Much of the philosophical literature on responsibility revolves around conceptual issues: what is responsibility? Is it a capacity, a suite of capacities, a trait of persons, a privileged subset of a person’s mental states? I don’t hope to resolve these debates here—maybe they’re irresolvable, if there turns out to be insufficient evidence on either side. What I do hope to accomplish here is to shift the focus of the debate in a different direction— away from a focus on the internal properties of persons, which are supposed to be constitutive of responsible agency, toward a focus on the environmental, public, or ‘ecological’ factors that cause, support, and arguably co-constitutive those capacities—a view that I call ‘methodological externalism,’ as opposed to ‘methodological internalism.’ I use these terms to distinguish two separable, but not entirely distinct, approaches. One focuses more on the internal states of agents—things like the structure of an agent’s desires or the robustness of an agent’s values—and the other focuses more on the social conditions in which an agent comes to have and exercise those capacities or structures.
Before proceeding, I should say something about what I take responsibility to be. We can demarcate the literature into two camps: the character view, on which responsibility is a matter of the agent’s deep self or most robust character traits, and the control view, on which responsibility is a matter of an agent’s control mechanism or capacity to respond to reasons. This is a popular way of systematizing the literature (see, e.g., Levy 2014). Rather than decide between these views, I’m going to assume that they both get something right: responsibility has something to do with character and something to do with control. But I will also focus on what I take to be a common thread between these views: they both tie responsibility to rationality. As Susan Hurley argues (2013), both models of responsibility—perhaps all models—presuppose that responsible agency consists of or is constrained by rational agency. If an agent is acting either against her own instrumental reasons (reasons she takes herself to have), or against her own substantive reasons (reasons she objectively ought to care about), then she is, to that extent and in that capacity, responsibility-impaired. For example, if someone is shopping compulsively and spending inordinate amounts of money on commercial products, he is rationality impaired and, by the same token, responsibility impaired. Put differently, the compulsive shopper is acting ‘out of character’—his compulsion is not part of his deep self; and he lacks control over his compulsive urges. His compulsion is ‘shallow’ and relatively uncontrollable, and this is in part because it is a rational defect.
I should also clarify out the outset that responsibility, as I understand it, is not simply a capacity or a mental state—although it is typically described in singular terms, which can be misleading. (In particular, it can lead one to assume that responsibility is a zero-sum proposition, rather than a spectrum concept, consisting of multiple parallel, largely independent mechanisms, rather than a monolith). I take responsibility to be “a suite of capacities,” as Vargas puts it (2015: 201), or a cluster of entangled mental states, as Sher describes it (2009). This leaves open the possibility that a deficit in one responsibility-constitutive capacity/state could leave intact or even enhance other, parallel responsibility-constitutive capacities/states. To give a simple example, having a mental health disability could impair one’s ability to function in certain areas of one’s life, while enhancing one’s artistic capacities—as often happens. This seems to have been the case for Van Gogh, who was diagnosed with schizophrenia and was hospitalized several times (Schindeler 1935), and yet managed to produce esteemed expressionistic artworks until his death. It seems perfectly coherent to say that Van Gogh was praiseworthy (and thus responsibility-apt) for his artistic accomplishments, even if they were completed while he was in a psychotic state that impaired some of his basic functional abilities.
In a similar vein, a member of a disadvantaged social group could be impaired in her ability to secure a job due to epistemic insecurity or material obstacles, and yet be responsibility-enhanced in terms of her ability to understand and appreciate the social factors that undermine her own agential capacities. This is what is called ‘epistemic privilege’ by feminist social epistemologists, and it refers to special epistemic authority on certain matters, as conferred by one’s social position. Often, social oppression engenders epistemic privilege vis-à-vis the social structures and mechanisms that support and sustain a privileged class, while oppressing historically disadvantaged groups. This might be especially salient to the victims-survivors of social oppression. In this case, the individual might be impaired in certain facets of her responsible agency by her context, while responsibility-enhanced by that same context in terms of her ability to grasp, appreciate, and critique dominant social paradigms. In this way, responsibility—conceived of as a set of dissociable capacities—can be something of a double-edged sword.
One of the implications of seeing responsibility as tied to rationality, I will argue in what follows, is that responsibility turns out to be relational, just because rationality is relational. More specifically, our rational interests are interpersonal interests—they are served by collaboration in pursuit of our joint projects. To give an example, if I’m interested in being a responsible philosopher, I will read as much relevant literature as possible. This includes literature by underrepresented groups. If I ignore this literature, I’m acting in such a way as to undermine my own rational self-interests, particularly my interest in producing objective research. At the same time, I’m undermining the career prospects and visibility of members of underrepresented groups, i.e., I am undermining their rational interests. Hence, when I undercut my own rational interests, I undercut those of other people by the same stroke. By acting irrationally, I am undermining my deep-seated values, and failing to respond to objective reasons that I ought to care about—that is, I am responsibility-impaired on both standard views. This is because, when I undermine my own rational interests, I thereby undermine my prospects for responsibility. And I undermine the possibility of ideal rationality, and responsibility, in the entire group.
Now, this example is about ‘philosophical agency’—arguably a kind of epistemic agency; but epistemic agency is also, I think, moral, in that it has robust moral implications. If you manifest implicit biases (by acting in a discriminatory way), you are also, by the same token, acting in a morally objectionable way. Epistemic flaws often have moral consequences, and are in this sense morally significant. Thus, while it is possible to conceptually distinguish epistemic reasoning and moral reasoning, they are, in fact, interrelated: moral reasoning is undermined by epistemic deficits such as implicit bias, and are served by (what Miranda Fricker calls) ‘epistemic virtue’—sensitivity to a person’s character. Unfairly discrediting someone on the basis of identity prejudice is an epistemic error, but it is also a moral infraction.
I should also clarify that moral responsibility, as I understand it, is related to our various social roles. Being a responsible philosopher, teacher, friend, co-worker, citizen, parent, teacher, etc., is part of what it means to be a morally responsible person. Impairments in any of these ‘role responsibilities’ thus constitute impairments in moral responsibility, or some facet(s) of morally-responsible agency. It’s hard to even fathom what it would mean to be a responsible moral agent independent of these roles, since being a moral agent just means being an agent embedded in a social network with various social roles—roles that generate responsibilities. Impairments in morally-relevant ‘role responsibilities’—those that have significant consequences for others—thus equate to impairments in moral responsibility.
Now, I’ve said that responsibility is tied to rationality, and rationality is relational. This is, in part, because rationality is a cooperative enterprise. Or, as John Doris puts it, optimal human reasoning is facilitated by cooperation (2015). This is perhaps most obvious when we think of global projects like preventing catastrophic climate change. No one person can make a substantive difference to the state of the global climate—which prompts Walter Sinnott-Armstrong to provocatively declare that ‘no one’ is responsible for climate change (2010). But if no one is responsible for climate change, that can only be because everyone is; and it is only by cooperating (e.g., introducing laws and regulations and social practises) that we can make an appreciable difference. If there is a cataclysmic climate disaster (as there is forecast to be if global temperatures increase by 2 degrees C compared to pre-industrial levels—1.2 degrees C more than the current temperature), we will all be impaired in our responsible agency, since we will either be killed, or geographically displaced, or otherwise encumbered by burdens that we didn’t previously face, which potentially undermine our ability to perform our familiar role responsibilities—as citizens, parents, workers, etc. The environmental case brings into relief how our rational interests are intertwined: we need to cooperate on large-scale, international environmental projects to ensure the continued survival of the species. If we fail in this particular joint project, we will all be responsibility-impaired at best, and dead at worst. So, this joint rational project is, at the same time, a responsibility-preserving project.
This is a big-picture example, but small-picture scenarios also illustrate the relational nature of responsibility. For example, without healthy interpersonal relationships, we cannot develop the capacity to function in any role responsibility whatsoever—we become quite anti-social. This is demonstrated, for example, by Harry Harlow’s inhumane experiments on monkeys (hopefully never to be repeated): Harlow showed that when monkeys are isolated from all social contact, they become anti-social and psychotic (Blum 2002). When re-introduced to other monkeys after a year, the social isolates were patently disturbed, incapable of playing, and two of them starved themselves to death. Similar psychological disturbances can be observed in prison inmates placed in solitary confinement—which ironically makes them more antisocial and more likely to recidivate (Grassian 1983). Attachment theory builds on this work, showing that children whose primary care-givers are unresponsive and insensitive tend to develop ‘insecure attachment’ (anxious, avoidant, or disorganized), which impairs their ability to have healthy adult relationships (Bowlby 1982, Ainsworth et al. 1978). Individuals with anti-social personalities and severe social deficits will not be particularly effective at taking on and adequately fulfilling their role responsibilities—in other words, they will be very substantially responsibility-impaired. Their social deficits may incite them to harm members of the general community, either be being freeloaders or positively harming others. A salient example of an antisocial agent is the psychopath, who suffers from deficits in cognitive empathy, making him extremely dangerous. Canadian psychologist Robert Hare says that psychopaths comprise only 1% of the human population but commit half of all violent crimes (2007: 18). These crimes can drive innocent victims into lives of self-harm, criminality, and mental illness (e.g. PTSD). That is, psychopaths’ harmful actions can deprive others of responsible agency. While some psychopaths are (according to Hare) untreatable, others (‘secondary psychopaths’) may respond to treatment; and many violent offenders are non-psychopathic, and more responsive to treatment. We have an interest as a society in rehabilitating offenders’ responsibility status (if possible), not only for their own sake, but to protect the responsibility-constitutive capacities of others. And we have a responsibility to incarcerate violent offenders who do not respond to treatment—those constitutionally incapable of responsibility—to protect the responsible agency—and other dimensions of agency and wellbeing—of others.
This explains how social deficits impair rationality. On the other side, socializing enhances rationality—and hence responsibility. For instance, collaborating with others on reasoning tasks tend to improve problem-solving outcomes (e.g., Laughlin 1965; Davis 1969; Maier 1970; Hill 1982; Shwartz 1995—cited in Doris 2015). Collaboration, in other words, enables us to be optimally rational. In turn, collaborating helps us to be optimally responsible, since it helps us to reason effectively and objectively (in a non-biased way) about problems that threaten our responsible agency. For example, if I am trying to decide how to address a student with putative behaviour problems in such a way that I am able to serve the rest of the class and the students’ interests at the same time—which is not an easy matter—it might serve me to consult with the student, the university’s counselling and disability services centre, and other relevant people, before coming to a decision about how to address the issue. Collaborating with the right people can help me come to a responsible decision in the matter—a decision that enables me to fulfil my role-responsibility as a teacher, in a way that also enhances the student’s responsible agency and preserves the baseline rate of responsible agency in the class.
Collaboration is thus good for responsibility. But not just any kind of collaboration is helpful, and this is important to note. First of all, we should collaborate with the right people—people who are implicated in the problem or have relevant expertise. But aside from this matter, we also want the collaborators to have the right demographic status and the right mindset—a point that should not be overlooked. Just saying that ‘collaboration enhances rationality’ is a misnomer, since some groups are incapable of cooperating and respecting each other’s contributions. We now know, on the basis of multiple research project on workgroup problem-solving, that demographic diversity within the workgroup’s composition improves the group’s performance on cognitive tasks. Early research, however, found exactly the opposite: that diversity decreased reasoning efficacy. This is because the early research failed to control for intervening process variables, such as the attitudes or ‘mindsets’ held by the group members. More modern studies, supported by comprehensive literature reviews, find that diverse group composition predictably improves reasoning outcomes, but only if certain group dynamics are in place—factors such as low ‘subcategorization’ (e.g., implicit bias), high cooperative interdependence, and high ‘diversity mindset’ (i.e., an appreciation of the value of diversity) (Kippenberg & Schippers 2007). This research lends support to the claim that collaborating can serve our responsible agency, as per the teaching example above, but we should be mindful to collaborate with the right people under the right epistemic conditions—conditions of receptivity to everyone’s epistemic authority.
To conclude this section: I submit that our rational goals are social goals, and thus relational goals. And responsibility is a function of effectively pursuing these rational-social goals—hence responsibility is relational. We cannot be ideally responsible, across the full panoply of our role responsibilities, if we shun others, discriminate against others, find ourselves (due to bad luck or by choice) in conditions of social isolation, deprivation, and social injustice, or if we destroy the planet or decimate the earth’s resources. While cataclysmic climate change will undermine our responsible agency in deep, potentially irreparable ways, more intimate harms, like discrimination, can also hamper our prospects for responsible agency.
One of the main implications of seeing responsibility as relational is that we all have an interest in distributing resources and care in such a way as to enhance the responsibility of every member of society, perhaps especially the least well-off, because responsibility-deficits do not remain isolated and cloistered: they spread around. There are no ‘gated communities’ for responsible agency: we’re all in the same boat.
One potential objection to this view is that a total recluse could be responsible, in spite of being completely socially isolated. I think that this is impossible: there is no responsible social isolate. This is because the social isolate is not capable of taking on social roles constitutive of responsibility; he is capable, at best, of a very limited range of capacities: perhaps obtaining food and shelter and feeding himself, but not much else. This person cannot obtain responsibility as I envision it. He is far from being able to acquire what I have been calling ‘ideal responsibility,’ viz., a range of context-sensitive social capacities that respond reliably to environmental cues. Moreover, I suspect that the complete social isolate is a fiction—though one that we are familiar with from the myth of the ‘state of nature’ and Romulus and Remus raised-by-wolves myths. In reality, humans evolved to live in societies. Hence, every real human being depends on community, and the products of joint efforts, for survival. Underestimating this fact can be deadly. The naïve and unfortunate John McCandless trekked into Alaska with nothing but a bag of rice, and summarily died of food poisoning (as described in ‘Into the Wild’ by Krakauer: 1993). Every successful ‘survivalist’ depends on human agricultural and manufactured products to survive in isolation; thus, even if they remove themselves from human society, they depend on the proceeds of this society for survival in the wild. But these wilful social isolates are not, on my view, as robustly responsible as members of society.
Now, one might alternatively object that philosophers already (tacitly?) take responsibility to be relational, so I’m not making a substantive point. But if they do, then why do so few responsibility theorists talk about relationships and social structures, and their impact on responsibility—aside from the many ‘nefarious neuroscientist’ examples?
In what follows, I intend to shed a spotlight on the role of the social in responsible agency. I will focus in particular on epistemic justice (and lack thereof).