Libertarians are properly skeptical of claims about government competence, despite the successful record Berman cites. However, one can explain the record without suddenly believing the government got something right. Considering that refugee applicants must undergo this high degree of scrutiny, the refugee program is unlikely to be the preferred route into the United States for would-be infiltrators; they would probably calculate they’re more likely to get caught applying for refugee status than by trying to accomplish their objectives in some other way.

“If ISIS wanted to attack the United States,” the Niskanen Center notes, “it need only dispatch one of the many foreign fighters who have come from the U.S. or the E.U. to do so”—or, as David Friedman reminds us, it could send some tourists. Actually, it would need only to mobilize so-called radicalized American citizens over the social networks. ISIS “would not need to attempt a two to three-year mission with a very low probability of success.”

Thus the government most likely succeeds in weeding out terrorists because they don’t present themselves to the authorities in the first place.

Of course the government’s role in scrutinizing refugees makes (most?) libertarians uncomfortable. (For one thing, it’s tax-financed, though it need not be.) But that’s the way it’s going to be for the foreseeable future. Nevertheless, we can take up the question of how a completely libertarian—by that I mean stateless—society would handle this matter. Would that society be vulnerable to people who mean it harm? I doubt it. Freed people are innovative, flexible, and entrepreneurial. We can be confident that a free society would devise methods of joint suretyship by which strangers could be vouched for, giving others confidence in dealing with them safely. In fact such mechanisms were devised long ago and would quickly be updated to be fully consistent with individual rights if the state were to leave the field.