Arguing with Unreasonable Persons

How Rawlsians should address Nozick, Trump, and Hitler

I wrote last week about a kind of applied political philosophy conundrum that people like Donald Trump create for those of us who endorse the idea of public reason. As polls and predictive models are now putting Trump slightly ahead in the presidential race, I would like to expand on this discussion briefly. In particular, I mentioned an article from the philosopher Baldwin Wong in Philosophia that develops an interesting argument about Rawlsian political liberalism can deal with “unreasonable persons” while showing them respect.[1] It is worth considering why this argument doesn’t seem to work in the current American political context – nor in the European one for that matter.

Let me briefly recall the setup in which this argument takes place. Some political philosophers accuse Rawlsian political liberalism of “sectarianism” on the ground that Rawls and his followers tend to consider that the coercive interference of the state should only be justified to “reasonable” citizens. Per definition, a citizen is reasonable if and only if she endorses the Rawlsian postulate that persons are morally free and equal beings who live in a society conceived as a fair system of cooperation. In turn, reasonable citizens accept liberal principles of justice that grant absolute priority to basic liberties and rights and promote a fair distribution of primary goods and opportunities in society. The point is that even if reasonable citizens disagree on almost everything else, there can still be a consensus over the basic structure of the society and on matters of basic justice. State coercion can therefore be justified and legitimate in a society where reasonable pluralism prevails. The problem – and this is where the sectarian accusation arrives – is that it is unclear how “unreasonable” persons (those who reject the Rawlsian postulate) should be treated. The risk is of course to treat them with condescendence (at best) or total lack of respect (at worse), as some sort of second-tier citizens.

Rawlsians have imagined several ways to circumvent the objection. The answer that I want to consider here is that reasonable citizens (including political officials) can address unreasonable persons through conjectures. The basic idea is to show unreasonable persons that, by adopting their own values and beliefs, they could also reasonably endorse the liberal principles and institutions that reasonable citizens are de facto imposing on them. This idea was already suggested by Rawls:[2]

“Observe here that there are two ideas of toleration. One is purely political, being expressed in terms of the rights and duties protecting religious liberty in accordance with a reasonable political conception of justice. The other is not purely political but expressed from within a religious or a nonreligious doctrine, as when, for example, it was said above that such are the limits God sets on our liberty. Saying this offers an example of what I call reasoning from conjecture. In this case we reason from what we believe, or conjecture, may be other people's basic doctrines, religious or philosophical, and seek to show them that, despite what they might think, they can still endorse a reasonable political conception of justice. We are not ourselves asserting that ground of toleration but offering it as one they could assert consistent with their comprehensive doctrines.”

Rawls goes on to note that for this to work, while it is not required that the citizen reasoning from conjecture actually accepts the unreasonable citizen’s values and beliefs, she must however offer a conjecture that is “sincere and not manipulative.” I put aside here the question of whether this is possible at all.[3] The issue that concerns me is whether reasoning from conjecture can work on all kinds of unreasonable citizens. In his aforementioned article, Wong contends it mostly targets “unreasonable but intelligible citizens” (UIC). UIC by definition do not accept the Rawlsian postulate but they (i) use reason to reject a law or a policy while not relying heavily on false empirical claims and (ii) they respect a basic code of civility, refraining to using violence and being willing to listen to others.[4]

Wong’s prototypical example is Robert Nozick. Nozick obviously didn’t accept the Rawlsian postulate and the resulting principles of justice. Actually, Nozick mostly disagreed with the second principle (about the fair distribution of opportunities and primary goods), but not so much with the first. A key point of disagreement is then whether the state is legitimate in using coercion to redistribute wealth from the rich to the poor. Nonetheless, Wong contends that we can design a conjecture offering a plausible argument that, from the premises of Nozick’s political philosophy, leads to the endorsement of a form of redistribution. I doubt that this is actually the case – and Wong refrains from affirming that the argument is conclusive, but let’s just stipulate that reasoning from conjecture can work often enough to make it a relevant strategy to deflect the accusation of sectarianism.[5]

Even in this case, there are two relevant concerns regarding the effectiveness of arguing through conjectures with unreasonable persons, especially with the current state of political affairs in mind. The first is that it explicitly cannot work with all unreasonable citizens, i.e., those who are “unintelligible” to use Wong’s wording. On the opposite end of the spectrum of unreasonableness from Nozick, figures someone like Hitler. Rawlsians agree that there is no point arguing with someone like Hitler. His beliefs are inconsistent and grossly based on false empirical claims. And, of course, he refrained from following any basic code of civility. In practice, it may remain complicated to determine how to deal with people like that, but in many countries, they are simply treated as persons who disrespect the law. But then, comes Trump. By Wong’s definition, Trump is unintelligible. He doesn’t care about the truth, makes blatantly false claims, threatens his political opponents by labeling them “enemies,” his changing his mind on some topics again and again. The problem is that while we have no problem accepting that Hitler is someone who should just be neutralized, we (the liberals who may endorse something like Rawls’s postulate) are not sure about Trump. Sure, he may legitimately go to jail for his fraudulent behavior, but not because his political ideas are dangerous. We recognize that it’s probably in vain to address Trump through rational arguments, but we don’t agree that this justifies putting him out of the political game, by force if necessary. What to do then?

How to argue with unreasonable persons: John Rawls addressing Adolf Hitler, Donald Trump carefully listening (generated with Dall-E)

Matters become worse when we realize that the reasoning-from-conjecture approach is appropriate when unreasonable persons are a small minority. Basically, it is just presupposing that we are living within the kind of public political culture from which Rawls says he is constructing all the core concepts of his political liberalism. But such presupposition cannot be kept for a society that has a fairly good chance to elect by democratic means an unreasonable and unintelligible citizen as its next president. It’s not to say that all Trump voters are unreasonable and Democrat voters are reasonable (though Rawls’s framework may imply something like that) but many Trump voters arguably just display the same lack of interest in truth as him. I’m not saying that Trump voters don’t have any reason to be displeased with the state of their country or their personal situation. As in France and in many other European countries, the vote for the extremes betrays real concerns and problems that governments of liberal democracies have failed to address over the past three or four decades. It doesn’t change the fact, that because of this and other reasons, the political culture of liberal democracies has evolved in a direction that makes the Rawlsian framework less and less relevant to account for political disagreement.

We may hope that there are other liberal frameworks that may be more helpful in sorting out things. If this is the case, they should probably adopt a more Weberian conception of politics, as I’ve already suggested here. This is, however, the topic not only for another post but for a full research program!

---

[1] Baldwin Wong, “Is It Sectarian for a Rawlsian State to Coerce Nozick? – On Political Liberalism and the Sectarian Critique,” Philosophia 51, no. 1 (2023): 367–87.

[2] John Rawls, “The Idea of Public Reason Revisited,” The University of Chicago Law Review 64, no. 3 (1997): 765–807, p. 783. Emphasis are mine.

[3] See Micah Schwartzman, “The Ethics of Reasoning from Conjecture,” Journal of Moral Philosophy 9, no. 4 (January 1, 2012): 521–44.

[4] Wong, “Is It Sectarian for a Rawlsian State to Coerce Nozick?”.

[5] Rawls’s discussion of reasoning from conjecture was mostly addressing disagreement related to religious beliefs and practices. I find the argument more compelling for this kind of cases.