The Trivium: Revisiting Ancient Strategies for Character Formation

The goal of this article is to examine virtue education tactics used in classical pedagogy, to provide a model for how virtue strategies can be applied as part of a coherent narrative of training, adjusted to the stages of emotional and intellectual development of the learner. By classical pedagogy, I mean the progymnas-mata, trivium, and quadrivium—the sequence of education rooted in the ancient paideia, refined by the Roman rhetoricians and the Scholastics, largely neglected for a time in favor of more practical modes of education, and revived with Dorothy Sayers’s 1948 essay, “The Lost Tools of Learning.” I describe the relevant features of classical pedagogy, and I focus on strategies related to the use of exemplars because imitation is central to this progression.

This article has two sections: (1) I first introduce the classical stages of education, focusing on the order in which virtue strategies appear. (2) Next I examine three specific strategy pairings applied: poetry and gymnasium; concept-learning and imitation; and imitation and discursive reasoning. These three strategy pairings all aid in maturing the emotion of admiration and helping the learner to emulate the right qualities as excellent so that she is well positioned to develop virtues of her own.

The significance of this article is twofold: First, it highlights the importance of order of education in virtue development. In the classical tradition, there is a logic not just to the application of certain methods, but also to the sequence in which they appear in one’s education based on the intellectual and emotional maturity of the learner. Second, this article provides practical guidance for how we might habituate reason in a learner. R.S. Peters describes a ‘paradox of moral education’ that results from attempts to combine both rational development and habituation in character education. The paradox is that habituation involves the extrinsic imposition of desires, attitudes, and actions in a learner, which undermines attempts to develop her own powers of critical reflection. The classical pedagogical sequence—and in particular the strategic pairing of imitation and discursive reasoning—may hold a key for how to successfully transition a learner from habituation to phronesis development so that she is well positioned to choose well for herself.

In Plato’s Protagoras, Socrates asks whether virtue can be taught. Protagoras says that it can and offers a myth to defend this belief. In this myth, all humans are given practical wisdom and fire to survive (Plato, 1997b, 321d). Subsequently, they are given a share of justice and shame so they can cooperate in a city. Virtue acquisition is open to everyone, and society punishes and rewards on this basis (Plato, 1997b, 323c-324d). For Protagoras, virtue is a type of wisdom and it is neither “natural or self-generated, but [is] something taught and carefully developed in those in whom it is developed” (Plato, 1997b, 323c). If virtue is wisdom, then it is teachable, and this sustains his claims to be able to teach it. Over the course of the dialogue, Protagoras is turned about and in frustration seems to deny his original claims about the teachability of virtue, claims which Socrates then takes up to defend. This conversation is full of important questions about the nature of virtues and the relationships among different virtues, and the cursory treatment of the dialogue provided here does not do it justice. But one clarifying and often overlooked moment in this exchange with Socrates is Protagoras’ description of his theory of education, a theory built around compulsion. Protagoras’ pedagogical sequence both makes his claims to teach virtue seem dubious and highlights a practical difficulty in virtue education—the problem of how to habituate young learners without “stultifying their psychological powers of critical reflection at a later stage” (Kristjánsson, 2006b, p. 102). Protagoras prioritizes rational development over habituation in his educational sequence, and the result is learners who are neither free nor good.

For Protagoras, moral education proceeds as follows: As soon as the child is able to understand what is being said to him, he is told what is honorable and not, what is holy and not, and what he should abstain from. If he obeys, then all is well. If he does not, he is straightened “with threats and blows, as if he were a twisted, bent piece of wood” (Plato, 1997b, 325d). He is sent to school to learn manners and reading, and to learn by heart the works of great poets “while sitting on a bench at school” (Plato, 2009, 325e). And if the boy is “temperate and gets into no mischief,” then he is rewarded with music as lyric poetry and harmony. The final step in the educational sequence Protagoras provides is gymnastics to add courage. Thereafter, the young man lives under the compulsion of laws of the state. “Why, Socrates, do you still wonder and doubt whether virtue can be taught? Cease to wonder, for the opposite would be far more surprising,” he concludes (Plato, 2009, 326e).

Protagoras’ pedagogical sequence is backward. Poetry and gymnastics are treated as education’s dessert, rather than used a propaedeutic to learning. Embodiment is largely ignored, except as an object of punishment. The student is made to sit still on the school bench while learning about heroes—the ones who should spark imitative action. Since virtue is acquired by acting, the fact that they are sitting still should give us pause. Furthermore, since moral virtue is acquired and not natural, one might wonder whether Protagoras has mistaken a docile natural temperament for virtue: He says that the temperate child is rewarded with music. What child is temperate? Lastly, the process both starts and ends with external rules, seemingly because internal change has not been made such that the student can live a well-ordered life. He writes that after they finish formal schooling, the learners are “not to act as they please,” indicating that posteducation, their desires are still at odds with good deeds (Plato, 1997b, 326d). Compulsion is necessary because, in this pedagogical sequence, virtue is unmotivated. The learner is told what is choice worthy but not formed in such a way that he sees it for himself.

In contrast, when Socrates describes education in the Laws and of the guardians in the Republic, he starts with poetry and gymnastics to form the learner’s affections, then proceeds to formal learning, like reading and writing (Plato, 1997c, 376-7). Play is used from the earliest stages (Plato, 1997a, 797a), and music serves as a training in beauty—aesthetic and moral—to both demonstrate harmony and to introduce certain heroes, those who are worthy of emulation (Plato, 1997c, 377a-c). The beauty of heroes shows students that the good life is attractive. Compulsion is used to a certain degree of course. For example, attendance is mandatory. But docility, a law-abiding spirit, and self-rule are more so the results of having the right affections in place than of fearing punishment. That is, the learner’s motivations for action are the love of beauty and goodness and of desiring these things for oneself, rather than from fear. This is important because fear of punishment may be helpful in eliciting the right actions, but this does not train the learner to act for the right reasons over the long term. The actions are neither free nor rightly motivated.

This comparison between Protagoras’ and Socrates’ pedagogical sequences is important because it highlights the fact that order of education matters in the formation of virtue. Interestingly, both Protagoras and Socrates have mostly the same ingredients of moral education—poetry, reading, writing, gymnastics, heroes, and compulsion. One exception is that “play” is conspicuously absent from Protagoras’ sequence in lieu of extra compulsion. Regardless, the teachability of virtue seems in large part to depend upon the order in which educational methods are used, rather than simply on whether those methods are used. I draw this out because, in much of the current literature on virtue pedagogy, it is not always clear at what age certain strategies are deployed or how they ought to be modified to accommodate developments in emotional and intellectual immaturity.

An example of this is that Zagzebski advocates “reflective admiration” of one’s exemplars (2013, p. 193). A person ought to refine the objects of her admiration by reflection in the context of her epistemic community to be sure she is admiring the right people as excellent. However, young children are the most dispositionally imitative and less circumspect in their admiring. They are the most shaped by aretaic role models, yet they lack the moral experience and intellectual maturity to engage in the reflective process of revision Zagzebski describes. Furthermore, most young children lack a virtue grammar to be able to articulate what is good about the people they admire. Without sufficient guidance, they are apt to admire the wrong people as good, and their admiration can make them vulnerable to the development of poor habits. Children require training of their emotion of admiration, just as other emotions are trained. Therefore, the use of exemplars for young children should be guided by those who know what is good and beautiful, and to the extent that it is possible to manage a child’s exemplar exposure, the role models who are placed before children should be those who genuinely instantiate the virtues (Fossheim, 2006, pp. 115-116). Over time, with sufficient moral experience and well-trained affections, learners can more independently reflect on who their role models are in order to emulate the right people, to practice virtue, to develop the right motivations, and to hopefully acquire virtues for themselves.

The classical model is coarsely structured to engage the body and emotions first, with a step-wise introduction of prudential and critical reasoning skills as the learner matures. There is an extensive literature on whether and how we might rationally participate in our own habituation based on textual evidence in Aristotle. (See Kristjánsson 2006b, pages 108-110 for an overview of the key players in this debate—Burnyeat, Curzer, Sherman, and Dunne, as well as Kristjansson’s own helpful commentary.) The classical model provides a framework for how reason can be introduced.

The basic classical sequence is derived from the explicit frameworks provided by Plato and Aristotle in their descriptions of paideia, an educational model that aimed to shape learners for virtue and citizenship. But this model was in no way complete in their own time. Henri Irenee Marrou describes how there is a time lag before a system of education adequately reflects the ideas held by a society. He writes, “That is why classical education did not attain its own distinctive form until after the great creative epoch of Greek civilization. We have to wait until the Hellenistic era before we find it in full possession of its own specific forms, its own curricula and methods” (1956, p. xiii). Classical pedagogy would be slow to mature, but when it did (in the “Hellenistic era and beyond” according to Marrou), it would provide insight into how to educate for the theories of virtue and citizenship Plato and Aristotle introduced (Ibid). For our own purposes, this means that in our renewed interest in virtue education, we should look not just to the explicit teachings of Plato and Aristotle for guidance on how to cultivate virtues but also to the period of education that followed them and attempted to answer the puzzles their theories created. One such puzzle is the habituated reason paradox introduced earlier.

The classical model as it is known today has three general phases: pre-education, trivium, and quadrivium. For a helpful overview of the model apart from my focused attention on virtue-formation, Dorothy Sayers’ 1948 essay, “The Lost Tools of Learning,” is a great resource.

Briefly, pre-education is poetry and gymnastics, which together served as a propaedeutic to formal education (Plato, 1997c, 376e). A student learned physical training for the practical end of developing competence in athletic or military contests, as well as to become disciplined enough to be teachable in a formal setting (Plato, 1997a, 654-5). Aristotle wrote, “Now it is clear that in education habit must go before reason, and the body before the mind; and therefore boys should be handed over to the trainer, who creates in them the proper habit of body” (Aristotle, 1997, 1338b).

Both Plato and Aristotle had concerns about how this stage was to be conducted. Plato worried about the competitive spirit’s temptations to seek empty glory (Plato, 1997a, 796a), and Aristotle worried about Olympic-caliber competitions that “exhausted their constitutions” (Aristotle, 1997, 1339a). Still, it is easy to see how moderate forms of athletic training can afford learners the opportunity to develop productive, emulative relationships with their peers through competition. Competition itself is a practice of outdoing or exceeding a peer in excellence, structurally analogous to Aristotelian zelos, a type of comparative emulation productive for encouraging virtue (Aristotle, 2019, 1388a). Further practice in emulation would also come from the imitation of their instructors, who were advanced athletes themselves (Marrou, 1956, p. 40). Also at this stage, poetry introduced learners to heroes. And, as in athletics, students learned from a poetic master whom they emulated to learn how to sing and to play the lyre (Plato, 1997a, 812). Poetry also taught students something like emotional competence and expression (Aristotle, 2009, 1447a5-30). Through harmonies, they participated in a range of emotions, partaking in and imitating these emotional experiences in songs and in dance—harmonies that reflected the aesthetic values of their culture (Aristotle, 1997, 1340a17-1341a7). Both Plato (1997c, 377b; 1997a, 810c-811b) and Aristotle (2009, 1448b4-25) took for granted that we learn through imitation and worried that we would imitate ignoble deeds, rather than good and fine ones.

Trivium means “three ways.” Quadrivium means “four ways” and was reserved for serious scholars. Together, the trivium and qua-drivium constitute the seven liberal arts. The trivium was the basic education model, consisting of three stages: grammar, logic, and rhetoric. Regarding Marrou’s observation about the lengthy maturation of classical pedagogy, the trivium appears to have been enriched by both the textual focus of the Hellenic period (Marrou, 1956, p. 95) and by the clear stages of imitation applied by Roman orators, such as Quintilian (2007, pp. 119-120). Grammar focused on concept-learning, improving a learner’s powers of observation and memory. Most of the imitation at this stage was done through recitations, by rote. The logic stage harnessed more prudence into the imitative process, while students also learned discursive reasoning, as I explain in the next section. The rhetoric stage invited spontaneous speech or invention on a topic that the learner selected as important to him, or as pertaining to his personal cares (Sayers, 1948, p. 17). After completing the trivium, a student might be invited to participate in the four subjects of the quadrivium.

While this is certainly a concise introduction to classical pedagogy, my focus is virtue strategies in particular. This overview should serve as sufficient background for introducing the three virtue strategy pairings in focus for the question of habituating reason.

The three strategy pairings are (1) poetry and gymnasium, (2) concept-learning and imitation, and (3) imitation and discursive reasoning. Poetry and gymnasium are paired and applied before formal education begins. Concept-learning and imitation are paired in the trivium at the grammar stage. Imitation and discursive reasoning are paired in the trivium at the logic stage. These are three sequential stages. Exemplars are used throughout moral education and at each of these three stages, but the way exemplars are used changes as students mature.

Before formal learning, students are prepared to be educated, and this happens through poetry and gymnastics, or exemplars plus athletics. By poetry, Socrates means the education of our heroes, which takes the form of story-telling and lyric poetry (Plato, 1997c, 376-8). A poet “by adorning countless deeds of the ancients educates later generations” (Plato, 1960, 245a). By gymnastics, he means various forms of physical exercise, such as wrestling, dancing, and sprinting (Plato, 1997a, 795d-796a, 813e-814e). Students are exposed to hero stories and to physical training at the same time, as a propaedeutic to education, to form their affections in preparation for formal schooling.

On the surface, this seems like not an especially fruitful virtue pairing, since stories and physical training are not obviously coordinated methods of moral education, nor are they regarded as very serious activities in early education. For example, a modern pre-school may present a combination of Dr. Seuss books and monkey bar games yet not understand these as pertaining to virtue education. (These activities may even be destructive to a child’s developing character, since the conventions of many playgrounds are somewhat Hobbesian.) This combination does not seem novel to the contemporary conscience because this is not too far off from what happens in preschools, introduced as story-telling and physical play. These are commonly paired activities. Although, in many cases, the process could be more directed and selective of the sorts of heroes portrayed and the types of exercises performed.

Poetry and athletics in fact constitute a powerful mixed-methods approach to virtue education because students can learn about heroes at the same time that they are encountering opportunities to be brave and to discipline their bodies. In general, exposure to exemplars like heroes is helpful in spurring a person to develop virtue because comparing oneself to an excellent other reveals a lack of excellence in oneself. Exemplars expose a “character gap”—a space between how we really are and the people of good character we should become (Miller, 2018a). Athletics can aid this process in a few ways.

First, physical exercise enables a learner to truly see her lack of virtue. Consider perseverance in an exemplar. If a learner is shown a hero with the virtue of perseverance and she sees that quality as desirable, if she is later presented with the opportunity to run, it becomes very clear whether or not she possesses that quality. In running, a person can measure the extent to which she can or cannot persevere in minutes or kilometers. She can practice this quality and measure her growth in objective terms.

There is more that must be said here to disentangle physical strength from perseverance. For example, some people can naturally run longer than others can without having first acquired through practice the excellence of perseverance. Still, as a general rule, physical training involves both physical work and character work to sustain an activity, and in the athletic context, it becomes very clear if a learner lack either of these types of work.

Further examples can be made with fortitude or prudence in ball sports. Performance suffers in a controlled “game” setting if a player lacks prudence or if she is not brave enough to act. In “How Admiring Moral Exemplars Can Ruin Your Life,” Thomas Alan, and colleagues describe how admiration can set one up for a distinctive kind of moral error—one may not realize the ways in which she is not an exemplar (2019, p. 241). That is, by holding an exemplar in mind, a learner can begin to think of herself in terms of the excellences the exemplar possesses, yet she does not actually have these excellences herself. Because of this, admiration seems to require that an actor have an adequate self-understanding in order to be motivated to improve. Otherwise, what may result is a type of self-glorification. Pairing athletics with exemplars can help avoid this error by showing a learner the ways in which she falls short, in measurable, performance-relevant ways.

Second, athletics models a means of improvement in the virtues. Taking again the example of perseverance, there is a practice-based approach to improvement in this respect: The athlete runs more kilometers on a consistent basis, on various types of terrain, at various speeds. Aristotle writes of virtue development that “men become builders by building and lyre players by playing the lyre; so too we become just by doing just acts, temperate by doing temperate acts, brave by doing brave acts” (Aristotle, 2011, 1103a-b). Virtues are acquired by repeated practice. If a learner desires to become virtuous, she has to practice the virtues. Structurally, then, participating in a sport is helpful for character formation because this is a domain in which we apply the logic of “practice.” Athletes wake up every day and repeatedly do the same things, with the intention of improving. Having exposure to exemplars who practice becoming better at a craft, and being invited to participate in that practice, can shape a learner’s expectations for how personal development occurs.

Third, athletics aids the force of exemplars by demonstrating that virtue formation is difficult. It is something that takes focused practice and a concerted effort. Continuing with the same example, perseverance often looks effortless in one who has it, and the concept of “perseverance” sounds easy. It just means to “remain,” or to keep doing what the person is already doing. But in running, the learner can quickly realize that “remaining” is difficult. When a person remains, she feels burning in her legs and in her lungs, and it takes labored practice to be able to persevere. In a formal classroom setting, students can learn about a virtue by definition, but they do not always experience training in that virtue or what it feels like to grow in virtue, and they are not often presented with the opportunity to build habits in a systematic way. Physical training provides this opportunity. It helps a person to appreciate the difficulty of acquiring a virtue exhibited by an exemplar, and it draws attention to the development the exemplar underwent to become excellent in that way.

Based on these three features of athletics— that physical training allows a person to see her lack of virtue, introduces a practice-based approach to acquiring virtue, and highlights the difficulty of acquiring virtue—using athletics in the context of exemplar exposure prepares one well to begin the process of virtue development. Of course, in this virtue pairing, the questions to ask are about the particulars of training and the qualities of exemplars. For example, Aristotle cautions against heavy exercises that might compromise intellectual development or stunt growth over the long term, as he characterizes the physical exercises of the Spartans as doing (Aristotle, 1997, 1338b). Moreover, there are certain qualities of exemplars that translate well in and out of an athletic context, such as fortitude and patience. But there are others that do not, like gentleness and truthfulness. If the hero in question were an outstanding computer scientist, for example, the complementarity of physical practice with exposure to that person might not work as well to draw out his excellences.

Furthermore, it is not obvious that if a child were to develop a local trait (or context-specific trait) of perseverance in running that she would necessarily become perseverant in all areas of life. Likely, she will still need to practice that virtue in various contexts to extend the virtue. Even so, this is just the beginning of the pedagogical process. If only a few virtues, or local versions of those virtues, are effectively exemplified and practiced at this stage, the student is made aware of the responsibility she has for working to develop her own character and she acquires structural knowledge of how to grow in virtue, through repeated practice.

In the trivium, intellectual virtues and linguistic abilities are developed so students can responsibly reason as orators and citizens. The first stage of the trivium is grammar. It is characterized by the development of memory and observation. Children at this age thrive in memorization and recitation, so chants and memory drills are common (Sayers, 1948). Quintilian describes this stage in this way: “Let us not waste the earliest years: there is all the less excuse for this, since the elements of literary training are solely a question of memory, which not only exists even in small children, but is especially retentive at that age” (Quintilian, 2007, p. 11). Many of the recitations performed at this stage are hero stories. Students also recite the works of literary masters, including those of their own teachers. Here is the second instance of exemplars being systematically applied in education. In memorizing these stories, students’ moral imaginations are shaped. And since students perform these exercises together, community standards are set by the excellences they collectively learn. At the grammar level, students also start to read and learn the basics (or grammars) of history, math, theology, and language (Sayers 1948).

As was introduced earlier, after students are exposed to poetry and gymnastics, they should be aware (in a genuine physical sense) of their shortcomings. They are introduced to stories of heroes, and they realize they do not yet have the excellences of these heroes. They know what it feels like to lack these excellences. For example, Odysseus would never give up on a journey, yet the learner gave up after a ten-minute jog. He must not be excellent like Odysseus is. But at this point in his formation, the learner is unable to articulate the concepts of the virtues he lacks.

In the recent virtue ethics literature, there is a debate about what is being called an “articulacy requirement” for virtue: Must we be able to articulate reasons for our apparently virtuous behavior, in order that it qualify as virtuous (Annas 2011; Stichter 2007a)? Julia Annas’ support for an articulacy requirement is made by analogy to other types of practical expertise. If a person is an expert, then he or she has skill (techne), which is distinguished from an “inarticulate ‘knack’ (empeiria)” by a knowledge of the conditions of the action (Annas, 2011, p. 20). An expert can “give an account of what he does, which involves being able to explain why he is doing what he is doing” (Annas, 2011, p. 20). Likewise, so the argument goes, the virtuous agent should be able to give an account for his or her actions.

There are a number of strong objections to the articulacy requirement. For example, Matthew Stichter argues that there are counterexamples in the domain of practical expertise: Not all experts are able to give an account for their actions (2007b). Another objection is one Annas herself raises—that certain people, such as those with Down’s syndrome, may not meet the minimum requirement of intelligence to provide reasons for virtuous actions. An articulation requirement for moral virtue might mean that those with limited intellectual abilities are excluded from acquiring mature moral virtues (Annas, 2011, p. 32).

I am not going to take a stance in this debate here because whether articulation is a necessary condition for moral virtue depends in part on how one conceives of virtues. Annas’ articulation argument is made by analogy to skills, and it is not clear whether the same conclusions can be drawn on a “virtues as global traits” view (Miller, 2018b), a “virtues as social intelligence” view (Snow, 2010), or for other models. Moreover, there are virtue accounts that accommodate intellectual difference. For example, Christine Swanton’s account includes both (1) mature Aristotelian virtues and (2) basic or stage-specific virtues (Swanton, 2016, p. 118). Basic virtues are excellences relative to the abilities of the learner, so a person might not be able to articulate reasons for actions yet not be excluded from virtue ascription altogether.

Nevertheless, in the specific context of acquiring virtue by way of the emulation of exemplars, the ability to articulate the excellence in question is of practical importance, regardless of the conception of virtue maintained. This is to assuage concerns that learners may be imitating the wrong people, or the wrong qualities in the right people. A considerable worry in using exemplars in moral education is that if there is no critical reflection on specifically named qualities set apart from the person, then we are just teaching students how to unreflectively conform to a charismatic leader (Kristjánsson, 2006a, p. 41). We may be training students to blindly imitate anyone they find admirable, without equipping them to critically reflect on what qualities they admire. Interestingly, the classical pedagogical sequence answers this concern by providing the virtue grammar state after poetry and gymnasium, and alongside the memorization of stories. Students are taught virtue concepts, which they memorize as a grammar to describe moral qualities and actions. This enables them to meet Annas’ articulation requirement. Moreover, even if there is no such requirement of virtue, if a student can articulate an excellence, this means her imitation will be more targeted than it would be if she merely recognized an amorphous sense of goodness in looking at a hero.

Sayers describes the grammar stage as involving “accumulation” rather than speculation (Sayers, 1948, p. 11). Students delight in acquiring words through recitations and rhymes, often performed with their classmates. They learn basic vocabularies and enjoy naming things, like natural specimens, historical dates, and people. At this period, acquiring a moral grammar might look like the memorization of virtues with pithy definitions. For example, students might learn that “prudence is right thinking plus action.”

English grammar lessons equip students to be able to articulate what is excellent about a good poem so they can acquire the tools to write one themselves. In the same way, the memorization of virtue concepts is a fitting complement to the hero stories and recitations they perform at this stage. This combination equips learners with a vocabulary to discuss what is excellent and choice-worthy, and it helps to draw explicit attention to the qualities one ought to imitate in an exemplar. This builds on the work done in poetry and gymnastics by naming the excellences in question, which well positions a learner to pay attention to virtues. In Seneca’s “On Liberal and Vocational Studies,” he describes the grammar stage in this way:

And if you inquire, “Why, then, do we educate our children in the liberal studies?” it is not because they can bestow virtue, but because they prepare the soul for the reception of virtue. Just as that “primary course,” as the ancients called it, in grammar, which gave boys their elementary training, does not teach them the liberal arts, but prepares the ground for their early acquisition of these arts, so the liberal arts do not conduct the soul all the way to virtue, but merely set it going in that direction (Seneca, 2007, 101).

Grammar “prepares the ground” for virtue by equipping students with an ability to name the imitable excellences in exemplary people. This does not take them all the way to virtue, but it sets them “going in that direction.”

The second stage of the trivium is logic, and in it we see a continuation of this graded approach to exemplar exposure, to accommodate growing emotional and intellectual maturity. At the logic stage, students are, unsurprising, introduced to logic, or formal reasoning. They analyze language, syntax, arguments, and laws, and they critically assess different positions on an issue (Littlejohn & Evans, 2006).

Exemplar exposure looks different at this stage because students have grown more capable of critical reflection. They engage in a process called imitatio, the imitation of a master. In imitatio, students memorize, translate, and copy great speeches and stories and are also expected to “imitate or emulate the heroic characters and principles described therein” (Kaplan, 2003, p. 241). Furthermore, unlike the recitations performed in the grammar stage, imitatio is not done entirely by rote. Students are also taught to abstract excellent qualities from a work and to apply them. In literature, this might look like the imitation of specific literary conventions used by a master. For example, students read Beowulf, then write their own short epic poems, using tools such as alliteration, in media res, and kennings. Sayers describes how, at the logic stage, “the pupil will learn to try his own hand at writing this kind of thing” and is instructed to pay special attention to the arguments contained in the work he is reading (Sayers, 1948, p. 15). In writing their own poems using the tools of the masters, students learn prudence by practicing the excellent qualities, rather than simply admiring them. They take ownership of these excellent qualities.

In the moral domain, imitatio can look like evaluating a hero and asking difficult questions about how the exemplar’s life differs from the learner’s life. A learner may be asked how to incorporate the fortitude of a hero in his own context. These conversations involve critical reflection on how a learner might incorporate a range of good actions in his daily life, in different situations, and over the long term. Again, these discussions happen in a community context.

Imitatio is practiced at the same time that students learn dialectic and formal reasoning skills. In this pairing of exemplars and formal logic, a learner’s prudence and rea- sons-responsiveness are refined while he is actively emulating exemplars in a more focused way. This stage prepares a student to move beyond emulation to virtue because he is equipped to think critically about his reasons or motivations for action.

Beyond logic is the rhetoric stage—the final level of the trivium. This stage is marked more by spontaneous action than by imitation. By this point, students will have learned a grammar of moral action, have internalized social and moral norms, and be oriented toward the right ends. In the humanities, this looks like the ability to take a stance and argue for a given position, making unique contributions to the intellectual life of the community. In moral terms, this looks like the ability to freely act in excellent ways, consistently and with cross-situational stability, for the right reasons. Sayers describes how, by the rhetoric stage, a learner’s “tendency to express himself windily or to use his eloquence so as to make the worse appear the better reason would, no doubt, be restrained by his previous teaching in dialectic” (Sayers, 1948, p. 2). This is in contrast to the preparations afforded by sophistical educations, like the one Protagoras described, which was structured in such a way as to produce more sophists like him.

Interestingly, the level of intellectual maturity required to participate in the art of rhetoric is structurally analogous to the moral stage of development required for virtue. Students act from the rules of grammar and make arguments in accordance with good reasons, but these concepts, conventions, and rules are not held explicitly in mind in the act of oration. They are internal to it. Furthermore, Quintilian describes how rhetoric students “have neither the time nor the inclination” for the sort of deliberative work that precedes this stage (Quintilian, 2011, p. 118). The sort of preparation of reason required at the grammar and logic stages, while pleasurable and appropriate for the young, is time-consuming and too great of a cognitive load to continually bear. But if a student is formed in the right ways, these conventions have become second nature. The learner can work from well-formed rhetorical habits.

Likewise, Aristotle describes how the virtuous person acts for the right reasons but from rightly-ordered emotions. Deliberative reason (prohairesis) is upstream of virtue as a necessary condition for becoming virtuous, but deliberative reason remains at the level of continence. During critical reflection, the action still needs to be chosen and has not yet become a habit, whereas the virtuous person’s reasons are internal and firmly held, but no longer explicitly revisited in daily actions. Nancy Snow calls this post-deliberation stage of habituation “goal-directed automaticity” (Snow, 2010, p. 52).

Interestingly then, education starts with a type of automaticity (unreflective instinct) and ends in a type of automaticity (well-formed habits). The pedagogical stages in between work to ensure that the automaticity of the second sort consists in habits of the right kind.

Marrou describes how there is a time-lag before a system of education adequately reflects the ideas held by a society. For our purposes, this means that in our renewed interest in virtue education, we should look not just to the explicit teachings of Plato and Aristotle for guidance on how to cultivate virtues but also to the period of education that followed them and attempted to answer the puzzles their theories created. One such puzzle is R.S. Peters’ paradox of habituated reason.

In this article, I evaluated strategies in the classical sequence for practical guidance in how to answer this Peters’ paradox—to combine habituation and practical wisdom development in ways that will well position a learner to develop virtues. What I found were three virtue growth pairings which together constitute a graded maturation of critical reason, in the context of exemplars. These stages provide practical guidance for how to invite critical reflection into the process of habituation so that by the end of the trivium, students are equipped to move productively in the direction of closing their character gap—to act freely and consistently in accordance with virtue.

Sabrina Little is a full-time instructor of philosophy at Morehead State University in Morehead, Kentucky. She received her doctorate in philosophy from Baylor University in 2020.