In this case, we reach the dénouement of the lawsuit that Dr. Joseph H. Levenstein has been pursuing against three officials of the medical school of the University of Illinois located in Rockford, Illinois. When the case was last before this court, we affirmed the district court's interlocutory ruling that the defendant university officials were not entitled to qualified immunity. Levenstein v. Salafsky, 164 F.3d 345 (7th Cir.1998) (Levenstein I). Eventually, the district court conducted a bench trial, after which it found that Levenstein had not shown that he had been constructively discharged, nor had he established an equal protection violation. The court accordingly entered judgment in favor of the defendants. This time it is Levenstein who is appealing. Although we have no trouble imagining how a trier of fact might have ruled in his favor, that is not the proper standard of review at this stage of the litigation. The district court's findings of fact were not clearly erroneous, nor do we find legal error in its opinion. We therefore affirm.

* Levenstein joined the medical school faculty of the University of Illinois in Rockford in 1990, as a tenure-track professor, after a distinguished career in South Africa that we described in Levenstein I, 164 F.3d at 348. Before the events giving rise to this lawsuit, he served as the head of the Department of Family and Community Medicine (the Department), first for the Rockford campus, and later for all four University of Illinois campuses. He was granted tenure in 1992. In 1994. Dean Bernard Salafsky, one of the defendants and Levenstein's immediate supervisor, recommended Levenstein for a Faculty of the Year Award and nominated him for the Pugh Charitable Trust "Primary Care Achievement Award." In 1995, Levenstein became the chair of the University of Illinois College of Medicine Primary Care Institute.

Starting around the middle of 1994, however, two events began to unfold: first, Levenstein and Salafsky had a falling-out over the financing and management of the University's Medical Service Plan (MSP), which was a fund designed to cover varlous medical school operating expenses like faculty salaries and maintenance; second, in the spring of 1995, defendant Patricia A. Gill, the Deputy Chancellor of Affirmative Action Programs (AAP), received an anonymous letter from a medical student alleging that Levenstein had sexually harassed her. Levenstein saw the University's response to the sexual harassment complaint as a veiled effort to undermine his efforts to understand deficits in the budgets of the MSP and the Department and to expose financial mismanagement.

Briefly, the financial dispute began when Salafsky's office revealed in mid-1994 that the MSP might be facing a cumulative deficit of an amount significantly greater than its operational losses of approximately $280,000 in 1993 and $260,000 in 1994. Over Salafsky's objection, Levenstein led the formation of an executive committee to serve as a "watchdog" over Salafsky's administration. Later that year, Salafsky proposed the construction of a new specialty clinic to be called the East Side Clinic. Levenstein and the other watchdog committee members agreed that such a clinic would be desirable, but questioned whether the MSP could afford it and whether it could be justified from an educational standpoint. As time went on, Levenstein asked more and more questions about the $400,000 MSP deficit and the projected $500,000 departmental deficit, but Salafsky was unresponsive. On April 19, 1995, the two had a confrontation at an executive committee meeting. Dr. Frank Chmelik, Chair of the Rockford MSP, approached Salafsky in an effort to resolve the problem. Ominously, Salafsky told Chmelik that "there was a much larger problem than this that would become more obvious over the ensuing weeks."

Salafsky was referring to the anonymous harassment complaint, which Gill had received only two days earlier, on April 17. On April 27, Gill forwarded a copy of the letter to Levenstein and to Salafsky. She advised Levenstein that there would be no investigation unless the anonymous complainant came forward. On May 1, a physician at Rockford who had worked with Levenstein contacted Gill to report inappropriate conduct by Levenstein. She, too, did not reveal her identity to Gill until later, but she did pass along her complaint to her immediate supervisor. On May 8, two more similar complaints came in, this time from two of Levenstein's departmental support staff. Salafsky was told about the later complaints.

On May 11, Salafsky and the Associate Regional Dean, Dr. Donald Wortmann, met with Levenstein to discuss the sexual harassment complaints. They told him that if he did not resign effective 5:00 pm that day, he would be suspended with pay pending the outcome of an investigation. As of the time of that meeting, no formal complaints had yet been filed, although Salafsky knew both the identities of some of the complainants and that they wished to pursue formal charges. After hearing all of this, Levenstein telephoned Gill to confirm what had been happening. At the end of the day, Salafsky suspended Levenstein with pay as promised.

The next day, Salafsky called a meeting of the faculty at which he announced Levenstein's suspension and the circumstances that had prompted it. He said, inaccurately, that Gill had received four letters in the last month; in fact, she had received only the one anonymous letter. Formal letters were not long in coming, however. On May 17, the author of the anonymous letter came forward, wrote a new letter, and signed a formal complaint. The two support staff members did likewise, and on May 21, the three women filed a joint "Request for Action" form. With this in hand, Gill opened an investigation.

On May 24, Gill informed Levenstein about the specifics of the complaints and outlined the investigation process. He was permitted to respond in writing to the allegations, which he did, denying any sexually offensive conduct. Gill completed her investigation on July 23. She submitted a report concluding that Levenstein had violated the University's sexual harassment policy to her supervisor, the head of AAP. The supervisor agreed and made two recommendations in the alternative: either Levenstein should be restricted for three years from exercising authority over female subordinates, with the exception of restrictions on classroom instruction, lecturing, and patient treatment; or, if it was not possible to create an arrangement that permitted him to perform fully as a member of the faculty, he should be terminated.

Levenstein submitted an appeal from the AAP's findings and recommendations on August 9 to defendant David C. Broski, the Chancellor of the University. Broski assembled a three-person Faculty Appeal Panel with help from Gill to handle the appeal. The panel conducted its own investigation; it gave Levenstein the opportunity to comment on various witness statements and to make a final statement. Its report, issued on November 20, essentially agreed with Gill's initial conclusion that Levenstein had engaged in offensive conduct. It made no recommendations for corrective action, but it observed that the earlier recommendation was contradictory and could not be carried out.

On December 21, 1995, Broski recommended to the president of the University, James J. Stukel, that Levenstein's tenure should be reviewed "with the intent to revoke." Broski also advised President Stukel that Levenstein had been placed on paid leave pending the out-come of the investigatory proceedings. At the same time, Broski notified Levenstein that he was relieved of his duties as department head and that Broski had recommended to Stukel that dismissal proceedings be commenced.

Stukel accepted that recommendation on January 6, 1996, and concluded that Levenstein should remain suspended. He communicated this to Levenstein, described the dismissal procedure, and indicated that he would seek the advice of the Faculty Advisory Committee after he reviewed the relevant materials. On February 7, 1996, he asked for the Committee's recommendation whether tenure revocation proceedings should begin. He also ordered that Levenstein be given other duties, without changing his compensation (which had remained in place throughout this period). Charles Lane Rice, the Vice Dean of the College of Medicine, assigned Levenstein the menial task of reviewing and evaluating medical videotapes for content and accuracy.

In the meantime, Levenstein, frustrated by what he perceived to be the slow pace of the proceedings, bias on the part of decisionmakers, and the pretextual nature of the actions being taken against him, brought charges of academic misconduct against Salafsky and other administrators at the Rockford campus. The Faculty Advisory Committee advised President Stukel that they would take up the question of Levenstein's termination after they had assessed Levenstein's charges against Salafsky.

The Committee's result was partially favorable, partially unfavorable to Levenstein. Although it reported on April 15 that it found no merit in Levenstein's charges against Salafsky, the next day it recommended to Stukel that he should not pursue termination of Levenstein's tenure. Instead, the Committee thought, the less drastic measures of reassignment and rehabilitation were appropriate. In coming to this recommendation, the Committee said that it perceived "no room for serious contention" that "several" of the charges against Levenstein were valid; nonetheless, it also indicated that some of the charges were "not particularly persuasive," some of the conduct was not necessarily wrongful, and in some instances Levenstein's actions "may be subject to an innocent interpretation." The Committee reported its strong reservations on the question whether the allegations proved that Levenstein's future performance would be "professionally substandard." The Committee did not copy Levenstein or any other administrators on its recommendation to Stukel, nor did Stukel tell Levenstein the gist of the Committee's findings.

Before Stukel had a chance to make a decision on the case, Levenstein pretermitted the process by tendering his resignation in a letter dated April 23, 1996. Stukel accepted. Had Levenstein not resigned, Stukel would have been free to accept, reject, or modify the Committee's recommendation. If he had chosen to pursue tenure termination, University statutes provided for a hearing at which Levenstein would have had an opportunity to present his case and for an appeals process.

On September 27, 1995, while the University proceedings we have just described were still underway, Levenstein filed this lawsuit under 42 U.S.C. § 1983 against the Board of Trustees of the University of Illinois, Salafsky, Gill, and Broski. He later dropped his claims against the Board of Trustees and filed an additional complaint in the district court, but the cases were eventually consolidated both in the district court and here, and so we treat them all as one.

After this court affirmed the district court's decision rejecting the defendants' claim of qualified immunity in Levenstein I, the case continued on to the merits. On February 4, 2002, the district court issued a memorandum and order in which it found that genuine issues of fact existed on three questions: whether Levenstein's procedural due process rights had been violated; whether his suspension was really the result of the defendants' animus and the procedures a sham; and whether he had been the victim of selective enforcement in violation of his equal protection rights. In the same memorandum, however, the court found that the defendants were entitled to qualified immunity insofar as these claims were brought against them in their individual capacities. It did so on the ground that Levenstein could not show under these circumstances that the defendants should have known that their actions were violating his rights. See Saucier v. Katz, 533 U.S. 194, 201, 121 S. Ct. 2151, 150 L. Ed. 2d 272 (2001). It properly noted that this court's opinion in Levenstein I was premised on the fact that the defendants' initial motion was based exclusively on the pleadings, and thus did not preclude the district court from revisiting the immunity issue on the basis of a fuller record. See Behrens v. Pelletier, 516 U.S. 299, 307, 116 S. Ct. 834, 133 L. Ed. 2d 773 (1996).

After making that ruling, the court went on to hold a bench trial. Although it did not spell out what relief was still possible, the only thing could have been the prospective injunctive relief of reinstatement that Levenstein had requested. Levenstein originally brought his suit against the three University defendants in both their official and personal capacities. The qualified immunity ruling spelled the end of his personal capacity claims for damages. As officials of the University of Illinois, which is functionally the State of Illinois for purposes of § 1983, the defendants were not "persons" who could be sued in their official capacity for damages. See Will v. Mich. Dep't of State Police, 491 U.S. 58, 71, 109 S. Ct. 2304, 105 L. Ed. 2d 45 (1989). Nevertheless, under the well-recognized theory of Ex parte Young, 209 U.S. 123, 28 S. Ct. 441, 52 L. Ed. 714 (1908), Levenstein was entitled to pursue injunctive relief against them for actions they took in violation of his constitutional rights. See Verizon Md., Inc. v. Pub. Serv. Comm. of Md., 535 U.S. 635, 645, 122 S. Ct. 1753, 152 L. Ed. 2d 871 (2002); Kentucky v. Graham, 473 U.S. 159, 167 n. 14, 105 S. Ct. 3099, 87 L. Ed. 2d 114 (1985). On September 5, 2003, the court ruled that Levenstein had not demonstrated that he had been constructively discharged and that he had not met his burden of showing a "class of one" equal protection violation.

On appeal, Levenstein first attacks the district court's conclusion that he failed to prove that the defendants were "out to get him" and that the process they used was a sham. He also argues that the court erred by failing to find that his forced idleness, coupled with the sham procedures, amounted to a constructive discharge, and by rejecting his equal protection claim. Finally, he cursorily takes issue with the court's decision with respect to the defendants' qualified immunity.

* We begin with the qualified immunity argument, since Levenstein's chances of any damages relief against the individual defendants are dependent on this point. The district court's explanation of why it was conferring qualified immunity was regrettably brief. Unfortunately, so was Levenstein's treatment of this point in his brief on appeal. Indeed, it was so abbreviated that we are tempted to say that he has forfeited the right to pursue this argument.

We have no need to explore the forfeiture issue, however, for a straightforward reason. In the end, as the defendants point out, it makes no difference to this case whether the district court's immunity ruling was correct or not. A proper treatment of the qualified immunity issue requires the court first to decide whether the officer's conduct violated a constitutional right, taking the facts in the light most favorable to the party asserting the injury. Saucier, 533 U.S. at 201, 121 S. Ct. 2151. If and only if so, then the court must determine whether the right was clearly established in the light of the specific context of the case. Id. Here, the court went on and conducted a full bench trial for purposes of deciding whether the defendants violated any of Levenstein's constitutional rights. It concluded, as we have already noted, that they did not. If this decision was correct, then there is nothing left of any potential damages claim in any event. We therefore have no need to decide how clearly the right to be free of sham proceedings and trumped-up charges was as of 1995 and 1996. Levenstein's case rises or falls on the soundness of the district court's analysis of the merits.

We begin our consideration of the merits with a reminder about the standard of review, which is the biggest hurdle that Levenstein must surmount. After a full bench trial, the district court's findings of fact may not be set aside unless they are clearly erroneous. Cerros v. Steel Techs., Inc., 288 F.3d 1040, 1044 (7th Cir. 2002). This, as we have often vividly reminded parties, is a highly deferential standard. We review questions of law de novo; mixed questions of law or fact that do not involve constitutional rights are normally reviewed for clear error. See Thomas v. General Motors Acceptance Corp., 288 F.3d 305, 307 (7th Cir.2002).

Levenstein's first argument applies, as he recognizes, to both his due process and his equal protection claims. He takes issue with the proposition that the defendants were proceeding against him in good faith. It is true that the district court made some findings that might be read as a finding that the investigation was unobjectionable. It noted, for instance, that the University was aware of a history of informal sexual harassment complaints against Levenstein, even though none of those complaints had resulted in formal proceedings. (Findings of Fact ¶¶ 26-30.) It found that the Faculty Appeal Panel (which did not include any of the defendants) "consistently found Levenstein's version of events to be less credible than the complainants [sic]." (Id. ¶ 53.) Furthermore, and closer to the point, the court found that the University provided many procedures that Levenstein chose not to use (id. ¶ 77), and that Levenstein was aware of those procedures. (Id. ¶ 78.) In the end, however, the court never came to any final conclusion about the underlying motivations for the investigation.

Instead, the court found that in the final analysis, it was not the University that deprived Levenstein of any property right protected by the Constitution. It was he who chose to submit his resignation just before President Stukel was about to decide how to proceed. He elected to forgo numerous procedures before entities including the University's Committee on Academic Freedom and Tenure and the Board of Trustees. Levenstein offers no reason for us to think that the Board, which had the ultimate responsibility to decide what to do, was biased, or that either the Committee or the Board would have refused to listen to his arguments about pretext and sham.

The district court observed that Levenstein was not bringing an independent claim based on his suspension. The fact that he was receiving full pay throughout the period of the University's investigation may explain this decision. No matter. Upsetting and unpleasant though it surely would be to be the victim of false accusations of sexual misconduct from someone whose secret agenda was eliminating a critic, those accusations had not yet led the University to terminate Levenstein's position. The only question remaining with respect to the due process part of the case, to which we now turn, is whether the district court was mistaken to reject the constructive discharge argument.

The Supreme Court recently had occasion to address the topic of constructive discharge in the context of a Title VII hostile-environment claim. See Pennsylvania State Police v. Suders, 542 U.S. 129, 124 S. Ct. 2342, 2347, 159 L. Ed. 2d 204 (2004). The question whether the plaintiff had suffered a constructive discharge was important because it determined whether the employer was entitled to raise an affirmative defense to the charge. Suders held that in order to establish a constructive discharge, such a plaintiff had to prove both harassing behavior sufficiently severe or pervasive to alter the conditions of her employment, and that "the abusive working environment became so intolerable that her resignation qualified as a fitting response." Id. at 2347. Although Levenstein's case relies on the due process clause rather than Title VII, the general approach to constructive discharge outlined in Suders is equally applicable here. For him, the central issue is whether, as he asserts, it was the University that deprived him of his protected property interest in his tenured position on the faculty, or if, as the district court concluded, he chose to leave.

Levenstein relies on this court's decision in Parrett v. City of Connersville, 737 F.2d 690, 694 (7th Cir.1984), in which we held that an employee is constructively discharged if "his working conditions were made so miserable that he was forced to quit." In Parrett, we found that standard met when a police detective was literally forced to sit in a windowless room that was formerly a broom closet and spend his entire shift with nothing to do. In Levenstein's view, his situation was identical in all material respects: he was put in a state of enforced idleness for almost a year, he was physically barred from campus, he was humiliated by Salafsky's statements and his removal as department head, and he was given the demeaning task of reviewing dusty old medical videotapes. He concludes that all of this, coupled with the alleged sham investigative procedures, amounted to a constructive discharge.

We put to one side the investigative procedures, which we have already found had not run their course and thus cannot be labeled as "sham" taken as a whole. We see a subtle but important difference in the way that the Supreme Court phrased the test for a constructive discharge in Suders and the phrase Levenstein has taken from our Parrett opinion. One could read the excerpt from Parrett as establishing a purely subjective test for constructive discharge, because only the employee would know how miserable the working conditions made her feel. The Suders test, in contrast, is an objective one, as the Supreme Court made clear:

Under the constructive discharge doctrine, an employee's reasonable decision to resign because of unendurable working conditions is assimilated to a formal discharge for remedial purposes.... The inquiry is objective: Did working conditions become so intolerable that a reasonable person in the employee's position would have felt compelled to resign?

124 S.Ct. at 2351 (citation omitted).

A number of facts lead us to answer that question in the negative. First, even Levenstein concedes that the allegations had been made against him, and that investigations were proceeding. Once such a process starts, for whatever reason, it is incumbent on the University to decide whether the complaints have merit, and if so, to decide what remedy is appropriate. If the complaints were made in bad faith or cannot be supported by the facts, the accused professor will be exonerated, which is the best that can be done even if he might have wished that the process had never begun in the first place. Second, unlike the policeman in Parrett, Levenstein's reassignment was explicitly temporary, pending the outcome of the investigation. Third, while eleven and a half months may have seemed like an eternity to Levenstein, in the real world it was not so long that his resignation could be called reasonable. The Supreme Court has acknowledged that at some point a delay in the termination process might become a constitutional violation. See Cleveland Bd. of Educ. v. Loudermill, 470 U.S. 532, 547, 105 S. Ct. 1487, 84 L. Ed. 2d 494 (1985). There, however, the Court found that a nine-month adjudicatory period was not per se unconstitutionally long, and it noted that the delay appeared to stem in part from the thoroughness of the procedures. The latter can also be said about Levenstein's case, and we do not see a difference of constitutional magnitude between nine months and twelve months, in the context of the elaborate procedural protections enjoyed by tenured university professors.

We conclude that a person who is on leave with pay, with a temporary (though unsatisfying) reassignment pending an investigation of serious job misconduct, who resigns rather than waits for the conclusion of reasonable prescribed due process procedures of the institution, has not from an objective standpoint been constructively discharged. Indeed, to treat any reasonable pre-termination suspension with pay as a constructive discharge would create considerable tension with Supreme Court decisions like Loudermill, 470 U.S. at 544-45 (expressly stating that a suspension with pay prior to a pre-termination hearing is a valid option for a state employer). See also Gilbert v. Homar, 520 U.S. 924, 929, 117 S. Ct. 1807, 138 L. Ed. 2d 120 (1997) (holding that suspensions without pay are also possible under some circumstances). However distasteful his temporary situation was, Levenstein's working conditions had not become so intolerable that a reasonable employee in his position would have felt compelled to resign. There is an inherent tension for any employee who is undergoing the kind of investigation that follows complaints of sexual harassment or other inappropriate conduct. Nothing in Suders equates this kind of tension with the intolerable working conditions that give rise to a constructive discharge or undermines the general Loudermill rule.

We therefore agree with the district court's conclusion that Levenstein was not constructively discharged from the University. He voluntarily resigned, and thus the defendants acting in their official capacities did not deprive him of any protected property interest.

Levenstein's final theory relies on the "class of one" equal protection cases. See Village of Willowbrook v. Olech, 528 U.S. 562, 120 S. Ct. 1073, 145 L. Ed. 2d 1060 (2000). Olech recognizes that a plaintiff states an equal protection claim where she "has been intentionally treated differently from others similarly situated and ... there is no rational basis for the difference in treatment." Albiero v. City of Kankakee, 246 F.3d 927, 932 (7th Cir.2001) (citing Olech, 528 U.S. at 564, 120 S. Ct. 1073, 145 L. Ed. 2d 1060). We have recognized that a person may also state a claim under a "class of one" theory by showing that "the government is treating unequally those individuals who are prima facie identical in all relevant respects, and that the cause of the differential treatment is a `totally illegitimate animus toward the plaintiff by the defendant.'" See Nevel v. Village of Schaumburg, 297 F.3d 673, 681 (7th Cir.2002); Albiero, 246 F.3d at 932; Esmail v. Macrane, 53 F.3d 176, 180 (7th Cir.1995).

The district court rejected this theory because Levenstein failed to identify another similarly situated individual who was treated differently. He asserts that there is no need to point to such an individual if one is proceeding under the "illegitimate animus" approach. This, however, mistakes what we said in Albiero, and (worse) is inconsistent with the Supreme Court's description of the theory in Olech. Olech requires a showing of how the other similarly situated individuals have been treated. Albiero follows that guidance by requiring the plaintiff to demonstrate how he was unfavorably treated as compared with others who are otherwise identical to him. After the plaintiff shows the differential treatment, he must then prove that it flows from an illegitimate animus, not from inadvertence or some kind of permissible governmental classification.

Even if we credited Levenstein's testimony that numerous other individuals who were accused of sexual harassment were not suspended pending an investigation or were not investigated at all, that would not be enough to prove this claim. First, this is far from a showing that these other individuals were identical to him in all relevant respects. See, e.g., Grayson v. O'Neill, 308 F.3d 808, 819 (7th Cir.2002). Second, the district court found that the University had ample reason to investigate him apart from any illegitimate animus, in the form of the oral and written complaints it had received, both in the spring of 1995 and earlier. He has not convinced us that the district court clearly erred in finding those facts.

When all is said and done, both Levenstein and the defendant university officials had their day in court. With all the evidence before it, the court concluded that Levenstein was not constructively discharged, and thus that no action of the University deprived him of a protected property right. It also concluded that Levenstein had not proven that he was singled out for unfavorable disparate treatment in violation of his equal protection rights. We therefore AFFIRM the judgment of the district court. We also DENY the defendants' motion to strike Levenstein's brief as unnecessary.