Rehearing and Rehearing En Banc Denied April 4, 1986.
POSNER, Circuit Judge.
This class action charges the State of Illinois with sex discrimination in employment, in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e, and the equal protection clause of the Fourteenth Amendment. The named plaintiffs are two associations of nurses plus 21 individuals, mostly but not entirely female, who work for the state in jobs such as nursing and typing that are filled primarily by women. The suit is on behalf of all state employees in these job classifications. The precise allegations of the complaint will require our careful attention later, but for now it is enough to note that they include as an essential element the charge that the state pays workers in predominantly male job classifications a higher wage not justified by any difference in the relative worth of the predominantly male and the predominantly female jobs in the state's roster.
Comparable worth is not a legal concept, but a shorthand expression for the movement to raise the ratio of wages in traditionally women's jobs to wages in traditionally men's jobs. Its premises are both historical and cognitive. The historical premise is that a society politically and culturally dominated by men steered women into certain jobs and kept the wages in those jobs below what the jobs were worth, precisely because most of the holders were women. The cognitive premise is that analytical techniques exist for determining the relative worth of jobs that involve different levels of skill, effort, risk, responsibility, etc. These premises are vigorously disputed on both theoretical and empirical grounds. Economists point out that unless employers forbid women to compete for the higher-paying, traditionally men's jobs — which would violate federal law — women will switch into those jobs until the only difference in wages between traditionally women's jobs and traditionally men's jobs will be that necessary to equate the supply of workers in each type of job to the demand. Economists have conducted studies which show that virtually the entire difference in the average hourly wage of men and women, including that due to the fact that men and women tend to be concentrated in different types of job, can be explained by the fact that most women take considerable time out of the labor force in order to take care of their children. As a result they tend to invest less in their "human capital" (earning capacity); and since part of any wage is a return on human capital, they tend therefore to be found in jobs that pay less. Consistently with this hypothesis, the studies find that women who have never married earn as much as men who have never married. To all this the advocates of comparable worth reply that although there are no longer explicit barriers to women's entering traditionally men's jobs, cultural and psychological barriers remain as a result of which many though not all women internalize men's expectations regarding jobs appropriate for women and therefore invest less in their human capital.
On the cognitive question economists point out that the ratio of wages in different jobs is determined by the market rather than by any a priori conception of relative merit, in just the same way that the ratio of the price of caviar to the price of cabbage is determined by relative scarcity rather than relative importance to human welfare. Upsetting the market equilibrium by imposing such a conception would have costly consequences, some of which might undercut the ultimate goals of the comparable worth movement. If the movement should cause wages in traditionally men's jobs to be depressed below their market level and wages in traditionally women's jobs to be jacked above their market level, women will have less incentive to enter traditionally men's fields and more to enter traditionally women's fields. Analysis cannot stop there, because the change in relative
Against this the advocates of comparable worth urge that collective bargaining, public regulation of wages and hours, and the lack of information and mobility of some workers make the market model an inaccurate description of how relative wages are determined and how they influence the choice of jobs. The point has particular force when applied to a public employer such as the State of Illinois, which does not have the same incentives that a private firm would have to use labor efficiently.
It should be clear from this brief summary that the issue of comparable worth (on which see the discussion and references in Paul Weiler, The Uses and Limits of Comparable Worth in the Pursuit of Pay Equity for Women, Discussion Paper No. 15, Program in Law and Economics, Harvard Law School, November 1985) is not of the sort that judges are well equipped to resolve intelligently or that we should lightly assume has been given to us to resolve by Title VII or the Constitution. An employer (private or public) that simply pays the going wage in each of the different types of job in its establishment, and makes no effort to discourage women from applying for particular jobs or to steer them toward particular jobs, would be justifiably surprised to discover that it may be violating federal law because each wage rate and therefore the ratio between them have been found to be determined by cultural or psychological factors attributable to the history of male domination of society; that it has to hire a consultant to find out how it must, regardless of market conditions, change the wages it pays, in order to achieve equity between traditionally male and traditionally female jobs; and that it must pay backpay, to boot. We need not tarry over the question of law presented by this example because as we understand the plaintiffs' position it is not that a mere failure to rectify traditional wage disparities between predominantly male and predominantly female jobs violates federal law. The circuits that have considered this contention have rejected it, see Spaulding v. University of Washington, 740 F.2d 686, 706-07 (9th Cir.1984); Lemons v. City & County of Denver, 620 F.2d 228 (10th Cir.1980); Christensen v. Iowa, 563 F.2d 353 (8th Cir.1977), and the AFSCME case discussed below; we shall see shortly that this rejection may be compelled by the Supreme Court's decisions in the Davis and Feeney cases.
The next question is whether a failure to achieve comparable worth — granted that it would not itself be a violation of law — might permit an inference of deliberate and therefore unlawful discrimination, as distinct from passive acceptance of a market-determined disparity in wages. The starting point for analyzing this question must be County of Washington v. Gunther, 452 U.S. 161, 101 S.Ct. 2242, 68 L.Ed.2d 751 (1981). Women employed to guard female prisoners were paid less than men employed to guard male prisoners. Since male prison inmates are more dangerous than female ones and since each male guard on average guarded ten times as many prisoners as each female guard, the jobs were not the same. Therefore, paying the male guards more could not violate the Equal Pay Act of 1963, 29 U.S.C. § 206(d), which requires equal pay only for equal work. The issue was whether it could violate Title VII, and the Court held that it could. A comparable worth study figured in this conclusion. The plaintiffs had alleged (and the allegation had to be taken as true for purposes of appeal, because the complaint had been dismissed, as in this case, for failure to state a claim) that the county had conducted a comparable worth
All that this seems to mean, as the dissenting Justices pointed out, is "that even absent a showing of equal work, there is a cause of action under Title VII when there is direct evidence that an employer has intentionally depressed a woman's salary because she is a woman. The decision today does not approve a cause of action based on a comparison of the wage rates of dissimilar jobs." Id. at 204, 101 S.Ct. at 2265 (emphasis in original). The relevance of a comparable worth study in proving sex discrimination is that it may provide the occasion on which the employer is forced to declare his intentions toward his female employees. In Gunther the county accepted (it was alleged) the recommendation of its comparable worth consultant regarding the male guards — decided to pay them "the full evaluated worth of their jobs" — but then rejected the recommendation regarding the female guards and did so because of "intentional sex discrimination," that is, because they were female, not because they had easier jobs or jobs that, for any reason, the market valued below the guarding of male prisoners (however a comparable worth consultant might value them).
The State of Illinois asks us to limit the teaching of Gunther to cases where the employer has accepted the recommendation of the comparable worth consultant with respect to the male job classifications. But that would be to take undue liberties with the Supreme Court's decision. The dissenting Justices pointed out that in limiting the Equal Pay Act to cases of equal work Congress had deliberately rejected liability based on the concept of comparable worth, and they argued that Congress had not intended to reverse field on the issue when it enacted Title VII a year later. The majority rejected this argument but left open "the precise contours of lawsuits challenging sex discrimination in compensation under Title VII." 452 U.S. at 181, 101 S.Ct. at 2254. It used the facts alleged in the case to argue that the dissenting Justices were exaggerating the impact of the decision on employers, but did not suggest that its holding was limited to cases with the same facts. So limited, its only effect would be to discourage employers from commissioning comparable worth studies.
Gunther suggests the type of evidence that is sufficient but perhaps not necessary to establish sex discrimination in wages for different work. A more recent case out of the State of Washington, American Federation of State, County & Municipal Employees (AFSCME) v. Washington, 770 F.2d 1401 (9th Cir.1985), suggests the type of evidence that is insufficient. The state's traditional policy had been to pay state employees the prevailing market rates of pay. Beginning in 1974, however, the state commissioned a series of comparable worth studies, each of which found that employees in predominantly female job classifications were paid about 20 percent less than employees in predominantly male job classifications judged to be of comparable worth. Eventually the state passed legislation providing for the phasing in over a decade of a wage system based on comparable worth. The suit charged that the state's failure to act sooner was a form of discrimination. The case was tried and the plaintiffs won in the district court, but the Ninth Circuit reversed. It held that a decision to pay market wages is not discriminatory, that "comparable worth statistics alone are insufficient to establish the requisite inference of discriminatory motive," id. at 1407, and that "isolated incidents" (id.) of intentional
The AFSCME case resembles our hypothetical case of the firm accused of sex discrimination merely because it pays market wages. AFSCME shows that such a case is not actionable under Title VII even if the employer is made aware that its pattern of wages departs from the principle of comparable worth to the disadvantage of women (plus the occasional male occupant of a traditionally woman's job) and even if the employer is not so much a prisoner of the market that it cannot alter its wages in the direction of comparable worth, as eventually the State of Washington did. The critical thing lacking in AFSCME was evidence that the state decided not to raise the wages of particular workers because most of those workers were female. Without such evidence, to infer a violation of Title VII from the fact that the state had conducted a comparable worth study would, again, just discourage such studies.
The plaintiffs can get no mileage out of casting a comparable worth case as an equal protection case. The Supreme Court held in Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976), that the equal protection clause is violated only by intentional discrimination; the fact that a law or official practice adopted for a lawful purpose has a racially differential impact is not enough. The Court applied this principle to sex discrimination in Personnel Administrator v. Feeney, 442 U.S. 256, 99 S.Ct. 2282, 60 L.Ed.2d 870 (1979). Massachusetts had a law giving preference in state employment to veterans. The preference was applicable to female as well as male veterans but of course most veterans are male. But as the purpose of the law was to benefit veterans of either sex rather than to favor men over women, the plaintiff's constitutional challenge failed. "`[D]iscriminatory purpose' ... implies more than intent as volition or intent as awareness of consequences. It implies that the decisionmaker ... selected or reaffirmed a particular course of action at least in part `because of,' not merely `in spite of,' its adverse effects upon an identifiable group." Id. at 279, 99 S.Ct. at 2296 (citation and footnotes omitted).
These holdings cast additional light on the contention that intentional discrimination can be inferred from the state's failure to eliminate wage disparities shown by the comparable worth report. Knowledge of a disparity is not the same thing as an intent to cause or maintain it; if for example the state's intention was to pay market wages, its knowledge that the consequence would be that men got higher wages on average than women and that the difference might exceed any premium attributable to a difference in relative worth would not make it guilty of intentionally discriminating against women. Similarly, even if the failure to act on the comparable worth study could be regarded as "reaffirming" the state's commitment to pay market wages, this would not be enough to demonstrate discriminatory purpose. To demonstrate such a purpose the failure to act would have to be motivated at least in part by a desire to benefit men at the expense of women.
Neither Davis nor Feeney were Title VII cases, a point emphasized in Davis. See 426 U.S. at 238-39, 96 S.Ct. at 2046-47. But when intentional discrimination is charged under Title VII the inquiry is the same as in an equal protection case. The difference between the statutory and constitutional prohibitions becomes important only when a practice is challenged not because it is intended to hurt women (say), but because it hurts them inadvertently and is not justified by the employer's needs — when, in short, the challenge is based on a theory of "disparate impact," as distinct from "disparate treatment" (= intentional discrimination). The plaintiffs in this case, however, have said that they are proceeding on the basis of disparate treatment rather than disparate impact. Their decision is understandable. In the usual disparate-impact case the plaintiff challenges
Another point is that the Bennett Amendment to Title VII (the last sentence in 42 U.S.C. § 2000e-2(h)) authorizes employers to pay different wages to men and women provided that the difference would be lawful under the Equal Pay Act, which allows unequal pay for equal work if the inequality results from "any ... factor other than sex," 29 U.S.C. § 206(d)(1)(iv). The Supreme Court in Gunther assumed without quite deciding that the Bennett Amendment allows an employer charged (necessarily under Title VII rather than the Equal Pay Act) with paying unequal wages for unequal work to defend by showing that the inequality is based on something other than sex, even if the result is a disparate impact. See 452 U.S. at 171, 101 S.Ct. at 2249. This reading would confine the scope of Title VII in a case such as the present to intentional discrimination.
So if all that the plaintiffs in this case are complaining about is the State of Illinois' failure to implement a comparable worth study, they have no case and it was properly dismissed. We must therefore consider what precisely they are complaining about. Our task would be easier if the complaint had been drafted with the brevity that the Federal Rules of Civil Procedure envisage though do not require. Before the era of modern pleading ushered in by the promulgation of the rules in 1938, a plaintiff to survive a motion to dismiss the complaint had to plead facts which if true showed that his legal rights had been invaded. The problem was that without pretrial discovery, which ordinarily could not be conducted before the complaint was filed, the plaintiff might not know enough facts to be able to make the required showing. For fact pleading the federal rules substituted notice pleading. The complaint would have to indicate the nature of the plaintiff's claim with only enough specificity to enable the parties to determine the preclusive effect of a judgment disposing of the claim ("a short and plain statement of the claim showing that the pleader is entitled to relief," Rule 8(a)(2)). The Appendix of Forms to the federal rules illustrates with a complaint for negligence that, so far as the invasion of the plaintiff's legal rights are concerned, says only: "On June 1, 1936, in a public highway called Boylston Street in Boston, Massachusetts, defendant negligently drove a motor vehicle against plaintiff who was then crossing said highway." Form 9, ¶ 2; and see Rule 84 ("the forms contained in the Appendix of Forms are sufficient under the rules and are intended to indicate the simplicity and brevity of statement which the rules contemplate"). The plaintiff was expected to use pretrial discovery to gather the facts showing the defendant's negligence and the defendant could serve contention interrogatories on the plaintiff to learn the theory behind the claim. See Rule 33(b), and Note of Advisory Committee to the 1970 amendment thereto. When discovery was complete, a pretrial order would be issued formulating the issues for trial; this order would perform many of the functions of the complaint in a system of fact pleading. See Rule 16. See generally Wright, The Law of Federal Courts § 68 (4th ed. 1983).
The idea of "a plain and short statement of the claim" has not caught on. Few complaints follow the models in the Appendix of Forms. Plaintiffs' lawyers, knowing that some judges read a complaint as soon as it is filed in order to get a sense of the suit, hope by pleading facts to "educate" (that is to say, influence) the judge with regard to the nature and probable merits of the case, and also hope to set the stage for an advantageous settlement by showing the defendant what a powerful case they
A plaintiff who files a long and detailed complaint may plead himself out of court by including factual allegations which if true show that his legal rights were not invaded. Kaiser Aluminum & Chemical Sales, Inc. v. Avondale Shipyards, Inc., 677 F.2d 1045, 1050 (5th Cir.1982); Associated Builders, Inc. v. Alabama Power Co., 505 F.2d 97, 100 (5th Cir.1974); Orthmann v. Apple River Campground, Inc., 757 F.2d 909, 915 (7th Cir.1985) (dictum); 5 Wright & Miller, Federal Practice and Procedure § 1357, at p. 604 (1969). The district judge thought the plaintiffs had done that here. Let us see.
The key paragraph of the complaint is paragraph 9, which reads as follows:
If this were the entire charging part of the complaint, there would be no question of dismissing it for failure to state a claim. The paragraph initially charges the state with intentional discrimination against its female employees, because of their sex; and this, standing alone, would be quite enough to state a claim under Title VII. It continues, "and because of their employment in historically female-dominated sex-segregated job classifications," and then adds a claim on behalf of male employees in those classifications. The continuation could be interpreted as an allegation that the state's failure to adopt a wage scale based on the principle of comparable worth violates Title VII, and if so fails to state a claim. But the mention of "sex-segregated" blurs the picture. If the state has deliberately segregated jobs by sex, it has violated Title VII. Anyway a complaint
Subparagraphs (a) through (g) present a list of particular discriminatory practices; and since they are merely illustrative ("not limited to"), the complaint would not fail even if none of them were actionable. Some are, some aren't. If (a), the "use of a sex-biased system for pay and classification which results in and perpetuates discrimination in compensation against women employed in historically female-dominated sex-segregated job classifications," just means that the state is paying wages determined by the market rather than by the principle of comparable worth, it states no claim. But if it means to allege that the state has departed from the market measure on grounds of sex — not only paying higher than market wages in predominantly male job classifications and only market wages in predominantly female classifications, but keeping women from entering the predominantly male jobs ("sex-segregated") — it states a claim. Subparagraph (b) adds nothing. If the state is discriminating against women by maintaining unwarranted wage differentials between predominantly male and predominantly female jobs, any men who happen to find themselves in predominantly female jobs will be, as it were, dragged down with the women — will be incidental victims of a discrimination targeted against others.
Subparagraph (c) is an effort to fit the case to the mold of Gunther. The critical difference, however, is that here the state is not alleged to have "evaluated" any of the predominantly female job classifications as being of comparable worth to predominantly male classifications. The Illinois Commission on the Status of Women, a public body, commissioned a comparable worth study which found the same sort of disparities that other comparable worth studies have found. The state itself — meaning the officials responsible for setting wage rates — has yet to reconfigure its wage system in accordance with the findings of the study, which is attached as an appendix to the complaint and is the evaluation to which paragraph 9(c) refers. But as we said earlier, the failure to accept the recommendations in a comparable worth study is not actionable. Paragraph 9(c) thus fails to state a claim — as does (d), which is the same as (c) except that it, like subparagraph (b), complains on behalf of male occupants of predominantly female jobs.
Subparagraphs (e) and (f) are inscrutable. If they complained about payment of unequal pay for the same work they would state a claim under the Equal Pay Act. But that Act is not cited in the complaint, perhaps deliberately, and the substitution of "work of equal skill" etc. for "equal work ... of equal skill" etc. may also be deliberate. The intention may be to claim that different pay for different but comparable work violates Title VII — and if so this is a comparable worth claim by a different name, and fails. However, when a defendant is unclear about the meaning of a particular allegation in the complaint, the proper course is not to move to dismiss but to move for a more definite statement. See Fed.R.Civ.P. 12(e); United States v. Employing Plasterers Ass'n, 347 U.S. 186, 189, 74 S.Ct. 452, 454, 98 L.Ed. 618 (1954).
That leaves subparagraph (g) — "Discrimination in classification." This could be a reprise of the comparable worth allegations or it could mean that in classifying jobs for pay purposes the responsible state officials had used the fraction of men in each job as a factor in deciding how high a wage to pay — which would be intentional discrimination.
Maybe the allegations in paragraph 9 are illuminated by subsequent paragraphs of the complaint. Paragraph 10, after summarizing the comparable worth study, says, "Defendants knew or should have known of the historical and continuing existence of patterns and practices of discrimination in compensation and classification, as documented at least in part by the State of Illinois Study." All that the study "documents," however, is that 28 percent of the
"Willfully" is, however, a classic legal weasel word. Sometimes it means with wrongful intent but often it just means with knowledge of something or other. Willful evasion of taxes means not paying when you know you owe tax. After reading the comparable worth study the responsible state officials knew that the state's compensation system might not be consistent with the principles of comparable worth ("might" because there has been no determination that the comparable worth study is valid even on its own terms — maybe it's a lousy comparable worth study). But it would not follow that their failure to implement the study was willful in a sense relevant to liability under Title VII. They may have decided not to implement it because implementation would cost too much or lead to excess demand for some jobs and insufficient demand for others. The only thing that would make the failure a form of intentional and therefore actionable sex discrimination would be if the motivation for not implementing the study was the sex of the employees — if for example the officials thought that men ought to be paid more than women even if there is no difference in skill or effort or in the conditions of work.
Paragraphs 11 through 31 of the complaint present the specific claims of each of the 21 named individual plaintiffs. In materially identical language each paragraph complains that the wage rate in the plaintiff's "job classification is discriminatorily depressed because it is historically female-dominated and because of defendants' continuing use of a sex-biased system of pay and classification," and that the "Defendants continue to pay discriminatorily low wages to plaintiff [name] because she [in one case he] works in a historically female-dominated job classification." Standing alone these allegations would appear to state merely a comparable worth claim. But they do not stand alone. In light of paragraphs 9 and 10, each of paragraphs 12 through 31 may conceivably (if somewhat improbably) be intended merely to identify each of the named plaintiffs as a victim of the discriminations alleged in the earlier paragraphs.
We have said that a plaintiff can plead himself right out of court. But the court is not to pounce on the plaintiff and by a crabbed and literal reading of the complaint strain to find that he has pleaded facts which show that his claim is not actionable, and then dismiss the complaint on the merits so that the plaintiff cannot replead. (The dismissal would preclude another suit based on any theory that the
Furthermore, a complaint is not required to allege all, or any, of the facts logically entailed by the claim. If Illinois is overpaying men relative to women, this must mean — unless the market model is entirely inapplicable to labor markets — that it is paying women at least their market wage (and therefore men more), for women wouldn't work for less than they could get in the market; and if so the state must also be refusing to hire women in the men's jobs, for above-market wages in those jobs would be a magnet drawing the women from their lower-paying jobs. Maybe the references in the complaint to the segregation of jobs by sex are meant to allege such refusals but if not this pleading omission would not be critical. A plaintiff does not have to plead evidence. If these plaintiffs admitted or the defendants proved that there was no steering or other method of segregating jobs by sex, the plaintiffs' theory of discrimination might be incoherent, and fail. But a complaint does not fail to state a claim merely because it does not set forth a complete and convincing picture of the alleged wrongdoing. So the plaintiffs do not have to allege steering even if it is in some sense implicit in their claim.
Before concluding that the district court should not have dismissed the complaint we must consider whether anything happened between the filing of the complaint and the district court's decision to show that the plaintiffs really were pleading just a comparable worth case. The first thing that happened was that the state filed its motion to dismiss or in the alternative for summary judgment. Attached to the motion were two affidavits, one from a state official attesting that the Illinois Nurses' Association had not pressed for comparable worth in collective bargaining negotiations involving some of the job classifications to which the named plaintiffs belong, the other from an economist attacking the concept of comparable worth. In response the plaintiffs argued that their case was not a comparable worth case but an intentional
But there was more to the plaintiffs' memorandum in opposition. Several pages were devoted to arguing that there was additional evidence of discrimination besides the results of the comparable worth study and the refusal to take corrective measures in accordance with the study. If these pages had been submitted in the form of a motion to amend the complaint, then, even though the judge's leave to amend would have been required (the plaintiffs had amended the complaint once, and that was the only amendment they were permitted to make without the judge's permission or the written consent of the defendants), he would have had to grant it. Leave to amend the complaint "shall be freely given when justice so requires," Fed.R.Civ.P. 15(a), and there could have been no prejudice to the defendants here, when the motion would have come so soon after the filing of the original complaint and before discovery had begun.
The plaintiffs attached to their memorandum more than 200 pages of exhibits. Much of this documentation is devoted to showing that the fraction of women varies across jobs and that the higher the fraction the lower the wage, and thus duplicates the comparable worth study. But some of it is relevant to showing deliberate discrimination. Some of it indicates or asserts that women are not classified in accordance with the work they actually do and as a result are paid less than the state's own rules entitle them to; that appointing "hand picked" successors when jobs become vacant and disseminating news of job openings by word of mouth rather than by public notice prevent women from competing for the higher-paid, predominantly male jobs; that women have been excluded from some job categories altogether, even though they are capable of performing the jobs; that predominantly female jobs have sometimes been abolished when layoffs were necessary, in order to spare men from being laid off; and that sometimes the same jobs have been given two classifications (e.g., Prison Clerk and Clerk Typist), one for men and one for women, with the former being paid more even though the work is identical. Much of this "evidence" would not be admissible at trial; some is outside the statute of limitations (and hence "merely an unfortunate event in history which has no present legal consequences," United Air Lines, Inc. v. Evans, 431 U.S. 553, 558, 97 S.Ct. 1885, 1889, 52 L.Ed.2d 571 (1977)); some is from before 1972, when Title VII was made applicable to state employees — and there is no legal duty to undo the effects of previous discrimination (see Hazelwood School District v. United States, 433 U.S. 299, 309, 97 S.Ct. 2736, 2742, 53 L.Ed.2d 768 (1977)). But we are not at the moment interested in whether the plaintiffs have produced any evidence of unlawful discrimination. That is the issue with regard to the propriety of summary judgment. The present issue is whether, even though the judge could not properly dismiss the complaint because it did not state a claim on its face, the plaintiffs may have put themselves out of court by their subsequent filing — a filing that might be compared to a response to a motion for a more definite statement under Rule 12(e) — which showed that the complaint was narrower than it might have seemed. Our answer is "no." The exhibits, treated not as evidence but as an elaboration of the complaint (as in a Rule 12(e) filing), confirm, what the language of the complaint suggests, that the complaint charges intentional discrimination of the sort made actionable by the Gunther case, as well as a mere failure to
The next question is whether the dismissal can be upheld on the alternative ground urged by the state, that its motion for summary judgment should have been granted. Although we are reluctant, without having the benefit of the district court's view on the matter, to uphold the dismissal of a complaint on the ground that the defendant was entitled to summary judgment, Martinez v. United Automobile Workers, Local No. 1373, 772 F.2d 348, 353 (7th Cir.1985), we would do so if it were quite clear that the plaintiff had failed to raise a genuine issue of material fact on an essential element of its claim. See Fed.R.Civ.P. 56(c); Hooks v. Hooks, 771 F.2d 935, 945 (6th Cir.1985); Experimental Engineering, Inc. v. United Technologies Corp., 614 F.2d 1244, 1247 (9th Cir.1980); cf. Jewel Cos. v. Pay Less Drug Stores Northwest, Inc., 741 F.2d 1555, 1564-65 (9th Cir.1984). The fact that discovery is not complete — indeed, has not begun — need not defeat the motion. A defendant may move for summary judgment at any time. Rule 56(b). If the plaintiff has not had time to do enough discovery to respond to the motion, his proper course is to ask for a postponement of the hearing on the motion. See, e.g., United States v. Light, 766 F.2d 394, 397-98 (8th Cir.1985).
Since the plaintiffs made no request for additional time, the defendants are entitled to summary judgment if as the record now stands the plaintiffs have failed to raise a genuine issue of fact regarding intentional discrimination. See, e.g., Powers v. Dole, 782 F.2d 689, 696-97 (7th Cir.1986); Bryson v. Royal Business Group, 763 F.2d 491, 495 (1st Cir.1985); Over the Road Drivers, Inc. v. Transport Ins. Co., 637 F.2d 816, 820-21 (1st Cir.1980); Brown v. Chaffee, 612 F.2d 497, 504-05 (10th Cir.1979). That might well be the situation had the defendants submitted affidavits or other evidence denying that they had engaged in intentional sex discrimination, but they did not. Their affidavits, being directed exclusively to the comparable worth theory of the complaint — being designed to show that the theory was not sound, either in general or in the circumstances of this case — did not address any other form of discrimination. This doesn't mean that the state was asleep at the switch; for when the state filed its motion it was not clear that the plaintiffs were seeking to press any other theory of discrimination besides comparable worth. Anyway, motions for partial summary judgment are permitted. See Fed.R.Civ.P. 56(a), (b). Nor will there by anything to prevent the state from filing a further motion for summary judgment supported by sworn denials of any intentional discrimination within the scope of the complaint. Lindsey v. Dayton-Hudson Corp., 592 F.2d 1118, 1121 (10th Cir.1979); Weit v. Continental Illinois Nat'l Bank & Trust Co., 478 F.Supp. 285, 292 n. 4 (N.D.Ill.1979), aff'd, 641 F.2d 457 (7th Cir.1981); Jones v. Wike, 654 F.2d 1129 (5th Cir.1981) (per curiam). But normally a motion for summary judgment unsupported by affidavits or materials of equivalent evidentiary standing that negate an essential element of the plaintiff's case cannot be granted with the effect of dismissing the entire complaint. Backes v. Valspar Corp., 783 F.2d 77, 79 (7th Cir.1986).
There is a conflict between the circuits over the question whether a defendant can ever get summary judgment without submitting evidence. Compare Catrett v. Johns-Manville Sales Corp., 756 F.2d 181 (D.C.Cir.), cert. granted under the name of Celotex Corp. v. Catrett, ___ U.S. ___, 106 S.Ct. 342, 88 L.Ed.2d 285 (1985), with Fontenot v. Upjohn Co., 780 F.2d 1190 (5th Cir.1986). We agree with Judge Bork, dissenting in Catrett, and with Judge Rubin, who wrote the opinion in Fontenot, that if the plaintiff fails to obtain admissible evidence with regard to an essential issue that the plaintiff has the burden of proving, the defendant can obtain summary
The state asks us, finally, to uphold dismissal on the ground that the difficulties of remedying wage discrimination between different jobs are insuperable. It points out that the district court would have to decide what the wages in the different jobs would have been but for discrimination in order to know how much backpay to award. But if difficulties of remedy, unless completely insurmountable, were a proper reason for throwing out a complaint at the pleading stage, Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954), would have been decided in favor of allowing public schools to continue to segregate the races. Even if, as the state argues, a court could not reasonably impose comparable worth on an employer even as a remedy for blatant discrimination, except in a case such as Gunther (or indeed AFSCME, had the plaintiffs there prevailed) where the employer has in some sense approved a plan for comparable worth, this would just mean that some other, though perhaps less effective, remedy would have to be substituted. Maybe the plaintiffs would have to be content with an injunction that would knock down any barriers to women's being hired in the traditionally men's jobs, thus allowing the market to eliminate any sex-linked wage differential not justified by a difference in demand and supply, as women gravitated into the higher-paying men's jobs. It is premature to conclude that there is no worthwhile remedy for the intentional discrimination that consists of overpaying workers in predominantly male jobs because most of those workers are male. We emphasize, however, that proof of this causality is essential and is not to be inferred merely from the results of a comparable worth study and from the refusal of the employer to implement the study's recommendations. We do not want to arouse false hopes; the plaintiffs have a tough row to hoe. They may lose eventually on summary judgment if discovery yields no more evidence than is contained in the unsupported assertions and stale and seemingly isolated incidents in the plaintiffs' exhibits. But the plaintiffs are entitled to make additional efforts to prove a case of intentional discrimination within the boundaries sketched in this opinion.
REVERSED AND REMANDED.