Meaning in Mentoring: More Than Meets the Eye/“I”

Introduction

This conceptual article extends theoretical understanding of the key Human Resource Development (HRD) intervention of mentoring. The article moves research in a new theoretical direction by introducing and developing a particularly fruitful theoretical perspective to the mentoring literature. By adopting a systematic and critical literature review, we introduce aspects of the identity and psychodynamic literature into the established canon of mentoring literature. It will be shown that what “meets the eye” in mentoring, such as knowledge acquisition and skills development, is only part of the story. A significant but as yet neglected part of the story is, the article argues, the conscious and unconscious meaning-making process occurring within mentoring relationships as both mentors and protégés strive to gain self-understanding and to construct an acceptable sense-of-self, a sense of “I.” This “meaning in mentoring” is examined as the key to understanding how mentoring really works. The new theoretical perspective developed from the identity and psychodynamic literature also serves to unsettle the comfortable status quo of mentoring being seen as unquestionably a “good thing.” It will be seen how mentoring can inadvertently work to constrain rather than enable personal development. The article concludes by examining the implications of the new insights for mentoring in practice and for future empirical inquiry.

Mentoring is a perennially popular personal and career development method (Ragins & Kram, 2007) that has recently burgeoned as a formal HRD intervention within organizations (Corner, 2014; Ghosh & Reio, 2013). However, informal mentoring has also become important in light of the growth of fluidity of work (Petriglieri, 2011) and the associated boundaryless careers (Fullick-Jagiela, Verbos, & Wiese, 2015). These changes in employment patterns have increased the importance of less structured and nonformal workplace learning methods (Hamlin & Stewart, 2011; Mizzi & Rocco, 2013). Concomitantly, all forms of mentoring receive a lot of empirical research attention with Ragins and Kram (2007) noting an “explosion of research interest that crosses disciplines” (p. 4) and Allen, Eby, O’Brien, and Lentz (2008) finding a “tremendous surge” in mentoring research (p. 343). The “sheer volume of literature on mentoring” (Ghosh, 2013, p. 144) has supported several substantial literature reviews. So, given that mentoring has been extensively studied, why is yet another article needed? This article contributes by making two distinctive contributions to understanding mentoring and to informing mentoring practice.

First, the article contributes by taking a critical perspective on workplace mentoring to deepen understanding and thereby enable more informed mentoring practice. Certain strands of the mainstream mentoring literature show some criticality in, for example, recognizing that mentoring can be dysfunctional with negative outcomes including overdependence and antagonism (e.g., Allen, Lentz, & Day, 2006; Young, Cady, & Foxon, 2006). However, while a more broadly critical take on mentoring has been called for (Opengart & Bierema, 2015), research is generally characterized by a positive, positivistic, and performative perspective (e.g., Ehrhardt & Ragins, 2019). In a comprehensive review of more than 200 mentoring studies, Allen et al. (2008) found almost universal reports of mentoring having only beneficial outcomes. For example, Fullick-Jagiela et al. (2015) noted the empowerment associated with mentoring, and mentoring gurus Ragins and Kram (2007) reported the “appeal of mentoring as a . . . transformational relationship” (p. 4). The practitioner-orientated literature reports in a similar tone, the “miracle” of mentoring (Ingleton, 2013) and the “magic” that can result (Ramalho, 2014, p. 177). While mentoring is less researched from the mentor perspective than the protégé perspective, studies of the mentor perspective similarly emphasize benefits such as how the relationship meets mentors’ “generative needs” (Kram, 1983, p. 610). Such findings align with the general tenor of mainstream HRD research as masculine and performance driven (Collins, 2012). By contrast, this article aligns with the emerging Critical HRD movement (Callahan, Rigg, & Sambrook, 2015) and particularly examines the power imbalance within mentoring. From our Critical HRD perspective, we respond to Collins’s (2012) calls to avoid complacency with the status quo, to be critical, and to avoid hegemonic agency in favor of change agency. We, therefore, aim to surface and to challenge prevailing assumptions, to foreground neglected themes, and to advance new and alternative ways of understanding and practicing mentoring.

Second, we contribute by moving beyond empirical description toward theoretical explanations. Despite theoretical advances in the social sciences in general, Ragins, Cotton, and Miller’s (2000) observation that mentoring research was predominantly empirical remains the case today. In their seminal handbook of mentoring, Ragins and Kram (2007) commented that scholars “continue to struggle with understanding . . . we know it works [but] we are still grappling with why, when and how” (p. 4; see also Young et al., 2006). More recently, in reviewing the literature of informal mentoring, Janssen, Vuuren, and Jong (2016) noted that the “literature . . . lacks theoretical explanations” (p. 507). Janssen et al. concluded that while findings were abundant, the underlying mechanisms remained a “black box” with theory-based studies being desperately needed (see also Ehrhardt & Ragins, 2019). Ghosh (2013) similarly remarked that although mentoring was “an often recognised HRD tool [this] does not imply that it is a deeply understood one” (p. 145). Moreover, where theoretical explanations of mentoring have been proposed, such as relational mentoring theory (Ragins, Lyness, & Winkel, 2010) or, of particular relevance to this article, professional identity theory (Allen, Eby, Poteet, Lentz, & Lima, 2004), these remain underdeveloped and have limited explanatory leverage.

Significantly, the article responds to Ragins and Kram’s (2007) as yet largely unanswered call to “bring new theoretical lenses to the discourse on mentoring” (p. 4). By theory, we mean quite simply, in relation to mentoring, a system of concepts that explains how mentoring “works” or fails to work and that is capable of predicting how and why mentoring might work or might go awry. The particular theoretical contribution made in this article is to show the relevance of social constructionist identity theorizing and of key facets of Lacanian psychodynamic theorizing for providing a better understanding of mentoring. Such theorizing responds to calls for an ontological shift from the mainstream focus of mentoring inquiry on the outcomes of mentoring toward a focus on the processes, that is, the dynamics, reciprocity, and mutuality of mentoring interactions (Janssen et al., 2016).

Traditionally, the focus of HRD, and workplace mentoring, has been upon enabling the acquisition of knowledge and fostering the development of skills required for occupational capability. However, it is increasingly recognized that occupational capability needs to be understood in more processual terms, as requiring the construction of new ways of understanding the self (Bryans & Mavin, 2003), that is, the construction of an effective occupational identity (Cameron & Grant, 2017; Muir, 2014). Poulsen (2013), thus, noted how the key and generic occupational capability of leadership “needs to be focused on leaders’ sense of identity” (p. 350). Identity theorizing is highly developed in the organization studies literature and is also used in the adult learning and educational literatures (e.g., Cameron & Grant, 2017), and the importance of identity formation in mentoring is widely acknowledged (e.g., Janssen et al., 2016; Muir, 2014; Mysyk, 2008). However, despite Ragins’s (1997) observation that identity was a fruitful area for further mentoring research, nearly 20 years later, Fullick-Jagiela et al. (2015) noted “the extent to which mentoring relationships contribute to a protégé’s sense of professional identity has not been examined thoroughly” (p. 498).

Therefore, the question guiding this theory-building research is, how can mentoring be understood for both protégés and mentors in terms of conscious and unconscious identity development processes? Young et al. (2006) suggested that understanding mentoring required multiple theoretical lenses, and we do not claim that the understanding developed here is an alternative that can replace established understandings. Indeed, our review begins with a critical analysis of established research and the associated theoretical understandings and then shows how the identity theorizing at the center of this article builds upon certain strands of established work. However, identity theorizing does offer additional important new insights, particularly into what are acknowledged to be less understood facets of mentoring such as the mutuality of benefits and dysfunctional outcomes (Ghosh & Reio, 2013; Ragins, 1997). Moreover, the theorizing offers leverage in understanding both formal and informal mentoring relationships.

The article now proceeds as follows. First, the literature review method is described. Second, a thematic overview of the corpus of mentoring literature is provided, and established ways of theorizing mentoring are discerned and evaluated. Third, the identity literature is analyzed, social constructionist theorizing of identity is developed in detail, and this theorizing is then applied to understanding mentoring. In particular, the identity-work literature is critically reviewed to understand how mentoring “works” for mentors and protégés. Fourth, the emerging literature adopting a psychodynamic perspective to understanding identity is critically reviewed to provide deeper insights into the identity processes invoked within mentoring relationships. Fifth, conclusions are drawn, showing how the identity literatures reviewed provide a fresh perspective on how mentoring relationships work. Implications for the practice of mentoring in organizations are developed alongside these conclusions. Finally, the implications for empirical inquiry are outlined.

Method

We adopted an integrative reviewing procedure influenced by the step-by-step framework for literature reviews in HRD outlined by Opengart and Bierema (2015). As explained in the introduction, the article critically examines the potential of identity and psychodynamic theorizing to extend the understanding of mentoring. However, it was initially necessary to overview the substantial corpus of mentoring literature built up across several disciplines over the past four decades to ascertain and evaluate established theorizations. This initial phase of our work was facilitated by the existence of several systematic reviews of the corpus of mentoring research. These reviews include Allen et al. (2008), Ghosh (2013), Fullick-Jagiela et al. (2015), and Janssen et al. (2016) and the comprehensive handbook of mentoring research edited by Ragins and Kram (2007).

Using these established reviews as guides, the EBSCOhost and ERIC databases were searched within the date range 1980 to 2018 for sources that theorized mentoring. An analytical matrix was produced in which each selected source was allocated a row and the adjacent column was then used to identify the type of theorization used. The matrix rows were subsequently organized so that sources using the same or similar theoretical approaches were clustered. In a second column, each researcher, working independently and based upon her and his personal reading of each source, commented on the extent of development of these theoretical strands. A third column was used to record assertions made by source authors of where theoretical development was still needed. The two researchers’ comments on the extent of theoretical development were then compared, anomalies examined and resolved, and judgments reached regarding the extent of development of particular theoretical strands. An agreed, verified, meta-matrix was compiled, which, as will be discussed, revealed that while diverse theoretical understandings of mentoring have been proposed, there is limited theoretical development of any one strand.

A key contribution of our review is, as noted, that it integrates identity and psychodynamic theorizing into the context of mentoring. As Ghosh (2013) noted, a focused, integrative review “can yield new insights and significant re-conceptualizations” (p. 147). Therefore, the next stage of our reviewing method was to focus on sourcing and reviewing those strands of the identity and psychodynamic literatures providing traction for understanding mentoring. Our searches on EBSCOhost and ERIC using search strings such as “mentor*” AND “identity*” within source titles and abstracts and with no date restrictions revealed just six sources (e.g., Muir, 2014; Mysyk, 2008). A snowball follow-up of references within the six sources yielded four further relevant sources. However, a search using “learn*” AND “identit*” revealed several hundred potential theoretical sources. Searching within these search results for items of relevance to learning within social interaction still left a vast corpus of identity and learning literature. To ensure a more manageable review, we used the insights provided within Black, Warhurst, and Corlett (2018). Black et al. (2018) showed how various traditions of identity theorizing deepen understanding of many areas of concern to HRD and pointed to the potential of the particularly well-developed area of social constructionist identity theorizing for providing fresh insights into mentoring roles and relationships. We, therefore, focused on this specific strand of identity theorizing, notably, literature conceptualizing identity-work and identity-regulation. We also concentrated on emerging studies that offer more critical insights deriving from psychodynamic perspectives that can be applied in understanding the identity processes invoked within mentoring. Sources were independently screened against the key criteria of their potential for understanding learning within social interactions. In total, 53 papers, including papers from the 12 most-cited researchers within this strand of identity theorizing and within organizational psychodynamic theorizing, were selected and form the basis of our contribution.

Mentoring Research: Empirical Strengths and Theoretical Limitations

Empirical research on mentoring is flourishing, as noted in the introduction. While this article makes distinctive contributions, to appreciate these contributions, it is necessary to understand the current and seemingly healthy state of mentoring research. This process has to begin with defining the construct of mentoring. Even after three or more decades of systematic research, a key challenge in examining mentoring lies in the fact that “a definitive answer to the question ‘what is mentoring?’ remains elusive” (Short, Cameron, Morrison, Piip, & Ebrahimi, 2014, p. 3). How mentoring is defined varies somewhat across the literature (Opengart & Bierema, 2015) and between different fields. Ghosh and Reio (2013) discerned more than 40 different definitions noting that mentoring is a multidimensional concept. Commonalities in the more established and widely cited definitions depict mentoring as a hierarchical developmental relationship, that is, as a sustained and purposeful one-to-one relationship between a mentor with advanced experience, knowledge, and wisdom and a protégé with less experience, for the purposes of professional development and personal growth. However, with the recognition of various distinct types of mentoring including informal mentoring (Ragins & Scandura, 1999), and peer and reverse mentoring (Megginson & Clutterbuck, 2006; Thurston, D’Abate, & Eddy, 2012), there has been a requirement for a broadening of the definition. Moreover, as societal norms have changed (Mysyk, 2008), the definition of mentoring has been evolving. Opengart and Bierema (2015), thus, discerned three common attributes of contemporary definitions: first, and differing most significantly from traditional definitions, reciprocity or mutuality of exchange; second, regular consistent interaction over time; third, developmental benefits. It is, therefore, recognized that mentoring is a two-way process, and that “mentoring may not be publicly recognised or observable” (Ragins, 1997, p. 484) but comprises “critical learning moments” between diverse partners (Muir, 2014, p. 369). Protégés may, thus, receive mentoring support from a network of individuals concurrently.

In parallel with the evolving definition of mentoring, research themes have ebbed and flowed. Several broad themes are discernible across the empirical literature. Some of these themes, such as investigations of mentoring outcomes for protégés, have persisted over time, whereas others, such as investigations into mentor benefits, have come to the fore more recently. As these themes are effectively examined in the literature reviews previously mentioned, for example, Ghosh (2013), we simply note here the significance of certain of the themes for the newer identity-focused understanding of mentoring that we will develop. In essence, the established literature can be characterized by examinations of “inputs” and “outcomes” with certain works also modeling the relationship between inputs and outcomes. Studies with a particular focus on “inputs” have examined mentoring types, such as informal and peer mentoring (e.g., Megginson & Clutterbuck, 2006; Thurston et al., 2012); mentoring functions, such as career and psychosocial functions (e.g., Ghosh, 2013; Kram & Isabella, 1985); phases or stages of mentoring (e.g., Kram, 1985); mentor attributes (e.g., Allen et al., 2008; Haggard, Dougherty, Turban, & Wilbanks, 2011); mentoring requirements at particular career stages (e.g., Kram & Isabella, 1985); the quality of mentoring relationships (e.g., Ragins & Cotton, 1991).

Studies with a focus on “outcomes” are found to be particularly prevalent (Janssen et al., 2016). Outcomes attributable to mentoring that are typically reported to be significant from the protégé perspective include upward career mobility, better reward packages, greater career satisfaction, enhanced engagement, and improved commitment (e.g., Ehrhardt & Ragins, 2019; Kram, 1985; Thurston et al., 2012). Importantly, too, in terms of outcomes, a considerable literature shows how mentoring serves to redress structural disadvantages such as obstacles to career progression experienced by women and minorities (e.g., Ragins & Cotton, 1991; Ragins et al., 2000; Sahoo & Lenka, 2016). While, traditionally, the mentor perspective has been neglected (Young et al., 2006), the early work of Kram and Isabella (1985) showed the complementarity of mentoring relationships and, as noted, more recently, research has emphasized the reciprocal and “mutual beneficence of the relationship” (Opengart & Bierema, 2015, p. 243). Specifically, Ragins (1997) found that mentors gained a sense of “generativity,” that is, “the sense of contributing to future generations” (p. 494; see also Jones, 2012). Finally, from the organizational perspective, it is widely reported that key outcomes attributable to mentoring include enhanced performance and productivity (e.g., Emelo, 2010).

It is typical for studies, particularly from the outcomes perspective, to conceptualize mentoring in terms of a provision-acquisition nexus. Thus, a widely reported finding is that mentors provide job-related advice and pass on information along with accumulated personal tacit knowledge and wisdom to protégés (e.g., Allen et al., 2006; Poulsen, 2013) and also enable protégés to “acquire critical technical skills” (Kram, 1983, p. 617). For example, Mysyk (2008) noted how mentors typically have built up a “library of lessons” from their experiences to “teach” their protégés (p. 212). Similarly, in reverse mentoring, less experienced colleagues teach more experienced colleagues, such as their official mentors, how, for example, to use new technologies (e.g., Megginson & Clutterbuck, 2006; Thurston et al., 2012). While such a provision-acquisition conceptualization is limiting, findings such as those that mentors provide acceptance and confirmation (e.g., Allen et al., 2004), access to useful networks (Ragins et al., 2000), and “appropriate job attitudes” (Mysyk, 2008, p. 212), are important pointers to the theoretical understanding that we develop.

While Empirical research on mentoring is flourishing, the theorization of mentoring is less developed. In their extensive literature review, Janssen et al. (2016) noted how “mentoring research has been criticised for one shot empirical data” (p. 507), and they highlighted the lack of conceptualization and “clear understandings of underlying mechanisms . . . [into] why mentoring processes are effective.” The dominant objectivist ontology of mentoring research has undoubtedly hindered the emergence of theory, with the subjectivist, processual ontological assumptions required for the emergence of theory being a distinct blind spot in the literature (Janssen et al., 2016).

The theorization of mentoring clearly requires strengthening, as this article will shortly turn to do; it has though to be acknowledged that certain theoretical advances have been made in recent years. First, mainstream organizational theorizing has been applied to mentoring. For instance, self-determination theory (Deci & Ryan, 1985) has been used to show how mentoring relationships help both parties fulfill the three basic human needs of autonomy, competence, and relatedness (Janssen et al., 2016). Similarly, theories of disadvantage and diversity provide a degree of leverage in understanding the utility of mentoring for groups that are underrepresented in leadership roles (Poulsen, 2013; Sahoo & Lenka, 2016). Second, theories have been developed specifically within mentoring inquiry such as “relational mentoring theory,” which is used to account for the mutuality of mentoring (Ragins et al., 2000). However, certain proposed theories, such as relational mentoring theory itself, fail to meet the defining criteria of theory that were outlined earlier. Criteria used in defining theory when applied to mentoring suggest a need to explain how and why mentoring works, to understand how the relationship between “cause” (mentoring) and “effect” (e.g., achievements) works and, particularly, to understand how mentoring can go awry. In what follows, we examine theorizing aligned with these defining criteria, in the form of practice-based learning theory, social constructionist identity theory with its associated concepts of identity-work and identity-regulation, and psychodynamic theories, to provide deeper insights into “how mentoring works” in practice.

Theorizing Identity to Explain the Meaning in Mentoring

Although the theorization and conceptualization of identity within the mainstream mentoring literature is still in its early days, that mentoring influences identity development is evident over the history of mentoring research. Therefore, while identity has yet to be effectively theorized in mentoring research, the contribution of mentoring to identity development is evident in findings that mentoring helps protégés to make sense of themselves in their roles (e.g., Chrobot-Mason & Thomas, 2002); to become accepted, validated, and confirmed in their roles (e.g., Kram & Isabella, 1985); and to develop self-confidence, self-esteem, and self-efficacy (e.g., Cameron & Grant, 2017). Further evidence that mentoring contributes to protégé identity formation comes, for example, in reports of mentors challenging and questioning protégés’ assumptions, providing feedback, and guiding their critical reflection so as to examine their values and philosophy of practice (Ghosh, 2013). That is, in essence, protégés learn lessons about the self from mentors and thereby find personal meaning in mentoring (Mysyk, 2008). For example, Muir’s (2014) empirical work demonstrated how leadership mentoring enabled each protégé to achieve a “deeper awareness of self” and thereby “know who one is . . . [to] fully realise that he or she is a leader” (p. 362). Protégés are also enabled to envision themselves in new and future roles (Fullick-Jagiela et al., 2015).

While the systematic theorizing of identity outcomes or processes associated with mentoring is underdeveloped, several theories used in mainstream mentoring inquiry form a useful bridge to the systematic social constructionist identity theorizing that we develop below. For example, social exchange theory has been cited extensively in mentoring studies (Young et al., 2006) and explains the mentoring relationship in terms of conferring rewards for both parties simultaneously. A further area of theorizing cited in established mentoring literature, social learning theory (Bandura, 1986), similarly moves toward social constructionist identity theorizing. In social learning theory, role modeling enables protégés’ assimilation of particular ways of being in relation to others in specific workplace communities (Warhurst, 2011). In particular, mentors can enable protégés to enter communities that might not otherwise be accessible and position protégés favorably within a nexus of relationships (Kram & Isabella, 1985). However, such theorizing suffers the same, and limiting, provision-acquisition explanatory nexus detailed earlier. For example, identity is understood as something that can be modeled and copied. Therefore, and following an extensive review of the mentoring literature, Janssen et al. (2016) called both for much greater clarity in the conceptualization of identity in mentoring and for more processual understandings. We now turn to this task.

Brown (2017) noted how the literature on identity is increasingly vast and heterogeneous (p. 297), and Corlett, McInnes, Coupland, and Sheep (2017) remarked on the “diversity of the field of identity scholarship” (p. 262). This growth reflects the increasing desire in neoliberal societies for individual autonomy (Brown, 2015) whereby identity is not so much ascribed but, rather, requires an individual struggle to create and sustain a preferred sense-of-self (Mizzi & Rocco, 2013). Such is the diversity of the field that in his seminal review article, Alvesson (2010) discerned seven images of identity in the literature, and Brown (2015) subsequently noted four key debates. Brown’s debates included whether or not identity is stable or dynamic and over- or under-socialized. These debates provide dimensions upon which a specific identity theory can be positioned. While not mutually exclusive, particular theories give more traction on specific HRD issues than others (Warhurst, Black, & Corlett, 2018), and given the sheer vastness of the literature, a tight focus will be needed on the theoretical perspective with most potential for understanding mentoring.

Identity and Learning Through Social Interaction

The potential of identity for comprehending learning, and, therefore, mentoring, became apparent with the rise of “practice-based” learning theory in the 1990s associated with Lave and Wenger’s (1991) seminal work. This theorizing has since had a “significant impact on the way in which learning is conceptualised” (Christensen, 2016, p. 125). To understand the contrast with traditional learning theories, Sfard (1998) differentiated learning theories as aligning with the metaphor of acquisition or of participation. Whereas traditional learning theories see occupational learning as involving the “acquisition” of knowledge and cognitive capabilities, newer “participation” learning theories see occupational learning as involving the assimilation of particular ways of knowing and doing intertwined with the appropriation of particular ways of being within specific communities of practice (Nicolini, Scarbrough, & Gracheva, 2016).

Learning, therefore, crucially, involves building identities (Lave & Wenger, 1991) with the learned being those who are able to identify themselves and be identified by peers as a capable person within the community or communities that sustain the practice in question (Cameron & Grant, 2017). Learning requires “legitimate peripheral participation” (Wenger, 1998), that is, a trajectory of gradual but progressive exposure so as to assimilate ways of doing, and, crucially, to appropriate ways of being, within the target communities (Gerken, Beausaert, & Segers, 2016). As seen earlier, mentors play an important role in enabling protégés to access key individuals and communities and to gain their acceptance. Mentors are, therefore, enablers of legitimate peripheral participation for identity formation. While important in moving the understanding of learning into new and productive territory, judging by the significant citation indices achieved by such works, situated learning theory has a couple of noted limitations. First, whereas learning typically involves a degree of conflict as established power balances are destabilized, situated learning theory is silent on such matters (Nicolini et al., 2016). Second, the processes through which the assimilation of a particular way of being, in other words, identity learning, occurs is inadequately theorized (Warhurst, 2006). However, what follows shows how the social constructionist theorization of identity and the associated constructs of identity-regulation and identity-work deepen understanding of power relations in learning and mentoring, and how identity learning occurs both for protégés and for mentors themselves.

Identity: A Social Constructionist Perspective

Identity can be a slippery concept (Alvesson & Robertson, 2016; Caza, Vough, & Puranik, 2018), and so a definition is required of what the concept means in social constructionist identity theory. Traditionally, identity was studied from a psychological perspective and was seen as developing through distinct stages into adulthood to become established and durable (Hutchins & Rainbolt, 2017). By contrast, the more contemporary micro-sociological, social constructionist perspective offers more traction in understanding mentoring (Muir, 2014) through bridging between psychological and sociological understandings of self (Watson, 2009). Specifically, an interpretivist poststructuralist approach is useful, and this approach “emphasizes the constructed and constructing view of identity” (Alvesson & Robertson, 2016, pp. 8-9). The self is understood not as a distinct and given psychological entity but, rather, as the way people perceive themselves, that is, their self-concept (Muir, 2014). Identity is an ongoing consciously reflexive project that involves a constantly constructed narrative (Gill, 2015). This narrative involves individuals making subjective claims both for themselves and for others regarding who they are (Caza, Moss, & Vough, 2017), and who they want to be with the aim of achieving existential coherence, continuity, and security. In other words, individuals seek meaning about who they are and who they can be. Individual “characteristics” and intentions undoubtedly influence who a person is. However, characteristics and intentions cannot entirely define the individual as the individual’s social and cultural contexts and temporal positioning both enable and constrain who the individual is or who he or she can become.

Therefore, the social constructionist identity perspective shows how identities are never secure and achieved but always “provisional . . . perpetual works in progress” (Clarke, Brown, & Hailey, 2009, p. 344) not least because individuals cannot simply claim and assert an identity. Identity is, rather, crafted (Watson, 2008) through interaction with, and for, others at specific times and in specific situations (Thomas, Hardy, Cutcher, & Ainsworth, 2014). Preferred identities are typically fragile and precarious being under constant scrutiny and threat from the judgment of others (Humphreys and Brown (2002) who challenge, contest, deny, or ignore those preferences (Alvesson & Willmott, 2002; Brown & Coupland, 2015). Negotiation with others in the community of practice is required to secure acceptance of preferred ways of being (Mizzi & Rocco, 2013). Identity claims are particularly influenced by various social identities or subject positions that might be imposed upon individuals by, for example, their occupational roles (Watson, 2008) and the associated expectations of others such as mentors of their protégés, and protégés of their mentors. Identity is, thus, intersubjective, and Caza et al. (2017) noted, “we come to know ourselves through our interactions with others” (p. 4) or, as Brown (2015) remarked, “the self is fuelled by the identities it feeds on” (p. 4). In sum, identities are created, defended, and shaped within networks of relationships at particular moments of time (Alvesson, Ashcraft, & Thomas, 2008). An individual cannot, therefore, simply become or be who she or he wants to be but is, rather, to some extent, defined by relations of power within groups.

Mentoring and Identity-Regulation

While power relations are not ignored in established understandings of mentoring, the treatment typically sees power, as with knowledge or skills, as an entity conferred by the mentor and acquired by the protégé. For example, Ragins (1997) reported that mentors help protégés develop their power resources through providing reflected power and improving protégés’ visibility, and their influencing capabilities. However, from the current social constructionist identity perspective, power can be seen as more pervasive in mentoring relationships. Not only are identities socially influenced and possibly constrained by the behaviors and actions of others upon whom a particular way of being depends, as just discussed, identities are also subject to social power relations in a more profound manner. Identity is dialogically constructed through internal soliloquy and with conversational partners (Brown & Toyoki, 2013) using the social resource of language. Following the “linguistic turn” in social sciences, language is seen as not merely representational but as constitutive (Reedy, 2008), which means, in the case of identity, that language does not mirror identity but creates or reveals identity (Sveningsson & Alvesson, 2003). In other words, people think and talk themselves into existence. As noted, a person’s identity lies in her or his “capacity to keep a particular narrative going” (Giddens, 1991, p. 54), and we are, quite simply, a story (Watson, 2009). Applying this insight to mentoring, it can be seen, for instance, that through narrating stories emphasizing their development of others, mentors form and evolve a satisfying sense-of-self (Mysyk, 2008). Through supporting such stories, protégés acquire a sense of being “learners” and of being secure in this learner identity, avoiding the excessive exposure that typifies many contemporary work roles.

However, many workers are found to be reliant for their sense-of-self on narrating only occupational stories (Petriglieri, 2011) and are thereby vulnerable to managerial control. Such stories draw upon discourses that are imbued with power that can limit, distort, or regulate the individual’s sense-of-self. For instance, Reedy, King, and Coupland (2016) found that “work organizations are . . . sites of determination rather than autonomy” where dominant identities are promoted, even imposed, and discourse is the primary medium of such control and power (Brown & Lewis, 2011; Collins, 2012). These discourses rule in and rule out certain ways of talking and of being, constructing individuals as certain kinds of people (Huber & Brown, 2017). In this way, discourses act as “technologies of the self” (Brown & Toyoki, 2013, p. 888), colonizing individuals from the inside in creating normalized, engineered selves (Clarke et al., 2009). Individuals succumb readily to social influences (Reedy et al., 2016) and are subjected unconsciously to the managerial control of subjectivity (Driver, 2009a, 2009b). Moreover, individuals might even be consciously complicit in their own subjugation if the identities on offer appear alluringly advantageous (Elsbach, 1999). Within high power-distance cultures (Hofstede, 2002), protégés may be particularly susceptible to such shaping or distortion. Such compliant ways of being may constrain more authentic ways of being that would be healthier for both individuals and organizations. To illustrate, an HR developer who has graduated from a School of Education may work to see herself and be seen by others as an “educator” first and foremost with a mission to enable employees to achieve their true potential. However, this idea and ideal of self may be challenged and constrained by the senior management of her firm who may define her as a “trainer” who is no more than someone who ensures that employees can meet tightly defined performance standards. The developer may be complicit in accepting this limiting self-definition on the basis that it gives her managerial status.

Through propagating and inculcating such dominant and defining discourses that serve to create individuals who are mere stencils (Alvesson, 2010) of managerially approved work identities, mentoring may serve to reinforce the regulation of protégés’ identities. The manipulative behavior discerned in certain mentoring relationships (Young et al., 2006) might amount to the suppression of autonomous and authentic, desired ways of being in favor of more docile, managerially acceptable ways of being (Cameron & Grant, 2017). Moreover, mentors are likely to be complicit in shaping protégés to be the sort of person they are themselves, that is, someone who fits harmoniously within the established organizational community. Mentors also typically enable access to networks and communities, and, in doing so, they may be contributing further to identity-regulation. Networks and communities offer their own specific set of readily available discourses as resources for individuals’ identity construction (Alvesson & Robertson, 2016). Although the vast field of social identity theorizing (e.g., Tajfel & Turner, 1985), and the related self-categorization theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) and positioning theory (Harré, 1998), were developed from a contrasting identity perspective, these theories add to our understanding of how individuals have a proclivity to draw on discourses from particular communities in defining themselves. A tendency is noted for individuals to “locate the self [with]in socially recognizable categories” (Mizzi & Rocco, 2013, p. 378) and to align with favored groups and disassociate from those deemed undesirable (Nazar & van der Heijden, 2012). In the example above, while the developer might adopt a limiting sense-of-self through redefining herself as a mere trainer, she may nonetheless gain a sense of security from identifying with her organization’s senior management. These related identity theories show the propensity of individuals to construct a sense-of-self in alignment with or positioned toward favored groups and then, once accepted within such groups, to exchange a private identity for one provided by the group (Brown, 2017). This process of individuals identifying with, and deriving, their sense-of-self from a group is particularly likely where the target group is perceived as having high social status (Gill, 2015).

Mentoring Enabling Protégés’ Identity-Work

Within social constructionist identity theorizing, the concept of identity-work explains how individuals are able both to resist regulating forces and to overcome, or at least cover over, insecurities and to thereby construct preferred identities (Brown & Coupland, 2015; Gill, 2015). This identity-work concept provides a particularly strong explanation of how mentoring works. As noted, contemporary identities are subject to both challenge and denial by others and are dependent upon affirmation from others. Moreover, at any one point in time, identity is fragmented (Garcia & Hardy, 2007) and kaleidoscopic (Humphreys & Brown, 2002). Therefore, while individuals might attempt to present themselves as confident and coherent, it is widely reported that identities are insecure, precarious, and fragile with vulnerability, anxiety, and inner conflict being widespread (Brown, 2015; Collinson, 2003; Hoedemaekers, 2010; Nicholson & Carroll, 2013; Reedy et al., 2016). While mentors can, as will be seen, assuage protégés’ insecurities, mentors can be vulnerable themselves. For instance, Kram (1983) found that mentors “may feel threatened” by a protégé’s “continued success or opportunity for advancement” (p. 622). Therefore, identity-work is also important for mentors themselves.

A widely cited definition of identity-work is that provided by Sveningsson and Alvesson (2003) who defined the process as one whereby people are “engaged in forming, repairing, maintaining, strengthening or revising the constructions that are productive of a sense of coherence and distinctiveness” (p. 1165). Through identity-work, people “create, adapt, signify, claim and reject identities from available resources” (Brown, 2017, p. 298) and attempt to “influence the various social identities which pertain to them” (Watson, 2009, p. 431) in attempting to secure a degree of “existential continuity and security” (Alvesson & Willmott, 2002, p. 622). Given the precarity of identities, identity-work is continually needed and is typically seen as a perpetual struggle (Clarke et al., 2009; Driver, 2018). A considerable literature has emerged demonstrating the utility of identity-work in understanding a broad range of individual and organizational phenomena, and in a recent comprehensive review of the identity-work literature, Brown (2017) discerned five distinct approaches to understanding identity-work (see also Caza et al., 2018). While the concept has yet to be systematically applied in understanding mentoring, within the discursive and psychoanalytical approaches to identity-work discerned by Brown (2017), certain types of identity-work (e.g., McInnes & Corlett, 2012) would seem to have particular traction. In particular, and reflecting the contemporary rise of individualism, “differentiating identity-work” (Huber & Brown, 2017) and “aspirational identity-work” (Thornborrow & Brown, 2009) seem germane. Such identity-work types reflect individuals’ craving for authenticity, distinctiveness, and a sense-of-achievement. Individuals see certain identities as prestigious, attractive, and, therefore, preferred, representing an ideal version of the self to be aspired to and worked for (Brown & Toyoki, 2013). Certain occupations or organizations that are perceived as prestigious, attractive, or distinctive provide forums for such identity-work (Learmonth & Humphreys, 2012). In facilitating acceptance within these occupations or organizations, mentors contribute to protégés’ differentiating and aspirational identity-work. Mentors might particularly contribute to protégés’ progression within such occupations and organizations through enabling the trialing of possible and provisional selves (Coupland & Brown, 2012).

Whereas, as seen, the individual might be, to some extent, limited, even determined, by the discursive resources that society and their organizations in particular make available to them, through identity-work, there is some scope for discursive maneuvering and of resistance. Discourses can be creatively exploited or transformed (Watson, 2009). Professionals and managers specifically might be particularly capable of reflexively appropriating and exploiting organizational discourses in constructing favored ways of being (Brown & Lewis, 2011). Importantly too, individuals draw upon discursive resources from beyond the workplace in their self-authorship (Driver, 2009a). Thus, whereas contemporary work is typically an important source of identity (Alvesson, 2010), with occupations and organizations providing resources for identity construction, individuals have a wealth of alternative resources for identity construction. Occupations may, therefore, be bypassed in individuals’ identity stories (Alvesson & Robertson, 2016). Mentors with a degree of reflexive awareness can assist protégés in resisting the power of managerially imposed discourses and enable protégés to discover alternative discourses in shaping the self through agentic identity-work.

Making sense of the self is a key element of all types of identity-work (Corlett & Mavin, n.d.), and mentoring interactions have been found to be invaluable identity-workspaces for such meaning-making (Muir, 2014). Identity-work for developing self-understanding is particularly needed in just those situations where mentoring is most likely to occur, such as when a protégé is in a new or evolving situation and is, therefore, likely to have suffered a loss of meaning. For instance, extending the example of the HR developer working to achieve her desired educator identity: Rather than simply passively absorbing the firm’s management discourse of “performance” and identifying herself as a trainer, the developer might, with the help of a mentor, actively adapt the management discourse of “performance” in constructing the desired educator sense-of-self. The mentor might enable the developer to recognize that employee “performance” is more likely to be secured if employees are not merely compliant but are creative and enterprising, and that these latter capabilities are far more likely to be cultivated by an “educator” than by a “trainer.”

Mentoring Enabling Mentors’ Identity-Work

Ghosh and Reio (2013) posed the question, “what might prompt individuals to volunteer their time for mentoring?” (p. 113). This question can be answered in terms of mentoring contributing to mentors’ own differentiating or aspirational identity-work. Being a mentor might in itself represent an aspired identity and certainly an identity carrying prestige and offering differentiation perhaps at a time when the mentor’s own career is plateaued or in decline (Black & Warhurst, 2018). Within the mainstream mentoring literature, one of the key “expected benefits of being a mentor” is found to be “recognition by others” (Allen et al., 2006, p. 274). Through mentoring and developing new talent within the organization, the mentor gains status (Ragins, 1997), recognition, and respect and receives self-confirmation (Kram, 1983). Moreover, being a mentor provides personal satisfaction (Young et al., 2006), feelings of fulfillment (Ragins, 1997), and “a sense of generativity” (Ragins et al., 2000, p. 1179).

Mentors, thus, gain personal meaning in mentoring through constructing a positive sense-of-self as a worthwhile and significant person. However, such an identity might not be particularly secure; mentors can, as noted earlier, feel displaced by their protégés (Ragins & Scandura, 1999) and threatened by protégés’ advancement (Kram, 1983). Therefore, mentors may respond by forever taking on further protégés in the hollow hope of compensating for their weakened sense-of-self. Moreover, mentors might themselves be prone to passively absorb dominant discourses in organizations, defining themselves using readily available language that serves to regulate who they are and that restricts who they might become.

Therefore, mentors might need to bolster their own sense-of-self by undertaking agentic remedial, repairing, or restorative identity-work (Knights & Clarke, 2014; Lutgen-Sandvik, 2008). Such identity-work processes will require critical self-examination of whose interests are ultimately being served by adopting, by default, prevailing ways of being (Alvesson & Willmott, 2002). Mentors might seek and evaluate different discourses for themselves and explore alternative, personally valued ways of being. In sum, identities are, for both mentors and protégés, “a continued dialectic of ‘structure’ and ‘agency’” (Brown, 2015, p. 7). However, further explanation is needed as to why, in the increasingly individualistic cultures characterizing developed societies and with reflexively aware mentors, agency does not simply trump structure in identity development. The second, and related, area of theorizing with particular utility for understanding mentoring, the psychodynamic approach to identity, provides a fuller explanation of the structural constraints and limits on agency.

Mentoring, Identity-Work, and the Unconscious

Identity comprises both conscious and unconscious elements (Hoedemaekers, 2010), and individuals are subjects of the unconscious such that conscious constructions can never totally define them (Driver, 2009b). Brown (2017), thus, noted “our inability to know ourselves” (p. 305). The psychodynamic approach to identity emphasizes the extent to which people are constrained by forces operating unconsciously that they might work to recognize and live with, but are incapable of controlling (Corlett et al., 2017). The role of the unconscious is touched upon in the established mentoring literature (e.g., Young et al., 2006) but understanding such facets of identity in mentoring has not received systematic attention. The psychodynamic tradition enables understanding of such unconscious facets of identity and is a useful supplement to the discursive identity tradition. Within the psychodynamic tradition, facets of Lacan’s extensive corpus of work (e.g., Lacan & Fink, 2002), and, particularly, Driver’s (2009a, 2009b, 2017, 2018) interpretations of Lacan’s work, provide a basis for understanding mentoring relationships in terms of identity-work.

Lacan’s work shows how the conscious narration of identity can never quite capture how identity is experienced. Driver (2009b), thus, noted, “we cannot say who we are” (p. 56). In Lacanian terms, this is explained through a distinction being drawn between the “symbolic order” and the “real.” The symbolic order is the domain of language through which identity is discursively constructed but such discursive constructions are not the “real.” A real, or prelinguistic, primordial “I” exists only outside of language. Knowing the self only through language locates existence outside of the self as in the child’s experience of the mirror image of itself as itself (Roberts, 2005). The self is both constructed and constrained within the symbolic order, within this limited and flawed medium for self-expression, and is thereby alienated from itself, never able to fully realize its desires (Driver, 2009a). That a real “I” cannot be expressed in the symbolic order, that is, in the words of others, results in an unconscious desire for an imaginary “I,” that is, for an “I” that is more real. For both mentors and protégés, discursive aspirational identity-work may contribute to the construction of an imaginary “I.”

This imaginary I, or self, though, is both alluring and entrapping. The imaginary self is a self-disciplined and controlled self that is thereby ultimately a subjugated self. Contemporary individualistic discourses of supposedly boundless possibilities fuel the belief that with sufficient striving and using, perhaps, alluring and prevalent symbolic resources from the corporate world such as those of “leadership” (Roberts, 2005), a satisfying imaginary self can be secured. Control of individuals thereby becomes, conveniently for those in power, subsumed within the self. Moreover, the conscious, desiring “I” that strives to construct an imaginary self is typically disrupted by the unconscious “I” (Driver, 2009b), thereby exacerbating a sense of inauthenticity and of failure to realize who we truly are (Driver, 2009a). Given the enormity of desire, identity-work to achieve an imaginary “I” is never finished. The individual is, therefore, forever prone to never quite conscious feelings of failure or of lack. Mentors typically “raise the bar” for protégés, encouraging their aspirations (Megginson & Clutterbuck, 2006). However, in the absence of understanding from psychodynamic approach, such actions may merely be fueling insecurity and lack.

“Lack” is a key concept in Lacanian psychoanalysis and has particular relevance for understanding mentoring. Primal fulfillment, the original prelinguistic sense of being whole, complete, and “real,” is inexpressible in the symbolic order of language. Therefore, the self is never adequately fulfilled, imaginary identity constructions necessarily fail, and “that which truly and uniquely marks the person is a loss or lack” (Driver, 2009b, p. 58). Individuals engage in relentless but fruitless attempts to “gloss over” this “irreducible kernel of lack” (Driver, 2017, p. 719). Indeed, Roberts (2005) cautioned of the “deathly desire to fix and stabilize identity” on the grounds that all there is in life is lack and the only way to a worthwhile life is to live with lack (p. 639). Might mentors, in supporting and guiding protégés’ pursuit of desired identities, be unknowingly complicit in fueling the delusion that the “deathly desire” to find wholeness and fulfillment can be conquered and that lack can be overcome?

Identity is, as noted, socially constructed, often contested, and thereby fragile. The psychodynamic perspective further elucidates this interdependency, showing how identity “desperately needs . . . and depends upon others” (Hoedemaekers, 2010, p. 381). The perspective, thereby, offers further insight into how mentoring relationships unfold. Individuals seek to compensate for lack through gaining others’ recognition and become rivals for recognition with actual or imaginary others (Roberts, 2005). Rivalry for recognition results in comparisons not only with others but also with old “versions” of the self. The individual may exaggerate differences with others, and with an old self. Perceived inadequacies are then unconsciously displaced into others or into past versions of the self in attempting to compensate for the unresolvable lack in life (Driver, 2018). However, an unreflected need for recognition in the eyes of others, or in comparison with previous selves, merely fuels fragility and anxiety. The important implication of this understanding is that unconsciously, mentors can be seen to “need” protégés as much as protégés “need” mentors for conquering lack and securing a seemingly satisfactory sense-of-self. However, “need” is not a sound basis for security. Attempts to cover over or to defeat unresolvable lack and to secure a desired and stable sense-of-self are precarious, and self-defeating, resulting merely in fantasies of wholeness, and the replacement of one desired, but elusive, imaginary way of being with another (Driver, 2017). By contrast, the only sustainable way forward is adopting a reflexive stance involving an acceptance and creative engagement with lack. The individual can then cease to strive to “become” someone and then someone else, and so on, but can, rather, find enjoyment in the liminal state of just “being” who they are (Driver, 2017). But what does this mean for mentoring in practice? In turning now to draw conclusions, we address this question.

Conclusion and Implications for HRD Practice

Mentoring in various forms is, as we have shown, widespread and is regarded as one of the most effective HRD interventions. Not surprisingly, therefore, the literature of mentoring is extensive. However, while it is clear that both protégés and mentors themselves derive tangible benefits from mentoring, it was seen how numerous scholars have been concerned by the absence of systematic theorization in the area. Moreover, the theorization of mentoring that does exist is typically descriptive, and the need for theorization that can contribute both to explaining how mentoring works and to predicting its outcomes is widely acknowledged (Ehrhardt & Ragins, 2019; Ghosh & Reio, 2013; Ragins & Kram, 2007). We also found that the mentoring literature is overwhelmingly positivistic, performative, and, suspiciously, positive. For instance, empirical research typically seeks a simple cause-effect relationship and purports to discern straightforward enhancements to both individual and organizational performance (the “miracles” and “magic”) that are attributable to mentoring.

Our research question arose from these concerns and our suspicions, and this question was, how can mentoring be understood for both protégés and mentors in terms of conscious and unconscious identity development processes? In answering this question, we avoided replicating the numerous, and perfectly satisfactory, reviews of the mentoring literature that have analyzed and mapped the corpus in various ways (e.g., Allen et al., 2008; Janssen et al., 2016). Instead, we completed a focused and selective review and adopted the new perspective of Critical HRD. Our review integrated key strands of the mentoring, identity, and psychodynamic literatures in responding to calls for the theorization of mentoring. We now highlight three key theoretical contributions that enhance explanations and improve predictions for how mentoring works. In extending the theoretical base of mentoring, these contributions have clear implications for mentoring policy and practice, and these are examined in what follows, with the implications for empirical research in mentoring being considered subsequently.

The first key contribution lies in our articulation of an identity perspective on mentoring whereby occupational learning is understood in terms of identity development. Through the systematic theorization of identity from a social constructionist perspective, we have moved the understanding of mentoring into a more fruitful, processual, direction (Janssen et al., 2016) where, in particular, the mutuality of mentoring is more effectively explained than has been the case to date. Drawing upon situated learning theory (Lave & Wenger, 1991), we established that while not denying the importance of acquiring knowledge and skill in occupational learning, a crucial, and as yet neglected, facet of occupational learning is the cultivation of ways of being, that is, of cultivating identities, within practice communities. Adopting an antirealist ontology that the social world is produced in and through interaction (Christensen, 2016), identity was conceptualized as socially constructed, relational, and intersubjective, being crafted, negotiated, and maintained within everyday communities, and at particular moments (Watson, 2008). As gatekeepers to communities, mentors can facilitate protégés’ cultivation of desired identities.

It was seen that identity-work is particularly necessary given the contemporary precarity of identity in organizations. Identity-work contributes, as was seen, to overcoming the insecurities and anxieties that arise nowadays from identity being less an ascription and more an achievement (Alvesson, 2010), to countering identity challenges, and to constructing a desired sense-of-self. It is thus predicted that mentoring will become more important than ever for the existential purpose of enabling individuals to ascertain a clear trajectory of becoming and to secure a desired way of being in work and, in turn, in life. A key implication for mentoring practice is the need for mentors to enable protégés to understand that a sense of vulnerability is as normal as a sense of security and that by acknowledging vulnerability, the protégé can become more receptive to learning and change. Furthermore, mentors need to recognize that being a mentor primarily involves enabling protégés to understand themselves, that is, who they are currently striving to be and who they could possibly be. In particular, mentors might scaffold protégés’ aspirational identity-work in their attempts to fashion distinctive and ideal versions of themselves (Thornborrow & Brown, 2009). Mentors can, moreover, assist with envisioning possible selves and trialing provisional selves (Coupland & Brown, 2012).

While, as our review showed, the effects of mentoring on mentors themselves has, until recently, been underresearched, we conclude that a key effect for mentors is that mentoring enables mentors’ own identity-work. Whereas, the mainstream literature has depicted mentors as providers of benefits to protégés (e.g., Allen et al., 2006), from the identity perspective, the benefits of mentoring are revealed to be reciprocal. Mentoring has the potential to bolster a mentor’s sense of security in her or his identity through conferring status and credibility among peers, and through providing a feeling of “generativity,” that is, of bringing on the next generation at a time when the mentor’s own occupational sense-of-self may be fragile and in need of rejuvenation. An important contribution of our inquiry, therefore, lies in showing that mentors need protégés as much as protégés need mentors; identity theorizing explains that the two parties are interdependent in constructing a satisfactory sense-of-self. The psychodynamic perspective extends this understanding, finding that the self is conceptualized as constituted in the gaze of others and that recognition by others is a condition of existence of the self (Roberts, 2005). Therefore, the proclivity of experienced employees to become mentors is more readily understood. The key implication of this contribution for mentoring as an HRD practice lies in mentors needing to acknowledge the reciprocity of benefits and to thereby cultivate circumspection and humility. However, that mentors might be threatened by their protégés’ successes points to a further implication for mentors in that support is needed for remedial, repairing, or restorative identity-work (Knights & Clarke, 2014); mentors may themselves need mentoring to secure a satisfactory sense-of-self.

The second key contribution arising from our theorizing predicts that mentoring might serve to entrap and regulate as much as enable identity development for both parties despite the best of intentions. As shown, identity claims are typically directly challenged, contested, or constrained by others (Alvesson & Willmott, 2002). For instance, subject positions are assigned according to role and often, too, according to race, class, and sexuality (Mizzi & Rocco, 2013). Being a protégé has the potential to constrain identity, locking the individual into a subservient subject position. However, being a mentor might also constrain identity, carrying connotations of being a leader rather than being a learner in a world where survival requires not merely leaders but lifelong learners. Identity is also constrained, or at least limited, indirectly and in more pernicious and pervasive ways by the unreflexive appropriation of dominant discourses. It was noted earlier that language is constitutive rather than representational (Reedy, 2008), and identities are conceptualized as constructed through narratives drawing upon available discourses. Brown and Lewis (2011) found that discourse is the primary medium of organizational power and control. Discourse is a technology of the self, being readily available, beguiling, and normalized. Managerially favored discourses colonize workers from the inside through being incorporated into self-narratives. The result can be docile, disciplined, and corporately scripted, stenciled subjects with a deluded and dependent sense of security (Alvesson, 2010; Huber & Brown, 2017). It is all too easy for mentors to be complicit with power, regulating protégés’ potential through encouraging the unquestioned absorption of limiting managerial discourses into their self-narratives (Collins, 2012), a process that Alvesson (2010) referred to as constructing identity for performativity.

Therefore, and in terms of practical implications, mentors should be critically self-reflexive (Cunliffe, 2016) as to who they themselves are attempting to be, through examining the discourses they weave into their own narratives of self. In turn, mentors then need to aid protégés to develop self-awareness (Opengart & Bierema, 2015) so as to be able to narrate more authentic, coherent, independent, and sustainable ways of being for themselves. Significantly, mentors can create identity-workspaces for themselves and their protégés (Muir, 2014). Identity-workspaces are spaces of resistance and for emancipation (Sambrook, 2017). In such spaces, individuals can experiment with counternarratives (Brown & Humphreys, 2006) and with narratives that draw upon new discursive resources so as to craft alternatives to organizationally prescribed identities. Mentors can, in other words, enable protégés’ identity-work in experimenting with possible future selves and in developing a “flexible repertoire of identity options” (Nicholson & Carroll, 2013, p. 1239). Quite simply, protégés need enabling to author new stories of self that draw upon more diverse discursive resources.

A third set of contributions emerges from the psychodynamic conceptualization of identity-work developed in the article. This conceptualization provides further explanations and predictions for how mentoring works and suggests further practical implications. The psychodynamic conceptualization shows how certain ways of being might be aspired but might also delude and entrap both parties. In adding a psychoanalytic perspective to identity-work, it was shown how the self is expressed in the conscious “symbolic order” that is only a partial and inadequate reflection of the “real” self and results in an unremitting but unconscious sense of lack. Identity-work to achieve a stable, secure, and aspired sense of self can therefore be seen as chimeric, because identity is never more than an imaginary and incomplete construction. This conclusion has far-reaching implications for mentoring. Acknowledging and accepting the inevitable failure of identity-work to achieve an imaginary identity opens up the liberating possibility not of overcoming lack but, rather, of giving up attachment and living with lack (Driver, 2017; Hoedemaekers, 2010). Therefore, while enabling protégés to author new stories of the self and achieve alternative ways of being, mentors also need to prevent protégés’ unthinking pursuit of imaginary, and inevitably illusory, ways of being and to enable protégés to, instead, achieve authenticity.

Finally, the limitations of the alternative theoretical perspective on mentoring that this article has advanced need to be acknowledged, and the directions for future research discerned. First, we are conscious that more attention needs to be given to examining the important area of mentoring for diversity from the identity perspective that we have developed. Examining this area of mentoring through an identity lens has particular potential for further refining the theoretical tools. Second, we acknowledge that identity development in relation to occupations and organizations may be of relatively minor importance for certain groups of workers for whom identity is derived from their roles as, for example, homemakers. Third, we recognize that identity may yield a better understanding of certain types of mentoring relationships than others and that identity-work may be more or less important depending upon the career stage of the mentor or protégé. Fourth, we do not suggest that identity is the only theoretical perspective required to answer “how” and “why” questions pertaining to mentoring. However, we have established that social constructionist identity theorizing combined with the psychoanalytical perspective have good traction in answering such questions and in informing the development of mentoring practice.

Nonetheless, in extending the utility of the understandings developed theoretically in this article, empirical investigations are required. Whereas the tradition of developing and refining mentoring scales has become an important empirical research methodology (Ghosh, 2013), the perspective developed here will require empirical inquiry using qualitative methodology. In particular, ethnographic approaches and associated methods designed to engender mentor and protégé narratives will be required. Thus, open interviews and dyadic observations of mentoring encounters in progress will be useful. Moreover, methods for ascertaining indicators of identity and particularly for uncovering unconscious identity processes will need to be deployed. For example, the close examination of narratives for unconscious slips and contradictions provide insights into unconscious facets of identity (Hoedemaekers, 2010; Winkler, 2013). Empirical research might, therefore, examine propositions such as “within accounts of mentoring experiences, to what extent can discourses be discerned that are indicative of identity-work?” And, within mentors’ accounts of mentoring, “how is language used to position the mentor in relation to her or his protégé/s?” However, despite the noted limitations and areas for further research, we conclude that there is “more than meets the eye” happening in mentoring relationships. Examining the “I” in mentoring holds the key to explaining how mentoring works. Through mentoring, mentors as much as protégés find meaning for themselves in a world where a desired identity has to be worked on but can readily go awry.