Corps de l’article

In his recent paper, Carr (2023) relies on an oddly limiting account of moral education based on behavioural discipline, and the narrative force of fictional novels and other stories, while launching an attack on the emphasis on role modelling as a deliberate educational strategy. He suggests that although role modelling might be inevitable throughout the course of moral education and human life, it could only be considered an auxiliary phenomenon rather than a reliable educational method. Carr is concerned about the possibility of indoctrination, the difficulty in generalizing the method across different students, and potential risks involving imperfect or negative models. 

Carr’s conceptualization of role modelling views professionals, for example, teachers or ministers, as responsible for taking on a modelling role concerning those they serve. While this may be the case, it is not the only source of role models. Parents, peers, friends, and fictional characters are all possible sources of role models (Annas, 2008; Kristjánsson, 2007, 2020; Sanderse, 2012; Sherman, 1987) that are not acknowledged by Carr. 

We see no reason to accept such a narrow conception of role modelling as the one suggested by Carr, but even if we did, an aspect of greater concern is his juxtaposition of an agent’s free personal choice with what he claims to be the “deliberate shaping, manipulation or coercion” (Carr, 2023, p. 71) involved in role modelling. As evidence of this understanding of role modelling, Carr discusses schools and their emphasis on discipline and rule-following. However, this conception of role modelling is neither theoretically sound nor supported by empirical evidence. Role modelling can include numerous expressions of an independent agency, from the recognition that growth is necessary, to the choice of the suitable role model, to a conception of what counts as a suitable role model, to an ongoing critical evaluation of the role model and their contribution to the agent’s education (Athanassoulis, 2022). Carr views role modelling as divorced from a normative perspective. Nevertheless, we have no reason to agree with him – the very choice to engage in modelling a particular person is normative.

In addition, recent research has pointed to the importance of autonomy and self-cultivation within moral education employing exemplars. Sanderse (revised and resubmitted) argues that teachers need to focus on helping students identify and learn from their moral role models rather than demanding that they emulate them. Researchers interested in moral role modelling are aware of the idea that the mechanism of moral role modelling should focus on students’ self-cultivation instead of passive reception and imitation of presented moral values and behaviours. In addition, the appropriate ultimate goal of moral education via role modelling, from the perspective of researchers, is the promotion of autonomy and self-identity, not indoctrination or unidirectional instillation of values and beliefs.

Furthermore, from a methodological perspective, we can also refer to previous research underscoring the use of autonomous instructional methods, not indoctrination, in moral role modelling. A recent empirical study by Han and Graham (2023) supports the idea that autonomy-supported aspects should be considered in moral education via moral role modelling. In the study, they synthesized data collected from previous exemplar intervention studies. They found that motivational and behavioural outcomes – that is, emulating presented exemplary behaviour – were maximized when presented exemplars were perceived as relatable and attainable, and when autonomy-supporting instructional methods, such as discussion and crafting, were used. They connected this to the self-determination theory that underscores support for autonomy, competence, and relatedness in education for promoting intrinsic, rather than extrinsic, motivation in the long run. Han and Graham (2023) demonstrate that students’ sense of autonomy should be supported while they are being provided moral exemplars. In addition to what Sanderse (2023) argues about self-cultivation being the goal of role modelling, Han and Graham (2023) suggest that the educational methodology should also be autonomous.

Effective role modelling via moral exemplars can only be done through autonomous instructional methods and by aiming at self-cultivation. In terms of this aim, indoctrination should not be pursued in moral role modelling, as discussed by Sanderse (2023). Furthermore, educational methods that make use of indoctrination cannot be effective in the application of moral exemplars, as shown by Han and Graham (2023). Therefore, it is clear that rather than being inimical to role modelling, self-reflection, deliberation, and endorsement are an essential part of it, and, contrary to Carr, indoctrination is not an inherent part of moral role modelling.

Carr is also concerned about the dangers of falling prey to the wrong role model. However, again, this is only a concern if we view role modelling as a blind process on the part of the student. Role modelling does not occur in a vacuum. Instead, it is supported by an appropriate upbringing, positive early influences, and the ability to slowly discover the direction one should be going in, all of which need to be in place before one chooses who will light the way on this path. Even if one were to choose unwisely, this in itself could prove to be an educational lesson. Imperfect role models may do an excellent job of enlightening imperfect students – for example, in the ways of resisting temptation, or in being a student of virtue rather than a perfectly virtuous agent (Athanassoulis, 2022). Even a vicious role model may have an educational role as an example of what not to do (Lockwood et al., 2002). Consistently, recent research on moral exemplars and role modelling has shown that imperfect exemplars, who are perceived to be more attainable and relatable compared with perfect moral saints, are more effective in promoting moral and prosocial motivation (Han & Dawson, 2023; Han & Graham, 2023; Monin, 2007).

Finally, we argue that moral exemplars possess unique values for cultivating virtues, specifically practical wisdom, that is, phronesis, which Carr consistently underestimates. Carr argues, “It is evident that both young and old often come to comprehend and/or acquire virtuous qualities in fairly diverse ways – including verbal or written moral explanations – that do not at all depend on role modelling or emulation of others” (Carr, p. 74). However, recent philosophical and psychological works addressing phronesis suggest that moral virtues and strengths function in a complicated cooperative manner, so using moral exemplars as holistic sources for moral education is recommended. In philosophical papers proposing the standard models of phronesis, Darnell et al. (2022) and Kristjánsson and Fowers (2022) argue that phronesis is a multifaceted virtue constituted by multiple functional components, and their interaction and coordination are also fundamental aspects. Moreover, Han (2024) has demonstrated that functional constructs in moral psychology – for example, reasoning, empathy, and identity – create a network structure, meaning the robustness of the network significantly predicts moral motivation and behaviour.

If the exercise of moral virtues with phronesis is structurally and functionally complicated, those using educational methods to cultivate them might end up explaining these methods insufficiently, without considering their inherent complexity. In light of this, moral educators require moral exemplars, who can demonstrate optimal or vicious moral functioning in a holistic way. In the cases of virtuous moral models, the exemplars are more likely to possess more sophisticated individual functions, such as reasoning, empathy, and identity, and demonstrate improved coordination among the functions compared with other people (Kristjánsson & Fowers, 2022). On the other hand, a network of suboptimal moral functioning might exist among vicious exemplars. Such exemplars can, inadvertently, present an example of the functional network of anti-morality that students might need to avoid. For instance, recent studies in clinical neuroscience report on the aberrant functional networks among brain circuitries associated with moral cognition, emotion, and motivation among psychopaths, white-collar criminals, and other types of people demonstrating anti-morality, without any intellectual or other cognitive issues, or even with above-average mental skills (Raine, 2019; Raine et al., 2012).

According to network science, a network consisting of connections between nodes possesses significantly more information than a mere sum of the nodes (Klein & Hoel, 2020). Connectivity, which represents interactions between nodes, in such a network provides unique additional information to the network beyond what can be explained by the components per se (see Seth & Bayne [2022] for a theoretical review). Given what has been suggested by network science, it would also be plausible to assume that moral exemplars present unique strengths and qualities emerging from the network of virtues and functionings. Hence, teaching individual moral skills, which Carr proposed without displaying concrete examples of moral functioning, either optimal or vicious via presenting networks in a holistic manner, is insufficient for promoting virtue development (see Damon & Colby, 2015). Emulating presented moral exemplars would be an irreplaceable and fundamental way for students to exercise, adjust, and fine-tune the network aspects of a virtuous person (see Han, 2023a), especially those constituting phronesis, such as coordinating individual virtues and functions, in moral education.

In conclusion, we suggest that Carr’s concerns regarding using models in moral education can be well addressed through updates in recent research. Moral educators can employ moral exemplars, including imperfect and even negative ones, as sources for moral development with autonomy-supporting instructional methods and self-cultivation. Furthermore, we argue that moral exemplars as moral paragons who can demonstrate optimal moral functioning in an integrative manner can be unique sources for cultivation of phronesis. Given all of this, moral exemplars in general should be considered irreplaceable sources for moral education.