DATE: May 18, 2000
TO: All Interested Parties
FROM: Daniel P. Jordan, President, TJF
SUBJECT: TJF Response to the Minority Report

At my request, Lucia C. Stanton, Monticello's Shannon Senior Research Historian, prepared the attached response to the Minority Report.

Ms. Stanton, a Harvard graduate, has studied African-American life at Monticello for almost thirty years and has authored, among other titles, Slavery at Monticello (1996). She is the foremost scholarly authority on the subject, and several other scholars reviewed her document before it was made available to all interested parties.

While fully respecting the right of the author of the Minority Report to express his views, I felt it was appropriate to respond to the issues he raised, and I'm grateful to Ms. Stanton for her clear analysis and commentary. Her views reflect my own -- and, I believe, those of the full committee. We have posted Ms. Stanton's document on our Web site (, where it joins the original TJF research report as well as the subsequent Minority Report. I encourage the reading of all three items.


Thomas Jefferson Foundation

Response to the Minority Report
Prepared by Lucia C. Stanton, Shannon Senior Research Historian

April 2000

The issues raised in the Minority Report were discussed at some length in the meetings of the Research Committee on Sally Hemings and Thomas Jefferson in early 1999. While Committee members recognized that some of the evidence cited in the Minority Report was inconsistent with the Research Report conclusion about Jefferson paternity, the Committee as a whole did not feel it was of sufficient weight to warrant a different conclusion. The five major points of the Minority Report appear below, with responses to them based on the Committee's meetings in 1999 and subsequent discussion among staff and scholars.


1. Jefferson denied the relationship (and by implication, Jefferson would not lie).

This is based mostly on the 1805 cover letter to Robert Smith, where Jefferson admits improper advances to Betsy Walker but adds that "it is the only one founded on truth among all their allegations against me." This has definitely been interpreted as a denial by Malone and some other historians, but it remains ambiguous in the absence of the missing enclosure. Federalist attacks in this period did include the Jefferson-Hemings relationship, but no one has yet identified exactly who or what Jefferson was responding to in this instance. Therefore, it is still impossible to say with certainty what "all their allegations" means.

I too would like to believe Jefferson never lied, but realize that would make him superhuman, and we know (from the work of Joanne Freeman and others) that he sometimes had an elastic conception of the truth, when he believed the stakes for the nation were high. Also, it is important to take into account the ethical values of the day and to consider the possibility that Jefferson might not have viewed a relationship with Sally Hemings as inconsistent with a virtuous life. Americans in general, as Robert McDonald has shown, were not very exercised about the Hemings issue. They were far more disturbed by the Betsy Walker "affair." We know Jefferson's rationalizing talents and can imagine ways he could find a fairly comfortable place for this relationship in his view of himself. As one Jefferson scholar has recently pointed out, Jefferson might have considered it necessary for his health (he had books on the subject of health and sexual activity).

Short answer: While the 1805 cover letter is a piece of evidence to be seriously considered, there is no way it can be viewed as a "direct" or "powerful" denial. It is fraught with ambiguity.


2. Edmund Bacon denied the relationship (and by implication, Bacon too would not lie).

Bacon named someone else (deleted by Rev. Pierson, presumably) as the father of Harriet Hemings. First we had to consider reasons Bacon might have had for absolving Jefferson of the Hemings connection (he was talking to a clergyman in 1860, when mores were decidedly different from those of 1800; he was deeply loyal to Jefferson and proud of his association with a great man, and so forth).

Then there is the issue of timing. Dr. Wallenborn suggests Bacon was at Monticello at the time Harriet Hemings was conceived, in 1800, when Bacon was fifteen. Bacon himself stated he first came to Monticello in 1800. Jefferson's records, however, indicate that Bacon began working at Monticello sometime in 1806, becoming overseer on Sep. 29. Nothing in the records indicates his presence at Monticello before this. Since Bacon's family lived not far from Monticello, it is certainly possible he was on the mountain on an occasional basis. But a regular early-morning presence at the mountaintop -- implied by Bacon's comments -- would suggest that he resided at and was employed at Monticello. Bacon's employment from 1800 to 1806 is extremely unlikely, however, since he is not mentioned in the Memorandum Books until Sep. 1806. After this date, he appears in the Memorandum Books with great frequency.

Short answer: Bacon's statement, because it is in direct conflict with the Madison Hemings account, is one of the stronger pieces of evidence against Jefferson paternity. At the moment, however, it stands alone -- since the Thomas Jefferson Randolph version is undermined by DNA evidence. Additionally, Bacon's (and Pierson's) partisanship and his misrepresentation of crucial dates weaken trust in his testimony.


3. Thomas Jefferson Randolph denied the relationship (and, by implication, Thomas Jefferson Randolph would not lie).

Thomas Jefferson Randolph's version of events was greatly weakened by the DNA evidence, which seems to explode the Carr brothers argument. Dr. Wallenborn adopts the view that only Eston Hemings had a Jefferson father, while Sally Hemings's other children were fathered by a Carr. The Report's findings explained the Committee's view on this issue: that birth patterns, Madison Hemings's account, the freeing of all Sally Hemings's children, and so forth, indicate that the possibility of multiple fathers is very unlikely. Randolph's accounts have the same problems of loyalty and timing as Edmund Bacon's; he told Henry Randall he was in charge of Monticello when Sally Hemings's children were born, years when in fact he was three to sixteen years old. Both Bacon and Randolph misrepresented their ability to know the situation, by changing the chronology.

Dr. Wallenborn puts a lot of emphasis (here and in #4 below) on the issue raised in Henry Randall's account: that Thomas Jefferson Randolph told him Martha Randolph pointed out to him that Jefferson and Sally Hemings "were far distant from each other" for fifteen months before the birth of the Hemings child that most resembled Jefferson. Dr. Wallenborn erroneously states that Randolph found this to be true when examining an old account book. It was Henry Randall who later found the birth date of the child in an old account book and was able to prove, "from well known circumstances," that it was impossible for Jefferson to have been the father. We know, however, from the Memorandum Book and other sources that Jefferson was at Monticello at the right time to father all of Sally Hemings's children. Recognizing this, Dr. Wallenborn suggests instead that it was Sally Hemings who was absent from Monticello. This supposition, however, is irrelevant in this instance. It would be impossible to know "from well known circumstances" about her absence from Monticello. So Randall's conclusion was either recalled incorrectly, misrepresented, or it dealt with a child of someone other than Sally Hemings.

Short answer: Thomas Jefferson Randolph's versions of events have been seriously called into question by the DNA evidence eliminating the Carr brothers and by inconsistencies and misrepresentations in his reports to others. The argument about Sally Hemings's possible absence from Monticello does not apply to the Randolph-Randall accounts.


4. There is insufficient information about Sally Hemings and other Jeffersons to make a valid statistical estimate of probability.

Even without the Monte Carlo simulation, the correlation of Jefferson's presence at Monticello and Sally Hemings's conceptions is striking, with no conception taking place in Jefferson's absences and four of them taking place within three weeks of his arrival back at Monticello. The Monte Carlo statistics merely give a probability percentage to that striking correlation.

The fifteen-month argument is again raised. Since there is nothing to indicate with any certainty which child Randolph and Randall considered most resembled Jefferson, there seems to be no particular grounds for applying the fifteenth-month argument to Eston Hemings. Even if one did, it would mean that Sally Hemings was absent from Monticello from July 1806 to September 1807, a case unusual enough to be mentioned in accounts or family and overseer's letters.

Dr. Wallenborn questions the validity of a Monte Carlo simulation made without comparative data on other Jefferson males, about whom little is known. While it is true that the movements of Jefferson's brother Randolph and his sons is little documented, the fragments of information that have survived do indicate that they were sometimes at Monticello during Jefferson's absences. Yet Sally Hemings never conceived in Jefferson's absence.

Short answer: It is highly unlikely that Sally Hemings was absent from Monticello for any extended period after 1789. The fragmentary documentation about other male Jeffersons provides as much information about their presence at Monticello during Jefferson's absence as during his presence.


5. Madison Hemings's recollections lack credibility because of the language used ("amazing" grammar and vocabulary) and his age (68).

While S. F. Wetmore's 1873 publication of Madison Hemings's memoir -- taken down before the days of audiotaping technology -- is probably not an absolutely verbatim transcript, but it does seem to be an accurate reflection of Hemings's statements. Other interviews Wetmore conducted are very different in language and tone, as is Wetmore's own introduction to the piece. In the case of Hemings's "remarkable knowledge of history," he was not remembering facts heard forty years before, as Dr. Wallenborn says, but, as Hemings states, he learned about Jefferson's public life in the years after he became free. Also, there are abundant examples of individuals with rudimentary educations who could speak or even write with great eloquence. Wetmore was obviously impressed by Madison Hemings, "who would have shone out as a star of the first magnitude," had he lived his life in different circumstances.

However much of the interview was Hemings's exact wording, his claim that Jefferson was his father is not something that bears on precise language. Hemings would not have forgotten who his father was, no matter his age.

Short answer: The details of language and historical facts are irrelevant to the main issue: paternity.