ERIE V. PAP’S A. M. (98-1161) 529 U.S. 277 (2000)
553 Pa. 348, 719 A. 2d 273, reversed and remanded.
[ O’Connor ]
[ Scalia ]
[ Stevens ]
[ Opinion of Souter ]
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version

Stevens, J., dissenting


No. 98—1161



[March 29, 2000]

Justice Stevens, with whom Justice Ginsburg joins, dissenting.

Far more important than the question whether nude dancing is entitled to the protection of the First Amendment are the dramatic changes in legal doctrine that the Court endorses today. Until now, the “secondary effects” of commercial enterprises featuring indecent entertainment have justified only the regulation of their location. For the first time, the Court has now held that such effects may justify the total suppression of protected speech. Indeed, the plurality opinion concludes that admittedly trivial advancements of a State’s interests may provide the basis for censorship. The Court’s commendable attempt to replace the fractured decision in Barnes v. Glen Theatre, Inc., 501 U.S. 560 (1991), with a single coherent rationale is strikingly unsuccessful; it is supported neither by precedent nor by persuasive reasoning.


As the preamble to Ordinance No. 75—1994 candidly acknowledges, the council of the city of Erie enacted the restriction at issue “for the purpose of limiting a recent increase in nude live entertainment within the City.” Ante, at 9. Prior to the enactment of the ordinance, the dancers at Kandyland performed in the nude. As the Court recognizes, after its enactment they can perform precisely the same dances if they wear “pasties and G-strings.” Ante, at 13; see also, ante, at 4, n.2 (Souter, J., concurring in part and dissenting in part). In both instances, the erotic messages conveyed by the dancers to a willing audience are a form of expression protected by the First Amendment. Ante, at 7.1 Despite the similarity between the messages conveyed by the two forms of dance, they are not identical.

If we accept Chief Judge Posner’s evaluation of this art form, see Miller v. South Bend, 904 F.2d 1081, 1089—1104 (CA7 1990) (en banc), the difference between the two messages is significant. The plurality assumes, however, that the difference in the content of the message resulting from the mandated costume change is “de minimis.” Ante, at 13. Although I suspect that the patrons of Kandyland are more likely to share Chief Judge Posner’s view than the plurality’s, for present purposes I shall accept the assumption that the difference in the message is small. The crucial point to remember, however, is that whether one views the difference as large or small, nude dancing still receives First Amendment protection, even if that protection lies only in the “outer ambit” of that Amendment. Ante, at 7. Erie’s ordinance, therefore, burdens a message protected by the First Amendment. If one assumes that the same erotic message is conveyed by nude dancers as by those wearing miniscule costumes, one means of expressing that message is banned;2 if one assumes that the messages are different, one of those messages is banned. In either event, the ordinance is a total ban.

The Court relies on the so-called “secondary effects” test to defend the ordinance. Ante, at 9—15. The present use of that rationale, however, finds no support whatsoever in our precedents. Never before have we approved the use of that doctrine to justify a total ban on protected First Amendment expression. On the contrary, we have been quite clear that the doctrine would not support that end.

In Young v. American Mini Theatres, Inc., 427 U.S. 50 (1976), we upheld a Detroit zoning ordinance that placed special restrictions on the location of motion picture theaters that exhibited “adult” movies. The “secondary effects” of the adult theaters on the neighborhoods where they were located–lower property values and increases in crime (especially prostitution) to name a few–justified the burden imposed by the ordinance. Id., at 54, 71, and n. 34 (plurality opinion). Essential to our holding, however, was the fact that the ordinance was “nothing more than a limitation on the place where adult films may be exhibited” and did not limit the size of the market in such speech. Id., at 71; see also id., at 61, 63, n. 18, 70, 71, n. 35. As Justice Powell emphasized in his concurrence:

“At most the impact of the ordinance on [the First Amendment] interests is incidental and minimal. Detroit has silenced no message, has invoked no censorship, and has imposed no limitation upon those who wish to view them. The ordinance is addressed only to the places at which this type of expression may be presented, a restriction that does not interfere with content. Nor is there any significant overall curtailment of adult movie presentations, or the opportunity for a message to reach an audience.” Id., at 78—79.

See also id., at 81, n. 4 (“[A] zoning ordinance that merely specifies where a theater may locate, and that does not reduce significantly the number or accessibility of theaters presenting particular films, stifles no expression”).

In Renton v. Playtime Theatres, Inc., 475 U.S. 41 (1986), we upheld a similar ordinance, again finding that the “secondary effects of such theaters on the surrounding community” justified a restrictive zoning law. Id., at 47. We noted, however, that “[t]he Renton ordinance, like the one in American Mini Theatres, does not ban adult theaters altogether,” but merely “circumscribe[s] their choice as to location.” Id., at 46, 48; see also id., at 54 (“In our view, the First Amendment requires … that Renton refrain from effectively denying respondents a reasonable opportunity to open and operate an adult theater within the city …”). Indeed, in both Renton and American Mini Theatres, the zoning ordinances were analyzed as mere “time, place, and manner” regulations.3 See Renton, 475 U.S., at 46; American Mini Theatres, 427 U.S., at 63, and n. 18; id., at 82, n. 6. Because time, place, and manner regulations must “leave open ample alternative channels for communication of information,” Ward v. Rock Against Racism, 491 U.S. 781, 791 (1989), a total ban would necessarily fail that test.4

And we so held in Schad v. Mount Ephraim, 452 U.S. 61 (1981). There, we addressed a zoning ordinance that did not merely require the dispersal of adult theaters, but prohibited them altogether. In striking down that law, we focused precisely on that distinction, holding that the secondary effects analysis endorsed in the past did not apply to an ordinance that totally banned nude dancing: “The restriction [in Young v. American Mini Theatres] did not affect the number of adult movie theaters that could operate in the city; it merely dispersed them. The Court did not imply that a municipality could ban all adult theaters–much less all live entertainment or all nude dancing–from its commercial districts citywide.” Id., at 71 (plurality opinion); see also id., at 76; id., at 77 (Blackmun, J., concurring) (joining plurality); id., at 79 (Powell, J., concurring) (same).

The reason we have limited our secondary effects cases to zoning and declined to extend their reasoning to total bans is clear and straightforward: A dispersal that simply limits the places where speech may occur is a minimal imposition whereas a total ban is the most exacting of restrictions. The State’s interest in fighting presumed secondary effects is sufficiently strong to justify the former, but far too weak to support the latter, more severe burden.5 Yet it is perfectly clear that in the present case–to use Justice Powell’s metaphor in American Mini Theatres–the city of Erie has totally silenced a message the dancers at Kandyland want to convey. The fact that this censorship may have a laudable ulterior purpose cannot mean that censorship is not censorship. For these reasons, the Court’s holding rejects the explicit reasoning in American Mini Theatres and Renton and the express holding in Schad.

The Court’s use of the secondary effects rationale to permit a total ban has grave implications for basic free speech principles. Ordinarily, laws regulating the primary effects of speech, i.e., the intended persuasive effects caused by the speech, are presumptively invalid. Under today’s opinion, a State may totally ban speech based on its secondary effects–which are defined as those effects that “happen to be associated” with speech, Boos v. Barry, 485 U.S. 312, 320—321 (1988); see ante, at 10–yet the regulation is not presumptively invalid. Because the category of effects that “happen to be associated” with speech includes the narrower subset of effects caused by speech, today’s holding has the effect of swallowing whole a most fundamental principle of First Amendment jurisprudence.


The Court’s mishandling of our secondary effects cases is not limited to its approval of a total ban. It compounds that error by dramatically reducing the degree to which the State’s interest must be furthered by the restriction imposed on speech, and by ignoring the critical difference between secondary effects caused by speech and the incidental effects on speech that may be caused by a regulation of conduct.

In what can most delicately be characterized as an enormous understatement, the plurality concedes that “requiring dancers to wear pasties and G-strings may not greatly reduce these secondary effects.” Ante, at 20. To believe that the mandatory addition of pasties and a G-string will have any kind of noticeable impact on secondary effects requires nothing short of a titanic surrender to the implausible. It would be more accurate to acknowledge, as Justice Scalia does, that there is no reason to believe that such a requirement “will at all reduce the tendency of establishments such as Kandyland to attract crime and prostitution, and hence to foster sexually transmitted disease.” Ante, at 10 (opinion concurring in judgment); see also ante, at 4, n. 2 (Souter, J., concurring in part and dissenting in part). Nevertheless, the plurality concludes that the “less stringent” test announced in United States v. O’Brien, 391 U.S. 367 (1968), “requires only that the regulation further the interest in combating such effects,” ante, at 20; see also ante, at 8. It is one thing to say, however, that O’Brien is more lenient than the “more demanding standard” we have imposed in cases such as Texas v. Johnson, 491 U.S. 397 (1989). See ante, at 8. It is quite another to say that the test can be satisfied by nothing more than the mere possibility of de minimis effects on the neighborhood.

The Court is also mistaken in equating our secondary effects cases with the “incidental burdens” doctrine applied in cases such as O’Brien; and it aggravates the error by invoking the latter line of cases to support its assertion that Erie’s ordinance is unrelated to speech. The incidental burdens doctrine applies when “ ‘speech’ and ‘nonspeech’ elements are combined in the same course of conduct,” and the government’s interest in regulating the latter justifies incidental burdens on the former. O’Brien, 391 U.S., at 376. Secondary effects, on the other hand, are indirect consequences of protected speech and may justify regulation of the places where that speech may occur. See American Mini Theatres, 427 U.S., at 71, n. 34 (“[A] concentration of ‘adult’ movie theaters causes the area to deteriorate and become a focus of crime”).6 When a State enacts a regulation, it might focus on the secondary effects of speech as its aim, or it might concentrate on nonspeech related concerns, having no thoughts at all with respect to how its regulation will affect speech–and only later, when the regulation is found to burden speech, justify the imposition as an unintended incidental consequence.7 But those interests are not the same, and the Court cannot ignore their differences and insist that both aims are equally unrelated to speech simply because Erie might have “recogniz[ed]” that it could possibly have had either aim in mind. See ante, at 14.8 One can think of an apple and an orange at the same time; that does not turn them into the same fruit.

Of course, the line between governmental interests aimed at conduct and unrelated to speech, on the one hand, and interests arising out of the effects of the speech, on the other, may be somewhat imprecise in some cases. In this case, however, we need not wrestle with any such difficulty because Erie has expressly justified its ordinance with reference to secondary effects. Indeed, if Erie’s concern with the effects of the message were unrelated to the message itself, it is strange that the only means used to combat those effects is the suppression of the message.9 For these reasons, the Court’s argument that “this case is similar to O’Brien,” ante, at 9; see also ante, at 13, is quite wrong, as are its citations to Clark v. Community for Creative Non-Violence, 468 U.S. 288 (1984), and Ward v. Rock Against Racism, 491 U.S. 781 (1989), ante, at 12—14, neither of which involved secondary effects. The Court cannot have its cake and eat it too–either Erie’s ordinance was not aimed at speech and the Court may attempt to justify the regulation under the incidental burdens test, or Erie has aimed its law at the secondary effects of speech, and the Court can try to justify the law under that doctrine. But it cannot conflate the two with the expectation that Erie’s interests aimed at secondary effects will be rendered unrelated to speech by virtue of this doctrinal polyglot.

Correct analysis of the issue in this case should begin with the proposition that nude dancing is a species of expressive conduct that is protected by the First Amendment. As Chief Judge Posner has observed, nude dancing fits well within a broad, cultural tradition recognized as expressive in nature and entitled to First Amendment protection. See 904 F.2d, at 1089—1104; see also Note, 97 Colum. L. Rev. 1844 (1997). The nudity of the dancer is both a component of the protected expression and the specific target of the ordinance. It is pure sophistry to reason from the premise that the regulation of the nudity component of nude dancing is unrelated to the message conveyed by nude dancers. Indeed, both the text of the ordinance and the reasoning in the Court’s opinion make it pellucidly clear that the city of Erie has prohibited nude dancing “precisely because of its communicative attributes.” Barnes, 501 U.S., at 577 (Scalia, J., concurring in judgment) (emphasis in original); see id., at 596 (White, J., dissenting).


The censorial purpose of Erie’s ordinance precludes reliance on the judgment in Barnes as sufficient support for the Court’s holding today. Several differences between the Erie ordinance and the statute at issue in Barnes belie the Court’s assertion that the two laws are “almost identical.” Ante, at 8. To begin with, the preamble to Erie’s ordinance candidly articulates its agenda, declaring:

“Council specifically wishes to adopt the concept of Public Indecency prohibited by the laws of the State of Indiana, which was approved by the U.S. Supreme Court in Barnes vs. Glen Theatre Inc., … for the purpose of limiting a recent increase in nude live entertainment within the City.” App. to Pet. for Cert. 42a (emphasis added); see also ante, at 9.10

As its preamble forthrightly admits, the ordinance’s “purpose” is to “limi[t]” a protected form of speech; its invocation of Barnes cannot obliterate that professed aim.11

Erie’s ordinance differs from the statute in Barnes in another respect. In Barnes, the Court expressly observed that the Indiana statute had not been given a limiting construction by the Indiana Supreme Court. As presented to this Court, there was nothing about the law itself that would confine its application to nude dancing in adult entertainment establishments. See Barnes, 501 U.S., at 564, n. 1 (discussing Indiana Supreme Court’s lack of a limiting construction); see also id., at 585, n. 2 (Souter, J., concurring). Erie’s ordinance, however, comes to us in a much different posture. In an earlier proceeding in this case, the Court of Common Pleas asked Erie’s counsel “what effect would this ordinance have on theater . . . productions such as Equus, Hair, O[h!] Calcutta[!]? Under your ordinance would these things be prevented … ?” Counsel responded: “No, they wouldn’t, Your Honor.” App. 53.12 Indeed, as stipulated in the record, the city permitted a production of Equus to proceed without prosecution, even after the ordinance was in effect, and despite its awareness of the nudity involved in the production. Id., at 84.13 Even if, in light of its broad applicability, the statute in Barnes was not aimed at a particular form of speech, Erie’s ordinance is quite different. As presented to us, the ordinance is deliberately targeted at Kandyland’s type of nude dancing (to the exclusion of plays like Equus), in terms of both its applicable scope and the city’s enforcement.14

This narrow aim is confirmed by the expressed views of the Erie City Councilmembers who voted for the ordinance. The four city councilmembers who approved the measure (of the six total councilmembers) each stated his or her view that the ordinance was aimed specifically at nude adult entertainment, and not at more mainstream forms of entertainment that include total nudity, nor even at nudity in general. One lawmaker observed: “We’re not talking about nudity. We’re not talking about the theater or art … . We’re talking about what is indecent and immoral… . We’re not prohibiting nudity, we’re prohibiting nudity when it’s used in a lewd and immoral fashion.” App. 39. Though not quite as succinct, the other councilmembers expressed similar convictions. For example, one member illustrated his understanding of the aim of the law by contrasting it with his recollection about high school students swimming in the nude in the school’s pool. The ordinance was not intended to cover those incidents of nudity: “But what I’m getting at is [the swimming] wasn’t indecent, it wasn’t an immoral thing, and yet there was nudity.” Id., at 42. The same lawmaker then disfavorably compared the nude swimming incident to the activities that occur in “some of these clubs” that exist in Erie–clubs that would be covered by the law. Ibid.15 Though such comments could be consistent with an interest in a general prohibition of nudity, the complete absence of commentary on that broader interest, and the councilmembers’ exclusive focus on adult entertainment, is evidence of the ordinance’s aim. In my view, we need not strain to find consistency with more general purposes when the most natural reading of the record reflects a near obsessive preoccupation with a single target of the

The text of Erie’s ordinance is also significantly different from the law upheld in Barnes. In Barnes, the statute defined “nudity” as “the showing of the human male or female genitals” (and certain other regions of the body) “with less than fully opaque covering.” 501 U.S., at 569, n. 2. The Erie ordinance duplicates that definition in all material respects, but adds the following to its definition of “nudity”:

“[T]he exposure of any device, costume, or covering which gives the appearance of or simulates the genitals, pubic hair, natal cleft, perineum anal region or pubic hair region; or the exposure of any device worn as a cover over the nipples and/or areola of the female breast, which device simulates and gives the realistic appearance of nipples and/or areola.” Ante, at 2, n. (emphasis added).

Can it be doubted that this out-of-the-ordinary definition of “nudity” is aimed directly at the dancers in establishments such as Kandyland? Who else is likely to don such garments? 17 We should not stretch to embrace fanciful explanations when the most natural reading of the ordinance unmistakably identifies its intended target.

It is clear beyond a shadow of a doubt that the Erie ordinance was a response to a more specific concern than nudity in general, namely, nude dancing of the sort found in Kandyland.18 Given that the Court has not even tried to defend the ordinance’s total ban on the ground that its censorship of protected speech might be justified by an overriding state interest, it should conclude that the ordinance is patently invalid. For these reasons, as well as the reasons set forth in Justice White’s dissent in Barnes, I respectfully dissent.


1. Respondent does not contend that there is a constitutional right to engage in conduct such as lap dancing. The message of eroticism conveyed by the nudity aspect of the dance is quite different from the issue of the proximity between dancer and audience. Respondent’s contention is not that Erie has focused on lap dancers, see ante, at 7 (Scalia, J., concurring), but that it has focused on the message conveyed by nude dancing.

2. Although nude dancing might be described as one protected “means” of conveying an erotic message, it does not follow that a protected message has not been totally banned simply because there are other, similar ways to convey erotic messages. See ante, at 11—12. A State’s prohibition of a particular book, for example, does not fail to be a total ban simply because other books conveying a similar message are available.

3. The Court contends, ante, at 14, that Ward v. Rock Against Racism, 491 U.S. 781 (1989), shows that we have used the secondary effects rationale to justify more burdensome restrictions than those approved in Renton and American Mini Theatres. That argument is unpersuasive for two reasons. First, as in the two cases just mentioned, the regulation in Ward was as a time, place, and manner restriction. See 491 U.S., at 791; id., at 804 (Marshall, J., dissenting). Second, as discussed below, Ward is not a secondary effects case. See infra, at 9—10.

4. We also held in Renton that in enacting its adult theater zoning ordinance, the city of Renton was permitted to rely on a detailed study conducted by the city of Seattle that examined the relationship between zoning controls and the secondary effects of adult theaters. (It was permitted to rely as well on “the ‘detailed findings’ summarized” in an opinion of the Washington Supreme Court to the same effect.) 475 U.S., at 51—52. Renton, having identified the same problem in its own city as that experienced in Seattle, quite logically drew on Seattle’s experience and adopted a similar solution. But if Erie is relying on the Seattle study as well (as the Court suggests, ante, at 16), its use of that study is most peculiar. After identifying a problem in its own city similar to that in Seattle, Erie has implemented a solution (pasties and G-strings) bearing no relationship to the efficacious remedy identified by the Seattle study (dispersal through zoning). But the city of Erie, of course, has not in fact pointed to any study by anyone suggesting that the adverse secondary effects of commercial enterprises featuring erotic dancing depends in the slightest on the precise costume warn by the performers–it merely assumes it to be so. See infra, at 7—8. If the city is permitted simply to assume that a slight addition to the dancers’ costumes will sufficiently decrease secondary effects, then presumably the city can require more and more clothing as long as any danger of adverse effects remains.

5. As the Court recognizes by quoting my opinion in Young v. American Mini Theatres, Inc., 427 U.S. 50, 70 (1976), see ante, at 13, “the First Amendment will not tolerate the total suppression of erotic materials that have some artistic value,” though it will permit zoning regulations.

6. A secondary effect on the neighborhood that “happen[s] to be associated with” a form of speech is, of course, critically different from “the direct impact of speech on its audience.” Boos, 485 U.S., at 320—321. The primary effect of speech is the persuasive effect of the message itself.

7. In fact, the very notion of focusing in on incidental burdens at the time of enactment appears to be a contradiction in terms. And if it were not the case that there is a difference between laws aimed at secondary effects and general bans incidentally burdening speech, then one wonders why Justices Scalia and Souter adopted such strikingly different approaches in Barnes.

8. I frankly do not understand the Court’s declaration that a State’s interest in the secondary effects of speech that “happen to be associated” with the speech are not “related” to the speech. Ante, at 12. See, e.g., Webster’s Third International Dictionary 132 (1966) (defining “associate” as “closely related”). Sometimes, though, the Court says that the secondary effects are “caused” by the speech, rather than merely “associated with” the speech. See, e.g., ante at 10, 12, 16, 19. If that is the definition of secondary effects the Court adopts, then it is even more obvious that an interest in secondary effects is related to the speech at issue. See Barnes, 501 U.S., at 585—586 (Souter, J., concurring) (secondary effects are not related to speech because their connection to speech is only one of correlation, not causation).

9. As Justice Powell said in his concurrence in Young v. American Mini Theatres, 427 U.S., at 82, n. 4: “[H]ad [Detroit] been concerned with restricting the message purveyed by adult theaters, it would have tried to close them or restrict their number rather than circumscribe their choice as to location.” Quite plainly, Erie’s total ban evinces its concern with the message being regulated.

10. The preamble also states: “[T]he Council of the City of Erie has [found] … that certain lewd, immoral activities carried on in public places for profit . . . lead to the debasement of both women and men … .” App. to Pet. for Cert. 41a.

11. Relying on five words quoted from the Supreme Court of Pennsylvania, the Court suggests that I have misinterpreted that Court’s reading of the preamble. Ante, at 9. What follows, however, is a more complete statement of what that Court said on this point:
“We acknowledge that one of the purposes of the Ordinance is to combat negative secondary effects. That, however, is not its only goal. Inextricably bound up with this stated purpose is an unmentioned purpose that directly impacts on the freedom of expression: that purpose is to impact negatively on the erotic message of the dance. . . . We believe . . . that the stated purpose for promulgating the Ordinance is inextricably linked with the content-based motivation to suppress the expressive nature of nude dancing.” 553 Pa. 348, 359, 719 A. 2d 273, 279 (1998).

12. In my view, Erie’s categorical response forecloses Justice Scalia’s assertion that the city’s position on Equus and Hair was limited to “[o]ne instance,” where “the city was [not] aware of the nudity,” and “no one had complained.” Ante, at 8 (concurring opinion). Nor could it be contended that selective applicability by stipulated enforcement should be treated differently from selective applicability by statutory text. See Barnes, 501 U.S., at 574 (Scalia, J., concurring) (selective enforcement may affect a law’s generality). Were it otherwise, constitutional prohibitions could be circumvented with impunity.

13. The stipulation read: “The play, ‘Equus’ featured frontal nudity and was performed for several weeks in October/November 1994 at the Roadhouse Theater in downtown Erie with no efforts to enforce the nudity prohibition which became effective during the run of the play.”

14. Justice Scalia argues that Erie might have carved out an exception for Equus and Hair because it guessed that this Court would consider them protected forms of expression, see Southeastern Promotions, Ltd. v. Conrad, 420 U.S. 546, 550, 557—558 (1975) (holding that Hair, including the “group nudity and simulated sex” involved in the production, is protected speech); in his view, that makes the distinction unobjectionable and renders the ordinance no less of a general law. Ante, at 9 (concurring opinion). This argument appears to contradict his earlier definition of a general law: “A law is ‘general’ … if it regulates conduct without regard to whether that conduct is expressive.” Barnes v. Glen Theatre, Inc., 501 U.S. 560, 575, n. 3 (1991) (opinion concurring in judgment). If the ordinance regulates conduct (public nudity), it does not do so without regard to whether the nudity is expressive if it exempts the public nudity in Hair precisely “because of its expressive content.” Ante, at 9, n. 6 (concurring opinion). Moreover, if Erie exempts Hair because it wants to avoid a conflict with the First Amendment (rather than simply to exempt instances of nudity it finds inoffensive), that rationale still does not explain why Hair is exempted but Kandyland is not, since Barnes held that both are constitutionally protected. Justice Scalia also states that even if the ordinance singled out nude dancing, he would not strike down the law unless the dancing was singled out because of its message. Ante, at 9 (concurring opinion). He opines that here, the basis for singling out Kandyland is morality. Ante, at 9. But since the “morality” of the public nudity in Hair is left untouched by the ordinance, while the “immorality” of the public nudity in Kandyland is singled out, the distinction cannot be that “nude public dancing itself is immoral.” Ante, at 10 (emphasis in original). Rather, the only arguable difference between the two is that one’s message is more immoral than the other’s.

15. Other members said their focus was on “bottle clubs,” and the like, App. 43, and attempted to downplay the effect of the ordinance by acknowledging that “the girls can wear thongs or a G-string and little pasties that are smaller than a diamond.” Ibid. Echoing that focus, another member stated that “[t]here still will be adult entertainment in this town, only it will be in a little different form.” Id., at 47.

16. The Court dismisses this evidence, declaring that it “will not strike down an otherwise constitutional statute on the basis of an alleged illicit motive.” Ante, at 11 (citing United States v. O’Brien, 391 U.S. 367, 382—383 (1968); Renton v. Playtime Theatres, Inc., 475 U.S. 41, 47—48 (1986)). First, it is worth pointing out that this doctrinaire formulation of O’Brien’s cautionary statement is overbroad. See generally L. Tribe, American Constitutional Law §12—5, pp. 819—820 (2d ed. 1988). Moreover, O’Brien itself said only that we would not strike down a law “on the assumption that a wrongful purpose or motive has caused the power to be exerted,” 391 U.S., at 383 (emphasis added; internal quotation marks omitted), and that statement was due to our recognition that it is a “hazardous matter” to determine the actual intent of a body as large as Congress “on the basis of what fewer than a handful of Congressmen said about [a law],” id., at 384. Yet neither consideration is present here. We need not base our inquiry on an “assumption,” nor must we infer the collective intent of a large body based on the statements of a few, for we have in the record the actual statements of all the city councilmembers who voted in favor of the ordinance.

17. Is it seriously contended (as would be necessary to sustain the ordinance as a general prohibition) that, when crafting this bizarre definition of “nudity,” Erie’s concern was with the use of simulated nipple covers on “nude beaches and [by otherwise] unclothed purveyors of hot dogs and machine tools”? Barnes, 501 U.S., at 574 (Scalia, J., concurring in judgment); see also ante, at 7 (Scalia, J., concurring). It is true that one might conceivably imagine that is Erie’s aim. But it is far more likely that this novel definition was written with the Kandyland dancers and the like in mind, since they are the only ones covered by the law (recall that plays like Equus are exempted from coverage) who are likely to utilize such unconventional clothing.

18. The Court states that Erie’s ordinance merely “replaces and updates provisions of an ‘Indecency and Immorality’ ordinance” from the mid-19th century, just as the statute in Barnes did. Ante, at 8—9. First of all, it is not clear that this is correct. The record does indicate that Erie’s Ordinance No. 75—1994 updates an older ordinance of similar import. Unfortunately, that old regulation is not in the record. Consequently, whether the new ordinance merely “replaces” the old one is a matter of debate. From statements of one councilmember, it can reasonably be inferred that the old ordinance was merely a residential zoning restriction, not a total ban. See App. 43. If that is so, it leads to the further question why Erie felt it necessary to shift to a total ban in 1994. But even if the Court’s factual contention is correct, it does not undermine the points I have made in the text. In Barnes, the point of noting the ancient pedigree of the Indiana statute was to demonstrate that its passage antedated the appearance of adult entertainment venues, and therefore could not have been motivated by the presence of those establishments. The inference supposedly rebutted in Barnes stemmed from the timing of the enactment. Here, however, the inferences I draw depend on the text of the ordinance, its preamble, its scope and enforcement, and the comments of the councilmembers. These do not depend on the timing of the ordinance’s enactment.