FOR THE SECOND TIME NOW, Michael Bellesiles, a historian at Emory University, is being accused of having relied on missing or nonexistent records for evidence in his Bancroft prize-winning book “Arming America.” And, for the second time, Bellesiles’s protestations and explanations have failed to vindicate his book, which sought to refute the “myth” of widespread gun ownership in antebellum America. Elementary fact-checking has destroyed the credibility of Bellesiles’s story of how he discovered various 18th century Vermont court records others say don’t exist.
The documents in question are Vermont court records from the late 18th century. The key passage appears on page 353 of “Arming America”: “During Vermont’s frontier period, from 1760 to 1790, there were five reported murders (excluding those deaths in the American Revolution), and three of those were politically motivated.” The endnote for this finding refers the reader to Superior Court records at the county courthouse in Rutland, Vermont. But as Ohio State University historian Randolph Roth has pointed out (and the court clerk in Rutland has confirmed), the volumes for 1782 to 1790 are not in the Rutland court’s holdings. Furthermore, the Superior Court did not exist before 1778, so it has no records for the period 1760 to 1777.
Bellesiles did not respond to this particular charge when it was raised in a refereed exchange in the spring issue of the William and Mary Quarterly. More recently, however, he went on the counterattack during a written exchange hosted by the History News Network, a non-specialist website. These records, Bellesiles wrote, “were missing until I found them while working on my dissertation in 1984. . . . I found those records in the abandoned jailhouse in Newfane. Those records and many others were stacked in boxes in the jail cells and I spent the next four months taking notes from those sources in unheated rooms by the light of a Coleman lantern. I used those records at length in my book ‘Revolutionary Outlaws,’ as Professor Roth knew, since he reviewed that book. The records are now in the Newfane Historical Society and their existence can be verified by Vermont’s State Archivist, Gregory Sanford.”
Reached for comment via e-mail, Sanford says he has “neither physical nor legal custody over Vermont’s court records” and that he lacks “in-depth knowledge of their content.” He could confirm that Bellesiles visited Newfane in the “mid-1980s” and that Bellesiles brought to his attention “the deplorable condition of the Windham County Court records.” And for this, Sanford says, “I owe Dr. Bellesiles a debt of gratitude.” Although the records in Newfane were a mess, according to Sanford, the state eventually got around to making an inventory. Fortunately, Sanford has a copy.
The only Supreme Court records listed in the inventory–remember it is Supreme Court records Roth says Bellesiles could not have read–begin in the year 1794, so they could not have been used to support any claim regarding the years 1760 to 1790. In fact none of the records listed in the inventory cover this exact period of time. The earliest of the records go back to 1766 and are County Clerk records, which would not tell us about murders, let alone how many had taken place throughout Vermont.
The inventory also lists Marlboro county probate records, beginning in 1784. Again, 1784 is too late; and probate records don’t tell us about murders; and Marlboro is only one of several counties that make up Vermont. Last, the inventory lists County Court records, starting in 1771, which again would only tell us about one county, and could not say anything about murder or anything else for the years prior to 1771.
The most obvious problem with Bellesiles’s latest claim is that neither the town of Newfane nor the surrounding county, Windham, appears in the original endnote of “Arming America,” suggesting that Bellesiles decided to manipulate this lavishly detailed account into a cover story when Roth went public with the accusation that Bellesiles was citing nonexistent records. Another question raised by Bellesiles’s statements is: If not Supreme Court records (also called Superior Court records), then what did Bellesiles “use at length in [his] earlier book ‘Revolutionary Outlaws'”?
A copy of that book, obtained at the Library of Congress, shows that Bellesiles did have some property and tax records for Newfane and the county of Windham (known earlier as Cumberland), both of which he lists in the Appendix. This would be consistent, incidentally, with the inventory supplied by Gregory Sanford. Bellesiles used many such records to document economic activity across Vermont towns and counties during the late 18th century. But his discussion went nowhere near any numbers about murders or crimes in general.
Curiously, ‘Revolutionary Outlaws’ does contain a handful of citations referring to individual Supreme Court records after 1782. But several of these are relatively well-known cases for which there would have been other forms of documentation, and the latest of these is dated 1785. None of these citations, however, mentions the “Newfane Historical Society” (which Bellesiles may have confused with the Windham Historical Society) as a location for the document. In fact, although Bellesiles takes the trouble to explain certain archival wrinkles in his appendix (like the fact that he found Marlboro county records in Windham county), he makes no mention of Supreme Court records found anywhere but in Rutland.
So, even if Bellesiles did find pre-1790 Superior Court Records for years other than 1778 to 1782–records that do not exist in Rutland county where he said they reside; records that are not listed in the Windham county inventory held by Gregory Sanford; records he did not refer to in “Arming America” or list in the long and detailed appendix of “Revolutionary Outlaws”–such records would still fail to cover the years mentioned in his original claim.
Bellesiles and his critics have been down this road before. Last year, a scholar investigating Bellesiles’s evidence discovered that San Francisco probate records Bellesiles had cited in “Arming America” had been destroyed in the earthquake of 1906. Bellesiles responded by saying he couldn’t remember where he saw those records, though it may have been one of several archival institutions in the San Francisco area. Journalists and scholars checking with those institutions–starting with the San Francisco Supreme Court, which Bellesiles cited in his book, and ending with the Contra Costa County Historical Society, the last of several places where he claimed to have seen probate records–discovered that such records were not in any of these archives.
Recently Emory University announced it was commissioning outside experts to investigate claims of widespread academic fraud in the Bellesiles case. Meanwhile, the National Endowment for the Humanities, which has been funding Bellesiles’s research on guns in America through the Newberry Library in Chicago, has demanded its own review of Bellesiles’s work.
David Skinner is an assistant managing editor at The Weekly Standard. His last piece on Michael Bellesiles, The Historian Who Couldn’t Shoot Straight, appeared in the February 25, 2002 issue.
