New doubts about gun historian

Research to receive hard critique today

By David Mehegan, Globe Staff, 9/11/2001

When Emory University historian Michael A. Bellesiles published his sweeping historical study of guns in Colonial America last fall, the reaction was electric.

His thesis that guns were relatively rare in Colonial households, and that the American ''gun culture'' didn't take hold until long after the Founding Fathers drafted the Second Amendment's ''right to bear arms,'' was immediately hailed by gun control advocates and by a host of historians impressed by his bold rewriting of conventional wisdom.

But even as publication of ''Arming America: The Origins of a National Gun Culture'' won Bellesiles plaudits - and, in April, Columbia University's prestigious Bancroft Prize for historical excellence - some of his academic doubters were poring over evidence Bellesiles cited and finding multiple instances in which he seems to have misused historical records.

Today, at Harvard Law School, Bellesiles's most adamant critic, Northwestern University law professor James Lindgren, plans to detail evidence that Bellesiles may have stretched or distorted the historical record in trying to prove his claim.

The Boston Globe has reviewed substantial portions of records Lindgren will cite: 18th-century probate records in Vermont and Rhode Island. The Globe has also checked into Bellesiles's claim to have studied certain records in San Francisco, records county officials say were destroyed by fire in 1906. In each case, the records appear to support Lindgren's accusation and suggest a disturbing pattern of misuse of data by Bellesiles in his book and in an article defending his thesis which he published on his Web site.

In telephone and e-mail interviews, Bellesiles stands by his research. ''I spent 10 years of my life traveling around to archives myself, without research assistance. I know how much work I put into it, and I stand by it.''

He does, however, concede that he apparently made an ''egregious error'' in his interpretation of some Vermont probate records cited in the Web-site essay. His transcriptions of those records repeatedly characterize weapons as ''old'' or ''rusty'' or ''broken.'' But the records themselves show no such notations.

Published by Alfred A. Knopf last September, ''Arming America'' drew immediate notice for its startling, and apparently copiously documented, finding that, contrary to common belief, ''gun ownership was exceptional in the seventeenth, eighteenth, and early nineteenth century, even on the frontier ... The gun culture grew with the gun industry.''

Bellesiles claimed to have examined more than 11,000 probate records of more than 1,200 counties, counting the number of guns listed in estate inventories. He found that between 1765 and 1821, not more than 17 percent of estate inventories listed guns. The gun ownership rate was even lower in the 1760-1795 period - about 14 percent, he said. And ''over half of these guns were listed as broken or otherwise defective,'' Bellesiles wrote.

Indeed, he wrote, one reason the Revolutionary War went on as long as it did may have been that the weapons available to the colonists were so scarce and in such poor repair. ''Probably the major reason,'' he argues in the book, ''why the American Revolution lasted eight years, longer than any war in American history before Vietnam, was that when that brave patriot reached above the mantel, he pulled down a rusty, decaying, unusable musket (not a rifle), or found no gun there at all.''

''Bellesiles has dispersed the darkness that covered the gun's early history in America,'' wrote historian Garry Wills in the New York Times Book Review. ''He provides overwhelming evidence that our view of the gun is as deep a superstition as any that affected Native Americans in the 17th Century.''

Firing back in a letter to the editor, National Rifle Association president Charlton Heston chided Wills for accepting ''Bellesiles's ludicrous argument,'' and the book has been denounced on gun-owner Web sites and by conservative reviewers in the Wall Street Journal and elsewhere.

After the book came out, Bellesiles reported a campaign of harassment, including abusive phone calls and what he said was a pattern of viruses sent to his computer. His Emory phone now refers callers to a mailing address, and his e-mail address, which he does not give out, is coded. (He is currently on a fellowship at the Newberry Library in Chicago, a historical research library.)

Some of the reaction has been so vociferous that the American Historial Association adopted a statement in June deploring personal attacks on Bellesiles.

Besides its fiery assault on America's present-day gun culture, the book was a lightning rod because of its potential to force a rethinking of the intent of the Second Amendment: ''A well-regulated militia, being necessary to the security of a free state, the right of the people to keep and bear arms, shall not be infringed.''

If, contrary to the familiar image of the sturdy yeoman with his trusty flintlock, few Americans had actually owned guns, it could be, as some gun-control advocates argue, that the amendment was never meant to apply to individuals.

The Bellesiles controversy also divides academic historians, for whom the Bancroft Prize is a singular honor. Past winners include such luminaries as Arthur Schlesinger Jr. and Allan Nevins.

''I have been a defender of the book,'' says Stanford historian Jack Rakove, who has written extensively on the Colonial period. ''It makes a number of arguments which one would have to challenge comprehensively in order to undermine its thesis.''

But Bentley College historian Joyce Malcolm, a Second Amendment specialist, says, ''The more I looked at it, the more disturbed I became. All historians can make mistakes and differ on interpretation, but in his case it's not just interpretation, or one or two points, but matters of fact and repeatedly.''

Lindgren, a specialist in probate law and statistical analysis (and a believer, he says, in gun control), became suspicious of Bellesiles's findings early on and began posting his objections on history discussion sites. He looked over some of Bellesiles's sources, and eventually wrote the academic paper, ''Counting Guns in Early America,'' which he will present today at Harvard and later at other institutions. The paper argues, among other things, that Bellesiles's data are grossly in error and that some of his conclusions are mathematically impossible. Lindgren also says that when he contacted Bellesiles, trying to get him to produce the details of his research, Bellesiles was unable to do so.

''In virtually every part of the book examined in detail,'' Lindgren told the Globe, ''there are problems ... An enormous number of people have become cautious. It's clear that this book is impressive to legal and social historians who do not check the background. Law professors and quantitative historians have been suspicious about the book since its release.''

Bellesiles says that he kept all his probate findings on yellow legal pads and that they were destroyed when a water pipe broke and flooded the history department offices at Emory last April, while he was in England. (There was a flood, an Emory spokeswoman says, and many history faculty lost books and papers. The spokesman could not say whether Bellesiles papers were among those lost.)

Lindgren also charges that Bellesiles could not have reviewed probate records in San Francisco for the 1840s and 1850s, as he claims to have done in his book and on his Web site, because all such records were destroyed in the 1906 earthquake and fire. Bellesiles's Web-site article, ''Probate Records as an Historical Source,'' lists the San Francisco Superior Court as the site where he did his research.

According to Ida Wong, deputy clerk of the San Francisco Superior Court, contacted by the Globe yesterday, ''All that we have here is 1907 and after. Everything before that was destroyed.'' Asked where surviving records might be, Wong said, ''I would not know where to refer you.''

In an interview, Bellesiles said he can't remember exactly where he did the California research, that he remembers going to the courthouse, but might have done it at the Bancroft Library at the University of California. He said he recalls finding 12 such probate files. Anthony Bliss, curator of rare manuscripts at the Bancroft, says it is possible there might be some lists in its collections, but could not say for certain. Certainly, he said, the great preponderance of such records were lost to the fire.

Serious questions have also been raised about an article Bellesiles posted on his Web site called ''Men with Guns'' which seeks to buttress the findings of his book. In it, Bellesiles discusses some Vermont probate files which list gun ownership. Lindgren alleges that Bellesiles's list misrepresents the content of the originals. A Globe examination last week of original records in the Rutland, Vt., probate court for the 1770s and 80s shows that Lindgren is apparently correct.

Six of many similar examples:

Bellesiles version: ''Cotton Fletcher, broken gun 6s [six shillings]"

The original: '' a gun @ 6 shillings.''

Bellesiles: ''Isaac Cushman, old gun 12 s.''

The original: ''one gun barrel and stock, 12 s.''

Bellesiles: ''Samuel Crippin, old gun 10 s.''

The original: ''one gun @ 10 s.''

Bellesiles: ''Asher Culver, 2 old guns.''

The original: ''firearm.''

Bellesiles: ''Jonathan Mayo, broken gun 6 s.''

The original: ''1 lb. gunpowder 6 s., 3 lbs leads 3 s.''

Bellesiles: ''Abel Moulton, 5 muskets, some old, two [pounds], 8 s.''

The original: ''Fire Arms, 2 [pounds] 8 s.''

When asked about the discrepancies, Bellesiles said he was mystified, ''I don't know. I am very upset about that. It's a mystery to me. I might have looked at a different record book. It's an egregious error on my part.''

Separately, in his review of Rhode Island records, Bellesiles writes in his book that of 186 estates of ''property-owning adult males'' in Colonial Providence, only 90 listed guns, and ''more than half of these guns are evaluated as old and of poor quality.''

Lindgren found that 17 of the estates were not of men but women. He also found that among 153 males whose estates included inventories, 94 mentions guns. But only nine of those are listed as old or in disrepair.

A Globe review of some of the Providence records, on file at Boston Athenaeum, appears to confirm Lindgren's findings. There were many estates of women among those Bellesiles cites, and few indicated guns in poor condition.

Asked why he characterized the guns that way, Bellesiles said that the low prices paid for the guns at auction indicate their poor quality.

Historian Alan Brinkley, chairman of the history department of Columbia, says no such questions have ever been raised about a Bancroft winner before, and since a new committee is convened each year to choose the winners, and dissolved afterward, there would be no clear way to reconsider a winner. As chairman, Brinkley was the presenter at the Bancroft ceremony, though he says he has not read Bellesiles's book. (The Bancroft carries a $4,000 cash prize.)

Brinkley stressed that what Bellesiles put on his Web site has no relevance to what is in the book. ''I don't think that a book prize would be rescinded on the basis of information on a Web site. A book is a book and needs to be judged on its own.'' Brinkley added, ''There is a difference between error and scholarly fraud. There are few books in which there are no errors. Any book that people set out to examine as this one has been would be found to have errors in it. Whether in this case they go beyond inadvertance and carelessness, I have no idea.''

Another leading historian said he finds the episode deeply troubling.

''There are many questions raised about his use of probate records and other materials,'' says Brandeis historian David Hackett Fischer, an authority on early America. ''They are very serious criticisms. It cuts to the very foundation of what he reports, and convincing answers are not coming from him.''