I. Introduction
Some of our most important theories of free will and moral responsibility highlight the significance of narrative. For example, John Martin Fischer’s highly influential semicompatibilist account emphasizes that the value of acting freely (in the sense required for responsibility) is a kind of self-expression. This self-expression is characterized as “depend[ing] for its meaning on a narrative structure.”Footnote 1 In acting freely and responsibly, the agent expresses “the meaning of a sentence of the book of his life” and this meaning “is fixed in part by relationships to other sentences in this book, that is, by the overall narrative structure of the life.”Footnote 2 Robert Kane, who develops one of the most important libertarian (that is, incompatibilist) positions, highlights narrative in terms of authorship. In discussing self-forming choices, he analogizes to the writing of a novel:
Imagine a writer in the middle of a novel. The novel’s heroine faces a crisis and the writer has not yet developed her character in sufficient detail to say exactly how she will react. The author must make a “judgment” (arbitrium) about how she will react that is not determined by the heroine’s already formed past, which does not give unique direction. In this sense, the author’s judgment of how she will act is “arbitrary,” but not entirely so. It has input from the heroine’s fictional past and, in turn, gives input to her projected future.
In a similar manner, agents who exercise free will are both authors of, and characters in, their own stories at once.Footnote 3
The fact that influential views on both sides of the compatibilism/incompatibilism divide emphasize and eloquently spell out this feature of our agency demonstrates its significance and its resonance with our experience. We experience ourselves as authors and care about being the tellers of our stories. An appeal to narrative is used, in both accounts, to illustrate both what we value and what we are doing in paradigm cases of free and responsible behavior; we are expressing ourselves or we are developing our character. We are writing our books. It is a powerful appeal.
One of the reasons the above appeal is so powerful and insightful is that narrative is embedded deeply in human cognition, emotion, and behavior. Thus, philosophers and psychologists have rightly discussed this capacity in numerous contexts. But could it be that narrative plays a further role in moral responsibility? Maybe narrative is not just what we value and do as morally responsible agents, but is, in a sense, how we are able to do it. Thus, my main aim in this essay is to argue that what I call “narrative capacity” is a genuine feature of our mental lives that enables us to become fully responsible agents.
I approach the issue from the standpoint of reasons-responsiveness. Reasons-responsiveness theories, which have been developed in a number of nuanced and insightful ways by different theorists, purport to explicate the “control” condition of moral responsibility.Footnote 4 They center on the idea that moral responsibility requires sufficient sensitivity to reasons. My claim is that our capacity to understand and tell stories has an important role to play in this sensitivity. But although I suspect that many actions for which we are responsible require at least a rudimentary narrative capacity (and I will offer some considerations for thinking so in what follows), my aim here is not to argue that narrative capacity is required for every such instance. Instead, I argue that narrative capacity plays a vital role in our becoming full-fledged responsible agents. Without such skill we would be cut off from the full range of reasons to which moral agents need access or we would have diminished ability to properly weigh the reasons that we recognize.Footnote 5 After arguing for the relevance of narrative skill, I argue that understanding the connection between reasons-sensitivity and narrative confers additional benefits. It illuminates important psychological structures and helps to explain some cases of diminished blame.
II. Narrative Capacity
Psychologists and philosophers discuss the uniquely human cognitive capacity to understand and tell stories. This capacity is marked by a number of features that make it difficult to reduce to other abilities.Footnote 6 Primarily, narrative is a way of making sense of events.Footnote 7 But while many kinds of reasoning might be regarded as “sense-making,” there are features of narrative that appear to set it apart from other cognitive processes. Psychologists characterize narrative reasoning as both goal-structured and focused on human or at least human-like behavior.Footnote 8 The typical problem-resolution structure that we are apt to think of when thinking about “story” relies on our ability to understand the motives, intentions, and actions of the characters. According to psychologists, children recognize typical action sequences in infancy.Footnote 9 But it is not until age five or so that children do more than follow common scripts and actually “structure events into stories . . . and [construe] events in terms of an initiating problem and its resolution.”Footnote 10 Children continue to develop beyond merely understanding the “landscape of action” to understanding the inner workings of characters, that is, the “landscape of consciousness,” as they grow into adulthood.Footnote 11
But narrative capacity involves more than the ability to recognize or identify the internal workings of characters. It also involves the ability to identify with them. Psychologists discuss narrative reasoning as involving “judgments of empathy.”Footnote 12 And there is typically an emotional element—we care about what happens to the characters and about how the story turns out. In general, emotion is an important feature of narrative. Stories resonate emotionally. David Velleman uses an example from Aristotle’s Poetics to illustrate the emotional import of narrative. Aristotle tells the story of Mitys. There are two relevant events to the story. First, Mitys is murdered. Then, the murderer himself is killed by a falling statue—of Mitys! Velleman suggests that what connects these two events into a story is not any sort of cause-effect sequence, but a familiar emotional cadence. It resonates with the audience emotionally and therefore “makes sense.”
The story of Mitys also points to another important feature of narrative. And this is the sense in which narrative involves a “meaning-affecting” relation.Footnote 13 The meaning of the second event (the death of the murderer) depends on its relation to the former event (the murder of Mitys). The connectedness thought to exist between events in a story is often cashed out in terms of temporal and causal relations. But the story of Mitys demonstrates that meaning-affecting is not always strictly or directly causal. It can be emotional or thematic.Footnote 14 The central idea is that the meaning of events is conditioned by their context within the narrative.Footnote 15
There is also a social or cultural aspect to narrative. There are several dimensions. In developing narrative skills, we come to understand the behavioral and emotional narrative patterns of our culture. Ultimately, by the time our narrative skills are fully developed, we will structure the narrative of our own lives according to familiar cultural signposts. But long before that happens, narrative skills are learned as caregivers “scaffold” them by co-constructing them with their infants and children.Footnote 16 For example, as developmental psychologist Robin Fivush explains, a child of sixteen to eighteen months might spontaneously provide a word pertaining to a recalled event (“berries”), to which the mother replies, “’That’s right, we had strawberries for breakfast, didn’t we? They were delicious.’”Footnote 17 Daniel Hutto argues that through this scaffolding process, narrative is how we come to understand folk psychology:
By participating in this kind of narrative practice children become familiar with the way the core propositional attitudes, minimally belief and desire, behave with respect to each other and their familiar partners: emotions, perceptions, etc. More than this, in such stories a person’s reasons are shown in situ; against appropriate backdrops and settings. For example, children learn how a person’s reasons can be influenced by such things as their character, history, current circumstances and larger projects.Footnote 18
We also apply narrative to our own lives by mentally traveling through time. Mental time travel is our ability to remember ourselves in the past and project ourselves into the future. It involves “autonoetic awareness” and a directing of attention forward or back.Footnote 19 Although not always explicated in terms of narrative, I claim that mental time travel has a narrative structure. It requires placing ourselves in the context of a story rather than merely having a “snapshot” memory with no features of connectedness or context.Footnote 20 It need not extend over vast expanses of time or provide accurate or detailed temporal ordering. Young children, for example, can mentally time travel but tend not to be able to sequence past events properly for quite some time (everything is “yesterday”).Footnote 21 To this extent then, it may sometimes only involve a limited or rudimentary capacity for narrative.
Beyond the ability to mentally time travel is the ability to see one’s whole life as a story.Footnote 22 Psychologist Dan McAdams is one of the pioneers of the “life story model of identity.”Footnote 23 According to McAdams, this capacity does not develop until at least adolescence. It is probably spurred at this time by the adolescent’s tendency to begin to look inward and think about who he or she is. Other psychologists have since tested his claim and have found that children asked to tell their life stories really do not appear to be able to do so in a fully coherent way before adolescence, and the coherence increases at least until age twenty. A study by psychologists Tilmann Habermas and Cybele de Silveira found that different kinds of coherence increase the most between various age groups. Between the ages of eight and twelve years old they found the biggest increase in temporal coherence. For causal coherence it was between twelve and sixteen years old, and for thematic coherence, between sixteen and twenty years old).Footnote 24
Given the above considerations, I think we should regard narrative capacity as a know-how, an interpersonal and practical skill that progresses throughout typical human development from rudimentary ability to full-fledged proficiency. I am inspired here by the work of Victoria McGeer, who discusses know-how in the context of folk psychology. She argues that our ability to understand folk psychology is a “psycho-practical know-how” or “insider expertise.”Footnote 25 As she points out,
Theoretical knowledge simply involves coming to know about how other things are or what other things do. Practical know-how, by contrast, consists in the development of a skill—the ‘internalization’ of methods for doing something oneself which are normatively guided by considerations of what constitutes doing it well.Footnote 26
At one point, she compares psycho-practical know-how to the “insider expertise” of linguistic fluency, which is a capacity with two different aspects: speaking and listening.Footnote 27 My suggestion is that we should understand narrative capacity along similar lines.Footnote 28 Through cognitive development, scaffolding, and social interactions, we internalize the methods of “reading” and “writing.” We learn to “read” the world and ourselves, while simultaneously learning to “write” our stories.
III. Narrative Capacity and Reasons-Responsiveness
There are a number of important and well-developed reasons-responsive accounts on offer. I will discuss one of the most influential—that of John Martin Fischer and Mark Ravizza. The basic underlying idea of a reasons-responsive account is that in order to control our behavior in the sense required for moral responsibility, we need to be sufficiently sensitive to reasons. If my behavior, for instance, is impervious to the sway of reasons then it seems that I am not appropriately subject to attributions of moral responsibility. Reasons-responsiveness is often understood in terms of reasons-receptivity and reasons-reactivity.Footnote 29 We not only need to be able to recognize reasons for action (receptivity), we need to be able to appropriately respond to the reasons that we recognize (reactivity). How sensitive one is to reasons is often understood in terms of counterfactuals—if presented with sufficient reason to do otherwise, would the agent do otherwise? Fischer and Ravizza discuss this in terms of looking to other possible worlds (relevantly similar to the actual world and in which the same agential “mechanism” is at work) in which there are sufficient reasons to do otherwise. In how many, and in which, of these worlds would the agent recognize and react to these reasons?
IV. Receptivity
With respect to receptivity, Fischer and Ravizza argue that there must be regular receptivity and an understandable pattern to the reasons an agent would recognize. Sensitivity to reasons means that the
response varies as a function of the strength of the reasons presented. In testing responsiveness in different possible worlds, we expect that, as the strength of reasons is increased, a point will be reached at which the agent, acting on the actual mechanism, will respond differently; moreover, as one moves beyond this threshold, it is assumed that increasingly strong reasons will also cause the person to do otherwise.Footnote 30
Strange patterns of receptivity would incline us to think that the agent is not actually sensitive enough to reasons to be held responsible. So if the agent would do otherwise for an incentive of a hundred dollars but not for an incentive of two hundred dollars, we might surmise that there is not an understandable pattern and the agent is not sufficiently responsive to reasons.
What kinds of reasons must an agent be able to recognize? There is some debate about whether reasons-receptivity requires recognition of moral reasons. Fischer and Ravizza argue that if an agent is incapable of recognizing moral reasons it seems problematic to hold such an agent responsible for failing to align his behavior with them. Psychopaths are a common example. Psychopaths recognize reasons for action and align their behavior with these reasons, but do not see others’ interests as reasons for acting. Along with psychopaths, Fischer and Ravizza also list smart animals and very young children, claiming that none of these three categories contain moral agents.Footnote 31
My claim here is that narrative capacity is crucial to becoming a fully reasons-receptive agent. Although recognizing reasons requires a number of different kinds of knowledge about the world, without narrative skill we would not be able to recognize the right kinds of reasons in a number of important cases. Recall that narrative allows us to understand both the “landscape of action” and the “landscape of consciousness” (see above). We come to understand the intentions, desires, beliefs, and behaviors of others (and ourselves) and so can recognize these as among our reasons for acting. Notice that some elements of our interpersonal relations are only explicable through narrative. It is only through narrative that I can, for example, understand your action as revengeFootnote 32 or see my action as forgiveness. And without the ability to mentally time travel, it is difficult to see how an agent could recognize what she has most reason to do, given that our reasons often relate to our goals and plans.Footnote 33 Furthermore, the context of a narrative can illuminate reasons that would not otherwise be recognized. What I see myself as having reason to do may depend on how a potential action’s meaning is “conditioned” by past and future. So perhaps our history together provides reason for me to help you. I cannot recognize this as a reason without narrative thought. The reason itself might be capturable in terms of so-called “snapshot” properties. It could be, for example, built in to the properties of the current circumstances that we have this history. But my claim here is not that responsibility itself is historical. My claim is that my ability to see a reason that is there now requires my ability to think narratively.
Given Fischer and Ravizza’s claim that receptivity must involve recognition of distinctively moral reasons, I think we should say more here about how coming to understand these narrative “landscapes” yields such reasons. After all, psychopaths purportedly understand the minds of others but often use this as a means of manipulation; they fail to see this knowledge as providing reasons for acting in the interest of others and so they fail to recognize moral reasons.Footnote 34 But do psychopaths really understand the landscape of consciousness? It is, in a sense, misleading to say that they do. Consider the fact that psychopaths are highly deficient in affect—that is, they do not have typical emotional responses and have trouble recognizing such responses in others.Footnote 35 So even if they are capable of predicting another’s behavior or intentions, they fail to understand emotional reactions or motivations in the way that one with “insider expertise” understands such reactions and motivations. They have a kind of “outsider status” whereby other people are not regarded as “fully real.”Footnote 36 Although they are able to predict and explain behavior in terms of desires, beliefs, and intentions, they nonetheless fail to “read” the world properly, since such reading requires understanding typical emotion.Footnote 37 Thus, the infamous psychopathic failure to empathize is closely connected to an inability to fully comprehend others and to fully recognize important reasons for acting.
Emotional empathy and care are relevant to receptivity in another, perhaps more fundamental, way. In order to recognize the reasons I have for acting, I not only need to be able to think about myself in the future. I also need to be able to care about myself in the future. It needs to matter to me how my story will go. This is highly relevant for moral agency. If I cannot care how my story will go, then I cannot see moral improvement or the securing of moral connections to others as reasons for acting. This element of caring about oneself in the future has also been discussed within the context of psychopathy. Psychopaths not only lack empathy for others, they lack it for their future selves.Footnote 38 This affects both receptivity and reactivity (more on reactivity below). Psychopaths are a real-world example of Frankfurt’s “wanton.”Footnote 39 They are cut off from moral agency not just because of their disregard of others but also because of their wantonness. The psychopath cannot worry that he has failed to “[live] up to [his] values and principles,”Footnote 40 and is thus unable to recognize an important range of reasons for acting. But it is also worth noting that even if we argue, as some have, that recognition of moral reasons is not necessary for responsibility (that is, that being incapable of feeling empathy toward others does not exempt one from moral responsibility), we might still worry that psychopaths are incapable of the appropriate patterns required for reasons receptivity. An inability to reason prudentially might be enough to exempt them.Footnote 41
It is tempting here to suppose that these reflections on psychopathy really show that empathy, rather than narrative, does the primary work with respect to reasons recognition.Footnote 42 I do not take issue with the idea that empathy plays a crucial role. But of course empathy is merely necessary and not sufficient for this recognition. And although we might seek to understand reasons-recognition as involving a set of separate necessary (and jointly sufficient) capacities, considering them independently in this way does not seem to capture what we are doing when we successfully recognize reasons. My claim is that when we do so, we are exercising a distinctive kind of know-how. The successful practitioner of narrative skill utilizes a number of capacities, but these capacities are unified in virtue of being exercised together toward the end of “reading” the world.
But what about reactivity? In addition to seeing what reasons there are, we must also be able to respond to these reasons. On Fischer and Ravizza’s account, reasons-reactivity need not be as strong as receptivity. If the agent can react to one sufficient reason to do otherwise, then the agent is reactive enough to reasons to be held morally responsible. The exact level of reactivity is a point of some controversy, but for now we can say that reasons-responsiveness does have a reactivity component such that agents may be exempted from responsibility if they are relevantly unable to conform their behavior to the reasons they recognize.Footnote 43
It is perhaps a bit more difficult to see why narrative might be required for reactivity. But our ability to conform our behavior to our judgments about what to do does require some narrative skill.Footnote 44 Once an agent has recognized what reasons she has and has made a judgment about what to do, she must form an intention that will guide her behavior. Carrying out the intention requires keeping the goal at the forefront of attention and working memoryFootnote 45 until the goal reflected in the intention has been achieved. This will not always require narrative. After all, intentional action seems possible without narrative. But when an action is protracted over time, or there is a temporal distance between decision and action, mental time travel may be required. In such cases, I will need to remember the intention and I will need to imagine the goal.
The attitude of care that is involved in narrative is also relevant with respect to being able to carry out an intention. It is because I care about how my story goes that I am motivated to try to act in accordance with what I have intended. Care will influence what reasons I recognize, but it will also influence my motivation to act in accordance with these reasons. Again, we might look to psychopaths as a kind of empirical support for this notion. Psychopaths are notoriously impulsive and, as Kennett and Matthews put it, “have trouble maintain[ing] a narrative thread . . . . [M]ental visits to the past and future engaged in by psychopaths tend to be haphazard and relatively unconstrained by the facts.”Footnote 46 They appear to act on whichever desire is strongest, even if this means abandoning a long term project.Footnote 47 Further support for the connection between care for one’s future self and reactivity comes from a recent study in neuroscience. When the “empathy” center of subjects’ brains was disrupted, the subjects were less able to delay gratification.Footnote 48
On my account, mental time travel and empathy for one’s future self involve at least rudimentary narrative skills.Footnote 49 But it is worth noting that more robust narrative skills also have a role to play in bolstering our ability to respond to the reasons that we have. A full life-story capacity, for example, may make it easier for us to respond to the reasons we have by making these reasons stronger. That is to say, the contextual elements of one’s life narrative condition the weightings of one’s reasons. Thus, someone without robust narrative skills may be deficient with respect to moral agency. This will be discussed in more detail in a later section.
V. Advantages of a Narrative Account
I have tried to show that becoming a full-fledged responsible agent requires narrative capacity. Now I would like to turn to some additional advantages of the narrative account. In discussing these advantages, I am not suggesting that other reasons-responsive views cannot account for the phenomena that I will discuss. Reasons-responsive views come in all kinds and have rich and nuanced developments in most cases. I aim instead to make the less ambitious claim that narrative capacity can help to illuminate the phenomena in question and thus fill in a reasons-responsive account in a helpful way.
A. Disharmony and self-correction
In a recent and very interesting paper, Michael McKenna and Chad Van Schoelandt look at the competing categories of reasons-responsiveness and so-called “mesh theories” of responsibility. Mesh theories are those theories that explain the control condition of responsibility, not in terms of our sensitivity to reasons but in terms of internal psychological structure—we lack control when inner harmony is lacking. For example: “On Frankfurt’s (1971) well-known hierarchical version of a mesh theory . . . an agent acts of her own free will when the first-order desire issuing in her action is one with which she identifies and, at a high-order, desires to be effective in leading her to action.”Footnote 50 McKenna and Van Schoelandt discuss the main difficulties with each category of views and then propose a hybrid view as a solution. For our purposes we need only look at their proposed difficulties for reasons-responsiveness.
They cite two general and related problems. First, they argue that reasons-responsive theories “offer no natural way to carve at the joints . . . the major architectural elements comprising the internal psychic elements of free persons of the sort we take ourselves to be.”Footnote 51 Reasons-responsiveness accounts do not explain the freedom-undermining effects of alienation from one’s own motives.Footnote 52 Furthermore, such accounts do not explain the potential for “self-correction” and “self-learning” that arises on some occasions when we are trying to decide how to form ourselves. In Robert Kane’s “self-forming actions” (for instance, his famous example of the businesswoman trying to decide whether to stop and help or to get to her meeting), it could be that sometimes we are not just deciding between actions but deciding between visions of our future self—deciding, that is, which motivations we want to be effective. McKenna and Van Schoelandt point out that reasons-responsiveness can cover all these cases insofar as the reasons we can be responsive to will reflect these inner workings.Footnote 53 The problem, though, is that
[a] good theory of human agency that accounts for freedom and responsibility should not just capture the extension of exercises of free agency. It should capture them in explanatorily illuminating ways, and it should identify relevant structure, if it really is there and really does have a causal role to play in how free actions are generated. By failing to attend explicitly to the internal operation of the complex structure of the agency of persons such as ourselves, pure reasons-responsive theories seem to fall short of capturing something that needs to be captured.Footnote 54
So although reasons-responsiveness highlights one important dimension of the control needed for responsibility—namely the proper connection to external reality—it remains relatively silent on the inner structural features.
B. Degrees of blame and difficulty ranking reasons
This points to another potential question about reasons-responsiveness. Because it tends to focus outwardly, as McKenna and Van Schoelandt note, we might wonder how it explains certain kinds of differences in our responsibility assessments. Responsibility theorists note that our practices and intuitions are not all-or-nothing. They reflect degrees, either of responsibility, or of blameworthiness (and praiseworthiness). One plausible principle regarding what accounts for these degrees is based on difficulty: the harder it is for an agent to do what she ought, the less we blame her for failing to do so.Footnote 55 We might, that is, blame an agent less if, under the circumstances, responding to reasons is more difficult than in the usual case. Sometimes it is harder based on features of the external circumstances. But other times what makes it more difficult are things that are internal to the agent. A Fischer-Ravizza type view can certainly accommodate this in various ways, but to echo the McKenna-Van Schoelandt concern, does it illuminate it? It is interesting to note that Fischer and Ravizza suggest the notion of degrees of responsibility with respect to children:
Consideration of children underscores the sense in which moral responsibility is not a threshold concept; our ordinary ascriptions of responsibility do not seem to be “all-or-nothing” judgments with no allowance for degrees of responsibility. In the case of children it is most natural to think of young individuals gradually developing an understanding and responsiveness to a range of reasons, including those that stem from moral demands. And it is this gradually expanding range of responsiveness that indicates the class of actions for which the child is properly held accountable.Footnote 56
Similarly, David Shoemaker suggests that children become more answerable for their actions as they come to recognize “an increasing variety of “reasonish” facts.”Footnote 57 While I agree that children recognize a narrower range of reasons, this fact may not cover all relevant cases. In many cases, it seems not to be the range that is at issue, but the weighting given to each identified reason. Developmental psychologists Monisha Pasupathi and Cecilia Wainryb put it this way:
Developmentally, children develop a sense of moral concerns as distinct from other types of concerns quite early in life . . . Even very young children view harming others as wrong . . . , show evidence of empathy with others’ experiences . . . , and consider moral issues as distinct from other domains of social cognition . . . . However, that is not to say that developmental changes have no implications for the nature of judgments in situations of competing concerns. Children weight competing concerns in mixed situations differently across different developmental periods . . . , making developmental changes one of the critical considerations in just what children and adolescents make of complex situations involving harm.Footnote 58
So it seems that children may sometimes fail to act as they ought, not because the range of reasons is smaller (that is, not because they lack the relevant moral reasons) but because they fail to give them their proper weight in the particular circumstance. But one might wonder if we can appeal to the range of reasons by further specifying what the reasons are. So, for example, David Shoemaker suggests that for moral answerability, an agent needs to recognize not just reasons for acting, but “instead of” reasons. The agent needs to be able to justify her behavior by saying why she did A instead of B.Footnote 59 Maybe this is what children cannot do when they have trouble balancing competing interests. In order to have justification, there needs to be “a contrast class of reasons relative to which one’s judgment may be better or worse.”Footnote 60 Perhaps a child cannot justify her behavior by explaining why she acted on self-interest instead of helping someone in need.
While I think it is plausible that children may have more difficulty justifying their behavior in this way on many occasions, I still do not think this explains all the relevant cases. One group of developmental psychologists examined children’s moral development by asking children to tell about a time they harmed someone. Here is an example from one of their interviews (the interviewer questions are in all capitals):
DO YOU THINK IT WAS OKAY OR NOT OKAY FOR YOU TO NOT LET JORDAN PLAY WITH THE BIONICLES? I don’t think it was okay, it was not okay, but I just didn’t have enough stuff [i.e., “Bionicles”]. SO WHY WAS IT NOT OKAY? Because she was just being left out and I was just playing with my friend and my brother. She might feel sad and it’s not right. But the choice was to leave her out or destroy my game because we couldn’t play if she played too.Footnote 61
The child can clearly explain why he or she chose to leave another child out in terms of “instead of” reasons. To this extent we might be inclined to hold the child responsible. But we might still feel like it was harder for the child to make the right choice and blame the child less. It appears that the failure in this case really was in weighting reasons properly rather than in failing to recognize the appropriate range of reasons (which includes “instead of” reasons).
On its face, traditional reasons-responsiveness does not distinguish between the agent who culpably fails to weight reasons properly and the agent who, through some relative deficiency, has more difficulty doing so. Suppose we look again to Kane’s businesswoman. Suppose that she decides to go to her meeting and is blameworthy for not helping. She recognizes reasons for helping and reasons for going to her meeting, but at the end of the day, she has failed to give her reasons for helping the weight that she should have given them. It seems to me that we are inclined to blame the businesswoman to a greater degree than the child who also improperly weights these kinds of reasons. I suspect this is because we regard children as lacking in some of the resources that would enable the right evaluations.
We might be tempted to think that “regular receptivity” can do all the work we need. That is, maybe the child does in fact fail the requirement of an appropriate pattern of recognition based on the objective strengths of reasons. But this need not be the case. This “objective strength” test is a test of rationality, not a test of moral insight. It is looking to see whether an agent is rational enough to know that a stronger reason of the same type would be a sufficient reason if the original version was sufficient. It is not testing how well agents weigh competing kinds of reasons against each other in particular contexts. Why suppose that the child is irrational in the sense tested by appeal to appropriate patterns? Suppose the child would have done the right thing were she presented with an additional reason, such as a reward. And suppose that she shows the correct pattern in the sense that she would do the right thing in virtue of rewards of increasing value. This seems like a perfectly coherent and plausible scenario. The child’s rationality does not seem to be in question, nor does her ability to recognize moral reasons.
Notice that reasons responsiveness cannot require that one in fact weighs reasons properly, or else the businesswoman would not be blameworthy. But it does seem to require that one be able to weigh reasons properly. And it seems to be because the child is nonculpably deficient in this capacity (relative to a typical adult moral agent) that we might be inclined to assign her less blame. But how can a reasons-responsiveness theory capture or explain this ability to properly evaluate reasons? If this is the ability that is required, what does it mean to say that an agent possesses it? Typically, the reasons-responsiveness theorist wants to capture whatever ability is required in terms of sensitivity to reasons by appealing to counterfactuals. My being able to respond to reasons means that if I were to have sufficient reason to do otherwise, I would do so (in a suitable number of worlds with the right conditions, and so on). But how do we measure the ability to weigh reasons properly? We might say that if the agent weighs them properly in at least one other world (holding fixed the appropriate background conditions) then she was able to do so. But the problem is that it seems plausible to think weighing is significantly more difficult for a child, even if she might get it right in a close enough world. And we cannot necessarily solve the problem by claiming that the degree of blame is based on the number of worlds in which the agent is able to do the right thing. We cannot do this because we want to be able to say that sometimes even a very weak-willed or morally corrupt agent is fully blameworthy. Such agents might weigh reasons improperly in many worlds. We might even think that it is therefore in some sense harder for such agents to do the right thing. But then perhaps it is not clear that all kinds of difficulty mitigate blame.Footnote 62
C. Explaining these phenomena in terms of narrative capacity
Can narrative help explain what’s going on in the cases above? Can it help explain the freedom-undermining alienation from one’s own motives? Can it explain our potential for self-correction in the sense discussed by McKenna and Van Schoelandt? I think it can, in both cases. The first worry was that the reasons-responsive view does not illuminate what has gone wrong when freedom is undermined by an unharmonious mesh. Narrative is able to illuminate this phenomenon by providing the internal structure that was thought to be missing from a reasons-responsive view. If my freedom is undermined because I am alienated from one of my motives, that means my desire is a reason for acting, but it is not a reason that I want to be effective. The unwanted desire is playing a role in my story that I do not want it to play.Footnote 63 As Frankfurt famously argues, it is because we can reason and care about our will that freedom of this sort can be a problem for us.Footnote 64 Having a narrative capacity means caring about how one’s story goes. It also means having a perspective from which to evaluate and a contextual background against which our motivations can be evaluated. It is also this that allows for an explanation of our ability to self-correct when we are reflecting about our motives during difficult choices. The businesswoman, for example, cannot consider who she wants to be when deliberating, if she does not have robust narrative skills.Footnote 65
What about the other concern regarding degrees of blame? Here again I think narrative capacity can fill in a reasons-responsive account. Think again of the child who recognizes the right kinds of reasons but has trouble assigning them the morally appropriate strengths. My claim is that it is in light of the less-than-fully developed nature of the child’s narrative capacity that she is less capable of assigning the proper weights. This makes it harder for her to do the right thing. As a “meaning-affecting” relation, narrative involves recognizing important contextual relationships. The same reason can appear more or less powerful based on how we interpret its relationship to other things. Children may have more difficulty doing the right thing sometimes even when they are able to pick out the relevant reasons, because they may be less able to appreciate their force. In some instances, it could be that an adult can appreciate this force because he is able to reflect upon why it is an important reason.Footnote 66 And this better accounting stems from narrative capacity. For example, the child might think: “I should not exclude her because it will make her sad and it’s not nice to make her sad.” But an adult can give more weight to this by thinking: “I don’t want to be the kind of person who makes others sad” or “I care about my relationship with her and do not want to undermine it just for the sake of my current enjoyment.”Footnote 67 This might seem like we can simply appeal, once again, to the range of reasons (discussed above). But more often, I think, an adult moral agent’s assessment of strength will be due to his narrative capacity without relying on inferences like these. He may be able to feel the strength of a reason based on how it resonates emotionally within the context of the story.Footnote 68 Or he will just know how to assess it without actively considering why it is stronger. We learn how to interpret behavior and to recognize the strength of reasons in context without necessarily knowing why we assess them as we do. So an adult agent typically need not rely on propositions about why he gives these reasons the weights that he gives them.
VI. Taking Stock
In the foregoing, I have tried to show why narrative capacity is important to reasons sensitivity and responsible agency. One way of summing up my attempt is to say that sometimes what we want to know in order to understand responsibility is what the capacities look like that allow us to become responsible agents. In other words, it is not always enough to say what we are doing when we act morally responsibly. We might also want to say something about how we are doing it.Footnote 69 On my view, narrative skills give us the ability to recognize a broad scope of reasons and provide them with their proper weights. I have aimed to show that agents without these skills are at a disadvantage and thus are not (yet) full-fledged responsible agents.
VII. Problematic Narratives
In closing, I would like to consider a potential worry for my view. One might wonder whether narrative actually undermines our moral agency and our ability to do what we ought to do. When we become competent storytellers, we might undermine our moral agency in ways not open to those without such capacities. We might construct narratives that actually constrain our behavior or block us from recognizing important reasons for acting.Footnote 70
For an extreme example we might look to someone like Dostoevsky’s Underground Man. In an important scene toward the end of the novel, after befriending and connecting emotionally with Liza, a prostitute, the Underground Man suddenly degrades her by giving her money. He describes his motivations for this behavior:
The thought of doing it occurred to me while I was running up and down my room and she was sitting behind the screen. But this much I can say with certainty: although I did this cruelty on purpose, it came not from my heart, but from my stupid head. This cruelty was so affected, so much from the head, so purposely contrived, so bookish, that I myself could not bear it even for a minute . . . .Footnote 71
He acts badly in this moment as the result of this “bookish” and “purposely contrived” scenario. He develops a narrative based on various literary tropes and ideals. It appears that thinking in terms of constructing a narrative and being hyper-aware of such construction has led him astray. But what is fascinating about this case is that the Underground Man is also highly aware of what he ought to have done. And we certainly hold him responsible for his behavior. His failure is not in using narrative reasoning, but in using it poorly and incorrectly. He constructs a narrative that actually serves to suppress the natural narrative skill (or “insider expertise”) that would (and sometimes does) allow him to connect with others and to do what he ought to do. The narrative he superimposes on his life is full of caricatures, not real people. It is devoid of genuine emotion (though he is not).
There are also more mundane cases, and cases in which we do not suppress the natural and interpersonal narrative skills in question, but in which we still lead ourselves astray. We might, for instance, insist on seeing ourselves in a particular way such that we are not able to notice reasons that count against this self-conception. Or we might convince ourselves about our lack of reactivity (“I can’t help myself”) and therefore undermine our own motivations to resist. Another dimension of this worry is that those with the best narrative skills might also be the most liable to use these skills in an agency-undermining way. The better one is at constructing narratives, the better one might be at self-deception, rationalization, and so on.
I readily admit that it is not far-fetched to think that we really can and do get carried away with telling the story we want to tell, and this can hinder us from acting well. But problematic narratives need not undermine the position defended here. Problematic narratives just go to show that it is part of the irony of our moral lives that the things that make us capable of moral agency can sometimes be the very things that make it harder for us to do what we ought to do. Although emotionally empathizing with others may be required for moral agency, it can also lead us to revenge and cruelty.Footnote 72 Being highly intelligent creatures allows us to reflect and act on how best to preserve and serve the interests of others, but it also, as Dostoevsky frequently implies, sometimes interferes with compassion and love of individuals. So although these capacities can lead us astray, without them, moral agency will not be possible.
