Choose a weapon
Fudged data. False identities. No holds are barred in the academic duel over guns and violence
John Lott Jr. is accustomed to criticism. When his book, More Guns, Less Crime, was published in 1998, the American Enterprise Institute scholar was fiercely attacked for his provocative argument--that more guns in the hands of civilians would reduce crime. Indeed, he has been accused of shilling for gun manufacturers, tweaking his data, and even conjuring up numbers out of thin air. But through it all, he has had Mary Rosh.
Until now, that is. Two weeks ago, a resourceful Weblogger named Julian Sanchez decided to double-check the identity of Rosh, one of Lott's most passionate online defenders. She claimed to be a former student of Lott--indeed Lott was "the best professor that [she] ever had." Rosh even posted a glowing review of Lott's book on Amazon.com. But when Sanchez looked deeper, he found that the "only purpose to [Rosh's] existence seemed to be to defend John Lott." A bit more detective work revealed that Lott and Rosh were one and the same. "It was a foolish thing to do," Lott now concedes, "but I get attacked a lot and I don't want to spend all of my time defending myself."
Welcome to the academic world of gun research, which can resemble a hall of mirrors at times. Indeed, Lott is not the first in this small scholarly subspecialty to attract controversy. Just last year, Emory University historian Michael Bellesiles resigned after accusations of fraud in his book, "Arming America," which challenged the idea that American culture has always been gun-oriented. Bellesiles was thoroughly discredited for distorting data and was eventually stripped of the prestigious Bancroft Prize.
Unlike Bellesiles, Lott is at least considered a credible researcher and a fine scholar, even by critics who think he is wrong. At the heart of Lott's argument is a data set of unparalleled size and scope. He has cobbled together crime figures for all 3,054 U.S. counties from 1977 to 1992. His provocative conclusion: Guns make the world safer. "If the rest of the country had adopted right-to-carry concealed-handgun provisions in 1992," he maintains, "about 1,500 murders and 4,000 rapes would have been avoided." His research shows these guns do not noticeably increase the number of accidental deaths--in 1999, 88 kids under age 15 were accidentally killed by firearms--and points out that permit-holding gun owners hardly ever commit crimes with their weapons.
Many academics come to a different conclusion. "There's no doubt Lott deserves a lot of credit for helping the world see that the effects of these [concealed-weapons] laws are not as harmful . . . as many people had feared," says John Donohue, a law professor at Stanford University who recently published a study critical of Lott's findings. But five more years of data--including an additional 13 states with concealed-weapons laws--have put Lott's conclusions to the test. Using Lott's same analytical methods, Donohue argues in a new book, "Evaluating Gun Policy," that it's now possible to see these laws as increasing crime.
Quick to fire. Other academics agree with Donohue that Lott is sometimes hasty in his conclusions. Even his supporters acknowledge that a data set of this size--with so many unstable and conflicting variables--cannot produce such dramatic and clear-cut conclusions. His findings, says Daniel Polsby, associate dean at George Mason University, spring from "a unique combination of [Lott's] abnormal psychology and quite conventional econometric overclaiming."