The Experts Vindicate in Part
"DNA Racial Classifications Refuted!"

A Vindification of the Brit-Am
Case Study: "The Cohen Gene"

In our article
DNA Racial Classifications Refuted!
We gave good reasons based on conventional scientific DNA research as to why the DNA "Racial" Classifications have only limited applicability.

This article of ours was ridiculed and unfairly criticised.
The critics appear appear to have been blinded by prejudice and not to have understood the subject matter.
Letters criticizing the article were published followed by Brit-Am answers to them.
This Correspondence has since been removed, for the time being at least, due to the request of one of those who had written.
It appears that people sometimes write and speak without always intending what they say or realizing its implications.
In place of the letters and answers to them we have below corespondence that in part confirms Brit-Am Conclusions and in part strongly opposes them. This correspondence is somewhat long and technical.Below is a continuation of the clarifications made by James Heald and Sasson Margoliot.

This may be somewhat long and technical for some who if nevertheless are still interested may go to our summary at the bottom of the page.



 and Discovery 

Facing A Revision of DNA Theory

Very Important and Interesting Books - Read More -
Order Now!

 Click Here 

Biblical Truth Joseph Origin
Scandinavian Secrets Role to Rule

 For all Brit-Am Publications 
 Click Here 

The Tribes


Independent Sources Confirm
Brit-Am Understanding in regards to CMH!

These letters deal with the same subject and more or less wrap the subject up for the time being.

James Heald

Date: Mon, 09 Jul 2007 22:42:06 +0100
From: James Heald
Subject: Re: [DNA] CMH and DNA Refuted?


Thanks for posting.

You need to understand the difference between Y-SNP markers ("single nucleotide polymorphisms") and Y-STR markers ("short tandem repeats").

Y-SNP mutations are rare; so that we can usually assume that 2 people with the same Y-SNP mutation both inherited it from the same ancestor and the mutation event. That is why they are sometimes called UEPs ("unique event polymorphisms")

On the other hand, Y-STR changes happen comparatively often, so Y-STR haplotypes can float about all over the place (well, they can over several thousand generations, anyway). In particular two people can end up with the same Y-STR haplotype, even though they come from completely different ancestries. (This is called convergence).

The haplogroups F, G, H, I etc are based on SNP mutations -- so every member of haplogroup J1 decends from a single ancestor, in whom the mutation characteristic of J1 first occurred. But the CMH is based on Y-STR mutations. So 2 people can both end up with the CMH, even though their ancestries are completely different. This is particularly true because the CMH is based on only six Y-STR markers, which makes these kind of coincidences very possible. In contrast modern surname studies often use up to 67 markers, specifically to avoid this kind of accidental confusion.

So there really is no contradiction.

Haplogroup J1 is associated with a big spead of different Y-STR values. So is haplogroup J2. The two spreads overlap, and the CMH is found in that overlap. So either J1 ancestry or J2 ancestry may lead to someone having the CMH values.

(And note also -- it's not just Cohens who have the CMH. Many other groups, completely unrelated, have also ended up with it as their Y-STR signature, quite independently).

Hope that helps,

All best,

James Heald.

Sasson Margaliot

Date: Tue, 10 Jul 2007 01:30:47 +0300
From: Sasson Margaliot Subject: Re: [DNA] CMH and DNA Refuted?

Dear Yair,

Let me try to explain in layman terms what is wrong in your article.
You ask:

" Is the CMH the result of a one-time only mutation (UEP, Unique Event Polymorphism) as conventionally explained? "

The problem is, no DNA expert ever said that the CMH is a Unique Event Polymorphism!

CMH is a combination of 6 STR markers (called "haplotype") and can be derived from literally dozens of different "haplotypes".

You say:

> *What applies to CMH is pertinent to all DNA groups*

Yair, this is simply not true. CMH is defined by a combination of six integer numbers, which could be arrived at in many different ways. The major genetics groups (like R1b, J2, etc) are not like that at all. They are defined by genetic changes at the level of *single nucleotide* .

A person who has all the markers of Haplogroup R simply cannot have a great-grandson in the Haplogroup J, because this would require a large number of reverse SNPs and repeated SNPs in just three generations. The probability of such combination of low probability events is *astronomically * low.


James Heald Clarifies the Issue from the Conventional Points of view:


From: James Heald
Subject: Re: [DNA] James Heald and CMH
"while the CMH can arise separately in different population groups, the same is /not/ true for haplogroups like F,G,H,I or J1 and J2.

...when you write, on the basis of the CMH, that "changes in /haplogroups/ may occur repeatedly at different times" - you do understand that this is not correct?

James Heald wrote:
Secondly, I think your language about the CMH "emerging spontaneously" needs care. Whereas SNP mutations are notable binary ON-OFF changes, the changes in STR haplotypes are very small subtle changes which leave the haplotype almost exactly where it was before. That's why it's more appropriate to talk of a haplotype "drifting" slightly in a lineage of a couple of hundred generations, rather than suddenly coming into existance;

What do you mean by "drifting"? (a) Do you mean that through heredity the bearers of CMH increased?
Or do you mean
(b) That the non-bearers of CMH (who however were close to CMH) had a tendency IN MORE THAN ONe INSTANCE to become CMH?
Would you not agree that according to the scenario you describe this second explanation is at least a possibility?

James Heald wrote:
Yes, (b) is what I mean.

Note that as you go forward through the generations, some lineages may even start close to the CMH, drift away, and then drift back again.

So different lineages (sometimes in different geographical parts of the world) can end up at the CMH independently of each other.

If it is a possibility could it not be that some environmental cause triggered the "drifting"?

James Heald replied:
It's not really down to some identifiable "cause", it's just the accumulation of random errors in the DNA copying.

So you have a group of lineages which start all at the same haplotype, that of the common ancestor; but as the errors build up, the descendent's haplotypes slowly diverge, drifting apart from each other.

..the spread of haplotypes associated with descendents of an individual tends to "spread out", like a cloud. CMH haplotypes don't appear out of nowhere. They represent a very slight change from something very similar. Because the changes are so small, I think most people would say that the CMH /is/ "exclusively hereditity". What it is /not/ is exclusively indicative of a particular ancestry.

Yair Answered:
 I do not see the difference.  I do not believe there is one.  How can it be "exclusively hereditity" (are you making fun of my  spelling mistakes?)  and not "exclusively indicative of a particular  ancestry"?
 It sounds like the same thing.
 If there is a difference,  Please explain carefully.
 "Talk slowly, I am a natural blond".


James continued:
Well, it depends what you mean by "exclusively heredity", when you wrote:

>>> ONLY this time did I get answers admitting that CMH is NOT exclusively heredity. >>>

It was your expression, not mine. I was not entirely clear what you meant. (But sorry for "exclusively hereditity" - my copying mistake, no making fun intended). If you meant, " The CMH is not "exclusively indicative of a particular ancestry", then that is correct. (But you do understand that the same statement made about say Haplogroup J1 would not be correct, don't you?) However, if you had meant that some other process was at work, other than ordinary inheritance, and an occasional small change in a Y-STR marker, then that would be unlikely to be correct.

The historical sequence was probably something like this. The MRCA [Most Recent Common Ancestor] of haplogroup J probably had a (6-marker) haplotype quite close to the CMH.  His descendents, the MRCAs of haplogroups J1 and J2 appear also to have had (6-marker) haplotypes quite close to the CMH. Over time since then, the range of haplotypes of their descendents have spread out, like two clouds; so comparatively few now have the exact CMH haplotype. (I  think your webpage says only about 3%). However, the much more recent ancestor of the J1 Cohens /did/ have the CMH haplotype, and (as that wasn't so long ago), most of his descendents /still/ have a 6-marker haplotype which is very close the CMH.


The figure is about 50% of Cohens with J1 having CMH. So ca half  retained it and half lost it, as Giacomi proposes?


Our current best guess is that those 50% share a recent common ancestor (though without higher-resolution results based on more markers, we can't be absolutely definitively sure of it). Maybe 5% "lost" it -- they still have haplotypes which clearly belong to the same cluster, but are maybe 1 point different when you look at the particular 6 markers used to define the CMH. The other 45% or so never had it. Their lineages were never J1 CMH, and they probably don't have a common ancestor with the J1 CMH Cohens in the last 10 or 20,000 years.

Interestingly there are also some Cohens in J2, and some of these also have a 6-marker haplotype which is close to the 6-marker CMH. (Though at 12-markers their haplotypes do become rather different to those of J1 Cohens).

The J1 Cohens and the J2 Cohens likely haven't shared a common patrilineal ancestor in 20,000 years - as becomes clear when you look at more than 6 markers. But just by chance, they are quite close when you only consider the 6 markers that Dr Skorecki et al first looked at.

The conclusion is that matching on 6 STR markers are not a very sufficient indicator of recent common ancestry. In contrast matching on SNP markers /is/ a strong indicator of shared "deeper time" ancestry. And /not/ matching on particular SNP markers is a strong indicator of /not/ sharing close "deeper time" ancestry.

It seems to me that, with these misconceptions clarified, what the material you cite on your page /isn't/ doing is refuting anything about >>how population Y-DNA evidence is interpreted by people who understand >>the difference between SNPs and STRs.

Rather, what the page underlines as incorrect is a populist misconception that the CMH is a "Cohen gene" that "came into being"


 You are saying more or less (like Giacomi) that CMH was the original haplotype or close to it but over time a good portion of its first bearers  "drifted away" genetically?


Basically, yes. If you go back 20,000 years, long before Aaron (if he ever existed), back to the patrilineal ancestor of the whole of haplogroup J, his 6-marker haplotype was probably close to the CMH-6. As time went on, one of his descendents had a mutation which defined Haplogroup J2. Another, on a different line had a mutation which defined Haplogroup J1. Descendents of these two account for pretty much all the current descendents of the original J man. Pretty much all other of his patrilineal descendent lines either died out or daughtered out. Both the original J2 man and the original J1 man probably had haplotypes which were (at 6 markers anyway) close to that of the original J man. Time passed. The haplotypes of a good proportion of descendents of the original J2 and J1 founders have drifted away genetically. Some may have drifted away and then drifted back. We don't know. But there are some which are still close to what is believed to be the original (at least on the 6 markers which were used to define the CMH; the evidence is they appear to have drifted on other markers instead). It appears that about 50% of Cohens are descended from one such line. A large number of Kurds in Iraq also descend from such a line. But there is no particular reason to believe they both descend from the same line. (Actually, I think those Kurds are J2 rather than J1, so that would definitely be a different line).

Summary: The same Y-STR /haplotype/ can be associated with different lineages, not recently related, particularly if only rather few markers have been tested. On the other hand, a Y-SNP /haplogroup/ is always associated with a single original founder, and a single original mutation. Does that help ? -- James. -


Summary by Yair Davidiy:


Contents by Subject Home
Site Map
Contents in Alphabetical Order
This Site


It is admitted that the CMH is not indicative in all cases of a common ancestry.
The CMH does seem to have a special affinity to the Cohens since both J1 and J2 Cohens tend towards it. James Heald seems to suggest two tendencies:
Both J1 and J2 were close to CMH.
The ancestor of the Cohens was especially close.
Even those who drifted away tended to drift back.
This "drifting" was NOT that of a single cases but of a group.
In other words it was not a "unique event polymorphisms (UEP)" but rather a string of
"multiple event polymorphisms (MEP)".
The MEPs appear to occur several times over according to a pattern.

This in a back-handed kind of way that Brit-Am understanding of mass DNA changes taking effect within groups is admitted or at least implied though not explained.
James Heald in effect attributes the CMH a very great antiquity and thus negates one aspect of the Brit-Am argument on the one hand though strengthening it greatly on the other.

For source material concerning the above polemic go to:
DNA Racial Classifications Refuted!

Main Page

Offerings and Publications

Return to Question and Answer
Table of Contents