Verstehen in the above sense is inaccessible to the reader, who, after a careful reading of Browning’s book, will know essentially what Browning knows about the situation of the policemen. But few readers will have lived vicariously in the policemen’s situation to the extent and depth that Browning did on account of his extended immersion in its details in the course of researching and writing his book.

The lesson that I want to draw from this examination of Browning’s reflections is that Verstehen is not just a form of understanding social action, it is also an attitude that one can take towards the deliverances of social scientific explanation and understanding. When we contemplate the findings disclosed by social science on the lives and actions of wrongdoers outside of our familiar milieus, we should strive to adopt a Verstehende attitude towards these perpetrators, wherein we fervently try to imagine ourselves into their circumstances. If we succeed in this, the judgemental outcome needn’t be that we then have to excuse the perpetrators. As with explanation, Verstehen won’t tell us how to judge the culpability of others. But it does, I think, show that in order to judge others fairly and justly, we should seek ‘reflective equilibrium’ whereby we try to judge others as we would want ourselves to be judged by others (for our possible, albeit unwitting, participation in socially structured wrongdoing). This might issue in judging others less severely than we are initially inclined to do, allowing them some degree of excuse. Or conversely, it might issue in judging oneself and one’s fellow citizens more severely than we are inclined to do, acknowledging our over-eagerness to reach for too-easy excuses.

Browning evidently does not excuse the Order Policemen, despite having achieved a deep Verstehende appreciation of their action situation. Therefore, his confession that he doesn’t know if he would have been a ‘killer or an evader’, and his contention that no one else knows what they would have done either, implies that he should think that himself, and all of us, deserve harsher judgement than we’d like to admit. ¹¹ But how would that harsher judgement be manifested, given that we haven’t done what the policemen did, but may well do similar were we to find ourselves in a comparable situation? The question raises deeper philosophical questions than I can enter into here. ¹² Suffice to say, we should be deeply disturbed by our knowledge of what they did and Browning’s explanation of how they came to do it. We should be deeply disturbed not only out of moral shock and disgust at what they did but also because of what their actions might reveal about us.

Conclusion

I hope to have shown that, and how, social science explanation can provide crucial input into assessing perpetrator responsibility. But that assessment does not follow automatically from the substance of the explanation (though it might be causally impacted by it), no matter how powerful and insightful the explanation might be. Assessment of perpetrator responsibility requires moral reflection and judgement in the light of what explanation reveals about their circumstances of action. How that is to be done depends on factors that lie outside of explanation. How one judges responsibility is a function of what kind of moral judge one is, what moral standards one invokes, and how one applies them. It depends also on how demanding one takes moral obligation to be. Most pertinently for the social sciences, I argued that how one applies these standards should be shaped by how one compares one’s own circumstances of action with what social science explanation reveals about those of others. I sought to illustrate how a non-propositional form of Verstehende reflection may be of pivotal service in this process. If this suggestion has merit, a welcome by-product of the moral exercise might be that it reveals something of importance about Verstehen itself, namely, that to do it well is as much an effort of will as it is of intellect.