On Wednesday, August 30, 2017 at 1:05:28 AM UTC-7, Andrew Lancaster wrote: > On Tuesday, August 29, 2017 at 10:42:42 PM UTC+2, wjhonson wrote: > > > My point is that Y is overused to the point of *inventing* details about ancestors whom you do not really share at all. > > Every type of genealogical evidence gets used wrongly sometimes. But I am not sure what type of examples you are describing here as being typical of Y DNA studies. I have mentioned a couple of Y DNA problems already, that I think many people would recognize, but they do not seem to match what you are seeing: 1. false assumptions about ethnic origins (Roman DNA and all that) and 2. over-reliance on bad/rough dating estimations for last common ancestor. In my opinion both derive partly from marketing strategies. I'm speaking of customers, not scientists. Customers testing their own Y and that of a purported sixth cousin, finding a 3 marker mismatch and saying "that's good enough" It's a house of cards built on marshmellows. If you have *not* built up your tree using close relatives and Autosomal DNA, than that 3 marker mismatch might be fifty years ago, or five hundred years ago, there is no way to tell.

On Wednesday, August 30, 2017 at 5:14:36 PM UTC+2, wjhonson wrote: > I'm speaking of customers, not scientists. > Customers testing their own Y and that of a purported sixth cousin, finding a 3 marker mismatch and saying "that's good enough" > > It's a house of cards built on marshmellows. > > If you have *not* built up your tree using close relatives and Autosomal DNA, than that 3 marker mismatch might be fifty years ago, or five hundred years ago, there is no way to tell. But the value of knowing you are in the same long term male line is NOT nothing, surely? (Often a big line can be broken down to much more recent branches because of unusual markers of course, but let's put that aside, and assume worst case for your argument: we only see that a bunch of Lancasters or Johnsons are in the same old male line.) It partly depends on the paper trail itself of course. For example if you have confidently identified one branch using a strong paper trail (let's say that for example one branch are in Australia and the other are in Scotland, and you know the immigrant) then you have one date after which the common ancestor is unlikely to have lived between that branch and others. You have to go through looking for things like that. Most or all genealogy involves trying to build up the probabilities, but knowing you will never reach 100%. Every increase in confidence is worth something.

On Wednesday, August 30, 2017 at 9:25:56 AM UTC-7, Andrew Lancaster wrote: > On Wednesday, August 30, 2017 at 5:14:36 PM UTC+2, wjhonson wrote: > > > I'm speaking of customers, not scientists. > > Customers testing their own Y and that of a purported sixth cousin, finding a 3 marker mismatch and saying "that's good enough" > > > > It's a house of cards built on marshmellows. > > > > If you have *not* built up your tree using close relatives and Autosomal DNA, than that 3 marker mismatch might be fifty years ago, or five hundred years ago, there is no way to tell. > > But the value of knowing you are in the same long term male line is NOT nothing, surely? (Often a big line can be broken down to much more recent branches because of unusual markers of course, but let's put that aside, and assume worst case for your argument: we only see that a bunch of Lancasters or Johnsons are in the same old male line.) > > It partly depends on the paper trail itself of course. > > For example if you have confidently identified one branch using a strong paper trail (let's say that for example one branch are in Australia and the other are in Scotland, and you know the immigrant) then you have one date after which the common ancestor is unlikely to have lived between that branch and others. You have to go through looking for things like that. > > Most or all genealogy involves trying to build up the probabilities, but knowing you will never reach 100%. Every increase in confidence is worth something. Like I said, relying on Y matches *with* a "strong paper trail" is the quest of fools and old sponges. If you have not used Autosomal DNA to even show that you are related to your own "relatives", than you need to start from scratch.