One of the fascinating subplots in the controversy over whether Michael Bellesiles committed scholarly fraud in his book, Arming America, is what he claims happened to some of his research notes. Bellesiles asserts that water damage from a broken sprinkler pipe in Bowden Hall, where the Emory University History Department's offices are housed, flooded the building and"pulped" and"destroyed" the yellow legal pads on which he says he recorded in penciled"tick marks" the 11,170 probates he says he researched. As a result, he has stated repeatedly, it was impossible for him to provide any evidence drawn from that disputed research to support his findings of low levels of gun ownership in antebellum America. In a recent communication to HNN, he explained that the yellow legal notes pads recording his probate research, some twelve in all, were sitting in a box with other papers and index cards,"when the ceiling of my office collapsed under the waters (our pipes are in the ceiling), [and] the water seems to have come down right on top of that box of pads. Everything in that box was destroyed, as was the chair under it (I was not there, so I cannot be certain)."

In that same communication, Bellesiles said the flood in Bowden Hall took place in May 2000, and that he"announced" his loss publicly in August of that year. He implied that the reason he delayed informing historians about his destroyed notes before his probate numbers came under critical scrutiny from scholars seeking to replicate them, was that"I was in Europe in the intervening months," suggesting that he was away around the period the flood occurred, since he had traveled to Europe sometime in May and remained there until August.

Bellesiles has told many confusing and contradictory stories about his probate research and the status of all his research notes, some of which I have gone into extensively in a previous article. But none is more fantastic or demands a greater suspension of disbelief, than the tale he has spun and is still spinning about his missing probate notes.

THE FLOOD

Always, in beginning that tale, Bellesiles places the date of the Bowden Hall"flood" which supposedly"pulped" his notes either in May 2000 or in the summer of that year, although it happened on April 2, 2000, on a Sunday evening. He first mentioned his"water damaged" notes in an e-mail -- one of several e-mails whose contents he disavowed only last month -- to one of his leading academic critics, James Lindgren, on Sept 19, 2000, and not in August as he only recently claimed; and in another e-mail, on Nov. 30, 2000, he dated that damage to"the Bowden Hall flood this summer [italics added]." Several weeks later, on Dec. 21, 2000, in a posting on H-NET (H-OIEAHC) he again repeated that his notes were"in a sorry state because of the great Bowden Hall flood of 2000 (the pipes in Emory's History building burst this summer)." In his article in the OAH Newsletter,"Disarming the Critics" (Nov. 2001, p.3), Bellesiles left the time of the"flood" vague, writing,"As I mentioned to many historians after my book appeared in September 2000, my notes of these probate records had been destroyed" during the Bowden Hall flood, and though he doesn't give a date for it, refers readers to the Emory Report, vol. 52, 8 May 2000, for confirmation of the damage it caused. But in his recent response to his critics in the William and Mary Quarterly (WMQ), he was once again precise about when the flood happened, and once again he was precisely wrong about the date, stating"it is now well known that my notes were destroyed in the May 2000 flood of Bowden Hall at Emory University" (WMQ, Jan. 2002, p.247). And in his latest communication to HNN, on April 9, 2002, he once more placed the"flood" in May. Never has Bellesiles, either privately or publicly, acknowledged that the"great Bowden Hall Flood," as he likes to label it, occurred on the date it actually happened: April 2, 2000.

How can one account for Bellesiles's repeated lapse of memory in assigning the flood to either May 2000 or to the summer, rather than the correct April 2 date? How was it possible for him to clearly recall the exact spot in his office where among his"mountain of paper," as he has described his research notes, his twelve yellow legal pads containing his controversial probate research lay on a chair under cascading water, but he can't remember the month or season of the year when he claims they became useless pulp? What is at work here, I believe, is not some bizarre form of selective memory disorder but an effort by Bellesiles to create a false impression of where he was and who was responsible when his disputed research notes apparently disappeared -- if they ever did, indeed, disappear or never existed. Since Bellesiles left for Europe in May and remained there until August, teaching at Oxford during the summer, his mantra that the"great Bowden Hall flood" took place in that time span appears to be a deliberate attempt to invoke an image of being absent amidst that turmoil. And if he was far removed from Emory when the waters tumbled from his office ceiling pulping his yellow note pads, then it just might have been possible, as he always implies, for unnamed others, presumably clumsy Emory custodians or incompetent employees of the firm Emory brought in to clean up and dry out the building, to have tossed them out with the trash before he could intervene.

What academic critic of Bellesiles's scholarship hearing of his grievous loss would be so bold, however far-fetched his tale of"pulped" notes might appear, to challenge Bellesiles's story, particularly when he began giving interviews and publicly informing everybody who would listen that salvaging them was a task wholly beyond his control. As he wrote HNN last month,"I was not there, so I cannot be certain" what happened. Still, even though he says he was not"there," he claims with absolute authority (it should be noted, however, he has given flagrantly contradictory and inconsistent statements about this issue) that"All of my notes were destroyed in boxes on the floor of my office, and they got drenched, pulped," and suggests with a knowing nod and wink, but never openly, that Emory's workmen discarded them (quoted in Melissa Seckora,"Disarming America", National Review Online, Oct. 1, 2001).

But however much Bellesiles tries to create the opposite impression, one important fact is irrefutable. Bellesiles was"there" at Emory in April 2000, teaching two courses on a Tuesday, Thursday schedule, one of which met in room 116 in Bowden Hall (from 1 to 2:15 p.m.) before the flood and was moved to another building following it. A student who took one of Bellesiles's courses, History 231,"Foundations of American Society to 1877," clearly recalls that term papers were due the first week of May and that final examinations took place early that month, long after the deluge of April 2.

And if Bellesiles was at Emory, teaching on a Tuesday, there is no reason why he could not have been in his office on Monday trying to save his research notes, that is, if he ever had probate notes on yellow legal pads to save. Like all Emory professors with offices in Bowden Hall, Bellesiles was notified as soon as possible about the water damage. Affected faculty were told to come in, inspect their premises, clear out their belongings, and make arrangements with the experts Emory provided to restore damaged intellectual property like books and notes, or to submit insurance claims for any losses. Before anybody arrived, water had been pouring for about 25 minutes from a burst connector on a sprinkler main on the third floor of the building. According to Chuck Norris, director of Emory's facilities management, whom I interviewed over the telephone, after the water was turned off, custodians were immediately dispatched with mops and buckets to soak up the water, which was about two inches high in some locations. Once Norris and others determined how extensively the building was affected by the water, Emory called in Disaster Services Incorporated, a company with specialized drying equipment and expertise to handle the matter.

Company employees entered Bowden Hall within two or three hours after the water was turned off, and began the task of drying out the wallboards, carpeting, and other water damaged interior components of the building. Heading the operation was David King, a senior vice president of the company, whom I also interviewed by phone several times. As he vividly recalls, only a few faculty members came in immediately to survey the premises and clear out their desks and offices. Bellesiles was not among them. Those who responded then and later, had to get passes from guards who were posted around the building. Once faculty arrived, they were provided with egg crates with which to carry out whatever books, notes, and other materials they wanted to remove. They were also given trash bags for anything they thought should be discarded, and only faculty made that decision. But before any items could be thrown out, the individual bags holding them were tagged and their contents were meticulously examined by specialists from a firm of salvors who inventoried what they contained. Claims adjustors were on the premises on the Monday after the flood, April 3, to verify for insurance purposes what was damaged, what was still usable, and what was lost beyond redemption. Also on hand were members of Emory's Library Preservations Office, ready to restore any books and papers faculty brought to them, a job which they proved extraordinarily adept at since only seven items out of more than 500 books and other materials were beyond saving.

At no time during all of this, said David King, who was on the premises almost around the clock directing his employees, did he ever knowingly come into contact with Bellesiles. Asked if his employees would ever throw out any paper or books on their own volition, he adamantly denied even the possibility of anybody doing so. Offices, he said, were locked, and guards were posted in the hallways. After the first night, only faculty members had access to their offices. And though he could recall by name several professors whom he helped in the process of drying or moving their belongings, Bellesiles was not one of them. Like everybody else at Emory, King never heard a word about Bellesiles's notes being"pulped" and thrown away until six or eight months later. Chuck Norris also never learned about Bellesiles's claims about his notes until the controversy over his research arose. And Bellesiles has admitted he never submitted an insurance claim for his"destroyed" notes at the time of the flood or since.

Asked whether Bellesiles's yellow legal pads could have been"pulped" or"destroyed" by water flowing out of the pipe in his office ceiling, as he has asserted, King dismissed the idea. The burst connector on the sprinkler main pipe was on the third floor, while Bellesiles's office was on the second. All main water pipes, moreover, were in the hallway, not in the office ceiling, as Bellesiles has stated. What water cascaded from the third floor to lower floors, came through gaps in the wall tiles and ceiling panels, and though some ceiling panels came down, they are not heavy or formidable enough to ruin anything. More importantly, he said, yellow legal pad paper would never be"pulped" or"destroyed" by water falling through the ceiling panels to the chairs or floor below.

THE WATER EXPERIMENT

Intrigued by what King, an expert on water damage, explained to me, I remembered the famous test that the late Nobel laureate physicist Richard Feynman conducted during the inquiry over the disaster that befell the Challenger spacecraft. He took a rubber O Ring, that he speculated might have been compromised during the cold weather that preceded the Challenger's takeoff, and inserted it into a glass of ice water to test its properties. The rubber in the O Ring became more rigid and less flexible as a result of the cold, establishing with that common-sense experiment that the cold weather and the rubber's reaction to it was probably responsible for the Challenger explosion. Couldn't a similar simple experiment with a yellow legal pad determine whether the paper would deteriorate and be"pulped," as Bellesiles has repeatedly asserted his probate notes were, by the water streaming down from his office ceiling?

After purchasing a twelve pack of yellow legal note pads from Staples for less than five dollars, I marked a pad arbitrarily with notations in a #2 pencil and placed it under a shower running at full force for 30 minutes and let it sit in the tub as it filled with at least two inches of water. As David King predicted, after the 30 minutes were up, the pencil notations were as legible as when I inscribed them and only the top sheets of the pad were completely soaked. Nevertheless, they remained intact and virtually unscathed, while the majority of the pad, except for the margins of the paper, was almost dry. It was if the top pages of the pad acted as a waterproof barrier against water affecting the inner pages. I then separated the pages and let the pad dry out overnight on a granite kitchen-top. Still slightly damp the following morning, I placed the pad outside in the sun to dry. Within hours, though wrinkled and puffed up, the pad became eminently suitable for writing once more. And the penciled notations remained as clear as when I earlier inscribed them.

I tried the same experiment, running the tub faucet, as opposed to the shower head, at full force over another pad for a half an hour, and then six of them stacked on top of each other, for another half hour. None of the pads were badly damaged. All remained completely usable after drying and the pencil marks were easily legible. I would have carried on the shower experiment for hours, except for the fact that the Northeast is now experiencing a water shortage. But even had I done so, I'm convinced none of the pads would have been"pulped." One thing is clear: it is impossible for a shower or a bath tub faucet running at full blast, simulating and even surpassing in force water dripping from a ceiling for 25 minutes, to"pulp" or"destroy" a yellow legal pad unless it was manufactured from tissue paper. David King was right. And anybody who doubts what my simple experiment demonstrated, can try it out for themselves. Indeed, I invite Bellesiles's most ardent supporters, one of whom recently posted a comment on H-NET (H-OIEAH, May 10, 2002), complaining about"wild charges of fraud. . . [that] have been leveled against Bellesiles," to attempt this experiment themselves.

Some of Bellesiles's apologists might say that this experiment merely indicates that Staples sells yellow legal pads that stand up to water. And so it does. But it also indicates much more than that. In particular, it demonstrates, conclusively in my opinion, that Bellesiles's story of"pulped" and"destroyed" notes is fictional. It strongly suggests, moreover, that Bellesiles invented the story of"pulped" yellow note pads to cover up the fact that he did little of the probate research -- other than perhaps examining a couple of counties in Vermont, and published records in Providence, R.I. -- he said he carried out.

The"great Bowden Hall flood" provided Bellesiles with a convenient excuse to explain why he could not back up his probate numbers when Lindgren e-mailed him on Nov. 28, 2000 that his counts of Providence's published probate records were highly dubious and that there were no microfilmed probates in the East Point, Georgia Federal Archive, where Bellesiles had previously asserted he read them. Before that fateful November e-mail from Lindgren, Bellesiles had claimed that only some of his notes were slightly damaged and that all of them were sitting in boxes -- probably the same egg crates David King's company provided -- stored in his attic. But with Lindgren now privately telling Bellesiles of the serious problems he found with his probate research, Bellesiles was faced with the strong possibility that his alleged scholarly misconduct would quickly be exposed. So -- I surmise -- he embarked on two interconnected strategies to confuse and hinder further inquiries.

First, in an e-mail to Lindgren on Nov. 30, 2000, Bellesiles now contended, contrary to his previous assertions about reading almost all his probate records on microfilm in the East Point federal archive, that he had actually carried out most of his probate research in original documents housed in about 30 county courthouses and archives scattered around the country, locations and research trips which he had never mentioned or recalled before. But because Bellesiles took only a couple of days to come up with this new tale for fear Lindgren might go public with his concerns before he could counter them, he provided inaccurate locations for a number of probate records, especially those that were not listed in the easily available published source he appears to have drawn many of them from, Alice Hanson Jones' compilation of probate inventories, American Colonial Wealth: Documents and Methods (1978).

In his haste, Bellesiles's most glaring error was to locate the San Francisco probates in the Superior Court of that city, when it was common knowledge among virtually all researchers familiar with its archives, but not to Bellesiles, that those records were destroyed in the 1906 Earthquake. Other conspicuous errors Bellesiles made in his rush to develop his new story, as Lindgren points out, were incorrect locations for probates in Los Angeles County, Ca., and Windsor County, Vt. And further confirming how pressed for time Bellesiles was to construct his story in order to counteract what Lindgren had discovered and might soon reveal, he claimed to have read the probate inventories for Knox County, Ind. on microfilm. The blatant error here was that they are not on microfilm. Instead, if Bellesiles had actually used those inventories, he would have had to spend weeks or months combing through the paper originals in Vincennes, Ind., a research junket he could hardly ever have forgotten had he actually undertaken it (James Lindgren,"Fall From Grace: Arming America and the Bellesiles Scandal", 111 Yale Law Journal [YLJ](June 2002), manuscript, 116-120; interview with James Lindgren, May 18, 2002). To compound these serious mistakes, Bellesiles has yet to correct his inaccurate locations for Los Angeles and Windsor, which are still on Bellesiles's website reports as this is written. And his incorrect location for Knox County remained on his website for months before he recognized -- or somebody recognized for him -- that error.

Bellesiles's second apparent strategy, coordinated with his first, was to make it as difficult as possible for his academic critics to prove that he had falsified his probate research. In pursuing this goal, Bellesiles began to state for the first time, both privately and publicly, that his yellow note pads, on which he said he had recorded all of his probate findings, both archival and on microfilm, had been turned into unreadable pulp by the"great Bowden Hall flood" and tossed out by parties unknown. As a result, he explained, he could not provide the data that his potential replicators requested and that the canons of the American Historical Association require all historians to make available once their works are published and questions arise concerning their reliability. And without that data, says Bellesiles, who is to say he is wrong? He made this argument when Lindgren proved (and historian Randolph Roth confirmed) that his main table of probate numbers was"mathematically impossible." How could Lindgren tell? Bellesiles wondered,"Since neither he nor I have the numbers, which were lost in the Bowden [Hall] flood [in 2000], I am at a loss to grasp his omniscience" (quoted in Lindgren,"Fall From Grace", YLJ, 136, fn. 206, 207, in which Lindgren clearly explains the simple math exercise Bellesiles failed to"grasp").

Unfortunately for Bellesiles, the tales he has concocted about where and how he did his probate research are belied by the incontrovertible evidence of his e-mails to Lindgren, and no midnight conversion in which he disavows the words they contain will change that basic fact. Also, the fantasy he has spun about his"pulped" notes is punctured by the reality of when the Emory flood actually took place, as against when Bellesiles wanted others to think it happened, his inconsistent and contradictory explanations about what happened to those notes, and by the demonstrable evidence that those mythic yellow note pads resist water far better than he ever imagined was possible. Obviously, this is another instance when Bellesiles did not do his research before making his claims.

THE GRADUATE STUDENT

But if there remain any doubts whether Bellesiles's tale about his"pulped" probate note pads is a product of his imagination, a startling new piece of information that only became available as this article was being prepared for publication should help lay them to rest. On May 17, new problems about his notes story were raised on an online"blogspot" run by Professor Michael Tinkler (the so-called"Cranky Professor"), who teaches at Hobart and William Smith Colleges in New York and who received his Ph.D. in European History from Emory University. Tinkler related the experience of a student who served as Bellesiles's graduate assistant in 1988. As the former student explained in a series of e-mails to Tinkler, she spent her year as Bellesiles's research assistant in the basement of Emory's library reading microfilmed probate records for him,"counting guns and entering the numbers into a [Lotus] spreadsheet" [italics in the original]. Taken aback by this information, Tinkler prevailed on her to get in touch with the Emory History Department. After listening to what she informed him, that contrary to what Prof. Bellesiles has always insisted about doing his probate research alone and marking yellow legal pads with penciled tick marks, she entered the data on a Lotus spreadsheet, the Department Chairman concluded that she had likely been working on his earlier book, a biography of Ethan Allen, which was published in 1994, and not Arming America.

Whether this is the case is not really that important. What is significant about this disclosure is that Bellesiles used a computerized database to record and manipulate his probate and other numbers. He has repeatedly told the press that he worked only with pencil on yellow legal pads. In his Sept. 19, 2000 e-mail to Professor Lindgren Bellesiles said in unequivocal words that his research was"All done by hand on a legal pad," and that"I did all my work alone. I had no assistants. . . ." In his Aug. 30, 2000 e-mail Bellesiles wrote,"I conducted the probate research before discovering the joys of statistical analysis on computers." Given that Bellesiles directed his graduate research assistant in counting guns in probate inventories into a Lotus spreadsheet for a year starting in 1988, how is it possible to explain his assertions that he never used a computer and had yet to discover"the joys" of employing computer databases while he was researching Arming America? The only reasonable interpretation is that he denied processing his probate data with a computer while working on Arming America because he feared what that data might demonstrate once his research came under scrutiny. So rather than provide quantitative scholars like Lindgren access to his database for evaluation (that is, assuming Bellesiles did the probate research he claims to have done, which is assuming a lot), Bellesiles posed as an"old fashioned" historian who recorded his numbers primitively with pencil on a yellow legal pad , the better to explain how that evidence disappeared in the waters of Bowden Hall.

Some might argue that some of this is speculative. And some of it is. But there is no doubt that Bellesiles consistently gave false information about the timing of the flood and very little doubt that he gave false information about his presence at Emory at the time of the flood. Yet even the most speculative portions of this essay are far more rational than the confusing and contradictory stories Bellesiles has given about the way he conducted his probate research and what happened to it. And in view of Bellesiles's well-established ability and readiness to weave fanciful fictions, he might yet concoct another one, such as arguing that his yellow note pads, unlike those that Staples sells, were manufactured out of tissue paper. And, I would expect, as in the case of the more gullible consumers of pulp fiction, there will be those, especially among his most ardent apologists, who might believe him.