Ancestry DNA

I recently had an Ancestry DNA test done. The process is clean, well-designed and private: 10ccs of spittle in a plastic container sent off in a pre-addressed package identifiable to the testers only by a reference number. Once the test is complete – only a matter of weeks – the results are available using the reference number.

Until recently, most genealogical DNA testing focused on the Y chromosome for direct male father-to-son descents, or mitochondrial DNA (mtDNA) for mother-to-daughter descent. In both cases, the tests seek out small inherited mutations. By comparing the results of large reference groups, it is possible to infer roughly when and where each mutation first arose and, by extension, the most recent common ancestor of everyone whose DNA includes that mutation. The rate of change on the Y chromosome is much faster than for mtDNA, and the inheritance from father to son mimics European surname inheritance, so Y DNA testing has long been the most genealogically useful.

ancestrydna-welcomeWhat Ancestry does is different. It is an autosomal DNA test, taking samples from across the entire genome, all 23 chromosomes, rather than just part of a single sub-chromosome. They test 730,525 points, over 7% of the entire genome. Even three years ago, industrial DNA-testing on this scale was barely imaginable, and it makes possible broad-brush comparisons on a completely new scale.

What do those broad-brush comparisons tell you? Ancestry uses the results in two ways. First comes an “ethnicity estimate”,  giving a percentage match to each of 26 ethnic groups. Second is a search across other Ancestry test results to uncover unsuspected 2nd, 3rd and 4th cousins.

So when I got the email saying my test was complete, I was like a child on Christmas morning. Would I be 40% Iberian, 10% Eastern European, 20% Finnish? Third cousins in Hawaii, please, please, please.

No. Apparently, I’m 98% Irish. After trying to choose which of my treacherous Brit fingernails to pull out in order to make the 100%, I took a closer look at the way the results are created. There’s a margin of error of +/- 5%. So I could actually be 103% Irish – Hiberniores Hibernicis ipsis, more Irish than the Irish themselves. Wrap the green flag round me, boys.

pathetic 95-percenters
Pathetic wannabe 95-per-centers

More seriously, that 98% points up the limits of autosomal testing. It is very broad, but also very shallow, its accuracy limited to four generations, great-great-grandparents. This is because you get half your DNA from each parent, who each got half from their parents and  so on, halving with each generation. As a result, the amount you share with any of your great-great-grandparents is just 6.25%.

I already know who all my great-great-grandparents were and where they lived, all along the Galway/Roscommon border, God help them. Not a Finn or a Polynesian among them. So Ancestry was telling me nothing I didn’t know.

Ethnicity is also much slipperier and more dangerous terrain that Ancestry seems to realise. In the US, being “Irish” or “Italian” or “Polish” is an interesting twist on basic American-ness: it might be pistachio or vanilla, but it’s still ice-cream. In Europe, we spent most of the 20th century fighting genocidal wars that revolved around a toxic mix of  ethnicity and nationalism. Not ice-cream at all.

The second part of the Ancestry DNA experience is genuinely useful, though. On Ancestry itself, I found four third-cousin families, all of whom have extended family histories about which I knew nothing.

And when I uploaded my results to the open-source, I found even more, and closer connections.

The Ancestry test costs £99 and is well worth it, especially if your main interest is the broader extended family. Just take the ethnicity side of things with a grain of salt.



Epic Ireland

On Saturday I loaded up on all the scepticism I could muster and headed down for a sneak preview of the Epic Ireland visitor attraction in the CHQ building on Dublin’s Custom House Quay. (Full disclosure: the reason for the invite is that the Family History Centre attached to Epic has licensed some of the software from my site)

Custom-built “visitor attractions” are not generally high on my wish-list: being told what to see, even in velvety PR-speak, gets my hackles up. And after all the fruitless ballyhoo a few years back over a National Diaspora Centre, I was afraid this private-sector version might go for the paddywhackiest of paddywhackery. So I entered the CHQ vaults with a clenched heart and some trepidation.

And two hours later emerged with my heart melted, a lump in my throat and my eyes out on stalks. The place is simply extraordinary. First, and most important, it is honest. The reasons for leaving and the lives left behind, the individual stories, the huge chronological and geographic span of migration from Ireland, are all
presented straight.

The barrel-vaults in the basement go on and on

But the wonderful use of touch-screens, hi-definition projectors, motion-sensors and especially of the barrel-vaults of the building itself make it possible for a visitor to skim or go deep, to linger over the role of the Irish in Bordeaux wine-making or the battle of Fredericksburg, to whip through Riverdance or be hypnotised by the spectacular animations illustrating the history of Irish science.

In the end, it was one of the most moving museum experiences I’ve ever had. I suspect anyone with Irish blood will find it just as emotional.

Quibbles? Of course: There’s not enough about the awkward Other Irish, Northern Presbyterians, responsible for the winning of the American War of Independence, a fact worth bigging up. I found the passport to be stamped as you go from section to section just a tad on the hokey side. The sheer scale can be a bit overwhelming. And there is some mission creep – it covers aspects of contemporary Ireland with only the most tenuous links to the Diaspora.

The passport at least gives a sense of the sheer scale of it all – each of these areas could take half-an-hour or more to explore fully

But, all in all, it is breathtaking.

Epic Ireland opens to the public on Saturday next, May 7th. I’ve been to the current top attraction in Dublin, the Guinness Storehouse, and Epic is much better. If there’s any justice, it will be a runaway success. And so will the Family History Centre.

The cowboy taxonomist rides again

One of the perennial problems I face is responding to the question “What do you do?”

“Genealogist” doesn’t quite cover it (and constitutes a dangerous open invitation to tell me about Granny Murphy). “Software developer” isn’t right either, because all the software I do is connected to Irish genealogy and heritage. Same for “writer”, same for “teacher”.

Poster - Destry Rides Again (1939)_01
That’s me, in the red hat

Twenty years ago, out of sheer badness I occasionally used to answer, “I’m a cowboy taxonomist”.  Taxonomy is the science of pigeon-holing, and at the time I was deeply involved in developing the software that underlies almost all of this website. It entailed dozens of interrelated categories and sub-categories, all ready to store information about genealogical sources.

To my surprise, I found I enjoyed it. As I put it to a poor, puzzled soul I once cornered as a party: “I have no interest in looking at every single gravestone transcript in Co. Tyrone. But I really, really want to know where they all are.”

When I originally collaborated with The Irish Times in putting the results of this cowboy taxonomy online, the plan in the back of my mind was that I’d use that income to keep on expanding and populating those pigeon-holes. And to some extent, that’s what happened. To misquote John Lennon, now I know how many pigeon-holes it takes to fill the Albert Hall.

But many other things got in the way over the years, and the expansion slowed way down.

The collaboration with The Times is now ending, and one of the unexpected side-effects is that the classifying juices have started flowing again.

Already, the way newspaper transcripts is handled has changed completely – you can now compare, county-by-county, what years of what relevant newspapers are online (look at Waterford, for example). General Register Office records likewise, now down to local registrar’s district level.

And I have plans. Map the surnames in Pender’s Census. Rejig the old step-by-step wizard to reflect the revolution in online access. Master the explosion of new gravestone transcript sites. More record-images, more maps –  Irish research ever simpler, ever more transparent.

Which suggests another response to that what-do-you-do question: “OCD genealogist”. Or is that just tautology?

The crocodile got dentures: parting ways with ‘The Irish Times’

After 18 years, the Irish Times/Irish Ancestors sub-site is winding down and transferring here,

Then …

From this morning, the Times is no longer accepting payment for any reports, instead offering a link that will take the user here, to a free version of each report. From next week, some of the legacy services will be redirected, telling users (and Google) that the service has moved permanently. A week-by-week process will increase the number of services redirected until May 23rd, when the Irish Times sub-site will cease to operate.


All the services on this site will remain free until May 11th,  after which a soft pay-wall will ask for a small monthly subscription after five free daily page-views. All users who have paid for Irish Times/Irish Ancestors reports will continue to have full free access here.

Obviously, I have mixed feelings about the change.  From the very outset in 1998, it was clear that a genealogy sub-site was out of place on a newspaper website, especially a genealogy sub-site not fully owned by the newspaper. I always thought of the relationship as symbiotic: I was the little bird that picked the crocodile’s teeth.

Ironically, it’s the success of the internet itself that has finally produced this parting of the ways. While the Times website was the Cinderella of the organisation, hard-working but neglected, my sub-site was welcome. But as the online side came front and centre, the Ancestors bit fell behind,  increasingly clunky and out-of-date.

By last year, it was clear that there were only two options. Either the sub-site got serious attention, with redevelopment and some marketing muscle, or it had to go.

In the end, it’s entirely understandable that, if the choice was between using scarce resources for online journalism or for redeveloping a genealogy sub-site, there could only be one result.

The crocodile got dentures.

You and me and Kim Jong-Un

Most people have very limited horizons when they think about their ancestors. It’s hard to feel a direct personal connection with anyone more remote than a great-grandparent. Eyes glaze over when you try to tell people of earlier generations, and one good reason is that the numbers inflate so rapidly, to the point of disbelief. How can you possibly have almost 33,000 direct ancestors just five centuries back? (The answer, of course, is that you can’t: think cousin marriage. Then think of something else.)

The very man

But when you lift your eyes to the geological timescale things start to get really peculiar. A simple, striking, scientific fact is that every single life-form so far examined shares the same ancestor. You, me, Kim Jong-Un, bacteria, jellyfish, dinosaurs, mushrooms and slime mould all descend from a single, original, living being. It has even acquired its own acronym: LUCA, short for Last Universal Common Ancestor. Current theory posits it as a small, single-cell organism, estimated to have lived some 3.5 to 3.8 billion years ago.

One of the implications is that the beginning of life on earth seems to have been a unique occurrence in the 4.2 billion-year history of the planet. Understandably, this makes many scientists squeamish – such an event is so vanishingly unlikely it begins to look like evidence for some kind of outside intervention, and legions of microbiologists are busy positing alternatives – unfound alien lines, multiple lines that were outcompeted by ours, cross-species sharing of genetic material. But the strongest evidence is still for a single, unique origin, as Darwin put it in The Origin of Species, “some one primordial form, into which life was first breathed”.

Hence my favourite of the many excuses for unsuccessful genealogical research: “We’re all brothers anyway, man.” Or at least 500th cousins.

The normal laws of time and space do not apply to genealogy

How much can we rely on family oral tradition?

The question is unanswerable without legions of ifs and buts. Take the undying belief of large numbers of families in the west of both Scotland and Ireland that one of their ancestors was a survivor of the wreck of the Armada in 1588. As many as 24 ships were indeed wrecked on the Irish coast, but the conditions that greeted anyone who made it ashore were hardly conducive to procreation. The English military force in Ireland, too small to fight off an invasion, compensated with savagery, slaughtering every Spaniard who couldn’t be ransomed.

Historians of the episode emphasise how few survived by pointing out that an envoy of Phillip II sent to Ireland in 1596 to seek out survivors could find only eight individuals.

Streedagh strand in Sligo, site of one of the bloodiest massacres of shipwrecked Spaniards in 1588

But hold on. What about those eight?

If just one of them had two surviving offspring who stayed in Ireland, who in turn each had two offspring – not an unreasonable survival rate – and this rate of reproduction continued in each of the 13 or so generations down to the present, that single survivor would now have 8192 descendants, plenty to provide a basis for a family tradition. Though, of course, the tradition remains impossible to prove.

That’s the difference between genealogy and academic history. Because we focus on individual stories, the strangest statistical flukes crop up again and again. The unlikely Balthazar McGuffin will appear nowhere in the records of the 1870s, where you expect him. But dozens of Balthazar McGuffins will then begin to crop up in the records of freed slaves, or medieval guild rolls, or at the court of Catherine the Great. “Black swan” events like these give historians attacks of the vapours, and very understandably.

As for quantum physics, so for history: at the smallest level, the normal laws of time and space do not apply.

Why are there no genealogical records in the Irish language?

There are two sure-fire ways to make someone brought up in Ireland squirm. The first – especially effective if you’re North American and wearing tartan trousers – is to ask about “the leprechauns”. Watch them furtively check out the nearest exits.

Aidan Doyle’s excellent ‘History of the Irish Language’

The second is to ask if they speak Irish. Almost everyone in the country has undergone fourteen or fifteen years of daily lessons in the language and almost everyone can just about come out with a few fragments of token pidgin (the cúpla focail).

The result is a profound, squirming ambivalence about the language. On the one hand, it’s dinned into us from an early age that Irish is “part of who we are”, an ancient tongue uniquely evolved to embody the Irish national character, the official language of the state. On the other hand, our real mother tongue is English, and has been the mother tongue of our ancestors back more than six generations. And the Irish we were taught in school was not a language like French or German, a medium of communication. It was a form of advanced obedience training, a series of phenomenally complicated grammatical hoops to be jumped through.

Aidan Doyle’s  A History of the Irish Language (Oxford, 2015) sheds much light on how this state of affairs came about.  The author, a lecturer in Irish and linguistics in UCC, approaches Irish from the point of view of historical linguistics rather than with the customary deference of Irish-language revivalism. He concludes that as early as the mid-eighteenth century the language was doomed as an evolving, living, European vernacular. The annihilation of the Irish-speaking propertied classes in the wars of the previous century meant it had no chance to develop “print-capitalism”: no newspapers, pamphlets, contracts, novels, advertisements. It had been made the language of the dispossessed, spoken only, beautiful and utterly powerless.

But Doyle’s most extraordinary finding is that the language as it now exists was virtually re-invented in the late 19th century. The revival movement started and led by Douglas Hyde was an Irish version of the contemporary Europe-wide mania for nationhood, combining crude ideas about evolution with near-racist notions of ethnic purity.

It was a very strong mixture. Here is Hyde writing about his relationship with Irish in 1886: “…the language I spoke from my cradle, the language my father and grandfather and all my ancestors in an unbroken line leading up into the remote twilight of antiquity have spoken … “.

Douglas Hyde c. 1885

Hyde’s family came to Ireland in the Elizabethan era and were granted land in Cork in return for services to England. He learned Irish in his teens, and was almost certainly the first member of his family ever to speak it. The power of his desire for a Gaelic Irish identity to match Magyar Hungarian and Bohemian Czech must have been almost overwhelming. It was certainly blinding.

The language became a potent symbol of national identity, more important than any linguistic facts. And that is what led to its reinvention. For the enthusiasts of Hyde’s Gaelic League, the actual language still spoken in parts of the remote west and south was impure, debased by generations of contact with English. So they purified it. The end result was a small group of urban Irish-language experts whose mother-tongue was English teaching the Irish language to larger groups of students whose mother-tongue was also English. The inevitable result that English-language idioms, grammar and syntax seeped into the “revived” tongue.

Note the discreet non-verbal lumpeen on the left

The results are still with us. My own favourite example is the Irish-language sign in my local park urging dog-owners whose pets foul the grass to “Glan suas é”, “Clean it up”, an utterly idiomatic English phrasal verb translated word by word.  Imagine a sign in French that says “Nettoyez-le en haut.”

So the answer to the question “Why are there no genealogical records in Irish?” is that there are no written records of any description in Irish before the twentieth century.

Is mór an trua é.


How good are the new Ancestry/FindMyPast Catholic transcripts?

That hybrid “Ancestry/FindMyPast” might be unfamiliar to some. After all, these are the Coke and Pepsi of online commercial genealogy, fierce capitalists supposedly competing for every advantage. But the 10 million or so transcripts separately released on March 1st last  on and are indeed two copies of the same recordset.

The Irish Times, March 2nd. I think Ancestry won.

Some time last July, after the National Library of Ireland made its Catholic parish register microfilms freely available (at, Ancestry looked into FindMyPast’s eyes and FindMyPast looked back at Ancestry and they each said, “You’re going to transcribe these, aren’t you?” Very sensibly, they agreed a joint project and split the cost. And then, as the release date approached, did their best to get the jump on each other, with FMP announcing the release well in advance and ancestry hogging the TV and newsprint publicity on the day itself.

I think ancestry won, even though they had RTE and The Irish Times announce the records on the non-existent “”.

At the time I was looking for a family called “Lonergan” in Mitchelstown baptismal records between 1825 and 1845. There are good transcripts for Mitchelstown on, courtesy of the Mallow Heritage Centre, so I extracted all their Lonergans (and variants) over those years, 46 records in total.

MItchelstown baptisms 1833, from the microfilm. Maybe one reason the new transcripts are flawed?

I have subscriptions to both ancestry and FindMyPast and couldn’t resist the chance to compare what they would find. The results were hair-raising. Using the broadest surname variant options on each, the same search produced 24 records on FindMyPast and just 16 on ancestry. (You can download an Excel file with the results here).

The difference can’t all be down to transcription errors and omissions. Much of the gap between the 46, 24 and 16 appears to be caused by the three sites’ differing surname variants systems: apparently ancestry doesn’t consider “Londergan” a variant of Lonergan.

But there is no doubt that Mallow picked up many entries that the new transcription missed. Those old non-image-linked transcripts were made from the original registers, not the sometimes-dodgy microfilms. Rootsireland, all is forgiven.

Of course, no definitive verdict is possible on the basis of such a tiny sample. But the moral is clear: if you’re using any transcript, continue to sup with a long spoon. And maybe you should add a few inches to the handle for the new transcripts.

Irish Heraldic Bling

It’s not just me, or the time of year that’s in it. There really are many, many more shops selling heraldic key-rings in Ireland than anywhere else. Why should this be?

The arms of Henry Ussher (1550-1613), Archbishop of Armagh

There are two sorts of heraldry, the genuine stuff, Norman, feudal, replete with sables, bezants and stags rampant, where a coat of arms is personal property transmitted down the generations through the eldest male child, just like any other feudal possession.


Then there’s “souvenir” heraldry, where everyone of the same (or remotely similar) surname is apparently entitled to bear identical heraldic tea-towels for all eternity. Elsewhere in the world, the distinction between the two is clear. In Ireland, the border is a bit hazier.

A key-ring to be borne in perpetuity by anyone called McDonnell who coughs up €3.99

The main reason is the way heraldry has been regulated in this country. Up to 1922, Ulster King of Arms, based in Dublin Castle, regulated the bearing of coats of arms in Ireland. After independence, his Office languished in post-colonial limbo until 1943, when it became the Genealogical Office, and Ulster was replaced by the Chief Herald of Ireland. The first Chief Herald of Ireland was Dr. Edward MacLysaght, a staunch republican given charge of the most intensely Anglo of all Anglo-Irish institutions. With understandable zeal, he set about creating Gaelic Irish versions of some of the Anglo traditions. One result was the debacle of the Chiefs of the Name of a few years ago. Another was the concept of “sept arms”, where every member of the same (rather loosely-defined) kin-group is entitled to bear the same arms.

Although the impulse to claim equal status for the Gaels was understandable, and perhaps necessary, the truth is that heraldry just did not exist in Gaelic Irish culture. The distinction between arms as personal property and as group or surname property was almost erased in MacLysaght’s classic Irish Families. Hence the proliferation of heraldic coasters.

Cats and Genealogy

My cat has no need of genealogy.


She has a sense of the past, for sure. Every time she sits on my lap, purrs and kneads my thigh I’m reminded of the explanation for her behaviour that I read a few years ago (and which now, despite trying, I can’t forget). Apparently, the kneading derives from her memories of massaging her mother’s teats to express milk when she was a kitten. However queasy the thought, it points to some dim connection between past security and present comfort hard-wired into her tiny brain.

I think a similar hard-wired sense of the past is widespread in nature, or at least in mammals – it’s hard to imagine a shark feeling homesick. That instinctive use of memory to shape the present is the impulse that lies at the root of our own need for history. In pre-literate societies, the intricate stories that were passed on and elaborated from generation to generation provided explanations of ancestry, making the present more intelligible by colouring it with the glow of the past. The durability and sophistication of these stories already took us a long way from instinct.

Language is the medium that made possible that accretion of social memory spanning multiple generations. And written language is what allows social memory to become truly accumulative, with each new generation standing on the shoulders of its predecessor, learning from its failures, expanding its discoveries.

Yet another reason to know our ancestors better, and yet another reason to ensure that their records are well-preserved and widely available. And one more reason why the cat is on my lap and not vice versa.

She seems perfectly content to do without accumulating social memories. Her memory of her past may be dim and purely instinctual, but she seems to get a lot of comfort from it.

Or perhaps she’s just tenderising a large prospective dinner.