James R. Roebuck, a disappointed academic tenure candidate, sued his former employer, Drexel University, alleging that but for discrimination on account of his race, he would have acquired tenure. The suit was grounded on both 42 U.S.C. Sec. 1981 (1982) and Title VII of the Civil Rights Act of 1964, 42 U.S.C. Secs. 2000e to 2000e-17 (1982).

In the Sec. 1981 suit the jury found for Roebuck, but the district court granted judgment n.o.v. for Drexel, holding that the evidence was insufficient to support the verdict. Concomitantly the district court found for Drexel on the Title VII claim (tried without a jury). The district court also denied Roebuck's motion to alter or amend the Title VII judgment to conform with the jury's findings on the Sec. 1981 claim. Roebuck appeals.

For the reasons that follow, we find just enough evidence of discrimination to conclude that Drexel was not entitled to judgment as a matter of law. We nevertheless agree with the district court's assessment that the jury's verdict in favor of Roebuck was against the clear weight of the evidence. We therefore will reverse the grant of judgment n.o.v., but affirm the district court's alternative grant of a new trial.

We disagree, however, with the district court's ruling that it was not bound to conform its Title VII judgment to avoid inconsistency with the jury verdict. We instead follow the great weight of authority in other circuits and hold that principles of collateral estoppel and jury supremacy preclude a district court from issuing a judgment at variance with the jury's findings. We therefore will vacate the Title VII judgment in favor of Drexel, with instructions to the district court to await the jury's verdict upon retrial before rendering judgment on the Title VII claim.

Dr. Roebuck, a black man, was hired in 1970 by Drexel as a Lecturer in the History-Politics Department of the University's College of Humanities and Social Sciences. In 1977, upon completion of his Ph.D., Roebuck was promoted to Assistant Professor and began his six-year probationary period for tenure. In 1983, after an extensive review of his credentials, Roebuck was denied tenure.1

Drexel employs a multi-level tenure review process and has published detailed standards for granting tenure. Because it is important to the result, a description of the process and the standards follows.

A faculty member is eligible for tenure after completing six years of teaching at the assistant professor level. In the year preceding the year in which a faculty member will be considered for tenure, the candidate meets with the dean of his college and the head of his department for a "pre-tenure review meeting," at which the candidate is given specific suggestions for improving his record and his chances for tenure. The candidate then prepares a tenure dossier, which forms, in essence, his application for tenure, including all relevant materials (scholarship, teaching evaluations, etc.).

The candidate next meets with a departmental committee formed to review his tenure application. The committee considers the entire dossier as well as the views of external peer reviewers who have been sent samples of the candidate's scholarship. The committee prepares a report, including a recommendation, and forwards it to the department head, who performs an independent evaluation of the candidate's credentials. The department head's recommendation is then considered, along with the committee report and other materials, by the Collegial Committee, which is composed of two representatives from each department in the candidate's college and is chaired in a non-voting capacity by the dean of the college. The Collegial Committee performs a de novo review of the material and prepares a recommendation for the dean. The dean, and subsequently the University Vice President for Academic Affairs, perform their own independent reviews. If the Vice President's decision is negative, the candidate is denied tenure, but is entitled to appeal.

The appeals process allows the candidate to reargue his case at each level of the tenure review at which he was unsuccessful. Should he fail to gain reconsideration, the candidate can then choose to petition a standing appeals committee, which issues an advisory report to the University President, or the candidate can appeal directly to the President himself.

In Roebuck's case, his department committee unanimously recommended him for tenure; however each subsequent recommendation in the process was negative. Upon denial of tenure by the Vice President, Roebuck appealed directly to the President but tenure was denied.

At each level of the process, the candidate's credentials are reviewed in three discrete areas--scholarship, teaching, and service. In order to gain tenure, the candidate's work must be rated at least "satisfactory" in all of the areas, and "outstanding" in at least one.

In reviewing a candidate's scholarship, a reviewer is expected to consider the work within the framework of the following priority scale: top priority--refereed publications; second priority--non-published but refereed work (such as conference papers); third priority--non-published work with a clear potential for refereed publication; and lowest priority--non-refereed publications.2 Each tenure candidate's scholarship is reviewed not only by the tenure committees and advisors themselves, but also by external peer reviewers in the candidate's field, whose reviews are available for consideration at each stage of the process.

Drexel's standards for review of a tenure candidate's teaching ability are vague, and there is little in the record to illuminate the requirements or the assessment process. It is apparent, however, that review is based primarily on evaluations by students, peer and department head recommendations (of both current and former, and both internal and external, faculty), and evaluation of the nature and substance of the courses taught by the candidate.

The standards for assessing service are exceedingly vague and the parties disagree about the application of these standards. A document entitled "Departmental and Collegial Review Procedures for Tenure" states that the tenure decision involves consideration of the candidate's "service to Drexel," and then explains that decisionmakers should evaluate the candidate's "record of performance in ... University, collegial, and departmental service, and community service relevant to the mission of the institution." J.A. at 1354. There is dispute, however, over what type of service is "relevant to the mission of [Drexel]," and over the extent and breadth of service necessary to qualify as "outstanding." Drexel argues that outstanding service requires leadership, rather than mere participation, or alternatively requires "extramural activity promoting the University." Brief for the Appellee at 13. However, as we will explain in some detail, infra, Roebuck points to contrary evidence, which we must credit for purposes of this appeal.3

Although a formal assessment of an applicant's rating on each of the three prongs is not required at each level of review, from testimony at trial it is apparent that Roebuck was denied tenure because his teaching and service were found merely satisfactory and his scholarship was determined to be unsatisfactory.

James Roebuck was and is a resident of West Philadelphia, a predominantly black community where Drexel, a predominantly white institution, is located. At the time of his hiring, Drexel's relationship with that community was somewhat uneasy, and Roebuck was hired in no small part because of his perceived ability to help "develop a more positive presence in West Philadelphia [and] to work more closely with the community." J.A. at 43 (testimony of Roebuck). As Roebuck explained in his testimony at trial,

[w]hen I was recruited to come to Drexel, one of the things that was said to me was that Drexel was in the process of trying to reach out to the West Philadelphia community, that I as a West Philadelphia resident would be a particular value to the Drexel community in my ability to interact with that community.... That was the basic underlying commitment from the University to me when I became a member of the faculty.

J.A. at 273.

With that understanding, Roebuck devoted himself unfailingly to community service. His resume at the time of his tenure review read like a directory of community organizations, including membership on numerous boards of directors and the chairmanship of several.4 Moreover, Roebuck plainly had participated in at least his fair share of intra-University service, primarily through his work on various faculty committees.5

Roebuck's feedback on his service activities conformed with his understanding of his role at Drexel. As he explained,

I got consistently positive feedback on the things I was doing. That came not only in terms of comment by various administrative officials, it came in terms of letters that I received from some of those officials indicating I had done a good job, I had helped the University. It came in terms of my annual meetings of my department chairmen.

J.A. at 58; see also J.A. at 59, 272 (Roebuck's testimony on additional positive feedback he received from the Drexel administration). There is thus evidence in the record to support Roebuck's contention that he was hired in large part for his ability to project a positive image for the University in the West Philadelphia community, that he performed that role vigorously and successfully, and that, at least until the time of his tenure application, his service was warmly received by the University.

Roebuck's teaching record, though not nearly as well documented as his service contributions, similarly reflects quality and commitment. He was warmly praised by many of his students, his student evaluations reflected general satisfaction on the part of his pupils, and he received several positive letters of recommendation about his teaching prowess.

His record of scholarship, however, was significantly less impressive. Although he listed numerous "scholarly papers and public lectures" on his resume, see J.A. at 1341, in his Statement on Tenure and Promotion he referred to only five items produced during his twelve years at Drexel in support of his claim of satisfactory scholarship for tenure purposes, J.A. at 1405-06.6

Roebuck conceded on cross-examination that not one of these items fell within either of the top two priority categories for scholarship--refereed publications or non-published refereed work. J.A. at 121-23. He claimed that one paper, which was based in large part upon his master's thesis, had a "clear potential for publication in [a] refereed journal[ ] or press[ ]," the third priority category; however he did not submit this piece to the tenure review committees as part of his tenure dossier, J.A. at 123-24, and hence it was not considered as part of his application. Of the two full-length papers that he did submit with his tenure application, he conceded that only one even fell within the University's priority scale, and that it fell in the lowest priority category--publication in a non-refereed journal, J.A. at 131; the other he conceded was unpublishable in its current form. J.A. at 125-27.7

Roebuck also submitted two biographical entries, totalling five typewritten pages, which had been published in a biographical dictionary and which he conceded were derived in large part from other scholarly works and from his own doctoral dissertation, rather than from any new original research. J.A. at 128-29. Finally, Roebuck submitted a contract he had signed with a publishing house in which he agreed to produce a bibliographical volume on U.S.-East Asia relations in the nineteenth and twentieth centuries, Roebuck's area of academic expertise, to be included in a fifteen to twenty-five volume bibliography on American foreign policy and diplomacy; no work on this volume, however, was submitted for his tenure review.8

In October, 1982, a committee of faculty members of the History-Politics Department convened to address Roebuck's tenure application. The deliberations of the committee reflected some difficulty with the question, but the committee ultimately recommended unanimously not only that Roebuck be granted tenure, but also that he be promoted from assistant to associate professor.9 The committee rated Roebuck's service and teaching "outstanding," and his scholarship "satisfactory."

With regard to his on-campus service (faculty committees, service to student organizations, etc.), the committee found Roebuck's activities to have been "numerous, willingly given, and constructive." J.A. at 1361. With regard to his extramural service activities, however, the committee report simply glows: "Dr. Roebuck's contributions have been not only numerous and excellent, but almost impossible to duplicate if Drexel is to achieve recognition as a well-regarded neighbor in the West Philadelphia community." Id.

The committee found "[t]he sheer number of Dr. Roebuck's service contributions to campus life [to be] impressive; few Drexel faculty of any rank could list as many." The report noted "his willing acceptance of assignments, his responsible performance of them, [and] his keen interest in all departmental matters." Id. Moreover, the committee recognized the unique value to Drexel that Roebuck's service activities provided. The report quoted favorably from several letters of recommendation, which noted: " 'Because he serves as a role model for people who do not ordinarily identify with institutions of higher learning, Dr. Roebuck has become a valued resource reflecting Drexel University's service to this city'.... 'His participation has enlarged the perception of the academic community [in] West Philadelphia.' " J.A. at 1362-63. The committee thus inferred that "Drexel's reputation is enhanced by [Roebuck's] work." Id.

The committee was similarly impressed with Roebuck's teaching, which they also rated "outstanding." The report recounted the course evaluations submitted by Roebuck's students, who generally spoke favorably of his teaching. Moreover, the committee found Roebuck's grading fair and in line with the rest of the faculty, the variety of his courses impressive, and his willingness to assist other instructors admirable. Hence they concluded that Roebuck was likely to continue to grow as a teacher.

The committee's assessment of Roebuck's scholarship, however, was of another order, and plainly reflected compromise of an internal struggle. The report stated:

The committee judges Dr. Roebuck's scholarly performance to be satisfactory. It means by that term not "satisfactory" in relation to some established and fixed standard such as might prevail in a department of long-established reputation for scholarship, but satisfactory in the light of a program to improve the quality of departmental scholarship without sacrificing other desirable goals.

J.A. at 1363.

The committee reviewed the comments of the four external scholarship reviewers, and found them mixed, at best, in their appraisal. According to the committee's assessment, one reviewer found Roebuck's published article on William Howard Taft to be " 'valuable,' " but found his "American Liberals" piece to need " 'additional work.' " J.A. at 1363. The committee reported that the reviewer "thought Dr. Roebuck gave promise of building a reputation in United States-East Asian relations." J.A. at 1364. Another reviewer, according to the committee report, found the Taft article " 'a rather good piece,' " but he, like the other reviewers, found the "American Liberals" piece lacking. Id. The final two reviewers were not particularly impressed by either article, with one giving a quite harsh appraisal of Roebuck's work. The committee was most impressed, however, with the first review. Id. Hence, the committee explained its decision to grade Roebuck's scholarship as "satisfactory," as follows:

The first way station [for the department] on the road to productive scholarship was seen to be activity, not necessarily quality, for it was thought that activity would generate a departmental tone out of which work of scholarly quality could thereafter emerge.... Seen in this light, the activity shown by Dr. Roebuck--the Taft article, the 1930s paper, the biographical entries, the contracted bibliography--is a step in a desired departmental direction.

Id. This was hardly a ringing endorsement. In the eyes of Drexel, the committee no doubt was damning him with faint praise. But for our purposes, taken in the light most favorable to Roebuck, this is at least a "satisfactory" review.

Moreover, the committee not only recommended Roebuck for tenure, but found his performance " 'measurably more impressive than that required for a favorable tenure recommendation,' " and thus, in accord with the college standards, recommended Roebuck for promotion to associate professor, as well. J.A. at 1365. However, this assessment was followed in the report by a somewhat cryptic comment:

One additional factor should be given weight. Over a decade ago Drexel University directed its administrative officers to bend every effort toward a search for qualified, competent black teachers. The need which generated that search has not expired. Meanwhile, Dr. Roebuck has exemplified on campus and in the community these past twelve years the qualities which make that extra effort worthwhile.

Id. Although it is unclear from the face of the report how much weight this "additional factor" was given by the committee, on review of a judgment n.o.v. we must give the benefit of every doubt to Roebuck and hence we read the report as recommending tenure to Roebuck on the basis of his own qualifications and not merely because he is black.

Throughout most of Roebuck's employment at Drexel, the History-Politics Department was chaired by Dr. Edward Arian, who consistently gave Roebuck high marks during their annual performance review meetings. However, in August, 1982, Dr. Phillip Cannistraro, a newcomer to Drexel, replaced Dr. Arian as head of the department. Based on Cannistraro's independent review of Roebuck's record, he recommended, in a letter dated January 17, 1983 to dean of the college Dr. Thomas Canavan, that Roebuck be denied tenure.

Cannistraro found Roebuck's record of service merely "satisfactory." He acknowledged that Roebuck's service "away from the Drexel campus has been long-standing and exemplary, particularly in the West Philadelphia community where he lives." Moreover, Cannistraro found that Roebuck's "on-campus service ... record compares favorably to that of most faculty members in the department." He nevertheless found Roebuck's overall service merely "satisfactory," because of his lack of service in the "professional sphere." "It is in this area," Cannistraro asserted, "that the faculty member brings to his department and university recognition in scholarly circles and enhances the institution's academic standing." Moreover, Cannistraro questioned whether Roebuck's "off-campus service should form the basis for judging professional accomplishment." He plainly was uncomfortable with the idea that such service in the West Philadelphia community could be "relevant to the mission of the institution." J.A. at 2035.

Cannistraro also found Roebuck's teaching to be merely "satisfactory." He questioned the departmental committee's extensive reliance on teaching evaluations culled only from Roebuck's most recent classes (apparently ignoring the fact that such evaluations were not readily available in prior years, J.A. at 174-76). Moreover, he questioned the breadth and the innovativeness of Roebuck's classes, and noted that in February, 1982, Dr. Arian had rated Roebuck's teaching merely "very good" even though he had rated two other professors "outstanding" that year. Finally, Cannistraro relied on Roebuck's failure to win any teaching awards that might have demonstrated his excellence in teaching.

Cannistraro took a decidedly dim view, however, of Roebuck's scholarship. He assessed the opinions of the external reviewers, determining that the positive reviews of the Taft article were actually either "cursory" or merely "luke-warm," and finding the negative reviews more on point. Moreover, he quoted extensively from the "uniformly negative" reviews of the "American Liberals" article and expressly agreed with those judgments. Given Roebuck's limited output since finishing his dissertation (one published article of "dubious value"), Cannistraro found no "promise of significantly improved productivity in the future." Thus he concluded that "there is no question that his performance in scholarship must be judged unsatisfactory."10 J.A. at 2036-37.

The Collegial Committee did not prepare a formal report after its deliberations, nor did it vote separately on the questions of Roebuck's service, teaching, and scholarship. However, unrebutted testimony at trial made clear that the Committee voted 6-0 to recommend the denial of tenure.

Because of the lack of a written record, the precise basis for that decision is unclear and, in fact, each member of the committee could have voted to deny tenure for different reasons. See J.A. at 389. Dr. Michael Sullivan, a member of the committee who testified for the plaintiff, conceded, however, that his impression was that the committee concluded that Roebuck's scholarship "left a lot to be desired," J.A. at 440, although he also insisted that the "judgment was that [Roebuck's scholarship] was mixed," J.A. at 438. According to Dean Canavan, who testified for Drexel, his sense was that the committee's vote was based on a finding of satisfactory teaching and service and unsatisfactory scholarship. J.A. at 559, 566, 568.

Like the Collegial Committee, Dean Canavan did not issue a written report on his recommendation. However, he testified that he found Roebuck's teaching and service satisfactory and his scholarship unsatisfactory, and hence he recommended to the Vice President that he deny tenure.

Canavan rated Roebuck's teaching "satisfactory" because he found that it fulfilled all of the University's "ordinary expectations," but was not sufficiently "innovative" to justify a rating of "outstanding." J.A. at 612-13. Canavan rated Roebuck's scholarship "unsatisfactory" because he found a significant lack of both quantity and quality. J.A. at 614-15. Finally, the dean rated Roebuck's service as merely "satisfactory" because he claimed that Roebuck had not "taken a leadership role of significant importance to have created some kind of change either in [his] department or in the college or in the University at large." J.A. at 616-17.

University Vice President for Academic Affairs Dr. Bernard Sagik made the final decision to deny Roebuck tenure. He testified that, based on his independent review, he rated Roebuck's service merely satisfactory, because he believed that "much of it was not related to the University." J.A. at 780.11 He rated Roebuck's teaching merely satisfactory, apparently because he found that it neither caused his students to change their field nor did it appear to have a "profound effect on their lives and their thinking." J.A. at 779-80. And he found Roebuck's scholarship to be unsatisfactory because he "did not see the level of refereed works that [he] would have hoped for," nor did he "see a draft of work in such form as to make [him] believe that with minor revision it was publishable." J.A. at 779.

At the suggestion of University President Dr. William Hagerty, Roebuck elected to appeal directly to Hagerty rather than resort to the appeals committee. By letter dated October 25, 1983, Hagerty denied the appeal.

In his letter, Hagerty explained that he found Roebuck's teaching satisfactory, though not outstanding, because of a failure of many students to provide comments on Roebuck's course evaluations; such failure, Hagerty explained, often stems from a feeling that the teacher is neither poor nor outstanding. J.A. at 2476.

Hagerty's assessment of Roebuck's service was somewhat unclear, as he did not grade it on the University's "satisfactory"--"outstanding" rating scale, and his explanation is somewhat cryptic. He evidently was impressed by the "enormous amounts of time and energy" Roebuck devoted to service activities, and thus he wrote: "Your service record is indeed above average. This is true with respect to the service to your department, to your college, and to the university as well as to outside areas." Nevertheless, Hagerty concluded: "Yet you are not seeking a public relations appointment, you are seeking a tenured faculty position and although service is very important, it does not substitute for either teaching or scholarship." J.A. at 2476-77.

Finally, Hagerty found Roebuck's scholarship to be unsatisfactory, apparently because of insufficient "scholarly publications," i.e., "publications which are reviewed by professionals for print in scholarly journals." J.A. at 2477. However, in announcing the ultimate denial of Roebuck's appeal, Hagerty seemed to go out of his way to explain why Roebuck's "above average" service could not justify a grant of tenure: "The primary matters affecting tenure as stated in the Faculty and Administrator's Guide, are scholarship and teaching. In at least one of our colleges, namely, the College of Science, service is a distant third." J.A. at 2478.12

Roebuck filed this suit in the district court for the Eastern District of Pennsylvania, alleging that Drexel, in denying him tenure and promotion, intentionally discriminated against him on the basis of his race in violation of 42 U.S.C. Sec. 198113 and Title VII of the Civil Rights Act of 1964, 42 U.S.C. Sec. 2000e to 2000e-17.14 In his prayer for relief, Roebuck sought compensatory damages (for emotional distress), punitive damages, and equitable relief in the form of back pay, reinstatement, tenure and promotion. The Sec. 1981 claim was tried to a jury and the Title VII claim was concurrently tried to the court, on virtually the same evidence.15

After a five-day trial, the jury returned a verdict for the plaintiff, answering the special interrogatories as follows:

1. Do you find that Dr. Roebuck has proven by a preponderance of the evidence that the Drexel University decisionmakers would have granted him tenure but for his race?

Yes.

2. If you have answered "Yes" to question no. 1, what amount of money do you find will fairly and adequately compensate Dr. Roebuck for damages shown to have been proximately caused by the denial of tenure to the present time?

$5,000.

J.A. at 17, 1273. At Roebuck's request, the district court waited to issue its decision on the Title VII claim until after the jury returned its verdict.16 After the verdict was in, the court issued a lengthy oral opinion, finding for Drexel on the Title VII claim. The court then entered judgment on the verdicts and denied Roebuck injunctive relief on the Sec. 1981 claim.

The parties thereafter filed post-trial motions, Drexel moving for judgment n.o.v. and Roebuck moving, under Fed.R.Civ.P. 59(e), to alter or amend the Title VII verdict to conform with the jury's Sec. 1981 verdict under principles of res judicata. The district court granted Drexel's motion for judgment n.o.v. on the Sec. 1981 claim17 and denied Roebuck's Rule 59 motion on the Title VII claim,18 hence the court entered judgment for Drexel on both claims. This appeal followed.

The proof required for a plaintiff to prevail under Sec. 1981 is identical to that required in a Title VII case--plaintiff must show that but for his race he would have been granted tenure. See Lewis v. University of Pittsburgh, 725 F.2d 910, 914-16 (3d Cir.1983) (simply proving that race was a "substantial factor" in an employer's decision is not enough; plaintiff must prove "but for" causation), cert. denied, 469 U.S. 892, 105 S. Ct. 266, 83 L. Ed. 2d 202 (1984); Lewis, 725 F.2d at 915 n. 5 (a Sec. 1981 claim "require[s] the same elements of proof as a Title VII action" (citations omitted)). In reviewing the record, it is important to keep in mind both what Roebuck must prove and what Roebuck need not prove in order to win his case.

In a typical discrimination case, involving a layoff or a failure to promote, the now-familiar McDonnell Douglas three stage procedure requires a plaintiff first to prove a prima facie case, i.e., to show that he is a member of a protected class, that he was qualified for the position but nevertheless was rejected, and that non-members of the protected class were treated more favorably than he. McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S. Ct. 1817, 1824, 36 L. Ed. 2d 668 (1973). However, we have explained that the precise elements of a prima facie showing will vary depending on the circumstances of the case. See Massarsky v. General Motors Corp., 706 F.2d 111, 118 n. 13 (3d Cir.), cert. denied, 464 U.S. 937, 104 S. Ct. 348, 78 L. Ed. 2d 314 (1983). In the context of a failure to grant tenure, we agree with the Court of Appeals for the First Circuit that, in order to satisfy the qualifications element of the McDonnell Douglas test, plaintiff

need only show that he was sufficiently qualified to be among those persons from whom a selection, to some extent discretionary, would be made. That is, he need show only that his qualifications were at least sufficient to place him in the middle group of tenure candidates as to whom both a decision granting tenure and a decision denying tenure could be justified as a reasonable exercise of discretion by the tenure-decision making body.

Banerjee v. Board of Trustees of Smith College, 648 F.2d 61, 63 (1st Cir.) (citations omitted), cert. denied, 454 U.S. 1098, 102 S. Ct. 671, 70 L. Ed. 2d 639 (1981).19

Moreover, because of the uniqueness of the tenure decision, in that an unsuccessful candidate is not necessarily replaced by a successful one, the third prong of the prima facie showing can be satisfied by showing that tenure positions "were open at the time plaintiff was denied tenure, in the sense that others were granted tenure in the department during a period relatively near to the time plaintiff was denied tenure." Banerjee, 648 F.2d at 62 (citations omitted).

Once plaintiff has made out a prima facie case, the burden shifts to defendant to articulate a legitimate, non-discriminatory reason for the complained-of action. See McDonnell Douglas, 411 U.S. at 802, 93 S.Ct. at 1824. Plaintiff then bears the burden of proving that defendant's articulated reasons in fact were merely a pretext. However, if the plaintiff presents enough evidence for a jury to find that the asserted reasons for the tenure denial were not the actual reasons, then the jury may infer that the employer actually was motivated in its decision by race; plaintiff is not required to provide independent, direct evidence of racial discrimination. See Duffy v. Wheeling Pittsburgh Steel Corp., 738 F.2d 1393, 1396 (3d Cir.) ("a showing that a proffered justification is pretextual is itself equivalent to a finding that the employer intentionally discriminated" (emphasis in original)), cert. denied, 469 U.S. 1087, 105 S. Ct. 592, 83 L. Ed. 2d 702 (1984).

As this court explained, sitting in banc, in Chipollini v. Spencer Gifts, Inc., 814 F.2d 893, 897 (3d Cir.), cert. dismissed, --- U.S. ----, 108 S. Ct. 26, 97 L. Ed. 2d 815 (1987), plaintiff retains the ultimate burden of proving discrimination. However, that burden may be met "indirectly by showing that the employer's proffered explanation is unworthy of credence." Id. at 898 (quoting Texas Dep't of Community Affairs v. Burdine, 450 U.S. 248 at 256, 101 S. Ct. 1089 at 1095, 67 L. Ed. 2d 207 (1981)) (emphasis omitted). Thus, "a plaintiff can prevail by means of indirect proof that the employer's reasons are pretextual without presenting evidence specifically relating to [race]."20 Id. at 898. If a plaintiff produces credible evidence that "the employer did not act for its proffered reason, then the employer's decision remains unexplained and the inferences from the evidence produced by the plaintiff may be sufficient to prove the ultimate fact of discriminatory intent." Id. at 899. Hence, a "defendant which is less than honest in proffering its reason for discharge risks an unnecessary ... discrimination verdict." Id.

Moreover, in spite of Drexel's claim to the contrary, a plaintiff in a discrimination case need not prove intentional discrimination at every stage of the review process. It is true that University guidelines and witness testimony support Drexel's claim that each successive evaluator performed a de novo review of Roebuck's candidacy. Nevertheless, it is also uncontroverted that at each stage of the process the evaluator had available and considered the reports and recommendations of each previous evaluator. See, e.g., J.A. at 550-51 (testimony of Dean Canavan) (Collegial Committee reviews all of candidate's tenure materials, including the department committee's and department head's recommendations); J.A. at 778 (testimony of Vice President Sagik) (in considering tenure applications, Sagik reads the various committee recommendations, although he attempts first to form an independent judgment); J.A. at 285 (testimony of Roebuck) ("In my conversations with Dr. Sagik, Dr. Sagik told me that [Dr. Cannistraro's letter] was the thing that killed me.... [it] was the factor that killed my bid for tenure"); J.A. at 2474 (letter from President Hagerty explaining denial of Roebuck's appeal) ("My review included all materials submitted to me by you, by Dr. Cannistraro and by Dean Canavan."). Hence it plainly is permissible for a jury to conclude that an evaluation at any level, if based on discrimination, influenced the decisionmaking process and thus allowed discrimination to infect the ultimate decision.

III. SUFFICIENCY OF EVIDENCE OF DISCRIMINATION

On appeal, Roebuck points to numerous items of evidence that he claims support the jury's verdict, focusing primarily on white candidates for tenure who Roebuck contends were treated more favorably than was Roebuck, hence showing disparate treatment. See generally Texas Dep't of Community Affairs v. Burdine, 450 U.S. 248, 101 S. Ct. 1089, 67 L. Ed. 2d 207 (1981) (setting forth the respective burdens and order of proof in a disparate treatment case). We do not find any of the white tenure candidates discussed by plaintiff to be sufficiently comparable to support his claim. See infra note 36. Nevertheless, although we freely concede that the evidence in support of the jury verdict is razor-thin, we find enough evidence in the record from which a reasonable jury, using the common sense lay wisdom for which juries are so often praised, could infer that Roebuck would have been granted tenure but for his race. See, e.g., Dreyer v. Arco Chem. Co., 801 F.2d 651, 654 (3d Cir.1986) (court upholds jury verdict in discrimination suit because, after reviewing the record "in the light most favorable to the non-moving party," the court did not find it to be "critically deficient of that minimum quantum of evidence from which the jury might reasonably afford relief") (citations omitted), cert. denied, --- U.S. ----, 107 S. Ct. 1348, 94 L. Ed. 2d 519 (1987). Hence, we next will discuss the evidence supportive of the verdict with respect to each of the three relevant tenure criteria.

Aside from the departmental committee, which rated his service "outstanding" and lauded his role as a bridge between Drexel and the West Philadelphia community as "almost impossible to duplicate," see supra page 721, reviewers at each successive stage of the tenure decision process rated Roebuck merely "satisfactory" in service.21 We believe, however, that Roebuck presented sufficient evidence of disparate treatment, in terms of more exacting standards and devaluation of his contributions, from which a jury could infer that, but for his race, Roebuck would have been rated "outstanding" in service.

A tenure candidate in Drexel's College of Humanities and Social Sciences is entitled to have his service judged according to the criteria published in the "Departmental and Collegial Review Procedures for Tenure." As we noted supra, Part I.B., this document states that the tenure decision involves consideration of the candidate's "service to Drexel" and then explains that decisionmakers should evaluate the candidate's "record of performance in ... University, collegial, and departmental service, and community service relevant to the mission of the institution." J.A. at 1354.

Dr. Cannistraro, the Department Chair, found Roebuck's service merely "satisfactory," largely because he questioned "the degree to which this kind of off-campus [community] service should form the basis for judging professional accomplishment." J.A. at 2035. Cannistraro's report appeared to assert that Roebuck's service was not "relevant to the mission of the University," and did not constitute "service to Drexel," as required by the University and College guidelines. Id. See supra Part I.D.2.

Such assertions, however, are plainly contrary to one of the very purposes for which Roebuck was hired. As explained supra, Part I.C., a jury could conclude that Roebuck was hired in large part because of his ability to interact with the surrounding West Philadelphia community, and that such service was uniquely valuable to Drexel because of the strained relations between Drexel and its neighbors. Moreover, many of Roebuck's community service activities were performed at the behest of Drexel administrators. See J.A. at 54-57, 856. Given the evidence discussed supra, Part I.C., that Roebuck's service was both appreciated and found valuable by the University,22 a jury could infer that Cannistraro's claim that Roebuck's service was not "relevant to the mission of the University" was not credible and was, in fact, a pretext.

In so concluding, we emphasize the case-specific nature of this determination, consistent with the standards we have described, supra Part I.B. Thus we do not doubt Drexel's right, in another case, to emphasize on-campus service and to relegate extramural service to a minor role. But, having set the ground rules for Roebuck's employment and thereby defined an important category of service, when Drexel acts as though those rules do not exist when the worth of Roebuck's stewardship is at issue, a pretext question may arise.

Cannistraro also emphasized Roebuck's lack of service "in the professional sphere," noting, for example, Roebuck's lack of work with professional committees in his field of specialization. Cannistraro's report asserted that "[i]t is in this area that the faculty member brings to his department and university recognition in scholarly circles and enhances the institution's academic standing," J.A. at 2035, implying that recognition gained by other types of service are not as valuable. Yet professional service is not listed in any University document as a relevant criterion in the tenure decision, and Dr. Michael Sullivan, an associate professor in Roebuck's department, explained that none of the categories of service listed by Drexel as relevant to tenure includes the category of service to the profession.23 J.A. at 415. Moreover, Sullivan testified that a requirement of professional service never before had been applied to any other member of the history-politics department, a factor that could give rise to an inference of pretext.24

Similarly, Dean Canavan testified that he could not rate Roebuck's service as outstanding because Roebuck had not "taken a leadership role of significant importance to have created some kind of change either in [his] department or in the college or in the University at large." J.A. at 616-17. However, nowhere in the tenure guidelines does there appear a requirement that either "leadership" or "some kind of change" be displayed in order to qualify for an "outstanding" rating. Moreover, none of the other evaluators appeared to apply this standard. Finally, a jury could infer from Roebuck's extensive resume, and particularly from his participation on numerous boards of directors, that Roebuck did, in fact, meet this standard by demonstrating leadership, see supra note 4, and similarly, that his activities did, in fact, cause a significant change in the University's community relations, see, e.g., supra note 22. Hence, a jury could infer that the dean's assertion did not reflect the true reasons for his unenthusiastic evaluation of Roebuck's service.

Additionally, President Hagerty, in denying Roebuck's appeal, conceded that Roebuck's service activities were "above average." Nevertheless, Hagerty belittled the value of these activities by noting that Roebuck was "not seeking a public relations appointment." Furthermore, he asserted, such activities could not "substitute for either teaching or scholarship," which he claimed were the "primary matters affecting tenure." J.A. at 2476-78. The University's and College's own guidelines, as Hagerty conceded, place equal weight on teaching, scholarship and service, and those guidelines allow a candidate to overcome merely satisfactory teaching and scholarship with an outstanding record of service. Such comments by Hagerty therefore could be interpreted either as a devaluation of Roebuck's concededly extensive service contributions or as yet another application of a thoroughly inappropriate standard to Roebuck's application.

Similarly, Vice President Sagik asserted that Roebuck's service was merely "satisfactory" because "much of it was not related to the University." Yet he conceded that service that redounds "to the advantage both of the University [and] the community, and makes use of the faculties' expertise" would be "related to the mission of the University." See supra Part I.D.5; J.A. at 780-81. The jury certainly could infer from the evidence that Roebuck's activities fit this bill precisely, hence any refusal to so admit constituted an intentional devaluation of Roebuck's service. Moreover, there was extensive evidence of Roebuck's on-campus contributions which the departmental committee found extremely impressive, see supra, Part I.D.1, but which Dr. Sagik and the other upper-echelon evaluators seem to have ignored.

In sum, we believe that a rational jury could infer from the cumulative weight of this evidence that Roebuck has presented sufficient evidence to cast into doubt the credibility of Drexel's assertions, and that the reasons proffered by the University for its failure to rate Roebuck's service as "outstanding" were mere pretexts. There is a slightly different, though cognate, way of looking at the matter. Under this approach, an inference that could have been drawn by the jury is that these reasons, though not pretexts, reflected a strained effort by Drexel to devalue Roebuck's service activities relative to more traditional forms of service activity. In this respect, Drexel's refusal to acknowledge the evident value of Roebuck's efforts to bring the University and the (black) West Philadelphia community together (efforts which Drexel itself recognized were successful primarily because Roebuck himself was a member of that black community) evinced an arbitrariness that constituted constructive racial discrimination.

We therefore conclude that there was sufficient evidence for a rational jury to find that, but for his race, Drexel would have rated Roebuck's service "outstanding."

Drexel concedes that Roebuck's teaching was "satisfactory." Roebuck, however, contends that he would have been rated an "outstanding" teacher, but for his race. Although there was enough evidence for a jury to conclude, independently, that Roebuck's teaching was outstanding, the question before us is whether there was enough evidence to conclude that Drexel's failure to rate his teaching outstanding was racially motivated.

Two of Roebuck's colleagues from the History-Politics Department testified that they believed Roebuck's teaching to be "outstanding." Dr. Richard Rosen, a member of the departmental committee that recommended tenure for Roebuck, testified that he had reviewed Roebuck's record and found him to be "an excellent teacher," which he equated with "outstanding." Moreover, he testified that Roebuck's teaching was on a par with his own, and Rosen was a past recipient of a University award for outstanding faculty. See J.A. at 344-45. Similarly, Dr. Raymond Lorantas, another history professor and a recipient of many teaching awards, both from the University and from student groups, testified that he knew Roebuck well and had co-taught a course with him, and he believed Roebuck's teaching to merit a rating of "outstanding." See J.A. at 525-26, 53538.25 Roebuck also produced evidence that on several occasions Dr. Arian, his former department chairman, had described his teaching as "outstanding," both in a letter of recommendation and in several annual evaluations.

There also is evidence in the record that at least one tenure reviewer applied an unfair, and perhaps insurmountable, standard to Roebuck's teaching credentials. Dr. Cannistraro, in his report recommending the denial of tenure, apparently refused to give much weight to the student evaluations. After noting the departmental committee's extensive reliance on these evaluations, he stated: "It would also be legitimate to ask what we know about Dr. Roebuck's teaching from a sampling of 74 students out of the almost 3000 he has taught, a sampling from four classes during the most recent of his twelve years of teaching." J.A. at 2034 (emphasis added). Yet Roebuck testified that the student evaluations included in his tenure dossier "covered the entire period in which there was a student evaluation procedure in placement." J.A. at 174. Moreover, he testified that tenure candidates were not supposed to supply the tenure decisionmakers with self-made evaluation forms; rather, they "were to use the collegial forms and the collegial forms only." J.A. at 175. Finally, it was apparent from the evidence that it was standard practice at Drexel to assess the teaching performance of tenure candidates based largely on these student evaluations. Hence, a jury could infer that Cannistraro's reasons were pretextual, particularly given his willingness to apply an inappropriately rigorous standard to Roebuck's service, as well, see supra, Part III.A.

We recognize that this evidence is very thin; any flaw in Roebuck's proof on this point, however, is not fatal to his case because a candidate for tenure need only be rated outstanding in one of the three categories. Given that we find enough evidence to support a finding that Roebuck would have been rated outstanding in service but for Drexel's discrimination, Roebuck need not prove that he likewise would have been rated outstanding in teaching.

The evidence to which Roebuck points to support a finding of discrimination in the rating of his scholarship is far weaker than that supporting the finding of discrimination in service and hence much more troubling. Roebuck's achievements in the realm of scholarship plainly were minimal, in terms of both quantity and quality. The University produced substantial amounts of evidence pointing to the deficiencies in Roebuck's scholarship. However, we do not sit as a super tenure review board, see Keddie v. Pennsylvania State University, 412 F. Supp. 1264, 1270 (M.D.Pa.1976), or as a super jury. It is not for us to weigh the evidence and determine whether we agree with the University's assessment or with the jury's verdict. We sit only to determine whether there exists that minimum quantum of evidence necessary to support the jury's finding that Roebuck's race was a determinative factor in Drexel's decision to rate his scholarship "unsatisfactory."

We reiterate that Roebuck need not prove discrimination by direct evidence. A "smoking gun" showing discriminatory intent will rarely be found, cf. Gavalik v. Continental Co., 812 F.2d 834, 852 (3d Cir.), cert. denied, --- U.S. ----, 108 S. Ct. 495, 98 L. Ed. 2d 492 (1987); discrimination today generally takes subtler and less apparent forms. See Chipollini v. Spencer Gifts, Inc., 814 F.2d 893, 899 (3d Cir.) (in banc), cert. dismissed, --- U.S. ----, 108 S. Ct. 26, 97 L. Ed. 2d 815 (1987); Jackson v. University of Pittsburgh, 826 F.2d 230, 236 (3d Cir.1987), cert. denied, --- U.S. ----, 108 S. Ct. 732, 98 L. Ed. 2d 680 (1988). Hence circumstantial evidence of discrimination, e.g., a showing of pretext, can suffice to win a verdict under Sec. 1981. Moreover, a jury is free to draw reasonable inferences from the evidence presented, using the common sense for which lay juries are most valued.

As a prerequisite for a finding of liability under Sec. 1981, a plaintiff first must present evidence that he was qualified for the position. See supra Part II. We believe that Roebuck presented enough evidence for a jury to conclude that he was at least qualified, in terms of scholarship, for an award of tenure. In spite of the substantial evidence to the contrary, there was evidence which the jury was free to credit to support a conclusion that Roebuck's scholarship was "satisfactory."

The departmental committee, composed of three faculty members in Roebuck's academic discipline, unanimously found Roebuck's scholarship satisfactory and recommended that Roebuck not only receive tenure but promotion as well.26 Even more significant, several external reviewers were favorably impressed with Roebuck's work. Drexel, on appeal, emphasizes the substantial negative commentary contained in these reviews. We, however, must examine these reviews in the light most favorable to Roebuck.

Of the six reviewers, three can fairly be read to recommend tenure and three to deny. Although his American Liberals paper was uniformly criticized, Roebuck's Taft paper was warmly received by four of the reviewers. The three who recommended tenure found it to be "a rather good piece," J.A. at 2028, to be "valuable," J.A. at 2029, and to demonstrate "both serious thought and analytical ability," J.A. at 2042. His work was described as "professional," J.A. at 2042, his writing was favorably reviewed, J.A. 2029, 2042, and the Taft piece was found to "provide[ ] useful information on what and who influenced [Taft's] thinking and positions," J.A. at 2029. Moreover, one of the reviewers who recommended denial of tenure found the Taft piece to be "satisfactory," "useful," and deserving of publication in a refereed journal; his recommendation was explicitly based on lack of quantity, rather than quality. J.A. at 2025-26. Another reviewer who recommended denial was highly critical of Roebuck's work, but then noted as a counterweight that the majority of the history faculty at Drexel were not even listed in "The Directory of American Scholars," implying that perhaps he was applying too high a standard. J.A. at 2033.27

Moreover, the jury had before it the scholarship itself, about which the jury, having heard testimony about academic standards, was free to make an independent assessment. We believe that the cumulative weight of this evidence was sufficient for the jury to find that Roebuck's scholarship was "satisfactory." The jury's assessment, based on its review and the reviews of others, that Roebuck's scholarship was satisfactory is, of course, not enough to support a finding of discrimination. It is, however, enough to support a finding that Roebuck met the minimum qualifications for an award of tenure, which, in combination with other evidence, can support the ultimate conclusion. That other evidence primarily takes three forms.

First, the record contains traditional pretext evidence. Dean Canavan testified that he relied heavily on the opinions of the external reviewers in formulating his judgment, and he declared that "the majority of them in my reading of these reviews took a negative position, both on Dr. Roebuck's achievement and potential." J.A. at 610. However, as already shown, the jury was free to conclude that these reviewers were at worst split in their assessments, and that even some of the detractors were somewhat tentative in their negative conclusions. Specifically, Canavan testified that one external reviewer, Dr. Norman Graebner, concluded that, "if the issue had to turn on scholarship, Dr. Roebuck would not be tenured." J.A. at 607. Yet Graebner's review in fact praised Roebuck's Taft article and spoke warmly of Roebuck's capabilities as a scholar. J.A. at 2028.28 He also stated:

[Roebuck's] elevation to tenure, which for you seems to require less than promotion, must be based on a combination of service and good teaching. I see no reason why these two areas of contribution, if the contribution is as high as the record seems to indicate, should not merit tenure and perhaps promotion as well.

Id. Although the statement is arguably more hortatory than surgically precise, Graebner was plainly familiar with Drexel's tenure standards, and his recommendation that Roebuck be accorded tenure because of his impressive record of service and teaching indicates that Graebner found Roebuck's scholarship satisfactory, a prerequisite for tenure at Drexel. At the very least, the jury was free to infer this implicit conclusion. The important point, however, is that the jury was also free to find Dean Canavan's assessment of this letter not credible and, in fact, a pretext.

Second, there is some direct evidence of a discriminatory attitude on the part of one key decisionmaker. Roebuck testified that in 1977 or 1978 he attended a meeting at which President Hagerty was asked why Drexel does not have more black faculty and why Drexel was not trying to get more black faculty. Hagerty's response, according to Roebuck, was that "blacks cost more." Hagerty then explained, in substance, that "in terms of comparable white faculty members ... blacks would cost Drexel more money to hire those black faculty members."29 Yet Roebuck claimed that, "to the best of my knowledge," his salary when hired was not out of line with whites in his department. Regardless of the truth of Hagerty's generalization (and Roebuck testified that, to his knowledge, it was not true), such a statement could be found to reflect a willingness on Hagerty's part to stereotype and make negative decisions based on race.

The jury was entitled to infer that Hagerty, as president of the University, had a significant influence on the attitudes and procedures of the tenure decisionmakers. Moreover, Hagerty had a direct involvement in the process, in that he made the final decision to deny Roebuck's appeal, culminating in a five page report. See J.A. at 2474-78. Although Hagerty's statements standing alone, occurring as they did over five years before the final denial of tenure, could not suffice to uphold a finding that Drexel discriminated against Roebuck, they do add support, in combination with the other evidence, to the ultimate conclusion.30 See Graefenhain v. Pabst Brewing Co., 827 F.2d 13, 20 (7th Cir.1987) ("to require ... that each piece of circumstantial evidence 'standing alone' be sufficient to support a finding of ... discrimination is to render meaningless the indirect method of proof").

Finally, the jury had before it the evidence of discrimination in the rating of Roebuck's service and teaching. These rating decisions were not made in a vacuum; the same persons reviewing Roebuck's service and teaching were simultaneously reviewing Roebuck's scholarship. In fact, at at least one level of the process (the Collegial Committee's review), separate votes were not even taken on each prong of the tenure decision; the committee merely voted yea or nay on the ultimate determination. Hence, once the jury concluded that a particular rating decision at any given level was motivated by discrimination and would have come out differently but for that discrimination, the jury also was free to infer that that discrimination infected other decisions concerning the very same candidate and made at the very same time.

For example, given Dr. Cannistraro's willingness to apply a more rigorous standard to Roebuck's service activities than that called for in college guidelines (or, alternatively, given a jury finding that Cannistraro's asserted explanation was merely pretextual), a jury reasonably could conclude that Cannistraro likewise was willing to treat Roebuck's scholarship more harshly than it otherwise would deserve. Put differently, once a jury has concluded that Cannistraro would have rated Roebuck's service "outstanding" but for his race, common sense allows the jury to conclude that Cannistraro was similarly willing to allow race to determine his rating of Roebuck's other qualifications.31 And, as already noted, each successive decisionmaker had before it the decisions and reports of the prior levels, hence any decision which itself was infected with discrimination in turn would have tainted the entire process.32

In sum, given the fact that a jury could find, on this record, that Roebuck was qualified for tenure, i.e., that his scholarship was deserving of a satisfactory rating, there was sufficient evidence--in the form of a pretextual explanation, a discriminatory statement and the possibly infectious discrimination in the rating of the other criteria--for a jury to conclude that Drexel's failure to rate his scholarship "satisfactory" was because of Roebuck's race.

We therefore conclude that, although we might well have decided the question differently de novo, there was sufficient evidence for a jury to conclude that, but for his race, Drexel would have rated Roebuck outstanding in service and satisfactory in scholarship,33 and hence would have granted him tenure. We therefore must reverse the district court's order granting judgment n.o.v.

Concurrent with its motion for judgment n.o.v. Drexel moved, in the alternative, for a new trial, asserting that the jury's verdict on the Sec. 1981 claim was against the weight of the evidence. Drexel asserts on appeal that the district court granted only the motion for judgment n.o.v., but did not rule on the alternative motion for new trial. It is true that the court's lengthy opinion addressed only the judgment n.o.v.; however, the relevant accompanying order reads, in pertinent part, "after consideration of defendant's motion for judgment notwithstanding the verdict or, alternatively, a new trial, and plaintiff's response thereto, it is hereby ORDERED that the motion is GRANTED." J.A. at 1304. We read this order as a grant of both motions, with the grant of a new trial plainly conditioned upon a reversal of the judgment n.o.v. This would be in accord with the requirements of Fed.R.Civ.P. 50(c)(1), which states that when a motion for judgment n.o.v. is granted, "the court shall also rule on the motion for a new trial, if any, by determining whether it should be granted if the judgment is thereafter vacated or reversed." See 9 C. Wright & A. Miller, Federal Practice & Procedure Sec. 2539, at 609-10 (1971).34

The consequence of the district court's alternative ruling is made plain by the rule: "In case the motion for a new trial has been conditionally granted and the judgment [n.o.v.] is reversed on appeal, the new trial shall proceed unless the appellate court has otherwise ordered." Fed.R.Civ.P. 50(c)(1). We therefore must determine whether we should allow the new trial to proceed, or otherwise order. We make this determination with due regard for our highly deferential scope of review. See American Bearing Co. v. Litton Industries, 729 F.2d 943, 948 (3d Cir.) (" 'The authority to grant a new trial ... is confided almost entirely to the exercise of discretion on the part of the trial court,' and will only be disturbed if the district court abused that discretion.") (quoting Allied Chemical Corp. v. Daiflon, Inc., 449 U.S. 33, 36, 101 S. Ct. 188, 191, 66 L. Ed. 2d 193 (1980)), cert. denied, 469 U.S. 854, 105 S. Ct. 178, 83 L. Ed. 2d 112 (1984). Such deference is peculiarly appropriate in reviewing a ruling that a verdict is against the weight of the evidence because the district court was able to observe the witnesses and follow the trial in a way that we cannot replicate by reviewing a cold record. See Semper v. Santos, 845 F.2d 1233, 1237 (3d Cir.1988).35

We also are cognizant of the precept that a new trial may be granted even when judgment n.o.v. is inappropriate. American Bearing Co., 729 F.2d at 948 n. 11; see, e.g., Rousseau v. Teledyne Movible Offshore, Inc., 812 F.2d 971, 972 (5th Cir.) (affirming grant of new trial even though there was "legally sufficient evidence to support the verdict, thus foreclosing a j.n.o.v."), cert. denied, --- U.S. ----, 108 S. Ct. 95, 98 L. Ed. 2d 56 (1987); 9 Wright & Miller Sec. 2539 at 608 ("In many instances the judge might grant a new trial on th[e] ground [that the verdict is against the weight of the evidence] even though he is constrained to refuse to order judgment [n.o.v.]."). Such a motion should be granted when, in the opinion of the trial court, the verdict is contrary to the great weight of the evidence, thus making a new trial necessary to prevent a miscarriage of justice. See generally 9 Wright & Miller Sec. 2531 at 575-76; 6A J. Moore, Moore's Federal Practice p 49.08, at 59-140 to 59-150 (2d ed. 1986).

We cannot say that the trial court abused its discretion in the instant case in granting the motion. Most of Roebuck's case focused on putatively comparable white candidates who were granted tenure. However, we agree with the district court that none of these candidates was sufficiently comparable to Roebuck to justify an inference of disparate treatment.36 Moreover, although we have discussed the record, as we must, in the light most favorable to Roebuck, we cannot ignore the overwhelming evidence that Roebuck's scholarship simply was not up to University standards.37

We must also acknowledge the extraordinary number of inferences that the jury must have drawn in order to reach the verdict that it did. Although we believe that each of the inferences that we have discussed are individually logically sound, we recognize that at some point too many inferences become mere speculation; this verdict plainly comes very close to that line.

Finally, a number of the arguments upon which we rely to uphold the verdict--in particular the argument that discrimination in rating Roebuck's service spilled over into the assessment of his scholarship--do not appear to have been made, in terms, by Roebuck himself at trial. We have exercised our obligation to comb the record for any evidence or inferences that could support the jury's verdict, and hold that a jury could have drawn the inferences and conclusions that we have extensively discussed, hence our reversal of the judgment n.o.v. However, we have no reason to believe that these inferences actually were drawn by this jury. Hence we are unwilling to allow the verdict to stand, particularly given the substantial emphasis placed at trial on the arguments which we view as unsupportable. See, e.g., supra note 36.

Thus for the foregoing reasons, although Roebuck presented enough evidence to the jury to preclude granting judgment against him as a matter of law, we will affirm the district court's grant of a new trial on the grounds that the verdict was against the great weight of the evidence.

Roebuck argued to the district court and renews the argument here that the jury's verdict in his favor on the Sec. 1981 claim is preclusive of an adverse judgment on his Title VII claim. The issue no longer is presented to us in precisely that form because we have affirmed the district court's decision to set aside the jury verdict and grant a new trial, hence it plainly cannot have preclusive effect at this time. However, the issue is still before us because, were we to decline to rule on it, we would leave in limbo the district court's Title VII judgment and leave open the possibility that that judgment itself could have preclusive effect upon retrial of Roebuck's Sec. 1981 claim.38

The question is one of first impression in this circuit; however, with one possible exception, every circuit to have ruled on the issue has held that the jury's findings on a Sec. 1981 claim are binding on the trial judge's resolution of a concurrently tried Title VII claim. See Ward v. Texas Employment Comm'n, 823 F.2d 907, 908-09 (5th Cir.1987); Williamson v. Handy Button Machine Co., 817 F.2d 1290, 1293-94 (7th Cir.1987); Garza v. City of Omaha, 814 F.2d 553, 557 (8th Cir.1987) (Sec. 1983 judgment preclusive in Title VII suit); Lincoln v. Board of Regents, 697 F.2d 928, 934 (11th Cir.), cert. denied, 464 U.S. 826, 104 S. Ct. 97, 78 L. Ed. 2d 102 (1983).39 We view these holdings as consistent with Supreme Court precedent, as well as with the respect that is properly accorded to a jury verdict in our system of jurisprudence.

Although not addressing this precise issue, the Supreme Court has stated quite clearly that where there are "separate and distinct statutory provision[s]," one of which authorizes legal relief and one of which authorizes equitable relief, "if a 'legal claim is joined with an equitable claim, the right to jury trial on the legal claim, including all issues common to both claims, remains intact.' " Tull v. United States, --- U.S. ----, 107 S. Ct. 1831, 1839, 95 L. Ed. 2d 365 (1987) (quoting Curtis v. Loether, 415 U.S. 189, 196 n. 11, 94 S. Ct. 1005, 1009, 39 L. Ed. 2d 260 (1974)) (emphasis added). A trial judge should not be allowed to ignore a considered decision of a jury merely because the judge views the issues differently, for otherwise the seventh amendment right to a jury trial would be significantly attenuated.40

Applying this principle of preeminence of jury verdicts to the instant case, we think it apparent that the issue of intentional race discrimination is "common to both [the Title VII and Sec. 1981] claims," and hence must be decided by the jury for purposes of both claims. Thus, because the elements of proof are identical in Sec. 1981 and Title VII claims, see Lewis v. University of Pittsburgh, 725 F.2d 910, 915 n. 5, we hold that a trial court is bound in a Title VII case to conform its verdict to the findings of the jury in a concurrently tried Sec. 1981 case.41

In this case, however, the district court has set aside the jury verdict as against the weight of the evidence, and we affirm this decision, supra, Part IV. Hence, the jury has not yet ruled conclusively on the question of discrimination and thus no findings exist to control the district court's decision. Nevertheless, Roebuck did present enough evidence to the jury to avoid judgment n.o.v. and therefore retains his right to a jury trial on the Sec. 1981 claim. Concomitant with that right is a right to a jury determination of all issues common to both the legal and equitable claims. We therefore hold that a district court should not render judgment on a Title VII claim prior to the return of the jury verdict on a related Sec. 1981 claim. Thus we will vacate the Title VII judgment in Drexel's favor with instructions to the district court to await the jury's findings on retrial before entering final judgment.

A contrary result would invite inconsistent judgments. Had the jury verdict been upheld, the district court would have been constrained to conform its judgment on the Title VII claim. We do not see why this should differ merely because of the fortuity that the jury verdict, in the opinion of the district court, could not stand. A refusal to vacate would also invite the circular argument that the Title VII judgment, left standing by this court, itself should preclude an inconsistent judgment on retrial of the Sec. 1981 claim.

We acknowledge that in Ritter v. Mount St. Mary's College, 814 F.2d 986, 990-92 (4th Cir.), cert. denied, --- U.S. ----, 108 S. Ct. 260, 98 L. Ed. 2d 217 (1987), the court held that a district court's findings in a Title VII suit are preclusive in a subsequent trial to a jury on an ADEA claim, even though the ADEA claim itself was filed jointly with the Title VII claim but had been erroneously dismissed by the district court.42 However, the Fourth Circuit appears to have cut back on its Ritter holding in Swentek v. USAir, 830 F.2d 552 (4th Cir.1987). In Swentek, a panel of the same court held that the district court's prior adverse findings in a Title VII suit claiming sexual harassment would not be binding in the jury trial of plaintiff's emotional distress claim, in spite of the concordance of the issues, where the Title VII and emotional distress claims initially had been tried together but the district court had discarded the jury's verdict for the plaintiff on the emotional distress claim and granted defendant a new trial. Id. at 562-63. We agree with the Swentek court that "[i]t would be anomalous to have a judge's award of a new trial nullified by the equitable findings of the same judge who ordered it." Id. at 563.43

Therefore, to avoid the problems faced by the Fourth Circuit in Ritter and Swentek, we believe that the better course is that followed by the Seventh Circuit in Volk v. Coler, 845 F.2d 1422 (7th Cir.1988). In Volk, the court held that where plaintiff had presented sufficient evidence on her Secs. 1983 and 1985(3) claims to allow the case to go to the jury, but the district court had improperly taken the case away from the jury, plaintiff was "entitle[d] to a jury trial on the [legal] claims before the trial court decides her Title VII equitable claims." Id. at 1438 (emphasis added). Hence, the court set aside the district court's premature Title VII judgment, and we do likewise. Cf. Hussein v. Oshkosh Motor Truck Co., 816 F.2d 348 (7th Cir.1987) (Title VII judgment will not be preclusive on retrial of plaintiff's Sec. 1981 claim; however, Title VII judgment will not be vacated because it was not appealed).44

For the reasons explained above, the district court's grant of judgment n.o.v. on the Sec. 1981 claim will be reversed, the court's grant of a new trial on that claim will be affirmed, the Title VII judgment will be vacated, and the case remanded for further proceedings.