My friend Tom is running, possibly for his life. It is a sweet summer evening in San Jose, and he and a colleague have just left work and are walking through a dicey neighborhood when they catch the eye of some young men, as many as 20 of them, sitting around an old car in a driveway. "Hey, you faggots!'' one of the young men shouts. Tom and his colleague walk past, quickly, but their persecutors rise like a flock of gulls and follow, shouting taunts and threats: "When we're done with you, they'll find your bodies!'' The two pick up the pace and the men come after them. "Run,'' says Tom, but the gang breaks into pursuit while Tom, trying to hold the pace, gropes in his backpack. The two reach a streetlight and there, where everybody can see, Tom suddenly stops, turns, and levels a semiautomatic handgun.


At that point, the young men chasing my friend lost their enthusiasm for blood sport. Tom and his colleague left the neighborhood as fast as they could. And if there had been no gun? "There's no question in my mind,'' says Tom, "that my friend and I would have been at least very seriously beaten, and maybe killed.'' I asked how the gang reacted to the gun. Tom says their leader demanded officiously: "Have you got a permit for that?''

Tom didn't have a permit, which is bad--but then he probably couldn't have gotten one if he had tried, which is also bad. California is among the states where, if you want permission to carry a concealed weapon, you have to prove that you are of "good moral character'' and that you have some special reason to carry. Tom could have shown that he was of good character, but he had no special reason. Until, of course, the reason arose one summer night.

As recently as a dozen years ago, almost every state was like California. Today, by contrast, almost half of all Americans live in the 31 states with so-called "shall issue'' laws, which require the authorities to approve a permit for (typically) anyone over 21 who is mentally sound, has no criminal record, pays a fee, and takes a gun-safety course. Florida began the stampede in 1987. Before then, about 17,000 people in the state had concealed-weapon permits; today, about 250,000 do. The Daytona Beach News-Journal notes that a number of local judges have permits. "I became convinced that some of these people might become dangerous,'' one judge told the paper. "So I took a firearms course, got the permit, and kept a weapon handy in the courthouse.''

Any time now, a natioanl majority will live in "shall issue'' states. Colorado, Missouri (with a referendum in April), Nebraska, and Ohio (with a new Republican governor) are all candidates to switch to "shall issue'' this year, with the noisy support of theNationalRifle Association. Grover G. Norquist, a conservative activist who expects to join the NRA board later this year, busies himself lobbying for concealed-carry when he is not busy lobbying for lower taxes--both issues, he says, being sides of the same conservative coin. "The more people who view themselves as independent of the state,'' he says, "the more people who are available to the center-right coalition.''

In effect, and to the horror of many liberals and gun-control groups, America is now running an uncontrolled nationalexperiment in self-policing. There are something like 70 million handguns in the country, and the odds have increased dramatically that today you passed one of them in the street.

So what do we know about the results of the experiment? First, that about 1 percent to 5 percent of a state's population typically take out concealed-gun permits. Second, we know what does not happen: America does not turn into the Dodge City of myth, with fender-benders becoming hailstorms of lead. (Actually, Dodge City, Kan., wasn't the Dodge City of myth. It was much safer than today's Washington, D.C., with homicides running to one or two per cattle-trading season and marshals mostly concerned, writes the historian Roger Lane, "with arresting drunks and other misdemeanants.'') People who are willing to register with the sheriff, pay a fee, and take a gun-safety course turn out to be unusually law-abiding, safer even than off-duty cops. In Florida, only 0.13 percent of concealed-carry licenses were revoked from 1987-97 for criminal activity of any sort. No murders seem to have been committed by people carrying licensed guns in public. The "shall issue'' law may actually deter bad behavior, since if you get into any sort of trouble you can lose your license.

What is harder to say is what we all want to know. Do concealed weapons, lawfully carried, reduce crime, or increase it?

The most comprehensive study of the subject is also the most controversial. John R. Lott Jr., an economist at the University of Chicago Law School, assembled data for all 3,054 U.S. counties over 18 years (1977 through 1994) and controlled for all sorts of variables. "This study uses the most comprehensive set of control variables yet used in a study of crime, let alone any previous study on gun control,'' he writes in his book More Guns, Less Crime, published last year by the University of Chicago Press.

The title gives away the punchline. ``When state concealed-handgun laws went into effect in a county, murders fell by about 8 percent, rapes fell by 5 percent, and aggravated assaults fell by 7 percent.'' (Note: This is after controlling for other variables.) "On the other hand, property-crime rates increased after nondiscretionary laws were implemented. . . . Criminals respond to the threat of being shot while committing such crimes as robbery by choosing to commit less risky crimes that involve minimal contact with the victim.''

Moreover, and maybe more surprising, Lott also finds that women and blacks--who are, for different reasons, disproportionately vulnerable to violence--benefit disproportionately from "shall issue'' laws. Thus, for example, each new concealed-gun permit issued to a woman increases women's overall safety three to four times as much as a new permit to a man increases men's safety.

In 1997, two economists, Ian Ayres and Steven Levitt, found that people who install LoJack radio-tracing systems in their cars pay about $600 each for devices that, by deterring auto theft, save society more than 10 times that amount. The trick is that LoJack is hidden; the thief doesn't know which car has it (steering-wheel clubs, by contrast, just displace theft to unprotected cars). If Lott is right, concealed guns have the same sort of effect. In Oregon, he finds, each new concealed-carry permit saves the state $3,500; in Pennsylvania, $5,000. So law-abiding people who pack heat are doing the rest of us a favor.

If Lott is right. Predictably, other scholars have found all sorts of things wrong with his work. "Lott's research is so fundamentally flawed in a number of different ways that his research really can't tell us anything about what the effects of these laws are,'' says Jon Vernick, of the Center for Gun Policy and Research at the Johns Hopkins University School of Public Health. Also predictably, and somewhat annoyingly, the results of concealed-gun studies tend to coincide with the authors' predisposition toward gun control generally. Gun nuts love "shall issue.'' Gun-control nuts hate it.

This is annoying because it is so obtuse. The question of whether guns should be available readily is completely distinct, logically and empirically, from the question of who should be carrying guns around. As it happens, I've lived in two countries with strict gun laws, Japan and Great Britain, and if I could press a button and make America's guns vanish, I would do so in a blink (and I'd repeal the Second Amendment while I was at it). It turns out that a country with few guns is a better place to live than a country with, say, a fifth of a billion guns.

But the fact is that America is awash with guns, and this fact is not going to change in my lifetime, and criminals carry guns already. A rational country would make guns harder for criminals to get (that's gun control) but easier for lawful citizens to carry (that's "shall issue''). By contrast, the current policy in states such as California--easy to get, hard to carry--is perversity incarnate.

In 1999, the debate is not about whether "shall issue'' causes much harm--by now we would have heard of any mayhem--but whether it does much good. So why in the world are liberals clinging to opposition to concealed-carry? No doubt because the gun debate has been infected by the culture wars, and people have taken up sides, and liberals feel obliged to revile any proposal supported by the likes of the NRA and Grover Norquist. This is a pity, as thinking with one's knees usually is. Liberals should be on Tom's side.