II. The Autonomy Argument for Speech and Expression

Free speech is typically taken to cover “the spoken word, but also the written word as well as conduct conveying a message and expression through symbols, demonstrations, and so on.”Footnote ⁶ There are almost no true free speech absolutists. Every theorist who characterizes herself as an absolutist removes specific kinds of speech or expression from protection. “Absolutism,” in this case, means “without exception given a list of exceptions” already built in to the definition of free speech some “absolutist” is advancing. I will be arguing that we should view privacy as one of these exceptions. Few would argue that instances of workplace quid pro quo sexual harassment, blackmail, slander/libel, or conspiracy to commit a crime should be protected speech or expression. Swinging a baseball bat at someone, no matter how expressive, is not something countenanced on free speech grounds. Sometimes regulations such as time and place restrictions will suffice.Footnote ⁷ Prohibiting someone from expressing themselves over a loud speaker at 2 a.m. in a residential neighborhood seems perfectly reasonable. Restrictions on nude dancing in a public park seem likewise agreeable, especially when noting that the ideas expressed could be voiced at another time or place.Footnote ⁸ Sometimes prohibiting speech based on the content of the expression is justifiable. Expressions that describe how to mass produce aerosolized smallpox is an example of expression that we may forbid.Footnote ⁹ While contentious and difficult to define, many free speech champions would agree that “sexual harassing speech” is justifiably prohibited.Footnote ¹⁰ Additionally, setting aside the exception of fair use, speech that violates intellectual property rights may be justifiably suppressed.Footnote ¹¹

Nevertheless, we are told, freedom of speech is the lifeblood of democracy. The famous First Amendment scholar, Alexander Meiklejon, noted “self-government can exist only insofar as the voters acquire the intelligence, integrity, sensitivity, and generous devotion to the general welfare that, in theory, casting a ballot is assumed to express … [citizens] must understand the issues which … face the nation, … pass judgment upon the decisions our agents make upon those issues, … [and] we must share in devising methods …” to solve these problems.Footnote ¹² Other free speech scholars offer different considerations. Consequentialists such as John Stuart Mill argue that free speech is essential for truth discovery, provides a check on the power of government, and promotes the virtue of toleration.Footnote ¹³ Deontological scholars such as Thomas Scanlon maintain that speech and expression are essential for individual autonomy, which in turn is necessary for the moral bindingness of contracts, including the social contract.Footnote ¹⁴

Wrapped together, these considerations fall nicely under Meiklejohn’s idea of self-government. To be a self-governor, individuals must have developed the required capacities to be rational project pursuers, capable of logical thought, planned actions, and considered judgments.Footnote ¹⁵ At its core, this aspect of self-government concerns the foundations of moral agency. Aside from the autonomy-building aspects of free speech and expression, there are also autonomy-protecting considerations. In order to make informed decisions and judgments about politicians, issues, and policy, individuals must be able to access and process the information needed. “Consent of the governed” must, in some sense, mean “informed consent.” If citizens are the source and ultimate arbiters of governmental power, then speech, expression, and access to the information necessary for self-government is important. First Amendment scholar Martin Redish largely agrees with Meiklejohn but substitutes the phrase “self-realization” for “self-government.”Footnote ¹⁶ Self-realization, according to Redish, includes the First Amendment values of truth seeking and power checking, as these are necessary for the development of an individual’s capacities, powers, and rational project pursuit. Additionally while the connections are less clear, autonomy arguments for free speech also include artistic expressions, such as works of fiction or paintings and sculptures, as important for self-realization and self-government.Footnote ¹⁷

III. Against the Autonomy Argument in Favor of Speech

But as noted in the opening, access to content is not a problem. There is more content available today than at any time in the past.Footnote ¹⁸ Obtaining a platform for expression is also not a problem. No longer wedded to the printed page or a narrow bandwidth of radio or television frequencies, individuals have a vast array of options for broadcasting their own expressions. Being heard beyond a fairly narrow group of family, friends, and acquaintances may be a worry, especially since the amount of content being produced and made available is staggering. But on the other hand, being heard by a larger group of people was not an issue in the 1950s, 1850s, or before. The autonomy-producing aspects of free speech do not seem to rely on reaching a mass audience unless we are considering a free press and government oversight.

Much has changed about content production and access over the past several decades. Speech, expression, and content used to be costly to produce and distribute. Those in production and distribution industries had to be fairly sure that what they were producing was worthy of the costs. Failure meant lost revenue, market share, and jobs. Quality controls and market analysis, including sales projections, served as gatekeeping mechanisms. Generally, content of all sorts was created by professionals. These individuals made a living, or attempted to make a living, by honing their craft and creating works that had commercial and artistic value.Footnote ¹⁹ These considerations, along with the fierce competition to secure production and distribution contracts, tended to ensure that the expressions created had some connection to quality.

Increasingly, anyone can instantiate and distribute virtually any kind of content. To instantiate an idea, set of ideas, feeling, theory, and so on, is to produce a physical copy—extended in space and time with a mass—external to the human mind. For example, without taking a stand on the metaphysics of ideas, when a recipe is instantiated it could be written down on paper or chiseled into stone. Managers, distributors, and other gatekeepers of quality are no longer needed, and in many areas of content production and distribution, they are irrelevant. As an example, consider the explosion of academic predatory publishing practices and “pay-to-play” journals and conferences.Footnote ²⁰ Even in a domain focused on quality of research, a domain dedicated to the quest for truth and the advancement of knowledge, we have increasing amounts of what could be called information pollution. A similar point can be made in various areas of news production.Footnote ²¹

While most kinds of speech and expression could serve self-government, either autonomy-building or autonomy-protecting, it is also true that these expressions are almost entirely fungible or replaceable.Footnote ²² There are numerous instantiations of most content forms that could be traded for a different instantiation or different content altogether without loss. Consider, for example, all the different cartoon shows available where one series could be substituted for another with no loss in self-government. Here I am not imagining that the creators of cartoons were told they could no longer produce a specific series, as that would impact self-government. Rather, I am imagining a counterfactual world where these creators had simply created, of their own volition, a different series of content. No one’s life would be obviously disadvantaged if The Roadrunner show had never occurred. Our entertainment time could have easily been filled up with Bugs Bunny cartoons, comic books, or other activities.

The claim being advanced is that almost all physical instantiations of ideas are fungible. Aside from the rare physical item like the Mona Lisa, the ideas, emotions, and expressions could be instantiated in a different way. Sometimes even a mere copy of an original instantiation will suffice. The replaceability of specific copies with different ways of expressing what was originally physically codified is also true of entire classes of content. Just imagine if motion picture technology had never been invented or if no one ever created progressive rock music. So, we would read more books, attend more plays, and immerse ourselves in jazz music. Yes, progressive rock is expressive, but this entire category of content could be replaced by expressions that equally serve autonomy and self-government. To put the point another way, there are vast areas of expression that you and I don’t consume. Suppose all you listen to is classical music and I don’t own a television or attend movies. Even without accessing this content our lives could go perfectly fine in terms of self-government. So it would seem that almost all physical instantiations of various ideas and emotions are fungible with respect to self-government.

This is not true for certain categories of ideas. For example, while there might be many different ways to express or convey the germ theory of disease, there is no equal substitute for this theory or set of ideas. It is not fungible. It is not as if human life would have been, more or less, unaffected had this theory never been discovered or distributed. Or consider the various ways someone could instantiate how antibiotics work within the human body. There could be textbook-written explanations, plays, musicals, and even pantomimes that instantiate these ideas. While each instantiation is fungible, the set of ideas is not. Currently there is no equally efficient or effective alternative to using antibiotics.

Fungible expressions of non-fungible ideas, theories, or feelings are generally necessary for self-government. This latter class includes scientific discoveries, philosophical theories, math, logic, and the like. For example, there are lots of ways to present Locke’s theory of consent as a legitimate source of governmental power, but this theory is not fungible. It is not like we could substitute a consent theory of governmental legitimacy with a theory of “might makes right” without loss to self-government. At one extreme we have political, philosophical, and social arguments or information clearly important for self-government while at the other we have communication with little or no value related to self-government. Also note that this argument against autonomy-based justifications for speech is not dependent on there being clear and distinct categories of fungible and non-fungible. Canonical examples in each category may give way to disputed borderline cases.

To determine the worth or value of an expression or a category of expression we ask what is lost in terms of self-government or autonomy if the expression or category did not exist. Again, if progressive rock did not exist we would be just fine. This is not true when we consider numerous areas of scientific endeavor that directly and positively impact human health and well-being. For example, in Barns v. Glen Theater, Inc. a U.S. court noted that while nude dancing is expressive, it is “marginally and on the outer perimeters” of First Amendment protection.Footnote ²³ On my view, the distinction between high-value, low-value, and no-value expression is similar to information that is important, less relevant, or irrelevant to democracy and self-government. Non-fungible ideas that promote or maintain human health and well-being would fall into high-value speech. Low-value and no-value speech would be fungible content that is irrelevant to the maintenance or promotion of human health and well-being. While entertainment, in general, might be important for health and well-being, any specific instantiation or area of entertainment could be replaced without loss.Footnote ²⁴

We must also be on guard to not fall prey to packing an otherwise low-value expression or category with important information. Imagine a cartoon that includes a joke about how finding a workable form of communism is like attempting to square the circle. The practicality of communism is an important consideration and one that is relevant to self-government. This content does not, however, have to be instantiated in a cartoon. Similarly, the expressive content found in swinging a baseball bat, no matter how profound, does not need to be instantiated in a swing toward someone’s head.

So, in general, it is simply an overstatement to claim that free speech and expression is necessary for autonomy and self-government. Vast areas of content are either fungible or totally irrelevant in service of this goal. As I will argue in the next section, most private information will fall into this latter category. Additionally, the information necessary for individual autonomy building and autonomy maintenance could be instantiated in ways that do not impact individual privacy rights. For example, pictures may be modified so that specific individuals cannot be identified.

A critic might counter with the view that there are instances of speech that should be protected even though the content is clearly fungible—for example, passing out Bibles on the street or publishing cartoons of a deity.Footnote ²⁵ First note that the fungibility argument being offered does not claim that those who wish to pass out bibles on the street or publish cartoons of a deity should be prohibited, unless these expressions violate privacy, property, or some other right. If the individual passing out bibles does not own the copyright and is distributing pirate copies, then this action can be prohibited. Artists who create depictions of a deity don’t necessarily violate any rights, and thus such expressions, while fungible, do reflect the autonomy of the artist.Footnote ²⁶ Second, these cases strengthen, rather than undermine, my point about fungibility and how autonomy arguments fail to provide overly strong support for free speech. If there were no individuals willing to pass out bibles on the street, we would still have access to religious ideas. Thus, our autonomy cannot be undermined by this particular lack of access.

Free speech champions might also counter with the view that by insisting on robust protections for all speech and expression that falls within the “protected” class, we guarantee individual autonomy and self-government. On this view, broad protection for free speech is sufficient for self-government. But imagine a society with numerous media sources and no government restrictions on speech. Each source, however, decides to publish “fluff” stories and avoid political or philosophical issues. In this case we could have total freedom of speech and shoddy democratic institutions built upon false information or no information. Imagine a society without serious books, plays, music, a society dominated by the drama of social media flame wars, or a society awash in information pollution. Individuals who inhabit these societies may have lots of content to consume, but little real autonomy. Robust freedom of speech is not sufficient to guarantee a robust democracy or individual autonomy.

Near absolute freedom of speech—speech that trumps privacy considerations, for example—is also not necessary for self-government. Consider our counterparts in the EU who are constrained by privacy laws and the right to be forgotten. The EU’s General Data Protection Regulation (GDPR) requires consent, breach notification, information access, data erasure, portability, and privacy by design. Data subjects must explicitly consent to the use and purpose of how their information will be processed. Data subjects must be notified when a breach, or unauthorized access, has occurred. Data subjects have the right to know if personal information about them is being held, used, or processed by data controllers. Individuals, with various exceptions, have the right to be forgotten. Data subjects can have “… the data controller erase his/her personal data, cease further dissemination of the data, and potentially have third parties halt processing of the data.”Footnote ²⁷ It would be difficult to demonstrate that the individuals living in the EU are impoverished in terms of autonomy and self-government by comparison to a superior American standard. Thus, the most a defender of speech can advance is that some areas of speech, expression, or content are important for autonomy and self-government while the vast majority of communication is largely irrelevant and replaceable.

IV. Privacy, Autonomy, and Contracts

While privacy has been defined in many ways over the last century, I favor what has been called a “control”-based definition of privacy. A right to privacy is a right to control access to, and uses of, places, bodies, and personal information.Footnote ²⁸ A serviceable definition of “personal information” is “any information relating to an identified or identifiable natural person … one who can be identified, directly or indirectly, in particular by reference to an identification number or to one or more factors specific to his physical, physiological, mental, economic, cultural or social identity.”Footnote ²⁹

Numerous privacy scholars have argued that there are close connections between privacy rights and human health and well-being.Footnote ³⁰ Household overcrowding studies and prison overcrowding research demonstrate that limiting an individual’s ability to withdraw causes an increase in aggression, depression, stress levels, recidivism, and all while suppressing the immune system.Footnote ³¹ Monitoring your children too much increases the probability of early drug use, dropping out of school, teen-pregnancy, and violence.Footnote ³² Businesses that use various covert and overt surveillance techniques tend to have higher employee turnover. Employees view various forms of electronic monitoring as harmful intrusions of privacy, and this perception increases aggression, destructive behavior, and employee turnover.Footnote ³³ Workplace drug testing deters highly qualified workers from applying, has a negative impact on workplace morale, and has been indicated in reduced productivity.Footnote ³⁴ While the rituals of coming together and leave-taking or asserting privacy are largely culturally determined, it would seem the need is not.

Additionally, many privacy theorists argue that privacy is associated, in some central way, with autonomy and self-government.Footnote ³⁵ To be sure, dealing with depression, aggression, and different sorts of risky behavior, has a negative impact on development and outlook. Or consider, for example, information-sharing practices that allow for the creation and maintenance of stable social relationships. What is shared with a lover is not shared with classmates or one’s children. These different relationships depend on norms of informational and physical access, and thus directly affect autonomy, self-creation, and identity. It is within these acts of control and maintenance that individuality forms. Moreover, being watched changes a given activity or behavior from one of private discovery, creation, or enjoyment and pushes it toward performance.Footnote ³⁶ Engaging in self-reflection and criticism, which is not typically conducted while others are watching, allows for greater self-knowledge.Footnote ³⁷ Thus, privacy also provides a check on political power and social pressure. Whereas free speech may provide sunlight on a politician’s behavior in the service of accountability and oversight, privacy rights mark out zones where access is left to the individual.

Privacy affords individuals the moral space to engage in autonomy-building practices. “Both animals and humans require, at critical stages of life, specific amounts of space in order to act out the dialogues that lead to the consummation of most of the important acts of life.”Footnote ³⁸

Infants are without privacy. As infants grow into toddlers and begin to communicate with language, they express wishes for separation at times. This process continues as children grow into adults.Footnote ³⁹ Toddlers and small children begin requesting privacy as they start the process of self-initiated development. More robust patterns of disassociation continue as children enter puberty. There is now fairly compelling evidence linking unstructured and unsupervised free time for kids with positive health outcomes.Footnote ⁴⁰ Finally, as more-fully formed young adults emerge, the walls of privacy have hardened and access points are maintained vigorously.

Critics have challenged the connection between privacy and autonomy, noting that violating someone’s privacy does not necessarily undermine autonomy.Footnote ⁴¹ Consider a case of covert surveillance (the covert peeping Tom) where the individual being watched will never come to know of the surveillance. First, note that the presumed disconnect between privacy and autonomy does not seemingly hold true of kids. Covert surveillance is perceived as parental disinterest and is associated with various health problems already mentioned. More generally, even in a case of covert monitoring, it would be hard to maintain that the surveillance target is leading a fully autonomous and self-directed life. For example, the target would have the false belief that she was not under surveillance—a fact completely controlled by the whims of the watcher. Moreover, as defined above, privacy is the right to control access to and uses of places, locations, and personal information. To the extent that our rights, like property and liberty rights, promote autonomy, any violation of these rights would undermine the target’s autonomous choices. Finally, there is no need to claim that privacy is necessary for autonomy in every instance. As with free speech, we could simply highlight how privacy is important for, and contributes to, individual autonomy.

In the remainder of this section I will argue that regardless of a commitment to strong rights to speech and expression, informational privacy rights should trump speech rights in a range of cases. Supporting the contract arguments that follow are the connections between privacy, autonomy, and flourishing, along with the assumption that individuals have a general right to bind themselves with agreements and contracts. The argument offered is not a hypothetical contract argument where we should treat personal information as property because, hypothetically, we would do so in a range of cases. Rather, what the argument shows is that it is possible to construct cases where privacy and contracts justifiably limit speech.Footnote ⁴² These actual contracts, like doctor/patient information sharing agreements, ground further restrictions that even free speech maximalists should countenance.

As noted in the opening, individual property rights may justifiably limit freedom of speech. Assume that I own a can of paint and a garage wall. Your right to free speech does not trump my rights to control the can of paint or the garage wall. If you were to seize the paint and start painting your original expression on my garage wall, you could be justifiably stopped and the expression removed. You may even be liable for damages. Similarly, your rights to speech do not trump my copyrights to a poem. Imagine I own a poem and you decide to paint my poem on your wall with your paint.Footnote ⁴³ Again, this activity could be justifiably stopped using appropriate levels of force if necessary.

Consider an even more extreme case. Suppose you wake up one morning to find yourself strapped down with me readying a tattoo gun. Struggling against your restraints you ask what is going on and I proclaim “I am exercising my rights to free speech. I am going to tattoo ‘Adam is God’ on your forehead.” Again, I think it is fairly clear that my rights to free expression do not trump your self-ownership or body rights. But rather than a case of compelled instantiation, where I force you to be a vehicle for my expression, imagine we contracted for this service. Suppose you freely agree that in exchange for allowing me to tattoo an expression on your forehead and your assurances that the expression would not be removed or altered, I would pay you one million dollars. With no other mitigating relevant factors, such as your being under duress, it would seem that the resulting contract would be morally and legally binding.

A sci-fi twist to this case reveals an important feature. Suppose that Smith has telepathically and temporarily seized control of your body. You are a helpless spectator as Smith controls his body and your body as well. Smith thinks the tattoo proposal for you is a great idea. Even better is when Smith has your body transfer the one million dollars to his bank account. While there might be numerous ethical aspects to this case, the one I am interested in has to do with the moral bindingness of the contract between you, as you are controlled by Smith, and me regarding the tattoo and the million dollars. Given that you did not agree to the terms of the contract—it was actually Smith—you are not morally bound to its terms.

Morally binding contracts presuppose prior entitlements or at least legitimate possession joined with rights to determine downstream uses. The difference between these cases, the difference that makes a difference in the moral bindingness of the relevant contracts, is that in the former case both parties had legitimate entitlements to the items in question. Assuming that individuals have a general right to make contracts and that contracts in certain conditions are morally binding, we may arrive at the view that in the first case your uncoerced agreement regarding an item you own or are entitled to control generates a morally binding contract. This is not true of the second case. One might be drawn to the view that mere legitimate possession could ground justified contracts or agreements. But, if I let you borrow my car you may legitimately possess it and yet lack the relevant rights to bind others with contracts that determine the downstream uses of the car. Similarly, if you legitimately possess something that is unowned—the ground you are standing on, suppose—it is not at all clear that your mere possession could ground contracts over downstream uses of the land you legitimately possess.

If private information is owned like your leg or a poem, then it would be possible to contract regarding access to, and downstream uses of, this information.Footnote ⁴⁴ Sometimes we allow access and limited use of our own personal information, but retain various rights. Other times, we transfer all our rights to control our own personal information to other individuals, groups, or even corporations. If private information is not owned, but rather legitimately possessed, including the relevant rights of future control, then it would also be possible to contract over access to, and downstream uses of, this information. These contracts could provide justified limits on freedom of speech on both private and public actors.

Elsewhere I have argued that informational privacy is a kind of intangible property right where allowing access does not entail the forfeiture or waiving of rights to control downstream uses of the information in question.Footnote ⁴⁵ Rather than repeating these arguments, I will consider a different justification. Imagine, once again, a world where individuals use anti-monitoring suits and completely sever the links between their real identity, both physical and informational, and their public-facing avatar. When I walk into a room, you don’t see me; you see a projection that I want you to see. You don’t hear my real voice, you hear an altered voice. You can’t tell my age, race, sex, gender, height, weight, or any personal information. All internet browsing patterns, phone GPS information, video surveillance records, and so on, are also de-linked from my real self. In fact, suppose that I could obfuscate all of this information with ever-changing avatars along with phone apps and browsers that broadcast misinformation. Perhaps upon giving you my de-encryption code you could see the real me while others nearby would see my avatar. Such secrecy may be odd, but it does not appear to be immoral. I could just as easily remain hidden on my thousand-acre estate, reject using any technology that could collect personal information, pay with cash, produce my own food, and limit interaction with outsiders to protect privacy. As long as I pay my taxes and obey the law, I would owe society no further access.

Now suppose that you have heard that someone with one of my avatars has had an interesting life and you would like to write a story. I agree to share some exploits but only on the conditions that you do not reveal my true identity and that I am paid a small fee. Being cautious, suppose I even insist on seeding my story with some irrelevant misinformation to defeat any re-identification algorithms that may be used to discover who I am. I sign, you sign, and it would seem that we have a binding contract. While there are many factors needed to generate contracts that are morally binding, including that no one is forced to participate, one important factor is that I own or have the relevant rights to the information I am sharing with you. The information is about me and it is not fictional or taken from someone else. Checking into some of the details I share, you conclude that there is, in fact, good reason to believe that I have shared authentic personal information. In this case, if you were to violate the terms of our agreement, use re-identification algorithms to determine my real identity, and publish my name and address along with my stories, you would have done something immoral and perhaps illegal.Footnote ⁴⁶

Imagine that the information was truthful but not about me. Suppose after weeks of recording my stories and taking notes you come to realize that I am a fraud.Footnote ⁴⁷ The stories I am recounting actually come from J. S. Mill’s autobiography. The moral bindingness of our contract would seem to be undermined and the fee need not be paid. Rather than being about Mill, imagine that the stories I recount were total fiction. Looking closely at the details you deduce that I am not just taking liberties with the story line, but that the events described could have never happened. Again, it seems that the moral bindingness of our contract will have been undermined. The contract, it would seem, would have explicitly stated or implicitly assumed that the stories shared were not fictional or about someone else.

The argument structure I have pursued is a kind of back-door argument for viewing informational privacy as a kind of property right. We start with a case where our moral intuitions lean the same way. For example, Fred shares his authentic medical history with Ginger, and she agrees to keep everything about the interaction private. What grounds the moral bindingness of this contract is that the bargaining situation is fair, there are no threats, coercive offers, fraud, and the like, and Fred owns or has the relevant rights over the information in question. Fred has the relevant rights that are needed for binding contracts over his personal information due to the connections between privacy, autonomy, and well-being. Note that in the typical case, the connections between privacy, autonomy, and well-being do not hold of Ginger and her control of Fred’s information.

Thus, in the typical case, when someone contracts to limit access to and uses of their own personal information, and when a second party agrees to various access and use limitations, we are drawn to conclude that a morally binding contract is created. One reason we might conclude this—perhaps an inference to the best explanation—is that both parties have the general right to make contracts, to bind themselves morally, and at least one party has the relevant moral entitlements over the object of negotiation needed to ground the contract. If so, then individuals negotiating over access to and uses of their personal information may ground a binding contract that limits downstream uses of this information. These limits could include limits on freedom of speech and expression. Thus, the law of confidentiality found in EU conceptions of privacy along with doctor/patient, lawyer/client, clergy/parishioner agreements are seemingly groundless without the assumption of prior entitlements.Footnote ⁴⁸

Establishing that individuals own or have the needed rights over their own personal information and can legitimately contract over uses of this information, grounded in autonomy and self-government, is not to establish that individuals have exclusive possession claims. Unlike the case of someone trying to use your body, where there will be interference with your use in the vast majority of cases, this is not true of personal information. Information is intangible and can be instantiated in different physical substrates at the same time. Thus, Crusoe’s use of Friday’s personal information does not necessarily interfere with Friday’s use. Imagine that Crusoe knows Friday’s personal stories because he found Friday’s diary inadvertently left on a public bench. Nevertheless, as already noted, mere possession or access does not automatically grant future use and control rights. Allowing a doctor to examine your knee is not also to waive all access and use rights over your knee. If so, invitations to dinner or offers to have sex would be weighty decisions indeed. Allowing a friend to read your diary would be a momentous decision. Moreover, given that access to personal information is largely dependent on technology, the position that access entails rights to control, use, and broadcast personal information about others, is dubious. Advancements in technology could inexorably lead to the content maximalist case envisioned in the opening paragraph. But just because all zones of privacy may vanish and technology may provide access to our most sensitive thoughts, emotions, and behaviors, it does not follow that such access is justified or that sharing this information is automatically permitted.

There is also an autonomy related difference that occurs and this may provide a compelling reason for why mere legitimate possession of information is not enough to justify downstream uses of and control over personal information, bodies, and locations. Depending on who takes notice and the medium used, Friday may be thrust into the public sphere against his wishes. Prior to the access and dissemination, Friday had a certain level of control over his reputation and public standing. For example, when Friday publishes an article on why forced taxation is immoral or why “safe spaces” on campus are worrisome, he is taking a public stand and is, presumably, aware of how this might affect his future opportunities. It is Friday who bears the costs and benefits of his choice. When Crusoe, on the other hand, obtains access to Friday’s private views about taxation or safe spaces and then publishes these views, it is Friday, not Crusoe, who will shoulder the downstream burdens or collect the benefits. This future is foisted, or perhaps forced, upon Friday, and his reputation cannot be “fixed” or replaced like other kinds of property. Arguably this foisting could be viewed as a morally relevant worsening that directly impacts Friday’s autonomy and self-government.

Cases that highlight the censorial aspects of copyright are useful when considering the issues that surround compelled speech. Like finding and publishing a private diary or cases of revenge porn, courts have noted and protected authorial autonomy and dignity. Footnote ⁴⁹ In these cases, an author keeps a diary or takes pictures of herself and in doing fixes an expression protected by copyright, which includes the rights of distribution and display.Footnote ⁵⁰ Those who copy or distribute copies of these protected works are in violation of copyright law. These sorts of cases infringe what Immanuel Kant called the right of the author, the “innate right in his own person, … to prevent another from having him speak to the public without his consent …”Footnote ⁵¹ For Kant, to publish a book is to speak with the public. The unauthorized publisher goes against the will of the author and undermines authorial autonomy and dignity.

While U.S. courts have upheld censorial copyright claims in a range of cases, a crucial element is that the dignitary interest of authorship can only be protected if the subject fixes the expression herself. For example, in the revenge porn case of Doe v. Elam, had Elam taken the pictures, Doe could not have brought a copyright infringement claim. In fact, had Elam taken the pictures, he would have held the relevant copyrights. Given that the expressions may be exactly the same in either case, whether Doe or Elam snapped the pictures, the fact of who fixed the expressions seems to be arbitrary and irrelevant when considering the dignitary interests involved. To put the point another way, if the dignitary and autonomy-based interests in these cases is strong enough to allow censorship, sometimes in the name of promoting overall speech and expression, then it would seem that privacy-based censorship would be likewise compelling.

Equally interesting are cases that promote free speech and expression by upholding anonymous publication or the use of pseudonyms. For example, in McIntyre v. Ohio (1995) the court noted “Anonymity … provides a way for a writer who may be personally unpopular to ensure that readers will not prejudge her message simply because they do not like its proponent … protect unpopular individuals from retaliation … at the hand of an intolerant society … and shield [authors] from the tyranny of the majority.”Footnote ⁵² Again suppose that via predictive analytics, internet browsing analysis, audio/video capture in public areas, facial recognition technology, and the like, we are able to capture virtually every fact about your daily life. It would seem that by capturing this information and broadcasting it we would be compelling you to speak to the public, just as if someone had published your diary against your wishes. It would also become impossible to gather privately, publish anonymously, use pseudonyms, or conduct certain sorts of research—denial of these activities would suppress speech and expression.Footnote ⁵³

V. What We Have No Business Knowing

Most privacy-intrusive information about individuals is irrelevant to speech-based autonomy and self-government and, thus, there would be no speech-based justification for overriding the property/autonomy-based contract argument presented above. Specific instantiations are fungible and generally not needed for the autonomy-building or autonomy-protecting aspects of self-government. For example, we may never know the medical or sexual histories of those living nearby and yet our lives will go perfectly well. Consider a case where your complete medical history has been made public and I possess a copy. Assume that in obtaining this copy I have violated no contracts, agreements, or rights. I, not illegitimately, possess a copy of your complete medical information. Or suppose I have the technological device mentioned earlier, one that could record everyone’s thoughts, words, and activities, and post searchable, updated, databases to the Web on a daily basis.

As with many of the cases already discussed, I think our intuitions lean the same way in this case as well. We don’t owe others access to, or control over, our medical information, sexual histories, private thoughts, and the like. Setting aside cases where the information target herself grants access and waives all downstream rights, there is something morally questionable about broadcasting this content independent of consent. First, individuals may have property rights in their personal information similar to copyrights. Second, such disclosures undermine the autonomy of the information target and are, in most cases, irrelevant to the self-government of content consumers.

A cased mentioned by Edwin Baker is instructive. “[O]ne person’s … autonomy might be enhanced by knowledge about her spouse, whom she thought loved her but who does not and who is actually having an affair …”Footnote ⁵⁴ In this case, there is a tension between autonomy and privacy—one person’s autonomy against another’s privacy. While Baker distinguishes speech claims from information access claims, he concludes that free speech includes the freedom to “expose any private information that a person knows…”Footnote ⁵⁵ If so, Baker may be driven to conclude that broadcasting medical or sexual histories, assuming they are not ill-gotten, is perfectly appropriate. I don’t think that not illegitimately possessing private information justifies downstream broadcasting any more than picking up a ball that rolls up to you makes it your ball, or being told a trade secret makes it yours to disclose. The cases already discussed support this view. Moreover, in the cheating spouse case, like examples of doctor/patient or lawyer/client nondisclosure rules, there is a prior explicit or implied contract in place governing access to and uses of private information. One would suppose that part of any spousal agreement there would be rules governing infidelity and information access.Footnote ⁵⁶ Additionally, as already noted, such contracts presuppose prior justified entitlement.

When focusing on non-governmental actors, those not employed to serve the public in some sort of official capacity, much of the information about them, and especially all of the private information, is fungible and unnecessary for autonomy and self-government. Even the most important or “real life” lessons that could be learned by consuming information about the private lives of famous actors, for example, could be instantiated and consumed without violating privacy rights.

Similarly, those who work in the public sector may be thought to agree to specific domains of privacy and accountability based on the relevant job descriptions and the context. For example, the public has a right to know if a city official is taking bribes, but no right to know the sexual orientation or medical history of the official. Admittedly, the autonomy-protecting aspects of speech related to government officials is tricky and to my mind should be spelled out in detail, or negotiated, before privacy intrusions occur. Beyond the information-sharing practices delineated in an employment contract, perhaps a rule like probable cause could be used to determine when incursions into private domains are acceptable.Footnote ⁵⁷ Engaging in criminal activity would seem, at first glance anyway, to sanction a closer look. Probable cause that an official is in violation of an employment contract could sanction the opening up of private areas as a condition of the contract itself.Footnote ⁵⁸

Consider a different case where context and contract set the terms of information-sharing and free speech practices. In the typical university classroom there is a subject matter being investigated along with various rules regarding speech, expression, and privacy. Speech rules typically include prohibitions against blurting out content even if this content is relevant to the class. Prying into the personal lives of students, teaching assistants, or professors is not acceptable behavior unless there is probable cause that a student or instructor is violating the class contract. The class contract sets the terms of interaction, the subject matter, and grading policy. Thus, it may be perfectly appropriate for a professor to closely monitor students while they are taking a test, but this same behavior would be inappropriate outside of class tests or outside of class.