Assumption of infallibility

I present here a somewhat more formal version of the famous “assumption of infallibility” argument from J.S. Mill’s On Liberty, with the conclusion that “it is never rational for a man, or government, to silence discussion of some opinion.”

I first shared it with others in 2020-07-23, in a Facebook group that is no longer active. At that time, I often made such versions of philosophical arguments I had found and shared them with that group, so that, with the all the premises distinguished and numbered, people could state their objections more precisely in the comments. It was a great learning experience.

This is probably my best work from that period, in terms of how much easier it is to follow than the corresponding original text. Five premises are asserted, and the conclusion is supposed to be deduced from them. After the formalization, I list the textual support that I had gathered for each premise, all of which was taken from the second chapter of Mill’s work. This is all unchanged from how it was first written in 2020, since I do not wish to read Mill’s work again to think more carefully about the quality of the formulation. After this, there will be some comments on the argument.

The argument

1. Every man well knows himself to be fallible. It is irrational to deny this.

2. Given that we are fallible, we are only justified in assuming one of our opinions as true for the purposes of action if we allow others complete liberty of contradicting and disproving it.

3. (a) A man, or government, who silences discussion of some opinion because he thinks it is false, is acting upon his belief that the opinion is false, and his belief depends upon reference to the opinion. (b) A man, or government, who silences discussion of some opinion for some other reason than it being false, such as it being useful to silence it, is also acting upon a belief of his which depends upon reference to the opinion the discussion of which he is silencing, such as, namely, that it is useful to silence it. (c) In general, all men who silence discussion of some opinion do so for some reason which depends upon reference to the opinion the discussion of which they are silencing.

4. People do not have complete liberty of contradicting and disproving an opinion if the discussion of it is silenced.

5. People also do not have complete liberty of contradicting and disproving a belief which depends upon reference to some opinion if the opinion which is referenced by the belief is silenced.

6. We may therefore infer, from #3–5, that whenever a man, or government, silences discussion of some opinion, he does so for some reason which others do not have complete liberty of contradicting and disproving. But to act in this way is to do what was described by #2 as incompatible with acknowledgment of his own fallibility. Therefore, from all the preceding statements, we may conclude that whenever a man, or government, silences discussion of some opinion, he denies his own fallibility, and therefore acts irrationally. Which is to say, it is never rational for a man, or government, to silence discussion of some opinion.

Textual support for the premises

Support for statement #1:

Unfortunately for the good sense of mankind, the fact of their fallibility is far from carrying the weight in their practical judgment, which is always allowed to it in theory; for while every one well knows himself to be fallible, few think it necessary to take any precautions against their own fallibility, or admit the supposition that any opinion, of which they feel very certain, may be one of the examples of the error to which they acknowledge themselves to be liable. (§2¶4)

Support for statement #2:

There is the greatest difference between presuming an opinion to be true, because, with every opportunity for contesting it, it has not been refuted, and assuming its truth for the purpose of not permitting its refutation. Complete liberty of contradicting and disproving our opinion, is the very condition which justifies us in assuming its truth for purposes of action; and on no other terms can a being with human faculties have any rational assurance of being right. (§2¶6)

Support for statement #3(a):

First: the opinion which it is attempted to suppress by authority may possibly be true. Those who desire to suppress it, of course deny its truth; but they are not infallible. They have no authority to decide the question for all mankind, and exclude every other person from the means of judging. To refuse a hearing to an opinion, because they are sure that it is false, is to assume that their certainty is the same thing as absolute certainty. All silencing of discussion is an assumption of infallibility. Its condemnation may be allowed to rest on this common argument, not the worse for being common. (§2¶3)

Support for statement #3(b), statement 4, and statement 5:

It is also often argued, and still oftener thought, that none but bad men would desire to weaken these salutary beliefs; and there can be nothing wrong, it is thought, in restraining bad men, and prohibiting what only such men would wish to practise. This mode of thinking makes the justification of restraints on discussion not a question of the truth of doctrines, but of their usefulness; and flatters itself by that means to escape the responsibility of claiming to be an infallible judge of opinions. But those who thus satisfy themselves, do not perceive that the assumption of infallibility is merely shifted from one point to another. The usefulness of an opinion is itself matter of opinion: as disputable, as open to discussion, and requiring discussion as much, as the opinion itself. There is the same need of an infallible judge of opinions to decide an opinion to be noxious, as to decide it to be false, unless the opinion condemned has full opportunity of defending itself. And it will not do to say that the heretic may be allowed to maintain the utility or harmlessness of his opinion, though forbidden to maintain its truth. The truth of an opinion is part of its utility. (§2¶10)

There is no textual support for statement #3(c), but I think that it is both implied and required. (Or rather, this is what I thought when I first wrote this, almost two years ago.)

Discussion

Few people said anything in the original group thread. In the poll that I made about which premises people disagreed with, #4 was the most popular to be doubted, which shocked me. As I said in a comment written the day after the thread, “Seriously? #4 should be the least controversial here. It is almost a definition.”

Alexander Brown’s paper

In a different thread – actually written before, in 2020-07-05 – I had expressed my heartfelt disagreement with Alexander Brown’s 2010 paper about the argument. It is supposed to defend Mill’s argument, but it denied premise #5 at the end, which I thought to be important. The result, I thought, weakened the argument and made it make less sense. I quote my summary and comments about it from the time:

Alexander Brown [...] discusses various interpretations of the argument. He concludes that not only is all silencing of discussion an assumption of infallibility, but also, silencing of discussion is a necessary condition for an assumption of infallibility to have been made, in Mill’s sense. He also concludes that such an assumption might be made even if censorship is undertaken without a judgment that the belief being censored is false, but with only a judgment that it is useful to suppress it, as in “Thomas Scanlon’s case of a government that acts against a misanthropic inventor who wishes to broadcast his household recipe for nerve gas on television.” Combining two formulæ given in the paper, it may be defined thus:

To make an assumption of infallibility is to feel sure either

(1) that one is right about the truth or falsity of an opinion or

(2) that one has a moral right to suppress an opinion (because of the utility of suppressing that opinion, or because of some form of democratic mandate, for example)

and to attempt to silence public discussion of that opinion on the basis of that feeling. The first case is called a ‘first-order’ assumption of infallibility, since it stems from a judgment about the belief itself being censored, while the second case is called a ‘second-order’ assumption of infallibility.

[...]

I think Mill’s argument is weakened by the denial, at the end of the paper, of the idea that one cannot judge second-order beliefs about an opinion if the opinion itself is being censored. The textual support for the inclusion of second-order assumptions – the claim that “there is the same need of an infallible judge of opinions to decide an opinion to be noxious, as to decide it to be false, unless the opinion condemned has full opportunity of defending itself” – indicates to me that Mill did hold this idea, and I think the argument does not make sense without it.

Without this idea, the argument does not make sense against a government that makes a “second-order assumption of infallibility” about some belief and does not make any attempt to silence contradiction of the second-order belief, such as one that silences denial of the Holocaust, believing that it is useful to do so, but does not silence discussion about whether it is useful to censor denial of the Holocaust – an example discussed at the end of the paper. The government is then acting upon an opinion which everyone has “complete liberty of contradicting and disproving”, and is therefore justified. Unless they really do not have this liberty if the first-order opinion is being censored, it then follows that one kind of “assumption of infallibility”[, the “second-order” kind,] does not make the mistake of denying human fallibility, which was supposed to be the core of the argument, and the reason for the name.

At the time, I felt strongly enough to also e-mail Alexander Brown about it; understandably, I got no reply.

My opinion about the argument

I still think that the argument, as it stands in my old presentation of it, is valid, and possibly sound. The most contentious premise, of course, is the statement #2, which is Mill’s interpretation of what human fallibility means. I accept it, and the rest of the argument, in a limited sense, as follows.

A belief which is “assumed true for the purposes of action” is almost always a probable judgment, i.e., not fully certain. Such a probable judgment, by its nature, does not represent certain knowledge of its object, and may be improved in that respect. So, I would act against the improvement of my own knowledge if I were to deny to others, who may even possibly have superior knowledge about the matter, the possibility of correcting me about it. Acting in this way would seem to be morally wrong, since knowledge is the perfection of man’s rational nature. I could still have no such qualms, however, about silencing a logically incoherent opinion, since pure reason cannot fail, and it is impossible to practically live as though it could.

I think that Catholics may agree with the entire original argument and still support such venerable Church institutions as the execution of heretics, and the Index of Forbidden Books. This may be done if they interpret the Church’s dogmas as being not “one of [their own] opinions”, as premise #2 requires, but rather one of God’s opinions, which are always infallibly correct, and may be “assumed true for the purposes of action” even if no one may discuss them. The Pope does not irrationally assume his own infallibility, but rather accepts it on faith in God, who cannot lie or make mistakes.

Censorship of all opinions other than dogmas would still remain proscribed by the argument. It is to be noted that the argument is concerned, in any event, with the censorship of opinions, and such as may be acted upon at that, so that it does not seem that the argument could ever be used against the censorship of, say, offensive language, or raunchy fiction. This seems to constitute the bulk of censorship, and with the censorship of heresy also excluded, the argument is left with a very limited practical scope.