Williams or McKenzie?
Reader David Williams asks: “I have tested at family tree dna at 67 markers. I have two exact matches with my surname. Looking at the paper trail I believe our common ancestor was 250 years ago. The question pertains to the other matches. I have 28 matches from distance 3-6. 24 of those matches have the surname of mckenzie, 4 have other scottish surnames and two did not reveal their names. From some of my research I believe it was a name change. With this many mckenzies is it reasonable to assume my paternal line is mckenzie and I should pursue that path of research?”
Hoo-boy. Quite a conundrum you’ve got there, isn’t it? Okay, let’s work this through.
Most Recent Common Ancestor Chart
The single most important thing to focus on is that you have tested a lot of markers and you have exact matches who share your surname. Let’s look at what each of those facts means for you.
First off, you’ve tested 67 markers, and testing that many markers greatly increases the information you have to see how long ago a common ancestor might have lived. If you click on the chart here (to enlarge it), you’ll see that having an exact match in only 12 markers gives you a 95% chance of a common ancestor within 29 generations; having an exact match in 25 markers drops that to 13 generations; 37 markers drops it to seven generations; and 67 markers drops it to six.3
Second, you have exact matches who share your surname. This is important because, in the western European culture your surname suggests you share, we’ve used surnames to identify people for roughly 500 years and for much of that time surnames have been passed from father to son. That’s part of the paper trail evidence that links you to your matches and has to be considered along with the DNA evidence.4
Taking those two things in combination, you have extremely significant matches. As explained by your testing company, Family Tree DNA, a 67-for-67 match within the same surname is evidence of a strong relationship:
A 67/67 match between two men who share a common surname (or variant) means they share a common male ancestor within the genealogical time frame. Their relatedness is extremely close.
All confidence levels are well within the time frame that surnames were adopted in Western Europe with the common ancestor predicted, 50% of the time, in 3 generations or less and with a 90% probability within 5 generations. Very few people achieve this close level of a match.5
Now let’s look at the McKenzie question. Here, you have no exact matches at all and you don’t share a surname. Just as both have to be factored in to see how closely you’re related to your Williams matches, both have to be factored in when considering how distantly you’re related to these folks.
A genetic distance of three to six at 67 markers even with a common surname means any common ancestor would be farther back in time.
• At a genetic distance of three or four, with a common surname, the common ancestor is “probably not extremely recent, but is likely within the range of most well-established surname lineages in Western Europe.”
• At a genetic distance of five or six, you and your matches “may … share a common ancestor within the genealogical time frame. The common ancestor is probably not recent, but may still be within the range of most well-established surname lineages in Western Europe.”6
I highlighted the word “may” in that last description, since it shows so well that, even if these folks were surnamed Williams, you might not be closely enough related to have any realistic chance to find a paper trail.
This is particularly true because of your haplogroup — the overall genetic population group you belong to.7 Your particular haplogroup — R1b — is “the most common haplogroup in European populations.”8 That means there are a lot of people in your same overall genetic population who share only distant ancestors with you.
Then you need to look at the specific markers where you and the McKenzies don’t match. Some mismatches are relatively unimportant, because some markers change more rapidly than others. FTDNA suggests that a mismatch in markers DYS458, DYS459, DYS449, DYS464, DYS576, DYS570, and CDY would be less significant in terms of how far back a common ancestor might be.9 And two of your mismatched markers with the McKenzies are in that group (DYS449 and CDY). But some of your other mismatched markers are not among these fast-changing markers, and that — together with the lack of a common surname — really suggests a more distant relationship than just the marker count alone.10
So is it reasonable to assume your paternal lineage is McKenzie? Nope. Not in any particularly recent time frame and potentially not within the time frame when there are genealogical records at all.
Bottom line for the short term — without any question — the best use of your time is going to be focusing on your paper trail research to see how far you and your Williams-surnamed matches can track back your common ancestry and to see whether you can actually identify a single common ancestor.
But in the long run… should you pursue the McKenzie line at all? You know, in your shoes, I would, even though I’d likely be looking at a common ancestor very far back in time, possibly even before the general adoption of surnames.11 That’s because, if I had to bet, I’d be putting my money on the chances that what you and all these other folks may have in common in a more recent time frame is a clan, rather than a specific ancestor.
Looking at the McKenzie surname group, your Williams matches are not the only folks who don’t have the McKenzie surname. There are five other surnames in the group that you match, and they all appear to have Scottish roots. So in your shoes, I’d want to look not at the McKenzie surname but at the Clan MacKenzie and the possibility of Scottish ancestry. Tulach Ard!12
SOURCES
- ISOGG Wiki (http://www.isogg.org/wiki), “Y chromosome DNA test,” rev. 23 Jul 2011. ↩
- Ibid., “STR markers.” ↩
- Chart from “How Many to Test? 12, 37, 67 Markers?,” FamilyTreeDNA.com (http://www.familytreedna.com : accessed 14 Apr 2012). ↩
- See “If two men share a surname, how should the genetic distance at 67 Y-chromosome STR markers be interpreted?,” FamilyTreeDNA.com (http://www.familytreedna.com : accessed 14 Apr 2012). ↩
- Ibid. ↩
- Ibid. ↩
- ISOGG Wiki (http://www.isogg.org/wiki), “Haplogroup/a>,” rev. 1 Jan 2011. ↩
- L. David Roper, “Y-Chromosome Biallelic Haplogroups,” Genealogy Pages of L. David Roper (http://www.roperld.com/ : accessed 14 Apr 2012). ↩
- “If two men share a surname, how should the genetic distance at 67 Y-chromosome STR markers be interpreted?,” FamilyTreeDNA.com (http://www.familytreedna.com : accessed 14 Apr 2012). ↩
- This analysis comes from comparing the two Williams-surnamed members of the FTDNA McKenzie surname project to the other-surnamed members of that project. ↩
- See generally Kimberly Powell, “Y-DNA Testing: When You Match Other Surnames,” About.com Genealogy (http://genealogy.about.com : accessed 14 Apr 2012). ↩
- The war cry of the Clan MacKenzie. See T.B. Johnston and James A. Robertson, Historical Geography of the Clans of Scotland, 3d edition (Edinburgh : W. & A.K. Johnston, 1899), 13; digital images, Google Books (http://books.google.com : accessed 14 Apr 2012). ↩
Judy, excellent closing point about looking at all the surnames associated with Clan MacKenzie! (I have surnames Morgan and MacRae, which are both associated with Clan MacRae. – Carolina Sandhills roots)
I was surprised by the sheer number of sept names for Clan MacKenzie — and what some of them were. I have loads of McKinney cousins, and it turns out that’s a Clan MacKenzie sept too.
Thank you for the information you provided.The wall I hit began after my dna test. Originally I tested at 37 markers. I matched at 0 distance to the 2 Williams and one Mckenzie, that is when I upgraded to the 67 marker test. I did the test to be a part of the Williams surname project.In it the Williams I match form our own group with no other matches.
I have traced my family name back to 1750 North Carolina.It is here where I believe from my paper trail that I find my common ancestor to my exact matches. Our 5x grandfathers were brothers. The mystery begins from here as I believe our 6x grandfather is not who some say he was, which was Joseph Williams son of John Williams the wealthy welshman. One, we do not match others in the Williams surname project who trace through him.Two upon the death of Joseph Williams, the son of John Williams,He does not mention any of the 6 sons and 2 daughters in his will but mentions other children. At least 6 of my uncles and aunts were still living.
Also I have located alot of family info in talking to the librarian at a library in Georgia where my family moved in the early 1800s. She has told me there is a large amount of info and she is trying to compile it to send to me. She has mentioned a reference to a strong Scottish influence and also to the battle of Culloden. So I agree in may be more to do with the Mackenzie clan than surname. But in e-mailing with one of the Williams I match he provided info from a Mackenzie that matches at distance 3 at 67 markers, which was that some of the Mackenzies were banished to the colonies and changed their names to fit into the Welsh community in Duplin North Carolina plus to avoid the British Army.
How much of this is accurate I do not know at present. But again thank you for your information. I believe you are accurate about the clan angle because in the Mackenzie clan project they have placed me in the core Mackenzie group.
Best of luck to you in nailing this down, Dave! You’ve sure got some interesting family history to work with.
David, my Roberts family also has a tradition of Scots descent (Clan Robertson or Donnachaidh) which was contradicted by an “internet cousin” as being Welsh descent – her common ancestor with me is 5 generations back (in North Carolina, although not Duplin County!) Interesting how these details are kept in some lines, but forgotten in others.
Betsy, I’m also a descendant of Robertsons from North Carolina, and we have conflicting oral traditions: either Welsh or Scots-Irish. Interestingly, our Robertson haplogroup is J2, and our closest YDNA matches (who are not very close) descend from Peter Robeson, a tavern owner in Philadelphia.