We have consolidated for decision two appeals reargued en banc on the same day. Although the makeup of the en banc court is slightly different in the two cases,^**** the similarity of the issues has persuaded us to treat the cases together. Unfortunately, a majority of the judges has not converged upon a single rationale for the resolution of all the issues in these cases. The purpose of this per curiam opinion is to describe the cases briefly, to announce the outcomes and indicate the basic lines of agreement and disagreement, to articulate the court's unanimous view with regard to the disposition of the state law issues, and to refer the reader to the separate, signed opinions that follow.

Both cases primarily charge sexual harassment of a female employee by a supervisory employee in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e et seq. And in both the plaintiff is appealing from a grant of summary judgment.

In Jansen, with which we begin, the plaintiff has additional claims--of retaliation in violation of Title VII and of intentional infliction of emotional distress in violation of the common law of Illinois. The entire court believes that the district judge was right to grant summary judgment for the defendant, Packaging Corporation of America, on both these claims. Some of the alleged acts of retaliation are outside the scope of Jansen's EEOC charge and are therefore waived. As for the other acts, PCA presented noninvidious reasons for them (for example, that Jansen was assigned a lunch hour--one of the alleged retaliatory acts--in order to assure that the phone in her department would be manned at all times). Since Jansen presented neither evidence that these reasons were mere pretexts nor any other evidence from which retaliatory intent could be inferred, the company was entitled to summary judgment. Dunning v. Simmons Airlines, Inc., 62 F.3d 863, 868-69 (7th Cir.1995). As for Jansen's claim of intentional infliction of emotional distress, it is preempted by the Illinois Human Rights Act, which confines claims of "civil rights violation" under Illinois law to proceedings under the Act. 775 ILCS 5/8-111(C). Sexual harassment is one of the civil rights violations specified in the Act, 775 ILCS 5/2-102(D), and Illinois' highest court has held therefore that common law tort claims that depend on allegations of sexual harassment may be brought only under the Act. Geise v. Phoenix Co. of Chicago, Inc., 159 Ill. 2d 507, 203 Ill. Dec. 454, 457-59, 639 N.E.2d 1273, 1276-78 (1994). Jansen's common law claim of infliction of emotional distress is supported by the identical factual allegations of her Title VII claims and is therefore preempted.

Jansen's principal claim is of sexual harassment in violation of Title VII. She was hired by PCA as a secretary to Al Antoni, the manager of the Tooling Services Department of PCA's Wheeling, Illinois plant. She has presented evidence that he subjected her to undesired and offensive sexual advances. He once intimated to her that he would hold up her raise if she didn't have sex with him. He held it up for a time, but eventually she did receive it and it was made retroactive. The incident with the raise is the core of her claim that she was subjected to what is known in the case law of sexual harassment under Title VII as "quid pro quo" harassment. In addition, she complains that Antoni's repeated advances created a hostile working environment. Reargument in Jansen was granted in advance of the release of the panel opinion.

Ellerth was employed in marketing in the Chicago office of the mattress-fabric division of Burlington Industries. Theodore Slowik, the division's vice president for sales and marketing, was not Ellerth's immediate supervisor, and was based in New York; but he was her supervisor's supervisor, and as such saw her in the course of business on a regular basis. He made sexual advances to her over a period of a year or so and from time to time intimated that she would not be promoted or otherwise do well at Burlington Industries unless she submitted to his advances. Ellerth argues that Slowik's conduct placed Burlington Industries in violation of Title VII on both a quid pro quo and a hostile-environment theory.

With regard to Jansen's claim of hostile-environment harassment, the entire court agrees that an employer who is negligent in the hiring, supervision, monitoring, or retention of the plaintiff's supervisor (Antoni) is liable for the supervisor's sexual harassment and that the plaintiff has submitted enough evidence of PCA's negligence to create a triable issue, so that summary judgment should not have been granted to PCA. All the judges with the exception of Judges Easterbrook, Rovner, and Wood believe that negligence is the only proper standard of employer liability in cases of hostile-environment sexual harassment even if as here the harasser was a supervisor rather than a coworker of the plaintiff. The view of these judges is set forth in Judge Flaum's opinion, which is joined by Judges Cummings, Bauer (as to No. 96-1361 (Ellerth)), Cudahy (as to No. 95-3128 (Jansen)) (with the reservations indicated in Judge Cudahy's separate opinion), Kanne (with the reservations indicated in Judge Kanne's separate opinion), and Evans; in Chief Judge Posner's opinion, which is joined by Judge Manion; in Judge Manion's opinion, which is joined by Chief Judge Posner; and in Judge Coffey's opinion. Judges Easterbrook, Rovner, and Wood, as explained in Judge Easterbrook's and Judge Wood's opinions, believe that the proper standard of employer liability in all cases of sexual harassment by a supervisor is respondeat superior, provided, however, that the harassment was committed by the supervisor in the course of exercising his actual or apparent supervisory responsibilities, was foreseeable, and subjects the employer to liability under the principles of the applicable state law. The view that the proper standard of care in cases of a supervisor's creation of a hostile working environment is negligence is thus the law of the circuit, as it is the majority's view.

Judge Flaum's opinion concludes that Jansen has a viable quid pro quo claim, as do Judges Easterbrook, Rovner, and Wood, though their route to this conclusion is different, as they do not believe that there should be any different standard for an employer's liability for supervisors' harassment depending on whether it is hostile-environment harassment or quid pro quo harassment. Chief Judge Posner and Judges Coffey and Manion disagree that Jansen has a viable quid pro quo claim, Chief Judge Posner and Judge Manion because they believe that strict liability for quid pro quo harassment should be limited to "company acts" (such as firing or demoting), as distinct from mere threats, and Judge Coffey because he rejects strict liability in quid pro quo cases and also because he deems Jansen to have waived her quid pro quo claim.

In Ellerth's case, the panel decision, reversing the grant of summary judgment, was issued, 102 F.3d 848 (7th Cir.1996), but, as is our practice, was vacated when rehearing en banc was granted. Judge Wood's opinion for the panel had held that Ellerth had presented enough evidence both of quid pro quo harassment, and of hostile environment harassment by a supervisory employee, to create triable issues, so that summary judgment should not have been granted to Burlington Industries either. All the judges except Judges Easterbrook, Rovner, and Wood believe that the hostile-environment claim was expressly waived by Ellerth in her briefs to the panel and that the dismissal of this claim should therefore be affirmed. All the judges except Chief Judge Posner and Judges Coffey and Manion believe that Ellerth's evidence of quid pro quo harassment was sufficient to create a genuine issue of material fact, thus precluding summary judgment, although the routes to this conclusion are different. As noted earlier, Chief Judge Posner and Judge Manion believe that an employer's liability for quid pro quo harassment should be limited to company acts, as explained in their opinions, as opposed to mere threats by the supervisor, and there were no company acts here. Judge Coffey believes, as also noted earlier, that there is no strict liability in a quid pro quo case and that there is no proof of negligence on the part of Burlington Industries with respect to Slowik's harassment of Ellerth.

The court's inability to forge a majority position with regard to the proper standard for evaluating an employer's liability for sexual harassment by a supervisory employee means that panels of the court that have similar cases in the future, and the district judges of this circuit on remand in these cases and in similar future cases, will have to determine and be guided by the narrowest grounds for the decisions in these two cases. Marks v. United States, 430 U.S. 188, 193, 97 S. Ct. 990, 993, 51 L. Ed. 2d 260 (1977). Perhaps in some future case this court will be able to forge a majority position; perhaps the Supreme Court will bring order to the chaotic case law in this important field of practice. In the meantime, it is important for the district courts of this circuit to recognize in this welter of opinions that certain views do command a majority within our court: in particular, that the standard for employer liability in cases of hostile-environment sexual harassment by a supervisory employee is negligence, not strict liability, and that liability for quid pro quo harassment is strict even if the supervisor's threat does not result in a company act. These principles will bind the panels of this court and the district courts of this circuit until the Supreme Court resolves the issues.

In accordance with the foregoing discussion, the judgment in Jansen is affirmed with respect to the claims of retaliation and intentional infliction of emotional distress, and is otherwise reversed and remanded; the judgment in Ellerth is affirmed with respect to the claim of hostile-environment harassment but is reversed with respect to the claim of quid pro quo harassment; both cases are remanded to the district court for further proceedings not inconsistent with this opinion.

FLAUM, Circuit Judge, concurring, joined by CUMMINGS, BAUER (in No. 96-1361), CUDAHY (in 95-3128), KANNE (in part), and EVANS, Circuit Judges.

Title VII makes it unlawful for any employer "to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual's race, color, religion, sex, or national origin." 42 U.S.C. § 2000e-2(a)(1). It is well established that this language encompasses a prohibition on sexual harassment in the workplace. Harris v. Forklift Sys., Inc., 510 U.S. 17, 20-22, 114 S. Ct. 367, 370, 126 L. Ed. 2d 295 (1993); Meritor Savings Bank, FSB v. Vinson, 477 U.S. 57, 64-65, 106 S. Ct. 2399, 2404-05, 91 L. Ed. 2d 49 (1986). In Meritor the Supreme Court instructed that traditional agency principles should guide employer liability rules in the harassment context. 477 U.S. at 72, 106 S.Ct. at 2408. The Court also cautioned, however, that "such common law principles may not be transferrable in all their particulars to Title VII." Id. Accordingly, while I believe that agency principles should inform and "limit" this court's analysis of employer liability, see id., I share Chief Judge Posner's concern that an undue reliance on the Restatement 2d of Agency ("the Restatement") could obscure the issues at stake.1 A frank discussion of the policies underlying this federal imposition of liability upon an employer is necessary.

The original objective of Congress in enacting Title VII of the Civil Rights Act was to achieve equality in employment opportunities through the eradication of discriminatory barriers. See Griggs v. Duke Power Co., 401 U.S. 424, 430-31, 91 S. Ct. 849, 853, 28 L. Ed. 2d 158 (1971). By definition, harassment that alters the terms and conditions of employment--as it must to be actionable, see Meritor, 477 U.S. at 67, 106 S.Ct. at 2405-06--denies its victims (almost invariably women) an opportunity to succeed equal to that of employees who remain free of such intolerable treatment. To further the goals of the Civil Rights Act, therefore, the imposition of liability under Title VII seeks to deter harassment, and our resolution of this issue, to the fullest extent possible, should be geared toward minimizing its occurrence. See 29 C.F.R. § 1604.11(f) ("Prevention is the best tool for the elimination of sexual harassment."). As Congress amended Title VII in the Civil Rights Act of 1991 to permit victims to collect compensatory and punitive damages, see Williams v. Banning, 72 F.3d 552, 553 (7th Cir.1995); cf. Dockter v. Rudolf Wolff Futures, Inc., 913 F.2d 456, 461 (7th Cir.1990), liability now serves to compensate victims of harassment as well. Yet the goal of ensuring equal employment opportunities for both men and women dictates that deterrence be the foremost objective.I.

With this premise as my starting point, I turn first to a discussion of the appropriate standard for employer liability in instances of quid pro quo harassment. Quid pro quo harassment occurs where "submission to sexual demands is made a condition of tangible employment benefits." Dockter, 913 F.2d at 461; see also Nichols v. Frank, 42 F.3d 503, 511 (9th Cir.1994) (quid pro quo occurs where supervisor "conditions a job, a job benefit, or the absence of a job detriment, upon an employee's acceptance of sexual conduct"). Pre-Meritor, we held in Horn v. Duke, 755 F.2d 599, 603 (7th Cir.1985), that employers were strictly liable for quid pro quo harassment perpetrated by their supervisory employees. Our holding in Horn was premised in part on agency law, see id. at 605, but was prior to Meritor's explicit directive. Until the panel decision in Ellerth, this court had yet to revisit our holding more fully guided by agency principles. See Baskerville v. Culligan Int'l Co., 50 F.3d 428, 431-32 (7th Cir.1995). In the period since Meritor, seven other circuits have stated that an employer should be held vicariously liable for a supervisor's quid pro quo harassment under agency principles. See Davis v. City of Sioux City, 115 F.3d 1365, 1367 (8th Cir.1997); Nichols, 42 F.3d at 513-14 (9th Cir.1994) (collecting cases); Bouton v. BMW of N. Am., Inc., 29 F.3d 103, 106-07 (3d Cir.1994); Karibian v. Columbia Univ., 14 F.3d 773, 777 (2d Cir.), cert. denied, 512 U.S. 1213, 114 S. Ct. 2693, 129 L. Ed. 2d 824 (1994); Sauers v. Salt Lake County, 1 F.3d 1122, 1127 (10th Cir.1993); Kauffman v. Allied Signal, Inc., 970 F.2d 178, 185-86 (6th Cir.), cert. denied, 506 U.S. 1041, 113 S. Ct. 831, 121 L. Ed. 2d 701 (1992); Steele v. Offshore Shipbuilding, Inc., 867 F.2d 1311, 1316 (11th Cir.1989).2 I believe this circuit should follow suit.

Agency law supports the imposition of vicarious liability upon employers for supervisory quid pro quo harassment. Section 219(a) of the Restatement provides that an employer is liable for torts of an employee committed while acting within the scope of his or her employment. We have previously explained that "[t]he deliberate act of an employee acting within the scope of his authority is the act of the employer, for an employer, at least where it is a corporation, acts only through agents." Shager v. Upjohn Co., 913 F.2d 398, 404 (7th Cir.1990); see also Lancaster v. Norfolk & Western Ry., 773 F.2d 807, 820 (7th Cir.1985), cert. denied, 480 U.S. 945, 107 S. Ct. 1602, 94 L. Ed. 2d 788 (1987); Horn, 755 F.2d at 605. Scope of employment, however, is an amorphous term. While Judge Wood in her examination of the subject would hinge the imposition of liability on "scope of employment" delineated by the "meaningful limitation[s]" of section 228 of the Restatement, see Wood, p. 573, I harbor reservations whether those limitations are indeed instructive.3 Section 228 appears to have little relevance in the sexual harassment setting. In my judgment, reliance on the Restatement in this instance is the type of formalism that does not get us far in practice.

In instances of discrimination, courts have relied upon the concept of delegated authority to flush out the term "scope of employment." (Agency purists may choose to recognize "delegated authority" as a wholly alternative theory.) When a supervisor wields the authority actually delegated to him to dole out job benefits and detriments in order to condition such employment consequences upon receipt of sexual favors, "the supervisor, by definition, acts as the company." Steele, 867 F.2d at 1316; see also Kotcher v. Rosa and Sullivan Appliance Ctr., Inc., 957 F.2d 59, 62 (2d Cir.1992) ("[T]he supervisor is deemed to act on behalf of the employer when making decisions that affect the [job benefits] of the employee. From the perspective of the employee, the supervisor and the employer merge into a single entity."); Bouton, 29 F.3d at 106-07 ("Scope of employment liability is often invoked in quid pro quo cases because the supervisor has used his authority over the employee's job to extort sexual favors"); Nichols, 42 F.3d at 514 ("A harasser is able to grant such job benefits or detriments only because he has ... authority to do so 'delegated to him by his employer' ....") (internal citations omitted). It is my belief that liability for sexual harassment should stem from the delegation of authority which can be used to harass.4

Although we did not formally adopt this "delegated authority" justification in Horn, we recognized and explained the argument:

[B]y delegating power to [a supervisor], the "employer" and [the supervisor] essentially merged; as long as the tort complained of was caused by the exercise of this supervisory power, [the supervisor] should be deemed as acting within the scope of his employment, and the employer should be held liable for the tort.

755 F.2d at 605. We have also noted the argument approvingly in several post-Meritor cases, including Shager, where we explained that "a supervisory employee who fires a subordinate is doing the kind of thing that he is authorized to do, and the wrongful intent with which he does it does not carry his behavior so far beyond the orbit of his responsibilities as to excuse the employer." 913 F.2d at 405; see also Hunter v. Allis-Chalmers Corp. Engine Div., 797 F.2d 1417, 1422 (7th Cir.1986). In sum, because a supervisor would be unable to engage in quid pro quo harassment without the authority and power furnished by the employer, the supervisor's conduct is properly imputed to the employer. See Bouton, 29 F.3d at 107; see also Meritor, 477 U.S. at 70-71, 106 S.Ct. at 2407-08 (outlining position of EEOC: "where a supervisor exercises the authority actually delegated to him by his employer, by making or threatening to make decisions affecting the employment status of his subordinates, such actions are properly imputed to the employer whose delegation of authority empowered the supervisor to undertake them").

These precedential statements and the Restatement answer only the question of whether agency law supports the imposition of vicarious liability. The question remains whether the imposition of liability on the employer for quid pro quo harassment serves Title VII's goal of deterring sexual harassment. Implicit in this question is whether a negligence standard better serves the objective of deterrence. The imposition of liability beyond that which is likely to deter sexual harassment serves no constructive purpose and unnecessarily imposes costs on employers, costs that are ultimately passed on.

In choosing an appropriate standard of liability, we must ask what it is that we expect employers to do in preventing their employees from engaging in harassment. If courts expect only that companies, once made aware of harassment, will take appropriate remedial measures, negligence would appear the appropriate standard. A negligence standard, which asks whether an employer knew or should have known about an employee's acts of harassment and failed to take appropriate remedial action, see Baskerville, 50 F.3d at 432; Juarez v. Ameritech Mobile Communications, Inc., 957 F.2d 317, 320 (7th Cir.1992); Guess v. Bethlehem Steel Corp., 913 F.2d 463, 465 (7th Cir.1990), necessarily focuses on the steps that employers should take in detecting and subsequently correcting harassment. Underlying a negligence theory is the assumption that employers should not be liable for harassment that they could not be expected to detect. Pursuant to a negligence standard for supervisory harassment, we would presumably require a posted anti-sexual harassment policy, a grievance procedure that allows a complainant to circumvent the supervisory chain of command, and prompt remedial action. See Baskerville, 50 F.3d at 432 (discussing "reasonable" company actions in co-worker harassment situation); see also Meritor, 477 U.S. at 72-73, 106 S.Ct. at 2407-09 (existence of discrimination policy and grievance procedure "relevant" to negligence analysis, but not "necessarily dispositive"). Liability would ensue where the policy or process was inadequate or the handling of an identified harasser was deficient. Regrettably, inherent in this scheme is that an harasser will always possess the opportunity to make at least one quid pro quo demand before an employer would incur liability.

While posted policies and grievance procedures are important, I believe that the remedial goals of Title VII demand more. Companies' efforts to deal with sexual harassment should be systemic and proactive, rather than discrete and reactive. We know that companies can implement grievance procedures and discipline wayward employees; but we also know that companies can hire, train, and promote employees with an eye toward preventing undesirable behavior. In the abstract, a negligence standard conceivably could account for a company's systemic efforts to promote a workplace free of sexual harassment. Employers who had not done enough to reduce the likelihood of harassment throughout the workplace would be found negligent, even if they had no notice that a specific employee was a harasser. In reality, a negligence standard tends to focus on a company's response to specific instances of harassment. Yet a company's reasonable response to a known harasser is not necessarily indicative of reasonable efforts to prevent harassment from occurring in the first place. By refusing to allow companies to escape liability, even in the first instance of supervisory quid pro quo harassment, we better serve the goal of deterrence.5

Chief Judge Posner warns that if the costs of preventing harassment exceed the harm of the harassment (measured in terms of the harassed employees' recoveries), companies will prefer to pay the occasional judgment rather than to take what he views as unreasonable measures to reform the workplace. See Posner, p. 511. This is an important point, for it suggests that there may be limits beyond which a judicially-fashioned rule, despite its good intentions, will have no power to prevent the discriminatory behavior that Title VII is designed to eradicate. I believe, however, that this concern is premised on an overly pessimistic view of companies' ability to influence the working environment. Cf. Coffey, p. 546. Companies have available a variety of measures--short of installing video surveillance cameras--through which they can both monitor and impact behavior patterns in the workplace. The costs of preventing future harassment essentially represent an investment in long-term freedom from liability. A company faced with recurring liability for harassment perpetrated by its supervisory employees will have an incentive to foster a culture in which harassment is not tolerated.

Defining an actionable quid pro quo, of course, is central to the liability standard. A minority of this court is of the opinion, as was the district court in Jansen, that an adverse job consequence is necessary to succeed on a quid pro quo claim (this idea takes the guise of "company act" in Chief Judge Posner's opinion). See Posner pp. 513-515; Manion, pp. 559-560; Kanne, p. 505. Thus, where the employee submits, or the supervisor turns out to have bluffed, or the harassment is reported before any follow-through on the quid pro quo threat, there is no cause of action under a quid pro quo theory. I cannot agree: a clear and serious quid pro quo threat alters the "terms and conditions" of employment in such a way as to violate Title VII and therefore can constitute an actionable claim even if the threat remains unfulfilled.

Principles of agency law support this construction. Making a clear and unambiguous threat of an adverse job consequence is an exercise of the supervisor's delegated authority, just as actually inflicting the consequence invokes that authority. Included in the supervisor's delegated power to hire, fire, promote, evaluate is the authority to threaten to use that power if certain conditions or requirements are not satisfied. For example, no one would question that a supervisor was exercising the authority entrusted to him if he informed an employee that the employee would be fired if his work did not improve. Thus, a supervisor's quid pro quo threat, i.e., a threat that clearly conditions concrete job benefits or detriments on compliance with sexual demands, can be imputed to the employer for purposes of liability. This position has a toehold in the case law.6 Cf. Posner, p. 514 (citing cases requiring adverse job consequence). It should be noted, that many of the cases stating that an adverse job consequence is an element of a sexual harassment claim were decided prior to the 1991 amendments to the Civil Rights Act, when only equitable relief was available. See, e.g., Chamberlin v. 101 Realty, Inc., 915 F.2d 777, 783 (1st Cir.1990); Hicks v. Gates Rubber Co., 833 F.2d 1406, 1414 (10th Cir.1987); Highlander v. K.F.C. Nat'l Mgmt. Co., 805 F.2d 644, 648 (6th Cir.1986). In that context, it arguably made sense to require some realized adverse action. Now that emotional damage is compensable, the scope of quid pro quo should encompass clear unfulfilled threats that cause serious emotional harm.

This construction of an actionable quid pro quo serves Title VII's goal of deterrence as well. If the aim is to guard against sexual harassment, there is no justification for distinguishing between those cases in which a supervisor carries through on his threats from those cases in which the employee either submits to the supervisor's demands or in which the supervisor does not act upon his threats. Liability in this situation serves the goal of preventing such abuse from occurring by creating incentives for companies to take steps in hiring and training their supervisors. And while compensation of the victim may be secondary in overall importance to the goal of deterrence, it is undeniable that an employee can be harmed by an unfulfilled quid pro quo. A supervisor's unambiguous communication that adverse job action is imminent if sexual favors are not forthcoming causes the employee real emotional strife. An employee who does not submit may well undergo anxiety, distress, and loss of productivity regardless of whether the threat is carried out. Further, holding employers responsible for threats that clearly condition tangible job benefits upon the receipt of sexual favors and result in injury, even if no adverse job consequences follow, is consistent with the fact that the employee who submits to such a threat is afforded a remedy. See, e.g., Nichols, 42 F.3d 503; Karibian, 14 F.3d 773. Employees who have the wherewithal to call the supervisor's "bluff" and suffer emotionally as a consequence should not have to go uncompensated, nor should a "bluff" so likely to cause harm go unrecognized by the law. The potential to succeed against a company on a claim of such an abuse of power is no "windfall."

That said, and cabined by Meritor's directive, we must recognize an exception to the rule of vicarious liability for unambiguous quid pro quo threats: no liability should result where a victim or plaintiff could not have reasonably believed that it was within the supervisor's power to affect the conditions of the plaintiff's job. As explained above, where the supervisor is using his actual delegated authority, liability may be imposed on the company. An employer may also be held liable where a supervisor's quid pro quo threat exceeds his actual authority, but the victim reasonably relies on the supervisor's threat because of his apparent authority.7 See Restatement § 219(2)(d). Apparent authority, appropriately described by Judge Wood as "the shadow of actual authority," see Wood, p. 575, exists only "to the extent that it is reasonable for the third person dealing with the agent to believe that the agent is authorized," and the third person actually believes the agent to be authorized. Restatement § 8 cmt. c. Thus, only where it would be reasonable for a victim of harassment to believe that the authority used to harass had been delegated to the supervisor would liability ensue.

I now turn to the employer's liability for a supervisor's creation of a hostile work environment. The Supreme Court has held that Title VII offers protection for a hostile work environment, defined as a workplace "permeated with 'discriminatory intimidation, ridicule, and insult' ... that is 'sufficiently severe or pervasive to alter the conditions of the victim's employment and create an abusive working environment.' " Harris, 510 U.S. at 21, 114 S.Ct. at 370 (quoting Meritor Savings Bank, FSB v. Vinson, 477 U.S. 57, 67, 106 S. Ct. 2399, 2405, 91 L. Ed. 2d 49 (1986)). Traditionally, plaintiffs alleging and courts assessing sexual harassment have differentiated between quid pro quo and hostile work environment claims. See Meritor, 477 U.S. at 65, 106 S.Ct. at 2404-05;8 Dockter, 913 F.2d at 458. It could be argued that no principled distinction can be drawn between the two types of harassment in cases where the harassment is perpetrated by a supervisor. In either instance, the supervisor may use the authority delegated to him to harass, for, as Justice Marshall observed in his Meritor concurrence, "[a] supervisor's responsibilities do not begin and end with the power to hire, fire, and discipline employees, or with the power to recommend such actions. Rather, a supervisor is charged with the day-to-day supervision of the work environment and with ensuring a safe, productive workplace." 477 U.S. at 76, 106 S.Ct. at 2410. A minority of this court takes this position and treats sexual harassment under a unified approach informed by the Restatement. See Wood, pp. 566-570.

I believe sound reasons exist to distinguish between liability standards for quid pro quo and hostile work environment harassment.9 My conclusion that harassment of the quid pro quo variety should give rise to employer liability is premised on the belief that companies can effectively communicate to their employees the message that such behavior will not be tolerated. Yet when it comes to the hostile work environment, the courts have been hard pressed even to define the phrase. See Harris, 510 U.S. at 24, 114 S.Ct. at 372 (Scalia, J., concurring) (" 'Abusive' (or 'hostile,' which in this context I take to mean the same thing) does not seem to me a very clear standard--and I do not think clarity is at all increased by adding the adverb 'objectively' or by appealing to a 'reasonable person's' notion of what the vague word means."); see also id. at 24, 114 S.Ct. at 371 (opinion of the court) ("This is not, and by its nature cannot be, a mathematically precise test ... but we can say that whether an environment is 'hostile' or 'abusive' can be determined only by looking at all the circumstances."); Baskerville, 50 F.3d at 430 ("It is not a bright line, obviously, this line between a merely unpleasant working environment on the one hand and a hostile or deeply repugnant one on the other."); Nichols, 42 F.3d 503, 510 (9th Cir.1994) ("Questions relating to love and sex are among the most difficult for society to answer--in or out of the workplace--and courts are hardly experts in the realm. Still, we must find ways to define sexual harassment."). Given this definitional difficulty, as well as the variety of views in society as to what constitutes abusiveness, see Nichols, 42 F.3d at 510, I have less confidence in employers' abilities to communicate a consistent message to their employees regarding what constitutes a hostile work environment. Further, it may not be possible for employers to take the measures necessary to eradicate all arguably offensive conduct from the workplace. Because " 'abusiveness' is to be the test of whether legal harm has been suffered, opening more expansive vistas of litigation," Harris, 510 U.S. at 24, 114 S.Ct. at 372 (Scalia, J., concurring), I believe that a more judicious approach is called for when it comes to imposing vicarious liability in hostile work environment cases.

While agency principles may not "limit" the imposition of a stronger standard of liability in the context of a supervisory hostile work environment, see Wood, p. 573,10 we should not lose sight of the purpose of imposing liability on supervisors: deterrence. Because a hostile work environment is difficult to define, employers may be unable to send an unambiguous message to employees. I believe the appropriate inquiry in dealing with this conduct remains whether the company has taken due care to prevent harassment and to respond to complaints of harassment. Traditionally, courts have viewed negligence as a close approximation of what the prevention of harm should cost to the company. See United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir.1947) (Learned Hand's famous test); see also McCarty v. Pheasant Run, Inc., 826 F.2d 1554 (7th Cir.1987) (collecting cases). By its conception, negligence possesses the flexibility to respond by degree to amorphous or variable harms, such as this one. We observed in Carr v. Allison Gas Turbine, "the greater the [sexual] harassment--the more protracted or egregious, as distinct from isolated ... or ambiguous, it is--the likelier is the employer to know about it or to be blameworthy for failing to discover it." 32 F.3d 1007, 1009 (internal citations omitted). Given courts' definitional uncertainty, the cumulative nature of such harassment, and my unwillingness to unfairly penalize an employer, it is best to let the standard of liability remain at negligence.

Further, a negligence standard dovetails with the Supreme Court's construction of hostile environment as "severe and pervasive," so that, in theory, a harassed employee will not be left without a remedy against a complicit company. In the overwhelming majority of cases, a plaintiff will have no cause of action until, at a minimum, two instances of harassment have accrued. If we assume prompt reporting (i.e., reporting at initial stages of harassment),11 employer liability will be triggered by the very creation of the hostile environment, as the employer will have allowed the harassment to reach the level of "severe and pervasive." As explained above, the law should not tolerate "one free" quid pro quo demand. Similarly, the law should not exhibit patience with an employer that permits harassment, once reported, to reach the point where it creates a hostile work environment.

This is not to say that we should ignore the reality that when the authority wielded by supervisors over subordinates is used to create a hostile environment, employees may be less likely to report such harassment. While the appropriate standard for both supervisor and co-worker hostile work environment harassment is negligence, I agree with Chief Judge Posner, see Posner, p. 511, that what constitutes due care may not be the same in both cases. Because employers entrust supervisors with authority over other employees who may be abused, harassment by a supervisor is more likely to affect the "terms and conditions of employment." A heightened duty of care should apply both in terms of the machinery available to employees to complain of harassment and in terms of a company's duty to investigate and respond effectively to such complaints. See Meritor, 477 U.S. at 73, 106 S.Ct. at 2408 (grievance procedure flawed because it required reporting to supervisor); cf. Carr v. Allison Gas Turbine, 32 F.3d 1007, 1009 (7th Cir.1994) ("It would be unrealistic to expect management to be aware of every impropriety committed by every low-level employee.").12 This intensified concern for supervisory abuse is implicit in the concept of negligence and in the Supreme Court's multi-factor approach, see Harris v. Forklift, 510 U.S. at 22-23, 114 S.Ct. at 371; it is manifest in the Supreme Court's reaction to Meritor's grievance procedure. Conscious of the power dynamic between supervisor and supervisee and eager for the realization of Title VII's goal, I stress it here.

I conclude with an application of my suggested approach to the facts of these cases, the elaboration of which I leave to the per curiam opinion, Chief Judge Posner, see Posner, at pp. 515-516, Judge Coffey, see Coffey at pp. 546-551, and Judge Wood, see Wood, at pp. 575-578. Under the standards set forth above, Jansen has created genuine issues of material fact as to whether she was subjected to quid pro quo harassment. Although it is by no means a foregone conclusion, a reasonable jury could find that by telling Jansen, "I haven't forgotten your review, it's on my desk," while at the same time patting his crotch, Antoni was conditioning a favorable performance review, and thus a raise, on Jansen's submission to his sexual desires. Because this alleged exchange could constitute a meaningful threat, and there is no evidence as yet that Jansen did not view the threat as real, a reasonable jury could conclude that Jansen was subjected to compensable quid pro quo harassment.

Jansen has also shown enough to get before a jury on her hostile work environment claim. Antoni allegedly subjected Jansen to months of sexually suggestive remarks and inappropriate acts and gestures, that could be considered "severe and pervasive." There is evidence in the record that PCA should have been aware of Antoni's campaign of abuse. While Jansen did not complain through official channels initially, she did complain to fellow employees. PCA was on notice that Antoni had allegedly harassed his two previous secretaries. Further, there is evidence that Jansen, like the two previous secretaries, was unaware of any sexual harassment policy at PCA. A reasonable jury, deliberating under the heightened standard of care applicable in the supervisory context, could find PCA liable.

Ellerth, like Jansen, has put forward sufficient evidence to survive summary judgment on her claim of quid pro quo harassment. Her pleadings submitted to the district court, specifically her motion opposing summary judgment together with its accompanying affidavits, while unartful, plainly articulate a basis for a quid pro quo claim. Cf. Saxton v. American Tel. & Tel. Co., 10 F.3d 526, 533 (7th Cir.1993). This basis is comprised of the following alleged statements made by Slowick to Ellerth. While in the lounge of a hotel where Ellerth had come at Slowick's request during a business trip, after an evening of allegedly ogling Ellerth, Slowick said, "you know, Kim, I could make your life very hard or very easy at Burlington." In a later phone conversation, when Ellerth asked permission to undertake a special project, he said, "I don't have time for you right now, Kim, unless you tell me what you're wearing." In a subsequent telephone conversation, Slowick denied her request and then asked her if she were wearing shorter skirts, as that would make her job "a whole heck of a lot easier." Slowick, at an interview regarding Ellerth's promotion during which he rubbed Ellerth's knee, said he had hesitations about promoting her because she was not "loose enough for him."13 From these essentially uncontested statements, a reasonable jury could conclude that Slowick was conditioning or threatening to base "the terms and conditions" of Ellerth's employment on Ellerth's catering to Slowick's sexual desires, and thus subjected to her compensable quid pro quo harassment.

Ellerth, while perhaps having shown enough to get before a jury on her hostile work environment claim, has waived the right to do so. Before the original panel, Ellerth argued that liability should be imposed for hostile work environment because Slowick "was aided in accomplishing the tort by the existence of an agency relationship." See Restatement § 219(2)(d). This argument alone would not prevent recovery under an alternate theory. See 5 Charles A. Wright & Arthur R. Miller, Federal Practice and Procedure § 1219 (1990) (theory of the pleading doctrine). Perhaps overly ambitious, however, Ellerth goes farther in her reply brief for the argument en banc and disavows any argument that liability should be imposed because of employer negligence. See Restatement § 219(2)(b). As I believe liability remains linked to negligence, I join Chief Judge Posner in his assessment that Ellerth has no remedy here. See Posner, at p. 516.

CUDAHY, Circuit Judge, concurring.

I participated in this en banc rehearing as a member of the panel that originally heard Jansen (with Judges Coffey and Flaum). Hence, my participation is limited to that case.

I join Judge Flaum's concurring opinion, and I write primarily to elaborate on what constitutes an appropriately "heightened" duty of care for employers whose supervisors engage in hostile environment harassment. For, as Judge Flaum indicates, there are good reasons for a heightened standard for supervisor misconduct in contrast to the same misconduct by fellow employees.

It may be that Judge Wood's analysis converges with Judge Flaum's on the bottom line. Judge Wood mentions circumstances in which supervisors may exceed the scope of their employment and therefore act without delegated authority. See Wood, pp. 574-575. In his negligence analysis Judge Flaum calls for a heightened standard of employer care. See Flaum, p. 502; see also Posner, pp. 511-512. I doubt that the result in Jansen, for example, would differ under one or the other of these approaches. Nevertheless, it is necessary to distinguish quid pro quo from hostile environment claims. The Supreme Court distinguished them in Meritor with an implication, at least, that they might better be treated differently. Meritor Sav. Bank, FSB v. Vinson, 477 U.S. 57, 72, 106 S. Ct. 2399, 2408, 91 L. Ed. 2d 49 (1986). Justice Marshall, concurring in the judgment in Meritor, read the majority opinion this way and objected to "a special rule, to be applied only in 'hostile environment' cases." Id. at 77, 106 S.Ct. at 2410 (Marshall, J., concurring in the result). But the opposite view seemingly prevailed, and we must take our guidance from that fact. The other circuits have taken this lead. See, e.g., Faragher v. City of Boca Raton, 111 F.3d 1530 (11th Cir.1997) (en banc); Nichols v. Frank, 42 F.3d 503 (9th Cir.1994); Bouton v. BMW of N. Am., Inc., 29 F.3d 103 (3d Cir.1994); Kauffman v. Allied Signal, Inc., Autolite Div., 970 F.2d 178 (6th Cir.1992); but cf., Karibian v. Columbia Univ., 14 F.3d 773 (2d Cir.1994).

That the Supreme Court distinguished the two kinds of harassment claims makes sense. In a quid pro quo situation, a supervisor acts with actual or apparent authority when he promises employment goodies or threatens their withdrawal to extract sexual "cooperation." Quid pro quo is always a creature of power. It is the classic paradigm of powerful males forcing their wills on vulnerable females. A hostile environment, on the other hand, can be created by persons of equal, or even inferior, status to the victim. Supervisory rank may facilitate it, and power may enhance it, but neither is essential, or even central, to the creation of a hostile environment.

In Jansen, the supervisor Antoni allegedly created a hostile environment. Power is not of the essence there, as it certainly is for quid pro quo harassment. One can imagine a lesser figure doing almost the same thing as Antoni. But his rank in the company was of considerable help to him in creating a milieu of abuse. It is not easy to articulate exactly how a supervisor's abusive conversation and posturing implicates his authority. Still, the intuition remains that the delegated power certainly helps--just as it helped Antoni. Just as it is the centrality of power and authority that dictates strict liability for quid pro quo harassment, it is the shadow of power and authority that dictates a heightened negligence standard for supervisory hostile environment harassment.

The negligence standard used to govern hostile or abusive environment claims involving a supervisor must be negligence as related to a special and demanding duty of care. The standard of care should be somewhat like that imposed on packers of parachutes or open heart surgeons. See W. PAGE KEETON ET AL., PROSSER & KEETON ON THE LAW OF TORTS § 38, at 208-9 (5th ed.1984); see also, e.g., Rustay v. Consol. Rail Corp., 775 F. Supp. 161, 163-65 (D.N.J.1991) (operating railroad requires greater duty of care). Judge Flaum has pointed out a weak link in a negligence approach: the assumption that a system can be built under which victims of supervisory sexual harassment will feel free to report the problem to the company. See Flaum, p. 502. There ought to be a presumption that in the reporting system under examination, victims could not reasonably be expected to report their problems (unless, of course, the victim actually did). The employer would have an opportunity to overcome this presumption. But, if no evidence were presented, the presumption would import liability.

Also, under the negligence standard, one of the principal liability issues is whether the employer had actual or constructive notice of the creation of a hostile environment by a supervisor. In Faragher v. City of Boca Raton, 864 F. Supp. 1552, 1563 (S.D.Fl.1994), the district court held that, if a situation in the company were sufficiently severe to amount to a hostile environment, it was by the same token sufficiently visible to put the employer on constructive notice. In general, this approach has merit. See Meritor, 477 U.S. at 72, 106 S. Ct. at 2408 (citing with approval Taylor v. Jones, 653 F.2d 1193, 1197-99 (8th Cir.1981) (holding employer liable for racially hostile working environment based on constructive knowledge)). I would create a rebuttable presumption that, if a situation attributable to a supervisor is severe enough to create a hostile environment, the employer is on constructive notice that remedial measures are required. The employer, of course, could offer evidence to the contrary.

These two presumptions give concrete meaning to Judge Flaum's call for a heightened duty of care. As I noted at the outset, this view might lead to verdicts not too different from Judge Wood's approach. These presumptions, however, would respect and reward prudent employers who (1) provide a genuinely protected channel for reporting harassment and (2) remain in close touch with the possibility of harassment by supervisors in their own organizations. The diligent employer will have an opportunity to adduce evidence to this effect and thereby to escape liability.

KANNE, Circuit Judge, concurring.

I join Judge Flaum's thoughtful opinion--except for reservations in two areas. First, with regard to quid pro quo harassment, I do not think an employer should be held strictly liable for mere threats made by a supervisory employee. I agree that our focus should be on deterring sexual harassment, but I believe a negligence standard will adequately do the job. A negligence standard will serve to deter quid pro quo threats because employers will rationally seek to avoid liability 1) by establishing policies and procedures to circumscribe improper conduct by supervisors, and 2) by imposing appropriate sanctions against errant supervisors once notified by an employee that a supervisor has made a quid pro quo threat. Beyond that, there is not much an employer realistically can do to put a stop to quid pro quo threats that do not result in any adverse job consequence. Imposing strict liability for such threats would therefore yield little (if any) greater deterrence and generate significantly greater litigation costs than a negligence standard. Moreover, quid pro quo threats may be ambiguous, and if such threats make an employer strictly liable, plaintiffs will attempt to turn all instances of supervisor sexual harassment into "implied threats" in order to take advantage of strict liability's easier burden of proof.

On the other hand, I cannot agree with Chief Judge Posner's position that in her threat-only situation (i.e., without an adverse job consequence) Jansen does not have a viable quid pro quo claim. Although strict liability is inappropriate for quid pro quo threats, Jansen should still be able to win on her quid pro quo theory if she can prove her employer's negligence in failing to obviate quid pro quo threats. Quid pro quo claims involving only threats deserve scrutiny for employer negligence because of the qualitatively more oppressive nature of quid pro quo harassment. Perhaps the facts here will demonstrate that the employer was not negligent, but Jansen's quid pro quo claim should be remanded along with her hostile environment claim so the district court can make the negligence determinations. Admittedly, in this and many cases involving mere threats, adding a quid pro quo claim to a hostile environment claim may not make much practical difference because the supervisor's threat will already be relevant to the hostile environment negligence analysis. In some cases, however, the quid pro quo claim will stand alone, and negligent employers should be liable for such quid pro quo threats.

My second reservation regarding Judge Flaum's opinion concerns his treatment of hostile environment harassment. I cannot agree with Judge Flaum's language suggesting that employers are under a "heightened duty of care" for instances of supervisor harassment. A negligence standard always entails a fact-specific inquiry regarding what care the parties reasonably should have taken. The training, authority, and control given supervisors by an employer will therefore always enter the negligence equation. Calling this a "heightened duty of care" may suggest to some that employers should be presumed liable for supervisor hostile environment harassment. I cannot endorse any such presumptions that take this type of case outside the normal rules of negligence and further complicate this increasingly difficult issue.

POSNER, Chief Judge, joined by MANION, Circuit Judge, concurring and dissenting.

These cases present the court with difficult and important questions concerning the application of Title VII of the Civil Rights Act of 1964 to sexual harassment by supervisory employees. I agree with the basic approach to these questions, and most of the specific answers to them, in Judge Flaum's concurring opinion. Our most substantial difference concerns the employer's liability, in "quid pro quo" sexual harassment cases, for what I shall call "noncompany acts"; these are illustrated by a supervisor's threat that is not followed up by termination, demotion, or other acts that change the contractual relation between the victimized employee and the employer. I shall explain this difference between us after setting forth a general approach that differs somewhat from his.

When the Supreme Court in Meritor Savings Bank, FSB v. Vinson, 477 U.S. 57, 72, 106 S. Ct. 2399, 2408, 91 L. Ed. 2d 49 (1986), said that "Congress wanted courts to look to agency principles" to determine the liability of employers sued under Title VII for sexual harassment of one employee by another, the Court was not ruling that Title VII incorporates the American Law Institute's Restatement of Agency. Obvious as this point should be to any reader of the Supreme Court's opinion--for the opinion states that "common-law [agency] principles may not be transferable in all their particulars to Title VII," 477 U.S. at 72, 106 S. Ct. at 2408--it hasn't sunk in all the way. One of the briefs in the cases before us actually states, "Since Congress placed a comma after the first clause of § 219(2)(d) [of the Restatement]...." The courts do not make this mistake; but the Restatement has become their template for analyzing the issue of employers' liability for harassment by supervisors. This is the wrong approach, and I am pleased that Judge Flaum does not follow it.

In saying that it is wrong to rely on the Restatement, I do not mean that it is wrong to bring agency principles into the application of Title VII. That is inescapable. Title VII creates a statutory tort of employment discrimination but imposes liability only on the "employer"; what is to count as the employer's discriminatory act, and so warrant a remedy under Title VII, is a question of agency that the federal courts must by default answer as a matter of federal common law. Often courts create a federal common law principle simply by borrowing a part of the law of the state in which the legal dispute arose. That would not be a sound approach here, and a majority of the judges of this court reject it; nor is it the tenor of the Supreme Court's reference in Meritor to the judicial role: "Congress wanted courts to look to agency principles." Given that the employer is the only person made liable by the statute, the question of agency is central to its administration, so that to defer to state law could introduce striking geographical disuniformities and indeed would empower a state to nullify Title VII.

It is true that the Supreme Court recently repudiated a comprehensive federal judicial power to create common law, but it was speaking of federal common law in "its strictest sense," Atherton v. FDIC, --- U.S. ----, ----, 117 S. Ct. 666, 670, 136 L. Ed. 2d 656 (1997), that is, of legal doctrine that could not be construed as interpretive. In Atherton it was a standard of care for directors of all financial institutions and had no more warrant than the fact that the institution was federally chartered or insured. Deciding what agency principles shall govern liability under a liability-creating statute such as Title VII is not free-wheeling common-law rulemaking; it is filling a statutory gap, a standard office of interpretation. There is no novelty in formulating federal principles of agency law in interpreting federal statutes that are silent on agency. See, e.g., Central States Trucking Co. v. J.R. Simplot Co., 965 F.2d 431 (7th Cir.1992); United States v. Balistrieri, 981 F.2d 916, 930 (7th Cir.1992). I have not found a case which suggests that liability under Title VII should depend on the agency law of the state where the dispute arose. The parties to the cases before us have not suggested that we take such an approach, and Judge Manion's opinion marshals compelling authority against it.

Both cases before us arose in Illinois, but the parties have not enlightened us about the Illinois law of agency. In fact Illinois courts almost always refuse to hold employers liable for the sexual misconduct of their employees, including their supervisory employees. See Deloney v. Board of Education, 281 Ill.App.3d 775, 217 Ill. Dec. 123, 128-29, 666 N.E.2d 792, 797-98 (1996) (truant officer who molested 16-year-old student); Randi F. v. High Ridge YMCA, 170 Ill.App.3d 962, 120 Ill. Dec. 784, 788, 524 N.E.2d 966, 970 (1988) (sexual assault by employee of day-care center on 3-year-old); Webb v. Jewel Cos., Inc., 137 Ill.App.3d 1004, 92 Ill. Dec. 598, 485 N.E.2d 409 (1985) (sexual assault by security officer on suspected shoplifter); Hoover v. University of Chicago Hospitals, 51 Ill.App.3d 263, 9 Ill. Dec. 414, 366 N.E.2d 925 (1977) (sexual assault by doctor on patient); Dockter v. Rudolf Wolff Futures, Inc., 684 F. Supp. 532, 535-36 (N.D.Ill.1988) (sexual harassment by manager) (interpreting Illinois law), aff'd, 913 F.2d 456 (7th Cir.1990). St. Paul Fire & Marine Ins. Co. v. Downs, 247 Ill.App.3d 382, 187 Ill. Dec. 130, 617 N.E.2d 338 (1993), found sexual molestation of a patient by a psychiatrist to be arguably within the scope of the psychiatrist's employment, but based this on the patient's psychological dependence on the psychiatrist, distinguishing the other cases that I have cited. Id. at 136, 617 N.E.2d at 344. Downs belongs to the class of cases, illustrated also by sexual molestation by police officers, as in our recent West v. Waymire, 114 F.3d 646, 649 (7th Cir.1997), in which the employer has entrusted the employee with unique power to molest or otherwise oppress subordinates or members of the public; the distinction between these cases and the ordinary case of sexual harassment by a supervisory employee is stressed in Deloney v. Board of Education, supra, 281 Ill.App.3d 775, 217 Ill. Dec. at 130 n. 5, 666 N.E.2d at 799 n. 5. In Deal v. Byford, 127 Ill. 2d 192, 130 Ill. Dec. 200, 537 N.E.2d 267 (1989), scope of employment was admitted, hence not an issue, and anyway the employee was clearly acting in furtherance of the employer's interests. See id. at 203, 206-07, 537 N.E.2d at 270, 272-73. And Pyne v. Witmer, 129 Ill. 2d 351, 135 Ill. Dec. 557, 543 N.E.2d 1304 (1989), did not involve an intentional tort.

Green Hills Country Club v. Illinois Human Rights Comm'n, 162 Ill.App.3d 216, 113 Ill. Dec. 216, 219, 514 N.E.2d 1227, 1230 (1987), does hold that an employer is strictly liable for any type of sexual harassment by a supervisory employee, but it bases this holding on an Illinois statute dealing with sexual harassment, 775 ILCS 5/2-102(D), and does not mention the common law of agency. Judge Easterbrook (and Judge Wood by implication) fastens on this statute as the ground for believing that the plaintiffs in our two cases have a good claim, noting the Supreme Court's reference in Meritor to the Restatement of Agency--though the Restatement is limited to common law, Restatement (Second) of Agency 3 (1958), so that Meritor's reference to it is not plausibly interpreted as authority for deferring to a state statute. Nor are Restatements confined to state common law, as Judge Easterbrook believes. The Meritor suit had not been filed in any state, but in the District of Columbia. The Supreme Court neither mentioned this fact nor suggested that the place to look for the agency principles applicable to the case was the law of the District of Columbia. My colleague is right that the "common law of agency has stood the test of time." But the observation is doubly irrelevant: he forswears reliance on common law principles of agency in favor of a statute that departs from those principles; and they were not, as I shall point out, developed with reference to the novel and atypical tort of sexual harassment.

His argument implies that by a stroke of the pen Illinois could eliminate all Title VII liability in that state that depended on the principles of agency law. This would mean all Title VII liability in which the employer was not a personal participant in the infringement of the employee's rights. A state statute that provided that a decision to fire or take other adverse action against an employee on the basis of her sex would not be imputed to the employer unless the employer's owner or chief executive officer had approved the action would do the trick. My colleague resists this implication, arguing that an attempt by a state to curtail Title VII liability would be preempted by the federal statute. But if as he believes Congress left the formulation of agency principles to the states, there is nothing in Title VII that could conflict with a state's decision to adopt the narrowest of such principles to govern sexual harassment or other forms of discrimination. He says we needn't fear that states might discriminate against Title VII plaintiffs by formulating narrow agency principles because the principles would apply to claims under state law as well as under federal law. But the state statute on which he relies for the agency principles applicable to Title VII sexual harassment suits filed in Illinois is not a general agency or tort statute; it is limited to discrimination. The logic of his position is that a state hostile to discrimination claims could draft a special, and especially narrow, set of agency principles to govern discrimination cases alone.

Just as I do not think that Title VII's silence with respect to the applicable agency principles points us toward the incorporation of state law, so I do not think that its silence can tell us anything about the structure of those principles. It thus cannot tell us whether, as I believe no one thinks (and the Supreme Court seems in Meritor to have nixed, see 477 U.S. at 72, 106 S. Ct. at 2408), the employer should be strictly liable for all acts of sexual harassment on or off the premises (in Ellerth's case, largely off) by all employees, or whether distinctions should be made on the basis of kinds of employee (such as supervisory or nonsupervisory) or kinds of act (such as demotion versus a sexual solicitation versus forcible rape), or some combination of the two.

The Second Restatement of Agency (there is no third) was promulgated 40 years ago, before Title VII was enacted and before the concept of sexual harassment had emerged as a distinct legal concept. There is nary a hint in the text or legislative history of Title VII that Congress intended to incorporate the Restatement by reference. It would have been loopy to do so. The Restatements are intended to provide a compact statement of common law principles. The essence of the common law method of rulemaking is suppleness and flexibility, facilitating adjustment to altered circumstances, and is inconsistent with treating any statement of common law principles as a petrified text.

The provisions of the Second Restatement of Agency are designed mainly for two types of case neither of which is before us. The first is the tort committed against a stranger, as where a truck driver employed by the defendant runs down a pedestrian. The second is the contract between a stranger and an agent of the defendant. The Restatement does not completely ignore torts committed by one employee against another, see § 473, but it subjects them to the "fellow servant" rule. § 474. This nineteenth-century doctrine excuses the employer from liability for negligent injury of one employee by another. Farwell v. Boston & Worcester R.R., 45 Mass. (4 Met.) 49 (1842). If the absorption of the Second Restatement of Agency into Title VII were taken seriously, courts would have to explore the implications of the fellow servant rule for the law of sexual harassment. This they rarely do (but see Guess v. Bethlehem Steel Corp., 913 F.2d 463, 465 (7th Cir.1990)), though, as it happens, the position they have arrived at with respect to harassment by coworkers is consistent with the rule. As I shall note in a moment, an employer is not automatically liable for coworker harassment; there is employer liability only if the employer is negligent.

The fellow servant rule might be thought inapplicable to sexual harassment because the rule is invariably stated in terms of the employee's negligent rather than intentional torts, and sexual harassment is intentional on the part of the harasser, that is, of the fellow employee. But the only reason the rule seems to be just about torts of negligence is that the liability of an employer for his employees' intentional torts is so much more limited than his liability for their unintentional (usually, negligent) torts that the employer would rarely have occasion to invoke the rule other than in the negligence setting. The fellow servant rule was applied to an intentional tort in Gabrielson v. Waydell, 135 N.Y. 1, 31 N.E. 969 (1892), and while it is an old case, most fellow servant cases are old, the fellow servant rule having largely gone down the drain when employers' liability acts and workmen's compensation acts superseded the common law of employer liability for industrial accidents. See Pomer v. Schoolman, 875 F.2d 1262 (7th Cir.1989), for a rare modern case.

If the fellow servant rule were sought to be applied to harassment by a supervisory employee, we would have to decide whether the harasser was a "vice-principal," W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 80, p. 572 (5th ed.1984), exercising a nondelegable duty of care, or merely a "superior ... servant." Restatement, supra, § 479. The employer would be liable for the vice-principal's harassment but not the superior servant's. The issue of classification would be made difficult to the point of unreality by the fact that the Restatement nowhere addresses sexual harassment of one employee by another, or indeed sexual harassment of any sort. It is impossible to tell from its discussion of nondelegable duties whether the supervisors in our two cases would have been thought vice-principals.

Sexual harassment differs from most other torts, and especially from the most common torts--the deliberate or accidental infliction of physical injury--in being "invisible" or largely so other than to victim and perpetrator. The employer will know whether one of his employees sustains a physical injury on the job; he is much less likely to know that one of his employees is being sexually harassed on the job. The feasibility of an employer's controlling the torts of his employees is, as I shall show, an important consideration in crafting agency principles, and one dependent on the particular tort.

I may be in a minority of judges in not believing (or at least in being willing publicly to deny) that the most fruitful way of approaching a novel issue of law is through seeking analogies in legal doctrines designed with quite other issues in mind. Nevertheless, had the Supreme Court told us to use the Second Restatement of Agency as the framework for evaluating sexual harassment under Title VII, I would bow to its command. It did not; but by citing the Restatement it gave lawyers and judges a straw to grasp at. The straw has broken in their hand. The Restatement turns out to be hopelessly vague in its bearing on the issue of employers' liability for sexual harassment, being vaguely worded and addressed to other issues. So judges can in good faith reach opposite results when they seek guidance in the Restatement to employers' liability for sexual harassment by supervisory employees. See, e.g., Faragher v. City of Boca Raton, 111 F.3d 1530 (11th Cir.1997) (en banc). The judges and lawyers who insist that the Restatement of Agency is The Way either are disingenuous, wishing to conceal their true grounds of decision, or are in the grip of the formalist belief that difficult cases can be decided by teasing out the meaning of words in a text composed with other problems in mind, without need to examine the social policies that the law might be thought to be serving. "We condemn Lochner as formalistic not because it involves a choice, but because it attempts to describe this choice as compulsion." Frederick Schauer, "Formalism," 97 Yale L.J. 509, 511-12 (1988). Since neither the text nor the legislative history of Title VII indicates what agency principles the authors of the statute had in mind, the formalist gropes for another text, finds the Restatement, and treats it inappropriately as a surrogate statute.

It is time that we threw away the crutch of the Restatement and, recognizing the differences between workplace sexual harassment and the actual subject matter of that antiquated screed, ask as an original matter--as I think the Supreme Court invited us to do in Meritor when it (under)stated that the common law of agency might not be fully transferable to sexual harassment--what the best regime of liability would be for these cases. It would be misleading even to speak of applying to this case "the law of agency." That would imply the existence of some ready-at-hand body of rules that we have only to apply to a new set of circumstances. The problem is not application; it is creation. I think it equally question-begging to ask whether a supervisor's conduct is "attributable" to the employer, or whether he may be said to "speak for" the employer, or whether an employer must be deemed to "delegate his common law duties" to his supervisory employees. These are just additional ways of asking whether the employer should be liable.

I understand the attraction of seeking a textual answer elsewhere, perhaps in local law, as Judge Easterbrook suggests; but this evasion of the occasional duty of judicial creativity will not work either, as I have explained and Judge Manion more fully. We thus cannot avoid the task of trying to create a set of agency principles that will deter sexual harassment without imposing an unreasonable burden on employers. I do not consider the burden of liability on employers a negligible consideration in formulating federal common law rules to govern aspects of employment regulation. Our labor markets are becoming choked by regulation, all well meaning but cumulatively an impediment to the efficient employment of the nation's most valuable economic resource, which is its workers.

I emphasize deterring sexual harassment rather than compensating its victims because, unlike many torts--and again the clearest illustrations are torts that inflict physical injury--sexual harassment does not usually bring about a significant change in the victim's wealth. The victim may be humiliated and deeply distressed by it; rarely will she (or the very occasional he) be impoverished by it. This is not to say that psychological pain should not figure in the calculation of damages. It should. All I wish to emphasize is that an award of damages in these cases is not primarily designed to protect the victim's standard of living. The payment of damages in the usual case of sexual harassment is an instrument for deterring future incidents of such harassment rather than for restoring lost earnings or for financing expensive curative or rehabilitative measures.

There is no novelty in grounding principles of agency on deterrence. A recognized objective of imposing "vicarious" liability for the torts of employees (the liability that goes by the name of "respondeat superior") is to reduce the amount of tortious behavior. Hartmann v. Prudential Ins. Co., 9 F.3d 1207, 1210 (7th Cir.1993); Konradi v. United States, 919 F.2d 1207, 1210 (7th Cir.1990); Alan O. Sykes, "The Boundaries of Vicarious Liability," 101 Harv. L. Rev. 563, 570 (1988). Our court has adopted (and it is not alone in doing so) the deterrent perspective as its guide to dealing with cases involving the creation of a hostile environment by the plaintiff's coworkers. The standard of employer liability in those cases is negligence: an employer who knew or should have known about the harassment and failed to take reasonable remedial measures is liable. McKenzie v. Illinois Dep't of Transportation, 92 F.3d 473, 480 (7th Cir.1996); Baskerville v. Culligan International Co., 50 F.3d 428, 432 (7th Cir.1995); Carr v. Allison Gas Turbine Division, 32 F.3d 1007, 1009 (7th Cir.1994); Doe v. Lago Vista Independent School District, 106 F.3d 1223 (5th Cir.1997); Rosa H. v. San Elizario Independent School District, 106 F.3d 648, 656 (5th Cir.1997); Splunge v. Shoney's, Inc., 97 F.3d 488, 490 (11th Cir.1996); Burns v. McGregor Electronic Industries, Inc., 989 F.2d 959, 966 (8th Cir.1993). The thinking is that if the employer has a realistic opportunity to prevent harassment, he must take it.

But if the goal is deterrence, why, it may be asked, is the standard negligence rather than strict liability? The reason is the infeasibility of an employer's stamping out this sort of harassment without going to extreme expense and greatly curtailing the privacy of its employees, as by putting them under continuous video surveillance. If the victim of sexual harassment complains to a supervisor, or a worker who notices what's going on complains to a supervisor, or the harassment is so pervasive (considering its nature, frequency, and number of victims and perpetrators) that the employer knows or should know about the harassment, the employer ought to take steps to correct the problem. He has or should have the information; he has only to act upon it. And since everyone knows by now that sexual harassment is a common problem in the American workplace, the employer ought in addition to take, in advance of specific cases of harassment, preventive measures against it, as by adopting and announcing a policy against sexual harassment and creating a discreet and convenient machinery by means of which victims can obtain relief without exposing themselves to retaliation. These are the responsibilities that a negligence standard imposes. They should deter most sexual harassment by coworkers without the courts' subjecting the employer to strict liability. A law that requires the employer to do more than is feasible to control harassment will impose costs without creating deterrent benefits. In the long run, these costs will be borne largely by consumers, in the form of higher prices for the employer's product, and workers, in the form of lower wages (because the higher costs are labor costs). Many consumers and workers are women, so women, who are the principal victims of sexual harassment, will pay a big part of the costs that employers incur as a consequence of excessively harsh principles of employers' liability.

In these circumstances, strict liability would not only be expensive and unnecessary, and possibly regressive as well; it would be futile. If the law imposes liability in harassment cases in which there is no reasonable measure that the employer could have taken to prevent the harassment, the only effect of the law will be to impose extra costs on employers and those with whom they are linked contractually. Employers will prefer paying the occasional judgment to incurring costs that, by definition, exceed the employer's foreseeable liability--by definition because, were the costs less than the expected liability, the failure to incur them would be negligence; it is only when they are greater, so that the employer would not be negligent for failing to incur them, that strict liability bites. The bite may come out of the hides of innocent workers and consumers, rather than deterring any more harassment than a negligence standard would do.

What this shows is that strict liability doesn't always have the happy result of just deterring a bad practice more effectively than the negligence standard. Sometimes the incremental contribution of strict liability to deterrence is nil or slight, and easily outweighed by the additional costs of a more expansive liability. This is implicit in the rejection of strict liability for coworker harassment cases, a rejection to which I believe all members of this court subscribe.

But the cases in which the issue is the employer's liability for harassment by coworkers are the easy ones. The difficult ones are those in which, as in the cases before us, the harasser is a supervisor rather than a line employee. We must distinguish between two types of sexual harassment by supervisors. In the first type, the supervisor uses or attempts to use his supervisory authority to obtain sexual favors from an employee. This is the domain of what has come to be called "quid pro quo" harassment. In the second type of case, the supervisor does not use or attempt to use his supervisory authority at all. He harasses an employee, but he does so in exactly the same way that an employee who had no supervisory authority would harass another employee. He uses unwanted terms of endearment; he fondles or rubs up against his victim; he displays sex toys or tells dirty jokes; he brags about his sexual skills; he proposes marriage; he threatens to kill himself; in the extreme case, illustrated by Martin v. Cavalier Hotel Corp., 48 F.3d 1343 (4th Cir.1995), he rapes her. These are modes of harassment equally available to a nonsupervisory employee. The proper standard of employer liability here is negligence, just as in the case of harassment by nonsupervisory employees, because it is as costly for the employer to police this kind of harassment by a supervisor as it is to police the identical harassment by a coworker. I believe that we all have qualms about Martin, the case that carries strict liability for the acts of a supervisory employee to its logical extreme.

Although the standard should be the same in the two types of hostile-environment case, the application of the standard is bound to differ as a practical matter. Even if the supervisor is scrupulous in forbearing to invoke his authority over the employee he is harassing, the existence of that authority is a fact of which the victim is bound to be constantly aware. She is more likely than in a case of coworker harassment to submit and less likely to complain. She may believe that because her harasser is a supervisor he is highly valued by the company and that as a consequence the higher-ups in the company will reject her complaint--will in fact fire her, not the harasser. Because such fears will often be reasonable, due care on the part of employers to prevent sexual harassment by its supervisory employees may require the creation of additional complaint machinery when the complaint is against a supervisor. Whether the employer had complied with this heightened duty of due care would be a jury issue in the first instance.

But I would not go further and impose strict liability on the employer for hostile-environment harassment by a supervisor. The "theory" that by making him a supervisor the employer had clothed him with the appearance of authority to harass employees is empty. Since everyone knows by now that sexual harassment is unlawful, a company that publicizes and enforces a stern policy against sexual harassment has done all that it can reasonably be expected to do to dispel its supervisors' "apparent authority" (to lapse into Restatement-speak) to harass. The adequacy of the policy in the circumstances (what if the harasser is the chief executive officer of the employer?) is a factor to be considered in making the required determination of negligence. There is no need to have a separate standard just because the harasser is a supervisor. Excessive complexity is the bane of American law; we have an opportunity to make it a little simpler.

So much for hostile-environment harassment. Let me turn now to quid pro quo harassment. We have to distinguish between two types of such harassment. In the first, the supervisor brings about a significant alteration in the terms or conditions of his victim's employment. He fires her, or denies her a promotion, or blocks a scheduled raise, or demotes her, or transfers her to a less desirable job location, or refuses to give her the training that the company's rules entitle her to receive. In all these examples the supervisor is using his delegated authority to do a company act. Strict liability is appropriate in this kind of case, because it is likely to deter this kind of sexual harassment more effectively than negligence liability would. The employer who is strictly liable will monitor the exercise of this delegated authority very carefully, knowing that it will be liable if the authority is abused. And this monitoring should be relatively easy to do. For it is usually a mistake for a firm, quite without regard to any potential legal liability, to give a supervisor unilateral authority to alter a subordinate's terms or conditions of employment significantly. In well-managed companies, decisions having such consequences are subject to rules, and to review by higher-ups in the company--the industrial equivalent of appellate review. The rules will be more carefully formulated and the supervisor's compliance with them in firing or otherwise hurting a subordinate more carefully reviewed by the supervisor's superiors if the employer is strictly liable for the supervisor's use of his delegated powers to harass subordinates. Courts applying a negligence standard would have great difficulty determining how closely the supervisor should be supervised. Such questions as how many tiers of review should be provided before an employee can be fired or demoted are not easily answered in terms of reasonableness or due care, the criteria of negligence. The regime of strict liability shifts the responsibility for deciding these questions to the employer, who knows more than judges do about how to control supervisory employees.

Could the argument that I have just sketched be made in support of strict liability for all sexual harassment by company employees, including harassment by coworkers? Employers know more than we do about how to extirpate that form of workplace sexual harassment too. True; but courts know, more or less, what is reasonable for the employer to do about hostile-environment harassment--institute a tough policy, disseminate it, establish a procedure by which a worker can complain without fear of retaliation (so don't require that the complaint be made to the man who is harassing her!), respond promptly and effectively to any report of possible harassment. Knowing what the employer should do, the courts have only to decide whether he did it, in order to decide whether he was negligent and should be liable. But when it comes to designing the optimum system for reining in the discretion of supervisory employees, the courts are at sea and it makes sense therefore to shift the responsibility entirely to the employer to create and administer an effective system for the review and control of company actions taken by supervisors in the exercise of their delegated authority.

Strict liability is inappropriate, however, when the supervisor merely makes threats, even if the threats are effective. This is why it is important to distinguish between the type of quid pro quo harassment in which the supervisor actually alters the terms or conditions of his victim's employment and the type of harassment in which he merely threatens to do so, whether or not the victim yields to the threats. Suppose the supervisor threatens to fire a subordinate unless she'll have sex with him, and she agrees--or refuses and he does not carry out his threat. In either case, because he has not used his delegated authority to commit a company act, there is no way in which a system for vetting such acts would catch him out. In a well-regulated company a supervisor who wants to fire a subordinate has to obtain the approval of higher-ups, as I have said, and they will have an opportunity therefore to determine the bona fides of his proposal. But if he doesn't propose to fire her, whether because she has submitted to his sexual extortion or called his bluff, there will be no proposed action to review. It will be no more feasible for the company to determine what is going on than it would be if the harasser were a coworker who had threatened to steal the victim's work tools if she didn't submit to him.

Romantic encounters, including romantic encounters between supervisors and supervised, are a fact of the workplace. Title VII does not purport to forbid them, and would be quixotic if it did. Many happy marriages have grown out of such encounters. Some of the encounters are abusive from the start, and some start well and turn ugly and engender charges of sexual harassment that sometimes have and sometimes lack merit. The words, the gestures, the other behaviors that differentiate the fully consensual relationship from the coercive relationship will often be invisible to the supervisor's superiors. The yielding to a threat will look no different from the yielding to a lawful proposal. It is only when the threat is carried out that the abusive supervisor does something, such as firing the supervised employee, that the employer will know about and should monitor. It is facile to suggest that employers are quite capable of monitoring a supervisor's actions affecting the work environment. Large companies have thousands of supervisory employees. Are they all to be put under video surveillance? Subjected to periodic lie-detector tests? Trailed on business trips by company spies? Even the EEOC does not go so far. The culminating recommendation in its "Guidelines on Discrimination Because of Sex" to employers is that they should "develop[ ] methods to sensitize all concerned" to the problem of sexual harassment. 29 C.F.R. § 1604.11(f). This is easy to say, but, as the armed forces have recently discovered to their sorrow, hard to do. It is unrealistic to expect that adopting strict liability in lieu of negligence, a relatively esoteric and marginal change in the law, will do much to create a corporate culture in which sexual harassment is unthinkable. Granted, the problem of monitoring the noncompany acts of supervisory employees is primarily a problem of large employers, for in companies that are at or near the statutory minimum of 15 employees, 42 U.S.C. § 2000e(b), noncompany acts will presumably be visible to top management, creating de facto strict liability. But it is the large companies that are the usual targets of litigation, because of their deep pockets.

The distinction that I am proposing between the harassing supervisor who commits a company act, such as firing, and the supervisor who merely threatens such acts, has at least a footing in the case law. See Gary v. Long, 59 F.3d 1391, 1396 (D.C.Cir.1995); Sauers v. Salt Lake County, 1 F.3d 1122, 1127 (10th Cir.1993); Kotcher v. Rosa & Sullivan Appliance Center, Inc., 957 F.2d 59, 62 (2d Cir.1992); Hicks v. Gates Rubber Co., 833 F.2d 1406, 1414 (10th Cir.1987). There are hints of it in Judge Easterbrook's and Judge Kanne's separate opinions. It is, for sure, not the orthodox approach, see, e.g., Karibian v. Columbia University, 14 F.3d 773, 778 (2d Cir.1994); Highlander v. K.F.C. Nat'l Management Co., 805 F.2d 644, 648 (6th Cir.1986); I would go so far as to concede that it has the curse of novelty; but the cases that take the orthodox approach do not explain why they do so, and the approach that I am proposing in its stead is not foreclosed by any decisions of the Supreme Court. The Court in Meritor said that Congress intended "to place some limits on the acts of employees for which employers under Title VII are to be held responsible." 477 U.S. at 72, 106 S.Ct at 2408. It didn't tell us what those limits are.

We can test my proposed approach by considering its application to other forms of discrimination forbidden by Title VII. The closest parallel is sex discrimination that does not involve harassment in the usual sense. Compare two cases. In one, a supervisor denies a raise to a subordinate because of her sex. In another, the supervisor slaps her because of her sex. In the first case, the supervisor has done a company act, and the company is strictly liable. In the second case, the supervisor has not done a company act, and there is no liability under Title VII unless the employer was negligent in hiring or failing to fire or supervise the employee. In Restatement-speak, the supervisor in the second case has not acted in the employer's interest; but that is just a conclusion. The practical reason why the employer is not liable is that the supervisor's act took place outside the channels in which action by supervisors is realistically controllable by their superiors.

The distinction between company and noncompany acts points the way to the proper handling of a theoretical third category of supervisor sexual harassment. That is where the supervisor uses his supervisory authority to create a hostile work environment, perhaps by encouraging his male subordinates to harass his female subordinates. If he does this without committing any company acts, then the case should be assimilated to the ordinary case of noncompany harassing acts by a supervisor, and the employer's liability should be governed by negligence. If the supervisor commits company acts, then the employer should be strictly liable. As a practical matter, the employer will be liable in almost any case in which a supervisor is orchestrating a campaign of sexual harassment, for the employer will either learn of the campaign or be negligent in failing to notice it.

I do not think it is an objection that my proposed approach might occasionally require applying different standards of employer liability to different conduct of the same supervisory employee, whose sexual harassment of a subordinate might include both company acts and noncompany acts. Suppose a worker got drunk at a company party and on his way out of the building bumped into and injured a customer, then got into his car and on his way home hit a pedestrian, both accidents being due to his drinking. The company would be strictly liable to the customer--respondeat superior would apply--but the pedestrian would have to prove the company negligent in order to establish its liability to him, because the drive was not within the scope of his employment.

My proposed approach is summarized in the following table. The term "company act" signifies an act that significantly alters the terms or conditions of employment of the victim of sexual harassment and "noncompany act" signifies hostile-environment harassment by coworkers or supervisors or the kind of quid pro quo harassment that involves only unfulfilled threats (either because the victim submits or because she calls the supervisor's bluff), so that no company act is committed. The difficult borderline case is that of constructive termination precipitated by a threat. The termination will look to the supervisor's superiors like a voluntary quit. But since there is always some paperwork involved in an employee's quitting, the higher-ups in the company will have some ability to monitor constructive discharges, and I would therefore impose strict liability in such cases.

By supervisor By coworker

"Company act" employer strictly liable N.A.

"Noncompany act" employer liable only if employer liable only if

negligent negligent

I now apply the suggested approach to the facts of our two cases. In Jansen there was no company act and therefore no quid pro quo violation. Antoni did hold up Jansen's performance review and raise for several months, but the review was favorable, the raise was made retroactive, and Jansen herself admits that delays of the length she encountered were commonplace. The delays she encountered were not significant alterations in the terms or conditions of employment--therefore were not "company acts" as I am using that term in order to denote the proper scope of quid pro quo liability. For they were not visible to and therefore corrigible by higher-ups in the company. The tenuous character and trivial consequences of Antoni's threat furnish a practical argument for the limited scope of quid pro quo liability that I am urging.

Jansen has, however, presented enough evidence to establish a prima facie case of hostile-environment harassment. What Antoni did to Jansen over a period of some eighteen months--a barrage of threats and sexual solicitations--was highly offensive, and not mere sexual banter or even the kind of mildly obnoxious vulgarity that we held in Baskerville v. Culligan International Co., supra, did not rise to the level of actionable harassment. The only issue is whether Packaging Corporation should have discovered what was going on. The answer is yes, a tentative yes but one that can be given with enough confidence to defeat summary judgment for the employer. Two previous complaints of sexual harassment had been lodged against Antoni. The second, it is true, was somewhat cryptic; but the company's response to it, which was to counsel Antoni about the company's sexual harassment policy, showed that Packaging Corporation "read" the complaint as one of sexual harassment, as it pretty obviously was. Antoni was also well known to be having sex with another worker, and while this relationship was apparently consensual, the fact that he was married to someone else at the time should have alerted the company to the possibility that he was sexually incontinent. The company's lackadaisical if not perverse reaction to Jansen's complaint (accusing her of sexual harassment) is further evidence of the company's negligence with regard to the harassment of female employees.

So the grant of summary judgment to Packaging Corporation with regard to the charge of sexual harassment should be reversed. But the only issue for determination by the jury on remand should be whether the company was negligent in failing to protect Jansen from Antoni. There is, moreover, a serious question whether she can obtain substantial damages, given her failure to complain for eighteen months. We have held that the doctrine of avoidable consequences, on which see, e.g., Outboard Marine Corp. v. Babcock Industries, Inc., 106 F.3d 182, 184 (7th Cir.1997); Ellerman Lines, Ltd. v. S.S. President Harding, 288 F.2d 288, 289-90 (2d Cir.1961) (Friendly, J.); Keeton et al., supra, § 65, pp. 458-59, is applicable to violations of Title VII as to other torts. Hunter v. Allis-Chalmers Corp., 797 F.2d 1417, 1427 (7th Cir.1986). The precise issue is whether if Jansen had complained sooner the company would have taken effective action. This is uncertain given the lackluster way in which the company responded when she did complain. The uncertainty creates an issue for a jury.

On my view of the scope of quid pro quo liability, Ellerth like Jansen is purely a hostile-environment case. Slowik was Ellerth's supervisor. It is of no consequence, so far as his supervisory authority over her was concerned, that there was another supervisor in between them; this just shows how much power he had over her, given that he was that much higher in the corporate hierarchy. But he never committed a company act against her, such as denying her a promotion, although he obliquely threatened to do so. Therefore she should have to show that the company was negligent in failing to detect and prevent the harassment. The evidence of negligence is too thin to create a jury issue--so it is no surprise that Ellerth doesn't even argue that Burlington was negligent. Although aware of the company's policy against sexual harassment, Ellerth did not complain about Slowik to anyone in the company's human resources department or to anyone else who might have taken corrective action; she did not complain to her supervisor. Unlike Jansen's employer, Burlington Industries had no other sources of information about its supervisor's misbehavior. The fact that Slowik told off-color jokes and commented on Ellerth's legs at a lunch in which a vice-president of Burlington was present could not be thought to put Burlington on notice of probable harassment. The vice-president had no reason to think Ellerth offended by Slowik's banter; not all women would be. (Slowik had also patted Ellerth's knee under the table at that lunch, but there is no indication that this was visible to the vice-president.) Although there is a bit of evidence that Slowik may have harassed another woman employed by Burlington, there is no evidence that this other harassment came or should have come to the attention of any higher-ups in the company. The fact that Ellerth complained to a few of her fellow nonsupervisory employees (and toward the end to a customer service manager) also did not put the company on notice. The precedents in our court and the other circuits do not authorize the basing of a finding of negligence on such scraps of evidence. See, e.g., Doe v. R.R. Donnelley & Sons Co., 42 F.3d 439, 447 (7th Cir.1994); Juarez v. Ameritech Mobile Communications, Inc., 957 F.2d 317, 320-21 (7th Cir.1992); Murray v. New York University College of Dentistry, 57 F.3d 243, 250 (2d Cir.1995). A case in which the plaintiff does not think enough of her ability to prove negligence to even raise a negligence claim would be a pretty dubious vehicle for going beyond the precedents.

But it is an even weaker case for negligence than I have yet let on. Ellerth's claim that she feared that complaining about Slowik would cost her her job is inconsistent with her claim that she quit the company because Slowik had made her working conditions unbearable. If the choice is to quit the company or complain, you obviously should complain, since if the complaint gets you in trouble you're no worse off than you would have been quitting--unless you are seeking a reference for your next job, and there is no evidence of that in Ellerth's case.

Her action in quitting rather than complaining underscores the importance of agency principles that place appropriate pressure on victims of sexual harassment to protect themselves from harassment by complaining to the harasser's superiors. Tort law in general allocates responsibility for the avoidance of nonintentional injuries (not just for their consequences) to both injurer and victim, rather than to the injurer alone. Sexual harassment is an intentional injury by the harasser, but an unintentional injury (in the usual case and in the two cases before us) by the harasser's superiors, or in other words by the employer itself. The employer is innocent; the victim is innocent; the optimal system of liability for minimizing sexual harassment requires cooperation by both. Once Ellerth decided that Slowik had made the workplace unendurable, she had nothing to lose and everything to gain by complaining. It follows that as between her and her employer, she was in the better position to prevent the misconduct of her supervisor. Slowik had a responsibility to his principal not to harass Ellerth, but Ellerth had a responsibility to her principal too--a duty to complain about Slowik through proper channels. With rights should come responsibilities.

In any event, Ellerth explicitly waived negligence as a ground for Burlington's liability, and so it is not available as a basis for reversing the judgment in Burlington's favor. In her reply to the petition for rehearing, repeating what she had said earlier in her brief on appeal, Ellerth states, "Appellant [Ellerth], in fact, agrees with Burlington that § 219(2)(b) does not apply to this case" (emphases in original). The reference is to section 219(2)(b) of the Restatement--the section that makes an employer liable for the torts of his employee if the employer is negligent.

COFFEY, Circuit Judge, concurring in part and dissenting in part.

I concur with the majority view, set forth in the Court's per curiam opinion, that the district judge properly dismissed Jansen's retaliation and intentional infliction of emotional distress claims. Per Curiam Op. at p. 493. I also agree with the majority that the district judge's grant of summary judgment in Jansen in favor of PCA should be reversed and remanded in order that a jury might address the limited question of whether PCA was negligent with respect to the alleged hostile work environment sexual harassment by Jansen's supervisor. Per Curiam at pp. 493-494. I write separately because I do not agree that an employer who lacks knowledge of harassment by one of its supervisors should be held strictly liable under "agency principles," in spite of the fact that the employer may have had every intention of eliminating such conduct from the workplace, and may even have formulated and publicized a complete sexual harassment policy. I am unable to join with Judges Posner, Flaum, and Diane Wood,1 each of whom, in differing degrees, urges that we use the cases before us as a vehicle for announcing bold and unprecedented Seventh Circuit law that will make employers strictly liable in Title VII cases involving allegations of sexual harassment by a supervisor. I am convinced that the creation of a strict liability standard is contrary to the Supreme Court's decision in Meritor Savings Bank v. Vinson, 477 U.S. 57, 106 S. Ct. 2399, 91 L. Ed. 2d 49 (1986), and at odds with the "agency principles" that the High Court has instructed us to consider in Title VII cases.

Strict liability for quid pro quo harassment by supervisors effectively amounts to almost "guilt by association," a concept that is fundamentally unfair as well as inconsistent with "agency principles." Under a regime of strict liability, the employer--similar to the taxpayer who is assessed a tax deficiency by the IRS--would be presumed guilty, and the traditional presumption of innocence cast aside; thus, the employer would be required to bear the burden of proving his innocence.2 Because the employer saddled with strict liability would not be permitted to argue that it lacked knowledge of the harassment or that it had measures in place to prevent the same, it would have only a limited opportunity to rebut this judicially-created presumption of guilt. In a quid pro quo case, for example, the employer could avoid liability only by proving that the alleged harassment did not occur, and the many factors that can and should weigh against a finding of negligence on the part of the employer (i.e., knowledge, preventive measures, sexual harassment policy) become irrelevant once strict liability is adopted. Judicially legislating a strict liability standard will also serve to increase the volume of Title VII litigation and impose a costly burden on American business that will obviously be passed on to taxpayers, workers, and consumers (as with judicial decisions that have continuously expanded tort liability in the products liability and professional malpractice areas).

With due respect for my fellow judges, the approaches which Judges Posner, Flaum, Easterbrook and Wood advocate disregard the need for judicial restraint and faithfulness to precedent in this highly-charged area of the law, and dramatically tip the very delicate balance of the scales of justice in favor of Title VII plaintiffs.3 As the Supreme Court noted in Meritor, when Congress enacted Title VII, it provided little guidance as to how the statute was to be applied in cases of sexual harassment:

The prohibition against discrimination based on sex was added to Title VII at the last minute on the floor of the House of Representatives.... The principal argument in opposition to the amendment was that 'sex discrimination' was sufficiently different from other types of discrimination that it ought to receive separate legislative treatment.... This argument was defeated, the bill quickly passed as amended, and we are left with little legislative history to guide us in interpreting the Act's prohibition against discrimination based on 'sex.'

Id. at 63-65, 106 S.Ct. at 2404 (citations omitted). However, "the fact that Congress might have acted with greater clarity or foresight does not give courts a carte blanche to redraft statutes in an effort to achieve that which Congress is perceived to have failed to do." United States v. Locke, 471 U.S. 84, 95, 105 S. Ct. 1785, 1793, 85 L. Ed. 2d 64 (1985). The Court recognized this in Meritor when it advocated "agency principles" and concluded "that the Court of Appeals was wrong to ... impose absolute [strict] liability on employers for the acts of their supervisors." Meritor, 477 U.S. at 73, 106 S.Ct. at 2408 (emphasis added).

As a federal appellate court, we must honor the Supreme Court's Meritor decision and, in an exercise of judicial restraint, hold that an employer is liable for a supervisor's acts of sexual harassment under the negligence standard only if and when the plaintiff can demonstrate that the employer knew or should have known of the alleged harassment and failed to take appropriate measures. Juarez v. Ameritech Mobile Communications, Inc., 957 F.2d 317, 320 (7th Cir.1992). Additionally, the imposition of vicarious liability might conceivably be proper in that rare situation (not presented in either Jansen or Ellerth) where a plaintiff is able to establish that the supervisor acted with actual or apparent authority to engage in such conduct. Restatement (2d) Agency § 219(2)(a) & (d).

A strict liability standard, in contrast with the more flexible and reasonable negligence standard, will preclude us from accounting for "the circumstances of [each] particular case," Meritor, 477 U.S. at 73, 106 S.Ct. at 2408, including the disparate corporate structures of modern business. As Judge Wood explains, Title VII runs the ambit in its application, reaching small firms with as few as fifteen employees, as well as large entities such as General Motors, with some 692,800 workers. Wood Op. at p. 572. For Judge Wood to suggest, however, that a supervisor who stands on the second lowest (the 692,799th) rung of the corporate ladder at General Motors "stands in the place of [his] principal," Wood Op. at p. 572, to the same extent that the lowliest supervisor (the fourteenth employee) does at a fifteen person firm is simply untenable. In my view, to impose strict liability merely because an employee is designated as a "supervisor" in my view disregards the realities and complexities of modern corporate America.

A. Title VII Is Not A Strict Liability Statute

When called upon to determine when an employer may be held liable under Title VII for sexual harassment by one of its supervisors, we must start with the Supreme Court's clear and unambiguous mandate in Meritor that we "look to agency principles." 477 U.S. 57, 72, 106 S. Ct. 2399, 2408, 91 L. Ed. 2d 49 (1986). The High Court in Meritor, far from endorsing an all-expansive interpretation of "agency principles," specifically rejected strict liability and made clear that agency principles provide a limitation on employer liability rather than a blank check for expanding liability. Id. Specifically, the Meritor Court stated:

We ... decline the parties' invitation to issue a definitive rule on employer liability, but we do agree ... that Congress wanted courts to look to agency principles for guidance in this area. While such common-law principles may not be transferable in all their particulars to Title VII, Congress' decision to define 'employer' to include any 'agent' of an employer, 42 U.S.C. § 2000e(b) surely evinces an intent to place some limits on the acts of employees for which employers under Title VII are to be held responsible. For this reason, we hold that the Court of Appeals erred in concluding that employers are always automatically liable for sexual harassment by their supervisors. See generally Restatement (Second) of Agency §§ 219-237 (1958).

Id. (emphasis added). Meritor's rejection of strict liability remains the law, and has not been superseded by any precedent from the Nation's highest court, although some courts, in attempting to apply "agency principles," have improperly concluded that strict liability is mandated in sexual harassment cases. Curiously, even Judge Wood, who champions a sweeping strict liability standard, concedes that "the Meritor Court viewed agency principles as limiting, not expansive, factors in determining liability." Wood Op. at p. 569.

The political winds of change have apparently affected the views of the EEOC regarding employer liability, for while the EEOC, as amicus curiae, argued for a negligence standard in Meritor, it now urges us to adopt a "strict liability" rule in cases where a supervisor has allegedly harassed a subordinate. The EEOC's former position was that:

If the employer has an expressed policy against sexual harassment and has implemented a procedure specifically designed to resolve sexual harassment claims, and if the victim does not take advantage of that procedure, the employer should be shielded from liability absent actual knowledge of the sexually hostile environment.... In all other cases, the employer will be liable if it has actual knowledge of the harassment or if, considering all the facts of this case, the victim in question had no reasonably available avenue for making his or her complaint known to appropriate management officials.

Meritor, 477 U.S. at 71, 106 S.Ct. at 2407-08 (quoting Brief for United States and EEOC as Amici Curiae). At oral argument, Judge Wood attempted to explain away this patent inconsistency between the EEOC's previous and current positions on the grounds that "Meritor is ten years old.... We have come, you know, a very long way in ten years." Meritor should be dispositive of the issues before us, and I disagree with the suggestion that we as judges on a United States Circuit Court of Appeals have a license to take the law beyond Meritor simply because that decision may not comport with our individual notions of social progress in 1997.

The Supreme Court in Meritor also recognized that the "existence of a grievance procedure and a policy against discrimination," while not necessarily dispositive, is "plainly relevant" and definitely a fact that should be considered in determining the extent of an employer's Title VII liability. Id. at 72-73, 106 S. Ct. at 2408. The Court's willingness to acknowledge the liability-mitigating effect of an anti-discrimination policy, even when the harasser is a supervisor, makes clear that in the interest of justice and fairness, we are bound by a negligence standard in all sexual harassment cases, regardless of the perpetrator's rank vis-a-vis his victim.

Because Meritor was a hostile work environment case,4 the Court did not definitively reach the question of an employer's liability for quid pro quo harassment by a supervisor. Nevertheless, there are courts which purport to follow Meritor and have "look[ed] to agency principles" as a vehicle to expand Meritor's holding and conclude that strict liability is the proper standard in supervisory quid pro quo cases. See, e.g., Nichols v. Frank, 42 F.3d 503, 513-14 (9th Cir.1994) (collecting cases); but see Nash v. Electrospace System, Inc., 9 F.3d 401 (5th Cir.1993) (Title VII is not a "strict liability statute for employers"). Our own post-Meritor decisions to date have not specifically addressed the issue of whether an employer is strictly liable for quid pro quo harassment by a supervisor, except to state in clear and unambiguous language that this remains an open question in light of Meritor. As Chief Judge Posner, writing for this court, explained just two years ago:

An employer is not strictly liable for sexual harassment of one worker by another unless, perhaps, the harassment takes the form ... of an abuse of authority, as where a supervisor threatens to fire a subordinate if she refuses to have sex with him. In such cases, a number of courts treat the supervisor as the employer, whether correctly or not we need not decide. (Neither the Supreme Court nor our court has had occasion to decide the question.) In all other cases, it is clear, the criterion for when an employer is liable for sexual harassment is negligence.

Baskerville v. Culligan International Co., 50 F.3d 428, 431-32 (7th Cir.1995) (emphasis added) (citations omitted).

Prior to Meritor, this circuit for a brief period adhered to a rule of strict liability, as set forth in Horn v. Duke, 755 F.2d 599 (7th Cir.1985). Although my respected colleague Judge Flaum relies to an extent upon the reasoning in Horn (see Flaum Op. at pp. 496-497), this reliance is at best confusing, for the Horn case was decided one year before the Supreme Court's Meritor decision, and Horn has since been called into question by this court. In North v. Madison Area Ass'n for Retarded Citizens, for example, we addressed the proper scope of employer liability in the context of a race-discrimination claim under Title VII, and, in light of Meritor, we disavowed Horn's rule of strict liability. North, 844 F.2d 401, 407 (7th Cir.1988). Although North, a post-Meritor decision concerned a case of race discrimination, as opposed to sexual harassment, it nonetheless addressed the proper scope of an employer's liability for the discriminatory acts of a supervisor. In North, we stated unequivocally that "in light of the decision in Meritor ... employers are no longer to be held to ... a strict standard in every case in which the actions of a supervisory employee are in question." Id. (emphasis added). This court's North decision held that in the absence of evidence that the supervisor acted with actual or apparent authority,5 the employer "could be held liable for discriminatory actions which were attributable to [the supervisor] only if it knew or should have known of those actions and failed to take reasonable remedial measures." Id. I am at a loss to understand why this court should wish to reverse course and re-instate a strict liability rule that we have held to be improper under Meritor.

This court has consistently rejected the strict liability principle in the context of sexual harassment cases involving non-supervisory employees. In Juarez, for example, a 1992 opinion Judge Cudahy authored, we cited Meritor for the proposition that "[e]mployers are not strictly liable under Title VII for sexual harassment engaged in by their employees" and proceeded to analyze the facts of the case under the accepted "knew or should have known" rubric (i.e., negligence). 957 F.2d at 320; see also Baskerville, 50 F.3d at 431-32.

In line with the principles enunciated in Meritor, we should apply a negligence standard in the cases before us, and in all cases where a plaintiff seeks to hold an employer liable for acts of sexual harassment allegedly committed by another employee. Notice to the employer (i.e., knowledge) is a key component of the negligence standard. As Chief Judge Posner very recently observed (in an opinion which I joined along with Judge Easterbrook), an "employer's duty to investigate and if appropriate take remedial measures is not activated until the employee complains of sexual harassment or information about the harassment comes to the employer's attention from some other quarter." Zimmerman v. Cook Co. Sheriff's Dept., 96 F.3d 1017, 1019 (7th Cir.1996).6 What constitutes "appropriate remedial action" depends on the gravity of the alleged harassment in a particular case, the information made available to the employer, and of course, the type of corrective action instituted. Baskerville, 50 F.3d at 432 ("Just as in conventional tort law a potential injurer is required to take more care, other things being equal, to prevent catastrophic accidents than to prevent minor ones, so an employer is required to take more care, other things being equal, to protect its female employees from serious sexual harassment than to protect them from trivial harassment.").

Before we even think about jettisoning the strict liability doctrine into the law of employment discrimination in this circuit, it is imperative that we closely examine the nature of the strict liability doctrine, as well as its origins and purpose in tort law. In the interests of justice, we must analyze and determine whether strict liability, whatever its arguable merits in other areas of the law, is well-suited to the Title VII context. Under a strict liability theory, the plaintiff need not establish that the employer was negligent or blameworthy, or that the employer authorized, condoned, or sanctioned harassment in the workplace, for the defendant under this theory is strictly liable, or liable "without fault." W. Page Keeton et al, Prosser and Keeton on the Law of Torts § 75, at 534 (5th ed.1984). Strict liability or "liability without fault" is the exception rather than the norm in Anglo-American tort law, for the doctrine has traditionally been confined to those situations where a defendant "engaged in some kind of activity which exposed others to a risk of harm...." Chicago, Milwaukee, St. Paul & Pacific Railroad Co. v. Union Pacific Railroad Co., 78 F.3d 285, 290 (7th Cir.1996) (discussing environmental statute which "impose[d] strict liability on an owner of property based merely on ownership without regard to what actions the owner took on the property to cause pollution."). For example, under the common law of many jurisdictions, a party who engages in "abnormally dangerous" or "ultrahazardous" activity that results in harm (such as engaging in blasting with dynamite in a residential area) may be held strictly liable, even though the party engaged in the activity exercised due care to prevent the resulting harm. See, e.g., Opal v. Material Service Corp., 9 Ill.App.2d 433, 133 N.E.2d 733 (1956); see also Restatement (Second) of Torts §§ 519 & 520 (1977).7

The rationale for strict liability is most often stated in policy terms, and specifically, economic terms. As the California Supreme Court explained in Greenman v. Yuba Power Products, Inc., a landmark products liability decision, "[t]he purpose of strict liability is to insure that the costs of injuries resulting from defective products are borne by the manufacturers that put such products on the market rather than by the injured persons who are powerless to protect themselves." 59 Cal. 2d 57, 63, 27 Cal. Rptr. 697, 377 P.2d 897, 901 (1963). In jurisdictions that recognize the doctrine of strict liability for defective products, a company which manufactures a faulty product and places it in the stream of commerce is presumed to be legally responsible if harm results from the use of that product, unless the defendant can establish a recognized defense such as assumption of risk or misuse of the product. Louis R. Frumer, et al. 2 Products Liability § 8.01 (1997).8 Strict liability literally shifts the burden of proof from the plaintiff to the defendant, reflecting a policy judgment that manufacturers, as a general rule, should bear the costs associated with defective products that cause harm.

As discussed below, I am convinced that the same cost-shifting rationale which may be argued to support a theory of strict liability for manufacturers of defective products (or those who engage in ultrahazardous activities such as blasting with dynamite) does not justify making employers "strictly liable" for the seemingly intractable social problem of sexual harassment. In my view, therefore, the adoption of strict liability in the employment context is not only contrary to agency principles, it also represents a novel, ill-advised, and harmful extension of the doctrine of strict liability that will have deep and lasting social and economic consequences. Policy decisions of this magnitude have traditionally been left to the legislative branch of government, and should remain the province of the people's elected representatives. In fact, Congress has not even addressed the issue of an employer's liability for sexual harassment, much less indicated that employers should be held strictly liable under a "cost-shifting" theory. Until such time as Congress--the Nation's true legislature--has spoken on this question, it would be more proper for us to refrain from imposing the far-reaching standard of strict liability by judicial edict.

Before analyzing the strict liability theories outlined by Judges Posner, Flaum, Easterbrook, and Diane Wood, let us initially define the term "agency principles." Our duty is to interpret the common law of agency, not to make new law. We therefore look to the principles set forth in the Restatement (2d) of Agency, which was cited in the Meritor decision, and in particular § 219 of the Restatement. As the trial judge in Jansen explained, § 219 delineates three potential theories for holding employers liable for the sexual harassment of their employees:

Section 219(1) holds employers responsible for torts committed by their employees within the scope of their employment.... Under § 219(2)(b), masters are liable for their own negligence or recklessness; in a harassment case, this is typically negligent failure to discipline or fire, or failure to take remedial action upon notice of harassment. Finally, under § 219(2)(d), if the servant relied upon apparent authority or was aided by the agency relationship, the master is required to answer.

Jansen, 895 F. Supp. at 1061 (quoting Bouton v. BMW of N. Am., Inc., 29 F.3d 103, 106 (3d Cir.1994)). With respect to the first of these three theories (respondeat superior liability), it is crucial to recognize that:

The liability of an employer for torts committed by its employees--without any fault on his part--when they are acting within the scope of their employment, the liability that the law calls "respondeat superior," is a form of strict liability. It neither requires the plaintiff to prove fault on the part of the employer nor allows the employer to exonerate himself by proving his freedom from fault.

Konradi v. United States, 919 F.2d 1207, 1210 (7th Cir.1990) (emphasis added). An understanding of these theories provides a basis for addressing my colleagues' respective positions, which I detail below.1. Judge Easterbrook's Approach to Employer Liability

Judge Easterbrook and Judge Wood are of the opinion that state rather than federal law should define the parameters of an employer's liability under Title VII, although none of the parties have urged us to take this curious and novel (indeed, unprecedented) approach. Out of the vast constellation of Title VII cases decided by the federal courts, including those that have addressed the issue of employer liability, I have been unable to locate a single precedent that supports the unique position of Judges Easterbrook and Wood on this choice of law question.9 In fact, Meritor provides a strong refutation of my colleagues' position, for the Supreme Court in Meritor did not qualify or limit the "agency principles" that federal trial courts were to "look to ... for guidance," much less require that "agency principles" be derived from state law. 477 U.S. at 77, 106 S. Ct. at 2410.10

Judge Easterbrook presents a somewhat abridged version of the Supreme Court's recent decision in Atherton v. FDIC, --- U.S. ----, 117 S. Ct. 666, 136 L. Ed. 2d 656 (1997), and cites that case for the proposition that "[w]hen a federal statute is silent, we obtain the necessary rule from state law, unless application of state law would undermine a federal norm." Easterbrook Op. at p. 553. Atherton does state that "normally, when courts decide to fashion rules of federal common law, the guiding principle is that a significant conflict between some policy or interest and the use of state law ... must first be specifically shown." Id. However, Atherton also makes clear that this "guiding principle" is only applicable to the creation of " 'federal common law' in the strictest sense, i.e., a rule of decision that amounts, not simply to an interpretation of a federal statute ... but, rather, to the judicial 'creation' of a special federal rule of decision." Id. (emphasis added). Similarly, in Northwest Airlines, Inc. v. Transport Workers Union, the Court noted that:

In almost any statutory scheme, there may be a need for judicial interpretation of ambiguous or incomplete provisions. But the authority to construe a statute is fundamentally different from the authority to fashion a new rule or to provide a new remedy which Congress has decided not to adopt.

451 U.S. 77, 97, 101 S. Ct. 1571, 1583, 67 L. Ed. 2d 750 (1981) (emphasis added); see also United States v. Little Lake Misere Land Co., 412 U.S. 580, 593, 93 S. Ct. 2389, 2397, 37 L. Ed. 2d 187 (1973) ("[T]he inevitable incompleteness presented by all legislation means that interstitial federal lawmaking is a basic responsibility of the federal courts."). I agree with Chief Judge Posner that "[d]eciding what agency principles shall govern liability under a liability-creating statute such as Title VII is not free-wheeling common-law rule-making; it is filling a statutory gap, [which is] a standard office of interpretation." Posner Op. at p. 507 (emphasis added).

Even if, for purposes of argument, I were to agree with Judge Easterbrook's premise that deciding the question of employer liability requires something more than traditional statutory interpretation, federal common law rules governing employer liability serve the basic "policy or interest" of uniformity. Interposing the statutes and decisional law of the fifty states, as proposed by Judges Easterbrook and Wood, would make a crazy-quilt of the law and thus grievously undermine our interest in fostering uniformity. Such an approach would make it most difficult, if not impossible, for an employer with plants in more than one state to comply with federal law, obviously necessitating that employers retain knowledgeable local counsel in several states, or even all of them.11 The Supreme Court has recognized that federal courts may "fill the interstices of federal remedial schemes with uniform federal rules ... when the scheme in question evidences a distinct need for nationwide legal standards." Kamen v. Kemper Financial Svcs., Inc., 500 U.S. 90, 97-98, 111 S. Ct. 1711, 1716-17, 114 L. Ed. 2d 152 (1991); see also O'Melveny & Myers v. FDIC, 512 U.S. 79, 88, 114 S. Ct. 2048, 2055, 129 L. Ed. 2d 67 (1994) (recognizing "federal interest ... in uniformity"); Wilson v. Garcia, 471 U.S. 261, 266-69, 105 S. Ct. 1938, 1941-43, 85 L. Ed. 2d 254 (1985) (resort to state law would result in "conflict, confusion, and uncertainty" concerning the appropriate statute of limitations in § 1983 cases). As Judge Manion points out most persuasively, Manion Op. at p. 564, the Court underscored the need for uniformity in Community for Creative Non-Violence v. Reid, 490 U.S. 730, 109 S. Ct. 2166, 104 L. Ed. 2d 811 (1989), which stated:

In past cases of statutory interpretation, when we have concluded that Congress intended terms such as 'employee,' 'employer,' and 'scope of employment' to be understood in light of agency law, we have relied on the general common law of agency, rather than on the law of any particular State, to give meaning to these terms.... This practice reflects the fact that federal statutes are generally intended to have uniform nationwide application.

Id. at 740, 109 S.Ct. at 2173 (emphasis added) (quotation omitted).

Judge Easterbrook suggests that adopting uniform federal rules of decision for employer liability under Title VII is somehow contrary to Erie R.R. Co. v. Tompkins, 304 U.S. 64, 58 S. Ct. 817, 82 L. Ed. 1188 (1938), because "there is no free-floating [federal] common law." Easterbrook Op. at p. 553. However, "[s]ince Erie, no decision of [the Supreme] Court has held or suggested that [the lower federal courts must] borrow[ ] state law to fill gaps in federal substantive statutes." DelCostello v. Teamsters, 462 U.S. 151, 160, 103 S. Ct. 2281, 2289, 76 L. Ed. 2d 476 (1983) (no "mandatory rule that we apply state law in federal interstices."); see also Sola Electric Co. v. Jefferson Electric Co., 317 U.S. 173, 176, 63 S. Ct. 172, 174, 87 L. Ed. 165 (1942). As Justice Jackson once explained:

In some cases [a federal court sitting in a non-diversity case] may see fit for special reasons to give the law of a particular state highly persuasive or even controlling effect, but in the last analysis its decision turns upon the law of the United States, not that of any state. Federal law is no juridical chameleon, changing complexion to match that of each state wherein lawsuits happen to be commenced.... It is found in the federal Constitution, statutes, or common law. Federal common law implements the federal Constitution and statutes and is conditioned by them. Within these limits, federal courts are free to apply the traditional common-law technique of decision and to draw upon all the sources of the common law in cases such as the present.

D'Oench, Duhme & Co. v. FDIC, 315 U.S. 447, 471-72, 62 S. Ct. 676, 686, 86 L. Ed. 956 (1942) (Jackson, J., concurring) (emphasis added).

I am in full agreement with the majority that the question of employer liability under Title VII should be treated as a matter of federal law, and that we should reject what Judge Manion aptly labels "a 50-state standard wrapped in local common law." Manion Op. at p. 564. The majority's approach, in contrast with that of Judge Easterbrook and Judge Wood, is faithful to the principle that "[l]egal rules which impact significantly upon the effectuation of federal rights must ... be treated as raising federal questions." Burks v. Lasker, 441 U.S. 471, 477, 99 S. Ct. 1831, 1836, 60 L. Ed. 2d 404 (1979).

2. Judge Wood's Approach to Employer Liability

Judge Wood (joined by Judges Easterbrook and Rovner) purports to follow "general principles of traditional agency law,"12 and insists that her approach "is not the 'strict liability' that the Supreme Court rejected in Meritor." Wood Op. at p. 566 n. 2. However, a careful reading of her entire opinion clearly demonstrates that she has seen fit to go far beyond Meritor by endorsing an overly broad strict liability standard that is based upon an unprecedented and (I believe) incorrect expansion of the respondeat superior doctrine.

Judge Wood asserts, without support in case law, that the "proper inquiry" in determining whether an employer may be held liable under the doctrine of respondeat superior is "whether the supervisor, acting at least in part with the purpose of serving the employer, committed the acts of harassment as he exercised the authority (actual or apparent) the employer conferred on him: hiring, firing, assigning work, disciplining other employees for harassing behavior, and the like." Wood Op. at p. 574 (emphasis added). I have two principal objections to the way Judge Wood has re-fashioned the respondeat superior doctrine for purposes of Title VII:(1) her analysis greatly expands the concept of "scope of employment," and (2) it fails to make clear who is (and is not) a "supervisor."

I agree with Judge Wood that, in determining the scope of an employer's respondeat superior liability under § 219, "[t]he first question ... should be whether the supervisor's action was within the scope of his employment." Wood Op. at p. 573 (emphasis added). But Judge Wood's definition of "within the scope of employment" is so broad and all-inclusive that it effectively imposes "automatic" or strict liability, contrary to the Meritor decision. Judge Wood argues that liability must be strict if a supervisor harasses another employee "as he exercise[s]" the actual or apparent authority delegated to him. The relevant conduct, according to Judge Wood, is "not the act of harassment," but instead the "broader course of action," i.e., the supervisor's performance of authorized duties. Wood Op. at p. 574. I disagree. If we were to adopt this reasoning, the legal phrase "within the scope of employment" would in effect become meaningless, for it would encompass a panoramic range of conduct, whether authorized or not. Without defined parameters, the phrase "within the scope of employment" no longer serves to limit the reach of respondeat superior liability and becomes a kind of legal pit of quicksand, drawing in all manner of activity. Applying Judge Wood's attenuated logic, a company could be held liable if its agent committed any kind of unauthorized conduct imaginable, short of rape,13 so long as the conduct somehow occurred "as [the supervisor] exercised" some kind of authorized job duty. For example, a television station could be held liable if one of its news reporters "lost his cool" and physically assaulted a person while recording an interview. Such an act would give rise to liability under Judge Wood's theory because it occurred "as" the reporter was reporting on news (authorized activity), but respondeat superior liability should not be imposed on these facts because the conduct in question clearly falls outside the scope of the reporter's legitimate duties and is not motivated by a purpose to serve the employer. Judge Wood's analysis converts § 219's theory of respondeat superior (or "the master answers") into a far more all-inclusive and unbending theory of strict liability or "guilt by association."

The concepts advisedly set forth in § 219(1) of the Restatement are not so elastic. An employee's conduct (i.e., sexual harassment) "is not within the scope of employment if it is different in kind from that authorized ... or too little actuated by a purpose to serve the [employer]." Restatement (2d) Agency § 228(2) (emphasis added); see also Denlinger v. Brennan, 87 F.3d 214, 216 (7th Cir.1996) ("intentional torts outside the scope of employment usually do not lead to an employer's vicarious liability"); Joel v. Morrison, 6 C. & P. 501, 172 Eng. Rep. 1338 (1834) (servant's conduct may not be imputed to the master when servant is "on a frolic of his own."). As Judge Bork observed when dissenting from the D.C. Circuit's decision in Meritor, "[i]ntentional torts involving sexual escapades almost never result in employer liability because they are personally motivated." Vinson v. Taylor, 760 F.2d 1330, 1332 n. 6 (Bork, J., dissenting) (emphasis added). Thus, a rule making employers liable for such escapades is "at odds with traditional practice." Id. Judge Wood asserts that the language in Restatement § 228, quoted above, "imposes a meaningful limitation on the scope of employment, thereby ensuring that employers will not be held liable for the 'frolics and detours' of their employees or for acts that are unforeseeable to the employer." Wood Op. at pp. 573-574. Yet for Judge Wood, it appears to be a given that unlawful quid pro quo harassment "will almost always fall within the scope of [a] supervisor's employment and thus result in employer liability" because it occurs so often "as" (or while) the supervisor is performing authorized duties. Wood Op. at pp. 565, 574.

Judge Wood's boundless definition of "scope of employment" contrasts sharply with the traditional understanding of that term. As set forth in a leading treatise on tort law, "scope of employment refers to those acts which are so closely connected with what the servant is employed to do, and so fairly and reasonably incidental to it, that they may be regarded as methods, even though quite improper ones, of carrying out the objectives of the employment."

W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 70, at 502 (5th ed.1984). While the general rule of respondeat superior embodied in § 219 holds that an employer may be liable for the tortious conduct of employees acting within the scope of their employment, it is crucial to note that:

not every act which an agent or servant may do while he is in the place appointed for the service, or during the time in which he is engaged in the performance, can be deemed to be within the course of employment, or within the scope of the authority. The test lies deeper than that; it inheres in the relationship which the act done bears to the employment. The act cannot be deemed to be within the course of employment unless, upon looking at it, it can fairly be said to be a natural, not disconnected and not extraordinary, part or incident of the service contemplated.

Hibma v. Odegaard, 769 F.2d 1147, 1168 (7th Cir.1985) (Coffey, J., dissenting) (quoting Cameron v. City of Milwaukee, 102 Wis. 2d 448, 456, 307 N.W.2d 164 (1981)). The Restatement makes clear that "not all physical acts of the kind authorized performed within the time and at the place of service are within the scope of employment, since only those which the servant does in some part for the purpose of giving service to the master are included." Restatement (Second) of Agency § 228, cmt. (a). If an employee has " 'stepped aside from the business of his principal to accomplish an independent purpose of his own,' " far removed from the scope and nature of his duties and responsibilities, the employer may not be held liable under the doctrine of respondeat superior. Hibma, 769 F.2d at 1168 (quoting Linden v. City Car Co., 239 Wis. 236, 239, 300 N.W. 925 (1941)).

In addition to broadening the meaning of "scope of employment," Judge Wood's analysis of respondeat superior liability for sexual harassment by supervisors fails to adequately address the definition of "supervisor." Judge Wood describes a supervisor as one who "exercise[s] the authority (actual or apparent) the employer conferred on him: hiring, firing, assigning work, disciplining others for harassing behavior, and the like." Wood Op. at p. 574. This "definition" of supervisor raises a number of questions: Is an employee who exercises any one of the duties which Judge Wood delineates a "supervisor" for purposes of imposing strict liability? May an employee be deemed a supervisor even if the authority delegated to him (over firing, for example) is shared with others? And what activities fall within Judge Wood's catch-all phrase "and the like," which follows her list of supervisory duties? If we are to lock ourselves into a strict liability standard for supervisory cases, as Judge Wood proposes, it is imperative that we define "supervisor" most precisely.

Our cases interpreting the statutory definition of "supervisor" in the National Labor Relations Act, 29 U.S.C. § 152(11) ("NLRA"), have examined the meaning of this term in some detail and recognized that "[s]upervision in the elementary sense of directing another's work" does not suffice to make an employee a "supervisor." N.L.R.B. v. Res-Care, Inc., 705 F.2d 1461, 1465 (7th Cir.1983). Rather, a supervisor "must have authority over another's job tenure and other conditions of employment." Id.; see also N.L.R.B. v. Winnebago Television Corp., 75 F.3d 1208, 1213 ("The test of a supervisor under § 152(11) is whether his judgment was dispositive in personnel decisions."). The need to define "supervisor" with precision--and to provide some meaningful, common-sense limitations on who may be deemed a "supervisor"--is even more imperative in cases of sexual harassment because Title VII, unlike the NLRA,14 fails to provide a statutory definition of this term. It would be helpful if Congress filled in this definitional gap in the law, particularly because judicial opinions expanding employer liability have relied so heavily on the distinctions between harassment by supervisors and harassment that occurs among co-workers. In the absence of a statutory or judicial definition setting forth parameters for the term "supervisor," sexual harassment plaintiffs will obviously have an incentive to characterize all alleged harassers as "supervisors," even in cases where the wrongdoer lacked significant authority over the plaintiff's hiring, firing, or conditions of employment.

In defining "supervisor" and addressing what conduct does (and does not) fall "within the scope of employment," we should recognize and understand that most employees, including those labeled as "supervisors," are subordinate to a number of other, higher levels of management, and lack the authority to make personnel decisions regarding whom to hire, fire, or promote. Even an employee who is just above the bottom of the employment "chain of command" may have the singular authority to "assign[ ] work" to other employees, which, according to Judge Wood, is one of the duties that transforms an employee into a "supervisor" for purposes of Title VII liability. For example, a "sanitation engineer" employed by a garbage and trash collection company may be charged with overseeing a crew of two or three workers assigned to the same truck, and authorized to "assign work" (i.e., decide on a given day who will empty trash bins and who will operate the truck). However, just because the employee has been authorized to perform such minimal duties, or just because his job description, having been elevated solely for payroll purposes, classifies him as a "supervisor," does not mean that he qualifies as a decision-maker in the corporate enterprise, much less that he falls within the upper echelon of responsibility at the firm.

A supervisor should only be considered as an agent of the firm when he is truly vested with more than a mere modicum of authority to act for, and on behalf of, his principal and when he is acting to advance the principal's interests, as when an individual with authority to hire and fire, promote and demote another employee actually exercises that authority, and objectively believes that his actions are in the best interests of the firm. In my view, Judge Wood is grasping at a very thin reed of "agency law" if she wishes to hold an organization strictly liable for an employee's intentional torts merely because that employee occupied some limited type of supervisory role within the firm, perhaps very far down in the company's hierarchy. For example, an individual classified as a supervisor of the housekeeping services detail within a corporation acts "within the scope of his employment" when he gives directions to his cleaning staff about what cleaning compound to use, or which brand of brush or vacuum to use, or when he directs that they sweep the floors at 7:30 PM rather than 8:30 PM. Would Judge Wood likewise hold a newspaper strictly liable if an assistant city editor, in a fit of anger, physically assaulted or even went so far as to shoot and fatally wound one of the reporters whose work he supervised? Should a correctional facility be held strictly liable if it allows a prisoner extra compensation for directing the activities of a clean-up crew and the supervisor/prisoner sexually assaults one of the inmates under his supervision? All of these intentional and unauthorized acts by "supervisors" represent a departure from the employee's legitimate supervisory role and therefore should not be imputed to the principal under the theory of respondeat superior. Yet under Judge Diane Wood's all-expansive interpretation of § 219(1) and her imprecise definition of the term "supervisor," liability could be imposed in each of these hypothetical situations, so long as the "supervisor's" misconduct occurred during the course of performing authorized duties.

Turning to the facts of the cases under consideration, the alleged conduct of Jansen's harasser (Antoni) and Ellerth's (Slowik) fell outside the scope of their employment. As the Tenth Circuit observed in Hicks v. Gates Rubber Co., "sexual harassment simply is not within the job description of any supervisor or any other worker in any reputable business." 833 F.2d 1406, 1418 (10th Cir.1987) (quotation omitted). Although Judge Wood seemingly acknowledges the limitations on respondeat superior liability by stating that the supervisor must "act[ ] at least in part with the purpose of serving the employer" Wood Op. at p. 574. In order for the employer to be liable, there is not a scintilla of evidence in either of the respective court records under review suggesting that the conduct of either Antoni or Slowik was motivated by a purpose to serve their employer. Like all of my colleagues on this court, I am unalterably opposed to true sexual harassment, which is vulgar, repulsive and uncalled for under any circumstances. However, I think it is obvious that those who engage in sexual harassment are driven by personal motives alone--motives that have nothing to do with furthering the aims and goals of their employers. As Chief Judge Posner, writing for this court, noted in a race discrimination case:

[H]arassment is an intentional wrong, and the doctrine of respondeat superior makes an employer liable only for those intentional wrongs of his employees that are committed in furtherance of the employment; the tortfeasing employee must think (however misguidedly) that he is doing the employer's business in committing the wrong. It would be the rare case where ... harassment against a co-worker could be thought by the author of the harassment to help the employer's business.

Hunter v. Allis-Chalmers Corp., 797 F.2d 1417, 1421 (7th Cir.1986) (emphasis added). In his scholarly review of post-Meritor sexual harassment law, Michael J. Phillips, a Professor of Business Law at Indiana University, has taken note of Judge Posner's observations in Hunter concerning the intentional nature of harassment, and argued cogently that "[e]mployers should virtually never be liable for sexual harassment if respondeat superior is the agency principle of choice because such behavior is rarely, if ever, within the scope of employment." Employer Sexual Harassment Liability Under Agency Principles: A Second Look at Meritor Savings Bank, FSB v. Vinson, 44 Vand. L. Rev. 1229, 1245 (1991) (emphasis added). An employee who sexually harasses another employee has not only violated the law, but in most cases (including the two before us today), he has likewise acted contrary to company policy. Because the harasser has clearly "stepped aside from the business of his principal" and acted beyond the "scope of his employment," his actions cannot be attributed to his employer under the theory of respondeat superior as set forth in § 219(1).

Judge Wood's opinion, with its elastic interpretation of the respondeat superior doctrine, represents a most dramatic expansion of employer liability. Her opinion is all the more sweeping because it goes on to suggest that we adopt a uniform mode of analyzing employer liability in all sexual harassment cases involving supervisors. Wood Op. at pp. 567-569. Not content to limit her version of strict liability to so-called quid pro quo cases, Judge Wood (joined by Judges Easterbrook and Rovner) would also extend that standard into each and every case in which a supervisor is alleged to have created a hostile work environment.15 Judge Wood's broad, all-inclusive, boundless net of strict liability, reaching quid pro quo and hostile work environment cases alike, finds no direct support in case law from any circuit,16 much less the Supreme Court. Those circuits that have adopted a strict liability standard limit the application of that standard to quid pro quo cases only, as Judge Wood concedes. Wood Op. at p. 569; Nichols v. Frank, 42 F.3d 503, 513 (9th Cir.1994); Bouton v. BMW of North America, 29 F.3d 103, 106 (3d Cir.1994); Sauers v. Salt Lake County, 1 F.3d 1122, 1127 (10th Cir.1993); Kauffman v. Allied Signal, Inc., 970 F.2d 178, 183 (6th Cir.1992); Steele v. Offshore Shipbuilding, Inc., 867 F.2d 1311 (11th Cir.1989).

Quid pro quo harassment does not fall "within the scope" of a supervisor's employment, as Judge Wood argues, because such conduct departs drastically from the supervisor's authorized duties and is not motivated by a purpose to serve the employer. Similarly, acts which create a hostile work environment (such as unwelcome sexual advances, touching, remarks which suggest an animus towards women, and the like) fall far afield from the scope of a supervisor's employment-related duties, as interpreted by the courts, and bear no relation to any legitimate activity that the supervisor is authorized to perform. As Judge Wood concedes in a footnote to her opinion, Wood Op. at p. 569, n. 4, the Eleventh Circuit recently rejected her view that employers may be held strictly liable under agency principles for acts of a supervisor which contribute to a hostile work environment. Faragher v. City of Boca Raton, 111 F.3d 1530 (11th Cir.1997) (en banc). In Faragher, two supervisory employees allegedly subjected the plaintiff, a female lifeguard employed by the City of Boca Raton, Florida, to "offensive comments, gestures, and touching." Ms. Faragher brought suit against the city under Title VII, alleging that the municipality, as the common employer of herself and the harassing supervisors, should be held vicariously liable for the hostile work environment to which she was subjected. In rejecting Faragher's claim, the Eleventh Circuit properly applied the principles of agency law, noting that when an "agent steps outside of his employment to do some act for himself which is not connected to his employer's business," then "the act is not within the scope of his employment." Id. at 1536. In other words, "if the agent acts from purely personal motives [i.e., sexual harassment], he is usually considered to have departed from his employment and his employer is not liable." Id. at 1537. The Court concluded that the harassers' offensive language, gestures and touching provided an "archetypical example of employees stepping outside of the scope of their employment and seeking to further personal ends." Id. In line with its proper understanding of agency law, the Eleventh Circuit, in an en banc decision, held that the city could not be held liable for the harassing conduct under a theory of vicarious liability. Id.

I am in agreement with the Eleventh Circuit that, if a supervisor or any other employee "acts from purely personal motives," as is inevitably the case in a hostile work environment scenario, he has "departed from his employment and his employer is not liable" under the principle of respondeat superior. Id. A majority of this Court agrees that strict liability is not warranted in hostile work environment cases, Per Curiam at p. 495, and the dearth of supporting case law for a "blanket rule" (extending strict liability to all cases involving supervisors) should certainly give pause to Judges Wood, Easterbrook, and Rovner, who are in a distinct minority on this issue.

3. Judge Flaum's Approach to Employer Liability

Like Judge Wood, Judge Flaum believes that "[a]gency law [and specifically § 219] supports the imposition of vicarious liability upon employers for supervisory quid pro quo harassment." Flaum Op. at p. 496. Judge Flaum reasons that when a supervisor uses the authority delegated to him to effectuate a quid pro quo, " 'the supervisor, by definition, acts as the company.' " Flaum Op. at p. 497 (quoting Steele, 867 F.2d at 1316)(emphasis added). Under this approach, the morally culpable harasser and the employer are treated as if they had somehow "merg[ed] into a single entity,"17 in spite of the fact that the employer lacked any knowledge of the harassment and took reasonable measures to prevent such conduct in the workplace. To my way of thinking, Judge Flaum's equating of employer and harasser is just a different colored facade on the same legal structure erected by Judge Wood, i.e., the theory that a "supervisor" (loosely defined) acts within the "scope of employment" (broadly defined) when he engages in quid pro quo harassment. Despite their differences, Judge Flaum and Judge Wood agree that an employer may be held strictly liable for quid pro quo harassment by a supervisor. I disagree with that proposition as a matter of both law and logic.

Although Judge Flaum advocates a strict liability standard for supervisory, quid pro quo cases, he states that the standard in hostile work environment cases should "remain at negligence." Flaum Op. at p. 502. Thus, Judge Flaum's approach is not quite as drastic as Judge Wood's, in that it does not propose strict liability for hostile work environment cases. Nevertheless, Judge Flaum asserts that employers have a "heightened duty of care" when it comes to preventing hostile work environment harassment by supervisors. Flaum Op. at p. 502. He does not define the meaning of this "heightened duty of care" standard, although he states that the "remedial goals" of Title VII require an employer to take "systemic and proactive" steps to prevent harassment by supervisors. Flaum Op. at p. 498. Judge Flaum also states (and I agree with his statement) that companies should "foster a culture in which harassment is not tolerated." Flaum Op. at p. 499. Unfortunately, this general statement fails to define or delineate his proposed standard, much less outline what employers must do in addition to the already widespread (and advisable) practice of adopting and publishing anti-harassment policies and instituting procedures for reporting harassment. If Judge Flaum's "heightened duty of care" standard were adopted, appellate courts across the country would soon find themselves in a quagmire, plagued by litigants who would (understandably) ask them to define and elaborate upon a standard that comports with neither traditional negligence nor strict liability. I am unable to join Judge Flaum's proposed modification of the traditional negligence standard for hostile work environment cases, and I am pleased to note that the court's per curiam opinion avoids the ambiguities of his "heightened duty of care" standard. I join with the per curiam opinion's straightforward holding that "the standard for employer liability in cases of hostile-environment sexual harassment by a supervisor is negligence, not strict liability...." ...." Per Curiam Op. at p. 495 (emphasis added).

4. Judge Cudahy's Approach to Employer Liability

Unfortunately, I do not find Judge Cudahy's reflections upon Judge Flaum's "heightened duty of care" standard particularly helpful; indeed, his analysis raises more questions than it answers. I will not address Judge Cudahy's suggestion that employers in supervisory hostile work environment cases should be held to the same standard as "packers of parachutes or open heart surgeons," Cudahy Op. at p. 504, for, in my view, neither heart surgery nor parachute-manufacturing are even remotely analogous to sexual harassment.

Judge Cudahy apparently believes that "vulnerable females" will not "feel free to report [sexual harassment] to the company," even in those firms and business enterprises that have a policy against sexual harassment and established procedures for reporting the same. Cudahy Op. at pp. 504-505. In fact, Judge Cudahy goes farther and asserts, for reasons unexplained, that we ought to presume that female victims of harassment "could not reasonably be expected to report their problems" using such grievance procedures. I disagree with his contention that female victims of harassment are always "vulnerable," nor should we assume that women (who have properly and successfully campaigned to be treated as equals in the workplace) are any less capable than men of sharing in the task of both monitoring and reporting sexual harassment. Furthermore, I am concerned that what Judge Cudahy may really be saying is that the existence of a sexual harassment policy (and reporting procedures) should have no bearing on employer liability in supervisory cases. An employer faced with such judicial commentary on the irrelevance of sexual harassment policies could well ask: Why establish, much less publish, a policy?

Another problem with Judge Cudahy's opinion is his assertion that whenever a sufficiently hostile or abusive environment exists (i.e., whenever the harassment is severe or pervasive enough to be actionable), there should be a legal presumption that the employer had "constructive notice" (i.e., "should have known") of the harassment. Cudahy Op. at p. 505. Here, I note that Judge Cudahy is improperly relying upon the reasoning set forth in the district court's opinion in Faragher, 864 F. Supp. 1552, 1563 (S.D.Fl.1994), which the Eleventh Circuit, on appeal, specifically and emphatically rejected in its en banc opinion. The Eleventh Circuit noted that:

[W]e do not agree with the district court's apparent belief that simply because conduct is pervasive enough to create an abusive work environment the employer should be charged with knowledge of the conduct. The question of notice to the employer is distinct from the question of the environment's abusiveness. Thus, the district court erred to the extent that it conflated the two inquiries.

Faragher, 111 F.3d at 1538. As recognized by the en banc court in Faragher, the traditional negligence standard (as opposed to the "heightened" version advocated by Judges Flaum and Cudahy) does, in fact, presently provide that an employer may be held liable if it knew or should have known about harassment in its workplace. In order that we might keep the two inquiries separate and thus avoid confusion, I agree with the Eleventh Circuit that the question of whether a specific employer, under a particular set of circumstances, "should have known" of harassment must remain separate and distinct from that of whether the harassment is of such a nature as to be actionable.18

5. The Need to Distinguish Quid Pro Quo and Hostile Work

I wish to note a final and very crucial area of disagreement with Judge Flaum's opinion as it relates to strict liability. As previously noted, Judge Flaum regards a quid pro quo as an essential prerequisite to imposing strict liability, and observes that "[D]efining an actionable quid pro quo ... is central to the liability standard." Flaum Op. at p. 499. However, neither of the cases before us, from a factual point of view, involved a true quid pro quo as previously defined in our case law.19 Judge Flaum has expanded the definition of quid pro quo harassment, possibly even beyond the bounds of reasonableness, in order to make the facts of Jansen come within the purview of a quid pro quo theory. Judge Wood, similarly, adopts an elastic approach to the definition of quid pro quo harassment, but this tactic is less crucial to her overall argument because, in her view, strict liability is the proper standard in any and all supervisory cases.

Quid pro quo harassment, as Judge Diane Wood noted in the very recent case of Bryson v. Chicago State University, "occurs in situations where submission to sexual demands is made a condition of tangible employment benefits." 96 F.3d 912, 915 (7th Cir.1996) (emphasis added). Harassment of the quid pro quo variety has been described as "the most oppressive and invidious type of workplace sexual harassment" because:

There can be no justification for requiring a worker to engage in sexual acts in order to obtain a job or job-related benefit, or to avoid a job-related detriment. Most workers subjected to sexual pressure in the workplace have little means of defense--other than the law. For economic reasons, most workers cannot simply abandon their employment--new jobs are hard to find.

Nichols, 42 F.3d at 510. The term quid pro quo literally means "something for something." Specifically, as Judge Rovner observed in Dey v. Colt Const. & Development Co., quid pro quo harassment involves an exchange that "link[s] an economic benefit to ... participation in conduct of a sexual nature." 28 F.3d 1446, 1453 (7th Cir.1994). In order to succeed under a quid pro quo theory, according to Judge Wood's opinion in Bryson, the plaintiff must demonstrate "what the 'quo' part of the quid pro quo was: [i.e.,] what tangible aspect of employment was affected." Bryson, 96 F.3d at 916 (emphasis added).20 In other words, a plaintiff must establish that she "has suffered a materially adverse employment action" for refusing to engage in unwelcome sexual conduct with the harasser. Id. (emphasis added). "Materially adverse employment action" can take many forms, but (as "materially" implies) such action must somehow affect a tangible, economic aspect of the plaintiff's job. To give but one example, this court did not hesitate to find a "materially adverse employment action" in the recent Bryson case, where the plaintiff, a university professor, alleged that she suffered the loss of her academic title and assignments to various administrative committees within the university after she refused to engage in sexual relations with a university official. Bryson, supra. We reasoned that both the title and the committee work contributed to the plaintiff's stature within the university administration and were "tangible employment benefits" because they "conferred prestige and [were] important to further professional advancement." We observed that "[u]niversities have few 'carrots' to dangle in front of tenured faculty members who reach full professorhood," and that the "reward structure" within a university often includes honorary or in-house titles and committee assignments, which can serve as the "building blocks" for a promotion within the university or the attainment of a more substantial position at another academic institution. Id.

While "materially adverse employment action" is a broad term, it is not without parameters and limitations established in case law. Specifically, "a supervisor's mere threat or promise of job-related harm or benefits in exchange for sexual favors does not constitute quid pro quo harassment." Gary v. Long, 59 F.3d 1391, 1396 (D.C.Cir.1995) (emphasis added). "[I]t takes more than saber rattling alone to impose quid pro quo liability on an employer; the supervisor must have wielded the authority entrusted to him to subject the victim to adverse job consequences as a result of her failure to submit to unwelcome sexual advances." Id. (emphasis added).

Hostile work environment harassment, by contrast with the quid pro quo variety, occurs "[w]hen the workplace is permeated with 'discriminatory intimidation, ridicule and insult,' that is 'sufficiently severe or pervasive to alter the conditions of the victim's employment and create an abusive working environment.' " Harris v. Forklift Sys., Inc., 510 U.S. 17, 21, 114 S. Ct. 367, 370, 126 L. Ed. 2d 295 (1993) (quoting Meritor, 477 U.S. at 65, 67, 106 S. Ct. at 2404-05, 2405-06). Although some of my fellow judges have proposed a dramatic expansion of employer liability under Title VII and have adopted a very elastic approach to quid pro quo harassment, I do not understand any of them to argue that we should abandon this Circuit's definition of what constitutes actionable hostile work environment harassment. Our decisions have emphasized that "isolated and/or trivial remarks of a sexual nature do not satisfy the definition of sexual harassment," Rennie v. Dalton, 3 F.3d 1100, 1107 (7th Cir.1993), cert. denied, 510 U.S. 1111, 114 S. Ct. 1054, 127 L. Ed. 2d 375 (1994), and that conduct is not actionable as sexual harassment if it is "too tepid or intermittent or equivocal to make a reasonable person believe that she has been discriminated against on the basis of sex." Galloway v. General Motors, 78 F.3d 1164, 1168 (7th Cir.1996) (holding that "safe harbor" exists for "low-level harassment."); see also Gleason v. Mesirow, 118 F.3d 1134, 1143-44 (7th Cir.1997). Hostile work environment harassment, like quid pro quo harassment, is a form of sex discrimination prohibited by Title VII, and obviously is not condoned under our present case law. But the two types of harassment are separate and distinct forms of discrimination and must not be co-mingled if we are to analyze and enforce sexual harassment law with coherence and clarity. In a hostile work environment case, unlike a quid pro quo case, the injury to the victim need not be directly linked to a tangible job benefit or detriment. See, e.g., Baskerville, 50 F.3d at 430 ("[t]he concept of [hostile work environment] sexual harassment is designed to protect working women from the kind of male attentions that can make the workplace hellish for women.").

Jansen claims that Antoni engaged in quid pro quo harassment when, in response to her inquiries about the late filing of her work performance review, he stated, "It's on my desk," while he performed a lewd gesture. Under the factual scenario presented in Jansen, I cannot agree with Judge Flaum that Antoni's alleged conduct is actionable as quid pro quo harassment. Flaum Op. at pp. 500, 503. The record reflects that Jansen was the secretary in the defendant company's tooling services department, and she worked primarily for Antoni, the manager of the department. Antoni exercised a measure of control over Jansen's working environment and was expected, as part of his duties, to contribute to the evaluation of the plaintiff's job performance. However, in his role as manager of the tooling services department, Antoni was not entrusted with the unilateral authority to hire or fire Jansen, nor was he the sole individual responsible for reviewing Jansen's performance.21 Contrast Highlander v. KFC Nat'l Management Co., 805 F.2d 644, 648 (6th Cir.1986) (would only impose strict liability for quid pro quo harassment if supervisory employee had "plenary authority over hiring, advancement, dismissal, and discipline" of the plaintiff). Since Antoni did not have final authority over personnel matters affecting the plaintiff, there is at least a legitimate question as to whether he was even in a position to engage in quid pro quo harassment of Jansen. In other words, I seriously question whether he ought to be classified as a "supervisor," and note again that we lack a clear definition of this crucial term to guide us in cases of quid pro quo sexual harassment.

Moreover, even assuming that Antoni could have subjected the plaintiff to quid pro quo harassment, the record is clear that he did not actually do so. Antoni never followed up his comment and/or gesture to the plaintiff with adverse job action, nor did Jansen ever comply with Antoni's alleged unwelcome sexual advances. As the district court made clear in granting summary judgment in favor of PCA on the quid pro quo claim, Jansen did receive her performance review and was given a satisfactory rating as well as a raise in salary. Jansen argues (and Judge Wood agrees) that the three-month delay in receiving her review/raise was itself an adverse job consequence, Wood Op. at p. 579, but a short delay in receiving a review/raise represents such a de minimis economic loss that it cannot reasonably or properly be interpreted as a "materially adverse employment action."22 Moreover, Jansen has failed to demonstrate a causal connection between her refusal to acquiesce in Antoni's alleged advances and the delay in her performance review. As Jansen well knew, it certainly was not an unusual occurrence for PCA to fail to conduct personnel performance reviews in a timely fashion. In fact, delayed reviews were so common that the company had in place a policy of making pay raises retroactive to the date when the review was due to be filed. Jansen's salary raise, although tardy, was made retroactive, in full accordance with this policy.

Jansen admitted, in her deposition testimony, that she neither acquiesced in Antoni's alleged demands nor suffered any tangible, adverse job consequences because of her failure to do so. Although Judges Wood and Flaum now would both hold that a mere threat of quid pro quo action is actionable as quid pro quo harassment, they have failed to cite any case law (nor am I aware of any) that would allow Jansen to proceed under a quid pro quo theory solely on the basis of unfulfilled threats (or perceived threats). See Gary, 59 F.3d at 1396 ("[I]t takes more than saber rattling alone to impose quid pro quo liability upon an employer."). One student of employment discrimination law has recently and properly summarized the law in this area, explaining that a quid pro quo claim:

requires that the victim either actually refuse the advances and suffer tangible job detriment or submit and retain a tangible job benefit. Quid pro quo theory appears not to encompass cases in which the supervisor threatens a job detriment but fails to carry through on his threat after the victim refuses to submit. There is thus no doctrine of attempted quid pro quo harassment; such unwelcome advances could only be challenged under the hostile work environment theory.

J. Hoult Verkerke, Notice Liability in Employment Discrimination Law, 81 Va. L. Rev. 273, 280 n. 15 (1995) (emphasis added).

Judge Flaum essentially maintains that there is a doctrine of "attempted quid pro quo harassment," and argues that his position "has a toehold in the case law." Flaum Op. at p. 499. He quotes language from three circuit court opinions for the proposition that a supervisor's unfulfilled quid pro quo threat may be deemed quid pro quo harassment: Nichols v. Frank, 42 F.3d 503 (9th Cir.1994); Karibian v. Columbia Univ., 14 F.3d 773 (2d Cir.1994); and Sparks v. Pilot Freight Carriers, Inc., 830 F.2d 1554 (11th Cir.1987). All three of these cases involved facts far more egregious than those set forth in Jansen. The plaintiffs in Nichols and Karibian, unlike Jansen, alleged that they actually acquiesced in their supervisors' demands for sexual favors in order that they might obtain or keep some kind of job-related benefit (in Nichols it was permission to take a leave of absence, in Karibian, the plaintiff maintained that her "raises, hours, autonomy and flexibility" were conditioned on her willingness to sleep with her supervisor). Sparks is also distinguishable from Jansen, for in Sparks the plaintiff alleged that she was discharged (a clear job-related detriment) "in retaliation for her refusal to accede to [her supervisor's] sexual demands." Id. at 1557.

I also find problematic Judge Flaum's assertion that Antoni's unfulfilled threat is actionable as quid pro quo harassment merely because it resulted (or allegedly resulted) in "real emotional strife" or "anxiety and distress" for the plaintiff.23 Flaum Op. at p. 500. Nor do I agree with Judge Diane Wood's speculation that because a plaintiff may experience both types of sex discrimination at the same time, the distinctions between quid pro quo and hostile work environment harassment (by now well-recognized in Title VII law) are somehow meaningless. Wood Op. at p. 567. Statements of this kind only serve to further cloud the law of sexual harassment by casting a haze over the well-recognized boundaries between quid pro quo harassment and hostile work environment harassment. Like the experienced trial judge in Jansen, 895 F. Supp. at 1066, we should follow the established and well-recognized definition of quid pro quo harassment in disposing of Jansen's claim, instead of merely assuming that there was a quid pro quo, or worse, twisting the definition of quid pro quo harassment to fit the facts of this particular case.

I also wish to emphasize that however objectionable Antoni's remark and gesture might have been, they do not, by themselves, rise to the level of actionable hostile work environment harassment as defined in our case law, for they did not "cross[ ] the line that separates vulgarity (not actionable) from harassment (potentially actionable)." Baskerville, 50 F.3d at 431. As this court has observed, "the concept of sexual harassment ... is not designed to purge the workplace of vulgarity." Id. at 430. Unfortunately, we do not live in the most cultured, virtuous or well-mannered time in world history, so that even in the workplace a certain amount of "vulgar banter, tinged with sexual innuendo," is inevitable today, particularly if one has the misfortune to work with "coarse or boorish" individuals who lack "refinement." Id.

In addition to arguing that a quid pro quo existed in Jansen, Judges Flaum and Wood claim that Ellerth also involved quid pro quo harassment. However, I agree with Chief Judge Posner that "Ellerth[,] like Jansen[,] is purely a hostile-environment case" and therefore must be analyzed under the negligence standard. Posner Op. at p. 516. The facts of Ellerth do not support the strained conclusion of my respected fellow Judges Flaum and Wood that quid pro quo harassment took place. Ellerth's contact with Slowik (who was based in New York, more than 500 miles away) included telephone conversations with her superior approximately once a week. During two conversations in early May 1994, the following exchanges allegedly took place (as summarized by the district judge):

[W]hen Ellerth was talking to Slowik to get special permission to do something for a customer, Slowik said something along the lines of "I don't have time for you right now, Kim, unless you're telling me--unless you want to tell me what you are wearing." Ellerth said she had to go and hung up. On a follow-up call, again to get permission, Slowik told Ellerth that she did not have permission for the project and he then said something along the lines of "are you wearing shorter skirts yet, Kim, because it would make your job a whole heck of a lot easier."

Ellerth v. Burlington Industries, Inc., 912 F. Supp. 1101, 1108 (N.D.Ill.1996).

Judge Wood, making the case for the existence of a quid pro quo, argues that "Slowik ... refused to give [Ellerth] special permission for work projects until she described her physical appearance to him." Wood Op. at p. 579. Referring to Slowik's question about whether Ellerth was attired in a long or a short skirt, Judge Wood seemingly embellishes the facts by asserting that Slowik forced Ellerth "to play the role of sex object in order to obtain desirable work assignments." Wood Op. at p. 579 (emphasis added). Once again, the definition of quid pro quo harassment is being stretched to fit the facts of this particular case. While asking a female employee to describe what she is wearing may be uncalled for and perhaps, if combined with other more egregious conduct, may give rise to a hostile work environment, such a comment is a far cry from asking a female employee to engage in sexual activity in order that she might keep her job or obtain a desired promotion. More importantly, Ellerth never did suffer an "adverse employment action" as a consequence of refusing to "play the role of sex object" (as Judge Wood refers to it). She was neither fired, denied a promotion,24 suffered a reduction in pay, nor did she lose a job title crucial to her advancement, much less was she transferred to another department with diminished responsibilities or fewer opportunities for promotion. Bryson, 96 F.3d at 916. Furthermore, Judge Wood has failed to persuade me that when Slowik denied Ellerth "special permission to do something for a customer," he was making "desirable work assignments" (i.e., assignments that would allow Ellerth to advance at Burlington) contingent upon willingness to go along with his sexual banter. I believe that Judge Flaum's and Judge Wood's journey in pursuit of a quid pro quo (on these facts) resembles a search for a diamond in a wagonload of hay. While they claim to have discovered the diamond (i.e., a quid pro quo), I have combed the record and can find nothing but the hay.

Judge Kanne has brought to our attention an additional and most important point concerning the dangers of expanding the definition of quid pro quo harassment (to include mere threats). He notes, in his opinion, that "quid pro quo threats may be ambiguous, and if such threats make an employer strictly liable, plaintiffs will attempt to turn all instances of supervisor sexual harassment into 'implied threats' in order to take advantage of strict liability's easier burden of proof." Kanne Op. at p. 505. A better approach would be to hold, consistent with our precedent, that plaintiffs such as Jansen and Ellerth may not proceed under a quid pro quo theory unless they can establish that they have actually suffered a tangible, job-related loss (such as being fired or denied a promotion) for refusing to go along with the harasser's sexual demands. Bryson, 96 F.3d at 916; Gary, 59 F.3d at 1396; Hicks, 833 F.2d at 1414.

Because our interpretations of congressional enactments such as Title VII are to serve as guideposts for the trial courts as well as for the legal profession as a whole (to say nothing of business and industry), it is crucial that we approach concepts such as quid pro quo harassment in a clear, logical fashion that adheres to precedent. The law of sexual harassment has become increasingly complex over the years, resulting in confusion and uncertainty as to what really constitutes sexual harassment in the eyes of the law. One observer has stated, without exaggeration, that "the web of sexual harassment [law] has truly become tangled in the mass of court opinions, agency guidelines, and law review articles on the subject." King, Sex, Love Letters, and Vicious Rumors, 9 B.Y.U. J. Pub.L. 341, 365 n. 5 (1995). The boundaries between "harassment" and normal interaction between the sexes are now so blurred that a six year-old first-grade student in Lexington, North Carolina who kissed one of his female classmates (also a mere six years of age) on the cheek, found himself suspended from school recently for engaging in "unwarranted and unwelcome touching" (i.e., "sexual harassment"). See Linda Chavez, Feminist Kiss Patrol is on the March, USA Today, Oct. 2, 1996, at 15A (common sense ultimately prevailed and the school lifted the suspension).

Mindful of the need to strike a balance between the rights of employers and employees, I am unable to join either Judge Flaum or Wood on their theory of what constitutes actionable quid pro quo harassment. By molding the facts of these cases, and attempting to make them fit within a newly-elusive quid pro quo theory, my colleagues are attempting to obliterate the very definition of quid pro quo harassment. This approach, should it become the law of this Circuit, will register a "7.5" on the Richter scale and shall certainly upset the balance between the rights of employers and employees while, at the same time, adding to the existing uncertainty and confusion in this complex, controversial, ever-expanding, and troubling area of the law.

6. Judge Posner's Approach to Employer Liability

Judge Posner has outlined a more limited approach to the question of employer liability, yet he still advocates--largely on policy grounds--that we venture part way into the realm of strict liability. He would only impose strict liability in a limited subset of quid pro quo cases, i.e., those where the supervisor successfully "us[es] his delegated authority to do a company act" as part of a quid pro quo; for example, by firing an employee, or denying a promotion/raise. Posner Op. at p. 512. Chief Judge Posner does not allow himself to fall into the trap of asserting, as do Judges Flaum and Wood, that a mere unfulfilled threat to take any of these actions is a sufficient basis for holding the company strictly liable. Posner Op. at p. 513. In fact, Judge Posner and I are in agreement that neither Jansen's supervisor nor Ellerth's engaged in a true quid pro quo, yet curiously Judge Posner theorizes at length as to why strict liability may be appropriate in some quid pro quo cases. Judge Posner apparently believes that if Jansen's supervisor had successfully blocked Jansen's promotion (because she refused to acquiesce in his sexual overtures), it would be proper to hold PCA strictly liable. I obviously agree with Judge Posner's rationale that the employer, knowing he is subject to strict liability, will have more of an incentive to monitor "very carefully" the personnel decisions of all of its supervisory employees. However, my colleague has failed to persuade me that this monitoring "should be relatively easy to do," or that it will bring about an effective system for the review and control of supervisory actions taken in the exercise of "delegated authority." Posner Op. at p. 512.

Initially, I wish to re-iterate that I am in agreement with Judge Posner that the questions before us are matters of federal common law and not state law. However, I do not agree with the Chief Judge that we are free to ignore traditional common law principles and "create" our own set of agency principles, guided solely by our personal views of what is sound policy in this area. Posner Op. at p. 510. Judge Posner refers to the common law principles set forth in the American Law Institute's Restatement (Second) of Agency as "antiquated screed," and throughout his opinion he relies heavily on policy arguments rather than on precedent. While I appreciate and always respect the Chief Judge's candor about his preferred approach, with all due respect, I submit that his method is inconsistent with our role as judges, and with the Supreme Court's directive in Meritor that we "look to agency principles" in this area (the Court even cited the Restatement as a source for such principles).

As noted at the outset, strict liability is based upon a policy judgment concerning the proper allocation of costs among various parties in society. At times, this court at times is called upon to exercise a policy-making function, especially when a federal statute is unclear or even silent as to a question of legal and social significance. As we fill the gaps in a federal statutory scheme such as Title VII, however, we should keep in mind the words of our distinguished colleague, Judge Jesse Eschbach, that we are "[w]riting ... on the shores of Lake Michigan rather than the banks of the Potomac...." Vail v. Board of Education, 706 F.2d 1435, 1445 (7th Cir.1983) (Eschbach, J., concurring). In exercising our very limited and circumscribed policy-making role, we must proceed with caution, and not lose sight of the fact that we are bound by the Constitution and by legal precedent. Above all, we must respect the doctrine of separation of powers, which is "essential to the preservation of liberty." Mistretta v. United States, 488 U.S. 361, 380, 109 S. Ct. 647, 659, 102 L. Ed. 2d 714 (1989). Under this doctrine, lawmaking authority is assigned to the elected branches of Government:

The Constitution wisely vests [the legislative] branch of the government with the power and authority of law-making, for the legislature is better equipped to carry out that task in that interested parties ... are able to present their respective positions before the legislative body in a more open, unrestricted, informal forum. Thereafter, the myriad of questions and problems, as well as their possible solutions, are brought before the entire legislative branch of government and are subject to the scrutiny of public hearings and debates.

Curtis v. Thompson, 840 F.2d 1291, 1305 (7th Cir.1988).

We should also remember that there are certain practical limitations on our capacity to address complex social problems. As Chesterfield Smith, a distinguished former president of the American Bar Association, once observed:

[C]ourts are being asked today to solve problems for which they are not institutionally equipped, or at least not as well equipped as other areas of government [such as the legislature].

As far as possible, judicial forums should be reserved for doing only that which cannot be done elsewhere.

The American public perceives the courts as a jack-of-all trades available to furnish the answers to whatever may trouble them. Shall a war be prosecuted or peace made? What is life, or when does death begin? Shall racial integration be achieved by ... busing of children to far away schools? How shall prisons and mental institutions be operated? Shall nuclear power plants be built, and if so, where? Shall the Concorde fly to these shores? Is affirmative action really inverse discrimination? Shall the snail-darter survive? [Should employers be held strictly liable for acts of sexual harassment by supervisors?] ...

The courts properly should be only the conflict-resolvers and not the problem-solvers of American society.

Wangen v. Ford Motor Co., 97 Wis. 2d 260, 324, 294 N.W.2d 437 (1980) (Coffey, J., dissenting) (refusing to join "judicial legislation" that would permit recovery of punitive damages in strict liability actions involving allegedly defective products). A decision issued by a panel of judges is at best a poor substitute for legislation enacted by a majority of the 100 Senators and 435 Representatives in the United States Senate and House respectively, after full hearings and debate, and signed into law by the President. The Congress, unlike this court, can act "with the benefit of citizen input ... [and] a trained staff to search out, weigh, balance, and comprehend" the far-reaching economic consequences of policy decisions of this nature. Id. at 324, 294 N.W.2d 437. "It is presumptuous for this court, which does not and cannot have the benefit of public hearings and constituent expression of opinion, to commend sua sponte any specific change [in legislation]," such as expanding Title VII to make employers strictly liable. Id. at 325, 294 N.W.2d 437 (quotation omitted). "If social welfare experimentation is to be conducted, it should be done by the legislature. The implications for the free enterprise system, and therefore the structure of our economy, are too disturbing to leave a decision of this magnitude to [a panel of] jurists." Id. at 331, 294 N.W.2d 437. As this court recognized in Curtis v. Thompson:

the fact-finding and policy-making capabilities of a court of law are far more limited and confined [than those of the legislature] due to the very being of a court of law, enshrined with its technical rules of evidence as to admissibility and materiality that often serve to limit a court's ability to make a clear, definitive, and thus reasoned pronouncement on matters of public policy.

840 F.2d at 1305. In carrying out our policymaking role, therefore, it is imperative that we "exercise [our] powers of review and decision making with discretion and reservation, giving due deference to the other branches of government." Id. Until Congress or the Supreme Court addresses more definitively the issue of employer liability under Title VII, we should approach the question cautiously, and not attempt to take the law beyond both the terms of the statute and existing case law by adopting a far-reaching standard of strict liability.

Judge Posner points out, and I agree, that the costs associated with a strict liability standard will ultimately and sadly be borne not only by the workers but also by the consumers, Posner Op. at p. 511; see also Excuse Factory, supra, at 296 ("workers can be expected to shoulder the bulk of the costs of a right to sue over things that go wrong in the workplace."). Nevertheless, having pointed out these costs, Judge Posner fails to take them into consideration when he proposes a strict liability standard that would mandate a highly centralized, potentially limitless, and expensive system for "supervising the supervisors." This system would require employers and CEOs to review any and all personnel and advancement related decisions with a jaundiced eye to ensure that they were not based on impermissible criteria. This kind of "monitoring" will not be "easy" in all (or even most) business settings. In fact, I am at a loss to understand how the CEO of any sizable company can be expected to monitor "very carefully" the personnel decisions of all of his company's supervisory employees, much less ascertain the subjective intentions lying behind them. I am also doubtful that supervision and training will succeed in eliminating the problem of workplace sexual harassment, whose origins are in social attitudes and beliefs that develop long before most people are old enough to draw their first paycheck.

Even if I were to agree with Judge Posner that multi-layered scrutiny of supervisory employees would prove workable in curtailing sexual harassment, this court is in no position to dictate management procedures for private businesses, nor should it impose costly legal requirements on these firms merely because we may individually be convinced that they are good policy. We judges are not "jacks of all trades," Wangen, 97 Wis. 2d at 324, 294 N.W.2d 437, much less sufficiently trained or experienced to be classified as experts on business management. Nor, as far as I am aware, have any of us successfully completed an M.B.A. program at a school of higher learning. We should refrain from tackling policy issues that lie beyond our province. Indeed, the Supreme Court has cautioned that "[c]ourts are generally less competent than employers to restructure business practices" and that "consequently, the judiciary should proceed with care before mandating that an employer must adopt" a particular business practice. Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 661, 109 S. Ct. 2115, 2127, 104 L. Ed. 2d 733 (1989) (superseded in other respects by the Civil Rights Act of 1991, Pub.L. No. 102-166 § 3, 105 Stat. 1071 (1991)). When an employer delegates the making of personnel decisions to its supervisors, and when the supervisor abuses that authority by engaging in quid pro quo harassment, fairness dictates that the employer should be judged in terms of what it knew about the supervisor at the time it delegated authority, and whether, in light of that knowledge, the delegation of authority was reasonable (i.e., negligence standard). Requiring more from industry (i.e., multi-layered bureaucratic structures to "monitor" supervisory employees "very carefully") ignores the realities and complexities of the modern business world and imposes an unreasonable and unworkable burden on employers.

D. Additional Concerns with Strict Liability

I am forced to disagree with our Chief Judge that expanding employer liability for sexual harassment represents but a "relatively esoteric and marginal change in the law." Posner Op. at p. 514. Making the employer strictly liable for the actions of an uncouth supervisor, without regard for any safeguards the employer may have put in place to prevent or deal with the harassment, will have deep and lasting consequences. The long-term problems that will result from a strict liability standard, in my view, far outweigh the dubious alleged gains associated with such a new and far-reaching standard. First of all, a strict liability standard will affect each and every facet of personnel decision-making, including, but not limited to, hiring, firing, and the employee review process, possibly bringing about a fundamental re-orientation of the employer-employee relationship. It will also have the effect of "chilling" social interaction in the workplace. Judge Posner notes, and I agree, that "romantic encounters, including romantic encounters between supervisors and supervised, are a fact of the workplace" and indeed, "[m]any happy marriages have grown out of such encounters." Posner Op. at p. 513. Today, however, as a result of the "constantly shifting" rules created by "evolving court opinions," the workplace is being transformed into "a nervous nest where people are afraid to say what they think and express honest emotions." Vincent J. Schodolski, Harassment Suits Curb Workplace Free Speech, Chi. Trib., June 23, 1997, at 1.

The detrimental impact of a strict liability standard will, by no means, be limited to social interaction in the workplace. We should consider, as an initial matter, the ramifications of strict employer liability for the federal judicial system, of which this court is an integral part. As a respected colleague of ours on the federal bench recently observed, "the federal courts are flooded with employment cases. We are becoming personnel czars of virtually every one of this nation's public and private institutions. The drafters of [Title VII] could never have intended the resulting consequences from what they deemed to be necessary, progressive legislation." Tschappat v. Reich, 957 F. Supp. 297, 299 (D.D.C.) (1997) (Sporkin, J.). Not surprisingly, since Congress amended Title VII in 1991 to permit the recovery of punitive damages (albeit only under limited circumstances),25 the number of employment cases filed in federal court has jumped from 10,771 (1992) to 23,152 (1996). Robert J. Samuelson, The Limits of the Law, Newsweek, Jun. 30, 1997, at 50. We should not kid ourselves: the adoption of strict liability by judicial fiat will likewise increase the volume of Title VII litigation and strain the resources of our already over-burdened federal court system. In recent years, Congress (in addition to legislating new civil causes of action) has created a significant number of new federal crimes, without conducting judicial impact studies to determine how such legislation will affect the justice system. See Kathleen F. Brickey, Criminal Mischief: The Federalization of American Criminal Law, 46 Hastings L.J. 1135 (1995). As Chief Justice Rehnquist has observed:

[I]n talking about the future of the federal courts, we must understand that Congress probably will continue to enact new legislation that provides new causes of action for litigants on the civil side of the docket and new federal crimes to be prosecuted on the criminal side of the docket. It is the federal district courts and the courts of appeals that are being hit hardest by this ever-increasing wave of litigation.... With the district courts, it is largely a question of having enough judicial manpower to adjudicate the incoming cases. The same is true to a large extent of the courts of appeals....

Keynote Address of Chief Justice William H. Rehnquist, Symposium: The Future of the Federal Courts (April 9, 1996), reprinted in 46 Am. U.L.Rev. 263, 265 (1996). In light of these developments, I am amazed that we would even consider expanding Title VII in a manner that will undoubtedly add to the already burdensome number of sexual harassment lawsuits that are brought before the trial courts and appealed to our own court.

The adoption of strict liability could also have significant, even worldwide, economic consequences, for it exposes American industry to a greater amount of Title VII litigation and thus adversely affects American business as it attempts to compete in an ever-expanding global marketplace. I join with thousands of American citizens who are of the opinion that employers today are over-regulated and saddled with an increasingly burdensome, indeed an overwhelming, amount of litigation, including Title VII litigation (much of which, as we realize, often is without merit). See Chrys A. Martin, Special Considerations in Sexual Harassment Claims, Federal Lawyer, Jul. 1996, at 35 ("Employment litigation is on the rise, particularly that involving claims of sexual harassment."). Indeed, the rising tide of employment-related litigation has prompted many firms to require--as a condition of employment--that employees agree to accept arbitration and forego their right to sue in court over employment-related disputes. Leslie Kaufman & Anne Underwood, Sign or Hit the Street, Newsweek, June 30, 1997, at 48.26

The clarion call of the social engineers of our day seems to be that courts should eradicate sexual harassment with instantaneous action and force employers to "pay the price" for alleged sexual misconduct through the imposition of a strict liability standard, whether or not they were even aware of such conduct. That clarion call may, in fact, be more akin to the first chime of the death knell for more and more American businesses and workers on this side of the globe. Unfortunately, we have been going down this road for some thirty years, as the ongoing project of contingency-fee lawyers has been to expand strict liability for deep-pocket defendants whenever possible. The expansion of products liability and punitive damages law has resulted in a windfall for contingency-fee lawyers who specialize in challenging deep-pocket corporate defendants, with the use of allegedly qualified "experts" who, in truth, courts have often found to be less than reliable, if not purveyors of so-called "junk science." All of this has created a tumultuous environment for American business, which is attempting to find its niche and compete in the global marketplace (thus generating American jobs) hampered by this albatross of court-created confusion over the scope of tort liability. See, e.g., Walter K. Olson, The Litigation Explosion (1991).

Through ill-advised decisions and judgments, our courts, egged on by activist groups, have created an almost hostile environment for business by permitting the recovery of crippling punitive damage awards and over-expanding doctrines such as strict liability for manufacturers of allegedly defective products. See, e.g., Wangen, discussed supra. In response to the general trend of making corporate "deep pockets" strictly liable (often for large punitive awards), many businesses have moved their operations (and many American jobs) out of the country (e.g., to Mexico) or even overseas,27 while others have been forced to close their doors or even file for bankruptcy protection.28 It is also well-documented that the added costs of insuring and defending against these lawsuits are passed along to the consumer. As noted recently by former United States Attorney General Richard Thornburgh, the "lawsuit tax" represents approximately $100 of the cost of a $200 football helmet, $20 of the cost of a $100 stepladder, $3,000 of the cost of a heart pacemaker, more than two-thirds of the cost of a childhood DPT vaccine, and as much as $500 of the cost associated with an average two-day maternity stay in the hospital. Richard Thornburgh, Lawsuit Loopiness: Congress, Lawyers Square Off Over Legal Reform, Harrisburg Patriot & Evening News, Dec. 28, 1995, at A11.

Numerous state legislatures have taken action to resolve the problems that have arisen since the establishment of punitive damages liability and its developing case law. The growing number of abusive and overzealous punitive damage awards--and the courts' reluctance to deal with the same--have prompted legislative reform efforts in states across the country. 1 Ghiardi & Kircher, Punitive Damages: Law & Practice § 6.36 (1997). Four states, Oregon, Georgia, Connecticut, and New Jersey, have enacted product liability statutes dealing specifically with the problem of burgeoning punitive damage awards. Id. at § 6.38. These four states have implemented a variety of strategies to deal with the problem, including the raising of standards for establishing liability, increasing the burdens of proof, and setting fault specific criteria to be considered in determining the size of awards. Id. On the other hand, the states of California, Colorado, Minnesota, Nevada, North Dakota, South Dakota, and Virginia, also in an attempt to address the issue, have enacted punitive damage statutes which likewise apply to products liability. Id. Many of these statutes contain elements of the Uniform Products Liability Act (UPLA), model legislation drafted by the Federal Interagency Department, while others have instituted caps on punitive awards. Id.29 Still other states, either by statute or by common-law, have seen fit to prohibit or sharply limit the recovery of punitive damages altogether.30

Concern over the adverse impact of large punitive awards has even prompted national efforts at reform. In 1996, for example, both houses of Congress passed the "Common Sense Product Liability Reform Act," which would, inter alia, have capped the punitive damages that a jury could award in cases involving faulty products. H.R. 956, 104th Cong. (2d Sess.). Although this reform legislation was vigorously opposed by the National Trial Lawyer's Association and eventually vetoed by the President, it nevertheless sent a strong message that a majority of Americans are very concerned about runaway judicial decisions in the products liability area, and the impact such decisions have on business, workers, and consumers. Wall St. J., May 3, 1996 at B11.31

Court-created liability in the area of medical malpractice (as in the products context) has served to increase dramatically both the size and frequency of malpractice awards and settlements, which in turn has driven up the cost of malpractice insurance premiums and contributed to the undesirable practice of "defensive medicine" (i.e., performing tests and procedures that are not medically necessary in order to safeguard against liability). See Theodore R. LeBlang, Medical Malpractice and Physician Accountability: Trends in the Courts and Legislative Responses, 3 Annals Health L. 105 (1994) (discussing recent judicial decisions that have broadened liability for malpractice, as well as the costs associated with expanded liability). I agree that a physician whose carelessness or mistake has caused injury should be held accountable under the law, and that justice requires that the victim be compensated. However, courts today are going well beyond such basic principles of traditional tort law. To cite but one example, it is increasingly common for the courts to permit a medical malpractice plaintiff to rely upon the doctrine of res ipsa loquitur in proving physician negligence, which, according to one critic, is nothing more than a "means of transforming the tort regime from a fault-based system to one of strict liability." Karyn K. Ablin, Res Ipsa Loquitur and Expert Opinion Evidence in Medical Malpractice Cases: Strange Bedfellows, 82 Va. L.Rev. 325 (1996) (emphasis added). Broad, judicially-created liability for malpractice, combined with the tremendous amount of paper-shuffling that is now required of doctors in order to satisfy mounting government regulations and requirements, has driven up the cost of healthcare and prompted many physicians to leave the practice. And, once again, who has suffered: the patient-consumer. Entangled in a legal morass created largely by ill-advised court and jury decisions, our nation's healthcare industry is now controlled from the insurance boardroom rather than the caregiver's office, and it is the patient who suffers under our "revolving-door" system of medical care (e.g., by being released early from the hospital as a cost-cutting measure). As with punitive damages in the products liability context, there has been a vigorous legislative response to skyrocketing malpractice awards in the form of statutory caps on non-economic damages in many states.32

Just as the ill-advised expansion of the law in the aforementioned areas has worked to the detriment of the average American citizen, so too will the expansion of employer liability in the Title VII context. By ignoring the Supreme Court's decision in Meritor and embarking on an ambitious judicial project to expand employer liability, we add to the already-onerous legal and regulatory burdens on American business, and send a message to employers that they will hereafter be considered as second-class citizens, whose "guilt" with respect to sexual harassment may henceforth be presumed, without even so much as a showing of knowledge much less fault. If history is any guide, many American businesses will respond to these developments by boosting prices for goods and services, or by transferring and re-locating their manufacturing operations across the waters, perhaps to countries in Asia or Central America where regulatory restrictions and the level of litigation generally are not as burdensome as in America. Or a firm might also choose to offset the increased costs associated with expanded Title VII liability by laying off workers and/or relying more upon machines and computers to do their jobs. One of my fellow judges even goes so far as to brush aside concerns about these potentially back-breaking costs, believing that costly prevention efforts will represent a corporate "investment in long-term freedom from liability." Flaum Op. at p. 499. However, as Judge Posner cogently explains, that "investment" is essentially a form of taxation that will be borne, not just by faceless corporate entities, but by all in society:

A law that requires the employer to do more than is feasible to control harassment will impose costs without creating deterrent benefits. In the long run, these costs will be borne largely by consumers, in the form of higher prices for the employer's product, and workers, in the form of lower wages (because the higher costs are labor costs). Many consumers and workers are women, so women, who are the principal victims of sexual harassment, will pay a big part of the costs that employers incur as a consequence of excessively harsh principles of employers liability.

Posner Op. at p. 511.

I agree that eliminating sexual harassment in the workplace is a worthy and commendable goal, but we must not live under the misapprehension that strict liability will somehow bring about a Utopian workplace, free of sexual harassment. Given their existing liability for negligence, employers already have an incentive to eliminate harassment, to say nothing of the fact that they might wish to eliminate such conduct for reasons independent of their potential liability under Title VII (e.g., because harassment is disruptive of the workplace and thus undermines productivity). Judges Posner and Flaum assume that strict liability will result in greater deterrence, but they fail to explain how this will happen. They effectively distill their argument into a cost-benefit analysis--that is, but for strict liability in quid pro quo cases, employers would presumptively prefer to incur the occasional cost of Title VII litigation rather than institute expensive measures to foster a "harassment-free" working environment. I am forced to disagree with this speculative theorizing, which is barren of support in empirical data.

In my view, the imposition of strict liability in the employment discrimination context represents a new and unwarranted extension of the doctrine of strict liability. The kind of cost-benefit reasoning that perhaps justifies strict liability for manufacturers of defective or faulty products is not convincing when applied to the employment arena. The manufacturer of a defective product may, and oftentimes does, profit from the sale of that product. When it does so profit, it should be held strictly liable once it establishes that a pre-determined fault or defect has caused harm. But obviously there is no gain to an employer when one of his supervisory employees engages in the sexual harassment of another employee.

In the products liability area, the imposition of strict liability is meant to impact on the economic decisions that manufacturers make about cold, tangible products made of metal, steel, iron, plastic and rubber, for "a manufacturer can anticipate certain hazards and 'guard against their recurrence' in a way that consumers cannot; similarly, consumers cannot investigate the 'soundness' of products." Marshall S. Shapo, 1 The Law of Products Liability § 7.05 (3d ed.1994). Manufactured products can be measured, analyzed, and tested by the manufacturer, and when a defect is identified, it can be remedied by replacing a dangerous machine or part with an improved or new design that is safer. The human "defects" in employees, by contrast, are largely hidden, elusive, and lie beyond the effective control of the employer. Only with great difficulty (if at all) can an employer measure and detect the human element that is "defective," i.e., the thoughts, feelings, and behavior of its employees or potential employees. Obviously, these "defects" are less susceptible to correction or "fixing" by the employer than is the case with defective products. Although Judge Posner argues that strict liability will result in "very careful" "monitoring" of supervisory employees, and thus a reduction in the amount of harassment in the workplace, I question whether such monitoring is feasible. Human behavior is influenced by myriad factors that lie hidden below the surface--beyond the reach of even today's surveillance cameras and x-ray machines. These factors include one's upbringing, moral beliefs, heredity, and environment. Additionally, as medical science is now discovering, the ebb and flow of chemicals within the human body (e.g., hormones and neurotransmitters) can and do profoundly affect behavior, sometimes resulting in depression, anxiety, and even anti-social acts such as sexual harassment. Substance abuse, which often goes undetected in the workplace, can upset the normal chemical balance within the human body, ofttimes contributes to behavior that is inappropriate, immoral, or even criminal in nature. Without prying deeply into the personal lives of its employees or potential employees, how is an employer supposed to have knowledge of, much less remedy, all of the factors that could possibly contribute to an employee's decision to engage in sexual harassment?

It is interesting to note that even under the existing negligence regime, an employer's attempts to avert liability by identifying and "weeding out" potential harassers in the workforce are often frustrated by state employment laws, which can interfere with the employer's ability to assess whether an individual is predisposed to engage in harassment (or other unlawful conduct). Following a case in which one supermarket employee was raped by another, a women's rights activist in Boston, Massachusetts (affiliated with a national organization) reportedly called for employers "to investigate the backgrounds of employees more thoroughly," opining that "[i]f an employer is not going to look out for the safety of employees, who is?" Walter Olson, "How Employers Are Forced to Hire Murderers and Other Felons," Wall St. J., June 18, 1997 at A23. Ironically, however, the kind of background investigation that might serve to protect other employees (and members of the public) is often prohibited by law, and can lead to the employer being sued for employment discrimination. Wisconsin law, for example, generally forbids an employer from asking job applicants or employees whether they have an arrest record. Wis. Stat. Ann. §§ 111.331, 111.32, and 111.335.33 Nor may a Wisconsin business deny employment to an individual because of a prior criminal conviction, unless the circumstances of the offense "substantially relate to the circumstances of the particular job." Wis. Stat. Ann. § 111.335.34 Whether an employee or prospective employee has a past arrest or conviction for, let us say, sexual assault, would be highly relevant to the individual's inclination to engage in sexual harassment or even rape, but in most instances such information is off limits to Wisconsin employers. Illinois, similarly, does not permit employers to use arrest information in making employment decisions. 775 ILCS 5/2-103.35 If strict liability is adopted as the standard in Title VII sexual harassment cases, the restrictions that bind employers in any states with laws similar to those of Wisconsin and Illinois will only be made more pronounced. Walter Olson, the author of a recent book aptly titled The Excuse Factory: How Employment Law is Paralyzing the American Economy (Free Press 1997), has described this bind or dilemma as yet another example of the "sued-if-you-do, sued-if-you-don't regime we impose on ... business." Olson, "How Employers Are Forced to Hire Murderers," Wall St. J., June 18, 1997 at A23. In the context of employer liability for sexual harassment, I ask: "Why should courts search for ways to tighten the noose around employers' necks (i.e., strict liability), and thus further 'paralyze' the American economy?"

We should recognize that "[m]en do not discard their personal qualities when they go to work," Hartford Accident & Indem. Co. v. Cardillo, 112 F.2d 11, 15 (D.C.Cir.1940), and that the vexing social problem of sexual harassment will very likely persist in the workplace, to some degree, regardless of the fact that an individual enterprise may devote considerable resources to its eradication. An individual's disposition to sexually harass, despite years of educational programs promoting gender sensitivity, as well as public awareness and the condemnation of harassment, is engendered in him long before he enters the workforce (indeed, our armed forces are currently struggling with this very problem, and finding it most difficult to root out deeply-seated social attitudes, despite their best efforts). Is it fair and equitable that we require business and industry (and their stockholders) to automatically bear the costs of remedying one of the more troublesome social problems of the present day; a problem that business and industry did not create, and that society as a whole has been unable to solve, notwithstanding tremendous effort for these many years? If any governmental body should adopt new law that would dramatically affect human interaction in the workplace, it should be Congress, which could act with the benefit of expert testimony, whether it be from human resource directors, social workers, psychologists, psychiatrists, or others with knowledge and experience in the behavioral sciences. New departures in the law of sexual harassment--if any--must be left to the thorough deliberation, study, hearings and decision of the legislative body, and not resolved by judicial fiat.

An employer should be held liable for true sexual harassment in the workplace if the employer had prior knowledge of the alleged harassment and failed to take appropriate and timely measures. However, in our litigious society, I fail to understand how this court can justify expanding sexual harassment liability beyond the traditional, tested, true, logical, and common-sense parameters of the negligence standard. Courts setting Title VII standards must proceed with caution and, above all, attempt to strike a balance between the rights, duties and obligations of employers and employees. Under a regime of strict liability, the unreported acts (or alleged acts) of even the lowliest supervisor would give rise to automatic liability for the employer. While the negligence standard is flexible and attempts to gauge the reasonableness of an employer's actions in a given set of factual circumstances, strict liability is a blunt legal instrument (a sledgehammer, if you will) that sexual harassment plaintiffs will wield to bludgeon employers for the improper conduct of their supervisory employees regardless of the circumstances. Delivering such a gift-wrapped weapon to Title VII plaintiffs would undoubtedly tip the very delicate balance of the scales of justice in favor of Title VII plaintiffs and against employers. Strict liability, in the context as set forth herein, amounts to overkill and fails to strike the balance we are obligated to continually strive to achieve if American industry and our never-ending search for justice are to co-exist and endure.

Strict liability in sexual harassment cases could also spill over into other Title VII contexts. For example, as the law in this Circuit currently stands, hostile work environment claims for harassment on the basis of race are analyzed pursuant to a negligence standard. See Daniels v. Essex Group, Inc., 937 F.2d 1264 (7th Cir.1991) ("Plaintiff must be able to prove that the employer knew or should have known of the harassment and failed to take prompt remedial action."). Insofar as Judge Wood explains that "[s]exual harassment is no different ... than other forms of discrimination," Wood Op. at p. 573, it is only a matter of time before we are implored to cast aside Daniels' negligence standard in favor of strict liability for racial harassment, so that businesses will be made to pay a price when racism rears its ugly and uncontrolled head in the workplace, unbeknownst to the employer, or perhaps in spite of the employer's best efforts to eliminate racism from the workplace.

I am of the opinion that in fairness to the American worker, consumer, and employer, we should reject the strict liability theories proposed by Judges Posner, Flaum, Easterbrook and Wood and hold employers liable in cases of true sexual harassment if and only when they knew or should have known of the harassment and failed to take proper remedial measures. In the employment context, liability should be based on fault, and the flexible negligence standard should be used to determine an employer's legal responsibility in Title VII sexual harassment cases, regardless of who allegedly engages in the harassment (supervisor vs. co-worker) or what type of harassment allegedly occurs (quid pro quo vs. hostile work environment). I fear that the approach advocated by some of my esteemed fellow judges will open the door to further interpretive problems and leave the law of sexual harassment even more inconsistent, complicated, and difficult to follow than it is today (indeed, speaking as a former trial judge, I question how trial judges will ever be able to make sense of and apply our disparate holdings in this en banc decision).

Jansen has introduced sufficient evidence of PCA's possible negligence in failing to protect Jansen from sexual harassment to warrant a trial on this issue, and summary judgment in favor of PCA should be reversed so that there can be a limited jury determination on the negligence question. As Judge Posner points out, the company did counsel Antoni about sexual harassment on at least two occasions prior to the alleged incidents involving Jansen. A jury will have to consider and weigh all of the facts and circumstances pertaining to PCA's alleged negligence, and a rational finder of fact could very well conclude, on the basis of the evidence in the record, that the company was on notice concerning Antoni's behavior and that its remedial measures (counseling) failed to prevent the problem from recurring in Jansen's case.

I reach the opposite conclusion with respect to Ellerth's case. As an initial matter, I concur with Judge Posner that Ellerth has waived her negligence claim, for Ellerth conceded, in her reply brief to the petition for rehearing, that § 219(2)(b) (negligence) "does not apply to this case." Posner Op. at p. 517. But even assuming that Ellerth's negligence claim has not been waived, summary judgment in favor of Burlington was proper and should be affirmed for the simple reason that the employer had no knowledge of the alleged harassment and was therefore not negligent. Judge Wood writes (and Judge Flaum agrees) that Ellerth's communications to "a number of Burlington employees" were somehow "sufficient to put the company on notice of the problem [with Slowik]." Wood Op. at p. 580. I disagree. Burlington's sexual harassment policy, which Ellerth received and read as part of the company's employee handbook, stated that the company would "not tolerate any form of sexual harassment in the workplace" and specifically directed employees to report their concerns about harassment to either (a) their supervisor or (b) the human resources department. Additionally, the employee handbook set forth the company's "open door" policy, which stated that if employees had concerns they did not feel comfortable discussing with their supervisors, they should raise those concerns with the human resources department.36 Although Ellerth claims to have mentioned her problems with Slowik to some co-workers, and to a manager in another division of the company, she admits that none of these individuals were her superiors at Burlington. In short, Ellerth failed to follow the procedures outlined in Burlington's sexual harassment policy by reporting Slowik's conduct to anyone at Burlington who was authorized to deal with such concerns, or to anyone designated to deal with such complaints by the sexual harassment policy. She failed to contact either (a) her immediate supervisor (Lawrence) or (b) the Human Resources Department. I agree wholeheartedly with Judge Posner that with rights should come responsibilities, including, in Ellerth's case, the responsibility "to complain about Slowik through proper channels," i.e., by using the procedures spelled out in the company's published policy against sexual harassment. Posner Op. at pp. 516-517.

Allegedly, Slowik told a number of off-color jokes (surprisingly, Ellerth cannot recall the content of even one of them) and allegedly made comments about Ellerth's legs, all in the presence of a Vice President of Burlington at a business luncheon. However, even assuming that these allegations are true, they are insufficient to establish that the company was on notice of sexual harassment, for, as Judge Posner notes, "the vice-president had no reason to think Ellerth offended by Slowik's banter; not all women would be." Posner Op. at p. 516. Indeed, one has to question whether Ellerth found the jokes offensive, for she cannot recall the content of even one of them.

II. JANSEN & "APPARENT AUTHORITY"

The discussion of Jansen's "apparent authority" claim in the Court's per curiam opinion and in the separate opinions of my fellow judges37 fails to provide adequate guidance to judges (including the trial judge on remand), lawyers, and affected parties. Although both Judge Wood and Judge Flaum discuss "apparent authority" in the abstract, it is unclear whether they conclude that Jansen has come forward with evidence sufficient to survive summary judgment on this theory. On the other hand, from my reading of the record, I am convinced that Jansen has failed to meet her burden of demonstrating that PCA, through its acts and omissions, created an appearance that Antoni was authorized to engage in acts of sexual harassment.

Apparent authority is "such authority as a principal intentionally or by want of ordinary care causes or allows [a] third person to believe that [his] agent possesses." Black's Law Dictionary 96 (6th ed.1991) (citations omitted); see also Restatement (2d) Agency § 219(2)(d). As Judge Flaum correctly notes, "apparent authority exists only 'to the extent that it is reasonable for the third person dealing with the agent to believe that the agent is authorized' and the third person actually believes the agent to be authorized." Flaum Op. at p. 500 (quoting Restatement (2d) Agency § 8 cmt. c)(emphasis added). When analyzing an apparent authority claim, the proper focus must be on the perceptions of the third party (in this case the plaintiff Jansen) and whether those perceptions were reasonable. One scholar has observed that "[i]n the sexual harassment context, the principal's actions must have given the victim a reasonable belief that the perpetrator of the harassment had authority to harass. For this reason, apparent authority seems an unlikely basis for employer sexual harassment liability." Phillips, Employer Sexual Harassment Liability, 44 Vand. L. Rev. at 1247-48.

Judge Wood co-mingles her discussion of "apparent authority," a concept set forth in Restatement § 219(2)(d), with her analysis of vicarious liability under § 219(1). Moreover, Judge Wood's interpretation of "apparent authority," like her analysis of respondeat superior liability, is far too broad. According to Judge Wood, who again reaches out to expand those concepts set forth in the Restatement, we should not ask whether the supervisor had apparent authority to harass, but whether he had the apparent authority to "hire, assign, supervise, etc." Wood Op. at p. 575. If the supervisor appeared to have such authority and engaged in harassment, then according to Judge Wood, "liability [for the employer] should result." Wood Op. at p. 575. It seems to me that the question Judge Wood would have a jury answer is far too broad, for the mere fact that an employee occupies a supervisory role, albeit perhaps even a very limited one, to some degree does not suffice to establish liability under an "apparent authority" theory. Rather, as the trial judge in Jansen explained:

liability is based upon the fact that the agent's position facilitates the consummation of the [tort], in that from the point of view of the third person the transaction seems regular on its face and the agent appears to be acting in the ordinary course of the business confided to him.

Jansen, 895 F. Supp. at 1066 (quoting Restatement (2d) Agency § 261 cmt. a) (emphasis added).

I have been unable to discover any evidence in the Jansen record from which a rational jury could conclude that Antoni, however improper his conduct might have been, "appear[ed] to be acting in the ordinary course of ... business" when he allegedly engaged in the acts of sexual harassment. Jansen claims that she was aware of various unsubstantiated rumors to the effect that: (1) Vicki Wiley left PCA because Antoni had "give[n][her] a hard time"; and (2) PCA tolerated workplace affairs and possibly even sexual liaisons (including one or two that supposedly involved Antoni), on the premises. Mere office gossip and banter falls far short of establishing that PCA either condoned or licensed sexual harassment, nor would unsubstantiated rumors permit a reasonable person to conclude that such harassment was part of the "ordinary course of business" at PCA. Although the plaintiff self-servingly claims that she was unaware of PCA's sexual harassment policy, the deposition testimony of Henry Weil (Director of Operations) and Antoni makes clear that the policy was posted conspicuously in at least one, if not two, locations prior to Jansen's complaint, including a bulletin board on the first floor near the entrance to the plant, which Jansen admitted passing "every single day." PCA's policy--which was explicit, detailed, in writing, and posted where employees could see it--made clear that harassment would not be tolerated at PCA and encouraged employees to report any and all instances of such harassment to "appropriate management." A rational jury, on the basis of this record, could not find that PCA failed to post its anti-harassment policy conspicuously enough, nor could it conclude that the plaintiff had a reasonable belief that PCA authorized, condoned, or sanctioned sexual harassment. Therefore, although I see merit in allowing Jansen's negligence claim to go forward, I do not believe that her "apparent authority" claim should proceed to trial because the facts of her case are plainly insufficient to support such a claim.

As the experienced trial judge recognized, the weak evidence adduced by the plaintiff to support her "apparent authority" argument is more than overwhelmed by evidence that Jansen knew Antoni "was operating outside the scope of his apparent authority in harassing her." Jansen, 895 F. Supp. at 1067. "If a person has information which would lead a reasonable man to believe that the agent is violating the orders of the principal or that the principal would not wish the agent to act under the circumstances known to the agent, he cannot subject the principal to liability." Restatement (2d) Agency § 166 cmt. a (quoted in Jansen, 895 F. Supp. at 1066). Thus, Jansen's knowledge or awareness that Antoni was operating outside the scope of his employment serves to defeat her "apparent authority" argument. In addition to the posted anti-harassment policy,38 there is further evidence of Jansen's awareness that sexual harassment was not authorized at PCA, for when Jansen did make up her mind to complain about Antoni (in mid-January 1993, at which time the alleged harassment had been going on for at least 15 months), she had no difficulty whatsoever in knowing how to proceed. Without hesitation, she contacted the proper individual at PCA (Human Resources Director Paul Migala), who took prompt action to address the situation. As the district court noted, the existence of a mechanism for dealing with sexual harassment not only serves to negate any finding of negligence on the part of the employer, it also quite obviously "removes the guise of apparent authority from the harassing supervisor." Jansen, 895 F. Supp. at 1067. In other words, because "Jansen plainly knew that she had some recourse at [PCA] ... she cannot now claim that she believed Antoni's conduct to be sanctioned by [PCA]." Id. The evidence in the record, considered as a whole, falls short of establishing that Jansen had reason to believe that PCA authorized, condoned, or sanctioned acts of sexual harassment. Thus, with respect to Jansen's apparent authority claim, we should affirm the trial judge's grant of summary judgment in favor of PCA.

"Apparent authority" is a legitimate "agency principle," see Restatement § 219(2)(d), but if it is interpreted too broadly, or applied without regard to the facts, it becomes the functional equivalent of strict liability. I agree with Judge Manion that "it is hard to imagine a scenario where an employee reasonably believes her supervisor's job description includes the right to sexually harass her." Manion Op. at p. 563 (emphasis added). This is particularly true in cases such as Jansen, where the employer had promulgated and publicized a clear policy forbidding sexual harassment in the workplace. See Phillips, Employer Sexual Harassment Liability, 44 Vand. L. Rev. at 1242-43 ("Claims that a supervisor had ... apparent authority to harass almost certainly will fail if, as is increasingly common today, the employer maintains an anti-discrimination or anti-harassment policy."); see also Manion Op. at p. 562 n. 4 (and cases cited therein).

In evaluating Judge Wood's and Judge Flaum's analyses of what constitutes quid pro quo harassment, I observed above that neither of the cases before us involved a true quid pro quo as defined in our case law. However, even if Jansen and Ellerth did suffer from quid pro quo harassment, they have waived their quid pro quo claims because they failed to raise them in their respective EEOC complaints or district court pleadings. In short, I disagree with my fellow judges, who either assume (or in Judge Wood's case, explicitly argue) that Jansen and Ellerth have preserved their claims of quid pro quo harassment for appellate review. Wood Op. at pp. 578-579.

Jansen did not raise a quid pro quo allegation until almost the overtime period in the lawsuit, more than a year after filing her complaint, when she submitted a memorandum and supporting affidavit in opposition to PCA's motion for summary judgment. She had not previously alleged that Antoni suggested any kind of explicit or implicit sexual quid pro quo in her report to PCA, her filing with the Equal Employment Opportunity Commission ("EEOC"), or her complaint filed with the district court. Nor was this allegation ever listed in Jansen's detailed and exhaustive handwritten "log" (allegedly chronicling Antoni's harassment of her during the fifteen-month period prior to her complaint). In her deposition, Jansen not only failed to make such allegations, she went so far as to concede that her performance review was "[n]either discriminatory nor retaliatory" (Jansen received a satisfactory evaluation and a raise in salary, retroactive to the time when the review was due). Jansen further testified, in her deposition, that she reported all alleged acts of sexual harassment to PCA and to the EEOC.

Jansen's belated assertions of quid pro quo harassment run afoul of well-established and important procedural rules, which, for reasons unexplained, the court has chosen to ignore. Title VII plaintiffs who wish to bring a claim in federal court must initially set forth their claim in a filing with the EEOC, or they are barred from pursuing the claim. Cheek v. Western and Southern Life Ins. Co., 31 F.3d 497, 500 (7th Cir.1994). This rule serves to "giv[e] the employer some warning of the conduct about which the employee is aggrieved," Id. (citations omitted), and permits the EEOC to conduct an independent investigation in order that it may assist the parties in determining whether court proceedings are necessary. 42 U.S.C. § 2000e-5(b). Jansen's EEOC filing made no allegation of quid pro quo harassment, nor did it set forth any facts that would support such a claim.

A civil plaintiff is barred from injecting new allegations into his or her case by raising them after the defendant has filed a motion for summary judgment, for such a strategy fails to give adequate or fair warning as to the claims the defendant must address and against which it must defend. Andree v. Ashland Co., 818 F.2d 1306, 1314 n. 11 (7th Cir.1987) (district court properly excluded a claim made for the first time in opposition to summary judgment). Chief Judge Posner, in an opinion joined by Judge Rovner, very recently noted that this court has been "critical of efforts to patch up a party's deposition with [her] own subsequent affidavit." Russell v. Acme-Evans Co., 51 F.3d 64, 67 (7th Cir.1995). Similarly, in an opinion authored by Judge Kanne and joined by Judge Flaum, this court has unequivocally stated that "parties cannot thwart the purpose of Rule 56 [summary judgment] by creating issues of fact through affidavits that contradict their own depositions." Darnell v. Target Stores, 16 F.3d 174, 177 (7th Cir.1994) (quoting Miller v. A.H. Robins Co., Inc., 766 F.2d 1102, 1104 (7th Cir.1985)). By raising allegations of quid pro quo harassment in her affidavit (after failing to raise them in her EEOC complaint and after conceding in her deposition that her performance review was not discriminatory), Jansen clearly attempted to use her affidavit for the improper purpose of "retract[ing] or explain[ing] away concessions that [she made] in [her] deposition." Russell, 51 F.3d at 67. Which of Jansen's three statements are we to believe: her EEOC filing (submitted in early 1993, shortly after the harassment allegedly occurred), her sworn deposition testimony, or her sworn affidavit?

According to Seventh Circuit case law, because Jansen failed to raise her quid pro quo harassment claims in a timely fashion, she has waived those claims on appeal. I am at a loss to understand how this court can ignore its own precedents and rely upon allegations that were, in all likelihood, created and conjured up by the plaintiff "at the last minute" in order to bolster her case. See Russell, 51 F.3d at 67 (observing that "[a]lmost all affidavits submitted in litigation are drafted by the lawyers rather than by the affiants."); Darnell, 16 F.3d at 177; Andree, 818 F.2d at 1314.

Many of my observations concerning procedure in Jansen apply with equal force to Ellerth, who also neglected to raise a claim of quid pro quo harassment in either her EEOC filing or her district court complaint. Until today, we have followed the principle, as set forth by Judge Rovner in Saxton v. American Telephone & Telegraph, 10 F.3d 526, 533 (7th Cir.1993), that when a Title VII plaintiff "first attempt[s] to articulate a basis for a quid pro quo claim in her appellate briefs ... that, of course, is too late." Id. (declining to reach the merits of plaintiff's quid pro quo claim because she relied "solely" upon a hostile work environment theory at the district court level).

What lessons are future litigants to derive from the court's abandonment of well-established procedural rules in the two cases before us today? Is the message that rules about waiver no longer exist? We would be well-advised not to cast aside the procedural rules in either Jansen's or Ellerth's case for a cause and also for the purpose of making the road wider and easier for future Title VII plaintiffs, nor should we use the cases before us as a vehicle for announcing a new strict liability standard that is specific to quid pro quo cases. This kind of approach will only serve to undermine respect for our precedents and encourage other plaintiffs (and their lawyers) to flout important procedural rules, to the detriment of the judicial process.

I concur with the majority that the district court's dismissal of Jansen's claims of retaliation and intentional infliction of emotional distress was proper, and I also agree that Jansen should be remanded in order that a jury may address the limited question of whether PCA was negligent in preventing Antoni's alleged hostile work environment harassment. However, as discussed, I am of the opinion that Jansen has waived her quid pro quo claim and do not believe that Jansen has come forward with evidence sufficient to survive summary judgment on her apparent authority claim.

Most importantly, I am not persuaded that this court should do an "end run" around the Supreme Court's decision in Meritor by adopting strict liability in quid pro quo cases, as proposed by Judges Posner, Flaum, Easterbrook and Wood. Factually, neither of the cases before us involved true quid pro quo harassment, and indeed, the plaintiffs have waived any such claims. Therefore, even if I were to agree that strict liability is the proper standard in quid pro quo cases (the common thread which unites Judges Posner, Flaum and Wood), the two companion cases before us are hardly the proper vehicle for announcing such a rule.

In the more than thirty years since Congress outlawed sex discrimination in the workplace (via a "last minute" amendment to Title VII), our nation's legislature has been silent concerning the meaning and scope of sex discrimination, and it has not, to date, enacted a statutory definition of sexual harassment. This congressional silence, possibly attributable to pressure from special interest groups, has left a vacuum to be filled by the courts, and as the opinions of judges across the land unfortunately demonstrate, a number of courts have been all too willing to interpret sexual harassment law expansively. The current state of affairs--legislative silence coupled with energetic judicial "gap-filling"--has done violence to the Framers' intent that lawmaking authority be vested in the legislative branch, and not the judiciary. It is my hope that Congress will address this issue and relieve the overburdened court system of the present and anticipated influx of oft-times unfounded sexual harassment litigation. Unlike a court of law, the Congress may investigate, hold hearings, and receive input from a wide range of experts, as well as those who are likely to be affected by changes in sexual harassment law. Congress, in the exercise of its wisdom and judgment, might provide for alternatives to strict liability that certainly would not be quite as far-reaching as strict liability. For example, Congress could provide more extensive remedies (e.g., double or treble damages) in cases where it could be established that (a) the plaintiff suffered true quid pro quo harassment, and (b) the employer knew of the harassment but failed to take appropriate remedial measures. Such an approach would retain the well-accepted and tested negligence standard, but increase the price tag for employer negligence with respect to quid pro quo harassment. This is but one example of how the nation's constitutionally designated lawmaking body might choose to deal with the problem of quid pro quo harassment, while still allowing employers to defend themselves by demonstrating that they lacked knowledge of the harassment and/or took reasonable measures to prevent the same.

In closing, I observe that "sexual harassment" is in the eyes of the beholder. One person's idea of friendly banter is another's idea of harassment. In the highly-politicized environment surrounding the volatile issue of sexual harassment in the workplace, neither employers nor employees have a clear understanding of what conduct the law allows, much less what it prohibits. Congress, as noted, has been silent. And despite a profusion of reported cases on sexual harassment, the judicial branch has failed to reach consensus about the meaning and extent of Title VII's prohibition against sexual harassment. Indeed, the divergence in opinion across our fellow federal circuits, as well as the eight separate opinions submitted by members of this court today, are powerful proof that the law in this area is unclear and unsettled. As it currently stands, the law of sexual harassment "tend[s] to avoid giving employers definite rules to obey but instead lay[s] out sweeping if vague aspirations that are given the force of law.... No one really knows where these concepts begin and leave off; all employers know is that if they guess wrong some future jury or judge may decide that they have broken the law." Olson, Excuse Factory at 3. In light of this uncertainty and confusion, I believe it is particularly unfair and most unreasonable to hold an employer automatically (strictly) liable for the alleged missteps of an alleged "supervisor," irrespective of the employer's lack of knowledge concerning the "harassment" and regardless of his efforts to discourage harassment in the workplace. Strict liability tips the time-honored and respected scales of justice in favor of the plaintiff.

EASTERBROOK, Circuit Judge, concurring in part and dissenting in part.

Chief Judge Posner's approach to this subject has many virtues, simplicity and the creation of desirable incentives prominent among them. I would join his opinion if I thought that federal judges were free to devise a law of vicarious liability for Title VII cases. Yet we are not. Title VII creates a national anti-discrimination rule but does not deal with all discrimination in the labor force. Only employers and their agents are covered. 42 U.S.C. § 2000e(b). Who speaks or acts for the employer? Who is an agent? Title VII does not say. It does not even hint. The answer must come from elsewhere. When a federal statute is silent, we obtain the necessary rule from state law, unless application of state law would undermine a federal norm. Atherton v. FDIC, --- U.S. ----, ----, 117 S. Ct. 666, 670, 136 L. Ed. 2d 656 (1997) (collecting cases); Turner/Ozanne v. Hyman/Power, 111 F.3d 1312 (7th Cir.1997).

Meritor Savings Bank, fsb v. Vinson, 477 U.S. 57, 72, 106 S. Ct. 2399, 2408, 91 L. Ed. 2d 49 (1986), directs the inferior courts "to look to agency principles for guidance in this area." "Agency principles" come from state law, as the Court implied by citing the Restatement of Agency, a distillation of decisions by state courts. "Agency principles" answer the question: who can make commitments, and whose acts can create liability, for a corporation? Like other questions about a corporation's organization and distribution of powers, and about the liability of its investors (who really pay judgments against corporations), agency depends on state law, for it is state law under which corporations exist. Corporations are complex webs of contracts, themselves specialties of state law. Thus Meritor tells us to "look to," not "make up," agency principles. One can "look to" principles only in an existing body of law, which means state law, for there is no free-floating common law. Erie R.R. v. Tompkins, 304 U.S. 64, 58 S. Ct. 817, 82 L. Ed. 1188 (1938).

Relations such as agency that are undefined by federal statute, like other elements of the background against which federal rules operate, come from state law--either directly, when the Rules of Decision Act, 28 U.S.C. § 1652, requires, or indirectly when federal law absorbs a needed rule from state law. (The difference between direct and indirect use of state law does not matter to these cases.) Although use of state law to determine the extent of vicarious liability entails disparity across state borders, even within a single firm, this is no more an objection than it would be for other torts or contracts. Suppose Slowik entered into a contract with Ellerth raising her salary, or made a defamatory statement about her (as she says he did, when he implied in conversation that she had capitulated to his demands). State law would determine whether the raise stuck (or whether someone higher up had the final word), and whether the slander was within the scope of employment--which is to say, state law determines who is the firm's agent, and for what purposes. Who speaks or acts on behalf of a corporation is a fundamental question of corporate organization, all but invariably resolved under state law, as Atherton shows even for firms chartered by the United States.

Absence of a national rule means that firms must answer for supervisors' acts in some states but not others. Many of my colleagues object to the possibility of variation, but if different treatment of the same facts in different states authorized the creation of federal common law, Swift v. Tyson, 41 U.S. 1, 16 Pet. 1, 10 L. Ed. 865 (1842), would hold sway today, United States v. Kimbell Foods, Inc., 440 U.S. 715, 99 S. Ct. 1448, 59 L. Ed. 2d 711 (1979), would have come out the other way, and a host of cases from the last 20 years using state law to decide subsidiary, but dispositive, questions in litigation under federal statutes would be overthrown. E.g., O'Melveny & Myers v. FDIC, 512 U.S. 79, 114 S. Ct. 2048, 129 L. Ed. 2d 67 (1994); Kamen v. Kemper Financial Services, Inc., 500 U.S. 90, 111 S. Ct. 1711, 114 L. Ed. 2d 152 (1991); Wilson v. Garcia, 471 U.S. 261, 105 S. Ct. 1938, 85 L. Ed. 2d 254 (1985); Marrese v. American Academy of Orthopaedic Surgeons, 470 U.S. 373, 105 S. Ct. 1327, 84 L. Ed. 2d 274 (1985); Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U.S. 630, 101 S. Ct. 2061, 68 L. Ed. 2d 500 (1981); Robertson v. Wegmann, 436 U.S. 584, 98 S. Ct. 1991, 56 L. Ed. 2d 554 (1978); Miree v. DeKalb County, 433 U.S. 25, 97 S. Ct. 2490, 53 L. Ed. 2d 557 (1977). Reverse the uniformity question and ask: Why should an act of sexual harassment by a supervisor be attributed to the firm under state law but not under Title VII (or under Title VII but not state law)? Federal common law achieves horizontal but not vertical uniformity.

Title VII is no different, in needing the assistance of state-law principles, from the antitrust, aviation, banking, securities, and other civil rights laws, the subjects of the seven cases just cited. O'Melveny and Kamen in particular use principles of state law to address questions about responsibility of a firm for acts of employees or agents, although in both cases liability would be based on national law.1 Employment is a contractual relation, see Hishon v. King & Spalding, 467 U.S. 69, 74, 104 S. Ct. 2229, 2233, 81 L. Ed. 2d 59 (1984), which implies that the line between an employee (covered by Title VII) and a partner or independent contractor (not covered) comes from state contract law. Cf. Ost v. West Suburban Travelers Limousine, Inc., 88 F.3d 435 (7th Cir.1996) (using common law principles to distinguish employees from independent contractors, but without discussing choice of law). The question whether a person can bind the employer by contract--for example, by settling litigation under Title VII--similarly depends on state law. Morgan v. South Bend Community School Corp., 797 F.2d 471 (7th Cir.1986). Cf. Kokkonen v. Guardian Life Insurance Co., 511 U.S. 375, 114 S. Ct. 1673, 128 L. Ed. 2d 391 (1994). In Peacock v. Thomas, 516 U.S. 349, 116 S. Ct. 862, 133 L. Ed. 2d 817 (1996), the Supreme Court held that state law determines when an investor is vicariously liable for a federal-law judgment against the corporation. These principles likewise point to state law to determine when the corporation is vicariously liable for an employee's acts.

None of the parties argues that federal statutory law specifies the extent of an employee's ability to bind the employer. Federal law requires complainants to file their grievances initially with state agencies, 42 U.S.C. § 2000e-5(c), the better to coordinate application of national and local anti-discrimination rules. Why undercut coordination by adopting for purposes of Title VII a set of agency rules that differs from those the state would apply to the same claim of discrimination? Preserving vertical uniformity is a strong reason not to have federal rules. Although the parties have briefed these cases on the assumption that federal law governs, we are entitled to disregard that forfeiture and to decide independently which body of law supplies the essential principles. Kamen, 500 U.S. at 99, 111 S.Ct. at 1717-18. See also United States National Bank of Oregon v. Independent Insurance Agents of America, Inc., 508 U.S. 439, 445-48, 113 S. Ct. 2173, 2177-79, 124 L. Ed. 2d 402 (1993).

State courts devising principles of vicarious responsibility are not stuck in the 1940s or condemned to use irrelevant analogies. Corporate law, actively monitored and adjusted by state courts, is one source of rules for determining when a manager's acts are attributed to the firm. Many states are developing distinctive bodies of law about workplace harassment, sometimes holding employers liable for acts perpetrated by supervisors. Some of these cases have been decided under state or local anti-discrimination laws. State courts may interpret their usual agency principles when adjudicating claims of discrimination, just as we have been invited to do, or they may devise a specialized body of agency doctrine for discrimination cases, as several of my colleagues prefer. Green Hills Country Club v. Illinois Human Rights Commission, 162 Ill.App.3d 216, 113 Ill. Dec. 216, 514 N.E.2d 1227 (5th Dist.1987), shows that Illinois holds a firm answerable for the kinds of supervisory conduct involved in our cases. A statute dealing with vicarious liability in discrimination cases supports its conclusion. 775 ILCS 5/2-102(D). State statutes and judicial decisions alike supply the law that we should use. (Recall what Erie held: that statutory and common law are treated the same under the Rules of Decision Act.) Illinois would hold both employers vicariously responsible for the conduct of Ellerth's and Jansen's supervisors.2 Federal law then makes that conduct actionable.

Illinois does not draw a line, for purposes of vicarious liability, between quid pro quo demands and hostile work environments. We should not superimpose such a distinction on a body of agency law that does not require it. Not unless Title VII itself requires that distinction, and it does not. Title VII forbids sex discrimination in employment. Insistence on performance of sexual services as a condition of receiving the pay or position to which a person is otherwise entitled is sex discrimination; so is the maintenance of an environment in which women are made to feel unwelcome or miserable, when men performing the same tasks are treated well. See Harris v. Forklift Systems, Inc., 510 U.S. 17, 114 S. Ct. 367, 126 L. Ed. 2d 295 (1993). Recognition that there are multiple ways to discriminate, in violation of federal law, does not imply different rules of vicarious liability.

Distinctively federal rules of vicarious liability in discrimination cases would be justified if application of the state's law of agency undermined a federal interest. State law then is preempted, and the court must fashion a substitute. In Community for Creative Non-Violence v. Reid, 490 U.S. 730, 109 S. Ct. 2166, 104 L. Ed. 2d 811 (1989), the Court devised a special agency rule for copyright cases, believing state rules to be inconsistent with one of the aims of the federal law. No one contends that Illinois' law of agency and Title VII are similarly incompatible. Cf. Johnson v. Fankell, --- U.S. ----, ---- - ----, 117 S. Ct. 1800, 1804-07, 138 L. Ed. 2d 108 (1997) (collecting cases about the incompatibility caveat to the use of state law). Chief Judge Posner's assertion that my approach "implies that by a stroke of the pen Illinois could eliminate all Title VII liability in that state that depended on the principles of agency law" (123 F.3d at 508) slights the doctrine that state law is used only if compatible with the federal norm. If a risk that state law may try to undercut the interest protected by federal law means that state law does not apply, then state law would never be used to resolve subordinate issues in federal litigation. A state's ability to set a one-day statute of limitations for personal-injury claims would allow it to wipe out liability for constitutional torts under § 1983, which in turn (on my colleague's view) would foreclose resort to state law no matter what its actual period of limitations was. What the Supreme Court has told us to do is different: we use state law unless its actual period of limitations is so short or so freighted with conditions that its application imperils vindication of the federal right. Burnett v. Grattan, 468 U.S. 42, 104 S. Ct. 2924, 82 L. Ed. 2d 36 (1984). See Morgan, 797 F.2d at 475; Amanda Acquisition Corp. v. Universal Foods Corp., 877 F.2d 496 (7th Cir.1989). A state may set a period of limitations as long as it likes, or apply lax tolling rules; a longer time to sue does not undermine the interest protected by the federal law. See Gosnell v. Troy, 59 F.3d 654, 656 (7th Cir.1995). Similarly a state may expand employers' liability, by imputing many supervisors' acts to the firm, without offending Title VII. My colleagues do not identify a single state agency rule, anywhere in the nation, that is incompatible with Title VII. Irony lies in adopting a novel federal rule--one less favorable to victims of workplace torts than actual state rules--in order to avoid a phantom "hostile" state rule.

Employers' (and employees') main source of protection from silly rules is generality: a state cannot affect claims under federal law without affecting claims under state law too. Nothing I have written here implies that federal courts should pay heed to a law such as: "For purposes of litigation under Title VII, supervisors' acts are attributed to the firm if...." Illinois' agency rules largely favor claims by employees. How far a state may move toward negating principles of respondeat superior is not a question we need engage today. Answering that question is impossible in the abstract, and hard in the concrete; some difficulties, at least, therefore are avoided if we disregard state law. But inventing federal common law is a dubious project for reasons both legal and practical. The legal reasons I have covered. The practical objections include the difficulty of formulating a good rule. Why should we think that our latest brainstorm will really be an improvement? Most mutations in biology and law alike are inferior. Common law of agency has stood the test of time in a way my colleagues' proposals have not. Any federal rule also creates vertical disuniformity: the same supervisory act is imputed to the firm for purposes of state but not federal law, or the reverse. Perhaps more surprisingly, a federal rule will do little to create horizontal uniformity. Thirteen federal courts of appeals comprising some 200 judges are bound to see these issues differently--the one proposition conclusively established by today's welter of opinions--and the Supreme Court lacks the time (and may lack the inclination) to superintend a national body of agency law. Admiralty, the most distinctive body of federal common law, has taxed the Court's resources and patience. Discrimination cases are more plentiful by two orders of magnitude, yet in the 33 years since Title VII was enacted the Supreme Court has heard only two involving workplace harassment. Prospects for a uniform federal common law in this field are dim.

Title VII does not preempt state anti-discrimination laws, and it is unwise to use one set of agency rules for state purposes and a completely different set for the same acts by the same supervisors under Title VII. Several of my colleagues' opinions endorse strict liability for supervisors' quid-pro-quo demands, even though Meritor rejected that approach for similar claims.3 Now it may be that agency principles often imply automatic corporate liability for supervisors' misconduct, as Judge Wood concludes (and as Green Hills Country Club shows), but following normal agency principles when they produce that outcome does not offend Meritor; this is what Meritor requires us to do. Judge Wood's opinion reaches sound conclusions about both the law and the facts; I join it in every respect but one.

The exception is that I would treat both Jansen and Ellerth as claims of hostile working environments. Although, on the approach Judge Wood and I take, there is no material legal difference between quid-pro-quo and hostile environment claims, this distinction matters to a majority of the court, so it is helpful to get the nomenclature straight. Slowik implied that he would hinder Ellerth's advancement if she did not provide sexual favors, but when she refused he did not carry through. He did not reduce her status or salary; indeed, he secured a promotion for her. Ellerth quit only because the working environment was spoiled by the sexual innuendo and offensive touchings. Antoni, like Slowik, also relented when his victim balked. The delay in Jansen's raise did not cost her anything; the employer provided back pay, and Jansen does not contend that she was entitled to interest on this sum. (For all we know, the employer paid interest.) Both plaintiffs therefore received their injuries exclusively from a supervisor's boorish conduct, rather than via the pay envelope or by submitting to unlawful demands. Both Slowik and Antoni solicited other persons to pay tribute (in the form of sexual acts) in order to receive benefits to which they were entitled. That is not only attempted extortion of the subordinate but also attempted theft from the employer, because Slowik and Antoni tried to use the company payroll to finance the acquisition of personal sexual services. Extortion and theft, whether completed or attempted, are crimes, and the incidence of this conduct might decrease if prosecutors treated them that way. What matters for current purposes, however, is that the attempts failed, leaving only the unpleasant working environments.

MANION, Circuit Judge, joined by POSNER, Chief Judge, concurring and dissenting.

Although I recognize that the possibility of any reader (other than members of this court) finding this opinion buried amid 200 pages is remote, I will nevertheless forge ahead. I write separately not to propose yet another standard of employer liability in cases of a supervisor's sexual harassment, but rather to condense some of the areas of agreement and dispute set out by the various writings submitted by other members of the court. At this point, four of the circuit judges have proposed distinct standards of liability in cases where a supervisor sexually harasses a subordinate. The central question is whether strict liability should be applied, and if so, for what? And if it is applied, should it be against an employer for all forms of sexual harassment in the workplace, or only for quid pro quo harassment? And then we need to define quid pro quo harassment--is it for some job action taken against an employee, or is the mere threat of a job action enough? Further, what level of supervisor can inflict the company with strict liability--a high-level supervisor who has the independent and unreviewable authority to execute, or a lower-level supervisor, even one who might be better described as a co-worker, whose input in job actions is limited to routine recommendations and job reports to a higher decision-maker?

If we were writing the law on a clean slate, I would be inclined to join Judge Coffey in imposing a singular, consistent standard for all types of harassment: negligence. That approach recognizes the unfairness in imputing vicarious liability to an employer for a supervisor's errant, personal and illegal acts that have nothing to do with the employer's business or the supervisor's responsibilities. But the slate is not clean. Unfortunately, as revealed by the volume of writings presented under the two cases decided today, the slate is marked with many confusing and even conflicting standards that give trial courts little guidance in this sensitive area. Nevertheless, I strongly endorse much if not most of what Judge Coffey has presented in his very detailed and thorough analysis. Although I am not able to join the bottom line of his concurrence, I strongly endorse his approach on many of the underlying issues.1

Given the state of the case law, it is my conclusion that Chief Judge Posner has fashioned an approach that is in line with Meritor Savings Bank, FSB v. Vinson, 477 U.S. 57, 106 S. Ct. 2399, 91 L. Ed. 2d 49 (1986), our own precedent in this area, and common sense. It is a practical and facile approach that draws needed bright lines (or at least brighter ones) in a dim area of employment law involving dimwitted behavior by offending employees. Judge Posner's "company act" approach helps sort out an issue that has confounded courts since Meritor was decided more than a decade ago. Its power lies in its underpinnings: there is no existent "law of agency" able to answer the questions before us today, and the Supreme Court itself recognized that common law principles are not transferable "in all their particulars to Title VII." 477 U.S. at 72, 106 S. Ct. at 2408.

Judge Flaum has drawn some lines that appear to have attracted at least a plurality of the other judges on the court. It is a rather back-to-basics approach that is commendable for several reasons. Primarily, it retains the distinction between hostile environment harassment and quid pro quo harassment, and it imposes a consistent standard of liability (negligence) in cases of hostile environment harassment no matter whether the harasser is a supervisor or co-worker. In that regard, however, I am less enthusiastic about the somewhat nebulous "heightened" standard of care in cases where the harasser is a supervisor rather than a co-worker. But in most respects Judge Flaum's approach is consistent with Chief Judge Posner's, the principal difference being--and it is a major one--how to treat a supervisor's threats of a job action if the employee does not comply with his sexual demand. I agree with Chief Judge Posner and Judge Kanne that threats without tangible job consequences ("company acts" such as a demotion or firing) should not impose on an unsuspecting employer the harsh penalty of strict liability.

Judge Wood's approach differs substantially from the other three by favoring one standard of liability in all cases of supervisory sexual harassment. Under that standard, an employer is liable if the supervisor committed harassment during the course of his supervision. If so, then the harassment is within the scope of employment even if his motivation was entirely personal and unrelated to the employer's business. If outside the scope of employment, then liability under this approach results if (1) the employer is negligent in failing to address the harassment; or (2) the supervisor has apparent authority to take the employment actions at issue. In addition, Judge Wood apparently has accepted Judge Easterbrook's standard of applying state laws of agency, so that an employer's liability under Title VII depends upon which state law a court applies. Without saying so, Judge Wood's approach effectively imposes strict liability for all sexual harassment, and with Judge Easterbrook's wrap of a 50-state standard, I find it unacceptable.

While these four distinct approaches do not cover every option, there are common threads running through the approaches of Judges Posner, Coffey, and Flaum that are consistent and important. Each favors a negligence standard of employer liability if the alleged harasser is a co-worker of the victim, which includes low-level supervisors who in practical terms are acting as co-workers and not supervisors when they harass. This is consistent with all of our recent decisions addressing this issue. See McKenzie v. Illinois Dep't of Transportation, 92 F.3d 473, 480 (7th Cir.1996); Baskerville v. Culligan Int'l Co., 50 F.3d 428, 432 (7th Cir.1995); Carr v. Allison Gas Turbine Div., General Motors Corp., 32 F.3d 1007, 1009 (7th Cir.1994). The alleged harasser in Baskerville was a low-level supervisor, and nothing in today's decision should be interpreted to mean that a negligence standard is inappropriate in cases where the harasser happens to outrank the victim in the company's hierarchy.

In addition, a majority of the court has decided to retain a standard of strict liability in cases of actual quid pro quo harassment, or for what Chief Judge Posner terms "company acts." This is consistent not only with Horn v. Duke, 755 F.2d 599 (7th Cir.1985), but the decisions of other circuits as well.2 Both Judge Flaum's approach and Chief Judge Posner's retain the distinction between hostile environment harassment and quid pro quo harassment. Given the state of the law, this distinction makes sense; a "company act" such as a supervisor's termination of an employee who refuses to have sex with him is much more severe than the general creation of a hostile environment (which typically consists of multiple acts of harassment actionable only in combination by creating an offensive work environment). The company act also is much more detectable than harassment committed outside the employer's vantage point, or at least it will become more readily detected once employers understand that they are strictly liable for these acts. When a supervisor wishes to fire a subordinate, it will behoove the employer to know the reasons and ensure that they are valid and non-discriminatory. Although this probably will require more detailed procedures, it is a better result than monitoring the workplace as if it were a prison, searching for inappropriate behavior and earning the disgruntlement and dissatisfaction of employees in return.

These are the areas of general agreement. But as the length and the separate writings in this "opinion" demonstrate, many differences remain. Four major areas remaining in dispute merit some discussion.

I. Quid pro quo harassment and a supervisor's unfulfilled threats

As I noted above, the court has retained the distinction between quid pro quo harassment and hostile environment harassment. In addition, under both Judge Flaum's approach and Chief Judge Posner's approach, an employer is strictly liable for a supervisor's quid pro quo harassment. And because of that harsh standard of strict liability, it is important and necessary to keep the definitions clear and narrow. Quid pro quo harassment occurs when the submission to a superior's sexual demands is made a condition of tangible employment benefits. Before strict liability attaches, however, the plaintiff must meet a significant burden.

Quid pro quo harassment is a special case of sexual harassment. This means that the plaintiff must establish the basic gate-keeping requirements of a sexual harassment claim in order to state a quid pro quo claim (e.g., unwelcome sexual advances motivated at least in part by the plaintiff's sex). However, in addition to these elements, the plaintiff must establish (1) an ultimatum or strong suggestion on the part of a supervisor that failure to comply with a sexual demand will cause an employment detriment; and (2) that when she refused to comply with the demand, the supervisor imposed upon her a tangible job detriment.

These two additional elements are necessary because they distinguish quid pro quo claims from hostile environment claims. The first element--an ultimatum on the part of the supervisor--may be either explicit or implicit. An example of an explicit demand is a supervisor who tells an employee she will be fired if she refuses to have sex with him. (This assumes, of course, that he has the power to unilaterally fire her.) But most cases fall into the "implicit" category, such as the supervisor who discusses the employee's job performance (and performance review) at the same time he broaches the subject of sexual favors. The closer the nexus between the discussion of the review and the sexual advance, the more likely the case involves actionable quid pro quo harassment. (This again assumes the supervisor's review has a dispositive effect on the subordinate's job.) But there is a dangerous trend to stretch the boundaries of this element so that any harassment by a supervisor who has authority to influence raises, leaves of absences, or other job benefits is quid pro quo harassment. Proof that the harassment is by a supervisor (who often has plenary power over the employee-plaintiff) is a necessary, but not a sufficient condition to establishing a quid pro quo claim. The plaintiff also must offer evidence that the supervisor, through his words or conduct, actually conditioned the employee's receipt of the benefits on his receiving sexual favors. Typically this occurs only when the supervisor is discussing job benefits with an employee at the same time he is discussing the sexual favors with her.

The second element of a quid pro quo case (that the plaintiff suffered a tangible job detriment as a result of not submitting to the advances) is equally important if we are going to impose strict liability in these cases. Sexual harassment is a tort, and strict liability is not imposed on tortious conduct except in rare circumstances, such as inherently dangerous activities requiring great care on the actor's part. (See Coffey concurrence, ante at 530-531.) Where those activities result in danger but not in damage, no liability exists. So too with quid pro quo sexual harassment. If the plaintiff has suffered no tangible job detriment as a result of not submitting to a sexual demand (e.g., she has called the supervisor's bluff), then the tort is incomplete and no cause of action can lie. While it is true that the adverse employment action (or tangible job detriment) may vary in form from case to case--a demotion in one case, a loss of title in another, Bryson v. Chicago State Univ., 96 F.3d 912, 916 (7th Cir.1996)--the job detriment must be material and tangible.

A threat without negative job consequences cannot support a quid pro quo claim. An employee who either has succumbed to her supervisor's sexual advance or has called her supervisor's bluff has suffered no tangible job detriment. In either case, absent a complaint by the employee, the company is in the dark. Even more so with belated claims of resulting psychological or emotional distress. Strict liability for injuries known only to her and her doctor is clearly unjust. Suppose the threat is "have sex with me or you're fired." Unless the employee complains, or refuses to have sex with the supervisor and thus is discharged, the company knows nothing. We can encourage companies to review discharges to make sure that they are legitimate (it would be unfortunate to require closer scrutiny when men fire women, but with this trend in the law it may come to that). But if in this example the employee was not fired, there is no "company act" to review. We all agree that the purpose behind vicarious liability of any kind is deterrence. A standard of strict liability on the employer's part will not deter the making of threats any more effectively than a standard of negligence because the employer is helpless to prevent them in the first place. No one favoring strict liability for mere threats has suggested that a company surreptitiously monitor its supervisors' conversations with employees to prevent threatening statements from being made.

Threats cannot constitute quid pro quo harassment for the same reason they could not fall within the supervisor's scope of employment--they are not authorized, they do not carry the employer's imprimatur, they do not speak on behalf of the company. In Chief Judge Posner's parlance, they are "noncompany acts"; in Judge Coffey's, they constitute "attempted quid pro quo harassment." In either case, they are personal and the employer has no reason to know they were uttered. If we treated threats as company acts, thus implicating strict liability, then we would be saying that any supervisor, acting on his own, has the power to bind an employer to liability based on tort, while at the same time few among us would agree that he could so readily bind his employer to a contract. And because so many complaints of sexual harassment contain threatening comments, quid pro quo harassment and strict liability would become the typical case rather than the narrow exception it should be.

The Supreme Court did not elaborate when it instructed courts "to look to agency principles for guidance" in the area of employer liability for sexual harassment. Meritor, 477 U.S. at 72, 106 S. Ct. at 2408. We are left to determine exactly what "common-law agency principles" it intended. Id. While the Restatement (Second) of Agency might look like a logical starting place, its usefulness is limited in cases such as this one. Issued almost forty years ago, nowhere does it mention sexual harassment; indeed, it predates Title VII. In fact, there is not much in the Restatement that makes any sense to apply to the sexual harassment context. The most applicable section probably is § 474, dealing with the so-called "fellow servant rule." But as Chief Judge Posner points out, applying that section to sexual harassment cases would be hopelessly confusing and formulaic. We would have to determine whether the harassing supervisor is a "vice principal" or a "superior servant." See, e.g., Restatement § 479. If the supervisor was classified as a superior servant, then another rule applies: "the master is not liable unless the superior servant's act is so close to that which he is authorized to do that it is within the scope of his employment." Restatement § 487, Comment a. This chain reaction is never-ending if we treat the Restatement as a surrogate statute.

Courts run into similar problems when trying to wring out a useable standard of employer liability from § 228 of the Restatement. That section contains a four-part test to determine whether conduct of a servant is within his scope of employment. I need not quote all of that section, for my colleagues do. It should be enough to quote § 228(2), which plainly states that "[c]onduct of a servant is not within the scope of employment if it is different from that authorized, far beyond the authorized time or space limits, or too little actuated by a purpose to serve the master." (Emphasis added.) Section 235 explicitly states that an "act of a servant is not within the scope of employment if it's done with no intention to perform it as a part of or incident to a service on account of which he is employed." Comment c could well be referring to an intentional tort such as sexual harassment in noting that "[t]he fact that an act is done in an outrageous or abnormal manner has value in indicating that the servant is not actuated by an intent to perform the employer's business."3 That Comment also recognizes that in many cases the servant (supervisor) is "merely using the opportunity afforded by the circumstances to do the harm." This is precisely what a supervisor who sexually harasses is doing--taking advantage of his circumstances, such as his proximity to the employee and authority over her. It makes him an opportunist and his employer a victim of circumstances beyond its control.

Against this backdrop some of my colleagues conclude that a supervisor's sexual harassment may well fall within the scope of his employment. That would seem to be the rare case indeed, as other circuits have held. See Harrison v. Eddy Potash, Inc., 112 F.3d 1437, 1444 (10th Cir.1997) ("sexual harassment simply is not within the job description of any supervisor or any other worker in any reputable business"); Andrade v. Mayfair Mgmt., Inc., 88 F.3d 258, 261 (4th Cir.1996) ("illegal sexual harassment is an illegitimate corporate activity, beyond the scope of supervisors' employment"); Gary v. Long, 59 F.3d 1391, 1398 (D.C.Cir.1995) (harassing supervisor acts outside the scope of his employment in creating hostile environment, and employer is not liable where it has adopted policies and implemented measures condemning sexual harassment); Bouton v. BMW of North Am., Inc., 29 F.3d 103, 107 (3rd Cir.1994) (sexual harassment is outside scope of employment and liability for supervisor's sexual harassment may only be imputed to employer where supervisor exercises actual or apparent authority; but plaintiff's belief that supervisor has apparent authority is unreasonable if the employer has a sexual harassment policy known to plaintiff).

My colleagues who embrace the Restatement and elevate it to the level of statutory force must accept the consequences as well. Under § 228, harassment only falls within the scope of a supervisor's employment if for some (inexplicable) reason, the supervisor believes that he is furthering the employer's business in the process. But more than this, the harassment also must approximate the type of conduct that the employer authorized, and obviously no reasonable employer would authorize such mistreatment of its own employees. Judge Wood believes that focusing on the conduct authorized by the employer is a "misstep," and that we should ask only whether the harassment occurred during the course of the employee's supervision. In other words, the issue would not be whether the tortious conduct creating the issue of liability was authorized, but whether it was performed by the supervisor in the course of his authorized activities. But if our reading of the Restatement is a misstep, then we are not alone, and the Supreme Court has stumbled too. See Varity Corp. v. Howe, 516 U.S. 489, ----, 116 S. Ct. 1065, 1073, 134 L. Ed. 2d 130 (1996) (citing Restatement's § 229(1) as "determining whether an activity is within the 'scope of ... employment' in part by examining whether it is 'of the same general nature as that authorized' ") (emphasis added).4

My third area of major disagreement concerns liability under a theory of so-called "apparent" authority. Apparent authority "is the power to affect the legal relations of another person by transactions with third persons, professedly as an agent for the other, arising from and in accordance with the other's manifestations to such third persons." Restatement (Second) of Agency § 8 (1957). Where a company publishes a policy against sexual harassment, as most employers do, there is no basis to support liability under a theory of "apparent" authority. This is in accord with the Court of Appeals for the Third Circuit, which has held that "[a] policy known to potential victims ... eradicates apparent authority the harasser might otherwise possess." Bouton v. BMW of North Am. Inc., 29 F.3d 103, 110 (3rd Cir.1994). Simply put, apparent authority must be based on an employee's reasonable belief that the agent (the supervisor) had authority, and no employee reasonably could believe that her supervisor was entitled to harass given a policy against such harassment.5

Apparent authority is the square peg that doesn't fit into this round-hole discussion. The theory mostly applies to contracts between a stranger and an agent of the defendant (the principal). Where the principal has led the stranger to believe that the agent has the authority to make the contract, the principal is bound by it. So the stranger's reasonable belief must be based on some affirmative act done by the principal. See Bourjaily v. United States, 483 U.S. 171, 189, 107 S. Ct. 2775, 2786, 97 L. Ed. 2d 144 (1987) ("the agent's authority must be traced back to some act or statement by the alleged principal"); see also Restatement § 8 Comment a ("Apparent authority results from a manifestation by a person that another is his agent, the manifestation being made to a third person and not ... to the agent."). Neither Jansen nor Ellerth contends that their employers informed them that their supervisors had the authority to sexually harass them. The facts demonstrate the opposite. For example, Ellerth concedes that Burlington had a policy against harassment and that she knew of it.6

Some of my colleagues apparently believe that an employer may be liable so long as it approves of the supervisor's general supervision, rather than his specific act of harassment. For them it is enough that the employer's general grant of authority to the supervisor "enabled" him, or put him in a position, to commit an intentional tort. See post at 579 ("The one action [Antoni] took that was clearly unrelated to his supervisory position was his alleged vandalism of [Jansen's] car. No authority that PCA gave him enabled him, helped him, or affected his ability to take knife to tire.") (Judge Wood, concurring) (Emphasis added). Under that approach, an employer nearly always is liable because the powers it bestows upon a supervisor "enable" and "help" him to harass a subordinate. Apparently, the only requirement is that he be on the job when he commits the tort. It is the wrong approach because it altogether shifts focus away from the agent's activities, and imposes liability on the employer no matter how dissimilar the conduct is to the activities the employer actually authorized and communicated to the employees as authorized. In a nutshell, an intentional tort such as sexual harassment is so outrageous (by now everyone knows it's illegal) that it is hard to imagine a scenario where an employee reasonably believes her supervisor's job description includes the right to sexually harass her. See American Society of Mechanical Engineers, Inc. v. Hydrolevel Corp., 456 U.S. 556, 566, 102 S. Ct. 1935, 1942, 72 L. Ed. 2d 330 (1982) ("Under an apparent authority theory, '[l]iability is based upon the fact that ... from the point of view of the third person the transaction seems regular on its face and the agent appears to be acting in the ordinary course of the business confided to him.' ") (quoting Restatement § 261, Comment a).

If there is an unusual aspect to any of the separate approaches issued today, it is one suggesting that we should look to state common law to determine when and under what conditions an employer is vicariously liable for a supervisor's sexual harassment. Presumably this would mean that we often would be faced with a choice of law question. We might impose liability on an employer in one case, but not another though the facts of the two cases were the same. We would explain this disparate impact by saying that though we are a federal court applying the civil rights laws of the nation (here, Title VII), and though Congress surely intended all citizens to have equal rights under that federal statute, we think it best to allow the particulars of local interests to control federal court decisions.

The suggestion that we look to state law to resolve sensitive issues under the civil rights laws contradicts the general assumption that " 'in the absence of a plain indication to the contrary, ... Congress when it enacts a statute is not making the application of the federal act dependent on state law.' " Mississippi Band of Choctaw Indians v. Holyfield, 490 U.S. 30, 43, 109 S. Ct. 1597, 1605, 104 L. Ed. 2d 29 (1989) (quoting Jerome v. United States, 318 U.S. 101, 104, 63 S. Ct. 483, 485, 87 L. Ed. 640 (1943)). "One reason for this rule of construction is that federal statutes are generally intended to have uniform nationwide application." Id. This is why in NLRB v. Hearst Publications, Inc., 322 U.S. 111, 64 S. Ct. 851, 88 L. Ed. 1170 (1944), the United States Supreme Court rejected an argument that the term "employee" as used in the National Labor Relations Act (Wagner Act) should be defined by state law. It explained: Both the terms and purposes of the statute, as well as the legislative history, show that Congress had in mind no ... patchwork plan for securing freedom of employees' organization and of collective bargaining. The Wagner Act is ... intended to solve a national problem on a national scale.... Nothing in the statute's background, history, terms or purposes indicates its scope is to be limited by ... varying local conceptions, either statutory or judicial, or that it is to be administered in accordance with whatever different standards the respective states may see fit to adopt for the disposition of unrelated, local problems. 322 U.S. at 123, 64 S.Ct. at 857. Because the NLRA was the model for much of Title VII, Lorance v. AT & T Technologies, Inc., 490 U.S. 900, 909, 109 S. Ct. 2261, 2267-68, 104 L. Ed. 2d 961 (1989), it follows that the reasons favoring a uniform national standard under the NLRA would support similar uniformity in our enforcement of Title VII. We should avoid the same "patchwork plan" of "varying local conceptions" governing civil rights laws that the Supreme Court sought to avoid under the NLRA. See id. ("[W]e have often observed that the NLRA was the model for Title VII's remedial provisions, and have found cases interpreting the former persuasive in construing the latter.").

Despite Congress' (and the Supreme Court's) seemingly clear preference that we favor a uniform national standard of employer liability rather than a 50-state standard wrapped in local common law, some of my colleagues would look to those local laws to determine when a supervisor acts within the "scope of his employment." But in interpreting the very same term under a different federal statute (the Copyright Act of 1976), the Supreme Court unequivocally (and unanimously) rejected that approach:

In past cases of statutory interpretation, when we have concluded that Congress intended terms such as "employee," "employer," and "scope of employment" to be understood in light of agency law, we have relied on the general common law of agency, rather than on the law of any particular State, to give meaning to these terms.... This practice reflects the fact that "federal statutes are generally intended to have uniform nationwide application." Mississippi Band of Choctaw Indians v. Holyfield, 490 U.S. 30, 43, 109 S. Ct. 1597, 1605, 104 L. Ed. 2d 29 (1989).

Community for Creative Non-Violence v. Reid, 490 U.S. 730, 740, 109 S. Ct. 2166, 2173, 104 L. Ed. 2d 811 (1989) (emphasis added). Meritor instructed us to look to "common law [agency] principles" to determine the scope of an employer's liability. Only three years later, Reid interpreted nearly identical language--the general "common law of agency"--as ruling out reliance "on the law of any particular State," which is exactly what some of my colleagues propose to do. For those who favor a 50-state standard of "scope of employment," Reid is seemingly insurmountable.

Finally, it is worth noting that already this year the Supreme Court has decided two cases of statutory interpretation under Title VII. See Robinson v. Shell Oil Co., --- U.S. ----, 117 S. Ct. 843, 136 L. Ed. 2d 808 (1997) (unanimously holding that the term "employees," as used in the anti-retaliation provision of Title VII, includes former employees); Walters v. Metropolitan Educ'l Enterprises, Inc., --- U.S. ----, 117 S. Ct. 660, 136 L. Ed. 2d 644 (1997) (unanimously favoring payroll method to determine how many employees employer "has" on day in question). In neither case did the Court look to state law to inform its interpretation of Title VII terms.

Ultimately, the scope and purpose of Title VII resolve the issue as to whether state common law should prevail. The primary purpose of the Civil Rights Act of 1964 was to rid the workplace of discrimination. See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 800, 93 S. Ct. 1817, 1823, 36 L. Ed. 2d 668 (1973) ("The language of Title VII makes plain the purpose of Congress to ensure equality of employment opportunities and to eliminate those discriminatory practices and devices which have fostered racially stratified job environments to the disadvantage of minority citizens."). I hardly think Congress intended to be successful only in so far as its efforts comported with state law. What happens when the Equal Employment Opportunity Commission, specifically empowered under 42 U.S.C. § 2000e-5 to enforce the provisions of Title VII by suing an employer on its own, files suit against an employer in federal court? Must the Commission turn to state law to support its claim of discrimination? It would seem odd that a federal agency's success in seeking the enforcement of a federal statute in federal court would depend upon state law.

Of all the approaches suggested by various members of the court today, given the present state of the law, Chief Judge Posner's approach would provide a consistent, uniform standard concerning when an employer is liable for a supervisor's sexual harassment. Where the supervisor commits a "company act," such as firing an employee who refuses to have sex with him, the employer is strictly liable. But when a supervisor's sexual harassment has nothing to do with his position (though technically it occurs during the course of his workday), the standard of employer liability should be one of negligence. The issue becomes whether the employer knew or should have known of the harassment, what it could have done to prevent it, and, of course, what it did to remedy it. We need not muddy the playing field by asking whether the supervisor was acting within the scope of his employment because it would be difficult to envision a scenario where he was (see supra, Pt.II). But even if we were to ask such a question, it would make little sense to condition the answer on the law of any particular state. In banning state literacy tests by passing the Voting Rights Act Amendments of 1970, Congress recognized that discrimination was a "national dilemma that touche[d] every corner of our land," and further determined "the way to solve the problems of racial discrimination [was] to deal with nationwide discrimination with nationwide legislation." Oregon v. Mitchell, 400 U.S. 112, 133, 91 S. Ct. 260, 269, 27 L. Ed. 2d 272 (1970). The same could be said about Title VII. Its standards ought to be accessible and uniform; in other words, they ought to be common. For all of these reasons, I join the Chief Judge's opinion.

DIANE P. WOOD, Circuit Judge, joined by EASTERBROOK and ROVNER, Circuit Judges, concurring and dissenting.

Today the en banc court establishes two propositions of great importance for all future claims of sexual harassment by a supervisor. First, a majority of judges agrees that both Kimberly Ellerth and Alice Jansen properly preserved all aspects of their claims of sexual harassment in their EEOC filings and district court complaints. (In addition to this opinion, see also opinions of Chief Judge Posner, Judge Flaum, and Judge Easterbrook.) Second, a majority of judges supports a rule under which employers will be liable when their supervisory employees, acting within the scope of their employment or their delegated authority, engage in "quid pro quo" sexual harassment. In keeping with the panel opinion in Ellerth v. Burlington Industries, Inc., 102 F.3d 848 (7th Cir.1996) (Bauer, Rovner, and D. Wood), I analyze this as a problem of agency law, under which acts of "quid pro quo" harassment will almost always fall within the scope of the supervisor's employment and thus result in employer liability under a respondeat superior theory. Other members of the court have offered their own explanations. Judge Flaum concludes that employer liability is "automatic" once the plaintiff shows that the supervisor conditioned employment consequences upon the receipt of sexual favors by wielding the authority actually delegated to him, and Chief Judge Posner has shown why this is an efficient outcome.

The important point is that a majority of the court agrees on several fundamental propositions. A plaintiff must prove that (1) the harasser had supervisory authority over her or him, (2) the employer authorized the supervisor to confer the tangible employment benefit in question (i.e., the "quid"), and (3) the supervisor demanded sexual favors as the price of that benefit. Furthermore, a majority of the en banc court agrees that the employer cannot escape liability in "quid pro quo" cases merely by showing that it has a policy against sexual harassment. These are important developments in this circuit's sexual harassment law, which should not be overshadowed by our remaining differences.

Finally, a majority of the en banc court agrees that both Ellerth (if she did not waive the point) and Jansen brought forward enough evidence to survive summary judgment on their claims that their respective supervisors created hostile work environments through acts of sexual harassment.1 (Whether Ellerth waived her right to rely on that theory is a separate question on which we disagree. I think not, for reasons I address in the final section of this opinion.) The majority has not, however, reached a consensus on the appropriate legal framework for this class of cases. In my view, the rules governing employer liability under Title VII for a supervisory employee's behavior do not depend on the kind of discriminatory or harassing conduct in which the supervisor has engaged; employer liability depends instead on whether the supervisor is acting within the scope of his authority, as I explain more fully below. I would therefore remand both cases for further proceedings.

The agency analysis I would use is consistent with any theory of discrimination that could be asserted under Title VII. Simply put, when a supervisory employee, acting within the scope of his authority as determined under the applicable state law, engages in discriminatory conduct, the employer should be liable under well recognized principles of respondeat superior, whether those acts are characterized for analytical purposes as "quid pro quo" demands, the infliction of a "hostile environment," or other discriminatory acts. The employer should also be liable when the supervisor acts within the scope of his apparent authority to take the employment actions in question: hiring, firing, assigning tasks, or controlling the workplace environment. Conversely, if the supervisor uses neither actual nor apparent authority to take the employment action, then the employer should be liable only if the plaintiff shows that the employer knew or should have known about the harassing behavior. This approach is consistent with both the statute and the Supreme Court's decisions in Meritor Savings Bank v. Vinson, 477 U.S. 57, 106 S. Ct. 2399, 91 L. Ed. 2d 49 (1986), and Harris v. Forklift Sys. Inc., 510 U.S. 17, 114 S. Ct. 367, 126 L. Ed. 2d 295 (1993). It gives the district courts the manageable task of deciding questions of scope of authority instead of imposing on them the impossible task of parsing a supervisor's every act to see if it belongs in the "quid pro quo" or the "hostile environment" category. Finally, it allows employers to contest the plaintiff's claim that the supervisor acted within the scope of his authority or abused his delegated authority. It therefore respects the Supreme Court's admonition in Meritor not to impose "strict liability" on employers in this area.2

* Before addressing the particular facts of the cases before us, it is useful to take a broader look at the legal principles that will govern them. Two fundamental issues arise. First, for purposes of employer liability, is there a principled distinction between "quid pro quo" and "hostile environment" claims? Second, under what circumstances will an employer be liable for the sexually harassing acts of its supervisory employees?

Title VII prohibits discrimination on the basis of an individual's "race, color, religion, sex, or national origin." 42 U.S.C. § 2000e-2(a). In both Meritor and Harris, the Supreme Court recognized that sexual harassment is a form of sex discrimination. It also recognized that sexual harassment is not confined to actions of the "quid pro quo" variety, in which someone demands sexual favors in exchange for a tangible job benefit or the promise of such a benefit. Sexually harassing actions that create a "hostile work environment" also fall within the statutory prohibition of sex discrimination. Similarly, courts have recognized that discrimination based on race might take the form of a hostile work environment, see, e.g., Rodgers v. Western-Southern Life Ins. Co., 12 F.3d 668, 673 (7th Cir.1993); Daniels v. Essex Group, Inc., 937 F.2d 1264, 1270 (7th Cir.1991).

In the real world, sexual harassment does not sort itself into tidy categories. The same woman might be subject to both kinds of harassment from her supervisor at the very same time: he might repeatedly order her to his office, close the door, and stroke her leg, and on each occasion he might tell her that promotions will occur more readily if she has sex with him. The former conduct (made possible by the fact of his authority over her) would fall within the "hostile environment" category, but the latter would clearly be "quid pro quo" harassment. A rule making the employer strictly liable for part of this encounter, but liable only for negligence with respect to the other part, would make neither legal nor practical sense. This is because, unlike the example Chief Judge Posner suggests of the drunken worker who first injures a customer and later hits a pedestrian, in the harassment case the offending supervisor inflicts injury on a single victim, through a single act or course of conduct, and thus violates only one legal right. When this supervisor bombards an unwilling subordinate with unwanted sexual images, touching, vulgar words, or denigrating comments, only the most committed formalist would feel confident in saying when those actions cross the imaginary line from "hostile environment" harassment to "quid pro quo" harassment. By the same token, when the supervisor makes constant demands for sex in exchange for job benefits (maybe in jest, maybe not), the victim surely suffers from a "hostile environment" at the same time as she endures the "quid pro quo" harassment.

As Judge Flaum and those who have joined his opinion recognize, a supervisor is able to harass in a variety of ways: he might attempt to use his supervisory authority to obtain sexual favors from a subordinate employee in a form of sexual blackmail; he might use his supervisory authority to create a "hostile environment" for the subordinate, requiring her presence while offensive language or behavior goes uncorrected or even encouraged; or he might offend in the same way as an obnoxious co-worker or stranger, in circumstances where the authority he wields in the company is irrelevant. The first two cases are both forms of sex discrimination made possible by the employer's delegation of authority to the supervisor; the third is harassment unrelated to the delegated authority. The critical question, for purposes of employer liability, turns on whether the supervisor has abused the powers the employer has entrusted to him; it does not turn on the way in which the supervisor has decided to abuse them.

A majority of this court would nonetheless make the employer's liability for a supervisor's actions depend on the artificial distinction between "quid pro quo" and "hostile environment" harassment. As the per curiam opinion reports, for actions that fall on the "quid pro quo" side of the line they would find the employer strictly liable, while for actions on the "hostile environment" side of the line they would find liability only if the employer was negligent. Such an approach relies on a line that will often be impossible to draw in supervisory harassment cases. Worse, it conflates two distinct inquiries: what counts as discrimination for purposes of Title VII, and when must the employer answer for it? There is nothing to justify the assumption that a supervisor's actions will be all one type or all another in each individual case; indeed, the facts of Jansen and Ellerth show that this is not likely to be true. If the statute itself spoke in terms of different types of sexual harassment and called for different liability rules, we would be bound to follow it and to do the best we could. But the statute does no such thing. Nowhere does it suggest that the question of who speaks for the employer depends on the type of discrimination (sexual harassment, other sex discrimination, race discrimination, etc.) at issue.

One argument for superimposing such a distinction on the statutory language rests on the empirical assumption that employers' ability to monitor will vary depending on the type of harassment at issue. Chief Judge Posner has suggested (and I agree) that a strict liability rule would generally cause an employer to monitor the supervisor's behavior more closely. Nonetheless, he attempts to limit the force of this observation to the "quid pro quo" situation, postulating that (1) employers are better able to police the "quid pro quo" variety of harassment (on the assumption that multiple layers of bureaucratic review will constrain a supervisor's "quid pro quo" actions), (2) these same layers of review will be ineffective for hostile environment claims, and (3) most claims will be filed against large companies. Putting to one side the assumption that the normal Title VII defendant will have all the layers of review on which he relies, there is no reason to believe that employers have so few tools against supervisors whose discriminatory actions inflict a "hostile environment" on subordinates. It is worth recalling how unbearable the workplace must become in order to be "hostile" as to be actionable under Title VII. The Supreme Court defined a hostile work environment as one "permeated with discriminatory intimidation, ridicule, and insult ... that is sufficiently severe or pervasive to alter the conditions of the victim's employment." Harris, 510 U.S. at 21, 114 S.Ct. at 370 (internal quotations and internal citation omitted). Compare Saxton v. American Telephone & Telegraph Co., 10 F.3d 526, 533-35 (7th Cir.1993) (co-worker on different occasions rubbing leg, kissing, and leaping out at employee from behind a bush not sufficiently severe or pervasive); Hennessy v. Penril Datacomm Networks, Inc., 69 F.3d 1344, 1353-54 (7th Cir.1995) (fellow employee's taking victim to striptease bar, shouting for her to get up and perform, comparing her breasts to those of the dancers, and propositioning her would not have been enough, standing alone, for a claim).

Employers are quite capable of monitoring a supervisor's actions affecting the work environment without engaging in Orwellian surveillance. The alternative requirements for liability of sufficient severity or pervasiveness give focus to the employer's efforts to ensure that its supervisors are not engaged in blatantly harassing acts or an ongoing pattern of significant sexual harassment, or for that matter acts of racial discrimination, conventional sex discrimination, or a host of other undesirable behaviors. Nothing systematic makes this kind of severe or pervasive condition harder to detect than a single, clearly unacceptable, act of "quid pro quo" harassment, which could easily occur behind closed doors with only the supervisor and the victim as witnesses. The relative advantage the employer has over the courts in bringing about a harassment-free workplace does not vary depending on the type of harassment. In addition, making it clear that employers must answer for their supervisors acting within the scope of the authority delegated to them will increase the deterrent power of Title VII, by leading employers to improve their training and supervision of these employees.

Just as the statutory language and empirical considerations do not support investing the distinction between types of harassment with the weight most of my colleagues would give it, neither do the Supreme Court's decisions in this field. In Meritor the Supreme Court announced that the lower courts were to look to agency principles to determine when an employer would be liable for a supervisor's sexual harassment. Meritor itself was argued as a hostile work environment case before the Court, but the record shows that the supervisor had actually engaged in both types of harassment. Notably, the Court did not distinguish between the two varieties of harassment when it came to discuss employer liability. Instead, it stated that:

Congress wanted courts to look to agency principles for guidance in this area. While such common-law principles may not be transferable in all their particulars to Title VII, Congress' decision to define "employer" to include any "agent" of an employer, 42 U.S.C. § 2000e(b), surely evinces an intent to place some limits on the acts of employees for which employers under Title VII are to be held responsible.

For this reason, we hold that the Court of Appeals erred in concluding that employers are always automatically liable for sexual harassment by their supervisors. See generally Restatement (Second) of Agency §§ 219-237 (1958). For the same reason, absence of notice to an employer does not necessarily insulate that employer from liability.

477 U.S. at 72, 106 S. Ct. at 2408 (emphasis added). This passage is phrased entirely in terms of Title VII as a whole, indicating that the Court was not creating a rule specific to sexual harassment cases--let alone "hostile environment" sexual harassment cases--but was instead construing the entire statute to require reference to agency principles. In addition, the Meritor Court viewed agency principles as limiting, not expansive, factors in determining liability, as compared with the strict liability rule the lower court had adopted there.

It is true that Judge Flaum's approach more nearly accords with that of the Third, Sixth, Ninth, and Tenth Circuits, which have not questioned the legal basis of the distinction between "quid pro quo" and "hostile environment" claims. That distinction is now some eighteen years old, dating from Catherine MacKinnon's pathbreaking book on Sexual Harassment of Working Women, published in 1979. Those courts, now joined by a majority of this court, have decided to make employer liability in this one area of Title VII law turn on the type of conduct at issue as well as the relationship between employer and employee.3 See Nichols v. Frank, 42 F.3d 503, 513 (9th Cir.1994); Bouton v. BMW of North America, 29 F.3d 103, 106 (3d Cir.1994); Sauers v. Salt Lake County, 1 F.3d 1122, 1127 (10th Cir.1993); Kauffman v. Allied Signal, Inc., 970 F.2d 178, 183 (6th Cir.1992). In my view, however, those courts have not taken to heart the Supreme Court's admonition in Meritor to look to agency principles in Title VII cases. What they have called "strict liability" for "quid pro quo" cases is, in fact, nearly identical to the result produced by properly applied agency principles. As such, I have no quarrel with the results these courts have reached. In fact, Karibian v. Columbia Univ., 14 F.3d 773, 780 (2d Cir.1994), and Faragher v. City of Boca Raton, 111 F.3d 1530, 1536 (11th Cir.1997) (en banc), suggest that the Second and Eleventh Circuits have moved toward a uniform mode of analyzing employer liability for all sexual harassment. In Faragher, the Eleventh Circuit held that:

an employer may be held indirectly, or vicariously, liable for hostile environment sexual harassment:(1) when a harasser is acting within the scope of his employment in perpetrating the harassment, see Sparks v. Pilot Freight Carriers, Inc., 830 F.2d 1554, 1558 (11th Cir.1987) (citing Restatement (Second) of Agency § 219(1)); and (2) when a harasser is acting outside the scope of his employment, but is aided in accomplishing the harassment by the existence of the agency relationship. Sparks, 830 F.2d at 1559-60 (citing Restatement (Second) of Agency § 219(d)).

111 F.3d at 1536 (footnotes omitted, parentheticals in original).4 In Sparks, cited approvingly in Faragher, the plaintiff's sexual harassment claim was factually similar to Ellerth's.5 In both Faragher and Sparks, the Eleventh Circuit relied on agency principles to assess the claims before it; it did not purport to impose "strict liability" on the employer.

B. Employer Liability for Supervisor's Actions

What, then, are the governing "agency principles" and where do we find them? As I indicate below, I agree with Judge Easterbrook that state law must furnish the rule of decision here, but even if we were creating a new federal common law of agency, it would be necessary to ground those rules in something. Chief Judge Posner has launched a broadside attack on the Restatement (Second) of Agency, noting that it is not a statute and need not be treated as "The Way." Few would disagree with that proposition. The Restatement nevertheless stands as a useful reference point for the common law of agency, and its principles have been adopted by many state and federal courts as they have engaged in the very common law process he describes. See, e.g., Haddon v. United States, 68 F.3d 1420, 1423-24 (D.C.Cir.1995); Lehmann v. Toys 'R' Us, Inc., 132 N.J. 587, 626 A.2d 445, 461-64 (1993); Hinshaw v. Board of Comm'rs of Jay County, 611 N.E.2d 637, 640 (Ind.1993); Olson v. Connerly, 156 Wis. 2d 488, 457 N.W.2d 479, 483-84 (1990); Deal v. Byford, 127 Ill. 2d 192, 130 Ill. Dec. 200, 537 N.E.2d 267, 272-73 (1989).

I see no reason to disregard the accumulated body of agency law that accepts the Restatement's propositions, any more than I would disregard state law developments that move beyond it. We owe respect to all of the varying sources of agency law to which the Supreme Court directed our attention in Meritor. The harder question, as Judge Easterbrook has pointed out, is to ascertain which body of agency law governs our interpretation of 42 U.S.C. § 2000e(b) (making employers liable for acts of their "agents," see Hunter v. Allis-Chalmers Corp., 797 F.2d 1417, 1422 (7th Cir.1986)):(1) a new federal common law of agency, applicable only to sexual harassment cases and invented out of whole cloth; (2) a federal common law of agency that borrows its rules from the appropriate state law; or (3) state law directly. As Judge Easterbrook has argued at greater length, the first of these options is inappropriate in light of the Supreme Court's recent reiteration of the rule that state law should be incorporated as the federal rule of decision (when a statute is silent) in the absence of a significant conflict with, or threat to, a national interest. See Atherton v. FDIC, --- U.S. ----, ----, 117 S. Ct. 666, 673, 136 L. Ed. 2d 656 (1997); see also United States v. Kimbell Foods, Inc., 440 U.S. 715, 740, 99 S. Ct. 1448, 1464-65, 59 L. Ed. 2d 711 (1979). It is hard to see how the need for national uniformity is any greater for the definition of the term "agent" in Title VII than it is in many other areas where federal courts routinely turn to state law for underlying property and contract rules. For example, the tax and bankruptcy laws--matters of intense national interest--rely on state laws to determine whether property rights have been created and the extent of those rights. See United States v. National Bank of Commerce, 472 U.S. 713, 722, 105 S. Ct. 2919, 2925, 86 L. Ed. 2d 565 (1985) ("[I]n the application of a federal revenue act, state law controls in determining the nature of the legal interest which the taxpayer had in the property"); Butner v. United States, 440 U.S. 48, 54 & n. 9, 99 S. Ct. 914, 918 & n. 9, 59 L. Ed. 2d 136 (1979) ("Notwithstanding this requirement as to uniformity the bankruptcy acts of Congress may recognize the laws of the state in certain particulars, although such recognition may lead to different results in different States.") State laws also determine if a state employee serves "at will" or not in § 1983 cases, see Border v. Crystal Lake, 75 F.3d 270, 273 (7th Cir.1996); and federal courts use state statutes of limitations for cases under 42 U.S.C. §§ 1981 and 1983. See, e.g., Johnson v. Railway Express Agency, Inc., 421 U.S. 454, 462, 95 S. Ct. 1716, 1721, 44 L. Ed. 2d 295 (1975); Robertson v. Wegmann, 436 U.S. 584, 590-91, 98 S. Ct. 1991, 1995-96, 56 L. Ed. 2d 554 (1978). See also the examples cited in Judge Easterbrook's opinion at 553-554.

I agree with Judge Easterbrook that the law governing the question who is an agent of an employer for Title VII purposes must be, under these cases, state law. To his discussion, I add only two observations. First, the administrative consequences of his point about "vertical" uniformity should not be underestimated. The existence of side-by-side, but inconsistent, federal and state rules on employer responsibilities for acts of their supervisory agents will make it exceedingly difficult for employers in Illinois, Indiana, and Wisconsin (not to mention other states) to establish a single coherent management structure. Second, the majority's rejection of respondeat superior liability for supervisory "hostile environment" cases means that the same woman who under state law will be able to hold her employer liable for a supervisor's violation of state civil rights statutes or for state torts will find her way to federal court barred. Victims of sexual harassment should be aware that the majority's approach has made them second-class citizens, both vis--vis their counterparts with state law claims and vis--vis other Title VII plaintiffs.

The law of agency, as the Supreme Court recognized in Meritor, is best suited to provide the answer to the question of when the employer must answer for the supervisor's actions, no matter what the supervisor's behavior. For present purposes, I discuss agency principles generally rather than undertaking a specific examination of Illinois's rules, both because I have no reason to believe that Illinois is an outlier on these matters and because in my view a detailed examination of Illinois's rules would be better performed in light of a more complete record. As reflected in Restatement § 219, agency law typically draws a fundamental distinction between acts within the scope of an agent's employment and acts outside the scope of employment. An initial question before us today is how, if at all, that structure applies to the relationship between a supervisory employee of a principal (the employer) and a subordinate employee of the same principal. A second preliminary question is whether this analysis operates differently in Title VII cases from the way it normally applies for other torts. When one considers the tripartite relationship among employer, supervisor, and subordinate employee, fully taking into account the responsibilities each has, the ability of each to take corrective action, and the incentive each has to act, the conclusions are inescapable that (a) general principles of traditional agency law provide a suitable analytical structure, and (b) nothing in the law of discrimination requires modification of this analysis.

The key relationship for the cases before us is the one between the employer and the supervisory employee, for the simple reason that the question is whether the employer should be liable for the supervisor's actions. Under established principles of law, the relationship between employer and supervisory employee is plainly that of principal to agent. See Moy v. County of Cook, 159 Ill. 2d 519, 203 Ill. Dec. 776, 640 N.E.2d 926, 927 (1994); Kansallis Finance Ltd. v. Fern, 421 Mass. 659, 659 N.E.2d 731, 733-34 (1996); Ruiz v. Conoco, 868 S.W.2d 752, 762-64 (Tex.1994); Western Elec. Co. v. Brenner, 41 N.Y.2d 291, 392 N.Y.S.2d 409, 360 N.E.2d 1091, 1094 (1977); Restatement (Second) of Agency § 218. In old-fashioned agency parlance, the employer is the "master" and the supervisory employee is the "servant." However denominated, the agency relationship exists for functional reasons: the employer hired the supervisory employee to perform specified tasks for the employer, and the employer controls or has the right to control the supervisor's conduct in performing these tasks. See Leon v. Caterpillar Indus., Inc., 69 F.3d 1326, 1333 (7th Cir.1995); United States v. Balistrieri, 981 F.2d 916, 930 (7th Cir.1992); Valenti v. Qualex, Inc., 970 F.2d 363, 368 (7th Cir.1992); Central States Trucking Co. v. J.R. Simplot Co., 965 F.2d 431, 433 (7th Cir.1992). Occasionally the question arises whether the "control" element is weak enough that the agent should be characterized as an independent contractor, see Kittlaus v. United States, 41 F.3d 327, 330 (7th Cir.1994), but that is plainly not an issue in these cases, and will virtually never be an issue in the modern business firm where the employee receives a salary for her services.

It is important to bear in mind that Title VII itself applies to all but the smallest business enterprises in our economy. Thus, when we consider the principal/agent relationship between employer and supervisor, we may be speaking of anything from a 692,800-employee company like General Motors, see Hoover's Handbook of American Business, vol. 1, at 659 (1996), with countless layers of management between the CEO and the line employees, to a firm that barely meets the jurisdictional minima required for Title VII. Cf. Walters v. Metropolitan Educational Enterprises, --- U.S. ----, 117 S. Ct. 660, 136 L. Ed. 2d 644 (1997) (describing how to compute whether an employer has 15 employees over a 20-week period annually, as required by 42 U.S.C. § 2000e(b)). Our approach must therefore be flexible enough to cover the incredible diversity of employers that fall within the ambit of the statute.

How, if at all, should the employer's responsibility for the supervisor's actions be affected if the victim of the supervisor's conduct happens also to be employed by the firm and under the supervisor's control? The employer's knowledge that one employee is under the supervision of another (to the extent dictated by the employer's own policies) certainly makes injury from the supervisor to the subordinate more foreseeable to the employer than injury to a stranger. That only underscores the importance of placing the responsibility on the employer to monitor the supervisor's job-related actions to ensure that he is not abusing the powers entrusted to him. Rather than assuming that subordinate employees have better access than, say, customers to a firm's internal mechanism for resolving discrimination complaints (although even that notion is questionable, since well-run companies normally offer complaint avenues for their customers) we should ask what job-related harm is foreseeable and who is in the best position to prevent it. Typically, we distinguish between persons who are able to contract around potential injury from a transaction and those who are strangers. On that basis, the subordinate employee stands in a position much closer to the stranger, since no amount of contracting can prevent her from becoming the victim of a supervisor's sexual harassment. See Alan O. Sykes, The Boundaries of Vicarious Liability: An Economic Analysis of the Scope of Employment Rule and Related Legal Doctrines, 101 Harv. L.Rev. 563 (1988); Alan O. Sykes, The Economics of Vicarious Liability, 93 Yale L.J. 1231 (1984). Even the common law "fellow servant" rule did not bar recovery against the employer when the injury occurred at the hands of a "vice-principal," who could be anyone to whom the master had delegated his common law duties. See Prosser and Keeton on Torts, § 80, at 572 (5th ed.1984). By analogy, the subordinate suffers injury from a "vice-principal"--literally, one who stands in the place of the principal--when a supervisor harasses her.6 Whether seen from the vantage point of foreseeability or in light of the turn-of-the-century "fellow servant" rule, employer liability follows.

If the injury flows from an intentional tort committed by the employee, such as the use of force or defamation, the employer will still be liable if the act was not unexpectable given the employee's duties. See, e.g., Deal, 130 Ill. Dec. 200, 537 N.E.2d at 272-73 (apartment complex owner liable for apartment inspector's attack on tenant); Aliota v. Graham, 984 F.2d 1350, 1358-59 (3d Cir.1993) (employer can be liable for defamation); Molton v. City of Cleveland, 839 F.2d 240, 249 (6th Cir.1988) (city potentially liable because police officer's assault and battery were not unexpectable); Rawling v. City of New Haven, 206 Conn. 100, 537 A.2d 439, 443 (1988) (genuine fact question whether police officer was acting within the scope of his employment when he committed a sexual assault); Daigle v. City of Portsmouth, 129 N.H. 561, 534 A.2d 689, 700 (1987) (tortious or criminal character of officer's acts does not automatically remove them from scope of employment). See generally Restatement §§ 245, 247. These cases demonstrate that an intentional tort is not as a matter of law unforeseeable or outside the scope of employment. (I note as well that the injuries suffered by victims of sexual harassment can include both severe psychological harm and pecuniary injury from decreased productivity and loss of job opportunities. We should not trivialize the harm that harassment can inflict.)

Neither this court nor any other has had any trouble applying these agency principles to the various claims of discrimination apart from sexual harassment that are cognizable under Title VII and related laws. See, e.g., Hennessy, supra, 69 F.3d 1344 (termination based on sex or pregnancy); Balistrieri, supra, 981 F.2d 916 (race discrimination in housing context); Miranda v. B & B Cash Grocery Store, Inc., 975 F.2d 1518 (11th Cir.1992) (disparate pay based on gender); Hunter, supra, 797 F.2d 1417 (race discrimination); Young v. Southwestern Sav. & Loan Ass'n, 509 F.2d 140 (5th Cir.1975) (hostile religious environment). Sexual harassment is no different and no harder to detect than other forms of discrimination. We know well that all forms of discrimination can be practiced subtly as well as blatantly. If an African-American employee charges racial discrimination and wants to prove it by showing that her supervisor routinely used racial epithets when talking to her, she would face the same evidentiary problems as she would if the charge were either "hostile environment" or "quid pro quo" sexual harassment. Evidentiary concerns are appropriately addressed through the summary judgment process, not by engrafting liability standards onto the law that the language of the statute does not support.

Agency principles therefore apply in Title VII cases for purposes of deciding when an employer must answer for its supervisory employees' discriminatory acts to the same extent they do elsewhere. The first question in any case dealing with a supervisor's harassment should be whether the supervisor's action was within the scope of his employment. In order to answer that factual question, criteria like those suggested in Restatement (Second) §§ 228 and 229 are useful benchmarks. Section 228 is worth quoting in its entirety, as a useful summary of the concept of "scope of employment" and how it differs from the idea of "strict liability":

(1) Conduct of a servant is within the scope of employment if, but only if:

(a) it is of the kind he is employed to perform;

(b) it occurs substantially within the authorized time and space limits;

(c) it is actuated, at least in part, by a purpose to serve the master; and

(d) if force is intentionally used by the servant against another, the use of force is not unexpectable by the master.

(2) Conduct of a servant is not within the scope of employment if it is different in kind from that authorized, far beyond the authorized time or space limits, or too little actuated by a purpose to serve the master.

In section 230, the Restatement repeats the well recognized rule that "[a]n act, although forbidden, or done in a forbidden manner, may be within the scope of employment." See City of Chicago v. Matchmaker Real Estate Sales Ctr., Inc., 982 F.2d 1086, 1096 n. 12 (7th Cir.1992). See also Aliota, 984 F.2d at 1358; Vlotho v. Hardin County, 509 N.W.2d 350, 354 (Iowa 1993); Walker v. Crigler, 976 F.2d 900, 904 n. 4 (4th Cir.1992); Huddleston v. Union Rural Elec. Ass'n, 841 P.2d 282, 292 (Colo.1992) (en banc); Ermert v. Hartford Ins. Co., 559 So. 2d 467, 476-77 (La.1990); Pyne v. Witmer, 129 Ill. 2d 351, 135 Ill. Dec. 557, 543 N.E.2d 1304, 1309 (1989); Yates v. Avco Corp., 819 F.2d 630, 636 (6th Cir.1987); Simmons v. United States, 805 F.2d 1363, 1369 (9th Cir.1986); Dickinson v. Edwards, 105 Wash.2d 457, 716 P.2d 814, 819 (1986) (en banc).

Contrary to Judge Flaum's suggestion in footnote 1 of his opinion, each of the four criteria described in § 228 imposes a meaningful limitation on the scope of employment, thereby ensuring that employers will not be held liable for the "frolics and detours" of their employees or for acts that are unforeseeable to the employer.7 His criticisms of these criteria all focus so exclusively on the harassing behavior that they would exclude all "mixed motive" cases, notwithstanding the fact that agency law has always recognized this type of liability. The focal point must be the conduct the employer has hired the supervisor to perform. It is here that Burlington and Packaging Corporation take a misstep. They argue that courts must ask whether the supervisor was employed to commit the intentional tort of sexual harassment. All an employer has to do to prevail under their view--either to show the lack of actual authority or apparent authority--is to show that the supervisor was not employed for the purpose of engaging in sexual harassment. Merely to state this is to show what a preposterous standard it would be. We have already expressly held that "[a] principal cannot free itself of liability by delegating to an agent the duty not to discriminate." Matchmaker Real Estate, 982 F.2d at 1096. That principle applies with equal force here.

In my view, the proper inquiry on this point is whether the supervisor, acting at least in part with the purpose of serving the employer, committed the acts of harassment as he exercised the authority (actual or apparent) the employer conferred on him: hiring, firing, assigning work, disciplining other employees for harassing behavior, and the like. It may help to consider the authorized time and space limits to define which incidents were work-related, although in today's mobile world they are hardly conclusive. The question whether the supervisor's conduct is actuated, at least in part, by a purpose to serve the master is undoubtedly of central importance. Agency law has always made it clear that the employer is not liable for acts of the agent that exclusively serve the agent's personal interests, just as it has also made it clear that employers can be liable if the motives are mixed. The conduct of the supervisor that must be assessed is not the act of harassment in isolation, but instead his broader course of action. If the supervisor is engaged in his own business or is acting on his own time, then his acts will normally not be within the scope of his employment unless he is acting with apparent authority. Finally, an intentional tort like sexual harassment plainly requires foreseeability. Some forms of sexual harassment are easily foreseeable, while other severe forms may be so far from the norm that the employer should not be held liable in the absence of negligence. In that connection, I believe that the Fourth Circuit may have gone too far in Martin v. Cavalier Hotel Corp., 48 F.3d 1343 (4th Cir.1995), in finding violent rape to be something undertaken at least in part with the employer's interests in mind, and thus something foreseeable to the employer.

If a court determines that a supervisor was acting within the scope of his employment or abused his supervisory authority, then employer liability follows under the well-recognized principles of agency law noted above. If the court finds to the contrary, then it must consider whether the employer is liable under any other theory. Two possibilities exist. If the employer was negligent in addressing the sexual harassment, then every member of this court agrees liability follows. See, e.g., Zimmerman v. Cook County Sheriff's Dept., 96 F.3d 1017, 1018 (7th Cir.1996); Baskerville v. Culligan Int'l Co., 50 F.3d 428, 432 (7th Cir.1995); Carr v. Allison Gas Turbine Div. Gen. Motors, 32 F.3d 1007, 1012 (7th Cir.1994) (Coffey, J., dissenting on other grounds). (I fear that employers may be baffled by the "heightened duty of care" that Judge Flaum would impose under the rubric of negligence for hostile environment cases involving supervisors: does this mean closer monitoring? a more flexible notice requirement? a duty to take more extensive preventive measures? If this becomes the law of the Circuit, I predict a considerable period of uncertainty while we answer these and similar questions.) Alternatively, as I have already noted, if the supervisor had apparent authority to take the employment actions at issue--again, not apparent authority to harass (which is impossible), but apparent authority to hire, assign, supervise, etc.--and no other limitation derived from agency law applies, then liability should result. Apparent authority is just the shadow of actual authority; it exists when third parties reasonably believe the principal gave the agent authority to conduct his affairs, even if no such agreement exists. (This means that the employer can take steps to negate the appearance of authority. Such steps might well show that the employee's belief in the supervisor's alleged power to promote or to fire, for example, would be unreasonable.) The existence of a policy against sexual harassment does not change this result, although it both can and should reduce the incidence of harassment and provide a mechanism for amicably resolving many claims.

No matter what the basis for employer liability, I do not disagree that a strong, credibly enforced policy against sexual harassment can affect a plaintiff's case at the damages stage. This is a highly fact-specific inquiry, however, which would arise only if a trier of fact found that liability existed for sexual harassment. It therefore has little bearing on the case before us today.

Proper application of the principles I have described requires an appreciation of the facts of the two cases before us. Both appeals come to us from grants of summary judgment on behalf of the respective employers. The following accounts of the facts are therefore taken in the light most favorable to the plaintiffs.

A. Jansen v. Packaging Corporation of America

Alice Jansen's experience as Al Antoni's secretary at the Tooling Services Department of Packaging Corporation of America (PCA) was not a happy one. Jansen was hired on August 2, 1991, to work for Antoni. As her supervisor, Antoni was responsible for evaluating Jansen's performance (which determined her raises) and providing input into her hiring and firing. Antoni's objectionable behavior ran the gamut from sexually suggestive remarks and unwanted touching to the delay of tangible job benefits (and the threat of further "quid pro quo" actions). At work, he frequently asked Jansen what she thought about "quickies;" he told her there was time for a "quickie" while patting his crotch; he requested oral sex; and he bragged about his former secretary's willingness to give him "quickies." In September 1992, Antoni was responsible for completing a performance review of Jansen's work. Instead of returning the review promptly to the personnel department, Antoni held it back (along with Jansen's raise) for several months. Jansen made several inquiries about the review, to which Antoni responded by patting his crotch and saying, "I haven't forgotten about your review, it's on my desk."

In the end, Antoni gave her a "fully satisfactory" review and she received her raise, retroactive to September (although with no interest, as far as the record shows), but his harassing behavior did not stop. On January 19, 1993, Jansen filed a complaint about him, informing PCA's Human Resources Director, Paul Migala, that Antoni had made unwelcome sexual comments to her. Later the same day Migala and Hank Weil (Antoni's supervisor) met with Jansen to discuss her complaint. At that meeting, Jansen suggested contacting Sonja Marciciak to corroborate her allegations and noted that she had heard that Antoni's previous secretary, Vicki Krompton Wiley, had been subjected to the same kind of harassment. Jansen named two other employees who had sexual relationships with Antoni and gave Migala the informal log she had kept of Antoni's harassing actions. At the conclusion of the meeting, Migala assured Jansen that the company would investigate her claims thoroughly.

The investigation that ensued, however, was not what Jansen had in mind. It began with an interview of Antoni, who denied all of Jansen's allegations and accused her of inappropriate behavior. At that point, rather than following up with the names of people Jansen had provided, PCA interviewed nine other employees (including Marciciak). Marciciak confirmed that Antoni had once made an unwelcome sexual comment to her; that she had seen Vicki Wiley crying because of Antoni's comments about "blow jobs" and the like; and that Jansen had complained about Antoni to her. She could not directly corroborate any of the incidents with Jansen, however, for the simple reason that she had not seen them. The other people that PCA interviewed (in lieu of Jansen's suggested witnesses) were similarly unable to back up Jansen's story directly. The investigation also focused on Antoni's accusations about Jansen. Ultimately, Migala and Weil decided that the evidence of Antoni's sexual harassment was inconclusive. They passed that word along to Jansen in a meeting, and at the same time they told her that their investigation had uncovered some potentially improper conduct on her part as well. At that time, they gave Jansen a copy of the company's sexual harassment policy. They also counseled Antoni about sexual harassment, notwithstanding their "inconclusive" investigation.

This was not the first time PCA had heard complaints about Antoni's behavior. Tami Long had been Antoni's secretary before Wiley. Long testified that soon after she started work at PCA, in December 1987, Antoni began making crude and suggestive remarks to her, eventually asking her to engage in oral sex. At one point, PCA's director of personnel called Long into her office to discuss reports of Antoni's sexual harassment. Long complied, but asked the director not to launch a formal investigation into the charges. Instead, the director "counseled" Antoni "without directly accusing him." The evidence also indicates that PCA knew that Wiley also had significant problems working for Antoni. Although she never complained directly to management about Antoni, Wiley told Migala when she left the company that it was "because of a personal conflict with Al Antoni," and that "another woman should not be put in that environment and have to go through what I did." Migala got the message. Following Wiley's resignation, Migala directly counseled Antoni (again) about sexual harassment. Jansen's complaint was thus at least the third time the company had heard about Antoni's proclivities.

After Jansen's complaint, Antoni began treating her differently. He took a "work to rule" approach to subjects like explaining her absence when she left her desk and made her the only secretary required to take her lunch break between 12:00 p.m. and 12:30 p.m. (the others could take their half-hour lunch breaks at their discretion). He stripped her of responsibilities such as answering his telephone calls and preparing the payroll, and took to locking his office, which made Jansen's job duties of delivering his mail and messages considerably more difficult. Although he restored the payroll duties when Jansen complained to Migala, Antoni delayed Jansen's performance review and her raise. Antoni also allegedly engaged in some "self-help" retaliation, slashing the tires of Jansen's car and wrapping her car antenna around her rearview mirror. Jansen ultimately filed a complaint with the EEOC claiming sexual harassment on February 26, 1993. In November 1993, the EEOC issued her right to sue letter, and she filed this suit on January 26, 1994, claiming sexual harassment and retaliation in violation of Title VII and raising several state law claims. Following discovery, the district court granted PCA's motion for summary judgment on all counts.

Kimberly Ellerth first interviewed for a marketing job at Burlington Industries in March 1993. After her initial interview with Mary Strenk Fitzgerald, a National Accounts Manager in the Mattress Ticking Division, she was invited back to interview with Theodore Slowik, the Vice President of Sales and Marketing for that division. Ellerth found her meeting with Slowik disturbing because he asked her highly personal questions such as whether she and her husband were planning on having a family and "practicing" at it, and he stared conspicuously at her breasts and legs. About a week after the Slowik interview, Burlington offered Ellerth a job in which she would report to Fitzgerald, who in turn reported to Slowik. Ellerth accepted the position and joined Fitzgerald as the only other employee in the Chicago office.

Given the management structure, Ellerth was required to see and communicate with Slowik on a regular basis. On average, she saw him one or two days every month or two, either in Chicago, in New York, or at other corporate locations. She spoke with him on the telephone approximately once a week. These encounters, both verbal and direct, were characterized by a constant barrage of sexual comments, innuendo, and occasionally more explicit threats. Before one of Ellerth's trips to New York in the summer of 1993, Slowik telephoned her and spoke suggestively to her. Once she got to New York, at least two offensive incidents occurred. First, Slowik told Ellerth a number of off-color, offensive "jokes" during a meeting in his office. Second, at a business lunch with Ellerth and Angelo Brenna, Burlington's Vice President of International Sales, Slowik told a number of sexually offensive "jokes" and rubbed Ellerth's knee under the table. She pulled her leg away when he touched it. As the three left the restaurant, Slowik walked several feet behind Ellerth with Brenna, commented "You have got great legs, Kim," and then asked Brenna what he thought. Ellerth complained about Slowik's behavior to two other Burlington employees as soon as she returned to the office.

Slowik continued making sexually offensive comments to Ellerth, both about her and about other women. These encounters were all in conjunction with business meetings. After one business dinner in Greensboro, North Carolina, Burlington's corporate headquarters, Slowik invited Ellerth to join him at the lounge of the hotel where they were both staying and she felt obliged to accept. While there, Slowik commented that the female band members had nice breasts, nice legs, and nice, skimpy outfits. Looking at Ellerth's breasts, he then said, "You are a little lacking in that area, aren't you, Kim?" When she did not respond, Slowik told her that she "ought to loosen up," and then, more threateningly, he said, "You know, Kim, I could make your job very hard or very easy at Burlington." Ellerth understood this to mean that she would have to have sex with Slowik to succeed at the company.

After the Greensboro trip, Slowik's telephone calls increased in number. Although they were brief, they normally included sexually harassing comments. Slowik would talk about Ellerth's body (particularly her legs) and would ask about her "practice" to have a family. On several occasions, Slowik refused to give her special permission to do things for customers until she described her clothing to him. He directly stated that her job would be much easier if she wore shorter skirts. On another occasion, he made comments in front of another employee that implied that Ellerth had been performing fellatio on him.

In addition to the harassing comments, Slowik repeated his unwanted touching of Ellerth on at least two occasions. He patted her rear at the annual company holiday party, and rubbed her knee during an interview for a promotion to the position of Sales Representative for the Midwest territory while asking her whether the frequent travel associated with the new position would make her husband miss her. After the interview, the previous Midwest sales representative told Ellerth that Slowik (who had final decisionmaking responsibility for the promotion) was reluctant to promote her because she was "arrogant." Patrick Lawrence, who was also involved in the interviews and also reported to Slowik, confirmed that comment. When Ellerth confronted Slowik with this information, he admitted calling her arrogant and told her he was hesitant to promote her. Ellerth asked, "[I]s it because I'm not loose enough for you, Ted?," and Slowik replied in the affirmative. Nevertheless, two weeks later Slowik called to tell Ellerth that she had received the promotion, although he could not resist adding a sexually inappropriate comment at the end of the conversation that reduced Ellerth to tears.

Burlington's employee handbook included a brief statement that "[t]he company will not tolerate any form of sexual harassment in the workplace.... If you have any questions or problems, or if you feel you have been discriminated against, you are encouraged to talk to your supervisor or human resources representative or use the grievance procedure promptly." In the next paragraph of the "Employee Relations" section, Burlington's handbook maintained that it had an "Open Door Policy":

Burlington's Open Door Policy means just what it says--management's door is always open to you whenever you have ideas, suggestions or concerns about work-related matters. Most often, you will want to speak with your supervisor first. But if you feel more comfortable talking with your human resources manager or department manager, please don't hesitate to do so. They will do their best to answer your questions or help solve a problem.

Defendant's Appendix, Exh. F at 3. Although Ellerth did not talk either to Lawrence or the human resources manager, she did make her complaints known in several ways. First, of course, she had complained directly to Slowik himself, who certainly stood in a supervisory relationship to her. Second, in early 1994, she complained to Donna Thibideau, Burlington's Customer Service Manager, that Slowik was sexually harassing her. Thibideau told her that Slowik also may have harassed another Burlington employee. Finally, Ellerth complained to several other Burlington employees, including Sherry Hester, a Customer Service Representative, and Brett Schneider, a Sales Representative.

In May 1994, Ellerth and her new supervisor, Lawrence, met to discuss ways in which they could improve office procedures. At that time, Lawrence made no mention of customer complaints about Ellerth's own performance. Shortly thereafter, Lawrence and Thibideau received two such complaints, which prompted a May 22, 1994, memorandum from Lawrence cautioning Ellerth to be more careful about returning telephone calls. On May 31, 1994, Ellerth left a message on Lawrence's answering machine telling him that she was quitting. She faxed him a letter to the same effect. Three weeks later, she sent Lawrence a detailed explanation of her reasons for quitting, which informed him directly for the first time about Slowik's harassing behavior, including the unwanted touching and lecherous looks. She explained that both she and her husband had feared that her job would be jeopardized if she had complained any more about Slowik's actions than she had done. On October 12, 1994, Ellerth filed charges of sexual harassment with the Illinois Department of Human Resources (IDHR) and the Equal Employment Opportunity Commission (EEOC). This suit followed after the EEOC sent its right to sue letter on November 30, 1994. As in Jansen's case, after some time for discovery the district court granted Burlington's motion for summary judgment on all counts.

Both PCA and Burlington raised an initial procedural objection to the plaintiff's ability to proceed on their "quid pro quo" theories at all. In Jansen's case, PCA claims that Jansen did not mention the factual basis for her "quid pro quo" claim in her EEOC complaint and thus that she should not be able to raise the claim now; it also argues that allegations about her "quid pro quo" claim were "utterly absent" from her complaint. Burlington similarly argues that Ellerth's Title VII complaint did not adequately give it notice that she was claiming "quid pro quo" harassment as well as "hostile environment" harassment. Neither PCA's nor Burlington's claims has merit even if we accept the distinction between "quid pro quo" and "hostile environment" harassment. Jansen's EEOC charge made it clear that she was complaining about harassment, and it was supported by her affidavit, which gave several specific examples of the harassing conduct. This suffices to show the kind of factual relationship between the conduct described in the EEOC charge and that raised in the lawsuit that our cases have required. See, e.g., Cheek v. Western and Southern Life Insurance Co., 31 F.3d 497, 500 (7th Cir.1994). Jansen's complaint also alleged facts sufficient to state a claim encompassing "quid pro quo" harassment under the liberal pleading standards of Fed.R.Civ.P. 8(a). See Vidimos, Inc. v. Laser Lab Ltd., 99 F.3d 217, 222 (7th Cir.1996); Bartholet v. Reishauer A.G., 953 F.2d 1073, 1078 (7th Cir.1992). Ellerth's EEOC charge and her Title VII complaint were similar. In the charge, she asserted that Slowik had sexually harassed her and that she had been "compelled to resign ... due to [the continued] hostile offensive and abusive work environment created by SLOWIK's harassment and my opposition to sexual harassment by SLOWIK." Her district court complaint alleged that she was "inappropriately touched and sexually harassed in her place of employment," and her brief in opposition to summary judgment argued that the harassment could be considered as the "quid pro quo" variety. Under my view, of course, both women's EEOC charges and complaints easily satisfy the burden of pleading that sex discrimination in the form of harassment occurred.

Under the standards outlined earlier in this opinion, both Jansen and Ellerth showed enough to survive summary judgment. Jansen's performance evaluation was held up for a significant period of time while her supervisor, Antoni (acting within the scope of his authority) both threatened her with an adverse job action if she did not have sex with him and imposed the immediate detriment of the delay when she refused to acquiesce. The threat was clearly enough to qualify as an adverse job action. The view espoused by a number of my colleagues that threats should never be enough to support liability, as a matter of law, would read out of the law all "quid pro quo" claims in which the victim acquiesces--a rule that would directly conflict with the Supreme Court's Meritor. See Meritor, 477 U.S. at 68, 106 S.Ct. at 2406. Furthermore, Jansen raised direct evidence of discrimination, see Troupe v. May Dept. Stores Co., 20 F.3d 734, 736 (7th Cir.1994). The mere fact that a jury might choose to believe PCA's alternative explanation for the delay--that it was just a consequence of bureaucratic inertia--is of no importance on summary judgment, where we must draw inferences in Jansen's favor. From Jansen's perspective, the delay acted as a continuing demand for sexual favors, which caused its own injury, and it appears to have deprived her of at least the interest she would have earned on a timely raise. See In the Matter of Milwaukee Cheese Wisconsin, Inc., 112 F.3d 845, 849 (7th Cir.1997) ("[c]ompensation deferred is compensation reduced by the time value of money").8

Jansen also showed that Antoni harassed her in numerous other ways that fell within the scope of his authority. He was going about his business for PCA, taking advantage of his supervisory powers, while he badgered her, offensively pestering her for "blow jobs" and the like, and manipulating her work assignments based on her reactions to him. The one action he took that was clearly unrelated to his supervisory position was his alleged vandalism of her car. No authority that PCA gave him enabled him, helped him, or affected his ability to take knife to tire; Jansen's remedy here lies in the tort law of Illinois, not Title VII. For the record, although it is unnecessary for me to reach the point, I agree with those who believe that Jansen's complaints to the company, combined with its previous experience with Antoni, were enough to preserve her rights under a negligence theory.

Ellerth also brought forward enough to defeat summary judgment. Her evidence indicates that Slowik, acting within the scope of his authority (and, as some would put it, using his delegated authority), refused to give her special permission for work projects until she described her physical appearance to him. Interpreted in light of his earlier comment that he "could make [her] job very hard or very easy at Burlington," a jury could find that she had to play the role of sex object in order to obtain desirable work assignments (and thus to advance in the company). The numerous incidents in which Slowik withheld permission for assignments until Ellerth satisfied his demands could, if believed by a factfinder, support a finding that harassment occurred.

Ellerth showed that Slowik, acting within the scope of his employment, poisoned the environment for her in other ways as well. As the district court conceded, the constant sexually suggestive remarks, the criticism of the size of her breasts, the importunements to be "looser," and the unwanted and uninvited touchings would have been enough to make work onerous for any woman. This court would look foolish indeed to the public at large if we appeared unable to distinguish between three friendly pats on the back and three acts of sexually oriented stroking. If a supervisor cannot tell the difference, he needs remedial counseling, not a free pass. No one wants to be a wet blanket on office romances, but the simple rule, reflected in every creditable sexual harassment policy in the country, is that unwanted touching is verboten. Would-be suitors in the office must just wait until their interest is reciprocated. Slowik knew that Ellerth did not welcome his advances, because she told him so. Obviously, there are two sides to this story, and the jury may not believe Ellerth in the end. But she is entitled on the facts she has shown thus far to try to persuade them that hers is the more credible account.

My conclusion that Slowik acted within the scope of his authority and abused his supervisory authority makes it unnecessary for me to reach the negligence issue. Those who are attempting to pigeonhole harassment cases have decided unfortunately that Ellerth waived this point by her statement disclaiming reliance on Restatement § 219(2)(b). I disagree. It is ironic that those who criticize undue reliance on the Restatement have seized on Ellerth's passing reference to its language as a way of finding that she waived the right to support her hostile environment allegations on a negligence theory; one can reject the Restatement and still turn to the Title VII rules on negligence. Indeed, her brief stressed the point that she notified a number of individuals at Burlington about Slowik's harassing behavior, which would be relevant only if she thought Burlington's knowledge might be relevant under an alternative theory of liability. If negligence were the standard for her "hostile environment" claim, I would find that Ellerth is entitled to reach a jury on this theory. Even if a company has a policy against sexual harassment, as Burlington did (though a skimpy one as these things go), it cannot put on blinders and pretend that corporate "knowledge" comes only through that channel. I do not understand the other members of this court to argue otherwise; a plaintiff is entitled to prove that the employer "knew or should have known" about the offensive conduct in any way that she can. Use of a policy is a particularly easy way to satisfy this burden, but it is not the only one. Ellerth complained to a number of Burlington employees about Slowik, including most notably Donna Thibideau, Burlington's Customer Service Manager. Thibideau thought this may not have been the first time Slowik had behaved inappropriately. Ellerth is entitled to the opportunity to prove at trial that her informal communications were sufficient to put the company on notice of the problem.

For the reasons stated above, I would REVERSE the order of summary judgment and REMAND both No. 95-3128, Jansen v. Packaging Corp., and No. 96-1361, Ellerth v. Burlington Industries, for further proceedings.