Edmund Bacon’s statement was not that he observed “_______” leaving Sally Hemings’ room at the time that Harriet was conceived. His statement is perfectly consistent with a conclusion that, having witnessed a man other than Thomas Jefferson leaving Sally’s room early in the morning on numerous occasions, that man was obviously Sally’s “lover” and presumably fathered her children. Bacon was certainly aware of the Callender charges and resulting rumors that Thomas Jefferson began a relationship with Sally in Paris that produced several children, and upon observing a different man leaving her room time and again—irrespective of whether this was in 1800 or after 1806—it was reasonable for him to conclude the charge against Jefferson was unfounded. While there may be some temporal ambiguity in his remark (which, we should keep in mind, was recorded by another), there is not the slightest hint of intentional deception or other dishonesty.
Far more important, our inquiry is not primarily about Harriet Hemings. There is no compelling17 evidence — DNA or otherwise—suggesting that Thomas Jefferson was the father of Harriet Hemings. The only Hemings child scientifically tied to a Jefferson father was Eston, who was conceived the year after everyone acknowledges that Edmund Bacon had assumed the duties of Monticello overseer. Edmund Bacon was born and grew up two or three miles from Monticello,18 and his brother William was overseer during Jefferson’s stay in Paris.19 Edmund worked for Thomas Jefferson in other capacities before becoming overseer in 1806. It is certainly possible that he did visit Monticello about the time Harriet was conceived and saw another man leaving Sally’s room. But it is also possible that his observations began in 1806 or 1807, and if so they are tremendously more valuable in the search for the father of Eston Hemings.
Finally, Edmund Bacon’s testimony is critically important because, if true, it totally destroys a fundamental assumption upon which the revisionist case is premised. If Edmund Bacon personally witnessed a man other than Thomas Jefferson leaving Sally Hemings’ room time and again in the early morning hours, it is a reasonable conclusion that his purpose was not to sell her encyclopedias or insurance, but to engage in sexual relations. And if that is so, either Sally Hemings was not monogamous, or she did not have a sexual relationship with Thomas Jefferson. This is obvious.
Before leaving the issue of Edmund Bacon’s recollections, it may be worth noting that both Professor Gordon-Reed and Monticello historian Lucia Stanton appear to attempt to avoid the inconsistency between Bacon’s comment that he put Harriet on a stage to Philadelphia and Madison’s assertion that Harriet lived in Washington, D.C., by paraphrasing Bacon’s account (making no reference to “Philadelphia”) and asserting that Harriet was “put on a stagecoach to freedom in the North at her father’s direction.”20
If Sally was not sexually involved with Thomas Jefferson, then our inquiry is at an end. If we assume she may have been involved with Thomas Jefferson, then she could not have been monogamous—and without that assumption, the “evidence,” such as it is, pointing to Thomas Jefferson as the father of Eston Hemings largely falls apart. As will be discussed, there is strong evidence that at least six other Jefferson males may have been present at Monticello when Eston was conceived. For a variety of reasons that will also be discussed, at least five of them would seem to be more likely suspects to have fathered Eston Hemings than President Thomas Jefferson.
Other than Madison Hemings’ allegation (about facts that occurred prior to his birth that he obviously could not have known with certainty), there is no reason to believe that Sally Hemings was monogamous. Madison reportedly admits that Sally’s mother had children by no fewer than four different men,21 and it is not at all clear that slave women were given a great deal of choice about their sexual partners. Nor is the case made by the Monticello Report for Sally’s monogamy persuasive:
While the DNA results bear only on the paternity of Eston Hemings, the documents and birth patterns suggest a long-term relationship, which produced the children whose names appear in Jefferson’s records. Even the statements of those who accounted for the paternity of Sally Hemings’s children differently (Thomas Jefferson Randolph, Ellen Randolph Coolidge, and Edmund Bacon) never implied that Hemings’s children had different fathers. Full-sibling relationships are further supported by the closeness of the family, as evidenced by documentation of siblings living together and naming children after each other.22
Some of the shortcomings of these strained arguments have already been addressed. Such unfounded suppositions hardly compare to the eyewitness account of Edmund Bacon. But they are necessary for the revisionist case, because without the assumption that Sally was monogamous the case against Thomas Jefferson quickly crumbles.
Professor Gordon-Reed attempts to discredit the Bacon testimony by asserting that he was motivated by an “interest in protecting his former employer’s reputation. … ”23 The only apparent basis for such an “interest” would be that Bacon genuinely admired and respected Thomas Jefferson, which presumably would be a consequence of evaluating everything he knew about the man. He said in 1862:
I am now in my seventy-seventh year. I have seen a great many men in my day, but I have never seen the equal of Mr. Jefferson. He may have had the faults that he has been charged with, but if he had, I could never find it out. I don’t believe that, from his arrival to maturity to the present time, the country has ever had another such a man.24
Bacon clearly appears to have respected Jefferson, but this does not establish a “bias” likely to affect his veracity. To suggest otherwise would be akin to saying that an individual who enjoyed a book or movie cannot be trusted to review it, as she or he will normally lie to others if asked if there were flawed passages. The kinds of biases that could undermine Bacon’s credibility might result from things like his still being employed by Jefferson’s family or having financial involvement with them. Edmund Bacon certainly may have lied, but there is no reason to assume that from what we know about the situation. He had less apparent reason to be biased than did Madison Hemings or any of Thomas Jefferson’s relatives.
I raised this issue with Dr. Daniel Jordan, president of the Thomas Jefferson Memorial Foundation, and he provided a response by Lucia Stanton which stated in part:
For many who have finally concluded that Thomas Jefferson was the most probable father of Sally Hemings’s children, Bacon’s account is the hardest bit of evidence to assimilate. So one needs to ponder what Bacon’s motives for misrepresenting the situation might have been. He too can be viewed as having “a stake in the outcome,” out of deep loyalty to Jefferson or pride in his association with the famous President. Then there are the circumstances of the interview to consider. Bacon was talking to a preacher in mid-Victorian 1860. …
For the Monticello committee, the issue boiled down to one lone voice against a chorus. A single dissenting opinion was not of sufficient weight to negate the preponderance of evidence on the other side.25
This sounds very much like the committee itself, having “finally concluded that Thomas Jefferson was the most probable father of Sally Hemings’s children,” doubled back to speculate “what Bacon’s motives for misrepresenting the situation might have been.” A more reasonable and objective starting point might have been to consider whether Bacon was misrepresenting the facts. It may be true that Bacon was “talking to a preacher in mid-Victorian 1860”—though it is also true that the minister in question was president of Cumberland College in Kentucky—but that did not prevent Bacon from discussing the young men who visited Monticello and became “too intimate with the Negro women” to suit one young visitor.26
Where is this “chorus” of witnesses alleging that Sally Hemings and Thomas Jefferson had a sexual relationship? As far as we know, there is not a single witness—among the hundreds of Monticello residents and perhaps thousands of visitors—who has left behind testimony that the two of them were seen alone together or in any kind of suggestive behavior. The “evidence” for the existence of the relationship consists of asse
rtions attributed to Madison Hemings concerning events that occurred long before he was born, surrounded by statements that are demonstrably untrue, and reportedly corroborated by Israel Jefferson whose account is clearly false on numerous material points. More importantly, unlike Edmund Bacon, Madison is not testifying to facts he could have personally observed. He is merely passing on assertions as they “came down to” him. Perhaps they came from Sally Hemings, perhaps from James Callender, or perhaps from others. We have no way of knowing.
Excluding Madison’s problematic statement, there is no evidence that Sally Hemings or any of her other children ever claimed that Thomas Jefferson was their father. When former Monticello slave Isaac Jefferson told his memoirs, he made mention of Sally Hemings but said not a word about any alleged sexual relationship with Thomas Jefferson. None of the other residents of Monticello claimed knowledge of this relationship, which was consistently denied by the Jefferson family. Where is this “chorus of voices” or “preponderance of evidence” on the other side? Presumably it is James Callender, John Hartwell Cocke, Thomas Gibbons, and Elijah P. Fletcher (as discussed in Chapter Nine)—none of whom even alleged they had personally witnessed any behavior by Thomas Jefferson or Sally Hemings that might support the allegation. Indeed, not one of them even alleges he was told of the relationship by someone who had witnessed anything more suspicious than the existence of light-skinned slaves at Monticello.
Finally, the statement that a “single dissenting opinion was not of sufficient weight to negate the preponderance of evidence on the other side” misses the true significance of the Bacon statement. To be sure, he does voice his opinion that Thomas Jefferson was not the father of Harriet Hemings. But that is not the critical part of his testimony. He states further that he frequently witnessed, with his own eyes, a man other than Thomas Jefferson coming out of Sally Hemings’ room in the early hours of the morning. That is not “opinion,” it is an assertion of fact—and it is probably the most reliable factual testimony we have to guide us.
Madison Hemings and others who allege that the relationship existed are discussing facts; but they are in reality merely voicing opinions about whether certain factual statements are true. Not one of them is testifying to material facts he personally saw, heard, or otherwise observed. It simply does not follow that a half dozen opinions by people with no personal knowledge of the key facts are of greater probative value than eyewitness testimony by one credible observer.
Nor, for that matter, is it close to being true that Edmund Bacon was “one lone voice” among Jefferson’s contemporaries in this matter. While Ms. Stanton is correct in observing that the Jefferson family explanation that Sally Hemings’ children were fathered by Peter and/or Samuel Carr has been shown by the DNA tests to have not been correct in the case of Eston Hemings, there is no reason to believe that they were wrong about some or all of Sally’s earlier children. They all had far more contact with Thomas Jefferson than any of the advocates on the other side (many if not most of whom had never set foot on Jefferson’s mountain); and while, as relatives, their testimony must be considered potentially biased, unless they were all lying their statements are probably of greater value than those of people who did not have an opportunity to witness events at Monticello over the decades.
As shown in Chapter Nine, even without removing the sources relied upon by Ms. Stanton and her colleagues who were clearly bitter enemies of Thomas Jefferson (like James Callender and Thomas Gibbons), and those who received their information on this matter from Jefferson’s enemies (like Elijah Fletcher), Jefferson’s contemporaries who voiced doubts about the accusations easily outnumber his critics. This list includes James Madison, Jefferson’s physician Dr. Robley Dunglison, Elder John Leland, Thomas Paine, Jefferson’s secretaries David Humphreys and William Burwell, and even prominent Federalist political opponents like John Adams, Alexander Hamilton, and General Light-Horse Harry Lee. Indeed, given the antipodal reputations of James Callender and Thomas Jefferson, it is likely that the overwhelming majority of Thomas Jefferson’s friends found the charge so preposterous as to not even warrant comment. Sadly, from Ms. Stanton’s response, it seems that the Monticello research committee began their inquiry having concluded that Thomas Jefferson was guilty.
The Accounts of Thomas Jefferson’s Relatives
It is perhaps not surprising that Thomas Jefferson’s children and grandchildren defended his innocence with respect to the paternity of Sally Hemings’ children. This could easily be explained as nothing more than family loyalty, and from all accounts they loved him dearly. Professor Gordon-Reed tells us their views “must be taken with a grain of salt” because “[t]hey loved him.”27
Nevertheless, their accounts ought not be lightly dismissed. For if this relationship had been going on regularly for decades, it is difficult to believe that they would not have learned about it. This is particularly true in the early days, when Thomas Jefferson is alleged to have begun a sexual relationship with a fourteen- to sixteen-year-old Sally Hemings in Paris. Sally was the maid to Jefferson’s two young daughters, and it seems unlikely that she would have had the discretion to keep such a relationship secret from them. Indeed, presumably it would have been in her interest for them to know that her status had been elevated from that of lowly “servant” to the beloved “partner”—to use the modern vernacular—of their powerful father. This point will be developed further in the next chapter.
Professor Joshua Rothman, who wrote his Ph.D. dissertation about race relations in antebellum Virginia, writes:
In early national and antebellum Virginia, standing sexual affairs between white men and African American women were nearly always open secrets. Divorce petitions in Virginia involving accusations of interracial adultery, for example, amply demonstrate that neighbors, friends, and relatives …always knew, sometimes for many years, about the illicit sexual conduct of both men and women in their families and communities.28
It is thus very difficult simply to assume that Jefferson’s relatives would have been ignorant of a relationship that allegedly lasted for decades.
Martha Jefferson Randolph
Martha (Patsy) Jefferson was by all accounts an unusually bright and perceptive young woman, and in terms of her values an attractive person.29 Shortly before Sally and Maria (Polly) arrived in Paris, Martha wrote to her father that “I wish with all my soul that the poor Negroes were all freed. It grieves my heart when I think that these our fellow creatures should be treated so terrible as they are by many of our country men.”30
She later obviously became very much aware of the allegations made against her father, but she does not appear to have believed them. Professor Joseph Ellis, for example, clearly found it difficult to reconcile his study of Martha Jefferson’s unpublished papers with his own newfound conviction that her father must have fathered children by Sally Hemings. Writing in the William & Mary Quarterly, he explains:
My own reading of those [Martha Jefferson Randolph’s] papers suggests that, much like her father, Martha Randolph possessed extraordinary powers of denial. She did not consciously cover up the ongoing sexual liaison so much as convince herself it did not exist. (How she managed this defies logic, but not in its new “Jeffersonian” version.)31
Perhaps one need not invent a new theory of “Jeffersonian logic” to explain what has been observed. Occam’s Razor would suggest that these same consequences might be explained by the conclusion that Thomas Jefferson and his daughter were not “in denial” at all, but rather they were simply aware that there was no truth behind the Callender allegations.
Consider also this account, from a letter from Jefferson biographer Henry S. Randall, to his colleague James Parton, written on June 1, 1868:
Mr. Jefferson’s oldest daughter, Mrs. Gov. Randolph, took the Dusky Sally stories much to heart. But she never spoke to her sons but once on the subject. Not long before her death she called two of them—the Colonel [Thomas Jefferson Randolph] and George Wythe Randolph—t
o her. She asked the Colonel if he remembered when ____ Henings [sic] (the slave who most resembled Mr. Jefferson) was born. …He said that he could answer by referring to the book containing the list of slaves. He turned to the book and found that the slave was born at the time supposed by Mrs. Randolph. She then directed her sons[’] attention to the fact that Mr. Jefferson and Sally Henings [sic] could not have met—were far distant from each other—for fifteen months prior to such birth. She bade her sons (to) remember this fact, and always to defend the character of their grandfather. It so happened when I was afterwards examining an old account book of the Jeffersons I came pop on the original entry of this slave[’]s birth; and I was then able (to know) from well known circumstances to prove the fifteen months separation …but those circumstances have faded from my memory.32
Now this account goes from Jeff Randolph through Henry Randall to James Parton—triple hearsay—and is subject to all of the dangers characteristic of such evidence. Martha Randolph could have been mistaken or knowingly not telling the truth. Her son Jeff could have fabricated the story or confused the details after many years. Henry Randall could have created it from whole cloth. Even James Parton could theoretically have fabricated the incident.
But if the basic account is accurate, Martha Randolph’s statement has some inherent credibility. If she believed her father guilty of the charges, and simply wanted to urge her sons to protect his memory, she would presumably not have asked Jeff Randolph to accept her account based upon his own recollection of certain facts. Because if Sally and President Jefferson had not been separated for fifteen months, Jeff Randolph would not be able to confirm her statement from his own knowledge. She could just as easily have fabricated some explanation that was not subject to confirmation (or, if she was lying, refutation) from his independent memory.
The Jefferson-Hemings Controversy Page 37