The Eston Hemings Family Oral Tradition Asserted
Thomas Jefferson Was Not Eston’s Father
Most significantly, the oral history from the four interviewed descendants of Eston Hemings—the only child of Sally Hemings indicated by DNA testing to have almost certainly been fathered by someone carrying the Jefferson DNA marker—is more consistent with a conclusion that Eston was fathered by the President’s younger brother Randolph or one of Randolph’s five sons.
The oral traditions of Eston Hemings’ descendants are explained in this quotation from Julia Westerinen in the book Jefferson’s Children:
I was told we were related to Jefferson’s uncle; that’s what was handed down in the family. I don’t know who started that lie. …It was 1975, after Fawn Brodie’s Thomas Jefferson: An Intimate History came out. My aunt recognized Eston’s name from our family tree and she called the author. Fawn did the research, came east and interviewed us, and said, “You’re directly descended from Thomas Jefferson, but through this woman named Sally Hemings, who was his slave.” So that’s when we first found out about it.4
Ms. Westerinen made the same point in an interview with People magazine in late 1998: “My parents told us we were related to Thomas Jefferson’s uncle, which made us cousins removed several times.”5
Now, it is clear that Eston could not have been fathered by either of Thomas Jefferson’s paternal uncles, as both had been dead for several decades when Eston was conceived. But there was a frequent Jefferson visitor to Monticello who was known as “Uncle” and who thus would seem to be a prime suspect when one considers both the DNA tests and the oral history of Eston’s family. For example, in a January 30, 1800, letter to her father, Martha Jefferson Randolph referred to the President’s brother as “Uncle Randolph,”6 and about four months before Eston was conceived, Thomas Jefferson received a letter from his granddaughter, Ellen Randolph, written from her family’s nearby home at Edgehill, which noted: “Uncle Randolph Jefferson is with us. … ”7
It thus appears that the oral history concerning Eston’s paternity is in error. He was not fathered by an uncle of Thomas Jefferson. One of the few things that we know with reasonable certainty about Sally Hemings is that she spent at least part of her life serving Martha and Maria, Thomas Jefferson’s two daughters who survived to adulthood. To Maria (who had passed away prior to Eston’s birth) and to Martha, Thomas Jefferson’s brother was “Uncle Randolph.” Professor Brodie reminds us that Madison and Eston would “have remembered no years without Martha Randolph in total charge”8 at Monticello.
Randolph was also known as “Uncle Randolph” by Thomas Jefferson’s grandchildren, and it would seem reasonable that Sally Hemings would also have identified him by this relationship. The DNA tests tell us that Randolph Jefferson was one of at least two dozen men, in addition to President Thomas Jefferson, who were equally as likely (considering only the DNA evidence) to have fathered Eston Hemings.
We also have reliable testimony by former slave Isaac Jefferson that Randolph was fond of spending his evenings at Monticello playing his fiddle with the slaves and dancing “half the night.”9 We have the oral history of Eston’s family asserting that his father was a Jefferson “uncle.” We have a letter from Thomas Jefferson, dated August 12, 1807, inviting Randolph to visit Monticello—and this was about a week before the start of the three-week window during which it is estimated that Sally conceived Eston. None of this, of course, proves that Randolph was Eston’s father; but, to the extent it is of value, the Eston Hemings family oral history obviously supports Randolph’s paternity far more than it does Thomas Jefferson’s. (It also shows, as discussed in other chapters, that Professor Gordon-Reed and others should have considered Randolph Jefferson in their studies.)
Of course, if one assumes that Professor Brodie was correct in her conclusion that Eston was, in fact, the son of Thomas Jefferson, then perhaps she performed a valuable service by so informing Eston’s descendants. But in so doing, she obviously corrupted the oral history process. Descendants of Eston Hemings who now proudly tell their children they are related to President Jefferson are not passing on eight generations of family “oral tradition,” but rather the far more recent conclusions of an unrelated Caucasian historian. This does not make their story wrong, but it should exclude it from consideration as legitimate Hemings family “oral history.” If their “source” is Professor Brodie, we should focus our inquiry on her scholarship.
We are now told that the oral history of Eston Hemings’ family was corrupted on a previous occasion, about a century ago, when white descendants decided to conceal their connection to a black slave woman. As recounted in the Monticello Report:
For Eston Hemings Jefferson’s descendants, the story of connection to Thomas Jefferson also remained alive, altered to protect their passing into the white world. They heard that they were descended from Jefferson’s uncle, and Eston Hemings’s name and the places the family had resided were changed, in order to sever their connection with Sally Hemings and African Americans.10
Two questions about this story come readily to mind:
First, one has to wonder why it was necessary to delete the name of the famous Thomas Jefferson in order to preserve a heritage of racial purity.11 Thomas Jefferson was not black, and excluding his name would be a rather drastic approach to “purifying” family genealogy. A far simpler and less speculative explanation for the original oral history would be that the story was true—that Eston Hemings was the son of a Jefferson family “uncle” who visited Monticello. And that explanation is far easier to reconcile with what we know about Thomas and Randolph Jefferson.
Second, and perhaps more important, one has to wonder how this century-old corruption of the family’s oral history would suddenly come to light now? There is abundant testimony on the public record from Eston’s descendants claiming the family history was “corrected” following a visit by Fawn Brodie in the mid-1970s. If the oral history prior to that time had included information about the earlier modification (“Honey, now that you are about to graduate from high school I want to tell you something important; you are the great, great granddaughter of Thomas Jefferson—except you are to tell your children it was his uncle . . .”), they would not have believed the “uncle” story before Fawn Brodie came calling to “correct” the record. This new allegation about an earlier corruption seems clearly to be someone’s creative speculation in an attempt to discredit the oral history of Eston’s family in order to strengthen their post-DNA claim to be descendants of President Jefferson. It serves as an excellent example of the weakness of oral traditions, as at any time as the story is passed down to a new generation it can be altered as the storyteller wishes.
Later, in response to one of my questions, Monticello historian Lucia Stanton endeavored to clarify the belief that the Eston Hemings oral history was modified prior to Professor Brodie’s involvement. In early December, 2000, she wrote:
Subsequent to the Report, we have learned that the oral history of Eston Hemings’s descendants was “altered” about sixty, rather than a hundred, years ago. Note that when Eston’s son Beverly Jefferson died in 1908, he was described in a newspaper as the grandson of Thomas Jefferson. In the 1940s Eston Hemings’s great-grandsons created a family tree that included descent from Thomas Jefferson’s uncle and completely omitted the Ohio portion of their ancestors’ lives. We think it is significant that this family passed on the story of descent from Thomas Jefferson for over a century, at probable risk to their lives as “passing” members of the white community (it would not have been hard for someone to learn that Jefferson had no surviving sons by his wife, Martha).
Candidly, the reason I had asked her to explain the basis of the allegation that the family history had been “altered” was that I suspected it might have been supposition based upon the fact that a friend of one of Eston’s sons had written to a newspaper following his death that Eston was Thomas Jefferson’s son. Is “oral history” regard
ed as so inherently suspect that a single letter to the editor, by an unknown writer who is not a family member and who provides not the slightest reason to explain the basis of his assertion, is automatically considered superior authority? One can only imagine the reaction of Hemings descendants if someone sought to introduce a comparable document in opposition to their claim. Nor, for that matter, was it relevant to anything that someone might have learned that Martha Jefferson “had no surviving sons.” Eston’s descendants would have equally been Thomas Jefferson’s descendants had they traced their lineage to Martha or Mary.
Might not this 1908 letter to the editor be equally explained in other ways? The allegation that Thomas Jefferson fathered children by Sally Hemings was hardly a great secret. Perhaps Beverly joked about it with one of his friends, or perhaps his friend knew of the rumors from other sources and elected to assert it at the time of Beverly’s death to make his deceased friend seem more important. It might be noted that a Madison, Wisconsin, obituary was entitled “Colonel Beverly Jefferson, Veteran Hackman.”12 Hemings family genealogist Judith Justus, in her book Down from the Mountain: The Oral History of the Hemings Family, asserts that military records reveal that Beverly Jefferson spent only three months in the Army after enlisting during the Civil War.13 Without explanation, she concludes that the word “Colonel” was likely a “courtesy” title; but in the absence of evidence to the contrary it might also suggest that Beverly Jefferson had embellished his background to impress his associates—perhaps telling some he was a former colonel in the Army and also the grandson of Thomas Jefferson.14 We have no way of knowing whether this is true, or not. The point is, there are many very simple and credible explanations for the letter to the editor without having to dismiss the Eston Hemings / Jefferson family (pre-Brodie) oral history; but, of course, in this instance that history is an impediment to the claim that Eston was Thomas Jefferson’s son.
One really need not resolve this dilemma, as alleged “oral history” that has clearly been corrupted once (by Professor Brodie), and possibly twice, is entitled to little weight. Further, the only way this history would be meaningful as evidence would be if Eston himself had been part of the chain. The fact that Eston does not appear to have ever clearly asserted15 that Thomas Jefferson was his father increases the risk that, if Beverly Jefferson or his family really did believe he was Thomas Jefferson’s grandson, the story might have entered the family traditions as an indirect result of someone having read the widely publicized articles of James T. Callender (or perhaps the account attributed to Madison Hemings16) and accepted them as true.
As already noted, Eston Hemings could not have been fathered by Thomas Jefferson’s “uncle.” Thomas Jefferson’s only paternal uncles, Thomas and Field Jefferson, died respectively in 1723 and 1765, and the DNA tests have ruled out paternity by a maternal uncle. Eston was not conceived until 1807. However, in keeping with the highly speculative spirit of this inquiry by advocates of Thomas Jefferson’s paternity, it is easy to imagine ways in which such an oral history story might have been started. For example:
President Jefferson’s brother Randolph was the paternal uncle of daughters Martha and Maria Jefferson, and thus was referred to as “Uncle Randolph” when he visited Monticello. It is not difficult to conceive how Eston might have been told that “Uncle Randolph” was his father. As the story was passed from generation to generation, with many years sometimes passing between each retelling, it would not be surprising for someone to forget the name Randolph and to recount only that Eston was the son of “Thomas Jefferson’s uncle.”17
Another possible explanation is that Eston was the son of Uncle Randolph’s third son, Peter Field Jefferson,18 who was sometimes known as “Field” and is thought to have been about 21 or 22 years old at the time of Eston’s conception. He presumably would have accompanied the rest of his family to visit Monticello about the time Eston was conceived. (As already mentioned, we know Randolph was invited to visit about that time.19) This third son of Randolph shared the name Field with Thomas Jefferson’s youngest uncle, who died decades before Eston was conceived. Over several generations, the word might have been passed down that “Field Jefferson” was Eston’s father. At some point, a curious descendant may have decided to do some genealogical research and discovered that “Field Jefferson” was Thomas Jefferson’s uncle, and may then have confused two individuals having the same name.
There is not the slightest bit of direct evidence to support either of these hypotheses, which places them in the same category with most of the other speculation that has driven “scholarship” on this topic in the absence of reliable historical data. The hypothetical examples prove nothing, but they do show how the oral history that Eston Hemings was a descendant of Thomas Jefferson’s “uncle” might have developed. Similarly, the “or cousin” part of the story20 might have emerged out of the confusion resulting from someone researching the family genealogy and realizing that both of Thomas Jefferson’s uncles died decades before Eston Hemings was conceived—and thus the legend that he was fathered by Jefferson’s “uncle” was clearly in error. If Eston’s father was not an uncle, they might have speculated, perhaps he was a cousin. Whatever the real explanation, the fact that Eston Hemings’ descendants passed down for generations the story that he was not the son of Thomas Jefferson is significant. Human nature suggests that people are far more likely to alter or embellish their “family tree” to claim descent from a distinguished individual than to conceal such a relationship.
Although it will not likely be a popular conclusion at Monticello, we might add that the oral history of the twenty-two descendants of Madison Hemings who were interviewed in the “Getting Word” project is also of little value to our inquiry. We know21 that Madison Hemings claimed that Thomas Jefferson was his father, and thus—unless the oral history adds more details—no benefit is gained by showing that Madison also told his children, they told their children, and so on for generations. If Madison was mistaken about, or intentionally misrepresented, the truth, or if Samuel Wetmore intentionally or inadvertently altered Madison’s account, the fact that others repeated this story for generations in good faith adds nothing to the probative value of his original account.
Shortcomings of Oral Traditions
A more general comment may be warranted about the inherent shortcomings of “oral history” (sometimes called “oral tradition” to distinguish it from verbal testimony later taken down from actual witnesses to historic events) as evidence—and this obviously applies equally to accounts passed down by white and black families. The reality was captured by Professor Jan Lewis, of Rutgers University—co-editor of the volume Sally Hemings and Thomas Jefferson, and a true believer that Jefferson fathered Sally’s children—in her introduction to a forum on this topic that appeared in the prestigious William & Mary Quarterly22 in January 2000. Discussing a contribution by Monticello’s Lucia Stanton, Professor Lewis writes: “Stanton reminds us that oral history is not intrinsically reliable but must be checked against other sources.”23
The very popular genealogy software “Family Tree Maker,” by Broderbund, provides this cautionary note to users evaluating stories passed down through the generations:
Tall Tales
Family tradition is another potential source of erroneous genealogical information. Frequently, such tradition holds that the family is descended from some particularly noteworthy or interesting ancestor, such as a high noble in England or a French Huguenot who fled persecution. In many of these cases, the relationship is fictitious, arising as the family history was passed down and embellished through the generations. In more extreme cases, the family may not have any roots in the supposed ancestor’s home country. Similar inaccuracies may also surround the history of the family surname and the deeds of various ancestors.
Once again, using a variety of sources to piece together the family tree is crucial for ensuring accuracy. Although family traditions may serve as useful clues for directing y
our search, they should not be taken at face value.24
In his own autobiography, Thomas Jefferson began by recounting that the “tradition of my father’s family was, that their ancestor came to this country from Wales, and from near the mountain of Snowden, the highest in Great Britain.” Then, noting that his father married into the Randolph family, he reports: “They trace their pedigree far back in England and Scotland, to which let every one ascribe the faith and merit he chooses.”25
Writing in History Now in 2004, Professor Gordon-Reed—while accusing white historians of racism for having given more weight to statements made to historians by Jefferson’s grandchildren than to the 1873 account attributed to Madison Hemings (discussed in Chapter Four)—argued that it was necessary “to subject claims made by masters and those made by slaves or the children of slaves to the same rigorous standards of proof.”26 One cannot quarrel with that statement. But even Professor Gordon-Reed recognizes a difference between “modern-day forms of oral history interviews in which scholars pose questions and record the answers of eyewitnesses to historical events,” and what she calls “oral history tradition” in which “stories are passed privately and informally from one generation to another.” She adds: “Scholars have been slow to accept oral tradition as historical evidence, for they are aware that stories can be changed or embellished as they pass from one generation to another. This concern that a story told to many people across decades may produce error cannot be lightly dismissed.”27 Yet one of her three “lessons” to be learned from the Jefferson-Hemings story, she tells us, is that “we must not privilege one form of evidence over another.”28
The Jefferson-Hemings Controversy Page 30