Popper On “Coercion”

In my previous post, I talked about the difficulties libertarians have even just agreeing with each other over what constitutes an ‘initial use of force.’ One libertarian might see traffic laws as violating ‘initial use of force’ and another may see ‘reckless driving’ as an ‘initial use of force.’ But no matter where a libertarian comes down on where to draw the line, I’ve never even once seen them have the slightest doubt that they drew the line in the right place. And they always believe that their line is so obvious that only an immoral person would disagree.

In other words, libertarians have a very consistent The Truth is Manifest problem when it comes to defining ‘initial use of force.’ They personally know what ‘initial use of force’ looks like when they see it and they are certain they are right and everyone else is wrong. There is no spirit of fallibilism that ever enters into the discussion. Every libertarian seems to believe they-should-rule when it comes to defining ‘initial use of force’ and no one else’s opinion can possibly be correct. This seems to be directly related to the fact that libertarians use the concept of ‘initial use of force’ in an easy-to-vary manner.

In this post, I’m going to explain why the concept of ‘initial use of force’ (at least in its current libertarian form) violates Popper’s epistemology and how the concept of ‘coercion’ actually fits within Popper’s epistemology. In addition, we’ll explore how libertarians might fix these issues and improve their political philosophy by incorporating Popper’s philosophies into their views.

Mask Laws Again

In the 4S email group, we’ve discussed mask laws quite a bit. The libertarian members seem all united in the idea that mask laws are immoral because they are coercive. The non-libertarian members are split up in their views. (Two non-libertarian members are in favor of them. I’m not in favor of them and think they are generally bad ideas.) [1]

The libertarian argument against mask laws tend to take a single unified approach that goes like this:

  1. Mask laws are coercive.
  2. Coercion violates Popper’s epistemology because it’s like gluing a piece of a puzzle in place. It removes possible error correction.
  3. Libertarian politics is the only political view against all coercion, thus it is the only political view consistent with Popper’s epistemology
  4. Ergo, mask laws are immoral because they violate Popper’s epistemology.

Even though I’m against mask laws (see below for arguments I favor), I believe that every single one of the above statements is fundamentally incorrect. In my opinion, this argument makes the argument against mask laws seem easy-to-vary and weak. And moreover, it’s at odds with Popper’s actual epistemology.

Why Finding “Initial Use of Force” Isn’t the Basis for Popper’s Epistemology

Now, of course, there are undisputed cases of “first use of force.” These are usually non-controversial cases that are no longer debated anymore. No one thinks killing people for fun is moral. [2] As such, no one is in favor of ‘initial use of force’ where the concept is clear cut. If all libertarians meant by ‘initial use of force’ were these clear cut cases, then this would be a trivially true view that all political views agreed upon.

The reason libertarians have problems agreeing on what constitutes an initial use of force is because they are insistent that it always exists in every circumstance. Since their philosophy is that all evil is the initiation of force, they seek an initial use of force even in cases where it is not clear cut or even doesn’t exist at all.

It is interesting to note that Popper himself never claimed that his epistemology was about finding something like a ‘first use of coercion’ in every disagreement and then opposing it. Popper famously opposed many fundamental libertarian ideas. For example, Popper told Thomas Szasz that his libertarian views on free trade of drugs were “silly” and he also advocated against unrestrained capitalism and for political interventions to protect the economically disadvantaged. (OS V2, p. 124-125)

So why did Popper not see Critical Rationalism as fundamentally about finding the ‘first use of coercion’ like libertarian critical rationalists do? The reason why is because Popper believed that it is common for people to simply find themselves at odds with each other because they have to live near each other and have different ideas on how to live their lives.

A Liberal Utopia – that is, a state rationally designed on a traditionless tabula rasa—is an impossibility. For the Liberal, Principle demands that the limitations to the freedom of each which are made necessary by social life should be minimized and equalized as much as possible (Kant). But how can we apply such an a priori principle in real life? Should we prevent a pianist from practicing, or prevent his neighbor from enjoying a quiet afternoon?

All such problems can be solved in practice only by appealing to existing traditions and customs and to a traditional sense of justice: to common law, as it is called in Britain, and to an impartial judge’s appreciation of equity.

Conjecture and Refutation, p. 472

For lack of a better term, we can call this phenomenon “simultaneous coercion” if you like. Neither side actually intended to coerce the other and neither initiated the problem. The problem just existed because people have to live near each other and have different ideas about how to live their lives.

Now given that this coercion is “simultaneous” it must be the case that it isn’t really “coercion” in the traditional sense at all — it’s really just externalities. It’s just a set of problems two (or more) people have that need to be solved due to the existence of each other. This is the correct Critical Rationalists way to look at the situation rather than by seeking “who is being coercive first” since neither was being coercive first.

If you are having an argument with someone over something, then you aren’t dealing with a non-controversial case – you’re dealing with one of the cases where it’s not obvious who “initiated the coercion.” That is why “initial use of force” or finding “who committed the coercion first” is not often helpful and will likely instead result in blinding yourself to your own coercive behavior via the truth is manifest error and anti-fallibilism.

This explains why libertarians are always so certain they have found the correct place to ‘draw the line’ on what counts as ‘initial use of force’ (even while never agreeing with even other libertarians!) It’s a case of every observation being ‘theory-laden.’ If you have wrongly based your political views around the idea that ‘initial use of force’ is the correct moral way to solve every problem you will now find an ‘initial use of force’ even if there isn’t one. And it will nearly always be a self-serving observation biased to whatever you happen to experience as coercion first. It will be very difficult for you to see how your own actions might be experienced by others as coercion.

Both Sides Feel Coerced

Consider the idea of mask laws. Suppose you don’t want to wear a mask. A law that says you must (or face a fine or jail time, or whatever) feels like a form of coercion to you. So if you are a libertarian that is against mask laws you will, naturally, feel like those that are in favor of mask laws are the ones being coercive first. You’ve now found your ‘first use of force.’ Ergo, mask laws are immoral.

But suppose you are a libertarian that is in favor of mask laws. You don’t want to be subjected to the risk of contagion during a pandemic by people acting negligently. So to you, the feelings of coercion start with the ‘idiot’ walking around without a mask during a dangerous pandemic and thereby creating an unnecessary risk of harm. You’ve now found your ‘initial use of force.’ Ergo, mask laws are moral because it’s the other guy that acted negligently and you’re just responding to their ‘initial use of force.’

But in reality, there is no pure coercion in either case. Just two people with opposing problems and no currently known solution. The search for coercion was false, misleading, and only made things worse because now you are both so certain you are morally correct you can’t even see that your own behavior is also ‘coercive.’ (i.e. simultaneously coercive.)

Libertarians are NOT the Only Political View Against All Coercion

This is also why libertarians are not the only political view that is against coercion. They are merely committed to putting their opinions into the language of coercion. But in reality, every political side is arguing that their views are correct because someone else did something immoral (coercive) first. This is true of even extreme examples: communists sincerely believe that the coercion started with the bourgeoisie taking advantage of the proletariat. Leftists sincerely believe that the coercion started with a ‘white supremacist system’ that was meant to put down people of color and take advantage of them. Never mind if they are correct or not. The key point here is how easy-to-vary the language of coercion actually is.

If all that matters is putting your claims into such language (as libertarians do) it would be trivial to take every single political discussion today and just make cosmetic changes to their arguments and suddenly all sides will be arguing over ‘initial use of force’ and not a thing will be resolved.

The Popperian Solution

The proper Popper-Deutsch way to look at disagreements like this is to ask “Are there any laws of physics that forbid us from coming up with a solution that both allows us to not wear masks and removes any chance of catching the virus?” Clearly, there is nothing in the laws of physics forbidding such a solution. Finding a vaccination for the virus is such a solution. 

If your solution does not solve both sides problems at the same time, then insisting on your solution and only your solution is coercive. And you are merely advocating for crypto-coercion, rather than against coercion.

No Need for a Special Definition for Coercion

To understand feelings of being ‘coerced’ we can define the term using Word Nominalism like this:

Coercion: Coercion is the word we use to describe the feelings we have when the mere existence of someone else with a different idea than me acts in a way that impacts me in ways I do not enjoy.

Basically, I’m accepting that the mere feelings of coercion is what ‘coercion’ means. The libertarians on 4S have argued to me that this definition isn’t useful because it defines ‘coercion’ so loosely as to apply to anyone. They misunderstand. That is the whole point of Word Nominalist definitions. Hanging special legal meaning on a word is Word Essentialism and is at odds with Popper’s epistemology. The very fact that you’re seeking to define ‘coercion’ in some deep precise way was the first sign you were off the rails in terms of Popper’s epistemology.

Popper points out that you should be able to express yourself without a need to precisely define your terms. Allow me to demonstrate that Popper was right by laying out the proper understanding of how to deal with coercion using Popper’s epistemology using no specially defined terms.

How to Handle Coercion Using Popper’s Epistemology

Suppose you and someone else have a disagreement. Let’s say one of you is in favor of mask laws and one of you is against mask laws. How should you go about resolving things:

  1. Can you convince one side that they are wrong?
    1. For example: maybe you do studies and find evidence that the pandemic is not at all dangerous. So those that are in favor of mask laws stop feeling like you not wearing a mask is negligent behavior. This would resolve the problem for both sides.
  2. If you can’t convince one side that they are wrong, can you creatively come up with a solution that fully satisfies both parties?
    1. As previously stated, a vaccination is one such solution. There may be others we haven’t yet thought of. Obviously, solutions like this require creativity and the creation of new knowledge.
  3. If you can’t come up with a solution that fully satisfies both parties, can you come up with a solution that both parties can live with?
    1. Here is an example: what if we don’t make mask laws in areas where the pandemic isn’t yet out of control, but instead ask people to voluntarily wear masks. But we make it clear that we’ll implement mask laws should this measure prove insufficient to avoiding an overrun of the medical system. This solution isn’t ideal for either side, but both sides might find it better than the other alternatives being offered. They may at least find this solution ‘tolerable.’  
  4. And finally, if you are unable to come up with a creative solution that satisfies both parties, then it is necessary to implement a solution that one or both parties still find coercive. So pick the least coercive of the available solutions.

The Principle of Least Coercion

Solutions 1 and 2 are both always (by definition) perceived by all parties as non-coercive. Solution 3 is perceived as coercive but tolerable. It’s the best we know how to do at the moment, but we know we can do better with the right kind of knowledge. As such, both sides should still seek a better solution (e.g. discover a vaccine) rather than stopping at Solution 3.

But here is the interesting point: if you’re having a political debate with someone, that implies there is no agreed-upon solution. So that means the realm of political debate happens to exist at solution level #4. That is why political debate is so contentious: because what we’re discussing is really who gets to coerce whom and by how much.

Political discourse lives at solution level 4. The very fact that you’re having the debate means that we have yet to creatively find a non-coercive solution that will satisfy everyone’s feelings of being coerced.

The Principle of Least Coercion: Political Debate to Reduce Coercion Until A Non-Coercive Solution is Found

So the correct Critical Rationalist replacement for the easy-to-vary concept of ‘initial use of force in all cases’ is instead the principle of ‘least coercion.’ In principle, we all agree that if you are at level 4 — because there is no known non-coercive solution known — then seeking the ‘least coercive solution’ is the right choice. What we’re not likely to agree upon is which solution is the least coercive.

This is really just a restatement of Popperian Fallibilism. The whole point of the debate is to find the least coercive solution. That we are debating it means we don’t agree on what it is. The debate exists to help us criticize proposed solutions and to force improvements to proposed solutions. With some luck, we’ll discover a ‘tolerable’ solution (i.e. Solution #3). But even that may not happen. And in this case, even not acting is a type of coercive action. So we simply have no non-coercive solutions available to use yet.

Put bluntly, liberatrians believe in the Truth is Manifest error. They wish to avoid political debate by simply finding who “obviously” initiated force. Conveniently, it’s always the other party. But the reality is that it’s not obvious who initiated force or even that that anyone initiated force. That’s why the political debate is warranted.

Why We Must Solve Problems, Even if Someone Feels Coerced

The principle of least coercion is the correct Popperian view of coercion. Coercion is not, by itself, anti-Popperian epistemology if all the other alternatives are more coercive. In fact, it’s a sign that we have yet to create the needed knowledge to make everyone satisfied. And given such a lack of knowledge, the correct answer is to try the best idea we can come up with and see what happens, then learn from it. The traditions and legal codes of an open society help us determine what thing to try first. But those traditions may be mistaken for this particular crisis and we may have to try something else. But the need to act and learn is paramount in an open society. And that is why I am against mask laws, but accept that society has every right to implement them. We should not upfront outlaw possible solutions to novel problems using easy-to-vary arguments like ‘initial use of coercion.’

Put plainly, trying out mask laws (or not trying them out) does not violate Popperian epistemology in and of itself. Any action to solve problems is better than not solving problems. This is why the glued puzzle piece analogy is incorrect. To gain knowledge, a Popperian open society has no choice but to take coercive actions – from the point of view of those that disagree with the action. The key thing that makes it Popperian is that it can be changed bloodlessly and something else tried later. This is exactly the opposite of gluing a puzzle piece down. This is why Popper has no issue at all advocating for laws that to libertarians “seem coercive.”

Would Popper Have Been In Favor of Mask Laws?

So does all of this mean that Popper would have been in favor of mask laws? Not even slightly. The realization that both sides are being ‘coercive’ doesn’t mean both sides are being equally coercive or that one side isn’t more correct than another.

Let’s put this another way. Let’s say you are against mask laws. So you make the following argument:

  • Mask laws are wrong because they are coercive while allowing people to not wear masks around those that don’t want to get sick is not coercive.

Compare this to the following argument:

  • Mask laws are more coercive than forcing people to get sick due people not wearing mask.

Neither argument is particularly good, but at least the second one has the benefit of not being known to be false. Additionally, the second one doesn’t fool you into thinking it’s a better argument than it is. That means you will be more capable now of improving your explanations. Some of the following might be even better:

  • We made so many draconian measures upfront that now we can’t use them when it really matters.
  • In some countries mask laws and lockdowns have lead to everyone needing to scan in everywhere they go, giving the government all one’s real-time location data. The loss of privacy and increased government control isn’t worth the cost.
  • Do mask laws really work or would it actually be more effective to make them voluntary?
  • Do we better serve the public by mandating anti-pandemic standards for every business owner or by letting each business creatively come up with their own voluntary measures and then take the consequences if they become a COVID hotbed? Might that not drive better innovation?

In other words, you can only actually make progress by getting to good hard-to-vary explanations that even someone in favor of mask laws and lockdowns can see might have merit. These are the right kinds of Popperian-style arguments for libertarians to be making. Merely calling mask laws wrong due to being coercive is both easily shown to be wrong and also a bad argument not likely to persuade anyone. Are you interested in making actual progress or not?


In conclusion, I have explained that Popper’s epistemology when applied to open societies is not correctly implemented as a (witch)hunt for ‘initial use of force’ but as an open-ended search for solving each other’s problems.

Mask laws are one possible approach to solving problems — I suspect a bad one, though I might be wrong. Convincing people via persuasion is another possible approach.

During the pandemic, there was (and still is) insufficient knowledge to understand how to best act, so there was no choice but to pick a course of action and try it, even though some people will inevitably find that solution undesirable and thus, for them, ‘coercive.’ And they are not wrong. “Coercion” is often merely a subjective feeling people feel when the fact that they have to live with others forces them to do things they’d prefer not to. And we should take their feelings seriously. We should try to come up with the most creative problem-solving solutions we can until, if possible, everyone is satisfied with the solution. (At which point no one will feel coerced any more.) This is what it means to live in a knowledge-creating open society and it is the only true meaning of removing coercion from society. Removing coercion requires knowledge-creation.

What we should not do is let ourselves become blind to other people’s legitimate problems. Deciding that mask laws are immoral on the grounds they are ‘coercive’ (or deciding that they are not coercive for the very same reason) both are cases of one side becoming blind to the reality of how their own preferred solutions impacts other people.

As Jon Haidt would say “morality blinds and binds.”  The only truly non-coercive solutions are the ones that satisfy everyone. Absent the necessary knowledge to do that, we are only left with coercive solutions. Popper’s epistemology says we should not hesitate to use ‘coercive solutions’ so long as all the known alternatives are more coercive. So we should pick the ‘least coercive solution’ available to us (though we might not always agree on what those are.) And even then, we should never be satisfied until we find an entirely non-coercive solution — which is just a fancy way of saying we keep seeking solutions that resolve all side’s concerns.


[1] However, I note that not all libertarians agree with those in the 4S email group and there are libertarians that insist mask laws make sense within a libertarian worldview. However, this post is specifically a response to the view expressed in the 4S group that masks laws are wrong because they are “coercive” and one need only identify a ‘first act’ of coercion to know that something is wrong and no further analysis is required to see if your own actions are also coercive from the other viewpoint.

[2] Not even serial killers believe that killing for fun is moral. They just don’t follow through with their beliefs. See Evil Genes by Barbara Oakley, p. 51-52 for discussion.

Leave a Reply

Your email address will not be published. Required fields are marked *