Abstract
It is widely accepted that the conflict between women’s family obligations and professional jobs’ long hours lies at the heart of their stalled advancement. Yet research suggests that this “work–family narrative” is incomplete: men also experience it and nevertheless advance; moreover, organizations’ effort to mitigate it through flexible work policies has not improved women’s advancement prospects and often hurts them. Hence this presumed remedy has the perverse effect of perpetuating the problem. Drawing on a case study of a professional service firm, we develop a multilevel theory to explain why organizations are caught in this conundrum. We present data suggesting that the work–family explanation has become a “hegemonic narrative”—a pervasive, status-quo-preserving story that prevails despite countervailing evidence. We then advance systems-psychodynamic theory to show how organizations use this narrative and attendant policies and practices as an unconscious “social defense” to help employees fend off anxieties raised by a 24/7 work culture and to protect organizationally powerful groups—in our case, men and the firm’s leaders—and in so doing, sustain workplace inequality. Due to the social defense, two orthodoxies remain unchallenged—the necessity of long work hours and the inescapability of women’s stalled advancement. The result is that women’s thin representation at senior levels remains in place. We conclude by highlighting contributions to work–family, workplace inequality, and systems-psychodynamic theory.
Keywords
The progress women made in the 1970s and 1980s in accessing positions of power and authority slowed considerably in the 1990s and has stalled in this century (Stainback and Tomaskovic-Devey, 2012). In 2013, 14.6 percent of executive officers in Fortune 500 companies were women, down from their 15.7 percent share in 2002 (Catalyst, 2006, 2014). In professional service firms, as well, women remain dramatically underrepresented in the partnership ranks, where they are 18 percent of equity partners in U.S. law firms (Rikleen, 2015) and 23 percent of partners and principals in U.S. accounting firms (Wilson-Taylor Associates, 2016), despite having reached parity with men at the associate level long ago and despite policies to counteract the problem. “Stagnation” is the word many use to describe women’s stalled movement into high-level positions that offer opportunities to wield power and influence (e.g., Catalyst, 2014). This stagnation is a key component of gender inequality in the labor force, but its persistence is still poorly understood.
Drawing on data from a global professional service firm, we generate a new explanation to help account for the persistence of this gender inequality. This explanation arose in the course of conducting research for a firm that, having already tried off-the-shelf solutions, sought our help in understanding how its culture might have been inadvertently hampering women’s success. Our study of the firm’s culture revealed that virtually all employees recited essentially the same narrative to explain this lack of success: the job requires extremely long hours, and women’s family commitments (but not men’s) conflict with these time demands; hence, women quit or fail to make partner. We call this explanation the work–family narrative.
We also observed disconnects that begged explanation. Whereas firm members attributed distress over work–family conflict primarily to women, we found high levels of distress among men as well; whereas the firm instituted accommodation policies to help women, we found that women who used them failed to advance; and whereas firm leaders’ rationale for requesting our help included higher turnover for women, we found equivalent rates. Moreover, when we proposed to firm leaders a fuller, empirically grounded explanation—that the firm’s long-work-hours culture was detrimental to both women and men but that women paid a higher price—these clearly well-intentioned, otherwise empirically minded professionals rejected out of hand the data and analysis they had requested, maintaining their belief that work–family conflict was primarily a women’s problem, that it explained women’s lack of success, and that any solution must therefore target women.
These disconnects not only cast doubt on this firm’s work–family conflict explanation and solution, they also replicated a puzzle found in the work–family literature. According to work–family research, a widely accepted explanation for the stagnation in women’s gains is that women’s family obligations conflict with professional jobs’ long hours (Ramarajan, McGinn, and Kolb, 2012; Ely, Stone, and Ammerman, 2014), and the widely championed solution has been policies offering flexible work arrangements designed to mitigate such conflict (Galinsky et al., 2010; Perlow and Kelly, 2014). Yet closer examination reveals holes in this logic and a solution similar to what we found in our firm. For example, while work–family accommodations have been shown to reduce women’s conflict (Ezra and Deckman, 1996; Madsen, 2003; Hill et al., 2004), they have done little to help women’s advancement prospects and often have hurt them by offering off-ramps that can stigmatize users and derail their careers (Kossek, Lewis, and Hammer, 2010; Williams, Blair-Loy, and Berdahl, 2013; Perlow and Kelly, 2014). Moreover, men increasingly experience work–family conflict and nevertheless continue to advance (Reid, 2015). Hence the popular work–family conflict explanation for women’s stalled advancement does not square with women’s or men’s lived experience, either in our firm or more broadly, and the currently popular solution to the problem has the perverse effect of maintaining it in both places.
We took these parallels between our firm and the research as the basis for our orienting research questions: why does work–family conflict persist as the dominant explanation for women’s stalled advancement despite evidence that calls it into question, and why do accommodations persist as the dominant solution despite evidence that they perpetuate the problem? In other words, why do organizations’ prevailing beliefs and remedies regarding women’s stalled advancement systematically, if unwittingly, maintain gender inequality?
In the spirit of abduction, a form of discovery that begins with “surprising” findings or “discrepancies” and “works backward to invent a plausible world or a theory that would make the surprise meaningful” (Van Maanen, Sørensen, and Mitchell, 2007: 1149), we used leaders’ resistance to engaging with the study’s findings as the starting point for our investigation. These data-driven analysts’ reflexive commitment to empirically dubious beliefs raised for us questions about the possible presence of an unacknowledged, hidden investment in preserving these beliefs.
While resistance could be explained by invoking more facile explanations (e.g., people do not like change), they fail to capture the complexity of what we observed. The widely shared nature of organization members’ belief in a questionable proposition—one that had the insidious effect of pointing to remedies that derail women’s careers—together with the intensity of leaders’ reactions when presented with data that challenged it, suggested to us that leaders’ resistance might be a smoking gun—a tip-off that something profound was at play. Working backward from this smoking gun, we sought to construct a “plausible world,” one in which systemic forces (not just individual ones) and emotions (not just cognitive and practical considerations) hold sway.
For a literature able to speak to the dynamics we observed, we turned to systems-psychodynamic theory, a multilevel, psychological theory of unconscious, emotional dynamics in organizations (see Jaques, 1955; Menzies, 1960; Long, 2006; Petriglieri and Petriglieri, 2015). This theory gave us direction for exploring the possibility of hidden motivations and the unconscious maneuvers organizations and their members collectively mobilize to keep them hidden. Our study tells the organizational and psychological backstory of how the work–family narrative, as part of a social defense system, functions to maintain gender inequality in the workplace. To begin, we turn to the literature on work–family, which points to our firm as classically culturally stymied—an exemplar—in its efforts to advance women, suggesting that building theory from our firm is a potentially fruitful endeavor that will yield generalizable insights.
The Work–family Explanation for Women’s Stalled Advancement
Although sociologists have critiqued the work–family explanation as simplistic (Stone, 2007; Damaske, 2011; Cha, 2013), it is culturally endorsed in many arenas. Its prominence in the national press rose beginning in 2001, eclipsing such themes as stereotyping, harassment, and the “old boys’ network” (Ramarajan, McGinn, and Kolb, 2012). An example is Lisa Belkin’s (2003)New York Times Magazine article, which argued that women fail to reach the top because they “find other parts of life more fulfilling” and which added “opting out” to the popular lexicon. The work–family explanation resonates with many managers and executives, as well. A 2013 study of over 6,500 Harvard Business School alumni from virtually every industry found that more than three-quarters overall—73 percent of men and 85 percent of women—attributed women’s blocked advancement to their having prioritized family over work (Ely, Stone, and Ammerman, 2014).
Work–family Accommodations as Intervention Strategy and Source of Stigma
Companies understand the problem similarly, and many have offered policies that feature flexible work arrangements designed to mitigate work–family conflict (Galinsky et al., 2010; Kossek, Kalliath, and Kalliath, 2012). Part-time work and a wide variety of “flex” options—including periodic and daily flextime, time off, leaves, and sabbaticals, among others (Galinsky et al., 2010)—are common among companies concerned about retaining and promoting highly qualified women professionals.
Research shows, however, that taking accommodations often creates a “flexibility stigma” that results in negative career outcomes (Glass, 2004; Stone and Hernandez, 2013; Williams, Blair-Loy, and Berdahl, 2013). Flexibility stigma is a bias against policy users that “causes the target to fall into social disgrace” (Williams, Blair-Loy, and Berdahl, 2013: 214). Taking an accommodation indicates an unwillingness to work long hours, and in a professional-work culture that valorizes virtually unceasing labor, seeking time away is stigmatized. Costs can be steep, negatively affecting wages (Coltrane et al., 2013; Goldin, 2014), performance ratings (Wharton, Chivers, and Blair-Loy, 2008), and promotion chances (Judiesch and Lyness, 1999; Cohen and Single, 2001). While research shows that costs can be mitigated by such factors as supervisors’ support (Blair-Loy and Wharton, 2002), supervisors’ attributions about users’ reasons for uptake (Leslie et al., 2012), having an organizationally powerful supervisor (Briscoe and Kellogg, 2011), and the regulatory environment (Blair-Loy and Wharton, 2002; Kalev, Dobbin, and Kelly, 2006), the general pattern is that people who take advantage of these policies are likely to find their careers derailed.
Theoretically, the “work-devotion schema” (Blair-Loy, 2003) is the bedrock that undergirds flexibility stigma. This schema is the cultural assumption that work “demands and deserves single-minded focus and allegiance” (Blair-Loy, 2003: 6; see also Moen and Roehling, 2005). The use of the word “devotion” is not accidental. Weeks (2011), drawing on Marx and Weber, also describes a moral component: Western society endorses an ideology that considers work “the highest calling,” a “moral duty,” and an “ethical practice.” In professional settings, workers with part-time schedules face the possibility of dishonor. Part-timers in law firms, for example, were considered “time deviants” for having broken the rule that lies “at the heart of what it means to be a true professional” (Epstein, 1999: 133). Among women financial professionals, Blair-Loy (2003: 184) noted that part-timers “are viewed as apostates, unworthy of advancement into the firm’s celestial ranks.” The real-world manifestation of the work-devotion schema is that employed Americans worked an average of 1,868 hours annually in 2007, an increase of 181 hours—more than one month—since 1979 (Mischel, 2013), and work hours for professionals are often longer (Jacobs and Gerson, 2004; Briscoe, 2007; Goldin, 2014). Both women and men are subject to the work-devotion schema, but women face an additional constraint: the expectation that they comply with a “family-devotion schema” (Blair-Loy, 2003).
Gendered Impact of the Work-devotion and Family-devotion Schemas
The family-devotion schema assigns women, not men, the primary responsibility of childrearing and housework and holds them accountable. Women are to find fulfillment in the intimacy of “intensive motherhood”—a child-centered, emotionally absorbing, and labor-intensive form of parenting (Hays, 1996)—and their devotion to family is expected to override all other commitments (Roth, 2006; Turco, 2010). Failure to do so may bring the sanction of being considered a bad mother (Blair-Loy, 2003). This prescription is at complete odds with the work-devotion schema, of course: in fulfilling the family-devotion imperative, professional women with children take more responsibility for childcare and thus are more likely to take accommodations (Ely, Stone, and Ammerman, 2014), become victims of flexibility stigma (Stone and Hernandez, 2013), and experience career derailment (Judiesch and Lyness, 1999; Cohen and Single, 2001). As a result, many women feel torn by the demands of these competing schemas (Stone, 2007).
For their part, professional men tend to follow the work-devotion schema. They thus typically do not face flexibility stigma, nor are they judged on the family-devotion schema (except insofar as showing too much family devotion evokes penalties; Berdahl and Moon, 2013; Coltrane et al., 2013). Even today, being a good worker is culturally compatible with being a good husband and father. In fact, the breadwinner ideal confers manly status on men who leave caregiving behind and put in long hours at work (Cooper, 2000; Kellogg, 2011; Berdahl and Moon, 2013). Nevertheless, men do experience the pull of family life. While the breadwinner contribution to family is still culturally honored, it is no longer the sole template, and many men today feel a pull toward greater involvement in home life and the frustration that can accompany that desire when work impinges on it (Winslow, 2005; Glavin, Schieman, and Reid, 2011; Humberd, Ladge, and Harrington, 2015; Reid, 2015, 2018).
Thanks to these deep-seated ideologies about work and family, organizations seeking to advance women face a conundrum: the dominant explanation for women’s stalled advancement points to an intervention strategy that, if taken, often derails their careers. In addition, the work–family explanation oversimplifies the problem by neglecting the fact that men also experience work–family conflict yet nevertheless advance. Work–family scholars have grappled with these problems, responding by, for example, testing new intervention strategies that seek to provide non-stigmatized flexibility to all workers (Kelly, Moen, and Tranby, 2011; Perlow and Kelly, 2014), but none has sought to understand why companies have become caught in this conundrum in the first place. Without such an understanding, explanations for the persistence of workplace gender inequality remain incomplete, and strategies for undermining it continue to be elusive. To find a better explanation, we turn to systems-psychodynamic theory.
Systems-psychodynamic Theory
To examine the possibility that the firm had an unacknowledged, hidden investment in maintaining the work–family explanation for women’s stalled advancement and accommodations as the solution, we turned our attention to unconscious dynamics, an area of renewed interest in the social sciences (Kahn, 1990; Barsade, Ramarajan, and Westen, 2009; Pratt and Crosina, 2016). The fields of social psychology, decision sciences, and behavioral economics are now replete with studies showing how individuals habitually, unwittingly, and without conscious awareness stray from rationality (Greenwald and Banaji, 1995; Newman, Duff, and Baumeister, 1997; Schimel, Greenberg, and Martens, 2003; Zhong and Liljenquist, 2006), and neuro-scientific evidence for the existence of unconscious processes bolsters claims about people’s possible motives for doing so (Westen et al., 2006; Schacter, Addis, and Buckner, 2007). As Barsade, Ramarajan, and Westen (2009: 144) concluded from their review of the literature on implicit affect in organizations, “[T]he notion that much of what we do is influenced by processes outside our conscious awareness is no longer a theoretical claim or the province of clinical observation.”
The theory of unconscious dynamics we draw on is systems psychodynamics (see Bion, 1955; Jaques, 1955; Menzies, 1960). This theory moves the analysis of unconscious dynamics from the individual to the meso level, drawing attention to the interplay between organizational arrangements and individuals’ emotions and emotion regulation. This perspective considers how the emotional needs of individuals shape structures, narratives, and practices in organizations and how these structures, narratives, and practices, in turn, shape the experiences of those individuals (Petriglieri and Petriglieri, 2010). Two central constructs of the theory are the social defense, an unconscious, organization-level defensive scheme, and psychological defense mechanisms, the unconscious strategies employees use to regulate their emotions.
Social defenses are defined as “collective arrangements—such as an organizational structure, a work method, or a prevalent discourse—created or used by an organization’s members as a protection against disturbing affect derived from external threats, internal conflicts, or the nature of their work” (Petriglieri and Petriglieri, 2010: 47). They function as “a collective psychopathology—a necessary evil—allowing the institution to hold together and pursue its task while at the same time limiting its flexibility and its members’ awareness” (Petriglieri, 2013: 1). We investigated the possibility that the firm’s work–family narrative functioned as a social defense.
A classic example of a social defense analysis is Menzies’ (1960) investigation of a hospital whose presenting problem was burnout and turnover among student nurses. According to Menzies’ analysis, this problem was the result of a social defense—practices the hospital had instituted for the unconscious purpose of helping nurses, both student and fully trained alike, fend off the more deeply disturbing emotions that arose daily when caring for sick and dying people. Such defensive practices included, for example, nurses’ use of bed numbers, diseases, or diseased organs in lieu of patients’ names (e.g., “the liver in bed 10”). These practices allowed depersonalization, an unconscious psychological defense mechanism, and together they kept at bay the primary anxiety—angst raised by repeated confrontations with illness and mortality. Yet they also kept nurses from developing meaningful caregiving relationships, depriving them of the very gratification that inspired many to join the profession in the first place. This deprivation gave rise to another problem—the presenting one: nurses’ burnout and turnover. While troublesome and anxiety-provoking, confronting the burnout-and-turnover problem was less threatening than confronting illness and mortality, and for this reason it served as a useful substitute focus for the organization.
The result of such dynamics is a system in which the social defense diverts attention away from deeply disturbing emotions arising from the organization’s work and, at the same time, creates or perpetuates a substitute (“presenting”) problem. Importantly, the substitute problem is unresolvable precisely because its sustaining mechanism—the social defense—must remain intact (and its functioning invisible) if it is to serve its protective purpose. The operation of the social defense is thus circular and self-reinforcing.
Unconscious defense mechanisms are emotion-regulation strategies people reflexively use to protect themselves from disturbing emotions (Barsade, Ramarajan, and Westen, 2009) and are often mobilized as part of a social defense system. In Menzies’ study, employees used depersonalization as an unconscious defense mechanism to distance themselves from the discomfort aroused by confrontations with disease and death. Other research has examined employees’ use of other defense mechanisms. In particular, studies have shown how employees use splitting, projection, and projective identification (Gould, Ebers, and Clinchy, 1999; Ashforth and Reingen, 2014; Petriglieri, Petriglieri, and Wood, 2018), defense mechanisms mobilized to ease the psychic tension that arises when experiencing two seemingly contradictory emotions or facing two desirable but seemingly contradictory choices (see Kernberg, 1985; Smith and Berg, 1987; Smith, 1989). As that experience is not unlike work–family conflict, we considered whether these defense mechanisms were in evidence in our firm.
Petriglieri, Petriglieri, and Wood (2018: 22, 24–25) offered the following definitions: Splitting occurs when people partition a set of conflicting thoughts or feelings into two distinct subsets and then claim one subset and ignore or disown the other. It is a psychological defense mechanism because it reduces the discomfort of ambiguity and ambivalence (Klein, 1959). . . . [Projection and] projective identification [are] defense mechanism[s] whereby members of one group project the split-off, disowned parts of themselves onto another group and distance themselves from it, while at the same time keeping the other group close enough that they can identify with it. It is defensive because the simultaneous dismissal, while not completely letting go of the other group, bolsters the self’s clarity and value while sustaining some semblance of wholeness through identification with the other (Petriglieri and Stein, 2012).
These defense mechanisms appear at the center of two recent systems-psychodynamic studies of organizations. In one study, members of a food coop experiencing tension between the social and financial parts of the organization’s hybrid mission formed two informal groups—idealists and pragmatists—splitting the mission between them. The social part of the mission was projected onto and owned by the idealists, and the financial part became the province of the pragmatists (Ashforth and Reingen, 2014). Each group then projectively identified with the other group, denigrating it but remaining connected to it, thus ensuring that both parts survived. A similar process was reported in a study of a business school with dual ideologies, leading students to split into camps aligned with either the instrumental ideology or the humanistic one while denigrating but still projectively identifying with the other (Petriglieri, Petriglieri, and Wood, 2018). In both studies, these unconscious defense mechanisms reduced the anxiety prompted by the prospect of holding two seemingly contradictory but essential parts of the organization, while also enabling the organization itself to function effectively. In these studies, we note that the groups were relatively equal in power and that both domains were indisputably central to the functioning of the organization.
Research has yet to examine how social defenses and unconscious defense mechanisms play out between groups of unequal power or in organizations in which one domain represents an organizationally devalued sphere, as was the case in our firm, where men had more power than women and where the family domain was devalued relative to the work domain. These omissions are surprising, given that most organizations are characterized by inequality and given that power dynamics often operate outside of conscious awareness (Smith, 1989). Thus taking a systems-psychodynamic perspective on the conundrum we sought to understand held promise for generating new insights into the theory. To this end, we pursued the notion that the conundrum might be the result of a social defense system at play in organizations.
This idea led us to a more refined set of research questions. Does the work–family narrative serve as a social defense? If so, what are the components of the social defense system, and how do they operate to maintain the narrative as the dominant explanation and accommodations as the dominant solution for women’s stalled advancement and thus the persistence of gender inequality? Finally, how does this new application of the construct to the problem of gender inequality, in turn, extend or refine theory about how social defense systems operate?
Method
Research Setting
We were invited by a mid-sized global consulting firm to investigate how aspects of the firm’s culture might have been inadvertently limiting women’s success and to design initiatives to stem the loss of women in pre-partner ranks, an industry-wide problem. We accepted this invitation in exchange for permission to use data from this investigation for research purposes and to collect data for additional research projects.
The firm, with offices located primarily in the U.S., provides advisory services in such areas as strategy, marketing, and finance. It draws its consultants from elite colleges and MBA programs, prides itself on its analytical rigor, and typically places high on lists of prestigious consulting firms. As is increasingly common among professional service firms, this firm had a clearly defined promotion path but did not adhere to a strict “up or out” system, and employees could remain despite not having been promoted. Employees could move from client- to internal-facing roles, although promotion to partner from the latter ranks was rare. Years to promotion into the partnership averaged 9 for men and 11 for women. Like other professional service firms, it was male-dominated, particularly at senior levels, with men constituting 63 percent of junior associates, 70 percent of associates, 77 percent of senior associates, and 90 percent of partners. Formal work–family accommodations were available to women and men but were individually negotiated and consisted of reduced-hours schedules, internal-facing assignments, and leaves of absence.
Participants
This paper draws on data collected in this firm over 18 months for three separate interview-based studies centering on gender-related research questions: the requested culture study, a study of men’s professional identities, and a study of women’s and men’s leadership identities. We refer to these latter two studies as tandem studies. Two senior leaders served as our liaisons to the firm and provided the contact information for potential participants. Samples for each study were randomly drawn from the same sampling domain—the four largest U.S. offices—except where noted. We relied on random sampling for our initial research questions, which was an appropriate strategy for yielding a set of broadly representative views and experiences related to the topics we set out to study (i.e., the firm’s culture and the two identity-related questions of the tandem studies). For the culture study, we especially wanted to ensure that the data were comprehensive and not biased toward any particular view of the firm. Serendipitously, the random sample allowed us to ascertain how widely shared the narratives and processes were that became the focus of our research. At the same time, it was sufficient for reaching saturation on the new constructs of interest (e.g., men’s and women’s experiences of the work–family nexus).
Our total sample included 107 consultants (including partners and associates) and five human resource (HR) personnel. Of the 33 women consultants interviewed, 8 were partners (5 married and 7 with children), and 25 were associates (16 married and 14 with children); of the 74 men consultants, 21 were partners (20 married and 20 with children), and 53 were associates (32 married and 18 with children). Most participants were white. We also interviewed the head of HR and four other senior HR personnel. For quotations in the findings section, we number each respondent and indicate sex using the designation “F” or “M” for female or male; we indicate consultants’ rank using “P” or “A” for partner or associate; and we use “HR” to designate human resource personnel.
Potential participants received an e-mail from our liaisons introducing the research projects, alerting them that they might be contacted, and assuring them that their participation was voluntary and confidential. Everyone who responded to our e-mailed invitation (only a handful did not) and who anticipated being available agreed to participate. Because of work schedules or unexpected travel, most interviews were rescheduled at least once and some as many as five times. The response rate (the number interviewed relative to the number solicited) for associates was about 70 percent, with non-participation largely due to insurmountable scheduling conflicts or leaves. The response rate for men partners was close to 100 percent. We invited all 11 client-facing women partners in the four largest U.S. offices to participate and interviewed 8 (73 percent). Of the three client-facing women partners we did not interview, one was on leave and two were unavailable during our visits.
Data
Data come largely from employee interviews. Interviews with consultants involved a series of open-ended questions covering, at a minimum, the following topics: perceptions of what it takes to be successful at the firm and particular challenges women may face; explanations for women’s slower advancement rate than men’s; and, for those participating in the tandem studies, personal accounts of how they experience themselves in their roles as professionals or leaders, including challenges they have faced. Online Appendix A presents a sample of these questions (http://journals.sagepub.com/doi/suppl/10.1177/0001839219832310).
Interviews with human resource personnel centered on the firm’s professional development system and on employees’ use of work–family accommodation policies. These HR personnel also supplied data on men’s and women’s turnover rates.
Interviews were mainly face-to-face on the premises, although some were in nearby venues, and interviews with overseas employees and with a few in distant U.S. offices were conducted via telephone. Most interviews lasted a minimum of an hour, and several lasted for two. Interviews were tape-recorded and transcribed.
We also considered as data the firm’s reaction to study findings that failed to confirm the work–family explanation for women’s mobility problems. In an experience not uncommon for field researchers (Berg and Smith, 1985), we became players in the organizational drama, caught up in the firm’s ongoing process of legitimating women’s stalled advancement as its primary problem and the work–family narrative as the primary explanation. In this case, upon learning that our proposed interventions were not targeted to women but instead would address the long-work-hours problem both women and men faced, the firm’s CEO lost interest in the project and ultimately terminated it (see also Bain, 1998: 419, on termination of the research relationship). Tandem studies continued without disruption.
This parting of ways on the culture study after a cordial and productive relationship was puzzling, and its timing on the heels of our feedback gave us pause: our analysis had upended the firm’s preferred explanation, and we were expelled. This sequence led us to a deeper level of analysis. Following Berg and Smith (1985), we analyzed firm leaders’ reaction as clinical data pointing to the possibility that a defensive operation was underway, specifically, that the firm was enacting a social defense in our midst.
Analyses
We moved iteratively between key concepts in the systems-psychodynamic literature, operationalized below, and the data, seeking to identify elements of the social-defense system as specified in existing literature and to reject, extend, or refine the theory upon encountering data that either did not fit or could not be easily accommodated. 1 Drawing on data about unconscious meanings and motives is necessary for analyzing psychodynamic processes such as social defenses. Capturing psychodynamic processes empirically, however, is not straightforward because they are driven by emotional conflicts sufficiently disturbing that people tend to experience and regulate them outside of conscious awareness and therefore cannot easily access or discuss them. Hence conventional content analysis of interview data, in which the meaning and significance of what people say are taken at face value, is, by itself, insufficient for this analysis.
To identify unconscious emotional dynamics in the interview data, we developed a more interpretive coding scheme, paying attention to both what interviewees said (or did not say) and how they said it, and we paid attention to coherence and contradictions. We flagged interview segments that reported accounts of work–family matters and that also contained signals of unconscious emotional dynamics. Work–family matters included interviewees’ references to their own and others’ circumstances, decisions, views, or feelings, as well as comparisons they made between women and men. Signals derive from noticing elements in interviews and transcripts that are typically ignored in ordinary social discourse, such as attempts to avoid distressing feelings, which can appear, for example, in “subtle shifts of topic when certain ideas arise” and other maneuvers (Shedler, 2010: 99). Signals include hesitations, stumbling, abrupt shifts, setting up stark contrasts, striking use of metaphors, equivocation, deflections, incoherence, and contradictions (Peebles-Kleiger, 2002; Shedler, 2006, 2010). These signals are visible, almost tangible manifestations of internal contradictions or feelings of distress. They serve as “tells” indicating that the content is potentially conflicted and warrants attention; they are “markers” similar to the “repetition, sequence, emotion, discontinuities, spontaneous communications, and idiosyncratic communications” of interest to therapists (Peebles-Kleiger, 2002: 69).
We then speculated about the emotional dynamics that gave rise to these signals by undertaking an iterative process of interpretation. One author or another would propose an interpretation of an interview segment, and the others would comment by agreeing, raising questions, or suggesting revisions. Sometimes the discussion led to multiple iterations, either face-to-face or in writing. A general criterion was how close or far the interpretation was to the data, with the solution always being to choose the interpretation that was closest. Examples of this iterative process, drawn from our analytic memos and notes from our conversations, appear in Online Appendix B.
Identifying these verbal behaviors and their meaning requires no deep knowledge of a person’s intrapsychic life, however, and the researcher need not have an ongoing therapeutic relationship to recognize them, because the purpose here is not to understand individual psychology per se but rather to identify patterns of emotional expression (and lack of expression) across people within a system. What is required is “clinical judgment” involving an “attuned clinical ear” (Shedler, 2006: 228) turned toward individuals but, in this case, for the purpose of identifying common themes and patterns in groups (e.g., among men, among women) and in the organization as a whole. The “clinical ear” listens to interviews on multiple frequencies and is especially attuned to the frequency that conveys unconscious strategies people use to manage their emotions (Wells, 1998). “Clinical judgment” is acquired, as is the case with any analytical skill set, through reading, experiential learning, and practice, which can be gained in programs and institutions—in this case, those that sponsor systems-psychodynamic training. 2
The content of unconscious emotional conflict—and hence, its signals—differed for men and women. Table 1 presents examples of unconscious emotional conflict in men’s interviews, which tended to revolve around feelings of guilt about time away from their families; table 2 presents examples in women’s interviews, where the conflict centered on ambivalence (examples 1–3) and self-doubt about their personal competence (examples 4 and 5). We did not expect different signals for men and women but simply noted what we observed and speculated about why. Men’s guilt, we reasoned, would be less likely to manifest itself as either the “back-and-forthing” or “stark contrasts” that women evidenced in describing their ambivalence or the “foreclosure of options” and “equivocating” they evidenced in describing their sense of competence. For their part, women, whose guilt is quite conscious, would face less need than men to engage in “deflecting feelings onto others” or other unconscious maneuvers that minimize feelings of guilt. This unexpected sex difference became data for use in developing new theory about how social defenses operate in contexts of structural inequality.
Examples of Unconscious Emotional Dynamics Relating to Guilt in Men’s Interview Data: Signals, Interview Excerpts, and Interpretations
Examples of Unconscious Emotional Dynamics Relating to Ambivalence and Sense of Competence in Women’s Interview Data: Signals, Interview Excerpts, and Interpretations
Once we charted patterns in the emotional landscape of men’s and women’s conflicts regarding work and family, we sought to identify what they did with these conflicted emotions—how they mobilized specific unconscious defense mechanisms to deal with them. Here, we saw evidence of employees engaging in splitting, projection, and projective identification—unconscious strategies activated at the intergroup level as a tacit arrangement between women and men.
To the extent that people experience emotional conflict stemming from internal pulls toward both work and family domains, while nevertheless describing the work domain as primarily men’s and the family domain as primarily women’s, we can say that they have engaged in splitting and projection at the intergroup level. Each gender takes the key parts of being a whole person and “splits” them in two—a committed parent and a committed worker. Consistent with cultural norms about the gendered division of labor (Ridgeway, 2011), the committed parent role is attributed to—”projected” onto—women, and the role of committed worker is projected onto men.
To the extent that women and men accept their assigned roles as natural and appropriate, we can say that each gender has “introjected” the projection (although not necessarily completely), which enables each gender to experience the other as having the characteristics associated with their assigned role, thus setting the stage for “projective identification.” Individuals engage in projective identification when they unconsciously identify with those who enact the disowned—split off—parts of themselves. In other words, they experience those enactments as emotionally gratifying (Klein, 1946), as, for example, when men project their emotional need for family onto women and women introject that need as their own, allowing men to experience their disowned need vicariously. 3
Men and women demonstrated splitting and projection when they described both work and family domains as personally compelling while nevertheless making repeated references to family as women’s domain and work as men’s (appearing, for example, as setting up stark contrasts and contradictions). The more firmly and consistently they located themselves and others in the gender-appropriate domain, the more wholly they were engaging in splitting and projection. Interviewees evidenced projective identification when their convictions about this gendered arrangement could be seen as an attempt to resolve the underlying conflict (appearing, for example, as abrupt shifts and deflections). The more widespread this dynamic across the firm, the more evidence we have that it operated at the intergroup level. The more fully it operates at the intergroup level, the more we can expect this arrangement to provide a collective experience of wholeness, giving employees some relief from the primary anxiety. The success of these strategies was variable, with some people seeming to experience more relief than others, suggesting more or less complete identification with the projection and, concomitantly, a more or less satisfactory resolution of the underlying conflict.
To identify the primary anxiety, we searched the interview data for discussions of work features that raised angst—what it was about the nature of work itself that might have given rise to this unconscious social defense system. The aspect of work that raised the greatest angst was the firm’s relentless demand for 24/7 availability, a demand that was widely regarded as problematic. The angst, or primary anxiety, was the internal conflict arising from that demand that daily forces a choice between love (family) and work, undermining individuals’ sense of human wholeness. 4
Finally, to identify any further elements of the social defense, we reviewed the data for evidence of additional firm-wide narratives, policies, and practices that reinforced the work–family narrative or that supported organization members’ unconscious psychological defense mechanisms.
The Work–family Explanation as a Hegemonic Narrative: One Firm’s Story
Below, we describe the plot elements of the work–family narrative and then, by recounting what we heard from interviewees and what we experienced while working with this firm, show how these elements cohere into a tale that constitutes a “hegemonic narrative”—a pervasive, status-quo-preserving story that is uncontested, even in the face of countervailing evidence (Ewick and Silbey, 1995). We used the hegemonic narrative concept to elaborate the qualities of the work–family narrative that made it an ideal social defense for our firm’s purposes and, in so doing, present the all-important context in which the work–family narrative unfolded as a social defense.
Plot Elements of the Work–family Narrative
The work–family narrative as articulated in this firm contained two “plot elements,” told and retold by members across the firm: (1) the job requires extremely long hours, and (2) these hours are impossible for women—but not men—who have family responsibilities; hence, women do not advance.
The story begins with tales of a job that demands long work hours and frequent travel. As is the case in professional service firms generally (Wharton and Blair-Loy, 2006; Briscoe, 2007; Goldin, 2014), hours at this firm were grueling. Interviews revealed no disagreements about the firm’s demands that employees put in long hours. Echoing many, one consultant described his work habit: “[S]hoot me something on Saturday by 10
The second plot element is that women’s commitment to family conflicts with the time demands of the job, thus hindering women’s career advancement. When asked why the partnership had so few women, partners indicted work–family issues, and this account filtered down the ranks. According to one partner, conflict is built in for women, making it hard for them to be seen as leaders: We have great intentions and I think pure intentions, genuine intentions, about getting the best involved regardless of gender, race, creed, religion, what have you. I frame it in the following way. What do I want people to worry about when they wake up first thing in the morning? So business development people, I want them to worry about business development. For project managers, I want them to worry about the project. . . . Women are . . . the project manager in the home, [so] it is hard for them to spend the necessary time, energy, and effort to be viewed here as a senior leader. (P19 M)
Some wove this narrative into their personal stories, as did a partner who described how his professionally successful wife “gave up work. . . . In the end, she decided—very difficult choices—that she wanted to have more time with the children” (P21 M).
Associates concurred with this analysis. One male associate opined, “I just think mothers have a different type of bond with their children . . . and it makes it that much more stressful and frustrating to be away . . . overnight” (A42 M). According to another, the conflicting requirements of motherhood and the job meant women reject the fast track: It’s just basic math, right? So you take 100 people. Fifty are women and 50 are men. Twenty-five of the women are going to have kids and not want to work. Twenty-five of the women are going to have kids and might want to work but won’t want to travel every week and live the lifestyle that consulting requires of 60- or 70-hour weeks. (A01 M)
This unrealistic picture is the extreme version of the pervasive storyline: motherhood renders women inadequate to the task and explains their relative lack of success. Note how after being introduced (“50 are men”), men never reappear in this narrative. It is a work–family problem, and it is women’s problem, not men’s. Note also that in this calculation all women are mothers, a conflation that was common in our interviews. Logically, the work–family narrative would acknowledge that childless women can succeed (because they can avoid the caregiving impediment the narrative highlights), yet this demographic group figured nowhere in the discussions of women’s advancement prospects and rarely in the stories we heard. It was as if all women were tarred with the brush of motherhood or incipient motherhood.
By and large, women associates agreed with the work–family narrative. According to one mother: There’s no overt sexism. Once you’ve proved yourself, people work with you. No one would hold me back from being on a hard-core partner track if I were willing to work 70-hour weeks and get on a plane every week. The issue is that women are choosing to have kids and be their primary caregiver. (A73 F)
The problem focus firm-wide was firmly on women, who were seen as less able or willing than men to compromise their family commitments. We note two word usages endemic to the narrative at the firm: “mother” acted as a stand-in for all women, and “family” acted as a stand-in for all personal (nonwork) commitments. These word choices represent archetypes that capture the power of the narrative to subsume all characters and settings into its storyline, even when they patently failed to fit.
The Work–family Narrative as Hegemonic
Numerous features of the work–family explanation for women’s stalled advancement suggested it may be a hegemonic narrative. According to Ewick and Silbey (1995: 200), a hegemonic narrative is a pervasive, uncontested, seemingly natural account that makes singular sense of an array of personal experiences and is resilient to countervailing evidence. It is an overarching strategic story that preserves dominant cultural meanings and power relations and reproduces them. In the case of the work–family narrative, the story preserves meanings and relations surrounding gender; the plot centers on the work-devotion and family-devotion schemas; and dominant meanings and power relations are reproduced by the career-derailing work–family accommodations that organizations take up as the ostensibly best resolution of this plot dilemma. Table 3 summarizes how the characteristics of a hegemonic narrative correspond to features of the work–family narrative in this firm. Most of the characteristics described in the table need little further explication, but two are crucial to the analysis and require more detail.
Characteristics of a Hegemonic Narrative and Evidence from One Firm*
Source: Adapted from Ewick and Silbey (1995).
First is the role work–family accommodations played in derailing women’s careers and thereby reproducing the gender status quo. According to the HR personnel we interviewed, the two most popular accommodations were switching from client- to internal-facing roles and working reduced hours, and women were more likely than men to do both. Although HR did not keep records of associates’ policy use, of the associates with children we interviewed, nearly half the women were taking one or both of these accommodations at the time of the interview, compared with only one of the men.
Women’s family demands were cited as the primary reason for their move to internal roles, a move that typically took one out of consideration for positions of real power. Because client-facing work is “travel intensive and time intensive [and] unpredictable, it’s harder for those with primary caregiver responsibility,” explained the head of HR. As a result, he continued, internal-facing roles are “disproportionately women because the hours are a little more predictable” (HR4 M). Almost 20 percent of women partners in the firm compared with 10 percent of men had completely transitioned to these non-revenue-generating roles. And while both men and women associates periodically took short-term staff assignments as part of their portfolio of responsibilities, it was different for women, who, many noted, bear the main responsibility for raising children and thus were more likely to remain in such roles.
Working reduced hours also damaged one’s prospects at the firm, according to HR personnel and many consultants. An associate noted that reducing availability—for example, by limiting travel—is “not necessarily held against you, but I think it’s just tougher to sort of prove your worth [to] a client” (A26 M). Yet a partner felt it could, in fact, be “held against you,” offering the following account of how flexibility stigma attaches to those who receive “special treatment”: You work as a team, so if you get any special treatment, your team members are going to feel you are not doing your share. No one wants to be working til midnight every night, so if you let women negotiate something special, it would be tough on the team. And the women will not want to be seen as not contributing, so it is a dilemma. (P08 M)
He went on to describe another consequence of taking accommodations: The special issue it presents for women is that sometimes senior leaders will know a woman has three kids and say, “I know this project is going to be a killer; I’d better not take a chance on her because she might ask for special accommodation or need to leave at 5:00 or something to go get her kids.”
The upshot for women as individuals was sacrifices of power, status, and income, and for women as a group was the continuation of a pattern in which powerful positions remained the purview of men, while women’s progress stagnated. Hence work–family policies flowing from the narrative helped reproduce the gender status quo, which meant that women constituted only 10 percent of partners and, among those who made it to partner, women’s track was two years longer than men’s.
The second feature of a hegemonic narrative that is crucial to our analysis is resilience—its ability to withstand a challenge—which was evident in events surrounding our feedback to the firm in which we pointed out disconnects between our findings and the work–family narrative, particularly that men (and not just women) experienced distress over long work hours. We found that firm leaders were unable or unwilling to consider the possibility that their work–family analysis was incomplete or that an alternative explanation for women’s stalled advancement might be valid, and we offer their resistance as evidence of the work–family narrative’s resilience. The sequence of events unfolded as follows.
In the feedback to firm leaders, we provided detailed data showing that men were at least as likely as women to say work interfered with their family lives. Among associates we interviewed who were parents, two-thirds of men reported work–family conflict compared with slightly more than half of women (nearly all of the remaining mothers were taking accommodations to ease the conflict). Men’s dissatisfaction with schedules that pulled them away from their families came through strongly in numerous statements such as this one: “I was traveling three days a week and seeing my children once or twice a week for 45 minutes before they went to bed. Saturday came, and I couldn’t go to my son’s soccer game. He burst into tears. I wanted to quit then and there” (A45 M). According to another, “Last year was hard with my 105 flights. I was feeling pretty fried. I’ve missed too much of my kids’ lives” (P09 M).
Thus, contrary to the work–family narrative’s exclusive focus on women, men too were troubled by the strain that long work hours placed on their families—and some of them left the firm as a result. According to one father, “I wouldn’t characterize myself as unhappy. It’s more overworked, and under-familied. If I were a betting man, I’d bet that a year from now I’m working somewhere else” (A34 M). And a year later, he was.
We pointed out this disconnect to the firm’s leadership, challenging the work–family narrative as oversimplified, and offered a broader, more nuanced analysis implicating unnecessarily long work hours and their disproportionately negative effect on women’s careers. Specifically, we presented interview data revealing a culture of overwork stemming from the firm’s practices of overselling and overdelivery. Regarding overselling, the sentiments of this interviewee were echoed by many: “[Some partners will] promise the client the moon . . . and not even think about what that means for their team. . . . ‘We’ll do X, Y and Z, and we’re going to do it all in half the time that you think it should take.’ And the client’s going to say, ‘Wow, that’s great! Why don’t we sign up?!’” (A26 M). And from another: “With no controls on how to scope a project, [a person] can sell anything whether it’s reasonable or not. If you kill your people, there is no cost to you” (A08 M).
The culture also valorized overdelivery, priding itself on delivering “110 percent” to clients and offering “smart” solutions to clients’ problems. Commenting on overdelivery, an associate complained: [A]ccount managers are very cerebral, and they’re all about the answer. And so they would always be like, “Oh, we should do this” and “Can we do this analysis? And can we do this?” Just because it would be interesting. . . . So [my team] worked all these weekends. . . . And I’ve been in the [same] spot as my team so many times where I just worked really, really hard and sacrificed family stuff, sacrificed my health for it, and at the end of the day, I look back on it, “Well, did we really have to do that? Probably not.” (A62 F)
Associates went along with overdelivery and overwork partly to stand out as stars in a pool of highly qualified people: “We do these crazy slide decks that take hours and hours of work. It’s this attitude of ‘I’m going to kill the client with a 100-slide deck.’ But the client can’t use all that! People do it so others on the team will see they’re smart” (A70 F).
While everyone suffers from the long-hours problem, we explained to the leadership, it disproportionately penalizes women because, unlike men, they take accommodations, which come with a career price. We noted that most men suffered in silence or otherwise made do, thereby leaving the woman-centric focus of the work–family narrative intact.
When we provided this feedback to the firm’s leaders, the CEO reacted negatively. The firm had requested an analysis of the firm’s organizational culture, but upon hearing that the gender problem was only a piece of a larger work-management problem and that the solution would involve a change in work practices that transcended work–family accommodations, he balked. Although we had interviewed over 70 men across all levels, many of whom had been there for decades, the firm head suggested we had not spoken to the “right” men but instead must have interviewed only new associates or uncommitted ones. He also questioned the intervention strategy that flowed from our analysis, which targeted overselling and overdelivery, on the grounds that it did not explicitly focus on women.
A few months later, our partner-liaisons presented the study findings to the rest of the partnership and, in an e-mail to us, reported that the partners and CEO remained “stuck on [understanding] how [our proposed interventions] were going to help women.” They shared with us the slides they had created, which were ostensibly based on our analysis and recommendations. We were struck by the absence of our data pointing to the dubiousness of their cultural assumptions about time (i.e., the necessity of 24/7 availability) and excellence (i.e., the necessity of overdelivery). Their presentation concluded by proposing such interventions as “Conduct joint research with the Center for Work–Life Policy’s Hidden Brain Drain Task Force,”“Actively engage with the Council for Women World Leaders,” and “Establish a head of Diversity & Inclusion who reports to the CEO,” none of which had any discernable relationship to the study’s findings. Without buy-in from the top, interest in the project flagged, and over the course of a year, e-mails about the project stopped and the engagement effectively ended.
If not for the work–family blinders, the firm might have seen and addressed work problems that hurt all employees. Instead, firm leaders maintained their original assessment—that the problem was women’s difficulty balancing work and family and that men were largely immune to such difficulties. Leaders’ unwillingness to engage with our evidence illustrates the resilient, Teflon quality of the work–family narrative, further highlighting its hegemonic nature, which in turn helps explain its stranglehold on the problem definition.
Yet this unwillingness on the part of evidence-driven analysts to engage the evidence begged further examination of the data. Upon revisiting company records, we learned that although one of the firm’s key concerns was “women’s higher turnover rate,” in fact women’s and men’s turnover rates did not significantly differ in any of the preceding three years. 5 We wanted to understand at a deeper level why data-oriented and clearly well-meaning firm leaders had failed to read the turnover data accurately and, in the face of feedback about the widespread problem of work hours, clung to their belief in the work–family narrative.
While the notion of resilience in a hegemonic narrative gave us some insight into why the work–family narrative was so tenacious, tracking the implicit emotional content in our interviews, guided by constructs from systems-psychodynamic theory, gave us further traction. We drew on this theory—specifically, the idea of a social defense—to consider the basis for participants’ unwavering conviction that women’s family lives were the ultimate obstacle to women’s advancement, a conviction that ultimately preserves the gender status quo.
The Work–family Narrative as a Social Defense
In the findings from our analysis of the emotional dynamics accompanying this conviction, the plot elements of the narrative thicken. A workplace socially constructed as requiring 24/7 availability gave rise to a primary anxiety: the threat of losing one’s sense of human wholeness prompted by the daily forced choice between love and work. To quell this anxiety, employees mobilized gender-based splitting, projection, and projective identification that, together with firm narratives, policies, and practices, reinforced the idea of women’s fitness for family and men’s fitness for work. Below, we describe men’s and women’s internal conflicts around the pulls of parenting and working, how they managed these conflicts, and how the social defense system functioned for each group.
The Problem for Men
The psychic tension men face is the demand that they have no identity other than as a labor commodity, which creates an internal conflict that must be resolved. Capitalism’s system of competition among firms compels overwork for many professionals. 6 This imperative sets up an ongoing demand that other, nonwork identities (and the needs generated within them, such as being a good parent) be contingent. With nonwork identities in the backseat, the identity that remains is that of the ideal worker: fully committed to work and fully available (Acker, 1990). Those striving to be the ideal worker must adopt the psychological stance of “my job is all-important.” But always chipping away at this stance is the raw reality of the requirement to stifle the demands of other identities.
These other identities—being a good parent, life partner, citizen—are contingent and expendable for the ideal worker. Yet for real people, these identities—particularly the parent one—are compelling. According to one man, “I definitely want my daughter relying on me. But she’s asking her mama, ‘Put me to bed,’ asking her mama, ‘Give me a bath.’ It’s because she knows that she can be relied on” (A25 M). Although he does not explicitly express it, his sense of guilt is palpable, a feeling many other men shared. While they were quick to describe missing their families, their defensive maneuvers, illustrated in table 1, typically intercepted the feeling and diverted the logical leap to feelings of guilt.
We show how projective identification keeps such feelings at bay by examining the psychological jujitsu one man demonstrated as he drew on the work–family narrative to explain women’s lack of advancement in the firm.
I believe deeply in my heart and soul that women encounter different challenges. There’s the collusion of society that it’s the woman who takes the extended maternity leave, and there are some biological imperatives, too. When my first child was born, I got to carry her from the delivery room to the nursery. It’s almost like I could feel the chemicals releasing in my brain. I fell so chemically, deeply in love with my daughter. I couldn’t imagine a world without her. I mean here it was in [just] the first eight minutes of her life. So I can understand, “How can I possibly give this up and go back to work?” (P20 M)
But back to work he went, and his take-away understanding was that women face problems with work–family. He could be said, from the standpoint of a defense analysis, to be splitting off his deep connection to his daughter, projecting it onto women in the firm, and projectively identifying with what he imagines to be the women’s emotional gratification. If he relinquishes that intense feeling of connection to his daughter, he has no need to feel sad and guilty about returning to work.
Unpacking the components of the last quotation helps clarify the defensive process. The narrative flow that seems smooth when he tells his story is in fact revealed as illusory when we track what happens to the intensity of his feelings of attachment to his daughter. The first two sentences make a distinction between women and men and link biology to motherhood. It is women and not men who have the parenting experience. He then says almost the exact opposite by abruptly shifting to his own biologically framed, intense emotional experience of parenting. In so doing, he is momentarily taking back the projection he just placed on mothers. His act of “understanding” women’s experience via his own, however, signals that projective identification has occurred. He is in effect saying, “I was having this experience, but it was transient, and now that I’ve sampled it, now that I’ve been a tourist in this emotional land, I have a way to understand what is happening to women in the firm.” The powerful emotional experience—with all the psycho-biochemical force he described—is no longer his. It is now theirs. He now knows it but is not governed by it.
In fact, he immediately shifts in his next statement to aligning himself not with women but with men, explaining how men are different from women. He continues, “I can’t think of a single instance where the fella took a six-month paternity leave to care for the baby while mom went back to work.” Speculating vaguely about how the firm works, he then says, “I think that we have a way of problem solving and a way of engaging with clients that doesn’t necessarily give a greater advantage to cowboy style or, kind of, the certainty that seems to be a social aspect of masculinity in North America. But . . . it’s clear to me there are clients who like that certainty.” He concludes by situating himself squarely in the male-dominated world of work: “You know, kind of like the—the work I do in the beer world. It [the brewery industry] is dominated by men, and I mean men slapping each other on the back and talking about golf and shit like that.” Thus he ends by placing himself where he began, in a different world from women’s, in the world of work (the “beer world”), where men and masculinity dominate. In this world, there is no room for the emotional experience of parenting. Here he is able to exist, however unhappily, unencumbered by the “different challenges” he ascribed to women in his opening statement. While this account was the most vivid and complete illustration of the dynamic of men projectively identifying with women in the firm, we argue that he is describing a work–family resolution that is not idiosyncratic but rather shared by other men.
This man drew directly on his understanding of women in the firm to achieve this resolution; more typically, men drew on their understandings of their wives, blurring the distinction among women and ultimately consigning them all to the private sphere. For psychodynamic purposes, the firm’s women become privatized and thus made indistinguishable from wives. Here is a man engaging in this process as he draws on the work–family narrative to explain women’s lack of advancement in the firm.
Consulting can be a bit more difficult for women. There’s a lot more traveling. It’s my personal—what I’ve seen—sometimes women are more attached to kids. They feel guilty. With my wife—. Sometimes they feel guilty if they’re taking time away from home, in a way that men don’t. You do travel a lot, you do work longer hours. So men don’t feel certain things that women do. (A08 M)
We note how in trying to talk about the difficulties the firm’s women face (“consulting can be a bit more difficult for women”), he feels the need to invoke his “personal”—shortly revealed to be his wife, a representative of the private sphere. We then see him moving back and forth in a fragmentary way between women in the private and public spheres of his life—his women coworkers and his wife: after invoking his “personal,” he returns to women coworkers (they’re “more attached to kids” and “feel guilty”), goes back momentarily to his wife (“with my wife”), and then reverts again to women coworkers’ guilt, a guilt men do not feel. In psychologically consigning the firm’s women to the private sphere, he paves the way for the splitting and projection appearing at the end of the passage: women carry the guilt associated with work; men (ostensibly) do not.
In light of this apparent interchangeability between the firm’s women and men’s wives, it is unsurprising that men often invoked their wives as the recipients of their projective identifications. An example of such a projective identification comes from a father of two as he was discussing how he managed work and family and clearly communicating how—in accord with the work–family script—he is exempt from the emotional pull of home. He said: So you tell the older kid, “Hey, get ready for school,” and he basically does. Not a hundred percent, but certainly ninety percent. And you even tell the little one—I mean he’s 7—he can get dressed. He can actually open his drawers and get the right clothes and get dressed. He can’t make his breakfast, but in a pinch, he can. The older one can definitely open up a thing of Pop-Tarts and pop them in the toaster. It’s not what you call healthy eating, but for this week, it’s fine. So, it’s—I mean—they don’t cry when I leave. Sometimes the younger one does. But—
We note that he breaks off very suddenly after mentioning the crying with the comforting thought that “it’s a lot easier” now that they no longer “both cry.” This move appears to be a deflection of guilty feelings about his children’s self-serve breakfasts and tears. In the next segment, he jumps to talk about sympathy with his wife having suffered from the boredom of being with children: “And the other thing is, she [his wife] wasn’t working, which is brutal. I mean it’s just boring. We love the kids—but it’s just not that interesting [to stay home with them].” His jump to his wife and his use of the word “brutal,” coming on the heels of the description of his interactions with his children, is suggestive. Does witnessing his children’s tears when he departs brutalize him? Perhaps. He may be eliding the emotions he feels about the morning routine—probably a bundle of guilt, shame, and sadness—by splitting them off and projecting them onto his wife: he conjectures that she is the one who feels brutalized, and he empathizes with that experience, thereby circumventing his own feelings and completing the projective identification. This interpretation raised for us the question: could they each feel brutalized by the loss of one domain—he the domestic, she the employment? Finally, he raises the idea of a shared boredom, perhaps numbing himself to his emotions. The problem of sadness and guilt vis-à-vis the children is gone. In all these men’s accounts, we see the invocation of the work–family narrative, which relieves them, at least to some degree, from the conflict.
Splitting off personal needs and feelings about family and projecting them onto women enables men to show up at work every day and fulfill both the cultural dictates about male breadwinning and the organization’s desire for the committed, ambitious workers it believes it needs to remain competitive. These arrangements provide men the illusion of a fulfilled life. The work–family narrative is crucial in supporting the psychological defenses that help men assuage the pain of losing the domestic realm, and the fact that the narrative is hegemonic ensures the palliative’s staying power. The splitting, projection, and projective identification keep the collective whole, but at the individual level, the wholeness is illusory and leaves men in a state of constantly grasping for what is called “psychic integration” (Menzies, 1960). Those unable to sustain this pursuit quit. These defenses work as Band Aids, but for most men, the reality of the on-the-ground, relentless demands of family continually poke through the defense.
The Problem for Women
The capitalist imperative for overwork creates a different psychic tension for women. While men construct at least the appearance of being ideal workers, fulfilling the demand for overwork and relegating nonwork identities to the backseat, women are asked to be ideal mothers, fulfilling the demands of intensive parenting and relegating the work identity to the backseat (see also Blair-Loy, 2003). Women striving to be ideal mothers must adopt the psychological stance of “my family is all-important,” yet jettisoning opportunities to contribute meaningfully beyond the domestic realm exacts costs (see also Stone, 2007). At a macro level, this splitting apart of work and family domains and assigning work to men and family to women allows the system of overwork to remain in place, however unsatisfyingly at the personal level.
For professional women like the ones in our firm, who have tasted success and reaped some of the rewards of their years of schooling—and who have persisted in the work sphere, despite messages that home is where they belong—this psychic tension is especially acute. Men struggle internally, often unconsciously, with the requirement to give up intimate connections, but they are at least conforming to cultural norms. For men, the parental role as breadwinner, as cut off from intimate connections as it may be, nevertheless goes hand in hand with their work commitment; indeed, for ideal workers, family devotion takes the form of breadwinning and is entirely compatible with overwork (see also Townsend, 2002). For women, however, the parental role as caregiver flies in the face of work commitment; for ideal mothers, family devotion takes the form of intensive parenting and is not only incompatible with overwork, it often compels reduced work (see also Ridgeway, 2011: 130). But for many ambitious women, both caregiver and worker identities are compelling (see also Stone, 2007).
Thus it is unsurprising that the professional women in our firm struggled openly with the push to split off the work component of their identity, even as they willingly complied with the family-devotion schema. Regarding the latter, one mother talked about her inability to shirk the home front, despite having a stay-at-home husband: I think there’s just a difference between the way a mother and a father look at their kids and the sense of responsibility that they feel. I don’t know, but I feel my male counterparts can more easily disconnect from what’s happening at home. . . . If I did sort of disconnect [from home], things wouldn’t fall apart. But I wouldn’t feel good about it, so it’s just not going to happen. (A57 F)
At the same time, her work commitment was strong: “If I make a commitment to this company and to this organization, which I’ve made, I’m going to do it. I don’t doubt myself.” Yet she did have doubts about whether her family commitment would allow her to do what it would take to develop professionally: I know I’ll fall down from time to time. I know I need to learn, and there are going to be things that—I don’t doubt myself . . . from a place of doubting myself—it’s more from a place of needing to learn and needing to grow. I doubt myself generally in being able to honor that, while also honoring the commitments I’ve made [to my family]. That is a constant worry.
The emotional conflict she feels is evident in her flip-flopping around the notion of self-doubt: “I don’t doubt myself . . . I doubt myself.” Whether she can handle it is, as she notes, “a constant worry.” Her unambivalent claiming of her mother identity, conveyed in the first part of the quotation, is not matched by an unambivalent repudiation of her work identity. Rather, she describes herself as both committed and doubtful when it comes to work. Table 2 presents further examples of women describing this ambivalence.
Thus women who embrace commitments to both family and work do not fully comply with the work–family narrative, and as a result, they are unable to reap all of its psychological benefits as a social defense. Specifically, although many women did seem to take in the family-caregiver projection handed them by their male coworkers and by the firm, thereby enabling men to identify vicariously with that split-off aspect of themselves, they did not seem to fully reciprocate by splitting off and projecting onto men their worker identity. Thus the psychological resolution many men found in projective identification was not fully available to these women, leaving them holding identities organizationally and socially constructed as contradictory. While men mourned the loss of the family-caregiver role, they had, in fact, given it up (more or less); women had not (yet) given up the worker role. The social defense could ease women’s dilemma only to the extent that they abdicated their worker identity—the way many men abdicated the intimate connections involved in caregiving. Unwilling to quit or substantially ratchet down their career aspirations, many women remained caught in this dilemma, struggling more openly than men did with whether and how they would be able to fulfill both their work and their family commitments. Thus by reifying the work–family narrative, the organization facilitated men’s resolution while remaining a thorn in women’s side, constantly reminding them that they were in the wrong place by being at work instead of at home.
These reminders appeared as three push factors women had to withstand to hold onto their work identity as ambitious professionals, all of which reinforced the work–family narrative as a social defense, solidified its hegemony, and further pressured women to split off their worker identity. First was the firm’s career-derailing work–family accommodation policies, together with the firm-wide practice of users being primarily women; second was a shared narrative about the mismatch between women’s selling style and the style the firm valued; and third was a shared narrative derogating women partners’ mothering. Research has documented these factors (see Williams, Blair-Loy, and Berdahl, 2013, on work–family accommodation policies; Kellogg, 2011, and Ibarra and Petriglieri, 2016, on mismatches between feminine stereotypes and valued work attributes; and Blair-Loy, 2003, on women professionals as bad mothers), but the idea that they function together as solidifying elements of a social defense is new.
The first push factor was the firm’s policies and practices that created the strong expectation that mothers take work–family accommodations, as elaborated above. Thus women had available a ready off-ramp from the path of overwork, which also meant, for women associates, an off-ramp from the fast track to partner and, for women partners, an off-ramp from the path to real power.
The second push factor was the firm’s narrative about the purported mismatch between women’s relational style of selling and the hard-charging style the firm venerates, qualities a number of women had difficulty embracing personally but easily attributed to men. Whereas the first push factor reinforced the work–family narrative by emphasizing women’s fitness for family, this push factor emphasized the flipside: their unfitness for work. This selling-style narrative loosened women’s identification with work and affirmed men’s, further easing the way toward women’s stepping back at work.
This push factor arose from the firm’s construction of the job of selling, the most valued skill in the firm. The biggest accolades and biggest sales come from selling to CEOs, and virtually without exception, people named men as the star rainmakers, whose style they described as hard-charging and unequivocal. Here’s how one of the firm’s most powerful women partners imitated the selling style of the person named by everyone as the firm’s most shining star.
He’ll walk into an executive meeting and say . . . “Okay, you want to achieve this. . . . It’s going to take an organizational intervention, an innovation . . . a broader corporate strategy. It’s going to be a multi-year program, let’s just be very clear. That will be . . . about $15 million over two years. Are you tracking, are you with me?” (P26 F)
In contrast, here is how she described her default style: “I walk into a client, I check in. I like to be friends with them. . . . I tend to say, ‘Here’s what I heard you saying, your agenda.’ . . . And I start to build, and then I hope that they get to this delightful conclusion that this is going to be a $15 million program.” Clearly she sees a mismatch between her reflexive relational style and the style the firm venerates and demands, a perception reinforced early in her career by a partner who warned her that relying on relationship building when making cold calls on prospective clients would communicate that “you don’t have a lot going on between your ears.”
Other women reported the same mismatch. Developing the selling skill as the firm defines it was especially hard for people who saw their chief strengths as the ability to be responsive to clients and to build relationships with them. According to one woman just below the partner rank: It’s hard for me. So much of it is based on this relationship that I develop. . . . I tend to form these extraordinarily close relationships with my clients, and I was going to meet with the person the next level up [whom I didn’t know]. I didn’t know how to create that impression with that person, because what my strong suit is, is making people feel listened to and trusted and cared for. (A59 F)
Thus this narrative about women’s relationality disqualifies them from sales super-stardom, pushing them away from work (and possibly toward home).
Compounding women’s presumed disadvantage in selling to CEOs was an unwillingness to exude a certainty they did not feel, an unwillingness they believed men did not share. As one woman explained: I think that in general men need to have this much knowledge to talk with authority [holds up thumb and index finger close together], and women need to have this much [opens them]. Right? And I think that my goal is to try to get away with this much [halfway between the two]. Not that I want to turn into a bullshitter, but giving myself slack on my internal burden of proof [would help]. (A67 F)
Another woman suspected that this unwillingness to make unfounded claims may be a problem for many women: “You also have to be a certain kind of woman to do well in [sales]. You need to be able to give advice about things you don’t know much about and be really confident, and maybe a lot of women don’t have those skills. I don’t see a lot of role models” (A58 F).
So the message is that the selling job is best done by men enacting a conventionally masculine style—hard-charging and unequivocal—and that enacting a conventionally feminine style—like being a “relationship builder”—risks the label of lacking something between the ears. Not on the radar screen was the notion that other selling styles—perhaps ones that maximize the trust that comes from cautiousness in making claims—might be effective in sales; also missing was any shared narrative that women could learn the venerated style. The firm had constructed effective selling in only one way, and women fell short. Thus the message that women are ill-equipped for sales bolsters the work–family narrative by giving further momentum to any inclination to scale back.
The third and final push factor was the firm’s negative message about women partners with children, a group that had held fast to their identity as ambitious professionals and had achieved recognized career success. These women had resisted ratcheting down via accommodations and had successfully overcome their supposed selling deficits, and we heard positive references to their professional success. The very existence of these successful women partners, although few in number, suggests the possibility of integrating worker and mother roles, and yet women partners with children were roundly condemned as bad mothers, undermining that possibility. To more-junior women witnessing such condemnation—both those who were mothers and those contemplating motherhood—career commitment would seem to exact a terrible cost.
We heard zero positive comments about women partners as mothers but many negative comments from both men and women. Women partners’ family lives were scrutinized and found lacking in a way that we did not hear about men, an observation noted by a woman associate: “When I look at a female partner, it does leak into my thinking: how do I think she is as a mother in addition to how do I think she is as a partner? When I look at men, I don’t think about what kind of father they are” (A63 F). Men were equally critical. According to one, “There is this one really senior woman . . . and she has two nannies. . . . I would never want my life to be like—my family to be like clients. That would just be horrible. . . . I look at the women in the office, and it’s like they’re either divorced or they work so much” (A47 M). A junior woman summarized the impact of such sentiments: “What I do not have is a positive role model of a working mom” (A73 F).
Another junior woman who was married and planned to be a mother told us the following about a woman partner: “She tells a story she thinks is funny about how her kid was surprised when she picked him up from school. He said, ‘I’m so honored that you came to get me.’ But I’m appalled by that story! That is not who I want to be!” (A55 F). Both the woman partner in telling this story and the junior woman in recounting it and repudiating the behavior reveal women’s struggle combining (or anticipating combining) the maternal role with professional ambition. What impels the partner to tell her junior colleague this potentially incriminating story? We speculate it is an unconscious attempt to exonerate herself of guilt. In making a public statement, she may be seeking to normalize her behavior, while at the same time implicitly giving the opportunity for condemnation. A response of silence or approval—the only two viable responses from a junior colleague—would allow her to feel exonerated, at least for a while. As for the junior woman, why is she impelled to relate this story to us? We suggest she may be making a public declaration of who she is not (but fears she could become if she were to live out her career ambition). In making this declaration, she paints herself into a corner: she either has to leave the firm or contemplate the possibility of being a bad mother. By declaring “that is not who I want to be,” she affirms her devotion to motherhood and guarantees a moment of emotional surcease.
The women in our study faced an unresolvable dilemma: responding to the pull of family and taking accommodations meant undermining their status at work, and retaining their ambition rendered them subpar performers (given the definition of superior performance as masculine) or—for women partners—subpar mothers. Thus the leitmotif around motherhood is that a woman cannot excel in this job and be the ideal mother. We speculate that when women fully take in the work–family narrative’s belief that motherhood and career commitment are incompatible, they ratchet down by cutting back their work hours or by taking an internal-facing role, or else they leave, offering “family” as the reason. Either way, women’s careers are derailed, and the work–family narrative gains even further support.
From a broader perspective, men’s and women’s internal struggles differed (see tables 1 and 2). For men, the firm’s social defense was more or less effective, allowing their internal conflict (guilt) to remain largely hidden—not just from themselves but also from the rest of the firm. But the social defense failed women, who openly struggled with their still-unresolved and conscious conflict (ambivalence), speaking of it often. The work–family narrative practically compels them to do so by making such expressions normative for women. Thus in a one–two punch, not only did women fail to receive the full anxiety-assuaging benefit of the work–family narrative as a social defense, but also their anxiety and the work–family narrative that encourages them to talk about it further freed up men to take on the identity of the ideal worker. All the while, the social defense system is helping supply the firm with a steady stream of seemingly amenable (male) workers. In short, work–family conflict was conscious for women, rendering it discussable, and by talking about it they effectively owned the conflict for the firm, reifying the notion that it is only a women’s problem.
A Systems-psychodynamic Explanation for Entrenched Workplace Inequality: Gender as a Case in Point
Figures 1 and 2 summarize our study’s findings and, by naming constructs and introducing new insights into how they play out between groups of unequal power, offer an analytically generalizable model (Yin, 2010) for using systems-psychodynamic theory to understand the entrenched nature of workplace inequality. The basis for the model’s generalizability derives from our firm’s similarity to other professional service firms in its demand for 24/7 availability (e.g., Turco, 2010); evidence for the broad cultural endorsement of the work–family explanation for women’s stalled advancement and its grounding in widely shared cultural schemas that assign women to family and men to work (e.g., Blair-Loy, 2003; Ely, Stone, and Ammerman, 2014); reports based on nationally and internationally representative samples of men’s struggles with work–family conflict (e.g., Shockley et al., 2017); and studies documenting the career-derailing effects of taking work–family accommodations (e.g., Williams, Blair-Loy, and Berdahl, 2013). These broadly shared characteristics suggest the transferability of our results to other professional service firms, at a minimum.
Operation of a social defense system that sustains intergroup inequality: Gender as a case in point.
Operation of an asymmetric social defense system.
Figure 1 presents an overview of the model. The generalizable components are the constructs, shown in capital letters, and the processes linking them, shown in italic; the remaining text in the figure displays our study’s findings. The model begins at the bottom of the figure, where the organization’s work context gives rise to a primary anxiety, leading employees to turn to unconscious defense mechanisms for relief. In our study, the unremitting demands of the 24/7 work culture gave rise to the wrenching anxiety of sacrificing either work or personal life—both integral components of human wholeness—leading employees to employ splitting, projection, and projective identification to reduce the internal conflict. These defense mechanisms operate at the intergroup level and in our study were based on the cultural association of men with work and women with family (see figure 2 for details about how this dynamic plays out). The organization-level social defense operates symbiotically with these unconscious processes. In our firm, the social defense was made up of several elements: discourses centered on the hegemonic work–family narrative that presents women as less-than-ideal workers, the supposed mismatch between women’s selling style and the optimal one, and the supposed bad mothering of women partners; policies that cohered around providing work–family accommodations despite their career-derailing effects; and practices that situated women as the primary policy users. Together, the social defense and the unconscious defense mechanisms mitigate the primary anxiety, although, as explained below, success is likely to be mixed in contexts of intergroup power disparities, as it was in our firm.
All the while, the social defense creates or perpetuates a problem that serves as a substitute focus for the organization. While disconcerting, the substitute problem is less threatening than the real problem and gives rise to a substitute and less distressing anxiety. Preoccupation with the substitute problem is a red herring; a tell-tale sign is that ostensible efforts to fix it are unconsciously designed to fail. In our study, the substitute problem was the firm’s inability to retain and promote women, and the tell-tale sign was that the firm’s efforts to resolve it—most notably, offering work–family accommodations—perpetuated gender disparities, giving firm leaders an unresolvable (and therefore always available) problem to worry about. By directing firm members’ focus, laser-like, on women’s worrisome retention and promotion rates—the substitute anxiety—the social defense system leaves the primary anxiety more or less languishing in the neglected, out-of-focus periphery. The result is an invisible, self-reinforcing, protective system that diverts attention away from the real culprit—in our firm, the 24/7 work culture.
Yet when intergroup inequality suffuses these dynamics, as was the case in our setting between women and men, the defensive pattern may well play out asymmetrically between the two groups, protecting the high-power group more fully than the low-power group. Whereas men were willing to split off their committed-parent identity and project it onto women, women, who readily took in that identity, did not reciprocate by fully splitting off and projecting their committed-worker identity onto men, despite the organizational push factors encouraging them to do so. This gender difference stems from the fact that men were more fully ensconced in their culturally prescribed domain (work), whereas women, because they were working, were at best only partially ensconced in their assigned domain (family), keeping one foot, however ambivalently, in the culturally proscribed one (work) as well. Because women complied less fully with this tacit arrangement, they experienced less psychic relief than men and thus enjoyed a less satisfactory resolution of the underlying conflict.
Generalizing from this finding, figure 2 unpacks the asymmetric operation of these processes between groups of unequal power. As in figure 1, the generalizable components are constructs, shown in capital letters, and the processes linking them, in italic; the rest of the figure text represents our findings. The center of the figure is arrayed around two orthogonal dimensions: the two groups under consideration—one high power and one low power—and their respective organizationally valued and devalued domains. Arrows depict the psychological processes linking each group’s members to one another and to valued and devalued domains. The outer ring depicts the social defenses in the organization we studied, arrayed to show which ones support each psychological process. The arrows between the circle and the outer ring indicate the nature of each group’s relationship to these elements, either compliant with or resistant to them.
To return to the center of the figure, while members of both low- and high-power groups have a fundamental connection to both domains, the work context is such that the two domains are experienced as contradictory, thus raising anxiety and setting in motion the social defense system. Systems-psychodynamic theory would predict that groups might manage that anxiety by engaging in a fully reciprocated process of splitting, projection, and projective identification; our findings suggest, however, that full reciprocation is unlikely in contexts of inequality where low-power groups have entered a high-status position (working women, in our study). In these contexts, whereas high-power groups (men) are more likely to comply with the defensive arrangement, low-power groups in high-status positions are more likely to resist. High-power groups would maintain a predominant cultural identification with the valued domain (work) while splitting off and projecting onto the low-power group the devalued one (family), where they can projectively identify with it. Meanwhile, members of the low-power group would continue to introject the devalued domain (family)—it is, after all, familiar and congruent with expectations. But they would also identify, however ambivalently, with the valued domain (work) associated with their high-status position and would resist the social defenses urging them to relinquish it. Thus the low-power group cannot parry the internal struggle stemming from the primary anxiety, while the high-power group receives some relief from it.
Discussion
Findings from a study of a professional service firm seeking to retain and promote women illustrated how the firm-wide explanation for women’s stalled advancement—that women’s work–family conflict keeps them from being able to work the requisite long hours—operated as an unconscious social defense, deflecting attention away from the firm’s culture of overwork.
This analysis invites speculation about what other functions the work–family narrative might serve in this organization and in others that similarly venerate long work hours. We suggest that the narrative also allowed the firm to deflect responsibility for women’s stalled advancement: firm leaders’ diligent efforts to solve the problem gave them an airtight alibi against any accusation that women’s failure to advance might be their fault. The narrative also justified the gender imbalance at senior levels: if women themselves prefer to be with their families, as the work–family narrative has it, leaders cannot be accountable for the glaring gender inequality in their senior ranks. Nor do they need to confront the disturbing possibility that they themselves might be biased or might have discriminated against women. Nor need women, for their part, confront the possibility that they might have been in any way ill-treated or victims of discrimination. To the contrary, in the course of detailing the work–family account, many participants of all ranks and both sexes went to great lengths to assure interviewers that women’s lack of advancement could not be the result of discrimination, suggesting that this unpleasant possibility existed, at some level, in their consciousness. The defense system, however, ensured that it was never seriously broached.
More fundamentally, leaders promulgated a work culture that pitted two fundamental social institutions—work and family—against each other, a culture at odds with the progressive and caring image the organization sought to cultivate. The solution to this contradiction again lay in the commitment to the work–family narrative, which protected the leadership from being seen as the source of the 24/7 requirements by instead deflecting blame to clients’ demands and industry norms. In showing concern for women’s problems, the narrative placed leaders on the side of “the people” rather than on the side of profits, rendering them innocent. This study thus shows how a social defense can maintain existing power relations while enabling those in power to appear as if they were invested in precisely the opposite.
This research makes three theoretical contributions. First, we deepen work–family scholarship by revealing the psychodynamic underpinnings of the central problems and constructs it examines. Work–family scholars have recognized the flexibility stigma that attaches to taking accommodations (Stone and Hernandez, 2013; Williams, Blair-Loy, and Berdahl, 2013) and the conundrum that lies at the heart of this problem: the work–family explanation for women’s stalled advancement points to an intervention strategy—accommodations—that, when taken, derails women’s careers. Our research is the first to offer a theoretical account for why this conundrum exists. In our analysis, women’s stalled advancement is a substitute problem that must be preserved because it deflects attention away from the problem of long work hours; accommodations, for their part, serve the unconscious function of keeping the substitute problem in play. This perspective suggests that scholars advocating accommodation strategies may be unwitting players in a game rigged against women from the start. While some research has begun to identify factors that mitigate the costs of accommodation-policy uptake, our findings suggest that disrupting core elements of the social defense, such as the gendered splitting of work and family, may be necessary. How organizations might systematically undertake such disruptions—and what might compel them to do so in the face of the powerful countervailing pressures we have described—would be worth investigating.
Recognizing the accommodation-policy conundrum—but not its systems-psychodynamic underpinnings—has led many work–family scholars to recommend sidelining accommodation policies in favor of a more broadly framed approach: changing companies’ culture of long work hours on the rationale that doing so will improve work processes more generally and enhance the personal lives of all employees (Ely and Meyerson, 2000b; Sturm, 2001; Rapoport et al., 2002; Bailyn, 2006; Perlow and Kelly, 2014). Such an approach, advocates argue, would enlist broad support for change and remove flexibility stigma while also eliminating an important source of gender inequality. This broader-based approach has been difficult to implement and sustain, however. One study showed that despite all employees being encouraged to work flexibly, women were more likely than men to take advantage of the opportunity and continued to be marginalized for doing so (Kelly et al., 2010). Moreover, the initiative was eventually terminated. Analogous culture-change initiatives have encountered similar problems (Ely and Meyerson, 2000a; Rapoport et al., 2002). Scholars have offered sociological accounts for these outcomes, highlighting the deeply entrenched nature of the masculinized ideal-worker norm (Bailyn, 2006; Kelly et al., 2010). Our findings take this explanation a crucial step further by demonstrating how that norm is embedded within a set of psychodynamic processes in which women and men—and the firm as a whole—unconsciously collude. According to our analysis, a sustainable, gender-equitable solution to the long-work-hours problem would require women and men to reclaim disowned, split-off parts of themselves and face the fact that their painful compromises are neither natural nor inevitable. While potentially a relief to some, this reclaiming would likely be psychologically intolerable to many, as evidenced by the widespread, unconscious commitment to splitting we witnessed, which protected employees from having to confront such a possibility. Yet absent this reclaiming (and firm leaders’ willingness to respond to it), the problem of long work hours will continue to reside squarely in women’s domain, ensuring that the substitute problem of women’s stalled advancement remains both front and center and unsolvable. In short, our study shows why these kinds of culture-change solutions, by themselves, are dubious at best.
Further, while our focus has been on the workplace, home and work are tightly connected, and our research has implications for how work–family scholars might further consider that connection. It stands to reason that home is part of a larger system that is integral to the functioning of the work–family narrative. Thus splitting dynamics undoubtedly traverse the work–home boundary, raising an intriguing question about whether social defenses exist in the home. For example, do norms, practices, or discourses in the home similarly divert attention from the forced choice between love and work (the primary anxiety) raised by the 24/7 work culture? And might childrearing approaches that mimic project management (see Stone, 2007) be one way that formerly professionally ambitious stay-at-home mothers quell the pain associated with the loss of work? Or does bringing home into the picture suggest a different, perhaps broader, primary anxiety? More generally, how does home join work as part of a larger, psychodynamically managed system that sustains women’s subordinate position and upholds traditional gender norms? These questions suggest that examining the systems-psychodynamic underpinnings of the work–home nexus may be worthwhile.
Second, our study contributes to scholarship on workplace inequality. Our findings point the way to a vast, unexamined territory—the role of unconscious, systems-level processes in sustaining inequality—and systems-psychodynamic theory is particularly well suited for examining this terrain. Inequality is fraught with ambivalence and anxiety, and it is the theory’s stock-in-trade to consider how people are motivated—and how the systems within which they are embedded are set up—to keep such feelings at bay. We note, for example, that while inequality has become increasingly objectionable across the globe, it persists as a system from which many nevertheless continue to derive benefits—however short term or highly compromised—across the power–inequality spectrum (Jost, Banaji, and Nosek, 2004). This contradiction implies a deep well of ambivalence underlying organizations’ conscious efforts to eliminate gender and other forms of inequality. Thus to suggest that such efforts may serve as a social defense is not farfetched and may help explain their frequent failure: the overriding purpose of consciously well-intentioned initiatives may be less about reducing inequality and more about fending off the disturbing emotions it raises while allowing it to persist.
As a case in point, we are struck by the proliferation of organizational initiatives, such as unconscious bias training, networking groups, and managerial diversity-performance evaluations, that endure despite evidence that they are ineffective and can even backfire (Kalev, Dobbin, and Kelly, 2006; Castilla and Benard, 2010; Duguid and Thomas-Hunt, 2015). A systems-psychodynamic perspective on these outcomes would investigate the emotional landscape—ambivalence, anxiety, and guilt—surrounding these initiatives. Are these initiatives inequality’s “insidious camouflage” (Carmichael, 2017)? If a company’s faith in its own progressivism can perpetuate inequality, as research suggests (e.g., Castilla and Benard, 2010), then might unconscious emotional conflicts be fueling this faith? In short, are companies’ good-faith efforts to eliminate inequality systematically and unconsciously blinding them to the errors of their ways? Our study illustrates not only how such covert processes can be empirically studied but also how failing to recognize them perforce maintains the status quo. This understanding begins to break the theoretical and practical gridlock scholars have faced when seeking to explain why the problem of inequality has been so resistant to remedy even in the face of remarkable social change on many fronts (see, e.g., Ridgeway, 2011).
The third theoretical contribution is bringing a power-based perspective to the systems-psychodynamic literature. We demonstrate how social defenses, like much else in collectivities, may work better for powerful groups, bringing greater nuance to scholars’ understanding of how social defenses operate. This insight is new; previous research has assumed that an organization’s social defense works equally well for all members of the system, whereas in our case, it worked better for men than for women, likely owing to men’s greater power.
To consider this possibility, we propose a hypothetical example based on racial inequality. This example roughly parallels the categories presented in figure 1. The work context in many organizations, and in this hypothetical one, is stratified, with people of color underrepresented in top positions and overrepresented in lower-level ones. Employees going about their daily work lives confront the situation of racial minorities’ virtual absence in upper-level jobs, reflecting a history and wider cultural context of white domination, which arouses an unconscious and unspoken primary anxiety centered on worries about one’s fundamental worthiness. In the case of whites, such an anxiety could take the form of “deep down, I’m a racist,” while for racial minorities in high-status jobs it could take the form of “deep down, I’m not good enough.” With employees unable to directly confront such deep-seated fears, the organization creates a social defense involving a narrative about concern for the organization’s lack of racial diversity as well as a set of practices designed to increase the pipeline of racial minorities. Meanwhile, with a focus on the pipeline (the substitute problem), racial disparities in the organization are left unaddressed, and anxiety about meeting recruitment goals largely displaces whites’ primary anxiety about their moral goodness; racial minorities’ anxiety remains intact because the organization’s narratives about diversity and efforts to increase pipelines (the social defense) fail to address their worries and may even exacerbate them (see Heilman, Block, and Stathatos, 1997). Such an understanding of organizations’ inability to move the needle on workplace inequality itself is new, as we noted above. Moreover, only whites’ primary anxiety about worthiness is reduced. As in our firm, social defenses operating in contexts of intergroup inequality seem to work better for high-power groups than for low-power ones, a new insight we bring to systems-psychodynamic theory. We propose that social-defense analyses would be enriched to the extent that they are sensitive to how intergroup power disparities may produce varying levels of protection for different groups.
Our finding that social defenses are especially effective at protecting powerful groups from confronting their unconscious emotional conflicts also speaks to how this new insight might inform a systems-psychodynamic approach to organizational change geared toward advancing equality. Powerful groups, who have the institutional means to undertake such change and yet the most to lose from it, may experience the greatest ambivalence and anxiety about inequality. Thus, like the leaders in our firm, they are positioned as key protagonists in maintaining any social defense designed to alleviate such anxiety. Future research on the covert organizational processes that sustain gender and other axes of inequality thus should be attentive to the unconscious dynamics underlying powerful groups’ weddedness to existing policies and practices.
More generally, these contributions involve linking the micro and macro realms. By elaborating how organizational features mediate between individual- and societal-level processes, our model of how organizations’ unconscious social-defense systems perpetuate inequality responds to the call for organizational scholars—especially those studying gender (Ely and Padavic, 2007)—to develop “meso”-level theory (House, Rousseau, and Thomas-Hunt, 1995).
We also make a methodological contribution. Scholars have recently called for greater attention to unconscious emotions in organizational settings (Barsade, Ramarajan, and Westen, 2009). Yet while experimentalists have developed tools for identifying and studying these emotions in the lab (e.g., Schimel, Greenberg, and Martens, 2003), field researchers have thus far lacked such a methodological toolkit. We begin to lay out a method for how field researchers might notice, bring to the surface, and interpret unconscious emotions in interview data. In reporting our methods, we paid particular attention to showing the moments in people’s interviews that we took to be signals of unconscious emotions (e.g., hesitations, contradictions, deflections), and we aimed for transparency in spelling out how we interpreted them.
We note that our work–family analysis is generalizable only to a particular stratum of the workforce—professionals in 24/7 work cultures—and to a particular manifestation of gender inequality—women’s lower representation in firms’ upper ranks. Hence the particular processes we uncovered may not apply to other groups of workers or to other sites of gender inequality. Workers at the bottom of the labor force, for example, are far more likely to face a shortage than a proliferation of work hours (Jacobs and Gerson, 2004; Lambert, 2008; Jacobs and Padavic, 2015) and are less likely to have access to work–family accommodations (Kelly and Kalev, 2006); these workers would therefore not suffer from overwork or its attendant anxieties, rendering the particular social defense we identified moot for this group.
Black women may be another exception. Firms’ upper ranks are overwhelmingly white, and the cultural expectation that (white) women enact the family-devotion schema might not extend to black women (Hurtado, 1989; Kennelly, 1999; Collins, 2004). Cultural prescriptions for black women dictate that working take precedence over family caregiving (Cuddy and Wolf, 2013). Moreover, black women are accorded more latitude than their white counterparts to enact at least some of the masculine behaviors normatively associated with leadership (Livingston, Rosette, and Washington, 2012). Accordingly, black women might be immune to organizational social defenses predicated on beliefs about women’s family primacy and lack of fit for leadership roles. Other social defenses, however, may help explain the persistence of black women’s underrepresentation in top ranks, and future research might examine this possibility.
Conclusions
We conclude with two thoughts. The first is what our findings mean for the larger project of gender equality in society. In her consideration of the cultural forces deployed to resist movement toward gender equality, Ridgeway (2011: 53) noted that both women and men have a “deep sociocognitive interest in maintaining . . . cultural beliefs . . . differentiating them” and thus “an interest in resisting a real erasure of gender difference.” We agree that the culture’s deep investment in cultural beliefs about gender differences is a key impediment on the road to equality (see also Ely and Padavic, 2007). Indeed, the wider culture is responsible for the creation of the work–family narrative. But we question the idea that people’s motivation is purely “sociocognitive,” with all the rational capacities implied by “cognitive.” Our findings suggest that psychodynamic desires and conflicts are also at play, making the path toward equality even more difficult to traverse. Thus if our analysis is correct, women’s advancement is slowed because of social defenses at the organizational level, along with the equally resistant-to-change wider cultural beliefs Ridgeway discussed.
There is hope, however, which brings us to our second point: the work–family narrative-based social defense suffers from weaknesses. First, hegemonic narratives may have staying power—that is what it means to be hegemonic—but that does not mean they are impossible to dislodge (Ewick and Silbey, 1995). As women and men employees continue to feel frustrated and as researchers point to productivity losses from long work hours, other accounts may displace it, making it less available for social defense purposes and creating space for other, less-entrenched formulations of the problem.
The other weakness of the work–family narrative as a social defense is that, like all social defenses, it is not completely effective. It fails to fully alleviate either the pain men feel over disconnection from family or the pain women feel over the stark choice they are handed between work and family. As Menzies (1960: 116) noted in her classic study of social defenses, the social defense system inhibited “self-knowledge and understanding” and thus “fail[ed] . . . its individual members desperately.” This was also the case with the employees in our firm. While the social defense was more effective for men than for women, it was not foolproof or absolute for them either, and discontents were ever-present for both sexes.
In closing, we note that our findings square with recent observations that progress toward gender equality will be slowed to the extent that efforts are focused exclusively on women (Coontz, 2011; Ridgeway, 2011; Joshi et al., 2015), but our findings also suggest that to be effective, expanding efforts to include men requires a broad vision. For example, a popular recommendation is to encourage men to use accommodation policies at a rate similar to women so as to level the playing field. We would argue, however, that accommodation policies alone—regardless of who uses them—will not dismantle the culture of overwork, nor will they dislodge the deep-rooted, multilevel, psychodynamically motivated association of women with family and men with work. They thus are unlikely to significantly advance the project of gender equality.
Instead, solutions require a thoroughgoing reconsideration of gender at work and at home, one that begins with exploring people’s “psychological investments in cherished identities” (Williams, Berdahl, and Vandello, 2016: 526). While this challenge may seem daunting, invoking Ridgeway and Correll’s (2000) metaphor of ocean waves moving a sandbar makes realizing a broader vision seem more possible (see also Butterfield and Padavic, 2014): one set of families embracing egalitarianism—one wave—has little effect; one company’s action to humanize work demands—another wave—similarly has little effect. But as the waves continue—as other families and companies follow—the old gender system will be eroded to the point of irrelevance. Only when women and men can pursue their lives so that the demands and gratifications of one domain—whether work or home—need not take precedence over the other will women achieve workplace parity with men.
Supplemental Material
DS_10.1177_0001839219832310 – Supplemental material for Explaining the Persistence of Gender Inequality: The Work–family Narrative as a Social Defense against the 24/7 Work Culture
Supplemental material, DS_10.1177_0001839219832310 for Explaining the Persistence of Gender Inequality: The Work–family Narrative as a Social Defense against the 24/7 Work Culture by Irene Padavic, Robin J. Ely and Erin M. Reid in Administrative Science Quarterly
Footnotes
Acknowledgements
We thank Joyce Fletcher, Debbie Kolb, Julia Moore, and Spela Trefalt for their contributions to the initial project that provided the impetus for this paper; William R. Earnest for his helpful psychoanalytic insights; and Ivan Barron, Sharon Mozgai, and Steven Shafer for their research assistance. We are also grateful to the following people for their helpful feedback on earlier versions of this paper: Elaine Backman, Lotte Bailyn, Hannah Riley Bowles, Emilio Castillo, Elizabeth Hansen, Larry Hirschhorn, Barbara Lawrence, James Krantz, Kathleen McGinn, Maureen Scully, Susan Silbey, Susan Sturm, and Christine Williams; Michael Pratt, Linda Johanson, and our three anonymous ASQ reviewers; and participants in the Systems Psychodynamics Gathering, INSEAD, December, 2015. Finally, we are grateful to the two senior leaders who served as our liaisons to the firm. This work was largely supported by the Division of Research at Harvard Business School. The firm funded data collection for the initial project; these data comprised 19 of the 107 total interviews we conducted for this research. The firm had no role in the design of the study, in the analysis or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.
Notes
Authors’ Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
