Of Slavs, Slaves, Vikings, and Genetics

I recently had a DNA test to help trace my ancestry, and the result surprised me. The larger story might shed light on one of the grimmest and most forgotten horrors of European history, an era of brutal slave trading.

By way of background, my known genealogy is very straightforward indeed. It shows close to 100 percent Welsh – not just Welsh, but one specific bit of south Welsh. That means mainly West Glamorgan, within a few miles of the city of Swansea, although with a couple of English guys in the 17th-18th centuries (In Wales, we call that “diversity”). I can identify all my ancestors through all lines back to about 1840, and far beyond that in some lines.

Hence, I am near pure 100 percent Welsh on all sides. However, through the years, I have faced a nagging question. Welsh people are stereotypically short and dark, which I am not. (I am 6’2”, and not dark). Nor were my uncles and aunts, who were all pretty tall people. At least they were on my maternal side, and I’ll explain in a moment why that distinction matters.

When I am in Europe, people all over the continent often have me marked as German, and address me as such. They greet a line of tourists like this: “Good morning sir!” “Good morning sir!” Then they come to me: “Guten Tag, mein Herr!” In Norway, the locals assume I am Norwegian.

What on earth is happening? Something was amiss. I was amiss.

Hence my inspiration to take the DNA test, and the result is fascinating. (I used FamilyTreeDNA). In total contrast to the genealogy, the DNA gives me as 90 percent British Isles origins and eight percent Eastern Europe, plus a smattering from south-east Europe. Now, it never pays to take such percentages precisely, but this is suggestive. And I can confirm that Eastern Europe linkage from another source.

More specifically, I had my mitochondrial DNA done, which only traces descent in the female line – mother to daughter to daughter, so that I cannot pass it on to my children. We measure this by the MtDNA haplogroup, of which there are a couple of dozen world-wide, and each is given a capital letter, so that for instance M is found among people in south east Asia, D in Japan, O in China, etc. There are also lots of subsets of those larger families. British Isles haplogroups are often J or T. The main MtDNA hapologroup in Wales is H.

My haplogroup, though, is none of the above, it is U, and specifically U4. It is in fact a striking (and quite rare) U4a1a, which points to Eastern Europe or the Baltic.


In the available commercial databases online, most people with that haplogroup tend to be Swedish, German, Danish, Polish …. Now, that statement is a bit slanted, as these databases only include people who have paid to get their DNA results done, so that would lead to a massive over-representation of wealthy northern Europeans. In no sense are these reliable scientific samples, nor do they say anything definitive about the actual distribution of U4 across Europe. Even so, U4 is not too common as a Welsh (or British) pattern. The furthest I can go back in my own female line is my great grandmother, and like all names on my chart, she was definitely south Welsh, with not a Pole or a Ukrainian in sight.

I should say by the way that I am not uncritically relying on these findings, which might be erroneous. But I have good reason to accept what I was told. In the paternal line, the results suggested individuals to whom I might be related, and I happen to know that those people and I share relatives with common surnames multiple generations back. There is no way a company could have cooked up such obscure, and uncannily accurate, findings. By extension then, I tend to trust the mitochondrial results.

My genealogy says one thing. My genetic information suggests something radically different.

Making life more difficult, in order to find the root of this genetic pattern, we would have to locate a woman, as only women carry MtDNA. We could not for instance assume an earlier woman in my ancestry who had a fling with a wandering Hungarian hussar in Napoleonic times. Nor, more seriously, can we invoke the many documented examples of skilled European workers traveling to Britain in early modern times, especially in pioneering Welsh industries like coal and iron. As far as I know, these visitors or migrants were all male.

So where does that U4 come from? I have an explanation – not the right one, necessarily, but an interesting speculation.

Wouldn’t it be wonderful if we could somehow place a woman from the Baltic/Slavic regions in South Wales in the pre-modern period, preferably very close to where my maternal family originated? Surely, that is a very tall order. Oddly enough, though, there is a historical window in which we can do something very much like that, and with remarkable geographical precision.

Over the past two centuries or so, my maternal family simply has not moved around much (Nor has my paternal line, but that is a different story). They have in fact remained within that small area of West Glamorgan, around Swansea and Neath, never really moving more than twenty miles or so in any direction. For the sake of argument, let us then assume they have in fact been in that small part of south Wales for centuries. Aha, but then we find an interesting connection. What we know about that area is that it is right next to one of the key regions of Scandinavian settlement in the British Isles in the Viking era, the ninth and tenth centuries.

The most important Scandinavian center in South Wales was Sweyn’s Inlet or Island, “Sweyns-eye” – that is, Swansea. It’s an open question whether Sweyn or Sveinn refers to a famous Norse king of that name, or just a lone adventurer. Near Swansea, Scandinavian names occur across the region of lower Gower, but not upper. The wonderful coastal landmark of the Worm’s Head in Gower is actually the head of the Ormr, Norse for a great Serpent (and it really looks like a sea serpent). A lot of the islands around the Welsh coast have pure Norse names like holm or -ey, as in Caldey. Flatholm is the Island of the Fleet. Such names scatter all across the coastal map to West Wales places like Tenby, another Scandinavian name.

Not only did the Norse name such places, but they and their descendants remained long enough to ensure that other people adopted and remembered the names. These areas were not just temporary camps: they were important enough to be real settlements, over decades or generations. At least along the coasts, the Vikings were there in force.

So what was the gender balance of these Scandinavian ventures? We tend to think of chiefs and raiders as all male communities, seizing local women, but there was more to the story. As settlements became more established, they might have brought wives or prospective (free) marriage partners. Over the years, archaeologists have recognized ever more examples of Scandinavian women buried in Britain. Now, one of these free women might just possibly have brought the specific U4a1a type direct from Denmark or Sweden.

But other, more sinister, factors were also at work, involving slave women. As I say, we are looking at the 9th-10th centuries. At this very time, one of the world’s largest slave trading operations was centered on the Baltic Sea, particularly seeking out slaves from the Slav and Baltic peoples. The word “slave” comes from “Slav,” but Finland was another great center for slaves. There is a huge scholarly literature on all this.

Recent scholarship suggests that slavery and slave trading were a major incentive for the whole Viking enterprise, from the eighth century onwards. In a polygamous aristocratic society, lower status men found it hard to obtain wives within their own communities, driving them to seek women elsewhere, by force. Initially, they did this in Baltic lands like Estonia, but then mightily extended their reach. Following the rivers, some pushed deep into Russia, while others ventured into the Atlantic realms, but the basic goals remained the same. Reporting one major raid in 821, the Irish Annals of Ulster note that the heathens “carried off a great number of women into captivity.”

Gradually, isolated slave raids evolved into a transnational business operation that ranged across Europe, and took many slaves to the Islamic lands. Captives would have been kidnapped and taken to one of the great Swedish slave markets at Birka or Gotland, or Denmark’s Hedeby. Scandinavians did much of the raiding, while Arab traders served as financiers and middlemen, and the distribution of these slave markets is indicated by the hoards of Arabic coins, dirhems, in trading centers like Birka.

A great many of those captives and slaves must have had U4 MtDNA. As we look at the map of lands where the U4 MtDNA pattern is most common, we also see the regions most heavily raided for their slaves precisely around this time.

Slave trading was thus a very large part of the economic life of the Viking world. Among other things, their enterprises ensured that large numbers of Irish and British slaves (thralls) ended up in early Iceland, where they have left a large genetic mark on the modern population. It would have been very natural for a Viking, maybe even the Sweyn who founded Swansea, to have had some slave girls along, whether as bed partners or as inventory for sale. In Iceland at least, some unfree women achieved the higher status of an acknowledged concubine, a frilla. Or possibly, a freeborn Norse woman brought along her unfree serving women and maids, even her nursemaid or her lady’s maid. A female slave, by the way, was usually called an ambátt rather than a thrall. Over time, slaves might be freed and join the mainstream community.

Let us suppose that those unfree women had daughters, who intermarried with local Welsh men – perhaps married, or else they were sexually exploited without their consent. They might have been sold, traded, or used as gifts. Whatever the exact process at work, any of these interactions would explain the importation of the U4 lineage into Wales.

Life for these slaves was as miserable as you might expect. In Norwegian law, slaves and thralls were described in the neuter gender: they were “it” rather than he or she, and were classified as just slightly superior to cattle. This is very much confirmed by the horrible portrayals of thralls that we repeatedly find in the large literature of the Icelandic sagas. But that observation leads to a major point about the nature of our historical evidence. Material evidence for free or aristocratic Scandinavian women is easy enough to find in the archaeological record, because they were deposited in substantial graves and accompanied by possessions such as brooches or other jewelry, or even weapons. Slave women, in contrast, owned nothing either in this world or in the grave, and their humble burials left very little for archaeologists to identify. You just did not bury rich grave goods when a slave woman – an “it” – died. We will likely never find material remains of Viking slaves in Britain. All they might have left – just conceivably – was their genes.

So could Baltic or Slavic girls have brought their MtDNA to South Wales? Very easily. Might my own maternal family even be descended from one of Sweyn’s slaves or concubines, someone from what we would now call Poland or Lithuania? I can’t prove it, but it is plausible. If not Sweyn himself, there were lots of other comparable chieftains, who might have had girls recently imported from Birka or Gotland.

My suggestion, then, is that “Slav-raiding” and slave-trading are the main means by which U4 MtDNA found its way to the British Isles, and perhaps to other parts of Western Europe.

I am still puzzled by that eight-plus percent figure for my own East European blood, which goes far beyond a single woman a thousand years ago. And as I say, that element must have entered the bloodline well before the mid-nineteenth century. (Modern Wales has plenty of later migrants from that region, but they are not the explanation). I wonder: maybe those Vikings in Wales imported other slaves from the Baltic and eastern Europe, whose descendants merged completely into the local genetic mix. Their descendants perhaps became local Welsh families, called Jones or Evans, or Williams, or even Jenkins.

Even a handful of slaves leaving offspring could make a sizable genetic impact in such a tiny overall population. How many people did the whole of Wales have in, say, 1000 AD? Barely 100,000 in all? And perhaps 5,000 in West Glamorgan? Those were very small genetic pools.

What I can say confidently is that those Slavs or Balts did not originally migrate of their own accord.


For the regional context, see Michael North, The Baltic (Harvard 2015). On Viking society generally, see Jesse Byock, Viking Age Iceland (Penguin 2001). Kirsten A. Seaver has a chapter on Viking women slaves in her “Thralls and Queens,” in Gwyn Campbell, Suzanne Miers, and Joseph Calder Miller, eds., Women and Slavery  (Ohio University Press, 2007), vol. 1: 147-167. See also Ruth Mazo Karras, “Concubinage and Slavery in the Viking Age,” Scandinavian Studies, 62 (1990) 141-162.

I have not read it yet, but Alice Rio has a forthcoming book on Slavery After Rome, on the period 500-1100 AD (Oxford University Press, 2017).



About Philip Jenkins