Archive for the ‘Evaluating Evidence’ Category

My great-grandfather John Smith was born in Georgia and migrated to Jacksonville, Florida sometime around the turn of the century. His roots in Georgia continue to be one of my greatest brick walls. I’ve been researching him in more depth recently, and I had a huge breakthrough on his wife’s family yesterday. I am so excited! This is an excellent case study in evaluating evidence.

John Smith married Georgia Harris, and for many years I knew almost nothing about her since she died at the age of 45. I had some success earlier with Georgia’s roots that was a big part of this new discovery. Georgia had two sets of children, one set with first husband Isaac Garner in Madison Cty, FL, and another with presumably her second husband John Smith in Duval Cty, FL. No one in my family knew about that first marriage. Oddly, the only censuses in which John and Georgia and children appear together is 1930 and the Florida state census of 1935. Clearly, they were if not married then at least having children together before then.

My grandmother wrote in her family bible that “Matilda Vickers” was the name of Georgia’s mother. THIS Matilda Davis, Georgia’s mother, migrated to Philadelphia, PA with her other daughter Ruth by 1920, as I detailed in the earlier post.  Matilda’s name on that 1920 census in Philly is “Garvin,” but it’s clearly the right woman since she is noted as being the “mother-in-law”. I assumed Matilda died there in PA–this is an important point to remember (and take note of all my assumptions, LOL).

By 1930, Matilda’s son-in-law Nish Torrence had remarried and was now living in Camden, NJ. This is the census tracking for Matilda Davis/Garvin, mother of Georgia:

Matilda Davis/Garvin, 1900-1920

When the 1940 census came out, I was looking for other family members in Jacksonville. When I looked up Georgia’s son Cornelius Garner, I was surprised to find this:

1940 Cornelius Garner

Cornelius was living with a “Matilda Vickerson” who is 73 and widowed. Cornelius’ relationship to Matilda is listed as “Roomer”. I decided to investigate this mysterious Matilda who kept popping up. Was she the Matilda mentioned in my grandmother’s Bible?

In 1935, a “Metilda Vickers” is living in Jacksonville, FL with Cornelius Garner & other family members. At some point in 1930-1931, city directories show “Matilda Vickers” as living in the house with John & Georgia’s family, although on the 1930 census Matilda is living with a woman named Katie Middleton and described as a Cousin:

City Directory

1930 Matilda Vickers

Before 1930, I could find no evidence of Matilda Vickers in Jacksonville, and I was unable to discover the name of her deceased husband. To recap Matilda Vickers visually:

Matilda Vickers, 1930s-1940

Matilda Vickers died in Jacksonville in 1944, and John Smith was the informant on her death certificate, but no relationship is given (Dagnabbit!).  After dusting aside some of my earlier assumptions, the key question was: is the Matilda Davis, mother of Georgia Harris, who by 1920 is living with her daughter’s family in Pennsylvania as Matilda Garvin, the SAME Matilda Vickers/Vickerson, who appears in Jacksonville by the 1930s?

The ages matched pretty consistently. It would answer why Matilda was associated with John Smith (she was his mother-in-law). It would answer why I couldn’t find that name before 1930 in Duval Cty, FL (she was living in Philadelphia). But I couldn’t explain the surnames. Incredibly, vital record searches solved that, along with a little creative thinking in terms of the names.

I found that a Matilda Davis married a man named Frank Gowen in Jacksonville in 1916.  Of course that was a transcription error–his surname was Garvin. He died in Jacksonville on 12 May 1918, leaving widow, Matilda “Garvin” on his death certificate. That’s why Matilda appears in Philadelphia with that name. And amazingly enough, in Philadelphia, I discovered a Matilda Garvin married Peter Vickers in 1920. And yes–he died there in June 1925. Matilda then moved back to Jacksonville before 1930.

I couldn’t find the records before because I was not looking under the correct surnames and also because “Gowen” did not turn up in a search for “Garvin.” Also, the fact that Matilda had 2 marriages in different cities with men who died shortly afterward added to the confusion. I am in the process of ordering the marriage records to confirm all of the above, but I feel very confidant in stating that:

Matilda Davis= Matilda Garvin=Matilda Vickers/Vickerson!

Keep in mind, I could only make the connection once I threw out my insistence that Matilda Davis had died in Philadelphia, and that the census taker had mistakenly entered her name as “Garvin.” Assumptions are fine, but remember that you made them and always be ready to re-examine them in the light of new evidence. One assumption was correct–my grandmother’s entry was wrong. She probably remembered that John Smith was related somehow to this Matilda, and assumed it was his mother. In fact, her mother-in-law died the year before she married her husband so she did not know her personally, so this mistake is perfectly reasonable.

I feel really good about solving this puzzle. I’m even now exploring a hunch that the missing marriage record for John Smith & Georgia Harris could be the one I see listed for John Smith and Florida Harris  in 1916.

Genealogy never stops being exciting for me. And possibly the best part of this is I added a new ancestor to my tree–my 4th great-grandmother, Virginia “Viney” Nealy, Matilda’s mother as shown on her death certificate.

Stay tuned for my next post where I explore just who was this Katie Middleton that Matilda was living with?

Read Full Post »

Inventoried Slaves

I talk alot on this blog about slave and slaveowner research because it’s one of my primary areas of interest. For those of us descended from enslaved ancestors, probate records are one of the first record sets we are taught to explore. If we’re lucky enough to discover that the slaveowner died before 1865, we may find our ancestors named in their will or listed in their inventories. As we advance in our skills, however, we’ve got to look even closer at probate records beyond just the will or inventory, not to mention the need to search beyond the slaveowner himself.

In this post, I want to show a recent example of how careful tracing through and understanding of those “other” probate records may provide a more complete picture of our ancestor’s path through the family. Familysearch has now posted probate record series for many states making this technique possible to do from home. Many Maryland counties are now up, which is what enabled me to explore this more fully.

First, I created a family tree of the slaveowner’s family. I encourage my students to use Rootsmagic or Family Tree Maker (or whatever software you have) and to create a separate file for the slaveowner’s family. This will be invaluable to your research. Many slaveowners married their first cousins, which makes keeping the names straight difficult (this is one practice Africans in general never imitated). It is imperative that you know at a minimum the parents of the couple, when/where the parents lived and died, all of the couple’s children, when and where they died, and especially who the daughters married.

As long as they died before 1865, start probate tracing with the slaveowner, then trace his wife if she outlived him, then their children if necessary. In a previous post, I talked about the various steps in the process, both for dying with a will (testate) or dying without a will (intestate).

Those who follow this blog know I’m a fool for charting. Take a look at the chart I made for Martha Willson, who died in 1837:

Magruder chart

Martha left a will (unlike the majority of people). I started with her date of death, and went to the probate book that covered those years. I went to the index, and easily found “Martha Willson, Will” on Page 164 of Volume V. Keep in mind that I am using the term “probate” to refer to these records in general. What they are actually called varies by state and locality—in the case of Maryland, these volumes are actually “Will Books [that also contain] Inventories and Accounts,” and are kept by the Register of Wills.

Back to Martha: my chart started with her Will, and noted any relevant phrases about her slaves. She specified that “Dick and Nelly” have their choice of going with either her son Robert or her son John. Dick and Nelly (from Martha’s inventory) are elderly slaves and were probably unable to do much if any work at ages 60 and 64. Martha specified that the rest of her slaves be sold at private auction.

The next important document in her estate probate is the Bond. Executors (in the case of a will) or Administrators (in the case of no will) must post bond with the State that they will faithfully execute their duties. It is important to know who is posting bond. They are usually family members. For example, Otho Magruder is Martha’s son-in-law. Also, a $20K bond told me this was a relatively wealthy estate.

Martha’s Inventory named 9 slaves. The next step after the Inventory were the Sales of her estatethis is where slaves can be missed! In these pages, the other 7 slaves are sold, but (because I know Martha’s family tree) they are all sold to her children. It seems that it was important to keep them “in the family.”

The next steps in Martha’s estate probate include a listing of Debts and periodic Accounting of the Estate. The number of Accountings (1st Acct, 2nd Acct, 3rd Acct, Final Acct, etc.) depends upon alot of things, like the size of the estate and whether or not minor children are involved. Those Accountings can also contain information about slaves, especially slaves being “hired out” for that year, so peruse them carefully. If minor children are involved, Guardianship records should also be traced, but may be handled in a different court.

Two other things I want to point out about Martha: Her estate probate spanned across 10 years. In the beginning of my genealogy research, I didn’t understand the need to trace forward decades after a death, but it is entirely not uncommon to find probates spanning large periods of time. I now trace at least 20 years forward after a death. As I mentioned, Martha was wealthy by standards of her time. Her final estate value of $11,098 in 1847 was roughly the equivalent of $303,000 today according to standard of living worth calculators.

I had already charted Martha’s husband, Zadock Magruder, who predeceased her in 1809:

Magruder Cooke Admin Slave Data_Page_3

As you can see, Zadock died without a will (intestate) in 1809. His estate probate spanned 11 years. Notice also that in his 1st Acct the value of his estate was calculated in pounds, not yet American dollars.

Zadock had 16 slaves in 1810 at the date of his inventory (The child Rezin, age 7, is likely my gggrandfather). It was clear that 27 years later, in his widow Martha’s estate in 1837, 6 of the slaves she then owned had originally belonged to her husband in 1810. Most likely, the rest of the slaves were split up and given to one or more of their 8 children. Trying to find who went where is why I started this whole exercise to begin with. Who got Mariah and Lucy and Beck and the others? Why was Jerry to be set free? Sadly, I still don’t have enough information from these listings to put together definitive family groupings.

Zadock Slaves, 1810

Another important point is this: the actual division of slaves, and to which children they went, is not always written in the official probate books. I have found them many times in original case files or loose papers (i.e., the papers that are apart of the probate proceedings but not necessary recorded in the official books). Always try to find that slave division. You can see from Zadock’s chart above that he owned 16 slaves. His wife Martha kept at least 6, so we know the others were likely divided amongst his children, but, that division is not recorded in the probate books.

This blog post was probably too long, but, hopefully I’ve highlighted a strategy you can use to get the most value out of probate records. Try it out on your slaveowning families, and see what you come up with. I’d love to hear about your finds!

(If you want to catch up on some of my previous posts on slave/slaveowner research, click on those topics in the right -hand “What I Talk About” box.)

Read Full Post »

I’m building a case that just got stronger. I have posted before on my long odyssey researching the Fendricks family, my maternal great-grandmother’s maiden name. I had a breakthrough in August 2009 and found a duplicate death certificate earlier this year.

In this line, I encountered the common roadblocks of moves across state lines and name changes, on top of the fact that these were rural African-Americans with enslaved parents, and the name they ended up keeping with was still complicated. Sheesh.

To summarize past research, I was stuck with my gggrandfather Mike Fendricks and wife Jane, who after many years I found newly married and living in Savannah, TN. The problem was that they were from Alabama and I had no idea what county. In 2009, the breakthrough came when I developed enough skills to really use cluster research  techniques. In short, this technique suggests researching the people your ancestor had close relationships with. Mike Fendricks, as an elderly man in 1920 was living in the house of a man named Dee Suggs, so, since I was stuck anyway, I decided to veer off and research this Suggs family. You can read the lengthier original post for more details, but the research led me to Lawrence County, Alabama, and this census  grouping in 1870:

1870 Census

My theory was—and has been—that this mysterious “Dee Suggs” is the same man shown on the census above named “Dewitt Suggs.” The evidence supports a conclusion that theMike” in the household is my ancestor and his brother, which is why they both migrated to Hardin County, why Mike was the witness on Dee’s marriage license, and why Mike is living with him in 1920.

Slowly I’m putting together a good case, but the fact that the 1870 census does not state relationships was a hindrance. I couldn’t find Dee Suggs anymore after 1920. I had a hunch recently that perhaps he went back home to Alabama and that hunch paid off when I checked the Alabama Deaths database on Familysearch. I found him, and his mother was indeed “Fronie Suggs” (Sofrona):

Dee Suggs

I was so excited! I couldn’t believe I found this. It’s not a smoking gun, as Mike’s death certificate in TN does not name any parents, but this lends significant support to my theory. I talked in a previous post about how sometimes all we can do is build a case.

It appears that a number of black “Suggs” were centered around Russellville, AL, and buried at New Home Cemetery (thus, a new research avenue). This death certificate also identified his father “Obe[diah] Gholston,” which illustrates another previous post topic, finding fathers who are not enumerated with the family in 1870.

I have not been able to find any large “Suggs” slaveowners in Northern Alabama, so finding out who may have owned Sofrona and her children will take some time.

One conflict in my theory is why in 1920, the census enumerator wrote that Mike was a “boarder” and not “brother”. However, using the Genealogical Proof Standard, this conflict is easily explained given the abundance of census errors.

Read Full Post »

This is a chart I like to share with my classes to illustrate the importance of collateral research. Like almost everyone, I was focused strictly on my grandparents and great-grandparents–all my direct ancestors–when I first started my research. Opening my eyes to include all siblings in each generation, and understanding the necessity of knowing the informant (and their relationship to the decedent) unlocked a world of information.

My great-grandmother, Beatrice Prather Waters (shown 5th from the left in the picture on the blog), had 8 siblings. The table below shows parent’s names gathered from her death certificate and 6 of her siblings (2 died in states where I can’t get copies of their certificates). I have also included who provided the information:

Prather Siblings

There’s a lot of room for confusion here, and the table makes that point clear. Had I stopped at just my great-grandmother, I would have been forever lost, because Beatrice’s son remembered “Eli” instead of the correct name “Levi.” And he didn’t remember the mother’s name at all. Beatrice’s mother was Martha J. Simpson and 4 of the 7 death records got it right. What’s interesting is that Margaret Simpson and Susan Simpson were in fact family members, but they were not the wife of Levi. People gave what they remembered at the time. Margaret was Martha’s stepmother, and Susan was her Martha’s sister. Of course, I can’t remember what I did this week.

All of this information was correlated with census, probate, deed and other record types to paint as clear a picture as possible of this family.

When you research, don’t forget to research all the siblings.

Read Full Post »

I read an article a few weeks ago that I think every single genealogist should read, and I was excited about sharing it with you all. It is a special issue of the National Genealogical Society Quarterly from September 2001 (Volume 89, No.3). This issue was completely devoted to discussion of the Thomas Jefferson-Sally Hemings affair that I’m sure everyone has already heard about. If you are a member of NGS (which I highly recommend) you can log in to their website and download this article from their NGS Quarterly archives immediately.

The esteemed Helen Leary, who is an extraordinary genealogist, tackles the subject in an article entitled,Sally Heming’s Children: A Genealogical Analysis of the Evidence,” which starts on page 165.  It is a 40-plus page article, long, but well-worth taking the time to print out and read. Helen illustrates use of the Genealogical Proof Standard to one of this country’s most enduring mysteries: Was Thomas Jefferson the father of Sally Heming’s children?

In Helen’s gifted hands, the evidence is laid out (truly massive amounts of evidence), every hypothesis tested, each conflict addressed and a clearer conclusion you won’t find anywhere. Helen is a masterful teacher, and a thorough researcher. I feel like I grew as a researcher just seeing how she approached the topic and addressed each and every concern. I will continue to apply these methods to my own research.

DNA testing performed in 1998 matched Sally Hemings youngest son Eston’s DNA to that of a Jefferson male. Along with the other evidence, I particularly enjoyed how Helen illustrated handling of bias on the part of researchers, and how that bias can negatively affect results. This article also showed how you can’t the play the game of “XYZ coulda happened” with research. Genealogy is not about coulda, woulda, shoulda.

I’ll leave you with a clip from the 1870 census that this article discusses that just blew my mind. In 1870, a census taker in Ross County, Ohio, enumerated Sally’s son Madison (most of whom went on to live as white people) and wrote the following notation into the census next to his name:

“This man is the son of Thomas Jefferson!”

1870 census

Now, that has got to make you say Wow. I’ve never seen anything like that before. I hope you’ll go read this article, come back here and let me know what you thought. I encourage you to read the entire issue: an article by Thomas Jones dissects the “official” report done by the Thomas Jefferson Scholars Commission (who continue to deny the pairing), and there is an excellent article by Gary B. Mills about proving children of master-slave relationships.

Read Full Post »

My ancestor Malinda Holt was enslaved by Giles Holt of Hardin County, Tennessee. Giles enslaved her along with one other woman, named Judy (sometimes written Julia) Holt. Both woman had multiple children of around the same ages. Although I will probably never know whether or not Malinda and Judy were actually sisters, I have decided to track Judy’s children as my relatives because it is obvious that their children had close kinship ties and considered each other family. I did a post sometime ago about Judy’s son James and his amazing life story. This rough chart shows each woman and their children:

One (of the many) wretched things about slavery is that often we trace back to that elusive female, listed as head of household in 1870 and we find no hint of a man. Our climb through the tree stops—there is no other branch to trace. Particularly if the children have light complexions, we wonder whether our ancestor was one of the millions who conceived children by white men in the community. We all know that slaves formed families with enslaved neighbors, but this relationship can be difficult to uncover if they are not found living together in 1870.

As I tracked Judy Holt’s children, a delightful surprise emerged. Judy’s son Henry Holt died during the Civil War, while he was a member of the 55th US Colored regiment. His mother Judy’s subsequent application for a pension in 1887 provided me with details of her children’s names and (approximate) birthdates. One of the depositions, from fellow soldier Richard Kendall, also included this little gem:

Richard “was well-acquainted with Henry Holt and knew his family. I do not know whether his father is dead or alive. His name [was] Sam Dixon.”

At last I found evidence of Judy’s relationship with a (presumably) black man. But where was he? For years I couldn’t find him because of my utter inability to be very creative with name spelling variations. But looking through Hardin County probate records recently led me to the will of one Elizabeth Dickson (note the spelling). That rang a bell in my mind, and sure enough, among the legacies she left to her daughter Jane was this:

“and she is to have my black man Samuel while…she lives single”

Racing back to Ancestry, there he is: Samuel Dickson in 1870, in the town of Savannah, right where he should be, and the right age, although he appears to be married to Lucinda now. Or perhaps Lucinda is a daughter.

1870 Sam Dickson

I got even luckier (I think its all luck at this point) when Judy also included in her pension file the fact that her daughters Sarah and Frances were both now surnamed “Davy”. Using that surname, I found, Judy’s daughter Frances’ (nicknamed Fannie) death certificate in 1917. Guess who was listed as her father? Sam Dickson.

Fanny Davey

While there is no way to know exactly how many of Judy’s children were fathered by Sam, the fact that I was able to uncover evidence for two of her children is pretty amazing. This is also a good example of using the technique of cluster research, to expand your vision and research the group of people surrounding your direct ancestors. The hunt for elusive enslaved fathers continues.

Read Full Post »

For the first time in my research, I have found two death certificates for the same person, filled out by two different people. As if the missing and inaccurate records aren’t enough to make me crazy, now this.

I was doing some research review and perusing the Tennessee Death Database when I found this death certificate for Miss Mary Ella Copeland:

Death Certificate 1

Mary’s parents migrated from Alabama to Tennessee, and I have never been able to find any document that states where in Alabama they came from. What struck me is that this death certificate listed “Tuscumbia, Alabama” as the birthplace for her parents. I thought, Now I know I have this death certificate already. Why didn’t I ever notice the town? Sure enough, when I pulled out the copy I had in my records, it only listed “North Alabama” for Mary’s parents:

Death Certificate 2

Both death certificates are indeed the same woman, my great grandmother’s sister Mary Ella Fendricks, who married Abe Copeland. Both list the same death date, February 9, 1930. However, one was completed by her husband, and one was completed by someone named James Casey, who I find associated with the family, but I’m unsure of his exact relation. One certificate lists the “Gant graveyard” as the burial place, while the other lists the “Savannah Colored Cemetery.”

They both illustrate the weaknesses inherent in so many records: the information is only as good as who gave it. They list different ages for Mary Ella. Her husband says her parents were Mike Fendricks and Kate Sharard. James Casey says her parents were Mike Fendricks and Kate Suggs.    

Mary Ella’s mother’s name (according to marriage license and census records) was “Jane Eliza”. But so many records consistently state “Kate”, that I’m starting to believe that she was actually called Kate or Katie. Even my grandmother remembered that name. She died at a relatively young age and no record suggests another marriage for Mike Fendricks.

I wonder what circumstance would cause someone to have multiple death certificates? Have any of you seen this? Sometimes I think the ancestors just like to MESS with us!

Read Full Post »

« Newer Posts - Older Posts »


Get every new post delivered to your Inbox.

Join 95 other followers