Ann B. Hopkins v. Price Waterhouse. Ann B. Hopkins v. Price Waterhouse

WILLIAMS, Circuit Judge,

dissenting:

The majority implicitly adopts a novel theory of liability under Title VII, but neither confronts the novelty of the theory nor gives it any intelligible bounds. Further, as it must to reach the result, it bends out of recognition this court’s holding in Toney v. Block, 705 F.2d 1364 (D.C.Cir.1983). These prodigies are necessary for the outcome because the district court’s judgment cannot be sustained under any hitherto accepted notion of Title VII liability. 1

The theory is one of sexual stereotyping. See, e.g., Majority Opinion (“Maj.”) at 465, 468, 469. An analysis grounding Title VII liability in such stereotypes may well be meritorious; but its articulation would require care. No one argues that Congress intended entirely to overturn Justice *474 Douglas’s observation that “the two sexes are not fungible.” Ballard v. United States, 329 U.S. 187, 193 , 67 S.Ct. 261, 264 , 91 L.Ed. 181 (1946). Dismissal of a male employee because he routinely appeared for work in skirts and dresses would surely reflect a form of sexual stereotyping, but it would not, merely on that account, support Title VII liability. Nor, I suppose, does anyone contend that use of the feminine pronoun “she” to describe a female is a forbidden “evaluatpon of] female candidates in terms of their sex.” Maj. at 468.

The court makes no effort to delineate the theory, to draw a line between permissible and impermissible. There is a good reason not to do so: the record here provided no causal connection between Hopkins’s fate and such stereotyping as went on among Price Waterhouse’s 662 partners. The evidence of sexual stereotyping 2 is carefully culled from a mass of critical comments on the plaintiff’s abrasiveness with no sex link whatever. The district court determined that these comments were well founded in fact, represented standards applied to men and women alike, and were the true basis of the firm’s decision. 618 F.Supp. at 1114-16. The questionable remarks consist, with one marginal exception, of two types. First, some of Hopkins’s supporters used such stereotypes in speaking of her or in voicing their speculations as to the workings of her opponents’ minds. Second, other partners had used such terms in other years in speaking of other female candidates. Thus, though some forms of sexual stereotyping can be discriminatory, the instances here, however they may be characterized, were at most “generalized discrimination within the employment unit,” Toney v. Block, 705 F.2d at 1367 , rather than discrimination “in the particular employment decision for which retroactive relief was sought," id. at 1366 (emphasis in original).

Under Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 252-53 , 101 S.Ct. 1089, 1093 , 67 L.Ed.2d 207 (1981), and Toney , this can do no more than establish a prima facie case of discrimination. In functional terms, it put upon the defendant the burden of showing that its stated reasons were not pretextual. Defendant met that burden. The district court made unchallenged findings that the reasons given were “not fabricated as a pretext for discrimination.” 618 F.Supp. at 1114. It also found that Price Waterhouse had “legitimate, nondiscriminatory reasons for distinguishing between the plaintiff and the male partners with whom she compares herself.” Id. at 1115. This clearly restored the burden to plaintiff to show that she was the victim of unlawful discrimination. Burdine, 450 U.S. at 256 , 101 S.Ct. at 1095 . The district court’s findings that Hopkins’s “conduct provided ample justification for the complaints [about her unpleasantness] that formed the basis of the Policy Board’s decision,” 618 F.Supp. at 1114, clearly shows that plaintiff did not meet that burden.

The district court summarized its view of the evidence of discrimination in these terms:

Discriminatory stereotyping of females was permitted to play a part. [1] Comments influenced by sex stereotypes were made by partners; [2] the firm’s evaluation process gave substantial weight to these comments; and [3] the partnership failed to address the conspicuous problem of stereotyping in partnership evaluations. [4] While these three factors might have been innocent alone, they combined to produce discrimination in the case of this plaintiff.

618 F.Supp. at 1120. I examine these elements in the same order.

1. Partner comments influenced by stereotypes. The bulk of the comments instanced as stereotyped are by Hopkins’s supporters. One said that opponents focused on Hopkins’s profanity “because its [sic] a lady using foul language,” another characterized her as “macho,” and another *475 said she had “matured from a tough-talking, somewhat masculine hard-nosed mgr. to an authoritative, formidable, but much more appealing lady partner candidate.” Id. at 1117. The majority evidently refers to these as “stereotypes ... brought to bear on [Hopkins’s] own candidacy,” Maj. at 469, but there is no reason to suppose they harmed it. 3 The psychological speculations of Hopkins’s boosters cannot by any stretch be “direct evidence that her gender was a significant motivating factor in her failure to make partner.” Id. at 470.

As for the “smoking gun” remark by her most ardent supporter, Thomas Beyer (“walk more femininely,” etc.), there is no reason to suppose that it represented any more than one partner’s speculations. The district court was clearly erroneous in characterizing the statement as having been made by Beyer in fulfillment of his “responsibility] for telling her what problems the Policy Board had identified with her candidacy.” 618 F.Supp. at 1117. Hopkins’s own testimony showed that it was Joseph Connor (Beyer’s superior) who bore the responsibility for informing Hopkins of the reasons for the decision and who did so. Tr. 87-97. Hopkins testified that Connor made no remarks about her sex. Id. at 95. After speaking with Connor she sought Beyer’s advice, along with that of several other partners. Id. at 98. Beyer had been on vacation, and he made no claim whatever to inside information on the discussions of the Policy Board: he and Hopkins “tried to guess who some of the [opposing partners] might be.” Id. at 89 (emphasis added). Neither her account of that conversation nor any part of her testimony contradicts the natural inference that his advice was just that: personal speculation as to possibly winning strategies. Id. at 102. Beyer’s testimony confirms this interpretation: Connor had said nothing to Beyer suggesting that Hopkins’s dress, walk, or any aspect of her personal appearance was a problem, but Beyer believed such a change might help. Id. at 168. He never articulated the basis for the belief.

The majority tries to shore up the misconception by imputing to Price Water-house an “artificial assumption that Beyer ... would be kept completely in the dark as to the Policy Board’s views on her candidacy.” Maj. at 466. No one assumes any such thing. The issue is whether Beyer was summarizing the Policy Board’s views or was offering his own helpful suggestions. The evidence of Hopkins and Beyer is clear that it was the latter. The only faint evidence the other way came from Roger Marcellin, a partner in another office who did field work for the Policy Board. Tr. at 305-07. He simply assumed (“ha[d] no doubt”) that Beyer would be reporting “where the problems were.” Id. at 316. Beyer’s and Hopkins's testimony on the subject makes clear that Marcellin’s guesswork was inaccurate.

In the majority’s most dramatic imaginative leap, the stereotyped language of Hopkins’s supporters is said, without a shred of supportive evidence, to have “len[t] credence to [stereotyped complaints of Hopkins’s critics] and unwittingly undermin[ed] the support they sought to provide.” Maj. at 466 n. 3. The creativity of the proposition is underscored by its building in an assumption that stereotyped critiques by Hopkins’s opponents exist — an assumption for which the majority identifies no record support.

The only remark by a Hopkins opponent that can be characterized as manifesting sexual stereotyping is the facetious suggestion that she should take a “course at charm school.” The smoke from this gun seems to me rather wispy. It was embedded in the following comment:

Contacts with Ann are only casual — several mtgs at OGS and MMGS sessions. However, she is consistently annoying and irritating — believes she knows more than anyone about anything, is not afraid to let the world know it. Suggest a course at charm school before she is con *476 sidered for admission. I would be embarrassed to introduce her as a ptnr.

Def.Exh. 27.

The substance of the remark has nothing to do with sex stereotypes. It fits with the many other characterizations of Hopkins (“too assertive, overly critical of others, impatient with her staff”; it required “diplomacy, patience and guts to work with her”; 618 F.Supp. at 1114) for which, the district court found, plaintiffs “conduct provided ample justification,” id. The objection, of course, is to the opponent’s silly phrase. The reference was doubtless sex-linked, and the majority is not unfair in characterizing it as a “somewhat derogatory colloquialism.” Maj. at 466 n. 4. Thus it may be more “sexist” than a comment, such as might be made of a young man, wanting in character, that he ought to be “sent to military school.” But to find discrimination by Price Waterhouse based simply on this remark is “utterly implausible.” See Bishopp v. District of Columbia, 788 F.2d 781, 786 (D.C.Cir.1986) (discussing criteria for disregarding district courts’ findings of fact in Title VII cases).

The district court and the majority take refuge in comments made by Price Water-house partners in evaluations of other women in other years. 618 F.Supp. at 1117; Maj. at 467-68. These included one plainly beyond the pale — a remark by a partner that he “could not consider any woman seriously as a partnership candidate and believed that women were not even capable of functioning as senior managers.” 618 F.Supp. at 1117. So we know that, at least at some time in the past, there was one male chauvinist pig rampant among the Price Waterhouse partners. But there is no evidence that this troglodyte ever influenced a single other partner. His comment was not repeated after 1981, perhaps because the informal atmosphere in the firm made such remarks unacceptable. In any event, no one claims he played any role in the relevant evaluation.

The other remarks (still relating to other evaluations in other years) are ambiguous. For instance, it had been said of one woman candidate that she acted too much like “one of the boys,” 618 F.Supp. at 1117, but this was apparently a criticism of her for socializing too much with the clerical staff and not enough with the professionals. Def.Exh. 64, tab 22. Even the majority recognizes that these remarks had no “direct impact on plaintiff’s candidacy.” Maj. at 468. But it takes them as evidence that partners at Price Waterhouse “often evaluate female candidates in terms of their sex.” Id.

In a case where alleged sexual stereotyping had a demonstrable connection to the plaintiff, a careful analysis of such remarks would be in order. Such an analysis would begin with the recognition that not all sex-based phrases are sexist. Our vocabulary is full of such phrases, some of which have gradually detached themselves from any genuine link to sex, or even switched sex. Thus “doll,” originally a slang phrase for a “conventionally pretty and shapely young woman, ... whose function is to elevate the status of a male and to inspire general lust,” see New Dictionary op American Slang 108 (R. Chapman ed. 1986), has come in some contexts to refer to any “notably decent, pleasant, generous person,” as in “Isn’t he a doll?” That is the way language evolves, especially in a lively, spontaneous culture such as ours. Words themselves are metaphors, and it is in their nature to acquire meanings completely detached from original, concrete detail, whether or not sex related. Thus the phrase “BS” clearly relies on no distinction between cows and bulls.

Here, the phrase “one of the boys” was used in a sex-neutral sense: it was used of a woman, and since it evidently referred to her camaraderie with clerical staff at Price Waterhouse, the statistical probability is overwhelming that they were predominantly women. The phrase’s connotation of easy familiarity (an “ordinary, amiable man ... without side or lofty dignity; = ORDINARY JOE: His Eminence was trying to be one of the boys,” id. at 305) easily escapes its masculine origins. The phrase does not manifest sexism, notwithstanding the solemn avowals of the plaintiff, the district court and the majority.

*477 But this case does necessitate a study of just what expressions Congress may have wished to wash from the American tongue. The remark related to another candidate in another year. It plainly was not “direct evidence that [Hopkins’s] gender was a significant motivating factor in her failure to make partner.” Maj. at 470.

In discussing sex stereotyping, the district court gave great weight to the testimony of Dr. Susan Fiske, a witness purporting to be an expert in that field. She claimed to be able to find forbidden stereotyping simply by reading partners’ comments — without information about the truth of the matters commented upon. Of course where the remarks themselves carry such a tint (if, for example, a commenter had said, “She’s too masculine”), anyone could do so. But (apart from the “charm school” remark) no Hopkins detractor said any such thing. Dr. Fiske’s expertise rose to the occasion. Her arts enabled her to detect sex stereotyping based largely on “the intensity of the negative reaction.” Tr. at 559. So if an observer characterized someone as “overbearing and arrogant and abrasive and running over people,” an expert such as Dr. Fiske could discern — and would, if the subject were a woman — that they stemmed from unconscious stereotypes. Dr. Fiske could do this without meeting the subject of the comment or making any inquiry into a possible factual basis. Id. at 569, 595-97. To an expert of Dr. Fiske’s qualifications, it seems plain that no woman could be overbearing, arrogant or abrasive: any observations to that effect would necessarily be discounted as the product of stereotyping. If analysis like this is to prevail in federal courts, no employer can base any adverse action as to a woman on such attributes.

2.The evaluation process gave weight to such comments. This generalization suffers precisely the defect of the first leg of the tripod of liability: it depends entirely upon comments that could not have adversely affected Hopkins. Either they related to other candidacies in other years, or they represented her supporters’ views or intuitions about her adversaries. All we have that connects in any potentially adverse way with Hopkins is the “charm school” remark. 4

3. Neglect of duty to address problem of stereotyping. Key to the district court’s finding of liability was Price Waterhouse’s failure to institute special programs for sensitizing partners to sex stereotyping, or otherwise to stamp it out of the evaluation process. 618 F.Supp. at 1120. This breaks new ground, blithely free of any effort to link it to any established legal principles. Nor is the new theory intelligibly defined. What set of facts triggers the duty? If such an omission is to ground liability, perhaps the plaintiff should bear an initial burden of demonstrating that gender stereotyping was more probably than not the cause of the adverse employment decision. The majority, like the district court, fails to clarify this important issue, perhaps because it is so clear that Hopkins failed to make such a showing.

From the facts here, it looks as though the duty to sensitize has a hair trigger. The implications are serious. The more delicate the trigger, the more completely this court has dropped the requirement of intentional discrimination out of the law. As few employers can say with confidence that those who run its hiring and promotion are one hundred percent free of what may later be characterized as forbidden stereotyping, the only safe course will be to institute programs of the sort approved by the district court. The rule turns Title VII from a prohibition of discriminatory conduct into an engine for rooting out sexist thoughts.

4. Innocent alone, the three factors combined to produce discrimination in the case of this plaintiff. Such alchemy is mysterious. Having found that specific complaints caused the Policy Board’s adverse decisión and that there was ample *478 justification for the complaints, the district court took up the allegations of stereotyping floating in the Price Waterhouse ether and the remarkable intuitions of Dr. Fiske. 618 F.Supp. at 1117-20. The court began on a cautious note — some negative comments on Hopkins “might be attributed to sex stereotyping.” Id. at 1118. It next determined that the commenters “may have been influenced by a sex bias.” Id. It then progressed from “might” to “did,” but never revealed how it reached the final ipse dixit.

The evidence here establishes at most the existence of sexist attitudes. Thus there can be no doubt that this court’s decision in Toney v. Block controls. The showing of “generalized discrimination” can at the most establish a prima facie case, requiring defendant to meet its burden of showing non-pretextual grounds for its action. The district court properly found those established, restoring the burden to plaintiff.

The majority would eviscerate Toney by a clever name change: calling the case one of mixed motive, the majority looks to precedents in related areas where a party acting with one permissible motive and one unlawful one may prevail only by affirmatively proving that it would have acted as it did even if the forbidden motive were absent; I have no quarrel with this principle. See National Labor Relations Board v. Transportation Management Corp., 462 U.S. 393, 403 , 103 S.Ct. 2469, 2475 , 76 L.Ed.2d 667 (1983); Mt. Healthy City School District Board of Education v. Doyle, 429 U.S. 274, 287 , 97 S.Ct. 568, 576 , 50 L.Ed.2d 471 (1977). But it has no relevance where, as here, discrimination has “not been specifically attributed to the employment decision of which the plaintiff complains.” Toney, 70S F.2d at 1366. 7b- ney does not permit a plaintiff to invoke the “mixed motive” concept whenever (1) he or she has shown only background evidence of some generalized discrimination and (2) defendant has proven that a non-pretextual reason “formed the basis” of the act. If this court is to deep-six Toney , it should do so en banc.

There is not enough evidence of intentional discrimination to support a verdict for Hopkins under any established approach to Title VII liability. The stereotype theory adopted by the district court should not be allowed to spring to life in a case where its occurrence is not plausibly related to the decision on plaintiff. If a court is to develop such a theory, it should do so in a context where it and the parties properly focus on what elements of sexual differentiation Congress may have sought to stamp out. If failure to provide sensitivity training is to be a ground of Title VII liability, there should be some illumination of the circumstances triggering the duty. And if Toney is to be overturned, it should not be by a panel of this court. I dissent.

. The majority’s treatment of the relief issues, however, seems correct.

. The line between legally permissible and legally impermissible stereotyping has yet to be drawn. When I use the term, I refer simply to whatever expressions have been so characterized by the district court or the majority.

. Cf. Maj. at 466 (“The comments of Hopkins’ supporters may or may not have harmed her candidacy....’’)

. If this leg is in any way based on the firm’s procedure of giving substantial weight to “no” votes, it is inconsistent with the district court's prior finding that "the firm's practice of giving ‘no’ votes great weight treated male and female candidates in the same way.” 618 F.Supp. at 1116. See Mitchell v. Baldridge, 759 F.2d 80 , 85 n. 3 (D.C.Cir.1985).