Abstract
The widespread embrace of virtue ethics in Catholic morality has not overcome sharp disagreement on particular moral issues. The authors argue that more attention be paid to developing the account of what a virtue is (“virtue theory”) in order to connect virtues with individual acts. Building on Thomas Aquinas, the authors suggest social cognitive theory provides key insights into the mechanics of agents “acting from character” and offers an empirical program that can further our understanding of moral disagreement.
Amidst much fragmentation, Catholic moral theology seems to be united on one thing: the importance of the virtues. Appeals to the virtues appear across the ideological spectrum, and animate discussions in both personal and social ethics. 1 Even Pope Francis’s approach to the moral life is understood as centering on the virtues by figures as prominent and diverse as Cardinal Walter Kasper and Bishop Robert Barron. 2
Can this consensus help resolve other aspects of conflict and fragmentation in the field? In 2014, David Cloutier and William Mattison, in a review article on the resurgence of virtue, noted that a “major pathway” for the further development of theological virtue ethics was a more “rigorous analysis of prudence.” 3 Without this, the authors warned, virtue ethics “tends to leave moral theologies tilting either toward legalism or toward appeals to a suspiciously non-rational ‘conscience.’” 4 Put another way, a careful understanding of virtuous practical reasoning (of which prudence is the central component) is necessary for making more precise connections between broad claims about the importance of the virtues and specific contested issues in moral theology. Their point seems prescient in light of the fact that, despite the broad claims about Francis and virtue cited above, debate over Francis’s exhortation Amoris Laetitia has descended into a law-versus-conscience duel. Appeals to justice and charity, as well as to more specific virtues like mercy and compassion, are made by those on both sides of this debate, even in support of diametrically opposed positions. How are we to understand the logic of such appeals?
What we need is a better understanding of virtues having a “logic.” We need some way to, as Daniel Westberg names the task, “describe the connection between the mind of the agent and his action” 5 —in a way that doesn’t reduce the process to agents simply facing a raw choice to obey or disobey a rule. This is the task of our article: to offer a more developed model of how virtue works by integrating the insights of a particular research program in psychology, Social Cognitive Theory (SCT). We do not propose to adjudicate particular moral disagreements, but instead argue that these disagreements should prompt us to do (neglected) work in developing a better understanding of virtue itself.
This article is an exercise in what philosophers have called “virtue theory.” 6 Philosophers have distinguished “virtue theory” from “virtue ethics,” by noting that “virtue ethics” involves some claim that concepts of virtue—rather than, say, law or utility—are foundational for moral evaluation. However, “virtue theory” tries to explain what virtue is; a thinker may have a virtue theory without being a “virtue ethicist.” 7 Virtue theory is peculiarly underdeveloped in Catholic moral theology, in part because the most prominent writers of recent years have, as we will show, chosen one of two paths: either hewing closely to Thomas Aquinas’ virtue theory or paying little attention to the psychological mechanics of virtue. This is particularly unfortunate because, as Elizabeth Anscombe noted in the article universally recognized as initiating the revival of virtue, the failure of other extant forms of philosophical ethics is due to their inattention to and unrealistic accounts of a “philosophy of psychology.” 8 She noted at the time that the field lacked “an account of what type of characteristic a virtue is . . . and how it relates to the actions in which it is instanced: a matter which I think Aristotle did not succeed in really making clear.” 9 Nearly fifty years after Anscombe, Julia Annas can still note that a perennial objection to virtue ethics is that it “cannot provide something we need for an ethical theory to be any good . . . ‘action guidance.’” 10
Even if Aquinas makes things somewhat clearer, Andrew Pinsert argues that “what Aquinas means by the term ‘virtue’” is still the “greatest difficulty” for a “satisfactory interpretation” of St. Thomas. 11 Thus, the work Anscombe prescribed has still not been done satisfactorily in moral theology. In the absence of a clear model of virtuous practical rationality, by which we can understand its psychological mechanics, we really cannot articulate what is going on when people disagree about “actions in which it is instanced.” Hence, the aforementioned tendency to default away from virtue language when arguing over specific issues. We suggest a model of virtuous practical rationality that integrates important elements of SCT with the “standard model” largely indebted to Aquinas. This “interactional” model of virtue enhances the standard model, especially by offering new insights on the relations between reason and the passions (or emotions), on the notion of practical reasoning itself, and on the idea of will. After developing the interactional model, we conclude with a few forward-looking suggestions for how the model might help us better understand specific disagreements.
What Is a Virtue? Beyond the Standard Model
Current discussions of virtue in Christian ethics tend to go in one of two directions when dealing with the psychology of the virtues: they work within Aquinas’ “standard model” for the underlying psychology or they shift attention away from any specific psychological model, typically in favor of an analysis of social relationships and social narratives. Charles Curran’s recent treatment of virtue ethics focuses on four thinkers, two of whom, despite some divergence, follow Aquinas closely (Jean Porter and Romanus Cessario) and two of whom, also in divergent ways, develop approaches that shift the focus of virtue to relationship (Stanley Hauerwas and James Keenan). 12 Similarly, in the Readings in Moral Theology volume Curran and Lisa Fullam compile on virtue, the first four essays (“virtue ethics in general”) offer a summary of Aquinas, followed by three contemporary narrative/relational approaches: MacIntyre, Curran himself, and Anne Patrick. 13 In Jennifer Herdt’s recent overview of “varieties of contemporary Christian virtue ethics,” she offers a somewhat (though not exactly) similar distinction between “natural law” and “particularist” approaches, the former largely following Aquinas and the latter emphasizing (in different ways) the importance of formation via social narrative. 14
It is true that most theological ethicists build on the basic definition that a virtue is a good habit. 15 For example, Curran begins his chapter on virtues by simply saying Aquinas understands virtue as “a good habit or stable disposition inclining the person toward the good.” 16 The rest of the chapter, however, pays little attention to more clearly specifying this psychology. Curran does note in passing that Aquinas sees the virtues as perfections of four specific powers, but says simply that “many today reject such a Thomistic ‘faculty’ psychology.” 17 Instead, Curran prefers a “relational” model, in which different virtues specify characteristics of different relationships.
Though differing from Curran in many ways, Stanley Hauerwas’s approach to character follows a similar trajectory away from the development of psychological models of moral agency. While his initial work promoted Aristotle and Aquinas as an alternative to both legalistic and situation ethics models, in part because they possessed a better understanding of the agent, he suggests “the analysis of character I provided there was far too abstract,” such that “the material content of character remained largely underdeveloped.” 18 MacIntyre’s important idea of an “intelligible action” and the discovery of John Howard Yoder’s exposition of Jesus helped Hauerwas to come to see morality and character in terms of “the ability to place our action within an intelligible narrative.” 19 He notes in a later essay that he “quit thinking about agency and began to think more about narratives that constitute our lives.” 20 Following this move, Hauerwas’s work increasingly focused on the social “community of character” that the church itself seeks to be.
Discussions of narrative and relationship are, of course, central to understanding virtue—even Aristotle’s ethics must be understood as embedded within a larger social vision, his accounts of politics and friendship. 21 Yet “virtues” themselves are not primarily attributes of relationships or polities; even the initial notion of virtue as a disposition or habit clearly places it within the psychology of persons. Moreover, the appeal to narrative can’t replace an account of agency without engendering a misleading passivity; rather, it calls for an account of agency that can make sense of what agents are doing when they interpret stories in relation to themselves in moral terms. Thus, even if there is an appropriate emphasis on the contexts (social, narrative, etc.) in which virtues are understood, this is not a substitute for a psychology of virtue, but instead a challenge to develop a model of virtue that incorporates these essential aspects.
Perhaps the “Thomistic faculty psychology” that Curran dismisses could do so? As Westberg’s account shows in detail, Aquinas’ psychology was considerably distorted in later moral manuals, especially by reducing it to a mechanistic set of complicated steps—a reductionism that is the likely target of Curran’s critique. 22 Westberg’s more subtle account defends Aquinas’ distinguishing of “faculties” of reason, will, and passion against a modern tendency in moral theology to simply talk about “the whole person . . . without trying to relate these activities to specific elements of the psyche” on the basis of the need to understand that the person is both “moved by reality” and also “active.” 23 Practical reason by definition engages the world, but its engagement must be one that is not mere “re-action.” This point is particularly valuable for any discussion of virtue, since on all accounts of it, these habits are supposed to be a definite enhancement of the agent’s ability to direct his or her own actions toward the good.
But how do they do this? When Aquinas speaks of virtues as habits, he understands these as “perfections” of “powers”—or as Jean Porter describes them, of “the capacities distinctive to us as creatures of a specific kind.” 24 Aquinas correlates virtues with specific human capacities, the ensemble of which must function well in order to achieve the good. Aquinas names these capacities as “desire” and “with qualifications, our intellectual and rational capacities.” 25 Aquinas divides our desiring capacities into will and the passions, and further subdivides passions into concupiscible and irascible passions. Thus, the cardinal virtues are fully accounted for as perfections of these four capacities: (practical) intellect, the will, and the two sorts of passions.
We should remind ourselves that “faculties” are not like physical organs located “inside” the person, but an intellectual model. As Ian Barbour notes, models are neither reality itself, nor pure imposition of a theoretical construct. The aim of a theoretical model is explanatory: “its chief use is to help one understand the world, not simply to make predictions.” 26 It is not simply a “useful fiction”; it should be taken “seriously” but not “literally.” 27 The return of virtue is in part a recognition of the limits of the explanatory value of other models of moral agency. Compared to a model of moral agency that is purely legal or one that is purely calculating of utility, this person-centered model seems to better explain reality.
Is the reason/will/passions model of powers qualified by virtues sufficiently explanatory? Models of virtuous agency face certain explanatory challenges from empirical studies of behavior. We will mention two, both of which invite an enriching of the standard model of virtues perfecting powers. One, the “situationist critique,” emerged in the past two decades in philosophical ethics, and at its extreme, claims that empirical evidence shows that human agents simply do not display “robustly reliable” character traits across time and situations, and so don’t have the “global character traits” virtues purport to be. 28 We will discuss this further later in the article, but at the very least, the critique requires us to give a better account of how virtues operate than simply asserting “Joe is honest” means possessing a “habit” by which he tends to do the honest thing. 29
A second challenge comes from increasingly persistent empirical evidence for the central role of emotions in making moral judgments. 30 It is true that a typical Thomist virtue model using the categories of reason/will/passions does offer an account: passions are subjected to the order of reason by the will. However, the mechanics of this model have been hotly contested. While Westberg eloquently suggests that Aquinas offers “an intimate combination of intellect and appetite” that involves a “dialectic of reason and desire,” 31 the specification of this combination has prompted significant debate, with a number of authors criticizing what has been dubbed a “spontaneity view” of virtuous passions. 32 In his recent monograph Aquinas on Virtue, Nicholas Austin has synthesized much of this recent work. Austin proposes a “moderate spontaneity view” of the emotions that “acknowledges the place of reason and will in all morally virtuous action but also finds a more positive place for the participation of habits subjected in the sensitive appetite.” 33 Austin’s account traces how a “participative” rather than merely “derivative” model of the rationality of the passions can be supported by attending to key distinctions in Aquinas. Most especially, Aquinas’ distinction between (disruptive) “antecedent passions” and (potentially virtuous) “consequent passions” can be read in such a way that renders the latter as “in a way rational.” 34 Jensen further develops this distinction by noting that “antecedent” and “consequent”—despite their obvious connotation—should not be understood as chronological but causal, suggesting a new category of chronologically prior but not antecedent passions, an addition that would maintain the proper subordination of passions to judgments of reason. 35
While this discussion certainly shows how attentive to subtleties Aquinas was, we wonder whether this esoteric and possibly unresolvable competition over Aquinas’ own texts is the best way forward. We suggest instead an enhancement of the basic model that better explains empirical insights. In particular, we think it would be helpful to recognize that modeling human psychology in terms of “reason” and “passions” may be too oversimplified and apt to mislead. The above-mentioned discussions tend to reify the dichotomy into two different “things,” often setting up a kind of “battle” between reason and emotion, in which each has to vie for political control of psychic “territory.” Try as he might, Austin’s sympathetic reading cannot avoid a recognition of a kind of master–servant relationship at the core of Aquinas’ explanations. There is a need for a model that helps us get beyond speaking as if these are two separate “things.” Paul Lauritzen, thirty years ago, was already alerting theological ethicists about the increasing evidence of the “cognitive” character of emotions. 36 Many recent examples repeat this call. Christopher Vogt draws on the work of Michael Spezio, who notes that reason and emotion should not be seen as “dichotomous, opposing systems,” but as a “complex interconnection of circuits.” 37 In another article relating virtue and neuroscience, Martin Rhonheimer suggests a similarly holistic idea when he describes virtues as “reasonableness inscribed in the tendencies of the senses.” 38 Jason King, using Bernard Lonergan and some scientific and philosophical work, suggests that theology “needs to update its emotional psychology” in a way that is “not hampered by faculty psychology” but can still draw upon it. 39 We will suggest that this debate is best approached not by pitting reason and feeling against one another as different “capacities” vying for superiority, but by developing a richer model of what is actually going on when we perceive and respond to the world, and then understanding virtues in terms of this model.
A good place to turn for such an enhanced model of the mechanisms of virtue would be to psychologists themselves. But both fields could benefit from some fluent “language-translation” on these questions. Psychologists have only recently begun to closely attend to the notion of virtue, and the engagement is very preliminary. 40 And ethicists appealing to (or sometimes against) psychology may neglect what they would never neglect in their own field: the diverse theoretical approaches of specific forms of psychology, and especially the intellectual genealogies of such forms. For ethicists, the question should not be “What can or can’t ‘psychology’ tell us?” but “Which psychology should we turn to and why?”
What Psychology Can Contribute: Toward an Interactional Model of Virtue
Social cognitive theory is a particularly promising psychology to which those interested in virtue should turn. As the theory’s name suggests, the distinctive emphases of the theory both on cognition and on social context fit but also provide precision for both the standard-model Thomists (who emphasize the place of rationality in virtue) and the narrative and relational virtue ethicists (who emphasize the social context of virtue).
Social Cognitive Theory: Moving Psychology beyond Traits
Some background on social cognitive theory will be helpful. The most central figures in its development are Albert Bandura and Walter Mischel. The historical context for their works illuminates their dissatisfaction with overly simplistic models of personality described by other approaches to psychology. Bandura developed his theory in part in response to the “one-sided determinism” suggested by then-dominant psychologies, such as radical behaviorism, whereby a single factor such as one’s environment was understood to be the central determinant of action. He proposed, instead, “triadic reciprocal determinism,” by which the environment, person, and behavior all reciprocally interact with each other. The context for his work was observational learning, that is, how we learn by watching the actions of others and the consequences of those actions. He argued that “human nature is characterized by a vast potentiality that can be fashioned by direct and observational experience into a variety of forms within biological limits.” 41 While humans have some limits based on their biology, beyond these limits they are capable of a great deal. The degree to which people live out this potential is the product of the examples they see others set and their own experiences.
The work of Walter Mischel also stemmed from a dissatisfaction with overly simplistic models of psychology—in his case, a dissatisfaction with the idea that personality could simply be described in terms of the manifestation of certain global “traits” that a person “possessed.”
42
In his words, Global traits and states are excessively crude, gross units to encompass adequately the extraordinary complexity and subtlety of the discriminations that people constantly make. Traditional trait-state conceptions of [persons] have depicted [them] as victimized by [their] infantile history, as possessed by unchanging rigid trait attributes, and as driven inexorably by unconscious irrational forces . . . A more adequate conceptualization must take full account of [the human person’s] extraordinary adaptiveness and capacities for discrimination, awareness, and self-regulation; it must also recognize that [people] can and do reconceptualize themselves and change, and that an understanding of how humans can constructively modify their behavior in systematic ways is the core of a truly dynamic personality psychology.
43
Against these oversimplifications, Mischel and his longtime collaborator Yuichi Shoda instead developed a model of “cognitive affective processing systems” (CAPS) as central to the person. Far more than simply exhibiting “traits,” people in any given context have encoding strategies (or ways of perceiving), goals, self-regulatory strategies, beliefs (such as about the consequences of actions), and affective responses. For instance, a student might initially perceive a course as a threat, develop the goal of not failing the course, develop self-regulatory strategies to avoid failure, believe that particular study habits will prevent that threatened failure, and feel anxiety. But, unlike a simple trait explanation, the student is not simply an “anxious” person; instead, in accord with triadic reciprocal determinism, all of these perceptions, feelings, and strategies are dynamic, changing in response to context and action, but also changing actions and context. Thus the student, in response to any number of factors, from a family crisis that suddenly makes the class relevant to unexpectedly interesting classmates, might come to perceive the class as an opportunity, develop the goal of understanding this interesting material, create new study habits so as to understand, believe these habits will facilitate understanding, and feel joy.
The crucial insight here is that people do not simply “have traits” and display some sort of consistent behavior across situations. Mischel and Shoda’s theory “takes into account the meanings that objective situations have for people.” 44 The student in the above example is not simply “anxious” or “studious” in relation to taking a class. Rather, the “situation” is always construed, and so “consistency” is not primarily to be understood as responsive to “objective” features, but to what they call the “psychologically salient” meanings that a given situation has for an agent. 45
For our purposes of modeling virtues, the key move is that both Bandura’s and Mischel’s theories display a central emphasis on interaction, especially between aspects of the person and aspects of a situation. As the aforementioned disagreements among virtue ethicists over “situationism” and the “territory” allotted to the passions illustrate, too often our models try to force an “either/or” choice, in which explanations of behavior must be given in terms of either some personal trait or some external situation. An interactionist model suggests this is a false choice; what we in fact need is a both/and explanation. Bandura’s emphasis on “reciprocity” means that “it is generally wrong to ask: ‘Which factor . . . is the cause?’” 46 In explaining this interactionist model of personality, Daniel Cervone and Brian Little refer to a distinction originally made by Kurt Lewin between “Aristotelian” and “Galilean” forms of explanation. In the former, entities are explained in terms of “inferred essential qualities” from which individual instances may deviate; however, in the latter, “actions are explained by simultaneous reference to qualities of the entity and environmental forces in play at the time.” 47 Notice that, for virtue ethicists, this does not mean scrapping claims about “dispositions” of agents, but it does suggest that all such claims should focus on what might be called the “dynamic” nature of those “qualities,” in that they are not fully specifiable apart from their interactional instantiations.
Some philosophers have already recognized the promise of social cognitive theory as an empirically robust model for what virtues are. Both Nancy Snow and Daniel Russell have proposed that, in order to get beyond the situationist critique, virtues be understood as “CAPS traits,” suggesting that persons have distinctive “behavioral signatures” that involve active cognitive-affective construal of situations they encounter. Snow defines virtues as “tightly-integrated cognitive-motivational-affective wholes in which the motivations intrinsic to virtues deeply shape the other constituents such that, were the motivations to be removed or replaced, the cognitive elements would change also.” 48 Her model also integrates the importance of relationality, characterizing virtues as particular forms of “social intelligence” that help us to work effectively with others. Similarly, Michael Lacewing draws on the social cognitive approach to develop a notion of virtues as “expert moral intuitions” in his attempt to overcome the dichotomization of reasoning and intuition that has resulted from the widespread adoption of Jonathan Haidt’s “social intuitionist model” of moral psychology. 49 Building especially on the work of psychologist Darcia Narvaez, Lacewing endorses the social-cognitive model since it offers “a broad conception of reason” that is always interactive with the “intuitions” at the heart of Haidt’s theory. 50
The KAPA Model: Explaining the Psychological Parts
However, this CAPS account is still a step short of being a genuinely explanatory model of virtuous agency. As Daniel Cervone has noted, CAPS functions as a “meta-theory” of personality, in which it is “conceived as a complex, dynamic organization of cognitive and affective elements”; however, “the exact nature and content of these elements remains unspecified.” 51 That is to say, CAPS doesn’t really offer an account of the “parts” of a moral psychology.
While Cervone entertains the possibility that “people may not have parts,” he argues that there are important gains to developing an explanatory model that will give more insight, particularly into “stable personality structures.” 52 His model, called “Knowledge and Appraisal Personality Architecture” (KAPA), is developed as a “Galilean” alternative to trait theory’s “Aristotelian” approach (recall the contrast from Lewin above). Among Cervone’s several critiques of typical trait methodology, we focus on two of them.
First, he indicates that traditional trait theories “disregard individual idiosyncrasy and instead describe abstract classes of person.” 53 To illustrate this, consider people who act in ways that might be labeled “agreeable.” For instance, they are typically gracious to their partner’s parents, avoid criticizing their partner’s driving, and try to resolve tension in the dating relationship. The statistical approach of trait theory would treat as equivalent one person who resolves the tension and avoids criticizing driving but who is not gracious to the partner’s parents and another who avoids criticizing driving and is gracious to the partner’s parents but does not resolve tension, rather than trying to understand the idiosyncrasies differentiating these two. KAPA is designed to understand those idiosyncrasies.
As a second critique of trait theory, Cervone notes that it makes “the claim that the statistical qualities correspond to real things that reside in people’s heads and that constitute the core of their human nature and function as the causes of their behavior.” 54 For example, it would take the historical average (e.g., that the people are gracious, avoid criticizing, and resolve tension 66% of the time) as corresponding to something “real” about them—that they are “agreeable,” and further assume that this thing about them causes their behavior. In contrast, KAPA, as a Galilean approach, would maintain “a basic principle of field theory” that “explains an entit[y’s] actions by reference not only to qualities of the entity, but to the entire situation within which the entity is acting. Explanation is interactional. One must understand relations between features of the entity and features of the situation.” 55 That is, causal explanations of behavior would not focus just on the person, but also on the entire situation in which the person is acting, and the person’s relation to that situation. 56
This twofold critique of trait theory seems especially important for those interested in exploring the relationship of virtue to modern psychology, because much of the conversation of virtue theory with psychology has been carried out with trait theorists. 57 As opposed to working with such a static model for possessing a virtue, KAPA does not seek to compare what people do “on average,” or their “overt behavioral tendencies,” as many other personality approaches do. Rather, it seeks “to assess patterns of stability and variability [in] overt personality functioning, as well as the contextualized personality structures and processes that contribute to these overt patterns.” 58 KAPA assessment methods are designed to be “sensitive to individual idiosyncrasy” and to “people’s capacity to contribute causally to their experiences and actions.” A focus only on averaging observed outward behavior—and then attributing such average behavior to a “trait”—completely misses the work agents do (and do differently from one another) in understanding their own behavior as theirs, as coming from them.
How then does Cervone propose to model behavior? His account is, of course, interactional—it involves the interaction between “enduring mental representations” (knowledge) and “dynamic evaluations of the meaning of encounters” (appraisals). 59 Modeling the agent’s internal “parts” in terms of knowledge and appraisals helpfully develops Aquinas’ insights about reason and emotions, but avoids some of the territorial dichotomization those terms can imply.
While “knowledge” is straightforward, the concept of “appraisal” will require some explanation for ethicists. It is drawn from appraisal theories of emotions, but is not simply another name for feelings or emotions. 60 Interestingly, the historical evolution of appraisal theory has roots in Thomism. In the 1940s, the “mother of appraisal theory,” Magda Arnold, was spurred by dissatisfaction with the understanding of emotion drawn from behaviorist and physiological approaches. 61 As she returned to Catholicism she was introduced to Thomistic psychology by the Jesuit psychology professor Fr. John Gasson, finding it especially appealing as an alternative to “the determinism of both Freudian theory and behaviourism” that, in her words, unlike “stimulus–response theories and their machine models of the human being,” explained how “all vital activities must have a source within the individual.” 62 This spurred her development of appraisal theory.
From an appraisal perspective, humans appraise, or judge, objects and encounters in light of their motivations. Appraisals prepare the body for action, but allow flexibility not present in reflex responses. This constellation of appraisal and bodily response to appraisal represents the emotion system. Thus, for instance, consider the potential emotional responses to noisy work outside one’s office. If humans were driven by reflex, the noise might lead automatically to specific action. But emotions allow flexibility in response. Some might not be distracted by the noise, and so judge the noise to be irrelevant to their motivations, thereby experiencing no emotion. Others might judge the noise to be inconsistent with what they want, for instance if they are trying to do work that does require careful attention. Some of these might also judge that other people are the cause of the noise, and this pattern of appraisal is likely to prepare them to confront those making the noise, and to the experience of anger. Others might judge that they cannot do anything to stop the noise, and so their bodies prepare to give up, and they will experience sadness. But, unlike with reflex, the angry person can also choose whether to act on the anger, and the sad person can choose whether to act on the sadness.
Returning to Cervone’s overall model, what difference can the KAPA model make for explaining the relationship of external behaviors and internal virtues? To illustrate its empirical approach to investigating characteristic actions, in one study Cervone 63 asked people to report their idiosyncratic strengths. Subsequently those strengths were mapped onto a variety of situations that reflect agreeability, such as resolving tension in a dating relationship and being gracious to a partner’s parents. Two subjects are illustrated in Figure 1 below. Both believed that their strengths were relevant to these “agreeable” behaviors, and judged themselves capable of engaging in those behaviors. But for one person the relevant strength was being nice, whereas for the other the strength was being able to manipulate people. Here is the key point: to a trait theorist, both might appear to be people high in the trait of agreeableness. But this would miss something fundamental and distinguishing about these people’s agency, that KAPA assessment notices: that the same actions can be motivated by different understandings of what one is doing in carrying out that action. This distinction in the agent’s own understanding is central to any account of virtue that does not reduce virtue to behavioral automaticity.
Example of KAPA assessment for two individuals. From Cervone, “Architecture of Personality,” 196.
Notice how the assessment exercises offer considerable insight into the process agents use in “putting together” the various components—and thus into the psychological dynamics of acting out of virtue. The exercises suggest that agents can actively (though differently) match personal characteristics and situational descriptions. By “actively,” we mean that this is a task that participants are able to do, but they do so differently. That they are all able to do it suggests (at the very least) that such an exercise more closely approximates their actual practical reasoning than does the construction of syllogisms! The exercises also suggest that agents have working conceptions not only of situations, but of the self, conceptions which affect how they respond to situations. 64
The Logic of Virtue: Related to Situation, Relevant to Self
How does this matter for modeling the “logic” of virtues? We suggest this model highlights two key interactions that ethicists should attend to in understanding virtuous practical rationality: a virtue must be construed by the agent in relation to an appraisal of a situation, and the virtue must be understood as relevant to the self. In both directions—toward the situation and toward the self—there are pieces of knowledge and related appraisals that can be used to explain the agent’s characteristic understanding of particular actions. The consistency of both of these interactions, held together by the agent in terms of a language of virtue, is what we would look for when we say that a person has a particular disposition. As Daniel Lapsley has put it, “moral identity” has a “dispositional signature” which is properly “located at the intersection of person–x–context interactions.” 65
Let us consider an example of these interactions: imagine two faculty members, both of whom see it as important to carry out their teaching responsibilities with “integrity.” Each might be experienced and careful, possessing sufficient knowledge to understand what they are doing—yet their construal of what sorts of actions and choices “integrity” calls for might differ. An instance one of us actually encountered involved two unquestionably committed teachers, one of whom believed that part of that commitment involved working on writing skills in discipline-specific courses, while the other saw this as something that a shared liberal arts curriculum was supposed to address, so that disciplinary courses could devote as much energy as possible to the discipline. It would not be right to say that one has “integrity” and the other does not; instead, the construal of integrity in relation to their respective appraisals of particular situations is where we should look to understand their different actions. Further, notice that the teacher who “doesn’t work on writing” might appear the same as a third faculty member who “doesn’t work on writing” out of laziness and for whom the virtue “integrity” is not self-relevant. For virtuous action to happen, you need two connections made: from situation to an understanding of the virtuous characteristic, and from an understanding of the virtuous characteristic to some conception of its place in one’s sense of self. In effect, the person must be able to say (even if they do not do so at the time of action) both “I have (or want to have) this virtue,” and “This virtue applies to my understanding of this situation.”
The Interactional Model in Relation to the Standard Model
Having developed the KAPA model as an explanatory one that is fine-grained enough to explain differences between agents performing externally similar particular actions and can also explain the dynamics that underlie agents’ consistency across actions, we can return to the question of how this approach enhances the standard model. The point is not to force a choice between the interactional model and the standard model, but to suggest targeted revisions of language that improve explanatory power. We highlight three contributions: an avoidance of a reason–passion dichotomy, an expansion of how we might understand practical reason, and an interesting explanation of what “willing” might look like in the virtuous agent. These contributions are broadly outlined here, and are meant as invitations to further development, rather than exhaustive treatments of the integrations.
KAPA and the Passions
First, the use of KAPA avoids the rigidification that plagued the later development of Aquinas’ faculty theory, especially by complexifying the reason–feeling relationship. As we have noted above, a knowledge-appraisal framework for understanding the mechanics of acting from character avoids parceling out discrete territory to “reason” and “emotion,” because it is less interested in separating interior “parts” and more interested in interior “interactions.” In appraisal theories of emotions, reason can give rise to bodily preparations for action (and the “feelings” associated with these preparations). Separating the reasoning from the feeling misses something fundamental. 66
But what, exactly? After all, the above idea of reason “preparing” for action may sound like what Aquinas means by a “consequent passion.” Yet, even beyond the still-debated language of “consequent passion,” the notion of appraisal offers interactionally insightful advantages. 67 One, it points specifically to the situation–self interaction; appraisal includes a notion of “relevance to self” that seems difficult to place in the standard model, but which is especially helpful in explaining why morally weighty emotional reactions may vary so widely. Two, by rejecting a reason–feeling territorial dichotomization, the notion supports Daniel Russell’s argument for a “unity of the virtues” that “depends on making phronesis . . . a central feature of every virtue.” 68 Russell’s articulation of this unity differs from the standard model. He rejects a solution to what he calls “the enumeration problem” of cardinality in terms of connecting specific virtues with specific “powers.” That is, he does not parcel out virtues by making discreet connections to specific and separate faculties, like reason or the passions. Instead, all virtues are individuated “according to the particular kinds of reasons to which each virtue is responsive.” 69 For example, in a social cognitive framework, fortitude need not be understood in terms of a virtue shaping a set of irascible passions, but as a cluster of related reasons for action that are connected by the agent to their self-understanding of their identity and to appraised aspects of situations. The virtue of practical wisdom is not explained by identifying it with a “distinct faculty” which appears to manage things from the outside—as in the “charioteer” model of prudence 70 —but instead as an internal “logic” of the cluster of cognitive-affective notions associated with any virtue.
KAPA and Practical Rationality
Second, the interaction of knowledge and appraisal avoids narrating “practical reason” in an overly narrow way. The reasoning that the different professors described above use to guide their action is not a matter of mere means–end calculative deliberation, and even less a matter of a deduction from axiomatic first principles. This richer account of reasoning is actually much closer to Aquinas than the deductivist approach of reason in the moral manuals. For example, Porter explains that Aquinas’ understanding of the “object” or “matter” of the virtue of justice involves a specification that looks much like Russell’s differentiation based on “particular kinds of reasons.” She suggests that, in the II–II, Aquinas offers “an extended account of the virtues as substantive normative ideals.” 71 This account, far from deductions from first principles, is constructed through selective interpretations of “traditional formulae” that allow for a statement of “the characteristic kind of action through which the virtue is expressed.” That is to say, along the lines suggested by more narrative accounts of the virtues, the “reasoning” involved in moving from virtues to specific moral issues is cognitive, but not purely formal or deductive; rather it is one that assembles a vision of the virtue through “stock images” and “paradigmatic kinds of actions.” 72 This process of moral reflection through the assessment and reassessment of paradigmatic examples can be understood in terms of the ongoing person/context encoding and construction described above. Moreover, it fits well the emerging consensus in philosophical ethics that possessing virtue involves “the wholehearted acceptance of a distinctive range of considerations as reasons for action” that accompany “the capacity to recognize, in any particular situation, those features of it that are morally salient.” 73 Notice how this definition uses terms like “considerations” and “salience” that evoke a more “sapiential” and less deductive account of practical reason.
Such an expansion of our understanding of moral reasoning has several promising implications for connecting virtue theory to other insights in contemporary moral theology, ones we can only mention in passing here. For example, it offers a richer way of integrating the vexed category of “subjective experience” into practical rationality, without appealing to it as an “alternative” to “reason.” Consider Margaret Farley’s treatment of experience as a source for ethics: she understands the problems of an appeal to “direct” experience, yet she concludes that experience must be seen as “given but not primitive, immediate but not innocent of interpretation, personal but not isolated, unique but not without a social matrix.” 74 This description comes perilously close to embracing contradictions; a social cognitive approach provides an effective analytic tool for understanding both the real impact of experience and yet also recognizing the already-existing processing systems by which the interaction of person and situation is “experienced.” Another connection would be incorporating exemplars in moral reasoning, especially the saints, an approach Patrick Clark explains by noting “virtues emerge only against a given horizon of the human good held in place by individual exemplars.” 75 This “irreducibility of exemplars” is their “essential” role in understanding how to enact virtues in various changing circumstances and cases, 76 fitting well with a cognitive-affective model requiring ongoing assessments of a relation of self to distinct situations. Note here Bandura’s original innovation was his emphasis on learning by observation of others. 77
KAPA and the Will
What of the third component of the standard model, the will? Careful readers will have noted its absence. Aquinas understands the will as “rational appetite.” Aquinas’ general theory of “appetition” explains why anything is moved “internally.” All animals (including humans) have sense-based “appetites,” but humans have an additional appetite, a “rational” one, whereby the agent moves toward the good not simply by a sense response, but by means of an intellectual grasp of the good as the good. 78 The agent is able to understand their object “under some universal formality of goodness,” 79 and not simply as “this particular good thing.”
What can psychology contribute here? In a social cognitive model, will would denote cognitive-affective schemas that include some effective reference to human goodness per se. Interestingly, Cervone’s model already includes a stand-in for this: “evaluative standards.” In contrast to “beliefs,” which “fit” the mind to the world, and “goals,” which seek to “fit” the world to the mind, Cervone suggests a third “analytically distinct” set of “cognitions [that] are criteria for judging the goodness and worth of an entity.” 80 While it might be suggested that such standards necessarily interact with the other two categories (e.g. one’s moral standards can and do rest on truths about the world), Cervone’s model incorporates the fact that just because something “is” does not mean it “should be” (we criticize the status quo all the time), but also that a person having a particular sincerely held goal does not immunize that goal from evaluation (it could be “wrong,” and one could come to discover that oneself). This suggests an interesting aspect of appraisal theories of emotion. One might appraise the goals one has, have emotions about those goals, and, as a consequence, have behavior arising from the appraisals of the goals. For instance, someone with the goal of watching a game might experience anger as a result of the judgment that someone else’s request of their time blocks that goal. But our fan might then judge that in the circumstance, watching the game is not consistent with “the good,” experience remorse about the anger and placing precedence on watching the game instead of the other’s needs, and change goals to a desire to help the other.
Even more interesting is Cervone’s recent collaboration with Brian Little, a psychologist interested in “personal project theory.” Personal projects are understood as “extended sets of personally salient action in context” that are “both embodied in action and embedded in a set of contextual affordances and constraints that can facilitate or frustrate their pursuit.” 81 The authors reference philosopher Bernard Williams’s similar concept of “ground projects” that are “the pursuits that if no longer viable would lead one to question whether it is worth going on at all.” 82 The viability of such projects is constantly assessed in terms of both “inner” and “outer” sustainability—the extent to which the project’s demands are compatible with aspects of the self and the extent to which the projects can be realized in one’s life situations. But what is most interesting about this work is the empirical evidence that sustaining such projects is essential for flourishing (at least as psychologists have construed flourishing); they function thus as a kind of rationally grasped overall “good.” Cervone and Little themselves note that a synthesis of KAPA and personal-project analysis allows for a “bridge between research on contextualized social-cognitive mechanisms and questions of higher order values, identity, and personal meaning.” 83 This is, we think, another way of saying that the synthesis allows us to understand the connections between the micro-dynamics of virtuous action and macro-questions about “the good.”
Thus, “having a will” in this model means having a comprehensive and personally compelling sense of the good, whether that be understood as some internal set of standards or some overall sense of life project. Possessing virtue would mean that these standards/projects are chronically accessible, 84 serving as genuine drivers of action.
The Revised Model and the Adjudication of Moral Disagreement
In conclusion, we want to return to the context that drove our search for a better, more precise modeling of the psychology of virtue: helping us better understand particular moral disagreements. It is beyond the scope of this article to apply the model. We also do not mean to suggest that KAPA can provide “solutions” to such disagreement. Instead, we briefly suggest how KAPA can offer a more careful analysis of what is going on in moral disagreement, especially when we have two reasonably conscientious Christian disputants on an issue, and both appeal to the same virtue—say, justice or compassion.
KAPA gives us some intelligible questions to ask regarding how each is making the connection from the virtue to the particular action, identifying more carefully where the disagreement lies. Analogously to our earlier example of professors’ sense of “integrity,” we could recognize that a sense of “sexual integrity” might be at work in different and potentially conflicting ways in disagreements over sexual choices, or a sense of “economic fairness” at work in two people with strongly differing views of particular economic choices. In some cases, we may discover that the conflict is really a battle between virtue and vice. But many of our disagreements are not really like that—and we cause considerable damage if we analyze them as if they are. 85 While such a recognition certainly doesn’t mean that both are correct, it does help us engage their actual cognitive-affective processes more accurately than one which simply dichotomizes between virtue and opposing vice.
Moreover, KAPA assessment offers empirical tools by which we might examine how such different conceptions of virtue operate—specifically, what interactions lie at the heart of the disagreement. This has importance for scholars of virtue ethics, insofar as they follow what David Elliot recently called the “turn to classification,” in virtue ethics, the arrival of studies focused on thick descriptions of virtues and vices themselves. 86 Even more so, such exploration is valuable pastorally, because it should be evident that disagreements among ordinary Catholics about, say, contraception or remarriage do not typically hang on anyone’s mastery of the technical arguments. Instead, moral theology needs a better understanding of how people are actually working these problems out from the agent’s own perspective.
Finally, the model highlights how an agent’s perspective involves the central place of chronically accessible virtue-based self-schemas whenever they are exercising prudential judgment about a particular issue. We are echoing what Julia Annas has defended as the “requirement of articulacy” for the possession of virtue. 87 Psychology gives us some interesting (and sobering) insight into how this works: the ubiquitous place of “dual-process models” in psychology suggests the ongoing effectiveness of self-schemas is almost certainly a matter of developing an automaticity of access, 88 but that such a development will take concentrated and sustained practice. 89 Moreover, especially in working through challenging or novel situations, the agent has limited working memory to do the active work of reconstructing and sustaining the development of multiple schemas. And pluralistic societies will increase the demands on the agent for these complex processes—perhaps leading to a richer sense of possibility for those who can handle the demands, but also potentially leading to the use of shortcut heuristics to resolve these demands. Thus, having what we described above as “a good will”—a chronically accessible, virtue-laden account of the good in general that is in constant interaction with the self and situations—appears crucial in understanding the “logic” of disagreements between virtuous agents.
At the outset of the essay, we suggested that an integration of social cognitive theory with virtue theory could offer promising insights into the practical reasoning of ordinary agents. The urgency of this task is not only to offer more precision in our analysis of what a virtue is, but more importantly, to enable an ethic of the virtues to connect more clearly and precisely to moral disagreements over particular actions. Social cognitive theory helps us name the specific work that agents might do in responding to the world they encounter, and how that work might come to be “virtuous”—that is, result in consistently excellent characteristic activity. Now the real work awaits: to apply this model to the specific areas of disagreement, in hopes that we can better understand each other and advance together toward the good.
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work was funded by a grant from the Happiness and Well-Being Project at Saint Louis University, which was funded by the John Templeton Foundation.