Going Offshore: Horseplay, Normalization, and Sexual Harassment COLUMBIA JOURNAL OF GENDER AND LAW GOING OFFSHORE: HORSEPLAY, NORMALIZATION, AND SEXUAL HARASSMENT CHRIS DIFFEE* On November 7 or 8, Oncale was present when a new galleyhand complained to the Chevron company representative in charge of the Chevron platform about Pippen kissing that galleyhand and Lyons telling that galleyhand that Lyons "loved him" and wanted to 'fuck him in his butt. " During a safety meeting the next morning, the Chevron company representative told Lyons and Pippen "not to mess with the galleyhand anymore," explaining that the galleyhand was "new" and "green to the field" and that the galleyhand "didn't know how it was offshore. INTRODUCTION When Oncale v Sundowner Offvhore Services was decided in 1998, many people undoubtedly "didn't know how it was offshore."'2 Like the galleyhand's abuse described above, Joseph Oncale's harassment occurred while working "offshore" as a roustabout aboard Chevron USA's Ship Shoal 266-A and B drilling platforms with seven other men. Between August and November 1991, Oncale was subjected to an escalating series of verbal threats by coworkers John Lyons, Danny Pippen, and Brandon Johnson.' On several occasions, these coworkers told Oncale, "[y]ou're going to give it to me," while at other times they described how they would eventually, in Oncale's words, "fuck me in my * B.A., 1993, University of North Carolina, Chapel Hill; Ph.D., 2001, University of California, Irvine; J.D., 2007, Stanford Law, School. For perceptive comments and longstanding support regarding this project, I would like to thank Barbara Fried, Bob Weisberg, Rich Ford, Heather Love, Amanda Goodin, Lisa Kinney Helvin, Rob Stockman, Nolan Reichl, Dan Sharfstein, and most especially Teemu Ruskola and Melissa Sanchez. I Brief for Respondents, Oncale v. Sundowner Offshore Srys., Inc., 523 U.S. 75 (1998) (No. 96-568), 1997 WL 634147, at *3 [hereinafter Brief for Respondents]. 2 Id. 3 Brief for Petitioner, Oncale v. Sundowner Offshore Srys., Inc., 523 U.S.75 (1998) (No. 96-568), 1997 WL 458826, at *3-6 [hereinafter Brief for Petitioner]. 4 Brief for Amicus Curiae in Support of Petitioner, Oncale v. Sundowner Offshore Srys., Inc., 523 U.S. 75 (1998) (No. 96-568) [hereinafter MacKinnon Brief], reprinted in 8 UCLA WoNMN's L.J. 9, 13 (1997). 302 24.3 COLUMBIA JOURNAL OF GENDER AND LAw behind."" Oncale was then the victim of three physical attacks on October 25 and 26, 1991 . In the first attack, Oncale was forced onto his knees by Pippen as Lyons unzipped his own pants, removed his penis, and stuck it onto the back of Oncale's head. The next day a similar incident occurred when Oncale was forced to his knees as Lyons placed his penis on Oncale's arm.' Later that night Lyons and Pippen attacked Oncale in the shower.' Pippen allegedly lifted Oncale off the ground as "John Lyons grab[bed] the bar of soap and rubbed it between the cheeks of my ass and t[old] me . .. they are fixing to fuck me."o After Oncale complained to his superiors, Lyons taunted him by saying: "You told your daddy, huh? Well, it ain't going to do you no good because Im going to fuck you anyway." Whereas Oncale's sexual harassment transpired literally offshore-"on the Outer Continental Shelf of the United States." as Oncale's brief to the Supreme Court recounts1 2 Title VII case law presents an itinerary of similarly offshore spaces. In Giddens v. Shell Oil Co., the backdrop was an oil refinery in Odessa, Texas, where Richard Giddens was harassed by coworkers Jack Tucci and Craig Harner.13 On one occasion, Tucci rubbed his crotch through his pants as he stood in front of Giddens. then invited Giddens into his office to see "how hard it would get."14 On another occasion. Tucci and Harner cornered Giddens in the men's room, removed their penises from their pants, and rubbed them on Giddens." Tucci and Harner later attacked Giddens in the refinery control room by hoisting him into the air and holding him upside down while Harner fondled Giddens's genitals and inserted 5 Id. 6 Id. 7 Id. 8 Id. 9 Id. 10 MacKinnon Brief, supra note 4, at 13. 11 Id. at 14. 12 Brief for Petitioner, supra note 3, at *28. 13 Petition for Writ of Certiorari, Giddens v. Shell Oil Co. (5th Cir. 1993) (No. 92-8533) (unpublished), 1994 WL 16099694, at *3. 14 Brief of Cross-Appellee and Reply Brief for Appellant, Giddens v. Shell Oil Co. (5th Cir. 1993) (No. 92- 8533) (unpublished), 1993 WL 13102593, at *4. 15 Id. 24.3 303 COLUMBIA JOURNAL OF GENDER AND LAw his finger into Giddens's rectum." In Mc Williams v. Fairfax Count) Board of Supervisors, the scene was an all-male county mechanics yard in Fairfax, Virginia, where Mark McWilliams's coworkers-known collectively as "the lube boys"-tied McWilliams's hands together, blindfolded him, and forced him to his knees as one man repeatedly inserted his finger into McWilliams's mouth to simulate fellatio.7 In Velnychenko v. 84 Lumber Co., the site was a building materials supply store in West Springfield, Massachusetts, where Leonid Melnychenko was repeatedly harassed by his coworker Richard Raab, who exposed himself. fondled Melnychenko's buttocks, asked Melnychenko for a "blow job," mimicked anal intercourse while standing behind Melnychenko, and told others that Melnychenko had performed fellatio on him.'" In Cummings v. Koehnen, the setting was a trucking company in Rosemount, Minnesota, where Richard Cummings's supervisor would daily ask Cummings, "how about sucking my little dick and mak[ing] it a big dick."" On another occasion, Cummings's supervisor approached Cummings from behind and rocked his hips in a simulation of anal sex as he whispered to Cummings, "Here, let me tell you how a real man takes it.""0 Later the same supervisor asked Cummings to "bend over so I can do you in the ass." 2 1 Oil platforms and refineries, mechanics yards. building supply stores, trucking companies: what binds these otherwise distinct spaces is the quality of being "offshore." Some of these workplaces are located literally offshore in geographically isolated settings like oil-platforms, while most are only metaphorically offshore in the sense of being culturally or institutionally estranged from the broader currents of social and sexual norms. It is also perhaps no coincidence that the majority of these workplaces are either exclusively or predominantly male, populated by men of similar racial. economic, and social backgrounds.22 In fact, even on this preliminary level, we encounter one of the primary contradictions that structures offshore activity. On the one hand, these workplaces recall a fiercely homogeneous workplace culture-typically white, male, working-class- 16 Id. 17 72 F.3d 1191, 1193 (4th Cir. 1996). 18 676 N.E.2d 45, 46 n4 (Mass. 1997). 19 568 N.W.2d 418,420 (Minn. 1997). 20 Id. 21 Id. 22 From this perspective, many of these workplaces could be productively understood as forming variations of what Erving Goffman termed "total institutions." See Erving Goffman, On the Characteristics of Total Institutions, in ASYLUMS: ESSAYS ON THE SOCIAL SrrUATION OF MENTAL PATIENTS AND OTHER INMATES 1 (1961). 304 24.3 COLUMBIA JOURNAL OF GENDER AND LAw that is almost anachronistic today. These workplaces are therefore insular enough to evolve their own informal norms and hierarchies without apology to unrepresented races. classes. and genders: there is no need to be politically correct offshore. But, on the other hand, alongside the homophobia, racism, and sexism that might be expected. these involuted workplaces also cultivate overtly sexualized practices between male workers that would be branded deviant and perverse by mainstream society. Although frequently dismissed as "horseplay," "rough-housing," or "locker-room antics," such terms cannot hide the actual severity, aggression, and sexuality evident in cases such as Giddens, Mc Williams, and Oncale. The result is, quite unexpectedly, a paradoxical expression of familiar masculine hierarchies through an explicitly sexual lexicon. Even as these deeply gendered, aggressive assertions of sexual power inevitably maintain a rejection of outright "homosexuality"- they are replete with simulations and invocations of anal intercourse, fellatio, and rape that, in the same gesture, mimic and disavow those practices-they nonetheless acknowledge that masculinity might be made, as well as unmade, through an encounter with traumatizing homoerotic abuse. In fact, part of the potency of these acts derives from their deliberate perversity. Rejecting the norms of the "onshore" world, these harassers enact a deviant, because desiring, masculinity-asking their victims for a "blow job" or inviting them to see "how hard it would get"-whose power to stigmatize comes from a manhood already hardened to, and perhaps also by, abuse. In the words of Richard Cummings's supervisor. a "real man"' can "do you in the ass" only because he knows himself "how a real man takes it.", Some men, however, have not taken it very well. Since the early 1980s, an increasing number of men have brought same-sex sexual harassment claims under Title VII of the Civil Rights Act of 1964.23 Originally, both courts and feminist legal scholars experienced some difficulty accommodating offshore activity in their theories of sexual harassment. Courts, for their part, rejected same-sex sexual harassment claims for a variety of reasons. Some courts reasoned that Title VII was historically intended to protect a "discrete and vulnerable group"' against discrimination by the majority.2 4 Hence, a successful plaintiff would need to demonstrate that an "anti-male environment" existed in the workplace. Other courts concluded that sexual harassment is principally unwelcome sexual advances, 23 Title VII of the Civil Rights Act of 1964, 42 U.S.C. ยง 2000e (2010), states that it is "an unlawful employment practice for an employer ... to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual's race, color, religion, sex, or national origin." 24 See, e.g., Goluszek v. H. P. Smith, 697 F. Supp. 1452, 1456 (N.D. Ill. 1988). 25 Id. 24.3 305 COLUMBIA JOURNAL OF GENDER AND LAW so the harasser's sexual orientation, not the victim's gender, is determinative." Even after the Supreme Court recognized same-sex sexual harassment claims in Oncale. it offered little guidance on how a plaintiff could prove such claims. Feminist legal scholarship has been positively reinvigorated by same-sex sexual harassment cases. Such cases prompted feminists to return to first principles and rethink the basis of sexual harassment as such. Originally, in the early theories of radical feminists such as Catharine MacKinnon, sexual harassment represented only the most undiluted form of the sex inequality pervading society7. However, it was not readily apparent what significance, if any, same-sex sexual harassment might have. Consequently, in the late 1990s. same-sex sexual harassment pushed feminist legal scholars to recognize that, rather than establishing the unilateral power of men over women, sexual harassment policed the gender roles of both sexes. According to scholars Katherine Franke, Vicki Schultz, and Kathryn Abrams, men who are perceived as gay, effeminate, or sexually inexperienced become victims of sexual harassment as punishment for nonconformity." Women are caught in an even more dire predicament because they are victimized both for fulfilling gender stereotypes and for deviating from them.2 9 Building on these accounts, courts have more recently agreed to extend the sexual stereotyping claim from Price Waterhouse to include male same-sex victims.3o Accordingly, over the intervening decade, relatively little new theoretical work has appeared as a practical and theoretical consensus formed that sees gender norms and sex stereotyping as the proper frameworks through which to approach sexual harassment. 26 See, e.g., Hopkins v. Baltimore Gas & Elec. Co., 77 F.3d 745,752 (4th Cir. 1996). 27 See, e.g., CATHARINE A. MACKINNON, SEXUAL HARASSMENT OF WORKING WOMEN: A CASE OF SEX DISCRIMINATION 1-2 (1979). 28 See Kathryn Abrams, The New Jurisprudence of Sexual Harassment, 83 CORNELL L. R\v. 1169, 1226 (1998); Katherine M. Franke, Whats Wrong with Sexual Harassment?,49 STAN. L. Rv. 691,766-71 (1997); Vicki Schultz, Reconceptualizing Sexual Harassment, 107 YALE L. J. 1683, 1774-89 (1998). 29 See Price Waterhouse v. Hopkins, 490 U.S. 228,251 (1989) (describing this predicament as the "catch-22" of gender discrimination); Franke, supra note 28, at 764-66 (analyzing examples). 30 See Prowel v. Wise Bus. Forms, Inc., 579 F.3d 285,290 (3d Cir. 2009); Vickers v. Fairfield Med. Ctr.,453 F.3d 757,762 (6th Cir. 2006); Dawson v. Bumble & Bumble, 398 F.3d 211,218 (2d Cir. 2005); Smith v. City of Salem, Ohio, 378 F. 3d 566,571-75,921 (6th Cir. 2004); Hamm v. Weyauwega Milk Prods., 332 F.3d 1058, 1062 (7th Cir. 2003); Nichols v. Azteca Rest. Enters., Inc., 256 F.3d 864, 874-75 (9th Cir. 2001); Spearman v. Ford Motor Co., 231 F.3d 1080, 1085 (7th Cir. 2000); Doe v. City of Belleville, 119 F.3d 563, 581 (7th Cir. 1997), vacated, 523 U.S. 1001 (1998) (remanding in light of Oncale). 31 As I discuss in great detail below, see infra Part IIA, there are several possible reasons for the lack of new theoretical work on sexual harassment. Typically, new theoretical work emerges in response to an 306 24.3 24.3 COLUMBIA JOURNAL OF GENDER AND LAw 307 Excluded from this consensus are Joseph Oncale and other victims of offshore activity. For, with a few notable exceptions, neither courts nor feminist scholars would find Joseph accumulation of instances that appear anomalous from the vantage point of existing conceptual frameworks. In the late 1990s, sex stereotyping provided both a theoretical and practical model able to accommodate otherwise inexplicable same-sex harassment, while also recognizing female sex agency and avoiding the pathologization of sexuality associated with MacKinnon's subordination account. Since the important work of Franke, Schultz, and Abrams, there has been a relative scholarly consensus (with two notable exceptions described below). The impetus for my own Article comes from several sources. First, there has been a significant body of scholarship concerning the performativity of gender and sexuality, perhaps most strongly associated with the work of Judith Butler but extended by numerous other scholars across other lines of identity and identification drawing on a variety of critical idioms. See, e.g., DAVID ENG, RACIAL CASTRATION: MANAGING MASCULINITY IN ASIN AMERICA (2001); JUDITH HALBERSTAM, FEMALE MASCULINITY (1998); Jost ESTEBAN MUNOZ, DISIDENTIFICATIONS: QUEERS OF COLOR AND THE PERFORMANCE OF PoLrrics (1999). Second, a separate body of scholarship had developed that explores the meaning of normalization as a complex historical, statistical, and social concept, often in conjunction with Michel Foucault's work on "govemmentality." See, e.g., MICHEL FOUCAULT, SECURITY, TERRITORY,POPULATION: LEcTuREs AT THE COLLEGE DE FRANCE 1977-1978 (Michel Senellart ed., Graham Burchell trans., 2009); IAN HACKING, THE TAMING OF CHANCE (1990); Francois Ewald, Norms, Discipline, and the Law, 30 REPRESENTATIONS 138 (1990). Finally, the practical and doctrinal limitations of sex stereotyping have become clear in the decade since Franke, Schultz, and Abrams published their accounts. See, e.g., Vickers, 453 F.3d at 763 (rejecting plaintiff's sex stereotyping claim after finding his harassment was based on "perceived homosexuality," rather than on "gender non-conformity"); Dawson, 398 F.3d at 222-23 (rejecting plaintiff's sex stereotyping claim after finding she provided no substantial evidence that her gender nonconformity led to an adverse employment action); Hamm, 332 F.3d at 1065 (rejecting plaintiff's sex stereotyping claim after finding he was harassed due to his perceived homosexuality, rather than his gender nonconformity); Simonton v. Runyon, 232 F.3d 33, 36 (2d. Cir. 2000) (same); Spearman, 231 F.3d at 1085-86 (same). However, against this backdrop, there have been two new lines of theoretical inquiry regarding sexual harassment. First, the most radical announcement of a "break" from feminism in general and feminist accounts of sexual harassment in particular would be JANET HALLEY, SPLIT DECISIONs: HOW AND WHY To TAKE A BREAK FROM FEMINISM 290-303 (2006). I would note, however, that the precise relationship of that break to the influential work of Franke and Schultz is less than clear when, for instance, Halley cites Franke and Schultz approvingly in her footnotes as representing significant interventions within feminist readings of Oncale. See id. at 385 n.9 (approving Schultz); id. at 386 n.10 (approving Franke). So while the reliance of Franke,Abrams, and Schultz on a model of gender hierarchy would clearly fall within Halley's minimalist definition (and criticism) of feminist models of power, it seems Halley also feels that the accounts of Schultz and Franke depart in significant ways from the theories of cultural and dominance feminists that are the primary targets of her critique. See id. at 17-18, 359-60. The second exception has been accounts of "second generation" discrimination arising from cognitive bias, structures of decision-making, and patterns of group interaction that work to exclude nondominant groups. See Tristin K. Green, Discrimination in Workplace Dynamics: Toward a Structural Account of Disparate Treatnent Theory, 38 HARV. C.R.-C.L. L. REV. 91 (2003); Linda Hamilton Krieger, The Content of Our Categories: A Cognitive Bias Approach to Discrimination and Equal Employment Opportunity, 47 STAN. L. REV. 1161 (1995); Susan Sturm, Second Generation Employment Discrimination: A Structural Approach, 101 COLUM. L. REV. 458, 465-74 (2001). However, the focus of these scholars typically falls on disparate treatment rather than sexual harassment and, to the extent that their theories analyze the cognitive foundations of unconscious stereotyping, the) can be understood as extending the work of Franke, Schultz, and Abrams. See, e.g., Krieger, supra, at 1189-98 (explaining the cognitive origins of stereotypes). COLUMBIA JOURNAL OF GENDER AND LAw Oncale to be the victim of sexual harassment. For many courts, all the sexualized language and eroticized acts in the world do not alter the fact that these are heterosexual men using sexual innuendo merely to tease a male coworker. Oncale's tormentors are not a group of "gay harassers" acting out of sexual desire, nor is Oncale himself part of a "discrete and vulnerable group"' deserving of Title VII protection. Likewise, for feminist scholars. Oncale is no "gender outlaw" being punished for his failure to conform to stereotypes of straight white masculinity. Instead, Oncale is disappointingly ordinary and, without visible evidence of gender nonconformity, would be unable to succeed under the sex stereotyping theory from Price Waterhouse. Oncale's harassment therefore falls within the interstices of both theories, lost in the no-man's land of cases routinely dismissed as "horseplay."' In this Article, I contend there is much to be gained by going offshore and placing cases of horseplay at the center of our thinking about sexual harassment. Due to their very marginality, cases of horseplay illustrate the limits that circumscribe our current legal and theoretical understanding of sexual harassment.3' And. to the extent that same-sex sexual harassment appears contradictory-an assertion of masculine norms via same-sex acts that violate those norms-it offers insight into the cultural contradictions that inhabit our contemporary social and political conjuncture. Rather than understanding sexual harassment as the expression of a socioeconomic system of gender hierarchy, we must instead situate it within the changing contours of white working-class masculinity. which is itself the product of a historically-conditioned set of social and economic relationships. Based on this intersectionalist vision of the perpetrators of horseplay, I argue that if the white male worker could emerge as the normative embodiment of the postwar U.S. social order (within a certain ideological configuration, at least), that status has come under increasing duress since the 1960s. Especially after the mid-1970s. white working- class men felt their fortunes stagnate and then decline with the end of the post-World War II economic boom and the disappearance of steady, high-paying union jobs. More fundamentally, as the nature, structure, and composition of labor has altered. the ability of the workplace to function as the site for the reproduction of white working-class manhood has been undermined. The industrial workplace has been superseded by forms of more flexible. affective, and information-based production, while the face of the working class has become female. non-white, and increasingly young. As a result, the position of white working-class masculinity has become even more contradictory than is usually recognized. Allan Bakke's famous 1974 lawsuit, for instance, is animated by more than the sense of forfeited power; its appeal to white male injury mimics the very post-civil rights social movements it would seek to displace. This historical perspective thus demonstrates Why 32 See Franke, supra note 28, at 694,762 (arguing that same-sex harassment cases highlight the underlying lack of a coherent theoretical framework for sexual harassment in general). 308 24.3 COLUMBIA JOURNAL OF GENDER AND LAw it would be reductive to identify sexual harassment with the confident assertion of male power in any direct or unmediated sense. Looking at same-sex sexual harassment more specifically, we can see how the offshore world of horseplay registers the contradictory position of working-class white manhood in the post-1960s era. As white male workers perceive themselves as increasingly marginalized both socially and economically, the fractures in manhood precipitate forms of acting out that attempt to consolidate threatened social prerogatives, but through seemingly perverse sexual forms of power. Yet such sexual acts appear perverse only if sexual norms are narrowly identified with the heterosexual/homosexual binary. Instead, we must recognize that contemporary sexuality contains numerous overlapping and contradictory definitional strands that have become submerged within, but not effaced by, the spread of the binary model of sexual identity in the postwar era. All-male environments, in particular, ranging from prisons to offshore workplaces, have played a key role in the transmission of such alternate forms of gender and sexuality, which represent a variety of cultural scripts available for opportunistic retrieval in the present. Same-sex horseplay may be understood as one such opportunistic reactivation: an attempt to negotiate the social contradictions of white working-class masculinity by reviving a specifically sexed and visceral enactment of workplace power. Furthermore, what happens offshore does not stay offshore. Horseplay, in fact, offers us a privileged site from which to rethink sex-based power beyond the hierarchical. structural. and categorical accounts that currently organize our theoretical understanding. First horseplay permits a shift away from a "top-down" emphasis on social structure towards a more nuanced understanding of how situated social actors seek to negotiate worlds not of their own making. Rather than viewing sexual harassment as a disciplinary practice that reinforces pre-existing gender norms, we can use horseplay to model how individuals actively appropriate and rework cultural materials for their own historical needs, needs which cannot be confined to the boundaries of sex. Those needs may take a highly visible (because contradictory) form in the case of horseplay, but they are no less present in more familiar forms of male-female harassment. Second, horseplay works to illustrate how sex- based practices invoke, inhabit, and rework broader cultural norms. which can themselves be understood as the sedimented products of social practice.3 3 At first glance, horseplay 33 References to "sex-based" or "sex- specific" practices are used throughout this Article to describe any conduct that either contains a specific reference to gender (for instance, gender epithets, statements regarding women's unsuitability for certain occupations, or graffiti that demeans women) or is overtly sexualized (for instance, physical groping of another's genitals, sexual propositions or descriptions of sexual practices, or acts that simulate sexual practices). These terms are intended to distinguish such conduct from sex-neutral conduct, including pretextual employment actions and nonsexual, gender-neutral harassment. 24.3 309 COLUMBIA JOURNAL OF GENDER AND LAw might appear to be a curious choice, for its sexualized acts appear to cut against the grain of our contemporary image of masculinity. Yet, as I will argue, it is a mistake to think that such acts are "outside power" just because they fail to fit within an already existing set of sexual norms.34 Rather, these are instances when the tensions and unevenness of the normative field are thrown into relief and the formation and deformation of cultural hegemonies made more transparent. Third, horseplay can offer an alternate perspective on the sex-specific acts appearing in both opposite-sex and same-sex harassment. As the historical analysis of horseplay demonstrates, such acts are not reducible to expressions of sexual identity or psychological desire, nor need they be motivated by deep-seated prejudices based on gender. They are, rather, a variety of cultural scripts opportunistically taken up and redeployed by individual social actors under certain historical exigencies. Here, the important doctrinal thrust of these theoretical arguments becomes clear. In complicating and challenging the accounts of Franke, Schultz, and Abrams regarding sexual harassment, this Article is not proposing that courts adopt an even more complex and historically- situated understanding of the dynamics behind sexual harassment. The crucial point is that, when courts evaluate whether the work environment discriminates on the basis of sex, they need not (and should not) attempt to decipher the reasons behind harassment at all. Instead, they typically need to consider only the objective, sex-based form of the work environment itself. From this perspective, the hostile environment, as a distinct paradigm of employment discrimination, does not seek to intervene in the motives or hidden prejudices behind the conduct. but to prohibit harassment (occurring for whatever psychological or cultural reasons) when it takes up a sex-specific form.35 Although a subsequent article will develop this doctrinal argument in necessary detail, the present Article offers a conceptual context for this legal claim by demonstrating how sex-based acts take up and rework broader contours of social power. But. at the onset. it is important to clarify that. from a legal and doctrinal perspective, the reasons that an individual has generated a sexually- demeaning environment-for instance, whether out of individual animosity or a drive to punish individuals who fail to conform to gender norms-is strictly irrelevant to the legal definition of sexual harassment. And here, again, we return to horseplay. For horseplay is exemplary in this regard: it is, by cultural definition, conduct that takes an explicitly sex- based form, but occurs for no discernible reason and even in violation of prevailing social logics. All sexual harassment should be regarded as just so much horseplay. 34 See infra Part II.C. 35 To be sure, not all unlawful workplace discrimination need take the form of sex-specific conduct. As Vicki Schultz, in particular, forcefully reminds us, Title VII equally prohibits sex-neutral forms of unequal treatment, including pretextual employment decisions and harassment that does not have an explicitly gendered content. See e.g., Schultz, supra note 28; see also Franke, supra note 28, at 709-10. 310 24.3 COLUMBIA JOURNAL OF GENDER AND LAW A subsequent article will develop at length the proper legal basis of the sex-based hostile environment claim." This Article works first to clear the ground for that account by intervening within the theoretical debates that have come to structure our contemporary understanding of sexual harassment. It consists of five parts. Parts I and II set the stage by reviewing the accounts of same-sex sexual harassment offered by courts and feminist legal scholars, respectively. Part I recounts the judicial response to the rise of same-sex sexual harassment. It argues that, rather than seeing all-male sexual harassment as something wholly new, courts retrieved various rationales embedded in more traditional sexual harassment jurisprudence in order to come to terms with sexualized abuse among men. The result demonstrated the underlying inadequacy and inconsistency still pervading traditional sexual harassment jurisprudence. In Oncale, the Supreme Court only deepened this disarray by recognizing a same-sex cause of action, but rendering it almost impossible to prove. Part II turns to the pivotal scholarship of Katherine Franke, Vicki Schultz. and Kathryn Abrams. who, in articles published nearly simultaneously in 1997 and 1998, provided the critical and legal framework missing from the Court's opinion in Oncale. By focusing on the cultural work of gender norms, these feminist scholars provided a much needed paradigm shift away from the model of unwelcome sexual conduct towards an understanding of how gender stereotypes and nonsexual forms of workplace discrimination work to reaffirm overarching structures of male power. Yet neither the legal recognition of the courts nor the innovative theories of feminists extended the protections of Title VII to Joseph Oncale and the forms of offshore sex-based power his abuse exemplify. Feminist accounts, in particular, have continued to define both gender norms and the cultural practice of sexual harassment in terms of a model of gender hierarchy. But retaining that model of power neglects what is most crucial in the performative theory of gender: the critical recognition that sex-based practices actively appropriate, inhabit, and rework the normative field and that these norms do not form a homogenous or unified whole-as the model of a hierarchy might be read to suggest-but consist of various contradictory and sometimes anomalous cross-currents that nonetheless participate in the constitution and deconstitution of power. The remainder of this Article attempts to take up the specific dynamics identified in these feminist accounts and to incorporate them within a revised theoretical and historical model of sex-based power. Part III reworks references to an abstract male normativity by tracing the redefinition of white working-class masculinity after World War II. If the particular historical normativity of white working men was due, in significant part, to their position in the broader articulation of Fordism and Keynesianism in this era-in which a system of mass production and mass consumption was underwritten by government policies aimed at stimulating national demand and ensuring social stability-then the economic, social, 36 For a summary of that argument, see infta Part V. 24.3 311 COLUMBIA JOURNAL OF GENDER AND LAw and political turmoil of the 1960s and 1970s wrought a decisive warping in that social identification. Regardless of the specific form of sexual harassment at issue-whether the anomalous conduct in Oncale or the stereotypical gender roles witnessed in Meritor Saving Bank-it is historically nafve to ascribe those acts to a monolithic normative masculinity severed from its relationship with specific economic, political, and technological regimes. Finally, in Part IV. I take up contradictory instances of same-sex horseplay in order to offer a revised theory of sexual harassment more generally. This section argues that horseplay represents one attempt to resolve the fraught position of working-class white manhood in the post-civil rights era. Abusive forms of sex dominance that appear deviant from the perspective of contemporary U.S. society can be understood, then, as the reactivation of certain modes of working-class male sexuality that have been historically associated with all-male environments. Such sex-specific acts of dominance seek to restore a sense of power in its most literal and bodily dimension in the face of perceived economic and social decline. Pointing to this convergence in cultural, historical, and sexual dynamics, I contend that horseplay can offer us a privileged cultural site from which to reconceptualize the theoretical and historical dimensions of sexual harassment more broadly in both its onshore and offshore forms. In a final Coda. I briefly outline the doctrinal implications of this theoretical model, arguing for a shift from the analysis of social function to the objective, sex-specific nature of certain workplace environments. I. Judicial Understandings of Same-Sex Sexual Harassment Prior to Oncale, a three-way split had formed among the federal appellate courts that had addressed in what ways, if any, same-sex sexual harassment could satisfy Title VII's causality requirement that discrimination occur "because of sex."Yet that split had less to do with same-sex sexual harassment than with the overlapping rationales that had converged within traditional, male-female sexual harassment claims as each circuit came to base its understanding of same-sex sexual harassment on one strand of traditional sexual harassment jurisprudence. The result was a disparate set of conflicting approaches to same- sex sexual harassment that revealed the incoherence of sexual harassment jurisprudence more broadly. The Fifth Circuit was the first federal court of appeals to consider the issue. and in Giddens v. Shell 0I,7 then Garcia v. ElfAtochen North America, 8 and finally Oncale," 37 12 F.3d 208 (5th Cir. 1993) (per curiam) (unpublished). 38 28 F.3d 446 (5th Cir. 1994). 39 83 F.3d 118 (5th Cir. 1996). 312 24.3 COLUMBIA JOURNAL OF GENDER AND LAW it categorically rejected same-sex claims as forms of "horseplay"40 that did not present the kind of gender discrimination Congress sought to prohibit through Title VII. In Mc Williams, the Fourth Circuit seconded that analysis when the conduct involved heterosexual men.41 but found in Wrightson v. Pizza Hut that unwelcome sexual advances of a "gay harasser" violated Title VII as a form of disparate treatment.42 The Eighth Circuit bracketed difficult issues of motivation in Quick v. Donaldson Conpany, Inc.4 3 and simply insisted that plaintiffs provide comparative evidence that men and women were treated differently within the workplace, a model that plainly ignored the problem of single-sex environments. Finally, in Doe v. City of Belleville,44 the Seventh Circuit held that plaintiffs might prove causation either through the explicitly sexual nature of the conduct (as routinely occurred in numerous male-female cases) or by demonstrating the conduct's basis in sex stereotyping (based on Price Waterhouse). Such was the state of same-sex sexual harassment jurisprudence in 1998 when the Supreme Court decided Oncale. In a pithy, six-page opinion, Justice Scalia reversed the Fifth Circuit's categorical ban on same-sex harassment claims and held that such claims were actionable under Title VII. 5 Yet whatever guarded optimism scholars initially felt towards Scalia's opinion was quickly replaced with a tally of its insufficiencies. These scholars noted that the Court failed to adopt the sex-stereotyping theory proposed in Doe and advocated by amici;46 that it set forth a presumption of heterosexuality that would permit an inference of sexual harassment in male-female claims but not in same-sex claims; 4 7 and 40 Garcia, 28 F.3d at 448. 41 72 F.3d 1191, 1195-96 (4th Cir. 1996). 42 99 F.3d 138, 143 (4th Cir. 1996). See also Year) v. Goodwill Industries-Knoxville, Inc., 107 F.3d 443 (6th Cir. 1997); Fredette v. BVP Mgmt. Assocs., 112 F.3d 1503 (11th Cir. 1997) (finding that unwelcome sexual advances by a "gay harasser" present one possible basis on which a find-finder might infer causality, but not deciding if other bases might also exist). 43 90 F.3d 1372 (8th Cir. 1996). 44 119 F.3d 563 (7th Cir. 1997), vacated sub nom. City of Belleville v. Doe, 523 U.S. 1001 (1998). 45 523 U.S. 75,79-80 (1998). 46 See, e.g., David S. Schwartz, WAhen Is Sex Because of Sex? The Causation Problem in Sexual Harassment Law, 150 U. PA. L. REV. 1697, 1742-43 (2002). The sexual stereotyping argument was placed squarely before the Court by Franke and Nan Hunter in the Brief of Law Professors as Amici Curiae in Support of Petitioner, Oncale v. Sundowner Offshore Servs., Inc., 523 U.S. 75 (1998) (No. 96-568) [hereinafter Franke & Hunter BriefJ. 47 See, e.g., Schwartz, supra note 46, at 1745-47. 24.3 313 COLUMBIA JOURNAL OF GENDER AND LAw that by explicitly embracing the Fourth Circuit's theory of the "gay harasser," the Court had opened the door to homophobic claims in which avowedly straight employees charged gay coworkers with sexual harassment as the result of their own fantasized projections of same-sex desire.4 In retrospect, what is significant about Scalia's opinion is not merely its inadequacies, but also the profound contradiction that it represents. For even as Scalia offers legal recognition to same-sex sexual harassment claims, he also deliberately refuses to provide a viable evidentiary framework for those same claims. As David Schwartz has argued, perhaps the most satisfying way of understanding Scalia's opinion is to read Oncale as refraining sexual harassment as reverse discrimination. 4' For, as Schwartz notes, the final winners of Oncale are not gay men or lesbians harassed for the ways they enact their genders and sexualities, but straight white men who are preyed upon by a "gay harasser.""o Moreover, Scalia's analysis unsurprisingly places sexual harassment within the most conservative paradigm ofTitle VII discrimination: the disparate treatment ofmale and female employees. Such a paradigm is emphasized, for instance, when Scalia identifies sexual harassment with Justice Ginsburg's oft-cited statement from Harris: "The critical issue, Title VII's text indicates, is whether members of one sex are exposed to disadvantageous terms or conditions of employment to which members of the other sex are not exposed.""' But with gender discrimination defined exclusively as the unequal treatment of men and women, sexual harassment between members of the same gender becomes difficult to conceive. This lack of theoretical recognition leads to the near practical evisceration of same-sex sexual harassment claims, which is demonstrated by the evidentiary routes that Scalia offers for same-sex claims.5 Scalia outlines three such routes: first. credible evidence 48 See, e.g., Mary Coombs, Title VII and Homosexual Harassment After Oncale: Was It a Victory?,6 DUKE J. GENDER L. & POL'Y 113, 144-45 (1999); Schwartz, supra note 46, at 1746-47; HALLEY, supra note 31, at 296. 49 Schwartz, supra note 46, at 1747-48. 50 Id. Oncale can thus be understood as a part of a more general trend to establish straight white men as the legitimate victims of discrimination and therefore to further elide the actual inequalities that continue to structure the U.S. workplace. See generally Reva B. Siegel, Equality Talk: Antisubordination and Anticlassification Values in Constitutional Struggles Over Brown, 117 HARV. L. REV. 1470 (2004) (using Brown v. Board of Education to trace the development of antisubordination and anticlassification interpretations of the Equal Protection Clause). 51 Oncale v. Sundowner Offshore Srys., Inc, 523 U.S. 75, 80 (1998) (quoting Harris v. Forklift Sys., Inc., 517 U.S. 17, 25 (1993)). 52 See also Schwartz, supra note 46, at 1736-37 (discussing the deficiencies of these routes). 314 24.3 COLUMBIA JOURNAL OF GENDER AND LAw that the harasser was homosexual; second, evidence that the harassment was motivated by general hostility to one gender, as shown. for instance, by sex-specific derogatory language; and, third, direct comparative evidence revealing disparate treatment of men and women in a mixed-sex workplace." The limits of these evidentiary routes are readily apparent. The latter two routes scrupulously avoid the underlying problem of Oncale: how to prove discrimination in a single-sex environment. Comparative evidence is literally impossible in such workplaces, and a general hostility towards men in general presumes a virulent self-loathing not often found in men. As a result. the only remaining way of proving sexual harassment in all-male workplaces is through evidence of the harasser's sexual orientation-providing valuable protection to the straight male population from the predatory desires of gay men. The limited reach of these evidentiary routes is amply demonstrated in subsequent case law.54 For instance. in VcCoiin v. St. John Health System, the plaintiff. James McCown. was allegedly sexually harassed after being transferred to an all-male projects shop as a construction worker. There, McCown's supervisor, Lloyd Soller, allegedly committed a series of inappropriate acts, including "grabbing McCown by the waist, chest, and buttocks; grinding his genitals against McCown's buttocks in simulated intercourse; telling McCown to 'squeal like a pig, or a woman.' and making other lewd comments; attempting to stick the handle of a shovel and a tape measure in McCown's anus; and kicking McCown in the buttocks.""' Looking to Oncale's evidentiary routes, the Eighth Circuit found no evidence of sexual harassment.1 First, the court rejected even the idea of "comparative evidence" regarding Soller's treatment of men and women, noting that the projects shop was an exclusively male environment and no evidence of differential treatment was possible." Second, the court failed to find evidence that Soller harbored any general prejudice towards men." Indeed, to the extent that the conduct constituted 53 Oncale, 523 U.S. at 80-81. 54 See Schwartz, supra note 46, at 1734-36 (describing how, on remand, Oncale's motion for summary judgment was rejected after he relied on two of the Court's proposed evidentiary routes). 55 349 F.3d 540, 541 (8th Cir. 2003). 56 Id. at 541-52. 57 Id. at 543. 58 Id. at 543-44. 59 Id. at 544. 24.3 315 COLUMBIA JOURNAL OF GENDER AND LAw "teasing" or "roughhousing," it was a sign of male camaraderie."o Third, the court found no evidence that Soller was "homosexual and motivated by sexual desire for Soller," despite the explicitly sexual nature of the conduct." AcCown thus makes explicit what Oncale had suggested: that the abstract recognition of same-sex sexual harassment. especially in all- male workplaces, was not supported by the necessary theoretical or practical framework it would need. It was up to feminist legal scholars, writing almost at the same time as the Oncale Court, to provide that absent framework through the concept of gender norms. II. Feminist Theories of Gender-Based Harassment A. Sexual Harassment as Gender Normalization Whereas many courts displayed uncertainty and not a little hostility towards same-sex sexual harassment claims up through Oncale, feminists used these claims as the basis for a more profound rethinking of why sexual harassment constituted gender discrimination. Such reconsideration occurred in 1997 and 1998 on the eve of Oncale through the almost simultaneous publication of articles by Katherine Franke, Vicki Schultz, and Kathryn Abrams. Although each of these articles explicitly addressed the problem of same-sex sexual harassment-seeing it as a symptomatic limitation in prior theories of sexual harassment-each essay also sought to correct more general theoretical problems that had arisen in the understanding of sexual harassment over the previous two decades. There were perhaps three main theoretical problems these scholars sought to redress. First, MacKinnon's formulation of sexual harassment, which seemed to rest upon a rigid and categorical understanding of female oppression, had long been criticized for being essentializing and deterministic." MacKinnon frequently spoke of male supremacy as a seamless and totalizing system of power in which the only positions were domination by men and subordination for women. But what had appeared as provocative and radical 60 Id. 61 McCown, 349 F.3d at 544. 62 See, e.g., Abrams, supra note 28, at 1201-02, 1213-14; Franke, supra note 28, at 759-62. See also Angela P. Harris, Race and Essentialism in Feminist Legal Theory, 42 STAN. L. REV. 581, 590-601 (1990); Judith Butler, Disorderly Woman, 53 TRANSITION 86, 90-95 (1991). See generally Katherine Franke, Cunning Stunts: From Hegemony to Desire: A Review of Madonna's Sex, 20 N.Y.U. L. & Soc. CHANGE 549, 555-61 (1993-94) [hereinafter Franke, Cunning Stunts]. 63 See Catharine A. MacKinnon, Feminism, Marxism, Method, and the State: Towards a Feminist Jurisprudence,8 SIGNs 635,638 (1983) ("This defines our task not only because male dominance is perhaps the most pervasive and tenacious system of power in history, but also because it is metaphysically nearly perfect."). 316 24.3 COLUMBIA JOURNAL OF GENDER AND LAW in 1979 felt unnecessarily reductionistic by 1997, if not earlier."4 Second, the courts had never developed a coherent and principled account of why sexual harassment constituted sex discrimination.6 Early courts had concluded that the sexual orientation of the harasser guaranteed that sexual harassment would discriminate between men and women.66 Yet this account framed sexual harassment as misdirected expressions of sexual desire, downplaying or ignoring nonsexual forms of gender-based harassment. 1 Other courts, unwilling fully to endorse MacKinnon, nonetheless found that the presence of sexuality in the workplace was detrimental to women for often unstated reasons." The subsequent inability of courts to respond coherently to same-sex sexual harassment claims brought this underlying lack of a satisfactory theoretical account to the fore.' Third, in describing the harm of sexual harassment. both MacKinnon and courts tended to pathologize sexuality as a force uniquely harmful to women.70 Ever since Barnes, courts had embraced the idea that sexual harassment stemmed from a clueless man's unwelcome expression of sexual desire that positioned its recipient as the passive object of male desire. MacKinnon, of course, rejected the idea of sexual harassment as misdirected desire and insisted that sexuality constituted the means of male domination." Yet with the alliance of MacKinnon and the religious right in support of the Minnesota and Indianapolis anti-pornography ordinances in 1983 and 1984. many feminists recoiled from what they felt were the puritanical implications of 64 See, e.g., John A. Miller et al., Comments on MacKinnon s "Feminism, Marxism, Method, and the State," 10 SIGNS 168 (1984). For a response to some of these points, see Catharine A. MacKinnon, Points Against Postmodernism, 75 CHI.-KENT L. REV. 687 (2000). 65 Franke, supra note 28, at 729-62; Schultz, supra note 28, at 1710-55. 66 Franke, supra note 28, at 706-12, 730-31; Schultz, supra note 28, at 1701-04. 67 Schultz, supra note 28, at 1732-37. 68 See Franke, supra note 28, at 692 (describing the "avoidance technique" of courts unwilling to embrace MacKinnon, but convinced that sexual harassment represents a form of gendered power); id. at 714-25 (discussing the claim that sexual harassment is discriminator) because it is sexual). 69 Id. at 694-95. 70 See, e.g., Abrams, supra note 28, at 1201-02; Franke, supra note 28, at 714-25, 730-47; Schultz, supra note 28, at 1698-99, 1704-05. See also Kathryn Abrams, Sex Wars Redux: Agency and Coercion in Feminist Legal Theory, 95 CoLum. L. REV. 304, 307-14 (summarizing challenges by pro-sex feminists to dominance feminism). 71 See Barnes v. Costle, 561 F.2d 983, 988-92 (D.C. Cir. 1977); see also Franke, supra note 28, at 730-47; Schultz, supra note 28, at 1701-04. 72 MacKinnon, supra note 63, at 646-47; Catharine A. MacKinnon, Feminism, Marxism, Method and the State: An Agenda for Theory, 7 SIGNs 515, 531-33 (1982) [hereinafter MacKinnon, Agenda for Theory]; CAHIARINE A. MAcKINNoN, Sexuality, in TONWARD A FEMINIST THEORY OF THE STATE 126 (1989). 24.3 317 COLUMBIA JOURNAL OF GENDER AND LAw MacKinnon's radical feminism. 3Bv the late 1990s. a growing number of scholars felt that both the accounts of MacKinnon and the courts hinged on an essentialization of sexuality as something foreign and dangerous to women that ignored their actual sexual agency.74 The accounts of Franke. Schultz, and Abrams each sought to address these problems in different ways. But one key feature that runs across their responses is an emphasis on gender norms and the cultural work of normalization in understanding sex-based power. Gender norms allow these scholars to reimagine inequality in three crucial respects. First, gender norms offer a new vision regarding the structure of male dominance. In place of any simple identification of female subordination with the biological dominance of men over women, an emphasis on norms highlights the impersonal structure of gender stereotypes and their coercive ability to shape the enactment of both masculinity and femininity). Norms can then explain how both men and women are subject to gender regulation and consequently how even same-sex sexual harassment works to entrench these stereotypes.76 Second, gender norms permit a new understanding of the cultural work performed by sexual harassment. On a local level, norms perform the work of gendering by performatively constructing the genders of both men and women, affirming those who conform with gender expectations and punishing those who deviate. On a more systemic level, gender norms distribute men and women within an overarching sexual hierarchy through the valorization of masculine norms and the devaluation of feminine norms. Third and finally, norms allow a more open approach to the forms that gender harassment 73 See, e.g., Lisa Duggan et al., False Promises: Feminist Anti-Pornography Legislation, 38 N.Y.L. SCH. L. REV. 133 (1993); Paul Brest & Ann Vandenberg, Politics, Feminism, and the Constitution: The Anti- Pornography Movement in Minneapolis, 29 STAN. L. REV. 607 (1987). 74 See Abrams, supra note 70, at 314-24 (describing responses by the Feminist Anti-Censorship Task Force); Franke, supra note 28, at 741-42, 746-47 (noting radical feminism risks ignoring the sexual agency of women). See generally Katherine M. Franke, Theorizing Yes: An Essay on Feminism, Law, and Desire, 101 COLUM. L. Ruv. 181 (2001) (analyzing the tendency of legal feminists to frame discussions of female sexuality in terms of either heteronormative reproductivity or sexual danger from men); Franke, Cunning Stunts, supra note 62 (discussing post-structuralist alternatives to essentialism). 75 See, e.g.,Abrams, supra note 28, at 1209. 76 See Abrams, supra note 28, at 1225-1229; Franke, supra note 28, at 762-72; Schultz, supra note 28, at 1774-98. 77 See Katherine M. Franke, Gender, Sex, Agency and Discrimination: A Reply to Professor Abrams, 83 CORNELL L. Rv. 1245, 1252 (1998) [hereinafter Franke, Gender, Sex, Agency and Discrimination]; Franke, supra note 28, at 764-72; Schultz, supra note 28, at 1774-77. 78 See Abrams, supra note 28, at 1209; Franke, Gender, Sex, Agency and Discrimination, supra note 77, at 1251-52; Schultz, supra note 28, at 1759-61. 318 24.3 COLUMBIA JOURNAL OF GENDER AND LAw may take. Because gender norms have no necessary link to sexual conduct, they permit a more elastic understanding of the plural social dynamics and workplace practices that may affirm stereotypical gender roles, undermine female competency, and entrench male norms., Indeed, many of the differences in the approaches of Franke, Schultz, and Abrams can be attributed to the different dynamics each scholar emphasizes as part of her account." Having provided some sense of the shared situation and conceptual stance of these accounts. I want now to provide a more nuanced description of their respective theories of sexual harassment. Franke provides the most explicitly theoretical approach to sexual harassment, seeking to offer courts a principled account of Why sexual harassment constitutes gender discrimination under Title VII. For Franke, sexual harassment is a regulatory practice that actively constructs gender roles within a broader structure of subordination she terms, borrowing from Francisco Valdes, heteropatriarchy." As Franke explains, heteropatriarchy describes an interlocking structure of subordination resulting from the ideological conflation of anatomical sex, gender roles, and sexual orientation. It permits Franke to emphasize that power is neither rooted in a set of physical attributes nor possessed by one social group to the exclusion of others. but circulates within a more general structure of regulatory norms that intersect and reinforce other lines of subordination such as compulsory heterosexuality.' It can thus explain how the harassment of gay men and lesbians, as well as "gender outlaws" more broadly functions to perpetuate familiar gender hierarchies.8 Sexual harassment represents one of many regulatory practices that reproduce this system of norms, operating, in Franke's words, as "a disciplinary practice that inscribes, enforces, and polices the identities of both harasser and victim according to a system of gender norms that envisions women as feminine, (hetero)sexual objects, and men 79 Abrams in particular argues that a pluralized understanding of these dynamics is crucial to a revised theory of sexual harassment. See Abrams, supra note 28, at 1202, 1215, 1217. See also Franke, supra note 28, at 730-47; Schultz, supra note 28, at 1769-74. 80 See, e.g., Franke & Hunter Brief, supra note 46, at *14 (asserting that practices that perpetuate sex-role stereotyping are paradigmatic of sex discrimination); Abrams, supra note 28, at 1215 (summarizing the plural dynamics constituting sexual harassment); Schultz, supra note 28, at 1762-69 (defining sexual harassment as "broadened to cover all conduct that is rooted in gender-based expectations," including nonsexual conduct). 81 Franke, supra note 28, at 696,763,739 n.247. See also Franke, Gender, Sex, Agency and Discrimination, supra note 77, at 1251-54 (discussing heteropatriarchy). 82 Franke, supra note 28, at 739 n.247. See also Franke, Gender, Sex, Agency and Discrimination, supra note 77, at 1251-52. 83 Franke, Gender, Sex, Agency and Discrimination, supra note 77, at 1253. 84 Id. at 1250-53. 24.3 319 COLUMBIA JOURNAL OF GENDER AND LAw as masculine, (hetero)sexual subjects."" Sexual harassment may reinforce gender norms in one of two ways. First. sexual harassment polices gender boundaries by punishing men and women who fail to satisfy stereotypical ideals of their respective genders." Second, sexual harassment can serve to reaffirm the existing status quo when it occurs between individuals who already conform to gender expectations: an aggressive male boss enacts and confirms his predatory masculinity when he sexually dominates his female subordinate. who is then socially positioned as weak. vulnerable, and sexually objectified.7 Furthermore, Franke's emphasis on the reflexive dimension of this performance allows her to break from MacKinnon's "transitive" model of power." Whereas MacKinnon tended to invest power within men as a discrete social group that unilaterally deploys that power to perpetuate male domination." Franke can demonstrate how the practice of sexual harassment reflexively constructs the genders of both harasser and victim according to heteromasculine norms.o Schultz, working in a more sociological vein, focuses her "competence-centered" account on gender-based job segregation and the ways that gender harassment maintains the male monopoly over occupational skill. In contrast to the "desire-dominance" paradigm adopted by early feminists and courts," Schultz contends that "a drive to maintain the most highly rewarded forms of work as domains of masculine competence underlies many, if not all, forms of sex-based harassment on the job."92 According to Schultz, the focus upon sexual conduct has obscured more prevalent patterns of nonsexual harassment that undermine the real and perceived competence of female workers. Men preserve a monopoly over the most valued forms of work by maintaining the association of masculinity with work competence and authority:' 4 gender-based harassment positions women as well as nonconforming men as unmanly, incompetent, and unsuited to performing a "man's 85 Franke, supra note 28, at 693. 86 Id. at 764-65. 87 Id. at 766. 88 Id. at 761. 89 Id. at 760. 90 Id. at 693-94,770-71. 91 Schultz, supra note 28, at 1692-1710. 92 Id. at 1755. 93 Id. at 1762-69. 94 Id. at 1756-61. 320 24.3 COLUMBIA JOURNAL OF GENDER AND LAw job."'" Whereas the "desire-dominance" model focused on sexuality as reinforcing gender inequality, Schultz emphasizes the everyday, nonsexual workplace practices that disadvantage female workers, such as providing female employees with patronizing forms of assistance, withholding training or information required to perform an occupational task satisfactorily, and characterizing women as appropriate for only certain stereotypical tasks.'" And because the workplace plays a critical role in the wider reproduction of social power-a nexus for the distribution of economic gain, psychological empowerment, political recognition, and social prestige-the effects of workplace discrimination ripple outward from employment to perpetuate gender inequality across society. Abrams accepts many aspects of Franke and Schultz's accounts but seeks to restore female subordination to the center of our understanding of sexual harassment." As Abrams explains, the important criticisms that feminists have lodged against MacKinnon's account regarding its essentialism and neglect of female sexual agency are not indispensable elements of that theory." Rather, by defining sexual harassment as "a phenomenon that serves to preserve male control and entrench masculine norms in the workplace." Abrams argues that we can continue to place female subordination at the center of our understanding, while still embracing the most diverse forms of workplace conduct as potential sexual harassment."' To define those forms of harassment, Abrams turns to an examination of the workplace as a specific institutional and historical site of gender struggle.I' According to Abrams, particularly prior to the 1960s, a strict sexual division of labor was observed between "pink collar" jobs and the rest of the labor market where male attitudes and expectations were firmly entrenched' 02 Between the 1960s and the 1980s. as women demanded expanded occupational opportunities and entered the workplace in greater numbers, the most blatant forms of sexual harassment emerged as a means of asserting male control. Women who entered traditionally male fields were subject to severe physical or sexual aggression or 95 Id. at 1762, 1769-89. 96 Id. at 1762-66. 97 Schultz, supra note 28 , at 1691-92. 98 See, e.g., Abrams, supra note 28, at 1172. 99 Id. at 1201-02. 100 Id. at 1172. 101 Id. at 1194-98. 102 Id. at 1194-95. 24.3 321 COLUMBIA JOURNAL OF GENDER AND LAw persistent verbal abuse in order to drive them from the workplace.os There were also more subtle ways of reasserting male control: supervisors might sexualize female workers in order to position them as sexual objects rather than competent workers or discipline a woman who behaved in a nonfeminine manner. 1 04 In addition, some men sought to affirm the primacy of male norms, for instance, by engaging in roughhousing and sexualized talk or bringing pornography into the workplace.o An appreciation of the diverse historical forms that workplace harassment has taken allows Abrams to maintain the centrality of female subordination while avoiding the determinism plaguing previous feminist accounts. B. Theoretical Limitations: Hierarchy, Norms, Performativity Undoubtedly one of the most crucial innovations in the shared work of Franke. Schultz. and Abrams has been their ability to free sexual harassment from an essentializing and deterministic fraimework of male power. They have, instead, each emphasized the impersonal nature of gender norms as a regulatory force affecting men as well as women and highlighted the diversity and contingency of workplace practices that perpetuate these norms. Yet, despite these innovations, the accounts of Franke, Schultz. and Abrams continue to rely upon a highly traditional hierarchical model of power.1 o6 In fact, these advances are achieved, not in spite of, but because of their central adherence to a model of gender hierarchy. For each of these accounts defines sexual harassment, not in terms of the form or content of that conduct (sexual or otherwise), but in terms of its "purpose or effect" in reproducing the system of gender hierarchy.o Consequently, each account is free to emphasize a different set of gender dynamics and workplace practices that contribute to the maintenance of hierarchy-heteropatriarchy and gender stereotyping for Franke, job 103 Id. at 1206-07. 104 Abrams, supra note 28, at 1207-08. 105 Id. at 1210. 106 As more than a decade has passed since these articles were published, one imagines that Franke, Schultz, and Abrams would themselves modify their own accounts in certain ways. Yet, because their accounts have assumed a central place in the theoretical understanding of sexual harassment, an) reassessment must be situated in relation to their theories. 107 See, e.g., Abrams, supra note 28, at 1172 (defining sexually harassing actions as those that "preserve male control and entrench masculine norms in the workplace"); Franke, supra note 28, at 772 ("On this account, sexual harassment is sex discrimination precisely because its use and effect police hetero-patriarchal gender norms in the workplace."); Schultz, supra note 28, at 1762 ("[WJe should also recognize that much of the behavior that creates a hostile work environment is conduct that has the purpose or effect of undermining the perceived or actual competence of women (and some men) who threaten the idealized masculinity of those who do the work."). These examples are discussed by Schwartz, supra note 46, at 1767-68. 322 24.3 COLUMBIA JOURNAL OF GENDER AND LAw segregation and the denigration of female competence for Schultz, male control and the entrenching of male norms for Abrams. While "hierarchization" does an admirable job as a heuristic able to highlight a diverse array of localized gender-based dynamics. gender hierarchy assumes a more extensive conceptual role in patterning how Franke, Schultz, and Abrams understand sex-based power. First, in terms of social structure, a relatively consistent socioeconomic structure can be identified across these accounts that. while not simply controlled by any biological group, nonetheless distributes social prerogative to the benefit of certain men and the detriment of women and nonconforming men. For Franke and Abrams, gendered power takes the explicitly structural forms of "heteropatriarchy"'os and "male control," respectively, whereas for Schultz. it takes the more sociologically-inflected form of the sex segregation of work.' 9 Second, in terms of social practice, gender norms explain the cultural work performed by individual acts of sexual harassment. Again, the nuances vary, with Franke emphasizing the reflexive construction of both harassers and victims around normative gender roles,110 while Schultz stresses the material implications of establishing an idealized masculine personae that can be mobilized to exclude women, youth, and effeminate men from economically and socially prized forms of work."' Finally, it is important to note the explicit causal relationship that cements structure and practice together in these accounts. For, in each case, a circuit of social reproduction is delineated, in which socially entrenched structures come to renew themselves by organizing the everyday lived relationships of individuals in the workplace, thereby reinforcing the normative status of those same structures." 108 See Franke, supra note 28, at 696, 763, 739 n.247 (heteropatriarch)); Franke, Gender, Sex, Agency and Discrimination, supra note 77, at 1251-54 (heteropatriarchy). 109 See Schultz, supra note 28, at 1756-61. Schultz is careful to reject the existence of a monolithic set of male interests. See id. at 1755 n.387 (defining harassment as "a medium through which some men seek to defend their view, of masculine interest and identity against contesting visions proposed by other men"). 110 See Franke, supra note 28, at 693 (defining sexual harassment as a "practice that ... envisions women as feminine, (hetero)sexual objects, and men as masculine, (hetero)sexual subjects"); Schultz, supra note 28, at 1756-74 (defining sexual harassment as a practice "designed to undermine a woman's outward image of competence and sense of self-confidence as a worker" while maintaining an idealized masculine image). 111 See Schultz, supra note 28, at 1774-77. See also Abrams, supra note 28, at 1218-19. 112 See, e.g., Abrams, supra note 28, at 1205 ("Sexual harassment ... functions as a means of establishing male control and expressing or perpetuating masculine norms in the workplace."); Franke, supra note 28, at 763 ("[TJhe wrong of gender-based subordination lies in its power as an overarching regulator) practice that has as its goal the production of feminine women as (hetero)sexual objects and masculine men as (hetero) 24.3 323 COLUMBIA JOURNAL OF GENDER AND LAw From this set of analytical roles emerges a definitional one as well. For the perpetuation of gender norms also acts as a legal test that courts and factfinders can use to identify sexual harassment for purposes of Title VII. For Franke, the reinforcement of the overarching gender order is decisive in defining sexual harassment, distinguishing, for instance, legally- cognizable forms of sexual harassment from other conduct that may also take an explicitly sexual form.' For Schultz as well, gender norms form a critical reference point in deciding whether conduct satisfies Title VII's causality requirement: for female harassment, Schultz instructs courts to inquire into whether the conduct "embodies gender-based expectations," using Price Waterhouse as a touchstone,' 14 whereas, for male victims, Schultz points courts to whether the harassment has "the purpose or effect of denigrating the harassee's manhood." whether through "[a]nti-gay ridicule" or "other types of evidence [that] suggest that the harassees were targeted because something about them threatened the dominant workers' views about the suitable masculine image for those who hold the job."'" Finally. for Abrams. male-female conduct would be legally "based on sex" when it is "connected with the enforcement of a sex and gender hierarchy.""' Same-sex harassment receives more differentiated treatment by Abrams. but in each of its specified forms-as the punishment of gender nonconformists, the practice of horseplay, and the unilateral imposition of sexual sexual subjects."); Schultz, supra note 28, at 1760 ("[HJostile work environment harassment is an endemic feature of the workplace that is both engendered by, and further entrenches, the sex segregation of work."). It is important to distinguish the descriptive use of gender hierarch) from the explanatory use. It is certainly legitimate to use gender as a categor) through which to analyze a particular social formation to reveal the degree of gender-based social stratification. In such a framework, gender hierarch) is an analytical construction resulting from a prior methodological choice. It is quite different, however, to translate this analytic result into a causal explanation for these inequalities, in effect using the existence of gender stratification as evidence to support the existence of a unitary gender structure ("patriarch)," "male control"), which then is understood to produce those very inequalities. See, e.g., Anna Pollert, Gender and Class Revisited; or, the Poverty of 'Patriarchy', 30 SOCIOLOGY 639 (1996) (criticizing feminist models that conflate descriptive and causal uses of gender hierarchy). For a still valuable overview and critique of hierarchical models of gender power, see R.W. CONNELL, GENDER & POWER: SOCIETY, THE PERSON, AND SEXUAL POLITICs 41-61 (1987). 113 See Franke, supra note 28, at 766-68 (excluding both same-sex horseplay and the demands of the "gay harasser" from the legal definition of sexual harassment). 114 Schultz, supra note 28, at 1800. 115 Id. at 1801. Schultz does not address the possibility that anomalous same-sex conduct, such as that in Oncale, could constitute an injury to the plaintiff's experience of his masculinity without reinforcing more general gender norms. This problem also weakens the otherwise excellent arguments in Hilary S. Axam & Deborah Zalesne, Simulated Sodomy and Other Forms of Heterosexual "Horseplay": Same Sex Sexual Harassment, Workplace Gender Hierarchies, and the Myth of the Gender Monolith Before and After Oncale, 11 YALE J.L. & FEMINISM 155 (1999). 116 Abrams, supra note 28, at 1223. 324 24.3 COLUMBIA JOURNAL OF GENDER AND LAw demands-"the entrenchment of male norms" is again central to the legal definition of the claim.'" As the role of hierarchical gender norms expands in these accounts, the corresponding significance of gender performance shrinks. Such a diminished role is surprising, given that gender performance has become a central theoretical tool in the rethinking of gender and sexuality, particularly through the work of Judith Butler."' While the work of Franke, Schultz. and Abrams all make reference to Butler's writing,"' performativity in these accounts is restricted to the largely negative role of countering the determinism and essentialism of earlier feminist accounts by emphasizing how sexual harassment actively molds both harassers and victims to fit normative gender roles."20 For instance, as Franke explains regarding the punishment of gender nonconforming men, "same-sex harassment has a performative effect as well: it authenticates the harassers' status as 'real men' and exiles [the nonconforming victims] from the domain of men.""' Performativity is thus imagined as a one-way process of "embodiment" 2 in which social norms are stamped upon individuals. whether in the mode of valorization ("authenticates") or abjection ("exiles").1> There is never a sense that the "performative effect" of harassment might itself go awry or 117 See id. at 1226-28. 118 See JUDITH BUTLER, GENDER TROUBLE: FEMINISM AND THE SUBVERSION OF IDENTITY (1990); JUDITIH BUTLER, BODIEs THAT MATTER: ON THE DISCURSIVE LIMITS oF "SEX" (1993) [hereinafter BUTLER, BODIEs THAT MATTER]; JUDITH BUTLER, THE PSYCHIC LIFE OF POWER: THEORIES IN SUBJECTION (1997); JUDITH BUTLER, ExCITABLE SPEECH: A PoLrrlCS OF THE PERFORMATIVE (1997); UNDOING GENDER (2004). See also Judith Butler Symposium, 21 COLUM. J. GENDER & L. (2011). For a parallel, practice-based theory of gender, see CONNELL, supra note 112, at 61-90. 119 See, e.g., Franke, supra note 28, at 735 n.220,763 n.380,765 n.390,766,771 nn.411-12; Schultz, supra note 28, at 1686 n.8. See also Abrams, supra note 70, at 357-58. 120 See Franke, supra note 28, at 763 (criticizing subordination theorists for presuming that "male and female subjects come to the sexually harassing workplace fully constituted: he as subordinator/colonizer, she as subordinated/colonized"). 121 Id. at 770-71. See also id. at 771 ("What is important about these technologies of gender is that the) operate 'on us' all the time, constantly reinforcing and creating feminine women and masculine men, thereby normalizing a set of gender roles."). 122 See, e.g., id. at 693 (describing sexual harassment as "the specialized instantiation of a sexist ideology" in which "the act embodies fundamental gender stereotypes"); id. at 747 ("Sexual conduct in the workplace has a special sting for women, not because our sensibilities render us particularly vulnerable to sex, but because the conduct literally sexualizes us. It embodies stereotypic gender norms that become true by virtue of their enactment."). 123 See, e.g., id. at 696, 763, 764, 770-71 (describing how the reflexive constitution of gender roles reproduces already existing gender stereotypes). 24.3 325 COLUMBIA JOURNAL OF GENDER AND LAw even fail-producing illegible or anomalous forms of sex such as the offshore activities witnessed in Oncale-or that these acts could effect a reciprocal transformation in the norms themselves. Rather, in this vision of power, the system of gender norms-hierarchical and oppressive-deploys sexual harassment as a tool to insinuate and reproduce itself in the everyday relationships of men and women. Although the next section will consider in detail the practical consequences of these features, a brief overview might be useful. On an analytical level, these accounts tend to ascribe gender norms with an autonomy and efficacy that they might not otherwise possess. In terms of autonomy. gender hierarchy emerges as the exclusive horizon in which the operation of gender, sexuality, and power are imagined to unfold. Power relations are seen to emerge out of overarching gender hierarchies (for instance, job segregation, male power. and gender stereotypes), which then replicate themselves by reasserting those same hierarchical relations within the workplace (for instance, by sexually objectifying women, denying them the same level of training as men, and harassing gender nonconformists). Although each of these accounts is intellectually sophisticated enough to incorporate a broader historical, social, and disciplinary context in its discussion, these inquiries are inevitably subsumed within the logic of gender normalization. Never is there a sense that these wider historical or social currents might introduce contradictory tendencies or tensions into the analysis. Furthermore, in terms of efficacy, Franke, Schultz, and Abrams attribute a degree of success to the normalizing work of sexual harassment that those norms may not necessarily possess. The cultural work of sexual harassment consequently takes on the circular role of "reinforcing,"' "entrenching," and "policing" already established gender norms.124 Problems exist on a more practical level as well. Because these regulatory structures are used to define in advance what constitutes gender-based power (both analytically and legally), the only manifestations that are legible are those that reproduce the system of gender norms. As a result. an array of gender-related dynamics are ignored because they cannot be assimilated to the model of gender normalization, including-as I will show in the following section-acts of same-sex "horseplay"' as well as the sexual acts of the "gay harasser." More basic still, even identifying individual practices with those categories presumes a highly interpretative act that leads Franke. for instance, to categorize conduct 124 See, e.g., Franke, supra note 28, at 696 (arguing that sexual harassment "perpetuates, enforces, and polices" gender norms.); id. at 728 (arguing that sexual harassment "replicates and perpetuates" sexual hierarchy.); id. at 771 (arguing that relevant case law illustrates how sexual harassment "as regulatory practice inscribes, enforces, and polices" a particular view of who men and women should be.); Schultz, supra note 28, at 1760 ("[HJostile work environment harassment is an endemic feature of the workplace that is both engendered by, and further entrenches, the sex segregation of work."). 326 24.3 COLUMBIA JOURNAL OF GENDER AND LAw based on whether it expresses sexual desire or polices gender norms. Furthermore, the key factor that distinguishes gender normalization from mere horseplay in these accounts is not the conduct itself, but whether some characteristic of the victim suggests he or she was punished for a failure to conform to sex stereotypes. Yet, often times, such characteristics are merely a convenient pretext for harassment, at most marking the victim as vulnerable to harassment rather than explaining the often extreme nature of the harassment itself. C. Practical Limitations: Same-Sex Sexual Harassment The theoretical issues we havejust traced-in particular, the tension between hierarchy and performativity that runs through the accounts of Franke, Schultz. and Abrams-also shape in critical ways how these scholars understand same-sex sexual harassment. The results are important, both because they determine where the boundaries of Title VII protection are set and because they allow us to better appreciate the practical stakes of these theoretical debates. As Franke, in particular, has underscored, same-sex sexual harassment cases offer a crucial perspective on all forms of harassment because they suspend the assumptions we normally bring to issues of gender, sexuality, and power."' While these cases may remain statistically marginal as instances of sexual harassment, they are conceptually important in pushing us towards a more coherent and comprehensive account of sexualized and gendered power. Perhaps appropriately, then, Franke provides the most explicit categorization of same- sex sexual harassment cases into three types: (1) the gay quid pro quo or hostile environment cases where the harasser is shown to be gay and his actual sexual desire for the plaintiff is not challenged; (2) nongay hostile environment cases where a man harasses another man in the workplace, though not because he wants to have sex with or desires his victim; and (3) nongay hostile environment cases where a man in the workplace is targeted for harassment of a sexual nature because he fails to conform to hetero-masculine norms.126 Within this categorization, the nonconforming man-the third and culminating type of same-sex harassment-occupies an unquestioned place of honor. For, in the accounts of Schultz and Abrams as well as Franke, the nonconforming man is paradigmatic of the 125 See Franke, supra note 28, at 694-95. 126 Id. at 766. 24.3 327 COLUMBIA JOURNAL OF GENDER AND LAw underlying harm of sexual harassment: the affirmation of gender norms. For Franke, the punishment of the nonconforming man presents the only true instance of sexual harassment. for it alone enacts the punishment of gender deviance and the reflexive authentication of the harasser's heteromasculine status.'2 Similarly, for Abrams, the harassment of the nonconforming man presents the "clearest case for actionability" under Title VII because "[s]anctioning men who do not manifest prototypical, (hetero-)sexualized masculinity is an important way of entrenching masculine norms in the workplace."' 28 For Schultz as well. harassment of the nonconforming man is exemplary of gender harassment in general in that it seeks to reinforce an idealized image of male competency. As Schultz explains, just as men reinforce gender-based job segregation "by perpetuating the belief that only those who possess certain idealized masculine qualities are competent to perform traditionally segregated jobs," so. too. do men sustain that idealized image "by engaging in harassment that drives away men who fail to conform to the desired image of masculinity or that incorporates them as weak and inferior workers."' In all three accounts, then, the nonconforming man stands at the conceptual center of same-sex sexual harassment. More interesting is what happens as these accounts encounter instances of same-sex sexual harassment that do not fit the paradigm of the nonconforming man. In Franke's typology, there are two such instances, which I want to discuss in turn. First, there are the "gay quid pro quo or hostile environment cases where the harasser is shown to be gay and his actual sexual desire for the plaintiff is not challenged.""no These cases center on the alleged acts of the "gay harasser." Second. there are what Franke somewhat obliquely (and negatively) describes as "nongay hostile environment cases where a man harasses another man in the workplace, though not because he wants to have sex with or desires his victim."'13 Here Franke is referring to cases involving what is typically described as "horseplay." In both sets of cases, the question posed is whether the conduct constitutes sexual harassment if it does not. on its face, appear to reinforce gender norms. In each case, the analyses of Franke and Abrams reveal key assumptions concerning how they understand sexual harassment and the enactment of sex-based power."' 127 Id. at 770-71. 128 Abrams, supra note 28, at 1226. 129 Schultz, supra note 28, at 1776. 130 Franke, supra note 28, at 766. 131 Id. 132 I have omitted reference to Schultz because her discussion of same-sex sexual harassment revolves exclusively around the punishment of the nonconforming man as a means of reinforcing the link between work 328 24.3 COLUMBIA JOURNAL OF GENDER AND LAw 1. The "Gay Harasser" First there is the "gay harasser."' For Franke, the gay harasser does not engage in sexual harassment at all. It is a mistake, she explains, to believe that either the abuse of workplace authority for private gratification or the sexual objectification of others is sufficient to constitute sexual harassment.' However unethical such acts may be, it is the perpetuation of gender norms, and that alone, that defines conduct as sexual harassment. Same-sex desire obviously departs from such heteromasculine norms. As Franke explains: "When a gay male supervisor requests sexual favors of a male subordinate no larger cultural gender orthodoxy is being policed, perpetuated or enforced. Rather, the supervisor is exploiting a position of power in order to satisfy his carnal desires."'" The gay harasser thus functions as a conceptual foil to the nonconforming man. Whereas the abusive acts of the gay harasser are severed from the systemic relays of power and are therefore unable to do the cultural work of normalization, the punishment of the nonconforming man plays a critical regulatory function in erecting heteromasculine norms, norms from which the gay harasser is already, by definition, exiled. By contrast. Abrams finds that the actions of the gay harasser do, at least potentially, count as sexual harassment. Abrams agrees with Franke that such cases do not serve to discipline gender nonconformists, but departs from Franke in embracing the possibility that other gender dynamics-whether based in sexual desire or not-might render such conduct actionable because they reinforce male norms.m "The critical question," explains Abrams, "is whether the perpetrator acts on his desire in a way that was unilateral and without reference to the desires of the target."13 6 Provided that conduct "affirms traditional norms regarding male sexual subjectivity" due to its unilateral quality, Abrams would categorize those acts as sexual harassment even if their perpetrator is gay or otherwise fails to fit the heteromasculine mold."m competency and an idealized image of masculinity. Schultz, supra note 28, at 1774-89. Unlike Franke and Abrams, Schultz does not address cases involving either "horseplay" or the "gay harasser." Id. Presumably, Schultz would find such cases to fall outside the protections of Title VII. 133 Franke, supra note 28, at 766-67. 134 Id. at 767. Franke notes that victims of the "gay harasser' can claim disparate treatment under Title VII because the harassment would not have occurred "but for" the gender of the victim (due to the harasser's sexual orientation). Id. at 767 n.398. 135 Abrams, supra note 28, at 1227-28. 136 Id. at 1228. 137 Id. 24.3 329 COLUMBIA JOURNAL OF GENDER AND LAw Both approaches dramatize the truncated role that gender performance plays in the articulation of regulatory norms. For both scholars, the conduct of the gay harasser is approached in the same analytical way: by inquiring whether specific acts work to reinforce normative gender roles. Abrams answers positively because she adopts a more elastic definition of male control in terms of unilateral desire. Franke answers negatively because, to her mind, the sexual desires of a gay man cannot serve to reaffirm the boundaries of heteronormative masculinity. Yet, in each case, the regulatory norm is imagined as preceding and framing the acts of particular individuals, set forth as a test to determine what conduct counts as an enactment of sex-based power. For neither Franke nor Abrams is there the possibility that individual practices might take up and transform the structure of gender norms. Nor is there any sense that gender norms might themselves represent the sedimented effects of such practices-indeed, that there might be no gender norms independent of the signifying effects of reiterated gender practices. Although Franke. for instance, stresses the performative and reflexive work of sexual harassment in the case of the nonconforming man.'" any similar dimension of performativity disappears when she turns to the gay harasser. From her perspective, performativity-at least, to be real and take hold-must emerge out of an already existing power structure. The gay harasser is locked out of such structures. exercising a coercive sexual power that leaves unchanged the broader arrangement of gender and sexuality. But if we understand gender performance as actively constructing the roles of "gay harasser" and the "nonconforming man," then it makes no sense to turn to these same categories to determine which sexualized acts participate in the operation of power and which do not. In truth, each of Franke's categories performatively fuses sexuality and power, and only a retroactive ideological operation installs the untenable dichotomy between (nonregulatory) "private desire" and (nonsexual) "social discipline" as an apparently objective fact. Every case and fact-pattern is already framed by a series of overlapping performative and interpretative acts. On the part of the harasser and victim, there is the sexualized workplace encounter itself. a scene in which, as Franke notes, each participant attempts to impose his or her own construction of sex upon the other individuals involved.' For subsequent factfinders and scholars asked to evaluate the dispute, the categories of 138 Franke, supra note 28, at 770-71. 139 Sexual harassment thus must be understood, not as the unilateral imposition of masculine authority or gender-related norms, but as itself a collectivized scene of gender performance, in which the participants actively seek to solicit and rearticulate the sexual significance of that enactment. For a suggestive analysis of rape that seeks to resist the narrative of passive victimization, see Sharon Marcus, Fighting Bodies, Fighting Words: A Theory and Politics of Rape Prevention, in FEMINISTS THEORIZE THE POLITICAL 385 (Judith Butler & Joan W. Scott eds., 1992). 330 24.3 COLUMBIA JOURNAL OF GENDER AND LAW sexual identity. desire, discipline, and conformity to stereotypes are also not inert facts. but the effects of further interpretative framings capable of either reifying or reworking these constructions. For instance, in AcWillians, the factfinder was presented with evidence that, on several occasions, the defendant Douglas Witsman followed the plaintiff Mark McWilliams into the restroom "where Witsman, with one hand in his unzipped fly, would put his arm around McWilliams or invite him into a stall."140 How is such conduct to be categorized without already taking sides in the performative contestation this conduct implies? If. for instance. we understand Witsman as a "gay harasser"-unafraid of satisfying his carnal desires though coercion and harassment, but himself a sexual minority-we have already inserted Witsman and McWilliams into a deeply ideological and identitarian framework. If, on the other hand, we focus on McWilliams and the ways that he deviates from masculine norms, does not that interpretation side with his harasser, cloaking Witsman's perversity within the guise of normality in order to collude with his positioning of McWilliams as abnormal and deviant?141 And beyond the identities there are the acts themselves: if we declare Witsman's conduct to be "sexual." why must it be severed from the enactment of social power unless it is willing to be forced back into the closet? If we take the performativity of gender seriously, then there are no unequivocal facts but rather a set of embedded performances that cut across and rework categories such as "gay harasser" and "nonconforming man." Particularly if a notion of gender norms proves too unsubtle to track these interchanges, we must be careful not to reintroduce a normative logic in precisely those sites witnessing its incipient unraveling. Abrams's approach encounters a similar problem by failing to recognize the performativity of these acts. In her discussion of the "gay harasser," Abrams explicitly rejects Franke's "either/or" of desire and power, declaring that the presence of desire does not, by itself. either mandate or foreclose a finding of sexual harassment.14 ' She shifts away from desire to ask whether the conduct reinforces "male sexual subjectivity" through the unilateral imposition of the harasser's desires "without reference to the desires of the target." 14 3 Male sexual subjectivity, notes Abrams, may be entrenched as a workplace norm through unilateral sexual conduct even when its participants do not fit the model of 140 McWilliams v. Fairfax Cnty. Bd. of Supervisors, 72 F.3d 1191, 1199 (4th Cir. 1996). 141 Franke and Schultz, for instance, interpret McWilliams's harassment as punishment for his gender nonconformity. See Franke, supra note 28, at 739-40; Schultz, supra note 28, at 1781-82. 142 Abrams, supra note 28, at 1227-28. 143 Id. at 1228. 24.3 331 COLUMBIA JOURNAL OF GENDER AND LAw a heterosexual man harassing a woman.144 Even a gay man can do it. So long as Abrams imagines norms as external structures that precede individual acts, it should not matter who precisely enacts those structures; it is only a question of whether those acts affirm the underlying norms. Yet having taken the opposite course from Franke in order to analytically separate norms from the sexual identities of the individuals who enact them, Abrams is also rigorous enough not to ignore the questions that result. For instance, are we to assume that the unilateral sexual conduct of a lesbian towards her female coworker simply affirms a generic "male sexual subjectivity"? Do the performer's own gender and sexuality not, in some fashion, necessarily alter the terms and effects of such performance? In the closing pages of her article, Abrams seems to recognize that it might, in fact, matter a great deal whether "male sexual subjectivity" is performed by a person who is male or female, straight or gay, and further that the contours of that performance might actively transform and inflect the norms themselves. Abrams symptomatically circles these issues through a series of "difficult questions" that remain unanswered at the conclusion of her article. 145 First, Abrams notes that, in her current approach, finding the acts of a "gay harasser" actionable as sexual harassment would imply a "direct parallel" between coercive heterosexual and homosexual dynamics. 146 But Abrams is troubled by that suggestion. She notes that theorists are divided on whether such parallels exist: Whereas MacKinnon interprets lesbian sadomasochism as an internalization of oppressive heterosexual dynamics, Joan Nestle contends that butch/ femme relationships are irreducible to the heterosexual model.147 Leaving that question open, Abrams next turns to an equally "difficult question"' concerning the dynamics of male-male sexual harassment.148 One of the central wrongs in male-female harassment is the disregard of female agency in general, notes Abrams. But if such a disregard does not occur in the context of male same-sex harassment-for men as a group continue to 144 Id. 145 Abrams concludes her essay by declaring that these issues must be resolved through "a fact-specific inquiry." Id. at 1228-29. But, as with Franke's categorical divisions between the "gay harasser" and the "nonconforming man," facts cannot resolve such theoretical issues because those facts are themselves already framed by a specific conceptual lens that renders them intelligible. To inquire, for instance, whether certain conduct has "entrenched in the workplace stereotypic notions of male sexual subjectivity," id. at 1229, is merely to beg the question whether there is a unitary "male sexual subjectivity" to be "entrenched," and whether the conduct in question might perform an) other cultural work beyond that of "entrench[ing]" pre-existing norms. 146 Id. at 1228. 147 Id. at 1228 n.309. 148 Abrams, supra note 28, at 1229 n.311 332 24.3 COLUMBIA JOURNAL OF GENDER AND LAW be socially recognized as sexual agents-then where does the wrong lie?14' Her answer is that either respect for another individual's agency is a "normative good" regardless of gender, or else male-male disregard of agency is secondarily "derived" from male- female harassment and therefore indirectly perpetuates the harm to women."' Yet, under either theory, the specificity of male-male sexual harassment is lost. Again, leaving that question open, in a final footnote, Abrams encounters another "potentially complicated question": whether unilateral sexual conduct between women could entrench norms of male sexual sub jectivity."' Abrams admits this question is "tricky" because, on the one hand, a unilateral sexual approach might be "interpreted differently" when undertaken by a woman, but, on the other hand, some sexual conduct by a lesbian might be "strikingly reminiscent of the behavior manifested by heterosexual males in some sexual harassment cases.""5 This, too, is a question that Abrams raises but does not resolve. All three of these concluding questions are "difficult" for Abrams because each suggests that "male sexual subjectivity" does not exist outside of the individual practices that reiterate and reproduce it. Rather. these sexualized practices-whether performed by a gay or lesbian supervisor, a butch/femme couple, or a man sexually abusing another man-actively take up, inhabit, and rearticulate the disembodied social norm associated with male sexual subjectivity. 2. "Horseplay" The second category that traces the limits of same-sex sexual harassment is that of "horseplay." As we have seen, for Franke and Abrams, horseplay euphemistically describes a category of harassment in which an aggressive heterosexual manhood is paraded through the workplace as crude phallic rituals. The result, as Franke notes, is a "fraternity-type culture" in which "a male employee teases other male employees with plastic penises, uses language like 'suck my dick,' or where talk of penis size, sexual conquests, or the touching of crotches" are a daily occurrence.' While such conduct can result in sexualized 149 Id. 150 Id. 151 Id. at 1229 n.312. 152 Id. Abrams's repeated use of the language of resemblance is symptomatic of her search for a more performative model of sexual harassment. For instance, Abrams repeatedly finds conduct to "parallel," "resemble," or "derive" from male norms because she recognizes that this conduct takes up-perhaps parodically, perhaps sincerely-already existing social norms and deploys them for purposes of its own. The next step would be to inquire how the cultural appropriation of these norms also entails their active resignification in ways that cannot be determined or controlled in advance. 153 Franke, supra note 28, at 768. 24.3 333 COLUMBIA JOURNAL OF GENDER AND LAw abuse. it does not appear to fit into either the category of the gay harasser or that of the nonconforming man. While Franke and Abrams appear to agree on the description of horseplay, they diverge in their responses. For Franke, horseplay superficially resembles sexual harassment because it involves "unwelcome or obnoxious sexual conduct." 15 4 Yet, as Franke persuasively argues elsewhere,'"' sexual content alone-even if "obnoxious" or "inappropriate"-does not provide adequate grounds for a finding of sexual harassment because that conduct (without more) "does not play a role in the regulation and enforcement of hetero-patriarchal gender norms.""" Nonetheless. Franke does explain how horseplay might, in certain instances, come to assume a regulatory function. In its original form, horseplay represents a narcissistic and autoerotic theater of manhood, not truly abusive or demeaning but merely populated with "plastic penises" and "the touching of crotches." If anyone becomes offended or harassed in such an environment, it is, at least initially, only by accident. But if an "enlightened man" objects to the phallic horseplay of his coworkers and subsequently finds himself harassed as a result, he has assumed the mantle of the nonconforming man.1 17 In that instance, Franke explains, horseplay has ripened into full-blown sexual harassment and presents a cognizable Title VII claim."' Abrams, by contrast, presents a more straightforward approach to horseplay. Because Abrams is willing to categorize a wide variety of dynamics as sexual harassment, she does not regard horseplay as merely "obnoxious sexual conduct" but. more specifically, as an expression of masculine norms. As Abrams explains, even when such expressions do not immediately locate a victim, they indirectly work to "express conceptions of normative masculinity" and consequently secure the "entrenchment of masculine norms,""' especially 154 Id. at 769. 155 See id. at 730-47. 156 Id. at 769. 157 Id. at 768-69. Franke calls this the "double bagging" rule, referring to the common practice of "bagging," in which male workers grab, or feign grabbing, one another's testicles. Id. at 769 n.406. The first time the "enlightened man" is bagged, he might be offended by such sexualized workplace conduct, but he is not being disciplined for his deviance from heteromasculine norms. Yet, after he complains to his coworkers about this conduct, the second time he is bagged, he is being punished for his refusal to accede to the sexual orthodoxy of the workplace. Thus the first bagging is an accident; the second bagging is sexual harassment. 158 Id. at 769. 159 Abrams, supra note 28, at 1226. 334 24.3 COLUMBIA JOURNAL OF GENDER AND LAw in workplaces in which those norms have recently been challenged.1"o As long as horseplay entrenches male norms, Abrams's expanded definition of sexual harassment can easily accommodate such heteronormative antics as indirectly contributing to the subordination of women and nonconforming men. The only question that remains for Abrams is whether such conduct invariably leads to an effect that the law is prepared to recognize as an employment-related disadvantage."' Rather than imposing a "double bagging"' rule around the complaints of the "enlightened man,"' Abrams more simply returns to the traditional pervasiveness requirement in order to ensure that the underlying conduct is sufficiently severe for a Title VII claim. So long as male horseplay can be shown to have "unreasonably . . . interfere[d] with [the] individual's work performance," Abrams has no difficulty including it within her account of sexual harassment."' Despite these differences in approach, the accounts of Franke and Abrams are equally hampered by their initial understanding of what horseplay represents. Both Franke and Abrams treat horseplay as an objective category for a certain kind of conduct and therefore never analyze how the category itself has been ideologically deployed both by courts and within U.S. culture more broadly. As with Franke's other categories, this is not principally a question of factual accuracy, for horseplay, as that term has been culturally mobilized, is not primarily a positive description at all. It is, rather, a category for the uncategorizable-a category that does not describe any specific set of facts, but which collects those cases that refuse to fit elsewhere. Horseplay does, of course, contain one positive feature: sexualized conduct that demeans or humiliates a person of the same sex.' But horseplay is otherwise characterized by its absences-specifically, the absence of both erotic and regulatory motivesl6 4-rendering it a negative designation for otherwise inexplicable male behavior.' Branding conduct as "horseplay" repackages and considerably narrows its significance by casting it as a kind of crude masculine joking or set of "locker-room antics." Such branding does three related things. First, it marginalizes the sexual content as accidental, unintended, and inessential. Second, it resituates the aggressivity as a form of male bonding ("play," 160 Id. at 1210-11. 161 Id. at 1226. 162 Id. (quoting Meritor Say. Bank v. Vinson, 477 U.S. 57, 65 (1986)). 163 See Franke, supra note 28, at 697-98. 164 See id. at 697-98. Of course, Franke and Abrams both see horseplay as expressing masculine norms but not in a way that directly regulates others. 165 See, e.g., id. at 766 (describing horseplay as "nongay hostile environment cases where a man harasses another man in the workplace, though not because he wants to have sex with or desires his victim"). 24.3 335 COLUMBIA JOURNAL OF GENDER AND LAW "antics") rather than targeted abuse and degradation. Third, it minimizes the severity of the conduct by framing it as "good-natured joking" accidentally "gotten out of hand." Abrams and Franke continue this trend by regarding this conduct as a kind of juvenile, if obnoxious. display of heteromasculine norms-as Franke explains, "inappropriate and offensive," but not in any fundamental way sexual, degrading, or severe.'' The inadequacies of that framing are evident. however, if we look at the facts of these cases. Oncale, of course, was a case routinely described as "horseplay"'7 and, indeed. Oncales treatment is difficult to otherwise place among Franke's categories. Certainly Lyons and Pippen fail to fit the mold of the "gay harasser," while Oncale appears too ordinary to play the "nonconforming man."' Yet the harassment of Oncale went far beyond "inappropriate" or "obnoxious" sexual conduct to include instances of simulated fellatio, repeated threats of rape, and at least one creditable rape attempt in the oil-platform shower. Even in the cases mentioned by Franke as examples of horseplay, a similar repression is evident. For instance, in Hart v. National Mortgage & Land Conpany-the case that Franke dismisses as involving "teas[ing]" with "plastic penises" -- John Hart complained of more serious forms of sexualized humiliation, including instances when his male supervisor "would grab Hart's genitals, grab Hart around the waist and try to mount him and make sexually suggestive gestures, accompanied by crude remarks."1 no While the court's euphemistic references to "crude remarks" and "sexually suggestive gestures" obscure the actual facts of Hart's sexual degradation, it is clear that Hart is not Franke's "enlightened man" morally offended by his supervisor's crudeness. 166 While same-sex sexualized abuse in the workplace is often labeled "horseplay," same-sex sexualized abuse in schools and other institutions (as well as certain isolated workplaces) is frequently described as "hazing." Yet by viewing this conduct as a form of initiation, the objectively severe sexual nature of the degradation can similarly be minimized. For examples of such abuse, see, e.g., Patrick Healy, L.I. District Is Criticized In Hazing Case, N.Y. TIMES, Sept. 23, 2003, at BI (describing allegations that varsity football players sexually assaulted younger teammates by inserting broomsticks, pine cones, and golf balls into the anuses of younger boys); Cioffi v. Averill Park Cent. Sch. Dist.,444 F.3d 158, 161 (2d. Cir. 2006) (describing allegations that a fourteen-year-old high school football player had been "tea-bagged" by other football players). 167 See Brief for Respondents, supra note 1, at *33 (describing Oncale's abuse as "sexually charged hazing in the workplace"); Franke, supra note 28, at 697 n.17 (listing Oncale among horseplay cases). 168 Franke did not reconsider the meaning of horseplay even after encountering the complete facts of Oncale's case. See Franke & Hunter Brief, supra note 46. 169 Franke, supra note 28, at 768. 170 189 Cal. App. 3d 1420, 1424 (1987). 336 24.3 COLUMBIA JOURNAL OF GENDER AND LAW Certainly, part of this repression stems from the difficulty that courts and scholars experience in understanding the sexuality present in these cases. Both Franke and Abrams, for example, purge horseplay of any trace of "real"' sexual motivations, preferring to keep such "authentic" sexual conduct firmly moored to sexual orientation and the gay harasser. Franke's model of the accidentally offended observer exemplifies that repression: the enlightened man of Franke's narrative is not originally the target of abuse, but only a third party offended by the mere display of sexuality, whether as a topic for conversation ("talk of penis size") or as a source of representation ("plastic penises"').1 1 Yet targeted sexual acts recur throughout cases of horseplay.172 In fact, in order to understand these cases properly, it is vital to dispense with the assumption that sexual orientation determines whether conduct is sexual. Horseplay proves that "straight" men can, in certain circumstances, adopt explicitly sexualized forms of abuse with the clear knowledge that this sexual dimension will render the abuse more damaging to its victims. Perhaps, in Oncale, Lyons merely rubbed his penis against Oncale's head rather than placing it in Oncale's mouth, and perhaps Lyons pushed a bar of soap-rather than a part of his own anatomy-into Oncale's anus" while telling Oncale that he was "fixing to fuck" him.1 74 Nonetheless, the absence of those motives undoubtedly did not lessen the sexualized humiliation that Oncale experienced. While such sexual abuse does not easily fit within our understanding of sexual identity, it does not mean that this conduct is not a form of sex-based power. Moreover, the homoeroticism of Oncale's harassment was not an accidental accompaniment to sporadic events on the Sundowner oil-platform, as if Oncale's abuse was an aberrant moment when 171 Franke, supra note 28, at 768-69. 172 See, e.g., James v. Platt River Steel Co., 113 F.App'x 864, 865 (10th Cir. 2004) (unpublished) (harasser jumped on plaintiff's back, stuck his tongue in plaintiff's ear, grabbed plaintiff's "crotch and private parts," and referred to plaintiff as his "bitch"); Rene v. MGM Grand Hotel, Inc., 305 F.3d 1061, 1064 (9th Cir. 2002) (en banc) (harassers caressed and hugged plaintiff and repeatedly "grabbed him in the crotch and poked their fingers in his anus through his clothing"); Leake v. Ryan's Family Steakhouse, 5 Fed. App'x 228, 230 (4th Cir. 2001) (per curiam) (unpublished) (harasser made "sexual remarks" about plaintiff, touched plaintiff's buttocks and genitals, and pushed plaintiff back into a bathroom as plaintiff was leaving, where harasser groped plaintiff and said, "Come on, let's do this here."); Shepherd v. Slater Steels Corp., 168 F.3d 998, 1001 (7th Cir. 1999) (harasser "play[ed] with himself" in front of plaintiff, "wagged" his exposed penis in front of plaintiff's face, and said he was "liable to crawl lip on top of [plaintiff] and fuck [plaintiff] in the ass"); Fry v. Holmes Freight Lines, Inc., 72 F. Supp. 2d. 1074, 1076 (W.D. Missouri 1999) (harassers kissed plaintiffs neck, "pounded" plaintiff 's buttocks to simulate intercourse, and asked plaintiff, "Whose dick have you been sucking?" "Do you want to take it in the butt?' and "Do you want to suck my dick'?"); Breitenfeldt v. Long Prairie Packing Co., 48 F. Supp. 2d 1170, 1172 (D. Minn. 1990) (harassers mimicked performing oral and anal acts on plaintiff and asked whether plaintiff preferred "it up the ass or down the throat."). 173 83 F.3d 118 (5th Cir. 1996). 174 MacKinnon Brief, supra note 4, at 13. 24.3 337 COLUMBIA JOURNAL OF GENDER AND LAW normally good-natured rough-housing "got out of hand." As the epigraph to this Article suggests, such behavior was simply "how it was offshore."" As in other all-male work environments, aboard Chevron's drilling platforms homoerotic abuse appeared as the abiding medium for the enactment of male power. A final element in the repression of horseplay concerns the counter-normative quality of that conduct. As we have seen, at the core of the leading theoretical accounts is a vision of a regulatory gender system performatively constructing both men and women in accordance with heteromasculine roles. Yet what is most striking about horseplay is that it directly contradicts the heterosexual norms that Abrams. in particular, understands that conduct to celebrate. I7Yet it is difficult to understand instances of simulated fellatio and threats of anal rape as consolidations of masculine stereotypes. Although Oncale's abuse. for instance. is clearly an imposition and enactment of power, the medium of this enactment-highly charged sexualized humiliation and abuse-defies the social norms of straight masculinity. There is power, but it is not a power of the normative: it does not reinforce the stereotypes of masculinity, at least as these rely upon a strict prohibition upon sexualized acts between men. Indeed, as the very term "offshore" suggests, such mobilizations of same-sex erotic abuse are deliberate deployments of a sexualized power energized by its counter-normative sexual content. When Lyons promises Oncale, "Fm going to fuck you anyway," 1 or when McWilliams is bound and blindfolded as Witsman forces his fingers into McWilliams's mouth to simulate fellatio,7 the anti-normative sexual quality of that power is not merely contingent. but one of many convenient ways of establishing a workplace hierarchy. While the desires of Lyons and Witsman might be difficult to understand from the perspective of sexual identity. their power is indissociable from a willingness, in Lyons's words, "to fuck you anyway"-to enact a brutal demonstration of power in spite of social norms and masculine conventions, in defiance of taboos on same-sex intimacy, and in direct violation of the structures of heteronormativity. As should be clear, the failure to recognize the severity, sexuality, and perversity of horseplay has broader theoretical and practical consequences as well. Theoretically, inscribing horseplay within a theory of gender normalization serves to hide those contradictions, cloaking Oncale's harassers in the guise of normality: not only are they 175 Brief for Respondents, supra note 1, at *3. 176 But see Abrams, supra note 28, at 1203 n.188 (questioning the extent with which masculine norms should be identified with compulsory heterosexuality). 177 MacKinmon Brief, supra note 4, at 14. 178 McWilliams v. Fairfax Cnty. Bd. of Supervisors,72 F.3d 1191, 1193 (4th Cir. 1996). 338 24.3 COLUMBIA JOURNAL OF GENDER AND LAw taken as the enforcers of social norms, but they are, by definition, normal themselves. Practically, this perspective encourages courts and legal scholars to ignore the perpetrators of abuse and, instead, to probe the victims of sexual harassment. The sex stereotyping claim, for instance, is essentially premised upon such an inquiry. Was it the victim's effeminacy. prudishness, or sexual inexperience that solicited the abuse? Yet seeing the victim's gender performance as the cause of harassment can easily slide into a pathologization of that behavior, becoming still another way to blame the victim for his or her abuse. Furthermore. the focus upon the victim's gender deviancy can place upon plaintiffs the additional evidentiary burden of proving that they visibly depart from gender stereotypes, leaving men such as Joseph Oncale without a viable claim.1 7 In the following section, we will put pressure on the normativity of the men who commit sexual harassment by offering a more historical account of white working-class masculinity in the postwar era. In place of the abstract socioeconomic structure of "heteropatriarchy" or "male control." we must position sexual harassment within a political economy of white working-class masculinity: the network of economic relations, social institutions, and political investments that have permitted certain men to assert their "normalness" against others in the workplace. Having established the historical conditions of that normality. we can then probe in what ways the changing conditions of that political economy since the 1960s may have rendered that normativity precarious or impossible today. As I hope to make clear, it is not a question of whether changed historical conditions have "caused" certain kinds of workplace harassment, for sexualized abuse between men existed long before the postwar era. Rather, it is a question of the altered social and historical meaning that these acts necessarily assume by dint of today's vastly different social and cultural landscape. Acts of same-sex humiliation might have occurred in various early twentieth- century settings, but when these same acts are taken up and repeated in the workplace today, their significance is irrevocably altered. Indeed, in this war of signification, the ability of Title VII to brand same-sex acts as "sexual harassment" rather than "horseplay" or "hazing" might prove to be a highly effective instrument of social change. 1II. The Political Economy of White Working-Class Masculinity As the previous section showed, the leading feminist accounts have tended to imagine sexual harassment as perpetrated by the enforcers of an abstract male order who embody 179 See, e.g., Franke & Hunter Brief, supra note 46, at *29-30 (proposing that, even in cases of severe sexualized abuse such as that of Oncale, courts should require specific evidence of sex stereotyping, such as "whether Mr. Oncale was singled out for harassment because of his failure to live up to some standard of masculinity, . . or ... because he objected to the hypermasculine environment"). 24.3 339 COLUMBIA JOURNAL OF GENDER AND LAw the norms of an aggressive and capable heterosexual masculinity. Yet, as I have argued, particularly in cases of horseplay, the perpetrators of harassment take up an explicitly counter-normative mode of sexualized power towards other men. In order to understand the social dynamics behind such counter-normative acts, we must go beyond the horizon of gender hierarchy in order to explore the social contradictions lodged within the contemporary articulation of white working-class masculinity. To do this, we must turn back to the working-class male workplaces described in this Article's opening in order to better understand the shifting cultural dynamics and social investments that have shaped and reshaped these workplaces, as well as these workers, over recent decades. But it is important to note in advance that the intersectional analysis of white working-class man is intended to be exemplary rather than unique: it serves to demonstrate that any appeal to an abstract male order ignores the multiple, overlapping, and contradictory forms of identification that unfold along lines of race, gender, class, and sexuality (to name only the most obvious). Furthermore, this analysis is not designed to explain the historical causes of horseplay or any other form of sexualized abuse, but to apprehend the social meaning that such gendered acts have within our own historical context. With those caveats in mind, this section offers a historical account of how a specific organization of race, class, and gender associated with white working-class men has been reworked in conjunction with the broader economic, political, and social dynamics of the twentieth century. The white male worker took definition at the start of the century within the Fordist and Keynesian socioeconomic order that stretched from the New Deal to the Great Society and which helped to enable the institutions and attitudes characteristic of white working-class life. With the economic crisis of the late 1960s and early 1970s, this provisional order underwent a prolonged crisis that led to a fundamental restructuring of U.S. society, including a reimagination of the component parts of these workers' identities: virility, whiteness, and labor. The white male worker would survive into the 1980s as the political face of backlash. betrayal, and reactionary anger. Yet the white worker was also the unacknowledged offspring of the very post-1960s identity politics he appeared to decry, constructed from the same cultural template as black power, second-wave feminism, and post-Stonewall gay liberation. As a point of social identification, then, the "white worker" now self-consciously bears the scars of economic decline, political disenfranchisement, and social victimization that position him, certainly no longer as a representative of gender norms, but as one of many competing figures for political redress in the post-civil rights era. As a result, even for white working men who remain within familiar working-class and unionized occupations, the broader social meaning of "white working-class masculinity" has inescapably changed due to the general dismantling of white industrial masculinity and its reanimation within a politics of resentment. Indeed, for a broad range of social 340 24.3 COLUMBIA JOURNAL OF GENDER AND LAw actors that are not themselves white. working-class, or even male, a generic male power might continue to circulate as a symbolic resource: a cultural repertory of gender-specific acts able to provide a visceral (if temporary) sense of power when other avenues of social power appear foreclosed.1so A. The Rise and Fall of the Industrial Working Class In order to grasp the growing contradictions in white working-class masculinity, we need to step back to understand how this particular segment of U.S. society could be seen as "normal" at all. For, in the decades after the 1880s as U.S. industrialization began in earnest, the large-scale arrival of Jewish and Catholic immigrants from Eastern and Southern Europe stoked nativist fears of race suicide and led to these immigrants' exile to the outermost margins of U.S. society-forced into irregular employment in exchange for poverty wages, working in abject conditions by day and living in overcrowded one- room tenements by night."' Yet, following the social, political, and economic upheavals of the Great Depression, the New Deal, and World War II, the postwar era presented a time of long-awaited prosperity for this and other working-class groups. Under the emergent Fordist-Keynesian regime, the postwar settlement between labor and capital generated a period in which rising corporate productivity and profits were linked to cyclical increases in worker wages overseen by collective bargaining and the threat of work stoppage."' In contrast to the disastrous economic policies of the 1920s. which kept worker wages at subsistence levels and thus precipitated the crisis of overproduction in the 1930s, the postwar era created a dynamic system that synchronized mass production with worker 180 For the argument that sexual harassment should be understood as the opportunistic mobilization of gender-specific cultural scripts, rather than the unilateral enforcement of gender norms, see infra Part IV.A. But viewing male power as a symbolic social resource open to an ambivalent process of appropriation opens up a range of productive questions, such as the ways that African-American men might appeal to gendered forms of power to compensate for perceived racial subordination or how certain women may take up cultural scripts of masculine power. 181 See JACKSON LEARS, REBIRTH OF A NATION: THE MVIAKING OF MODERN AMERICA, 1877-1920, at 71-91 (2009). See generally LIZABETH COHEN, NAKING A NLw DEAL: INDUSTRIAL WORKERS IN CHICAGO, 1919-1939 (2d ed. 2008). 182 See MIKE DAVIS, PRISONERS OF THE AMERICAN DREAM: POLITICS AND ECONOMY IN THE HISTORY OF THE U.S. WORKING CLASS 190-91 (1986); DAVID HARVEY, THE CONDITION OF POSTMODERNITY: AN ENQUIRY INTO THE ORIGINS OF CULTURAL CH-ANGE 121-40 (1991); JAMES T. PATTERSON, GRAND EXPECTAFIONs: THE UNITED STATEs, 1945- 1974, at 61-81 (1997) [hereinafter PATTERSON, GRAND EXPECTATIONs]. See also Katherine Van Wezel Stone, The Postwar Paradigm in American Labor Law, 90 YALE L.J. 1509 (1981) (discussing the ideological model of collective bargaining in the postwar period). 24.3 341 COLUMBIA JOURNAL OF GENDER AND LAw wages and mass consumption.13 The result was a massive reshaping of the white working class. During this period, perhaps a quarter of the U.S. population achieved middle-class status, including those semi-skilled white ethnic working families that formed the backbone of the industrial workplace.14 Mass consumption of cars. houses. and household appliances was underwritten by national policies supporting home loans, tax relief for mortgages, and the subsidization of the federal highway system, which then rendered possible the exodus into the suburbs during the 1950s and 1960s.185 New possibilities for educational mobility also reshaped the white working-class family through the Servicemen's Readjustment Act of 1944, better known as the G.I. Bill, which allowed almost eight million World War II veterans-half of all those serving in the war-free college education.1 "6 By the 1960s. over half of the high school graduates in more affluent states like California were going to college.1 1As a result of these postwar changes, by the early 1960s the social and economic realities of large sections of the white working-class had been significantly altered by their incorporation in the expanding cycles of Fordism. Whereas, only a generation before, working-class families had been trapped in dead-end, casualized jobs and living in rented tenements, a large proportion now enjoyed high-paying union jobs and homeownership, as well as the promise of a college education and even greater upward mobility for their children. Unfortunately, much of that advance stalled or was undone following the crisis of the late 1960s and early 1970s. The first indicators were economic. As early as 1966, U.S. corporate productivity and profitability had begun to decline and would continue to do so for the next fifteen years.'" Domestic manufacturing, in particular, began to face steep challenges from Japan and Germany as these countries completed their postwar economic recoveries and sought to expand their export markets into the United States." Oil embargoes imposed in 1973 in retaliation for support of Israel during the Yom Kippur 183 See DAVis, supra note 182, at 190-91; HARVEY, supra note 182, at 121-30; PATERSON, GRAND EXPECTAFIONs, supra note 182, at 61-81. 184 See DAVIs, supra note 182, at 191; PATTERSON, GRAND EXPECTATIONS, supra note 182, at 64. 185 See DAVIs, supra note 182, at 190; PATTERSON, GRAND EXPECTATIONS, supra note 182, at 64-81. 186 See DAVIs, supra note 182, at 191; PATTERSON, GRAND EXPECTATIONS, supra note 182, at 367. 187 See DAVIs, supra note 182, at 192. 188 See HARVEY , supra note 182, at 141; Robert Brenner, Structure vs. Conjuncture: The 2006 Elections and the Rightward Shift, 43 Nuw LEFr RIv. 33, 41-42 (2007). 189 See HARVEY, supra note 182, at 141; Beth Bailey & David Farber, Introduction, in AMERICA IN THE SEVENTIES 1, 3 (Beth Bailey & David Farber eds., 2004). 342 24.3 COLUMBIA JOURNAL OF GENDER AND LAw War caused U.S. oil prices to spike 387 percent.1 ' The following year, the short-term oil shortage was institutionalized as a long-term energy crisis as OPEC raised the price of oil among its member nations. causing the price of overseas oil to jump from $1.77 a barrel in October 1973 to $10 a barrel in early 1974.' In the mid-1970s, as the keenest pangs of the oil crisis receded and the sharp recession of 1973 passed, the country settled into the economic malaise of "stagflation." coupling the stagnant output of goods with the unprecedented growth of unemployment and inflation. Taken together, these changes signaled the growing inability of the Fordist-Keynesian system to remain master of the contradictory tendencies of the emergent global economy.' The corporate capitalism of postwar Fordism required large-scale fixed capital investments in the mass production of relatively uniform consumer goods-automobiles, televisions, refrigerators, washers and dryers-that presumed the insatiable growth of consumer demand. But after suburbanization slackened and the domestic durable market became saturated,"' international competition eroded the remaining domestic market for American companies. By 1971 the United States imported more goods than it exported for the first time in the twentieth century.194 For white working-class men, the effects of this crisis were particularly acute, as they eroded many of the hard-won gains of the postwar decades. As corporations attempted to restore profitability, the postwar compact between labor and capital was predictably abandoned. Initially, organized labor pushed back against inflation, new forms of automation, and the loss of rank-and-file control over workplaces."5 But labor militancy could succeed only so long as there were profits to be shared. As the 1970s progressed and profitability continued to decline, American companies had little to lose from labor action. Facing economic stagflation at home and steep competition from abroad, American manufacturing 190 JAMEs T. PATTERSON, RESTLESS GUNT: THE UNITED STATEs FROM WATERGATE To BUSH V. GoRE 7 (2005) [hereinafter PATTERSON, RESTLESS GUNTJ. 191 EDWARD L. AYERS ET AL., AMERICAN PASSAGES: A HISTORY OF THE UNITED STATES 814 (4th ed. 2009); HARVEY, supra note 182, at 145. 192 HARVEY, supra note 182, at 145. 193 Id. at 141-42. 194 Bailey & Farber, supra note 189, at 3. 195 See Jefferson Cowie, "Vigorously Left, Right, and Center": Crosscurrents of orking- Class Arnerica in the 1970s, in AMRICA IN THE SEVENTIES, supra note 189,75,76-78 [hereinafter Cowie, "Vigorously Left, Right, and Center"]. 24.3 343 COLUMBIA JOURNAL OF GENDER AND LAw fled to the open shops of the West and South.'1 At the same time, the number of workers in the manufacturing industry as a whole-the traditional core of union power-shrank as a percentage of the national labor force, further weakening union influence, even as the number of women, part-time, and service employees, largely outside of unions, continued to rise.1 1 Unionization rates in the private sector dropped to twenty-two percent by 1979, and between 1973 and 1979 the number of days that strikes disrupted businesses fell approximately one-quarter.' As American businesses restructured and retrenched, they explicitly sought to use the economic malaise to weaken or break the power of unions, first through the use of replacement workers'" and later through the lay-offs, plant closings, and bankruptcy proceedings that rendered labor militancy irrelevant.200 The result was a growing tide of concessionary bargaining and labor give-backs.20' If any doubt remained regarding the relevance of unions, President Reagan put them to rest by summarily firing over eleven thousand striking air traffic controllers in 1981.202 The early 1970s thus formed the turning point in the fortunes of white male workers: after seeing their wages rise forty- two percent in the 1960s to reach a peak in 1972, the crisis of 1973 would precipitate a decline that would continue for the next twenty-five years.203 196 See BRUCE SCHULMAN, THE SEVENTIEs: THE GREAr SHIFT IN AMERICAN CULTURE, SOCIETY, AND POLrICS 111 (2002). 197 See PATTERSON, RESTLESS GIANT, supra note 190, at 65. 198 See Brenner, supra note 188, at 42. 199 See generally JULIUs GETMAN, THE BETRAYAL oF LOCAL 14: PAPERWORKERS, POLITICS, AND PERMA'LNENT REPLACEMENrS (1998); WILLIAM B. GOULD IV,AGENDA FOR REFORM 185-88,202-03 (1993); JosephA. McCartin, "Fire the Hell Out of Them": Sanitation Workers'Struggles and the Normalization of the Striker Replacement Strategy in the 1970s, LAB., Fall 2005, at 67. 200 See PHILIP NICHOLSON, LABOR'S STORY IN THE UNITED STATES 291-92 (2004). 201 Id. at 292. See also KIm MOODY, AN INJURY To ALL: THE DECLINE OF AMERICAN UNIONISM 155 (1997) (describing the new union posture as "a partner in corporate triage and salvation"); Katherine Van Wezel Stone, The Legacy of Industrial Pluralism: The Tension Betveen Individual Employment Rights and the New Deal Collective Bargaining System, 59 U. CHI. L. REv. 575 (1992) (surveying the role of labor law in the union decline of the 1980s). 202 See generally SCHULMAN, supra note 196, at 232-34; Cowie, "Vigorously Left, Right, and Center", supra note 195, at 100-02; Bernard D. Meltzer & Cass R. Sunstein, Public Employee Strikes, Executive Discretion, and the Air Traffic Controllers, 50 U. CHI L. Rv. 731 (1983). 203 JEFFERSON COWIE, STAyIN' ALIVE: THE 1970S AND THE LAST DAYS OF THE WORKING CLAss 12 (2010) [hereinafter COWIE, SAyIN' ALIVE]. 344 24.3 COLUMBIA JOURNAL OF GENDER AND LAw B. From the "Silent Majority" to Bakke: An Identity Politics of the White Working Man In 1974, the cultural resentments of working-class white men found an unlikely hero in the figure of Allan Bakke: exemplary white male, ex-Marine, Vietnam veteran, and plaintiff of a lawsuit against the University of California alleging that the special admissions program of the medical school at Davis operated to exclude him from the school on the basis of his race in violation of the Fourteenth Amendment.204 Bakke's lawsuit also usefully summarizes the more complex cultural transformation in the meaning of the white working man from untroubled representative of mainstream America to marginalized and angry icon of "reverse discrimination." One source of that cultural transformation was, of course, the economic changes already surveyed. Under the Fordist-Keynesian regime of the postwar era, the white working man had come to embody a particular hegemonic form of masculinity around which was woven a broader symbolic image of U.S. society, including the nuclear family. the working class home, the union hall, and the Catholic Church.20 5 However, with the social turbulence of the 1960s and the economic crisis of the 1970s, this social order appeared to many to be imperiled. In the wake of the civil rights and free speech movements, Vietnam and the anti-war movement, and stagflation and the rustbelt, the contours of white working-class life were reshaped as the meaning of work, gender, and race all underwent transformation.206 The postwar model of white working-class identification became difficult to maintain and increasingly available for cultural rearticulation.20 By the early 1970s, Allan Bakke's isolated struggle for admission to medical school could be felt as culturally representative of the general dispossession of working white men from the center of U.S. life, as well as the incipient formation of a new politics intended on channeling this anger into a larger political and cultural agenda. 204 Regents of the University of California v. Bakke, 438 U.S. 265, 277-78 (1978). 205 See, e.g., JOSiUA B. FREELAN, WORKING-CLASS NEW YORK: LIFE AND LABOR SINCE WORLD WAR 11 (2001). 206 See generally COWIE, STAYIN' ALIVE, supra note 203; KENNETH DURR, BEHIND THE BACKLASH: WHITE WORKING-CLAsS POLITICS IN BALTIMORE, 1940-1980 (2007). For the historical roots of these changes, see THOMAS J. SUGRUE, ORIGINS OF THE URBAN CRISIS: RACE AND INEQUALITY IN POSTWAR DETROI (2005); Thomas J. Sugrue, Crabgrass-Roots Politics: Race, Rights, and the Reaction against Liberalism in the Urban North, 1940-1964, 82 J. Am. HisT. 551 (1995). 207 See generally STANLEY ARONOWITZ, THE POLITICS OF IDENTITY: CLASS, CULTURE, SOCIAL MNOVEMENrs (1992) (tracing the displacement of class-based identities after the rise of "disorganized" capitalism in the 1980s); Geoff Eley & Keith Nield, Farewell to the Working Class?, INT'L LAB. & WORKING-CLAss HisT., Spring 2000, at I (tracing the uneven recomposition of working-class collectivity after the 1970s). 24.3 345 COLUMBIA JOURNAL OF GENDER AND LAW From the mid-1960s through the end of the 1970s, various avatars of this disgruntled and disenfranchised white working man circulated in U.S. culture, prefiguring Bakke's historic discrimination claim.2 0 Significantly, these images typically evoked a now obsolete Cold War masculinity whose assertion of normality derived, not from widespread social investment, but from an oppositional stance towards the burgeoning counterculture-an opposition to opposition, so to speak. In 1969, for instance, journalist Pete Hamill painted an explosive portrait of the "white working class" as a "people on the edge of open, sustained and possibly violent revolt," squeezed between "[t]axes and the rising cost of living" and consequently resentful of government hand-outs to undeserving minorities demanding reparations.2 09 Similarly, in his November 3, 1969, speech, Nixon made his famous appeal to the "silent majority" as a counter-image to the "vocal minority" who wished to "impose [its opinion] on the Nation by mounting demonstrations in the streets."2' 0 Such figures appeared not only in the media but increasingly on the streets as well. Such a spectacle occurred on May 8, 1970, when two hundred construction workers in downtown New York descended upon students protesting the recent Kent State killings."' Later christened the "hard-hat riots"' b the media and turned into a series of increasingly staged demonstrations of pro-Nixon support in the months to follow, in its original, less orchestrated form, construction workers chased down and beat student protesters for more than two hours, leaving seventy people injured. Another shocking example occurred in the fall of 1974 during the anti-busing protests of South Boston. Charleston, and Hyde Park, as working- class white protesters attacked school buses, chanted racist slogans, and fought riot police in an effort to halt mandatory school desegregation.2 12 Against this cultural backdrop, the 208 See PATERSON, GRAND EXPECTATIONS, supra note 182, at 637-77 (describing the "backlash" of the late Sixties); Cowie, "Vigorously Left, Right, and Center", supra note 195, at 77, 84-91, 95. See generally DuRR, supra note 206. 209 Pete Hamill, The Revolt of the White Lower Middle Class, N.Y. MAG.,Apr. 14, 1969. See also Jefferson Cowie, Nixon's Class Struggle: Romancing the New Right Worker, 1969-1973, 43 LAB. Hisi 257, 260-61 (describing the influence of Hamill's essay on the Nixon administration's hopes to forge a "New Majority" based on white working-class rage) [hereinafter Cowie, Aixon's Class Struggle]. 210 Richard M. Nixon, Address to the Nation on the War in Vietnam, November 3, 1969, in LANDMARK SPEECHES ON THE VIETNAM WAR 142, 155-56 (Gregor) Allen Olson ed., 2010). See SCHULMAN, supra note 196, at 23-52 (on the broader cultural meaning of the Nixon presidency). 211 PATERSON, GRAND EXPECTATIONs, supra note 182, at 755-56. See also Joshua B. Freeman, Hardhats: Construction Workers, Manliness, and the 1970s Pro-War Demonstrations, 26 J. Soc. His. 725 (analyzing the cultural image of the 1970s construction worker as "politically reactionary, pathologically violent, and deeply misogynist"); Cowie,Nixon 's Class Struggle, supra note 209, at 264-66 (describing the Nixon administration's appropriation of the protests). 212 See JACK TAGER, BOSTON RioTs: THREE CENTURIES OF SOCIAL VIOLENCE 196-208 (2000). See generally 346 24.3 COLUMBIA JOURNAL OF GENDER AND LAw resentments of marginalized whites could coalesce around Bakke's own grievances and the image of white masculinity as a source of injury. Such images were symptomatic of the many ways in which the vanishing world of the steady union pay check and the all-white neighborhood were actively being reworked as a point of identification for a broader political and cultural movement seeking to arouse and harness working white disaffection. Given the power vacuum created by the unraveling of the New Deal political and legal order in the 1970s, it is unsurprising that conservative elements within the Republican party would be the first to formulate a new hegemonic project claiming to make sense of troubled times.21 That project emerged in the form of the New Right, mixing economic free-market liberalism with reactionary social conservatism and thereby coming to redefine the framework of mainstream political discourse to the present.2 14 Within the cultural landscape of post-New Deal, post-civil rights America, the white working-class man was cast as a central actor in the New Right's reinterpretation of the 1960s. in which the "liberal Democratic establishment" (to borrow George Wallace's phrase) had abandoned the bread-and-butter economic issues and social values of ordinary working people in order to accord preferential treatment and special constitutional protections to various "special interests," including racial minorities, prisoners, women, gays and lesbians, students. criminal defendants, and the mentally ill.2" In this retelling taxation was the illegitimate redistribution of wealth away from the hard-working white working class to "undeserving" blacks and Latinos who enjoyed the benefits of public housing, Medicaid, food stamps, and other government welfare programs without contributing to the public coffers. 16 In different ways, Howard Janis's Proposition 13. Reagan's hatred of "big government," and Clinton's promise to "end welfare as we know it" were all rooted within this post-1960s political landscape. RONALD FORMISANO, BOSTON AGAINST BUSING: RACE, CLASS, AND ETHNICITY IN THE 1960S AND 1970s (2d ed. 2003); Barry Stuart Roberts, The Extent ofFederal Judicial Equitable Power: Receivership ofJSouth Boston High School, 12 NIw ENG. L. REV. 55 (1976) (discussing the 1975 district court decision to place South Boston High School under receivership). 213 See Brenner, supra note 188, at 45. 214 See DAVIS, supra note 182, at 157-80; SCHULMAN, supra note 196, at 193-217. See generally LISA McGIRR, SUBURBAN WARRIORS: THE ORIGINS OF THE NEw AMERICAN RIGirr (2002). 215 See generally THOMAS BYRNE EDSALL WITH MARY D. EDSALL, CHAIN REACTION: THE IMPACT OF RACE, RIGIS, AND TAXES ON AMERICAN PoLITIcs (1992). 216 Id. 24.3 347 COLUMBIA JOURNAL OF GENDER AND LAW Yet it is not simply, as the New Right political narrative implies, that the white working class shifted political allegiances in response to liberal betrayal. It is also not simply, as the counternarrative on the Left has sometimes charged,1 that white workers were the victims of political manipulation that used social issues to push white workers to vote against their own economic interests. Whatever the truth of these two arguments, the fact is that white working men were placed in a highly contradictory social position during this period. In the 1940s to 1960s. the Fordist-Keynesian regime of the postwar period had incorporated white working-class men within the dominant mode of regulation and built around them the broader set of political and social investments already described. The white working male had come to symbolize this contingent alignment of forces and to experience that relationship as a basic feature of his selfhood-derived from his whiteness, his masculinity. and his status as a worker. When this investment was withdrawn after the 1960s and 1970s. the white male worker did not simply dissolve into a different configuration of alliances and identifications, but was seized upon and given a second, unnatural life by the New Right. The New Right worked to facilitate the general cultural and political investment in a largely anachronistic figure-the image of the 1950s working-class family and the "Golden Era" of postwar America "-which could channel social disaffection into a post-New Deal order." White workers were thus placed in an anomalous position: maintained as an active political figure, yet only through an identification with social and economic relations that no longer existed?220 Hence the contradictory position described earlier-suspended in the gulf between the symbolic identification with power that was embodied in their status as white, working class, and male. and their real and perceived exclusion from the actual social, political, and economic arrangements of the last four decades. 217 See, e.g., THoMIAS FRANK, WIAr'S THE MATTER WITH KANSAS? How CONSERVATIVES WON THE HEART OF AMERICA (2005). See also Tom Mertes, A Republican Proletariat, 30 NLw LEFI REV. 37 (2004) (discussing Frank). But see Larry M. Bartels, Vhat's the Matter with WVhat's the Matter with Kansas?, 1 Q. J. POL. Sci. 201 (2006) (criticizing Frank based on empirical evidence of voting patterns). 218 See STEPHANIE COoNTz, THE WAY WE NEVER WERE: AMERICAN FAMILIES AND THE NOSTALGIA TRAP (1992) (analyzing the "family crisis" of the 1980s and the idealized image of the middle-class family that it presumed); Matthew D. Lassiter, Inventing Family Values, in RIGIWARD BOUND: MAKING AMERICA CONSERVATIVE IN THE 1970s 13 (Bruce J. Schulman & Julian E. Zelizer eds., 2008) (analyzing the New Right investment in "family values"). 219 See generally Cowie, Nixon's Class Struggle, supra note 209 (discussing Nixon's "blue-collar strategy"). 220 Cowie notes the ironic fate of class in 1970s public discourse: just as the national discussion of class "teetered toward extinction," U.S. society began to experience unprecedented social inequality and income polarization. Jefferson Cowie, Portrait of the W1orking Class in a Convex Mirror: Toward a History of the Seventies, LAB., Fall 2005, at 93, 93-97. 348 24.3 COLUMBIA JOURNAL OF GENDER AND LAw Perhaps this contradictory position is best captured by the discourse of victimization and "reverse discrimination" that emerged in the late 1960s and 1970s.2' Bakke, in fact, represented something potentially very different from the other symbols of white working discontent in the period: for, in contrast to Nixon's "silent majority" or the construction workers who brutally beat anti-war protestors-voices of cultural resentment that nonetheless asserted their rightful place as the normative center of American society- Bakke couched his claims in the language of race. identity, and injury that owed as much to the civil rights movement as to the virulent racism of Boston's anti-busing demonstrators. And, in fact, Bakke's lawsuit can been seen to embody the more complex cultural work through which working-class "whiteness" was racialized and rendered available as a form of post-civil rights identity politics.2 ' As cultural historians have shown, part of that work was done through the dramatic resurgence of "white ethnicity" in the 1970s, spawning white ethnic organizations such as the Italian American Civil Rights League, popular quasi-sociological tracts such as Michael Novak's The Rise of the Unmeltable Ethnics, and films like The Godfather.2' Yet, despite its superficial nature, white ethnicity provided the necessary social cover for the deeper cultural replacement of the nineteenth-century idea ofAnglo-Saxon whiteness with the "recovered" idea of multiple, ethnicized Italian. Polish. and Irish white identities.124 Then, after the fad of white ethnicity passed by the mid-1970s. these multiple identities could be reconsolidated into a single racialized form of working- class whiteness that was available without regard to actual cultural heritage.225 Reference to "white ethnicity" became, from this perspective, redundant because whiteness as such had been reimagined as an ethnicity. 221 See generally DAVID SAVRAN,TAKING IT LIKE A MAN: WHITE MASCULINITY, NASOCHISM, AND CONTrEMPORARY AMERICAN CULTURE (1998) (providing a theoretical account of white male identification with victimization since the 1950s). 222 See generally Ian F. Haney Lopez, "A Nation of Minorities": Race, Ethnicity, and Reactionary Colorblindness, 59 STAN. L. REV. 985, 1021-46 (2007). 223 SCHULMNN, supra note 196, at 80-84. See generally MATHEW FRYE JACOBSON, Roo's Too: WHITE ETHNIC REVIVAL IN PosT-CIVIL RIGi rs AMERICA (2006). 224 On nineteenth-century whiteness, see NOEL IGNATIEV, How THE IRISH BECAME WHITE (1995); MATHEW FRYE JACOBSON, WHITENESS OF A DIFFERENT COLOR: EUROPEAN IMMIGRANTS AND THE ALCHEMY OF RACE (1999); DAVID R. ROEDIGER, THE WAGES OF WHITENEss: RACE AND THE NIAKING OF THE AMERICAN WORKING CLASS (1991). 225 For instance, although Allan Bakke did not belong to a discernible white ethnic group, several white ethnic organizations submitted amicus briefs in support of his position. See Brief of the Polish American Congress, The National Advocates Society and the National Medical and Dental Association as Amici Curiae, Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978) (No. 76-811), 1977 WL 204793; Brief of Amicus Curiae for the Order Sons of Italy inAmerica for Support of Respondent, Regents of the University of California v. Bakke, 438 U.S. 265 (1978) (No. 76-811), 1977 WL 189539. 24.3 349 COLUMBIA JOURNAL OF GENDER AND LAw This racialization had. in turn, been rendered possible by a number of more profound historical transformations. Most directly. white ethnicity was created from the template of the militant black cultural nationalist movement, which had forsaken the integrationalist ideals of the civil rights movement and gained its most popular representatives in Malcolm X and the Black Panther Party. 26 Less directly, the influx of non-European immigrant groups from Asia and Latin America after 1965 permitted whiteness as well as blackness to appear as two racial identities among many others.227 In a more fundamental sense, white ethnicity drew its energy from the cultural and political authority that only a claim to race could offer. In the wake of the civil rights movement, overtly racist language had been officially delegitimated, forcing working-class whites to seek a new vehicle for the expression of political demands.2' Building on the black nationalist paradigm, white ethnicity offered working-class whites a way to retrieve a history of oppression, common suffering, and shared discrimination that would allow them to compete with African Americans, Asian Americans. Latinos. and others in the struggle for political recognition. 2 9 It is in this specific sense that the white working man was reimagined as a form of identity politics. The result was an internally contradictory racial politics, to say the least. On the one hand, angry white ethnics became the public face of the backlash and the incarnation of Nixon's silent majority. On the other hand, white ethnicity was itself a distinctly post-1960s phenomenon, constituted through a reimagination of white working-class masculinity upon the model of black cultural nationalism, feminism, and post-Stonewall gay rights.23o Although that racial politics is only one part in the post-1960s reworking of masculinity, it is emblematic of the ways that race and class interceded in this period to transform the normative meanings associated with white working-class manhood. 226 See SCHULMAN, supra note 196, at 58-63; Thomas J. Sugrue & John D. Skrentny, The White Ethnic Strategy, in RIGHTWARD BOUND, supra note 218, 171, 177-78. 227 Eric Porter, Affirming and Disaffirming Actions: Remaking Race in the 1 9 70S, in AMERICA IN THE SEVENTIES, supra note 189, at 50, 59-60; SCHULMAN, supra note 196, at 68. 228 See Sugrue & Skrentny, supra note 226, at 178. 229 See id.; Howard Winant, Behind Blue Eyes: Reconstructing the White University, 225 NLw LEFT REV. 73,74-75; Micaela di Leonardo, White Ethnicities, Identity Politics, and Baby Bear's Chair, 41 Soc. TEXT 165, 167-78 (1994). 230 See di Leonardo, supra note 229, at 175; Sugrue & Skrentny, supra note 226, at 178-80; Winant, supra note 229, at 75. 350 24.3 COLUMBIA JOURNAL OF GENDER AND LAw C. Industrial Masculinity in the Era of Flexible Labor and Feminized Work The previous two sections have sketched the broader social and economic transformations of the postwar period and their effects on white working-class men. As I have argued, the equation of white working-class men with the general hegemonic structure of postwar U.S. society was itself a historically specific product that was no longer possible in the same way following the shifts of the 1960s and 1970s. Now I want to turn more specifically to how working-class masculinity was itself placed into crisis through the redefinition of "work" after the decline of Fordism. Labor historians have demonstrated the many ways that the workplace acts as a key site in the construction of masculinity and male identity. For instance, historian Steven Maynard has argued that, in the early twentieth century, the transition from craft production to mass production precipitated a crisis in previous ways of defining working-class manhood as the former association of masculinity with craft independence and control over the work process was stripped away by corporate Taylorism.231 In dialectical response, a new "rugged" industrial manhood emerged that defined masculinity in terms of the new and previously despised conditions of work, brute strength, and constant exposure to danger.232 In a similar manner, social historians have noted that. since the 1970s, the social conditions necessary for the maintenance of male working-class identity have eroded as the industrial work of the Fordist assembly-line has been superseded by and subsumed within a new paradigm of labor. As with the racialization of whiteness, the effect has been to transform the relation of white working masculinity to the hegemonic norms of American society. Central to this transformation have been changes to the nature of "work" after the decline of industrial Fordism. As we have seen, the 1970s witnessed a steady decline in 231 Steven Maynard, Rough Work and Rugged Men: The Social Construction of Masculinity in Working- Class History, 23 LABOUR/LE TRAVAIL 159, 160-66 (1989). See also Freeman, supra note 211; Steve Meyer, Rough Manhood: The Aggressive and Confrontational Shop Culture of U.S. Auto Workers During World War 11, 36 J. Soc. Hisi 125 (2002). For an overview of work in this vein, see Ava Baron, Masculinity, the Embodied Male Worker, and the Historian's Gaze, 69 INT'L LAB. & WORKING-CLASs HiST. 143 (2006). Frequently historical studies of masculinity have relied on a model of masculine "crisis" in which masculinity is imagined as suddenly thrown into disarray by changes in the structure of ownership, the division of labor, the incursion of women and/or immigrants, and fears of national decline. But, as historians have come to recognize, that narrative of crisis-which finds no counterpart in the construction of femininity-is itself an integral part of the construction of masculinity, leading to predictably reactionary calls for "remasculinization." See Judith A. Allen, Men Interminably in Crisis? Historians on Masculinity, Sexual Boundaries, and Manhood, 82 RADICAL Hisi Rv. 191 (2002); Bryce Traister, Academic Viagra: The Rise ofAnerican Masculinity Studies, 52 AM. Q. 274 (2000). 232 Baron, supra note 23 1, at 145-47; Maynard, supra note 231, at 162-66. 24.3 351 COLUMBIA JOURNAL OF GENDER AND LAw the Fordist order of mass production and mass consumption characteristic of the postwar boom. Initially, the decline of traditional working-class male employment was seen as a sign of deindustrialization and the complementary rise of the so-called "service sector." But in retrospect it is clear that these changes were only symptomatic of a more thorough- going restructuring of capital. 33 Part of that restructuring entailed the emergence of an informational mode of production, which was distinguished from industrial work by the new prominence of knowledge, information, affect, and communication. 34 Manufacturing itself has not merely declined but has been incorporated and recast within a new informational paradigm.35 In place of the Fordist model of uniform goods mass produced upon on an assembly-line, manufacturing is now typically structured around the extensive use of computerized robotics and just-in-time inventory systems that permit small-batch production of specialized goods without large fixed capital investments.36 Alongside this transformation in traditional manufacturing has been the collapse of the very distinction between production and service through the growth of what economists call "producer services": businesses that cater to other firms, for instance, by offering banking and finance, insurance, real estate, legal, accounting, and auditing services. While these services generate a select number of highly remunerative jobs, the vast majority are low-wage, low-skilled positions that constitute the invisible substratum of the contemporary firm. 238 Beyond this informational component, another important aspect of immaterial labor is the new centrality of affective and care-based work built around the commodification of human contact and social interaction."' The burgeoning healthcare industry, for instance, centers on the provision of care, whether in hospitals, clinics, doctors' offices, retirement or 233 See generally MICHAEL HARDT & ANToNIO NEGRI, EMPiRE 280-303 (2000) (describing the "informationalization" of production); HARVEY, supra note 182, 141-97 (describing the transition from Fordism to flexible accumulation); Mark Barenberg, Democracy and Domination in the Law ofWorkplace Cooperation: From Bureaucratic to Flexible Production, 94 CoLum. L. REV. 753, 879-928 (1994). 234 See generally MANUEL CASTELLS, THE RISE OF THE NETWORK SOCIETY (1996); HARDT & NEGRI, supra note 233, at 284-86. 235 See HARDT & NEGRI, supra note 233, at 289-90. 236 CASTELLS, supra note 234, at 166-80; HRDT & NEGRI, supra note 233, at 289-90; HARVEY, supra note 182, at 155-56; Barenberg, supra note 233, at 882-83. 237 See SASKIA SASSEN, THE GLOBAL CITY. NEW YORK, LONDON, TOKYO 90-197 (2001); John Tschetter, Provider Services Industry: Why Are They Growing So Rapidly?, MONTHLY LAB. REV., Dec. 1987, at 31. 238 HARDT & NEGRI, supra note 233, at 291-92. 239 Id. at 292-93; Michael Hardt, AJective Labor, BOUNDARY 2, Fall 1999, at 89. 352 24.3 COLUMBIA JOURNAL OF GENDER AND LAw convalescent centers.'40 Likewise, the entertainment and consumer technology industries have become co-partners in providing a nomadic public with endless ways of consuming communication and social interaction, most notably through the explosive rise of social networking technology.241 Whereas working-class men once constructed their masculinity around the role of producers capable of braving the danger and monotony of the industrial workplace, they now find themselves recast as seners dedicated to the ephemeral provision of affect or information. In addition to changes in the nature of working-class "work," the structure of the labor market and the labor process has been radically transformed since the 1970s. As we have seen. from World War II through the 1970s, the core of the white working-class community was in many ways founded upon the availability of the union job, carrying with it decent wages, long-term job security, and the promise of a pension. But in the wake of two bouts of inflation, unprecedented rates of unemployment, and the rollback of trade union influence, a very different organization of the labor market and the labor process itself have wrought changes in all sectors of the economy. In particular. since the 1970s, the labor market has been restructured to maximize flexibility through the replacement of full-time union employees with a variety of part-time, temporary, subcontracted, and outsourced workers able to satisfy shifting demands without long-term commitments for ongoing employment, healthcare benefits, or pensions.242 The result is a segmented labor market in which a small employee core enjoying guaranteed full-time employment, job security. and generous benefits is flanked by a larger contingent of peripheral employees susceptible to rapid downsizing in response to changing market conditions.24 3 These changes, in turn, have wrought shifts in industrial organization. Internally, the bureaucratic mode of corporate management has been replaced by nonhierarchical forms of "cooperative." project-driven teams that are then pitted against one another in an internalized, results- oriented market system.'44 Externally, the proliferation of outsourcing and organized subcontracting has revived atavistic systems of domestic, familial, and artisanal labor, 240 HARDT & NEGRI, supra note 233, at 292-93. 241 Id. 242 HARVEY, supra note 182, at 150. See generally Katherine V.W. Stone, The New Psychological Contract: Implications of the Changing W1orkplace for Labor and Employment Law, 48 UCLA L. REV. 519, 539-49 (2001). 243 CASTELLS, supra note 234, at 216-303; HARVEY, supra note 182, at 150-51. 244 CASTELLS, supra note 234, at 169-72; Barenberg, supra note 233, at 884-90 (describing the objective emergence of self-managing teams), 904-18 (describing the possibilities of domination attending self- managing teams). 24.3 353 COLUMBIA JOURNAL OF GENDER AND LAW sweatshop and casualized production, as well as the growth of unprecedented informal and underground economies.'~ *As a result, working-class solidarity and self-organization have been undercut through the intercession of family and domestic labor systems that replace economic relations of labor and capital with alternative forms of obligation and control.24 On a different but no less important symbolic level, the breadwinning white male is no longer the face of the working class. Women and persons of color have, of course. always been part of the American. working class. But they have never occupied the high-paying unionized positions that white men held in key industries such as manufacturing, mining and transportation.2 47 Yet with the deindustrialization of the Northeast and the rise of a predominantly white-collar service sector, the white and male working class became, by the end of the 1970s, a predominantly female, non-white. and non-union working class. Already by 1980, the American working class was more than 60% female and non-white; when white men under the age of twenty-five are added, then more than 70% of the working class was composed of women, minorities, or youth.2 4 8 Equally importantly, women and non-whites have been "proletarianized" in vastly greater proportions than white males. By 1980. 65% of black women held working-class jobs, compared to 64% for black men. 52% for white women, and only 38% for white men. 249 As a result of this recomposition, even on a demographic level, the earlier identification of the working class with white masculinity is no longer readily available.2' 0 245 HARVEY, supra note 182, at 152-53; SASSEN, supra note 237, at 289-305. 246 HARVEY, supra note 182, at 153. See generally Stone, supra note 242. 247 See DURR, supra note 206, at 104-09 (discussing racial barriers within Baltimore unions during the 1950s); DENNIS A. DESLIPPE, RIGHTS, NOT ROSES: UNIONS AND THE RISE OF WORKING-CLASs FEINISM 1945-80 (discussing the postwar emergence of working-class feminism in U.S. unions and the resistance it had to overcome). 248 Erik Olin Wright et al., TheAmerican Class Structure,47 AM. Soc. REV. 709,724 (1982). See also Mitra Toossi, A Century of Change: The U.S. Labor Force, 1950-2050, MONTHLY LAB. RAv., May 2002, at 15 (the number of women participating in the labor force rose from eighteen million in 1950 to sixty-six million in 2000). 249 Wright et al., supra note 248, at 724. 250 See David Roediger et al., The End of Whiteness? Reflections on a Demographic Landmark, NEw LAB. F., Spring/Summer 2001, at 49. See also Julie Bettie, Class Dismissed? Roseanne and the Changing Face of Working-Class Iconography, 45 Soc. TEXT 125 (1995) (tracing shifts in the popular cultural representations of the working class); Marion Crain & Ken Matheny, Labor's Identity Crisis, 89 CAL. L. REV. 1767 (2001) (analyzing how labor law, among other factors, has continued to marginalize race, ethnicity, gender, and other social identities in labor ideology); Eley & Nield, supra note 207 (tracing the uneven shift from the industrial male proletariat to a new working-class formation centered around female white-collar workers and public 354 24.3 COLUMBIA JOURNAL OF GENDER AND LAw It is the force of these accumulated changes in the nature, structure, and composition of labor since the 1960s that accounts for one aspect of the contradictory position of white working men today. In the most basic terms, the restructuring of work under globalized post-Fordism has radically transformed the ability of the workplace to function as a viable site for the construction of white working-class masculinity. While the working class still exists today, and is undoubtedly larger than ever, the structure of work that helped to affirm an industrial working-class masculinity, and the broader social norms it represented, is no longer available. Rather, post-Fordist labor-affect-based, casualized, and de facto embodied by women, minorities, and the young-is now constructed around a feminized subject regardless of the actual gender of workers. As cultural theorist Donna Haraway has argued, work has been "redefined as both literally female and feminized, whether performed by men or women."251 As Haraway explains: "To be feminized means to be made extremely vulnerable; able to be disassembled. reassembled. exploited as a reserve labor force; seen less as workers than as seners; subjected to time arrangements on and off the paid job that make a mockery of a limited work day; leading an existence that always borders on being obscene, out of place, and reducible to sex."252 As a result of these shifts. even for those men who remain within traditional working-class occupations-those offshore spaces on oil-platforms or automotive assembly lines-being a "White working- class man" names a bygone era. In this sense, even white working-class men are no longer "white working-class men" because the social and economic framework that gave that term its broader meaning no longer exists. It is with this altered framework in mind that we must now retheorize sexual harassment. IV. Retheorizing Sexual Harassment As the previous section has argued, sexual harassment should be understood, not in relation to an abstract order of gender norms and heteropatriarchy, but more specifically in terms of the varying historical constructions of manhood and the shifting normative status they have assumed. The prior section attempted to map out some of these shifts by tracing how white working-class masculinity has been redefined and culturally mobilized since the 1960s. In this section, I want to track the impact of these historical deformations on our employees). 251 DONNA HARAW4Y, SIMIANS, CYBORGS, AND WOMEN: THE REINVENTION OF NATuRE 166 (1991). 252 Id. See also HARDT & NEGRI, supra note 233, at 293 (arguing that the new prominence of affect in contemporary labor represents the generalization of a model of production previously identified as-and culturally understood to be-"women's work"). 24.3 355 COLUMBIA JOURNAL OF GENDER AND LAw understanding of sexual harassment through a discussion of those contradictory instances of same-sex sexual harassment in the workplace known as horseplay. But first it might be helpful to describe why I believe horseplay can be useful for a general rethinking of sexual harassment. First. as a form of sexualized abuse, horseplay resonates with a far wider set of conduct that courts continue to regard as prototypical of sexual harassment. Feminist theories have attempted to offset this presumption in various important ways-for instance, by shifting the focus to nonsexual forms of discrimination or by showing that the harm lies not in sexuality but in how sexuality "spills over" to reinforce stereotypical sex roles. But these strategies do not directly confront the sexualized nature of this conduct. Horseplay, however, forces a fundamental reinterpretation of sexualized harassment by providing examples of extreme sexual abuse that cannot be reduced to expressions of sexual interest. Horseplay presents, in naked form, a sexuality in the service of power. Second, horseplay permits us to understand more fully the relationship between power and normalization. In its deliberate violation of taboos on homoeroticism, horseplay breaks with prevalent cultural norms. Yet, even as it refuses to be reduced to the enforcement of already existing norms, horseplay is unquestionably the exercise of a specifically gendered and sexualized form of power. Rather than regarding horseplay as simply "deviant," then, we need to go beyond a unitary notion of gender norms to imagine a more diverse. uneven, and dynamic normative field, in which various conflicting representations compete for influence across different sectors of society. The apparent abnormality of horseplay offers a fresh perspective from which to study the active constitution and deconstitution of regulatory norms within this shifting field, as well as the fact that power need not always assume a normative cast. Third, the contradictions that run through horseplay might be paradigmatic of the ways that all sexual harassment is articulated within a broader circuit of social power, such that the invocation of gender-based power represents an attempt to compensate for losses that have accumulated along other social registers. Rather than imagining sexual harassment as the circular "reinforcement"' or "reproduction" of existing gender hierarchies, then, horseplay demonstrates that power is a relation rather than a commodity and that some of its most visible manifestations might be best seen as stagings of power underwAritten by perceptions of loss, deprivation, and crisis. Indeed, from this perspective, some enactments of sexual harassment might directly result from its perpetrators' inability to accede to the norms of masculinity. 356 24.3 COLUMBIA JOURNAL OF GENDER AND LAw Finally, horseplay ruptures the sheen of "normality" that frequently cloaks the perpetrators of sexual harassment. If horseplay's anomalous cast dramatizes the compensatory circulation of gender-based power, then it may usefully typify the conflicts, compensations, and contradictions that energize all acts of sexual harassment, even in their most stereotypical forms. In almost all such enactments, perpetrators deliberately hide these contradictions through an invocation of "confident masculinity." Treating horseplay as paradigmatic allows these displacements and compensations to be placed back at the center of sexual harassment, so that the ruptures that traverse contemporary masculinity are rendered visible and can be exploited as strategic sites for intervention and change. A. Sexualized Abuse and the All-Male Workplace The "abnormality" of horseplay stems from its perpetrators' transgression of general social prohibitions against same-sex sexual conduct. Therefore, as a first step, we must dislodge the persistent idea that the heterosexual/homosexual binary represents the only form of sexual norm. Historians of sexuality have shown that the stark division between homosexuality and heterosexuality apparently so central to contemporary American society is, in fact, a surprisingly recent invention that received widespread cultural currency only after World War 11.253 Prior to this period, the meanings and linkages among sexual practices, sexual identities, and sexual objects have been historically understood and articulated in a variety of ways.'54 The concept of sexual orientation was taken up from a more diffuse popular self-understanding and formalized at the end of the nineteenth century in the work of sexologists Richard von Krafft-Ebbing and Havelock Ellis.2'5 They postulated the existence of two distinct psychological types of person, applicable to the entire adult 253 See, in particular, GEORGE CIALUNCEY, GAY NEW YORK: GENDER, URBAN CULTURE, AND THE MAKING OF THE GAY MALE WORLD, 1890-1940, at 13-23 (1995). 254 See JOHN D'EMnIo & ESTELLE B. FREEDMAN, INTIMATE MATTERS: A HISTORY OF SEXUALITY IN AMERICA (2d ed. 1998); Robert A. Padgug, Sexual Matters: On Conceptualizing Sexuality in History, in PASSION AND POWER: SEXUALITY IN HIsioRY 14 (Kathy Peiss et al. eds., 1989); George Chauncey, "Wlhat Gay Studies Taught the Court": The Historians 'Anicus Brief in Lawrence v. Texas, 10 GLQ 509 (2004). See generally MICHEL FOUCAULT, HISTORY OF SEXUALITY: AN INTRODUCTION: VOLUME 1 (Robert Hurley trans., 1978). 255 On the medical model of homosexuality and its vicissitudes, see George Chauncey, From Sexual Inversion to Homosexuality: The Changing Medical Conceptualization ofFemale "Deviance, " in PASSION AND POWER, supra note 254, at 87 (discussing female "deviance" in nineteenth-century U.S. medicine); JONATHIAN NED KArz, THE INVENTION OF HETEROSEXUALITY (1995) (tracing the emergence of "heterosexuality"); JENNIFER TERRY, AN AMERICAN OBSESSION: SCIENCE, MEDICINE, AND HOMOSEXUALITY IN MODERN SOCIETY (1999) (tracking the shifting medical discourse of "homosexuality" as a response to the broader anxieties of U.S. society); JEFFREY WEEKS, COMING Our: HOMOSEXUAL POLITICS IN BRITAIN FROM THE NINETEENTH CENTURY TO THE PRESENT (1979). 24.3 357 COLUMBIA JOURNAL OF GENDER AND LAw population, in which the gender of an individuals preferred sexual object was the primary organizing principle of desire. It took another half-century for this medical concept to gain broader cultural acceptance as a form of self-understanding, initially among middle-class urban men and then in other sectors of the population.' Hence, horseplay might be better understood, not simply as anomalous or transgressive, but as the retrieval of an alternate set of gendered norms. Indeed, rather than the language of norms, it might be more productive to speak of a more fluid set of roles, positions, and statuses that sexuality might be capable of brokering within certain periods, cultures, and settings. If sexual orientation has itself emerged as one dominant norm in postwar American culture, it is even more important to recall that there have always been other models of gender and sexuality that have formed-and continue to form today-the fluid medium of sexual experience. Early twentieth-century working-class culture epitomizes one such alternate organization of sexuality and gender around active and passive sexual acts. As historian George Chauncey has shown. in this period, men were commonly identified by their gender status, which was only secondarily expressed through the roles that each person played in sexual acts.25 7 "Normal"' men could engage in sexual intercourse with other men and remain masculine so long as they did not assume the passive or receptive roles.' Men who assumed passive roles in same-sex encounters--variously called "fairies," "pansies," "sissies," "queers," and "queens"-were so categorized, not because they were seen as possessing a distinct sexual desire, but because their proclivities for anal and oral receptivity were deemed unmasculine and effeminate.25 Positions in sexual practice reflected underlying gender roles, and object choice was only a symptom of the more fundamental assumption of a different, "inverted" gender role.260 Horseplay can be seen to draw from this early repertory of sexual models as its perpetrators assume sexually aggressive positions towards other men while retaining, and even affirming, their masculine status. Horseplay thus finds particular resonance with the 256 See CIIAUNCEY, supra note 253, at 90-130. Other cultures continue to organize their dominant understanding of gender/sex roles outside of the heterosexual/homosexual binary. See, e.g., ROGER N. LANCASTER, Lii Is HARD: MACHISMO, DANGER, AND THE INTIMACY OF POWER IN NICARAGUA 235-78 (1992) (describing how, in Nicaraguan popular culture the passive anal-receptive partner is stigmatized as the cochdn, or "queer," whereas the active-insertive partner is seen as a "real man"). 257 See CHAUNCEY, supra note 253, at 13-23. 258 Id. at 13-23,70-71. 259 Id. at 14-23. 260 Id. 358 24.3 COLUMBIA JOURNAL OF GENDER AND LAw pre-war category of the "wolf," denoting a hyperbolically aggressive and virile man Who habitually seeks out other men-typically not fairies, but men otherwise distinguished by age, build, ethnicity, or some cultural vulnerability-for sexual partners. 6 ' The wolf's masculinity was, in fact, often positively reinforced through sexual relations with other men. 2 6 In contrast to the indifferent, if receptive, attitude of manly "trade,"' or the more solicitous and conjugal relations of "husbands" or "jockers" towards "fairies"' or "lambs," the wolf evinced a predatory same-sex appetite willing to seduce and perhaps physically coerce other men for his satisfaction.263 Consequently, the wolf's relationships were often organized less around gender roles, as with husbands and fairies, than on the basis of status hierarchy and power.26 4 The wolf thus represents one culturally available script, markedly distinct from that of sexual orientation, in which a man could actually enhance his masculine status through the eroticized domination of other men. Furthermore, as the example of the wolf illustrates, these alternate understandings of gender and sexuality have not vanished with the rise of the heterosexual/homosexual binary, but have continued to coexist, bleed into, and inflect contemporary forms of sexual practice and identity. Formerly, historians of sexuality tacitly assumed that different modes of sexual understanding fell into a succession of historical paradigms, often shifting from isolated sodomitical acts to the consolidation of a homosexual identity.'65 Yet horseplay is not simply the atavistic return of an anachronistic sexual persona. For, obviously, the postwar prohibition on same-sex intimacy as unmanly still exerts a powerful influence over 261 Id. at 87-95. See also PETER BOAG, SAME-SLx AFFAIRS. CONSTRUCTING AND CONTROLLING HOMOSEXUALITY IN THE PACIFIC NORTHWEST 25-26, 31-35 (2003) (describing wolf/lamb relationships within the turn-of-the- twentieth-century U.S. sexual system); REGINA KUNZEL, CRIMINAL INTIMACY: PRISON AND THE UNEVEN HisT'ORY OF MODERN SEXUALITY 62-71 (2008) (describing the wolf/punk relationship as part of the U.S. prison sexual culture in the 1920s and 1930s). 262 See CHAUNCEY, supra note 253, at 87-95. 263 Id. Of course, many of these categories were frequently conflated, with "wolf" and "jocker" in particular being used interchangeably to indicate otherwise masculine men who maintained the active-insertive position in sex in contrast to the punk or lamb. See CLAUNCEY, supra note 253, at 87; KUNZEL supra note 261, at 66; BoAG, supra note 261, at 25-26. 264 BoAG, supra note 261, at 88. 265 Undoubtedly, the opposition between "acts" versus "identities" was galvanized by Foucault's History of Sexuality. See FOUCAULT, supra note 254, at 43. But see David Halperin, Forgetting Foucault: Acts, Identities, and the History of Sexuality, REPRESENTATIONs, Summer 1998, at 93 (exploring the complexities of this opposition in Foucault and in the historical formation of earl) Greek sexuality). In an) case, the general tendency to narrativize the construction of sexuality goes beyond the influence of Foucault's work. See EVE KoSOFSKY SEDGWVICK, EPISTEMOLOGY OF THE CLOsET 44 n.41 (1990) (discussing accounts that narrativize the development of sexual paradigms with modern homosexuality as their conclusion). 24.3 359 COLUMBIA JOURNAL OF GENDER AND LAw this conduct, generating the odd mixture of verbal threats and bizarre physical enactments seen in Oncale and Mc Williams: brutal acts of sexual humiliation that both invoke and avoid actual sex. To understand how horseplay takes up but redeploys, for instance, the sexual role of the wolf. we must forsake linear historical narratives in order to recognize the heterogeneity and diversity of sexual knowledges and practices coexisting down to the present. This point has been forcefully made by literary and cultural theorist Eve Sedgwick, who argues that the historical progression from acts to identities has tended to reinforce the notion that the heterosexual/homosexual opposition represents a totalizing and coherent definitional system monopolizing the present.'66 Observing "the unrationalized coexistence of different models" 6 of sexuality in various historical periods, Sedgwick argues that contemporary sexuality can more productively be viewed as a performative "space of overlapping, contradictory, and conflictual definitional forces"268 that have become sedimented through time. Extending Sedgwick's argument, David Halperin has contended that there is no unitary history of male homosexuality.'6' Halperin, for instance, traces the coexistence of at least four provisional systems of sexual categorization whose histories and interactions have been obscured, but not superseded, by the emergence of the heterosexual/ homosexual binary.270 According to Halperin, homosexuality and heterosexuality do not name the concluding paradigm in the history of sexuality, but rather denote "an effect of this cumulative process of historical overlay and accretion" caused by the sedimentation of these models of male sexuality and gender deviance which pre-date the ascendency of sexual orientation.' Once we cease viewing these forms of sexual practice as organic expressions of an individual's sexual identity or a unified historical discourse, we can, instead, begin to appreciate their existence as disparate cultural scripts still available today for opportunistic retrieval. Moreover, in order to understand the particularity of all-male workplaces, we must recognize the ways these scripts move along continuities of space, geography, and environment rather than identities or desires-constituting a kind of "situational sexuality" 266 SEDGWICK, supra note 265, at 45-48. 267 Id. at 47. 268 Id. at 45. 269 DAVID HALPERIN, How To DO THE HIsTORY oF HOMOSEXUALITY 109 (2002). See also id. at 10-13 (discussing Sedgwick's criticism of his earlier work). 270 Id. at 109-10. 271 Id. at 109. 360 24.3 COLUMBIA JOURNAL OF GENDER AND LAw emerging within particular contexts and institutions.' Historian John Howard, for instance, has argued that there is a distinct regional and spatial distribution of sexuality that only comes into focus once the rigid logic of homosexual/heterosexual self-identification is relaxed.273 Howard, for instance, examines the dispersed networks that sustained possibilities of same- sex desire in Mississippi from 1945 to 1985 fostered by the bus stations. hotel restrooms. bars. and automobiles in which men met and mingled.'27 Chauncey likewise stresses how the specific erotic system of wolves. jockers, and punks was widespread among groups of men who found themselves separated from the family and neighborhood systems that regulated sexual norms, particularly prisoners, seamen, and itinerant workers. ' Historian Peter Boag develops this insight by exploring the Pacific Northwest prior to World War II. where he uncovers two contrasting homoerotic subcultures, coexisting in time but separated in space .2 6 The first was a working-class world rooted in the labor camps that seasonally sprung up around the logging, fishing, mining, and railroad jobs of the region and which drew together migrant workers, causal labors. and "hobos" numbering in the thousands.2 77 In these transient spaces, a distinctive sexual culture developed in which adult jockers engaged in sexual relations with younger male punks, who rejected oral sex in preference to interfemoral intercourse.2 78 A second, quite distinct sexual culture emerged among the growing population of urban white-collar workers-clerks, salesmen, teachers, small shop owners-who met and mingled in the citys' parks, hotels, and saloons.279 These middle- class urban spaces fostered a very different culture of discrete same-sex relationships between adult men of the same age that did not include the taboo on fellatio that existed among jockers and punks.' 272 See generally KUNZEL, supra note 261, at 8-9 (arguing that the situational nature of prison sexual culture suggests the dangerously "situational" nature of all sexual identity). 273 See JOHN HOVARD, MEN LiKE TnAr: A SOUTHERN QUEER HISiORY 3-17 (2001). 274 Id. 275 CHAUNCEY, supra note 253, at 76-91. See also Maynard, supra note 231, at 167-69 (noting that male bushworkers in the logging camps of northern Ontario in the late nineteenth and early twentieth century engaged in same-sex relations). On the sexual significance of the open seas, see HANs TURLEY, Rum, SODOMY, AND THE LASH. PIRACY, SEXUALITY, AND MASCULINE IDENTITY (1999). On prison sexual cultures, see KUNZEL, supra note 261. 276 BOAG, supra note 261. 277 Id. at 15-44. 278 Id. at 25-28, 35-39. 279 Id. at 89-124. 280 Id. at 117-20. 24.3 361 COLUMBIA JOURNAL OF GENDER AND LAw Cases concerning same-sex sexual harassment demonstrate that sexualized conduct continues to pervade and structure certain male-dominated workplaces to the present. Such conduct assumes many forms and intensities. For instance, in the mascot production shop in Ocheltree v. Scollon Productions, Inc., male employees fostered an aggressive sexual culture of boasting. mutual taunts. crude erotic enactments, and general one-upsmanship undoubtedly intended to see who would be the first to betray offense."' One male worker told coworkers about his wife "sucking his dick and swallowing and letting it run down the side of her chin," while. on another occasion, the shop supervisor described his enjoyment of intercourse with young boys.28 On a different occasion. employees speculated on homosexual relationships between various male employees and even on how one employee was having sex with a dog.28 3 The same mixture of camaraderie and belligerence can be detected in the workplace practice of "bagging," in which male workers grab and squeeze each other's testicles. For instance, in Quick v. Donaldson Co.. Phil Quick complained that he was bagged approximately a hundred times over the course of a year, including one incident in which a coworker grabbed Quick by the arms and lifted him off the ground while calling on another worker to squeeze Quick's testicles hard enough to produce swelling and bruising. Although such behavior was dismissed by the district court as "non-sexual aggression" aimed at Quick's genitals,285 in truth bagging functions as a formalized, low- level practice by which men demonstrate their masculinity through an ability to give and take sexualized abuse. On the more extreme end of this spectrum, there is the brutal and carefully targeted sexualized abuse of Dennis Breitenfeldt in Breitenfeldt v. Long Prairie Packing Co. 6 Working as a "boner" in a Minnesota meatpacking plant, Breitenfeldt alleged he was daily the victim of sexually violent behavior that he and his coworkers termed "violating" -consisting, in Breitenfeldt's case, of being jumped by coworkers and held down in bins of raw meat or troughs of blood while simulated acts of oral or anal sex were performed upon him; having his testicles grabbed or hit by coworkers; having a steel rod used for knife sharpening forcibly rubbed between his legs; and being asked whether 281 308 F.3d 351, 353-54,367-69 (4th Cir. 2002). 282 Id. at 369. 283 Id. 284 90 F.3d 1372, 1375 (8th Cir. 1996). 285 895 F. Supp. 1288, 1296 (S.D. Iowa 1995). 286 48 F. Supp. 2d 1170 (D. Mimi. 1990). 362 24.3 COLUMBIA JOURNAL OF GENDER AND LAw he preferred "it up the ass or down the throat."2 As if this were not enough, when a light above Breitenfeldt's workstation flickered, his coworkers would yell "blue light special," jump him, and again simulate sex acts. 8 8 When he complained, Breitenfeldt's supervisor dismissed the conduct as "just in good fun" and refused to have the light bulb changed28 9 Breitenfeldt's fate follows those of plaintiffs in Oncale, McWilliams, and Giddens in offering perhaps the most extreme and difficult to explain instances of male-male sexualized abuse. Together, these cases justify a long-standing cultural suspicion that male-dominated environments are capable of fostering "deviant" forms of gender and sexuality. In the next section, I want to draw together these insights to develop a more coherent understanding of both horseplay and sexual harassment more broadly. B. Horseplay, Contradiction, and Acting Out Having dislodged the notion of a single sexual norm, as well as the related idea of a totalizing sexual system, we can now present a theory of horseplay that more fully accounts for its counter-normative and compensatory elements. As we have seen, the sexual acts typical of horseplay remain inscrutable only if sexual norms are narrowly identified with the heterosexual/homosexual binary. Once we acknowledge that sexual orientation is a relatively recent and incomplete form of sexual self-understanding, we can recognize the diverse array of models of gender and sexuality that continue to inflect contemporary sexual experience. Rather than invoking a unitary system of gender norms, we might, instead, more productively speak of a variety of scripts through which gendered and sexed practices are circulated and reshaped. Early twentieth-century working-class culture offers many examples of how same-sex relationships can be organized around scripts centered upon gender status (fairies, trade, husbands), age differences (jockers and punks), or differentials of power, position, and physical force (wolves and lambs). These alternative scripts have not simply disappeared today but have continued to circulate and resurface in both familiar and unfamiliar forms. All-male environments, ranging from prisons to offshore workplaces, have played a key role in the cultural transmission of these alternate forms of gender and sexuality. which have remained within the institutional memory of these sex-segregated spaces as a repertory of sexualized practices, initiatory acts, and gender scripts available for appropriation and enactment in the present. 287 Id. at 1172. 288 Id. 289 Id. 24.3 363 COLUMBIA JOURNAL OF GENDER AND LAw Consequently, horseplay may represent an attempt to resuscitate a specifically sexed form of power to compensate for the deficiencies otherwise felt to exist among certain groups of men. In fact, the very extremity of these sexual acts bears powerful witness to the profound needs animating these occurrences. One way to understand these needs, I have argued, is by situating them within the broader transformation of white working-class manhood since the 1960s. The dominant political narratives of the post-New Deal era cast these social and economic changes in terms of the betrayal and marginalization of white working-class men. But that narrative of social disenfranchisement represents a cultural screen memory projected upon, but also rendered possible by, the more fundamental historical crisis and dissolution of the Fordist-Keynesian regime which had underwritten the material and cultural organization of postwar white working-class life. In the aftermath of the crises of the 1960s and 1970s, working-class white men experienced an increasingly fractured relationship to the ways they had formerly articulated their gender and racial identities. The result was a sense of social marginalization and political disenfranchisement that was channeled by the New Right into a highly particular economic and political agenda. The figure of "white victimization" epitomized, for instance, by Allan Bakke offers one representation of this broader cultural narrative and the corresponding identity politics of white male injury it has come to sustain. Consequently, rather than seeing horseplay as an extreme or anomalous form of conduct. we might use it as a lens through which to view apparently more normative instances of male-female harassment. For the contradictions of horseplay are contradictions only if we presume that its perpetrators are, in fact. invested in the social norms they violate- specifically, social taboos on same-sex conduct. But, as the onshore world recedes, the imagined investment in such abstract norms recedes and more pressing needs surface. Indeed, sexual harassment may not be rooted primarily in an imagined investment in an abstract gender order at all, but in the possibility that sex offers an intensely bodily experience of power that, for certain segments of men. at least. might compensate for less tangible perceptions of social loss. Even in "normative" instances of male-female harassment, then, "patriarchy" is not an overarching power structure that men are tasked to defend. but a symbolic resource that they actively call upon through talismanic acts such as sexual harassment.2"o Horseplay in fact permits us to see all sexual harassment as 290 Raewyn Connell employs the concept of the "patriarchal dividend" in order to explore the ways that individual men actively enjoy the benefits derived from "patriarchy" as a symbolic structure, even when they do not directly engage in harassment or discrimination. R.J. CONNELL, MASCULINITIEs 79 (1995). My account argues that the reverse relationship may also apply. Whereas the patriarchal dividend focuses on the hidden or indirect participation (and benefit) of men in the symbolic structure of patriarchy, I want to emphasize how men may overtly seek to invoke (or dramatize) their own participation in "patriarchy" precisely because their share of the patriarchal dividend is missing or withdrawn. 364 24.3 COLUMBIA JOURNAL OF GENDER AND LAw a species of cultural script driven not by sexual desire or the ideological dictates of gender norms, but by the deeply felt promise of sexualized abuse to shore up and push back an array of social anxieties and contradictions. Horseplay thereby permits us to see all sexual harassment as the invocation of a specifically gendered power that must be situated within a more complex compensatory economy of racial, political, and economic marginalization that is part political fantasy and part social fact. The resulting demonstrations of sex-based power may, at times, align with normative ideas of gender (for instance, in opposite-sex harassment), while at other times it may violently transgress that system of norms (for instance, in horseplay). Yet, once we reject the priority of a gendered social structure. we can see that, in both same-sex and opposite-sex cases, we find situated social actors attempting to make their own histories out of what lies closest at hand-their own masculinities. This account also helps to show the insufficiency of any theory that defines sexual harassment strictly in terms of gender norms. The powerful contradictions that have altered the self-identification of white working-class men since the 1960s attest to the inadequacy of appealing to an abstract socioeconomic model of power such as heteropatriarchy to explain harassment. Early feminist legal scholars, such as MacKinnon, appeared to conflate sexuality and power within a single structure of male domination and female subordination with little space for female sexual agency or historical change. The influential work of Franke. Schultz. and Abrams has sought to right those accounts by positioning sexuality as only one tool that male workers may deploy in maintaining gender hierarchies and the male monopoly over more highly rewarded jobs.' Yet perhaps it is power. not sex. that is the problem. The solution is not to desexualize power, but to break with the static and hierarchical model of power that continues to animate certain feminist analyses of sex- based power. We must then understand both power and sex differently. On the one hand. in understanding power, we must break from a hierarchical model that imagines power as a commodity possessed by one group to the detriment of another, as well as from a structuralist model by which an impersonal system reproduces itself via a coercive social discipline. Rather, we must recognize that power is itself the provisional effect of numerous contingent alliances among disparate social forces, which only retroactively proclaims its naturalness and permanency. On the other hand, in understanding sexuality, we must break from both the nafve biological conception of sex as exclusively an expression of erotic desire as well as the radical feminist view of sex as solely a form of subordination. Instead, 291 See Abrams, supra note 28, at 1215 ("There is sexual harassment that involves the expression of sexual desire, sexual harassment that involves sex but no sexual desire, and even sexual harassment that involves no sex."); Franke, supra note 28, at 729 ("[SJex is the method, but sexism is the meaning of sexual harassment."); Schultz, supra note 28, at 1797 ("We need an account of hostile work environment harassment that recognizes that sexuality is only one tool that male workers can deploy in a struggle to maintain the masculine composition and image of more highly rewarded jobs."). 24.3 365 COLUMBIA JOURNAL OF GENDER AND LAw we must recognize sexuality's historical and social character as well, including the variety of now submerged idioms associated with the all-male workplace and its alternate repertory of sexualized scripts. These sexual scripts offer men a range of flexible forms through which to enact gender-based power. Yet the figure of "roles" and "scripts" must also not be read as implying a voluntaristic subject, able to adopt and discard erotic personae at will. Rather, these sexual models represent fields of cultural and individual investment that become lodged in and as us-as sources of loathing, fear. desire, or apathy. sometimes projected onto others and sometimes felt to be our innermost selves. At times, their gravities pull us from our familiar orbits as we find ourselves drawn into acts and appetites that we might not recognize as our own. Indeed, there might be an even more striking point to be made regarding harassment and gender norms. In contrast to the assumptions underpinning gender norms, this account suggests that the most hyperbolic dramatizations of gender hierarchy might. paradoxically be driven by the very inability of certain men to identify with male norms and enjoy the social entitlements that masculinity is felt to entail. Sociologist Raewyn Connell, for instance, has located a similar set of dynamics driving what she terms "protest masculinity."' While conducting extensive life histories of working-class men, Connell found that some working- class men adopt extreme forms of highly aggressive, socially defiant masculinity, not as a result of having passively internalized social norms regarding manhood, but as a mean of compensating for a more profound experience of social marginalization rooted in class.2 Due to their working-class backgrounds, these men have found their identifications with hegemonic masculinity difficult or impossible to maintain-many are barely literate-and the normal advantages that men possess over women and other social "others"' in terms of employment and job advancement are largely closed to them.2 94 Their response is to fully identify with the marginality that defines their social position and to transform that status into a hyperbolic display of defiant, often violent, virility." The resulting masculinity is extreme but also disembodied and external-a formal means for mounting social resistance and therefore maintained only on the level of outside persona, for instance, as 292 CONNELL, supra note 290, at 109-19. See also Michelle Fine et al., (hs)Secure Times: Constructing White Working-Class Masculinities in the Late 20th Century, GENDER & Soc'v, Feb. 1997, at 52 (drawing on ethnographic studies of white working-class adult and teenage men from the late 1980s and 1990s to trace how their eroded racial and class status leads to a characteristic oppositional stance based on gender and race that seeks to displace the realities of class). 293 CONNELL, supra note 290, at 109-19. 294 Id. at 115-16. 295 Id. 366 24.3 COLUMBIA JOURNAL OF GENDER AND LAw an obsession with face.29 6 Even as these young men aggressively assert gender boundaries, it is the sharply defined sense of difference, not any specific investment in the content of masculinity, that is crucial to propping up their imperiled sense of self.' Consequently, gender difference becomes experienced as paroxysms of sexuality and violence, narrowing masculinity to almost literal bodily experiences."' Indeed, as others have argued, in certain circumstances, outright sexual violence may offer such men a visceral experience of gender difference, enacting on the level of performance a masculinity missing on the level of identity.299 Building upon such work, it might be useful to replace the concept of gender normalization with the psychoanalytic notion of "acting out" in order to describe this fractured psychic and social economy. In its psychoanalytic valence, acting out denotes the often violent expression of psychic contradictions that defy conscious recognition and so manifest themselves through symptomatic and agitated acts that fail to resolve the underlying conflicts. 3 o0 These acts may temporarily relieve the built-up tensions within the self, but because they do not alter the underlying issues, tensions unavoidably mount once more. In parallel fashion, sexual harassment can be seen as a species of cultural acting out that responds to the contradictions inhabiting the contemporary identities of white male workers. Rather than viewing sexual harassment as a confident assertion of heteropatriarchal power, acting out suggests how sexualized acts in all-male institutions may be opportunistically seized upon as culturally available scripts for the attempted consolidation of personal male power, even when-and even because-same-sex acts transgress the hegemonic rules of heteronormativity. Such a paradoxical economy of masculinity might go some distance towards explaining why, in cases such as Oncale, men are willing to adopt socially stigmatized and counter-normative forms of domination. For men who find themselves cut off from the male prerogatives that they feel are rightfully 296 Id. at 109-12,116. 297 Id. at 109-11. 298 Id. 299 See, e.g., MIARK SELTZER, SERIAL KILLERS: DEATH AND LIFE IN AMERICA'S WOUND CULTURE 89-104 (1998); KIAUS THEVELEIT, MVALE FANTASIES, VOL. 1: WOMEN, FLOODS, BODIES, HISToRY (Stephen Conway et al. trans., 1987). 300 See Sigmund Freud, Remembering, Repeating, and Working-Through, in THE STANDARD EDITION OF THE COMPLETE PSYCHOLOGICAL WORKS OF SIGMUND FREUD XII 150 (James Strachey ed., 1953-74); Jean Laplanche & J.B. Pontalis, Acting Out, in LANGUAGE OF PSYCHO-ANALYSIs 4-6 (Donald Nicholson-Smith trans., 1974); Alan Rowan, The Place ofActing Out in Psychoanalysis: From Freud to Lacan, PSYCHOANALYTISCHE PERSPECTIEVEN, Oct. 2000, 83. 24.3 367 COLUMBIA JOURNAL OF GENDER AND LAw theirs, perverse sexualized acts may be appealing on at least two, inconsistent fronts: first, in registering the sense of perversity, marginalization, and stigma already felt by such men due to their historical situation; and, second, in simultaneously attempting to compensate for these feelings of powerlessness by using that same perversity to exercise masculine sexual dominance over other men. Yet, in the end, this Article advocates an agnosticism towards both the historical causes of this conduct and the psychological motives of these actors. Certainly, the prior discussion of the postwar transformation of white working-class masculinity, juxtaposed with our exploration of horseplay as the revival (or survival) of an earlier sexual system, suggests a historical narrative in which the accumulated social pressures upon a certain sector of men has resulted in these aberrant forms of workplace harassment. Yet such a causal narrative encounters a number of obstacles, not the least among them being the occurrence of same-sex sexualized abuse in a wide range of contemporary social settings, from college fraternities to military units to prisons. Moreover, as the discussion of early-twentieth century sexuality shows, not only are these same-sex acts not new, but they represent lines of continuity (if. at times. submerged) that have endured as part of the institutional memory of certain spaces and settings. Even speculation on whether these historical changes have increased or decreased the number or severity of same-sex incidents misrepresents the precise argument I am trying to make. In conclusion, I would like to make a series of corrective points regarding the scope and meaning of these arguments. First, the relationship I want to stress between horseplay and the altered economy of white working-class masculinity is not one of causation, but of social meaning. For, as the historical relationships among gender, race, and work have altered over recent decades. a set of pre-existing sexual practices and knowledge has been culturally appropriated, invested with new significance, and mobilized for the divergent purposes of particular social actors. While the practices themselves may endure, their meaning takes shape within a historical conjecture that is irrevocably different. A past repertory of sexualized scripts can today be harnessed to address cultural anxieties not present in earlier decades regarding perceptions of political and cultural marginalization. Indeed, Title VII itself offers an important means of contesting the significance of these acts: challenging the definition of such sexualized conduct as "horseplay" or "rough- housing"' and branding it, instead, as "sexual harassment"' gives this conduct a meaning that it likely does not have for the men who commit it."' Despite the overtly sexualized nature of Joseph Oncale's own harassment, his harassers would undoubtedly be shocked 301 See Cass R. Sunstein, On the Expressive Function of Law, 144 U. PA. L. Rv. 2021, 2032-33 (1996) (offering a general discussion of the law's ability to change social nonms). 368 24.3 COLUMBIA JOURNAL OF GENDER AND LAw to consider their conduct as sexual harassment-something that men do to women as an expression of desire. Altering the legal and cultural significance of conduct may in fact prove to be one of the most effective means of halting the conduct itself. Second, my discussion of white working-class masculinity is intended to challenge the idea of an abstract male order that lies at the center of the accounts of Franke, Schultz, and Abrams. Rather than seeing the conduct as intended to enforce gender norms, I argue that it should be understood as so many cultural scripts opportunistically taken up by particular social actors for any number of reasons. The difference in these approaches becomes clearer if we consider how they lead to particular perspectives on sexual harassment. If we see sexual harassment as the means of reinforcing a confident heteromasculine order, then it is logical to identify sexual harassment through an examination of the perpetrators and victims of sexual harassment. The result is both over-inclusive and under-inclusive. For whereas all male abuse of a female coworker would, without more, violate Title VII (itself an outcome no court would endorse), a male plaintiff such as Joseph Oncale-subject to brutal. sexualized abuse, but too ordinary to fit the profile of the "gender outlaw" finds himself out of luck. But if we see sexual harassment as a cultural script deployed for a variety of overdetermined cultural and personal reasons, then it is the sex-specific form of the harassment that becomes dispositive. From this perspective, the purpose of our account of white working-class manhood is largely negative: it serves to demonstrate that the precise reasons that individual men might turn to sex-specific conduct are complex and often founded on an overdetermined set of historical contradictions and compromises that belie the fiction of a unitary masculine order. While I have presented a detailed historical account of these shifting conditions, for purposes of Title VII, I am not advocating that courts move from one already complex theory of sexual harassment to a second, even more complex theory of sexual harassment. On the contrary, the legal upshot of my historical argument is that we should shift from an analysis of social function to an analysis of the objective content of harassment. Rather than inquiring into Whether conduct has the "'purpose or effect"' of reproducing gender norms, courts should ask if conduct takes a sex- specific form that discriminates on the basis of gender. Yet. in addition to the practical exigencies of the legal claim, there is also an important theoretical lesson that horseplay can offer us regarding the possible transformation of gender norms. To the extent that gender norms are able to ensure their reproduction through practices such as sexual harassment, it is unclear precisely how such a structure might be altered. Title VII. of course, along with other antidiscrimination laws, offers a legal means to oppose specific instances of harassment. However, that intervention remains local and reactive. Furthermore, we must be vigilant of the ways that the legal system itself 24.3 369 COLUMBIA JOURNAL OF GENDER AND LAW may indirectly naturalize certain social categories, as well as formalize the attachment of politicized identities to injuries that the law is then imagined to redress.3o2 The sex stereotyping claim, for instance. may work to codify the binary opposition of the harasser's "normalized" masculinity and the victim's gender "nonconformity," either taking these roles as straightforward facts or-if they are seen as social constructions-affirming the efficacy of the disciplinary process that produces them. By contrast, a focus on horseplay can provide a different perspective on the forms that intervention may take. The very imperative that gender norms materialize themselves through the embodied practices of individuals also necessarily generates the complementary possibility that such acts may go awry, becoming illegible or even unviable.a Rather than representing a rare exception to the dynamics that typically govern sexual harassment, horseplay may, in fact, exemplify an instability lodged within every enactment of gender. Horseplay may thus uncover the fault lines of a masculinity that is itself. already. less normal and less normative than one might assume and which therefore may act as the pressure point for the contestation and reelaboration of a broader set of regulatory practices. Admittedly, such a strategy of intervention might appear highly strange: it is not, for instance, located primarily in the nonconforming man who defies the normalizing violence of the system, but is instead lodged in the often violent and morally reprehensible acts of the perpetrators themselves. Yet this is the point: to replace a logic of resistance with the possibilities of transformation empowered by the definitional incoherence of the system itself. C. Horseplay in Practice: Doe v. City of Belleville The Seventh Circuit case of Doe v. Cit) of Belleville demonstrates why a focus on the dynamics of acting out presents a more productive theoretical approach to sexual harassment than is offered by concepts of gender norms and sexual stereotyping. 3 1o In Doe. the victims of same-sex harassment were sixteen-year-old twin brothers. referred to as "H." and "J.," who were hired by the city to work with a group of older men to cut weeds in the municipal cemetery. 0 ' Both boys were immediately harassed at work: J. for being overweight and H. for wearing an earring. .o6 But H. quickly became the main target of abuse, especially by 302 See WENDY BROWN, STATES OF INJURY 52-76 (1995). 303 See BUTLER, BODIES THAT NIATTER, supra note 118, at 10. 304 119 F.3d 563 (7th Cir. 1997), vacated sub noin. City of Belleville v. Doe, 523 U.S. 1001 (1998) (remanding in light of Oncale). 305 Id. at 566. 306 Id. 370 24.3 COLUMBIA JOURNAL OF GENDER AND LAW Jeff Dawe, "a former Marine of imposing stature," who referred to H. as "queer" and "fag." telling him to "go back to San Francisco with the rest of the queers"' and asking him if he were "a boy or a girl."o3 As the court recounts, Dawe soon took to calling H. his "bitch" and promising that he would take H. "out to the woods" and "get [him] up the ass."3 os Other coworkers joined in, encouraging Dawe to "get a piece of that young ass."' including H.'s supervisor, who once asked, in reference to Dawe's repeated rape threats, whether H. were "tight or loose"' and wondered "would he scream or what."o'0 Events came to a head one day when a coworker griped to Dawe that his "bitch" appeared grumpy and that Dawe should do something about it."' Somewhat inebriated after several lunchtime drinks. Dawe walked towards H.. saying, "'m going to finally find out if you are a girl or a guy."3 H. backed away to avoid Dawe. but found himself trapped against a wall. 312 Dawe then grabbed H. by the testicles and turned to his coworkers to announce. "Well, I guess he's a guy.""' H. testified that after this episode he became convinced that Dawe was willing and capable of sexually assaulting him in accordance with Dawe's repeated threats. 14 Both H. and J. quit their jobs and invented a story to hide what had happened. In many respects, H.'s abuse appears to fit comfortably within the rubric of gender normalization. The Seventh Circuit, in fact, reversed the district court's grant of summary judgment, finding that the brothers' abuse presented sufficient evidence for a Price Waterhouse sexual stereotyping claim.3 16 The court concluded that H. had been singled out for abuse because the way he projected the sexual aspect of his personality did not conform to his coworkers' ideas of appropriate masculine behavior.'1 In particular, H.'s abuse seems to have stemmed from the earring that he wore to work. which-accented by 307 Id. at 567. 308 Id. 309 Id. 310 Doe, 119 F.3d at 567. 311 Id. 312 Id. 313 Id. 314 Id. 315 Id. 316 Doe, 119 F.3d at 580-83. 317 Id. at 580. 24.3 371 COLUMBIA JOURNAL OF GENDER AND LAw his youth---deviated from the stereotypes of normative masculinity embodied by the work crew. Although H.'s harassment might have, at least in part, taken the homophobic form of gay-bashing, for the court, it was impossible and meaningless to distinguish whether H. was harassed because of his gender or some imagined sexual orientation. The crucial question, explained the court, was "whether H.'s gender would have been questioned for wearing an earring. . . if he were a woman rather than a man.""' To the Seventh Circuit, it was obvious that H.'s coworkers believed that "an earring is a feminine accoutrement not suitable for male adornment" and that these gender stereotypes were responsible for the abuse H. suffered.:" However straightforward the court's reasoning might be, such an abstract idea of gender norms ignores the specific social position of men like Dawe. It is easy, of course, to read Dawe as personifying the broader order of an aggressive and powerful masculinity, a former member of the military and obvious leader of the all-male city work crew. From this perspective, H. would be the natural foil to Dawe, both deviating from and defining the masculinity of the work crew. Yet casting Dawe as representative of an abstract male order neglects the social and historical contradictions that appear to drive the abuse. Dawe is not a generic white male; he is rather the once-proud Marine now revealed to be an uneducated and downwardly mobile leader of a group of similarly futureless men swept out of public sight and consigned to the upkeep of local graves. The domination of two sixteen-year- old boys appears as an attempt to shore up the imaginary lines of identification that link Dawe, however tenuously, to the broader structures of white, male heterosexual power in American society. Dawe bullies two teenage boys, then, not because of some hidden challenge they present to the heteromasculine order. but. quite the opposite, because of their extreme social vulnerability. Indeed, the fact that Dawe and the others are driven to tyrannize such helpless teenagers is a symptom of the underlying inability of these men to otherwise experience a sense of power they feel should be rightfully theirs. Gender norms also fail to explain the form and momentum of the harassment itself. Whereas gender normalization would explain H. 's abuse as a response to his nonconformity it is also relevant to inquire under what circumstances a teenager's earring must be answered by threats of rape. Initially, the work crew unimaginatively focused on the most convenient pretexts to harass each boy: H.'s earring and J.'s obesity. But once the abuse begins, the harassment of each teenager follows an independent track. J.'s harassment appears relatively inert, settling into stock insults concerning his weight. H.'s harassment, by contrast, shifts 318 Id. at 581-82. 319 Id. 372 24.3 COLUMBIA JOURNAL OF GENDER AND LAW from a convenient set of derogatory terms to an elaborate fantasy centered on Dawe and H. Not only does Dawe himself refer to H. as his "bitch," but other coworkers share in the fantasy, encouraging Dawe to "get a piece of that young ass"' and take H. "out into the woods." The abuse assumes a life of its own, far beyond any social logic of bullying, as it taps into a set of contradictory social and psychic needs that can be acted out in conjunction with the sexualities of its participants. As the collective sexual scenario gains an internal momentum, it is clear that the shared investment in H.'s subjection has gone far beyond any original response to his supposed lack of gender conformity. H. has rather become incorporated into an abusive narrative that lies beyond the control of any single member of the group, programmed according to a set of sexualized alternatives-whether H. is "tight or loose,"' "a guy or a girl,"' if he will "scream or what"-that say more about the masculinities of the work crew than that of the twins. In particular. the collective fantasy surrounding Dawe and H. uneasily shifts between two different systems of male sexuality. For his part, Dawe first appears as the swaggering, heterosexual male out for a little "fag bashing." Yet, as the abuse shifts from taunts to rape threats, Dawe. too. shifts from repressed homophobe to something like the predatory "wolf' bent on demonstrating his masculinity through an ability to "punk" younger and less powerful men. On the part of H., we witness a parallel transformation in status. H.'s harassment appears to have originally focused on his purported sexual orientation. For instance, Dawe and others quickly seize on the earring to brand the teenager a "fag" and a "queer" and to instruct him to "go back to San Francisco with the rest of the queers." However, once H. is placed in a passive and effeminized position through the active process of his harassment. he seems to drift into the role of the "bitch"' for the group's most hypermasculine member. The idiom thus shifts from sexual orientation to gender status and the adoption of active-passive roles in sexual practices. Under the latter discourse, Dawe begins to threaten to take H. out into the woods in order to "get [him] in the ass," while others speculate to Dawe if H. would be "tight or loose."' Within the collective fantasy that underwrites the group's harassment of the boys, H. is no longer distinguished by his exclusive attraction to other gay men, but by the receptive sexual position lie assumes towards "real" men such as Dawe. In short. H. has gone-in the group's vernacular- from "fag" to "bitch."' And whereas, before, H. was utterly excluded from the work crew's heterosexual masculinity and imaginatively exiled to "San Francisco,"' he now can be reincorporated into the group's male hierarchy as Dawe's feminized partner and potentially dragged off into the primitive psychic space known as "the woods." What starts as the group's homophobia ends in H.'s sexual subjection, pinned against a cemetery wall with Dawe's hand between his legs. 24.3 373 COLUMBIA JOURNAL OF GENDER AND LAw V. Coda: The Sex-Based Hostile Environment Claim Not only does Doe usefully illustrate the limitations of a gender normalization approach, but it also points the way forward to an alternate legal framework for the sexual harassment claim. For, in addition to evidence of sex stereotyping, the Seventh Circuit accepted the explicit sexual character of the harassment itself as an alternate basis on which to find that discrimination occurred "because of' the plaintiff's sex.320 Revising Doe slightly, I would contend that this latter evidentiary route, focused on the sex-specific cultural materials mobilized as part of workplace discrimination, offers a more satisfying basis for the hostile environment claim. Therefore. before concluding, I want to briefly discuss how the legal framework of the sexual harassment claim should be reoriented around the sex-specific practices of workplace discrimination. As we have seen. feminist accounts focused on gender norms have tended to define sexual harassment. not in terms of the form of harassment. but in terms of its systemic effects in reproducing gender norms and sex-segregated workplaces. But defining sexual harassment in terms of such effects fails to recognize a host of sex-based conduct that cannot be readily assimilated to an abstract normative gender structure. Cases of horseplay, in particular, remain illegible from this perspective. In this Article, I have argued for adopting a performative approach built upon the understanding that individual enactments of sex actively appropriate, inhabit, and transform the normative meaning of gender. Sexual harassment should be understood to contain a heterogeneous cross-section of sex-based dynamics that do not necessarily support a single structure of power ("heteropatriarchy") or a single means of enacting sex-based power (sexual objectification or the disciplining of gender nonconformists). Reversing the analysis, then, we must focus on the diverse forms that sex-based harassment may take in order to render Title VII capable of intervening within an equally varied set of situations that come to sustain and renew the regulatory field of gender. Instead of conceiving sexual harassment as a series of yes-or-no questions comparing practices to monolithic normative structures-Does this conduct reinforce heteropatriarchy? Do these acts affirm the association of masculinity and skill? Does this conduct perpetuate masculine norms and male control?--we must do away with the notion of a single, unified norm and, instead, discern how sex-specific practices accrue as a more fluid, uneven, and contradictory set of positions, roles, and relationships that take shape between individuals, both inside and outside the workplace. 320 Id. at 580. 321 As I noted above, this legal framework will be developed at length in a separate article. 374 24.3 COLUMBIA JOURNAL OF GENDER AND LAw Accordingly, to rethink the sexual harassment claim. we must conceptualize the causality requirement, in particular, without reference either to the "purpose or effect" of that conduct (as some legal scholars have done) or to the harasser's subjective motives or desires (as some courts have done). Rather, causality should be derived from the objective nature of the hostile environment itself, whether it takes the form of sex- neutral unequal treatment or sex-specific physical or verbal acts. A critical resource for that reconceptualization exists in the form of the race-based hostile environment claim. Arising out of Rogers v. EEOC in 1971.322 this line of cases developed slightly earlier than sexual harassment and, notably. without the significant cultural confusion introduced by the issue of sex. Importantly, these race-based claims led courts to conceptualize the hostile environment claim as a distinct paradigm of discrimination, separate from either disparate treatment or disparate impact. In contrast to the pretextual job decision at the center of the McDonnell Douglas/Burdine framework "-neutral in appearance, but motivated by the defendant's impermissible consideration of race, gender, or other prohibited category- the race-based hostile environment typically did not involve neutral conduct nor did it require plaintiffs to prove, directly or indirectly, the motives behind the conduct. Rather, the hostile environment enacted, directly through its objective content and meaning, the discriminatory basis prohibited under Title VII. The causal basis of racist graffiti, for instance, adhered to the objective historical meaning of race-specific terms rather than to the particular subjective reasons that it was introduced into the workplace. Consequently, the prior focus on the defendant in the disparate treatment claim shifted to a new focus on the plaintiff as the lens through which the hostile environment was constructed. For, as Harris instructed, the hostile environment was not a simple empirical fact, but rather a theoretical totality. 324 capable of aggregating disparate materials from various actors and irreducible to the sum of its parts. In place of the defendant and his or her hidden animus, then, the hostile environment claim became focused on the plaintiff as the active field of identification, capable of drawing together and reunifying spatially and temporally dislocated forms of conduct. Drawing on this renewed theoretical understanding, workplace sex discrimination can be thought to operate according to two different paradigms of discrimination: on the one hand, the model of disparate treatment, in which sex-neutral workplace practices are directed towards certain employees based on their gender, race, religion, or other impermissible 322 454 F.2d 234 (5th Cir. 1971). 323 McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973); Tex. Dep't of Cnty. Affairs v. Burdine, 450 U.S. 248 (1981). 324 Harris v. Forklift Sys., Inc., 510 U.S. 17, 23 (1993). 24.3 375 COLUMBIA JOURNAL OF GENDER AND LAw basis; and, on the other hand, the paradigm of sex-specific practices, in which particular employees are exposed to gender-specific materials that may not be specifically directed towards them at all, but which nevertheless negatively affect them on the basis of their sex. 25 From this perspective, the sex-based hostile environment claim no longer focuses on the discriminatory intent behind sexual harassment and breaks from any analysis of the "why" of harassment, whether as the specific psychological motive behind the conduct or the underlying purpose or effect it might manifest in reinforcing gender norms. From this perspective, horseplay might be paradigmatic of the sex-based hostile environment- estranged from the normative onshore world and surprising even to the perpetrators themselves, yet enacting what the Seventh Circuit in Doe described as the gender-specific experience of sexualized harassment." 6 Of course, the hostile environment should not be limited to instances of sexualized abuse; indeed, horseplay permits us to see sexualized conduct as merely one culturally available form of conduct that might be mobilized in the construction of the hostile environment. Horseplay thus "desexualizes" sexual conduct, revealing it as a sedimented cultural script susceptible to mobilization alongside, and in no wav different from, other forms of discriminatory conduct, including sex-neutral disparities in treatment. misogynistic language, anti-female sexually explicit representations. and direct sexualized abuse. Thus, whereas a theoretical focus on horseplay might initially appear to shift our focus from women's historical subordination, it actually provides a model upon which to reimagine the harassment of women without the overarching presumption of male dominance or gender hierarchy, problematic appeals to the social logic of normalization 325 As Vicki Schultz has cautioned, understanding sex discrimination as operating through two paradigms of discrimination does not mean that a plaintiff's claim should be divided into separate disparate treatment and hostile environment claims. For one of the distinguishing features of the hostile environment claim is that, in contrast to the disparate treatment claim which centers on a tangible employment action such as hiring or firing an employee, the hostile enviromment claim aggregates a variety of workplace conduct to construct an overall environment. By necessity, then, the hostile environment claim cannot revolve around a single defendant's impermissible motives, but must center on the plaintiff's experience of the hostile enviromment as both subjectively and objectively hostile. It is therefore within the context of the overall hostile enviromnent that sex-neutral conduct must be evaluated. Separating sex-neutral and sex-specific conduct directly violates the dictates of Harris to evaluate the hostile environment according to "all of the circumstances." See Harris v. Forklift Sys., Inc., 517 U.S. 17, 23 (1993). For exemplar) analyses of the hostile enviromment that combine content- specific and content-neutral conduct, see Jackson v. Quanex Corp., 191 F.3d 647, 659-62 (6th Cir. 1999) (stating that Harris' directive to consider "all of the circumstances" entails evaluating race-neutral acts of intimidation and ridicule alongside race-specific ones); Andrews v. City of Philadelphia, 895 F.2d 1469, 1486 (3d Cir. 1990) (instructing the trial judge to consider on remand "all of the incidents to see if they produce a work enviromment hostile and offensive to women of reasonable sensibilities," ranging from evidence of pornography and sexual objects displayed on desks to the disappearance of plaintiffs' work files, anonymous phone calls, and the destruction of their property). 326 119 F.3d 563, 578. 376 24.3 24.3 COLUMBIA JOURNAL OF GENDER AND LAW 377 (and the attending opposition of the harasser's idealized masculinity and the victim's stigmatized deviance), or the assumption of a unique female vulnerability to sexual injury. The often inscrutable character of horseplay, in effect. cleanses the hostile environment claim of the unnecessary and often unproductive distinctions that have attached themselves to the claim-whether the plaintiff or defendant is gay or straight, whether the victim visibly departs from gender norms, or whether the conduct reflects a desire for sex or a desire to humiliate and demean. Instead, horseplay helps return the claim to its original and proper focus: the objective features of the workplace environment and the multiple ways that the environment may discriminate based on the plaintiff's sex.