I’ve been long now in the trenches of original genealogical research spanning more than three decades (progressing, along with younger siblings, from elementary-schoolers accompanying our parents to oral history interviews, cemetery traipsing & tombstone transcriptioning in the early 1970s, to an interested teen hand-copying from my parental pedigrees & family group sheets, to an adult beginning my own intensive research in microfilms of original documents at genealogical repositories throughout the 1980s & making family history trips, viewing originals held in courthouses in the 1990s, supplementing & establishing family history by personal & telephone interviews, correspondence & transcription of tape-recorded contributions with my grandparents & Great-aunts, & my last surviving Great-grandmother, expanding into correspondence with other historians & researching at courthouses, libraries, and historical societies in ancestral states). During that time, I’ve seen the problem caused by inaccurate spellings, transcriptions, indexing, and erroneous summarizing in publications. It’s HUGE, with so many far-reaching tentacles of error being adopted as truth and fact when it is nothing but fiction caused by a fundamental mistaken reading of a document.
This is why it’s so essential to keep going back to the original (since, with each transcription or publication & re-publication, more chance for error creeps in and you’re that many more generations removed from the actual document made at the time the ancestor was present). Sometimes it’s only much later, once experience is gained in methodology, and with the family, that a document is seen & transcribed accurately in historical context.
And when doing genealogical detective work, one small error can completely change the foundation &/or goal of what you’re searching for and how you ought to view it or weight it’s value.
I’ll give two quick examples:
The first pertaining to research on Younger Herndon (1785 VA-1859 KY):
BACKGROUND (1): U.S. Federal Censuses are taken once every decade since the founding of the nation, beginning in 1790. From 1790 through 1840, the census only lists head of household with gender & age-span; and all household others are listed as tick marks in age & gender brackets, males, females, slaves. 1850 is the first to list all members of the family, but it does not indicate the relationship of the household members (parent, child, grandparent, grandchild, aunt, uncle, in-law, friend, cousin, servant, hired or slave laborer), 1860 in KY & TN seems to have been done in tense hurried times, with given names listed only as initials. The 1870 census seems to be the first one to list blacks by name. The 1880 census is the first to indicate the relationship of household members to the head (therefore the wife may not have been the mother of all the children, but could have been a successive marriage, hence the need to identify birth gaps). Accelerated Indexing Systems of SLC, provided pre-internet alphabetized census index books. The preface explained census rules, and how you cannot prove the accuracy of data supplied because, if the family wasn’t home after the ideal three attempts, info could be taken from a neighbor or children or a local business owner, anyone who could comment on the resident family. The 1940 census was the first to indicate identity of the informant (by a circled “x” next to the name). If the informant was a later spouse, they may not have known the ages or birthplaces of all family members and may merely have been approximating. FYI:This same problem occurs with death certificates, to which early ones (away from rigidity of modern law) particularly are prone, in that the informant may not even have been a relative, but a neighbor, and so far removed from actual knowledge of the birth & parentage data could have been & often is found in error.
EXAMPLE (1): The 1820 Census of Stewart Co TN, enumerates the family of my ancestor Younger Herndon (1785 VA-1859 KY). Problem is that the teens in one category appears to number 4. This was an error published in the AIS Census Index Book accessed in the early 1980s, and unfortunately hand-copied into my notes. (REF: http://search.ancestry.com/cgi-bin/sse.dll?indiv=1&db=tncen&h=30476709&tid=&pid=&usePUB=true&_phsrc=aMc2762&_phstart=successSource&usePUBJs=true&rhSource=7734. Original data: Jackson, Ron V., Accelerated Indexing Systems, comp.. Tennessee Census, 1810-91. Compiled and digitized by Mr. Jackson and AIS from microfilmed schedules of the U.S. Federal Decennial Census, territorial/state censuses, and/or census substitutes.) Sadly, it was a mistake taken from an 1820 census and incorrectly applied as an 1810 tax list when no 1810 census or substitute exists. (See: “Bogus 1810 Tennessee records” (http://dgmweb.net/Ancillary/Warnings/1810-TN.html). Yet, for over two decades, it was believed that there were four teens in that age bracket in the family. This was important in determining whether this was a household containing two child-bearing couples or only one. After digitization & indexing & transcription of it online at Ancestry.com; I was revisiting the problem, viewing the original online when my computer froze. I went to another computer (this one with a larger screen) and pulled it up again, this time taking in everything the indexer recorded, saying the numeral was not a four but a one—I thought how can that possibly be right?! Then I pulled up the digital image and poured over the whole page, not just the part pertaining to my ancestor. I compared all the numerals. To me it looked like there were two types of numeral four: a pointy-top closed four & an open-top boxy four. Turns out (by checking the sums calculated by the enumerator on the original document)—in every instance the pointy-top closed fours were in fact a deeply flagged numeral one overwriting a dash, leaving the appearance of a four, but being in actuality a one. Only the open-topped boxy fours were true fours. That was a consistent fact in every calculation, and the only thing that brought agreement with the calculations of the original enumerator, enabling the transmission of his definitions identifying his own numerals. Thank Heaven the indexer caught the true identity of the numerals! It was far from apparent and took a good deal of study or training on their part (in contrast to the hastily transcribed names which often are erroneous). This indexer was a God-send to me and this document proved there were no other teens old enough to be parents of any of the household, ensuring it consisted of one parent couple, Younger & presumably his wife, the only female above the age of !
The second pertaining to research on Benjamin Herndon (ca. 1760 VA-1814 SC):
BACKGROUND (2): This man, ancestor Benjamin Herndon, (http://person.ancestry.com/tree/53440451/person/13510502479/facts), is reported by his own grandson in 1891 (SAR application # 4434) as having served in the patriot cause in the American Revolutionary War; present in payrolls of Capt Samuel Hawes, Col Alexander Spotswood (later Capt James Upshaw, Col Christian Febigar)’s Company of the 2d Virginia Regiment (see authorities: Saffell’s 1894 3rd edition page 275; Gwathmey’s 1938, pages 350 & 372); from May-Dec 1777 Caroline Co VA, till joining the Horse Service on 30 Dec 1777 in Valley Forge along with five others from his unit on the same day; horse service identified in 1818 by surviving fellow-Private Obadiah Carter (viewable at http://revwarapps.org/w8585.pdf), as Colonel George Baylor’s Regt of horse called Light dragoons, the 3rd regiment of Cavalry, and confirmed by “A List of Men Listed in Col Baylor’s Regiment of Light Dragoons, transcribed & published in the Virginia Genealogical Society Quarterly in 1993, Vol. 31, No. 2, as being dated January 17, 1779. The author’s cited source was VSLA Personal Papers Coll Acc #22547.) Problem is that there were four Benjamin Herndon’s submitted as patriots to DAR; three of whom had some contact with Caroline Co VA (two were pre-Rev War; mine was post-Rev War).
Example (2): Rather than under Virginia State Library & Archives, the document was found at Library of Virginia, in Manuscript collections under title “Returns of Colonel George Baylor’s regiment, 1778-1779.” LVA Acc # 22547: “A List of Men Listed in Camp for Col Baylor’s Regmnt of Light dragoons, January 17, 1778.” (The document source is described by the repository as “Papers, 1778-1779, of the Virginia Office of the Quartermaster General, Public Stores in Fredericksburg, Virginia, consisting of a return, dated 1 March 1778, of arms and accoutrements in the Third Troop belonging to Colonel George Baylor’s (1752-1784) Regiment of Light Dragoons; a list, 17 January 1779, of men in camp for Colonel Baylor’s Regiment of Light Dragoons; and a roll, n.d., of horses purchased for Colonel Baylor by Miles Selden, Jr., of Henrico County, Virginia.” Much to my dismay, but in accordance with my wondering, although both the repository and Minor W. Weisinger, the 1993 author, both transcribed the date as 1779, which would be more beneficial to my proof needs, the truth is, that the final digit is not a 9 but an 8 which was partly obscured, written in the space within the overhanging lengthy lower loop of the “y” in the word “Baylor” which first inhabited the space in which the 8 was written—the strange 8 has no comparison anywhere on that page, but does not match the 9s throughout, and is formed much like the inverted v encompassed (pointed teepee-like) in the top half of a flip-flop sandal. (I’ve found other similarly-formed 8s in the Rev War documents digitized & indexed for 1778 at Ancestry.com; it’s as though they form the numeral upside down). I’ve suggested computer study of it with chromograpic or chromolithographic-type PhotoShop separation of color layers to verify it, but it really is discernible with the naked eye.
For my research, this costs an entire 12 months of my ancestor’s life, and what I’d rejoiced was presumably gained when I found the 1993 published transcription of the document, but by viewing the original myself, I learned the truth, which to me meant, that I could not use it as stand-alone proof to rule out another Benjamin who resided in nearby Goochland Co VA, made his will as sick of body on 17 May 1778 & died by 19 Oct 1778 when the will was presented & proven in court. This required use of supportive marriage & deed documents to establish that he had forfeited residency in Caroline County VA in the early 1770s, having purchased his first land in Goochland Co VA on 24 Nov 1770, married his only wife in Goochland on 26 Dec 1771, and as a resident of Saint James Northam, Goochland Co VA, sold 97 acres of Goochland land to his mother (who was coming from Caroline Co VA) on 19 Jul 1773; and again as resident of Saint James Northam, Goochland Co VA, he sold an additional 97 acres of the 388 acres on which the said Benjamin Herndon now lives. Thank goodness for the progression of marriage and deeds to prove residency, to be able to rule out the Benjamin who was no longer a resident of Caroline (& had not been there for five years before the start of the Rev War; & rather was a resident of Goochland 7 years before the 1777 Caroline Co VA pay & muster rolls & 8 years before the 17 Jan 1778 Baylor’s Camp return which listed MEN’s NAMES: Benjn Herndon; REGIMENT & COMPANY: 2[nd] V[A] R[egiment], Capt Upshaw’s Comp[an]y, PLACE OF ABODE: Living in Caroline County V[A].”
I recommend these online tutorials by webinar lecturer, Crista Cowan, to aid you in your search:
“Spelling Doesn’t Count: Tips for Finding Your Ancestors”
“Anyone Can Read This: Basic Paleography for Genealogists”
“What Does That Say? More Paleography Tips and Tricks”
Citation: By researcher, Rose H. Bonnell, 13 Aug 2016.
Provenance & Permissions: This blog post is the Intellectual Property of Rose Herndon Bonnell (https://wordpress.com/post/familydisambiguation.wordpress.
com/528) permission is granted for use when quoted in full, with proper attribution & citations.