We talked about one case of Mill’s libertarianism: freedom of thought and discussion.
Mill’s argument for this application of the harm principle turns on two propositions:
We began by talking about whether suppressing discussion makes sense only if those doing the suppressing presume that they are infallible.
In my opinion, the answer is no. It could make sense to suppress discussion if you think that doing so is more beneficial than not doing so, given the evidence you have at hand. That’s the standard we apply to all of our other actions. Why should this be a special case?
Perhaps it’s special because you’re shutting off information. Ordinary decisions don’t foreclose options for gathering more information. They are just taken without perfect information. Most of us are willing to revise our decisions in the light of new information. But that is the sort of thing that appears to be impossible here: no new information can come up. So you had better be sure!
I’m not completely convinced, however. Time and brain power are limited. So we only pursue so many kinds of information from so many sources. Don’t we implicitly write off lots of information? And in doing so, don’t we have to act on imperfect information?
It’s true that suppressing sources of information is more drastic than ignoring them. The latter course of action leaves the information around to be considered later. The former might not.
But doesn’t that just show that you had better have pretty good reason for suppression? The costs of toleration have to be pretty high and your evidence has to be pretty good. I don’t see why suppression requires infallibility.
After all, we do suppress some kinds of discussion. That was the thrust of the examples that Fowler gave. Not even Fowler knows for certain that the state should lock up anyone who tries to give Al Qaeda plans for a nuclear bomb. Maybe they would blow themselves up in trying to make one! No one can know for sure. But we must act with imperfect evidence and the balance of that evidence, imperfect as it is, favors suppression.
On the other hand, we didn’t really come up with an example in which it makes sense to censor an opinion as opposed to technical instructions. And, strictly speaking, Mill did only mention opinions. If a distinction between opinions and, say, information or instructions could be developed, perhaps Mill might avoid this objection.** Added April 25.
Mill claims that unfettered discussion is the best path to truth and understanding. I think Williams shows he’s wrong about that.
But I also agree with Fowler that this doesn’t fatally wound Mill’s argument. Mill used his broad claim about truth to support a narrower claim about the state: it should not hinder discussion. That narrower claim could be supported even if the broader one is abandoned.