On Coercion

Why It is Inescapable and Must Be Justified

A couple of weeks ago, I wrote about the claim that there is a presumption in favor of liberty. This claim seems to have a corollary. If there is a presumption in favor of leaving people to do what they want, it seems that there is also a presumption for not coercing people. On this understanding, coercion is just the other side of the freedom coin. In this essay, I want to explore this apparent symmetry. In the end, this symmetry is only apparent.

Coercion is a notoriously difficult concept to define. According to Friedrich Hayek, “coercion occurs when one man’s actions are made to serve another man’s will.”[1] This is very loose. A more specific definition could run like this: “A is coercing B if A is using force, or threatening to use force so that B does α while she would prefer to do β.” This more specific definition differentiates coercion from persuasion, manipulation, and other kind of non-coercive influence. Persuasion and manipulation do not entail the use of physical force. It consists of changing people’s preferences and beliefs by providing arguments and information (persuasion) or by deceiving them (manipulation) so that while one wanted initially to do α, she ends up preferring to do β. This also permits us to distinguish coercion from the related and close notion of power, often defined as A’s ability to make B do what A wants. Coercion is one of the ways power over others can be ascertained. This is not the only one, however. Consider how an autocrat can exert power on the population. The autocrat can use brute physical force, or the threat of it, to coerce people into doing things they don’t want to. But this is unlikely to be very effective, especially if society is large and monitoring costs are high. A better approach is to count on people’s fear and tendency to self-administrate punishment. If the autocrat is able (by persuasion or manipulation) to make everybody believe that anyone who displays unloyalty toward the regime will be socially boycotted because interacting with a disloyal person is to be disloyal, people will tend to display loyalty in public.[2]

That’s it for the definitional work. The reason why we may want a finer concept of coercion is that power is a too diverse and general phenomenon to be able to make interesting claims here. Sources of influence are multiple and therefore there are power relationships everywhere. That doesn’t mean that the phenomenon is not worth to be studied. It’s only to say that A has power over B is not really informative as long as you’ve not made a more specific claim about the source of A’s power over B. So, let’s focus on coercion as a medium of power. Liberals and libertarians have traditionally focused on state coercion. The very nature of the state makes it, in the words of Max Weber, the entity with a monopoly over the legitimate claim to use coercion on the inhabitants of a territory. The standard libertarian argument against the state is that coercion is wrong because it violates individuals’ natural rights. Only voluntary transactions are respectful of individuals’ rights. Since coercive taxation is required for the state to survive, the state is illegitimate, plain and simple.[3]

There are several problems with the libertarian argument. One is that, though the state is coercive, it’s arguably also the case of other social relationships that don’t involve the state. In particular, Marxists and other leftist critics may argue that market relations involve coercion. Market transactions transfer property rights. Property rights partly consist of a right for the owner of γ to exclude non-owners from the use and the benefits of γ. Correspondingly, property rights create obligations for non-owners to not use γ. Hence, property rights are coercive in the sense that they license the use of coercion to force non-owners to respect their obligations. That in state societies like ours it is the state that is legitimate to use coercion to enforce property rights doesn’t change anything. In an ideal libertarian society, coercion would be exerted by owners themselves or, more likely, by “protective agencies.” Coercion is no more legitimate per se in the latter than in the former case. You could say that the state is not legitimate in using coercion to collect taxes or enforce property rights because you haven’t voluntarily contracted with the state to give it these prerogatives. I could also say that I have no reason to regard the coercion exerted by a protective agency to enforce your property rights as legitimate because I haven’t voluntarily contracted either with the agency or you to give you this right.

In a recent post, Matt Zwolinski persuasively shows that this argument is however not convincing:

But what should we conclude from this? All legal property arrangements involve coercion. So is there no difference then between the coerciveness of a laissez-faire economy and a more heavily regulated one? Is an economy based on free labor just as coercive as a Stalinist, centrally planned one?

That seems silly. Even if all systems of ownership involve coercion, surely there are important differences between the degree and kind of coercion involved. We can start by noting, with Friedrich Hayek, that coercion based on general rules is very different from coercion based on the imposition of someone’s arbitrary will. People can learn the rules of property in advance, and plan their lives around them. A Stalinist dictator, in contrast, is much harder to predict - or avoid.

Coercion can also limit options that are more or less central to a person’s interests. It’s one thing for me to tell you that you’re not allowed to walk on my lawn. It’s another thing for me to tell you that you can’t practice your preferred religion. It’s yet another thing if everyone is telling you that you’re not allowed to walk on their lawn, or in their store, and doing so deprives you of access to essentially needed resources.

Zwolinski makes the important (Hayekian) point that we should distinguish between coercion based on general rules (like the rules of property) and coercion based on the arbitrary will of an individual. We should also look at the nature of the options that coercion prohibits. Some practices, goods, and values are arguably more important than others for different people. The use of coercion to make it impossible for individuals to receive information and to express themselves is arguably more problematic than the use of coercion to refrain you from using my car without my permission.

Another way to formulate things is that as soon as individuals have rights (property rights, rights of free speech, …), coercion is necessarily underneath. Otherwise, rights would not be enforced. This is the aspect that makes freedom and coercion the two sides of the same coin. There is no freedom without rights and no rights without coercion. Hence, there is no freedom without coercion. There is absolutely nothing paradoxical or even counterintuitive here. That complicates however the relation between the presumption in favor of liberty and the would-be presumption against coercion. Suppose that we decide that the just society is the one that maximizes the extent of individuals’ basic liberties, with the constraint that everybody must have the same liberties.[4] That does not automatically imply that we should minimize coercion. What we want to do is eliminate some forms of coercion, i.e., those that are arbitrary and not justified by general rules that everybody accepts (or has reason to accept). Moreover, in light of the fact that individuals are likely to have unequal abilities to fully realize their basic liberties, a liberal society can be relatively coercive, e.g., forcing some to pay relatively high taxes to fund a universal income. The bottom line is that we want coercion to be justified, not minimized.

Let me end by briefly mentioning an argument that market relations can be coercive in a different sense, i.e., that the take-it-or-leave-it nature of market relations can force some people into worse situations than if no market transaction were taking place.[5] The argument is basically like this. Suppose Alf and Betty do business together. Betty is also doing business with Chris. Now, assume that Alf revises the deal that he offers to Betty in such a way that Betty would be worse than not doing business with Alf. Obviously, Betty should refuse. But, as it happens, Alf also has the capacity to offer a very interesting deal to Chris,[6] but only on the condition that Chris stops doing business with Betty if Betty refuses Alf’s offer. Under some conditions, the threat can be credible, i.e., Chris will stop doing business with Betty to be sure to do business with Alf. In turn, Betty will have no choice but to accept Alf’s offer, even though it is pretty bad for her. Some may argue that Alf is coercing Betty into accepting a worse deal. Compare this with the situation where Alf is threatening to hurt Betty if she doesn’t give him her wallet. The outcome is strikingly the same. And, since market transactions are possible only because Alf, Betty, and Chris have property rights, the market arrangement is backed up by coercion.

My answer to this case would be that since the property rights of our three individuals are legit and backed up by general rules, there is nothing wrong with the outcome, even if it can be described as a form of economic coercion. The difference with the wallet scenario is that, in the latter, Alf would infringe on Betty’s property rights. That doesn’t mean that the outcome in the market scenario is justified. Obviously, Alf’s coercive power over Betty is due to the initial distribution of property rights. This distribution might be unfair. But the objection here doesn’t come from the use of coercion, but rather from the unfairness of the distribution.

---

[1] F. A. Hayek, The Constitution of Liberty: The Definitive Edition (Routledge, 1960 [2020]), p. 133.

[2] Timur Kuran, Private Truths, Public Lies: The Social Consequences of Preference Falsification (Cambridge, Mass.: Harvard University Press, 1997) is a good theoretical and historical discussion of this kind of mechanism. This is also close to how Vaclav Havel characterizes totalitarian power. See Vaclav Havel, The Power of the Powerless (Vintage Classics, 2018).

[3] Some libertarians like Robert Nozick have displayed some ingenuity to show that a minimal state can result from the voluntary choices of individuals. However, the logic of Nozick’s argument is not completely clear and it remains that, once the state has emerged, it’s coercive nature is likely to surface. Robert Nozick, Anarchy, State, and Utopia, 2e édition (New York: Basic Books, 1974 [2013]).

[4] This is similar to Rawls’s first principle of justice. John Rawls, A Theory of Justice (Oxford University Press, 1971).

[5] This argument is made notably by Kaushik Basu. Kaushik Basu, Prelude to Political Economy: A Study of the Social and Political Foundations of Economics (Oxford University Press, 2000), pp. 147-55.

[6] Or suppose that Alf and Chris are engaged in the intertemporal provision of a public good and that Alf credibly commit to provide the public good if Chris does not deal with Betty.