How is Virtue Cultivated?: Seven Strategies for Postgraduate Character Development

Despite renewed academic interest in virtue ethics and character development, institutions of higher education have largely avoided intentional efforts to form the character of university students.¹ Most character education programs focus on primary and secondary students, leaving university students without sufficient support to negotiate a crucial time in their moral development. This article seeks to address this gap by focusing on the character education of postgraduate students pursuing advanced academic or professional degrees. For many students, postgraduate education marks a time when they are beginning careers, learning new skills, and facing unexpected challenges in integrating their personal values with the values and expectations of their professions. These challenges are all the more significant during “emerging adulthood,” the stage of life between adolescence and adulthood—usually between ages 18 and 29—when many postgraduates typically begin advanced degrees (see Arnett, 2000, 2014a, 2014b). Psychologists have identified emerging adulthood as “a time of great instability,” perhaps even “the most unstable stage of the life span” (Arnett, 2014b, p. 158). Because emerging adults are marrying and having children later, holding a higher number of jobs, changing residences more frequently, and pursuing higher education and professional training over a longer period of time, this developmental stage has become a time of intense “identity explorations” as emerging adults experiment with new ideas, roles, and relationships and consider the kind of person they hope to become (Arnett, 2000, 2014b). For these reasons, emerging adulthood is a “transformative period of self-development,” when “character traits,” “character strivings,” and “character prospection” are important features (Noftle, 2015, pp. 490-493; cf. Arnett, 2000; Lapsley & Hardy, 2017). While more empirical research is needed to analyze these aspects of emerging adulthood, existing scholarship suggests that this developmental stage is an important time for character formation, which makes the absence of postgraduate character education programs all the more concerning.²

This article seeks to make two contributions to the theory and practice of postgraduate character development. First, we provide an accessible synthesis of recent research in education, philosophy, and psychology on seven Aristotelian strategies of character development: (1) habituation through practice, (2) reflection on personal experience, (3) engagement with virtuous exemplars, (4) dialogue that increases virtue literacy, (5) awareness of situational variables, (6) moral reminders, and (7) friendships of mutual accountability. While scholars have begun to direct more attention to each of these strategies, few examine how multiple strategies fit within a holistic program for character development. Second, in the discussion of these strategies, we show how each can be effectively integrated and applied in a practical program for character development. Here we move beyond recent work that offers theoretical and empirical support for individual strategies, and beyond the task of summarizing promising strategies that practitioners might employ, to supply practical examples from a 3-year case study (2014-2017) of the Oxford Global Leadership Initiative (GLI), an initiative of the Oxford Character Project. The GLI is a voluntary, extracurricular program for character development, comprised of select groups of University of Oxford postgraduates from a range of disciplines, professions, and backgrounds. The GLI has continued to grow since 2017, but our focus here is restricted to the first 3 years of the program.³ By providing both a theoretical framework and practical examples of how to integrate each strategy, we hope to offer guidance for educators who aspire to design character education programs informed by educational theory and practice.

The theoretical framework that informs our approach to character development is supported by a long tradition that dates back to Aristotle and continues in contemporary philosophy and psychology.⁴ On this broadly Aristotelian account, “character” is conceived as the collection of stable, deep, and enduring dispositions that define who we are and shape how we characteristically think, feel, and act (see Annas, 2011; Aristotle, 1999; Zagzebski, 1996). This Aristotelian tradition distinguishes moral and intellectual dispositions of character. It typically identifies positive dispositions as “virtues”—habits that dispose us to think, feel, or act “at the right times, about the right things, toward the right people, for the right end, and in the right way”—and negative dispositions as “vices”—habits that dispose us to think, feel, or act at the wrong times, about the wrong things, toward the wrong people, for the wrong end, or in the wrong way (Aristotle, 1999, 1106b22-23, 1109a20-b7; cf.1106a16-1109a19). Between these two poles are dispositions that cannot be classified as either virtues or vices. These intermediate states include “continence”—when we know and do the right thing, but lack the firm and settled disposition of character to do it reliably and consistently—and “incontinence”—when we know the right thing to do but fail to do it (Aristotle, 1999). The aim of Aristotelian character education is to encourage people to develop more stable and enduring virtues of character in the face of temptations and difficulties (see Burnyeat, 1980; Sanderse, 2012; Wilburn, 2007).

Given this developmental and aspirational conception of virtue (Annas, 2011), the GLI focused its efforts on helping postgraduates cultivate relevant virtues and resist corresponding vices. Since it would be difficult to develop and measure the entire suite of virtues within the span of 1 year, during the 3 years of this case study (2014-2017), the GLI focused on four specific virtues: (1) a sense of vocation, (2) a commitment to service, (3) gratitude, and (4) humility.⁵ Although an entire suite of virtues is necessary to promote individual and communal flourishing, this more focused approach aligns with recommendations to address a select number of virtues rather than trying to develop all of them within a short time (Meindl et al., 2018; Miller, in press).

Based on the experience of educators leading the program, the GLI chose its focal virtues for four reasons. First, all four virtues are generally recognized as morally admirable by a wide range of philosophical, moral, and religious traditions, including the traditions with which most GLI’s participants identify.⁶ Addressing these four virtues thus provides a substantive focus for the program while also supplying sufficient scope for diversity.

Second, these four virtues are widely seen as essential for wise and effective leadership (e.g., van Dierendonck & Patterson, 2015; Kempster et al., 2011), particularly for individuals expected to progress to the positions of authority and influence for which their postgraduate education trains them. Yet these four virtues tend to be neglected in most postgraduate programs. Unlike virtues such as resilience or persistence, they cannot be assumed to develop naturally through rigorous selection, intellectual study, or professional training. A practical program focused on these virtues thus meets an important developmental need for students who voluntarily join the program.

Third, all four virtues help to shift attention away from narrow self-interest toward the common good and the character required for significant impact in a pluralistic society (Kempster et al., 2011; van Dierendonck & Patterson, 2015). This shift is especially important for universities and postgraduate programs that identify social impact and public service among their explicit educational aims.

Finally, all four virtues respond to specific challenges that accompany emerging adulthood. Most emerging adults, for example, are involved in intense exploration of their identity and vocation. In one poll of 18-to 29-year-olds, 77% agreed that “This is a time of life for finding out who I really am” (Arnett, 2014b, p. 158). A program that encourages postgraduates to develop a sense of vocation can help them achieve this objective. Similarly, cultivating a commitment to service may help to direct their activities and aspirations toward others during a period that psychologists have identified as a “self-focused age” (Arnett, 2014b, p. 159). In one survey, 71% of emerging adults affirmed that “This is a time of my life for focusing on myself” (Arnett, 2014b, p. 159). Whether such self-focus reflects the narcissism of “Generation Me” (Twenge, 2006) or a temporary developmental stage that shifts once emerging adults enter more stable jobs and relationships (Arnett, 2010), cultivating a deeper commitment to service can help to prevent an age-appropriate self-focus from hardening into a vice. The same applies for gratitude and humility. Philosophers have long held that gratitude disposes us to recognize our dependence on others, which, in turn, discourages us from seeing achievements as simply our own and thereby encourages humility (see, e.g., Keys, 2006). Humility, in turn, disposes us to recognize our limitations and the need for others’ assistance, which encourages gratitude. Recent work in social psychology confirms this mutually reinforcing dynamic, showing that gratitude and humility contribute to an “upward spiral” that decreases self-focus, increases positive affect, and encourages other-regarding behavior (Kruse et al., 2014). If increased self-focus is a defining feature of emerging adulthood, then the four focal virtues may be effective antidotes.

The GLI intentionally structured its efforts to cultivate these virtues within a leadership-based program for several reasons. First, many universities, including Oxford, aspire to educate global leaders in diverse fields but, with the exception of programs in some professional schools, most do not provide specific leadership training within their postgraduate curricula. A voluntary leadership and character development program can thus supplement curricular offerings and is easier to implement than programs that consume class time or require formal departmental approval across multiple faculties. Second, most postgraduate leadership training focuses on cultivating leadership skills rather than virtues, so a character-based initiative helps postgraduates consider the ends, purposes, and manner in which their leadership skills might be used. Finally, many postgraduates aspire to become “leaders” in their fields and thus seek opportunities that will help them find meaningful work or secure positions of authority or influence. The GLI tapped into this existing motivation to recruit students who might not be as attracted to an initiative focused solely on character. This approach fits with research suggesting that indirect forms of moral education that engage student motivation, tap into social influence, and embed efforts within an existing culture are often the most effective (Lapsley & Yeager, 2012; Meindl et al., 2018).⁷ It also aligns with one approach to virtue development that suggests that participants who cultivate virtues with instrumental value for a particular role (such as “good leader”) may come to recognize the constitutive and intrinsic value of acting virtuously, regardless of its instrumental value for any particular role (Snow, 2018).

To cultivate such leaders of character, the GLI employed seven Aristotelian strategies of character development grounded in theoretical and empirical research in education, philosophy, and psychology. A brief explanation of each strategy and the GLI’s application of it highlights how academic research can inform practical programs for postgraduate character development (for a summary see the Appendix).

In the Nicomachean Ethics, Aristotle (1999) differentiates “virtues of thought” from “virtues of character” by how they are acquired: virtues of thought arise “mostly from teaching,” while virtues of character emerge through “habit” (1103a15-18). Since Aristotle conceives virtues by considering how they are acquired, an Aristotelian account is “essentially developmental” (Annas, 2011, pp. 4, 38; cf. Broadie, 1993; Burnyeat, 1980; Russell, 2015).

Aristotle’s distinction between teaching and practice is important for postgraduate character development. Postgraduates cannot learn how to improve their character simply by reading a book or applying an abstract formula they learn in class.⁸ Rather, they must learn virtues of character, in part, by doing virtuous actions. Indeed, for a good disposition to become a stable and enduring virtue, it must become a kind of habit—a deep, reliable, and entrenched disposition of thought, feeling, or action (Aristotle, 1999; cf. Broadie, 1993; Burnyeat, 1980). On an Aristotelian account, we acquire such habits through practice—by repeating appropriate thoughts, feelings, and actions over and over again until we gradually become disposed to think, feel, and act in the right ways at the right places at the right times, as if by second nature (Aristotle, 1999; cf. Annas, 2011; Broadie, 1993; Burnyeat, 1980; McAdams, 2015; Narvaez & Lapsley, 2005; Russell, 2015; Wilburn, 2007). Aristotle compares this process to that of acquiring skills: “we become builders, for instance, by building, and we become harpists by playing the harp. Similarly, then, we become just by doing just actions, temperate by doing temperate actions, brave by doing brave actions” (1103a34-1103b2). Just as a musician becomes a “virtuoso” by practicing an instrument, repeating certain movements, and playing pieces over and over again until they become second nature, so too a person becomes morally virtuous by repeating or perfecting good actions, thoughts, and feelings until she develops stable and enduring dispositions of character (Annas, 2011; Aristotle, 1999; Russell, 2015).

Recently, scholars have shown how this Aristotelian account of habituation has empirical grounding. Drawing on work in cognitive and social psychology, Nancy Snow (2010, 2016) highlights how habitual actions align with what psychologists call “goal-dependent automaticity,” a form of cognitive processing that enables us to act intelligently yet automatically toward particular goals, even when we do not have those goals consciously in mind. On this account, certain roles, activities, or situations activate representations of virtue-relevant goals that prompt particular thoughts, feelings, and actions, and the repeated activation of these thoughts, feelings, and actions cultivates habits that dispose us to think, feel, or act accordingly, even without conscious deliberation (Snow, 2010; cf. Narvaez & Lapsley, 2005; Walker, 2000; Weng et al., 2013). This process parallels Aristotelian habituation.

As Julia Annas and others emphasize, however, mere automaticity is not sufficient for virtue. Virtues and skills are not routine habits—mindless or unthinking responses to stimuli—but intelligent habits, capacities that involve dynamic, higher order judgments about the salient features of a particular situation and how to act appropriately within it (Annas, 2011; Broadie, 1993; Damon & Colby, 2015; Snow, 2010). While such judgments need not always be consciously front of mind, both virtues and skills depend on the exercise of practical wisdom and reflect underlying reasons for action (Aristotle, 1999).

Yet Aristotle (1999) is careful to distinguish virtues and skills in several ways (cf. Zagzebski, 1996). Two distinctions are most relevant. First, a virtuous person deliberates toward morally good ends and acts with the intention to achieve those ends, whereas a skilled person can employ skills toward any end, whether moral or immoral. A virtue, by contrast, is directed toward morally good ends and thus “makes a human being good and makes him perform his function well” (Aristotle, 1999, 1106a23-24; cf. 1105a27-b6). Second, a virtuous person acts from a “firm and unchangeable state,” not just from a fleeting feeling or inclination (Aristotle, 1999, 1105a34-35). In other words, she acts “in the way in which just or temperate people do,” namely, out of a settled disposition of character (Aristotle, 1999, 1105b7-9). A skilled person, by contrast, does not necessarily have to be in the “right state” of character to perform an act or skill well (Aristotle, 1999, 1105a26-35; cf. Russell, 2015).

These differences are particularly important for postgraduate character development. Most postgraduate programs focus primarily on cultivating leadership skills, from developing a strategy and communicating effectively to delegating responsibility and managing an institutional hierarchy. While skills are undoubtedly important for effective leadership, they do not necessarily contribute to ethical leadership. Ethical leaders must not only act with knowledge of what they are doing but direct their actions toward morally good ends and develop the settled dispositions of character to perform virtuous actions consistently in diverse circumstances. A leadership program that focuses only on skills without attention to virtues ignores the importance of character for ethical leadership.

If postgraduates become virtuous in part by performing virtuous actions—by leading well—then a character development program should ideally provide opportunities for students to practice leading in various contexts. Service-learning courses, group projects, and cocurricular activities can provide such occasions. However, coordinating these opportunities can often be time-consuming and labor-intensive, especially for programs such as the GLI that have a small staff. Moreover, the GLI and other voluntary programs cannot demand too much of postgraduate students who are simultaneously enrolled in full-time study.

Even with such constraints, however, the GLI applied Aristotelian insights about habituation in four ways. First, it encouraged particular practices of habituation in relation to some of the focal virtues. For example, to prepare for the discussion of gratitude, the GLI asked participants to keep a gratitude journal, a practice that has been shown to help to cultivate the virtue (Emmons, 2007). As an empirical assessment of the GLI program revealed, participants were more likely to report increases on focal virtues when conceptual analysis was paired with practical exercises to cultivate virtue, which helps to explain why participants reported quantitative increases on gratitude and service but not on vocation and humility (Brant et al., 2020). Given this finding, recent iterations of the GLI have done more to encourage habituation of the other focal virtues, for example, by asking participants to identify a virtue they are eager to cultivate and commit to adopting correlate practices over the course of a month.⁹ Benjamin Franklin’s attempt to cultivate a specific virtue each week provides an example of a historical leader who adapted a similar strategy (Franklin, 2005).

Second, the GLI challenged the “heroic” myth of leadership that identifies paradigmatic leaders as those who occupy positions of institutional power or authority (Haslam et al., 2011). Instead, it highlighted the complex and often blurry relationships between “leaders” and “followers” (Keohane, 2010) and introduced participants to more collaborative models of leadership that do not require a leader to occupy an institutional role. If individuals can practice leadership in more informal settings, then conversations that programs like the GLI organize can themselves provide occasions for leadership, opportunities for students to discern when to speak up or step back, when to challenge an opinion or raise an alternative view, and when to collaborate with diverse colleagues to pursue common objectives. In this way, the GLI supplied informal opportunities to practice leadership, even if not in an institutional setting.

Third, the GLI sought to introduce participants to particular practices relevant for good leadership. The GLI, for example, structured its program with readings and discussions centered on practices of reflection, friendship, and accountability, with the aim of helping participants adopt these practices in their daily lives. Moreover, facilitators intentionally modeled practices central to social cooperation, such as active listening and respectful engagement, which are especially important for a global leadership initiative that seeks cross-cultural cooperation. By encouraging and repeating these practices throughout the program, the GLI helped participants habituate certain ways of acting, thinking, and being.

Finally, the GLI offered opportunities for participants to reflect on the leadership they have practiced in other contexts. This is one advantage of working with postgraduate students: many pursuing advanced degrees already have significant leadership experience from their undergraduate education or prior work experience, and many maintain leadership roles while pursuing postgraduate degrees. These experiences provide participants with raw material on which to reflect and apply the conceptual and imaginative practices developed through the GLI. By recruiting and selecting applicants who have previous experience in significant positions of leadership or are currently occupying leadership roles, the GLI parlayed prior practice to encourage reflection on personal experience, which points to a second strategy.

If, as Aristotle argues, virtue requires knowledge of why and how we act in particular circumstances, and if this knowledge comes through reflection on repeated actions, then reflection on experience will be central to character development. In particular, reflection on experience can help us develop the virtue of “practical wisdom” (phronesis), the capacity to discern morally salient features of situations and deliberate about how best to act (Aristotle, 1999; cf. Hursthouse, 2006; Kristjánsson, 2014b; Russell, 2009; Schwartz & Sharpe, 2010; Zagzebski, 1996). The relationship between experience and practical wisdom is one reason why Aristotle describes the “wise” or “prudent person” as someone who is usually of advanced age or maturity: the “young person” lacks the experience to make good judgments, while the wise “see correctly because experience has given them their eye” (1999, 1141b10-24, 1142a12-15, 1143a20-b14; cf. Annas, 2011; Broadie, 1993; Burnyeat, 1980; Colby & Damon, 1992; Prior, 2007).

Empirical research supports the importance of reflection in developing practical wisdom. Several psychological models and measures include a “reflective” capacity as a fundamental dimension of wisdom (Ardelt, 2004; Gluck & Bluck, 2013; Webster, 2007), while many scholars and practitioners have made “reflective practice” an essential component of personal and professional development (Schon, 1987), even using an Aristotelian conception of practical wisdom to conceptualize it (Hart & Cooper, 2015; Kinsella & Pitman, 2012; Schwartz & Sharpe, 2010). Educators, for example, have advocated integrating reflection and practical wisdom into law (Cantrell, 2003; Cantrell & Sharpe, 2016), medicine (Epstein, 2008; Hart & Cooper, 2015; Wald & Reis, 2010), and education (Cooke & Carr, 2014; Lunenberg & Korthagen, 2009; Shulman, 2007). An academic journal, Reflective Practice, now exists to explore different conceptualizations and applications in various fields.

The GLI sought to prompt structured reflection on personal experience in several ways. In its recruiting events and application, the GLI asked applicants to reflect on their own experiences to identify virtues that are essential for good leadership and leaders who exemplify them. This reflection on personal experience continued in the first set of formal discussions, where participants spend several minutes identifying and sharing personal examples of good or bad leaders, the character traits that define those leaders, and how they felt interacting with them. Throughout its programming, the GLI repeated such exercises—both through written reflection and reflective discussion (Colby et al., 2007)—to prompt reflection on personal experiences where participants have exercised, or failed to exercise, specific virtues, or where they have experienced the effects of others’ virtues or vices. This repetition, in turn, helped to foster a reflective practice that will inform their leadership and character beyond their time in the program (cf. Cantrell & Sharpe, 2016).

Reflective exercises have several pedagogical benefits (Colby et al., 2007). First, they build on Aristotle’s suggestion that “we ought to begin from things known to us,” including the “experience of the actions in life” (1999, 1095b4, 1095a2-4; cf. Annas, 2011; Burnyeat, 1980). Only when we have experience of what is virtuous and vicious can we discern how and why to act. Moreover, research shows that starting from what is “familiar” is especially important for adult learners who tend to process information and make connections more readily when they can draw on prior experience and connect what they are learning to real-life situations (Knowles et al., 2015; cf. Cantrell & Sharpe, 2016).

Second, intentionally setting aside time for reflection before the substantive conversation begins helps to ground more abstract conversations in concrete, lived experiences. This aligns with research suggesting that engaging experiences that are more “realistic,” “relevant,” and “salient” is more effective than analyzing abstract issues or hypothetical dilemmas (Walker et al., 1995, pp. 375-378, 403-404). Thus, the GLI provides readings, exercises, and discussions that encourage participants to integrate ideas about virtue, character, and leadership with their own values, experiences, and aspirations. To use Aristotle’s words, the primary purpose of the GLI is not “to know what virtue is, but to become good” (1999, 1103b27-30; cf. 1179b1-1180a6).

Finally, sharing personal reflections helps to build community, which, as we argue below, is a constitutive component of character formation. Sharing personal experiences gives participants a chance to know each other and develop bonds of trust, which is essential in a community where peers learn from each other and more experienced exemplars.

In the Nicomachean Ethics, Aristotle elevates the “wise” or “prudent person” (phronimos) as the model and standard of virtue, an exemplar whose character enables others to determine which action is virtuous and how to perform it (1999, 1107a1-3, 1140a25-1141b23). Philosophers since Aristotle have affirmed this emphasis on exemplarity. Recently, Linda Zagzebski (2017) has even developed an entire moral theory that places exemplars at the center.

Exemplars can serve several pedagogical functions. First, by embodying particular virtues, values, and ideals, exemplars offer role models to admire and emulate, which, as research shows, can elevate our moral vision, increase our motivation, and inspire us to emulate the actions, attitudes, or characters of those we admire (Algoe & Haidt, 2009; Colby & Damon, 1992; Colby et al., 2007; Cox, 2010; Immordino-Yang & Sylvan, 2010; Miller, 2018; Vianello et al., 2010; Zagzebski, 2017). Second, exemplars can serve as “counterfactual models” that prompt us to imagine how an exemplary person would act in a similar situation, which can help us discern how we should act (Miller, 2014, pp. 230-231; cf. Colby & Damon, 1992; Miller, 2018).¹⁰ Third, exemplars can deepen our cognitive understanding of how particular virtues, values, or ideals can be instantiated or realized in particular contexts, helping us understand what the virtue is and how it can be practiced (Miller, 2014, 2018). Fourth, exemplars can supply moral reminders that make norms salient and offer concrete, living proof that abstract ideals or virtues are actually possible to embody or attain (Colby & Damon, 1992; Colby et al., 2007; Damon & Colby, 2015). Seeing an exemplar exercise moral courage in the face of great obstacles, or make significant sacrifices to show compassion in the midst of suffering, proves that such acts are not only admirable but possible. Finally, exemplars may even “reshape our moral imagination” (Miller, 2018, p. 201), helping us see situations in new ways, encouraging us to challenge an unjust status quo, or offering “new insight into the ethical demands of ideals and how they can motivate people to do the right thing” (Engelen et al., 2018).

Exemplars, of course, can take many forms. Those living in our immediate context— parents, peers, teachers, coaches, and community leaders—are perhaps the most direct examples, but news reports and mass media also supply access to exemplars whom we can learn from and critically emulate, discerning what to emulate or avoid as we reflect on our own conduct and character (Annas, 2011; Miller, 2014, 2018). Moreover, exemplars need not be living. There is a long tradition of elevating examples from history, politics, science, literature, and the arts to encourage audiences to enact specific thoughts, actions, or virtues or develop a more ethical way of life (see Annas, 2011; Ivanhoe, 2007; McAdams, 2015; Miller, 2014, 2018; Zagzebski, 2017).

Empirical studies have affirmed the importance of exemplars for moral development.

Psychologists, for example, have shown that admiration of another’s virtue is a “profoundly motivating” emotion, fostering a “desire to be virtuous and to accomplish meaningful actions despite difficult obstacles” (Immordino-Yang & Sylvan, 2010, p. 110; cf. Haidt & Seder, 2009). Other studies have demonstrated that moral exemplars elicit feelings of “elevation” that increase motivations to “do good things for other people, become a better person oneself, and emulate the virtuous role model more generally” (Algoe & Haidt, 2009, p. 123). Studies on helping have shown that subjects exposed to examples of helping behavior are more likely to contribute or cooperate in follow-up scenarios than those in control groups.¹¹ In particular, researchers have demonstrated that “relevant” and “attainable” exemplars—those who are perceived to share similarity or proximity in age, stage, gender, cultural affiliation, values, interests, or profession—are particularly effective at increasing moral motivation and emulation (Han et al., 2017; cf. Colby et al., 2007; Lockwood & Kunda, 1997; Vos, 2018). Although scholars are still trying to determine precisely how exemplars inspire followers to habituate good character—whether through cognitive processing mechanisms or emotions of “elevation” or “admiration” (Algoe & Haidt, 2009; Kristjánsson, 2017; Miller, 2014; Vianello et al., 2010; Zagzebski, 2015)—there is a widespread consensus that emulating a good role model—actual or fictional, living or dead—is an effective way to inculcate good character.

The GLI provided access to exemplars in six ways. First, the GLI elicited reflection and conversation on personal exemplars, which aligns with research that recommends engaging participants with relevant and attainable exemplars before presenting stories from more extreme or distant exemplars (Han et al., 2017; Zagzebski, 2017). As mentioned above, the GLI’s recruiting conversations and application, along with the first substantive discussion, invited participants to reflect on exemplars of good and bad leadership from their own lives. Similar questions helped to ground discussions of the four focal virtues in light of personal exemplars.

Second, the GLI combined a focus on personal exemplars with readings from or about historic exemplars, including biographies, narratives, and letters that elevate specific leaders as examples of good or bad leadership. Readings, for example, included an excerpt from a biography of Nelson Mandela; analyses of historical or contemporary leaders in politics, business, law, and medicine; and letters and personal narratives from influential thinkers and leaders. On another occasion, the GLI provided admission to the Churchill War Rooms and asked participants to reflect on examples of leadership and character in the museum and imagine how they would respond if they were in Churchill’s shoes, utilizing the exemplar as a counterfactual model.

The GLI also assigned readings that highlight exemplars of the four focal virtues, deliberately selecting readings from a diversity of fields, backgrounds, and traditions to encourage relevance and attainability while facilitating discussions that honestly acknowledge leaders’ flaws, which prevents exemplary stories from being deflating or dispiriting. For vocation and service, for example, readings included a letter from the poet Rainer Maria Rilke (2001), an essay on vocation by the writer Annie Dillard (1994), a poem by Naomi Shihab Nye (1994a), and an excerpt of a speech by Martin Luther King, Jr. (1986). For gratitude and humility, participants read “Ozy-mandias” by Percy Bysshe Shelley (1977), a letter from Nelson Mandela (2010) on practices to cultivate humility, generosity, and service, and a reflection from Albert Einstein (2010) with insights relevant to all four focal virtues. By illuminating the larger context of exemplars’ lives, highlighting how practices, institutions, and communities influenced their moral development, and evaluating how they responded to particular challenges, these readings provided examples of virtues to emulate and challenged participants to reflect on how they might respond in similar situations.

Third, in line with Aristotelian insights on the importance of literature and the arts in moral formation (Carr, 2005, 2006a, 2006b; Sanderse, 2012), the GLI provided access to exemplars through the arts. During a day retreat in London, the GLI organized a guided tour of the National Portrait Gallery with a former director who led participants through various “portraits of leadership” to explore what we can learn from careful attention to how artists portray leaders from politics, science, sports, the military, and popular culture. The GLI also used exemplars from the stage and page to provoke reflection and conversation. For example, it organized a role-playing workshop on “King Lear and leadership” with a Shakespeare scholar before watching a live performance and discussing moral lessons from the play. It designed an optional “Ethics Through Fiction and Film” reading group to explore how characters in various novels and films grapple with moral complexity and engage our moral imaginations. And it organized an interactive workshop on “leadership and the wisdom of jazz” to consider what we can learn from watching members of a jazz quartet lead and follow, improvising and collaborating to achieve a collective goal they could not envision in advance.¹²

Fourth, the GLI invited speakers whose leadership might be relevant and attainable for postgraduate participants (Colby et al., 2007). GLI speakers included a nonprofit leader, an economist, a lawyer, a novelist, a former member of parliament, a senior business executive, university leaders, and professors in various fields. Since its students are so diverse, the GLI intentionally invited speakers from a diversity of countries, cultures, professions, ages, races, and genders to increase relevance and attainability, and it hosted speaker dinners in someone’s home to make the environment more intimate, private, and conducive to personal storytelling. In its invitations, the GLI explicitly asked speakers to share challenges and struggles to make their examples more relevant and attainable for students.¹³ The Q&A that followed often had the same features as the in-vivo, “assisted autobiography” that Colby and Damon use to analyze exemplars (1992, pp. 8, 17-18, 321-324).

Fifth, the GLI connected participants with mentors in their respective professional fields who met two to three times over 7 months to discuss their work, share challenges, and offer relevant guidance. The GLI ensured relevance by working directly with each student to identify potential mentors whom they admired or wanted to learn from based on their background, age, profession, or personal story (see Colby et al., 2007). Mentors included members of parliament, former cabinet ministers, professors, philosophers, journalists, peace-builders, and a former Archbishop of Canterbury. To focus mentoring conversations on character and the four focal virtues, the GLI supplied mentors and participants with a list of questions to guide their conversations. These questions make expectations explicit and encourage both parties to focus on morally relevant issues rather than simply on professional development.¹⁴

Along with professional mentors, the GLI also paired each postgraduate with a mentor from its staff, who met regularly with participants to discuss the program and reflect on values and vocation. Staff mentors, usually in their 30-40s, provided different forms of relevance and attainability, including similar educational backgrounds and greater proximity in age and stage of life (Colby et al., 2007). This two-tier mentoring structure enabled the GLI to address the various needs of participants while ensuring their mentoring conversations focused on character and the focal virtues. It also ensured continuity and depth of connection if professional mentors became busy and were unable to meet regularly (Colby et al., 2007).

Finally, GLI leaders sought to model the virtues they desired to cultivate in participants. The GLI specifically trained facilitators to model the behavior they hoped to elicit, such as respectful engagement, personal storytelling, vulnerability, and the four focal virtues.¹⁵ Such modeling increases relevance and attainability and aligns with research on how peer role models can actively encourage prosocial behavior (Cox, 2010; Han et al., 2017). Combined, these efforts provided various opportunities for participants to critically engage various kinds of exemplars.

A fourth strategy involves discussions of particular virtues and how they can be practiced in concrete circumstances. Dialogue is among the most common ways that contemporary people engage moral issues (Walker et al., 1995), and Kristjan Kristjánsson (2014a) has shown how it is central to the Aristotelian tradition. Open-ended dialogue provides opportunities to discuss complex moral issues, share practical experiences, test theoretical ideas, and learn from others’ perspectives. In particular, dialogue can help us understand why specific virtues are important and how they can be developed, practiced, or applied in diverse contexts. It can increase our cognitive understanding of character, deepen our emotional awareness and practical wisdom, and provide occasions to practice specific virtues and skills in conversation with others (Athanassoulis, 2016; Kristjánsson, 2014a; Noddings, 1994).

Empirical research supports the role of dialogue in moral formation (Athanassoulis, 2016; Berkowitz & Bier, 2004; Higgins, 1980; Schlaefli et al., 1985; Schuitema et al., 2008; Snarey & Samuelson, 2014). Influenced by Lawrence Kohlberg’s model of ethical reasoning about complex cases, educators have long used discussions of difficult ethical dilemmas to improve moral judgment and development (DeHaan et al., 1997; Higgins, 1980; Schlaefli et al., 1985; Snarey & Samuelson, 2014). Dilemma discussions can be helpful for illuminating moral complexity, identifying general principles, and considering the implications of various ethical issues, but reasoning about rare, abstract, and often hypothetical dilemmas tends to overemphasize the cognitive aspects of moral development and obscure the more ordinary ways that everyday relationships, contexts, and communities shape our character.¹⁶

In recent years, character educators have directed discussions away from analyzing ethical dilemmas to increasing “virtue literacy,” the “capacity to know and understand the necessary language and virtue concepts required to evaluate morally salient situations” (Arthur et al., 2017, p. 94). This emphasis fits with Aristotle’s approach to analyzing specific virtues in the Nicomachean Ethics. “For if we discuss particular aspects of character one at a time,” Aristotle writes, “we will acquire a better knowledge of them” (1999, 1127a16-17). Helping participants understand the “moral and conceptual grammar” of a particular virtue such as gratitude, for example, can highlight how the virtue differs from the simple emotion, attitude, or act of gratitude and reveal how the virtue can be practiced appropriately and discriminately in diverse circumstances (Morgan et al., 2015, p. 108). Such discussions are particularly important within an Aristotelian approach that requires doing the right action in the right way for the right reason at the right time and place (Morgan et al., 2015).

The GLI incorporated such dialogue in various ways. First, it structured engagement with visiting speakers not as one-sided lectures but as open-ended conversations where speakers could share personal experiences in a safe and intimate setting and where participants could ask questions that elicit discussion about how they have exercised particular virtues in particular contexts. Second, the GLI structured one-on-one mentor meetings as dialogues where both mentors and participants dialectically explore how they have responded to challenges and exercised leadership and character in their respective roles. Finally, and most significantly, the GLI organized its program around face-to-face, structured dialogues about leadership and the four focal virtues. Since research indicates that “the most productive discussions (whether peer or teacher-led) are structured, focused, [and] occur when students hold the floor for extended periods of time, when students are prompted to discuss texts through open-ended or authentic questions, and when discussion incorporates a high degree of uptake” (Soter et al., 2008, p. 373, cited in Kristjánsson, 2014a, p. 348), the GLI selected a number of short readings to structure small group discussions on the virtues. Often, these readings served different functions, such as deepening moral and conceptual understanding, generating an affective connection, highlighting a relevant exemplar, or providing a critical analysis that helped participants make judgments about when a particular virtue is appropriate. In line with recommendations from Morgan et al. (2015) regarding gratitude, for example, the GLI assigned poems by Mary Oliver (1992) and Rumi (2004) that foster self-awareness and affective connections, along with a more critical article by Barbara Ehrenreich (2015) that highlights the potential dangers and limits of gratitude and its relation to other concepts. With these texts as prompts, facilitators engaged participants dialectically rather than didactically, providing an opening question or exercise to prompt conversation but otherwise encouraging students to share their experiences and perspectives. While facilitators occasionally followed up by asking an open-ended question, making a connection, or offering insights on a conceptual aspect of a particular virtue, they positioned themselves not as teachers but as coparticipants in a shared learning community, and they tried to create conditions where participants feel safe and empowered to share their own personal experiences, ideas, or disagreements. This dialogical approach allowed all participants to deepen their understanding of the relevant focal virtue, make connections to their personal experiences, and coconstruct meaning, understanding, and community with others in the conversation.

One way to expand this dialogical approach would be to actually teach participants how to facilitate constructive dialogues about virtues. Between 2014-2017, the GLI primarily focused on either facilitating or modeling constructive dialogue, but more explicit instruction on how participants can facilitate dialogue themselves could help them develop a crucial practice of leadership and increase their knowledge of particular virtues. For example, the GLI could introduce participants to a particular model of dialogue—such as “deliberative dialogue” (Harriger & McMillan, 2007), “sustained dialogue” (Saunders 2001; Sustained Dialogue Institute, 2019), or “transactive discussion” (Berkowitz & Gibbs, 1993)— that illuminates various stages and modes of effective discussion. Then the GLI could invite each participant to facilitate at least one dialogue with their friends or classmates on a particular virtue to gain practice of how to lead dialogues and deepen their knowledge of the virtue. As educators know from experience, being required to teach or lead a discussion about a topic increases knowledge and familiarity, which, in this case, could aid virtue literacy.

A fifth strategy for cultivating good character involves fostering awareness of how situational variables, cultural influences, and institutional incentives shape character and behavior. Around the turn of the century, a number of philosophers and psychologists began drawing on empirical studies to suggest that global character traits may not even exist. According to these “situationists,” our moral thoughts, feelings, and actions do not reflect stable global traits of character, but the influence of specific and often subconscious situational variables and biases that condition our behavior more than any underlying disposition (Doris, 2002; Harman, 1999, 2000). Over the last decade, scholars have provided convincing replies to situationism, showing various ways in which the empirical studies that underwrite situationism do not preclude the existence of stable dispositions of character.¹⁷ While these scholars resist situationism as a global critique of virtue, they often affirm its cautionary lessons. Even if situational variables and biases are not the sole determinants of moral emotions, attitudes, and actions, they are one such determinant, and recognizing their influence encourages us to take more care in attributing particular virtues or vices to individuals and recognizing the impact that situational factors can have on our character (Adams, 2006; Annas, 2011; Athanassoulis, 2016; Herdt, 2015; Miller, 2014; Samuels & Casebeer, 2005; Snow, 2010). As Christian Miller (2014) argues, “If there are a number of psychological processes which (i) often operate subconsciously or outside our conscious awareness, (3) have important implications for moral behavior, and (3) can prevent that behavior from having moral worth or can even lead to the performance of morally forbidden actions, then a natural strategy to use in trying to become a more virtuous person is to first become better aware of and familiar with these processes” (p. 233). According to Miller (2014), awareness of these psychological processes can enable us to “be more attuned to situations in which they might be activated, and work to compensate for, correct, or counterbalance them” (2014, p. 233; cf. Snow, 2010; Webber, 2016).

Such a strategy goes back to ancient Athens. To become virtuous, Aristotle argues,

Implicit in Aristotle’s advice are two strategies supported by empirical studies. The first is what Miller describes as “getting the word out,” which involves informing participants about how particular psychological tendencies or situational variables—such as the tendency to overestimate our abilities or be influenced by the behavior of those around us—might encourage or inhibit ethical action without our conscious awareness (2014, pp. 233-236; 2018, pp. 209-214; cf. Athanassoulis, 2016; Doris, 2002; Grant, 2010; Samuels & Casebeer, 2005; Walker, 2000; Wilburn, 2007). One study, for example, shows that college students who heard a lecture on the “bystander effect”—the tendency to avoid helping someone when others are present—were significantly more likely to help someone in a staged emergency later that day (67%) than those who did not hear the lecture (only 27%) (Beaman et al., 1978; cf. Miller, 2018). A follow-up study showed that the difference (42.5% compared to 25%) persisted when the staged emergency was 2 weeks after the lecture (Beaman et al., 1978; cf. Miller, 2018). These and other studies suggest that getting the word out can alert individuals to situational variables and biases that they can correct or resist (Samuels & Casebeer, 2005).

A second and related Aristotelian strategy is “selecting our situations,” where we “actively seek out those situations which are going to inspire us to act well, while actively avoiding those situations that are fraught with temptation and other pitfalls” (Miller, 2018, p. 204; cf. Athanassoulis, 2016; Doris, 2002; Webber, 2016). This strategy is implicit in Aristotle’s advice to “steer clear” of vices or situations that might be most dangerous for us (1999, 1109a25-35). Citing a number of philosophers and psychologists who endorse this strategy, Miller (2018) suggests that we can, for example, actively seek out friends, role models, and communities that might have a positive influence on us, while avoiding those whose example and influence might be harmful.

Since we possess various biases and encounter endless numbers of situational variables every day, character education programs must be selective about how they tailor these strategies to the developmental stages and situations of participants. Since the GLI worked with postgraduates training to be leaders in their respective professions, it focused on professional expectations and incentives embedded in different occupations and institutional cultures. To begin the conversation, the GLI showed participants a TED talk by a social psychologist who uses a rigged Monopoly game to reveal how positional inequalities of wealth, power, and privilege—even when determined by random chance—change how participants behave within a particular setting (Piff, 2013). Participants then discussed readings that analyze how particular institutional incentives, cultures, and situations shape individual character, including an analysis of the “occupational hazards of working on Wall Street” (Lewis, 2014), a summary of research showing that political leaders become less empathetic and more “coldhearted” as they become more powerful (Inzlicht & Obhi, 2014), and an analysis of “demoralizing institutions” in medicine, law, and banking (Schwartz &Sharpe, 2010, pp. 197-228). Afterward, the GLI divided participants into groups according to their profession to discuss the “occupational hazards” that accompany their chosen field. By “getting the word out” about how situational variables and institutional incentives condition their character, either positively or negatively, the GLI sought to help participants correct or counteract them and become more intentional about selecting their situations.

Immediately following this discussion, the GLI explored particular practices that can enable participants to reflect on their tendencies, hold themselves and others accountable, and develop the habits needed to resist or reform their institutional cultures. For practices of solitude and reflection, participants read a poem on the “art of disappearing” (Nye, 1994b), a lecture on “solitude and leadership” (Deresiewicz, 2010), and a letter from Nelson Mandela (2010) explaining how 15 minutes of daily meditation was crucial to his personal development. For practices of self-development and self-accountability, participants reflected on a poem about the stages of self-knowledge necessary to overcome bad habits (Nelson, 2007), a letter from Kurt Vonnegut (2015) on the power of art, music, and poetry to “make your soul grow,” and an excerpt from Benjamin Franklin’s autobiography (2005) that charts his efforts to cultivate 13 virtues of character. With these examples as inspiration, participants then discussed the practices they have used in the past to gain critical distance, resist or reform their institutional cultures, or make difficult decisions in morally fraught situations. Here, conversations about situational variables incorporate other character development strategies such as habituation through practice, reflection on personal experience, engagement with exemplars, and moral reminders. The GLI could build on these conversations by drawing on Mary Gentile’s action-oriented approach in Giving Voice to Values, which suggests students can develop the confidence and “moral muscle” to overcome situational pressures by framing their “self-story as individuals who want to voice their values” and constructing and practicing personal “scripts” that function as appropriate responses to questionable practices in their professional fields (2010, pp. 169, xiii).

Finally, to connect “getting the word out” and “selecting situations” with an emphasis on exemplars, the GLI arranged a lunch conversation with a former member of parliament who rose to power as a cabinet minister but later committed perjury to save face during a scandal and served 7 months in prison. He spoke candidly about his original motivations for entering politics, how the trappings of power deformed his character, and how his time in prison enabled him to see the potentially distorting effects of political power. This wide-ranging, personal engagement enabled participants to connect the morning’s discussions of institutional incentives and personal practices with the practical experience of a relevant and attainable exemplar.

Throughout the program, the GLI also discussed how other implicit biases and assumptions—such as those related to gender, race, religion, and culture—influence views about leadership or the focal virtues. By including readings, speakers, and exemplars from diverse perspectives and traditions, the GLI sought to highlight how leadership and character can be understood and enacted from various points of view. In this way, the diversity reflected in GLI programming helped to prompt increased awareness about how biases and assumptions might affect the understanding and practice of leadership and character in contemporary society.

Another effective way to challenge the effects of situational variables and cultivate good character is by providing moral reminders that make particular norms salient (Athanassoulis, 2016; Mazar et al., 2008; Miller, 2014, 2018; Shu et al., 2011). By increasing the salience of norms and recalling commitments to maintain particular values as members of specific communities, moral reminders alert us to the morally salient features of situations, “call our attention to our moral commitments,” and “make it much more difficult in our own mind to justify doing the wrong thing” (Miller, 2018, p. 134). In other words, moral reminders create psychological barriers to self-justification and self-deception. Because we want to see ourselves as virtuous people, being reminded of our values, standards, and commitments makes it psychologically difficult to violate them without updating our “self-concept,” recategorizing the situation, or rationalizing the behavior to avoid cognitive dissonance (Mazar et al., 2008; Miller, 2014; Shu et al., 2011). For this reason, moral reminders help to counteract temptations that might arise from situational variables and reinforce a commitment to acting in accordance with our internal standards. Indeed, one of the values of professional codes of conduct is that they provide moral reminders that encourage professionals to act in accordance with widely shared ethical norms.

Miller (2014, 2018) synthesizes empirical research on academic cheating to highlight the effectiveness of moral reminders. In one study, seating students directly in front of a mirror and playing a recording of their own voice (in contrast to a random voice) reduced cheating on an intelligence test by reminding students of who they see themselves to be (Diener & Wallbom, 1976). In the “self-aware condition,” only 7% cheated, compared to 71% in the “non-self-aware condition” (Diener & Wallbom, 1976, p. 110). Other studies show that students who were asked to recall the Ten Commandments or sign an honor pledge before taking a test were significantly less likely to cheat when given the opportunity than those not provided with such reminders (Mazar et al., 2008; Shu et al., 2011). The importance of making norms salient highlights one important function of honor codes on college campuses: honor codes not only provide enforceable rules to live by but remind students of their values and commitments to live up to them (Mazar et al., 2008; Miller, 2014, 2018; Shu et al., 2011).

The GLI sought to promote self-awareness and make norms salient from the time students applied to the program. Most significantly, the GLI’s extended schedule of readings, reflections, discussions, and speakers every week or two provided regular and consistent reminders to participants about the importance of character and the four focal virtues, increasing participants’ self-awareness and making it more likely that participants aspire to be leaders of character and less likely that they act unethically when situational variables arise. The value of making norms salient thus provides an important reason to extend a character development program over several months rather than condense it into a weekend retreat. Regular readings, discussions, and speakers provide consistent moral reminders that help participants strengthen their moral commitments, resist temptations, and internalize norms around character. The GLI could further incorporate moral reminders by offering more extensive discussions of relevant professional codes of conduct and by encouraging students to adapt moral reminders in their efforts to habituate virtue through practice, for example, by using sticky notes, calendar reminders, or digital notifications to prompt gratitude journaling or habituate their chosen virtue for a month.

The influence of moral norms highlights a final feature of moral formation: character development never occurs in isolation. We inevitably shape and are shaped by the culture and community around us, whether we recognize it or not (Annas, 2011; Aristotle, 1999; Colby & Damon, 1992; Damon & Colby, 2015; Hunter, 2000). This is why Aristotle (1999) holds that even the most self-sufficient people need good friends.

On an Aristotelian account, friendships contribute to the “cultivation of virtue” in several ways (Aristotle, 1999, 1170a12). For example, friendships embody and inculcate particular practices, values, and narratives, which shape friends’ understanding of moral character (Aristotle, 1999; Brewer, 2005; Kristjánsson, 2014a). By encouraging specific behaviors and making moral norms salient, friends help each other recognize and develop a shared conception of a good life and the virtues needed to pursue it. In particular, friendships provide a context for the “sharing of conversation and thought,” which enables friends to determine what kind of action or character is “choiceworthy” (Aristotle, 1999, 1170b6-19; cf. Brewer, 2005; Kristjánsson, 2014a; Sherman, 1987). As friends discuss ideas and experiences with each other, they may come to know aspects of each other better than they know themselves, which can prompt reflection and provide a mirror that helps them see themselves in a new light (Aristotle, 1999; cf. Brewer, 2005; Kristjánsson, 2014a; Millgram, 1987; Sherman, 1987; Wilburn, 2007). Relatedly, friendships provide an “opportunity for beneficence” that enables friends to practice acts of virtue (Aristotle, 1999, 1155a6-10; cf. Sherman, 1987). With occasions to “do good” with and for others, friends come to habituate good thoughts, feelings, and actions and thus develop a more virtuous character through regular practice (Aristotle, 1999, 1171a22-27, 1171b13-28; cf. Sherman, 1987).¹⁸ Moreover, friendships provide access to relevant and attainable exemplars who can teach us how to improve our character. Since “we are able to observe our neighbors more than ourselves, and to observe their actions more than our own,” the virtuous person needs “virtuous friends” (Aristotle, 1999, 1169b34-1170a4; cf. Sherman, 1987). Friends also supply assistance and support when others encounter difficulty since “we have our pain lightened when our friends share our distress” (Aristotle, 1999, 1171a22-1171b4; cf. Sherman, 1987). Finally, friends supply “mutual correction” when the other goes wrong (Aris totle, 1999, 1172a11-14).¹⁹ Accountability contributes to moral development directly by countering certain forms of behavior and indirectly by shaping norms, values, and expectations, for “each molds the other in what they approve of, so that ’[we learn] what is noble from noble people”’ (Aristotle, 1999, 1172a13-14; cf. Sherman, 1987).

As this last quote implies, simply having friends or being a member of a community is not enough to acquire good character: the moral quality of these relationships matters. As Aristotle writes, “the friendship of base people turns out to be vicious.… But the friendship of decent people is decent, and increases the more often they meet. And they seem to become still better from their activities and their mutual correction” (1999, 1172a9-14; cf. 1172a4-8). The influence of common activities is among the reasons Aristotle prioritizes friendships based on virtue rather than “utility” or “pleasure” (1999, 1156a6-1157b14, 1158a3-10; Cooper, 1977).

Recent empirical studies on “group effects” support these Aristotelian insights. Studies on helping behavior show that when individuals are surrounded either by strangers, or by participants who do not assist a “victim” in the experiment, they are less likely to help, but when they are surrounded either by friends or participants who actively assist the victim, they are more likely to help (for overviews, see Miller, 2013, 2014, 2018). While these studies measure behavior rather than dispositions, they suggest that friendships can either “inhibit” or “enhance” certain behaviors by providing models of moral behavior, generating norms that promote accountability, and encouraging common activities (Miller, 2013, pp. 142-149). One neuroscience study even suggests that brain waves—and the patterns of perception, interpretation, and behavior they reflect—are highly similar among friends (Parkinson et al., 2018). Since research suggests that relationships with family, colleagues, and friends provide the most common context for ethical decisions and the most common source of moral exemplars (Walker et al., 1995), friendship becomes an especially important form of moral development.

Friendship is especially relevant for a leadership initiative that seeks to challenge “heroic” models of leadership and shift attention to the everyday contexts in which we might exercise influence. One of the most striking features of the moral exemplars studied by Colby and Damon (1992) is that they were actively receptive to, and shaped by, the influence and feedback of their followers and friends. Friendships provide a developmental context for ethical leadership, facilitating various forms of cooperation and exchange, providing increasing levels of support, accountability, and congruence, and encouraging more collaborative, egalitarian, and transformational forms of leadership (Boyd & Taylor, 1998; Romero-Iribas & Martínez-Priego, 2011; Shushock, 2008). To challenge heroic, hierarchical, and individualistic models of leadership, one scholar has even proposed the metaphor of “leadership as friendship” to encourage a more “participatory,” “interdependent,” and “relational conception and practice of leadership” (Perrault, 2005).

Building on these insights, the GLI encouraged meaningful friendships in four ways. First, the GLI provided a community in which members had access to teachers, exemplars, and peers who share similar values and commitments and whose example can indirectly encourage ethical thought and action. These relationships provided opportunities for support and emulation and helped to make communal norms around leadership and character more salient.

Second, the GLI intentionally organized opportunities for participants to develop strong friendships by pairing formal discussions with informal opportunities to talk or share a meal before or after an event. This provided occasions for participants to share the “conversation and thought” that Aristotle identifies as conducive to friendship, which in turn encourages good behavior and offers occasions to practice particular virtues.

Third, the GLI led a discussion on friendship as an intentional practice of leadership. Participants were invited to reflect on what they value about their most meaningful friendships and then discuss several assigned readings, including an excerpt from Augustine’s Confessions (2009) that highlights friendship’s various functions, a poem by Oriah Mountain Dream (1999) that encourages honesty and vulnerability, and a letter from Clementine Churchill (2013) that exemplifies a friendship of accountability. These readings offer both content about what friendship entails and an example of how it is enacted by relevant exemplars.

Finally, the GLI provided occasions for participants to practice mutual correction and accountability. By enacting vulnerability, inviting questions during discussion, and structuring conversations as dialogues among peers, GLI facilitators sought to create a warm and trusting space that promoted mutual engagement and encouraged openness, vulnerability, and a diversity of perspectives, which challenges participants to reflect on their views in light of others’ experiences and adjust their assumptions accordingly. Moreover, during group conversations and mentoring meetings, GLI leaders gently challenged particular perspectives from participants and highlighted moments when participants’ professed values or commitments might not match their practice. Further, specific readings provided examples of how participants can practice such candor and correction among themselves. One of the most interesting aspects of Benjamin Franklin’s efforts to cultivate virtue, for example, is that he added “humility” only after a close friend told him that he was “generally thought proud” (2005, p. 73). Future cohorts of the GLI could further expand this approach by creating “accountability partners” among participants, requiring each participant to meet monthly with one other participant to offer candid feedback, provide support, and practice accountability.

The purpose of this article has been to present seven strategies of character development grounded in theoretical and empirical research and supply practical examples of how this research can inform a program for postgraduate character development. Preliminary evidence from a case study of the GLI in its first 3 years (2014-2017) suggests that, when these seven strategies are combined and applied consistently over time, they are productive in helping students understand and cultivate important character virtues.²⁰ Since its inception, the GLI has continued to iterate on its practical programming, expanding the number of participants, increasing the number of focal virtues, and developing new ways to integrate these seven strategies. Similar work at Wake Forest University has integrated these seven strategies in both undergraduate and professional courses and programming, presenting new opportunities both to apply these strategies of character development and to assess their impact on students (Lamb et al., 2021). More theoretical and empirical research is needed to refine and assess each strategy of character development as it applies to emerging adults and to evaluate how these strategies can be combined in a comprehensive program to cultivate virtue. Such programs will inevitably need to be adapted to fit different institutional, educational, and cultural contexts and the developmental needs of diverse participants, but we hope to have provided some theoretical and practical resources that can aid educators who seek to support postgraduate students at a crucial time in their moral development.²¹