Full opinion text
MURNAGHAN, Circuit Judge: When the Department of Religion at the University of North Carolina declined to promote or to reappoint Assistant Professor Mary Carroll Smith, she, after unsuccessfully pursuing administrative avenues of relief, filed an action in the United States District Court for the Middle District of North Carolina against the University and several faculty members. Alleging that various University members had violated her federal and state constitutional and statutory rights when they decided not to renew her teaching contract, she sought declaratory and injunctive relief, damages, and either reappointment or promotion with tenure. While numerous issues were raised and decided by the district court, the only remaining issues of concern to us are whether the University discriminated against Smith because of age, in violation of the Age Discrimination in Employment Act (the “ADEA”), and whether the University discriminated against her because of sex and religion, in violation of Title VII of the Civil Rights Act (Title VII). The age discrimination issue was tried before a jury which rendered a verdict in favor of the University. Earlier motions by both parties for summary judgment as to the age discrimination claim were denied by the court as was Smith’s later motion to enter a judgment notwithstanding the verdict. The sex and religious discrimination charges were tried to the court. It too found for the University. The final results in the district court, on the merits of each issue raised, therefore have amounted to total victory for the University, total defeat for the plaintiff. As an interim measure, pending a decision on the merits, this Court had reversed the district court’s refusal to enter a temporary restraining order and preliminary injunction. We ordered entry of an injunction to continue in effect “until a trial of the case is had and a decision entered upon the merits.” Under the injunction, the University was required (a) to continue the plaintiff’s employment; (b) to refrain from hiring a replacement for the plaintiff; (c) to maintain in effect the plaintiff’s teaching position; (d) to pay the plaintiff as theretofore; and (e) to refrain from disseminating information injurious to the professional or personal reputation of the plaintiff. Although she failed in the end to prevail on the merits on any of her claims, the court, pursuant to 42 U.S.C. § 2000e-5(k), awarded Smith $8,300 in attorneys’ fees and $302.06 in related litigation expenses which represented the costs incurred in obtaining the preliminary injunction. Smith now seeks our review of the proceedings below, contending that: (1) the court’s instructions to the jury were erroneous with respect to the standard applicable to age discrimination claims and to her burden of proof; (2) the court erred when it refused to grant her motions for summary judgment and judgment notwithstanding the verdict on the age discrimination claim; and (3) the court’s findings regarding the sex and religious discrimination claims were clearly erroneous. Dissatisfied by the award of attorneys’ fees, the University filed a cross-appeal in which it argues that Smith, although successful in obtaining a preliminary injunction, was not a “prevailing party” within the meaning of the pertinent statutory provision and, therefore, is ineligible for attorneys’ fees. Factual Background In early 1973, Mary Carroll Smith was 38 and a recent recipient of a Ph.D. from Harvard University in Sanskrit and Indian Studies. Her predominant area of scholarly concentration, interest, and research was the computer study of Sanskrit texts. Pri- or to receiving her doctorate,. Smith had earned Masters degrees in English and Sanskrit and, for seven years, had taught, first, at a small college and, then, at a community college. In addition to her formal academic religious education, Smith, for ten years, had been a member of a Roman Catholic Sisterhood. During early 1973, the Religion Department at the University of North Carolina was striving to develop a graduate program in religion. Because the faculty was small and the budget limited, the Department faculty preferred a graduate program of general religion studies which integrated the interests and talents of all department members, rather than a program aimed at specialization. To create an academic balance within the Department, the faculty embarked on a search for a person knowledgeable in nonwestern religions-an area in which the Department perceived itself as weak. Professor Ruel Tyson, a professor of Religion at the University and then acting chairman of the Department, testified that, while they were searching for a specialist in nonwestern religions, they were looking for specialists of a particular type who could move from “the base ... of their expertise in one religion ... and contribute to other courses and other discussions in the Department having to do with their understanding of basic issues and the Study of Religion . .., ” Through academic channels, Tyson became aware that Smith had training suitable to the faculty vacancy. He, by letter, inquired of Smith whether she would be interested in interviewing for the position. She replied affirmatively. Thereafter the interviewing process commenced. After interviewing Smith in Boston, Massachusetts, Tyson, desirous for advice, forwarded Smith’s resumé and supporting materials to Professor John Schütz, the chairman of the Department, who was in Europe conducting research. In his letter, Tyson wrote that he was very favorably impressed with Smith and proceeded to enumerate her strengths. Despite his positive impression of her, Tyson did have a few reservations which he expressed as follows: Her age is sore cause for worry, 38 years in June. However, to be a first woman in the Dept. I place some premium on experience and maturity, a person not susceptible to fads and fashions, and on that score, I have no worry. . . . The other question concerns the status of her review two years hence for reappointment or tenure, etc. Will we be under pressure to promote after just three years, given her previous experience etc.? ... I should also have added into this question-equation, the factor of age. She is a former nun, but seems to have worked all that out nicely, anyway, all that is some years back. Reflecting upon these remarks at trial, Tyson said the only reason he mentioned Smith’s age was because he, as acting chairman, was a novice in the reappointment process and was unsure what considerations were important. He claimed that he “took an affirmative view toward [Smith’s] age,” believing that, in her case, it was an indicator of “maturity and experience.” With regard to the question about promotion at first review, Tyson again attributed it to his new role and also to Smith’s specific inquiry directing his attention to the issue. Responding to Tyson’s letter, Schütz was “less than dazzled by Ms. Smith on paper.” His primary concern was that she had “a technical interest in texts which happen to have religion as their subject matter,” as opposed to having “an interest in religion as a subject matter.” Also troubling Schütz were the periods of time in which Smith failed to explain her activities. Tyson, in a letter, attempted to address and dispel Schutz’s concerns. After much deliberation, Professor Tyson, in April 1973, extended an employment offer to Smith. Tyson explained to Smith that the University was offering her an initial 3-year probationary contract and that during the second year of the probationary period, the department chairman would make a recommendation to reappoint, to promote, or not to renew her contract. Smith accepted the offer and, by letter, stated, “the Faculty Legislation on University Government does not mention the promotion of an assistant professor at the first review .... I am accepting the position in the hopes of promotion at the first review.” Responding to Smith’s desire for early promotion, Tyson expressed his understanding of the promotion process as follows: Again, because my experience is not deep in these matters, I think I can safely say that a faculty member may be promoted at any time: such promotions are not tied to first review, second review, etc. I mean by this that the legislation you refer to demands the reviews indicated there, but does not pre-empt early [sic] action than required. Again, departments differ and I have only impressions, but I estimate that over half the assistant professors do serve two terms before being reviewed for promotion. However, so many things enter here it is not safe to generalize. The study of tenure currently underway will produce hard data on this sort of question. While I am certainly in no position to give you any assurances regarding your hopes for promotion at first review, it is a fact that this is not unprecedented. In July 1973, Professor John Schütz returned from England and resumed his duties as department chairman. The next month Smith assumed her teaching duties. She was the first full-time woman faculty member in the history of the Department. On October 1, 1974, Professor Schütz wrote to Smith informing her that it was time to review her teaching contract. Reiterating what Tyson had stated in his offer of employment, he explained that the review process would have one of three possible outcomes: reappointment; promotion; or nonrenewal. To permit a fully informed review, he suggested that she might desire to submit evidence including scholarly work and a written statement of goals and achievements. With respect to the issue of promotion, Schütz added: Because you have had more teaching experience than is common for someone in the first term of an Assistant Professorship, I want to return to the matter we discussed last spring. You mentioned then what you also mentioned in an earlier letter to Ruel Tyson in April, 1973: that you would hope to spend only one term at the rank of Assistant Professor before promotion. You will recall that our discussion last spring centered on the fact that such a promotion would automatically bring with it a permanent appointment, and that therefore it would more likely reflect a judgment about long-range potential for both effective teaching and a scholarship than it would reflect reward for length of past service as a teacher in general. Or to put the matter another way, the tenure decision is a professional decision about you as a teacher and scholar in your field of special interest and competence. (Emphasis added). To enhance her possibilities of reappointment or, hopefully, promotion, Smith submitted several research papers in addition to a statement of her scholarly goals. A meeting of the tenured faculty was held on November 5, 1974 to discuss the renewal of Smith’s teaching contract. Present at the meeting were Professors Schütz, Boyd, Dixon, Long, Tyson, Sasson, and Peck. All had been furnished copies of the materials submitted by Smith and the student evaluations of her performance. Professor Schütz testified that the meeting commenced with a review of Smith’s background. After that introduction, Professor Tyson recalled that the meeting had two distinct parts. The first issue addressed was whether promotion was appropriate. According to Schütz, the criteria guiding the promotion and tenure decision were: scholarship; teaching; service to the Department and the University; and the ability of the person to satisfy the goals and needs of the Department. Mindful of those criteria, the faculty, after much debate, decided unanimously that it was premature to consider Smith for promotion and tenure. Having foreclosed the possibility of promotion, the faculty spent the second part of the meeting discussing whether reappointment was desirable. Schütz testified that the criteria for reappointment were identical to those for promotion and tenure, the difference being the weight assigned to each criterion. In his words, “[t]he demands would be higher for promotion and tenure than they would be for reappointment.” The conclusion of most of the faculty members was that reappointment should not be offered. For example, while Professor Peck believed that Smith was a good Sanskrit scholar, he felt she was critically deficient in her abilities to perform the wider responsibilities of her teaching position. He questioned her capacity for growth and development. Professor Long voiced criticism about Smith’s inability to recognize the implications of her work. He also was concerned over the narrowness of her work; Smith channeled most of her energy into the technical aspects of the Mahabhárata, ignoring the nonwestern religions and the basic issues, common to all religions. Similar concerns were raised by Professor Tyson who in addition noted that she was doing little comparatively with her knowledge of the epics nor did she seem able to recognize the implications of her narrow focus nor to transfer them to other areas of religion. Recalling his impressions at trial, Tyson testified that he noticed that Smith had difficulty teaching religion and discerning the subject matter which should be discussed. Furthermore, he found her arguments and discussions in meetings to be difficult to follow, and frequently they were not pertinent to the issues being addressed. Having had numerous occasions to engage in scholarly discussions with Smith, Professor Dixon was distressed about her responses to his questions. In his words: Professor Smith would be inclined to give me a question [sic] as though it were a memorized answer for a course, the closest thing she could come to an answer to my question, but it didn’t fit. The result was that very quickly I stopped seeking this kind of help because it wasn’t-I wasn’t getting any help, and this obviously began to raise questions in my mind about her ability to deal even with the areas that we had assumed were those of her training and competence when she is trying to communicate with other members of the Department. Moreover, when she participated in faculty meetings, she dominated them with irrelevant issues that other faculty members did not desire to discuss. It was his perception that her nonresponsiveness extended to her teaching activities; her students were asking to be taught but she was not honoring their request. Of particular concern to the faculty, in this meeting, was, if Smith were reappointed, what were her prospects for tenure when the next review occurred in three years. As Schütz testified, the feeling was that if the possibility of tenure were low, then there was a strong justification for not renewing the contract. Although noting that reappointment did not commit the University to promote at the next review Schütz stated that it “certainly [tells] a person that you hope and you believe that [she] will be ready for another review in three years for tenure.” It was in such a context that Professor Dixon, seeing fatal flaws in Smith’s work, uttered the comment that Smith was too old to change. A discussion then occurred about the likelihood that Smith would progress in light of her development over the prior two years; it was acknowledged that she was a beginning scholar. Schütz recalled that many decisively negative remarks were made. Schütz recalled that Smith’s age arose in two contexts: her potential for growth and development and the professional consequences to Smith if the Department took either action. Professor Schütz testified on cross-examination that Smith’s background as a member in the Roman Catholic teaching order played no role in the evaluation and review of her. Her training as a Roman Catholic was neither relevant as experience nor as an individual religious preference considered in the employment decision. When the meeting ended, the prevailing mood was that an offer of reappointment should not be extended. Professor Sasson was the only member harboring doubts on that issue. He was concerned that they were being too harsh in their criticisms of Smith. In his November 6, 1974 letter to Professor Schütz, he expressed his dissatisfaction with the meeting: [I] would ... like to register a protest at the nature, the direction, and the level of discussion which unfolded during our evaluation. I feel that our criteria became over-refined as the discussion unraveled, and in many cases edged on irrelevancies. We seem to have demanded gratification in areas which, in other candidates, were left uninvestigated. In such a sensitive matter, we owe it to ourselves that we base our decision (if negative) on firmer reasons than the ones we seem to be offering. Professor Long also expressed concern about the review procedures. He felt that a desperate need existed to establish departmental criteria. In his letter to Professor Schütz he recommended that the criteria should be “teaching, research and writing, general faculty responsibilities, [and] compatibility within the intellectual orientation of the faculty.” Applying his recommended criteria, Long proceeded to evaluate Smith. With respect to her research and writing Long opined: What I find missing in her work is the context for a discussion of these issues. Hardly any use is made of the work of Georges Dumezil, who over the last thirty years has devoted himself to an analysis of the Mahabhárata and the religious cultural life of the Indo-Europeans.... I mention these texts-not because they are not mentioned in her work but because she has chosen not to take account of previous scholarship in an area in which she wishes to work and thus the context for her work is not established as part of a cultural humanistic endeavor. I am therefore suspicious of her use of computer analysis in the absence of materials of the sort mentioned above for it may be an indication that she does not understand the real scope of her proposal. Noting that her teaching was difficult to evaluate, he nevertheless criticized her for her “mystical attitude” towards teaching. Long, reviewing one of her examination questions, felt it to be “absolutely naive.” Nor did Long have favorable comments about Smith’s compatibility within the intellectual context of the faculty. He observed that: Professor Smith does not know what we are about most of the times. It appears too often that she does not understand the meaning of the university and the type of discourse appropriate to it. In most cases she tends to stifle the discussion of her colleagues in obscurantist manners regarding topics that are purely personal and stated in a personal rhetoric. Her presence has never enhanced any faculty discussion I have attended. In concluding that Smith should not be reappointed, Long raised these general considerations: In many respects Professor Smith is only at the very beginnings of her career. She seems to be competent in the Sanskrit language but she appears to be very naive regarding the context in which this competence might make sense. The most immediate context should be that of defining an almost “sacred” language in religious terms. She has great difficulty coming to terms with what is and ought to be the study of religion, and she does not yet know how to go about making other disciplines and areas of scholarship amenable to and aspects of a scholar’s work in the University community. . . . I do not think that she fulfills the kinds of intellectual needs of this Department and that it will be some time before she becomes a mature scholar. When I compare her with students who took their doctorates within a two year period of the awarding of her degree, I find that she is far behind them, and I refer to those who have also had to learn difficult languages. I think that should we reappoint her such reappointment would be tantamount to promotion to tenure three years hence. I cannot see the germs of growth in her such that she should or could be considered for tenure after three more years. Schütz responded to Long’s letter on November 7, 1974. While acknowledging that the Department did not have specific procedures and criteria, he noted that there were “very few departments which [could] reduce these factors to precise equations .... ” Moreover, the issues raised by the Smith case had never arisen before. Aware of Long’s dissatisfaction with the first meeting and Sasson’s feelings that they were being too harsh in their criticism of Smith, Schütz thought it advisable to convene a second meeting. He believed that a second meeting would afford the faculty another opportunity to evaluate Smith after having had time to reflect following the first meeting. Perhaps more importantly, he wanted to insure that the initial criticisms in the first meeting had not merely served as a catalyst to list Smith’s negative traits, without attention being properly paid to her strengths. The meeting was held on November 12th. Retracing his thinking at the second meeting, Tyson remembered that his conviction had intensified that there was a strong probability that “Smith would not develop sufficiently in the areas that most concerned [him] to present [the Department] with a strong candidate for tenure three years hence....” To facilitate his analysis, Schütz, after the second meeting, made notes. In the notes he listed Smith’s strengths and weaknesses. When comparing these characteristics with those of Professor Sanford, “[i]t was clear [to him] that it was a generally shared perception that she was not very sophisticated in Religious Studies.” Schütz read into the record the following remarks he made in his notes: “[I]s reappointment possible?” Below that is “No.” Below that indented is “Her age, strong precedents, we do not expect a positive decision down the road. 2) Is this fair? Are we treating like cases alike?” Below that I wrote, “Jack.” Jack is Professor Sasson.... Number 3 is “Are Affirmative Action matters at stake-, then I wrote “Not that I can see.” And 4, “Are her interests better cared for by one course of action than by another?” When asked at trial what he intended by the reference to Smith’s age, Schütz responded: Well, that is considering the whole question of reappointment. There had been one mention of her age that I think that I’ve already mentioned in my testimony about Professor Dixon, and I wrote down her age, the age because it had arisen in the discussion in one other way. There was no reference to chronological age at the meeting, the first meeting or the second meeting or any other, and I don’t know really anyone in that room would have known of the age, but the discrepancy between her maturity and stage of professional development and her career had been brought out and that was in connection with the whole question of whether or not she was as far along as she ought to be for someone two years out of her Ph.D. and what we might do in three years and all the things I mentioned. So I put that down and it struck me that there really wasn’t anything there to be said-that is, it wasn’t a relative consideration. (Emphasis added.) With respect to the question whether like cases were treated alike, Schütz explained that, to ensure fairness to Smith, the question was designed to force him to retrace the steps taken in the reviews of Sanford and Smith. He concluded that the considerations present in the two reviews were the same and that the reviews progressed similarly. The affirmative action remark was simply to force him to analyze and determine whether the review complied with the University affirmative action guidelines. Pursuant to that goal, Schütz considered whether his recommendation would be different if Smith were a white male. He concluded that his decision would not differ. By the last remark, Schütz explained that he was concerned with two considerations. First, if there only was but a slim possibility that if Smith were reappointed, she would be promoted at the next review, it would seem fairer to her both personally and professionally to make the decision at the first review rather than delaying an adverse determination for another three years. Apart from that concern, Schütz also had to decide what was in the Department’s best interests. Being a small department with limited funding, it was of particular importance that each professor perform a function in the Department’s growth and development plan. If a professor was not satisfying the expectations of the department, its goals would be thwarted. Following much deliberation, Schütz, on November 15, 1974 recommended to Dean James Gaskin of the College of Arts and Sciences that Smith’s teaching contract not be renewed. To support his recommendation, Schütz wrote: It is in the area of scholarly research and orientation that questions arise about renewal of the appointment. Such questions are less directly concerned with the quantity and quality of publication than with the range of Professor Smith’s scholarly interests and their appropriateness for this Department.... The issue facing the Department is not whether to promote to Associate Professor, as Professor Smith has formally requested, but whether to renew or terminate the present appointment as Assistant Professor. Renewing the contract would give more time for consideration of Professor Smith’s work, but there seems little disagreement in the Department’s analysis of the evidence. More time might offer opportunity for Professor Smith to extend her work in ways the Department would find appropriate, for she is a new scholar in her field. But her strength and weaknesses are clear, and the latter are by no means absolute shortcomings. In part they are a function of her location in this kind of Department. She is a good Sanskrit scholar and only less impressive in matters of the history of religions. Clearly there are departments and schools where her work would be appropriate and she would prosper. Considering her age and all other factors it seems best to encourage her to seek a new position now. Finally, some of the deficiencies sensed in Professor Smith’s approach to her field are particularly acute needs of the Department at this stage in its development. Schütz informed Smith of his decision in person and by letter dated November 15, 1974. He told her that, after considering her work “as a teacher, departmental and University citizen, and as a publishing scholar, ‘the faculty questioned, despite its “strong affirmation of much of [her] work,” ’ . . . the overall appropriateness of [her] work for [the Department] and [its] needs....” Shocked by the decision, Smith wrote a note to Schütz asking him to elaborate upon the reasons why she was not reappointed. Schütz explained the decision by writing: Many issues discussed with negative impact seem to me to fit under the larger question of the appropriateness of your skills, interests and approaches to the total task which must be accomplished within this Department by whoever might bear primary responsibility for a field such as Indian religions. Your technical command of the kinds of scholarly tools necessary for Sanskrit studies such as you have performed on the Mahabhárata, for example, is in no way in doubt. Your interest in religion in general is well known to all your colleagues here. What has seemed missing both on paper and in conversation is a command of middle range materials both primary and secondary which move from the particular focus of your specialized research through the whole range of scholarly issues and interest comprising the current status of Indian religious studies in general and then beyond to the area sometimes known as Religionswissenschaft. Your enthusiasm in these areas sometimes seems to outrun your command of the materials. On November 21st, after consulting with Dean Gaskin of the College of Arts and Sciences, Smith requested the Department to reconsider its determination. Schütz alerted the faculty to the reconsideration request. Not entirely certain as to what constituted a reconsideration, Schütz sought advice from Dean Gaskin. Gaskin told Schütz that a reconsideration was simply a thorough and thoughtful reevaluation of the Department’s decision. To enable an uninhibited discussion, Gaskin suggested to Schütz that he should absent himself from the reconsideration meeting. Schütz agreed. He scheduled the reconsideration meeting for early December. When the reconsideration meeting commenced, Schütz instructed the faculty that it had two tasks: first, to reconsider its earlier review of Smith and, second, to review his recommendation. The only new material submitted for their consideration was a December 3rd memorandum submitted by Smith. After his introductory remarks, Schütz, pursuant to Dean Gaskin’s recommendation, informed the faculty that he was going to absent himself from their discussions. In Schutz’s absence, Professor Boyd chaired the meeting. Despite initial confusion over the novelty of Schutz’s absence, the meeting eventually assumed structure. The tone of the meeting, however, did not deviate from that of the earlier sessions. Following the reconsideration meeting Professor Boyd sent a letter to Schütz and, because it met with “immediate and enthusiastic approval,” enclosed a copy of Professor Long’s list of reasons for terminating Smith. Boyd summarized Long’s position as, apart from a judgment on Smith’s professional competence, “she [did] not fill [the] place within the plans of the department which was envisioned at the time of her employment.” Boyd concurred with Long’s assessment that a general statement of reasons for termination, in contrast to a specific enumeration of reasons, would be more immune to attack as being discriminatory. Because some confusion persisted as to the legitimacy of the reconsideration meeting due to Chairman Schutz’s voluntary absence, Schütz called a fourth meeting. The purpose of the meeting was to clarify the reasons for his absence in addition to insuring that Smith’s reconsideration had been thoroughly discussed and understood. Recognizing their decision would be reviewed, Schütz stressed to the faculty members the need to focus on standards “early rather than late.” Also of importance were the impact of Smith’s expression that she did not want consideration of her long term welfare to influence their resolution of the problem and “the relatively unusual nature of a nonrenewal decision in the first term.” These considerations were to be balanced with those initially evaluated: “1) long range promise as a scholar-teacher with several facets, skills and dimensions; and 2) [the Department’s] current needs during the next five years.... ” The conclusion reached in the fourth meeting remained unchanged from the previous sessions. Accordingly, Schütz, in January, informed Smith that the reconsideration proceedings had not altered his earlier determination. Having exhausted the review possibilities within the Department, Smith, on October 2, 1975, submitted her case to the University Grievance Committee. The Committee interpreted Smith’s grievance as alleging due process deficiencies in the University review procedures and discrimination because of sex, religion, and the exercise of First Amendment rights. The Department, in summarizing its reasons for its decision to the Committee, stated: She is on the one hand unable to relate her special area of scholarship to the broader issues of the study of religion, and, on the other hand, she has no coherent language or systematic mode of dealing with basic theoretical questions in the area of religious studies. It was primarily on grounds related to this issue that she was not reappointed. The Committee, after conducting a thorough investigation, rendered its report on February 17, 1976. Among its findings were that: Smith was the first full-time female faculty member in the Religion Department and the first member of that Department not to be reappointed or promoted during the review process; the Department did not have a formal list of criteria for reappointment and promotion; and Smith was a victim of discrimination. It was their opinion that age was a factor in the Department’s decision especially since Smith, by requesting promotion, had imposed a greater seriousness in the Department’s deliberations than if only a reappointment decision was involved. The Committee, after finding that Smith had demonstrated a prima facie case of sex discrimination, found that the Department’s reliance upon Smith’s deficiencies in the history of religion was a protect for sex discrimination. To justify its conclusion the Committee relied on two items of evidence: the reason for not reappointing Smith did not relate to the Department’s established formal or informal criteria; and although Smith’s weakness was present in some male members of the faculty, it was not examined in their reappointment or promotion decisions. In summarizing its findings, the Committee held: In essence, an academic department with an admitted history of discrimination against women hired its first female faculty member with a narrowly defined and clearly understood speciality for her teaching and research, which she fulfilled in the classroom and in her publications; some time thereafter the department instituted a new evaluation criterion, [competency in the history of religion], for continued employment but did not communicate it to her as it had done with other personnel criteria; the department then fired her for not meeting that new criterion while at the same time reappointing a male colleague in equivalent status without even evaluating him in light of the new criterion; and the department in taking that action considered elements of sex discrimination among the undifferentiated impressions that influenced its course. Thus, all the categories of evidence that need to be explored in analyzing a claim of sex discrimination, save one-intent to discriminate-yield information that requires the committee to uphold the grievance. The Committee recommended that the Department reconsider its action and offer Smith reappointment with the understanding that tenure would not be automatic after the expiration of the three-year reappointment contract. In spite of the Grievance Committee’s recommendation, the Department, after several meetings, refused to alter its prior determination. Professor Tyson, the new chairman of the Department, informed Smith of his determination on March 24, 1976. Smith’s appeal of the decision to all levels of the University was unavailing. Thereafter, on July 16, 1976, she filed a complaint in the United States District Court for the Middle District of North Carolina. It is in the context of the detailed factual situation which has been described with particularity, that the several legal contentions advanced by the parties must be analyzed and dealt with. It should be observed, that following a full-scale trial, with findings of both jury and judge, the University, as the winning party, became entitled to a statement of facts as strongly favorable to its contentions as the proceedings would permit. We have stated the facts not as we necessarily should have found them, acting de novo, but as the factfinder (jury or judge) reasonably, and with sufficient support, was entitled to find them, or, when articulated, did find them. I. The Jury Instructions A. Introduction In instructing the jury on the age discrimination claim, the court explained: The Age Discrimination in Employment Act makes it unlawful to not reappoint or promote any individual who is covered by the Act if age is a causal factor in the non-reappointment or non-promotion of the individual. The Plaintiff has the burden of proof on these issues. To meet this burden, the Plaintiff must prove each of the following things by a preponderance of the evidence-first, that she was an employee covered by the Age Discrimination Act, and, ladies and gentlemen, you’re instructed that as to this element, the undisputed facts in this case establish that the Plaintiff, Mary Carroll Smith, was an employee covered by the Age Discrimination Act, and second, that the Defendant University refused to reappoint or promote the Plaintiff, and thirdly, that the Defendant University in so refusing did so because of the Plaintiff’s age. In determining whether the Plaintiff has proven these elements, you are instructed that the Age Discrimination in Employment Act prohibits employers, including Universities, from discriminating against their employees on the basis of age.... In order to establish the third element-that is, that the Defendant University in so refusing did so because of the Plaintiff’s age in order to establish that element, the Plaintiff must show by preponderance of the evidence that the University refused to reappoint or promote the Plaintiff because of her age. The Plaintiff is not required to prove that the refusal to reappoint or promote her was based solely on her age. Rarely can it be said that a person making a decision is motivated solely by a single concern or even that a particular purpose was the dominant or primary one. The Plaintiff is required to prove only that her age was a substantial or motivating factor in the decision not to reappoint or promote. If on either the reappointment issue or the promotion issue, the Plaintiff fails to satisfy you that each of the three elements exists, then you would answer “no” with respect to that issue. However, if you decide that the Plaintiff has carried her burden in that respect to either or both the reappointment or promotion issue, you must then decide whether the University would have made the same decision not to reappoint or promote the Plaintiff even if the University had not considered the Plaintiff’s age or some other illegitimate factor such as her sex or religion. In answering this question, the burden is on the University to show by a preponderance of the evidence that it would have reached the same decision not to reappoint or promote the Plaintiff if it had not considered the Plaintiff’s age or some other illegitimate fact. If after considering this question you decide the University has failed to satisfy you by a preponderance of the evidence that it would have reached the same decision, you should answer the issue “yes” in favor of the Plaintiff; however, if the University has satisfied you that it would have reached the same decision, you should answer the issue “no” in favor of the University. (Emphasis added.) At trial and on appeal, Smith finds fault with the instructions because they, in her opinion, did not set forth with sufficient clarity “the extent to which age discrimination must be involved in the employment decision for it to be actionable.” In particular she finds missing from the instructions the information that the justifications offered by the University could be pretextual and that age discrimination could be indirect or unintentional. Furthermore, Smith argues that the court erroneously stressed to the jury that age must be the sole criterion for refusing reappointment or promotion and that the University could escape liability for age discrimination if it could show that it would have reached the same decision if age had not been considered. It is Smith’s position that once age has been shown to have been considered, the University has the burden to show that age was not a factor influencing its decision. We commence our review by recognizing that an appellate court, when assessing the adequacy of jury instructions, is guided by the rule that the instructions should be viewed as a whole. “If the judge’s instructions properly present the issues and the law as applicable, it is no ground for complaint that certain portions, taken by themselves and isolated, may appear to be ambiguous, incomplete or otherwise subject to criticism.” See Laugesen v. Anaconda Co., 510 F.2d 307, 315 (6th Cir. 1975). In an ADEA action, the plaintiff has the burden to show that his dismissal was attributable to age discrimination. Loeb v. Textron, Inc., 600 F.2d 1003, 1017 (1st Cir. 1979). When an ADEA case is tried by a jury, “[t]he central issue, which the court must put directly to the jury, is whether or not plaintiff was discharged ‘because of his age.’ ” Id. Many courts in setting fourth the elements of proof and allocating the burdens of proof in ADEA cases have borrowed the principles enunciated in the Title VII case of McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). To establish a prima facie case of discrimination under McDonnell Douglas, a plaintiff must show: (i) that he is a member of the protected class; (ii) “that he ... was qualified for [the] job [and his performance satisfied his employer’s expectations] . . .; (iii) that, despite his qualifications [and performance], he was [dismissed]; and (iv) that, after his [dismissal], the position remained open and the employer continued to seek applicants from persons of complainant’s qualifications.” Id. at 802, 93 S.Ct., at 1824; see Loeb v. Textron, Inc., 600 F.2d at 1011-14; Schwager v. Sun Oil Co. of Pennsylvania, 591 F.2d 58, 60-61 (10th Cir. 1979); Kentroti v. Frontier Airlines, Inc., 585 F.2d 967, 969-70 (10th Cir. 1978); Price v. Maryland Casualty Co., 561 F.2d 609, 612 (5th Cir. 1977). Under the approach of McDonnell Douglas, when a plaintiff is successful in proving a prima facie case, the second stage of proof requires the employer only “to articulate some legitimate, nondiscriminatory reason” for its action. McDonnell Douglas Corp. v. Green, 411 U.S. at 802-03, 93 S.Ct. at 1824. He is not required at the second stage to prove absence of discriminatory motive by a preponderance of the evidence. Board of Trustees of Keene State College v. Sweeney, 439 U.S. 24, 99 S.Ct. 295, 58 L.Ed.2d 216 (1978). Once the employer satisfies this production of evidence by showing a legitimate nondiscriminatory reason for its decision, the plaintiff must move to the third stage to prove that the employer’s reason, which on its face appears an adequate justification for its action, is really no more than a pretext for age discrimination. McDonnell Douglas, 411 U.S. at 804,93 S.Ct. at 1825. A plaintiff prevails who succeeds in convincing the trier of facts that the justification is a pretext for discrimination. Examining the instructions which were given to the jury in the present appeal, it is apparent that the court did not recite the full McDonnell Douglas formulation. The court instructed the jury that first, it had to find that Smith was covered by the ADEA; second, that she was denied reappointment and promotion; and third, and most importantly, that the University’s decision was because of Smith’s age. With respect to the first element, the court told the jury that it was undisputed that Smith was covered by the Act. While the court did not explicitly instruct the jury that it was to find for the plaintiff as to the second element, obviously it was a point as to which there was no dispute; everyone acknowledged that Smith had not been reappointed and that she had not been promoted. The plaintiff was excused from the burden of proving the other two elements, under McDonnell Douglas, of a prima facie case. Having mentioned two necessary elements of proof as to which indisputably Smith had made out her case, all that remained to be proven under the charge was that the reason the University refused to reappoint or to promote was because of her age. Age need not be the sole reason but it must have been a substantial or motivating reason-a determinative reason-in the University’s determination. Loeb v. Textron, Inc., 600 F.2d 1003, 1019 (1st Cir. 1979). That determination, as mentioned previously, is the central concern in all age discrimination cases. The court then instructed the jury that, once Smith had proved the three elements, she would be entitled to recover unless the jury should determine that the University had demonstrated, by a preponderance of the evidence, that “it would have reached the same decision not to reappoint or promote the Plaintiff if it had not considered the Plaintiff’s age or some other illegitimate fact.” Unless the University satisfied its burden-again it must have done so by a preponderance of the evidence-the jury was instructed that it must render a verdict for Smith. Again, therefore, the McDonnell Douglas formula was curtailed in a manner favorable to Smith. Instead of requiring a two step proof, with the University required to shoulder the burden of articulating a legitimate nondiscriminatory reason and the plaintiff then compelled to adduce evidence of pretext, the district court placed the final burden directly on the University of demonstrating by a preponderance of the evidence that age discrimination did not cause the action of which plaintiff complained. It, therefore, essentially gave the jury the benefit of a finding that pretext had been made out. Technically, Board of Trustees of Keene State College v. Sweeney, 439 U.S. 24, 99 S.Ct. 295 (1978), was violated, but since it was the University, as the party who prevailed under the jury verdict, who was injured, not Smith, any error is not grounds for reversal. To recapitulate, comparing the instructions with the McDonnell Douglas formula we note that the instructions: (1) omitted some of the elements of a prima facie case as to which the burden under McDonnell Douglas is assigned to the plaintiff; (2) rather than merely requiring the defendant to articulate a legitimate, nondiscriminatory reason, it required it to prove, by a preponderance of the evidence that it would in fact have reached the same decision if it had not considered the plaintiff’s age; and (3) plaintiff was not required to prove pretext at all in order to have the University forced to assume the burden of showing its decision was not the product of Smith’s age or other illegitimate fact. For reasons that will soon become apparent, the instructions, although not completely conforming to the McDonnell Douglas requirements, were adequate and any errors that were committed favored, rather than harmed, the plaintiff. B. Degree of Required Identity with McDonnell Douglas Instructions Recently, the First Circuit, in Loeb v. Textron, Inc., 600 F.2d 1003 (1979) had occasion to address the applicability of the McDonnell Douglas principles, composed for a court trial under Title VII, to a jury trial of an ADEA claim. Noting the similarities in purpose and provision between the ADEA and Title VII and the absence of any indication in the legislative history that Congress intended “age discrimination . . . to be subject to different standards and methods of proof than race or sex discrimination,” the court concluded that the McDonnell Douglas formulation was appropriate for ADEA cases tried by a jury. Although appropriate for ADEA cases, the court was careful to stress that the jury need not be instructed as to all the McDonnell Douglas elements. It stated: McDonnell Douglas was not written as a prospective jury charge; to read its technical aspects to a jury, as was done here, will add little to the juror’s understanding of the case and, even worse, may lead jurors to abandon their own judgment and to seize upon poorly understood legalisms to decide the ultimate question of discrimination. Since the advantages of trial by jury lie in utilization of the jurors’ common sense, we would have serious reservations about using McDonnell Douglas if doing so meant engulfing a lay jury in the legal niceties discussed in this opinion. But we do not equate use of McDonnell Douglas with a requirement that the full formulation be read in baec verba to the jury. McDonnell Douglas is to a large extent an analytical framework enunciated post hoc, in light of a given set of facts, to give judges a method of organizing evidence and assigning the burdens of production and persuasion in a discrimination case. In light of this and the fact that the defendants’ burden is one of production rather than of persuasion, only the factual determinations necessary to the underlying rationale of McDonnell Douglas need be made by the jury-the burden-shifting can and should be monitored by the judge. Moreover, the term ‘prima facie case’ need never be mentioned to the jurors; ... McDonnell Douglas should be used to identify the important factual issues, and these can be set out in the charge, or in special questions, divorced from legal jargon. Id. at 1016. Depending on the nature of the evidence, the jury may never need to be educated as to the elements of a prima facie case. Recognizing that “direct evidence of discrimination is likely to be unavailable, and . . . the employer has the best access to the reasons that prompted him to fire, reject, discipline or refuse to promote the [employee],” the prima facie case requirement was designed to assist the employee in surmounting these evidentiary obstacles. Loeb v. Textron, Inc., 600 F.2d at 1014. Once an employee makes out a prima facie case “an inference of discrimination [arises] only because we presume these acts, if otherwise unexplained, are more likely than not based on the consideration of impermissible factors. . . . And we are willing to presume this largely because we know from our experience that more often than not people do not act in a totally arbitrary manner, without any underlying reasons . . ..” Furnco Construction Corp. v. Waters, 438 U.S. 567, 577, 98 S.Ct. 2943, 2949, 57 L.Ed.2d 957 (1978). Thus, “[i]f the principle [sic] ingredients of plaintiff’s case are inferences that are to be derived from the underpinnings of the McDonnell Douglas-type prima facie case, the jury should be asked to make these findings.” Loeb v. Textron, Inc., 600 F.2d at 1018. However, as suggested by the court in Loeb, when a plaintiff introduces direct and/or circumstantial evidence of discrimination, instructing the jury about the elements of a prima facie case, in addition to being superfluous, may be confusing. That court stated: [It is likely that most cases will not] fit a ‘classic’ or ‘pure’ McDonnell Douglas paradigm .... At least two other types of cases seem likely. One is a case which simply does not fit the mold of the McDonnell Douglas formula, as where plaintiff’s evidence of discrimination is significantly different (for example, where plaintiff relies chiefly upon direct evidence of discriminatory motive in a letter or on an admission from defendant). The court should not force a case into a McDonnell Douglas format if to do so will merely divert the jury from the real issues; rather it should use its best judgment as to the proper organization of the evidence and the charge. In cases of this type, the best charge may simply be one that emphasizes that plaintiff must prove, by a preponderance of the evidence, that he was discharged because of his age-with adequate explanation of the meaning of the age statute, the determinative role age must have played, etc., Another case would be one in which proof of the McDonnell Douglas elements is a significant part of plaintiff’s total evidence, but where there is also other evidence, direct or circumstantial, that might support an inference of discrimination. ... Here the district court will again have to exercise its judgment. Strict adherence to McDonnell Douglas, with instructions as to all elements of the particular prima facie case there described, may not be required. If, for example, there is substantial other evidence to warrant a finding of discrimination, it may be superfluous to tell the jury that it must find each element of the prima facie case. Rather, as in the previous case, the question of discrimination can be put more directly and simply. (Emphasis added.) Id. at 1018. Thus Loeb teaches first, that, while McDonnell Douglas may be appropriate to a jury trial of an ADEA case, all the elements need not be recited to the jury, and second, that, despite its appropriateness to age discrimination claims, McDonnell Douglas is not the only permissible standard; other formulations may be proper, especially when direct and circumstantial evidence of discrimination have, as here, been introduced. We believe that Professor Smith’s case is covered by the second aspect of the Loeb lesson. Most of the evidence that Smith introduced consisted of notes taken by and correspondence among Religion Department professors. In such a situation, the jury was not concerned so much with the inferences possible from the McDonnell Douglas type prima facie case as it was with whether the direct evidence that Smith introduced, demonstrated that the University’s decision was unlawfully motivated by her age. As we construe the court’s language, the three elements that it required Smith to prove by a preponderance of the evidence constituted a variety of a prima facie case of discrimination although not expressed in classic McDonnell Douglas terms. Once Smith had satisfied the proof, the court placed the burden of rebuttal on the University. The right of rebuttal consisted of the right to produce for the jury evidence that, even though prima facie case had been made, still, in point of fact, the refusal to reappoint or to promote actually proceeded frorn other reasons unrelated to age-that plaintiff would not have been reappointed or promoted whatever had been her age. Smith contends that she was prejudiced by this portion of the court’s instruction. She argues that once she demonstrated that age was a determinative factor in the University’s decision, the burden was on the University to show that it was not a substantial factor in its decision. Smith would have us engage in a meaningless semantic exercise, for that instruction was in substance given: “In answering this question, the burden is on the University to show by a preponderance of the evidence that it would have reached the same decision not to reappoint or promote the Plaintiff if it had not considered the Plaintiff’s age or some other illegitimate fact.” Even though at an earlier stage the court has determined that a prima facie case has been made out to the effect that age was an impermissible factor in the determinations not to promote, that does not end the matter. It does not shut off the defendant’s right of rebuttal. That right encompasses an opportunity to prove by a preponderance of the evidence that in fact age was not a causative factor. As stated in Laugesen v. Anaconda Co., 510 F.2d 307 at 317: However expressed, we believe it was essential for the jury to understand from the instructions that there could be more than one factor in the decision to discharge him and that he was nevertheless entitled to recover if one such factor was his age and if in fact it made a difference in determining whether he was to be retained or discharged. (Emphasis added.) While the intermediate answer is that the evidence suffices to permit a conclusion that age was impermissibly relied on, it does not foreclose a final determination, based on evidence advanced by the defendant, that actually age had not played a part in the determination not to reappoint or promote-that the very same result would have come about if age discrimination, prima facie or otherwise, had never existed. Cf. Mt. Healthy City Board of Education v. Doyle, 429 U.S. 274, 286-87, 97 S.Ct. 568, 576, 50 L.Ed.2d 471 (1977): Initially, in this case, the burden was properly placed on respondent to show that his conduct was constitutionally protected, and that this conduct was a ‘substantial factor’-or to put it in other words, that it was a ‘motivating factor’ in the Board’s decision not to rehire him. Respondent having carried that burden, however, the District Court should have gone on to determine whether the Board had shown by a preponderance of the evidence that it would have reached the same decision as to respondent’s reemployment even in the absence of the protected conduct. While Mt. Healthy arises under the First and Fourteenth Amendments, both in it and in our ADEA case, the question of causality is involved in essentially the same fashion. As the statute makes clear, it prohibits discrimination “because of such individual’s age” (Emphasis added). Appellant freely acknowledges at p. 2 of her Reply Brief: “We admit causation is the standard, and that, in fact, is the thrust of our argument.” Mt. Healthy was also articulating standards with respect to a “rule of causation.” Mt. Healthy, 429 U.S. at 285, 97 S.Ct. at 575. As for not mentioning pretext, the district court treated that issue as decided in Smith’s favor, moving directly to require the University to prove that the failure to reappoint or promote was not causally linked to considerations of age. There was, consequently, no occasion for an instruction as to pretext. Having imposed upon the University the burden of showing that, even without any age considerations being present at all, the plaintiff would not have been reappointed or promoted, there was no need also to instruct about pretext. For, as the Supreme Court noted in Board of Trustees of Keene State College v. Sweeney, 439 U.S. 24, 99 S.Ct. 295 (1978), placing the burden of proof upon the employer at the second stage also incorrectly places the burden upon him to show that the reason was not a pretext. See generally Runyan, Employment Decision-Making in Educational Institutions, 26 Wayne L.Rev. 955, 1011-13 (1980). The court in Loeb v. Textron, Inc. held that: We think it now clear that McDonnell Douglas leaves the burden of persuasion at all times with the plaintiff, and that the employers’ burden to ‘articulate’ a legitimate, nondiscriminatory reason is not a burden to persuade the trier that he was in fact motivated by that reason and not by a discriminatory one. Rather it is a burden of production-i. e. a burden to articulate or state a valid reason, following which the complainant must show that the reason so articulated or stated is a mere pretext or ‘cover-up’ for what was in truth a discriminatory purpose. This, indeed, is only logical. If an employer were to prove that he was motivated by a legitimate reason, there would be no room left for showing that the reason was a ‘pretext,’ as pretext is ‘a purpose or motive alleged or an appearance assumed in order to cloak the real intention or state of affairs.’ Webster’s Third New International Dictionary (1971). To say, as the court did here, that the defendant must prove that its action was based on a legitimate reason and that the plaintiff must ‘then’ prove that it was not, is contradictory. 600 F.2d at 1012. The instruction was altogether fair to the plaintiff; the jury in finding for the University obviously concluded that the University had met its burden. C. Other Alleged Errors in the Jury Instructions Addressing Smith’s remaining challenges to the instructions we find them all devoid of merit. That the instruction failed to mention that age discrimination may be direct or unintentional does not render it reversible error. In Laugesen v. Anaconda Co., 510 F.2d 307, 314 (6th Cir. 1975), the court also omitted an instruction that the employer did not have to have the specific intent to discriminate, but rather, as the court did here, when referring to the reason for the University’s actions employed the term “because of the plaintiff’s age.” Id. We think that, although language precisely relating the nature of the requisite intent was not present, the instruction managed to convey to the jury, in understandable terms, the applicable law which was that the discrimination could be either intentional or unintentional; the important consideration was whether the decision was influenced by the employee’s age. “. .. the judge’s instructions substantially charged the jury in the manner desired by plaintiff without using the spe