This study should begin by my making it clear that my own DNA profile does not match either of the chiefly lines. However unlike one interlocutor this does not leave me with an inferiority complex (leading to an underlying desire to ‘level down’) – so I have no axe to grind here.

Get Involved!! Sign up at familytreedna.com

Despite the cautionary tales related below, I would urge any Grant to get involved. The Grant project has been very successful overall and the data is broadly speaking the data. There is discussion to be had about some of the interpretation of the data, but then participants can formulate their own if they wish to.

Family Tree DNA (FTDNA) has been extremely wise in the way it operates. It offers and carries out the testing and leaves the interpretation to volunteer project coordinators. This frees them from any responsibility for the conclusions the volunteers reach. This makes very good business sense because reliable interpretation would be very expensive and so probably not commercially viable. My own view is that this is not problematic – for any participant is free to seek that advice from any avenue they see fit. However, like any other company in the field, FTDNA makes money by selling tests – so the more tests the more profit.

The philosophy adopted by FTDNA was that if you shared a surname and if your DNA (as judged by 37 STR markers) matched then there was a fair chance that you were related within a historical time frame. That was all – the rest was down to you to get on with the research. As time went on the number of markers offered increased dramatically – offering a lot more “accuracy”, but also facilitating misdirected “group think”…..

Beware Sophomores

When Charles Darwin popularised the Theory of Evolution he was building on existing thinking, particularly advanced by the likes of Charles Lyell who had been arguing that there had been continuous incremental evolutionary change – and that that change was uni-directional. This philosophy (it was not at all grounded in experiment or observation) mired the development of the understanding of evolution so that it is only within living memory that due attention has been paid to catastrophe. Modern science shows 5 planet-wide mass extinction events in Earth’s history – each of them followed by an explosion of genetic diversity. [There have, of course, been many more such events more localised in their range and scope.]

A parallel group-think dogmatism has infected the mostly sophomore amateurs who have voluntarily taken on the task of offering interpretation of DNA results.

Dogma #1: mtDNA

It is a “well known fact” that mitochondrial DNA is passed down exclusively from the mother. There is even a “scientific” explanation for this: it is recognised that sperms themselves do carry mitochondrial DNA (presumably needed to energise their swimming!) but the section of the sperm’s tail where it is located is jettisoned as soon as the head of the sperm meets penetrates the surface of the egg. Errr….. except when it isn’t.

Studies have demonstrated that albeit in rate cases some of the male mtDNA does indeed penetrate the egg and can them supply a varying proportion of the embryo’s mtDNA. More than 50% has been recorded, but normally it would be less.

You can read this up for yourself. I would recommend starting with these:

So although it is a reasonable rule of thumb to suppose that there is no contamination of the transmission of the mother’s mtDNA it is NOT absolute and the intelligent person will remain alive to the possibility just outlined.

Dogma item #2: SNPs

The dogma is well illustrated by a comment in an email to me in early 2016 from Geoff Grant, one of the Clan Grant DNA project administrators:

“Once an SNP occurs, it is passed on without fail to every following male generation.”

Err… except when it is not. wikipedia.org / Haplogroup_R1b says

“in 2016, R-P25 was removed completely as a defining SNP, due to a significant rate of back-mutation”

So hardly was the virtual ink dry on Geoff’s email when the greater DNA community was refuting his claim so hard that they had to change their protocol; clearly, they had known about this for some time – it is just that it was rare enough that they had thought it was not a problem. So we can see that Geoff’s ‘certainty’ was misplaced.

Now this was just one SNP which happened to have been being used as a criterion reference. There are others which are also used but have not suffered this back mutation. Conversely there are other SNPs which are not used as criteria in this grand way, but which nevertheless have a “relatively high rate” of back mutation. So what this tells us is that not all SNPs are the same: some have a greater tendency to recombine than others and we have no option but to suppose that some are more prone to breaking than others – and this enhances the possibility that the same SNP could occur in two separate individuals, and logically this is the more likely to happen when the rest of the Y Chromosome is the more similar.

Nevertheless, Geoff persisted. In July 2021 he wrote to me

The Grant Surname SNP tree is in complete agreement with the work of Alex R. Williamson. If you have a problem with this, email him as he’s a much bigger fish than we are. Be sure to tell him all about these magical rogue SNPs that you’ve found.

The very next day I replied:

Thank you so much for alerting me to the work of Alex R Williamson.
at https://www.ytree.net/ he says
Backmutations or recombination may occasionally reverse SNPs that earlier generations may have carried.
which is precisely what I have been saying for 8 years or more.
Sorted.
So this reference was extremely helpful and supportive.

The following day the founder of the Grant DNA project made this comment:

If that theory is true, then it has the potential to invalidate the results of every other person who has ever been tested.

Well yes and no. What is invalidated is the faux absolute certainty. We can accept that this sort of thing is very rare. But it DOES occur. So, we need a little caution and a little humility and less tunnel vision. And it shows that Geoff was careless with the source he claimed to rely on.

Dogma #3 YDNA

Parallel to the mtDNA discussion above is the “well known fact” that the Y DNA passes from father to son undamaged and unaffected by anything and so only subject to small random incremental changes – copying errors. Just because this is observed to be the case does not make it so, because the danger is that the process of interpretation becomes a self-fulfilling. In other words, at the moment any discrepancy which is more than minimal is currently assumed to imply that there is no genetic connection between the two samples or that the common ancestor lived thousands of years ago, whereas in reality at least some of these cases will be explained best by a “catastrophic” change, the principal trigger for which will be the environment into which the Y chromosome finds itself.

Indeed we know that is true in extreme cases because there are some men who are unable to fertilise some women in circumstances where the man is fully capable of fertilising other women and the woman is easily fertilised by other men. In the case of this mutual infertility we do not know the exact cause (although some women have been “immunised” to try to prevent her body rejecting her husband’s genetic material), but if this can happen in the extreme it is simple logic to suppose that in other cases there are significant, but lesser, effects.

In the project we already have an instance of two brothers being tested whose YDNA profile is NOT identical – yes only a small difference, but indicative of just how subtly mutation can be triggered.

Here I am going out on a limb which I am aware is speculative, but the logic is inescapable: we should not make Lyell’s assumption. Of course most mutation will be incremental and slow, but we must not exclude examples best explained by relatively sudden, substantial change, catastrophic if you will, whereby the Y chromosome undergoes change which is greater than the slow-and-steady merchants will allow.

Dogma #4: Time to Common Ancestor/Divergence Point

We noted above the example of brothers in the Clan Grant project whose YDNA is not a full match. Now let us hypothesise that the difference is the P25 noted above. The fact is that these brothers have a common ancestor less than 100 years ago, but if it is a “well known fact” that the P25 SNP was established (for sake of argument) 4,000 years ago then the assumption of the sort laid down as fact by Geoff is that these two brothers could not possibly have a common ancestor within the last 4,000 years. Now that we understand that SNPs CAN “repair” this dogma must be consigned to the bin. It may be a useful starting point and could remain the best guess in default of any other information or logical case – but it is no more than that.

Case Study #1: Our late Chief Sir James Grant of Grant

Sir James devoted many years of his life to the clan, taking more seriously innovative initiatives than his peers. He was an enthusiastic supporter of the DNA project and was very quick to allow his own profile to be identified. He may have been the first clan chief to do so. This was a boon to the project as it allowed – and continues to allow – participants to establish whether or not they are “of the blood”. He also consented to his sample being subjected to the “big Y” analysis which assisted in discriminating between cadet branches.

However, this has not been entirely without controversy. All the MS Grant histories assert that the original Viking line of chiefs failed to an heiress in the early 1300s and she married an Andrew Stewart who then took the name Grant. The story was somewhat garbled so it took a little while for me to identify who this man was – the eldest (and illegitimate) son of James Stewart the 5th High Steward.

Sir James’ DNA profile was close to that of the Royal Stewarts’ but distant enough for the slow-and-steady merchants to claim that the link was “impossible”. In January 2009 I contacted FTDNA to ask about the probability for the two samples sharing a common ancestor in the mid 1200s. The answer came back at roughly 40% - meaning that in 10 such pairs 4 would have and 6 would not have. The matter is further complicated by there being Grants whose Royal Stewart ancestry is not in any doubt. Into this mix we should add that Andrew Stewart and Maud Grant’s eldest son was known as “Patrick Beg mac Mauld” – ie he was noted for being especially small.

There are several different constructions which could be placed on this:

  1. James Stewart’s mistress was “playing away” so that Andrew was not really James’ son at all. This seems particularly unlikely not only because this is not the way mistresses, especially of someone so important generally behaved, but also it would be truly remarkable that she would choose as her ‘other’ lover someone whose DNA was so close to James’.

  2. The Grants who are recognised as Royal Stewarts were not really Grants at all, but were either Royal Stewarts who took the name Grant (but, if so, why???) or were the products of Grant wives “playing away”. This is extremely unlikely as the mores of the time was that illegitimate children were recognised, accepted and provided for by their fathers (as both Stewart and Grant history amply demonstrates).

  3. The traditional story is correct. Andrew Stewart is who the histories say he was and that Chief Patrick ‘Beg’ was the result of a “catastrophic” event (whose aetiology is yet to be understood) which affected both his Y chromosome (whence the variation) and others (whence his diminutive size). In this scenario the Grants with unequivocally Royal Stewart DNA descend from younger children of Andrew and Maud (her pregnancy with Patrick adjusting her body for future pregnancies).

The Lurg Parallel: Mrs Nora Parry Davies was a great servant of the clan who took charge of family history research for many years. Her close relative and another DNA participant matched fairly closely and both had paper trail backs back to the Grants of Lurg (for which see Fraser’s family tree). The same comparison process done by FTDNA said that the chances were only 60% - but in this case we KNOW that it was correct.

Conclusion: The fact is that the connection claimed in the MS histories is NOT excluded. On the contrary the science is supportive of the claim even if that support is not as unequivocal as we all might wish. I am left in no doubt that the claim is correct. For some 250 years it has been a casualty of the “Good” Sir James Grant attempt to rewrite history.

Case Study #2: The Late Dr Douglas Grant of Tyne and Wear.

The death of Douglas Grant at the young age of 56 from Covid19 came as shock to all – but in retrospect it should not have as in 2018 he had advised me that he had an inherited immune disorder. He is also a huge loss to Grant history research.

When I saw his DNA results, I made contact with him because it was my opinion that he had an albeit weak connection to the chiefly line and had similarities with another result which apparently mapped to Glenlochy in Stratha’an. Names in the Glenlochy Pedigree suggested that this was the Mac Robbie Grants. Douglas told me about his descent from horse breeders – specifically a breed he called the Strathavon horse. He was staggered when he learned of Stratha’an in Strathspey and the direct connection to horse breeding in that area. He set to work with a will and soon established many connections including to John Grant hanged for his part in the Gunpowder Plot and to Shakespeare and the Spensers!

Subsequent testing of the other sample showed that the best fit was with the Grants of Tullochgorm – and this connection is confirmed in some of the MS histories. So the most likely narrative is that Glenlochy had indeed been MacRobbie territory, but that at some stage their main line had failed and a Tullochgorm lad had come in to take up the reins, almost certainly marrying the relevant MacRobbie heiress.

This left Douglas. There was no argument, but there was a stand off – the key issue being a single SNP which other chiefly line descendants carried but he did not. It was on this basis alone which gave Geoff the confidence to argue that Douglas could not possibly be of the chiefly line.

There are two major flaws in this assertion:

  1. As we have seen SNPs can recombine – so on that basis alone the argument falls.

  2. The oldest cadet profile we have is Tullochgorm – from Patrick, Sir Duncan’s illegitimate half-brother. But all the viable scenarios suggest that Douglas’ line broke off before this one. So (a) we have no idea when the SNP was formed (b) there is ample space to accommodate Douglas’ line between Sir John Grant and Ian Roy and (c) there is no problem with the SNP recombining. But it is even worse than that, because it is likely that Sir John settled his brother Alan on his English land (just as he took his nephews to France with him).


Chart showing a suggested lineage based on DNA evidence

Case Study #3 The Clan Allan

In Sir William Fraser’s 1883 “Chiefs of Grant” (see Library) the most egregious hatchet job was to build on the work of “The Good” Sir James in rubbishing the Viking Origin story of the Grants. The second such job was on the Allanach or Clan Allan Grants. Instead, it was suggested that these people had nothing to do with the Grants but adopted the name because it was convenient – thereby dismissing all the MS history. Sadly, Geoff Grant bought into this story. The Clan Allan have a pedigree going back so far and although there is a considerable lacuna behind this they have always claimed descent from Allan a younger brother of Sir Lawrence Grant, the 3rd Chief. So it should come as no surprise and as a wonderful vindication when they turned out to be of Haplogroup R1a1 – just what we should expect of Viking ancestors (just like the MacDonalds), whereas the Stewart/Grant line is standard Western European R1b. There are two matching samples here – one from the acknowledged family of the Chieftains, the other from a member whose ancestors come from the correct place but for whom we do not yet have a paper trail connection to a location on the tree.

Yet again Geoff’s airy dismissal is not at all scientific and is not based on any understanding either.

Case Study # 4: The “Non-Parental Event” (NPE)

One member of the project has a beautifully laid out and very well researched family tree leading back to a cadet branch of the chiefly line. The problem was that the DNA was no match whatsoever for that line. This was not a matter of a few markers here or there, it was a completely different line. More bizarrely still the sample WAS a match for another member of the project.

Looking back through the family tree an ‘obvious’ explanation did spring out: a couple married in the usual way for people of that status (he 21, she presumably c14) but no child was produced for nearly 20 years. Had the mother been of a mind to ‘play away’ then there would be no good reason for such a long time gap. On the contrary the suggestion is one of faithfulness despite the man being away serving in the military for prolonged periods. My interpretation is that they realised that the mother’s biological clock was ticking and so it was mutually agreed that she would try someone else – and with some more research we could probably pin down the “sperm donor” or at least his family.

Given what we have noted above, this is by no means necessarily a case of “the father” being “infertile”. Because the man was faithful, we cannot be sure. All we can say is that the couple was not mutually fertile.

The MacDonalds

Mention was made above to the MacDonalds. I may have been the first Grant to be tested by Family Tree DNA because when I did there was no Grant project and I was very kindly hosted by the MacDonalds who already had their project up and running. However, at that time none of their chieftains had participated and so the project leaders were clinging to the idea as promulgated in MS1467 that they descended from an Irishman called Colla Uais. I did try to advise then that this was wrong but they were adamant. Some considerable time later the Chieftains did participate and the DNA wonks had no option but to climb down. The MacDonalds were indeed Vikings as I had told them.

Beware Rogue Companies

There was one company offering DNA testing based in Scotland which was forced to close because the advice being offered was so inaccurate – these companies make money by attracting participants and all to often the best way of doing this is by telling them something they want to know or something intriguing that will become a talking point.

Some years ago, I was in contact with a Black Caribbean Grant who had been tested by another company. The pedigree was known and was all black apart from the Grant who had impregnated a black girl several generations ago. The company did not know this, of course and when the results came back the family was advised that they were 10% Finnish! The major flaw in this argument was that even if the Grant had been 100% Finnish (which, clearly, he was not) his genetic contribution after even so few generations was less than 10% (it would have been 6.25% max.).

Conclusions

  1. Despite the assertions to the contrary, the DNA results vindicate both the Viking origin narrative of the Grant Chiefs and both the interposition of Andrew Stewart and his likely place in the Royal Stewart line. [Although the Stewarts were not yet Kings of Scots, nevertheless they descend from Romano-British royalty for which see “Scottish Clans….”.]

  2. DNA is still a very young science and, as with almost all science, intelligent students are alive to nuance and very careful about trading in or relying on “absolutes” and “certainties”.

See also: Response to DNA article from Standfast magazine on this site.