When does the application of a law raise a First Amendment concern? Though seemingly an elementary question, the answer often turns out to be quite ambiguous. If a law's application does implicate the First Amendment, courts employ various tests to determine whether there is a First Amendment violation. Content-based regulations usually trigger very exacting scrutiny;(1) but if the targeted speech falls within certain narrowly-defined categories, the scrutiny decreases.(2) Content-neutral provisions generally provoke a more intermediate level of review.(3)
To trigger any of these tests, however, the law's application must as a threshold implicate the First Amendment at all. Understandably, laws that aim at speech or that in most applications affect expressive activities will always raise a First Amendment issue, Coptent-based provisions fit in this category, as do content-neutral laws that directly target speech (such as laws prohibiting leafleting,(4) handbill distribution,(5) or posting of signs(6)) or that inevitably burden expression(7) (such as laws imposing a special tax on paper and ink,(8) banning honoraria for speeches,(9) or barring criminals from selling their stories(10)). The doctrinal confusion arises with respect to another type of content-neutral provision: "generally applicable" laws that primarily aim at nonexpressive activities,(11) but that in some applications "incidentally" restrict speech.(12) Free speech doctrine is unclear when such incidental restrictions raise a First Amendment concern.
Intuitively, some incidental restraints seem to implicate the First Amendment. For example, the use of a noise ordinance to halt a political rally may seem to warrant First Amendment review, even though the law applies generally to both noisy speeches and noisy jackhammers.(13) On the other hand, an increase in general corporate tax rates would impede the publishing activities of a book company, but enforcing the rate change against the publisher may seem not to raise any First Amendment issue.(14) Although intuition might suggest that the noise law but not the tax law should trigger First Amendment scrutiny, free speech doctrine is far from clear why (and whether) the laws do in fact have different constitutional implications. Specifically, in what circumstances do incidental effects like those produced by the tax and noise laws implicate the First Amendment? The answer is significant, for it determines whether government must come forward with justification sufficient to withstand First Amendment scrutiny. Yet the Supreme Court has paid the question little attention,(15) and its cases examining incidental restraints seem to point to conflicting conclusions. A closer analysis, however, reveals an explanation that reconciles these cases, and that in general rationalizes the Court's approach to incidental restrictions of speech.
The Supreme Court could of course adopt either of two blanket rules, both of which are plausible-that incidental restraints should never raise a constitutional concern, or that they should always trigger some level of First Amendment scrutiny. The Court has embraced the former view in its free exercise jurisprudence: Employment Division v. Smith holds that the incidental effect of enforcing a generally applicable law against religiously motivated action never concerns the Free Exercise Clause.(16) Justice Scalia, who wrote the Court's opinion in Smith, would also completely except incidental restrictions of speech from First Amendment coverage.(17) Adopting his position would signal that the Court's primary concern when dealing with incidental restraints is with legislative motivation: Since generally applicable laws by definition target nonexpressive activities, their legislative purpose most likely is not related to suppressing speech; thus, if legislative motivation is the central consideration, their application in a way that incidentally affects expression would never implicate the First Amendment.(18)
But the Court has not subscribed to this blanket rule. Sometimes, laws not aimed at speech must nevertheless satisfy a First Amendment balancing test if their application incidentally burdens expressive activities. For instance, a law that prohibits sleeping in public parks normally does not affect speech; but in Clark v. Community for Creative Non-violence, the Supreme Court found that such a regulation must satisfy some level of First Amendment scrutiny when enforced against demonstrators who camp overnight to protest homelessness.(19) And in its famous decision in United States v. O'Brien the Court made clear that "incidental limitations of First Amendment freedoms" at least sometimes raise a First Amendment concern,(20) and set forth a test--essentially a form of intermediate scrutiny-that it would apply to incidental restraints.(21) O'Brien, Clark, and other cases subjecting incidental restrictions to First Amendment scrutiny verify that the Court does not regard all generally applicable laws outside the scope of First Amendment review,(22) and that legislative motivation is not the only consideration guiding the Court's treatment of incidental restraints.
A different consideration, speech-restrictive effect, would justify the opposite blanket rule--that incidental restraints should always trigger some level of First Amendment scrutiny. The speech restrictive effect of incidental restrictions can be quite substantial. In fact, incidental restraints can produce precisely the same consequences for expressive activities as direct, even content-based, restrictions.(23) For instance, the general regulation banning camping in public parks has the same restrictive effect on overnight demonstrations as does a law that specifically prohibits protesting the plight of the homeless by camping, even though in the former case the impact is incidental while in the latter it is direct. A similar comparison can be made whenever a general law incidentally restricts speech.(24) In this light, a focus on speech-restrictive effect suggests that incidental restraints should always raise a First Amendment concern. Justice Souter, joined by three other Justices, seemed to endorse this view in his dissent in Cohen v. Cowles Media,(25) and Justice Blackmun voiced a similar view on behalf of two other Justices in his dissent in Arcara v. Cloud Books.(26)
The Court, though, has not adopted this blanket rule either, sometimes refusing to apply any First Amendment scrutiny to laws whose application incidentally burdens speech. It has, for instance, upheld enforcement of labor, antitrust, and nuisance provisions against newspapers and bookstores without engaging in any balancing analysis whatsoever, despite the restrictive effect on expressive activities.(27) Its reluctance to subject every incidental restraint to First Amendment scrutiny stems from a fear that this would require subjecting an enormous range of laws to a constitutional balancing analysis. In some sense, in fact, every law could incidentally restrict speech in certain applications:(28) A parking ordinance incidentally restricts speech when applied against an illegally-parked newspaper delivery van; the tax code burdens speech when used to tax a book publisher. A view that would subject all incidental burdens to First Amendment scrutiny thus could effectively obligate courts to apply a First Amendment analysis to every law.
In the Court's view, such profound implications for judicial review suggest that incidental restraints simply cannot always trigger First Amendment scrutiny--the resulting deluge of First Amendment claims could overwhelm the courts with constitutional balancing inquiries.(29) "Any other conclusion," Justice O'Connor maintains, "would lead to the absurd result that any government action that had some conceivable speech-inhibiting consequences, such as the arrest of a newscaster for a traffic violation, would require analysis under the First Amendment."(30) This concern with "First Amendment overload" is somewhat unclear, for the balancing inquiry need not be an involved one in every case; the Court could quickly dispense with situations like the arrest of a newscaster for speeding by engaging in only a pro forma balancing.(31) Perhaps the Court's uneasiness with excessive balancing stems from a fear that overuse of First Amendment scrutiny would trivialize the significance of applying First Amendment protections. Or perhaps the Court is reluctant to subject all generally applicable laws to a "least restrictive means" examination (there is no rationality review in First Amendment analysis),(32) even the weaker form of that analysis associated with the O'Brien standard.(33) Or perhaps the Court simply, and rather understandably, thinks it inappropriate to include every incidental effect--even if as remote as that produced by a speeding law--within the reach of the First Amendment.(34) But whatever the reason, the Court clearly desires some limiting principle, and so while incidental restrictions sometimes trigger First Amendment scrutiny, they do not always.
The Court thus has not adopted an all or nothing philosophy-incidental restraints sometimes implicate the First Amendment and sometimes do not. From a doctrinal perspective, identifying the specific circumstances in which they trigger a First Amendment analysis is plainly significant: The difference between applying some scrutiny and no scrutiny whatsoever is, at bottom, a doctrinal distinction, that not only determines whether the government must at the very least justify the incidental burden under some balancing test, but also in a broader sense delimits the scope of the First Amendment. From a practical perspective, the distinction may seem less critical, since the Supreme Court thus far has applied the O'Brien analysis with little bite when examining incidental restrictions.(35) Yet the distinction is not without practical significance. The Court would likely engage in a more searching inquiry of incidental effects in some circumstances, for instance if the speech occurs in a public forum;(36) and the presence of a public forum only affects the level of scrutiny, not the threshold question whether a law's application implicates the First Amendment at all-after all, an individual arrested for speeding could no more demand a First Amendment balancing, might be more inclined (and have more opportunities) than the Supreme Court to find incidental restrictions inconstitutional, even if the the Supreme Court--applying the same analysis--would have reached a different conclusion;(37) they of course would not have that opportunity if the Supreme Court made clear that no First Amendment scrutiny should apply in the first place.
But what in fact has been the Court's message in this regard? While we now know that the Court rejects both of the blanket rules, and that consequently incidental restrictions sometimes trigger First Amendment review and sometimes do not, what are the particular circumstances in which they raise a First Amendment concern? In Arcara v. Cloud Books,(38) the Court undertook its most meaningful examination of this question.
Arcara involved a nuisance statute authorizing closure for one year of any facility used for prostitution.(39) Although typically the nuisance law would not affect expressive activities, in Arcara the statute was enforced against an adult bookstore at which prostitution and various illicit sexual activities had taken place.(40) In these particular circumstances, consequently, the closure order encumbered the store's First Amendment protected bookselling activities. In response to this incidental effect, the New York Court of Appeals subjected the nuisance provision to O'Brien scrutiny, and found the law's application to the bookstore unconstitutional under that standard.(41)
The Supreme Court, however, concluded that applying the nuisance law to the bookstore did not raise a First Amendment concern at all, and that the lower court therefore erred by subjecting the law to a balancing analysis in the first place. According to the Court, general laws that do not target expressive activities (and instead aim at "nonspeech"--prostitution in this case),(42) only merit a First Amendment inquiry "where it was conduct with a significant expressive element that drew the legal remedy in the first place, . . . or where [the law] has the inevitable effect of singling out those engaged in expressive activity."(43) Since prostitution, not bookselling, provoked enforcement of the nuisance law against the bookstore,(44) and since the closure provision applied to expressive and nonexpressive entities alike,(45) neither of these circumstances existed in Arcara.
The Court had never before articulated the rule that a law which does not aim at speech only implicates the First Amendment if "conduct with a significant expressive element ... [draws] the legal remedy" or if the law "has the inevitable effect of singling out ... expressive activity."(46) The last part of the rule (the "inevitable effect" portion) deals primarily with content-neutral laws whose impact on speech is fairly direct.(47) The first part, though, represents the Court's most significant delineation of the circumstances in which incidental restrictions merit First Amendment review. Under this aspect of the Arcara rule, the First Amendment implications of incidental restraints apparently depend upon the character of the activity that "draws" the law's application in any particular case: Only if this enforcement trigger has "a significant expressive element" does the law raise a First Amendment concern.
The rule is somewhat ambiguous. First, it is unclear when the behavior that provokes the law's enforcement has a "significant enough expressive element" to warrant a First Amendment analysis. That is, the rule fails to specify precisely what must be the relationship between the enforcement trigger and the suppressed speech. For instance, is it enough that the conduct that draws the law's enforcement facilitates speech in some manner? Or must the enforcement trigger be necessary to some particular expressive act? Or, most narrowly, must the activity that draws the law's enforcement itself be expressive? More generally, it is somewhat puzzling why the Court only cares about incidental effects when the activity that draws the law's application has expressive aspects--the speech-restrictive effect may be far more pronounced even when the enforcement trigger is wholly nonexpressive.(48)
To address these ambiguities, it is useful to examine the cases the Court cited in Arcara as supporting its newly fashioned rule. The Court first identified two decisions involving "symbolic conduct," the draft card burning case, United States v. O'Brien,(49) and the expressive camping case, Clark v. Community for Creative Non-violence.(50) According to the Court, it applied First Amendment scrutiny in O'Brien and Clark because in both cases the behavior that drew the law's enforcement had a sufficiently "expressive element." O'Brien evaluated the constitutionality of a federal law prohibiting destruction of draft certificates. The law aimed at conduct, draft card destruction, which often can be nonexpressive.(51) But when, as in O'Brien, protesters burn their draft cards to display their opposition to the draft, the conduct that triggers the law's enforcement becomes expressive. Similarly, although the activity that drew the law's enforcement in Clark (sleeping in a park) seldom relates to speech, it becomes expressive when, as was the situation in Clark, demonstrators camp to protest the plight of the homeless.
The Court's use of O'Brien and Clark to illustrate the Arcara rule's meaning suggests that the rule should be construed narrowly, so that incidental restraints only concern the First Amendment when the activity that draws the law's application is itself used to express a message. This interpretation conforms the treatment of incidental restrictions with general First Amendment doctrine: Expressive activity also triggers the application of both content-based laws and content-neutral provisions that directly affect speech.(52) The relevant questions thus become: Is the activity that draws the law's enforcement itself expressive, or on the other hand, is it possible to distinguish a nonexpressive enforcement trigger? Both expressive draft card burning and expressive camping implicate the First Amendment under this test: Since it is analytically impossible to distinguish their expressive and nonexpressive aspects, one cannot isolate a wholly nonexpressive activity that draws the law's application.(53) The enforcement trigger, therefore, is undeniably expressive. This interpretation of the Arcara rule not only explains why the Court applied First Amendment scrutiny in O'Brien and Clark; it also demonstrates why the Court finds no First Amendment issue in other situations. In Arcara itself, for instance, the activity that drew the nuisance provision's enforcement (prostitution) lacked any expressive element,(54) and the restricted speech (the bookstore's expressive activities were it not closed) was completely distinguishable from the enforcement trigger. Likewise, application of a speeding law to a newscaster would not raise a First Amendment concern:(55) The nonspeech that triggers the law's enforcement (speeding) is not itself expressive, and the speech that the law incidentally restricts (the story the newscaster was rushing to cover) is analytically distinct from the enforcement trigger. For similar reasons, the rule explains why the Court finds no First Amendment issue with applying general corporate tax rates to bookstores,(56) or with enforcing labor or antitrust laws against newspapers.(57)
This interpretation of the Arcara rule, however, has difficulty explaining another case the Court cited in Arcara for support, United States v. Albertini.(58) Albertini concerned an ordinance that prohibits individuals from reentering a military base after having been barred by military officials.(59) The defendant, Albertini, had destroyed secret documents at a military installation.(60) In response, the military (pursuant to the antireentry provision) forbade him from entering the base again. Albertini nevertheless reentered the base during an open house several years later, where he and some companions peacefully protested against nuclear arms proliferation.(61) Recognizing Albertini, military officials halted the demonstration and arrested him for violating the reentry law.(62) Albertini challenged his subsequent conviction on First Amendment grounds.
Since Albertini did not begin his protest until after he had illegally reentered the base, the conduct that drew application of the law (the initial reentry) was not itself expressive. The situation therefore differs from O'Brien and Clark. Moreover, not only was the enforcement trigger non-expressive, it was not even contemporaneous with the restricted speech (unlike, for example, a trespass that occurs during a demonstration).(63) Because it is possible to identify a nonexpressive enforcement trigger completely separate from the suppressed speech, application of the reentry law seemingly should not implicate the First Amendment under the Arcara rule. The Court at one point seemed as though it would subscribe to this reasoning, stating: "Respondent was prosecuted not for demonstrating at the open house, but for reentering the base after he had been ordered not to do so."(64) Ultimately, however, the Court subjected the reentry provision's application to O'Brien scrutiny,(65) and even though it eventually found no First Amendment violation, it at least established that the law raised a First Amendment concern.
Albertini indicates that incidental restraints may trigger a First Amendment analysis even if the behavior that draws the law's application is not itself expressive.(66) There are of course more expansive ways to interpret the Arcara rule so that it would accommodate Albertini. For instance, the activity that draws the law's application may be deemed to have a "significant expressive element" not only when it is itself expressive,(67) but also if it facilitates some expressive activity, or, more narrowly, if it is necessary to some particular speech act. Since illegally reentering the base facilitated Albertini's protest, and was also necessary to allow his on-base demonstration, these broader constructions of the rule could explain the Court's decision to subject the reentry provision to an O'Brien analysis.
But the Court could not have meant in Arcara to invite First Amendment scrutiny whenever the activity that triggers the law's application facilitates speech or is necessary to some expressive act. Otherwise, enforcing a speeding law against a news reporter might require a First Amendment analysis; it is not difficult to envision situations in which speeding both facilitates reporting a particular story and is necessary to report the story. As a result, expanding the Arcara rule to accommodate Albertini would also gerroneously sweep in the speeding law (and a host of other laws).(68) This suggests that Albertini must be an exception to the Arcara rule, with the rule retaining its narrower interpretation. But why is the speeding law also not an exception?--after all, both speeding and illegal reentry can facilitate speech. That is, why should an arrest for illegally reentering the base to protest trigger First Amendment scrutiny, when an arrest for illegally speeding to the base to protest would not? And can any justification for this distinction be reconciled with the Arcara rule, allowing a general rationalization of the Court's approach to incidental restraints?
One conceivable justification for the differential treatment of the speeding law and the reentry law might focus on speech-restrictive effect. The speeding law might often only delay expressive activity, while the reentry law might completely foreclose it.(69) For instance, if Albertini had been arrested for speeding while on his way to the base, he might well have eventually reached the base and still engaged in a demonstration. On the other hand, arresting Albertini for illegally reentering the base did not merely delay his protest, but stopped it forever. Thus, perhaps Arcara means that incidental restraints normally require First Amendment review only if expressive activity triggers the law's enforcement, but that situations in which enforcement of the law produces substantial speech-restrictive effect will also provoke a First Amendment analysis, even if the activity that draws the law's application is nonexpressive.(70) This approach could possibly explain why the Court applied First Amendment scrutiny in Albertini but would not in the speeding situation.
The Court, though, should be reluctant to supplement the Arcara rule with an inquiry into speech-restrictive effect. Since all generally applicable laws can have speech-inhibiting consequences,(71) this approach could require determining in countless cases whether the restrictive impact is sufficiently substantial to concern the First Amendment. It should therefore arouse the same concerns with excessive judicial review as a view that would subject all incidental burdens to First Amendment scrutiny.(72) Moreover, there is a measurement problem associated with calculating the degree of speech-restrictive effect in different situations. Calibrating such effects along a common metric seems unworkable, which makes comparing their extent in different circumstances largely guesswork. As a result, courts would have difficulty differentiating a "substantial" speech-restrictive effect from an insubstantial impact, and the calculation could easily become question-begging.
Nevertheless, even if it were possible to distinguish substantial and insubstantial speech-restrictive effects in a principled manner, and to do so in a way that avoids the "First Amendment overload" problem, an effects-based explanation cannot adequately account for the Court's different views of the speeding and reentry laws. This is because the speeding law could have the same speech-inhibiting effect as the reentry provision, but the Court still would not subject it to any First Amendment scrutiny. For instance, if in Albertini the base's "open-house" lasted for only a few hours, arresting Albertini for speeding could completely foreclose his opportunity to protest at the base, equaling the effect of the reentry law. But enforcing the speeding law presumably still would not provoke a First Amendment analysis: Otherwise, whether Albertini could present a satisfactory claim would depend on how late he was to the open-house. Accordingly, the Supreme Court's decisions make clear that the First Amendment implications of incidental restraints do not depend on the degree of speech-restrictive impact.(73) To the extent that these effects can be compared at all, they surely were greater in Arcara than in Albertini,(74) yet of the two cases only Albertini raised a First Amendment concern.
There is, however, an alternative explanation that better distinguishes the speeding law from the reentry law, and that in my view rationalizes the Supreme Court's entire approach to incidental restrictions of speech. This view concentrates on the danger of a speech-suppressive administrative motive rather than the degree of speech-restrictive effect. Although generally applicable laws promote legislative objectives that presumably are unrelated to restricting speech,(75) government may nonetheless use general laws to suppress expression; in this sense, a speech-restrictive administrative motivation can trump a speech-neutral legislative motivation.(76) A concern with speech-suppressive administrative motivation (or at least with circumstances evincing the possibility of illicit motive)(77) seems largely to explain the Court's treatment of incidental restrictions. While not explicit, this concern is inferable from the decisions examining incidental restraints, and it elucidates why some incidental effects (such as those produced by the reentry provision in Albertini) trigger First Amendment scrutiny, and others (such as those caused by a speeding law) do not.
A significant characteristic of speeding laws in this regard is the substantial unlikelihood that an intent to suppress speech motivates their application. This is due to the nature of a typical speeding arrest: The arrest normally occurs without any prior knowledge of the driver's identity, intentions, or destination. As a result, it is rather improbable that the motive in enforcing speeding laws in any given case relates to restricting the speeder's speech.(78) For instance, if Albertini had been cited for speeding while traveling to the military base, it would have been extremely unlikely that a desire to restrict his expressive activities provoked his arrest. Except under peculiar circumstances, his plans to go to the base and his intent to engage in a protest there would have been unknown to the arresting agent.
In contrast, there is some danger that the military applied the reentry law against Albertini in order to suppress his speech, or at least that the decision to apply the law was a reaction to his demonstration. Even before officials recognized Albertini as an illegal entrant, they were irritated by his protest and hoped to terminate it.(79) And although military police did not approach Albertini until after identifying him as being barred from the base, the base commander had already ordered military police to stop Albertini's demonstration before realizing that he was an illegal entrant.(80) In these circumstances, enforcement of the reentry provision against Albertini may have been provoked by a reaction to his speech-a desire to halt his demonstration or to punish it. The possibility of such an illicit, speech-restrictive administrative motive might explain why the Supreme Court engaged in a First Amendment analysis in Albertini.
To see the centrality of motive in the Court's decision to apply First Amendment scrutiny, suppose that the military enforced the reentry law more like a speeding law, carefully examining everyone wishing to enter a base to keep out illegal entrants. Were this policy in effect, Albertini (and all other illegal entrants) would never make it inside. Albertini still might contest the application of the hypothetical reentry policy against him on First Amendment grounds, claiming that it restricted his planned speech inside the base. But it is highly doubtful that the law's application would raise any First Amendment concern: Otherwise, every illegal entrant stopped at the gate could qualify for O'Brien by scrutiny simply claiming that the law's application restricted his impending speech.(81) The most significant distinction between these circumstances and those in Albertini, and the most plausible explanation for their differential constitutional implications, relates to the likelihood of an illicit motive. In the hypothetical situation it is substantially less probable that a desire to restrict Albertini's speech could provoke application of the law. Since he and all other visitors to the base would simply be checked before entering as a matter of course (much like passengers passing through a metal detector before boarding a plane), the danger of an improper motive would be very remote.
The motive explanation not only clarifies why the reentry and speeding laws receive different First Amendment treatment, but it also allows reconciling Albertini with Arcara. Recall Arcara's rule that for an incidental restriction to trigger a First Amendment analysis, the activity that draws the law's enforcement must itself be expressive. This requirement can be understood as a rough proxy for identifying circumstances containing danger of a speech-suppressive administrative motivation. Almost by definition, when expressive rather than nonexpressive activity provokes application of a law, there is an increased likelihood that a reaction to speech motivates the law's enforcement. Conversely, when it is at least possible to identify a distinct, nonexpressive violation that triggers application of a law, an objective unrelated to suppressing speech more likely induces the law's enforcement.(82) In this sense, determining whether speech or non-speech draws a law's application can be thought of as a surrogate, allowing an almost instinctive assessment of the danger of improper administrative motivation.
Compare, for instance, the enforcement of a speeding law against a newscaster with the enforcement of a camping ban against expressive campers. The speeding law's application, to which the rule would not apply any First Amendment scrutiny, contains an insignificant danger of a speech-restrictive motive. It is possible to identify a nonexpressive enforcement trigger--speeding-and there is ordinarily little reason to doubt that speeding (and not speech) in fact motivates the law's enforcement, especially since the speech has not yet occurred. This should normally be the case when nonexpressive activity draws the law's application (unless, as in Albertini, the circumstances suggest otherwise).
With the expressive camping involved in Clark, in contrast, it is impossible to identify a wholly nonexpressive activity that triggers the law's enforcement; the expressive and nonexpressive components precisely coexist. There is then no conceivable opportunity to apply the law against its nonspeech element without knowledge of its speech element.(83) As a result, it is less certain that only a desire to restrict nonspeech motivates the law's application, and more probable that the law's enforcement is a reaction to the campers' speech. The absence of a nonexpressive enforcement trigger at least enhances the possibility that a speech-restrictive motivation has played a part in the enforcement decision, explaining the application of First Amendment scrutiny to ensure that the substantiality of the governmental interest (and the means chosen to further it) vindicates the speech-suppressive impact.
The recent case of Cohen v. Cowles Media(84) confirms that a concern with speech-restrictive motivation underlies the Arcara rule. In Cowles Media, Cohen, an associate of a political candidate, gave several newspapers damaging information regarding an opposing candidate, on the condition that the papers keep his identity confidential.(85) But because Cohen's affiliation with the opposing candidate was itself of public interest, the papers revealed his identity.(86) Cohen then sued the newspapers under a promissory estoppel theory. The papers argued that enforcing promissory estoppel law against them would unduly inhibit truthful reporting of newsworthy items (such as Cohen's identity), and that this incidental effect would violate the First Amendment.(87) The Court disagreed, declaring the incidental effect "constitutionally insignificant" and finding that enforcement of promissory estoppel principles against the press did not even raise a First Amendment concern.(88) But this seems to violate the Arcara rule (which the Court never cited): The activity that drew application of promissory estoppel law--the publication of Cohen's identity--was itself expressive.(89)
The Court's stated rationale for not engaging in a balancing analysis was that promissory estoppel law applies generally to all activities, expressive and nonexpressive.(90) But incidental restrictions by definition involve generally applicable laws (the camping ban applies to both expressive and nonexpressive campers), so the fact of general applicability cannot itself explain the lack of a First Amendment concern. Rather, since private litigants, not the government, decide when to initiate promissory estoppel actions, the unstated underpinning of the Court's decision may well be the impossibility of an illicit administrative motive. In this light, Cowles Media is the converse of Albertini: In Albertini the Court applied First Amendment scrutiny even though nonexpressive activity drew the law's enforcement; in Cowles Media the Court did not engage in a First Amendment analysis even though expressive activity triggered the law's application. The reason for deviating from the Arcara rule in both is a concern (Albertini) or lack of concern (Cowles Media) with improper motivation.
* * *
Thus, although seemingly incompatible on their face, Arcara, Albertini, and Cowles Media together suggest that a concern with speech-suppressive administrative motivation rationalizes the Supreme Court's approach to incidental restrictions. The Arcara rule--a proxy for identifying circumstances that contain a threat (or at least possibility) of speech-restrictive motive-represents the general default. It subjects incidental restraints to a First Amendment analysis only if the activity that draws the law's application is itself expressive. But even if the enforcement trigger is not expressive, Albertini illustrates that a First Amendment concern may nevertheless arise if there exist special reasons to suspect a speech-restrictive motive. At the same time, even if technically the enforcement trigger is expressive, Cowles Media demonstrates that First Amendment scrutiny may nevertheless not apply if the danger of a speech-restrictive motive is especially trivial.
More generally, this analysis provides an answer to our initial question: If a law that neither targets speech nor inevitably affects expression is applied in a way that incidentally burdens expressive activities, in what circumstances does the incidental restriction trigger a First Amendment analysis? The correct response is not always, as an emphasis on speech restrictive effect might suggest; nor is it never, as a focus on legislative motivation might direct.(91) Instead, the Supreme Court seems to have chosen a fairly coherent and sensible (albeit somewhat unexplicit) middle path: Incidental restrictions sometimes raise a First Amendment concern, depending on the likelihood of a speech-suppressive administrative motivation. And the loose set of rules that flow from Arcara, Albertini, and Cowles Media represent the Court's implementation of this approach. (1.) Turner Broadcasting System v. F.C.C>, 114 S. Ct. 2445 (1994). Content-based laws are "`related to the suppression of free expression.'" United States v. Eichman, 496 U.S. 310, 314 (1990) (quoting Texas v. Johnson, 491 U.S. 397, 410 (1989)). For general commentary on the distinction between content-based and content-neutral laws, see Geoffrey R. Stone, Content Regulation and the First Amendments, 25 Wm. & Mary L. Rev. 189 (1983). (2.) See generally Stone, 25 Wm. & Mary L. Rev at 194-96 (cited in note 1). categories of speech that receive distinct treatment and the First Amendment, 25 Wm. & Mary L. Rev. 189 (1983). (3.) See United States v. O'Brien, 391 U.S. 367, 377 (a968). Content-neutral laws can be "`justified without reference to the content'" of the restricted speech. Ward v. Rock Against Racism, 491 U.S. 288, 293 (1984)). In actuality, the Supreme Court apllies several varieties of scrutiny to conten-neutral restrictions, most of which fall somewhere between strict and deferential review. see Geoffrey R. Stone, Content-Neutral Restrictions, 54 U. Chi. L. Rev. 46, 48-54 (1987). (4.) Schneider v. State, 308 U.S. 147 (1939). "Time, place, or manner" regulations generally fit in this category, since they aim directly at speech, albeit without reference to its content. See Clark v. Community for Creative Non-Violence, 468 U.S. 288, 294 (1984). (5.) Martin v. City of Struthers, 319 U.S. 141 (1943). (6.) Members of the City Council of Los Angeles v. Taxpayers for Vincent, 466 U.S. 789 (1984). (7.) Laws that "inevitably burden expression" may not directly penalize speech per se, but they have such a predictable and disproportionate effect on expressive activities that their application nearly always burdens speech. See Arcara v. Cloud Books, 478 U.S. 697, 704-05 (1986). (8.) Minneapolis Star & Tribune v. Minnesota Comm'r of Revenue, 460 U.S. 575 (1983). (9.) United States v. National Treasury Employees Union, 115 S. Ct. 1003 (1995). (10.) Simon Schuster, Inc. v. Members of N.Y. State Crime Victims Bd., 502 U.S. 105 (1991). (11.) Because they do not target expressive activities, and instead apply to speakers and nonspeakers (as well as expressive activities and nonexpressive activities), laws that incidentally affect speech are often termed "generally applicable" regulations. See, e.g., Cohen v. Cowles Media, 501 U.S. 633, 669 (1991). (12.) See Gerald Gunther, Constitutional Law 1225 (Foundation Press, 12th ed. 1991) (distinguishing content-neutral laws "focusing specifically on expression" from those that are "aimed at a wider range of behavior and [have] only an 'incidental' impact on free speech". (13.) Cf. Grayned v. City of Rockford, 408 U.S. 104 (1972). (14.) See Minneapolis Star, 460 U.S. at 581. (15.) Arcara v. Cloud Books, 478 U.S. 697 (1986), discussed in Part II, contains the Court's only serious discussion in this regard, and Arcara is somewhat ambiguous in its own right. There is also little legal commentary on the First Amendment implications of incidental restraints, which is surprising in light of the voluminous literature on free speech issues. Among the few exceptions are Susan H. Williams, Content Discrimination and the First Amendment, 139 U. Pa. L. Rev. 615,722-28 (1991); Stone, 54 U. Chi. L. Rev. at 105-14 (cited in note 3); Frederick Schauer, Cuban Cigars, Cuban Books, and the Problem of Incidental Restrictions on Communications, 26 Wm. & Mary L. Rev. 779 (1985). (16.) 494 U.S. 872 (1990). (17.) See Barnes v. Glen Theatre, 501 U.S. 560, 576-79 (1991) (Scalia, J., concurring). (18.) See id. at 578. (19.) 468 U.S. 288 (1984). While the Court in Clark ultimately found that the camping regulation satisfied First Amendment scrutiny, the important point here is that it applied any scrutiny at all. (20.) 391 U.S. 367, 376 (1968). (21.) The O'Brien test states that "a government regulation is sufficiently justified if it is within the constitutional power of the Government; if it furthers an important or substantial governmental interest; if the governmental interest is unrelated to the suppression of free expression; and if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest." Id. at 377. Although the O'Brien test was apparently designed to examine incidental restraints, commentators have contended that it applies to content-neutral laws in general, see, e.g., John Hart Ely, Flag Desecration: A Case Study in the Roles of Categorization and Balancing in First Amendment Analysis, 88 Harv. L. Rev. 1482 (1975), and the Court seems largely to have followed suit. See Clark, 468 U.S. at 298 (finding that the O'Brien test is "little, if any, different from the standard applied to time, place, or manner restrictions"; Keith Werhan, The O'Briening of Free Speech Methodology, 19 Ariz. St. L.J. 635 (1987). (22.) See, e.g., United States v. Albertini, 472 U.S. 675 (1985) (law prohibiting reentry on to military base); Wayte v. United States, 470 U.S. 598 (1985) (law requiring registration with the Selective Service); Marsh v. Alabama, 326 U.S. 501 (1946) (trespass ordinance). (23.) See Stone, 54 U. Chi. L. Rev. at 105-07 (cited in note 3). (24.) For several such examples, see Gunther, Constitutional Law at 1217 n.12 (cited in note 12). By the same logic, content-neutral laws can have the same speech-inhibiting impact as content-based laws. Some commentators have questioned the doctrinal distinction between content-based and content-neutral provisions on precisely this ground, and have proposed that all laws-whether content-based or content-neutral-instead be analyzed under a single "unitary" framework that emphasizes speech-restrictive effect. See Martin H. Redish, The Content Distinction in First Amendment Analysis, 34 Stan. L. Rev. 113 (1981). (25.) 501 U.S. 663 (1991). The facts of Cowles Media are spelled out in Part IV. Justice Souter, speaking for the four dissenting Justices, suggested that he would engage in a balancing analysis whenever a law incidentally burdens First Amendment activities. He found "'nothing talismanic about neutral laws of general applicability,"' because "such laws may restrict First Amendment rights just as effectively as those directed specifically at speech itself." Id. at 677 (Souter, J., dissenting) (quoting Employment Division v. Smith, 494 U.S. 872, 901 (1990) (O'Connor, J., concurring)). As a result, he would "articulate, measure, and compare the competing interests involved in any given case to determine the legitimacy of burdening constitutional interests." Id. (26.) 478 U.S. 697, 708-10 (1986) (blackmun, J., dissenting) (contending that generally applicable laws which affect speech should always face First Amendment review). See also Konigsberg v. State Bar of California, 366 U.S. 36, 68-69 (1961) (Black, J., dissenting). (27.) See Arcara, 478 U.S. at 697 (application of nuisance law to bookstore); Citizen Publishing Co. v. United States, 394 U.S. 131, 139 (1969) (application of antitrust laws to the press); Associated Press v. United States, 326 U.S. 1 (1945) (same); Associated Press v. NLRB, 301 U.S. 103 (1937) (enforcement of labor laws against the press). See generally Cowles Media, 501 U.S. at 669-70. (28.) See Barnes v. Glen Theatre, 501 U.S. 560, 576 (1991) (Scalia, J., concurring); Larry A. Alexander, Trouble on Track Two: Incidental Regulations of Speech and Free Speech Theory, 44 Hastings L. J. 921, 927-31 (1993). (29.) See Arcara, 478 U.S. at 705-06; Glen Theatre, 501 U.S. at 576 (Scalia, J., concurring) ("It cannot reasonably be demanded, therefore, that every restriction of expression incidentally produced by a general law regulating conduct pass normal First-amendment scrutiny...."). (30.) Arcara, 478 U.S. at 708 (O'Connor, J., concurring). (31.) See id. at 710 (Blackmun, J., dissenting) (contending that incidental restrictions should always trigger a First Amendment balancing, but acknowledging "[a]t some point, of course, the impact of state regulation on First Amendment rights becomes so attenuated that it is easily outweighed by the state interest" . (32.) See Arcara, 478 U.S. at 706. Unlike equal protection cases, where the Court can use rationality review in less troublesome cases, the lowest standard of review in First Amendment cases-O'Brien scrutiny-is akin to intermediate scrutiny: The O'Brien standard requires a "substantial" government interest and means "no greater than is essential to the furtherance of [the government) interest.' United States v. O'Brien, 391 U.S. 367, 377 (1968). Since the Court lacks a suitable doctrinal tool with which to examine incidental effects that seemingly should not trigger a searching analysis (such as the application of a general tax law to a bookstore, or a speeding law to a newscaster), it may prefer applying no scrutiny at all to applying a form of intermediate scrutiny. See Glen Theatre, 501 U.S. at 576 (Scalia, J., concurring). (33.) See United States v. Albertini, 472 U.S. 675, 689 (1985) (finding that the O'Brien standard's narrow tailoring component does not require that there be no less restrictive alternative, but only that the "government interest ... would be achieved less effectively absent the regulation" ; Ely, 88 Harv. L. Rev. at 1484-90 (cited in note 21) (comparing strong and weak forms of a less restrictive means analysis and discussing O'Brien's embracing of the latter). (34.) CL Schauer, 26 Wm. & Mary L. Rev. at 787 (cited in note 15) (expressing similar concerns with subjecting all incidental restrictions to a First Amendment analysis). (35.) See Stone, 54 U. Chi. L. Rev. at 110-11 (cited in note 3); Schauer, 26 Wm. Mary L. Rev. at 787-88 (cited in note 15). (36.) See Hague v. (CIO, 307 U.S. 496,515-16 (1939); Stone, 54 U. Chi. L. Rev. at 113-14 (cited in note 3); Schauer, 26 Wm. & Mary L. Rev. at 788-89 (cited in note 15). (37.) See Arcara v. Cloud Booky, 478 U.S. 697 (1986) (New York court of Appeals finds incidental restriction unconstitutional under O'Brien scrutiny, but the Supreme court reverses, ruling that lower court should not have engaged in any First Amendment analysis at all). (38.) Id. (39.) Id. at 699-70. (40.) Id. at 698-99. (41.) Id. at 701-02. (42.) Id. at 706-07 n.3. (43.) Id. at 706-07. (44.) Id. at 706-07 n.3. (45.) Id. at 705. (46.) Id. at 706-07. (47.) Laws that have the "inevitable effect" of restricting speech impose a "disproportionate burden" on expression, so that they effectively "singl[e] out those engaged in expressive activity." Arcara, 478 U.S. at 704, 707. Incidental restraints, on the other hand, affect expressive activities only in certain applications, and normally do not affect speech at all--their effect on speech is hardly "inevitable." As such, when laws "inevitably" affect expressive activities, they are more akin to direct restraints than incidental restraints, and it is no surprise that they trigger First Amendment scrutiny. See notes 7-10 supra and accompanying text. For instance, in Arcara the Court identified the special use tax provision it struck down in Minneapolis Star & Tribune v. Minnesota Commr of Revenue, 460 U.S. 575 (1983), as an example of a law that "inevitably" affects expressive activities. Arcara, 478 U.S. at 704, 707. The use tax was imposed on sales of paper and ink, in a manner that it effectively only applied to a few large newspapers. When a law's burden falls so disproportionately on First Amendment activities, it seems rather doubtful that the law aims primarily at "nonspeech" and only incidentally affects expression. Indeed, in Minneapolis Star the Court suggested that the objective of the tax law must have been related to suppressing expression. Minneapolis Star, 460 U.S. at 585. In this light, such a provision's effect on speech is not merely incidental, but is rather direct. See Stone, 54 U. Chi. L. Rev. at 108-09 (cited in note 3). For examples of other laws whose application, although not directly triggered by speech, nevertheless "inevitably" affects expression, see text accompanying notes 8-10 supra. (48.) Compare, for example, the application of the nuisance law to the bookstore in Arcara with enforcement of a generally applicable trespass law against a lone protester. In the latter situation, the activity that triggers the law's enforcement, demonstrating on private property, clearly has expressive elements. Yet enforcement of the law has a relatively minor effect on speech: First, there is only one protester, whose message may not be heard by anyone; and second, the individual can continue his protest just outside of the property fine. In the Arcara situation, in contrast, the restrictive effect on speech (closure of a bookstore for one year) is likely more substantial in a quantitative sense, even though the sexual activities that triggered enforcement of the nuisance law are nonexpressive. (49.) 391 U.S. 367 (1968). (50.) 468 U.S. 288 (1984). (51.) Although it seems somewhat dubious that individuals would often destroy their draft certificates for nonexpressive reasons, see Laurence H. Tribe, The Mystery of Motive, Private and Public: Some Notes Inspired by the Problems of Hate Crime and Animal Sacrifice, 1993 S. Ct. Rev. 1, 34 (suggesting that violations of the ban on draft card destruction will almost always be expressive, except for "the odd soul who bums a draft card just to stay warm or to light up his campsite"), the Court characterizes the law as one that aims primarily at nonspeech conduct. See Arcara, 478 U.S. at 702 (contending that "the destruction of a draft card is not ordinarily (52.) This is somewhat autological--it means nothing more than that expressive flag burning triggers the enforcement of a content-based ban on burning the flag, see Texas v. Johnson, 491 U.S. 397 (1989), and that distributing leaflets provokes the application of a content-neutral provision barring leafleting, see Schneider v. State, 308 U.S. 147 (1939). (53.) See Ely, 88 Harv. L. Rev. at 1495-96 (cited in note 21) ("[B]urning a draft card to express opposition to the draft is an undifferentiated whole, 100% action and 100% expression. It involves no conduct that is not at the same time communication, and no communication that does not result from conduct." Therefore, "to outlaw the act is ... necessarily to regulate both elements. (54.) Arcara, 478 U.S. at 705. Prostitution could conceivably be expressive in some situations, but the Supreme Court likely considers prostitution to be conduct that, as a matter of law, can never be expressive. See Wisconsin v. Mitchell, 113 S. Ct. 2194, 2199 (1993); Barnes v. Glen Theatre, 501 U.S. 560, 570 (1991); City of Dallas v. Stanglin, 490 U.S. 19, 25 (1989); O'Brien, 391 U.S. at 376. (55.) See Arcara, 478 U.S. at 708 (O'Connor, J., concurring). Several commentators have also used this as an example of a situation that would not trigger any First Amendment scrutiny. See Laurence H. Tribe, American Constitutional Law [sections] 12-23 n.2 (2d ed. 1988); Schauer, 26 Wm. & Mary L. Rev. at 790 (cited in note 15); Williams, 139 U. Pa. L. Rev. at 723 (cited in note 15). (56.) See Minneapolis Star & Tribune v. Minnesota comm'r of Revenue, 460 U.S. 575, 581 (1983). (57.) See, e.g., Citizen Publishing Co. v. United States, 394 U.S. 131, 139 (1969) (antitrust laws); Associated Press v. NLRB, 301 U.S. 103 (1937) (labor laws). (58.) 472 U.S. 675 (1985) (cited in Arcara, 478 U.S. at 703). (59.) Albertini, 472 U.S. at 677-78. (60.) Id. at 677. (61.) Id. at 678. (62.) Id. at 678-79. (63.) See Marsh v. Alabama, 326 U.S. 501 (1946). (64.) Albertini, 472 U.S. at 686. (65.) Id. at 687-89. (66.) In this light, the Arcara rule does not entail "evaluat[ing] whether the conduct [that draws the law's enforcement) is inherently more like expression, for example, draft card burning, or inherently more like conduct, for example, prostitution." Jonathan David Selbin, Bashers Beware. The Continuing Constitutionality of Hate Crimes Statutes after R.A. V., 72 Ore. L Rev. 157, 171 (1993). This brings to mind the speech-conduct distinction, which ultimately is largely indeterminate. See Ely, 88 Harv. L. Rev. at 1494-96 (cited in note 21). Rather, Albertini demonstrates that the Court may engage in a First Amendment analysis even if the behavior that triggers the law's application-here, entering a military base-is not inherently expressive at all. (67.) Arcara v. Cloud Books, 478 U.S. 697, 706 (1986). (68.) Violations of almost any generally applicable law could, in certain circumstances, facilitate expressive activities. Cf. Williams, 139 U. Pa. L. Rev. at 723-24 (cited in note 15). (69.) See Schauer, 26 Wm. & Mary L. Rev. at 790 (cited in note 15) (suggesting one possible way of limiting the First Amendment implications of incidental effects is to distinguish between a mere "restriction" of speech and an "absolute prohibition" . (70.) Such an explanation resembles the one offered in Stone, 54 U. Chi. L. Rev. at 114 (cited in note 3). (71.) See text accompanying note 28 supra. (72.) See Part I supra. The only difference is that this approach requires determining initially whether the extent of restrictive effect merits First Amendment review, whereas the other would subject all incidental restraints to a balancing inquiry--either way, every incidental restriction receives a nontrivial analysis. (73.) See Arcara v. Cloud Booky, 478 U.S. 697, 706 (1986) (noting "we have not ... subjected every ... sanction imposed through legal process to [First Amendment] scrutiny simply because each particular remedy will have some effect on the First Amendment activities of those subject to sanction"). The Court's recent ruling in Alexander v. United States, 113 S. Ct. 2766 (1993) demonstrates its reluctance to focus on speech-restrictive effect. In Alexander, the Court would not apply any First Amendment scrutiny to a RICO provision that authorized the forfeiture and destruction of entire lines of businesses, even though this required permanently destroying several speech-related entities (such as bookstores and movie houses) and entire inventories of First Amendment protected items (including books and films). (74.) The closure order in Arcara prevented the bookstore from disseminating all of its books for a period of one year. The reentry law in Albertini prevented one protester from engaging in a demonstration at the locale of his choice. While it is true that the bookstore could have sold its books at another location, see Arcara, 478 U.S. at 705, it is equally the case that Albertini could have engaged in his protest elsewhere, presumably including just outside the military base. (75.) See text accompanying note 18 supra. (76.) See Cox v. Louisiana, 379 U.S. 536, 555-57 (1965). (77.) There is an important distinction between merely perceiving the danger of a speech-suppressive administrative motive, which is what I mean to discuss here, and proving outright that an intent to restrict speech motivated a law's enforcement. With respect to the latter, if the government selectively prosecutes speakers for violating a generally applicable law or enforces the law as a pretext to suppress speech, the speaker could attempt to prove that the selective enforcement violates equal protection principles. See Wayne v. United States, 470 U.S. 598 (1985); Arcara, 478 U.S. at 707 n.4. That is distinct, however, from situations where the circumstances merely suggest an increased threat of a speech-suppressive motive. The argument here is simply that incidental restraints raise a First Amendment concern where there is a danger of a speech-restrictive motive. For the speaker to "meet the heavy burden of proving selective prosecution," Arcara, 478 U.S. at 712 (Blackmun, J., dissenting), would require not merely establishing the possibility of an illicit motive, but proving that such a motive triggered the law's enforcement. (78.) This of course would not be the case if government enforces speeding laws as a pretext to restrict speech. Without evidence of selective application, however, the circumstances typically surrounding enforcement of speeding regulations do not suggest danger of a speech restrictive motive. (79.) See United States v. Albertini, 472 U.S. 675, 678-79 (1985). (80.) Id. at 690. (81.) Cf. Arcara v. Cloud Books, 478 U.S. at 706 (suggesting prisoners cannot claim their confinement infringes their right to speak in the future). (82.) Cf. Alexander v. United States, 113 S. Ct. 2766, 2784 (1993) (Kennedy, J., dissenting) There is a "vital difference between a punishment imposed for a speech offense and a punishment imposed for some other crime. Where the government seeks forfeiture of a bookstore because of its owner's drug offenses, there is little reason to surmise, absent evidence of selective prosecution, that abolishing the bookstore is related to the government's disfavor of the publication outlet or its activities."). (83.) See Ely, 88 Harv. L. Rev. at 1496 (cited in note 21). (84.) 501 U.S. 664 (1991). (85.) Id. at 665. (86.) Id. at 666. (87.) Id. at 671. (88.) Id. at 672. (89.) If a paper violates antitrust laws or labor laws, see Citizen Publishing Co. v. United States, 394 U.S. 131, 139 (1969) (antitrust laws); Associated Press v. NLRB, 301 U.S. 103 (1937) (labor laws), the behavior that draws the law's enforcement might facilitate expression, but would not itself be expressive. In Cowles Media, in contrast, publicizing Cohen's identity--the conduct that drew the law's application--is itself expressive, and this is why the Arcara rule would presumably apply here as opposed to the antitrust or labor situations. (90.) Cowles Media, 501 U.S. at 670. (91.) See Part I supra.
(*) Law clerk, the Honorable J. Harvie Wilkinson, III, U.S. Court of appeals dor the Fourth Circuit, J.D. Stanford University, 1995; .B.A. Stabford University, 1994. Thanks to Professor Gunther and to Brad Joondeph for their comments on previous drafts.…