A. Rights

Following Shiner, I use the term “moral right” to refer to rights that are justified within a plausible moral system, a system that may or may not be rights-based.Footnote ⁹ This leads to the first of two distinctions drawn by Meir Dan-Cohen, the contrast between utility rights and autonomy rights.Footnote ¹⁰ Utility rights are rules justified by their role in promoting social welfare or other collective goals. Autonomy rights, on the other hand, are grounded in respect for persons, where respect requires protection for individual choice. The second relevant distinction is between original rights and derivate rights. A's right, X, is an original right if X is recognized out of direct concern for A's interests or choices. X is a derivative right if it is afforded to A in order to protect the original rights of other persons. I refer to original autonomy rights as OA rights and derived autonomy rights as DA rights.

Again, Shiner maintains that unlike natural persons, corporations cannot hold OA rights. Corporations, however, may still argue for constitutional recognition of their speakers’ rights as DA rights. Corporation speakers’ rights would be recognized to protect natural persons’ OA rights to receive commercial expression. Shiner, however, in order to undermine the case for corporation DA rights, argues that respect for autonomy does not support a general OA hearers’ right.

B. The Grounds of Freedom of Expression

According to Shiner:

For Shiner, the argument above is the “standard” one in that it summarizes the case for freedom of expression that is common to autonomy liberals such as John Rawls (the early Rawls, at least), Will Kymlicka, and Joseph Raz.Footnote ¹² According to Shiner, the wider variant of this argument, while it can include various forms of artistic expression and ideas about the good, cannot be extended to include commercial expression.

While he concedes that a precise definition of commercial speech is difficult to specify, Shiner points to “I will sell you product X at price Y” as a paradigm example.Footnote ¹³ I cannot provide in this paper a lengthy discussion of whether commercial expression promotes autonomy in a way relevant to freedom of expression. It may be that commercial information does not make the right kind of contribution to autonomy. Arguably, it does not involve the development of ideas about justice, beauty, or the good. Likewise, a paradigm example of two-way communication in commercial speech is bartering over price, which seems far removed from discussions of political, moral, and aesthetic values. If commercial expression merely facilitates the efficient execution of one's life plans, not the development of values that determine such plans, then perhaps it does not fall within the scope of freedom of expression. I am dubious of this line of argument, but beyond a few comments in Subsection C, I do not address it at length.Footnote ¹⁴

C. Speakers and Hearers

While Shiner's main concern is corporate commercial expression, he also denies that mere conveyance of commercial speech enhances the autonomy of a natural person speaker. He does, however, admit that commercial expression may enhance the autonomy of hearers (if commercial expression tells me where the cheap beer is, I can use my beer savings for other pursuits), though he later denies that this contribution to hearers’ autonomy is relevant to freedom of expression.Footnote ¹⁵

Here I see no relevant distinction between the autonomy interests of speakers and those of hearers. Consider an analogous example of political expression. Bob books a public hall to read from his favorite works of political philosophy. He reads the texts in a monotone voice, providing no commentary and answering no questions from the audience. He has decided that there is no way that he can improve upon the relevant works, but he thinks that reading from them will help to draw them to people's attention. However, a ban on Bob's reading would still interfere with his chosen participation in the discussion of political values. The same could be said of a law that prevented him from reprinting political works already in the public domain. Presumably, Bob has an OA right to pass along political speech that he regards as valuable.

Now consider Ben's distribution of price advertising. Ben reflects on advertisement information and reaches a decision parallel to that of political parrot Bob. He concludes that the information is worth passing on to hearers (it may help them with the efficient pursuit of their projects) and that he cannot improve upon this information. So he simply passes it along. Now if Bob's decision to pass along information that he deems valuable is protected by OA rights, then it is hard to see that Ben's decision is not. Neither adds to the relevant information, but both make a decision regarding that information's value to potential hearers. The political parrot enjoys OA rights protection for his chosen contribution to public discourse—passing along what he judges to be worth passing along. If this is enough to generate OA expression protection for the political parrot, it should be enough to generate the same protection for the commercial speaker.

To block the extension of freedom-of-expression protection to commercial expression, Shiner must fall back on his criticism of its informational content. If Shiner is wrong and this content does fall within the scope of freedom of expression, then both speakers and hearers, so long as they are natural persons, have OA rights to commercial expression. Of course, most commercial expression is produced by corporations. Again, if corporations lack OA speakers’ rights, they may claim to hold DA speakers’ rights that protect natural persons’ OA hearers’ rights to receive commercial expression. Here Shiner counters by arguing that there are no general OA hearers’ rights in freedom of expression.

D. The Unnecessary Duplication Principle

According to Shiner, hearers’ rights will be “robust” enough to support the commercial-expression doctrine only if they fall outside the scope of the following principle:

For Shiner, hearers’ rights are unnecessary where our intuitions concerning freedom of expression are already explained by speakers’ rights. Here they are the mere “obverse” of speakers’ rights and they provide nothing for corporations to “borrow.”

While Shiner maintains that his argument requires only that the Unnecessary Duplication Principle (UDP) apply in the context of commercial expression, he thinks that UDP is likely a sound general principle. One reason for its extension “at least within legal systems, is that legal systems abhor unlimited extensions of liability, and unwarranted extensions of standing to bring litigation.”Footnote ¹⁷ Says Shiner:

But what does it mean to say that hearers’ rights add nothing? Shiner's assessment seems to depend on the assumption that removing the relevant statute is the only suitable remedy for violations of freedom of expression. If we are open to other remedies—such as monetary damages—then recognition of hearers’ rights certainly adds something to the picture, as this would increase the state's potential liability. Even if we set this issue aside, it is clear that allowing persons to challenge a law as hearers as well as speakers opens a law to more challenges than it would face if only speakers were afforded standing. So the addition of hearers’ rights alongside speakers’ rights makes for some kind of changes to the legal system.

Presumably, Shiner thinks that such changes to the legal system are unnecessary in that they exceed what is required for full protection of OA expression rights. There are three ways of understanding this claim. I argue that each construal of UDP is either false or otherwise incapable of supporting Shiner's argument against hearers’ rights.

First, one might claim that both hearers and speakers have coextensive freedom-of-expression OA rights but still deny the need for general recognition of hearers’ rights at the constitutional level. Perhaps speakers’ constitutional rights already justify removing any law that impedes hearers’ OA rights. But this does not marginalize hearers’ rights in the way that Shiner wants. It concedes their general OA status, denying constitutional recognition only if such recognition is unnecessary for their protection as OA rights. This raises the question of whether hearers’ OA rights actually do receive full protection from speakers’ constitutional rights. I discuss whether hearers require general constitutional rights for full protection of their OA rights in Subsection H.

UDP might also be interpreted as reducing hearers’ rights to mere DA rights. On this view, only speakers ever enjoy OA rights. Thus hearers should have constitutional rights only where such rights are needed for complete protection of speakers’ OA rights. For example, imagine a speaker who is unwilling to challenge a statute that infringes her OA rights. Allowing an interested hearer to contest the statute could provide advanced protection for future speakers’ OA rights. This version of UDP seems to serve Shiner's purposes. If corporate speakers lack OA rights, then there is nothing in corporate speech for hearers’ constitutional rights to protect. Hearers’ constitutional rights would be recognized only for the protection of natural persons’ OA speakers’ rights. However, where Shiner does accept hearers’ rights, he treats them as OA rights, not mere DA rights. Moreover, in Subsection G, I argue that Shiner's “standard” argument precludes him from reducing hearers’ rights to DA rights.

Finally, one might hold that hearers’ OA rights exist, but only where speakers’ OA rights are absent. In all other cases, hearers’ rights are the mere “obverse” of speakers’ rights. On this view, infringements of freedom of expression sometimes occur without speakers’ rights, and in these cases, the infringement is to hearers’ OA rights. This seems to be what Shiner has in mind with UDP. However, as I argue in Subsections E and F, the idea that hearers’ OA rights can emerge only in cases that lack speakers’ OA rights is not cogent. Moreover, even if hearers’ OA rights required the absence of speakers’ OA rights, the commercial-expression doctrine would be unaffected. In corporate-expression cases there are no speakers’ OA rights to duplicate, so constitutionalizing hearers’ rights to receive commercial expression does not run afoul of UDP.

This last point shows Shiner's assertion that UDP need apply only in the context of commercial expression to be seriously confused. If commercial expression's content falls outside the scope of freedom of expression, then there is no work for UDP to do. But if commercial expression's content is covered by freedom of expression, UDP cannot provide Shiner with a backup argument. In this case, commercial expression is protected by speakers’ OA rights when speakers are natural persons and by hearers’ OA rights when speakers are corporations.

In the following sections, I argue that hearers have general OA rights to receive expression. I conclude Section I by discussing how constitutional recognition of hearers’ rights would affect standing and liability.

E. The Subject Focus Principle

Hearers’ OA rights, if they exist at all, cannot disappear from a case just because speakers’ OA rights are added to the mix. My argument for this claim is based on the principle that determining whether a subject holds an OA right requires us to focus on features of the subject and that the question of whether a subject's right is infringed must focus on the treatment of the subject. Suppose that subject A is held to have an OA right R that is infringed by action X. If so, then subject B has the same right R infringed by action Y unless we can point to morally relevant differences between subjects A and B or between actions X and Y with respect to their effects on A and B. Call this the Subject Focus Principle (SFP).

For OA rights, the question of whether an individual holds a given right is answered by determining whether she holds the relevant intrinsic properties. I hold such a right because I am a certain kind of agent, not because attributing that right to me serves some social purpose. Likewise, whether my OA rights are infringed is a question about what is done to me (and, if my rights have been waived, what I have done to make this the case). In what ways are my liberties or opportunities affected? If my rights are infringed in one scenario but not in another, then there must be a change from the subject's perspective. Whether a rights infringement is justified, all things considered, may turn on questions of broader social interests, but the question of whether the right has been infringed concerns what is done to the subject.Footnote ¹⁹ So we should be suspicious of a principle that makes the determination of rights infringements turn on the question of whether such recognition is “necessary.” OA rights are not utility rights; they are not determined by social cost-benefit analysis.

Of course, what is done to one person is often bound up with what is done to others. I may be deprived of an opportunity to communicate with another person through direct interference with her actions. Still, if an action infringes my rights, then, at least prima facie, much the same action infringes the rights of another similarly situated person—or me again, similarly situated on a different occasion.Footnote ²⁰

Suppose that Bob lacks his own OA right to distribute his expression. However, we determine that preventing Sarah from hearing Bob is an infringement of Sarah's OA rights; preventing her from accessing Bob's ideas cannot be squared with respect for Sarah as a self-determining autonomous agent. Change just one thing here. Bob does in fact have his own OA speakers’ rights. A simple change in Bob's status should not alter our assessment. Why should Bob's being wronged, if he is wronged by a restriction on Sarah's liberties, imply that these restrictions are no longer a wrongful interference with Sarah's autonomy? Here the state of being wronged is not like an object that only one person can own. Both are wronged if both have suffered an objectionable loss to their autonomy.

Of course, Shiner may think that hearers have general OA rights but that general constitutional recognition is not necessary for their protection. However, Shiner seems to assume that removing the relevant statutory limitation always provides full remedy in free-expression cases. This might be the case for utility rights. Here what is most important is moving forward with laws that optimize the future benefits of freedom of expression. Removing the relevant statute satisfies this goal. However, for OA rights, we cannot assume that removing the statute fully remedies the losses an individual suffers from a rights infringement. Even if removing the limitation is the only suitable remedy for statutory infringements, monetary damages could still be appropriate where governmental interference lacks statutory authority (say, with wrongful police action or regulatory controls that exceed statutory authority). I say more about this issue in Subsection H.

I now turn to the example-based argument that Shiner uses to defend both UDP and his marginalization of hearers’ rights, using SFP to refute that line of argument. I then turn to a more detailed discussion of the grounds of hearers’ OA rights in Subsection G.

F. The Barren Marketplace

Shiner concedes that speakers cannot exercise meaningful freedom of expression in the absence of an audience, what Justice Brennan called “a barren marketplace of ideas.” In Shiner's words:

Shiner provides three examples that are intended to drive a wedge between the audience's importance to speakers’ OA rights and the stronger notion of “robust” hearers’ rights:

Shiner maintains that while it is the audience's physical liberty that is constrained, it is Oppol's OA expression rights that are infringed.Footnote ²³ I accept that the constraint on the audience's physical liberty denies Oppol an opportunity for meaningful exercise of his freedom of expression. However, I deny that only Oppol's OA expression rights are infringed here. Consider the next two examples:

While time, place, and manner restrictions on speech are often deemed legitimate, (2) is clearly intended to prevent the speaker from reaching any significant audience. Moreover, according to Shiner, as in (1), the government infringes Oppol's expression rights through limitations on the physical opportunities of his audience. Only in Scenario (3) does Shiner acknowledge hearers’ OA rights:

Shiner claims that (3) cannot infringe Oppol's OA right to speak, as this would imply that freedom of expression subtends a right to choose one's audience.Footnote ²⁶ Because he finds this implausible, Shiner concludes that (3) infringes the hearers’ rights of Oppol's party members.

There are two problems with Shiner's account of these examples. First, a right to choose one's audience is actually quite plausible. Of course, freedom of expression does not subtend guarantee of the audience of one's choice. My right to freedom of expression is not infringed just because my preferred audience refuses to listen. Nor is it infringed just because the government fails to assist those who wish to listen to me. However, if the government seeks to prevent me from reaching some of my willing audience, does this not infringe my freedom of expression? Here Shiner cannot complain that the restrictions are imposed on the audience's physical liberty, as he holds that restrictions on the liberties of others can infringe one's freedom of expression.

Shiner might respond that restrictions on the liberties of Oppol's audience do not infringe Oppol's rights until those restrictions deprive him of any significant audience (as the restrictions in (1) and (2) do). But surely things do not have to go this far before there is some infringement of his speakers’ rights. Shiner might argue that Oppol's OA rights are infringed once he is deprived of a reasonably sized willing audience. However, it is far from clear what criteria we should use to determine the threshold for a reasonably sized audience. It makes more sense to accept that a speaker's freedom of expression is infringed whenever the government attempts to prevent him from communicating with willing hearers. The difference between (1) and (3) is a matter of degree, not kind.

Now for the second problem. Accept for the sake of argument that hearers’ rights, those of Oppol's party members, are infringed in (3). What is done to the members of Oppol's party? The government imposes restrictions on their liberties intended to prevent them from hearing Oppol's speech. But this is precisely what is done to all citizens in (1). Thus, if we accept SFP, it is not possible for hearers’ rights to be infringed in (3) but not in (1). There is no morally significant difference between how hearers are treated in (1) and how they are treated in (3).Footnote ²⁷ If a restriction on my liberties or opportunities infringes my freedom of expression, this cannot change simply because the government imposes those restrictions on additional people. Presumably, Shiner would find an arrangement that was intended to impede the opportunities of only some hearers to be an infringement of their OA hearers’ rights. Here again, however, the infringement of hearers’ OA rights cannot disappear just because more hearers are subjected to the same restrictions.

Where does this leave us? (1) and (2) are fairly comprehensive restrictions on Oppol's right to freedom of expression. However, (3) must also be considered a restriction on his freedom of expression. Likewise, (3) is a restriction on the audience's right to freedom of expression, but so are (1) and (2). Hearers’ and speakers’ rights are coextensive in these cases.

SFP shows that hearers’ OA rights, if they exist, must be general ones. Of course, Shiner might counter by endorsing the version of UDP that treats hearers’ rights as mere DA rights. But this move cannot be squared with his “standard” argument.

G. The Basic Status of Hearers’ Rights

Consider one of the cases in which Shiner is willing to recognize hearers’ rights. In Lamont v. Postmaster General of the United States, the late Corliss Lamont launched a freedom-of-speech challenge to U.S. Post Office regulations that placed procedural restrictions on publications identified as “communist political propaganda.”Footnote ²⁸ The government counterargued that the material in question had been mailed by foreign governments, which hold no constitutional rights under U.S. law. However, Lamont's challenge succeeded, as have similar hearers’ challenges in which the would-be speakers, as foreign nationals, could not claim constitutional rights.Footnote ²⁹ As Shiner recognizes, it is hard to make sense of the result in cases such as Lamont without recognizing OA hearers’ rights. In cases such as Vasquez (discussed below), affording hearers standing protects the constitutional rights of future domestic speakers. But in Lamont, the restrictions target only speakers who lack such rights. If there is a rights violation in Lamont, it must be to Lamont's hearers’ rights.

However, according to Shiner, the peculiar circumstances of cases such as Lamont preclude them from supporting general OA hearers’ rights. Now, there is something a bit odd about such cases. While their speakers lack domestic constitutional rights, surely we attribute OA expression rights (albeit nonenforceable ones) to these speakers if they are natural persons. However, this complication can be eliminated with a simple variation, an example used by Alexander. Suppose that the government bans a book by a deceased author. The author has no rights now, so the rights violation, if there is one, must be to the OA rights of hearers. As Alexander points out, the banning of books is a paradigm free-expression case. A plausible theory of freedom of expression must account for this example.Footnote ³⁰

Now, one might try to defend the exclusivity of speakers’ OA rights by endorsing the continuing rights of deceased speakers. But this would ignore the obvious. Even if the dead speakers’ expression rights continue, it should be clear that the status of banning books as a violation of freedom of expression does not depend on the continuing rights of the dead. The banning of books infringes the expressive autonomy of hearers, and this infringement can be understood in terms of Shiner's “standard” argument. So even if the author no longer has any rights, the activity is protected because the reader has an OA right to read the book.Footnote ³¹

Free expression concerns participation in the social exercise of discussing values and ideas. Speakers may assert their ideas in an attempt to gain the assent of other persons, but their expression rights are also protected in order to give them the opportunity to gain critical feedback. If we deny them this opportunity, we place an objectionable constraint on their expressive autonomy. Likewise, hearers may simply be looking for someone to give them the “right” answers, but they may also be seeking input for critical reflection on their standing views. Thus restrictions on freedom of expression limit the autonomy of speakers and hearers in similar ways; they impede the shared enterprise of value development.

Freedom of association works much the same way. If the government passes restrictions that prevent two individuals or groups of individuals from meeting or working together for some social purpose, it restricts the freedom of association of both parties.Footnote ³² Such a restriction could be imposed through threats of criminal sanction on only one of the two parties. Even here, however, it is easy to understand the restriction as one that interferes with the associative freedom of both parties. Now, one might say that freedom of association is different in this regard—there is symmetry between parties in freedom of association, unlike freedom of expression, where the role of the speaker seems more basic.

But this is misleading. Within free associations there may be asymmetrical relations based on hierarchical authority structures. And free expression may involve a relationship of exchange where one moves back and forth between the role of speaker and hearer. (Suppose that Oppol planned to take questions from his audience.) The leaders of a hierarchical association are not the only ones whose freedom of association is infringed if that association is hindered by legal restrictions.Footnote ³³ Their rights to freedom of association are not more basic than those of others within their group. The same holds for freedom of expression. Where speaker and hearer wish to communicate, inappropriate legal restrictions infringe the OA expression rights of both parties.

Freedom of association protects the sharing of activities, while freedom of expression concerns the sharing of information and ideas. As technology advances, our ability to transfer ideas between persons at great distances is increased, as is our ability to preserve the contributions of persons now dead. But freedom of expression is still in many ways a form of associative freedom that protects links between persons who seek to exchange ideas.Footnote ³⁴

If hearers are wronged when the government prevents them from receiving the messages of dead authors, then they are also wronged when they are prevented from receiving the messages of living ones. There is no relevant difference in the treatment of hearers between the two cases. The fact that speakers’ rights are infringed in the latter kind of case shows merely that these cases are closely aligned with infringements of freedom of association.Footnote ³⁵

Of course, not all free-expression cases are closely aligned with freedom of association. For Baker, the value of self-realization is often achieved through “solitary” or self-expressive speech.Footnote ³⁶ A key example would be writing in a diary that is not intended for the eyes of other persons. Shiner does not appeal to such cases in his argument for the primacy of speakers’ rights, and, as Alexander's discussion shows, there are cases of “solitary” hearers (examples similar to the above case of the reader of a deceased author's book). I discuss the relevance of what I call pure speakers and pure hearers in Section II.

H. Constitutional Redundancy

A. The Defense of Speakers’ Rights

According to the “standard” argument, freedom of expression provides a protected space for the choices needed for individual autonomy and self-realization. As I argue above, Shiner fails to appreciate the way in which this argument supports hearers’ basic rights. The “standard” argument holds that respect for individual autonomy requires that individuals be allowed to participate in the development of values. We might view this as an idiosyncratic project—individuals simply need freedom from interference so they can work out what they wish to do with their own personal lives. But the “standard” argument holds that respect for autonomy requires that individuals be permitted to interact with others in order to espouse, develop, and reflect upon both personal and social values. It should be plain enough that speakers’ rights are also required for participation in the development of values.

Suppose that speakers’ rights are never violated except in cases where hearers’ rights are also violated. Suppose further that there are cases in which hearers’ rights are violated even in the absence of a rights-bearing speaker. This may “suggest” that speakers’ rights are merely derivative ones. But this cannot, by itself, displace the intuition that a silenced speaker suffers an objectionable constraint on her autonomy. Such a speaker is prevented from making a contribution to public discussion, denied the opportunity to contribute to other persons’ thoughts. Of course, hearers are also wronged—they are deprived of a desired connection with another person. In this context, freedom of expression is a kind of associational freedom—restrictions on it wrong both parties. The “standard” argument, in emphasizing participation in the development of values, captures this associational aspect of freedom of expression. Of course, if there are pure-hearer cases, we must account for these as well, or our account of freedom of expression will be incomplete. But accounting for pure-hearer cases should not lead us to marginalize the associational cases, as these are an important part of freedom of expression. If we find no pure-speaker cases, then it turns out that there is a nonassociational aspect to free expression that concerns only hearers’ rights.

One could strain to show otherwise by insisting on a derivative account of speakers’ rights in speaker/hearer examples. However, the fact that speakers’ rights protect hearers’ rights does not show that the former have only instrumental value.

Proving that speakers’ rights are mere DA rights would require the adoption of one of the following two principles (call these elimination principles), both of which should be rejected. First, we might invoke a simplicity principle, arguing that an account of freedom of expression should recognize as few basic rights as possible. But whatever value simplicity may have, it cannot require us to suppress important aspects of moral experience; the intuition that a silenced speaker is wronged should not be set aside just for the sake of simplicity. Second, we might hold that the only OA rights in freedom of expression must be ones that can stand by themselves (in the absence of another party's correlative rights). But strict adherence to such a principle distorts our understanding of a right that much of the time concerns exchange between persons; it ignores the social aspect of our liberties. The freedom to read is a paradigm example of freedom of expression, but it is not the paradigm example of freedom of expression. A town-hall meeting where people speak, listen, and respond is also a paradigm example. We cannot achieve a full understanding of this example by focusing only on those elements that it shares with the reading case (and as I argue in the next section, the reading case is not quite a pure-hearer case).

I am not saying that Alexander actually endorses one of the elimination principles. My point is that he needs one of these principles to dispose of speakers’ rights. Moreover, even if we accepted an elimination principle, it would dispose of speakers’ rights only if there were pure-hearer cases and no pure-speaker cases. There are pure-hearer cases, though only one of Alexander's examples qualifies as such. Moreover, there are genuine pure-speaker cases as well.

B. Alexander's Criticism of Track Two Jurisprudence

Alexander argues that if there is a moral right to freedom of expression, that right is a negative right of hearers to receive information. For him, the analysis of important free-expression examples reveals that what we value in freedom of expression is hearers’ freedom to receive information. However, Alexander argues that there is no moral right to freedom of expression. In his view, an ideal of “evaluative neutrality” is crucial to freedom of expression; if the government decides free-expression cases by assessing the value of information, then it decides what people should hear. Yet few think that the right to freedom of expression is absolute, so free-expression decisions require us to balance the value of freedom of expression against other values. To preserve evaluative neutrality, we must assign to freedom of expression either a constant overriding weight or zero weight when weighing it against other values, making it either absolute or impotent. The only other approach is to engage in a kind of contextual balancing where the expression in question is weighed against other values, but this does violate evaluative neutrality.Footnote ⁶⁰

Once again, Track One regulations are intended to determine what ideas are expressed or received. Track Two regulations, while they interfere with expression, are concerned with noncommunicative aspects of the regulated conduct. If the government bans hate expression to prevent the spread of racist ideas, then the ban is a Track One regulation. On the other hand, if the government bans the distribution of paper pamphlets on city streets in order to prevent excess litter, the pamphlet restriction is a Track Two regulation.Footnote ⁶¹ According to Alexander, a principled Track Two jurisprudence is not possible; freedom of expression can be concerned only with Track One laws (though for the reasons outlined above, Track One jurisprudence also fails).Footnote ⁶²

As Alexander explains, the U.S. Supreme Court's Track Two jurisprudence concerns two kinds of cases: public-fora cases and symbolic-speech cases. Public-fora cases concern the use of public facilities, such as buildings, parks, and streets, while symbolic-speech cases concern the use of nonlinguistic forms of communication, the burning of draft cards, for instance.Footnote ⁶³

According to Alexander:

For Alexander, there is no reliable distinction between conditions that are “facilitative” of expression (“time, place, and manner”) and communicative aspects of expression (aspects of the message itself). When government regulates the time, place, and manner of speech, it affects aspects of the message itself; here the choice of a place to speak may be just as important as the use of particular symbols or expressive activities. This makes the adequate alternatives element of the Track Two test problematic, as all governmental regulation affects the content or communicative aspects of expression.Footnote ⁶⁵

More importantly, Alexander also argues that there should be “no distinction between Track Two laws directly regulating speech activities and all other laws.”Footnote ⁶⁶All Track Two laws regulate expression indirectly in the sense that the government's goals are not related to communication. However, as Alexander points out:

According to Alexander, a law that precludes pamphleting in airports may deprive travelers of certain religious or political messages, but at the very least, this law conveys information about what it is like to travel in airports without litter and traffic congestion.Footnote ⁶⁸ Likewise, any law that prevents speakers from communicating certain messages is likely to enable other speakers (or the same speakers) to communicate different messages. Laws banning actions such as pamphleting on public streets place obvious restrictions on expression, but laws regulating matters of distributive justice must have profound message effects.

A Track Two claim is a claim that a law X that impedes freedom of expression (by impeding the transmission of information Y) be removed. But while the removal of X would facilitate the transmission of Y, the removal of X would also impede or prevent the transmission of some other information, Z. Thus, assigning a constant weight to the value of freedom of expression is pointless, as free-expression interests are always aligned against each other in Track Two cases. Both the legal status quo and the removal of X cost hearers information, so the free-expression constant appears on both sides of the dispute. The only way to break this tie is to evaluate X's informational consequences, weighing the value of the information lost against the value of the information gained.

Here we face two problems. First, we cannot know what information would be produced under an alternative legal status quo; we can only speculate. Second, even if we could know this, assessing the relative value of X's information would violate evaluative neutrality. Again, Track Two decisions would be decisions about what information people should receive. (A Track Two decision would also require us to assess the value of the information relative to costs to nonexpression values such as security, property, and privacy.)Footnote ⁶⁹

C. Speakers’ Rights and Track Two Jurisprudence

Recall Alexander's main contenders for the role of freedom of expression's central principle:

1. (4) Freedom of expression is implicated whenever conduct intended to communicate a message is suppressed or penalized with the result than an audience is prevented from receiving the message.

2. (5) Freedom of expression is implicated whenever an activity is suppressed or penalized for the purpose of preventing a message from being received.Footnote ⁷⁰

(4) and (5) provide the basis for Track Two and Track One jurisprudence respectively. For Alexander, (5) is the superior principle, and this entails that free expression should be restricted to Track One jurisprudence (though he argues that even this version of free expression fails). Again, my own view is that free expression cannot be captured by a single principle. (4) and (5) capture different aspects of freedom of expression, though neither succeeds as a “whenever” principle.

(5) is closely aligned with arguments put forward by Thomas Scanlon and David A. Strauss.Footnote ⁷¹ In general, Track One laws violate respect for autonomous persons because these laws use coercion to control and suppress thought itself. The objectionable character of thought control explains why Track One laws implicate the activities of pure speakers and pure hearers as well as communicative expression. Of course, (5) still requires some qualifications for legitimate message-targeting laws: for instance, protection for privacy, tort rules concerning defamation, and perhaps even regulation of hate speech. Now, according to Alexander, we cannot provide a satisfactory account of the distinction between legitimate message-targeting laws and Track One laws that violate the commitment to freedom of thought and expression. Obviously a full defense of free expression would require an answer to Alexander on this point, and I cannot provide that in this paper. For now I simply note that subject to the exceptions noted above, all expressive activity enjoys protections against Track One laws.

On the other hand, allowing pure speakers and pure hearers to advance Track Two claims would make virtually all laws Track Two laws. Whenever an action is precluded by law, an avenue for pure-speaker self-expression is closed, and an avenue for pure-hearer experience/information reception is closed as well. Thus a Track Two jurisprudence for these expressive activities threatens to transform freedom of expression into a more general right to liberty of action. We lose the notion of a distinctive liberty right to freedom of expression.

This is the problem faced by Baker's account of free expression. Baker would extend freedom of expression to all nonviolent or noncoercive expressive activity. Now, I do not dispute the legitimacy of his restriction—permitting violent expressive activity cannot be squared with respect for other persons as autonomous agents. Baker's restriction is fine as a necessary condition for free-expression coverage, but it cannot be a sufficient condition. Any sane account of a general right to liberty would also be subject to a qualification of nonviolence, so Baker's account still fails to distinguish freedom of expression from a general right to liberty.Footnote ⁷²

All laws restrict individual liberty to some degree, but laws that mediate the conflicting pursuits of individuals are necessary in a society that provides room for autonomous lives. If a distinct free-expression claim is to be pressed against such laws (which are not Track One laws), it must be something more than a claim to the exercise of nonviolent liberty.Footnote ⁷³

So we should restrict Track Two jurisprudence to cases involving communications between willing speakers and willing hearers, both with free-expression rights. For all intents and purposes, interactions with young children would be treated as either pure-hearer or pure-speaker cases, as young children lack the rights required to engage the associative aspects of freedom of expression. This entails that laws intended to secure parental rights of control and custody over children do not implicate freedom of expression. Young children, lacking their own expression rights, are merely one source of information that hearers might wish to access—restrictions on access to children would not raise freedom-of-expression concerns unless they were Track One laws. One might argue that private unobtrusive expression also requires a form of Track Two protection, but this is not necessary. Precisely because such private expression does not present an immediate problem for others (such as traffic congestion or litter), it is unlikely to be directly constrained by Track Two laws. Government's concern with private expression is far more likely to target the ideas involved in such expression and how an individual might subsequently act on those ideas. This would invoke Track One protection.

Hearers do not have a Track Two claim to information in general—they have a Track Two claim against interference with their communication with willing speakers. With this in mind, consider again the example of banning the distribution of literature in airports:

On Alexander's terms, freedom of expression requires us to weigh the pamphleteers’ information against that provided by an uncongested airport. But if Track Two jurisprudence is restricted to communication between willing speakers and willing hearers, the latter information falls outside the scope of freedom of expression. It is a pure-hearer claim that would be implicated only by Track One laws motivated by the goal of suppressing access to this information.

Of course, one might reformulate Alexander's concern as a question about the costs of respecting freedom of expression. If freedom of expression is not absolute, then presumably we must still weigh the value of the uncongested airport's information against that of the pamphleteers to determine if the cost of free expression is too high. However, to raise serious concerns about the costs of permitting expression, we need to do more than assert that the expression restricting regulation provides “something that it is like” information. It is true that there is “something that it is like” to move about in an uncongested airport. But there is also “something that it is like” to move about in an airport in which people go about the business of both communication and travel; I receive this “something that it is like” information even if I choose not to engage with the pamphleteers myself. Since there is “something that it is like” information on both sides, the simplest approach is to treat them as canceling each other out in assessments of freedom of expression's costs.

Recognizing speakers’ rights also requires us to reconsider the idea that all laws are Track Two laws. It is true that all laws affect what is said, when it is said, and to whom it is communicated. Now, again, if we think that hearers’ claims to receive information, and only those claims, ground freedom of expression, then any law that “affects” what is said implicates freedom of expression. Again, this would be all laws. However, if hearers have Track Two claims only to information from willing speakers, then laws that merely “affect” what a speaker says do not implicate hearers’ rights. If the government opens travel to previously restricted parks, this may affect my speech in that I then choose such travel over speaking opportunities. It would be different if I was prevented from addressing hearers by a law that penalized my giving a public address. However, if regulations play a noncoercive role in influencing my speech, potential hearers have no legitimate complaint—they have a claim only to what I choose to communicate.

Alexander accepts that the right to freedom of expression must be a negative right—hearers have a right of noninterference in their reception of what speakers chose to communicate—not a positive claim that speakers communicate with them. However, his expansive sense of Track Two laws implies at least a weak version of a positive right for hearers—hearers have a claim against the government that it encourage speakers to provide genuinely beneficial information (at an appropriate cost in terms of other goals). All laws are Track Two laws because all laws “affect” speech—the implication here is that government should attempt to “affect” speech in an optimal way.Footnote ⁷⁵

Of course, this distinction between coercion and merely affecting needs development, and this brings us back to the distinction between “direct” regulation of speech and other laws that constrain speech. It seems obvious that if the government penalizes my speaking on the street, it restricts my freedom of expression. Yet other laws penalize not expression itself but activities or use of resources that support the exercise of expressive liberty. According to Susan Williams:

Alexander maintains that this line cannot be drawn. Laws that impede activities “closer” to speech may seem more important than other laws, but the latter—rules concerning property ownership and wealth distribution, for instance—can have even more profound message effects than the former. This problem is not removed just by recognizing speakers’ rights—here laws do not merely affect what hearers receive; they determine what speakers are able to say.

The solution, I think, lies in emphasizing that freedom of expression is a negative right; it does not call for rectification of distributive injustice (in terms of wealth, property, and resources). Background inequalities do have profound effects on what speakers are able to express, but building concern for such effects into freedom of expression pushes it in the direction of a positive right. To resolve this problem, we could stipulate that Track Two jurisprudence does not respond to claims that require rectifications to the current distribution of wealth, property, and assets. Instead, Track Two jurisprudence could require that government avoid penalizing activities and resources that are generally open to persons regardless of their distributive share. This approach would leave the government free to regulate resources more commonly at the disposal of better-situated citizens, and it would require sensitivity to the ways in which persons with lesser distributive shares make use of their freedom of expression.

However, this takes us only so far. There are many laws that impose penalties on expressive conduct itself, and these laws need not be motivated by the goal of suppressing expression. Surely not all of these laws implicate the right to freedom of expression. Again, the distribution of pamphlets in airports or on public streets adds to the costs of sanitation. So it is reasonable to expect that penalizing pamphleting would conserve resources that could be spent on security, health, or even education. Thus to permit pamphleting and protesting (which requires additional sanitation and policing expenditures) subsidizes these expressive activities at the expense of other projects. In these cases, we are usually confident that permitting expression is worth the costs, but again, how do we distinguish these cases from the speeding driver? For Alexander, Track Two jurisprudence must treat maintaining the 55-mile-per-hour speed limit as a refusal to subsidize another expressive activity; in this case, safety and traffic efficiency are deemed more valuable than this loss of expression, a decision that violates evaluative neutrality.

Again, we could deal with this problem by treating the driver as a pure-speaker or a pure-hearer case, both of which are restricted to Track One protection. But this is not sufficient, as we can imagine a version of the speeding driver case that includes a willing audience. Principle (4) makes this version of the speeding driver a freedom-of-expression case, which is unacceptable. So restricting all of freedom of expression to Track One jurisprudence may seem like the best approach here. However, restricting all freedom of expression to Track One would also entail that the refusal to permit public protest does not implicate freedom of expression if it is motivated by conservation of the resources needed for security and not a desire to suppress messages. This may seem counterintuitive, but Alexander sees no principled way of distinguishing speeding-driver cases from conventional expression cases.

We need principles that distinguish the speeding case from conventional examples of protected expression in a way that avoids evaluating the relevant information. Here one might appeal to safety, arguing that speeding is impermissible because it poses risks of physical harm to third parties. However, even a 55-mile-per-hour speed limit carries some risk of injury. Likewise, a political protest generates some risk to third-party safety as well—participants in the protest may turn to violence, as may persons deeply offended by the protesters’ message. Yet others, who have no concern with the message, may use crowding and tension as tools for facilitating violent mischief.

We might say that speeding takes us above the normal level of safety, but the “normal” level of safety set by the 55-mile-per-hour speed limit is a product of a policy choice and subject to change. It is hard to see how this level of safety/risk functions, by itself, as a morally significant baseline. Likewise, a political protest may elevate the risk of harm beyond the “normal” risk found in the absence of such a protest.

There is, however, a significant difference in the way that these different expressive activities elevate the risk of harm. Note that in the parade protest case, it is not so much the expressive activity itself that raises the risk of harm but rather the ways in which people respond to it. There is also the problem of protesters themselves exceeding the bounds of mere protest, using the parade as an opportunity to organize violent actions. All of these risks concern responsible wrongdoing that may occur along with the legitimate expressive activity, which, by itself, need not elevate the risk of injuries. A similar point applies to pamphleting. While allowing pamphleting will likely increase the level of litter in an airport or on a public street, the litter still results from morally responsible decisions—the choice not to take the trouble to dispose of pamphlets properly. The speeding case is different. Speeding increases the risk of injury, but here the increased risk has little to do with the likelihood of subsequent responsible wrongdoing. By driving faster the driver is simply pushing against the limits of his own ability to prevent accidents and against the same limits of other drivers and pedestrians.Footnote ⁷⁷ Here, unlike the protest case, there is an important sense in which the expressive activity itself increases the risk of injury.Footnote ⁷⁸

The considerations above support the following principles. (1) Track Two claims do not apply to expressive activities that in and of themselves elevate physical risks for persons with no desire to participate in those activities.Footnote ⁷⁹ Allowing such Track Two claims would in effect draw persons into associations that they do not wish to join.Footnote ⁸⁰ (2) Risks of physical injury that would be the result of wrongful actions by responsible persons should be discounted in assessing the costs of expressive activity in Track Two analysis. This is bound to be controversial, and I have no clear answer regarding the extent of such discounting. But this principle seems to reflect our commitment to freedom of expression—we are inclined to pay predictable costs that cannot be directly linked to the responsible use of freedom of expression itself. Together, these principles explain why we do not take the speeding case seriously as a freedom-of-expression case and why we are inclined to permit conventional expressive activities.

This does not answer all the problems of Track Two jurisprudence. First, what I am putting forward does not completely eliminate the need for evaluating information in expression cases. I am suggesting that the costs of expression be discounted in those cases where the costs do not result from the responsible exercise of freedom of expression. That still leaves unanswered questions about when these costs are high enough to justify Track Two laws. Second, nothing here answers Alexander's argument concerning the effects of time, manner, and place regulations on the content of speech. If Alexander is right, then seemingly innocent regulations can undermine the speaker's ability to communicate his message. Third, what I am suggesting here may be deemed too modest, as I deny that the right to freedom of expression requires the government to expand opportunities for expression (by addressing the way distributive injustice affects opportunities for expression). However, in my view, a modest Track Two jurisprudence is the best that we can do, and recognizing speakers’ rights goes some way towards answering Alexander's criticisms of this part of freedom of expression.