Edwin B. Allaire, Philip White v. Lorene L. Rogers

TATE, Circuit Judge: The plaintiffs, eight politically active tenured professors at the University of Texas at Austin (UTA), brought this 42 U.S.C. § 1983 action against UTA President Lorene Rogers and the UTA Board of Regents after being denied the full salary increases for which they had been recommended for the fiscal year 1975-1976. The plaintiffs alleged that this salary action was taken in retaliation against their exercise of their first amendment right to freedom of speech and association. After a bench trial, the district court ruled that the plaintiffs were not entitled to the requested injunctive, declaratory, and monetary relief. Contending that the district court’s finding that the plaintiffs had failed to prove retaliatory intent was clearly erroneous, three of the plaintiffs, Professors White, Gavenda, and Shepley, now bring this appeal. We find that the district court’s decision was clearly erroneous only with respect to plaintiff-appellant White. Therefore, the decision must be affirmed in part, reversed in part, and remanded for further proceedings in accordance with this opinion. *1056 I. The Factual Background A. The Controversial Activities of Professors White, Gavenda, and Shepley In 1975, Professor Gavenda was president, and Professor White was legislative representative and past president, of the UTA chapter of the Texas Association of College Teachers (TACT). This organization was involved in lobbying the state legislature for additional funds for faculty salary increases. To further this lobbying effort, White prepared a report entitled “Excellence at UT Austin,” in which he demonstrated, among other things, that UTA faculty salaries had rapidly declined in comparison with other leading public universities, and that UTA had the worst student-faculty ratio of any major university in the United States. The report was distributed to the members of the UTA Board of Regents and to legislative leaders. Gavenda, along with Professor Hill, one of the plaintiffs below who have not appealed, attempted to further the cause by appearing before the state legislature’s budget board and by lobbying several individual legislators. There is some question how effective the TACT lobbying efforts were. The defendant President Rogers testified that certain legislators and others had complained to her that Gavenda and Hill had presented misleading data. These individuals claimed that Gavenda and Hill had included teaching assistants in determining the average salary at UTA but had not included them in determining the salaries of the other universities with which UTA was being compared. (Although Rogers could not recall any criticism of White, presumably Gavenda and Hill were using the figures from his report.) Furthermore, some legislators questioned how supposedly full time faculty members could be spending so much time at the legislature. Regardless of whether the TACT lobbying efforts had any positive or negative impact, in 1975 the Texas legislature allocated UTA money for faculty merit raises for the first time in six years. At the same time these events concerning faculty salaries were occurring, the plaintiff-appellant Professor Shepley became involved in an unrelated controversy with the UTA administration. At that time Shepley was serving as president of the Austin Chapter of the American Association of University Professors (AAUP), an organization dedicated to promoting academic freedom and preserving teachers’ rights. After being contacted by members of the faculty of the University of Texas Nursing School at Austin, he wrote several letters on behalf of the nursing faculty with regard to alleged violations of the rules regarding promotion to tenure. Shepley admitted, and it is clear from the content of the letters, that he wrote both as an individual faculty member and as president of the AAUP. He had these letters typed by a physics department secretary, on UTA stationery, and then transmitted the communications through the university mail system. President Rogers considered these actions to be violations of a university rule limiting use of university services and supplies to “official budgetary units.” 1 She informed Paul Olum, the dean of Shepley’s department, of the violation and asked him to investigate the possibility of disciplinary action. Dean Olum responded that in his opinion no disciplinary action should be taken in light of the vagueness of the rule and Shepley’s good faith misunderstanding as to what was permissible thereunder. In addition to the above incidents, all three of the plaintiffs had long been involved in other controversial political activities. They were the leaders of TACT and AAUP, organizations that often were critical of the administration. Also, Shepley and Gavenda had been involved in the university-wide protest concerning the firing of Dr. Steven Spurr, the defendant Rogers’ predecessor as UTA president. B. The Controversial Decisions of President Rogers President Rogers was responsible for determining the raise each faculty member *1057 received, subject to final review by the Board of Regents. She reached her decisions after receiving recommendations from three sources: departmental budget councils made up of faculty members, the chairpersons of each department, and the deans of each college. The budgetary process was conducted in two phases. In the first phase, both the recommendations and the president’s budget decisions were made on the assumption that each department would receive 5% more for faculty salaries than it had the year before. However, due to the additional money allocated by the legislature for merit raises, a total of 6.8% actually became available. Therefore, in the second phase, President Rogers distributed the amount over 5% in the form of $400 merit raises to approximately two-thirds of the faculty. New employees and those already receiving over $35,000 a year were automatically excluded from consideration for the $400 raise. Rogers’ salary decisions concerning the three plaintiffs-appellants gave rise to this lawsuit. Although each plaintiff-appellant did in fact receive a raise, the increases they received were reduced by Rogers to less than the amounts recommended by their deans, department chairmen, and budget councils. Shepley’s dean, department head, and budget council all recommended that he receive a $1000 raise; President Rogers approved a $400 increase with no merit raise. The budget council in Gavenda’s department recommended a $1500 increase for him while his department chairman and dean thought $1000 appropriate; President Rogers gave him a $500 increase and the $400 merit raise for a total of $900. Shepley and Gavenda thus became two of the 78 faculty members, out of a total of approximately 2000, who had their recommended salaries reduced by the president. Rogers’ actions concerning Professor White were somewhat more atypical. First, she approved the full $1900 increase recommended by White’s departmental administrators, and, in addition, she awarded him the $400 merit raise, which meant that White would receive a total raise of $2300. A few weeks later, however, Rogers learned that, although the Board of Regents had the authority to override all budget decisions, they would review only increases of over $2000 and that she might be required to justify raises of such magnitude. Subsequently, she withdrew her award of the $400 merit raise to White so that White’s total increase amounted to only $1900, safely below the $2000 figure that would trigger Board scrutiny. Of the more than 200 faculty members who were initially awarded increases greater than $2000, White was the only one whom Rogers reduced after she learned of the Board’s policy. Moreover, White was the only member of the history department who did not receive the $400 merit raise. Even three members of the department who were recommended for no increase at all and two who were on “terminal appointment,” that is, to be fired at the end of that school year, were found by Rogers to be worthy of this supposedly “merit” award. C. The Proceedings Below White, Gavenda, Shepley, and the five other professors who have not appealed, felt that Rogers had reduced their recommended salaries in retaliation for their political activities. After efforts by various intermediaries to persuade Rogers to change her mind failed, the eight professors brought this suit. After a two-day bench trial, the district court found that the plaintiffs had done little more than “hypothesize” that Rogers had retaliated against them for engaging in the controversial activities. Accordingly, the court granted judgment in favor of the defendants. II. The Applicable Legal Principles A. The Substantive Law Applicable to the Plaintiffs’ Claim The sole issue on appeal is whether the district court’s conclusion that the plaintiffs failed to prove retaliatory intent is clearly erroneous. The professors concede, as they must, that they are entitled to relief only if President Rogers reduced their recom *1058 mended salaries in retaliation for the exercise of constitutionally-protected rights. The professors do not claim any form of contractual or legal right to receive the maximum possible raise, or, for that matter any raise at all, and they do not dispute the fact that Rogers could lawfully have taken the action she did for a number of reasons, no matter how questionable or irrational, or for no reason at all. As the Supreme Court has noted, In the absence of any claim that the public employer was motivated by a desire to curtail or to penalize the exercise of an employee’s constitutionally protected rights, we must presume that official action was regular and, if erroneous, can best be corrected in other ways. The Due Process Clause of the Fourteenth Amendment is not a guarantee against incorrect or ill-advised personnel decisions. Bishop v. Wood, 426 U.S. 341, 350, 96 S.Ct. 2074, 2080, 48 L.Ed.2d 684 (1976). The legal principles applicable to the plaintiffs’ claims are not in dispute. Both sides agree that the case is governed by Mt. Healthy City School District v. Doyle, 429 U.S. 274, 97 S.Ct. 568, 50 L.Ed.2d 471 (1972), and subsequent decisions of this court. Pursuant to the principles enunciated by the Supreme Court in Mt. Healthy, the plaintiffs have the burden of proving that their conduct was constitutionally protected and that this conduct was a “substantial” or “motivating” factor in Rogers’ decisions to reduce their recommended raises. Once the plaintiffs meet this burden, the defendants have the burden of proving that the same salary decisions would have been made even in the absence of the protected conduct. See Mt. Healthy, supra, 429 U.S. at 287, 97 S.Ct. at 576. See also Bickel v. Burkhart, 632 F.2d 1251, 1254 (5th Cir. 1980); Van Ooteghem v. Gray, 628 F.2d 488, 491 (5th Cir. 1980), on rehearing, 654 F.2d 304 (5th Cir. 1981) (en banc). 2 Without deciding whether the plaintiffs’ conduct was constitutionally protected or whether the same salary decisions would have been made regardless of the controversial conduct, the district court ruled in the favor of the defendants, because it found that the plaintiffs had failed to prove that their arguably protected conduct was a “substantial” or “motivating” factor in the president’s decisions. B. The Standard of Review The appealing professors do not question our holding in Van Ooteghem, supra, that the district court’s findings cannot be disturbed unless they are “clearly erroneous” within the meaning of Federal Rule of Civil Procedure 52(a). 3 Therefore, we cannot reverse the district court finding that President Rogers did not lower the recommended salary raises of the plaintiff professors in retaliation for their exercise of constitutional rights of free speech, unless, after careful review of all the evidence, we are “left with the definite and firm conviction that a mistake has been committed.” Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 123, 89 S.Ct. 1562, 1576, 23 L.Ed.2d 129 (1969), quoting United States v. United *1059 States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948). See 9 Wright & Miller, Federal Practice and Procedure § 2585 (1971). Although if we were the trier of fact we might have reached conclusions that differ from the district court’s findings with respect to Shepley and Gavenda, we are not able to say that these findings are erroneous with the requisite “definite and firm conviction.” On the other hand, we have no doubt that the court’s findings are clearly erroneous with respect to White. III. The District Court’s Findings Regarding Shepley and Gavenda Are Not Clearly Erroneous As is frequently the case in disputes of this nature, there was no direct proof that President Rogers sought to retaliate against the plaintiff professors. As the district court noted, the plaintiffs who testified at trial could chiefly “hypothesize” that Rogers acted in response to their political activities. Rogers, as might be expected, adamantly denied that the plaintiffs’ political activities had had any effect on her decisions. She testified that she was not opposed to Gavenda’s lobbying efforts to obtain money for salary increases, and that her only complaint about Shepley’s AAUP activities was his use of campus facilities to prepare and distribute the letters he wrote on behalf of the nursing school faculty. Rogers claimed that she reduced Gavenda’s recommended raise because he was already receiving a relatively high salary in comparíson with others in his department and because, during the ten years she had known him, she had seen little evidence of outstanding teaching ability or scholastic productivity. As to Shepley, she testified that his recommended raise was lowered partially as a penalty for his abuse of the faculty mail system and partially due to the general impression she had of his academic performance. Despite the absence of a “smoking gun” that definitely proves retaliatory intent on the part of Rogers, a considerable amount of circumstantial evidence suggests this conclusion. A rather high percentage of the professors who were adversely affected by Rogers’ decisions were outspoken activists, and an inference of retaliatory motive might be drawn from this fact. The dean of the UTA College of Humanities commented in a letter he wrote to President Rogers that, “when all is said and done, the list of persons affected together with their functions of academic and professional activities speaks too plainly.” A faculty senate fact-finding committee concluded that “there exists a correlation between certain faculty activities and lack of salary reward, a pattern which could not have been accidental.” In the five departments examined by the committee, eleven of the twenty-four individuals who were either not awarded the $400 merit raise or did not receive the recommended salary increase, or both, were engaged in controversial political activities. 4 *1060 Other evidence has also led us to have some question as to the correctness of the district court’s findings regarding Shepley and Gavenda. Dr. Griffy, chairman of the UTA physics department of which Shepley and Gavenda were faculty members, testified that he spoke by telephone with President Rogers shortly after she made her salary decisions. Griffy protested the cuts in Gavenda’s and Shepley’s salary increases. Griffy made notes of the conversation within a month or so after it had occurred; he testified that the notes were an accurate reflection of what was said. When questioned about Gavenda and Shepley, Dr. Griffy’s notes showed that Rogers responded that it “[wjould be inappropriate to reward with substantial raises those individuals whose actions during the past year were at times not in the best interest of the university.” Although, according to Griffy’s notes and testimony, AAUP and TACT were never explicitly mentioned during the conversation, Griffy apparently was left with the strong impression that Rogers’ actions were intended as retaliation against Shepley and Gavenda for their activities on behalf of these organizations. 5 Rogers denied ever having made such a statement. She testified that Griffy told her that it seemed strange that the presidents of both TACT and AAUP had their recommended salaries reduced, and she replied that their activities had “nothing to do” with her salary decisions. 6 While we feel sure that President Rogers was testifying to her best recollection as to the conversation that had occurred several years earlier, the record reflects no reason to question the accuracy of Dr. Griffy’s notes of the conversation, recorded soon after the event, or his credibility in so testifying. Rogers repeatedly denied that the plaintiffs’ political activities had any impact on her salary decision. However, she did admit at her deposition that she was unsure whether she had been able to disregard totally the information she had received about Gavenda’s purported misstatements to the legislature: Q. You had at least formed the impression that he [Gavenda] was presenting misleading information to the Legislature? A. Right. Q. Can you say you were able to discard that completely from your decision making process? A. No, I don’t know whether I could or not. *1061 Furthermore, she could not deny at the deposition (although she did deny at trial) that she may have felt that Gavenda’s TACT activities had diminished his effectiveness as a faculty member: Q. Did you take into account, in forming your general impression [of Gavenda], his activities with TACT and whether that in fact might have diminished his effectiveness as a member of the faculty? A. I do not know that I can answer that with any degree of specificity. Rogers Dep., at 38-89. In view of all this evidence that Rogers did in fact act with retaliatory intent, if we were making a de novo factual determination, we might well conclude that the plaintiffs had carried their burden of proving that their political activities were a “substantial” or “motivating” factor in Rogers’ salary decisions. However, we are unable to find that the district court’s contrary conclusion was clearly erroneous. The primary reason why we have determined not to disturb the district court’s findings with respect to Gavenda and Shepley is that the statistical evidence of record is simply too inconclusive to permit us to form the necessary “definite and firm conviction” that the district court’s findings were erroneous. For example, while the faculty senate committee report indicates that eleven of the twenty-four adversely affected teachers in the five departments examined were engaged in controversial political activities, there is no indication of the percentage of the total number of faculty who were politically active. Obviously, if 11/24 or more of the faculty as a whole were politically active, then no inference of retaliation can be drawn from the fact that 11/24 of the faculty members who received cuts were active. 7 Even within the physics department no definitive inferences can be drawn. Gavenda and Shepley were but two of the four full-time physics professors whose recommended raises were reduced. 8 Professors Moore and Sudarshan were the other two; they received cuts larger in amount than either Shepley or Gavenda. 9 Gavenda speculated that Moore was cut because of his abrasive personality and contemptuous attitude towards administrators, but Rogers testified that Moore was a friend and a strong supporter of hers. No explanations for the reduction in Sudarshan’s recommendation appear in the record. The physics department chairman apparently held both Moore and Sudarshan in higher regard than either Shepley or Gavenda. The chairman made his salary recommendations in order of priority. Moore was at the very top of his list, ranked number one in priority, and Sudarshan was rated eleventh of the twenty-seven physics faculty members. On the other hand, Gavenda and Shepley were near the bottom of the chairman’s list, ranked twenty-first and twenty-sixth, respectively. The physics department budget council likewise rated Moore and Sudarshan higher than both Shepley and Gavenda. The budget council made its recommendations in four priority groupings. Both Sudarshan *1062 and Moore were included in the “first priority” group, while Gavenda was deemed worthy of only the second priority group and Shepley was placed in the third, or next to last, priority classification. There thus do not appear to be any academic reasons why Gavenda and Shepley should have been accorded treatment more favorable than that which was accorded to their apolitical colleagues Moore and Sudarshan. Shepley makes two other arguments that require only brief discussion. Rogers stated that if Shepley had written the controversial letters as a faculty member rather than as president of the AAUP, he would not have, in her opinion, violated the university regulations. Therefore, Shepley asserts, because it was merely his association with AAUP that put him in violation of the rules, Rogers violated his right to freedom of association by punishing him for writing the letters on behalf of the AAUP. If we understand this argument correctly, Shepley claims that because he has the right to use the campus mail and other campus facilities for faculty business, he has a constitutional right to use these facilities for AAUP business. But although Shepley has the constitutional right to associate with the AAUP and to write complaints to the UTA administration on behalf of the AAUP, we do not believe it is seriously arguable that the constitution requires the university to subsidize Shepley’s exercise of his constitutional rights. The university has the right to protect its property, and to this end it may adopt reasonable regulations with respect to the time, place, and manner in which speech-related activities are conducted. See Healy v. James, 408 U.S. 169, 192-93, 92 S.Ct. 2338, 2351-52, 33 L.Ed.2d 266 (1972); Dayan v. Board of Regents, 491 F.Supp. 138, 140 (M.D.Ga.1979), aff’d., 620 F.2d 107 (5th Cir. 1980). Shepley also contends that, even if the university could constitutionally prohibit him from using the university mails on behalf of the AAUP, the regulation that he was said to have breached was unreasonable because of its vagueness. Although the written rule (which is reprinted at note 1 supra) is hardly a model of clarity, Shepley admitted that he had been well aware for several years that the rule was being “very strictly applied” to the AAUP. Apparently he considered the controversial letters to be university business rather than AAUP business; however, close examination of the letters leaves little room for doubt that he was writing as president of the AAUP, not as an associate professor of physics. 10 In our view, the only aspect of this controversy that gives rise to a substantial question is not whether Rogers could have punished Shepley for violating the rule but whether she did in fact do so. In other words, did she truly intend to punish Shepley for his violation of the university mail rule or was this merely a pretextual excuse intended to cover the fact that she was punishing him in retaliation for his substantive work on behalf of the nursing faculty and the AAUP? Inadvertent violation of a rule such as this, which we suspect is often honored only in the breach, 11 hardly seems an appropriate ground for subjecting Shepley to a sizeable reduction in his recommended raise. However, Shepley has neither contended nor offered any evidence that shows that the president’s concern over the violation was pretextual, 12 and Rogers testified that this was just one of many factors in her decision to reduce his recommendation. Accordingly, because v/e do not feel a definite and firm conviction that a mistake has been made, we defer to the district court’s finding that Gavenda and Shepley *1063 did not meet their burden of proving that their activities on behalf of TACT and AAUP were a “substantial” or “motivating” factor in President Rogers’ salary decisions. IV. The District Court’s Findings Regarding White Are Clearly Erroneous Although we have determined that the district court’s findings concerning Shepley and Gavenda must not be disturbed, we reach the opposite conclusion with respect to White. Our determination is based primarily upon the admissions of President Rogers at her deposition and at trial. Based on her admissions, we conclude with little difficulty that White’s activities on behalf of TACT were a substantial factor in Rogers’ decision to withhold the $400 merit raise she had initially awarded him. As noted previously, Rogers originally gave White the full $1900 raise recommended by his department, plus the $400 merit raise, for a total increase of $2300. However, after Rogers received word that the Regents might call upon her to justify raises of more than $2000, she decided to withhold the $400 merit raise from White to keep his total raise below $2000. When asked at her deposition whether she would have given White the full $2300 but for the fact that this decision would have been subjected to special review by the Regents, Rogers answered: “I suspect that he would have, yes.” (At trial, she changed her reply to “He might have, but I can’t be certain of that because, as I said, I continued to go back through that budget.”) After learning about the Regents’ policy on raises greater than $2000, Rogers testified that she decided to reduce White’s recommendation for two reasons: There were two reasons: One, he had been considered for promotion, had been promoted that year, so this material concerning him was very fresh in my mind at that point. The other thing was because he had been rather prominent, I figured his name might be more likely to come up than some of the others. It appears that Rogers made a bipartite analysis. Firstly, she realized that the Regents would be more likely to focus upon White due to his lobbying efforts on behalf of TACT. As she testified at her deposition: Q. Well, you did hear criticism of the TACT operation, did you not? A. Right. Q. And you knew that he [White] had been at least in attendance at the Legislature? A. Right. Q. And you told me that you heard Legislators say to you, “How is it that if these people are to be hired out there to be teaching that they can be spending that time down here talking to us?” A. That’s right. Q. And, initially, I suppose you heard that criticism of Dr. White? A. Right. Q. Did you take in consideration then that that might conceivably be difficult for you to justify an increase for him in excess of $2,000? A. I don’t think that that would make it more difficult to justify. It might have made it more likely that that name be brought up. Q. If you were called upon? A. If I were called upon. As a matter of fact, no one’s name was brought up. Q. But at least in your mind there was the fact that he was around the Legislature, and it made it more likely that if someone down at the System Board was looking through your salary recommendations that they might have fixed on his name as being one where you would be pressed to justify; is that right? A. That’s right. And as she testified at trial: Q. You testified among other things, that he had been active in TACT, he had been at the Legislature. *1064 A. Right. Q. And because of that, at least, or in part because of that you thought it was likely that his name might attract attention of the System Office; is that right? A. ‘ It seemed to me that if anyone was going to question names, it would be names that they knew. Q. Names of those who had been busy and around or busybodies? A. Well, names of people who had been active enough that the name would be known. I doubt if anyone would just pick out a Joe Doe and say, “justify this.” After coming to the realization that White’s activities made it more probable that his name would attract attention, Rogers then reevaluated White and determined she was not certain that, if called upon to do so, she could justify giving him a $2300 raise. She testified: We had just been through the promotion process with Dr. White in December. There had been a question as to whether he should be promoted or not. The average time for a person to move from the associate professor level to the full professor level is about eight years. He had been an associate professor for fifteen years, and we finally made the decision to go ahead and promote him, but it was on that basis that I was not certain that I could say, “Well, he’s one of our really outstanding scholars,” if anyone called on me to justify that. In our opinion, it is irrelevant whether, after making her reevaluation of White, Rogers decided to reduce White’s raise below $2000 because of the controversy over his promotion (as Rogers contends), or because she anticipated that the Regents would be hostile to White due to his TACT activities (as the plaintiffs contend). In Rogers’ own words, the fact that White’s name was “more likely to come up” due to the fact that he had been “rather prominent” was one of her “two reasons” for withdrawing the $400 raise from White. Therefore, because Rogers clearly subjected White to stricter scrutiny because of his political activities, we conclude that White’s activities were a “substantial” or “motivating” factor, see Mt. Healthy, supra, in her decision to withhold the $400 salary increase. White’s prior political efforts were essentially the reason why Rogers felt that she had to reevaluate White’s proposed salary increase. If White had been an apolitical “Joe Doe,” we are convinced from her testimony that President Rogers would not have reconsidered recommending the full salary increase. The statistics appear to corroborate this conclusion. Of the several hundred faculty members whom Rogers initially granted increases greater than $2000, White was the only one whose recommendation was subsequently reduced. This is so despite the fact that White was more highly regarded within the history department than any of the other history professors for whom Rogers approved increases greater than $2000. In the departmental priority rankings for faculty salary increases, White was ranked third out of forty. Nevertheless, Rogers awarded Professors Graham, Divine, Rad-key, and Lasby increases of greater than $2000, even though each of them was ranked significantly lower in priority (eighth, eleventh, twelfth, and twenty-third, respectively) than White. 13 Furthermore, White was the only member of the history department who did not receive the $400 merit raise. Despite White’s high departmental ranking, Rogers chose to give every other history teacher, including those scheduled to be fired, higher “priority” with respect to these so-called “merit” raises than Professor White. Accordingly, we hold that the district court’s finding that White has not met his burden of proving that his political activities were a substantial factor in Rogers’s decision is clearly erroneous. *1065 Because the district court found that White had failed to prove that Rogers was influenced by White’s political activities, the court did not determine whether White’s conduct was constitutionally protected or whether Rogers carried her burden of proving that she would have made the same decision even if White had not engaged in the protected conduct. See Mt. Healthy, supra, 429 U.S. at 287, 97 S.Ct. at 576. We have already held that “but for” White’s political activities, Rogers would not have subjected White’s recommendation to more intense scrutiny and would not have decided to rescind her grant of the $400 merit raise to him. And the defendants do not seriously dispute the fact that White’s lobbying efforts were entitled to first amendment protection. Absent proof of false statements knowingly or recklessly made, White’s right to speak on issues of public importance is constitutionally protected. See Pickering v. Board of Education, 391 U.S. 563, 574, 88 S.Ct. 1731, 1738, 20 L.Ed.2d 811 (1968). Because White has proved that his constitutionally-protected conduct was a substantial factor in Rogers’ decision to revoke the grant of the $400 merit raise, and because Rogers has failed to prove that she would have made the same decision even in the absence of the protected conduct, all elements of the Mt. Healthy test are satisfied and White is therefore entitled to relief. We shall remand the case to the district court for a determination of the relief to which he is entitled. Conclusion With respect to Professors Shepley and Gavenda, the judgment of the district court is AFFIRMED. With respect to Professor White, the judgment of the district is REVERSED. The case is REMANDED to the district court for further proceedings in accordance with this opinion. AFFIRMED IN PART, REVERSED IN PART, AND REMANDED. Footnotes [1] The rule provides: “Both campus mail privileges and stenographic and supply services shall be limited to official budgetary units.” [2] Mt. Healthy involved a teacher who allegedly had been discharged in retaliation for his exercise of first amendment rights, but it is now established that the same principles apply when retaliation takes place in the form of altered employment conditions rather than termination. See Bickel v. Burkhart, supra, 632 F.2d at 1255 n.6 (denial of promotion); Swilley v. Alexander, 629 F.2d 1018 (5th Cir. 1980) (placement of letter of reprimand in teacher’s personnel file); McGill v. Board of Education, 602 F.2d 774 (7th Cir. 1979) (retaliatory transfer). [3] In Van Ootegehm, the panel applied the clearly erroneous standard, rejecting the appellant’s argument that, as in Title VII discrimination suits, we should treat the district court’s findings on employer’s motivation as “ultimate facts” subject to plenary review. We found that it would subvert the clear mandate of Fed.R.Civ.Pro. 52 to apply the phrase “ultimate fact” to important factual questions simply to justify plenary review. 628 F.2d at 491-92. The en banc court also applied the clearly erroneous test, but it did not state any reasons for rejecting the appellants’ argument. 654 F.2d at 305. Rule 52(a) provides in pertinent part: “Findings of fact shall not be set aside unless clearly erroneous, and due regard shall be give to the opportunity of a trial court to judge of the credibility of the witnesses.” [4] The committee report stated in pertinent part: In making this particular analysis we considered no terminal or modified service appointments. In the five departments, with a total of 177 faculty members, the following patterns emerged: (1) there appear to have been 24 cases where the individuals either were not awarded the presidential merit increase or did not receive the recommended salary increase or both; (2) of these 24 seven were either presidents of AAUP or TACT or members of the Counsel on Academic Freedom and Responsibility or chairperson of a major faculty organization; (3) in addition, three of the 24 were administrative officers of budgetary units which may have been controversial; (4) one of the 24 was an outspoken critic of administrative practices; (5) of the 24 one had a comparatively high salary; and (6) of the 24, 12 were either unexplainable (perhaps because of insufficient data) or they fell into other categories such as that of having an abrasive personality. These 24 cases constitute about 14% of the 177 faculty members in the five departments, and the seven cases cited in (2) above constitute about 29% of the 24. In addition to these 24 cases wherein the committee has unmistakable evidence, there are numerous other cases of alleged irregularities not included above for one of two reasons: (1) the professor involved asked that his case not be made a matter of public record; or (2) the data available were not so complete. [5] Gavenda testified that Griffy was “obviously very upset” after his conversation with Rogers. At trial, Griffy, who then still headed the UTA physics department, was somewhat more circumspect as to his conclusions regarding his conversation with Rogers: Q. In her conversations did she ever lead you to believe that she had taken some type of retaliatory measure toward these two individuals? A. I’m afraid I will have to ask from some clarification of the word “retaliatory”. Q. Did you feel that she had reduced their salary increases because of their activities either with TACT or AAUP or whatever their activities would have been in regards the University? A. I do not ever remember — I do not remember ever hearing the words “TACT,” or “AAUP” used during our conversations. Her response was that she felt that it would not be appropriate to reward activities which she felt was not in the best interest of the University. Now it’s a question of whether you consider not rewarding retaliatory. [6] Rogers’ verbatim testimony regarding the conversation with Griffy was as follows: Dr. Griffy called me and said that he thought it was striking or strange, I’m not sure of the word he used, that two of the people in his department whose recommended salaries were reduced, was the president of TACT and the president of AAUP. I assured him that their activities there had nothing to do with that, with my setting of salaries, and he then said that he thought especially strange in Dr. Gavenda’s case because it was his understanding that Dr. Gavenda and Dr. Hill had been most instrumental in getting this salary raise money for that year, and at that point I told him that was not my understanding, that I had had negative feedback from the data that they had presented from members of the Legislature as well as some of the people of the system. I do not recall exactly who it was. I went on to say that this was not my reason why I cut this, but at the same time that I did not see that it was a reason for raising salaries. [7] It is clear that many professors who were at least as controversial and outspoken as the plaintiffs did receive the full recommended salaries. For example, Rogers testified without contradiction that she did not cut back on the recommendation for Dr. Jim Sledd, even though he had talked to legislators more than anyone else, accusing the administration of outright dishonesty in the reporting of faculty work loads. Rogers also mentioned four other “activist or outspoken” professors whose recommendations she did not reduce. On the other hand, it is equally clear that there were many completely uncontroversial faculty members whose recommendations Rogers did cut. [8] Rogers also reduced the recommendations of two other professors who had joint appointments with the physics department and another UTA department. [9] For Moore, the budget council recommended a $2000 raise, the chairman $4000, and the dean $3000. Rogers gave him $1500, plus the $400 merit award, for a total raise of $1900. For Sudarshan, the budget council, the chairman, and the dean all recommended a $2000 increase. Rogers cut this to $1000. Sudarshan was automatically excluded from consideration for the $400 merit award, because his salary was greater than $35,000. [10] For example, one of the letters expressly begins, “As President of the local chapter of the American Association of University Professors, I have been asked to look into another case involving tenure at the School of Nursing.” [11] Shepley testified that he regularly received personal and other unofficial communications through the campus mail. Other evidence in the record is of similar effect. [12] In fact, it appears from the record that Rogers was under some pressure from the University of Texas System Office to enforce the rule. [13] Graham and Divine received $1800 raises, plus $400 merit raises, so that each received a total increase of $2200. Radkey received a total increase of $2100 ($1700 plus the $400 merit award) and Lasby received $2134 ($1734 plus the $400 merit award).