Seana Shiffrin's Speech Matters Footnote ¹ is a remarkably rich, original, and insightful work, casting new and sometimes surprising light on a wide range of issues in law and morality.

Chapter 1 provides a careful and revealing analysis of the wrongness of lying, framed as an investigation of the permissibility, in Kant's famous example, of lying to the murderer at the door in order to protect the intended victim, who is hiding in your house. Shiffrin believes that it is permissible to lie in such a case, but her theoretical aim in the chapter is to argue that in general there is a distinct moral objection to lying, as opposed to deceiving a person by other means. I examine below the forceful if somewhat elusive arguments she offers for this conclusion.

Chapter 2 takes up the question of the validity of coerced promises, continuing the theme of whether and how the wrongful actions and intentions of others can affect what it is permissible for us to do. Shiffrin argues that in many cases we are not bound by coerced promises but that we should respond to them in ways that leave open a path through which valuable moral relations with the coercer can be restored.

In Chapter 3, Shiffrin sketches a novel account of freedom of speech, based on our needs as thinkers, and contrasts this with more familiar speaker-based or audience-based theories. Chapter 4 considers whether, on this account, lies should have freedom-of-speech protection. Laws against lying do not, she argues, directly threaten core free-speech values, although they may be open to objections of other kinds. Chapter 5 argues, on grounds of equality, that the law should “accommodate” certain moral and prudential lapses by, for example, not prosecuting some lies about one's past. Finally, in Chapter 6, Shiffrin argues against claims that the goals of institutions such as universities, governmental agencies, or police forces justify restrictions on the freedom of speech of those working within them or justify lies told in order to promote those ends.

Shiffrin defends specific positions on many of the moral and legal issues dealt with in these chapters. But her emphasis throughout is on the values that are at stake, frequently identifying values that have generally been overlooked or showing that familiar values have a different shape from what is commonly supposed and therefore different implications for what should be done. Investigation of this kind is one of the ways that philosophy can provide the greatest insight, and it makes the book an outstanding example of fruitful interaction of philosophy and law. A second distinctive feature of the book is that the values Shiffrin emphasizes are in many cases not just considerations that are relevant to moral rightness and wrongness because we have nonmoral reasons to care about them but, rather, moral values in a stronger sense—things that we have reason to care about because of their moral value, such as individuals’ moral character and moral agency, and the moral value of their relations with one another. Readers may vary in whether they find this emphasis exciting and enlightening, or unnecessarily moralistic, or, as in my own case, sometimes one and sometimes the other.

There is a great deal to be learned from close scrutiny of every chapter of Shiffrin's book, but in this review I focus on Chapter 1, on the morality of lying, and Chapter 3, on freedom of speech.

The subtle analysis of the wrongness of lying that Shiffrin offers in Chapter 1 is framed as an investigation of the permissibility, in Kant's famous example, of lying to the murderer at the door in order to protect the intended victim who is hiding in your house. Shiffrin believes that it is permissible to lie in such a case.Footnote ² But the more general theoretical question she investigates is whether there is a particular moral objection to lying, distinct from whatever objections there may be to deceiving a person by other means. It might seem that there is no such objection. Suppose you were able to avoid being confronted by the murderer by misleading him about the whereabouts of his intended victim in some other way, such as by making footprints in the fresh snow leading from your house to a nearby bus stop. It might seem to be mere moral prissiness to breath a sigh of relief because at least you did not have to lie to the murderer. Shiffrin argues that this would not be mere prissiness, because there is indeed a distinct moral objection to lying. Her argument for this conclusion proceeds through a number of steps.

The first thing that is special about lying has to do with content. One can deceive a person about many facts about the world. Lying can do this, but it does so by deceiving the person about one's beliefs. Shiffrin's second point is that being able to come to know what others believe is distinctively important for us. We have, she says, “foundational interests in knowing the contents of each others’ minds” (9). This access is important not only for our personal relations with people who are close to us but also for the political relations between citizens and, Shiffrin argues, is necessary for our own development as moral agents (26). This claim about the development of our moral agency seems to me a little overblown and in any event not necessary to the force of Shiffrin's argument. So I focus on the importance for our relations with one another having reliable knowledge of what others think and feel.

The next step in Shiffrin's argument is that (truthful) communication is a uniquely important means for doing this. We have no good means of coming to know what others think and feel if we cannot rely on what they tell us. The norm of truthfulness about one's mental states is, she says, analogous to the convention in war of using a white flag to indicate an intention to surrender. It is impermissible to misuse this convention even to defeat an evil enemy, because to do so is to undermine a crucial means for avoiding massacre and opening a door to peace (24–25). The moral objection to lying is therefore not just that it may mislead the person to whom the lie is told. Whether it misleads anyone or not, deliberately false speech “does damage to our collective testimonial framework by giving us reasons to doubt that a person's word is reliable as such and that somber testimonial speech provides us with warrants to take what is offered as representing what is believed” (136).

This argument may seem to appeal to the effects of individual acts of lying: that they undermine the valuable norm of veracity by giving others reason to doubt that this norm is generally accepted. Some things that Shiffrin says, such as the passage just quoted, may support this interpretation, but she also puts her argument in more Kantian terms. She says, for example that the liar “wrongs humanity by acting on a maxim that could not serve as a public principle of action, because it would undermine the rational reliability of warrants altogether and with it, our ability to pursue our joint moral ends” (24). This version of the argument appeals not to the consequences of an individual action but to what things would be like if the maxim of that action were generally regarded as one that it is permissible to act on.

The objection based on the effects of a particular lie has force in some cases. It is wrong to behave in ways that undermine important social institutions. But the force of this objection depends on how serious these effects are likely to be. The objection is strongest in the case of public figures, whose lies are likely to be widely known and whose example is more likely to influence the opinions of others about the permissibility of violating the norms in question. Effects of this kind can make a lie wrong whether or not anyone is actually deceived by it. But such effects are not the only moral objection to lying. A lie is not permissible just because the person telling it is known to be so idiosyncratic and iconoclastic that his attitude toward social norms will not be taken by anyone as indicating that these norms are not taken seriously. So, although objections to lies based on their likely impact on social norms can be weighty, they cannot be the whole story.

The Kantian version of Shiffrin's argument might provide the needed complement here. That argument depends on the claim that if it were generally regarded as permissible for people to misrepresent their mental states, then we would lack reliable access to the truth about these states. The question I want to raise is whether this is true.

Do we generally have reason to believe what people tell us about their beliefs and feelings only because we take it to be generally accepted that it is impermissible to misrepresent what one believes and feels? I doubt that this is so. It does not seem to me that my belief that there is such a norm is my only or my main basis for believing what the person next to me on the airplane tells me when he insists on unburdening himself about his children or about what he thinks about Hillary Clinton or Donald Trump. Nor is such a norm my basis for believing my friend and colleague when he reveals his feelings about our department or his worries about dying. In neither situation is it the case that in the absence of a norm of veracity I would have no reason, or even much less reason, to believe what the stranger or my friend tells me about his beliefs and feelings. In most cases I have ample reason to believe what people tell me about such things because I have good reason to believe that they want me to know what they believe.

There are of course cases in which this is not true. We sometimes have good reason to think that our interlocutor has or may have strong reason to misrepresent what he or she believes. In such cases our grounds for believing what we are told can depend on our confidence that the person speaking recognizes and takes seriously a norm requiring one to tell the truth in situations of the kind in question.

The case of letters of recommendation for students who are on the job market, which Shiffrin discusses later in the book (190–191), is a good example. Letter writers want their students to get jobs and hence have strong reasons to make them appear to be strong candidates. In the absence of an accepted norm of veracity, this could lead to an arms race of escalating exaggeration, which undermines the usefulness of this important form of communication. Puffed-up letters that exaggerate the writers’ opinions of a candidate's merits can thus be wrong for at least three reasons. They may be wrong because they deceive their readers. Whether they are believed or not, they can be wrong because they have the effect of degrading the communicative environment by, among other things, encouraging others to believe that they cannot be frank about the merits of their students without undermining their chances of getting a job. Third, inflated letters violate a norm that, as Shiffrin says, is needed to preserve the usefulness of this form of communication. (I argue below for a fourth reason.)

As said above, however, most cases in which a person tells us what she believes or feels are not like this. This marks an important difference between truth telling and promise keeping. Promises have a role in exactly those cases in which we think that a person has good reason not to do the thing we are counting on him or her to do or may have such reasons in the future. A promise is intended to rule out acting on these reasons, and the norm of promise keeping is grounded in the reasons promisees have to want to be able to rely on assurances they are given in such circumstances. But only some instances of truth telling have this structure. So I do not find the Kantian version of Shiffrin's argument a persuasive explanation of the wrongness of lying in general. I do, however, agree with Shiffrin that lies can be wrong even when they are not going to deceive anyone and even apart from the effects that they may have on a social norm of veracity. Here is a sketch of an alternative argument for that thesis, similar to but, I believe, distinct from the one she offers.

We all have strong reason to want reliable information about many matters, including, for some of the reasons Shiffrin cites, the beliefs and feelings of others. We thus have good reason to object to a principle that would permit others to lie to us about such things, unless they have some particularly good reason for doing so. This seems to me sufficient to explain why it is wrong to lie to someone, without some very good reason, about matters that that person has good reason to want know about. The qualifications in this rough statement of a principle provide room for exceptions. They explain why it is not wrong to lie to the murderer at the door who, as Shiffrin says, has no good reason to have information that he needs only for a plan that he should not carry out. They also explain why it need not be wrong to lie in order to deflect intrusive questions about one's private life, which one has good reason to want not to reveal and which the other person has no sufficiently good reason to want to know.

Like Shiffrin's argument, this argument is based on our need for certain information. It is therefore broader than the prohibition against deception, since people can violate this principle by lying about their beliefs even if what they say does not deceive anyone, since no one is convinced by it. Because it appeals to our reasons for objecting to being told particular lies, this argument does not depend on the effects that lies may have on the general norm of truth telling, nor does it depend on a claim that without a general norm against lying, we would generally lack reason to believe what we are told. For these reasons, this argument seems to me preferable to Shiffrin's argument, interpreted in either of the ways suggested above. But I would be happy to learn that this is the argument she in fact had in mind.

Shiffrin sees a deep connection between the ethics of lying and freedom of thought and speech. They are both, in her view, founded on the same interests that we have, as rational and moral agents, in communication with others. The interests she identifies as underlying freedom of speech can be summarized as follows:Footnote ³

1. a. Developing a capacity for practical and theoretical thought: Each thinker has a fundamental interest in developing her mental capacities to be receptive of, appreciative of, and responsive to reasons and facts in practical and theoretical thought.

2. b. Apprehending the truth: Each thinker has a fundamental interest in believing and understanding true things about herself, including the contents of her mind, and the features and forces of the environment from which she emerges and in which she interacts.

3. c. Exercising the imagination: In addition, each thinker has a fundamental interest in understanding and intellectually exploring nonexistent possible and impossible environments. [This enables them better to conceive the future and to think of what could have been the case in the present and past.]

4. d. Moral agency: Each thinker has interests in acquiring the relevant knowledge base and character traits as well as forming the relevant thoughts and intentions to comply with the requirements of morality. [This may already be contained in a to c.]

5. e. Becoming a distinctive individual: Each thinker has interests in developing a personality and engaging more broadly in a mental life that, while responsive to reasons and facts, is distinguished from others’ personalities by individuating features, emotions, reactions, traits, thoughts, and experiences that contribute to a distinctive perspective that embodies and represents each individual's separateness as a person.

6. f. Responding authentically: Each thinker has interests in pursuing a to e through processes that represent free and authentic forms of internal creation and recognition. [This includes an interest in forming thoughts, beliefs, and other mental contents in ways that are independent of distorting influences and interests in being able to share these contents with others at their discretion.]

7. g. Living among others: Each thinker has a fundamental interest in living among other social, autonomous agents who have the opportunities to develop their capacities in like ways. [This is not only because of the value of such relationships in themselves but also because they are crucial for the forms of development mentioned above in a to e.]

8. h. Appropriate recognition and treatment: Each thinker has fundamental interests in being recognized by other agents for the person she is and having others treat her morally well.

These interests support freedom of speech because they provide strong reason to object to governmental actions and policies that interfere with the free development and operations of people's minds or are based on rationales that are incompatible with valuing this protection (such as “adopting a restrictive litter policy on the grounds that it may have the effect of discouraging protests” [92]).

Shiffrin contrasts her thinker-based approach to freedom of speech with accounts of other familiar types, such as accounts based on the interests of speakers in having access to audiences in order to express their views, accounts based on the interests of audiences in having access to the views of others in order to make up their own minds what to think, and democracy-based accounts, which recognize both citizens’ interests in having access to information in order decide how to vote and their interests in having opportunities to influence others, and thus apply only to a particular range of issues concerned with politics and public policy.

It is an important strength of Shiffrin's approach that in contrast to democracy-based theories, it applies to the widest rage of forms of expression, including not only discursive (written or spoken) expression on any topic we have reason to be concerned with but also nondiscursive expression, including art, music, dance, and so on (81).

Shiffrin's thinker-based approach may seem to be based only on the interests we have as audiences in having access to facts and being exposed to the opinions of others. But she argues interestingly and for the most part convincingly that this is not the case (89–91). It is, she says, essential to our development as thinkers that we be able to “externalize” our thoughts by expressing them to others and that we be able to interact with others about the content of these thoughts. Shiffrin illustrates this need by connecting the effects of censorship with the ills suffered by prisoners who are subjected to extended periods of solitary confinement. They “suffer terrible depression, despair, and anxiety; moreover, their emotional control and stability wanes, and their abilities to interact with others atroph[y]” (91). Thus, she says, “to forbid or substantially restrict free expression is not tantamount to solitary incarceration but lies on a spectrum with it: it is to institute a sort of solitary confinement outside of prison but within one's mind.”

Looked at in one way, this seems an exaggeration. As Shiffrin says, those subjected to solitary incarceration can suffer impairment of their rational capacities. Those subjected to severe censorship, such as dissidents in the Soviet Union who were not imprisoned, need not, or so it seems to me. The serious objections to the restrictions they were subjected to lie instead in interference with their more specific interests in being able to express and discuss with others their views on particular subjects important to them, including art and religion as well as politics. These interests are, however, thinker based. So Shiffrin's observation seems plausible if the “spectrum” is just the range of cumulative strength of various different thinker-based reasons.

Shiffrin says that her thinker-based approach provides “a more comprehensive, unified foundation for freedom of speech protection” than alternatives based on audience or speaker interests alone (85). The claim to greater unity might be questioned, given the diversity of the thinker-based interests I list above and given the fact that as Shiffrin acknowledges, free-speech protection may serve other interests as well. But the degree of unity is not the central consideration here. What is most valuable about Shiffrin's approach is the distinctive shape that her interpretation gives to the values it identifies as at stake in freedom of speech and thought, and the connections that this interpretation builds with other rights.

One example of this strength already mentioned above is the breadth of Shiffrin's account of freedom of speech, applying as it does to discussions of aesthetics as well as politics and to nondiscursive expression as well as to speech and writing. Whether the importance of these different activities is unified in the value of moral and rational agency or based instead on the independent reasons we have for wanting to be able to engage in each of them and to have others do so as well, it is an important strength of her account that it recognizes all of these considerations as relevant to freedom of speech.

A second significant feature of Shiffrin's emphasis on our interests as thinkers is its implications for the relations between freedom of speech and other rights, such as freedom of association. According to a view based on the interests of speakers, the ability to associate with others is important as a means for making our views better known. By organizing in political parties, for example, we can provide a megaphone for our views. Similarly, according to a view based on the interests of audiences, advocacy groups can be important because they bring a wider range of views to our attention. Shiffrin's account, because of its emphasis on our interests, as thinkers, in developing our own ideas, brings out the fact that participating in such associations is a valuable part of the process of formulating one's own thoughts.

A third important aspect of Shiffrin's approach is its implication for the difference between expression by individuals (and associations of individuals of the kinds just mentioned) and expression by for-profit corporations. Corporations do not have thoughts that they have an interest in developing and communicating, and for-profit corporations are not important means for the development and expression of individuals’ thoughts. So the primary values underlying freedom of speech are not at issue in the legal regulation of expression by for-profit corporations (although audience interests in access to accurate information may play a role).

Shiffrin's account also has distinctive implications for the free-speech rights of children and the mentally disabled (104–105). In contrast to views that make some level of competence an eligibility condition for these rights, Shiffrin's view recognizes that children and those suffering from mental illness or disability have many of the thinker-based interests she lists. What is different about these cases lies in the fact that what is required to respond adequately to these interests may vary, depending on the type of individual in question. In the case of children, for example, their interest in developing their rational capacities justifies compulsory education of a sort that would be impermissible for adults. And shielding children from some forms of expression may be justified if exposure to such forms of expression would impair their development in sufficiently important ways.

Elements of accepted freedom-of-speech doctrine, such as the prohibition of state censorship, are crucial means for the protection of the interests that Shiffrin identifies. But adequate protection of these interests requires much more, including education, access to information, and many other conditions. What is needed to serve these interests thus goes well beyond standard free-speech doctrine, which is, as she says, “predominantly negative” in its provisions and “encompasses only state action” (106). One might ask, therefore, whether what she provides is the basis of an account of freedom of speech or of something much broader and more demanding.

This is not, however, an objection to Shiffrin's view, for several reasons worth noting. First, this greater reach is not a feature peculiar to Shiffrin's account. On any plausible account, the interests and values underlying free-speech rights will have further implications as well. Even a narrow democracy-based account would have such implications. In order to decide how to vote, we need access to facts about what our government is doing, of the kind provided for by the Freedom of Information Act. But it is not an objection to this account of the interests underlying the right to freedom of speech that this implication goes beyond what that right itself (at least as interpreted in current constitutional jurisprudence) requires.

The limits of the right to freedom of speech as a constitutionally recognized requirement can be explained in terms of the limited institutional competence of courts as means of securing rights and also perhaps in terms of the history of particular institutions. But of course the limits of current doctrine need not be accepted. It is a virtue in an account of the interests underlying free-speech rights that it provides a basis for pushing these rights beyond their current understanding. Perhaps something like the Freedom of Information Act should be recognized as required by the First Amendment. Perhaps a way should be found to make that Amendment apply to suitably defined forms of private censorship as well as to state censorship.

It is a very good thing—and a testimony to the importance of the values protected by the right to freedom of speech—that this right has found recognition in the constitutional jurisprudence of so many countries. But we should not let our thinking about freedom of speech be circumscribed by the current limits of this jurisprudence. It is one of the many virtues of Shiffrin's rich book that it presents the values of freedom of speech and freedom of thought in an expansive way that pushes these boundaries.