One Last DNA Attempt

 

With an ancestral family comprised of many siblings, it is possible to test DNA connections through more than one branch of the family line. With my hypothetical father of second great-grandmother Sarah Catherine Laws Davis, I've already had the chance to check out DNA cousins who descended from Larkin Laws and Pine Dexter Laws—with more promising results coming from descendants of Larkin than those of his younger brother.

There was one more set of DNA matches to attempt unraveling: matches whose ancestral connection was through Larkin's and Pine Dexter's sister Elizabeth Laws. Tracing the line from Elizabeth to the ten supposed DNA cousins who descend from this branch of the Laws family has proven more tangled than my attempts with Pine Dexter's line.

To trace the line of Elizabeth Laws' descendants, I had to go back to some of the documents I had found at first this month. Reviewing these, though, opened my eyes to other family issues. If, for instance, Elizabeth was the head of this line of descent for this set of my Laws DNA matches, I needed to find a marriage record for her—or, at the very least, some records attributing births to this mother.

According to some of my DNA matches, their ancestral Elizabeth had children born as early as 1856. However, when I returned to the census records which contained her supposed father, William Laws, she was still listed in his household. If she had been married before the arrival of that child born in 1856, I would not expect to see a twenty two year old Elizabeth still in her father's household in the 1860 census. Granted, there are three very young children also in that household at the time, and enumerations from that time period did not explain relationships among those sharing the living quarters—but I believe I have discovered a different explanation for those three youngest children, none of whom have a different surname listed in the record.

Likewise for the 1870 census, where I found William Laws' household entered just up a few lines from that of his hypothesized daughter Sarah Catherine Laws Davis. Once again, there is a woman named Elizabeth in the Laws household. Though her age has advanced from twenty two in the 1860 census to only twenty eight in the 1870 census, such anomalies I have spotted before in other family lines.

The point in all this exploration is that it is likely that the proposed children for Elizabeth—at least according to the trees of those DNA matches—actually belong to someone from another line of the Laws family.

That, of course, leaves me with one fairly convincing exploration of DNA matches—that of Larkin Laws' seventeen descendants who match my DNA results—plus a somewhat less convincing set of matches from Pine Dexter's line, and a definite no-go for their sister Elizabeth's line. If I take the same approach as I did yesterday, though, and look both ways on that DNA street, perhaps that exploration may offer up some siblings or cousins for William Laws, himself, possibly pointing me to who William's parents may have been, as well.

Before I reach that far beyond my goal for this month, today's exploration pointed out a few additional details I need to examine about William Laws, himself. While I am still on the hunt for a will—or at least burial records—for William, I'll take some time next week to read between the lines on what I noticed about the census records over the decades for this William Laws household.

When Crossing the DNA Street, Look Both Ways

 

Finding DNA matches who confirm my connection to Larkin Laws, the possible brother of my second great-grandmother Sarah Catherine Laws, was encouraging. My goal is to determine whether Larkin's father William was also father of Sarah Catherine. But I can't just assume that the one line's connection is enough. After all, regardless of how unusual a name like Larkin might have been, there could be another sibling set out there paired with the same names.

To bolster my confidence in this conclusion about Larkin and his dad William, I decided to look at DNA matches who descend from Larkin's brother with the equally-unusual name, Pine Dexter. Although Larkin gifted me with seventeen DNA matches according to AncestryDNA's ThruLines tool, there were only four matches for Pine Dexter—but four, I thought, would be enough. Enough, that is, until I surveyed the situation more closely.

Starting from the top of the list—a DNA match who shared only forty two centiMorgans with me—it didn't take me long to realize the ThruLines lane that led from Pine Dexter to my match had a big road block two generations in. Pine Dexter I could document, and his son I could find records for, but the suggested grandson in this match's proposed line of descent was not one I could replicate through documentation.

Sometimes, we miss details about a family's immediate family. As I saw yesterday, it can be easy to miss one child out of the family, if changes are made mid-decade between census enumerations. But try as I might, I could not locate this person ThruLines claimed as being Pine Dexter's grandson. 

Since the entire line of descent is usually diagrammed in the ThruLines tool, I decided to reverse research directions. Instead of looking down the line of descent from Pine Dexter all the way to my DNA match, I looked the other way. Starting from the first person listed in the line of ascent with dates of both birth and death, I looked for documentation to confirm that relative, then move upwards from there, building my own tree.

There was a second reason for trying that approach: the second DNA match identified as a descendant of Pine Dexter also claimed that undocumented person as an ancestor. For this second roadblock, it was far easier to trace upwards, based on documentation. There, I could clearly see which line of the Laws family this DNA match should be claiming as an ancestor: someone named Aaron Laws, son of Erwin, who like my William also was born in North Carolina.

Whether these two DNA matches descend from a sibling of my William or one of William's cousins, it is clear that the matches—if the connection is based on this part of their family tree—are much more distant than are those descending from Larkin's line.

I'm not sure I'm ready yet to grapple with that relationship tangle. This exploration, however, did remind me not to rely too confidently on my past tree-building work. The reason I've developed such a broad, "bushy" tree is that I purposely include all collateral lines in each generation, then chart all their descendants.

With a tool like that, my natural inclination is to trust using what I've built to look downward through the generations, when in reality, I still need to be prepared to look upwards from the current generation's identified parents. That, of course, means building trees for matches from a reliable starting point, but with the ThruLines assist, that is easily accomplished, even for those subscribers with private trees who have opted in to sharing with matches.

Yes, for a tree like mine, it is easy to look up matches in one direction: from ancestor downwards. But I need to look at this as a two way research street, and look both ways. Sometimes, we need to look backwards in time from the present, to double check the work of our DNA match.

The Winding Path of DNA Descent

 

Genealogical puzzles, fortunately enough, can be approached in more ways than one. That is a good thing in my current case, while I'm puzzling over just who it might have been that was my second great-grandmother's father. The woman I was told was Sarah Catherine Laws, wife of Thomas Davis, showed up in so many records simply by the given name Catherine. I knew her maiden name was Laws—but which Catherine Laws? And which father would be the one for the right Catherine?

While I've spotted a possibility with William Laws—who, thankfully, gave three of his children un-missable names—it has been a rough go to pin him down as clearly the father of Catherine. For one thing, as we've seen, William moved around a lot. That pattern seemed to be adopted by his sons, making their unusual given names even more valuable to me. Although I can't seem to pin William down as to a date of death or place of burial yet—not to mention, a will would still be helpful—I realized there is another approach we can take: follow the winding path of his sons' DNA descent.

As you can imagine, even that approach has its challenges, as the family seemed to have a multi-generational way with moving from place to place, or appearing in multiple marriage records. But I decided to try my hand once again at the chase based on genetic genealogy this time.

If William Laws was my third great-grandfather, then right now I'd be seeking DNA matches at the approximate level of fourth cousin. A relationship that distant isn't likely to share much genetic material—the range, according to Blaine Bettinger's helpful Shared centiMorgan chart, could be up to 139 centiMorgans at the most, but 35 for a mid-range. Then again, it could be that my fourth cousins and I share absolutely no genetic material at all.

In the case of the ThruLines suggested DNA matches for my Laws ancestor at Ancestry.com, my top match for William's son Larkin's line shared a bit over ninety centiMorgans with me. The next closest match shared fifty five—and the further from the top of the list I got, the more precipitously the numbers dropped.

There was another problem with the Laws matches I found: that top match with over 90 cMs shared genetic material with me in four separate segments. I would have been far more encouraged to see those centiMorgans all contained within one segment. As the total centiMorgan count decreased, those many segments seemed to dilute the possibility of finding a verifiable match through corresponding documentation. But I tried my hand at it anyhow.

My first step was to look for Laws matches who shared one larger segment with me, rather than four or five smatterings of genetic material. For two matches who each shared one segment measuring twenty three cMs with me, I was able to confirm their line of descent. That was encouraging.

Then, I tried my hand at the names closer to the top of the list, the ones with the many segments shared. I had tried tracing those lines on paper several years ago without any success, but thankfully this year, the documentation was there to confirm the connections. Not that this was easy; many moves, many marriages, many incidents causing me to backtrack to pick up events hidden in between the census years made me wonder whether I'd ever find satisfactory confirmation of the connections.

So far, I've verified four out of seventeen matches in the ThruLines readout for William's son Larkin. I'll eventually work my way through the rest of that list. The work will certainly go more quickly, now that the trailblazing effort has laid out the patterns and the locations. From that point, I'll move on to confirm the four ThruLines matches descending from William's son Pine Dexter—a name which can't be missed, although one that has suffered much recorded abuse.

Then, unless I can find contradictory evidence that there was another William Laws out there during the same time period in that same northeastern Tennessee area—or another set of brothers named Larkin and Pine Dexter—I think it will be safe to assume my second great-grandmother Sarah Catherine was daughter of William Laws. It all comes down to doing the grunt work of building a solid paper trail—but DNA can certainly help point the way. 

Moving Along Now

 

There may have been a reason why William Laws and his sons Larkin and Pine Dexter were difficult to find, even though using the FamilySearch Labs' Full Text search. Perhaps they might have been just a shade over the side which in more recent times would have prompted law enforcement to warn, "Move along, now."

I went back to the FamilySearch Labs website for a second attempt at finding William—or at least his sons—mentioned in a will in either North Carolina or Tennessee. Bottom line: despite reversing my technique to use "Larkin Laws" as the keyword and William as the main search term, I couldn't find any last will—or even any property records. I looked in both North Carolina and Tennessee.

However, I did stumble upon something else.

Though I am not yet convinced that William Laws was my second great-grandmother Sarah Catherine Laws Davis' father, I have found her family living quite close to the household of a man by that name in the 1870 census. An added perk comes our way thanks to DNA matches who descend from those distinctively-named possible brothers, Larkin and Pine Dexter.

Though my second try at digging up wills or deeds has failed dismally—not that I'm quitting just yet—I did run across some odd records. 

Try this mention from the Yancey County, North Carolina, court minutes for the July term in 1855. Remember, I had spotted a William Laws family in Yancey County for the 1850 census, with a family which included names such as "Larken" and "Pendexter."

According to the North Carolina court minutes, there was a note about a judgement in regard to a case labeled "State vs. Larken Laws." No further detail, but looking for additional records, an earlier entry from the January 1854 term of the same court noted that the "defendant Larkin Laws," though called, failed to appear at that session.

Perhaps the no shows might have been owing to the family's decision to remove from the state for more comfortable quarters across the border. Remembering that William's possible daughter had married by 1856 and was living in Washington County, Tennessee—a match that always had me wondering about how the connection was initiated—I wasn't surprised to find the next mentions of the Laws family in that new location. The court minutes from Jonesborough, the county seat, at the beginning of 1857 provided a list of names for the "insolvent poll tax" for 1855 and 1856. If you guessed William Laws was mentioned, you are spot on. Bonus points if you wondered if his son Larkin was mentioned as well.

Looks like it was time to move on once again.

Some Choice Terms

 

It's the moment of truth: I'm using the FamilySearch Labs Full Text search to see if I can find any trace of what became of my second great-grandmother Sarah Catherine Laws Davis' potential father, William Laws. My hope is to easily find a will drawn up by William among the legal papers in the state of his last residence, Tennessee. 

The trick, apparently, is to choose my search terms wisely—date, place, record type, keywords. There is a reason for this. No—there are several reasons for this. Let me list the problem with the choices.

First is the obvious: looking for someone surnamed Laws in a collection of legal records may be difficult, if a good number of those records include the oft-use phrase, "according to the laws of...." So I thought of a preemptive strike: couple my search for William Laws with a keyword. I have two easy possibilities, each of them having to do with his penchant for selecting unusual names for his sons: the name Larkin, and the name Pine Dexter.

Well, Pine Dexter didn't help much. Apparently, there are several mentions in legal records in the state of Tennessee including the word pine. Why, I'm not sure, but the count of hits for this choice provided by the FamilySearch Labs search page was rather discouraging. And surprisingly, Larkin was a surname used enough times in Tennessee to hamper my search progress using the other keyword option. I have yet to try out the name of the third son—Wiley—but my gut tells me that name wouldn't serve me much better.

Pinpointing the location for a possible will for William also didn't help. I couldn't rely on his remaining in the same Tennessee county where I had found him for the 1870 census—Greene County—because I had already found him moving quite a bit for the enumerations I had traced him to in the past. However, it looks like it might be worth my while to return to the program to do a separate search for each county location.

It can make all the difference which search terms we use to find our ancestors. Even if we are using a tool as effective as the Full Text search, the approach sometimes requires us to experiment with the terms we use. Then, again, I have no way of knowing whether William stayed in the last Tennessee county where I found him, or backtracked through his previous locations during the prior decades in Tennessee—or, worse for my search, returned home to North Carolina where he had been born.

Then, too, it helps to use the online system when it is not displaying a warning that some features "may not be available" while the website operators are "making improvements" to the site. Perhaps, among other records, any details about William Laws simply weren't available during my late-night research sessions—one tacit way to put in a good word for the genealogical early birds among us.

Time to Make Choices

 

An email I received midweek reminded me: it's time to make my choices about the upcoming RootsTech conference. Yes, I know it doesn't begin until March 6 this year, but with 113 online workshops to choose from over the three full days of the event, this sounds like an opportunity well worth some advanced consideration.

Just looking over the RootsTech calendar of online events, I can see there are several sessions focused on technology—and especially the idea of integrating the use of artificial intelligence in genealogical research. Sessions on using DNA testing figure somewhat prominently in the workshop offerings—including a talk by now-retired Family Tree DNA founder, Bennett Greenspan, called "Bennett's Y-Chromosomal Time Machine."

Because of the worldwide focus of the organizers of RootsTech, it's no surprise to find several workshops focused on ethnic research. Everything from several sessions on addressing the many challenges of African-American research or Jewish genealogy to exploring Native American research can be found on the schedule. But that's not all. There are sessions on exploring Central and South American roots, Chinese ancestry, as well as the expected European origins such as Danish, Norwegian, Scottish, Irish, and French.

I was particularly encouraged to see a couple sessions being offered on Polish research, by fellow genealogy blogger Julie Szczepankiewicz of "From Shepherds and Shoemakers." It isn't often that I find any workshops on that topic in general conferences, despite the United States including a sizable Polish diaspora. Since Polish research figures prominently in my goals for my Twelve Most Wanted each year, I'd like to see more of such training made available.

Besides Polish research, another annual focus for me involves following my father-in-law's Irish ancestors through Canada to the U.S. In one particularly tricky branch of his family, I have been partnering with two Canadian researchers who are also interested in this same set of Irish surnames. We have lately been looking at emigration schemes from Ireland to Canada, some of them pre-famine, and for that I was delighted to find a RootsTech offering for this year on that very topic: American Ancestors' Melanie McComb discussing "Escaping the Famine: Exploring Irish Settlements in Canada."

Finally, for my mission to continue sharing family stories—especially those difficult-to-face episodes in our families' past—I found genea-blogger Laura Hedgecock's session, "Navigating Sensitive Topics," to be essential for those of us who need to put "difficult chapters of the family's history" into writing.

There is, of course, no way I could absorb each one of these 113 sessions offered at this year's RootsTech in a week's sitting. No problem, though: many workshops will be available on-demand for up to three years after the conference's close. Scheduling my watch list not only to plan my attendance but for later viewing will probably be a wise move, just in case last-minute Life happens. It's good that room for flexibility has been scheduled into the choices for the RootsTech plan, too.  

Outliers

 

"Ninety eight percent of women will experience...."

How many times do we see statistics that seem to put us in a box—and assume without further examination that that is where we belong? Not that I want to launch into a dissertation about standard deviations and obtuse calculations, but I have lately been reminded of a certain detail about my relationship to assertions about percentages: I am an outlier. If everyone else is part of the ninety eight percent, I'll be camped out with the two percent.

What that means is when those who know far more than I do about statistical calculations come up with their proclamations of the way the average person should be, I'm usually not in that group. In other words, when everyone else zigs, I zag. 

When it comes to family history, that scenario holds true. Several years ago, when my siblings, cousins, and I finally broke through the mystery brick wall concerning my Polish patriline, I kept on the research trail until I could pin down the precise location of where my paternal grandfather's parents might have lived: it was a specific region in what is now the country of Poland but was, back in that century, a part of Prussia known as Pomerania.

Shortly after that discovery about my roots, my husband and I traveled back east, where I got the chance to visit with my eldest paternal cousin and reveal what I had found.

"So we're Pomeranians?" my cousin quipped.

Not to be outdone, my husband—for whom I have almost pinpointed his maternal ancestry as coming from the region of Alsace-Lorraine, now in France—retorted, "Yeah, and I'm an Alsatian."

We seldom talk about these historic regions any more. They are swallowed up by larger nations without any thought about how the previous regions may have had differing characteristics which are now lost to the record. The heritage of these isolated people groups may have been lost to us, but could characteristics we think of as outliers actually owe their origin to these lost-to-time people groups?

This is not a dynamic isolated to those northern regions of Pomerania or Alsace, but expand to include other regions in Europe which also had a history as a distinct group. Catalonia as part of Spain comes to mind—as does the nearby Basque region, as well. The same holds true for the other continents, as well, as witnessed by the ongoing search to more accurately represent such people groups in the reference panels used by geneticists—and, by extension, genetic genealogists.

Conclusions based on results of studies drawn up using subjects from specific regions (or reference panels) may not apply as accurately to people who descend from a different isolated people group. In other words, a study based on the population of Japan, for instance, might not produce findings as accurate for descendants of people groups from Africa. In one way, we become each other's outliers.

On the other hand, as more of us from around the world participate in DNA testing, we broaden that universe by including our own "norm"—the outlier for those other groups—into the mix. I've read comments concerning the lack of medical studies including, for instance, indigenous populations of North America and how that could impede the ability to reach conclusions that would be more helpful for those "outliers." Likewise, I wonder how more applicable medical developments might be for those of African descent if individuals could someday pinpoint not only the regions of origin on the continent, but their specific ancestral people groups.

We may all be part of humankind, but we contain a world of difference between us. With discoveries in DNA, the more we all test, the more we can all discover about that fascinating genetic mosaic—the details we can now only chalk up to being part of a vague concept like "outliers."