I agree with Ramesh. The toughest thing for me in this debate is that we should want an accurate accounting of everyone who is in the country. But, as Ramesh notes (and as I’ve acknowledged before), you can’t get that without offering illegals some kind of legal status. I could swallow hard and go along with that if (a) the problem were not so large (you can’t responsibly do it with 12 to 20 million people) and (b) I were satisfied that the government was both capable of enforcing and determined to enforce the immigration laws — I don’t think either has been demonstrated, and I doubt that either could be. (For example, no matter how earnest and competent I think Mike Chertoff is, and I know him to be top-shelf on both counts, he cannot convince me that a DHS under a President Hillary Clinton would enforce the laws when he is gone.) Since the problem is too large, and offering legal status is likely to make it worse, that means the cost is too high to justify the benefit of accounting for everyone who is here — no one is saying that wouldn’t be a real benefit; it’s just not one we can afford.

The main reason I favor the attrition strategy (see, e.g., here) is that, contrary to comprehensive reform, it recognizes our immigration situation for what it essentially is: a crime problem. Crime problems are not “solved,” they are managed. Why do we object to treating terrorism as if it were a crime problem? Because when something is regarded as merely a law-enforcement matter, that implicitly means we tolerate some instances of it. We never round everyone up. It is not desirable or possible to eradicate all instances of crime — a government that tried such a thing would have to intrude over-bearingly on the freedom of innocent people. We only tolerate such crisis-time intrusions when dealing with something like terrorism. Terrorism has to be stopped because the cost of tolerating even some few instances of it is too high. But drug trafficking and fraud, to cite two common, contrary examples, are not in that category. They, like immigration, are problems we manage. We don’t pretend to arrest every offense; we enforce aggressively and strategically enough to deal with the most serious offenders and discourage the rest. The enforcement is modulated until the level of crime is tolerable.

Regarding immigration like a crime problem is the succinct answer to the comprehensive-reform crowd’s best rhetorical device: The searing question, What are you going to do about 12 million people? The answer is: The same thing I do about the millions of drug felonies that happen yearly.

I don’t expect to deal with every one, much as I might like to. I don’t have the resources and the public would not support such a sweeping program. So, I am going to go after the low-hanging fruit (smugglers and felons who are apprehended by federal state and local authorities violating criminal laws), I am going to deal aggressively with the root of the problem (the employers who knowingly hire illegals — if I had my druthers, I would also address access to public welfare services and too-easy citizenship qualification, but that is probably too much to ask). And I am going to strengthen my border and external enforcement to try to prevent the flow from getting here in the first place.

But as for the rest, I am going to tolerate their law-breaking — not approve it or give it amnesty, but grudgingly tolerate it — as I do with the gazillion instances of people who use illegal drugs. I am going to make life difficult for them; I am going to round up the occasional bunch of them so they don’t get too comfortable in their law-breaking; and I am going to encourage other measures (analogous to drug testing in schools and certain industries) which, short of arrest, discourage the behavior. But I am going to realistically accept that I can’t, and don’t want to, round-up every offender. Illegal immigration is a crime problem to be managed and reduced, not exterminated.