In 1974 and 1975, two classes of plaintiffs, represented by the Alliance to End Repression and the American Civil Liberties Union respectively, brought suit under 42 U.S.C. § 1983 against the United States and the City of Chicago. The suit charged that the FBI’s Chicago office and the Chicago Police Department’s intelligence division were violating the class members’ First Amendment rights by overly intrusive and improperly motivated investigations of alleged subversive activities. In 1981, before a trial could be held, the defendants agreed to a consent decree, which was approved by the district court the following year, imposing detailed restrictions on the defendants’ investigative authority.
In 1997, long after this court had interpreted the decree as imposing fewer restrictions on the FBI than the district court had thought it did,
The City appealed from the refusal to modify the decree. It pointed out that it had complied with the decree throughout the entire period of almost two decades in which it had been in force, that during this period the Supreme Court and this court had become ever more emphatic that the federal judiciary must endeavor to return the control of local governmental activities to local government at the earliest possible opportunity compatible with achievement of the objectives of the decree that transferred that control to the federal courts, and that the culture of law enforcement in Chicago and the character of the threats to public safety by ideologically motivated criminals had so far changed as to make much of the decree obsolete. We agreed with the City and ordered that its motion be granted in its entirety.
All this is by way of background to the present appeal (and cross-appeal, in which the ACLU, however, does not join), which is by the City from an award by the district court to the class representatives of more than $1 million in attorneys’ fees. That is on top of at least $450,000 in fees awarded (and paid) for legal services rendered earlier in the proceedings; the true figure is undoubtedly higher but the parties’ records are incomplete, doubtless because of the great age of the case.
The $1 million award is for the following legal services rendered between 1994 and 2001: two proceedings for contempt of the decree, which failed; the opposition, just described, which also failed, to the modification of the decree; and efforts, which also bore no fruit so far as anyone cari say, to monitor the City’s compliance with the decree — no fruit, that is, except the failed contempt proceedings. The order is ap-pealable now because the activities for which the fees were awarded are complete in a sense that will become clearer in our discussion of the merits.
The awarding of attorneys’ fees under 42 U.S.C. § 1983 is governed by 42 U.S.C. § 1988(b), which authorizes such awards only to the prevailing party. See
Buckhannon Board & Care Home, Inc. v. West Virginia Dept. of Health & Human Resources,
In only two classes of case governed by section 1988(b) or similar fee-shifting provisions (see
Hensley v. Eckerhart,
And likewise in
Turner
the court made clear its belief that the postdecree efforts had been valuable in inducing compliance with the decree, even though they, too, had failed in detail.
In the second and more numerous class of cases, attorneys’ fees incurred in efforts to monitor compliance with the consent decree are said or assumed to be compen-sable even if no postjudgment order results from the efforts.
Eirhart v. Libbey-Owens-Ford Co.,
The catalyst idea was that if a lawyer’s effort produces a good result albeit not an actual judgment or other judicial relief, it should be compensable. The Court described it as the idea that a fee-shifting statute “allows an award where there is no judicially sanctioned change in the legal relationship of the parties,”
The plaintiffs are left to argue that the fact that the failed proceedings were offshoots of a pending case rather than entirely free-standing lawsuits should make a difference. Some cases have suggested that if postjudgment proceedings are “inextricably intertwined” with the original decree, in the sense of involving the same facts and legal issues, they can be considered part of the original case in which the plaintiff prevailed.
Cody v. Hillard, supra,
In any event, the postjudgment proceedings here, coming as they did so many years after the consent decree went into effect, are clearly separable from the proceeding that led up to the entry of the decree. Just as a bankruptcy proceeding provides a venue and procedural framework for prosecuting discrete claims (“adversary proceedings”) by and against the debtor — for example, a tort claim by the debtor that were he not in bankruptcy would be prosecuted as a civil suit in state court — so a consent decree will often, and here does, provide a venue and procedural framework for prosecuting discrete claims. For example, persons complaining of violations of the
Shakman
decree (actually decrees,
Shakman v. Democratic Organization of Cook County,
We do not think that our plaintiffs would argue that if a member of the class went to a lawyer who does not represent the class, and that lawyer filed a motion for contempt on behalf of his client and lost, the lawyer would be entitled to a fee, on the ground that the class member was a prevailing party by virtue of the consent decree. They would attempt to distinguish the case on the ground that the lawyers who negotiated the decree, or their successors (it has been, after all, more than two decades since the decree was entered), were “duty-bound” to enforce the decree— to monitor it, to file contempt proceedings on the basis of information obtained through the monitoring or otherwise, and to oppose any effort to lift or alter the decree. The attempt would fail. The decree imposes no such duties; nor does the law. The original decree contained provisions for audits and other methods of assuring compliance with the decree that were to be administered by the Chicago Police Board, while the modified decree requires annual audits by the Chicago Police Department and one audit to be performed by a national independent public accounting firm within five years. Neither the original nor the modified decree imposes on these lawyers any duty of operating the compliance machinery. They could have walked away from the case as soon as the consent decree was approved confident that a compliance machinery in which they had been given no role had been established. They would not have been letting down the class had they done so.
This brings out an important difference between this case and
Plyler
and
Turner.
Consider
why
the decrees in those cases but not in this one authorized the award of attorneys’ fees for postdecree litigation without requiring that the plaintiffs prevail in that litigation. The answer is that they must have been expected to be the enforcers of the decree. This is explicit in
Turner:
“the consent judgment provided that the representatives of the plaintiff class would establish the PMC [Plaintiffs’ Monitoring Committee, the entity seeking the award of fees] to, as its name suggests, monitor and ensure the continuing enforcement of the consent judgment.”
We are not suggesting that the lawyers for the class had to walk away from the *773 decree once it was entered. But they could not appropriate for themselves a guaranteed lifetime income by bringing and losing a series of actions to enforce the decree and charging the expense to the City and thus to the taxpayers. The class-action device is not intended to be a lawyers’ gravy train. The law does not reward lawyers for losing cases, or reimburse them for a prefiling investigation (which is the character of the plaintiffs’ monitoring of compliance with the decree) that leads to the filing of a losing suit. The plaintiffs sue and lose, the City defends and wins — but, we are told, the City is to pay its own expenses plus the plaintiffs’. We do not see the sense of that. The plaintiffs argue that unless they are reimbursed when they lose, their incentive to sue will be diminished because there is always a danger of losing a suit. Yes, and if section 1988 said that losing plaintiffs were entitled to awards of attorneys’ fees, they would have a good argument.
Their strongest claim is to fees for defending the decree against the modification sought by the City. It is strongest because they were responding rather than initiating and because modification is perhaps not as readily conceptualized as a separate suit adventitiously embedded in the underlying class action as is a proceeding for contempt or what we have characterized as a prefiling investigation. But we reject the claim for three reasons. First, we have conceptualized postdecree litigation, including collection litigation, broadly as a discrete phase analogous to a free-standing suit.
Resolution Trust Corp. v. Ruggiero,
Second, as we have noted already, the plaintiffs had no duty — statutory, contractual, or ethical — to oppose modification. Third, and further demonstrating the absence of even an ethical duty to oppose, their opposition verged on the unreasonable. The decree had been in force for 15 years when the City asked that it be modified.
If not limited to reasonable and necessary implementations of federal law, remedies outlined in consent decrees involving state [or, we add, local] officeholders may improperly deprive future officials of their designated legislative and executive powers. They may also lead to federal court oversight of state programs for long periods of time even absent an ongoing violation of federal law.... Rule 60(b)(5) allows a party to move for relief if “it is no longer equitable that the judgment should have prospective application.”... In Rufo v. Inmates of Suffolk County Jail,502 U.S. 367 ,112 S.Ct. 748 ,116 L.Ed.2d 867 (1992), the Court explored the application of the Rule to consent decrees involving institutional reform. The Court noted that district courts should apply a “flexible standard” to the modification of consent decrees when a significant change in facts or law warrants their amendment. Id., at 393,112 S.Ct. 748 .
Frew v. Hawkins,
— U.S. —,
Another point and we are done. The fees awarded in this case are for work dating back to 1994, a dozen years after the decree had been entered. This suggests the unwisdom of “conduct,” or “regulatory,” decrees (that is, equitable decrees that do not merely tell the defendant not to do or to stop doing something, but instead regulate his behavior and so impose a continuing duty of judicial supervision) that contain no sunset provision. Beginning in the Clinton Administration and continuing in the present Administration, the Justice Department and the Federal Trade Commission have both decided to include sunset provisions in their regulatory antitrust decrees. U.S. Dept. of Justice, Antitrust Division, Antitrust Division Manual ch. 4, p. 55 (3d ed.1998); Federal Trade Commission, Final Rule, “Duration of Existing Competition and Consumer Protection Orders,” 60 Fed.Reg. 58514, 58515 (Nov. 28, 1995). The absence of such a provision from the decree involved in the present case is, in retrospect, a considerable defect. It has meant that the plaintiffs have been encouraged to pursue enforcement efforts bound to fail because the decree, a response to the turmoil of the Vietnam War era now almost forgotten, is merely a relic.
The
modified
decree has a quasi-sunset provision. It provides that upon the completion of the independent audit that the modified decree orders but in any event no later than 2006, the district judge is to consider whether to dissolve the decree. We urge expedited completion of the audit, to clear the way to a prompt consideration of whether the decree has indeed outlived its usefulness, as in
People Who Care v. Rockford Board of Education,
The fee award is reversed. The plaintiffs are entitled to no fees for the legal services rendered during the period in issue, and a fortiori the Alliance is not entitled to the additional half million dollars sought in its cross-appeal.
REVERSED.
