James W. Myers, III, P.C., Crown Point, Ind., for plaintiffs-appellants.

Nick Katich, Lucas, Holcomb & Medrea, Merrillville, Ind., for defendants-appellees.

Before CUMMINGS, POSNER, and MANION, Circuit Judges.

POSNER, Circuit Judge.

This appeal requires us to decide a difficult and important question of free speech. Because the case was dismissed on the pleadings, we take as true the facts stated in the complaint. On May 5, 1987, the City of Hobart, Indiana, held its Democratic primary election. The mayor and several members of the city council were defeated. Many of the city's police had worked for the opposition during the primary campaign, and the lame-duck mayor and his lame-duck council were furious. A few weeks after the primary, the council enacted an ordinance requiring all city employees to work 2,080 hours a year (an average of 40 hours a week for every week in the year), minus sick leave and vacation. Anticipating the ordinance, the mayor had already issued an executive order directing that the ordinance be put into effect immediately upon enactment. Although the ordinance made no difference to the status of city employees who already worked a regular five-day week, it made a big difference to Hobart's police. They had been working a 5-2, 5-2, 5-3 schedule, meaning that they got three days off (rather than the regular two) every three weeks, and as a result averaged fewer than 39 hours a week. The ordinance required the police department to change to a regular five-day-a-week schedule.

Several weeks after the ordinance was enacted, the Hobart police union and two of its members filed this suit against the mayor and others, charging that the ordinance had been enacted to punish the police for opposing the mayor and council members in the primary. Laid under the ubiquitous 42 U.S.C. Sec . 1983, the complaint alleges that the ordinance violates the First Amendment, which the courts have, of course, held applicable to state (including municipal) action by virtue of the Fourteenth Amendment. The only relief sought in the complaint was an injunction against the ordinance. The district judge dismissed the suit under Fed.R.Civ.P. 12(b)(6) (failure to state a claim), on the ground that evidence concerning the motives of a legislative body is inadmissible to demonstrate a violation of the First Amendment. The plaintiffs asked to be allowed to amend the complaint to add a challenge to the executive order and a request for damages; the judge refused, on the ground that the switch of the plaintiffs' point of attack from the ordinance to the executive order was an attempt to evade the force of his ruling on the original complaint and that the suit had no merit whatever the relief sought. The plaintiffs appeal from both the dismissal of the original complaint and the refusal to let them file the amended complaint. We were told at argument that the ordinance has been repealed by the new city council--the insurgent Democratic slate that won the primary having gone on to win the general election as well. The request for an injunction is therefore moot, and the plaintiffs' only remaining stake is the cost to them of working longer hours during the period when the ordinance was in effect; for that is the measure of the harm they suffered from the defendants' alleged retaliation. See Ustrak v. Fairman, 781 F.2d 573, 579 (7th Cir.1986).

The defendants do not question that the First Amendment entitles Hobart's police to support candidates in a primary election, or that retaliation by public officials against the exercise of First Amendment rights is itself a violation of the First Amendment. See, e.g., Soderbeck v. Burnett County, 752 F.2d 285 (7th Cir.1985); Altman v. Hurst, 734 F.2d 1240, 1243 (7th Cir.1984) (per curiam); McGill v. Board of Education, 602 F.2d 774, 780 (7th Cir.1979); cf. Elrod v. Burns, 427 U.S. 347, 96 S.Ct. 2673, 49 L.Ed.2d 547 (1976); Branti v. Finkel, 445 U.S. 507, 100 S.Ct. 1287, 63 L.Ed.2d 574 (1980). The second proposition is vividly illustrated by our decision in Bart v. Telford, 677 F.2d 622 (7th Cir.1982), a case where the act of retaliation--which we held to be actionable under the First Amendment--was ridiculing a city employee for bringing a birthday cake to an office party; the ridicule was retaliation for her presumption in having run for the mayoralty against the incumbent. The posture of the present case prevents us from questioning that the only motive for the enactment of the ordinance was to punish the police for having opposed, and by opposing contributed to the defeat of, the very officials who enacted the ordinance. The council members failed to invoke their absolute legislative immunity (on which see, e.g., Supreme Court of Virginia v. Consumers Union of the United States, 446 U.S. 719, 732-34, 100 S.Ct. 1967, 1974-76, 64 L.Ed.2d 641 (1980)) in the district court, and have therefore waived it in this court. Finally, and it might seem dispositively, it is easy to name cases where legislators' intentions have been used to prove the unconstitutionality of their handiwork. See, e.g., Wallace v. Jaffree, 472 U.S. 38, 56-60, 105 S.Ct. 2479, 2489-92, 86 L.Ed.2d 29 (1985). Indeed, the Supreme Court has held that proof of discriminatory intent is indispensable to demonstrating that a state statute violates the equal protection clause, Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976); and what is indispensable is also admissible.

Despite all this, we agree with the district judge that the complaint states no claim under the Constitution. Cases like Washington v. Davis do not supersede--not entirely at any rate--the principle that courts "will not strike down an otherwise constitutional statute on the basis of an alleged illicit legislative motive." United States v. O'Brien, 391 U.S. 367, 383, 88 S.Ct. 1673, 1682, 20 L.Ed.2d 672 (1968); see also Palmer v. Thompson, 403 U.S. 217, 224-25, 91 S.Ct. 1940, 1944-45, 29 L.Ed.2d 438 (1971). This principle is founded not only on the difficulty of determining by forensic methods the motives of a collective body, but also on respect for the political process and on simple comity between departments of government. Its logic extends to a case like this, where, the offending ordinance having been repealed, the plaintiffs are seeking damages against those responsible for its enactment. The principle has certainly been qualified, however, see Washington v. Davis, supra, 426 U.S. at 244 n. 11, 96 S.Ct. at 2050 n. 11; we have to decide how much it has been qualified.

We think it survives undiminished in cases such as this where the statute or ordinance does not single out particular individuals or groups for benefits or burdens and is not challenged as discriminating on invidious grounds such as race, religion, and sex. In Vandenplas v. City of Muskego, 753 F.2d 555, 560 (7th Cir.1985), on which the plaintiffs in this case rely heavily and which contains a dictum that the motives of the members of the city council "would presumably be relevant" in "a case where facts implicating the First Amendment were properly alleged" (emphasis in original), the council had refused to pass a resolution that would have rescinded a "raze order" for the destruction of two buildings that the plaintiffs owned. Unlike this case, the council's action was directed against the plaintiffs and no one else. And in Feeney v. Massachusetts, 442 U.S. 256, 99 S.Ct. 2282, 60 L.Ed.2d 870 (1979), the statute unsuccessfully challenged as discriminating against women gave job preferences to veterans. It was apparent from the statute that one group, veterans, was being benefited, obviously at the expense of another group--nonveterans--to which the plaintiffs belonged.

No outside observer reading Hobart's 40-hour-a-week ordinance would suppose it directed against the police or any other definable group. It does not mention police, whereas the statute in Feeney mentioned veterans. It appears to be an utterly commonplace personnel regulation. Of course appearances can deceive; many laws as innocuous as Hobart's ordinance do harm particular groups, and often these are groups that are on the political outs with the enacting body. But who can be surprised that legislators reward their supporters and mulct their opponents? What else is new?This brings us to the second, and more cogent, ground of distinction between this case and cases such as Davis and Feeney where legislators' motives have been deemed relevant to the constitutionality of enactments. The political process is not impaired when legislators are merely forbidden to engage in invidious discrimination. It is impaired when legislators are forbidden to favor their supporters and disfavor their opponents. For that, after all, is politics, or rather an important part of politics. Politics is not merely, or even primarily, the disinterested pursuit of the public interest. It is a power struggle. See Grossart v. Dinaso, 758 F.2d 1221, 1236 (7th Cir.1985) (dissenting opinion). The First Amendment does not forbid the characteristic operation and outcome of representative government. Cf. Eastern Railroad Presidents Conference v. Noerr Motor Freight, Inc., 365 U.S. 127, 81 S.Ct. 523, 5 L.Ed.2d 464 (1961). It does not outlaw the interest-group state; it does not purge politics from politics. The desire to reward friends and punish enemies is not the ethic of the Sermon on the Mount, but neither is it on the same dismal plane with desiring to make entitlement to government benefits, or liability for the burdens of government, depend on immutable personal characteristics such as race or sex. We have expressed our reluctance to suppose, without clearer guidance from the Supreme Court, that the First Amendment is properly understood to have banished political considerations from public employment. See LaFalce v. Houston, 712 F.2d 292 (7th Cir.1983); see also Danenberger v. Johnson, 821 F.2d 361, 364 (7th Cir.1987); Horn v. Kean, 796 F.2d 668, 674-75, 678 (3d Cir.1986) (en banc); Avery v. Jennings, 786 F.2d 233, 237 (6th Cir.1986).

A pragmatic ground for rejecting the principle urged by the plaintiffs is that its acceptance would put at hazard a vast amount of routine legislation--federal, state, and local. Legislation passed by lame-duck legislatures, legislation passed by newly elected legislators--all would be subject to invalidation by a federal court upon evidence that the legislation, though on its face concerned only with the most ordinary matters of governmental administration, had actually been intended to punish the legislators' political opponents, or reward the legislators' friends with largesse obtained by taxes on their enemies. It would not even be necessary to show that the legislation had cost someone his job; these plaintiffs did not lose their jobs. The expansion of judicial review of legislation would be breathtaking. Yet the enlargement of the marketplace of ideas would be slight--maybe nonexistent. For every beneficiary of governmental favoritism (as the plaintiffs probably were before the ordinance was passed; for why should the police work fewer hours than other city employees?) who might be discouraged from speaking out against the political establishment there is a seeker of governmental favoritism whose zeal in advocacy and political campaigning will be dampened if legislators are prevented from rewarding supporters and punishing opponents.

Yet to refuse to take the step urged by the plaintiffs, large as it is, may seem to create two arbitrary distinctions: one between executive retaliation and legislative retaliation (O'Brien was a case about legislators' motives, not about the motives of executive officials, and we need not speculate on the applicability of its principle to executive action), the other between legislation that openly reveals its harmful effect and legislation that artfully conceals its effect. However, these closely related distinctions are not arbitrary. Executive retaliation is easily pinpointed against particular, and powerless, opponents. It is retail in character: not only because its target can be as narrow as a single person, but also because the executive does not face the same coordination problem as a multi-member body, a legislature. Legislative retaliation is doubly wholesale, and therefore much clumsier: The enactment of a statute or ordinance requires the concurrence of a majority of a plural body, and is therefore more difficult to accomplish; and a statute or ordinance affects a class of people, not just a particular, named individual, with the result that the victims, being more numerous, are somewhat more likely to have political remedies. Coniston Corp. v. Village of Hoffman Estates, 844 F.2d 461, 469 (7th Cir.1988). Against this it can be argued that executive retaliation is easier to spot and prove. But despite this point it is accepted that where a statute or ordinance does not have the generality characteristic of legislation--as in the resolution at issue in Vandenplas, and as in private bills and bills of attainder--a more beady-eyed examination of motive is appropriate. (Indeed, no legislature, state or federal, may pass a bill of attainder. See U.S. Const., art I, Secs. 9-10.) Cases like Washington v. Davis can be understood as allowing the broadened inquiry appropriate to private bills when the statute or ordinance that is challenged is directed against (or in favor of) a specified group and when the discrimination claimed goes beyond the usual bounds of political competition and becomes invidious.

The ordinance challenged in this case is not pinpointed against a named individual or group; it is general in its wording and impact. Policemen were not the only employees of the City of Hobart who were not working a regular five-day week before the ordinance; firemen were in the same boat. (The parties were not able to tell us at argument how the ordinance affected firemen.) And the ordinance did not hurt only the police who had opposed the incumbent mayor and councilmen; it hurt all the police. This made it a blunt retaliatory tool--perhaps one that only a lame duck would wield. And lame duck retaliation is the least intimidating kind (well, not completely, because the lame ducks have nothing to lose and are therefore beyond the reach of political remedies); it is likely to be undone by the sound ducks when they take office--as happened here. Most important, the ordinance is not attacked as invidious discrimination. Its target is not a racial, religious, or sexual group, but simply a group of political opponents. Cf. Grimes v. Smith, 776 F.2d 1359, 1366-67 (7th Cir.1985).

If the ordinance had been directed just at the police or just at opponents of the incumbents, it would have been more effective as retaliation because less likely to harm others besides those intended to be harmed. But by the same token it would have been subject to a more exacting scrutiny. Legislators' motives would be admissible, as we have seen. In the case of legislation not obviously disadvantageous to the plaintiff or his group, the principle that legislators' motives are inadmissible to demonstrate a violation of the First Amendment is compelling, in the absence of a demonstration that the sort of abuse of legislative power alleged in this case is widespread and is making serious inroads into the protection of First Amendment values.

If the mayor and police chief, who are defendants along with the City, the city council, and its (former) members, had procured the passage of the ordinance by fraud, as the police officers in our recently decided Jones case procured the plaintiff's prosecution by fraud, they could not take shelter behind the ordinance. See Jones v. City of Chicago, 856 F.2d 985, 993-94 (7th Cir.1988). The plaintiffs could obtain damages whether or not the ordinance would stand. (The ordinance might be fine, regardless of the motives for its enactment--to repeat an earlier point, why shouldn't Hobart's police work 40 hours a week?) But there is no allegation of fraud. The city council wanted to pass the ordinance, well aware of its retaliatory purpose and intended effects; the council members were not gulled like the prosecutors in Jones. If as we believe the ordinance is constitutionally invulnerable, executive officers cannot be punished for assisting without deception in its passage.

The city council of Hobart is small, the mayor played an important role, and the group primarily affected by the ordinance has only 24 members (the number of police in Hobart). So there is an argument for an exception to the general principle that legislative retaliation embodied in an enactment that is not invidious and does not purport to grant benefits to a defined group or impose burdens on such a group is not actionable. But it is desirable to keep law simple when that can be done without serious damage to important social interests, and the case for an exception is not a strong one here. All city employees are potentially affected by the ordinance, firemen particularly--and we do not know how many of them there are, nor how they might be affected. The city council is small, and the separation of powers is less elaborate on a local than on a state or federal level, but these points cut two ways: not only do they blur the distinction between executive and legislative conduct, which helps the plaintiffs; they also demonstrate the remoteness of the case from the central concerns of the First Amendment. The power of the government of Hobart, Indiana to stifle free speech is slight--not only because the writ of that government does not run wide but also because dissatisfied residents or employees of Hobart can leave Hobart at lower cost, psychic as well as financial, than a person oppressed by his national government can escape oppression by emigration. That is a virtue of our federal system.

The remaining question is whether the plaintiffs should be allowed to attack at least the mayor's action in issuing an executive order implementing the ordinance. The amended complaint not only dropped the council members as defendants but shifted the focus of attack from the ordinance to the executive order. The district judge regarded the amendments as a transparent effort to circumvent his ruling dismissing the original complaint, and we agree. The mayor had no authority not to put the ordinance into effect, given that it was--we have just held--a lawful ordinance. It is as if the plaintiffs had sued the council's messenger for carrying the draft of the ordinance to the printer.

The behavior charged in the complaint may be unedifying (though that depends on one's theory of the political process), but we do not see how the federal courts could remedy it without exceeding reasonable bounds on federal judicial intervention in the characteristic operations of the political process. We are not the city council of Hobart. The judgment of the district court dismissing the action and refusing leave to file an amended complaint is

AFFIRMED.