This case involves a dispute over George Washington University's refusal to promote a faculty member to full professor during the 1970's. The primary question is whether the professor's claims are barred by the statute of limitations. For the reasons that follow, we affirm in part and vacate in part and remand for further proceedings.

* For many years, Nicholas Kyriakopoulos has served as an Associate Professor in the Electrical Engineering and Computer Science Department of George Washington University's School of Engineering and Applied Science. After assuming his present position in 1970, appellant sought promotion to full professor beginning in 1973 and continuing each academic year thereafter until 1978. Each year, his effort failed. After successive rejections, Kyriakopoulos appealed the adverse decisions rendered in 1975 and 1976 through the University's internal grievance procedure. A Hearing Committee was empaneled and in due course upheld the denial of promotion. The Hearing Committee's decision was sustained, in turn, by the Faculty Senate Grievance Committee in 1978. Kyriakopoulos also appealed his 1977 denial of promotion, which a Hearing Committee again upheld.

Of particular relevance to this appeal is appellant's effort in 1978 to secure a full professorship. During the course of what proved to be protracted internal grievance proceedings with respect to appellant's 1978 application, the Faculty Senate Hearing and Grievance Committees eventually recommended that Kyriakopoulos be promoted to full professor as of the fall of 1983. Although the Hearing Committee reached no conclusion as to whether appellant merited promotion, it concluded that the departmental Personnel Committee had improperly applied the governing criteria in considering Kyriakopoulos' 1978 application. An Appeals Panel of the Grievance Committee concurred in both conclusions.

The matter then went before the University's Board of Trustees, which determined to remand the case for further consideration and findings on the specific issue whether appellant merited promotion. The reason for the Trustees' remand was that the faculty grievance proceedings had failed to yield a finding that Kyriakopoulos indeed merited elevation to a full professorship. As appellant puts it in his brief, "[t]he University's trustees were advised by one of its attorney/members that it was powerless under the contract to promote plaintiff without a finding of merit under any circumstances." Brief for Appellant at 18 (emphasis in original). The matter accordingly went back to the departmental Personnel Committee for reconsideration.

In the remand proceedings, consideration of Kyriakopoulos' application was to be based on the criteria in effect as of 1978 and on his record as it stood that year. After further consideration, the Personnel Committee once again concluded that appellant did not merit promotion under the operative criteria, namely that the candidate "must possess outstanding ability." Joint Appendix ("J.A.") at 137. Kyriakopoulos was informed of the Personnel Committee's adverse action in November 1985. Eschewing any further internal grievance proceedings, Kyriakopoulos filed suit in United States District Court in March 1986.

The complaint set forth various claims. The first three causes of action sought specific performance (promotion), and both compensatory and punitive damages for breach of contract in failing to promote Kyriakopoulos in 1978 (and earlier) and for injury to his reputation by virtue of his being passed over for promotion. The fourth claim sought damages (both compensatory and punitive) for the Trustees' failure in 1985 to follow the recommendation to promote Kyriakopoulos to full professor (notwithstanding the absence of a finding of merit). The University responded with a motion for summary judgment, which the trial court granted in part in September 1986 by dismissing the first three causes of action as barred by the statute of limitations. As the parties were continuing to litigate over the fourth cause of action, appellant filed a second amended complaint which added two new claims. The fifth cause of action challenged the 1985 actions of the Personnel Committee, which considered the promotion issue on remand from the Trustees. Among other things, appellant alleged that the Personnel Committee had created and applied "inoperable" criteria to his application and had failed to provide findings in support of its adverse decision. Finally, Kyriakopoulos complained (in his sixth cause of action) that the University had breached its duty to act in good faith in performing the contract.

The District Court ruled entirely in favor of the University. In its Memorandum and Order filed September 30, 1986, the trial court held that appellant's first three causes of action were barred by the statute of limitations. In so concluding, the court stated:

It is, of course, highly desirable that academic disputes be settled through University grievance mechanisms. Administrative remedies should be exhausted before resort to the courts is permitted. However, the Court of Appeals ... has plainly suggested that administrative remedies need not be exhausted in a case of this nature, and that the statute of limitations is not tolled by resort to such procedures.

Kyriakopoulos v. George Washington Univ., 657 F. Supp. 1525, 1534 (D.D.C.1987), J.A. at 275.

In April 1987, the District Court filed a Supplemental Pretrial Order addressing appellant's fourth cause of action. As the trial court described this portion of Kyriakopoulos' claim, "[i]t is the essence of plaintiff's position on count 4 that when the Grievance Committee and the Appeals Panel recommend promotion of a person to professor, the Board is obligated to authorize the promotion." Id. at 1525, J.A. at 779. The court rejected this position, opining that the pertinent provision of the Faculty Code "contemplate[d] that the Board [of Trustees] will make the final decision, informed, but unfettered, by the decisions of the subordinate committees." Id. at 1526, J.A. at 780. The District Court went on to condemn as "frivolous" Kyriakopoulos' contention that the Board's declination of the promotion recommendation was in bad faith. Id., J.A. at 781. The court put it this way:

The Board has no legal duty to approve automatically the recommendation of the Grievance Committees.... The fact that on one occasion more than 10 years ago the defendant promoted to professor a faculty person ... who had been the victim of legally cognizable sex discrimination without a finding that she merited promotion is not material evidence that the Board's decision here was either discriminatory or in bad faith. Count 4 fails to state a claim on which relief can be granted.

Id.

Finally, in an Order and Memorandum filed September 22, 1987, the District Court granted summary judgment in favor of the University as to appellant's fifth and sixth causes of action. In so doing, the trial court held that Kyriakopoulos had failed to adduce any competent evidence that he in fact merited promotion. J.A. at 704. In particular, the District Court emphasized that appellant had failed to meet the publishing requirements as enumerated by the Personnel Committee. In the trial court's words, "[Kyriakopoulos] repeatedly applied for promotion only to be told more or less the same thing: he needed to publish in order to be promoted. Yet, it is undisputed [that] he has not responded to the University's suggestion that he publish." Id. at 707. As for appellant's proffered evidence of expert testimony by an engineering professor (Professor Robert Newcomb of the University of Maryland), who would testify to the effect that Kyriakopoulos was in fact qualified for promotion at George Washington, the trial court rejected that evidence, without elaboration, as incompetent. Id. at 706. The trial court also stated, in the wake of this court's decision in McConnell v. Howard University, 818 F.2d 58 (D.C.Cir.1987), that "if there is any institutional academic prerogative protected by the First Amendment, as [the University] persuasively contends there is, it ought to protect a University's academic decision-making process from the kind of lay intrusion plaintiff seeks." J.A. at 706.

This appeal followed.

Appellant's first three causes of action founder on the statute of limitations. As to those, Kyriakopoulos seeks relief for the University's failure to promote him to full professor as late as 1978. The trial court correctly granted summary judgment for respondent holding appellant's claims to be time barred. See Kyriakopoulos, 657 F. Supp. at 1532-34, J.A. at 270-77 (attachment to subsequent order).

The statute of limitations is designed to prevent exactly the sort of litigation embodied in the first three causes of action--protracted quarrels over long past events, based on stale records, allowing the parties no repose and preventing the healing of old wounds. For these well-understood reasons, among others, the District of Columbia delimits recovery upon actions such as those before us--"on a simple contract, express or implied"--to three years. D.C.Code Ann. Sec. 12-301(7) (1981); see Carter v. Washington Metropolitan Area Transit Authority, 764 F.2d 854, 858 (D.C.Cir.1985) ("[T]he general rule that one can glean from all the cases discussed above is that courts should apply the [District of Columbia] statute of limitations strictly, even though barring actions often seems arbitrary and inequitable.") (and cases cited, id. at 855-57). The statutory period begins to run when the party harmed reasonably has notice of the breach. See, e.g., Prouty v. National R.R. Passenger Corp., 572 F. Supp. 200, 205 (D.D.C.1983); Ehrenhaft v. Malcolm Price, Inc., 483 A.2d 1192, 1198-99 (D.C.1984).

Here, appellant had notice of the events underlying his principal allegation at the latest when he received a letter in April 1979 informing him that he had once again been denied promotion. Yet, Kyriakopoulos did not file this suit until March 1986, an interval that (obviously) far exceeds three years. It cannot reasonably be gainsaid that the statute thus bars claims, such as the first three causes of action, predicated upon such long-past events.

Faced with this formidable impediment, Kyriakopoulos argues that the statute does not bar these particular causes of action because the Board of Trustees in 1985 directed the Personnel Committee to make a "retroactive decision" evaluating his qualifications as of 1978. See Reply Brief for Appellant at 5-7. We disagree. Neither the renewed ability to recover losses dating from 1978 nor the 1985 direction to reconsider the decision avoids the bite of the limitations provision. Neither factor makes the first three causes of action rest upon a breach other than the initial denial of promotion, which, as we have seen, occurred more than three years before Kyriakopoulos repaired to court. Rather than "relating back" and reviving earlier claims, the later actions, properly analyzed, act upon the earlier actions merely as a failure to cure any previous breach of contract--a failure that neither constitutes a new breach nor saves appellant's claims from operation of the limitations bar. See Fitzgerald v. Seamans, 553 F.2d 220, 230 (D.C.Cir.1977) ("[T]he mere failure to right a wrong and make plaintiff whole cannot be a continuing wrong which tolls the statute of limitations, for that is the purpose of any lawsuit and the exception would obliterate the rule."); accord Press v. Howard University, 540 A.2d 733, 735 (D.C.1988) (rejecting theory of continuing breach based on an ongoing suspension, finding the statute to run from the initiation of the breach).

If Kyriakopoulos' contention is viewed, generously, as a claim that the University's grievance procedures toll the statute of limitations, that argument likewise falls short. The contract itself determines whether the parties have annointed contractual remedies as the exclusive means to redress alleged breaches (and thus bar access to the courts until contractually ordained procedures are exhausted). Without that explicit agreement, however, the statute of limitations normally begins to run despite the existence of grievance procedures. Cf. Delaware State College v. Ricks, 449 U.S. 250, 261, 101 S. Ct. 498, 505, 66 L. Ed. 2d 431 (1980) ("[T]he pendency of a grievance ... does not toll the running of the limitations periods. The existence of careful procedures to assure fairness in the tenure decision should not obscure the principle that limitations periods normally commence when the employer's decision is made.") (citations omitted) (Title VII context); Fitzgerald, 553 F.2d at 227 ("The choice of whether to pursue one remedy or the other, or both, may be difficult.... But the administrative and judicial remedies were independent. And the limitations period ran on the judicial remedy independently of the [Civil Service Commission] proceeding.").

As the trial court rightly observed, the Faculty Code makes mandatory, at most, the preliminary grievance procedures; it manifestly does not require institution of formal proceedings. See Kyriakopoulos, 657 F. Supp. at 1533-34, J.A. at 273-75. More importantly, nothing in the contract suggests that the grievance procedures constitute an exclusive remedy. Thus, as the District Court correctly held, appellant could have repaired straightaway to the courthouse, rather than exhausting contractually provided avenues of redress.

Appellant's fourth cause of action seeks relief for the Trustees' failure to follow the favorable recommendation of the Hearing Committee and Grievance Committee. Because the filing of the complaint followed within three years of the Board's action, no statute of limitations problem attends this allegation. However, the District Court correctly concluded that Kyriakopoulos failed to establish that the Board's disposition breached any contractual right.

The Faculty Code provides that, upon the decisions of the Committees, "the record of the case, including the decisions of the Hearing Committee and the Grievance Committee, shall be transmitted to the President and the Board of Trustees for final disposition." Faculty Code, Procedures for Implementation of the Faculty Code E.5, at 26, J.A. at 129. As the trial court observed, nothing in this provision dilutes the traditional discretion reserved to the Trustees or mandates that the Board mechanically accept the committees' recommendations. Basic principles of contract interpretation, buttressed by the District of Columbia's insistence that a contract's written terms establish the rights and obligations thereunder, underlie our reluctance to impose duties upon the Board not imposed by the Code or any other practice or agreement.1 To be sure, the Board's "final disposition" must be informed by the record of the case, and the reasoning and conclusions of the committees are to be before the Board. These provisions are intended to ensure that the grievance procedure is not a sham and that the Board acts on the basis of full information. But the contract does not transmogrify the Board into a rubber stamp of committee decisions.

There is yet another, fact-bound reason for declining appellant's interpretive approach. The Grievance Committee's recommendation (that Kyriakopoulos be promoted without the satisfaction of being deemed to merit promotion, Opinion of the Appeals Panel of the Grievance Committee 29, 33-34, J.A. at 93, 97-98) was in the nature of a compromise. That is, there was no specific finding that Kyriakopoulos merited promotion. Yet, as appellant himself recognizes, the Board knew itself to be contractually bound by the Code's stipulation that "promotions ... shall be made solely on the basis of merit." Faculty Code IV.C, at 8, J.A. at 120. So long as it based its decision on criteria specified in the Code and complied with other contractual provisions, the Board could simply have rejected the compromise proffered by the Grievance Committee and denied Kyriakopoulos promotion. Instead, the Board's "final disposition" mandated further inquiry into whether appellant actually merited promotion. See Letter from President Lloyd Elliott to Prof. Nicholas Kyriakopoulos et al. (June 28, 1984), J.A. at 377. Not a shred of evidence suggests that the Board acted for improper reasons or otherwise did not permissibly exercise its discretion. Under these circumstances, summary judgment appropriately lay. See Celotex Corp. v. Catrett, 477 U.S. 317, 106 S. Ct. 2548, 91 L. Ed. 2d 265 (1986).

Kyriakopoulos' fifth and sixth causes of action, which principally challenge the Personnel Committee's consideration of his application for promotion in 1985 upon remand from the Board of Trustees, give us greater pause.2 On reflection, we are persuaded that the District Court too narrowly construed appellant's claim and, in consequence, did not evaluate parts of the record which might reasonably be viewed as tending to support that claim. The trial court's grant of summary judgment in favor of the University on these two claims provides little articulation for the decision. Yet, we are reluctant to conclude that the trial court's disposition was clearly in error. In view of the desideratum of the orderly and considered conduct of civil litigation, we are loathe to embark upon a plenary evaluation of the record without the benefit of the trial court's analysis. Accordingly, we find ourselves constrained to remand these causes for further consideration, but in the light of concerns which we hereinafter set forth.

Appellant's contractual claim in this particular is a narrow one: that the University disposed of his appeal of the denial of promotion, in the Board's 1984 remand and the Personnel Committee's subsequent decision in 1985, in a manner that the contract did not permit. Specifically, appellant would make out a valid, non-time-barred contractual claim if he were able to show that the University disposed of his claim in 1985 for reasons unrelated to whether he actually merited promotion (e.g., bias, use of improper criteria and the like). That is, an extraneous basis for the decision not to promote Kyriakopoulos would, it appears, violate the Faculty Code's guarantees that "promotions ... shall be made solely on the basis of merit" and "shall be dependent upon professional competence as evidenced by teaching ability, productive scholarship, participation and leadership in professional societies, service to the University, and public service." Faculty Code IV.B.1, IV.C, J.A. at 120.

Unlike appellant's fourth cause of action, this sort of claim requires only that the court examine the express terms of the contract to discern the nature of the University's obligation. The pertinent Faculty Code provisions establish, it seems to us, a continuing duty on the part of the University; action which contravenes that contractual command would therefore establish a separate, independent breach. The contract barred the University from considering factors other than Kyriakopoulos' scholarly merit in the initial promotion decision, and from considering those extraneous factors in its execution of the contractually guaranteed grievance procedure. Thus, if upon remand from the Board the Personnel Committee acted upon Kyriakopoulos' application by using criteria other than those dictated or permitted by the contract, the Committee would thereby have committed an independent breach. For example, if in 1985 the Personnel Committee had determined to deny promotion on the basis of appellant's skills on the Department's softball team (as of 1978), the Committee would have done more than simply fail to cure past breaches. This "fresh" breach--punishing appellant's lack of softball prowess--would be comparable to any other failure to comply with a specific contractual provision, such as the failure to follow a procedural requirement. See McConnell v. Howard Univ., 818 F.2d at 66-67 (failure to transmit report of Grievance Committee to Board, as required by Faculty Code, would constitute violation of contract). In short, although these two claims (unlike appellant's first three causes of action) avoid the limitations bar, they do so only because recent actions, and recent actions alone, provide the basis for the asserted breach.

The District Court, however, characterized appellant's claim as one seeking promotion, see J.A. at 704, and as a claim of retaliation. Id. at 706. Neither characterization, we think, fully captures the gist of the claim, namely that the Personnel Committee in 1985 disposed of Kyriakopoulos' attempt to secure promotion for reasons other than his scholarly merit, which were the only grounds ordained by the contract. The trial court focused upon Kyriakopoulos' scholarly merit, which may be important to a finding of breach (though proof of merit may only be necessary to establish the particular remedy sought). However, additional evidence, some of which the District Court mentioned, see id. at 706-07, could possibly be interpreted as supporting the argument that the Personnel Committee in 1985 defeated Kyriakopoulos' promotion upon an improper and impermissible basis (e.g., the history of the department's hostility toward Kyriakopoulos, see generally, Opinion of the Appeals Panel of the Grievance Committee, J.A. at 63; the overlapping membership of the earlier, allegedly tainted Personnel Committee and the 1985 Personnel Committee, see J.A. at 202, 380, 418; the unchanged nature of the Personnel Committee's treatment of Kyriakopoulos' application in 1985, see Minutes, Full Professor Personnel Meeting (Nov. 6, 1985), J.A. at 380-81; Letter from Prof. R.H. Lang, Chairman, E.E. & C.S. Dep't, to Nicholas Kyriakopoulos (Nov. 11, 1985), J.A. at 414; and evidence that could be viewed as supporting the proposition that Kyriakopoulos did in fact merit promotion, see infra pp. 446-47). In our view, a grant of summary judgment to the University should, in addition to articulating why no reasonable fact-finder could conclude that Kyriakopoulos merited promotion, demonstrate why the proffered evidence does not raise genuine disputes over material issues of fact.

Of especial concern is the trial court's treatment of whether Kyriakopoulos in fact merited promotion. Throughout its opinion, the District Court focused on its conclusion that Kyriakopoulos did not merit promotion, and thus could maintain no contractual claim based upon the University's allegedly evaluating his applications on improper grounds. See J.A. at 704-07. In reaching this conclusion, the court failed to set forth a basis for its evaluation of several items of evidence that might allow a reasonable person to reach a "factual conclusion other than that reached by the district court," Tymshare, 727 F.2d at 1155, and thus possibly entitle Kyriakopoulos to survive the University's motion for summary judgment.

First, the basis for the District Court's disparagement of Prof. Newcomb's proffered expert testimony as not competent and inadmissible is not at all clear. See J.A. at 704, 706. Contrary to the trial court's intimation, First Amendment concerns do not render incompetent the evaluation of Kyriakopoulos' merits as a scholar, see infra pp. 446-447. Less globally, any incompetency flowing from a putative expert's lack of familiarity with the specific criteria applied by the University cannot, in reason, extend to the entire testimony addressing Kyriakopoulos' standing in the field, which provided support for Kyriakopoulos beyond the solitary statement the trial court cites. See generally Deposition of Dr. Robert Newcomb, J.A. 448. In addition, as the Supreme Court held not long ago, the nonmovant's evidence need not be entirely "in a form that would be admissible at trial in order to avoid summary judgment," Celotex, 477 U.S. at 324, 106 S.Ct. at 2553, but need only establish the existence of a genuine issue as to material facts. See Fed.R.Civ.P. 56(c).

Second, the trial court did not address other, seemingly pertinent evidence. The opinion is silent with respect to various aspects of the record (e.g., Kyriakopoulos' record of writing and related scholarship; annual evaluations of Kyriakopoulos' and others' work and teaching; records of those promoted; comparisons of various professors' records, directed to illustrating the relevant criteria governing promotion) that could be taken to suggest Kyriakopoulos to be as qualified as other scholars who secured promotion. See, e.g., "List of Plaintiff's Exhibits by Type" 1-5, J.A. at 341-45; "Statement of Plaintiff of Material Facts as to Which There is a Genuine Issue, at the Very Least" 1-3, J.A. at 407-09.

Third, and fundamentally, the trial court should now consider the record fully, uninhibited by a conception that the First Amendment (or the special nature of a university) shields the University's promotion decision from the scrutiny necessary for the protection of contractual rights. Contrary suggestions in both the memorandum opinion granting summary judgment on the fifth and sixth causes of action, see J.A. at 706, and comments accompanying the grant of summary judgment on the fourth cause, see Kyriakopoulos, 657 F. Supp. at 1529, J.A. at 787-89, fail fully to reflect this circuit's law. In McConnell, 818 F.2d at 67-71, this court explained that contractual rights are to be enforced as diligently (and are valued as highly) in a university setting as in any other: "[W]e do not understand why university affairs are more deserving of judicial deference than the affairs of any other business or profession.... [E]ven if there are issues on which courts are ill equipped to rule, the interpretation of a contract is not one of them." Id. at 69. McConnell 's reasoning readily encompasses the promotion setting as well as the disciplinary one. We can discern no principled reason why it should be delimited to the latter setting, which gave rise to the McConnell litigation. Of course, we do not here extend McConnell 's scope to all aspects of the promotion decision. This case does not involve a judicial recalculation of the University's evaluation of a professor's scholarly merit. The factfinder's scrutiny need extend only far enough to ensure that the University perform its contractual duty--which applies to the grievance procedure as well as to the initial promotion decision--to dispose of a professor's promotion application on no basis other than scholarly merit. As we see the issue, the University cannot claim the benefit of the contract it drafts but be spared the inquiries designed to hold the institution to its bargain.

Fourth, notwithstanding the District Court's contrary suggestions, see J.A. at 703-04; Kyriakopoulos, 657 F. Supp. at 1528, 1530, J.A. at 786, 791-92, we are of the view that the University should derive no comfort from the various committees' failure, prior to 1985, to find that Kyriakopoulos merited promotion. The reason is two-fold: first, the appeals committees expressly disavowed the Personnel Committee's adverse determination in 1978, see Opinion of the Appeals Panel of the Grievance Committee 15-28, J.A. at 79-92; second, the two committees concluded that because the University had so erred in its initial consideration of Kyriakopoulos' professional merit, the record allowed no conclusion to be drawn about whether he in fact merited promotion. See id. at 7, J.A. at 71; Report and Recommendation of the Hearing Committee 9, J.A. at 111.

Finally, our analysis of the record persuades us that the District Court's emphasis on Kyriakopoulos' lack of refereed publications is misplaced. The trial court appears to have accepted the University's argument that Kyriakopoulos' lack of refereed publications clearly indicated that he did not merit promotion. See J.A. at 707; Kyriakopoulos, 657 F. Supp. at 1529, J.A. at 789. However, the University's own Grievance Committee faulted as arbitrary departures from its printed criteria both the Personnel Committee's emphasis upon refereed publications and its failure to consider Kyriakopoulos' extensive non-refereed publications. See Opinion of the Appeals Panel of the Grievance Committee 16-17, J.A. at 80-81. Yet, on remand the Personnel Committee again highlighted Kyriakopoulos' lack of refereed publications in its less than fulsome discussion of why he did not merit promotion. See Letter from Prof. F.H. Lang, Chairman, E.E. & C.S. Dep't, to Prof. Nicholas Kyriakopoulos (Nov. 11, 1985), J.A. at 414. That misplaced emphasis could lend itself to the unflattering interpretation that the Personnel Committee in 1985 merely ratified its prior, impermissible methodology in considering the promotion question. At the least, the extent of Kyriakopoulos' publications and the propriety of the Personnel Committee's renewed focus upon refereed publications seem, at this juncture, to be material facts in dispute, weighing against rather than supporting a grant of summary judgment.

* * *

* * *

With all respect, we believe that the dissent confuses analysis of alleged repeated breaches of a continuing duty on the one hand, and, on the other, the evaluation of when an action establishes a single breach that triggers the statute of limitations. Continuing duties may arise directly from a contract (e.g., a provision providing that a university consider each professional application concerning promotion by committees composed of not fewer than 10 members) or from other legal sources (e.g., the prohibition of defamatory statements). If, in this hypothetical case, a university committee of five members considers a professor's application and issues a report libeling him, and subsequently does so again as part of a grievance proceeding several years later, the later actions are actionable even though they constitute a repetition of the earlier contract breach and libel. No issue of tolling arises, even though the statute of limitations may bar a suit based upon the committee's earlier actions.

The issues of tolling and continuing breach do arise, however, when the court is called upon to decide when an act constituting a single breach of contract triggers the statute of limitations. The portion of Ricks cited by the dissent, Diss. Op. at 449 (quoting Ricks, 449 U.S. at 261, 101 S.Ct. at 505), addresses the question whether the initial denial of tenure triggers that breach's limitations period. Press involved a similar inquiry, because "the alleged breach of contract--the suspension--occurred only once; Press was not resuspended at regular intervals during the three-year period." 540 A.2d at 735.

However, neither Ricks nor Press (nor indeed the entire notion of tolling) suggests that repeated and distinct breaches of contract do not trigger their own limitations periods. Ricks states that despite the time-barred, allegedly discriminatory denial of tenure, Ricks could have maintained a suit for discriminatory discharge if he had "allege[d] and prove[d] that the manner in which his employment was terminated differed discriminatorily from the manner in which the College terminated other professors who also had been denied tenure." 449 U.S. at 258, 101 S.Ct. at 504. Press, too, suggests that the court would have found distinct breaches (with distinctly triggered limitations periods) if it had been presented with "a series of repeated acts, not a single act extending over a long period" (or, as indicated, if presented with re-suspensions). 540 A.2d at 735.

Attending to this distinction underscores the fact that our analysis in no way eviscerates the statute of limitations. As we are all agreed, Professor Kyriakopoulos is barred from litigating the alleged 1978 breaches. See supra pp. 442-43. Any University action that merely declines to remedy that breach, so long as that action independently breaches no applicable provision of the contract, gives rise to no action. The Board's decision provides such an example, as would any grievance decision that complied with other contractual provisions (e.g., the grievance committees' actions in the K-1 and K-2 proceedings involving Prof. Kyriakopoulos). This result will be the usual one, ensuring that no "exception" swallows the limitations rule. Yet, if in the course of the grievance proceedings, the University breaches a continuing duty, that act would not be time-barred merely because the breach resembled a separate, earlier breach.

In this case, we have interpreted the contract to establish a continuing duty, and we direct the trial court to determine whether the University's 1985 actions independently breached that continuing duty. Under our approach, policies underlying both the statute of limitations and contract law are served: the statute of limitations bars all old claims, but the University is held accountable for any recent, distinct breaches. That, as our lights lead us, faithfully reflects the decisional law of this jurisdiction. The District Court acted in accord with this approach when it proceeded to consider the fourth through sixth causes of action after interpreting the statute of limitations to bar the first three counts. Similarly, the University's own argument assumes that a separate contractual breach in 1985 would trigger its own limitations period. See Brief for Appellee at 20 (For the fifth and sixth causes of action, "appellant's claims ... are unsupported by any contractual requirement imposed upon the University" [a premise that this opinion rejects], and "[a]s a result, the claimed breaches are again [like the fourth cause], at most, a refusal to cure the 1978 breach, and not a separate breach of contract."). If we err in our interpretation of D.C. decisional law, we are in fine company indeed.

JUDGMENT ACCORDINGLY.

STEPHEN F. WILLIAMS, Circuit Judge, concurring in part, dissenting in part:

In 1978 George Washington University, acting through the Personnel Committee of its Department of Electrical Engineering and Computer Science, refused to promote Professor Kyriakopoulos to full professor. For seven years thereafter, Kyriakopoulos pursued his grievance through the University's internal procedures. The Faculty Senate Hearing and Grievance Committees heard the matter, then the University Board of Trustees, and finally the Personnel Committee considered it on remand in 1985. It again decided against promotion. Kyriakopoulos filed this diversity suit in 1986.

This court affirms the district court's decision that the claims based directly on the 1978 denial must be dismissed under the District of Columbia's three-year statute of limitations. The court finds that under District law the pendency of grievance proceedings after the 1978 denial did not toll the statute. Maj.Op. at 443. So far, I agree.

The court, however, treats as a new breach the Personnel Committee's 1985 refusal to alter its 1978 decision. This solution, however appealing, wholly undermines the District's rule that the pendency of grievance proceedings does not toll the statute.

A discrete body of case law has interpreted statutes of limitations in the context of employment decisions, particularly in the realm of university-faculty relations. See Maj.Op. at 442-43. In Delaware State College v. Ricks, 449 U.S. 250, 101 S. Ct. 498, 66 L. Ed. 2d 431 (1980), a college professor alleged that his denial of tenure violated both Title VII of the Civil Rights Act of 1964, 42 U.S.C. Sec. 2000e et seq. (1981), and 42 U.S.C. Sec. 1981 (1982). The Supreme Court held that the applicable statute of limitations began to run at the time the College made its decision to deny tenure, rejecting the professor's argument that the pendency of grievance procedures tolled the running of the limitations period:

[E]ntertaining a grievance complaining of the tenure decision does not suggest that the earlier decision was in any respect tentative. The grievance procedure, by its nature, is a remedy for a prior decision, not an opportunity to influence that decision before it is made.

... The existence of careful procedures to assure fairness in the tenure decision should not obscure the principle that limitations periods normally commence when the employer's decision is made.

449 U.S. at 261, 101 S. Ct. at 505-06.

The District of Columbia cases have taken the same view. See, e.g., Clark v. Scott, 329 A.2d 442, 445 (D.C.1974) (pursuit of permissive remedies does not toll statute). In Fitzgerald v. Seamans, 553 F.2d 220 (D.C.Cir.1977), this court, interpreting the D.C.Code's three-year statute of limitations, offered an explanation for the rule: "[T]he mere failure to right a wrong and make plaintiff whole cannot be a continuing wrong which tolls the statute of limitations, for that is the purpose of any lawsuit and the exception would obliterate the rule." Id. at 230. Cf. Press v. Howard University, 540 A.2d 733 (D.C.1988) (rejecting theory of continuing breach, but the grievance procedures available did not appear to address the claim on which plaintiff sought relief in court).

The panel reads these cases as rejecting any theory that the pendency of permissive grievance procedures tolls the statute and thus finds Kyriakopoulos's claims concerning the 1978 decision to be barred. Although the rule is not self-evidently wise, and certainly sounds oddly in the ears of judges accustomed to dismissing the claims of litigants who have failed to exhaust their remedies, such appears to be District law.

But the majority negates the rule by instructing the district court to consider on remand the merits of the claims regarding the 1985 decision. As the flaws that plaintiff finds in the 1985 decision add nothing new to 1978 process, the situation is one in which the defendant has simply failed to right a wrong. See Fitzgerald, 553 F.2d at 230. To assert a valid and timely claim, Kyriakopoulos must allege that some new illegality, distinct from the mere failure to cure the 1978 decision, caused the University's 1985 persistence in its original disposition.

Application of this principle will be easy in some cases. For example, Kyriakopoulos's fourth count, alleging that the Board of Trustees violated University procedures when in 1984 it overrode the recommendation of the Hearing and Grievance Committees, see Maj. Op. at 443-44, alleges a distinct error which (if a wrong at all) would constitute a new breach.1

At the other end of the spectrum, if Kryiakopoulos had claimed that in 1978 the Personnel Committee denied his promotion because of improper insistence on refereed publications and then in 1985 did so for the same reason on remand, both claims would be time-barred. The same would of course be true if the identical illegality were committed by a second university body. Unless both allegations are considered to be untimely, every appeals panel that affirms the original decision will automatically create a new breach, thereby negating the rule articulated by Ricks, Fitzgerald and the other cases.

These hypotheticals are easy to classify because the errors involved are either identical or clearly dissimilar. The more difficult cases arise when the allegations involve different degrees or manifestations of the same type of error. For example, Professor Kyriakopoulos alleges that one reason his merit was not properly assessed in both 1978 and 1985 is that the Personnel Committee was motivated by bias or bad faith. When might the claims concerning bias in the 1985 decision escape the bar of the statute of limitations? If the evidence indicated that a new professor joined the Personnel Committee between 1978 and 1985 and that she was biased against Kyriakopoulos, I can see no reason why her bias would not constitute a new wrong, on which Kyriakopoulos could prevail if he could prove that the new member's presence determined the adverse outcome in 1985. Even an increase in the bias of one or more continuing committee members could create a new claim, so long as the plaintiff could show that it was the increase that caused the committee to persevere in its error.

Similarly, even where the two errors take the form of relying on "reasons unrelated to whether he actually merited promotion," Maj.Op. at 445, the error in the later phase could start the statute running anew so long as whatever was novel in it brought about the continued rejection. If the Personnel Committee rejected Kyriakopoulos in 1978 because of undue insistence on refereed publications, but in 1985 straightened out that error and went on to reject him because of an excessive reliance on classroom evaluations, I would have no difficulty finding a claim originating in 1985.

On this analysis, which appears essential to preserve the District's rule, Kyriakopoulos's fifth and sixth causes of action are time-barred. These claims resemble the hypothetical of successive university determinations that make the same substantive error. In his complaint, Kyriakopoulos claims that in 1978 and in 1985 the relevant university authorities focused on improper criteria. Neither he nor the majority identifies any alleged error made in 1985 that is distinct from the 1978 flaws. On the contrary, the majority states that "[o]f especial concern [in the resolution of the fifth and sixth causes of action] is the trial court's treatment of whether Kyriakopoulos in fact merited promotion," Maj.Op. at 446, and its primary concern seems to be the plaintiff's allegations of undue reliance on refereed publications, Maj.Op. at 447. These allegations are no different from those arising from the 1978 decision. See Second Amended Complaint at 4 (alleging that in 1978 the Personnel Committee "ignored expressly stated criteria for promotion in plaintiff's case, such as that stating that a balance between teaching and research was desirable, while instead requiring that plaintiff meet ... higher levels of performance"); see also id. (defendant refused to promote plaintiff "for ... other reasons unrelated to the merits"); id. at 6 (defendant breached implied covenant of good faith by "failing to consider and grant plaintiff's promotion on the merits").

The majority suggests that Delaware State College v. Ricks supports its view. See Maj.Op. at 448. There the Court held that the statute started to run on the tenure denial, not the discharge a year later; the discrimination occurred at the tenure denial, and the discharge at the end of the next year was simply an ineluctable consequence. 449 U.S. at 257-58, 101 S.Ct. at 503-04. When the Court observes that for the statute to start anew with the discharge, "Ricks would have to allege and prove that the manner in which his employment terminated differed discriminatorily from the manner in which the College terminated other professors who had also been denied tenure," 449 U.S. at 258, 101 S. Ct. at 504, it says only that for this event to restart the statute, it must involve a new discrimination, one not inherent in the original decision. That seems to fit my position exactly.2 The majority's reading of the sentence wholly undercuts the Court's clear statement that the pendency of grievance proceedings does not toll the statute. Id. at 261, 101 S. Ct. at 505.

The majority seems to posit a class of cases where some university body "merely declines [in the grievance procedure] to remedy [the initial] breach," as opposed to instances where it "independently breaches" a provision of the contract. Maj.Op. at 448. It cites the action of the Board of Trustees here as an example. But of course here the Board made no substantive decision at all; it remanded to the Personnel Committee (and we all agree that that act, if a contract violation, would constitute a fresh breach).

What I find hard to conceive is a substantive failure to remedy that would not constitute a new breach under the majority's view. Suppose that on remand here the Personnel Committee simply said, "We find no error in the initial decision." Presumably that would represent an endorsement of the fact-finding and reasoning of the initial decision, and thus be infected with its errors. Suppose they were even more taciturn, and said only "No relief." Perhaps that would qualify as falling short of an independent breach with respect to the substantive errors, although strictly as a logical matter it would seem to depend on the panel's approval of the initial decider's approach. But this is thin solace for the university; if the contract gave the employee a right to review by this university body, I strongly suspect that a court might find the silent treatment to be a breach of the university's procedural obligations. Moreover, even when the review panel is under no legal obligation to explain its decision, it is likely to do so out of the natural and decent impulse to tell the loser why he lost. Thus, it seems to me virtually certain that the majority's mode of analysis provides a method by which plaintiffs can circumvent the District's application of its statute of limitations to grievance procedures in every case where those procedures are employed.