Comments on: English ? Scottish ? Irish ? What's the difference ?

English ? Scottish ? Irish ? What's the difference ?

y2karl — Fri, 09 Mar 2007 23:59:23 -0800

...Historians teach that they are mostly descended from different peoples: the Irish from the Celts and the English from the Anglo-Saxons who invaded from northern Europe and drove the Celts to the country's western and northern fringes. But geneticists who have tested DNA throughout the British Isles are edging toward a different conclusion. Many are struck by the overall genetic similarities, leading some to claim that both Britain and Ireland have been inhabited for thousands of years by a single people that have remained in the majority, with only minor additions from later invaders like Celts, Romans, Angles, Saxons, Vikings and Normans. The implication that the Irish, English, Scottish and Welsh have a great deal in common with each other, at least from the geneticist's point of view, seems likely to please no one.

A United Kingdom? Maybe
See also Myths of British ancestry
In the words of one well known Basque cultural icon: HA Ha!

By: lattiboy

lattiboy — Sat, 10 Mar 2007 00:20:29 -0800

As an almost even split of 3rd generation Scott, Irish, and English (and who freely refers to myself as "American"); I no longer feel so special. In fact I feel quite antiquated.

By: Abiezer

Abiezer — Sat, 10 Mar 2007 00:52:51 -0800

What's in a name?--Britons, Angles, ethnicity and material culture from the fourth to seventh centuries

By: ClanvidHorse

ClanvidHorse — Sat, 10 Mar 2007 02:11:44 -0800

Lattiboy, who can you get an almost even three way split? And, its Scot (with one t!). Anyway, not that it matters much. I am a 2nd generation Scot from Irish descent and apparently my dark hair derives from when the Spanish were banging all the women in sight during the armada. Roy Keane is the same apparently.

By: DenOfSizer

DenOfSizer — Sat, 10 Mar 2007 02:22:18 -0800

Once they do the same with the French and the Germans, my German Irish inlaws will finally be driven completey nuts. Go science! Y2Karl, why are you up this late? And are you going to change your name to DSTKarl?

By: mr. strange

mr. strange — Sat, 10 Mar 2007 02:22:36 -0800

Anything that reminds us that we're all basically the same has to be a good thing. In other news: Humans are discovered to share more DNA with other humans than with other animals.

By: sebastienbailard

sebastienbailard — Sat, 10 Mar 2007 02:47:29 -0800

In other news: Humans are discovered to share more DNA with other humans than with other animals. Well that just means we're going to have to try harder, doesn't it? Here kitty-kitty-kitty-kitty...

By: spiderwire

spiderwire — Sat, 10 Mar 2007 02:49:16 -0800

Jesus I feel like I just got smacked upside the head with a blockquote tag

By: jack_mo

jack_mo — Sat, 10 Mar 2007 03:12:18 -0800

The implication that the Irish, English, Scottish and Welsh have a great deal in common with each other, at least from the geneticist's point of view, seems likely to please no one. On the plus side, if you meet anyone enraged by the news, you'll know they're a nationalist dickhead. Also, we discussed this last September when the news broke. Well, re-broke - seems that back in 1320, the Scots had a fairly good idea of their origins:

Most Holy Father and Lord, we know and from the chronicles and books of the ancients we find that among other famous nations our own, the Scots, has been graced with widespread renown. They journeyed from Greater Scythia by way of the Tyrrhenian Sea and the Pillars of Hercules, and dwelt for a long course of time in Spain among the most savage tribes, but nowhere could they be subdued by any race, however barbarous.

By: Ethereal Bligh

Ethereal Bligh — Sat, 10 Mar 2007 03:41:53 -0800

"...[how] can you get an almost even three way split?" "Almost" is the key word. If one of his parents was half one and one-quarter of each of the others, and the other the same but a different mix, then he's close to a third each. Specifically, he's 3/8 of two and 1/4 the third. True thirds aren't possible, but can be more and more approached by doing the same sort of thing successively. As to the linked article, it sounds promising except for the linguistics part of it. Perhaps I'm selling these statistical geneticists short, but my minimally informed opinion is that they are approaching their attempt at statistical linguistics with a great deal of intentional naivete that borders on willful ignorance. It seems to me there's two serious problems with it. The first, and what linguists will probably say, is that how languages change is quite a bit (sufficiently) different from how a genome changes to make such simplistic statistical analysis worse than useless. The second, and more my opinion, is that we lack any real understanding of language sufficient to have almost any confidence at all such that a statistical analysis like this would be trustworthy. Most of the really interesting things in genetics is still mysterious and yet at least it's chemistry. We don't have any sort of underlying physical model for the processes of language to build upon. It's all top-down—linguistics is still at the Mendelian stage. And turning that analogy around, you can see that it's arguable that these geneticists using statistical analysis on phonemes would be as if they were using the same sorts of analysis on some attempt at a quantification of the variation and inherited relationships of phenotypes. Such an analysis would give some results that happen to be right, and others that are profoundly wrong. The reason statistical analysis of population genetics works so well is because, basically, what they're looking at and what they're interested in is very, very simple. The relationship between randomness, order, and change is simple. They're looking at the genetics at a very low level, close to the reductionist bottom. We're a long, long way from that with language.

By: Miko

Miko — Sat, 10 Mar 2007 04:58:03 -0800

What? The Irish the same as the English! Begorrah! *shakes shillelagh* A fine St. Paddy's-day-week post. Genetics is discovering some really fascinating things about the history of human migrations, it seems. In folklore, there's a lot of interest in watching how genetic data affects the origin stories of various peoples. When your tradition tells you that your people came from high in the mountains and were led into your valley by a benevolent creator, but your genes tell you you're actually a descendant of the historical enemy group across the river, there's some deep and often uncomfortable reflection to be done. Clanvid horse: I grew up hearing about the Black Irish, too, and noting the differences between the dark-haired, blue-eyed people of Irish descent who looked quite different from the auburn-haired, hazel-eyed ones in my family. But the consensus is that the story of the Black Irish descending from invaders during the Spanish Armada is myth, and possibly an American one, at that. It seems that the occurence of the dark hair is just more evidence of the Iberian migration.

By: sour cream

sour cream — Sat, 10 Mar 2007 05:31:39 -0800

If the pre-Celt-Anglo-Roman inhabitants of the British isles were speaking Basque, then what happened to all the Basque place names that would inevitably have remained from such a sub-stratum? Is there evidence of any Basque place names in Britain or Ireland?

By: Astro Zombie

Astro Zombie — Sat, 10 Mar 2007 06:10:37 -0800

I've always been told that I was dark Irish. Now I find out we all are.

By: Miko

Miko — Sat, 10 Mar 2007 06:16:53 -0800

Sour cream, are you a good enough speaker of either Irish Gaelic or Basque to know that Irish place names are definitely not related to Basque place names? I'm not a linguist, of course, but I'm not sure I'd be able to tell my Inchicronan from my Imarcoain, my Leagui from my Laois, or my Meagas from my Maigh Nuad just by the name. Though I have no basis for saying so, the languages don't seem so very dissimilar that there could be no relationship. But this is a job for languagehat.

By: jonmc

jonmc — Sat, 10 Mar 2007 06:24:09 -0800

But the consensus is that the story of the Black Irish descending from invaders during the Spanish Armada is myth, and possibly an American one, at that. It seems that the occurence of the dark hair is just more evidence of the Iberian migration. I am one very confused black haired Irish-American right now.

By: Faze

Faze — Sat, 10 Mar 2007 06:36:37 -0800

I love this piece of news, old as it is. What now of Sir Walter Scott, Robert Louis Stevenson and Mel Gibson? If this doesn't shake the British Isles to their very foundations, what will? (Of course, we outsiders have always noticed the pasty, knobby, pustuled similarity among all of you and wondered at the cluelessness of your competing nationalisms.)

By: languagehat

languagehat — Sat, 10 Mar 2007 06:41:15 -0800

Perhaps I'm selling these statistical geneticists short, but my minimally informed opinion is that they are approaching their attempt at statistical linguistics with a great deal of intentional naivete that borders on willful ignorance. No, you're absolutely right. Read the Language Log takedown of the idiocy involved. Why can't specialists stick to their own fields, or if they want to go cross-disciplinary, check with somebody who actually knows something about the other discipline? Bah.

By: jonmc

jonmc — Sat, 10 Mar 2007 06:41:21 -0800

We can still be a bunch of brawling drunks though, right?

By: malthas

malthas — Sat, 10 Mar 2007 06:56:04 -0800

Ahh, so Sean Connery playing the Spanish guy in Highlander does makes sense.

By: Coventry

Coventry — Sat, 10 Mar 2007 06:58:27 -0800

Reminiscent of Greg Egan's story, Mitochondrial Eve.

By: pyramid termite

pyramid termite — Sat, 10 Mar 2007 06:59:10 -0800

there's nought else to do, jonmc ... especially since the potato chip factory burned down and took my life savings with it

By: bonaldi

bonaldi — Sat, 10 Mar 2007 07:09:26 -0800

Seriously, are there still people who think that genetics is what makes one set of people charming and with a tendency to produce great literature and another set pasty with a tendency to eat chips?

By: pyramid termite

pyramid termite — Sat, 10 Mar 2007 07:30:35 -0800

what about the chippy people who eat pasties?

By: ROU_Xenophobe

ROU_Xenophobe — Sat, 10 Mar 2007 07:53:38 -0800

attiboy, who can you get an almost even three way split? 16 great-grandparents. 5 English, 5 Irish, 5 German, and one thylacine.

By: b1tr0t

b1tr0t — Sat, 10 Mar 2007 07:53:39 -0800

I'm sorry, but WBP doesn't quite roll off the tongue like WASP does.

By: ROU_Xenophobe

ROU_Xenophobe — Sat, 10 Mar 2007 07:54:43 -0800

s/German/Scottish/g I was thinking of my own ancestry there for a minute. Especially the thylacine... gramma she-wolf was hot.

By: serazin

serazin — Sat, 10 Mar 2007 08:08:08 -0800

Awesome. I sent this to my Scottish dad.

By: Foosnark

Foosnark — Sat, 10 Mar 2007 08:47:12 -0800

Clan MacLaren legend is that the founders of the clan were Irish invaders and native mermaids. (And I assumed there might be a grain of truth in it, if you substitute "Picts, who carved mermaids on things" for "mermaids.")

By: flashboy

flashboy — Sat, 10 Mar 2007 08:49:33 -0800

I hate the current trend for which the core principle is that no science story is worth reporting unless it involves Big Fuckoff Paradigm Shift. Because it leads people - the journalists, the scientists and the scientists' press people - to entirely mischaracterise the current state of knowledge in order to make the new research seem far more world-changing than it actually is. The notion that the prevailing model is a replacement-based one - where the current population of the British Isles is neatly divided up into groups who have entirely different, invasion-based ancestries - is a fiction, pure and simple. Oppenheimer weasel-wordedly acknowledges this, but buried deep within his article. Talking of the genocide-replacement model, he admits: "With the swing in academic fashion against "migrationism" (seeing the spread of cultural influence as dependent on significant migrations) over the past couple of decades, archaeologists are now downplaying this story, although it remains a strong underlying perspective in history books." It's not an "academic fashion", as though its just this season's whimsical fad - it's the actual state of knowledge, based on evidence, in the discipline. The actual genetic research is looks an interesting contribution (although far from conclusive) to the study of the messy, complicated, history of the British Isles and Europe generally. But it's not a Big Fuckoff Paradigm Shift, and I wish people wouldn't pretend it was. And let's not get onto how profoundly stupid an opening sentence "Britain and Ireland are so thoroughly divided in their histories that there is no single word to refer to the inhabitants of both islands" is.

By: y2karl

y2karl — Sat, 10 Mar 2007 08:54:53 -0800

By: Mick

Mick — Sat, 10 Mar 2007 09:33:15 -0800

We drink and we fight and we fight and we drink ...and crrrrrry

By: meehawl

meehawl — Sat, 10 Mar 2007 09:38:43 -0800

What this proves: Populations, especially in pre-modern eras, were porous and exogamous. People will shag anything, and the motion of chromosomes can best be seen as a random walk. This is old news, yet the revelation that people are horny and get lots of play constantly seems to surprise some people. I suspect this is because many of these people don't get out much, and are crap at exponential maths and geometric series. Look at the long run, over hundreds of generations, and a slight early mover advantage will tend to distribute haplotypes in a Gaussian distribution throughout the population. Somewhere between 1 and 5% of children do not have the male parent they think they have - compound that over a few hundred generations... Cultures (or at least, the fittest ones that survive) spread in totalizing waves that impose notions of "us" and "them" on highly genetically heterogeneous substrates. I still haven't seen an alternative explanation for the rapid spread of recessive blondies and gingers in northern Europe, however, except that it was through definite sexual selection and lusty campaigns.

By: lattiboy

lattiboy — Sat, 10 Mar 2007 09:51:48 -0800

This "Almost" is the key word. If one of his parents was half one and one-quarter of each of the others, and the other the same but a different mix, then he's close to a third each. Specifically, he's 3/8 of two and 1/4 the third. True thirds aren't possible, but can be more and more approached by doing the same sort of thing successively. answers this: Lattiboy, who can you get an almost even three way split? And this: posted by lattiboy at 12:20 AM PST Answers this: And, its Scot (with one t!). Bow to my cut and paste skillz!

By: Termite

Termite — Sat, 10 Mar 2007 09:59:22 -0800

From the NY Times article: "the principal ancestors of today's British and Irish populations arrived from Spain about 16,000 years ago, speaking a language related to Basque." I'm not even an amateur linguist, but this seems like an absurd statement to me - that there could be anything that could be called "related to Basque" 16,000 years ago, and that we could have any idea what language people in a particular place spoke 16,000 years ago - some 10,000 years before the oldest known written records.

By: infinitewindow

infinitewindow — Sat, 10 Mar 2007 10:10:48 -0800

When your tradition tells you that your people came from high in the mountains and were led into your valley by a benevolent creator, but your genes tell you you're actually a descendant of the historical enemy group across the river, there's some deep and often uncomfortable reflection to be done.

True dat. I have a buddy who was proudly Armenian-American and fairly anti-Azerbaijani. When he found out his name meant his ancestors were from Azerbaijan, he had some deep thinkin' to do. For my part, it's nice to know I'm a mutt whose racial and social origins are pretty much unknown thanks to poor record-keeping. At the same time, it would be nice to know that I may develop Tay-Sachs, sickle-cell anemia, alpha-1 antitrypsin deficiency or one of many other nasty genetic diseases. The unavoidable thing for me is that these conditions are closely associated with a specific ethnic community. I can't predict my future without uncovering my past, and the implications of that sadden me.

By: languagehat

languagehat — Sat, 10 Mar 2007 10:26:06 -0800

I'm not even an amateur linguist, but this seems like an absurd statement to me It is. See my link above.

By: dbarefoot

dbarefoot — Sat, 10 Mar 2007 15:10:27 -0800

I refer you to the Great (but inaccurately named) British Venn diagram.

By: bingo

bingo — Sat, 10 Mar 2007 19:07:48 -0800

Ahh, so Sean Connery playing the Spanish guy in Highlander does makes sense. You'll be disappointed to hear that his character was Egyptian, then.

By: Miko

Miko — Sat, 10 Mar 2007 20:26:38 -0800

dbarefoot: That is the simplest, most wonderful thing I've seen in a long time. If only that had been around when I got my BA in English Lit. It clears up a whole lot of basic questions.

By: flashboy

flashboy — Mon, 12 Mar 2007 16:04:24 -0800

According to someone who know more about genetics than me, this qualifies as "bad historical population genetics".

By: languagehat

languagehat — Tue, 13 Mar 2007 09:02:15 -0800

Mark Liberman in Language Log: Not just bad linguistics, but shoddy historical genetics too? Sounds like the whole thing is a crock.