Introduction: is phronesis a developmentally inadequate construct?

The most uniquely identity-conferring feature of Aristotelian and neo-Aristotelian forms of character education – understood philosophically as the educational incarnation of virtue ethics (Annas 2011) – is the invocation of phronesis or practical wisdom as an intellectual meta-virtue that orchestrates the moral and civic virtues and ultimately guides all mature decision-making. This uniqueness is best seen by comparing Aristotelian character education with other philosophically inspired forms, such as Platonic or Thomistic that, contra-Aristotle, both rely on a specific master virtue among the moral virtues trumping others (justice in the former and love as agape in the latter), or with the reigning social scientific virtue paradigm of the day, positive psychology (Peterson and Seligman 2004), which makes do without either a master or a meta-virtue for adjudication.Footnote 1

For some reason, this prominence of phronesis has not – until very recently at least – been reflected in the vast practical literature on character development and education; and pre-2020 no instrument to evaluate phronesis progress existed that could be used for pre-and-post testing of educational interventions (Darnell et al. 2022). At best, educated guesses can be offered about this lacuna. First, much more teaching materials have been produced for the primary and lower secondary levels of the school system than for the upper secondary and early-college levels where phronesis development is usually considered to emerge. This could be inter alia because it is easier to focus on cultivating individual virtues, such as gratitude and compassion, if the thorny issue of virtue conflicts can be avoided. Second, most of the copious professional ethics literature on phronesis has not been about the standard Aristotelian concept but rather, on the one hand, MacIntyrean (MacIntyre 1981) and, on the other, intuitionist-cum-postmodern (e.g. Kemmis 2012) versions of it. However, both of those concepts happen to prioritise other problematics over developmental ones. Third, as outspoken as Aristotle himself was about methods of early-years moral education, especially habituation and role-modelling, he was singularly unhelpful in delineating useful strategies for phronesis development: to the extent of simply trading in platitudes. For example, it is not very illuminating to assert that phronesis is developed through ‘teaching and experience’ (Aristotle 1985: 33 [1103a14–16]); we want to know what kind of teaching? – what sort of experience?

In the last couple of years, various factors have contributed to a heightened interest in the developmental and educational conundrums surrounding phronesis (Kristjánsson et al. 2021). The most significant of those is perhaps the recent turn in traditional wisdom research in psychology towards a practical, as distinct from a theoretical, concept of wisdom (Grossmann et al. 2020), which has brought the developmental wisdom discourse in psychology into better alignment with the standard character-education literature (Huynh and Grossmann 2020). This practical turn has then been aided by a more general burgeoning of interest in character-and-virtue research within psychology (Fowers et al. 2021; Wright et al. 2021). At the same time, educational philosophers have continued to explore the concept, using the tools of their trade (Harðarson 2019; Burbules 2019). Yet, at the end of all of this, when an attempt is made to collate what we actually know about phronesis development and education, what emerges is at best a long series of received wisdoms, assumptions and hypotheses,Footnote 2 mostly yet-to-be-confirmed empirically (Kristjánsson 2021a).Footnote 3

Psychologist Dan Lapsley (2021) has argued recently that the dearth of any empirically credible developmental account of phronesis is symptomatic of a ‘developmental inadequacy’ in Aristotelian character theory that makes it surplus to requirements for contemporary developmental science. Meanwhile, Silverstein and Trombetti (2013) think that Aristotle has a rich developmental account that urgently needs to be mined. The truth probably lies somewhere in a ‘golden mean’ between these two suggestions. On the one hand, it is important to mine and reconstruct what the academic father of phronesis had to say about its development, and to subject it to scrutiny; on the other hand, it is equally important to venture beyond Aristotle’s texts and to explore what insights can be added to his developmental story in the light of contemporary research. Those are, indeed, the twin aims of this article.

I begin, in the section ‘Developing towards virtue: mining and transcending Aristotle’, by piecing together the developmental trajectory that Aristotle suggests towards full virtue, up to and including phronesis development. I also briefly list ten potential weaknesses of this picture, considerations of (some of) which constitute the bread and butter of current developmental science. In the section ‘The functions of Phronesis, and what can be achieved through teaching’, I present a reconstructed Aristotelian model of phronesis and discuss how the teaching element of phronesis education could be executed, with an illustration from an ongoing phronesis intervention for UK police-science students. In the section ‘Teaching about “What Happens to the Heart”’, however, I dampen somewhat the enthusiasm about this ‘taught’ component, by explaining how relatively small the ‘zone of proximal development’ is that can be targeted by scaffolded teaching in this area. Finally, the section ‘Concluding remarks’ elicits some educational implications of the conclusion that most of phronesis development will need to be ‘caught’ from gradually unfolding personal and professional experiences. Some of those implications will most likely apply to all taught forms of advanced moral education, extending well beyond Aristotle and police ethics.

Developing towards virtue: mining and transcending Aristotle

To begin a study of how phronesis develops and the ways in which that development can be facilitated, the natural starting point is Aristotle himself, upon whom almost all contemporary explorations of phronesis draw. However, Aristotle obviously wrote his texts long before the advent of developmental psychology, and ideas such as the Kohlbergian one about distinct stages of moral development through which everyone needs to pass in a certain order were foreign to him. Aristotle does describe people at different levels of moral maturity and some scholars have tried to adapt those into an ‘Aristotelian stage theory’ (Curzer 2012; Sanderse 2015). I have come to believe, however, that Aristotle did not think everyone needed to pass through all the levels he describes and that he is, rather, presenting two possible but distinct routes to a fully virtuous life, as depicted in Fig. 1.

Fig. 1
figure 1

A Neo-Aristotelian Model of Moral Development (Jubilee Centre 2017)

This model includes two trajectories towards moral development. The upper trajectory in the model, which we could name Plan A, is for those fortunate enough to have been brought up by good people (as moral exemplars), exemplifying moral habits and endowed with sufficient material resources. Those fortunate children are the ones most amenable to moral development. They internalise moral habits by copying what they see being done by their role models, and gain virtue knowledge and understanding through both ‘caught’ and ‘taught’ methods, adapted to their temperamental dispositions.Footnote 4 Guided by emulated mentors, they become, step-by-step, ‘just by doing just actions,’ ‘brave by doing brave actions,’ etc. (Aristotle 1985: 34 [1103b1–2]). What is more, to draw on an analogy from the field of nutrition, they eat their ‘greens’ because they enjoy their taste; they do not need to force themselves to ‘eat’ the right things, and their emotions harmonise with their action choices. At some stage, then (Aristotle is mostly silent about when this happens, though one would presume in late adolescence and early adulthood), the young gradually begin to develop critical thinking and reflection and revisit critically the traits with which they were originally inculcated: subjecting their merely habituated virtues to revision. They now learn the value of moral goods ‘sought,’ in addition to simply being ‘caught’ and ‘taught,’ and advance towards the stage of full autonomous virtue, which Aristotle calls ‘phronetic’ (i.e., guided by the metacognitive capacity of phronesis). While some of their actions will be guided by externalist reasons (e.g., prudential motivations towards peace and sociality), most of their actions will be internally motivated by the conception they have developed of virtuous traits of character being constitutive of their identity: their second nature, so to speak. Some people – endowed with extraordinary personal strengths and/or spurred by unusual social circumstances – will progress even further than simply being phronimoi (persons with phronesis-infused virtues), towards the level of heroic virtue. Those are the Nelson Mandelas and Martin Luther Kings of this world – but Aristotle does consider heroic deeds supererogatory and not necessary for counting as fully virtuous.

The lower trajectory in the model, which we could name Plan B, is for those slightly less fortunate, brought up under more mixed moral conditions and hence less amenable, originally, to character-virtue development. Given that they will still have some moral exemplars in their environment to emulate – even if those happen to be outside of their immediate family – they will develop a conception of the morally good. However, because of the patchy ways in which this conception is strengthened via ‘caught’ methods, these children will arguably lack self-regulation. To return to the nutritional analogy, they may understand the value of eating their ‘greens,’ but they lack the self-control to do so as they do not really love the taste. This is the stage that Aristotle calls ‘incontinence’ (akrasia). Unfortunately, the majority of people, according to Aristotle’s fairly dim view of his contemporaries, stagnate at this stage, or between incontinence and the next stage of ‘continence’ (enkrateia) (1985: 190 [1150a15]). Through practical habituation – either motivated by friends/mentors or their own powers of insight – a significant group of people progress towards being well self-regulated morally; and that is a considerable moral achievement. However, it still falls short of full virtue because even if the continent now actually ‘eat their greens,’ they still do not particularly enjoy the experience. In other words, they have to force themselves to be good. What they end up doing may be behaviourally indistinguishable from the actions of the truly virtuous, but it is not phronetically motivated in the same way, but rather mostly instrumentally or extrinsically driven. Yet, some of the continent agents may succeed in climbing up to the level of full virtue (the upper Plan A-trajectory), especially if they are fortunate enough to be in the company of close friends occupying that level.

This ancient model of development towards virtue tallies surprisingly well with contemporary understandings of moral development (see e.g. Fowers et al. 2021), especially if one adds to it Aristotle’s more general remarks about the social context required for either of the above developmental trajectories to be open to moral leaners in the first place: remarks that undermine some current individualistic and politically conservative accounts of character development.Footnote 5 Nevertheless, as could be expected, many aspects of this model are subject to ongoing criticism, both from contemporary neo-Aristotelian and non-Aristotelian educators. Since many of those criticisms and concerns are relevant to the specific topic of this article – the development and education of phronesis – it is instructive to offer a brief list of them here.

First, Aristotle assumes that children are born morally neutral (i.e., without any budding dispositions to be either moral or immoral), meaning that moral development is fundamentally a result of upbringing and education. In contrast, quite a lot of current empirical evidence indicates that new-born children are endowed with empathy and even more advanced moral capacities, perhaps as an evolutionary adaptation (Hoffman 2000). Aristotle makes the much more modest claim that children are born with the potential to become moral, in the sense of possessing capacities to internalise moral traits, but that ‘none of the virtues of character arises in us naturally’ (1985: 33 [1103a17–19]).

Second, another controversial assumption is Aristotle’s ‘early-years determinism’ about moral development, suggesting that children brought up under terrible conditions and not receiving any significant moral stimulation in their early years will never progress towards virtue at all (1985: 292 [1179b11–31]; cf. Kristjánsson 2015: chap. 5). Notice that there is no ‘Plan C’ in the above model for those children. Again, there is considerable evidence indicating that Aristotle may have been wrong about this and that radical moral conversions, if rare, are possible later in life, even for those not brought up in a minimal sense as moral agents (Kristjánsson 2020: chap. 6).

Third, and related to the second worry, Aristotle does not consider ‘moral elevation’ – understood as attraction to abstract moral ideals – as a feasible source of moral motivation. Intent on killing off the idealism of his mentor Plato, and only wanting to acknowledge emulation of the moral characteristics of persons (role models, mentors, close friends) rather than high-minded ideals, Aristotle’s developmental theory runs the danger of appearing disenchanted and deflated (or at least too cheaply practical) regarding possible sources of moral awe and inspiration (Kristjánsson 2020: chaps 5 and 7).

Fourth, the ‘paradox of moral education’ that Peters (1981) famously identified looms large here: how one can be trained heteronomously to become autonomous. This paradox arises because the ‘caught’ early-habituation stage in Aristotle’s Plan A mainly seems to involve arational imitation and repetition only. How does one get from there to autonomously (i.e., phronetically) ‘sought’ moral goodness?

Fifth, a related worry is more about a gap than an explicit weakness in the model. Why does Aristotle say so little about the proper timing of, and best methods for, phronesis education – apart from the previously mentioned platitudes about learning it through teaching and experience? Most neo-Aristotelians take it for granted that he is talking about late adolescence–early adulthood and that the ‘teaching’ relates to critical discussions with mentors and peers (as ‘character friends’). But those assumptions must largely be read into Aristotle’s texts.

Sixth, Aristotle seems to be overly demanding about the intellectual nature of moral decision-making post-habituation. Doing the right thing does not have any moral value anymore unless it is done for the right (phronetic) reasons and from the right motives (1985: 40 [1105a30–34]). Mere prosociality (good social outcome) does not seem to matter at all. So Aristotle appears to move here between two extremes: from a habituation period that is all about socialisation and correct habit-formation towards a period of complete autonomy where habituation and socialisation play no role anymore (yet contrast Sanderse 2020).Footnote 6

Seventh, Aristotle – despite his aversion to idealism – seems to have an overly idealised view of the psychological unity of the fully virtuous person, and to lack a sense of the residues of pain that may remain even after a fully phronetic decision has been reached: say, the pain that Sophie (in Sophie’s Choice) continued to experience after choosing her son (for good reasons) over her daughter in the Nazi concentration camp (Kristjánsson 2022).

Eighth, Aristotle does not explain what intellectual virtue the merely continent draw upon to solve conflicts and remain functionally self-controlled morally. It is not phronesis (which is only for the virtuous), and it is not mere calculation (deinotes), which is completely instrumental and amoral.

Ninth, Aristotle paints a rather disconcerting picture of moral heroes: those who go beyond full virtue in devoting their lives to philanthropy or public benefaction. People seem to be able to do this only at the expense of both the exhibition of mundane, everyday virtues and any deep engagement with aesthetic activities – being consigned to a life of self-sacrificial philistinism (Kristjánsson 2020: chap. 4).

Tenth, after extolling all the advantages of the phronetic life, Aristotle suddenly changes gear in Book 10 of the Nicomachean Ethics, now telling us that actually an even better life is one of mere contemplation, away from the cut and thrust of everyday activities (Kristjánsson 2015: chap. 5). Although it is possible to interpret him as saying that this sort of life is only possible once the phronetic life has already been achieved, most morally committed readers will find this message deflating and potentially counter-productive from the point of view of moral development and education. Does such development and education then only have value for the sake of something else that transcends it? For someone as practically minded as Aristotle, one would also presumably need to factor in opportunity costs and the law of diminishing returns for those aiming at this highest supra-moral level.

The functions of phronesis, and what can be achieved through teaching

Let us, in the remainder of this article, focus our attention on the phronesis-related weaknesses in Aristotle’s developmental model and what can be done to ameliorate them. It seems reasonable to suppose that the ‘experience’ required for phronesis development is experience of the sort of quandaries that phronesis is meant to solve: namely, by gradually becoming more adept at figuring out what to do about them and why. Obviously, those experiences cannot be induced artificially or pre-empted through any educational interventions,Footnote 7 so let us confine our attention to the ‘teaching’ element. What is it precisely that we can teach students that helps them build up this intellectual virtue?

To answer that question, we need to say a little bit more about what phronesis is (see further in Kristjánsson et al. 2021). As first elaborated upon historically by Aristotle, phronesis is considered a meta-virtue of holistic, integrative, contextual, practical reflection and adjudication about moral issues, leading to moral action. When an important decision is required, one can decide well or poorly about how to act. A perspicacious description of good decision-making about crucial moral issues therefore seems vital. As a ‘virtue’, phronesis refers to excellence in such decision-making. As a ‘meta-virtue’, it includes metacognitive considerations of the injunctions of different moral virtues, especially when those conflict, to reach a measured decision. Phronesis is metacognitive in that the phronimos (person endowed with phronesis) reflects on and evaluates his or her cognitions, emotions and actions in terms of their wisdom, desirability and harmony.

Neo-Aristotelians tend to agree that phronesis contains discrete components, performing a set of inter-related functions. While they do not always agree fully on the number and nature of those, I simply introduce here the reconstructed Aristotelian model that happens to have been chosen to ground an intervention for police-science students in the UK, to be discussed as a case in point below. This model includes the following four functions (Kristjánsson et al. 2021).

Constitutive function

Phronesis involves the cognitive ability to perceive the ethically salient aspects of a situation and to appreciate these as calling for specific kinds of responses. This ability can be cultivated and amounts to the capacity to ‘read’ a situation by seeing what is most important or central. We can also refer to this function as moral sensitivity.

Integrative function

Through phronesis, an individual integrates different components of a good life, via a process of checks and balances, especially in circumstances where different ethically salient considerations, or different kinds of virtues or values, appear to be in conflict and agents need to negotiate dilemmatic space.

Blueprint function

The integrative work of phronesis operates in conjunction with the agent’s overall understanding of the kinds of things that matter for a flourishing life: the agent’s own ethical identity, aims and aspirations, her understanding of what it takes to live and act well, and her need to live up to the standards that shape and are shaped by her understanding and experience of what matters in life. This amounts to a blueprint of flourishing.

Emotional regulative function

Individuals foster their emotional wellbeing through phronesis by bringing their emotional responses into line with their understandings of the ethically salient aspects of their situation, their judgment, and their recognition of what is at stake in the moment. We can also refer to this function as infusing emotion with reason.

As already indicated, professional ethics constitutes one of the growth areas in the recent resurgence of phronesis research. Phronesis is typically seen there as offering an antidote to a 20th-century deontologically motivated obsession with written codes and rules (Schwartz and Sharpe 2010). While professional codes of ethics tend to be essentially rule-based, the UK police Code of Ethics (College of Policing 2014) constitutes an exception. Although it employs the language of police ‘principles’, 8 of those 9 ‘principles’ happen to be virtues. Moreover, the Code produces a so-called National Decision Model for police officers that bears a striking resemblance to an Aristotelian model of phronesis. The Code is replete with warnings about a belief in the unproblematic codifiability of police activities, and it foregrounds what we could call the three D’s of phronesis – discernment, deliberation and discretion – although it does not refer directly to the concept of phronesis. Recent scandals in UK policing have prompted a renewed focus on the ethical basis of the police force, and a new report on character virtues (or a lack thereof) in policing in England and Wales has provided considerable enlightenment about the current state of play in this area (Kristjánsson et al. 2021).

To cut a long story short, the Report makes various recommendations about how police ethics education can be improved. Simultaneously, the Code of Ethics is undergoing a revision which – at least if the advice from the Report is heeded – may lead to its becoming even more explicitly virtue ethical and phronesis-focused than before. At the time of writing, the authors of the Report (including the present author) are trialling an intervention for police-science students at a number of UK universities, aimed at cultivating phronesis through teaching. The actual outcome of the intervention, and how it is assessed, will be addressed in a separate paper in due course. What matters for present purposes, however, are some of the considerations and reflections that went into preparing the lesson plans for the relevant students. Those will be used, in the remainder of this article, as a platform from which to view the possibilities – but, no less importantly, the limitations – of teaching as a method to facilitate advanced moral functioning, such as that embodied in the (presumed)Footnote 8 intellectual virtue of phronesis.

The first constraint that we encountered was that of time. The most we could negotiate with the universities in question was an intervention taking up 4 × 45 min of class time spread over 4–6 weeks.Footnote 9 Typically, ‘professional ethics’ was not a special subject in those police-science students’ timetables so we had to negotiate an entry into ‘related’ taught subjects in the curriculum (such as ‘Accountability’), where time is already limited. This may seem like a fortuitous and non-philosophically relevant consideration, but in the light of Aristotle’s own insistence that the development of good character is a practical rather than a theoretical enterprise, it is a highly relevant one. Given the way professional education is conducted, it is unrealistic to assume that the time allocated to direct teaching about phronesis will be anything other than strictly limited.Footnote 10 We also decided that, since we could not assume that the students had any background in virtue theory, we would need to devote the first class mostly to teaching towards what is nowadays referred to as ‘virtue literacy’: simply explaining to them what terms such as ‘character’, ‘virtue’, ‘virtue ethics’ and ‘phronesis’ mean, and how those might be related to the police Code of Ethics and the National Decision Model.

This left us with three classes for the ‘substantive’ part of the intervention. We decided to devote those mostly to a deep discussion of topical police dilemmas. The choice of those dilemmas was tricky – they would have to be relevant, realistic and relatable. However, we were aided here by the work of an Expert Panel who had already helped us create dilemmas for a previous study (Kristjánsson et al. 2021). Here is an example of one of the dilemmas chosen:

You work in a police response unit. A new female student officer of South Asian heritage joins your team under the degree holder’s entry programme. Whilst you are alone with a close and experienced white, male colleague, he refers to the new officer and comments, ‘I bet her parents are disappointed she's a copper. A Paki with a degree; there’s not many of them – she should have done Law and become a lawyer, or Finance and been an accountant.’ You have never previously heard your colleague express views like this and have worked alongside him for a number of years. What would you do?Footnote 11

Through a guided discussion, the students are asked to discuss and reach a conclusion about various questions, including: (1) Which virtues or values are competing here and steering the police officer (later: ‘you’) in different directions? (2) What are the pros and cons of each action option? (3) Is the police officer experiencing strong emotions prior to the decision? (4) If so, what are those emotions? (5) What should the police officer do, in your view?

At the close of the intervention, the students are asked to relate their answers to the police Code of Ethics and the National Decision Model – as well as reflecting further upon how those texts are related to the phronesis model, and what the possible synergies of those might be. A post-test with the phronesis measure (Darnell et al. 2022) is then meant to gauge whether progress had been made during the intervention in phronetic decision-making (compared to a pre-test with the same measure), with respect to one or more of the components/functions of phronesis.

Teaching about ‘What Happens to the Heart’

The intervention, briefly sketched above, is no rocket science. The aim is, somewhat obviously, to help students develop the different components of phronesis in the Aristotelian model, by taking them through some of the considerations that motivate and (ideally) strengthen each component. The method of teaching is a guided discussion about relevant dilemmas: a method that has a long history in approaches to moral education as distinct as those of Kohlbergianism, neo-Kohlbergianism and virtue-based character education. Without wanting to underplay the potential strengths of this approach, which are fairly well documented within these three traditions (see e.g. Thoma et al. 2013 – and we would not have chosen it if we did not consider it valuable), I will in the remainder of this section focus on its limitations.

To couch the rationale of the intervention in a slightly more academic educational language, it is set within what we deem to be the police-science students’ ‘zone of proximal development’ (ZPD) as phronesis learners. In line with Vygotsky (1978), we understand that zone to sit between two other zones, of (1) what students can learn by themselves without going through the actual future experiences and (3) what students will have learned after going through the actual future experiences. The ZPD marks the in-between zone of (2) what students can learn prior to the actual experiences through ‘scaffolded teaching’ by a skilled tutor.

Here is the first problem. Policing is – along with professions such as medicine, nursing, teaching and the military – a burdened profession in the sense of one in which practitioners are likely to encounter various psychologically charged, and even life-changing, situations that are impossible to explain to students in sufficient depth before they encounter them. These are also professions with a high rate of burn-out, perhaps because of various factors that gradually seem to sap the practitioners’ original moral purpose in entering them (Arthur et al. 2021). Ideally, in order to elicit the necessary trust between the tutor and the student, the former should be a trusted mentor or a ‘character friend’ who has already gone through some of the experiences that are being related to students. Otherwise, there is danger that the ‘scaffolding’ effect will not be activated. In our case, we simply did not have access to police-science lecturers with sufficient knowledge of applied virtue ethics to steer a phronesis-intervention themselves. The tutors here are therefore academics without grounding in police science and without practical experience in working within a police force. That is a notable limitation – although, on the other hand, having the same tutors deliver the intervention across different cohorts/universities offers a layer of methodological robustness.

The second problem is that the dilemmas presented to the students involve experiences that are in a fundamental sense embodied. I am not using the term here in any obscure philosophical sense, but simply as referring to the fact that the experiential context will involve physical processes and feelings as well as mental reflection. To give an analogy, it is almost impossible to explain to a young child what sexual jealousy feels like and how those feelings will affect moral decision-making once the relevant adolescent hormones have kicked in. The child will perhaps know what sibling jealousy feels like, and analogies can be drawn with those experiences; but they are not the same as the experiences of sexual jealousy. Similarly here, some of the dilemmas presented to the police-science students involve situations that are bound to elicit strong physical and emotional reactions – but ones which cannot be known ‘in one’s skin’ prior to the event.

In short, we are dealing here with a ZPD that is severely circumscribed by the fact that the situations for which the students are being prepared are experientially conditioned and embodied. All that can be achieved within the ZPD is an intellectual exercise that may, at best, stimulate certain discrete components of phronesis but can only partially account for the context in which the eventual decision will be set.Footnote 12 If we venture further than that, in attempting to expand the ZPD, two perils await us. One is developmental naivety, in which complex experiences are reduced to an intellectual exercise in an attempt to articulate something that is inarticulable out of context – possibly inducing the infamous Dunning-Kruger effect.Footnote 13 The second is paternalism, in which we cavalierly ignore the students’ need to engage in their own Millian ‘experiments in living’ prior to becoming capable of making autonomous moral decisions, be those professional or personal.

That said, the temptation is very strong to expand the ZPD, especially if the tutors have gone through some of those experiences themselves and perhaps made mistakes that they want to pre-empt in students. The educational dilemma created here is a well-known one,Footnote 14 with implications far beyond any interventions to cultivate Aristotelian phronesis. On the one hand, we may have tutors who know in their own skin what typically ‘happens to the heart’ in the relevant professionFootnote 15; on the other hand, we have budding professionals who have not gone through those experiences and are full of idealism about their future work. The tutors do not want to curb the students’ idealism; but they also want to convey to them a sense of the challenges ahead. One of Leonard Cohen’s most compelling lyrics (from the song ‘Happens to the heart’, released posthumously in 2018) is about this very dilemma – although it is set in a personal rather than a professional context. The song ponders the question of whether we should ‘tell the young messiah / what happens to the heart’. Cohen charts his youthful exuberance in ‘meeting Christ and reading Marx’. However, he describes the perils he encountered in trying to ‘double park’, namely pursue two incompatible values (in his case, women!) at the same time, and how that left ‘an ugly mark’ on his heart. Gradually, he became more aware of how the exigencies and vicissitudes of life undermine one’s unity and how such experiences stifle and stunt: ‘It ain’t pretty, it ain’t subtle / What happens to the heart’.

The educational message of this song is clear, even if implicit – and it was made even clearer in the evocative video that was produced to accompany its release in 2018, two years after Cohen’s death. The video, however, suggests a deus ex machina: a Buddhist elevation where the ‘young messiah’ gains insight through a divine intervention, rather than the sort of mundane academic intervention that I have been describing in this article. However, there is nothing in the lyrics to suggest such an outcome. Rather, it remains an open question how much one can or should tell the young apprentice beforehand about ‘what happens to the heart’.

Concluding remarks

Contrary to what Aristotle told us, direct teaching is probably quite a useful – even if limited – method of character education in early age. For example, teaching young children what the words ‘gratitude’ and ‘grateful’ mean – and how they can be distinguished from words such as ‘appreciation’ and ‘appreciative’ – almost certainly helps them to understand the nuances of the terms and (possibly) to internalise the relevant virtuous emotions in their proper instantiations (cf. Gulliford 2018). This article has, however, focused on the age of early adulthood where Aristotle himself presumably saw the greatest need for teaching – namely as a stepping stone towards phronesis – although he failed to tell us what form such teaching should take.

Drawing on an Aristotelian model of phronesis, I have described an attempt to prepare young police-science students for the vagaries of police work through a short course on phronetic decision-making: a course attempting to hit a Vygotskyan zone of proximal development, paving the ground for the embodied experiences that the students are likely to encounter as police officers. I have aimed in this article at dampening enthusiasm about the effectiveness of any such intervention by explaining how tantalisingly small this growth zone will be. That should not surprise those who have studied the more general development of mature adult thinking in the professional realm, of which phronesis development is simply the morally informed incarnation.Footnote 16 Those studies tend to revolve around constructs such as ‘metacognition’, ‘identity formation’, ‘professional reflection’, ‘skill acquisition’, ‘expertise’ and ‘tacit knowledge’ (see e.g. Kallio 2020), and the growth of all of these is considered to be fundamentally experiential and sluggishly cumulative. It would be a miracle if phronesis development presented any radically different features.

Even the two most vocal champions of phronesis as part of professional ethics education claim that it ‘is not something that can be taught’ (Schwartz and Sharpe 2010: 271) – although they probably understand the term ‘teaching’ more narrowly in this context than Aristotle did. While I would not go as far as Schwartz and Sharpe, it is worth reminding readers of the well-known Chinese fable of the farmer who impatiently tried to pull up his rice shoots to make then grow faster, as a result of which they lost their rootedness and withered away. Young police-science students, for instance, need to be fed a diet that does not exhaust their capacities for digestion – which is not the same as saying that they should not be provided with an intellectual initiation into some of the tough and discretionary choices that await them, and with a stark warning that no police rule book will relieve them of the responsibility for making those choices themselves in the line of duty.

Just as the discourse on another of Aristotle’s core concepts, that of flourishing (eudaimonia), carries with it a risk of blandness and banality if the underlying variables are not populated with sufficient specificity (Kristjánsson 2020; cf. Carr 2021), the phronesis discourse loses practical resonance when it is pitched at too high a level of philosophical abstractness. I agree with Lapsley that no philosophical concept in the area of moral psychology and education ‘is completely understood until the process by which it attains mature form is mapped out’ (2021: 138). What Lapsley and I disagree about is whether Aristotelian phronesis satisfies this ‘principle of developmental adequacy’. I would argue to have shown above both how phronesis fits into a plausible trajectory of moral development (in the section ‘Developing towards virtue: mining and transcending Aristotle’) and how its underlying logical and developmental components can be unpacked (in the section ‘The functions of phronesis, and what can be achieved through teaching’). What remained, however (in the sect. ‘Teaching about “What Happens to the Heart”’), was a certain scepticism about the extent of the role that teaching can play in the development of phronesis, as illustrated with the example of police-science students.

If this scepticism is well-grounded, it carries implications that go beyond any educational theories of phronesis cultivation. What the two mutual historic nemeses of all moral education, Kohlberg and Aristotle, seem to agree about is the reduced role of the environment or ethos for development as the moral learner matures, and an increased role for (what Vygotskyans would call) scaffolded teaching and for autonomous decision-making.Footnote 17 In simple terms, caught elements of moral development are meant to be replaced with taught and, in particular, personally sought ones, as the students leaves behind the ‘courtyard’ of moral development and enters the ‘palace’ – to use Peters’s (1981) well-known metaphor.

In contrast, I hazard to conjecture that, at least in the area of professional ethics education, such education never ceases to rely heavily on situational cues and other uncodifiable and essentially unpredictable lessons picked up in the cauldron of professional work: an essentially unsystematic moral environment. However well students are prepared through taught lessons – and however well motivated they are to seek the good of their own accord through varied work experiences – norms and values caught on the job will remain a powerful source of moral motivation. Perhaps, rather than either ignoring the caught elements, or completely giving in to passive ‘moral situationism’ about them (Doris 2002), phronesis education needs to prepare students for critically evaluating those influences. In the context of police ethics education, for example, that means teaching students to be alert to norms issuing from the so-called ‘blue code’ (Westmarland and Rowe 2018) and to learn to subject those to scrutiny before they become an integral part of their moral identity.