Harvard Law Review Harvard Law Review Harvard Law Review

Harvard Law Review Forum

Emotional Compelled Disclosures

Government-mandated warnings can be emotionally manipulative as well as factually wrong

There is much that Professor Rebecca Tushnet and I agree on when it comes to emotion and compelled speech.1 Most fundamentally, we agree that emotion and reason are inextricably linked, and therefore government-compelled disclosures that provoke emotional responses are not automatically suspect. Consequently, the D.C. Circuit was wrong to strike down mandatory cigarette warnings on the ground that they were emotional rather than purely factual.2 We also both agree that the government should not be able to compel emotional speech, or really any speech, that is factually erroneous or misleading. For this reason alone, the government should not be able to require disclosures that suggest abortion causes depression or an increased risk of suicide, as these claims are patently false.3

For Tushnet, compelled disclosures, whether they be graphic cigarette warnings depicting the gruesome consequences of smoking or abortion counseling detailing the state’s moral view on abortion, are fine as long as they are accurate and nonmisleading.4 Her thesis — that from a Free Speech Clause perspective, only factually wrong or misleading emotional compelled speech is problematic — has the virtue of clarity. But it also has its limits. First, it defines deception too narrowly and overlooks that you can deceive and mislead with emotion as well as with facts. Second, if deceptive compelled speech triggers concern because it fails to respect the autonomy of its audience, then the government’s goals, and not just its means, merit examination. Finally, a complete analysis of compelled disclosures must also consider the autonomy of the compelled speaker.

I. Deception

Tushnet rightly condemns deceptive disclosures that are factually false or misleading. I would, however, define deceptive more broadly to also encompass compelled speech that is, for want of a better phrase, emotionally false or misleading. By that I mean compelled disclosures that intentionally exploit certain heuristics, including affect heuristics. A heuristic is a cognitive shortcut. Because our minds are overwhelmed with information, they have developed these shortcuts to help process all that information. While often accurate, heuristics can also lead to predictable errors. For example, one shortcut advertisers regularly use is to make you like something not by virtue of its own merits, but by associating it with something else that you already like. It is the difference between trying to persuade you to buy a toaster with a beautiful picture of the toaster, and trying to persuade with a picture of a beautiful woman draped over the toaster. The latter is intentionally taking advantage of a cognitive shortcut. Just as it is manipulative to intentionally provide false information (making people think positively about the toaster by falsely stating *chefs love this toaster*), it is manipulative to intentionally exploit certain cognitive shortcuts (making people think positively about the toaster by associating it with a *sexy woman*). Both compelled disclosures lead the audience to draw conclusions they would not otherwise have drawn.

I think most, if perhaps not all, of the tobacco images proposed by the FDA fall into the legitimate use of emotion category. That is, any emotional response is due to the merits of the claim, rather than intentional exploitation of an affect heuristic. You don’t recoil from smoking because it has been associated with something you already view negatively and you are transferring these negative emotions onto smoking. Rather, you recoil from smoking because the image of the diseased lung or cancerous mouth depicts the actual consequences of smoking. Any negative reaction is not mapped onto smoking from another source, but stems from the smoking itself.

Tushnet points to my hesitation in finding that all FDA-mandated images satisfy this criterion to conclude that this distinction is unworkable. It is true that the line-drawing may be difficult. Nevertheless, I am not comfortable having no line at all. Tushnet believes that government mandates are not likely to associate completely unrelated things, like the harms of smoking and maggot-infested meat. I am not sure I trust the government that much, especially in the context of states’ abortion regulations. If the government may intentionally exploit affect heuristics and put images of weeping women or maggot infested meat on cigarette packages, it opens the door to allowing the government to force women seeking abortions to look at images of weeping women or maggot-infested aborted fetuses. This deliberate exploitation undermines the decisional autonomy at the heart of free speech jurisprudence.

II. Goals of Compelled Disclosures

If decisional autonomy is what is at stake, then the state’s goals should matter. Tushnet disagrees: Provided the compelled disclosures are not deceptive, the government may, for example, equally seek to deter abortion as smoking. As long as the audience can ultimately decide whether to abort or to smoke, their autonomy remains intact.

I suspect her thesis is again propelled by the problem of line-drawing. Without doubt, there are difficult lines to draw. I even agree that some, such as the distinction between compelled disclosures meant to inform and those meant to persuade, are impossible. At the same time, I think Tushnet is too quick to reject that autonomy might also be implicated by government interference with the decisionmaking process as well as the ultimate decision. After all, decisionmaking is not limited to making the final decision but includes choosing what to consider in making that decision. Consequently, I think there are differences, and differences that matter in terms of decisional autonomy, between the government trying to persuade someone not to smoke because smoking is harmful and the government trying to persuade someone not to have an abortion because abortion is tantamount to murder.5

First, the two goals differ in that one is based on uncontested facts and one is based on a contested moral proposition. Tushnet argues that my distinction between controversial and noncontroversial is untenable because in this day and age, everything is contested.6 I disagree that everything is up in the air. Such an argument is akin to claiming there is no difference between truth and falsity because everyone disputes everything.

The smoking disclosures attempt to persuade audiences to avoid smoking because smoking will make them (and those around them) sick. That smoking is addictive and causes various illnesses is not disputed. It is a well-established scientifically based fact. The abortion disclosures, on the other hand, attempt to persuade audiences to avoid abortion on the grounds that abortion kills an unborn baby. (Here, I have in mind laws like those in South Dakota that require doctors to tell women that an abortion will “terminate the life of a whole, separate, unique living human being.”7) This is not uncontested fact. Rather, it is a moral if not religious viewpoint over which the country is deeply divided.

Second, the two goals differ in that one promotes autonomy and one potentially undercuts it. Given that smoking is toxic and addictive, odds are it will enhance your autonomy to eschew it.8 Unlike smoking, an abortion can boost a woman’s autonomy. As Justice Ginsburg has explained, the availability of abortion ensures “a woman’s autonomy to determine her life’s course, and thus to enjoy equal citizenship stature.”9

Compared to forcing people to take note of undisputed facts, forcing people to listen to the government’s moral/religious views resonates differently for autonomy, as does urging a course of action that might actually work against the audience’s autonomy. Furthermore, if avoiding state orthodoxy is a prime free speech goal, then compelled disclosures that have the goal of foisting state orthodoxy onto a captive audience ought to set off free speech alarm bells.

III. Speaker Autonomy

That the government’s goal is based on controversial moral considerations rather than uncontroversial facts also raises a compelled speech issue that Tushnet chose not to address in her paper, which is the free speech rights of the compelled speakers. State-mandated speech implicates the individual autonomy of the compelled speaker, as the right to control your speech can be violated as much by being forced to speak as by being silenced. And as the Supreme Court emphasized in West Virginia State Board of Education v. Barnette,10 which held that the state cannot force students to recite the Pledge of Allegiance, and Wooley v. Maynard,11 which held that the state cannot force drivers to carry the state’s ideological message on their automobiles, the insult to autonomy is greater when the speaker does not share the government’s viewpoint.

The compelled tobacco disclosures do not present this concern. There is no risk of infringing upon the individual autonomy of tobacco companies because they are corporations, not people. Forcing people to convey the government’s message risks treating them as a means to an end and compromising the inherent dignity with which all humans are endowed. Although corporations are sometimes considered legal “persons,” they are not, of course, actual people or like actual people.12 Corporations are not ends in themselves but instrumental entities — legal fictions created to facilitate economic growth. Nor do corporations possess an inviolable dignity such that we need to respect their autonomy. On the other hand, the compelled abortion disclosures, which force physicians to convey an ideological if not religious message, do violate individual autonomy.13 In fact, compelled ideological speech is anathema to free speech for precisely this reason.


Tushnet and I agree that the appeal to emotion in compelled disclosures is not necessarily problematic in and of itself. We both agree that the state should not compel false or misleading speech. But this alone is too modest a proposal. When the government intentionally exploits emotion heuristics, it compromises the audience’s autonomy. When it mandates an ideological message, it undermines the autonomy of both speaker and audience. Tushnet’s concerns about line-drawing have merit, but do not, in my opinion, warrant granting the government an (almost) free pass as to how it conveys its message and what goals it may choose.

Nonetheless, Tushnet’s insistence that we recalibrate our skepticism to focus more on deception and less on emotion might have prevented one of the most puzzling aspects of these compelled speech cases, which is the way the court of appeals decisions transformed the facts of smoking into ideology, and the ideology of abortion into facts. In striking the compelled tobacco disclosures, the D.C. Circuit found that the emotional disclosures necessarily equated to ideological ones.14 In upholding South Dakota’s abortion disclosures about terminating the life of a whole, separate, unique living human being, the Eighth Circuit held that because the statute defined “human being” as member of the species of homo sapiens, it was a factual statement.15 Different conclusions might have obtained had the former court accepted and understood emotion more and the latter court tolerated deception and manipulation less.

* Professor of Law, University of Miami School of Law.

1. See generally Caroline Mala Corbin, Compelled Disclosures, 65 Ala. L. Rev. (forthcoming 2014).

2. See R.J. Reynolds Tobacco Co. v. FDA, 696 F.3d 1205 (D.C. Cir. 2012).

3. See Rebecca Tushnet, More than a Feeling: Emotion and the First Amendment, 127 Harv. L. Rev. 2392, 2415 & n.110 (2014).

4. Id. at 2432.

5. Both goals seek to avoid harm, either to oneself or to others. Where the goals differ is in how the government reached the conclusion that the discouraged behavior (smoking or abortion) is harmful.

6. Tushnet, supra note 3, at 2427.

7. S.D. Codified Laws § 34-23A-10.1(1)(b) (2013).

8. Granted, a few outliers may enjoy smoking precisely because of its danger, but these exceptions do not defeat the general proposition that avoiding addiction is autonomy-enhancing.

9. Gonzales v. Carhart, 550 U.S. 124, 172 (2007) (Ginsburg, J., dissenting).

10. 319 U.S. 624 (1943).

11. 430 U.S. 705 (1977).

12. Nor do corporations equal an association of people. While voluntary associations and their members are essentially alter egos, corporations and the people who comprise them are not. On the contrary, the point of incorporation is to create a distinct legal entity so that, for example, the owners are not liable for the corporation’s debts.

13. The concern with speaker autonomy is particularly misplaced when dealing with corporation’s commercial speech, which is protected not for the sake of corporations but for the sake of audiences: “[T]he extension of First Amendment protection to commercial speech is justified principally by the value to consumers of the information such speech provides.” Zauderer v. Office of Disciplinary Counsel, 471 U.S. 626, 651 (1985).

14. See R.J. Reynolds Tobacco Co. v. FDA, 696 F.3d 1205, 1212 (D.C. Cir. 2012).

15. Planned Parenthood Minn., N.D., S.D. v. Rounds, 530 F.3d 724, 735 (8th Cir. 2008) (en banc). The Eight Circuit Court of Appeals also upheld disclosures about depression and suicide on the (shaky) grounds that they merely referred to correlation rather than causation. See Planned Parenthood Minn., N.D., S.D. v. Rounds, 686 F.3d 889, 905 (8th Cir. 2012) (en banc).