Mr. Bellesiles’s Emails
After this article was posted, Mr. Bellesiles provided HNN with a response to Mr. Lindgren's article. Click here to read Mr. Bellesiles's response.
On the day before the editor of History News Network posted Professor Michael Bellesiles's unfortunate comments about his emails and about me, the editor asked me to write a response. I told him that I did not have the time for a general response, but I wrote this answer quickly to just one of Bellesiles's new claims--that he did not"recognize" or"acknowledge" the emails that he wrote to me. More reflective responses to some of his other novel claims will have to await a break in my scholarship (on issues not involving Arming America). FACTUAL CORRECTION OF A CLAIM BY MICHAEL BELLESILES
On August 30, 2000, Professor Michael Bellesiles wrote to me by email:The probate records are primarily on microfilm in the federal archives, though I also looked at several counties in their undigested form so as to control for wills and any earlier transfer of guns.
[ Click here to read the email.]
On September 19, 2000, Bellesiles again wrote to me by email:
The probate records are all on microfilm in the National Archives. I went to the East Point, Georgia, federal center to read these microfilms. The films are organized by county. . . .[ Click here to read the email.]Finally, all the original notes from this research are in boxes in my attic, where I moved them after they were damaged by a flood in Bowden Hall at Emory. I do not have the time right now to search for those notes and to lay out the water damaged pages for drying. However, I repeat, the material is easily available at any federal repository.
After I informed Bellesiles in late November 2000 that the reference librarians at the East Point archives said that they had no such state probate records, Bellesiles wrote me again by email on November 30, 2000:
I may not correctly recall where I read the microfilms. I thought it was at East Point, but it hardly matters.He then proceeded to claim that he did most of his research, not on microfilm in one library in East Point as he had claimed, but rather mostly using original (paper) records in over 30 different state and county archives throughout the country.
Now Bellesiles is quoted by History News Network as claiming:
I object strenuously to your publishing those email messages over my name . . . . You may certainly print them over Lindgren's name, but I do not recognize those words as mine, nor acknowledge them. They are incorrect and contradict what I have written many other places as far back as 1991, and what I have discussed at many scholarly conferences over the 1990s.
This denial is not quite what it seems. Someone following this affair must be very careful when interpreting Bellesiles's claims to read what he said, not what he appeared to say. Bellesiles did not actually say that he didn't write the emails that he wrote to me. He says that he does not"recognize" or"acknowledge" their words, not that they are inaccurate representations of what he wrote. And the emails are, indeed, substantively and devastatingly"incorrect" (to use his word). Bellesiles suggests that he didn't write them, but stops short of unambiguously making that claim. Most people, however, would and did read him as saying that. Further, although he does not say that I fabricated these emails, he implies as much.
In his response to Professor Sternstein, Bellesiles goes further. He disputes a passage that appears in substance in both his August 30, 2000 and his September 19, 2000 emails to me:
I must note that I have never said that"the inventories [I] researched were primarily on microfilm in the National Archives." That is just plain false.
Yet I have Bellesiles's disavowed emails on one or both of my home or work computers in their condition as they were in August and September 2000, the days he sent them to me. And I don't know how to alter them without showing alterations. Perhaps he was hoping that I didn't retain his emails unarchived, but I did.
What is perhaps more devastating to his new claim is that I confronted Bellesiles with his crucial 9/19/2000 email when we both appeared with Randy Barnett and Jack Rakove on WBEZ public radio on Jan. 16, 2001. The audio from this program is still available online. On the radio, I read back to him a passage that he now flatly denies ever having said--where he claimed that the probate records were all on microfilm at the East Point, GA National Archives and that he went there to read them.
On WBEZ, he defended, reinterpreted, and excused having made those statements in his earlier email. He endorsed most of the very statements that he now disavows, repeating on WBEZ that the probate inventories"are all available on microfilm," and that he"did in fact, ah, look at them at the National Archives." Both assertions are impossible since at least 7,000 of the probate inventories in his sample counties and years were never microfilmed (Bellesiles gives his total of all his samples as 11,170, which appears to be much too small for what he claims to have counted in Arming America, p. 445). He did appear to recognize that there were some problems with what he wrote to me (presumably that the East Point archives have no state inventories in their collection), pleading as a general excuse:"I must admit I was in a hurry to be done with your email . . . ."
When confronted with his email claim that all the probate records were in the East Point archives and that (for all but ancillary documents in several counties) he had read the inventories at East Point on microfilm, he did not deny that he wrote the email. In fact he defended most of what he wrote in it.
Click here to listen to the January 16, 2001 radio interview yourself.. Click here to read a partial transcript of the interview.GENERAL COMMENTS
This dispute didn't have to happen. Late in 2000, before presenting my scholarship at scholarly conferences in 2001, I privately offered to help Bellesiles in making corrections to his probate work. What Justin Heather and I had found was so devastating that I thought it best for everyone involved if its revelation came in part from Bellesiles. I emailed him in late November 2000, copying two other professors, privately informing Bellesiles that I was finding serious problems in his counts, and asking him to begin to try to reconcile them. I even offered to lend him one of my copies of the Providence records he originally used.
In late 2000, when I offered him my help in making corrections, I still fully expected that Bellesiles would soon be taking the usual scholarly route of grudgingly thanking my co-author and me in public for pointing out his errors. I was trying my best to get him to focus on his errors so that by announcing them first himself, he could mitigate the public embarrassment.
Every time Bellesiles does something like his latest disavowal of his own emails, making claims that are so easily checked, he loses supporters and gets himself in deeper. I hear people who just a couple months ago were agnostics on the dispute saying that the disavowed emails are the latest in a long string of"smoking guns."
It appears to me that Bellesiles's editors and friends have failed him. As with the recent Contra Costa County records fiasco, why didn't Bellesiles run his latest claims by others who could perhaps have exercised more judgment? It became clear over a year ago, when an early draft of our scholarly manuscript that will appear in the William & Mary Law Review began being circulated, that his main 1765-90 probate data were mathematically impossible, that he claimed to count guns in over a hundred wills in Providence that never existed, that he repeatedly counted private guns as owned by the government, and that he repeatedly counted women as men (all errors that have been recently verified by Randolph Roth for the William & Mary Quarterly).
All of these claims would have taken only a few hours to check in a good research library. If in January 2001, Bellesiles's editors at Knopf (or his close friends) had just checked these errors and impressed on him the necessity of pulling the book for major revisions if these and similar criticisms were true, Bellesiles's career would not now be in shambles. He could have just said that he made some terrible mistake--which he could not determine because of his story about the loss of his records in the flood. Eyebrows would have been raised, but it would have stopped there. Most of his colleagues would have been sympathetic, since he would have made the decision to withdraw the book for major corrections while it was still known to be a candidate for academic prizes.
Either Bellesiles was too stubborn to listen or--more likely--there were no such wise friends or editors when it could have done him any good. By blindly supporting Bellesiles or initially criticizing those of us whose scholarship found basic errors, his friends and editors did absolutely the worst thing they could have done for Michael. Surely, his friends and editors had a right to expect more from Bellesiles (and some historians have privately expressed their anger or disappointment at having been taken in). But Bellesiles also had a right to expect more real help from them--especially from his editors at Knopf, who could have saved the academic career of one of their authors if they had simply done their job.
WBEZ Odyssey, hosted by Gretchen Helfrich
Guests: Michael Bellesiles, Jack Rakove, Randy Barnett, and James Lindgren
Jan. 16, 2001
Note:The Jan. 16, 2001 WBEZ show Odyssey is an hour long, but the most interesting exchanges are in the first 33 minutes; the crucial exchange is in the last few of those first 33 minutes. Besides the email question, there are other interesting revelations, such as where I catch him asserting on the radio that he didn't use published probate sources and where Bellesiles says in passing that my counts are accurate (attributing our differences to his false claim not to have used published probate sources). On the air, I read back to him his claim in Arming America to have used the published records of Providence, a claim that he now reconfirms on his website and in the paperback edition of his book. When confronted with this and with discrepancies with his earlier false claim to have used microfilm probate records stored in the National Archives, he seemed to be considerably chastened.
[TRANSCRIPT]
[At about 29:45:]
James Lindgren: Now as far as the published records versus other records, ah, Michael, you know what you wrote to me. I'm sitting here looking at your email to me. And when I asked you for these things, as I repeatedly asked you for your cites--I will read from your email to me:
The probate records are all on microfilm in the National Archives. I went to the East Point, Georgia, federal center to read these microfilms. The films are organized by county. [This is the 9/19/2000 email from Bellesiles to me that Bellesiles now disavows.]
In your emails to me, until I sent you an email saying that your data is way off and that you need to go and recheck the Providence records--only after that did you claim that you looked at the sources in the original courthouses at the, at the primary source. Now in one of your early emails to me, you said that you looked at se-, ah, several counties in the courthouses. But this is a completely new statement made only in the last month . . . .
[At about 31:00]
Gretchen Helfrich: Michael Bellesiles. [long pause] Do you want to respond?
Bellesiles: Yes, yes, I'm just. [long pause]
Hmm, now a lot of this can be stretched out over a great deal of time, um, and I'm sure your listeners would become very bored with it. Let me first say that when I received Mr. Lindgren's first email I had no idea who he was. I did not realize he was such a significant, um, legal scholar.
...
As for the, um, microfilms, yes, those microfilms are available, the inventory lists, not the complete files, but the inventory lists are all available on microfilm, um, and I emailed you the addresses--um, it's the Mormon collection, usually shorthanded, um, the Latter Day Saints have collected these. I did in fact, ah, look at them at the National Archives, because, um, ordinarily I look at material at the Georgia Archives in Atlanta, but they were doing renovations when I had gone over there. This seems like a very minor point and I must admit I was in a hurry to be done with your email because I had so many and I didn't know who you were.
AUGUST 30, 2000 EMAIL TO JAMES LINDGREN
The statistics are in the appendix to my book. But to answer your
I'm afraid that these answers do not serve your purpose. The sort of
Good luck, Michael B SEPTEMBER 19, 2000 EMAIL TO JAMES LINDGREN
My apologies for the long delay in responding to your email. I have been MR. BELLESILES'S RESPONSE TO JAMES LINDGREN, 4-17-02
There is an almost perverse insistence on the part of some people to misread what l have written. Here is my message to you: You will see that no place do I deny responding to Mr. Lindgren's email messages. What I deny is that I ever said that the probate records were located at the National Archives in East Point, Georgia. I have said it often before and will say it again, though I know that those who are biased will insist on misreading it yet again: I told Mr. Lindgren that I read some probate records at the National Archives. It is possible that this statement was misunderstood. However, I have never said that “the inventories [I] researched were primarily on microfilm in the National Archives.” That is just plain false. Neither my book nor my web site give that location for the probate materials, most of which are in local archives, particularly county courthouses. I did use the National Archive’s microfilm readers to read some probate records that I brought with me. I trust that you will do me the courtesy of clarifying to your readers that I have never denied responding to Mr. Lindgren's emails, but I do deny and refuse to acknowledge misinformation to the effect that probate records may be found in the National Archives in East Point, Georgia. Sincerely,
Subject: Re: request for data
X-Phforward: V2.1@relay
To: jlindgren@nwu.edu
Cc: mbelles@emory.edu
From:"Michael A. Bellesiles"
questions in a general way, let me note that I conducted the probate
research before discovering the joys of statistical analysis on computers.
The probate records are primarily on microfilm in the federal archives,
though I also looked at several counties in their undigested form so as to
control for wills and any earlier transfer of guns. All of my note taking
was on legal pads. I simply went through each probate record looking for
guns, recording each and every occurence of any type of gun or gun parts
in any condition. As for photocopies--I did not have the money for such an
effort (tens of thousands of pages), nor can I conceive of the purpose of
such a mass photocopying of material that is available in public archives.
advanced research project you propose would require graduate student
assistants and major funding.
Subject: Re: request for probate info and citations
X-Phforward: V2.1@relay
To: jlindgren@nwu.edu
From:"Michael A. Bellesiles"
X-MIME-Autoconverted: from 8bit to quoted-printable by
relay.it.northwestern.edu id JAA02926
X-MIME-Autoconverted: from quoted-printable to 8bit by
casbah.it.northwestern.edu id JAA28858
inundated with hundreds of emails over the past week, many of them less
than friendly, and am just now finding the time to go through them all.
Please allow me to explain my research to you. I did all my work alone. I
had no assistants and received only a pair of two-month fellowships from
the AAS and the Huntington to support my efforts. Repeatedly, agencies
from the Guggenheim to the NEH rejected my proposal as too far outside the
standard view of American history to be worthy of support.
The probate inventories were the spur to my research, not its purpose or
entirety. NRA attacks aside, Arming America does not base its argument
solely on the 11,170 probate records examined. In a terrible error, when I
removed the long discussion about the reliability of the records, I also
removed reference to that number, which will appear on table one in the
very next printing.
The probate records are all on microfilm in the National Archives. I went
to the East Point, Georgia, federal center to read these microfilms. The
films are organized by county. My sample set is listed in the note on
table one. I counted a county as frontier for the thirty years after
settlement, then moved it into the regional category. My method was
primitive by social science standards. With every new record I put a tick
in my total column. If there was any sort of gun or gun-part I counted
that. That’s it. All done by hand on a legal pad. I also went to several
county court houses to look through wills that were not microfilmed to see
if guns were listed as already given or to be given. I attempted to err
always on the side of the gun being present; thus my counting gun-parts.
As anyone who has used probate inventories will tell you, sometimes the
writing is dreadful. A “musket” might actually be something completely
different, such as a “mop bucket,” a “gun” could be “gin.” I always
assumed it was a gun.
I also looked at court records, business records, militia records,
ordinance records, Army records, account books, personal letters, diaries,
travel accounts, newspapers, magazines, legislative records, petitions,
novels, short stories, illustrations, and anything else which might
contain a reference to guns. All the counting I did in all of these
sources was done by hand much as with the probate inventories. Thus I
looked through boxes of lists of guns in armories and the hands of
individual militia units and did not enter them into any sort of data
base; I simply recorded the numbers so that I could finish this book in my
lifetime. I used no regression analysis, nothing more complicated than
arithmetic. I tried to express these numbers in easy terms for a general
reader. It seems to me that an excellent exercise for graduate students
would be to set them to a roll of probate records and see what they find.
The Institute for Early American History and Culture had a long discussion
of guns in probate records after I gave my first academic talk on the
subject six years ago. That discussion may still be catalogued on their
web site. Nearly all of those who participated reported reaching the same
findings I did.
Finally, all the original notes from this research are in boxes in my
attic, where I moved them after they were damaged by a flood in Bowden
Hall at Emory. I do not have the time right now to search for those notes
and to lay out the water damaged pages for drying. However, I repeat, the
material is easily available at any federal repository. Anyone can spend a
great deal of time reading those thousands of pages of lists. I can almost
guarantee that their numbers will differ slightly from mine, as they may
see a “gun” that I did not and miss several that I counted as “guns”
because it looks like a different word. This is why history is not a
science. Best wishes, Michael B.
Michael Bellesiles
Department of History
Emory University
Atlanta, GA 30322
404) 727-4467 fax (404) 727-4959
mbelles@emory.eduI finally read through the rest of that ill-tempered response. Yes I object strenulously to your publishing those email messages over my name. You may certainly print them over Lindgren's name, but I do not recognize those words as mine, nor acknowledge them. They are incorrect and contradict what I have written many other places as far back as 1991, and what I have discussed at many scholarly conferences over the 1990s. I prefer the old fashioned standard: if you want to know what an author says, read what he and she wrote, not what others say was written.
Professor Michael Bellesiles