post by Paul Kelleher
Brad DeLong has a piece in the Boston Review symposium on libertarianism. T. M. Scanlon, who wrote the symposium's lead essay, says in a follow-up that "I agree with almost everything Brad DeLong says." I am surprised to hear this.
Scanlon's essay was a methodical attempt to pick apart three potential arguments for libertarianism. In contrast, DeLong focuses not on the weakness of this or that argument, but on what he sees as a telling weakness in the people who make these arguments:
When the chips are down and libertarians are faced with the potentially monstrous consequences of overvaluing individual autonomy or rights to non-interference, they almost invariably...retreat to claiming not fiat libertaria ruat caelum ["Let (libertarian) justice be done though the heavens fall"] but rather that such situations would never arise in practice, because libertarian policies do produce the greatest good for the greatest number.
DeLong goes on to suggest that we don't need to take libertarianism seriously, because people who call themselves libertarians are systematic self-deceivers about the ancillary benefits of libertarianism or about the imagined bad effects of alternative political arrangements. In Logic 101, we teach students that this type of argumentation is called argumentum ad hominem, and that in most cases it's a logical fallacy. For instance, suppose I argue that theism is false because believers in God shockingly claim their faith grows even stronger when God permits great evil to befall them. The correct reply to my argument is that nothing about the existence of God follows from facts about the psychology of the people who believe in him. Unfortunately, this is precisely the fallacy DeLong commits in his essay.
Take DeLong's central example, the libertarian philosopher Robert Nozick. He notes that immediately after concluding that "the minimal state remains as the most extensive state that can be justified," Nozick remarks:
Our normative task in these two chapters is now complete, but perhaps something should be said about the actual operation of redistributive programs. It has often been noticed, both by proponents of laissez-faire capitalism and by radicals, that the poor in the United States are not net beneficiaries of the total government programs and interventions in the economy. Much of government regulation of industry was originated and is geared to protect the position of established firms against competition, and many programs most greatly benefit the middle class. The critics (from the right or the left) of these government programs have offered no explanation, to my knowledge, of why the middle class is the greatest net beneficiary.
Here is DeLong's appraisal of this passage:
This statement [that the poor are not net beneficiaries of redistributive policies] is empirically false...The major government programs...benefit the middle class and the poor, largely by raising those who would otherwise be poor into the middle class, and protecting the middle class from the danger of falling into poverty.
Given the passage’s empirical falsity, the interesting question is why Nozick felt that he had to make it. Weren't his arguments strong enough without it? Wasn't he happy saying that a libertarian society is a just society—even if (or rather though) it grinds the faces of the poor in the dirt and makes their lives nasty, brutish, and short? He was not. Nozick felt that his argument for libertarianism was incomplete without the empirical claim that libertarian society is good for the poor.
DeLong's diagnosis is that Nozick couldn't stomach the possibility that a libertarian state might leave the poor worse off than some more interventionist state. So Nozick deceived himself into thinking that libertarianism will always be better for the poor. But, DeLong suggests, if libertarians have to deceive themselves like this, libertarianism can't be a theory worth taking seriously.
Suppose we grant DeLong's premise about Nozick's self-deception. Even so, the argument falls apart if we can find some other libertarian (or--gasp!--two) who's willing to own up to the bad consequences that libertarianism may have for the poor. This more stalwart libertarian will simply say that Nozick's backbone was no match for his intellect. And DeLong has given us no ammunition with which to reply to this. That is the problem with ad hominem arguments.
As it turns out, I highly doubt that anything about Nozick's backbone is revealed in the passage DeLong jumps on. If you actually read the entire short section in which it appears, you get the impression that the passing remark about net beneficiaries simply provides an excuse for Nozick to address "the puzzle" of why the bottom 51% of the voters don't just "vote for redistributive policies that would greatly improve their position at the expense of the best-off 49 percent." That, of course, is actually an interesting question, and no self-deception about libertarianism is needed to see it as something worth investigating.
The great irony here is that it's the user of ad hominem arguments that comes off looking desperate. Such arguments signal that one has no substantive reply to uncomfortable conclusions, which in DeLong's case isn't actually true. But the fact that DeLong has substantive things to say about Nozick's libertarianism makes it even more curious why he is so keen on the ad hominem. So keen is he, in fact, that this is not the first ad hominem he has pressed against Nozick's libertarianism:
[A]rguments that are deployed only when they count on the side one has already pre-chosen are not really arguments. [...]
This point was brought home to me most powerfully when I discovered that the critique of Robert Nozick's Anarchy, State, and Utopia that was most effective in providing my students with the proper critical distance from the book were not any of the critiques of Nozick's philosophy...but was instead a little piece, "Anarchy, State, and Rent Control" in the New Republic...about how Robert Nozick used the Cambridge Rent Control Board to squat in Eric Segal's apartment until he had extorted $30,000. [...]
Whenever I teach "Anarchy, State, and Rent Control," odds are that the class will then have a furious argument over whether Nozick's failure to take his own libertarian principles very seriously should or should not be taken as a reason to distrust Nozick's book. I have, after much thought, decided that it should be taken as a strong reason to distrust Anarchy, State, and Utopia.
I'm sorry, but this really is a waste of valuable class time. When DeLong's students later encounter someone who is attracted to something like Nozick's libertarianism--maybe this libertarian has never even heard of Nozick--nothing constructive will come from the student's conviction that Nozick was some sort of cheat or scoundrel. If the conversation is to be constructive, it will proceed as Scanlon's lead essay does: point by point, logical move by logical move. It will proceed by avoiding ad hominem arguments.