1 Introduction

Our sensitivity to moral reasons and the likelihood that we act morally is strongly influenced by our social environment. This is a key insight of ecological accounts of moral agency, which present a compelling picture of moral psychology and provide an attractive rationale for holding people responsible. On these kinds of accounts, the social environment influences moral reasons-responsiveness in numerous ways. Our social environment determines what moral considerations are salient and what moral reasons we are sensitive to (Vargas 2013, 2021; Hurley 2011). According to an influential recent account by Victoria McGeer (McGeer and Pettit 2015; McGeer 2019), our reactive attitudes and practices of holding each other to account are crucial for people’s responsible agency: internalising the reactions of a moral audience makes us more responsive to moral reasons and thus more likely to act morally. Reasons-responsiveness, on this account, relies on feedback from moral audiences and is scaffolded by the social environment. We can appropriately be held responsible if we are moral agents that possess this audience dependent reasons-responsiveness.

In this paper, we discuss McGeer’s (2015, 2019) scaffolded reasons-responsiveness account in the light of two concerns and suggest how it should be revised to address them. The first concern is that some agents may be less attuned to feedback from their social environment but are nevertheless morally responsible agents – for example, autistic people. At first pass, agents who struggle to pick up on certain typical forms of feedback would seem to pose problems for any account that posits that audience sensitivity, especially the ability to recognise and internalise a moral audience’s likely reactions, is essential to moral action. There is good evidence that despite difficulties understanding social feedback, as a group, autistic individuals are no more likely to behave immorally or criminally than neurotypicals (Maras, Mulcahy, and Crane 2015, Dempsey et al. 2020). This indicates that autistics recognize and respond to moral rules, and thus we need to question whether moral feedback is in fact as crucial as McGeer’s account postulates.

The other problem with ecological accounts we address is a practical one. In the real world, moral audiences can work to undermine reasons-responsiveness if one’s social environment condones immoral actions. In this case, attention to one’s audience may actually support immoral action. An ecological account of responsibility needs to explain how agents can guard against the corrupting effect of misguided moral audiences. We need an explanation of how people can resist certain audiences and choose between different moral audiences.

We propose to resolve these problems as follows: We argue that autistic people, too, rely on audience feedback in developing moral reasons-responsiveness, but this feedback often needs to be more explicit. We claim audience feedback plays both an informational and a motivational role in responsible agency and that the difficulties agents with atypical sensitivity to feedback face tend to result from the former. These difficulties can be countered by audience feedback taking a different shape depending on the traits of the person it is addressed to. Children rely on explicit moral feedback in the form of praise and blame in their moral development. (Some) autistic persons may continue to be particularly dependent on explicit audience feedback in novel contexts or concerning behaviour which might unintentionally be hurtful or frightening to others. However, there is also a lot to be learned from ways in which autistics are less dependent on moral validation by their environment regarding specific moral judgments and actions, even though they generally care about being able to morally justify their actions to others. Application of moral rules and a certain rigidity can help autistics avoid being corrupted by their social environment’s questionable application of moral rules.

These lessons then inform our discussion of the problem of corrupting moral audiences. Audience sensitivity is not all that matters to reasons-responsiveness; it needs to be tempered by a consistent application of moral rules. McGeer’s scaffolded reasons-responsiveness account needs to make explicit allowance for that. Furthermore, given neurotypicals’ strong dependence on moral feedback, we need to make sure that we seek out new moral audiences and moral interchange, so we do not become complacent and entrench possible moral error. In particular, we need to pay attention to the reactions of those affected by our actions and consider whether our actions are justifiable to them.

Here is how we will proceed: in Sect. 2 we will lay the conceptual groundwork, present ecological accounts of moral agency and responsibility, and explain why we focus on McGeer’s account. In Sect. 3, we discuss the challenges posed to the account by autistic individuals and how these can be resolved. We address the second problem for the account, corrupting moral audiences, in Sect. 4 and explain how it can be resolved using insights from Sect. 3. Section 5 concludes.

2 Ecologically Supported Reasons-Responsiveness

What does it take to be a responsible agent, the kind of person who can justifiably be blamed or praised, punished or rewarded? One influential account of moral responsibility proposes that an agent is responsible if they are reasons-responsive (Fischer and Ravizza 1998). Reasons-responsiveness encompasses the ability to recognise (moral) reasons for action and respond to them at least some of the time (Fischer and Ravizza 1998). In other words, an agent needs to meet certain criteria of moral understanding and control to be morally responsible. Positing the need to recognise reasons for action presupposes that there is some standard for correctness when it comes to reasons for action. Responsibility theories are normally silent on what this standard is, leaving the details to normative ethics. We will do likewise and not espouse a theory of normative ethics; however, we will say more about the process of adjudicating between different moral perspectives when we discuss corrupting audiences. Further, on a reasons-responsiveness account, morally responsible agency requires the ability to evaluate actions morally and have some level of control over what actions one performs. Here, we will work with a folk psychological understanding of control and leave the metaphysics of free will to one side.

Ecological accounts of moral and responsible agencyFootnote 1 stress the extent to which an individual’s reasons-responsiveness depends on their social environment. Susan Hurley (2011) points out that the social environment can both undermine responsible agency, for example through manipulation, and support it by helping people counteract cognitive biases. Manuel Vargas notes that people’s sensitivity to moral considerations and their ability to act on considerations they are sensitive to depends on the circumstances they find themselves in. Under oppression, the oppressed may find it more difficult to do the right thing; and the oppressors may be unaware of the wrongness of some of the practices they are involved in. Both may therefore not be (fully) responsible, in one case because of lack of control, in the other because of ignorance. Vargas summarizes the dependence of responsible agency on one’s social environment as follows:

“One’s status as a responsible agent or not, and whether one is blameworthy or not, are not facts prior to, or independent of the normative concerns and particulars of circumstance that structure our agency. Social context, in the form of individual or structural oppression, can impair or even undermine responsible agency.” (Vargas 2018, p.127)

Kevin Timpe (2019) takes things one step further by saying that not only morally responsible agency, but agency more generally is dependent on one’s social environment.

One specific way the social environment affects responsible agency is through the practice of holding individuals responsible itself. On Vargas’s account, the effect of responsibility practices on agents is conceived quite generally: praise and blame and the moral evaluation of behaviour communicate the norms and expectations of a society and are part of what shapes people’s responsiveness to moral reasons (Vargas 2021). However, he does not draw a very tight link between being held responsible and being a responsible agent. While our practices of blame and praise are justified by their effects on people’s reasons-responsiveness generally, individual instances of praise and blame are justified by an assessment of an agent’s reasons-responsiveness, not by likely effects of holding responsible.Footnote 2

Victoria McGeer (2015; 2019; McGeer and Pettit 2015) draws the most explicit and direct link between being held responsible and being a responsible agent. On her view, being susceptible to moral influence is constitutive of reasons-responsive agency (McGeer 2015). McGeer and Pettit (2015) highlight the way reasons-responsiveness depends on one’s social environment: the reactions of a societal audience to one’s moral and immoral actions shape our sensitivity to moral reasons. People can be directly sensitive to reasons; but they can also be indirectly sensitive by reacting to a societal audience’s moral feedback. In this way sensitivity to our social audience reinforces or expands direct sensitivity to reasons. Chloe’s awareness of a possible audience and their reaction to her actions will make her think not just about what she takes to be the best course of action, but about how others would morally evaluate her action and whether they would condemn or praise her behaviour. Because people are motivated to fulfill the moral expectations of those who they accept as moral interlocutors, they are also motivated to attend to the reasons those interlocutors take to be relevant in a certain situation.

“[A]s a reasons-responsive agent you can be expected, not only to be moved by the reasons you confront in making a choice, but also by the audience, actual or prospective, that you confront (…) your responsiveness to reasons may be the product in part of quite a different sort of sensitivity as well. This we describe as your sensitivity to others, in the role of actual or prospective audience, rather than simply to reasons” (McGeer and Pettit 2015 p.170).

The proposed mechanism of ecologically supported reasons-responsiveness can be illustrated further by an example. Imagine a police officer, Officer Jane, is trained in proper protocol for use of force and thus is made abundantly aware of moral reasons related to appropriate use of force. While working as a beat officer, Jane becomes less sensitive to these reasons because she tends to let her emotions get in the way of a careful calibration of her use of force in the heat of an arrest. Jane could be reminded of the reasons related to use of the right level of force – those she might use to justify her actions to others – if she becomes more sensitive to her wider societal audience. This audience may become salient to Jane in various ways: she might attempt to imagine members of the community reacting while watching her make a particularly rough arrest; she might look around and remind herself of the actual physical crowd watching her make an arrest (and recording it on their phones); or, Jane may remember that she is wearing a body camera so that her actions are recorded for official review. We can imagine that if Jane is asked to wear a body camera for some months, she might gain a heightened sensitivity to reasons related to appropriate use of force due to heightened sensitivity to her audience, and that this sensitivity might then become more automatic – that is, she may no longer need the camera to be appropriately sensitive to reasons for appropriate use of force, because her first-order disposition has been recalibrated.

This account provides an attractive picture of moral psychology, where reasons-responsiveness is supported by being part of a community that will hold us responsible for our actions. More than that, it is also supposed to explain why we are justified in holding people to account in ways that are painful for them – in doing so we scaffold their moral reasons-responsiveness. With this picture in place, however, there are important clarifications to make. It’s tempting to think of the way moral behaviour is supported on this account in quite a cynical way. Specifically, we may think that the result of moral feedback is people behaving well in front of their peers solely in order to gain social approval (Haidt 2013). That the social environment is a powerful motivator for conformity to societal norms is an insight stretching from the thought experiment of the ring of Gyges in Plato’s Republic to current social psychology. However, on McGeer’s account we don’t act in line with the moral expectations of our audience merely for the sake of audience approval. We are not just concerned with curating the impression others have of us. Rather, our audience creates a heightened responsiveness to moral reasons themselves by making them more salient. While the examples focus on the reinforcement of norms, moral feedback also plays an important role in giving explicit feedback on the norms that people are expected to observe. This is particularly important in moral development or in novel contexts. Agents pay attention to, and are thus responsive to, moral reasons in part because they have a moral audience. By thinking about how our behaviour may be justifiable to people whose moral views we respect and care about, we become more aware of, and more motivated to act upon, the right moral reasons. McGeer (2019) stresses that morally responsible agency is therefore not a binary trait that people have or lack. It is instead something akin to a skill that is developed through the give and take of moral reasons and the practice of being held to account for our actions. Holding others morally responsible therefore has a proleptic element; we are holding people to account on the basis of moral reasons that we expect them to come to accept.Footnote 3

What different ecological accounts have in common is that they stress that we cannot be good moral agents on our own: we rely on social scaffolding. Our societal audience is important to learning about and understanding moral standards, refining this understanding, and adhering to these standards. This emphasis on the social determinants of responsible agency also provides an attractive rationale for why we are morally justified in blaming and punishing others. In doing so, we develop their reasons-responsiveness. McGeer’s account thus gives a descriptive account of how morally responsible agency is developed and sustained, and makes a normative claim about (morally) responsible agency. We are justified in holding others to account because this aids their reasons-responsiveness; conversely, people who are unable to respond to their moral audiences are not morally responsible agents (McGeer and Pettit 2015; McGeer 2015, 2019).

3 Scaffolded Reasons-Responsiveness and Persons with Atypical Sensitivity – Autism

The scaffolded reasons-responsiveness account assumes that being responsive to a moral audience is a core mechanism for becoming and remaining responsible. It highlights the way in which people’s reasons-responsiveness depends on real and imagined feedback from our social environment, and the way in which we feel obligated to justify our actions in response. But this raises the question of what to make of individuals whose responsiveness to their social audience is, as it were, atypical. This is the case in individuals with autism. Autism is characterized by difficulties with social interactions and with gauging the thoughts and emotions of others.

The diagnostic label ‘autism’ covers a growing and extremely varied group of individuals who can have very different cognitive-emotional profiles (Fletcher-Watson and Happe 2019). It therefore makes more sense to consider the effect of certain psychological features commonly found in autism on moral responsibility, rather than making blanket statements about the moral agency of autistic people. Autism is a neurodevelopmental condition characterized by (1) problems in social interaction and communication, and (2) rigid or restricted behavioural patterns (APA 2013). In discussions of autism and moral responsibility, the so-called ‘theory of mind’ or ‘empathy’ deficit and ‘executive function deficit’ theoriesFootnote 4 have been particularly influential (Richman 2018; Stout 2016). These map fairly neatly onto the two DSM criteria for an autism diagnosis. Whereas difficulties in mind-reading (accessing others’ mental states) seem to account for problems in social communication and interactions; differences in executive function might account for rigidity of thought and restricted, repetitive patterns of behaviour or activities.

However, recent research suggests that autistic persons’ problems with social interactions or communication are not best understood as an ‘empathy deficit’. It seems that the autistic persons’ difficulties with anticipating and reading the neurotypical majority’s emotional reactions have created the impression in some that they lack the ability to empathize. However, the term ‘empathy’ is too coarse grained to be of much use here, as it subsumes mind-reading abilities, the tendency to care about others’ wellbeing, and the disposition to act in ways that contribute to others’ wellbeing (Stenning 2020). Empathy has both cognitive and affective elements. Autistic persons do not seem to care any less about others’ wellbeing than neurotypicals and can offer an emotional response appropriate to another’s mental states (Rueda, Fernandez-Berrocal, and Baron-Cohen 2015). Nevertheless, problems in reading and anticipating one’s (normally neurotypical) social audiences’ reactions are common for autistic individuals. This means that autistic persons can struggle to notice and make use of social feedback.

We are now in a position to see why responsible agency might be thought to be compromised in autistic individuals on McGeer’s scaffolded reasons-responsiveness account: if we must be able to rely on our society’s moral feedback to be reasons-responsive, it would seem that individuals with autism are at a (moral) disadvantage. However, autistic individuals are generally moral and rule-abiding (Dempsey et al. 2020) and they are not over-represented in e.g. criminal statistics (King and Murphy 2014). Furthermore, autistic individuals do care about others’ wellbeing and are motivated to avoid causing harm. Autistic children tend to go through a process of moral development that appears to be the prerequisite for this kind of moral care for others. A recent meta-analysis found that children with ASD were capable of forming selective, secure attachments (Teague et al. 2017). Sue Fletcher-Watson and Francesca Happé note that “attachment, empathy and affection are not reduced in autism - although they may not easily recognise another’s thoughts, autistic people are no more likely to be callous than non-autistic people” (Fletcher-Watson and Happe 2019, p. 144). Autistic children often exhibit an ability to make moral judgments on par with other children (Leslie, Mallon, and DiCortia 2006).

Importantly, autistic individuals make roughly the same moral judgments as neurotypical persons concerning which actions are right and wrong, although there is a tendency to be more strongly focused on outcomes than intentions (Moran 2011). In one study, high-functioning autistic children actually judged harming others as morally worse than control children did (Li et al. 2014). The authors of this study speculated that the autistic children they tested might have more rigid criteria for what constitutes morally wrong action because they were more rule-oriented than controls. The autistic children in the study were also just as likely to cooperate with others as controls. After finding no significant differences between high-functioning children with autism and neurotypical children in moral judgment, Leslie et al. (2006) concluded that such judgment “may function and develop to an interesting degree independently of theory of mind” (Leslie, Mallon, and DiCortia 2006, p. 279).

Autistics’ successes as moral actors indicate that differences in mind-reading can be compensated for. Despite their problems reading other people’s mental states, autistic persons are adequately sensitive to moral reasons. Looking at the way this happens in the case of autism is instructive for understanding the mechanism of audience sensitivity more generally and refining our understanding of ecological reasons-responsiveness.Footnote 5 In an earlier paper, McGeer (2008) suggests that autistic individuals have a different type of moral agency, which is more strongly guided by what she calls a concern for cosmic order and by rules. But she also expresses the worry that ‘a good part of the behaviour we identify as manifesting moral sensibility among individuals with autism may stem from a need to abide by whatever rules they have been taught without sharing our understanding of the ends these rules are meant to serve’ (McGeer 2008, p. 240). To our knowledge, McGeer has not taken a stand on how her account of autistic moral agency and the scaffolded reasons-responsiveness account interact. But, as we explain below, we disagree with the more pessimistic comments on moral agency and autism quoted above. While we think adherence to rules plays an important role in autistic individuals’ moral agency, we don’t think this is blind adherence, but part and parcel of a cognitive style that is in some respects more rational and less prone to bias (Rozenkrantz, 2021). A closer look at the different mechanisms underlying responsiveness to moral reasons in neurotypicals can help to develop McGeer’s scaffolded reasons-responsiveness account and paint a more optimistic picture of the role of rules for responsible agency, both for autistics and neurotypicals.

We claim moral feedback plays two roles in an ecological account: a motivational and an informational one. The informational role of moral feedback tells us what moral reasons we should respond to. By seeing what others consider right and wrong we can better understand which actions are immoral and harmful versus moral and helpful. We gauge actual audience reactions and internalize the moral judgments of our societal audience. For example, we often imagine how people might react when considering an action, and this imagined reaction highlights moral reasons to perform or not perform the act. If a person considers taking their neighbour’s car without permission because theirs is broken and they have an important appointment, one might imagine their neighbour’s reaction. Imagining the neighbour’s moral anger and feelings of betrayal emphasizes the moral wrongness of the act.

The motivational role works through our desire to gain the approval of our moral audience and avoid disapproval. We don’t want our neighbour to be angry with us, and we don’t want their moral condemnation. However, this this motivation goes beyond merely wanting approval and being aversive to disapproval. We also want to be able to justify ourselves to others. Justification can importantly come apart from actual approval when we think about whether others should approve of what we are doing. When we justify our actions, we claim our spot in the moral community and as moral agents, even in the face of disagreement or disapproval. For example, I might share out departmental burdens in a way that I know some of my colleagues will be unhappy with. Nevertheless, I am motivated to do this in such a way that I feel answerable to them in terms of fairness and am able to morally justify what I am doing. Similarly, if I needed my neighbour’s car to rush my child to the hospital, I might feel that I can justify taking it and that my neighbour should understand.

We are now in a position to give an account of likely differences in audience sensitivity for autistic persons, and how this might affect their reasons-responsiveness.

3.1 Autism and the Informational Role of Audience Feedback

Let us look at the informational role of audience feedback first. Difficulties in mind reading and picking up on social cues or predicting one’s audience’s likely reaction mean that an important source of moral information is sometimes missing. An autistic person might also be very focused on a specific task, and, due to attentional differences, miss cues that neurotypicals might take to be reasons for concern and action. In this case, we can see how executive function problems might pair with difficulties in reading social cues to create insensitivity to audience. For example, an autistic person focused on organizing his study materials at the college library might miss the library announcement that it is closing in 10 min, and not notice the increasingly anxious looks from the librarian who needs to pick her kids up after closing. He may not understand that when the librarian asks, “Are you almost finished?” while pointing at the clock that this really means, “You need to finish now” – he is insensitive to her implicit feedback regarding his behaviour. And he may not be capable of throwing his things in a bag and leaving quickly when the librarian, already late to close the library and make pick-up, finally tells him explicitly that he must leave. This shows how in some cases a failure to pick up on implicit cues can impair sensitivity to audience feedbackFootnote 6.

Feedback will need to be more explicit for persons less attuned to social feedback in certain situations, especially feedback that has to do with how actions affect people’s feelings. Fortunately, moral feedback can be and often is very explicit. This is regularly the case with young children during the processes of moral development, so that moral norms can be internalised. Parents often say things like, “Jill is sad that you took her toy. It is wrong to take something that isn’t yours without asking. You need to give it back.”

Autistic individuals can and do make use of explicit feedback. When autistic actor Rick Glassman was interviewed by Trevor Noah on the Daily Show, Mr. Glassman said his diagnosis gave him permission to “teach people how to communicate with me. I would often say ‘Listen, if you think I’m weird, just tell me, I’m not going to pick up on it.’” Telling people that he needed explicit feedback if he was “being annoying” made people feel safe, says Glassman, so much that they “didn’t even seem to care” about their differences.Footnote 7 Autistic persons’ use of explicit feedback to apply a moral rule such as “don’t annoy others” does not mean that local problems won’t still occur; there may be cases where audience reactions are missed or read incorrectly, especially in novel situations. No one may be to blame for this; or, if someone is to blame, it may be the person who could have given explicit feedback (more so than the person who needed explicit feedback). Where this happens, it provides grounds for a local excuse, especially where the problem is new and unexpected. Exempting autistic people from responsibility altogether, rather than taking the condition to provide local excuses, amounts to a form of disrespect to them as they are not seen as fully capable moral interlocutors (Brandenburg 2022).

Further, it is important to note that problems in accessing other’s mental states are not just a problem that autistic people have. Neurotypicals who are tired, stressed, or even self-absorbed can fail to take note of the mental states of persons impacted by their behaviour (Richman 2018). Furthermore, just as autistic people can find a neurotypical difficult to read, neurotypicals can have problems empathizing with autistic individuals (this is called the “dual empathy” problem, cf. Milton 2012, Walker 2014, Chapman 2020). Where sets of persons have different cognitive styles, there’s a two-way obligation to compensate for lack of easy mind-reading. This problem may also occur when people are from different cultures. The German first encountering understated British reproach may not pick up on the message sent by her moral audience, because she is not attuned to the accepted way of telling people they have just made a gross misstep. What the example of autism teaches us is that audience responses need to be tailored to the recipient, and that often, and especially regarding behaviour that is harmful, audience feedback needs to be explicit. This is not just a psychological claim about how to best communicate expectations. It is also a normative one, because communicating one’s moral expectations is a way of taking someone seriously as a moral agent and a starting point for negotiating moral norms.

3.2 Autism and the Motivational Role of Audience Feedback

People, neurotypical and autistic, want to be recognised and respected as moral judges and decision makers. In working on this paper and others, we have repeatedly seen that autistic interlocutors understandably get quite annoyed by research that appears to question their reasons-responsiveness or their responsible agency. The need for moral recognition is strong. By age 10 most children have become aware that others evaluate them morally (Permer and Wimmer 1985). Children make use of this knowledge in order to present themselves favourably and attribute negative emotions to self as wrongdoer (Nunnar-Winkler 1998). Very young children seem to feel principally answerable to their primary caregivers, but as they get older the circle to whom they answer grows wider. One of the outcomes of the process of moral development is becoming indoctrinated into a societal-level moral community.

Autistic persons, too, are motivated to act morally and be taken seriously as moral interlocutors. However, there is some anecdotal evidence that approval of individual moral decisions or outlooks may be less relevant to them. Autistic persons can be keen to apply moral rules in a categorical way even where moral feedback indicates that an exception to a moral rule would be socially acceptable. For example, while a neurotypical might think it is alright to “splash out” and fly the family to a distant location for a vacation to celebrate a big birthday, an autistic person might not think a birthday trumps the moral judgment that such a trip is harmful to the environment. Greta Thunberg reports that she finds non-autistic people morally strange, especially in their inconsistency of professed attitudes and actual behaviour on climate change. So, it would seem that autistic persons may be less reliant on their moral audience reinforcing a moral rule, once it’s been adopted.

While autistic persons feel generally answerable to their moral audience and obligated to justify their actions (morally), they may be less inclined to respond to moral feedback within a specific context about a particular act. We speculate that this greater reliance on moral rules and one’s own moral judgments rather than on specific moral feedback may be partially the result of finding synchronous moral feedback harder to access. If I am less certain of how my moral audience is likely to react, I will, for better or worse, have to rely on my own moral judgment to a larger extent.

Plausibly, there are other factors at play as well. Recent findings have shown that autistic individuals are less prone to a number of cognitive biases such as framing effects, optimism bias etc. (Rozenkrantz, D’Mello, and Gabrieli 2021). This means that autistic’s moral judgments may be more rational and consistent with professed values than those of neurotypicals. In other words, autistic persons are less likely to embrace irrational exceptions to moral rules.

The desire to be able to justify one’s actions to a moral audience and the ability to gauge the likely reaction of said moral audience regarding a particular act can come apart. In this case, autistic individuals seem to err on the side of justifiability rather than actual audience approval. In some cases, this approach will produce a better outcome, morally speaking. Thus, we propose an addition to the scaffolded reasons-responsiveness view to the effect that adherence to moral rules is sometimes as important, or even more so, than audience sensitivity for moral reasons-responsiveness. However, it is worth noting that all individuals (autistic or not) are still reliant on moral audiences in the acquisition of moral rules. Thus, even when we rely on moral rules, these are initially derived from moral audiences. But this reliance on a moral audience for norm acquisition is compatible with internalised rules taking precedence to current audience feedback. For example, where a societal audience inconsistently applies moral rules in ways that serve individuals self-interest, autistic moral actors would seem to be better off than neurotypicals, who tend to be highly swayed by contextualized moral feedback (Jefferson and Sifferd 2022). This highlights our second worry for accounts which claim that moral audiences support reasons-responsiveness. What can such accounts say about cases where the moral audience serves to undermine agents’ ability to be reasons-responsive?

4 Corrupting Moral Audiences

McGeer and Pettit present us with an ideal scenario, in which the audience supports moral agency and moral progress by blaming and praising us when this is appropriate. Real audiences are different. They sometimes get it right, but sometimes they don’t. Often this will take the shape of inconsistent commitment and adherence to values which are endorsed within a society. Behaviours that violate these values may not discouraged - and may even be encouraged - by the moral audience. If our reasons-responsiveness depends on our moral audience, and that audience is too permissive or not sensitive enough to moral reasons, we can feel justified in doing what is morally wrong. Poor moral audiences can lead to moral error.

We are familiar with this problem from the context of climate change. Despite societal recognition of climate change as a serious moral problem, many of us carry on as before, taking several flights a year and eating significant amounts of meat because that is socially accepted behaviour. Another example of immoral behaviour that is accepted in certain circles of society is tax evasion. Here, audience feedback, or lack of negative audience feedback, can reinforce bad behaviour. People may justify cheating on their taxes by claiming that their peers are behaving in the same way: in their subset of society, tax evasion is “normal”.

There are also more extreme cases where audiences can contribute to immoral behaviour and warped values. For example, Jake Davison, part of the incel - involuntarily celibate - movement, killed five people (including his mother) before shooting himself. His misogynistic views concerning women and his ideas of what constituted justifiable behaviour were formed and reinforced by the online subculture of incels he was part of. While other factors will have played a role in this case, there is no doubt that his moral audience had a corrupting effect on his reasons-responsiveness.

How can the threat that corrupting audiences pose to our moral reasons-responsiveness be addressed? Jules Holroyd (2018) notes the problem that audiences with morally bad values pose for ecological accounts of responsibility such as McGeer and Pettit’s and makes the following suggestion: we should not only be sensitive to our actual moral audience but also to idealized audiences. If we expand our understanding to include audiences we can conceive of, and who “populate the moral community, but whom in fact one is unlikely to face in one’s actual experiences” then we can have greater confidence that our perceived audiences will enhance our sensitivity (Holroyd 2018, p. 144).

This would in principle be a normatively satisfying solution, but it isn’t one that captures the psychological mechanisms very well, and it undermines the appeal of the scaffolded reasons-responsiveness account. If persons are able to adjust their moral compass by imagining an idealized moral audience, it is not clear why they need a real one in the first place. Finding out how an idealized audience would react would seem to involve finding out what the right action is by thinking about underlying moral principles we endorse and how to apply them consistently, without bias or self-interest. But as stated, if we are already able to do this, then idealized audiences seem to constitute an unnecessary further step. For that matter, we might just imagine what a perfectly good deity might say in response to an action and scaffold our reasons-responsiveness that way. In this case there is no need for persons to consider a human moral audience at all.

While we don’t see the solution as idealized moral audiences, we do agree that the scaffolded reasons-responsiveness account needs to be developed to explain how the moral agent evaluates their moral audiences and the feedback they offer. To return to a point made earlier, one thing we do when we are deciding on the moral course of action is to think about what we can justify. It is possible to think about justification in terms of an idealized moral audience. But we suspect that what is really going on here is that we think about what we can justify to ourselves, given the moral values and norms we already endorse. The decision or action should then be justifiable to any other person holding the same values. The examples of autistic moral decision making and the potential harm done by moral audiences show that consistent application of moral principles or rules is a further essential ingredient in moral decision making.

Nevertheless, moral audiences are still crucial for the initial development and the maintenance of reasons-responsiveness. Agents learn morality from their social audience, and audience feedback can enhance reasons-responsiveness to a level an actor would not be able to achieve on their own by justifying their actions through appeal to their own values. The audience does not merely serve as something to which we must justify ourselves morally, thereby motivating right action – it also plays an informational role by telling us what sorts of behaviours are seen as unacceptable or harmful and why, especially in novel contexts. They give us insight into how our actions affect others.

Idealized audiences will not always take us as far as we need to go in our moral thinking, because of limitations to our moral imagination. Sometimes imagining what it is like to “walk in another’s shoes” - especially when we have good information about what is like to be that person - will provide important moral information. But it is often better to get such information from the person themselves, especially if their situation is so different from ours that our ability to put ourselves into their shoes is limited. It is here that the example of autism and the double empathy problem become instructive again. We need others’ explicit feedback to see how our behaviour looks from a different perspective and what its effects are. In cases when there are significant differences in the way we experience the world or our moral outlook, this can only be done by feedback directly from the people affected.

In more extreme cases, the very fact that we don’t take certain groups of people seriously as moral interlocutors is part and parcel of the problem. In situations of oppression, the people who have grounds to blame the oppressor do not have the power to do so as they are not consulted and their views are dismissed as uninformed and unimportant. (Think of the example of slavery.) They are therefore not part of the moral scaffolding, which means that an important perspective is missing. We need to seek out and listen to the people who are affected by our behaviour, in addition to trying to imagine ourselves into their position.Footnote 8 Real but different audiences will often provide a better moral corrective than the ones we can imagine based on our own conceptual resources.

There is a further question here about moral responsibility and culpable ignorance that we can only touch upon: if the moral audiences we are responsive to are isolated individual subgroups of society, we may worry that some persons will be morally ignorant due to poor audience feedback and therefore have a ready-made excuse for their behaviour. This raises the question of whether such persons are responsible for harmful acts when they didn’t know any better. As we have seen above, Vargas (2018) argues that responsibility may be undermined when oppressors are ignorant of wrongdoing. We cannot adequately address this worry here. However, we feel some of our responses to the concern of corrupting moral environments imply that many persons who lack moral knowledge due to their moral audience are obligated to seek out this knowledge by exposing themselves to other relevant moral audiences. In this case such a lack of moral knowledge may result from culpable negligence (Sifferd & Jefferson 2022).

5 Conclusions

In this paper we raised two challenges to McGeer’s scaffolded reasons-responsiveness account: agents who are less attuned to social feedback such as autistics, and corrupting moral audiences. We found that, once we parsed the two roles that feedback from a moral audience play, autistics provide reasons to revise the scaffolded reasons-responsiveness account. We argued that autistic persons, like neurotypicals, wish to justify their behaviour to a moral audience and rely on their moral audience for feedback. However, autistic persons may need more explicit feedback when it comes to effects their behaviour has on others. They also compensate for difficulties they have in receiving information from the moral audience by justifying action through appeal to moral rules. This shows that McGeer’s view of moral agency needs to include observance of moral rules as a way of reducing reliance on audience feedback. We suspect that McGeer would approve of this proposal, as she mentions that an instance of blame can lead to vocal protest by the target, and a possible renegotiation of norms and rules for what constitutes acceptable behaviour (2019). Consideration of corrupting audiences highlights a different problem from that of resisting blame and renegotiating norms. It draws attention to cases where individual agents must try to go beyond what is accepted in their moral environment, a significant challenge for social beings who rely strongly on moral audiences in developing and calibrating their moral reasons-responsiveness. Resistance to a moral audience requires the capacity to evaluate the action differently; often this will be with reference to a moral rule or principle.

For both neurotypical and autistic individuals, consistent application of moral rules or principles can reinforce and bring back to mind important moral commitments when we are led astray by our own desires or specific (im)moral audiences. But moral audiences still play a crucial role to developing and maintaining reasons-responsiveness. First, they are essential to the development and maintenance of all agents’ moral sensitivity. Second, they can provide an important moral corrective where people may have moral blindspots, especially when they provide insights into ways in which a person has fallen short morally by not taking on board reasons that are not obvious to them. Often, these can be reasons which pertain to the respectful treatment of others who are in some important way different from that person.

Where moral audiences corrupt reasons-responsiveness, it may be that the feedback we are receiving is from a smaller group, as was the case with incel Jake Davison. To ensure that one receives high-quality feedback it is important to seek out different audiences, including persons who might be affected by one’s actions, to get explicit feedback. Another way a moral audience can corrupt is by failing to provide accurate feedback on specific behaviours given the values the moral community embraces, as often happens with the moral problem of climate change. Here the example of autistic moral actors is also instructive. They show that in order not to be swayed by permissive audiences, we need to think about whether we are consistent in following the moral values and rules we endorse.