Ancient DNA: Who We Are and How We Got Here

David Reich’s new book Who We Are and How We Got Here (OUP March 2018) is a revelation. Reich is a professor in the department of genetics at the Harvard Medical School, and runs a specialised lab that focuses on ancient DNA as a tool to study ancient populations and human evolution. His book is a readable popular science account of the extraordinary advances in the understanding of human prehistory that have taken place in just the last half-decade, and the full-scale revolution that is about to take place.

The techniques he describes are remarkable, involving what Elizabeth Kolbert has described as “reassembling a Manhattan telephone book from pages that have been put through a shredder, mixed with yesterday’s trash and left to rot in a landfill”. This is a long way from a simple cheek swab. Try disentangling 40,000 years of bacterial and fungal rot from a finger bone.

But Reich and his peers have mastered those techniques. The number of labs producing whole-genome ancient human DNA results is already growing rapidly and is set to grow exponentially over the next decade. The best analogy is one Reich makes himself, with radiocarbon dating. The ability it provided to accurately date any piece of organic material had a profound and continuing impact on archaeology, to the point where there are now more than 100 specialised radiocarbon dating labs in the world. As the number and location of ancient DNA results expands in a similar way, the prehistory of human population groups will come into focus ever more clearly.

Already, some of the findings he presents are startling. Western Europeans and Native North Americans share significant stretches of DNA, showing a relationship predating Columbus by thousands of years. Reich demonstrates unambiguously how both descend from the same ancestral population, dubbed “Ancient North Eurasians”, some of whom migrated across the Bering land bridge around 15,000 years ago, while others moved westward to contribute to European ancestry.

Even earlier, he uncovers clear evidence of our mixing with other variants of human, Neanderthal, Denisovan and undoubtedly many others, after we emerged from Africa around 70,000 years ago. All present-day non-Africans carry some of their genes. Evidently, we interbred with and exterminated them on our way to planetary dominance.

A bottle-neck Y-chromosome event sounds like a euphemism for a bar-fight. Reich describes these as recurring mass reproductions, where a single male and his offspring have vastly more descendants than their contemporaries. Examples in historic times would include the large numbers descended from an individual in fifth century Ireland, speculated to be Niall of the Nine Hostages, or the vast numbers across the thirteenth-century Mongol Empire bearing genes derived from another individual, speculated to have been Genghis Khan. These “Star-clusters” as they are known, happen again and again in prehistory. Though the mighty warlords’ memories have vanished, bits of their Y-chromosomes live on.

One aspect of Reich’s work rightly worries him. “Race” has been a sore topic in science for decades, and for good reason. Scientists have been enthusiastic enablers of some of the worst atrocities carried out in the name of racial purity. A denial of the significance of race or anything resembling is now obligatory in the social sciences.

The good news is that Reich buries the notion of racial purity six feet under and dances on its grave. Every ancient genome shows intense and repeated mixing of populations. “Whites”, for example. descend from a mixture of four ancient populations that lived 10,000 years ago, each as different from one another as Europeans and East Asians today. We are all mutts, to the  Nth  degree.

The bad news is that as DNA testing becomes ubiquitous it will simply no longer be possible to ignore average genetic differences between distinct populations. Jews have won over 200 Nobel Prizes, more than 20% of the total, despite making up less than 2% of the world’s population. Athletes of West African ancestry hold 95% of the top times in sprinting. Roll a barrel of beer into a roomful of Irishmen and 98% are guaranteed to start singing.

This is nothing like racism, any more than saying that men and women are different is sexism. People are different, groups of people are different and the important question is what we choose to do about that difference. Celebrate (and sing about) it, maybe?

Where’s me granny?

I recently filmed a segment  for an upcoming episode of “Who Do You Think You are?”  – don’t ask, can’t tell – and found the old itch acting up. Years back, I did two series of “The Genealogy Roadshow” on RTÉ and that same old urge is still there to jump up and down and shout, “Look at me, Ma! Look at me!”

“Look at me, Ma!” Out-take from The Genealogy Roadshow

Thinking back, watching the finished shows was very different indeed to making them. Stories that had been just problems to be solved or lines to be remembered during filming became intensely touching when the camera showed the depth of the feelings produced in the participants. I’m thinking of the astonishment and joy of the American family meeting a completely new branch of their family in Ireland, of the woman seeing a photograph of her grandfather for the first time and recognizing her own face in his, of the family finally imagining in dramatic detail how their grand-uncle fought and died in the First World War.

The real lesson of the series was one already known to anyone who has done any genealogical research, a lesson not treated with enough respect by shows that depend on celebrities to hold the viewers’ interest. There is an endless variety and a recurring fascination in the family stories that stretch back behind absolutely everyone, however humble. To retrieve and reconstruct these stories can evoke the dense skein of everyday history as if it were our own experience and let us feel its detail in ways that no other form of research can match.

On a personal note, though, I remember being a little disappointed. Years ago, children in Dublin used to annoy cyclists by shouting out helpfully as we pedalled past, “Hey Mister! Yer back wheel’s going round!” I had been hoping to hear some of them in the street shouting after me, “Hey Mister! Where’s me granny?” No such luck.

FindMyPast smuggles out birth and marriage records

If you do any research on Irish records you soon become aware of the great mosaic of online transcripts of what purport to be the same records. The National Library microfilms of Cork and Ross Catholic registers, for example, are transcribed on, and But IrishGenealogy didn’t transcribe the records of St Mary’s (North Cathedral) in Cork city, because they were doing their level best not to tread on the toes of any existing transcript-holders and the church itself has a transcript. Ancestry and FindMyPast need not be so bleeding-heart sensitive: the only online transcript of this huge collection of Cork city records is with them.

IrishGenealogy treads on the toe of Abdul Abulbul Amir

Similar part-overlapping record collections are also common in the state records of births, marriages and deaths. FamilySearch has a copy of the full all-island indexes up to 1922 and the Republic up to 1958, completely different, naturally, to the freshly-created indexes (to 1916, 1941 and 1966) on IrishGenealogy. FamilySearch also has a part-transcript of the first seventeen years of birth registrations, up to 1881, making it possible to search on the mother’s maiden name. Which you can only do on IrishGenealogy from 1900. But the register images on IrishGenealogy make it possible to trawl by hand through all your search results, only not before 1870 for marriages or 1878 for deaths. Clear?

And then, of course, has full transcripts of some of the local registrars’ records, different to the central copies used for IrishGenealogy and FamilySearch. And the Northern Ireland GRO has complete searchable transcripts of the local registrars’ records for areas now in Northern Ireland (to 1917, 1942 and 1967). Which overlap both some of rootsireland’s and some of IrishGenealogy’s transcripts. Whew.

Now FindMyPast has added to the merriment with a massive set of transcripts of IrishGenealogy’s register images for births and marriages.

They’ve transcribed the entire record. Any chance of a townland search?

There are flaws – Claire Santry recently pointed them out – but I still think they are seriously to be welcomed. Apart from adding another transcription (with mistakes and omissions, to be sure, but different ones), they open up the birth records before 1900, transcribe the entire record, not just personal names, and make it possible to confine a search to a local registrar’s district, a much more precise area than the Superintendent Registrars Districts available on IrishGenealogy itself. A boon if your Muphys married Murphys.

Most importantly, the service is free to anyone who registers with them.  This is one aspect of FindMyPast that isn’t nearly well enough known. Lots of their records are free, in particular the parish register transcripts and records digitised in collaboration with the National Archives of Ireland. As far as I can tell, nowhere on the site lists what’s free and what’s subscription-only.

I suspect the FMP researchers are trying to sneak free stuff out past the FMP lawyers and moneymen. So thank you, Brian and Fiona.

Irish surnames as historical evidence

Over the years, Irish surnames have received a good deal of careful attention, from Fr Patrick Woulfe’s Sloinnte Gaedheal is Gall (1923) to Edward MacLysaght’s Surnames of Ireland (1969) and most recently Seán de Bhulbh’s Sloinnte na hÉireann: Irish Surnames (1997). Ulster names have been particularly well served. Robert Bell’s Book of Ulster Surnames (1997) and Brian Mitchell’s The Surnames of North West Ireland (2010) both dig deeper than an all-Ireland approach allows.

All of them work to a similar format: summarise received wisdom about surname etymology and meaning; give rough geographic distributions; list well-known bearers of the name. They are essentially dictionaries focused on elucidating the surnames themselves, which makes them mainly of interest to bearers of the surnames and to local historians.

I’ve produced plenty of similar potted histories myself, and found it very hard. So hard that at one point in the early 90s I ended up inventing a ‘well-known’ bearer of a surname, complete with fictional back-story. And then forgot which surname it was. So somewhere, I think on this site, is a non-existent famous person. I believe (though I can’t be sure) that I was channelling The Scarlet Pimpernel.

More seriously,  the study of surnames, in particular surname distributions, can provide decent historical evidence, especially now that technology allows historic data to be mined and examined in novel ways. One example is  mapping surname variety across Ireland in the mid-19th-century Griffith’s Valuation census substitute. Simply take the number of distinct surnames listed as householders in each county and divide by the area of the county. The result is an average number of different surnames per area.

Distinct surnames per square kilometre

Unsurprisingly, Dublin has the densest concentration of names, but the area with by far the next greatest variety is the ancient tuatha of Oriel, comprising Armagh, Louth and Monaghan. The western seaboard counties (with the exception of Sligo) have surname densities far below average, even though they were the most highly populated areas. The northeastern counties, with their mix of Scots-Irish and Gaelic-Irish, have surname variety well above average. The clear conclusion is that surname variety or density is a respectable proxy for cultural diversity. Or cultural purity.

Plenty of onomastics like this can be found at the Society for Name Studies in Britain and Ireland,

The cousins! The cousins!

Between 1780 and 1845, the population of Ireland mushroomed from about three million to about eight and a half million. That much is well known and often cited as one of the reasons for the catastrophic devastation of the Great Famine of 1845-49. But such a vast increase must also have had a profound effect on family relationships.


Teenage marriage, cousins to dance with and only a potato crop to harvest. What could possibly go wrong?

Look at the numbers. The population tripled in size over less than three generations, a veritable explosion. And the growth wasn’t evenly spread, taking place disproportionately among the vast mass of Gaelic Catholic subsistence farmers in the West and the South. Someone born in those areas in the 1820s and 1830s was typically one of a very large family, ten or more, whose parents and grandparents would also typically have come from families the same size. So if your twenty aunts and uncles did their demographic duty, you could easily have two hundred first cousins, all almost probably living within walking distance.

And more distant cousins? Second cousins are all the descendants of a common set of great-grandparents. Taking the same average of ten per generation, you could easily have had more than four thousand of them, again almost all in the same geographic region.

Child mortality and (pretty much unavoidable) cousin intermarriage might reduce some of those numbers, but the point still stands. Immediately before the Famine, some areas of Ireland were populated by extraordinarily dense cousin networks.

Gaelic society had always been tribal, but this was tribalism on steroids. Everyone was literally closely related to everyone around them. No wonder Ancestry’s DNA service is so good at identifying genetic groupings in particular areas of Ireland – the period 1780-1830, the outer limit of its standard autosomal DNA test, is exactly the period when we very obligingly got together and married our cousins like rabbits.

The Grenhams hard at it in the 1830s

Cousin-density like this also left its mark on a more recent Ireland. Membership of the two main political parties, Fianna Fáil and Fine Gael, often appears to be inheritable, with great masses of relatives voting for (or knifing) each other .

And that great population bulge may be another reason why it’s so hard to get your ancestral line back beyond 1800. Four thousand Jams O’Donnell second cousins?




The Valuation Office: post-Griffith’s records

The 1852 Act that (retrospectively) provided the legal underpinning for Richard Griffith’s great survey also envisaged annual revisions. It was a live tax record, after all, and needed to reflect changes in occupier, size of holding, name of lessor  – anything that might affect the valuation and the tax payable. Some of Griffith’s local army of  valuators must have remained in situ, supplying notice of changes to the central Office in Ely Place, Dublin.

5 and 6 Ely Place, former home of the Valuation Office

The system in Dublin used a handwritten copy of the valuations, with amendments written over the original entry. Coloured pencils were employed to distinguish the year of the changes. (Unfortunately the microfilm copies made by the LDS Family History Library are in black and white.)

After a decade or so, the books would become increasingly difficult to read, so a fresh copy had to be made. The whole process then continued until that book in turn was cancelled and recopied.

Despite the original plans, full annual revisions were never a realistic prospect. The system eventually settled down to a natural rhythm, with a single revision (or “cancelled”) book covering fifteen or twenty years. For areas in the Republic, the revision process went on right up to the 1980s, when the local property tax was abolished, producing anything up to twelve separate books for a single area. In Northern Ireland the system changed in the 1930s and the most recent revisions are almost all for the 1920s.

Ballyminaunhill, Wexford, c. 1900

The usefulness of the records is obvious: they provide a continual time-lapse record of every piece of property listed by Griffith, unbroken over the following century-and-a-half. Circumstantial evidence of death, emigration, money problems, family disputes – the revision books can contain them all.

Most importantly, the massive shift in land ownership to small tenant farmers that took place around the start of the twentieth century, and the fierce land hunger that drove it, means that there is almost certainly a family relationship between those who own the land now and those recorded in the revision books then. In other words, the books can provide excellent evidence of living relatives.

Ballyminaunhill, final revision book. Things got complicated in the late twentieth century.

The books are available for areas in the Republic at the Valuation Office in the Irish Life Centre, Lower Abbey Street, Dublin. The VO is currently scanning the full collection with a view to making it available online. For the moment, the scans are only available onsite in the public office.  Areas covered  are Cos Cork, Donegal, Dublin, Galway, Kerry, Limerick, Mayo, Meath, Monaghan, Offaly, Roscommon. Sligo and Tipperary.

The Revision Books for Northern Ireland are in PRONI. An excellent PRONI sub-site allows online research by placename on the full set, with links to high-quality scans of the originals. Even for people without Ulster ancestor, the sub-site gives a great sense of just how useful these wonderful records can be.


The Valuation Office: pre-Griffith’s records

Irish research doesn’t have many under-explored collections, so you’d expect joy unconfined to surround the records of the Valuation Office. However, most researchers’ eyes glaze over when you mention them.  And for good reason. The Office is one of the oldest continuously functioning arms of public administration in Ireland, set up in 1826 and still operating today. That’s almost two centuries, long enough to produce entire Himalayas of paper, which include some of the gnarliest records imaginable.

The Irish LIfe Centre in Abbey St, Dublin, current home of the VO

The Office’s basic role was (and is) simple enough: to create property valuations on which local property taxes could be based. Its main achievement was Griffith’s Primary Valuation of 1847-1864. But the creation of Griffith’s was far from simple.

The initial Townland Valuation Act (1826) allowed for a complete assessment of the annual value of every parcel of land and building in Ireland, with the aim of producing a total for each townland that could be used as the basis of a yearly tax. But in 1826 nobody even knew what townlands were where, so the project became dependant on, and intertwined with, the Boundary Commission and the Ordnance Survey.

Griffith in 1854 with his beloved geological map of Ireland (Stephen Catterson Smith, National Gallery of Ireland)

The earliest OS maps were produced for Londonderry in 1831 and it was then and there that surveying began.  Richard Griffith, the Valuation Commissioner, quickly saw that the plan to value everything was far too ambitious for the resources available. While continuing to value all land, a  threshold of £3 annual value was adopted for buildings to be assessed, excluding the large majority of householders but still covering a significant number of dwellings and commercial premises, especially in towns. The valuation continued on this basis for the next seven years, covering eight of the northern counties. By 1836, however, it was clear that even with a £3 threshold the surveying would never be completed. In that year the threshold was raised to £5, covering only the most substantial buildings.

Londonderry 1831, where it all began

The year 1838 saw the introduction of the Irish Poor Law, a rudimentary system of relief for the most destitute, and its funding was  based on a separate property survey. It quickly became obvious that it made no sense to have two separate systems of local taxation, based on differing valuations of the same property. In 1844, with townland valuations complete for 27 of the 32 counties, Griffith was authorised to change the basis of assessment by dropping the £5 threshold and covering all property in the remaining counties, all of them in Munster.

The Tenement Valuation Act (1852) retrospectively allowed Griffith to extend the system throughout Ireland – a procedure he had in fact already begun. Over two decades of valuation he had assembled a veritable army of skilled employees, each with a precise role in his vast valuing mechanism. It allowed him to publish county-by-county surveys of the entire island between 1847 and 1864, generally with the southern counties earlier and the northern counties later.

In the course of this long-drawn-out saga, huge quantities of manuscript records were produced. Firstly, the pre-1836 valuation of the northern counties, based on the £3 building threshold, created local valuers’ “house books” and “field books”, the former including the names of occupiers, the latter in theory at least concerned purely with soil productivity. They remain particularly useful for urban or semi-urban areas in northern counties before 1838.

After the change in the basis of assessment in 1844, the main categories of valuers’ notebooks continued to be known as ‘house books’ and ‘field books’, but the distinction became more than a little blurred, with information on occupiers appearing in both. To add to the merriment, other classes of notebook were also created:

  • Tenure books, showing landlord and lease information;
  • Rent books, showing rents paid, as an aid to valuation;
  • Quarto books, covering towns,
  • Perambulation books, recording valuers’ visits, and
  • Mill books.

There are far fewer of these than of the ‘house’ and ‘field’ books.

The survival of pre-publication Valuation Office records is patchy, with some parishes having four or more books while others have none at all. The only way to find out what’s there is to examine the records themselves. There are two major collections, one in PRONI, the other in NAI. The PRONI records cover only the six countues of Northern Ireland and are only available onsite in Belfast. It’s not clear from the information they provide that any of the post-1844 books are publicly available. In 2003, the LDS microfilmed the entire collection then held by NAI and this is the collection now online at

Another set of records, held by the Valuation Office itself until 2003 and subsequently moved to NAI, is still being catalogued and conserved and so is not as yet publicly available. The word is that it includes thousands upon thousands of valuers’ hand-made maps, with the tantalizing prospect of detailed field and townland maps that will hold a magnifying glass to rural Ireland of the 1830s and 1840s.

NAI has done a wonderful job of making the earlier microfilmed records intelligible and accessible. Written by former director Frances Magee, the introductory text for each of the manuscript types is a master-class in the valuation process and the codes used by the valuers. She has a book on the Valuation Office collection in the works, due out some time in the next twelve months. Roll on the day.

One caveat about the database transcripts. Something funny seems to be happening on the NAI site. Compare the search results for a Martin Heavy on FindMyPast with the results on the NAI site. There should be no difference – they’re the same underlying datasets. For the moment, the FindMyPast search is the one I trust. It’s free and also permits browsing by parish, something not available on the NAI site.

Licknavar, outside Skibbereen, March 31 1853: four households

These records might seem marginal and messy compared with the published Valuation, but keep in mind that Griffith’s, far from being the record of a settled population, is a snapshot of the aftermath of a catastrophe, the Great Famine. In many areas enormous changes took place  between the original survey and final publication.  In the area around Skibbereen, for example. the pre-publication records are be the only surviving evidence of entire villages that were wiped out.

Licknavar, outside Skibbereen, July 24 1850: fourteen households


Techno-utopia or techno-dystopia?

Last week, RTÉ TV news ran a piece of techno-utopianism that really got my attention. A group based in Trinity College are going to create a digital 3-D model of the old Public Record Office that was destroyed in 1922.

Reconstruct this.

Fair enough. The centenary of that ignominious act is approaching and it certainly needs remembering.

Then the report went on to say that the group was also going to digitally reconstruct the records that had been destroyed, thus retrieving seven lost centuries of history and genealogy. Say again? They’re going to magic back into existence the ten million or so returns from the lost censuses of 1821, 1831, 1841 and 1851? How? By closing their eyes and clicking their heels together three times?

The group’s website tells a more nuanced story. Yes, there will indeed be a shiny, walk-through digital model, but the approach to reconstructing records is modest enough.  It centres on using Herbert Wood’s A Guide to the Records Deposited in the Public Record Office of Ireland (Dublin, 1919) to identify substitutes for what was destroyed, much as  the National Archives of Ireland has been doing since 1922 with, for example, its testamentary record substitutes. The difference is that copies of, or links to, the substitute material will be organised within the virtual model, thus recreating (kind of) the experience of using the old PRO.

It’s a good idea. Any opportunity to digitise and publicise surviving fragments and substitutes is very welcome. And it should also shine a light into a few areas that don’t get enough attention – the seven volumes of thirteenth to sixteenth-century excerpts from plea rolls, patent rolls and pipe rolls in the National Library Genealogical Office collection (GO 189-95), for one.

But I think that the end result is likely to be more dystopian than utopian. At last we’ll get to see, perfectly-rendered, the size and shape of the black hole at the heart of Irish history and genealogy.

The project site has a news page that links to all the attention they got last week. It wasn’t just RTE, nearly every other news outlet got the story wrong and reported the imminent return out of digital thin air of all the lost records. The team need to be a bit more careful about how they attract attention to what they’re doing.  Some of those news reports look distinctly fake.

Persistent and heavy falls of frogs

Irish whinging about the weather can get on your nerves. But my God the 2017-18  winter deserves all the whinging it gets. It’s toying with us, pretending to stop just long enough to get us to put away our thermal long johns, then whipping back into vicious life. Here’s how I comfort myself as I stand at a Dublin bus-stop with every piece of exposed flesh flayed by horizontal sleet.

Soft day, mind you.

First I remember how things were just a short generation ago, when central heating was a novelty in Ireland. For the four months of winter, life shrank down to the semi-circle around the fire. Many’s the seat of a pair of trousers I singed painfully in the quest for a biteen of heat. Ochón

Then I think back further. Imagine life in the Viking town around Wood Quay in Dublin, where the only heat would come from a fire without a chimney in the centre of a mud-and-wattle hut. Then just think what persistent Irish February rain could do to a mud-and-wattle hut.

“What didn’t kill them made them  stronger”? I’m sure Irish Februarys killed more Vikings than Brian Ború. So I feel thankful (and  hope that bloody bus comes soon).

rain locust blood boils footandmouth darkness.

My real heroes at this time of the year, though, are Irish Met Office forecasters.  Your heart has to go out to them. They have the most thankless task you could imagine, so they keep trying to soften the message:

“Scattered showers of locusts spreading from the West in the morning, turning to fiery hail by the afternoon and becoming persistent overnight. But it’ll be mostly dry on Friday!”

Weather in Dublin, February 1 -9, 2018


Resistance Genealogy

The current issue of The New Yorker has a lovely article about Jennifer Mendelsohn, an American genealogist who has taken to researching the immigrant ancestors of US politicians and pundits who make anti-immigration pronouncements. Time after time, she’s come up with ancestors who personify the supposed failings being denounced. I especially like her response to Fox News commentator Tomi Lahren who finger-wagged: “Respect our laws and we welcome you. If not, bye”. Tomi’s great-great-grandfather was indicted for forging citizenship papers.

The indictment. Disappointingly, he was acquitted.

America is obviously ripe for this kind of thing – as she says herself, “Unless you’re Native American or you descend from slaves who were brought here against their will, you are an immigrant in this country, or you’re a descendant of an immigrant in this country.” On the other hand, I doubt what she’s doing will have much effect. Respect for the truth is not a conspicuous characteristic of those she’s challenging. I’m sure the massed ranks of Fox News commentators are laughing at her naivety as they tuck in to their breakfast of broiled baby immigrant.

Is there any equivalent in Ireland? The only example I can thing of is Catherine Corless wading through Tuam death records to bring to light the evidence of  Mother and Baby Home abuses.  I’d be very interested in suggestions.

Then, of course, there’s the Irish contribution to the current state of the US. Or, as an article on the History News Network has it, “ Why are all the conservative loudmouths Irish-American?