I'm sorry, Jack, but I disagree with you on a number of counts
pertaining to both my understanding of autosomal DNA as well as the
ability to track past a person's great-grandparent. However, this is not
the platform for such a discussion. I do not know your scientific
background, but I know mine.
I will not be discussing this with you any further except to say that
in particular is not a company that I would recommend from
a scientific point of view.
On 12/02/2019 7:22 am, Jack Fallin wrote:
I can easily accept Carl’s emendation that his “Pseudo-Science”
reference, admittedly a bit stretched, was addressed only to the national origin part of
the typical autosomal DNA sales pitch. On that we can agree, the advertising severely
exaggerates the accuracy obtainable with those results, for reasons mentioned in my note.
Oddly, the one area where I believe the results to be most reliable are those that may be
the most controversial, e.g. sub-Saharan African [versus generally European] and
Native-American [sometimes paired with Siberian] - in those instances the differences are
so marked that it’s possible to have a higher degree of confidence in them. However,
claims that the results can be fine-lined to individual “countries” in Africa [just like
claims to be able to differentiate, say, Holland and Hollands-speaking Belgium] are highly
Ellish, I’m afraid you’ve completely missed the difference between using autosomal DNA
for relation-matching and using it for identifying national origin. Of course, people can
fib about national-origin, all they may be going on is rumors passed down orally in
garbled form from relatives who may have preferred to be from France [think about during
WW I] than from Germany. The stuff you are talking about, the inability to change our
actual DNA make-up relates to the entirely different process of matching DNA across
related individuals. As I think both Carl and I can agree - that process is generally
objective and yields results that, perhaps in rare cases, can be extremely useful in
The real “genealogical” problem with autosomal DNA [versus, say, Y-DNA analysis] is that
it typically gives you a vast number of people who are technically (and accurately)
related to you - but their surnames are completely unrecognizable. That’s a byproduct of
our historic, patrilineal, naming practices. You will typically know your mother’s and
grandmother’s maiden names, but after your great-grandmother, maiden names increasingly
become a mystery. More importantly, every sibling of every one of your direct male and
female ancestors will show as related to you, but that will involve a myriad of sisters
whose maiden names you will not know - and all of their children! That surname problem
shows up again, more directly, with mitochondrial DNA analysis where the mother to mother
connection is inexorable but you inevitable run out of recognizable maiden names.
In the end, the general identity of male surnames [barring name-change, adoption or
illegitimacy] is one of the primary reasons why Y-DNA analysis yields the simplest and
most verifiable genealogical results, albeit of only one limited path of relatedness.
> On Feb 11, 2019, at 1:02 AM, fermanagh-gold-request(a)rootsweb.com wrote:
> Send FERMANAGH-GOLD mailing list submissions to
> To subscribe via email send a message with subject subscribe and body
> subscribe to fermanagh-gold-request(a)rootsweb.com
> To unsubscribe via email send a message with subject unsubscribe and
> body unsubscribe to fermanagh-gold-request(a)rootsweb.com
> You can reach the person managing the list at
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of FERMANAGH-GOLD digest..."
> Today's Topics:
> 1. Re: DNA Analysis: "Pseudo-Science"? (Eilish Gmail)
> Date: Mon, 11 Feb 2019 19:43:30 +1100
> From: Eilish Gmail <eilish7(a)gmail.com>
> Subject: FERMANAGH-GOLD -Re: DNA Analysis: "Pseudo-Science"?
> To: fermanagh-gold(a)rootsweb.com
> Message-ID: <c73c9045-19a1-f55e-4f51-cf37d1ee1abd(a)gmail.com>
> Content-Type: text/plain; charset="utf-8"; format=flowed
> Ha! Carl Definitely (B).
> No snakes in my family tree (and surprisingly no snake oil salesmen).
> Makes the book I mentioned a few months back on genetics where numerous
> geneticists stated that we basically all have the same roots based on
> our DNA seem much more plausible.
> And if it's wrong and not repeatable under differing laboratory
> conditions, it basically is pseudo science. Science is based on
> repeatability and consistency between differing
> experimenters/laboratories and it's clear that none of the parties
> concerned are interested in the accuracy/precsion of their results,
> because it would be easy for them to consult each other. That's how
> science works. It's purely money making and their science roots have
> gone to the wind.
> Eilish, who actually is a scientist and frustrated by the lack of ethics
> in this particular industry.
> On 11/02/2019 6:12 pm, Carl wrote:
>> Well Jack and I obviously got different impressions from the CBC program.
>> Just to clarify, I got some valuable stuff from FTDNA's Ydna and
>> Family Finder, and I'm quite happy with that part of the testing.
>> I still think that national or geographic origin is your basic
>> marketing gimmick, separating you from your money. I guess "pseudo
>> science" was probably a little harsh, but really, you don't go
>> presenting something to your family as fact when its so full of
>> demonstrable errors.
>> Take comfort that, depending upon your belief, we ALL originated a.)
>> In an obscure garden where Eve sat fuming while Adam and the serpent
>> surfed on an Imac, or b.) in the Great Rift Valley or c.) Somewhere Else.
>> Love to all, Carl
>> On 2019-02-10 8:19 p.m., Laura May wrote:
>>> (Please forgive me if I am missing some context here—I appear to be
>>> among those that only receive some messages.)
>>> I have done a DNA test and I am a fan! In fact, we had been searching
>>> for over 30 years for where in “Ireland” our immigrant ancestor
>>> Robert Thompson came from before DNA pointed us to our Thompson
>>> cousins from Fermanagh.
>>> (If anyone would like to compare results with me for matches I would
>>> be interested.)
>>> I wanted to say that I think the issue Jack is mentioning is only
>>> with the samples and analysis that the companies are using as their
>>> comparative base for ethnic origins. So they are relying on samples
>>> they believe to be from a particular region to compare new tests
>>> against, but “knowing” the origins of DNA in terms of
>>> geographic/ethnic location is riddled with the potential for errors,
>>> as we have seen from the twins news and our anecdotal evidence. There
>>> is lots of room for improvement on this front and I personally would
>>> not recommend a DNA test to determine your ethnic background to
>>> anything more than a very general idea.
>>> The cousin matching is much, much more exact because, as you say
>>> Eilish, you are comparing one strand of DNA directly to another. This
>>> aspect of the testing is very reliable.
>>>> Il giorno 10 feb 2019, alle ore 21:28, Eilish Gmail
>>>> <eilish7(a)gmail.com> ha scritto:
>>>> "Fibbers" do not come into DNA testing. DNA "strands"
>>>> with other "strands". The longer the "strand" that is
>>>> higher the change that the persons concerned are related to some
>>>> degree. It's not a lie detector test, and nor will the
>>>> weeded out over time. When I did my own ill-fated DNA test, I was
>>>> not prompted as to what I thought was in the mix. Only the sample as
>>>> required, certainly not an attached family tree.
>>>> And if it's "sloppy" for identical twins, Lord help the
rest of us.
>>>>> On 11/02/2019 1:09 pm, Jack Fallin wrote:
>>>>> Well, I went to the program described and it said absolutely
>>>>> nothing about DNA analysis being “pseudo-Science.” This is a
>>>>> profoundly misleading description of what CBC was doing.
>>>>> It’s plain common sense that when a testing company tries to figure
>>>>> out “national” roots they are comparing the material you submit to
>>>>> that submitted by others and those others “self-identify” where
>>>>> their ancestors came from. Simple enough - if those people are
>>>>> fibbing the results are going to be skewed. Over time, as enough
>>>>> people submit material the outliers (fibbers) will be weeded out
>>>>> because their results are not lining up with the folks whose
>>>>> ancestors really did come from those places. This also means that
>>>>> each testing company will come up with slightly different results
>>>>> because they will be starting from a different set of informants
>>>>> who were related to individual locations at different points in
>>>>> time. As for why the twins hit slightly different “national
>>>>> origins” even within the same testing company, that, as the story
>>>>> points out, will vary with the large (but nonetheless limited)
>>>>> number of “sites” sampled, the “algorithm” employed to try to
>>>>> identify distinctive variations, and the precise mix of alternative
>>>>> samples at the specific date and time tested.
>>>>> “National origin” results will always be inherently sloppier than
>>>>> than actual DNA-relatedness tests [where the order of accuracy runs
>>>>> Y-DNA, Mitochondrial DNA, autosomal DNA] and those tests of
>>>>> relatedness are the real reason for genealogical DNA testing.
>>>>> The companies involved absolutely Did Not miss the fact that these
>>>>> people were twins - the only result that would justify the
>>>>> “pseudo-Science” label; the only result quoted easily found the
>>>>> two to be more than 97% related. So if you are looking for
>>>>> relatives (as opposed to what obscure country you may be attached
>>>>> to) the CBC results, if anything, verified the tests’ ability to
>>>>> identify matches.
>>>>> Jack Fallin
>>>>> Walnut Creek, CA
>> Email preferences: http://bit.ly/rootswebpref
>> Privacy Statement: https://ancstry.me/2JWBOdY
Terms and Conditions:
>> Rootsweb Blog: http://rootsweb.blog
>> RootsWeb is funded and supported by Ancestry.com
and our loyal
>> RootsWeb community
> End of FERMANAGH-GOLD Digest, Vol 15, Issue 44
Email preferences: http://bit.ly/rootswebpref
Privacy Statement: https://ancstry.me/2JWBOdY
Terms and Conditions:
Rootsweb Blog: http://rootsweb.blog
RootsWeb is funded and supported by Ancestry.com
and our loyal RootsWeb community