Joseph J. Ellis is the Ford Foundation Professor of History at Mount Holyoke College in Massachusetts. He is the author of American Sphinx: The Character of Thomas Jefferson (1996), which won the National Book Award. From American Sphinx by Joseph J. Ellis, 1996, pp. 303-307. Copyright 1997 by Joseph. J. Ellis. Used by permission of Alfred A. Knopf, a division of Random House, Inc.

The history of the story itself falls naturally, if not neatly, into three discernible phases. The first was the early nineteenth century, when James Callender published the initial accusations and the Federalist press spread them across the country. Callender's motives, all historians agree, were scurrilous and vengeful. He probably heard the rumors about miscegenation at Monticello while imprisoned in Richmond--it was a story that had been making the rounds in Virginia for several years--and felt no compunction about reporting the gossip as fact. His charges, while obviously motivated by the basest personal and political motives, derived a measure of credibility from three different factors. First, Callender had accurately reported on the adulterous affair between Alexander Hamilton and Maria Reynolds in 1797; while wildly irresponsible and blatantly smut-seeking, he tended to exaggerate rather than tell outright lies. Second, Sally Hemings did have several children who were obviously fathered by a white man and some of whom had features that resembled those of Jefferson. Third, Callender had correctly accused Jefferson of making unsolicited advances toward Elizabeth Walker, a married woman, when he was a young bachelor in 1768. Jefferson acknowledged the truth of this youthful indiscretion in 1805, made a public apology to her husband, John Walker, but claimed it was the only charge "founded on truth among all their allegations against me." Nevertheless, the accuracy of the Walker accusation lent a measure of credibility to the Sally story.

The next chapter in the story, which occurred in the middle decades of the nineteenth century, produced two new pieces of evidence, each important in its own right, but together contradicting each other. In 1873 Madison Hemings, Sally's next to last child (born in 1805), gave an interview to the Pike County (Ohio) Republican claiming that his mother had identified Thomas Jefferson as his father and, in fact, the father of all her five children. This claim was verified by Israel Jefferson, another ex-slave from Monticello, also living in Ohio at the time and a longtime friend of Madison Hemings's. The following year, in 1874, James Parton published his Life of Thomas Jefferson and reported another story that had been circulating within the Jefferson and Randolph families for many years--to wit, that Jefferson's nephew Peter Carr had been the father of all or most of Sally's children and that he had admitted as much to Martha Jefferson when she had confronted him with the charge. Sally's children looked like Jefferson, then, because they were related, but through Carr rather than Jefferson himself. This version of miscegenation on the mountaintop received partial corroboration from Edmund Bacon, the former manager of Monticello, who claimed in his interview of 1862 that he had seen another man leaving Sally's quarters "many a morning." There were now two different versions of the Sally story placed before the public, one rooted in the oral tradition of the Hemings family and the other in the oral tradition of the white descendants of the Jefferson Randolph family.

The third chapter of the story dates from the 1950s, when the scholarship on Jefferson, especially the massive publication project led by Julian Boyd and the authoritative six-volume biography by Dumas Malone, generated fresh evidence and a new and spirited debate about its meaning. Although the most dramatic episode occurred in 1974, with the publication of Fawn Brodie's Thomas Jefferson: An Intimate History, a national best-seller that argued in favor of the liaison and even claimed that Jefferson and Sally Hemings loved each other, the new evidence in fact came from Malone. Despite his own forcefully argued conclusion that the Sally story was a fictional creation of Callender and nothing more, Malone's research revealed that Jefferson was present at Monticello nine months prior to the birth of each of Sally's children. Since he was often away at Philadelphia or Washington, and since Sally never conceived in his absence, the timing of her pregnancies was compatible with his paternity. In 1993 the researchers at the Thomas Jefferson Memorial Foundation discovered a "missing" daughter, born in 1799, who had died soon thereafter. That birth was also compatible with Jefferson's residence at Monticello. Although Brodie's book ignited a raging debate within the scholarly world and then a proliferation of biographies, novels, films and popular magazine stories, it was in truth Malone's findings about the chronology of Sally's pregnancies that constituted the most tangible piece of new evidence to support the charge of a sexual liaison.

Where does that leave the matter? Well, unless the trustees of the Thomas Jefferson Memorial Foundation decide to exhume the remains and do DNA testing on Jefferson as well as some of his alleged progeny, it leaves the matter a mystery about which advocates on either side can freely speculate, and surely will. Within the scholarly world, especially within the community of Jefferson specialists, there seems to be a clear consensus that the story is almost certainly not true. Within the much murk[dotlessi]er world of popular opinion, especially within the black community, the story appears to have achieved the status of a self-evident truth. If either side of this debate were to file for damages in a civil suit requiring a preponderance of evidence as the standard, it is difficult to imagine an impartial jury finding for either plaintiff. Jefferson's most ardent defenders still live under the influence of what might be called the Virginia gentleman ethos (i.e., this is not something that a Virginia gentleman would do), which increasingly has the quaint and charmingly naive sound of an honorable anachronism. Meanwhile those who wholeheartedly endorse the truth of the story, either in Callender's original version as a tale of lust and rape, or in Brodie's later rendering as a tragic romance between America's premier biracial couple, have also allowed their own racial, political or sexual agenda to take precedence over the evidence. On the basis of what we know now, we can never know.

This means that for those who demand an answer the only recourse is plausible conjecture, prefaced as it must be with profuse statements about the flimsy and wholly circumstantial character of the evidence. In that spirit, which we might call the spirit of responsible speculation, after five years mulling over the huge cache of evidence that does exist on the thought and character of the historical Jefferson, I have concluded that the likelihood of a liaison with Sally Hemings is remote.

Two pieces of circumstantial evidence strike me as telling: First, Sally's last two children, Madison and Eston, were born after Callender's charges created the public scandal in 1802, and it is difficult to believe that Jefferson would have persisted in producing progeny with Sally once the secret had been exposed and the Federalist press was poised to report it; second, among Jefferson's contemporaries neither Alexander Hamilton nor John Adams, both of whom were political enemies who undoubtedly enjoyed the sight of their chief adversary being stigmatized by the kind of innuendo he had spread against them, found it possible to believe that Callender's accusations were true. Nor, for that matter, did Henry Adams, whose critical appraisal of Jefferson's character established the scholarly standard for ironic dissection.

What Hamilton and both Adamses understood about Jefferson, and what my own immersion in the historical evidence has caused me to conclude as well, is that for most of his adult life he lacked the capacity for the direct and physical expression of his sexual energies. Henry Adams put it most explicitly when he said that Jefferson's temperament was "almost feminine." When scholarly defenders like Dumas Malone and Merrill Peterson claimed that Jefferson was "not the kind of man" to engage in illicit sex with an attractive mulatto slave, they were right for reasons that went deeper than matters of male gallantry and aristocratic honor. Jefferson consummated his relations with women at a more rarefied level, where the palpable realities of physical intimacy were routinely sublimed to safer and more sentimental regions. He made a point of insulating himself from direct exposure to the unmitigated meaning of both sex and slavery, a lifelong tendency that an enduring liaison with Sally Hemings would have violated in ways he found intolerable. He obviously knew about the ongoing miscegenation at Monticello, but his powers of self-deception and denial protected him from facing these facts, and his urge to remain oblivious was considerably stronger than his sexual drive.

He was, to be sure, capable of living with massive contradictions, but his psychological dexterity depended upon the manipulation of interior images and personae; he was not that adroit at the kind of overt deviousness required to sustain an allegedly thirty-eight-year affair in the very center of his domestic haven. One could plausibly argue that his famous fear of racial mixing "adhered its intensity from his personal encounters with its sensual attractions." But nothing that we know about Jefferson supports the linkage between sex and sensuality. His most sensual statements were aimed at beautiful buildings rather than beautiful women. In sum, the alleged relationship with Sally Hemings, if it did exist, defied the dominant patterns of his personality.

It certainly did not defy the dominant patterns of racial interaction at Monticello or the slave quarters of many plantations throughout the antebellum South. The fact that only a small fraction of the black population in contemporary America is of purely African extraction constitutes the most palpable evidence that sexual relations between blacks and whites have a long, if secret, history. At the purely metaphoric level, then, the story of an enduring sexual connection between Jefferson and Sally Hemings conveys an elemental truth about the hidden history of race relations in America. At the same illustrative level, the very possibility of such a liaison accurately captures the inherently promiscuous character of Jefferson's elusive personality, especially when it encountered unpalatable realities. But metaphoric or illustrative versions of the truth are not the same thing as historical truth per se. For those predisposed to disagree with my conjectures about the unlikelihood of the relationship, the most disquieting piece of evidence they will need to absorb also relates to timing. For if the story is true, Jefferson's withdrawal from a leadership position in the antislavery movement began at the same time that the affair with Sally Hemings allegedly started.

For those who wish to explore the evidence pro and con, the best summary of the case against the alleged liaison is Virginia Dabney, The Jefferson Scandals: A Rebuttal (New York, 1981). The best synthesis of the other side is the book, by Annette Gordon-Reed, entitled Thomas Jefferson and Sally Hemings: Historians end the Enterprise of Defense, soon to be published by the University of Virginia Press, which represents an attempt to enhance the credibility of Fawn Brodie's interpretation in Intimate History without recourse to Brodie's psychiatric paraphernalia. It is worth noting that Gordon- Reed, who graciously allowed me to read her manuscript in process, is herself an agnostic on the question. So is Lucia Stanton of the Thomas Jefferson Memorial Foundation, who supervises the ongoing research into the Hemings descendants and probably knows more about the evidence than anyone else.