What Can y-DNA Testing Do for Me?
Deciding to Take a y-DNA Test for Genealogical Research
In the past decade or so, consumer DNA testing has exploded. Several companies advertise their tests as ways to peer into your past and reveal hidden ethnicities or family origins. While this popularity attracts people into the hobby of genealogy, it really doesn’t tell the participants any specific information.
The next step for people trying to connect with their family histories is usually the testing of their autosomal DNA (also called atDNA). This is a mix of DNA from both of your parents and can be used for cousin matching. Based on the amount of shared DNA material, predictions are made to help guide people as they build their family trees. In the best cases, this kind of testing is very helpful in filling gaps in your family tree. Another popular application for atDNA is for adoptees wishing to connect to biological family members. Sometimes. these two activities are combined, as in a person looking for great-grandparents if they know their grandfather was adopted without any further details being known.
The general usable range of atDNA is the first several generations behind you. Your parents are one generation back, grandparents are two, and so forth. In my family, most of the successful atDNA connections we’ve made (multiple testers) have been 4–5 generations back, especially on my maternal line where my aunt tested her atDNA after about two decades of serious genealogical research. In the gaps she had, mostly along great-something grandmothers for whom we had no maiden names, my aunt was able to combine her excellent tree with matches to distant cousins to find some of those missing families.
For me, the most obvious gap was in my Brewster line, where I had no information at all about my great-grandfather. All I knew is that my great-grandmother had remarried at some point and even though she was alive when I was born, I never heard any stories about her first husband. My grandfather himself had died before I was born, so I didn’t really have any Brewster family stories at all. I had no Brewster cousins outside of my dad’s siblings, so I was facing a huge blank space on that side of my family tree.
At this point, it might seem obvious to take a yDNA test. Because the y-chromosome is passed virtually unchanged through the generations from father to son, any man’s yDNA can reveal dozens of generations of patrilineal descent. When hundreds of these yDNA lines are compared to each other, some commonalities are revealed that actually can be traced to a single common ancestor. These similarities are typed together into haplogroups which define these larger groups by that common, very distant ancestor. My y-haplogroup is I1, also known as I-M253. This line is generally thought to be 3,000–5,000 years old, originating in mainland Europe. It is particularly prevalent in Scandinavian countries and is closely aligned historically with Anglo-Saxon and Viking invasions, so it is also relatively common in England.
I did not take the y-DNA test when I was starting my Brewster research. First of all, it is an expensive test. I was in the middle of a career change and graduate studies when I began my serious Brewster research, so a y-DNA test seemed to be an extravagance at the time. Also, I wondered how useful the results would be to my immediate problem. This is the issue that many potential y-DNA testers face and what I hope to do with the rest of his article is to help you make the right decision for your research.
Generation One: Father
Testing y-DNA cannot definitively find your father. Instead, the basic testing resulting in a paternal haplogroup can help rule out candidates. My brother tested his atDNA with 23AndMe before I tested my y-DNA with Family Tree DNA, and his I-M253 result gave me something to compare to. When my I-M253 result came up, it didn’t prove we had the same father. Instead, it allowed us to confirm that we did not come from different haplogroups, indicating different fathers. In some cases, a child and father will not match at this basic level, which is called a non-paternity event. Not all NPEs are the results of affairs, but if they are unexpected then they definitely reveal some family discrepancy or secret.
Generation Two: Grandfather
I knew my grandfather’s name, so in building my family tree, I had myself, my two parents, and my four grandparents. If my paternal grandfather were missing, how could y-DNA testing help? Assuming you’ve tested atDNA already and have an unknown male first-cousin match, then his y-DNA test should match yours almost perfectly since you presumably share grandfathers. There might be a mutation in the grandfather-father-son line, but the likelihood for a non-match is extremely small and can safely be disregarded. Realistically, atDNA matches between unknown first cousins are much more likely, and that should be where you focus your energy.
Generation Three: Great-Grandfather
This is where I was four years ago — looking for a man named Brewster who was the father of my grandfather Francis Brewster. The best source of information for this would be family, so if you have relatives who knew your great-grandparents then their information is first-hand, and even if they don’t quite remember all the dates and places, they can provide clues like burial information and marriage dates. Armed with this information, you can start searching the US Censuses for free. Currently, all are online from 1790 through 1940, except for the 1890 Census which was lost in a fire.
Just like with an unknown grandfather, the chance for family secrets increases, so your great-grandfather may have had a family before he married your great-grandmother (which happened in my maternal grandmother’s case). Maybe your great-grandfather went off to war while your great-grandmother was pregnant and the man your grandfather thought was his father was technically a stepfather. This is one way surnames change, or what actually happens, the surname of the original y-DNA is different from the next generation’s y-DNA. Eventually, a thorough scouring available records helped me find my great-grandfather, Eugene Brewster. On the 1910 US Census, when I used my great-grandmother and one of my grandfather’s older siblings as search terms, I found Eugene alive.
Generation Four: Great-Great-Grandfather
For many of us, this is part of the family tree the person was likely not living when we were first entering into genealogy. The lack of firsthand knowledge becomes more likely since your grandfather might have been young when his own grandfather died. In my Brewster research, finding Eugene Brewster’s father was a small bump in the road. Based on the lack of an 1890 Census, what may have been easily solved became a larger problem. Each of us have 16 great-great-grandparents, and when I began researching, I knew nothing about any of them besides two general localities. My Slovakian family I could assume was still in Slovakia one generation back from my great-grandparents. Francis and Eugene Brewster were both in Northern New York, in Clinton and neighboring Essex Counties. My assumption started with Eugene being born in the same relatively isolated, rural area he lived in when he was a young married man in 1910.
One of the expectations for y-DNA testing is to trace an unbroken male line to find geographic origins, and one of your main assumptions can be that people tended to be born, grow up, marry, and have children within a relatively small geographic area. Of course, in the United States there has been a westward expansion from colonial times, but it always makes sense to look close to where you know they already have been.
When you combine intensive geographical research with the geographical clues that are available with y-DNA (through a Surname Project, most likely), you have some chances at breaking through brick walls. Still, y-DNA testing alone is probably not enough. According to my results, a descendant of my 2x great-grandfather is not present in the databases (neither FTDNA matches nor in the Brewster Surname Project). What I needed at this point in my research was a little luck, which I found in the form of the 1892 New York State Census (when did your state do censuses?).
I hadn’t been very familiar with the types of documents available in 2014 when I first started looking for Eugene’s father, but I did find that New York had conducted a census in 1892, when Eugene was about 11 years old. He should still be living at home and indeed, I found him in the Beekmantown, NY household of Charles and Mary Brewster. Additionally, on the adjacent page, I found Eugene’s future wife, Sarah “Sadie” Thornton, age 7. The coinciding ideas of time, location and opportunity definitely work here.
Generation Five: 3x great-grandfather
The sweet spot for most atDNA cousin matching is at the 2–4 cousin range, so your 3x great-grandparents are accessible still through cousin-matching. In addition, this distance is a great place for the specialized y-DNA test because confirming or ruling out same-surname matches is possible. These 3rd cousins are men you may not have ever met, and in testing their y-DNA lines, you help fix their locations in your tree. This becomes even more critical when you are trying to rule out men of the same name or surname. In the early 1800s, there were three Daniel Robinsons in Delaware County, New York. One was the ancestor of my paternal grandmother, the other two are apparently unrelated. If one of my Robertson cousins did the y-DNA test, I would expect the possibility that these three lines could be differentiated in the Robinson/Robertson Surname Project, where there are over two dozen groupings based on haplogroups.
In 2014, when I was contemplating testing my y-DNA to “see behind” Charles Brewster, there were not many men I could compare with, and none that had any pertinent identifying information. The Brewster Surname Project’s main impetus for many years was attempting to establish DNA lines from Elder William Brewster of the Mayflower. A second prominent but unrelated line of Brewsters settled in Long Island — the Revolutionary War spy Caleb Brewster descends from that family. I was stuck with a man born in the 1850s, so neither the Long Island family nor William Brewster’s lines would help me because of the 200-year gap. While y-DNA can provide compelling evidence to paternal lineages, the test can only work for you if you can connect somehow with those lines, and at the time, I didn’t see the value of possibly connecting back to the 1600s if I couldn’t find the generations in between.
Had I not discovered Charles Brewster’s father, this story wouldn’t be very interesting. It took me about three years to discover the line, flesh it out, and rule out other compelling possibilities (usually all three taking place at once, over and over). Again, with a pool of many y-DNA testers, this process could have been streamlined, at least the ruling out of the wrong possible candidate part. With a y-67 match, some participants can find matches in this 5–8 generation range. With a y-37 match, a definitive connection cannot be made, but again, within a large pool of the same Surname, a close y-37 match can provide very strong clues.
My search for Charles Brewster’s father was difficult for several reasons. The first was that on the 1870 US Census, there is no Brewster family in Clinton County with a son Charles in his middle teens. The second is that none of the other Brewster families seemed to fit with any other corroborating information, of which I admit there was not much. So, I jumped back to 1860 and 1850, intending to sift through all the Brewster families in the area and see where they were in 1870 and 1880 so that I could rule them out. This process took me the bulk of two years, on and off, and involved looking at dozens of families.
From a historical perspective, Clinton County was essentially wilderness until after the Revolutionary War. The US Census data available from the 1790–1840 Censuses is limited to “head of household” names, usually a man, but sometimes a widow. Knowing that this data was nearly useless for my immediate purpose, I focused my energy on the families in the 1850 US Census. Even had I done a y-DNA test, because I had no idea where my Charles Brewster born about 1853 came from, I would have needed connections with Brewster Surname Participants at this 5th generation with excellent family trees tracing all descendants.
In the 1850 Census, I found the family of Oliver Brewster and his wife Sophia living in Mooers, NY, just north of Beekmantown in Clinton County. Oliver and Sophia had 19 year-old Oliver Jr. living with them as well as a Herman and Mariah Brewster and Jacob and Margaret Brewster — all in their teens and early twenties married and enumerated on the same page. I made the assumption that Jacob and Herman were Oliver’s sons. Oliver and Sophia also had several other children, and between them all, the older children were born in Canada, the middle couple in Vermont, and the youngest in New York. My first task was to find Herman and Mariah on the 1860 Census and I did, with three children: James, Charles, and Jane. Jane was listed there as five years old and Charles as nine months, but otherwise this was a solid possibility. Jacob and Margaret did not have a child named Charles in 1860, and neither did Oliver, Jr. Even with the incorrect age, I forged forward to 1870 where I could find none of the family in Clinton County. Eventually, I found a James and Charles of the correct age living with a Johiel Shutts and his family in a neighboring county. My presumption then became that this is the correct Charles (in the lack of any other choices) and that the boys became orphaned sometime in the past decade. Herman Brewster, born 1824 in Canada, became my suspected 4x great-grandfather.
Generation Six: 4x Great-Grandfather
Luckily, finding records for Oliver Brewster, born about 1800 in Canada was easier than finding records of his son Herman (New York had non-existent record-keeping in nearly all frontier areas until the late 19th-Century). I found Oliver on the 1831 and 1825 Lower Canada (Quebec) Censuses. Again, these early censuses did not contain names of spouses, but there were several other Brewsters in the area, so I felt good about being able to eventually trace further back. I broke through this logjam when I posted an inquiry about Oliver Brewster in a Facebook genealogy group. Someone checked The Brewster Genealogy 1566–1907 book by Emma C. Brewster Jones, a well-documented source, and found Oliver there.
Generation Seven: 5x Great-Grandfather
Oliver was listed as the son of a Lewis Brewster who married Sarah Ray. Lewis left Vermont sometime in the 1790s and by December 31, 1799 the couple had their first child, Oliver Brewster. At this point, I accepted that Oliver might be my correct ancestor or Herman might have been Oliver’s nephew or a close cousin, but in the absence of any other records, I decided to follow Occam’s Razor, that this simplest path was probably correct.
Generation Eight & Nine: 6x & 7x Great-Grandfathers
Eliphaz Brewster was born in Connecticut and moved north into Vermont with his father Charles and other family members in 1763. They were both in the Green Mountain Boys Militia in the Revolutionary War, and Charles served in various official capacities for the nascent State of Vermont before his death in 1790. In four years of searching, I have connected with two other descendants of Eliphaz through Lewis’s brothers, but neither have all-male lines, so I could only possibly match them with atDNA, which I do not.
Generation Ten: 8x Great-Grandfather
The line I found in The Brewster Genealogy ending with Oliver Brewster that traces back through Lewis, Eliphaz, and Charles is a Mayflower line going all the way back to Elder William Brewster who arrived in Plymouth Bay in 1620. The 10th generation back from me, Ebenezer Brewster, is Elder William’s 2x great-grandson. Ebenezer married Elizabeth DeWolf and had four children before he disappeared from the record in about 1729 after the birth of their last child, Elizabeth. Ten years later, in 1739, Elizabeth filed for divorce from Ebenezer, citing abandonment. The mystery of what happened to Ebenezer continued on until the late 1990s when the tradition of an Ebenezer Bruster, Mayflower Descendant, was tested with y-DNA. The two lines matched. Subsequently, Ebenezer’s movement from Connecticut to New York and ultimately Virginia has been documented, along with a second set of children after he left Elizabeth.
In December 2017, when I sent my cheek swab in to FTDNA for a y-67 test, it was my hope to match these descendants of Ebenezer, as this would provide me with genetic proof along with the circumstantial case I had built from New York back through Lower Canada and Vermont to Lebanon, Connecticut.
About a month later, I had results on FTDNA. In the interim, I had joined some online groups dedicated to y-DNA, trying to piece together the various components I expected would be able to support or refute my research. In between, my brother received results for an atDNA test through 23AndMe which provided a very general male haplogroup of I-M253. According to their analysis, it is a relatively common haplogroup, about 1 of 7 of their customers have it, but more importantly, it matched the Brewsters descended from Elder William Brewster. I tried hard to not get my hopes up, especially reading all the material about Non-Paternity Events (NPE), a very clinical term used to describe any situation where the surname and y-DNA do not match. This is common in the case of adoption or remarriage, but it also very bluntly reveals instances where the expected father is not the actual biological father. Looking at the men listed in the Brewster Surname Project, several men match descendants of Elder William but have different surnames. When my results came in, my project administrator assured me that those men I matched were biological Brewsters- one adopted, one whose ancestor changed his name from Brewster, a third who could trace a suspected affair to a Brewster neighbor. More importantly to me, though, was that through the y-67 matches, I definitely matched closest with documented descendants of Ebenezer Brewster and his father William (which would make me a descendant of Elder William’s oldest child Jonathan), as well as several descendants of Elder William’s other son, Love Brewster. Jonathan and Love’s brother Wrestling, the only other Brewster son of that generation, was never married and no documented lines descend from him.
Reliability of y-DNA Results
Happy to have DNA proof to underlie my paper research, I shared my story with several online groups. Most people were happy to hear that it all paid off, but a couple of knowledgeable people pointed out that y-DNA is not as precise as I wanted it to be. I would like to take some time to point out the possible holes in my conclusions and provide a template for others doing y-DNA analysis for genealogical purposes.
My haplogroup- I1, aka I-M253, is old, perhaps 15–20,000 years old, and is suspected to have originated in mainland Europe. This means, in all practical sense, that my paternal line goes back into time immemorial. Certainly, I cannot trace this line to one particular ancestor, especially when many Northern and Western European men belong to that general haplogroup. Fast-forward to me today, I can trace the Brewsters, but the reality is that the certainty for a y-67 connection is less than 100% in most of my cases in generations 1 through 20. Elder William Brewster is 14 generations back from me and his descendant Ebenezer is only 10. This is not to say that my y-DNA doesn’t make me related to this line, but the reality is that, at least in genealogical terms, both Elder William and I may only share a common ancestor like his great-grandfather, making him a many times great-uncle as opposed to a direct ancestor.
This is a common occurrence in many Surname Projects, where the common name stretches back, especially to Ireland or Scotland, and the y-DNA is not able to provide the needed degree of support in the most recent five or ten generations, just when much genealogical research tends to be far less available. The average person might not have as much information about that 8x great-grandfather, and so the y-DNA matches do contain a large degree of uncertainty. Indeed, joining a Surname Project should be the first step of maximizing the usefulness of your y-DNA test, where you are able to communicate with the administrators about the specifics and idiosyncrasies of your personal situation.
Still, my test results coupled with the very specific, targeted testing of Brewster lineages for Mayflower purposes does provide a higher degree of certainly. While there may have been random I-M253 Brewsters in England who were cousins and uncles to William Brewster, none of those people came to America and propagated. The major divisions in the Brewster line occurred in England prior to the 1500s and as those “other” Brewsters (mainly I-M223 and R-M269 haplogroups) did emigrate to North America, their lines are clearly distinguishable in the y-DNA record. The most reasonable, simplest conclusion, that I descend from Ebenezer Brewster, will be the one I consider correct until and unless contradictory evidence materializes.
What Next?
The doubts of the absolute validity of my test results sown by knowledgeable people caused me to seriously examine the whole process of y-DNA testing including its limitations and attainable goals. In my Brewster Surname Project, we have a host of older y-37 tests, some completed nearly 20 years ago which helped establish links between the General Society of Mayflower Descendants members and Brewsters like myself who cannot furnish the required documentation. There are several testers who have the full y-111 tests, including one who descends from my ancestor William, Ebenezer’s father. What I would like to see is more Brewsters tested so that we can round out potential male lines from Jonathan and Love. At this point, there are certainly thousands of candidates, but even the atDNA tests don’t have Brewster surnames in my or my brother’s results. The next test after the y-111 is called the “Big Y” which examines the haplogroup through SNP testing, as opposed to the STR testing of the numbered markers. This Big Y provides a refinement of the various branches, coming down through the generations to specific individuals or families. Some of the Brewsters I match have completed Big Y testing, but not many.
