Blame and the Objective Attitude (OA)
So far I have been speaking mainly about responsible agency, which I have described as a cluster of capacities constrained or partially constituted by rationality, which in turn is facilitated by conditions of social justice, cooperative interdependence, and diversity. Responsibility is thus a relational property, which is enabled and enhanced by these group dynamics, and impaired by factors that hinder them.
I have said less about praise and blame (as opposed to responsibility qua capacity). I want to clarify here that if someone exhibits rationality deficits (which are responsibility deficits on my view), it does not automatically follow that the person is not blameworthy for his overt behaviour. If this followed, then people would be blameworthy for very little, since many (if not all) moral infractions stem from rationality deficits. However, I reject this deflationary conclusion. The reason is that responsibility deficits in a person’s moral cognition merely dispose the person to problematic behaviour, they do not necessitate that behaviour. Implicit gender bias, for example, is a responsibility deficit on my view, since it hinders rational thinking and inclines one toward overt sexist behaviour, but it does not necessitate this behaviour. The overt manifestation of the implicit state is blameable, even if the possession of the state is not. This is true, on my view, even if the action was not recognized by the speaker as morally problematic. On some versions of character theory (e.g. Angela Smith’s), the person who acts on an implicit state is blameable insofar as the action expressed part of his evaluative agency; and on the most natural interpretation of the control view, the person is blameable in case he failed to take reasonable steps to remediate or suppress the expression of his implicit bias (see Jules Holroyd 2012). Both views allow us to blame people for expressions of implicit states.
But what if the person can’t control the expression of an implicit state because he doesn’t recognize that it’s wrong? This is where the views seem to come apart. Character theory (as described above) grants that a person can be blamed for uncontrollable behaviour, while control theory requires that the agent have control over his action. Fischer and Ravizza, for example, say that responsibility depends on “conditions at the time of action,” together with considerations about the agent’s personal history (2000). I’m inclined to think that both models get something wrong. Re. character theory, I’m not sure why we would want to blame people who are incapable of responding appropriately (i.e., with recognition, not just in a mechanical way) under any conceivable circumstance. We don’t blame cats and dogs (I presume), and non-responsive human beings are seemingly morally equivalent to these species, in which case blaming them is a category mistake. Re. control theory, I’m not sure why we would blame people only in response to their capacities ‘at the time of action’ and prior to that time. It’s common for philosophers to say things like, ‘S is blameable for p at t if S had control over p at t—which suggests that S isn’t blameable for p at t+n, regardless of how S’s cognitive architecture evolved in the interim. But given that our motivational structures are relatively domain specific (e.g., Thach et al. 1992), situation sensitive (e.g., Doris 2002), and subject to ego depletion (e.g., Hagger et al. 2010), our ability to effectively exercise control over our behaviour may differ from one context to another, and from one moment to the next. If so, then whether someone is blameable the control-theory sense may also be contextual.
To bring this into relief, suppose that Smith made an insulting remark to me yesterday in a fit of uncontrollable rage, after being trounced in a tennis match by his colleague and storming off the court. While he managed to repress his urge to batter his colleague with his racket, he couldn’t help making an insulting remark to me later. The urge was utterly uncontrollable (imagine) given his motivational hierarchy and the circumstances. He was utterly unrepentant at the time (also an uncontrollable state). Does this mean that today, when he’s in a calm and collected state, I can’t demand an apology from him? It seems as if I’m entitled to hold Smith accountable whenever he has the motivational capacity to respond appropriately—whether today or tomorrow or next week—irrespective of his motivational psychology at the time of action. This isn’t an endorsement of character theory, but it’s something of a departure from control theory as it’s typically construed, since it denies that blame rests on considerations about the time of action and prior times only.
This example is meant to suggest that our reactive attitudes should do something—they should serve a function. (See Vargas’ ‘moral influence account’ for a similar view ). But these attitudes can serve a function whether they track a person’s motives at the time of action, or her motives at the present time. It’s feasible to think that we sometimes blame people to alert them to the fact that they have a domain-specific rationality deficit that is sensitive to triggering contexts, and they should either avoid those contexts in future, or develop new cognitive strategies to deal with them. In this case, we’re holding the person responsible as a means of encouraging better behaviour in the future, by bringing to light intelligible moral considerations. Our blaming attitude serves a purpose if it is intelligible to the other person.
How can we know when someone is blameable? We can’t. We can suspect, infer, or hypothesize that someone is blameable, but we can’t know that someone is, for various reasons that I’ve discussed here and in previous posts. These include: (1) we don’t have direct access to another person’s motivational states, so every appraisal of a person’s quality of will is necessarily an inference; (2) these inferences are affected by epistemic injustice in a way that undermines the warrant of our moral claims; and (3) given that motivational states are context-sensitive, we can only infer that our blaming attitude will be intelligible in any given context. That said, we can, and often do, make fairly accurate predictions about people’s behaviour across a fairly broad range of contexts, especially if we know the person well. What we need to be wary of is blaming people hastily or over-confidently, though we should not suspend blame when there is sufficient evidence. We ought to blame people if we have good reason to think that they are responsive, could be responsive under different circumstances, or could use alternate cognitive strategies to respond better in future.
If someone is incapable of responding to blame, this is when we ought to take the objective attitude (OA) on Strawson’s view (see the last post). When we take this attitude, we don’t engage with the other person as a moral agent, but instead take steps or enact and enforce policies to control and/or alienate the person. Some people might be globally non-responsive, i.e., morally incapable across every conceivable circumstance: these people are psychopaths. They completely lack emotional empathy, which enables us to care about other people (see Hare 1999, 2002). Others have merely local responsibility deficits.
To illustrate how my account pans out, let’s do a case study. This case study will also hopefully shed light on how we should think about the current political situation from a moral perspective.
Donald Trump: A Case Study
How should we morally appraise Donald Trump, the current President of the United States? First, Trump could (in theory) be a good example of how rationality deficits do not automatically translate into non-responsibility. Trump has many rationality deficits, but this does not necessarily entail that he is not blameable for his overt behaviour. On the other hand, Trump has not shown signs of being capable of responding appropriately to blame, and he seems incapable of apologizing or showing remorse.
Let's home in on a specific example of overt behaviour as our point of analysis. In October of 2016, the media released a video of Trump bragging about sexually assaulting women (in words not worth repeating). The public was justifiably outraged. Trump responded by releasing a so-called apology, which wasn’t an apology so much as a deflection of blame. In his ‘apology’ video, he denies that his remarks reflect his real self, insinuates that he shouldn’t have to apologize in any case because the public’s response is an overreaction that only serves to deflect attention from the ‘real’ issues, and tries to downplay his remarks by describing them as ‘locker room talk,’ insinuating that there is a context in which it is appropriate to demean women.
In other words, instead of taking responsibility, he tried to turn the tables by accusing his critics of being overly sensitive and obscurantist. Now, even if he had offered a satisfactory verbal apology—which he certainly didn’t—it wouldn’t have passed as a proper instance of ‘taking responsibility’ unless he had also taken steps to atone for the damage done to the epistemic and political climate by his statements. He would have had to, e.g., volunteer at women’s safe houses, donate to women’s causes, and so on. But instead of doing any of these things, he made matters worse by offering a sad pretence of an apology, and then, as if to add insult to injury, proceeded to introduce executive orders that harm women, such as the order barring the U.S. from giving foreign aid to NGOs that provide abortion services or even discuss the possibility of abortion with clients.
This shows that Trump has heaping tons of rationality deficits (e.g., implicit and explicit gender bias), and he also hasn’t shown any sign of moral recognition, or of being capable of responding appropriately to blame. Instead of taking responsibility, he seems intent on blaming his critics and escalating his attacks against women's rights and interests.
In light of the evidence, I think that there are two viable ways of construing the situation. (1) We can see Trump as fairly morally impaired, but potentially capable of responding to blame under a limited range of circumstances, in which case blame is appropriate in limited contexts, but OA is the appropriate default response; or (2) we can see Trump as globally morally impaired and susceptible to OA across all contexts. To recapitulate from the last post, OA entails using policy, legislation, and exclusion to control and regulate a person’s behaviour, rather than trying to engage with the person as a competent moral interlocutor. OA is justified only toward people with severe cognitive deficits—though we tend to deploy it in perverse ways in light of pervasive stereotypes, as I pointed out earlier. However, if a person is genuinely morally impaired—either in local but irremediable ways, or globally—the situation warrants OA.
Brian Leiter recently appraised Trump as either a sociopath or a narcissist, but favoured the narcissist diagnosis because Trump doesn’t try to fly under the radar—he "craves attention and adulation" (2016). Both personality disorders cause substantial moral deficits, and both are relatively resistant to treatment, because they cause antisocial behaviour and do not typically cause psychological distress to the agent. While narcissists can respond to prudential considerations, their ability to respond to moral claims is very limited, and sociopaths similarly lack normal moral responsiveness. In both cases, the person might be a better target for OA than useless attempts at blame and moral reasoning, which (likely) will not be intelligible to him.
I’m personally more inclined to think that Trump is a psychopath, which overlaps with sociopathy and narcissism but entails a complete lack of emotional empathy. Psychopaths, however, have cognitive empathy—the ability to read others’ emotional states, and this can make them more cunning predators. Trump appears to be a living incarnation of Robert Hare’s profile of a psychopath from his book, ‘Snakes in Suits’ (2002), where he says that psychopaths comprise a disproportionate number of CEOs as well as incarcerated violent offenders—though psychopaths prefer to operate on "the shady side of the law" whenever possible (Hare 1995: 25). Although psychopaths comprise a small percentage of the human population (maybe as low as 1%), they cause a disproportionate amount of harm, in part because they are protected and even rewarded in unjust political systems that prize confidence and glib charm over moral character, and systematically fail to protect and defend the psychopath’s primary targets: vulnerable groups. This environment exacerbates the psychopath’s non-responsiveness and anti-sociality by rewarding and reinforcing it (especially, perhaps, in locker rooms?). Not all psychopaths are actively anti-social (e.g., James Fallon, the American neuroscientists, is a benign psychopath by his own reckoning); but those who do exhibit anti-social traits appear to be totally untreatable. For example, violent psychopaths who are placed in forensic treatment programs tend to come out either no different or more violent than before (Rice, Harris, & Cormier 1992). Psychopaths often use what they learn in treatment to harm people more effectively; and they tend to disrupt treatment sessions, spoiling them for more treatable patients (Hare 2002). For this reason, Hare believes that criminal psychopaths are the one group of prison inmates that should not be offered treatment, and should be incarcerated for longer than other groups.
If Trump is a psychopath, then he’s not blameable for any of his behaviours, including failing to remediate his psychopathic traits, given that they are situation-invariant. Hare says that the only reasonable way to deal with a malignant psychopath is to avoid the person at all costs—which is essentially an injunction to take the OA, rather than trying to engage with the person as if he were a moral agent. This is likely what P.F. Strawson would advise as well, and what he had in mind when describing OA. Instead of trying to reason with a dangerous psychopath, the reasonable course of action is to protect yourself, alert others, rally confederates against the person, and contain the person’s anti-social behaviour. These are reactions external to the reactive attitudes, which also function to protect the moral community. Unfortunately, whether Trump is a psychopath or merely a malignant narcissist as Leiter suggests, we may be forced to interact with him (indirectly at least) for four more years—in which case we can only hope that the formal mechanisms of American democracy can contain his recalcitrant behaviour.