I rarely agree with anything that NR’s John O’Sullivan and Mark Krikorian say about immigration, but they have set just the right standard for the debate: realistic or utopian. Which view is more realistic, theirs or mine?

The O’Sullivan-Krikorian stance, as argued in the September 12 issue, holds that we can seal our borders, do without immigrants, and easily drive the 11 million illegal immigrants already here to leave the country by enforcing existing laws. I believe that immigration is driven by market forces, and that by and large it is good for the U.S. Our challenge today is to find a way to control it - not pretend that it doesn’t exist or that we can simply wish it away.

Let’s start with the economy. I think that those who say we don’t need immigrants have their heads buried in the sand. Their claim may have been true 45 years ago when half of all American men did not finish high school and became unskilled laborers. But today less than 10 percent of American men drop out of high school - and, paradoxically, our increasingly automated service economy is more dependent than ever on low-skilled workers. The baby-boom generation is retiring from the workforce, and native-born fertility rates are in decline. The problem isn’t just that Americans don’t want to work out in the fields or up on roofs in the hot sun; employers can’t pay them enough to make that kind of job worthwhile for most people. The real problem is that there aren’t enough native-born workers to sustain the industries that rely on unskilled labor - from agriculture to food-processing to health care to construction - much less fuel the growth of these sectors, or related ones. John O’Sullivan calls me unrealistic, but his denial of this basic modern fact of life strikes me as the ultimate in irresponsible wishful thinking.

Then there’s the fantasy of dramatically reducing the number of illegal immigrants who enter the country each year. It’s not physically impossible - I’ll grant that. If I thought, as National Review does, that immigration was destroying the fabric of American life - if I believed that most of today’s newcomers were not assimilating and never would, and were instead congealing into a permanent underclass - then I too would favor forgoing the economic growth immigrants make possible. And I would also favor creating the virtual police state that would be necessary to reduce immigrant numbers in the dramatic way O’Sullivan and Krikorian advocate. But today’s newcomers are assimilating - learning English as fast as or faster than the immigrants of a hundred years ago, moving up the economic ladder, intermarrying at unprecedented rates.

And make no mistake about the kinds of draconian measures that would be required to stop the flow: a 2,000-mile wall on the border, maybe an electric fence, armed immigration agents empowered to stop and interrogate anyone at anytime, a national ID card, regular raids in American workplaces, barrio sweeps, highway checkpoints, and interior patrols. Remember Prohibition? It showed that we can suppress almost anything, if we’re willing to pay the price. But how likely would that be in the United States today? How realistic is it to think that most Americans would find it acceptable to live under a permanent state of siege?

That’s not to say that we can’t control immigration, or shouldn’t. On the contrary: Of course we need to retake control of our borders. Of course we need to close down the black market that currently sustains much of the American economy - 8 million illegal workers and growing. We also need to know the name of every foreigner in the country - need to know if he has a criminal record and where to find him if his name appears on an international watch-list. Most important, we need to reestablish the rule of law: to restore the credibility of our statutes and our national respect for the authority behind them. The real question isn’t if, but how. What’s the realistic strategy for retaking control? No less than the social order and our national security depends on finding a plausible, practical approach.

A realistic strategy would begin by acknowledging the fact of contemporary immigration: our labor needs, and the international supply and demand that generates a robust and continuing flow of foreign workers. Then, instead of forcing roughly a third of this flow underground, as we do today with unrealistically low immigration quotas, we would raise those ceilings to accommodate the workers we need. This isn’t just sleight of hand - making the illegal flow legal by legalizing it, as John O’Sullivan suggests. This is the only workable way to retake control: by replacing our unrealistic, unenforceable law with limits we can make stick, and then enforcing those new limits to the letter. It’s the difference between Prohibition and liquor licenses. Not only will it be much easier to enforce more honest quotas, but we will be far likelier to muster the political will to do so. Only when we break through our denial, and set our quotas more in sync with our labor needs, will we move beyond the nudge-nudge, wink-wink hypocrisy that now undermines enforcement of our immigration laws.

The new system would need teeth - real teeth - on the border and in the interior. Sens. John Cornyn and Jon Kyl weren’t far from the mark in the immigration-reform bill they introduced in July: more men, more money, more detention beds, more-aggressive prosecutions, enhanced technology, stiffer sentences, and some rethinking of the role state and local police have in enforcing immigration law. Above all - the sine qua non - we need to make it impossible for illegal immigrants to work in the U.S. And in order to do so, we must develop a mandatory electronic employment-verification system, something like point-of-sale credit-card verification - ideally by swipe-card - in every U.S. workplace. John O’Sullivan will say it won’t happen, will say I’m being unrealistic. But the fact is that something like this is proposed in every immigration bill on the table today, including the McCain-Kennedy legislation introduced earlier this summer. Even employers, most of whom would prefer a stable, legal workforce, are increasingly prepared to accept electronic verification - provided it comes with more realistic immigration quotas.

Finally, the third pillar of a more realistic approach: We can’t build a new airtight, lawful system on top of a rotten, illegal foundation. We need to do something about the 11 million illegal immigrants already in the country. Yes, of course, this is morally difficult; no one wants to reward or encourage unlawful behavior. (Not to mention, as Mark Krikorian points out, that it will be a vast, bureaucratic challenge to administer any process that allows several million people to adjust their status.) But that’s the thing about reality: It’s often difficult and intractable - and sooner or later we must face up to the reality created by 30 years of willfully blind immigration policy.

After years, and sometimes decades, in this country, many of the 11 million own homes and businesses. Others have U.S.-born children - citizens and illegal immigrants in the same family. Some will go home if prodded, as Krikorian suggests, but many - perhaps several million - would only be driven further underground by what he euphemistically labels an “attrition strategy.” Talk about make-believe: Even if it worked, attrition would be a disaster, all but crippling the economy. But it almost assuredly won’t work. Most of the 11 million would remain in the U.S., as illegal as ever and as much of a threat to our security. The only people who would benefit would be the smugglers and forgers whose intensified help the immigrants would need to stay ahead of the law.

John O’Sullivan and Mark Krikorian are right to warn us. The worst mistake we can make is to pursue an unrealistic policy. The choice is ours: We can retake control of the border and restore order in our communities. Or we can continue with our heads in the sand, denying reality and paying the price - eroding the rule of law and endangering our security.

MARK KRIKORIAN REPLIES: Tamar Jacoby is ubiquitous - from the Wall Street Journal and Washington Post to The Weekly Standard and New Republic, she has been the most enterprising and persuasive booster for an amnesty allowing 11 million illegal aliens to remain in the United States.

But she hasn’t been persuasive enough.

Let’s start, as she did, with the economy. She says that those of us who dispute the need for immigrant labor have our heads in the sand. If she means that our economy has managed to accommodate 11 million illegal aliens, and that their magical, overnight disappearance would be disruptive, then of course she’s right. But that’s not what she has in mind. Instead, her contention is that America is running out of a vital resource - high-school dropouts who become unskilled laborers - and thus needs to keep importing workers “to sustain the industries that rely on unskilled labor.”

When employers say this, we all understand that they’re seeking government-approved access to cheap labor outside the regular system, a government “subsidy” for their low-tech, labor-intensive operations. For instance, California tomato farmers testified in the 1960s that “the use of braceros [Mexican guestworkers] is absolutely essential to the survival of the tomato industry.” But that labor program was ended anyway, and illegal immigrants did not immediately pick up the slack - so the farmers concluded that their investment in lobbying hadn’t paid off, and instead they invested in harvest machinery. The result: a quadrupling of production over the following 30 years, and a drop in the post-inflation retail price of tomato products.

In fact, no economist has ever discovered a significant net economic benefit from immigration. Perhaps she knows something they don’t, but I suspect that Jacoby is just acting out the old economist joke about assuming something that doesn’t exist.

Her description of what would be needed to actually enforce current immigration law is just as inaccurate. She imagines a “permanent state of siege” requiring “draconian measures.” In reality, the consistent application of perfectly ordinary law-enforcement tools would be enough: an end to government’s winking at illegal employment; a requirement that new hires be verified through the simple online tool that the immigration service already provides (and that I use at my immigration think tank); streamlined deportation rules; Israeli-style fencing along the parts of the border most often used by illegal immigrants. This will not make the problem vanish overnight. It is instead a genuinely conservative approach to a long-neglected problem, an approach that allows us through attrition to shrink the number of illegals over time.

As in all things, we would eventually reach a point of diminishing returns, where further reductions in the number of illegals might require police-state tactics. Then Jacoby’s amnesty would at least be a legitimate topic for discussion. But amnesty shouldn’t even be on the table until after the authorities have demonstrated a commitment to enforce the law - a commitment that is almost entirely lacking in this administration.

Our immigration strategy should be clear: enforcement first.

JOHN O’SULLIVAN REPLIES: Start with the contradictions. In paragraph five, Ms. Jacoby warns against my “fantasy” of reducing illegal immigration on the grounds that it would require “a 2,000-mile wall on the border, maybe an electric fence, armed immigration agents empowered to stop and interrogate anyone at any time, a national ID card, regular raids in American workplaces, barrio sweeps, highway checkpoints, and interior patrols” - and would not work even then.

By paragraph six, however, she is insisting that her own policy would entail “clos[ing] down the black market that currently sustains much of the American economy - 8 million illegal workers and growing.” And when she reaches paragraph eight, she concedes this will require “more men, more money, more detention beds, more-aggressive prosecutions, enhanced technology, stiffer sentences, and some rethinking of the role state and local police have in enforcing immigration law. Above all - the sine qua non - we need to make it impossible for illegal immigrants to work in the U.S. And in order to do so, we must develop a mandatory electronic employment-verification system . . . in every U.S. workplace.”

Well, it’s a lady’s privilege to change her mind - but in the space of four paragraphs? Let me ignore that her description of the “attrition” method of reducing illegal immigration is a travesty, as Mark Krikorian establishes in his response. But even if it were accurate, why would her “draconian” methods work when mine would fail? Her explanation is that she would admit more legal immigrants in line with our economy’s “needs.” This isn’t just “making the illegal flow legal by legalizing it,” she assures us (though it is plainly that), but a balanced strategy. Having eliminated the “need” for illegal workers, she would then enjoy public and employer support for rounding up the few remaining ones.

Alas, this argument is another example of the “economics without prices” in which Ms. Jacoby specializes. Her newly legal workers would be entitled to the minimum wage and subject to Social Security and other tax obligations. That would price many of them out of the “black market” jobs they now do. Either those jobs would go undone - Ms. Jacoby’s main charge against Mr. Krikorian and me - or they would continue to be performed by illegal immigrants at lower prices. Almost certainly the latter.

Ms. Jacoby would therefore find herself employing draconian methods against an illegal immigrant workforce almost as large as now - and possibly larger since she has yet again ignored our experience that legal immigration acts as a magnet, not as a substitute, for the illegal kind. At the same time she (and President Bush) propose the mammoth bureaucratic tasks of legalizing the 11 million illegals already here and managing a much larger legal inflow. It is a recipe not for a more orderly job market but for maximizing chaos.

Interestingly, in her advocacy of the Bush immigration “reforms,” Ms. Jacoby does not claim that they will make native-born Americans better off. Nor even raise per capita income in the U.S. Nor improve the income and job prospects of low-paid Americans already badly harmed by mass immigration. Nor hold down the high and rising fiscal costs of illegal immigration.

These are not unimportant.

Tamar Jacoby is a senior fellow at the Manhattan Institute.