This case is before us for the second time. Our affirmance of the district court's decision that Sweeney's promotion to Professor of Education at Keene State College was delayed because of her sex, Sweeney v. Board of Trustees of Keene State College, 569 F.2d 169 (1st Cir., 1978), was vacated and remanded by the Supreme Court "for reconsideration in the light of Furnco (Construction Corp. v. Waters, 438 U.S. 567, 98 S. Ct. 2943, 57 L. Ed. 2d 957 (1978))." 439 U.S. 24, 25, 99 S. Ct. 295, 296, 58 L. Ed. 2d 216 (1978). We in turn remanded to the district court, which again found in Sweeney's favor. No. 75-182 (D.N.H. Jan. 29, 1979). Keene State College once again appeals.
From the beginning, Sweeney has sought to prove her claim of sex discrimination by the methodology in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S. Ct. 1817, 36 L. Ed. 2d 668 (1973). Under that case, an individual Title VII plaintiff may proceed by first establishing a "prima facie case" of discrimination; this then requires the defendant to "articulate" a legitimate, nondiscriminatory reason for its adverse action regarding the plaintiff. To prevail, the plaintiff ultimately must prove that the reason given is a pretext for discrimination. See 411 U.S. at 802-05, 93 S. Ct. 1817. Since the Supreme Court vacated our first Sweeney Decision, we have taken pains to point out that, under McDonnell Douglas, the defendant's burden is merely a burden of production, and that the burden of persuasion remains at all times with the plaintiff. Loeb v. Textron, 600 F.2d 1003 at 1011-1012 (1979).
The error that prompted the Supreme Court to vacate our original decision occurred in our discussion of defendants' obligation to "articulate" a legitimate reason for Sweeney's non-promotion once plaintiff had established a prima facie case. We stated erroneously that defendants were required "to prove absence of discriminatory motive." 569 F.2d at 177. In remanding the case to us, the Supreme Court reemphasized the actual language and rule of McDonnell Douglas, 411 U.S. at 802, 93 S. Ct. 1817, and Furnco, 438 U.S. at 578, 98 S. Ct. 2943, that a Title VII defendant need only "articulate" a valid reason, and indicated that defendants surely had done so. See 439 U.S. at 25 n.2, 99 S. Ct. 295. The Court was concerned that we had "imposed a heavier burden on the employer than Furnco warrants." at 25, 99 S. Ct. at 295-296.
On further remand from us, the district court manifested its understanding that defendants had met their limited burden of articulating facially valid reasons for not promoting Sweeney, and concentrated upon the ultimate question: whether Sweeney had proven by a preponderance that the reasons stated were pretexts for discrimination. The court concluded that Sweeney had met her burden in this regard:
"(Sweeney) proved to my satisfaction that the basic reason for the failure to promote her was because of her sex, that the reasons advanced by the defendants were pretextual, and that plaintiff would have been promoted in the academic year 1974-75 but for the fact that she was a woman."
See Fisher v. Flynn, 598 F.2d 663, 665 (1st Cir. 1979) (Title VII plaintiff must meet "but for" standard of proof); Cf. Loeb, 600 F.2d at 1019-1020 (same rule in ADEA cases).1
The issue now before us is whether the district court's decision in favor of Sweeney is clearly erroneous.2 Because of the procedural history of this case and the parties' disagreement over the issues before us,3 we have reviewed the record a second time in light of our current understanding of the law. We conclude that the district court's decision was not clearly erroneous and therefore affirm.
Sweeney initiated the promotion procedure in the fall of 1974; in November Dr. St. John, then Chair of the Education Department, wrote to Dean Davis that Sweeney wished to be considered for full professor and had the support of the department's Advisory Committee on Promotions, although he personally had mixed feelings about her case.4 In any event, she was considered by the 1974-75 Faculty Evaluations Advisory Committee (FEAC), which recommended against her promotion.5 Dean Davis then wrote to Sweeney that she would not be promoted and gave her the "pro forma" explanation that she had,
"not fulfilled the qualifications as stated in the Faculty Manual; namely, that your teaching and research has not been "marked by the perspective of maturity and experience, or by some creative attribute generally recognizable in the academic world as a special asset to a faculty.' "
In November 1975, however, after the Faculty Appeals Committee (FAC) had urged that Sweeney be given more specific reasons for the adverse decision, See 569 F.2d at 173, President Redfern conferred with Dean Davis and with former FEAC members and then met with Sweeney. The evidence shows that he told Sweeney that the reasons were largely personal ones: that the FEAC members thought that she "personalized professional matters," was rigid, narrow-minded, and inflexible, intolerant of students' views and "old fashioned" in her supervision of student teaching. Her alleged concern with the height of window shades was cited as an example. Redfern also said that her minutes of the graduate faculty meetings were thought not to be of professional caliber, and that she did not show a "give and take" spirit on committees. These reasons were brought out at trial, where they were supplemented by the testimony of Dr. Quirk, who was Chairman of the 1974-75 FEAC.
Dr. Quirk testified that Sweeney's case for promotion was "weak" and "mediocre . . . at best." The 1974-75 FEAC, which consisted of five men, considered five candidates for promotion to full professorships three men and two women. Only two were recommended: one man, James Smart (vote 3-2), and one woman, Janet Grayson (vote 5-0). The vote against Sweeney was five to zero. According to Quirk, the reasons for the vote were "varied."
"There were quite a few reasons. But it was probably just an extremely weak case. I think one has to look at the criteria involved in promotion, and when (sic) one has to keep in mind the fact that that is promotion to the top rank of the college, that is, to full professor, and when you view it that way, the case has to be a positive, a strong positive case for recommendation for promotion to this top rank. If you look at the categories involved, there are basically three categories in which we make a judgment. One of the categories is teaching effectiveness. The other is contribution to the college. And the third is scholarly qualifications."
Defense counsel on direct examination brought out that Sweeney had served on no campus-wide committees except the College Senate, to which she was elected by her department rather than on a campus-wide basis. Quirk also testified that he had served with Sweeney on the Admissions and Standards Committee in 1974-75 and that their relationship had been marked by "some disagreements" over reviewing students' applications to the Education Department for professional education. Quirk felt that Sweeney's advocacy of a "subjective interview" as a requirement for admission, allegedly without any criteria, and her statement, in response to his question, that convicts had "no place in front of the classroom" were examples of her "lack of maturity."6
The reasons given for the 1974-75 denial of Sweeney's promotion thus were, in essence, that Sweeney had a tendency to be narrow-minded and rigid, to personalize professional matters, and to be difficult to work with. Defendants did not state that Sweeney was lacking in scholarly qualifications, but suggested that she had made an insufficient contribution to the college, for example, to its committees, and that her personality interfered with her teaching and colleagueship.7
Sweeney applied for promotion again to the 1975-76 FEAC and was successful. In the interim she had filed charges of sex discrimination, and at trial she expressed the view that the 1975-76 promotion was in response to that action, as "(everything) else remained constant." Defendants emphasize, however, that in September 1975 Sweeney was made Director of the Education Department's reading program and that, in her November 1975 meeting with Redfern concerning the 1974-75 denial, Redfern said that her performance in that program might lead to her promotion. Defendants' position is that Sweeney had much stronger departmental support in 1975-76 than in 1974-75, in part because of her work in the reading program.
Defendants now contend, in essence, that Sweeney did not introduce evidence sufficient to prove that these reasons which on their face are legitimate and nondiscriminatory were pretexts for discrimination. Reminding us of the Supreme Court's admonition in Furnco, that proof of a prima facie case is not equivalent to a factual finding of discrimination, 438 U.S. at 576, 579, 98 S. Ct. 2943, defendants argue that Sweeney did no more than present a "generalized inference of discrimination," through statistics showing an imbalance of male faculty and the like, and that she failed to disprove specifically the reasons given for her non-promotion during 1974-75. While the case is close, we disagree: the record contains evidence sufficient to support a finding that the reasons advanced were not the real reasons for Sweeney's non-promotion in the year in question.
Contrary to the reasons that allegedly prompted the 1974-75 FEAC not to recommend her, Sweeney introduced significant evidence that she had worked well with a variety of people in a variety of roles and contexts, and that her personality did not impede her effectiveness as a teacher or as a member of the faculty. The College itself had granted her tenure in 1972, indicating that whatever attitudinal problems Sweeney had8 were not an obstacle to her becoming a permanent member of the faculty.9 In addition, several witnesses who were qualified to judge testified as to Sweeney's flexibility and skills. At least four persons who knew her well and had had occasion to work closely with her in group endeavors testified that she was open to new ideas, that she was not rigid, inflexible or intolerant, and that she did not personalize professional matters. Although her only college-wide committee membership was as departmental representative to the College Senate, it was brought out that by 1974 she had served on the Senate's Admissions and Standards Subcommittee, as secretary to her own department, as a member of her department's curriculum committee, and as faculty advisor to two organizations. She had been active in supervising student teaching and had been on the New Hampshire Board of Education's Professional Standards Board, for which she had served as chair of the Subcommittee on Appeals and as a member of the Subcommittee on Provisional Certification. A number of the witnesses who testified on her behalf had worked successfully with her on one or the other of these committees.
There also was testimony that Sweeney was at least as qualified as others who had been promoted to full professorship, and testimony as to a general perception of sex bias at Keene State. Various witnesses compared Sweeney's qualifications to those of other members of the faculty, both male and female, and indicated that, although she was not one of Keene State's "superstars," she did rank among the better members of the faculty.
The documentary evidence shows that when, in the next year, the 1975-76 FEAC recommended Sweeney for promotion, it did not, as defendants would have it, give special weight to her contributions to the reading program. Both Alfred Thomas, Chair of the Education Department that year, and FEAC mentioned this contribution as only one of many reasons for promotion; her personal attributes and qualifications in 1975-76 seem little different from those in 1974-75.10 The Dean, moreover, made no reference at all to the reading program, but rather spoke generally of Sweeney's good teaching evaluations and of her "good record of service to her department and to the College," and noted that she had served on the New Hampshire Professional Standards Board a position that she had held since 1970. See 569 F.2d at 178 n.18.
Defendants argue that Sweeney did no more than show that differences of opinion existed between members of the faculty at Keene and that she did not show that the 1974-75 FEAC acted out of sex bias. We fully agree that the issue is not whether Sweeney was qualified for promotion or should have been promoted in 1974-75 by some objective measure, but whether she was denied a promotion because of her sex. Loeb, 600 F.2d at 1014. The recommendation of the 1974-75 FEAC is entitled to stand even if it appears to have been misguided, unless it was sex biased. Loeb, 600 F.2d at 1012 n.6, 1014.
While Keene State's faculty members were entitled to hold different opinions as to Sweeney's qualifications, the evidence and testimony just reviewed suggests that more than just differences of opinion were involved. The defendants' alleged reasons border on describing Sweeney as, to quote plaintiff's brief, a "school marm." The focus on her alleged attention to the height of window shades in particular seems a trivial comment. In light of the evidence that Sweeney's personality was not as described by Redfern and did not interfere with her ability to work on committees or with people, the district court could have concluded that the five male members of FEAC would not have fastened upon such reasons had Sweeney been a man.
The nature of the reasons given, and the evidence introduced to show that they were either insubstantial or fictitious, stood with more general evidence suggesting that women at Keene State were evaluated by a stricter standard than their male colleagues, and that the institution generally was unresponsive to the concerns of its female faculty. Much of this evidence such as the statistical composition of the faculty11 and the attitude of the affirmative action officer is recounted in our original opinion, 569 F.2d at 178-79. While by itself it does not prove that Sweeney in particular was a victim of discrimination, it does add "color" to the decision-making process at Keene State and to the reasons given for Sweeney's non-promotion. Proof of a general atmosphere of discrimination is not the equivalent of proof of discrimination against an individual, but evidence of such an atmosphere may be considered along with any other evidence bearing on motive in deciding whether a Title VII plaintiff has met her burden of showing that the defendants' reasons are pretexts. See Furnco, 438 U.S. at 580, 98 S. Ct. 2943; Loeb, 600 F.2d at 1015 n.14. We think that it was open to the court to conclude from the totality of the evidence that the reasons given for Sweeney's non-promotion in 1974-75 were implicitly influenced by the fact that Sweeney was a woman.
Defendants emphasize that the promotion system at Keene relies on peer group support, and that Sweeney lacked the support of Dr. St. John, the Education Department Chairman,12 as well as of FEAC in 1974-75. They argue that there was no evidence that Sweeney did not have their support because she was a woman. We disagree. Although there was no direct evidence, we think that the district court could have inferred that FEAC and Dr. St. John were sex biased in light of the nature and weakness of the reasons given for her non-promotion coupled with the evidence of the statistical composition and general character of the institution and of the insensitivity of many including St. John to the concerns of the female faculty.13
Defendants make much of the fact that the 1974-75 FEAC recommended one man and one woman for full professorship and did not recommend two men in addition to Sweeney. They also call attention to the fact that the 1974-75 FEAC considered the applications of twenty-five men and six women for tenure and/or promotion, and acted favorably toward all women except Sweeney, but toward only twelve of the twenty-five men. This information obviously bears on the question of discrimination, but it does not render the district court's conclusion clearly erroneous. Of the five women promoted by the 1974-75 FEAC, only one, Janet Grayson, was promoted to full professor; a defendant's willingness to appoint women to lower ranks does not preclude a finding that a woman who sought promotion to a higher rank was discriminated against. Nor does the promotion of one woman of admittedly outstanding credentials to full professor and the denial of such a promotion to two men preclude a finding that another woman was denied the same position because of sex. One familiar aspect of sex discrimination is the practice, whether conscious or unconscious, of subjecting women to higher standards of evaluation than are applied to their male counterparts. The district court could have concluded consistently that Grayson merited promotion by any standard, that Sweeney was better qualified than the two men who were denied promotion, and that Sweeney would have been promoted had she been evaluated against the standard that was applied generally to men.
Defendants have persuaded us that this was a close case, but not that the district court committed clear error in concluding that Sweeney was denied a promotion because of her sex.
The judgment of the district court is affirmed.
* Of the First Circuit, sitting by designation.
** Of the District of Massachusetts, sitting by designation.
1. Although the district court acted before our decision in Loeb v. Textron, 600 F.2d 1003 (1979), was released, nothing in its opinion suggests that it applied a legal standard inconsistent with Loeb or, more importantly, with Furnco Construction Corp. v. Waters, 438 U.S. 567, 98 S. Ct. 2943, 57 L. Ed. 2d 957 (1978).
2. We indicated that we followed the clearly erroneous standard of Fed.R.Civ.P. 52(a) in our first Sweeney decision, Sweeney v. Board of Trustees of Keene State College, 569 F.2d 169 (1978). Defendants now urge us to abandon that standard in Title VII cases on the ground that a "factual" finding that a plaintiff was denied a promotion because of her sex is equivalent to a finding on the ultimate legal issue of discrimination. This argument has persuaded some circuits that appellate courts should make an independent determination of the question of discrimination and apply the clearly erroneous standard only to the district court's findings of subsidiary facts. E. g., Stewart v. General Motors Corp., 542 F.2d 445, 449 (7th Cir. 1976), Cert. denied, 433 U.S. 919, 97 S. Ct. 2995, 53 L. Ed. 2d 1105 (1977); Causey v. Ford Motor Co., 516 F.2d 416, 420-21 (5th Cir. 1975). We are not inclined to that approach. This circuit has applied the clearly erroneous standard to conclusions involving mixed questions of law and fact except where there is some indication that the court misconceived the legal standards. E. g., Burgess v. M/V Tamano, 564 F.2d 964, 976-77 (1st Cir. 1977), Cert. denied, 435 U.S. 941, 98 S. Ct. 1520, 55 L. Ed. 2d 537 (1978) (admiralty negligence); Raymond v. Eli Lilly & Co., 556 F.2d 628, 629-30 (1st Cir. 1977) (finding of reasonable diligence) (per curiam); Forbro Design Corp. v. Raytheon Co., 532 F.2d 758, 763 (1st Cir. 1976) (finding of "obviousness"); Cf. Senter v. General Motors Corp., 532 F.2d 511, 526 (6th Cir.), Cert. denied, 429 U.S. 870, 97 S. Ct. 182, 50 L. Ed. 2d 150 (1976) (court not bound by "clearly erroneous" principle where party contends wrong legal principle applied). We see no reason to depart from this course in discrimination cases; the opportunity for firsthand observation may be especially important in one such as this, where the issue is whether "personality" reasons were sexually biased. We shall look carefully, however, to detect infection from legal error, and of course the clearly erroneous standard does not shield findings that are unsupported or arbitrary. See generally United States v. United States Gypsum Co., 333 U.S. 364, 394-95, 68 S. Ct. 525, 92 L. Ed. 746 (1948); 9 C. Wright & A. Miller, Federal Practice & Procedure §§ 2589-2591 (1971).
3. Defendants ask us to reconsider the entire case and phrase the issue before us expansively. They ask what evidence is necessary to sustain a finding of discrimination when employment decisions are made through a system of peer group review, and to what extent evidence used to establish a prima facie case or general inference of discrimination can be relied upon to conclude that a defendant's action as to an individual plaintiff was discriminatory. Sweeney would have us construe the Supreme Court's remand order as no more than a request for clarification of the standard used in our first decision, and not for complete reconsideration of the case. She argues that the opinions below demonstrate that the district court and this court in its original opinion applied the proper legal standards, despite our misstatement as to the defendants' burden, and that we therefore need not review the evidence again. Alternatively, she argues that the evidence demonstrates "unequivocally" that the reasons for non-promotion were pretexts.
4. See note 12, Infra.
5. Keene State follows a promotion procedure under which candidates apply through their departments to the Faculty Evaluations Advisory Committee (FEAC). The departmental chairman forwards the application to FEAC, which makes a recommendation to the Dean. The Dean usually adopts FEAC's recommendation and, if it is positive, forwards it to the Trustees for final approval. See 569 F.2d at 172-73.
6. On cross-examination it was brought out that Quirk had had other differences with Sweeney because he had signed course registration cards for his wife, an education student who officially was Sweeney's advisee.
7. The reasons were summarized well by President Redfern. Called by Sweeney, he agreed during direct examination that the reasons were "subjective" and "judgmental," that they had no relationship to whether she had a terminal degree or to her scholarly research, that "in terms of the Faculty Manual criteria these might be more related to the process of maturity" and that they essentially dealt with Sweeney's personality "as reflected in the specified types of professional activities, such as committee work."
8. Dr. Blacketor, Chair of the Education Department in 1972, in recommending Sweeney for tenure did acknowledge some problems in "her personal and professional attitude" when she first joined the faculty, but stated that she had shown significant improvement in 1970-71 and 1971-72.
9. Keene State considers the tenure decision the most important in the careers of its faculty members, because it effectively confers lifetime employment.
10. In its memorandum recommending Sweeney's promotion to the Dean, FEAC said, "The committee has reviewed the promotion material presented in (sic) behalf of Miss Sweeney and was impress-d (sic) with her ability as a member of the instruction staff, indicated by her teaching index of 3.99 (out of 5). Her promotion carries the full endorsement of the Education Department F.E.A.C. She has been a member of the College Senate, has served on several college committees and has, in addition, been a member of the Professional Standards Board for the State of New Hampshire, serving as Chairman of the Professional Standards Appeal Board for four terms. "Miss Sweeney has published and is a willing contributor to faculty and students interested in research. Her professional activities include membership in several national and regional associations with participation in convention programs. She has also contributed to the College by successfully submitting a grant for the Right to Read Program. She has also been recognized for her professional competency by being selected as a member on accreditation teams at the state, regional and national levels." Aside from the reference to the Right to Read grant, substantially the same statement could have been made of Sweeney in 1974-75.
11. The statistical evidence was relevant, although it could not be conclusive of discrimination against Sweeney. The absence of women in the upper ranks at Keene State was not projected back in time and tied to their availability for appointment to the lower ranks in the different disciplines in earlier years. Moreover, the existence of a statistical disparity, while often helpful in establishing a prima facie case of discrimination, does not by itself meet an individual's burden of proving that the reasons given by the employer were pretexts, Furnco, 438 U.S. at 579-80, 98 S. Ct. 2943. Nevertheless, the statistics here were striking enough and covered a sufficiently long period of time (I. e., the usual time in rank for the various ranks) as to constitute some evidence of bias. See McDonnell Douglas, 411 U.S. at 805, 93 S. Ct. at 1825 ("statistics . . . may be helpful to a determination of whether petitioner's refusal to rehire respondent . . . conformed to a general pattern of discrimination against blacks"); Furnco, 438 U.S. at 580, 98 S. Ct. at 2951 (composition of workforce "is not wholly irrelevant on the issue of intent").
12. When Dr. St. John wrote to Dean Davis stating that Sweeney had asked to be promoted to full professor and that the Education Department Advisory Committee on Promotion had recommended her, he also stated that he had "completely ambivalent feelings" about Sweeney's application, listed six "pros" and six "cons"," declined to recommend personally for or against, noted that the Advisory Committee had been unable to obtain Sweeney's personnel file from the Dean, and stated that she was "entitled to a fair and impartial judgment." Defendants state repeatedly in their brief that the Education Department "did not consider" Sweeney in 1974-75, that Sweeney "did not have the positive support in 1974-75 of her department," and that in 1975-76, when she finally had the support of her department, department chairman and FEAC, she was promoted. We find this line of argument misleading. The record shows that applicants for promotion were considered by their department evaluation committees, which in turn made recommendations to the department chairmen, who forwarded the applications to FEAC. In both 1974-75, by Dr. St. John's own admission, and in 1975-76, Sweeney had the support of her department's evaluation committee. The significant difference was that the 1974-75 department chairman did not endorse the committee's recommendation, whereas the 1975-76 chairman did. The district court could have concluded that St. John undermined the committee's recommendation and, on the basis of the evidence reviewed herein, that his criticism of Sweeney was determined by a subtle, if unexpressed, bias against women faculty.
13. For example, Eleanor Vanderhagen testified that Dr. St. John was involved in the publication of a newsletter by the Education Department that carried an announcement about an all-male honor education fraternity. When Vanderhagen wrote to him "pointing out its role in professional advancement for careers for men and women" and tried to meet with him to discuss her feeling that this was "inappropriate for a college publication," St. John replied that he was "unavailable" and left a message that he "considered the whole thing trivial." There was also evidence, which the court below was entitled to credit, that St. John was condescending toward women and had been discourteous to Sweeney from the first time they met.