In December 1799, Sally may17 have had another daughter—possibly named Thenia—but the details are sketchy. This child may have died as an infant, and no physical description is known to have survived.18
In May 1801, Sally gave birth to “Harriet II,” who lived to adulthood and left Monticello in 1822. The only surviving descriptions of Harriet asserted that she was “white enough to pass for white,”19 “nearly as white as anybody,”20 and “very beautiful.”21 Once again, the sketchy descriptions are of little value in establishing her paternity beyond suggesting that her father may well have been a Caucasian. (Neither of Jefferson’s daughters by his wife who lived to adulthood was widely described as being physically “beautiful.”)
On January 19, 1805, Sally gave birth to Madison—the only one of her children believed to have left behind any kind of written biographical statement.22 Madison is the only one of Sally Hemings’ known children to have clearly claimed to be the son of Thomas Jefferson, so his case is particularly important. Unfortunately, Madison left behind no known photographs, sketches, or other detailed evidence of his physical appearance. He appears to have been darker in color than some of Sally’s other children,23 and was several inches shorter than his younger brother Eston. The issue of Madison’s height illustrates the uncertainty of the evidence in this matter. While Thomas Jefferson was generally said to be 6 feet, 2H inches tall,24 the official Albemarle County records gives Madison’s height at age twenty-six as 5 feet, 7K inches,25 a difference of greater than 7 inches. But Madison’s biographer—an anti-Jefferson Republican Party activist who is the source of most of what we are told is true about Madison—alleges Madison Hemings was 5 feet, 10 inches tall at age sixty-eight.26
Sally Hemings’ last known child was born on May 21, 1808, and was named Eston. He is her only child who has been linked by DNA testing to a male Jefferson father, and he may have had the closest physical resemblance to the President. Albemarle County records describe Eston in 1831 as “6 feet one inch high, Bright Mulatto.”27 Ohio journalists described him later as “very slightly colored, of large size”28 and “of a light bronze color, a little over six feet tall, well proportioned, very erect and dignified”29—asserting that “his nearly straight hair showed a tint of auburn, and his face, indistinct suggestion of freckles.” (Thomas Jefferson also had auburn hair and freckles.) Eston was also said to be “decidedly intelligent.”
There is also a report that two residents of Eston’s community in Ohio visited Washington, D.C., in the 1840s (approximated twenty years after Jefferson’s death), and upon observing a statue of the late President (see Figure 9 on the next page) remarked that it reminded them of Eston Hemings.30 There is no suggestion that either man had ever actually seen Thomas Jefferson, and the Monticello Report asserts that “local rumors” in Ohio at the time alleged Eston was Jefferson’s son.31 The alleged observers may also have been influenced by the fact that the statue in question was made of bronze, which was the term used to describe the color of Eston Hemings’ skin.
We do not have a specific date for this incident, but it was later reportedly mentioned to Eston—who had moved away from Ohio in 1852 at the age of forty-four—and the statue had been moved from the Capitol to in front of the White House five years earlier. Thus, they were presumably comparing the statue to an Eston Hemings in his late thirties or early forties.32
Figure 9. Statue of Thomas Jefferson in Washington, D.C., said to resemble Eston Hemings.
French sculptor Pierre-Jean David D’Anger had never personally seen Thomas Jefferson and made the statue in question entirely on the basis of a painting belonging to Lafayette that had been made in 1821—shortly before Jefferson’s seventy-eighth birthday.33 Thus, the Ohio visitors to Washington, D.C., were drawing a comparison between Eston Hemings and a statue of a man nearly twice Eston’s age made by a sculptor who had never set eyes on his subject.
Making an accurate, three-dimensional statue from a single, two-dimensional painting is no easy task, and several people who had known Thomas Jefferson complained that the statue was not a good likeness.34 The statue had been commissioned six years after Jefferson’s death by an independently wealthy American naval officer named Uriah Phillips Levy, who admired Jefferson greatly and offered the statue as a gift to Congress. Several legislators complained that it was not a good likeness of the former President, but Congress ultimately chose to accept the gift. Beginning in 1847, the statue stood in front of the White House, where it was presumably observed by the two Ohioans. At minimum, both Eston and the statue were tall and bronze.
There were other slaves at Monticello who were described as being tall and intelligent,35 but who were not even arguably Thomas Jefferson’s children. However, even Jefferson’s grandson, while vigorously denying the President’s paternity of any of Sally’s children, reportedly admitted that it was clear that some of Sally’s children had Jefferson blood in their veins.
Modern science has proven him right, as we now know from DNA testing that Eston Hemings was probably fathered by a member of the Jefferson family or by some other male carrying the Jefferson DNA marker. Whether that father was President Thomas Jefferson, or his brother Randolph, one of Randolph’s sons, or perhaps even a paternal cousin, any of them would have passed on genetic material that could easily explain the perceived physical resemblance. Similarly, Thomas Jefferson’s nephews Peter and Samuel Carr—sons of his sister Martha—would have carried sufficient “Jefferson” genetic material to possibly produce children bearing a likeness to the President; and they remain serious candidates for the paternity of any of Sally’s older children.
Before closing this discussion, perhaps another comment is in order about the alarming practice of advocates of the Jefferson-Hemings story trying to present the lack of information on an issue as “evidence” of something beyond the reality that we simply do not know a great deal of information after the passage of nearly two centuries. On page 217 of Thomas Jefferson and Sally Hemings, Professor Gordon-Reed notes Thomas Jefferson Randolph’s comment that Sally’s children obviously had Thomas Jefferson’s “blood” in their veins. From this, Professor Gordon-Reed asserts:
A declaration from a close Jefferson relative saying that it was obvious that Jefferson’s “blood ran in [the] veins” of Sally Hemings’s children and that one child could be mistaken for Jefferson must be regarded as strong evidence indeed. Declarations against interest are regarded as having a high degree of credibility. …
Randolph’s statements on the resemblance effectively narrowed the universe of white men who could have fathered Hemings’s children from that of any white man who visited Monticello during the years that Sally Hemings was giving birth. Had Randolph never said that the children looked so much like Jefferson, those who wanted to absolve Jefferson of responsibility could say …that the father could have been any one of hundreds of visitors to Monticello. Randolph’s assertions brought the possible fathers down to three men: Thomas Jefferson, Peter Carr, or Samuel Carr.36
As a factual matter, Professor Gordon-Reed’s conclusion that Jeff Randolph’s comment brought “the possible fathers down to three” is simply not true. It entirely ignores the possibility, for example, that Randolph Jefferson or one of his five sons could have fathered one or more of Sally’s children. Each of them would have carried “Jefferson’s blood” in his veins. Indeed, one of the many serious shortcomings in Professor Gordon-Reed’s highly praised volume is her total disregard of Randolph and his family in the first edition of her book. Although she lists in her bibliography Bernard Mayo’s Thomas Jefferson and His Unknown Brother Randolph37—and thus was presumably aware of the existence of Randolph Jefferson—she excludes Randolph and his family from her genealogical table of “The Jeffersons and Randolphs (Relevant Connections Only),”38 as well as from the nearly fifty entries in Appendix A to her book, entitled “Key to Important Names.”39 Nor are Randolph Jefferson or any of his children referenced in her index.40
One does not
like to speculate about the motives of others, and in this instance it really does not matter whether the exclusion of six key “suspects” from Professor Gordon-Reed’s volume was an error of ignorance, over-enthusiastic advocacy, or oversight. In reality, the list of “possible” fathers who could have produced children with Sally Hemings who had “Jefferson’s blood” in their veins numbered at least twenty-five,41 not “three” as Professor Gordon-Reed asserts; and any list of “serious suspects” certainly must include at least five of the six males of the Randolph Jefferson family, who we know were invited to visit Monticello shortly before Sally Hemings conceived Eston—the only child who we know with reasonable certainty carried “Jefferson’s blood” in his veins.
Nor do we have any information about the possibility that Thomas Jefferson’s paternal grandfather, who matured sexually more than a century before Eston Hemings was conceived, might have produced unknown children in or out of wedlock whose descendants might be candidates for Eston’s paternity. What we do know is that there were more than two dozen adult men in Virginia when Eston was conceived who carried the same DNA marker found in one of Eston’s descendants, most of them much younger and in better health than the sixty-four-year-old President. There may have been more.
In summary, the reported perceptions of physical resemblance of some of Sally Hemings’ children to Thomas Jefferson would seem probative of very little. Not enough information is known about most of her children to draw a serious comparison. Eston was tall (but not as tall as Thomas Jefferson), and the most commonly noted “likeness” of others appears to be their very light skin. Even in the case of Eston, a physical resemblance is readily explained if one assumes that he was the son of any Jefferson male, and that conclusion is strongly reinforced by the DNA tests. The physical resemblance does little to assist us in determining whether that Jefferson male was President Thomas Jefferson, his brother Randolph, one of Randolph’s five sons—all of whom were quite likely at Monticello when Eston was conceived—or even President Jefferson’s cousin George or one of more than a dozen other Jefferson men who might have paid an unrecorded visit to Monticello at that time.
* * *
Footnotes
1. Eric S. Lander & Joseph J. Ellis, “Founding Father,” Nature, vol. 396, Nov. 5 1998, at 13.
2. Thomas Jefferson Memorial Foundation Research Committee, Report on Thomas Jefferson and Sally Hemings at 8 (hereinafter referred to as “Monticello Report”). See also, id. Appendix F at 1.
3. Joshua D. Rothman, James Callender and Social Knowledge of Interracial Sex in Antebellum Virginia, in SALLY HEMINGS AND THOMAS JEFFERSON 87 (Jan Ellen Lewis & Peter S. Onuf, eds. 1999).
4. See Chapter One.
5. THE GARDEN AND FARM BOOKS OF THOMAS JEFFERSON 31 (Robert C. Barron, ed. 1987).
6. James A. Bear, Jr., The Hemings Family of Monticello, VIRGINIA CAVALCADE, Autumn 1979 at 81.
7. Letter from Henry S. Randall to James Parton on Jefferson and the “Dusky Sally” Story, in MILTON E. FLOWER, JAMES PARTON: THE FATHER OF MODERN BIOGRAPHY 236 (1951), included at Appendix E of Monticello Report.
8. Id.
9. See Chapter Two.
10. The President Again, RICHMOND RECORDER, Sept. 1, 1802, reprinted in Monticello Report, Appendix E.
11. Monticello Report, Appendix H at 6.
12. Id.
13. FAWN M. BRODIE, THOMAS JEFFERSON: AN INTIMATE HISTORY 31 (1974).
14. See Chapter One.
15. Quoted in Monticello Report, Appendix H at 8.
16. The 1830 Charlottesville census appears to identify Sally Hemings as a white woman. Whether she personally identified herself as a Caucasian at that time is unknown.
17. There seems to be no clear evidence that this was Sally’s child, or even that the child was named Thenia. See discussion on pages 91–93.
18. Quoted in Monticello Report, Appendix H at 10.
19. Id. (observation of Ellen Randolph Coolidge).
20. Id. (observation of former Monticello overseer Edmund Bacon).
21. Id.
22. The statement is attributed to Madison, but appears to have been written by journalist Samuel Wetmore. See Chapter Four.
23. Official Albemarle County records described him in 1831 as having a “light complexion” (Monticello Report, Appendix H at 12). Unlike some of Sally’s children, Madison appears not to have endeavored to live as a white person. Whether this was merely a consequence of personal preference or was influenced by his skin color is not known.
24. This was the description provided by his grandson in 3 HENRY S. RANDALL, THE LIFE OF THOMAS JEFFERSON 675 (1858).
25. Quoted in Monticello Report, Appendix H at 12.
26. Id. See also Chapter Four.
27. Quoted in Monticello Report, Appendix H at 13. Cynthia Burton informs me that, two years later, in an 1833 Special Census of Free Blacks in Charlottesville, “Eston Hemonds” was listed as “negro” and not “mulatto” as reported in the Monticello Report (id. at 16). I have not checked this, but mention it in case others wish to do so.
28. CHILLICOTHE LEADER, Jan. 26, 1887, quoted in Monticello Report, Appendix H at 15.
29. DAILY SCIOTO GAZETTE, Aug 1, 1902, quoted in Monticello Report, Appendix H at 15.
30. Id., quoted in ANNETTE GORDON-REED, THOMAS JEFFERSON AND SALLY HEMINGS: AN AMERICAN CONTROVERSY 15 (1997).
31. Monticello Report, Appendix F at 6.
32. It is also possible that the observers saw the statue before it was moved to the more prominent location on Pennsylvania Avenue; but that would have made Eston even younger and a meaningful comparison all the more difficult.
33. THE LIFE PORTRAITS OF THOMAS JEFFERSON 76–77 (1987).
34. MELVIN I. UROFSKY, THE LEVY FAMILY AND MONTICELLO 1834–1923 at 54 (2001); and MARC LEEPSON, SAVING MONTICELLO 63 (2001).
35. See, e.g., JEFFERSON AT MONTICELLO 23.
36. GORDON-REED, THOMAS JEFFERSON AND SALLY HEMINGS 217 (1997) (emphasis added).
37. Id. at 278.
38. Id. at xxii.
39. Id. at 239–44.
40. Id. at 285. Ironically, in one of her numerous attacks on the character of earlier Jefferson historians, Professor Gordon-Reed refers to the practice “of a historian rendering nonexistent some available evidence that hurt his position.” Id. at 49.
41. See, e.g., Monticello Report at 9.
8
Reassessing the Oral Tradition of Sally Hemings’ Descendants
* * *
The January 2000 report of a Research Committee of the Thomas Jefferson Memorial Foundation said that the “family history of Sally Hemings’s descendants, transmitted orally over many generations, states that Hemings and Thomas Jefferson are their ancestors.”1
Later, in talking with a United Press International journalist on the day in 2001 when the Scholars Commission report was originally released, Monticello President Daniel Jordan emphasized that the foundation “never said that DNA proved it. We said that the science and the history, especially the oral history, suggest a high probability that Thomas Jefferson fathered all of Sally Hemings’ children.”2
Before briefly discussing the importance of oral history—including its serious shortcomings as compared with some other sources of historical evidence—it is useful to clarify to what this statement is actually referring. A reader might mistakenly conclude that the foundation’s highly acclaimed “Getting Word” project—which has reportedly interviewed more than 100 descendants of Monticello slaves since 1993—has spoken to descendants of all of Sally Hemings’ children and consistently been told that they learned as children that they were descendants of President Thomas Jefferson. But that is not even close to being an accurate statement.
On the contrary—if one accepts the clear DNA evidence that Thomas Woodson was not Thomas Jefferson’s child3—the Getting Word project has located descendants of only two of Sally Hemings’ known children. They have no “oral history” from descendants of Beverly or Harr
iet Hemings. Indeed (again, excluding Thomas Woodson), only the descendants of a single child of Sally Hemings (Madison) have alleged that they were told that Thomas Jefferson was their direct ancestor prior to the publication of Professor Fawn Brodie’s 1974 biography making that allegation.
Once again, it is not my view that Thomas Woodson was not Sally Hemings’ child. I neither know nor greatly care about the answer to that question. Reportedly, the most powerful oral history seems to have come from statements by descendants of Thomas Woodson. One need not question the veracity of Thomas Woodson (who obviously had no first-hand knowledge of his own paternity) or any of his descendants to observe that seven or eight generations of very sincere oral tradition are less probative of the paternity issue than Dr. Foster’s extensive DNA tests. Whether Thomas Woodson was, or was not, the son of Sally Hemings, the oral history that he was the son of Thomas Jefferson seems clearly to be in error and thus probative of nothing relative to the current inquiry. Its primary utility at this point may be to emphasize the shortcomings of oral tradition as an historical source.
That Thomas Woodson may honestly have believed and asserted that he was the child of Sally Hemings and Thomas Jefferson is deserving of consideration. But its value as “evidence” is not enhanced simply because numerous generations of descendants from several of his children have retold the story in good faith and embraced it as true. Assuming that they have not at any point intentionally or inadvertently distorted his testimony, the modern “oral history” can be no more probative of the truth than the credibility of his original account. This reality may well be frustrating to families who must rely largely upon such accounts for information about their ancestry, but surely its truth is self-evident.
The Jefferson-Hemings Controversy Page 29