|DNA AND BROWN GENEALOGY
By James Armistead Brown, Jr.
Reposted with permission,
Brown Family Genealogy"
for other outstanding Brown Family Articles
A. Introduction and Summary
B. Technical Details about DNA
C. The First Brown DNA Match
D. The Second Brown DNA Match
E. The Third Brown DNA Match
F. Four Non-Matches and their Importance
G. Suggestions for Future Research
Appendix: A Speculation on DNA and Hypothesis-Friendly Genealogy
Note: Numbers in [brackets] refer to footnotes.
A. INTRODUCTION AND SUMMARY
DNA analysis as a genealogical technique first became practical for amateur researchers about four years ago. But so far, it has been used only by a tiny fraction of the millions who pursue genealogy as a hobby. In fact, it is probably safe to say that the overwhelming majority of family researchers still regard DNA testing as a mere curio -- to the extent they even know about it.
On the other hand, DNA analysis has already contributed positively and impressively to research on the large and widely dispersed Brown family descending from William Brown of colonial Augusta Co VA and early Knox Co TN.
The related groups in this large extended family have ranged from the Browns of Warren Co TN and St Clair Co AL; to the Browns of Pontotoc Co MS, Izard Co AR and Marion Co AR; to the Missouri and Kansas descendants of Gideon Leeper Brown; and to the descendants of Oregon Trail pioneer Hugh Leeper Brown, founder of the town of Brownsville, Oregon.
Beginning in 2002, DNA "surname tracing" has been used not only (1) to locate and confirm relationships among descendants of this family, especially in cases where the paper trail has been incomplete, but also (2) to show that certain relationships previously hypothesized are in all likelihood not valid.
In brief, the DNA tests enumerated below have advanced considerably our understanding of relationships between and among descendants of the William Brown in question. At the same time, it is important to realize that these findings have assumed their importance almost entirely within the context of a large body of traditional genealogy conducted by numerous researchers for this kinship group since at least the early 1960s.
In the author's opinion, the import of the DNA findings -- taken in essential conjunction with traditional documentary research -- is that there can no longer be any reasonable doubt as to the very close relationship of these scattered families, even though precise "paper trails" between and among them have not in all cases been satisfactorily documented with the sorts of primary sources that customarily have formed the backbone of serious genealogy.
Moreover the author submits that, in addition to their obvious benefits for research on the Brown family of this website, these DNA results and their interpretation can be a useful case study for genealogists working on other family lines, especially in the way that these results have both sustained and falsified hypotheses drawn from traditional genealogy.
B. TECHNICAL DETAILS ABOUT DNA
Most DNA analyses for genealogy have so far involved tests of the Y chromosome. This "yDNA" is possessed only by males, and it is passed from father to son over hundreds of years (and even thousands of years) with little or no change. As such, yDNA "traces" the surname in a fashion virtually ideal for genealogy. 
For basic "surname tracing" (which describes the methodology used for research on this website's Brown family) numerical yDNA scores are compared to determine if two or more living men share a recent common ancestor in their unbroken male-to-male lines. If the men in question have yDNA numbers that are identical or very close, then basic techniques of statistical inference may be used to estimate the probabilities that their common ancestor lived within a given number of generations.
So for example, if two identically surnamed men should match perfectly on 23 or more markers from the commonly used 25-marker DNA test, then one could have a high degree of confidence that their common male ancestor lived within the last few hundred years.
If on the other hand the two men in question should match on fewer than 23 markers, then it is highly likely either (1) that they do not have a common "male line" ancestor, or (2) that their common ancestor lived so long ago as to render "genealogy" -- as that term is commonly understood -- virtually useless for identifying him.
It is beyond the scope of this brief discussion to deal with the complex matters of genetics and mathematical statistics that sometimes arise when DNA science intersects with old-fashioned genealogy. Suffice it to say, however, that the simple and relatively powerful "yes-no" analysis described in this article represents only a first step along the new path of DNA genealogy. 
The following sections will discuss in detail the specific DNA matches and "non-matches" that have advanced significantly the genealogical research for the extended Brown family of this website.
C. THE FIRST BROWN DNA MATCH
Y-chromosome DNA testing was first used for genealogical research on our Brown line in mid-2002, when Vic Brown of Denison TX and Jim Brown (the author) both submitted DNA samples for testing under auspices of the large Brown/Browne/Braun DNA Study, an activity conducted by amateur genealogists via a contractor, Family Tree DNA Inc. of Houston TX, with actual lab work at the University of Arizona. 
Vic and Jim each had submitted his own DNA in complete ignorance of the other man's test. They had communicated once, about a year earlier, since they knew their Brown ancestors had lived near one another in early Warren Co TN. But they concluded that given the total lack of direct evidence on point -- evidence which had been sought by Brown researchers for years -- there was no realistic chance of either proving or disproving a relationship between their Brown lines.
Then about a month after the samples were submitted, Vic and Jim got the unexpected news of a perfect 12/12 DNA match. To make a long story short, the odds against such a match were huge -- unless of course their Brown ancestors in Tennessee actually were close kin. And as such proof of kinship, the DNA match took a large chink from the "brick wall" that had long blocked research by descendants of the Browns of St Clair Co AL, several of whom in the 1960s had traced their line to Warren Co TN (but no further).
Shortly before Vic and Jim discovered their surprising DNA match, and completely unbeknownst to them, fellow Brown descendant James E. Hargraves of Elk Grove CA had become convinced that his and Jim's mutual ancestor David Brown was closely related to a large family of Browns including not only the Revolutionary War veterans Thomas and Robert Brown of Warren Co TN, but also several clusters of Browns in pre-statehood east Tennessee and in 1820s Arkansas Territory.
So although Vic and Jim had been totally unaware of their kinship, their DNA match came to James Hargraves less as a surprise than as a powerful confirmation of relationships he had recently proposed (or "hypothesized" -- although he did not use specifically this word) as a result of his work in traditional "paper" genealogy. 
And while Vic did not have flawless documentation back to his likely ancestor, Revolutionary War vet Thomas Brown, Vic's and Jim's DNA match taken together with all of James Hargraves' exhaustive evidence and previous conventional genealogy left little or no doubt that their Brown ancestors were indeed from the extended Brown family of Warren County's well-known Revolutionary War veterans, the brothers Thomas and Robert Brown. Moreover, since these brothers had a solid paper trail back to Augusta Co VA, the findings were sufficiently persuasive in the author's opinion to carry the genealogy of his St Clair Co AL Browns all the way back to the same spot in 1750s colonial Virginia.
D. THE SECOND BROWN DNA MATCH
The next instance of DNA testing for our Brown line was the discovery in 2003 of a 12/12 match between Danny Brown of Branson MO and Vic Brown, together with a 24/25 match between Danny and Jim Brown. Danny is a documented descendant of the Thomas Brown who died in St Clair Co AL in 1819. 
Before Danny's DNA test, James Hargraves had proposed on the basis of circumstantial evidence that this Thomas was a brother of David Brown. Danny's DNA results, on the other hand, constituted direct evidence of close kinship between Thomas and David Brown -- evidence that had the additional virtue of being consistent with (and supportive of) a belief in their having been brothers.
In conclusion, although Danny's DNA results also are consistent with "cousinship" between David and Thomas Brown, it is the author's opinion that unless and until contrary evidence should appear, the most plausible scenario is one of siblingship -- so that James Hargraves' interpretation has been entirely appropriate in positing the two men as brothers
E. THE THIRD BROWN DNA MATCH
Most recently (June 2004) a DNA test with John Pershing Brown of Manhattan KS has revealed a 25/25 match with Danny Brown, a 12/12 match with Victor Brown, and a 24/25 match with Jim Brown. 
John is a descendant of Gideon Leeper Brown, whose family went more or less directly to Missouri from east Tennessee in the mid-1800s, without ever having resided in Warren Co TN. The upshot is that John's DNA match has been critically important in validating James Hargraves' proposal that the unusual name "Leeper" (chiefly as a middle name, but also as a surname) is a key for establishing relationships among a widely dispersed group of Brown families, who have by now been traced all the way from Virginia and the Carolinas to Tennessee, Alabama, Mississippi, Arkansas, Texas, Missouri, Kansas and Oregon. 
F. FOUR NON-MATCHES AND THEIR IMPORTANCE
No discussion of DNA genealogy should leave the impression that positive "matches" are the be-all-and-end-all for this new type of research. In fact, to the extent that genealogy borrows from well-established scientific methodology, then an ability to falsify proposed relationships ("hypotheses") is fully as important as "proof" -- whatever one's definition of the latter. 
In this regard, it's worth noting that in at least three recent instances, DNA tests have shown at high levels of statistical confidence that certain Brown lines with roots in early Warren Co TN or St Clair Co AL were not kin to the extended Brown family of this website.
In the first recent case, a DNA sample from a living descendant of the Absolom Brown family of early Warren Co TN has shown these pioneers were almost certainly not related to their neighbors, the family of the Revolutionary War veterans Robert and Thomas Brown -- settling a question that had long lingered among Brown researchers for the County. 
In the second recent case, a Brown-surnamed man with apparent roots in St Clair Co AL was found not to match the family of Thomas and David Brown -- in spite of the fact the test participant's ancestors had been listed in early censuses as children of Thomas Brown's widow. 
In the third case, a descendant of Daniel Brown of early Warren Co TN was found to have a perfect DNA match with a well-documented Brown family from neighboring White Co TN, a line clearly not related to the Brown family of this website -- in spite of a well-founded prior belief that Daniel Brown was from the extended family of the Revolutionary War veterans Thomas and Robert Brown. 
In the fourth recent case, a Brown family from Jackson Co TN has been found to have different DNA from the Browns of Warren Co TN -- in contradiction to a hypothesized relationship between the two families. 
Finally, tests under auspices of the large Brown/BrowneBraun DNA Study have already identified more than
350 separate Brown families with no genealogical relationship to the extended Brown family of this web site. Many of these other families, moreover, lived in early Virginia or the Carolinas -- the same general vicinity as William Brown of early Augusta Co VA and Knox Co TN. So if nothing else, the "negative" comparisons with these other families have narrowed considerably the range of possibilities for finding kinship relations to the William Brown line. 
G. SUGGESTIONS FOR FUTURE RESEARCH
Many puzzles concerning the Brown family of this website might eventually be addressed by DNA analysis. Here are four examples:
(1) The single DNA test carried out so far with a man from the David Brown line shows a very small mutation that is not present in the 25-marker tests for either the Thomas Brown line of St Clair Co AL or the Gideon Leeper Brown line of Missouri. Yet the two latter lines have identical DNA on the 25 markers in question.  Given the limitations of currently available data, it is not possible to determine whether this mutation arose with David himself, with his father, or with one of David's descendants. But if this same mutation should eventually be found in a Brown-surnamed man who is demonstrably not descended from David, then the results almost certainly would falsify the hypothesis that David and Thomas Brown of early St Clair Co AL were brothers -- and the same results would simultaneously necessitate the alternative explanation that they were cousins. On the other hand, if a test on a "non-David descendant" should not have this particular small mutation, then the hypothesis of brotherhood between David and Thomas would continue to be accepted.
(2) Settlers Francis Brown and Hugh Brown are mentioned prominently in the archives of colonial Augusta Co VA, often in proximity to a William Brown and his probable father-in-law, John Black.  Yet there appears little chance that conventional paper genealogy will ever settle definitively the questions of whether Francis and Hugh Brown were kin to the William Brown who died ca. 1807 in Knox Co TN. On the other hand, if we should locate living Brown-surnamed men with good paper trails back to Francis or Hugh, then DNA from these living men might either confirm or falsify the hypotheses of kinship among Francis, Hugh and William Brown.
(3) The will of William Brown, 1807 in Knox Co TN, mentions son John Brown.  There are also numerous records from early Roane County (previously a part of Knox) that show a John Brown as Sheriff, in close proximity to men who appear to have been William's sons Alexander, Robert, Thomas and William.  The author has encountered suggestions from other researchers that John Brown the Sheriff may have been the same man as John Brown of the will. But given all the men named "John Brown" in early Tennessee, the possibility of settling this matter by conventional genealogy are slim at best. On the other hand, an answer might be possible if a DNA sample could be obtained from a living Brown-surnamed man with an unimpeachable paper trail back to Sheriff John Brown: If this DNA should turn out to match that of our William Brown's descendants, then we could not exclude the possibility that John was William's son. A non-match, however, would almost surely lead us to conclude that the two men were from entirely different Brown families.
(4) Two families in present-day Texas, who have not been able to document "paper trails" to the Brown family of this website, have ancestors from Alabama known to them respectively as "Guionell L. Brown"  and "Leap Gian Brown."  Another Texas family has an ancestor reputedly from St Clair Co AL, named "Francis Marion Brown."  In all three instances, naming patterns and geographic proximity point to kinship with Browns descended from William Brown of early Augusta Co VA.  If living Brown-surnamed men from any of these contemporary Texas families could be located for DNA tests, then it is very likely they would match the four Brown men already tested. And if these Texas descendants should have minor mutations away from the DNA "scores" of the previously tested men, then the information might help eventually to establish a new picture of "branching" patterns in the Brown family tree.
A large body of conventional "paper" research on the extended William Brown family has been conducted over the past 40+ years by a number of descendants and their kin. In this regard, the discovery by James Hargraves of the "Leeper key" has been critically important, alongside prior work by a dozen or so other kin with whom the author has shared information since he first researched his Brown line in 1962. This number has included the late Marvin Ryan, the late Lois Barrett O'Kelly, Judy Voran and others. 
Without this important foundation, the specific DNA matches and mismatches described above would have had greatly diminished usefulness -- and apart from the initial and totally unexpected match between Vic Brown and Jim Brown, they might never have been found.
Likewise, without the rich body of "hypotheses" provided by this conventional genealogy, it would not have been possible to articulate the suggestions above for future research.
In the author's opinion, the upshot of the happy marriage between DNA analysis and traditional documentary genealogy described above has been to elevate from a "credibility based upon preponderance of the evidence" to a "credibility beyond a reasonable doubt" the following propositions:
(1) that the author's great-great-great grandfather David Brown of early St Clair Co AL was closely related to Thomas, William, Alexander and Guion Leeper Brown of the same county;
(2) that these men were from the same family as the Revolutionary War veterans Thomas and Robert Brown of Warren Co TN;
(3) that Gideon Leeper Brown of Missouri and Hugh Leeper Brown of Oregon shared a common ancestry with the St Clair and Warren County Browns; and
(4) that all these men were from the line of the William Brown who moved from Augusta Co VA in 1758 to the Waxhaws region of NC/SC and who died in Knox Co TN ca. 1807.
Moreover, as a general matter, DNA science and the closely related field of genetics can reasonably be expected to continue their rapid advances during years to come -- in such fashion that the application of DNA testing to genealogy can be expected to create an ever-improving methodology for both amateur and professional researchers. Therefore all family historians, whether or not related to the Brown family of this website, may justifiably hold an optimistic view as to the possibilities for continued cross-fertilization between the new discipline of DNA science and the old discipline of documentary genealogy.
APPENDIX: A Speculation on DNA and Hypothesis-Friendly Genealogy
The role of hypotheses and "hypothesis testing" in genealogy is a subject in its own right that deserves much more exploration than possibly can be provided here.
But since the new field of DNA genealogy has already borrowed insights and techniques from the "hard sciences" of microbiology and genetics, it seems appropriate here to discuss briefly how the scientific methodology of hypothesis testing may be applied to genealogy -- including not only DNA genealogy but also conventional "paper" genealogy, especially in cases where the latter may depend upon circumstantial evidence.
In brief, although perhaps the issues come to the fore most clearly with the example of DNA genealogy, the author believes all genealogy can benefit from an "hypothesis-friendly" approach. 
A common dictionary definition of the word "hypothesis" (plural "hypotheses") is:
"a proposition tentatively assumed in order to
draw out its logical or empirical consequences
and so test its accord with facts that are known
or may be determined." 
For purposes of the discussion here, a working definition of the term "genealogical hypothesis" is taken then as a plausible but tentative explanation of a kinship relation between or among individuals that has been put forth on the basis of (1) logical reasoning, combined with (2) evidence that is either circumstantial, or in the case of direct evidence, seriously incomplete.
It is obvious from this working definition that genealogical hypotheses are nothing new. Researchers have always relied on them, whether or not explicitly, and whether or not an identical terminology has been involved. But it appears to the author that conventional genealogy has generally tended to treat hypotheses somewhat as illegitimate children, not to be openly acknowledged and not to be relied on for important research findings.
In contrast, the approach advocated here is based in part on a belief that waiting for virtually flawless direct evidence can be counterproductive, by prematurely closing off potentially fruitful avenues of research.
In other words, an hypothesis-friendly methodology is more willing than conventional genealogy to use circumstantial evidence and/or seriously incomplete direct evidence, so as to "connect the dots" in logical and plausible patterns. Such patterns then may help not only to guide new research, but also may help us discern previously unseen relationships whenever we take fresh looks at old data.
In this view, a genealogical hypothesis is something to be sought and emphasized, not only with caution as to its limitations but also with an appreciation of its potential for advancing one's research agenda. In many cases, this process can help bring about what one genealogist has usefully described as an "interim analysis."  The interim analysis may then provide a valuable framework for seeking, organizing, and interpreting additional data.
And following upon the example of the hard sciences, an hypothesis-friendly approach may be undertaken with the reasonable expectation that an interim analysis often will evolve into a more broadly accepted interpretation -- moving in the best of all cases from a tentative application based upon "preponderance of the evidence," into an acceptance "beyond a reasonable doubt."
In author's opinion, the most useful research "advice" flowing from an hypothesis-friendly approach to genealogy may be encapsulated as follows:
When faced with a genealogical "brick wall" (defined as the absence of satisfactory direct evidence to support an essential relationship, without which a particular research path comes to a halt), one should attempt to establish one or more plausible hypotheses that serve as a tentative solution or solutions to the problem at hand -- stated openly, with full acknowledgment of any logical or empirical shortcomings. Then use these hypotheses to guide future research. Any new findings should eventually be employed to test the hypotheses. If new findings contradict any given hypothesis, the latter should be discarded without embarrassment. But if new findings are consistent with a hypothesis, then it may be held with greater confidence -- with the recognition that it will always remain subject to additional confirmation or falsification in the face of new evidence.
The methodology advocated here is not without its dangers. The reader is justified in a certain apprehension that it may increase the ease with which unskilled or careless genealogists jump to unsupported conclusions by misusing or misunderstanding the "interim analyses" of conscientious researchers, no matter how carefully the hypotheses may be hedged nor how cautiously the analyses may be presented.
In defense, the author submits that the unskilled and the careless will probably do their damage in any event, whether conscientious genealogists use, or whether they avoid, an hypothesis-friendly methodology.
At the same time, one may hope that the explicit presentation of genealogical findings as hypotheses, with forthright discussion of their ever-present logical and/or factual shortcomings, will in the long run improve rather than degrade our ability to combat erroneous genealogy.
An analogy to the Internet may be appropriate: Probably all experienced researchers know today how the Internet has allowed unsupported and downright incorrect genealogical information to spread with an ease and persistence that far exceed comparable problems when genealogy was restricted to conventional printed media. But just as the Internet's genealogical virtues overwhelmingly outweigh its vices, it seems reasonable to suggest that a wider appreciation of hypothesis-friendly genealogy can far outweigh the potential drawbacks.
Two important cautionary comments:
In order for findings derived from the hypothesis-friendly approach to have credibility, there should be a clear explanation of, and an emphasis upon, the inevitable limitations of the methodology and the inferences drawn with its help.
Equally, propositions resting upon this approach should be presented in such a manner that the reader will readily perceive not only that they remain open to refutation in the face of credible contradictory evidence, but also that the researcher proposing the findings is not "defensive" about their being tested in the light of possible new evidence, since all genealogical findings must to some extent remain tentative for all time. 
In the author's opinion, DNA analysis fits perfectly with hypothesis-friendly genealogy, since -- as it is absolutely critical to understand -- a DNA test generally cannot in and of itself "prove" a specific relationship: Rather, the results of a DNA test may either be consistent with a hypothesis that has been posited on the basis of some prior information, or alternatively, the DNA results may falsify such a posited relationship.
The case of Thomas Jefferson and Sally Hemings is useful to illustrate why a DNA test generally cannot "prove" a relationship: In the Jefferson-Hemings investigation, DNA samples were taken from two living men, one a descendant in an unbroken male-to-male line from one of Sally's sons, the other a descendant in an unbroken male-to-male line of Thomas Jefferson's uncle, Field Jefferson. The two DNA samples matched closely.
This result was entirely consistent with the hypothesis that Thomas Jefferson was the father of Sally's son, since Mr. Jefferson's yDNA signature would have been the same as that of his uncle. But the DNA test by itself cannot discriminate between this hypothesis and the alternative explanation that one of the uncle's sons, or Thomas Jefferson's brother or nephew, was the father of Sally's child. 
Beyond this point, DNA testing per se has little or nothing left to contribute, so that historians concerned with the Jefferson-Hemings matter must then fall back upon circumstantial evidence -- derived almost entirely from conventional "paper" documentation -- to decide which of the competing hypotheses has the greatest explanatory power.
So as shown by the Jefferson example, and equally by the example of the Brown family of this website, DNA genealogy's promise for research on any specific family is most likely to be realized within a context of plausible and well-organized hypotheses derived from conventional documentary research.
1. There are two other types of DNA analysis for genealogy, namely (a) research on the unbroken female-to-female line by means of mitochondrial DNA and (b) so-called "deep ancestry" research. These two investigative modes have not been used to study genealogy of the Brown line of this website and so will not be discussed here. For information on these techniques, see
2. For a basic example of mathematics applied to advanced research in DNA genealogy, see:
3. This research project started in early 2002 with the aim of building a data base of DNA results for ALL families carrying the Brown surname, not just the family of this website. The study also includes families whose surnames use variant spellings like Browne, Broun and Braun. Over 80 separate, non-related Brown lines had been identified within the study by mid-2004, based upon DNA samples from more than 100 living men, all of them believed to descend in unbroken male-to-male lines from Brown-surnamed male ancestors. For details, see:
http://brownsociety.org/browndna/results.htm. Specific results for Vic and Jim Brown are in Group 2, Kit 3996 and Kit 4020. Data on Vic Brown's genealogy supplied in Emails to author, various, from Victor Brown (VICBROWN at ev1.net) and Marilou Werrell (werrell at aolcom).
4. For details about James Hargraves' pathbreaking research, see:
5. Brown DNA Study, http://brownsociety.org/browndna/results.htm, Group 2, Kit 12719.
6. Brown DNA Study, ibid., Group 2, Kit 20243, plus Emails to author, various, from Evelyn Brown (brownjpej at msn.com) .
7. Hargraves' crucial breakthrough in using the "Leeper key" is described in the item cited above at Footnote 4. Valuable Information on the "Leeper middle named" Browns of Missouri, Kansas and Oregon may be found on the websites of (a) Linda Haas at
http://homepages.rootsweb.com/~haas/genealogy/browns.html, (b) the late Danielle Thompson at
http://members.aol.com/DThomp3907/brown.html, and (c) Cecil Houk at
http://freepages.history.rootsweb.com/~cchouk/oregon_trail/b1.htm. Although these websites state without support (and incorrectly, in the author's opinion) that the wife of William Brown of early Knox Co TN was Martha Bassett, this lapse does not detract from their value in tracing the Browns from Tennessee to Missouri and points west.
8. The prevailing opinion among students of scientific methodology is that one generally should not speak of hypotheses as being subject to "proof" or as capable of being "proven." The favored terminology, rather, is to speak of them as "not falsified" or "sustained" -- or similar words emphasizing that all hypotheses and indeed all scientific knowledge are to some extent tentative, and therefore never "proven." See the references in footnote 23. It is the author's opinion that genealogists should follow a similar practice. On the other hand, the words "proof" and "proven" are so often used in common parlance that any attempt to banish them from genealogy obviously would be futile, especially since in 1997 the Board for Certification of Genealogists formally adopted the "genealogical proof standard" to replace "preponderance of evidence" as their chief test of credibility. See Christine Rose, Genealogical Proof Standard (San Jose CA: Rose Family Association, 2001), p. 1; and, Board for Certification of Genealogists, The BCG Genealogical Standards Manual (Orem UT: Ancestry Publishing, 2000), pp. 1-2. On the other hand, the latter publication does acknowledge "that any statement about ancestors, even if it meets the GPS, is not absolute or everlasting." Ibid., p. 2. The author suggests that this caveat is to some extent inconsistent with the Board's determined use of the word "proof."
9. Brown DNA Study, http://brownsociety.org/browndna/results.htm, Group U, Kit 11127, Dan W. Brown results, plus Emails to author, various, from Dan W. Brown (DBrown at xtra.co.nz) and the website of Chuck Smith (Woodchuxmail at aol.com) on Absolom Brown at
10. Brown DNA Study, ibid., Group 7, Kit 11520, John Calvin Brown results, plus Emails to author, various, from Carolyn Masur (caromoo at nezero.net).
11. Brown DNA Study, ibid., Group 6, Kit 21131, William Wayne Brown results, plus Emails to author, various, from Norma Irene Morgan (Irnmorgan at aol.com) .
12. Brown DNA Study, ibid., Group U, Kit 17164, Joe Howard Brown results, plus Emails to author from Mary Lu Johnson (hellomlu at bellsouth.ne)t
13. Brown DNA Study, ibid., various results.
14. Brown DNA Study, ibid., Group 2, Kits 3996, 4020, 12719 and 20243.
15. See, for example, Lyman Chalkey, Chronicles of the Scotch-Irish Settlement in Virginia: Extracted from the Original Court Records of Augusta County, 1745-1800, reproduced at
16. Will of William Brown, August 7, 1803, recorded in Knox Co TN probate records for April 1807. Transcribed by Mary Lu Johnson (hellomlu at bellsouth.net) on July 1, 2000, from Knox County Probate Book 1, Jul. 1792-Oct. 1811, pp. 233-34, found at Tennessee State Library and Archives, Nashville, microfilm roll 155 for Knox County.
17. See, for example, the transcriptions of the 1802 and 1805 tax lists of Roane County, at
18. Emails to author, various, from Sammie Williams Pettitt, (SamLouPet at aol.com) .
19. Emails to author, various, from Linda Starr, (LSStarr at pilgramage.us).
20. Emails to author, various, from Betty Brown Mayo, (BMayo at ev1.net).
21. David Brown of Warren Co TN and St Clair Co AL, the author's great-great-great-grandfather, had two sons who carried the "Marion" name: William Marion Brown, who moved to Arkansas in the 1850s, and Lt. Marion Brown, who died of measles while serving in the Confederate Army.
22. It is not possible to name every single contributor to this process. But in addition to the researchers and Brown descendants cited in other footnotes here and in the text above, the author wishes to acknowledge helpful information supplied by the late May Pegues Girand, the late Alice Mae Morrow, Bennie Mitchell, Lisa Bishop, Susan Kelly, Gail McNeely, Ruth Barrett, Claudia Pegues Barlow, Ann Turner, Dan Page, Kay Vanderostyne, and Clint Joyce. Any omissions from this list are unintentional, and anyone thus omitted has the author's sincere apology.
23. A standard reference on the role of hypotheses in science is Karl Popper, The Logic of Scientific Discovery (various editions). For a summary of Popper's views and those of his critics, see
http://plato.stanford.edu/entries/popper/. "Popperian" methodology has been particularly influential in the social sciences. See for example, Milton Friedman, "The Methodology of Positive Economics," in Essays in Positive Economics (Chicago: University of Chicago Press, 1953). And a Google search on "Popper and sociology" yields about 21,600 entries. See also
http://www.sociology.org.uk/p1mp4n3d.htm. The author has not, however, located in print or on the Internet a systematic analysis of the use of hypotheses in genealogy, although bits and pieces are abundant. Perhaps the most definitive recent discussion of genealogical methodology is Elizabeth Shown Mills, Evidence!: Citation and Analysis for the Family Historian (Baltimore MD: Genealogical Publishing Co., 1997). Mills clearly recognizes the important roles of hypotheses and their testing in genealogy, but she does not present a specific analysis of the matter. Ibid., p. 47.
24. Webster's Third International Dictionary of the English Language Unabridged, 1973 edition.
25. The term is taken from Richard A. Clifford, "Tools Used in Analysis of a Century-Old Genealogical Research Problem," reprinted from The Genealogical Journal, Vol. 4, No. 4, December 1975, at
http://freepages.genealogy.rootsweb.com/~turnbo/tx/pg82.htm. The article describes how a hypothesis-friendly approach was profitably applied to genealogical research on Clifford's Turnbo family, contrasting his "interim analysis" with conventional genealogy's use of "direct evidence documents ... that do not require any reasoning process or inference to prove the family relationship of two or more persons."
26. The Board for Certification of Genealogists states "that any statement about ancestors, even if it meets the GPS, is not absolute or everlasting." Op. cit., p. 2. So even though the BCG does not use the word "hypothesis" in their Standards Manual, they have in essence acknowledged that all genealogy is, in the final analysis, a series of hypotheses. And similarly, Elizabeth Shown Mills states that "the case is never closed on a genealogical conclusion." Op. cit., p. 57.
27. For additional information on the Jefferson DNA controversy, see: