Very short summary: This essay discusses the distinction between the public and the private. This distinction has a particular value in liberal societies. I argue that publicity is a requirement of social morality. In the meantime, privacy is part of the rights that liberal social morality should guarantee. Unfortunately, there is no unconditional principle determining where one stops and the other begins.
A typical feature of modern liberal societies is the well-understood importance of the distinction between the public and the private. In saying this, I’m not implying that the distinction has been unknown to other types of human societies. For reasons I will discuss below, the reach of publicity is limited —just not everything can be public in society. On the other hand, as I will also explain, a modicum of publicity is required to permit social coordination. Complete privacy is impossible to obtain and, in any case, not desirable.
Nonetheless, the affirmation of the distinction and the understanding that many disagreements about how to organize society and how to live within it are with respect to where to draw the line between the private and the public sets liberal societies apart. Consider the contrast with totalitarian societies. A defining characteristic of totalitarianism is its Orwellian negation of privacy. The totalitarian ideal is complete transparency. It can be achieved directly through state surveillance (as in Orwell’s 1984 novel) or, more realistically, indirectly with the cooperation of the population itself.[1] More generally, human societies tend to pursue what I’ve called in a previous essay a normalization of human behavior. For the state, normalization is a source of power because it increases its ability to control the behavior of individuals by making it more predictable. Part of the normalization process consists in turning what was private into public information, for instance by making sure that people communicate in a language that is understandable by everyone else.
At the other extreme, some are fantasizing about fully private societies where individuals have complete control over the information they want to display. This old libertarian ideal is now enjoying a second life with the advent of technologies like smart contracts and the blockchain. They open the way to forms of life where most social interactions pass through encrypted data, which, for instance, conceals the identity of users. In particular, from the outset, the attraction of cryptocurrencies has been that they promised the possibility of escaping state surveillance and, thus, its control over transactions. Of course, these technologies are a double-edged sword, as they can also easily be turned against privacy, depending on who controls them.
Economic theory can offer an interesting perspective on the delicate balance between the private and the public. In undergraduate textbooks, it is sometimes suggested that “perfect information” is one of the conditions necessary to achieve an efficient allocation of resources. Writing this is actually more confusing than anything else. Indeed, a system of perfectly competitive markets only requires the publicity of prices. Consumers’ preferences and firms’ production costs, while relevant to determine the equilibrium prices, don’t need to be known by everybody, even less be “common knowledge.” What is true, however, is that if a planner wanted to infer the general equilibrium of this system —i.e., the vector of prices that equalizes supply and demand on every market, they would need to have perfect knowledge of preferences and production costs. As Friedrich Hayek pointed out, this information is private and costly to gather, making the task of the planner an impossible one.[2]
However, economics teaches us that information problems, such as information asymmetries, may impede market mechanisms, putting the planning issue aside. For instance, if consumers are unsure about the quality of products, they may be unwilling to pay too much, thus driving producers of high-quality but more expensive goods out of the market.[3] The roots of the problem are that some information (here, about the quality) is private rather than public, and that there is no easy way to credibly communicate it. More generally, coordination failures are often the result of incorrect expectations that are themselves grounded in imperfect information. For instance, when you drive in a country that is not yours, a major risk comes from the fact that you probably ignore some local rules about driving. Crucially, your ignorance is likely to be not about formal legal rules as they are public (everybody knows that people drive on the left side in the UK!), but rather about informal norms, e.g., that in some cities, drivers routinely cross at red lights. Even worse, while you may ignore some social norms (and even don’t know that you ignore them!), other individuals may wrongly assume that you know them —they don’t know that you don’t know.
This points toward the fact that systems of rules and norms, i.e., institutions, as well as culture, are intrinsically relying on publicity. To understand this point, let me quickly discuss some technicalities. We call an event E something that is happening in the world and that individuals can observe and on which they can form a belief or have knowledge of. E can be caused by some persons’ actions or be the result of some natural forces. Consider a population P of individuals. If everybody in P knows E,[4] then we say that E is mutually known in P (or mutual knowledge in P). As the example above suggests, the fact that one knows or doesn’t know E is not enough to account for people’s behavior. You also need to determine what people know (and believe) about others’ knowledge and beliefs, what they know (and believe) about others’ knowledge (and beliefs) about everyone’s knowledge (and beliefs), and so on. When it happens that everybody knows that everybody knows that… E ad infinitum, we say that E is common knowledge in P.
Now comes the interesting part. E is a public event, just when the fact of knowing E entails that you know that E is common knowledge. This means that upon knowing E, you infer that everybody knows E, that everybody knows that everybody knows E, and so on. In other words, if E happens and is known, it is necessarily common knowledge. This is likely to happen in physical settings where people can directly observe that something is happening and, because of the spatial configuration, can observe that others are observing too. Even better: suppose that you observe E and that others also observe E; and suppose that from E, you infer F and that others also infer F. For instance, you're in a football stadium and you observe Kylian Mbappé putting the ball into the net (event E) and, from this, you infer that France has just scored a goal (event F). Then, both E and F are public events![5]
This last point is very important because if common knowledge were possible only for events that can be directly observed, its scope would be fairly limited. However, in the previous example, the scope of knowledge that can be public in a population is largely expanded as soon as individuals share a form of inductive understanding of their world. In his small but insightful book Rational Ritual, the economist Michael Chwe argues that humans’ ability to coordinate within cultural systems is tied to the existence of public events and a shared inductive understanding.[6] Indeed, on Chwe’s account, culture itself is all about publicity and shared understanding.
This analysis of the role of publicity is relevant to understanding the roots of social morality, i.e., the set of social rules that settle what is morally obligatory, permissible, and forbidden in society. Social morality solves complex coordination problems by serving as an equilibrium selection device.[7] Morality is at least partly as conventional as how to drive on the roads in a specific country. To determine what is permitted or prohibited in some specific context, individuals need to know the relevant social rules. To know the rules, however, is to infer the same practical implications as everyone else from the same set of public events. In other words, social morality is part of the broader cultural framework of society. It relies on the same kind of publicity that helps to solve mundane coordination problems.
The public nature of social morality is especially relevant in social dilemmas where individuals are torn between pursuing their personal interests (or even their personal moral convictions) and following a rule of social morality that instructs them to give up some personal advantage for the common good. The classic situation of this kind is the prisoner’s dilemma. If the utility numbers in the matrix below represent individual interests or their personal moral convictions, then each player’s dominant strategy is to play D (i.e., they will achieve higher utility gains by playing D irrespective of what the other is doing). However, by doing so, they achieve an outcome that is strictly worse for everybody than if they all played C.
A well-known solution to this kind of social dilemma is to design it so that individuals have to play it repetitively for an unknown number of times (or infinitely). Indeed, there are good reasons to think that social morality would have never evolved if both biological and cultural life were not organized around repeated interactions between living organisms, including humans. In the simple 2-player case, if each player can perfectly observe what the other player did in the previous rounds, then a large range of strategies can potentially lead to the optimal social outcome. The most famous of those strategies is “tit-for-tat,” where the player starts playing C and then repeats the opponent’s choice from the previous turn. If, after each turn, there is a probability of p that the interaction lasts at least one more turn, then we obtain a new game.
If p is high enough (here, p > 1/3), the best response against someone who plays tit-for-tat is to do the same. Obviously, if the two players both play tit-for-tat, they will play C from the first to the last turn and achieve the social optimum. The same logic works in large populations where individuals play the prisoner’s dilemma several times in a row against different opponents. However, for it to implement the social optimum, players need to know how their opponents behaved in previous interactions with others. In other words, social morality can do its job only if there is an information mechanism that makes individuals’ behavior public. Biologists speak of “good standing” and economists of “reputation,” but the idea is the same. Your past behavior signals your moral worth (i.e., your propensity to follow the rules of social morality) and gives crucial information to others to help them determine what to do when they meet you. Suppose that based on this information, people play the “good-standing strategy:” if your opponent has ever played D in the past, play D; otherwise, play C. Over the long run, your prospects are similar to those in the second matrix above. Hence, you'd better play C at each turn.
More or less sophisticated versions of the good-standing strategy are everywhere in nature. Due to evolutionary pressures, non-human animals are genetically predisposed to adopt them. In human societies, things are a bit different. Though we also surely have genetic dispositions to cooperate, good-standing strategies are essentially cultural artifacts. They are made possible because publicity is socially engineered. Human societies are organized in such a way as to make public the kind of information required to make social morality work. In most cases, this is the unintentional outcome of cultural evolution, but in a few significant ones, it results from intentional social constructions, eventually with political roots.
For instance, consider the role played by reputation mechanisms in daily economic activities. The fate of many products and services depends on how users rate them. This is a more technologically advanced version of good-standing strategies that abound in nature. China’s social credit system is a generalization of the same logic at a bigger scale, except for the fact that it is also conceived as a quasi-Orwellian tool of social control.
René Magritte, “The Son of Man” (1964)
The last example brings me to what is a concern from a liberal perspective. Publicity clearly is a requirement of social morality. It is also, up to a point, needed for the exercise of legitimate political power. Coordination failures and social dilemmas cannot be solved without making part of people’s lives transparent enough so that their behavior can be anticipated, monitored, and controlled. On the other hand, liberal societies highly value privacy. Privacy is indeed among what is sometimes called the jurisdictional rights that define a private sphere within which individuals are fully sovereign.
In a recent post on his blog, Ethereum’s founder Vitalik Buterin offers a series of arguments that explain the value of privacy:
“Privacy is freedom: privacy gives us space to live our lives in the ways that meet our needs, without constantly worrying about how our actions will be perceived in all kinds of political and social games
Privacy is order: a whole bunch of mechanisms that underlie the basic functioning of society depend on privacy in order to function
Privacy is progress: if we gain new ways to share our information selectively while protecting it from being misused, we can unlock a lot of value and accelerate technological and social progress”
Buterin’s “privacy is freedom” argument directly echoes the ideas that a right to privacy is needed to protect one’s freedom. Full transparency is incompatible with freedom insofar as it gives to others, both individuals and political authorities, leverage for direct and indirect social control. Buterin’s second argument is important as it points toward an often-neglected point. When privacy is compromised and especially if your identity is not protected, a lot of resources are likely to be spent on “side games” when some are trying to influence the decisions of others, e.g., through bribes, social pressures, or physical threats. This is not only harmful social waste. It also somewhat paradoxically threatens the rules of social morality by increasing the personal cost of following them. In this sense, the recognition of the importance of privacy is undoubtedly moral progress.
At the most general level, it is not difficult to understand and accept that both publicity and privacy are needed. But this leads to many practical issues. Last year, I wrote about the apparently meaningless controversy around Taylor Swift’s private jet travel to attend the Super Bowl. In this essay, I was reflecting on John Stuart Mill’s peculiar view about privacy, which he tended not to regard as so important based on his account of individuality:
“The problem is that few of us meet the requirements of Millian individuality. The case of billionaires flying with their private jets is mostly anecdotal because they represent a small minority that has anyway means (including legal) to protect themselves from social pressures. But the rules of social morality must be amenable to universalization. So, if we consider Swift’s case as unproblematic, we should ask whether this remains true when applied to the average person. In a sense, the Millian view of privacy is compatible with the so-called “cancel culture,” minus the use of force to shut down individuals. The latter is largely based on the infringement on privacy and the use of intimidation and collective pressure to eliminate views from the public sphere. Here as elsewhere, liberals face a difficult tradeoff, meaning that we should expect them to entertain a range of different and partially incompatible views.”
Many people are inclined to think that Taylor Swift’s behavior must be made transparent, at least when it creates externalities. That makes sense. To enforce rules of social morality that regulate behavior producing external effects, we need to have enough information about this behavior. But the “universalization test” shows how this can easily turn into invasive demands that would compromise the privacy rights of everyone. Of course, we can argue that privacy rights may sometimes be overridden by other considerations, such as concerns about highly polluting activities. What is not permitted, in principle at least, is that different persons are treated differently on the ground that some are famous and not others.
Practical issues also arise with the way our personal data is collected and eventually used, both by private companies and state authorities. Data collection helps the state to enforce the law. At least in some respects, it can also permit private companies to improve their products and increase the satisfaction of their customers. On the other hand, they can be easily used to infringe on our privacy rights and control our behavior. This is even more true given the large availability of behavioral insights that makes now relatively easy to covertly influence people’s behavior without the use of any form of explicit coercion.
These examples illustrate that the separation of the private and the public is an important point of contention in liberal societies. There is no firm principle about its location, nor any algorithmic procedure to fix it on a case-by-case basis. From a liberal perspective, what must be accepted is that the precise contours of this separation are themselves determined by the rules of social morality, so that they also depend on conventions and contingent beliefs. What we must be attentive to is whether these rules can be universalized so that they are acceptable to all.
[1] See Vaclav Havel’s vivid account of “post-totalitarian societies” where social control is mostly exercised by the population itself. Vaclav Havel, The Power of the Powerless (Vintage Classics, 2018).
[2] F. A. Hayek, “The Use of Knowledge in Society,” The American Economic Review 35, no. 4 (1945): 519–30.
[3] George A. Akerlof, “The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism,” The Quarterly Journal of Economics 84, no. 3 (1970): 488–500, https://doi.org/10.2307/1879431.
[4] I.e., everybody knows that E has happened, or equivalently, that the proposition that describes E is true. For instance, everybody knows that the Oklahoma City Thunder are the 2025 NBA Champions (this event happened) or that the proposition “the Oklahoma City Thunder are the 2025 NBA Champions” is true.
[5] David Lewis is the first to have formally established this result. David K. Lewis, Convention: A Philosophical Study (John Wiley and Sons, 2002).
[6] Michael Suk-Young Chwe, Rational Ritual: Culture, Coordination, and Common Knowledge (Princeton University Press, 2003).
[7] Ken Binmore, Natural Justice (Oxford University Press, 2005). Gerald Gaus, The Order of Public Reason: A Theory of Freedom and Morality in a Diverse and Bounded World, Reprint edition (Cambridge New York,NY: Cambridge University Press, 2012).