In challenging the Supreme Court’s newly announced method of determining the coverage — and thus the boundaries — of the First Amendment, Genevieve Lakier demonstrates perceptively and effectively that some of the domains most widely discussed as being outside of the First Amendment’s coverage have long been understood to have First Amendment dimensions and implications.1×1. Genevieve Lakier, The Invention of Low-Value Speech, 128 Harv. L. Rev. 2166 (2015). Obscenity, commercial speech, libel, and fighting words, for example, the subjects most commonly discussed as being, now or in the relatively recent past, uncovered by the First Amendment, are, she shows, topics that have for generations attracted First Amendment discourse and been the subject of First Amendment–inspired legal doctrines. Consequently, the idea that the permissibility of content-based regulation turns on the designation of speech as high value or low value is, she argues, historically and doctrinally mistaken. For the Supreme Court to use a historical test for the designation of categories of speech as high or low value, as the Court says it has done in its recent Stevens2×2. United States v. Stevens, 130 S. Ct. 1577, 1584–86 (2010). and Entertainment Merchants3×3. Brown v. Entm’t Merchs. Ass’n, 131 S. Ct. 2729, 2734–36 (2011). decisions, is to rely on bad history to produce mistaken constitutional doctrine.
I have no quarrel with Lakier’s impressive historical analysis, nor with her critique of the Stevens–Entertainment Merchants approach. But her conclusions seem less surprising, and even expected, once we recognize that she is focusing exclusively on cases and issues at or near the borders. Borderline cases of First Amendment coverage will display attributes of coverage and noncoverage, just as borderline cases of almost anything will display attributes lying on both sides of the border. But once we move away from the border, the duality of the attributes diminishes, and so it is with uncovered speech. The speech that is plainly and obviously uncovered — that by no stretch of the imagination has anything to do with the idea of freedom of speech — has been treated no differently historically than it is now, and there is little reason to believe that the existence of the borderline cases has much to say about the existence or implications of those domains that are well outside the range of First Amendment interest.
I. Coverage 101
Lakier frames the issue in terms of high-value and low-value speech. But although that terminology has some provenance,4×4. See, e.g., Larry Alexander, Low Value Speech, 83 Nw. U. L. Rev. 547 (1989); Cass R. Sunstein, Pornography and the First Amendment, 1986 Duke L.J. 589, 591; Christopher M. Schultz, Note, Content-Based Restrictions on Free Expression: Reevaluating the High Versus Low Value Speech Distinction, 41 Ariz. L. Rev. 573 (1999). the real issue is not between high- and low-value speech, but instead between high-value and no-value speech, where “no-value” is a reference only to First Amendment value and not to value simpliciter.5×5. The culprit here is Justice Brennan’s plurality opinion in Jacobellis v. Ohio, 378 U.S. 184 (1964) (plurality opinion), where he introduces the idea that material cannot be deemed obscene unless it is “utterly without redeeming social importance,” id. at 191 (quoting Roth v. United States, 354 U.S. 476, 484 (1957) (internal quotation marks omitted)), a phrase reiterated in his plurality opinion in A Book Named “John Cleland’s Memoirs of a Woman of Pleasure” v. Attorney General, 383 U.S. 413, 418 (1966) (plurality opinion). Justice Brennan purported to be relying on language in Chaplinsky v. New Hampshire, 315 U.S. 568 (1942), as cited in Roth, 354 U.S. at 485, but Chaplinsky talks of “social value as a step to truth,” 315 U.S. at 572. The social value to which the Chaplinsky Court was referring is thus revealed as a specific kind of free-speech value. Consequently, it is important to recognize that something — actual sex, for example, and not depictions of it — might have a great deal of social value but no speech or free-speech value. In other words, the issue is about which forms of speech, in the ordinary language sense of that word, or which forms of communication or expression, will be understood as having nothing to do with the First Amendment. Speech and communication are at the center of contract law, for example, but it is laughable to suppose that all, most, or even much of contract law in any way implicates the First Amendment. So too, and equally obviously, with the law of wills and trusts, with prohibitions on perjury, with most of the law of evidence, with much of antitrust law, and with those countless other areas of law where the content of speech and communication routinely generates legal or regulatory consequences.
My own preferred terminology distinguishes between coverage and protection.6×6. Frederick Schauer, The Boundaries of the First Amendment: A Preliminary Exploration of Constitutional Salience, 117 Harv. L. Rev. 1765, 1769–74 (2004); see also Frederick Schauer, Can Rights Be Abused?, 31 Phil. Q. 225 (1981); Frederick Schauer, Categories and the First Amendment: A Play in Three Acts, 34 Vand. L. Rev. 265 (1981); Frederick Schauer, Codifying the First Amendment: New York v. Ferber, 1982 Sup. Ct. Rev. 285. The distinction itself has earlier origins with different terminology in articles by Harry Kalven7×7. Harry Kalven, Jr., The Reasonable Man and the First Amendment: Hill, Butts, and Walker, 1967 Sup. Ct. Rev. 267, 278 (distinguishing between the “ambit” and “level” of First Amendment protection). and Laurent Frantz,8×8. Laurent B. Frantz, The First Amendment in the Balance, 71 Yale L.J. 1424, 1444 (1962) (distinguishing the scope of the First Amendment from its strength). with the basic idea being that it is necessary to distinguish the domain of events with which the First Amendment is concerned from what the First Amendment will do with those events. Although speech explicitly inciting imminent acts of serious illegality under circumstances in which the illegality will likely ensue is not protected according to Brandenburg v. Ohio,9×9. 395 U.S. 444, 447 (1969) (per curiam). nonprotection is a consequence of the application of the stringent Brandenburg standard, a standard applicable precisely because speech publicly advocating unlawful acts is covered by the First Amendment. By contrast, the speech involved in a routine contract offer and acceptance, or in a conversation aimed at fixing prices between two corporate executives, or in the words used by a gambler to place an unlawful bet with an unlawful bookmaker is unprotected without application of a test of any stringency at all, and that is because these acts — all of which are “speech” in ordinary language — are simply not covered by the First Amendment, rendering the high degree of protection offered by the First Amendment inapplicable. The question of which forms of speech are covered by the First Amendment is thus distinct from the question of how much protection the speech that is covered will receive.
The distinction between coverage and protection is not only about the First Amendment and speech but is also a necessary feature of all rules. Any rule regulates some but not all behavior, and were a driver apprehended for driving above the speed limit to claim that the Rule Against Perpetuities governs his driving, we (and the arresting officer) would think him daft. The same structure applies to the First Amendment, but with the added complication that the First Amendment uses the word “speech,” even though many things that are speech in the ordinary language meaning of that word are no more covered by the First Amendment than driving above the speed limit is covered by the Rule Against Perpetuities.
Problems arise not only because of the patent noncoverage by the First Amendment of some acts that are literally speech but also because of the contested nature of some of the topics at the boundary between coverage and noncoverage. Commercial advertising is often thought to have been formerly uncovered and thus regulable as long as the regulation met the minimal standards of rational basis review,10×10. Valentine v. Chrestensen, 316 U.S. 52, 55 (1942). but is now subject to regulation only if the regulation meets the somewhat more stringent standards first set forth in Central Hudson Gas & Electric Corp. v. Public Service Commission.11×11. 447 U.S. 557, 564 (1980). The test was most recently restated in Lorillard Tobacco Co. v. Reilly, 533 U.S. 525, 554–56 (2001). So too with libel, which was understood to be uncovered12×12. Beauharnais v. Illinois, 343 U.S. 250, 254–58 (1952). prior to New York Times Co. v. Sullivan.13×13. 376 U.S. 254, 268–69 (1964). Fighting words14×14. Chaplinsky v. New Hampshire, 315 U.S. 568, 571–72 (1942). and obscenity15×15. Roth v. United States, 354 U.S. 476, 481–83 (1957). are, in theory, still treated as uncovered,16×16. On fighting words, see R.A.V. v. City of St. Paul, 505 U.S. 377, 384–86 (1992). On obscenity, see Paris Adult Theatre I v. Slaton, 413 U.S. 49, 54, 69 (1973). but the definition of those categories has become heavily infused with First Amendment values.17×17. Thus, although obscenity remains uncovered, Slaton, 413 U.S. at 54, 69, the definition of obscenity has been crafted to ensure that materials having First Amendment value are not placed in the uncovered category. See Miller v. California, 413 U.S. 15, 24–25 (1973). The same is true of fighting words, where the definition of fighting words has been substantially narrowed and clarified since Chaplinsky, with the narrowing and clarifying designed to exclude from the uncovered category of fighting words anything except face-to-face epithets likely either to cause injury by themselves or to provoke retaliatory violence. See Lewis v. City of New Orleans, 415 U.S. 130, 132–33 (1974); Gooding v. Wilson, 405 U.S. 518, 523, 528 (1972). Still, as the Supreme Court emphasized in Paris Adult Theatre I v. Slaton,18×18. 413 U.S. 49. materials found to be obscene under the Miller v. California19×19. 413 U.S. 15, 24–25 (1973). definition of obscenity may be regulated by the government for any number of reasons,20×20. Slaton, 413 U.S. at 60–63. many of which are pretty lame, just as with many of the reasons that are used to justify regulation under any other application of rational basis scrutiny.21×21. See, e.g., City of New Orleans v. Dukes, 427 U.S. 297, 304–05 (1976) (per curiam); Williamson v. Lee Optical of Okla., Inc., 348 U.S. 483, 487–88 (1955).
II. Enter Stevens
None of the foregoing is controversial, at least as a matter of legal doctrine, but things have changed as a consequence of claims in several recent Supreme Court cases. In United States v. Stevens, most prominently, the Supreme Court was asked to find that animal abuse videos were entirely uncovered by the First Amendment and thus regulable in the same way that Miller-tested obscenity is regulable.22×22. See United States v. Stevens, 130 S. Ct. 1577, 1581 (2010). This the Court refused to do, accompanying its refusal with the assertion that new categories of noncoverage would be recognized only if, like the existing categories of noncoverage, there was a history of permissible regulation without First Amendment scrutiny.23×23. Id. at 1584–86. This standard was reiterated and affirmed a year later in Brown v. Entertainment Merchants Ass’n,24×24. 131 S. Ct. 2729 (2011). dealing with the question of whether California could prohibit the sale of interactive violent video games to juveniles.25×25. Id. at 2732. Here, as in Stevens, the Court found the games covered by the First Amendment, refusing to expand the domain of noncoverage and insisting that the only uncovered communications were those that were historically subject to regulation.26×26. Id. at 2734–35.
Lakier challenges this claim, and her contribution is in showing, by deep historical investigation, that obscenity, libel, commercial speech, and fighting words, most prominently, have long been thought to be forms of communication whose control has First Amendment implications.27×27. Lakier, supra note 1, at 2179–92. As a result, she argues, the noncoverage of obscenity, for example, cannot itself be justified historically, making it a mistake to assume that regulability should be dependent on historical noncoverage.28×28. Id. at 2203–18. And if this is so, then regulation might be permissible even for those forms of speech, as with the speech in Stevens, that do not satisfy the Court’s newly invented historical test.29×29. Id. at 2211–22. Because that test misrepresents the actual history, Lakier concludes, a historical test is simply the wrong instrument to use. And although Lakier appears to acknowledge that some conception of First Amendment irrelevance in the design of First Amendment doctrine is necessary, inevitable, and desirable,30×30. Id. at 2229–30. this recognition is in considerable tension with her claim throughout the Article that the distinction between high-value and low-value speech — coverage and protection, in my language — has been in some way “invented” by the Supreme Court, first in Chaplinsky and then in Stevens.31×31. Id. at 2197 (“Inventing a Tradition”); id. at 2203 (“new doctrine”); id. at 2207 (“very new conception”); id. at 2211 (“reinvented”); id. at 2214 (“[T]he distinction between high- and low-value speech is a product of the twentieth century . . . .”).
III. The Boundaries of Coverage
Lakier’s critique of the Stevens approach to noncoverage is on target. I have no reason to question Lakier’s historical excavations and analysis, and I also agree with her in rejecting the idea that history is the only or even the best way to determine coverage and noncoverage. Violent video games are, after all, only one example of a form of speech that simply did not exist in the past, and thus to ask whether it was or was not subject to regulation in the past is akin to asking about the treatment of diseases of unicorns. Yes, we might draw inferences about how to treat unicorn diseases, were unicorns to exist, from how we treat the diseases of horses, which unicorns resemble, but once we get into determining the similarities and differences between real and imagined creatures it is apparent that the determination is heavily infused by our own judgments of just which similarities ought to matter and which should not. And so when we try to ask whether violent video games or animal crush videos would have been regulable in the past we are engaged in the same dicey business of picking the features of these acts that do or do not resemble past forms of regulation and nonregulation, and in doing so we are simply expressing our current view about what we think should be regulated and what not.32×32. As Justice Alito quipped during the Entertainment Merchants oral argument, “Well, I think what Justice Scalia wants to know is what James Madison thought about video games. Did he enjoy them?” Transcript of Oral Argument at 17, Brown v. Entm’t Merchs. Ass’n, 131 S. Ct. 2729 (2011) (No. 08-1448) (argued as Schwarzenegger v. Entm’t Merchs. Ass’n). Better, Lakier correctly argues, to make these judgments on the basis of a straightforward acknowledgement of the background ideas and ideals of the First Amendment, rather than expecting history to do all the work. If the Court is worried about expanding coverage, as she suspects, then a false reliance on mistaken history is a poor way to embody that worry.
On all of this Lakier is on sound footing, but things get slipperier once she ventures into broader speculation about the invention of the distinction between coverage and protection, between high-value and low-value speech. Lakier appears to believe that by demonstrating that the most salient subjects of noncoverage were in fact more covered in the past than the Court and most commentators have recognized, she has supported the conclusion that the distinction between coverage and protection — between high- and low-value speech — is in some way newly fashioned, and that it did not exist in earlier times, at least prior to Chaplinsky.33×33. See note 31, supra. But that is precisely where she stumbles, and she stumbles because she has made the common mistake of making too much out of borderline cases.
One of the most important features of commercial speech, fighting words, libel, and obscenity is that all were actually the subject of the above-described Supreme Court cases in the 1940s and 1950s affirming their noncoverage. But the very fact that these cases existed at all shows that these are cases in which someone found it worthwhile to claim First Amendment coverage, in which some lower court either found coverage or found the coverage question to be a close one, and in which the Supreme Court found the issue close enough to take the case and give it full plenary consideration. But as the literature on the selection effect tells us,34×34. See especially George L. Priest & Benjamin Klein, The Selection of Disputes for Litigation, 13 J. Legal Stud. 1 (1984). cases (or, more accurately, events) that are genuinely easy are rarely taken to court, and, if so, are rarely pursued to judgment, and, if so, are rarely pursued to appeal, and, if so, are rarely pursued to and accepted for Supreme Court plenary consideration.35×35. See Frederick Schauer, Easy Cases, 58 S. Cal. L. Rev. 399 (1985); Frederick Schauer, Judging in a Corner of the Law, 61 S. Cal. L. Rev. 1717 (1988). Conversely, the cases that do wind up in the Supreme Court are the skewed sample of only those disputes as to which opposing parties holding mutually exclusive positions each believe they have a chance of prevailing. And these disputes are systematically, and almost inevitably, those on the borderline.
And so it is with the distinction between coverage and protection. There has never been a serious Supreme Court discussion of why political advocacy is covered (albeit not necessarily protected) by the First Amendment because an argument to the contrary would simply not be taken seriously. Conversely, the arguments and cases dealing with the coverage of libel, obscenity, fighting words, and commercial speech have arisen precisely because the issues are difficult and the topics lie close to the borderline between the covered and the uncovered. But there has never been a Supreme Court or lower federal court or state court case even dealing with why the speech that makes a contract or a will is not covered by the First Amendment, and that is not because such speech really has been covered, as Lakier maintains with obscenity, libel, fighting words, and commercial speech, but because such speech is leagues away from the outer boundaries of plausible First Amendment coverage. Similarly, there appears never to have been a case in either the state or federal systems taking seriously the idea that in-court perjury is covered or protected by the First Amendment.36×36. Indeed, when perjury has even been mentioned in free speech context, it is only as a contrasting example of the kinds of things, many of which do not involve words, that are plainly regulable. See State ex rel. Crow v. Shepherd, 76 S.W. 79, 95 (Mo. 1903) (en banc) (noting that perjury is like conspiracy, receiving stolen goods, and disturbing the peace in being plainly subject to regulation), overruled in part on other grounds by Ex parte Creasy, 148 S.W. 914 (Mo. 1912) (en banc); Kintz v. Harriger, 124 N.E. 168, 170 (Ohio 1919) (using perjury as a reductio to explain that all uses of words are not part of the idea of free speech), overruled on other grounds by Stephenson v. McCurdy, 177 N.E. 204 (Ohio 1931) (per curiam); State v. Butler, 35 P. 1093, 1094 (Wash. 1894) (treating perjury as similar to escape in being in a different category from speech). And with respect to ordinary fraud, when freedom of speech issues have arisen about, say, picketing, fraud and violence have been grouped together as examples of categories that stand opposed to the categories about which freedom of speech is relevant.37×37. See, e.g., People v. Harris, 91 P.2d 989, 993 (Colo. 1939) (en banc); Busch Jewelry Co. v. United Retail Emps. Union Local 830, 9 N.Y.S.2d 167, 168 (Sup. Ct. 1939); see also St. Louis Sw. Ry. Co. v. Griffin, 154 S.W. 583, 585 (Tex. Civ. App. 1913) (treating fraud as a recognized abuse of free speech and not as anything involving the principle of free speech), rev’d on other grounds, 171 S.W. 703 (Tex. 1914); Wis. Labor Relations Bd. v. Fred Rueping Leather Co., 279 N.W. 673, 683 (Wis. 1938) (observing that not every use of speech implicates the freedom of speech). Similarly, if the parties to an otherwise unlawful price-fixing agreement even attempted to argue that their behavior was covered by the First Amendment and thus should be measured against First Amendment–inspired standards, their claim would be treated with incredulity, precisely because the noncoverage is so obvious.38×38. And that is why there is an important difference, even under a historical approach that both Lakier and I reject, between speech that has long been regulated and speech that has long been understood to be regulable. The question is not (only) one of what has been regulated, but (also) what, counterfactually, would have been understood to be regulable without constitutional constraint had a government sought to regulate it.
It is unthinkable that all human behavior is covered by the First Amendment, and almost as unthinkable that all human behavior involving words is covered. Only certain activities, and only certain verbal or linguistic ones, attract the full machinery of First Amendment analysis, tests, standards, and presumptions. That libel, obscenity, fighting words, and commercial speech have long been at least discussed in free speech terms, as Lakier valuably demonstrates, shows only that these are the topics as to which arguments for coverage are at least plausible. But there are other topics, even those involving speech, where the arguments for coverage are entirely implausible. For these topics, the First Amendment is not even within range. Once we recognize this, we can recognize that the distinction between coverage and protection is neither new nor invented. In its most obvious nonborderline manifestations, it has been invisible, but it has not been non-existent. The distinction is an inevitable feature of free speech doctrine, a feature that the identification of borderline cases does nothing to undercut.
* David and Mary Harrison Distinguished Professor of Law, University of Virginia.