Seldom are historical controversies resolved so abruptly and decisively as when DNA testing proved that Thomas Jefferson had lustful assignations with his slave Sally Hemings, the "Dusky Sally" of historical debate and popular rumor-mongering. At first, the news coverage focused on the genetic whiz-bangery that produced the insight. Next came the airy, thumb-sucking pieces on how the revelation might affect the Jeffersonian legacy. Was the Mount Rushmore icon an unregenerate hypocrite, tarnished beyond rehabilitation? Would he be reborn as an unlikely multi-culti icon?
So far, however, the press has largely let Jefferson historians off the hook -- virtually all of whom, until this month, vigorously dismissed the possibility of a liaison between the pair. Often, they did so using intemperate language that evoked the furious defense of a best friend unfairly maligned, rather than the careful weighing and sifting of evidence. Now, without fully acknowledging their mistakes, some -- in particular, Joseph Ellis, the Mount Holyoke professor and author of "American Sphinx: The Character of Thomas Jefferson," which won a National Book Award last year -- are stepping forth to interpret the episode for us afresh, as if they had never been wrong in the first place.
Time out. It seems that a period of self-analysis for the historical profession is in order -- at least under the portico that shelters Jeffersonian scholars. As much as this has been a story about DNA's ability to solve time-shrouded mysteries, it should be an occasion for reflections on the all-too-human failings of historians.
Ellis, the Jeffersonian expert du jour, is Exhibit A in how scholars have mishandled the Hemings question. His role is especially important and revealing because he was asked to provide historical context for the article in Nature, the scientific journal that broke the Y-chromosome story. The same week, he received a platform in U.S. News & World Report to pontificate further on the subject -- a subject, one hastens to add, on which he has been spectacularly wrong in the past. Since then, he has been quoted in every conceivable news outlet.
The word "character" is central in the title of Ellis' book. One might think that in the course of an examination of the character of our third president, arbiter of universal human rights, the question whether he had a 38-year liaison with a chattel slave might be salient. The issue, after all, has been around since 1802, when muckraking journalist James Callender first aired it. In 1873, Madison Hemings, one of Sally's sons, told white journalists that his mother had told him that he and his siblings were Jefferson's children. A black Monticello caretaker "confirmed" his tale. A year later, when a white biographer came knocking, Jefferson's white grandchildren argued that Jefferson's randy nephews Peter and Samuel Carr were the true miscegenators. No one ever doubted there were some light-colored slaves wandering around Monticello who looked an awful lot like our founding father.
In "American Sphinx," however, Ellis relegated the Hemings question, for the most part, to an appendix. In his prologue, he says that no jury could decide the case on the basis of the circumstantial evidence. In the body of the book he doesn't say much more than that after James Callender leveled the accusation, "this piece of scandal ... affixed itself to [Jefferson's] reputation like a tin can that rattled through the ages and pages of history."
In the appendix, however, he took a harder line. He treated the possible liaison less as a topic interesting in its own right than as a contemporary cultural conundrum. Why, he seemed to be asking, do people continue to believe this? "Within the scholarly world," he wrote, "there seems to be a clear consensus that the story is almost certainly not true." He went on to observe -- rather condescendingly, to my ears -- that "within the much murkier world of popular opinion, especially within the black community, the story appears to have achieved the status of a self-evident truth." The message was clear. Who would you believe? The "scholarly world"? Or black urban legend?
He explained how he came to this conclusion. "After five years mulling over the huge cache of evidence that does exist on the thought and history of the historical Jefferson," he wrote, "I have concluded that the likelihood of a liaison with Sally Hemings is remote." Why? "For most of his adult life he lacked the capacity for direct and physical expression of his sexual energies."
As evidence Ellis noted that Henry Adams had called Jefferson "almost feminine." And Jefferson's "most sensual statements," Ellis continued, "were aimed at beautiful buildings rather than beautiful women."
Historians make judgments based on fragmentary evidence; missteps are an occupational hazard. Yet when Ellis shifted from hard-line skeptic (1997) to interpreter of fresh evidence that definitively proved his old views wrong (1998), he didn't skip a beat. In Nature, he wrote: "With impeccable timing, Jefferson appears to remind us of a truth that should be self-evident. Our heroes -- and especially Presidents -- are not gods or saints, but flesh-and-blood humans." What happened to the godlike man who sublimated temptations of the flesh to sober appreciation of Doric columns?
And in the Nov. 9 U.S. News essay, Ellis suddenly depicts the scholarly world in a different light. The "clear consensus" evaporated. "Within the scholarly world," he wrote, "the acceptance of a Jefferson-Hemings liaison had been gaining ground over recent years."
Here the scholarly backpedaling seems a bit disingenuous. But forget the self-contradiction for a moment. This time, Ellis was correct the first time around. For years, historians had been disparaging anyone who dared to suggest that Jefferson and Hemings may have had a sexual relationship.
In "The Jefferson Image in the American Mind" (1960), Merrill D. Peterson, who held an endowed chair at the University of Virginia, attributed the Jefferson-Hemings legend to "the Negroes' pathetic wish for a little pride, and their subtle ways of confounding the white folks." Nice. Dumas Malone, an indefatigable biographer who devoted his professional life to Jefferson, wrote pointedly that he would have much preferred to ignore the whole sordid Hemings question if others had not had the bad taste to air it. In "Jefferson the President: The First Term" (1970), he sniffed that the charges "are distinctly out of character," and that adjudging them to be true "would be as absurd as to charge this consistently temperate man with being, through a long period, a secret drunkard."
Can these sentiments be dismissed as the dated ravings of the old guard? Hardly. In 1991, one Alf J. Mapp Jr., in "Thomas Jefferson: Passionate Pilgrim," called the charges "Federalist fancy." (The Federalists, of course, were Jefferson's sworn political enemies.) In 1993, Willard Sterne Randall, in "Thomas Jefferson: A Life" (a book read and praised by President Clinton), said the tale "must be put down as mere gossip about a great man published in the absence of journalistic standards, much less historical ones."
All this could inspire depression about the state of historiography, were there not a hero to the story, a woman who has gotten some due, but not nearly enough: Annette Gordon-Reed, a professor at New York Law School. (Princeton historian Sean Wilentz does give her a nice pat on the back in the current New Republic.) Her 1997 book "Thomas Jefferson and Sally Hemings: An American Controversy," while respectfully reviewed, was vastly overshadowed by Ellis'. Like an avenging torts professor straight out of "The Paper Chase," she laid bare -- in tight, unsentimental prose -- the shoddy logic and hidebound assumptions of the die-hard Jefferson defenders long before the Y-chromosome did.
Her main point was: Circumstantial evidence is all we have. If you're going to dismiss the testimony of Madison Heming (a black man) on the grounds that it is mere oral history, and possibly self-serving, shouldn't you do the same for the accounts of Jefferson's white grandchildren, who pinned the paternity on the family black sheep, the Carr brothers? To put it bluntly, Gordon-Reed wrote, the game is rigged when black and white testimony conflict. The white oral history was not just deferred to, it was ferociously defended -- despite the fact that testimony from both whites and blacks agreed that there were mulatto children who looked like Jefferson running around.
"The underlying theme of most historians' denials," she wrote, "is that the whole story is too impossible to believe." Jefferson, after all, was almost too good for this earth. Hemings was barely human. Scholars "shamelessly employ every stereotype of black people and distortion of life in the Old South to support their opinion," she concluded.
It would be unfair to tether Joseph Ellis and "American Sphinx" too closely to traditional Jefferson hagiography. In many ways, his book was a roundhouse right delivered to the Jefferson mafia -- those academics who saw their role as burnishing Jefferson's status as the exemplar of the Virginia gentleman. Ellis underscored the unfathomable mental gymnastics that enabled Jefferson to condemn the typical American revolutionary as a "stupendous" hypocrite for failing to see that one hour of bondage was "fraught with more misery than ages of that which [Americans] rose in rebellion to oppose"; and yet, in the next breath, to ditheringly argue that the issue should be left for "providence" to solve. Yet, on the Hemings question, did Ellis free himself entirely from the time-worn biases? "I don't think you can group all the biographers and historians who did not endorse the existence of this liaison in the same category," he told me. He points to the clear expressions of uncertainty in his prologue, and insists he did not come down as strongly in the appendix as I and others have read him.
Moreover, he says, his reasons for doubting the liaison were hardly flattering to Jefferson -- let alone racially tinged. After all, arguing that Jefferson was effectually impotent is hardly comparable to invoking "the Negroes' pathetic wish for a little pride."
I called up Gordon-Reed to see if she was enjoying her vindication. The answer, of course, is yes. But she's not entirely happy that the media has focused so extensively on the discovery itself, and not on historians' past handling of the matter. Although she doesn't want to pick a fight with Ellis, she was irked by Nature's selection of him as interpreter of the scientific findings. "It drove home the point of my book," she says, "which is that where Jefferson is concerned, the automatic person to talk to is the white guy." (She hastens to add that since the story exploded, she and other black scholars have gotten plenty of air time and column inches.)
Despite the National Book Award, Ellis had hardly distinguished himself on the Hemings question. It was ancillary in his book. If a scholar better versed in the Jefferson-Hemings lineages had written the Nature item, Gordon-Reed suggests, a more nuanced picture of the Jefferson-Hemings saga might have emerged in the popular press. Just one example: Thomas Woodson, Sally Hemings' alleged first son, may be a myth. There's no contemporary evidence that suggests Hemings had a son by that name, yet the Jeffersonian punditry repeatedly fails to make this clear.
There is one small sore point in the two scholars' personal relationship. Before Ellis had finished his manuscript, Gordon-Reed sent him a copy of hers. They had an amiable intellectual exchange, and he intimated that she had convinced him that the odds were roughly 50-50 that an affair had taken place. He even wrote her a nice blurb. When "American Sphinx" came out, however, Ellis reverted to the position the relationship was a "remote possibility." That stunned Gordon-Reed. "He pretty much gave me the back of his hand," she says. "I'm not so good at math, but a 50-50 chance is not a 'remote possibility.'"
Ellis says there is no contradiction here. The circumstantial evidence, he agrees, was inconclusive. He was swayed by his "reasonably intensive understanding of Jefferson's voice in his correspondence with women." He disagrees with Gordon-Reed's assessment of the depth of racial bias in recent Jefferson scholarship. "While I think that race was a variable earlier in the historiography," he says, "I don't think it has been over the last 25 years."
Turning error into triumph, he argues that far from undermining his book, the new findings strengthen its basic argument. "It seems to me the new evidence deepens and darkens the Ellis version of Jefferson as an inscrutable sphinx," he says, "and moves him from being not a person living a paradox, but a person living a lie."
Gordon-Reed was an amateur in the classic sense: A non-historian who followed up on a lifelong fascination with Jefferson, whom she found "fascinating and horrifying." And maybe only someone from outside the historians' guild could identify the double standards that have governed discourse on this topic for so long. Another outsider, Conor Cruise O'Brien, the Irish intellectual and diplomat who has lately chipped away at Jefferson's status as poster boy for liberal democracy, went on record three years ago to say he believed the affair happened. "I think the Jefferson of the biographers -- for whom such behavior would be unthinkable -- is a fictional construct." That statement now seems remarkably prescient. Historians scoffed when an earlier champion of the Jefferson-Hemings story, historian Fawn Brodie, pointed out in the '70s that Jefferson often used the French word for "mulatto" to describe the color of French soil during his visits there. Racial mixing was much on his mind, she concluded. Maybe that interpretation was psycho-biographical drivel. But is it any more so than Ellis' musings that Jefferson displaced his sexual urges onto neoclassical half-domes and arches?
Finally, how much distance is there, really, between the pronouncement by Dumas Malone that "no professional historian has ever believed" the Hemings story and Ellis' assertions about the "clear consensus" among scholars? "It is my belief," Gordon-Reed wrote in "Thomas Jefferson and Sally Hemings," "that those who are considered Jefferson scholars have never made a serious and objective attempt to get at the truth of this matter."
Now that the historical question has been resolved, it's the historiographical one that nags. Not: Was Jefferson chaste? But: Were the historians who were investigating the issue honest and objective in their searchings? And, if the consensus is that they were not, will they confront their failings? Or will they insist there was never a problem and in the same breath confirm that they have solved it?