Recently, the preprint for a massive new study by Morten Allentoft and colleagues went online over at bioRxiv.org. It is a paper full of important and illuminating new data from all over Stone Age Eurasia. However, this article will mainly focus on the new data from Scandinavia. The newly sequenced ancient genomes from Scandinavia are many, and many of the samples are of excellent quality. The study provides us with the much-needed ancient DNA from Mesolithic Denmark, and features samples from the old Mesolithic cultures of Maglemose and Kongemose, as well as from the Late Mesolithic Ertebølle culture.

Furthermore, the researchers also provide a good sample size from the Neolithic Funnelbeaker culture (both from Denmark and southern Sweden) as well as the Single Grave culture, giving us a clear overview of the autosomal and uniparental transitions that took place. In addition to that, Allentoft et al. 2022 delivers some fascinating genomes from Late Neolithic Scandinavia and from the Nordic Bronze Age. In short, what we have here is probably the most elaborate study ever carried out on the topic of Scandinavian archaeogenetics. Now, let us get deeper into the contents of the study.

This article can be seen as a follow-up to Genetics of the Nordic Bronze Age and The Alvastra pile-dwelling: farmer-hunter interactions in early Middle Neolithic Scandinavia.

The Mesolithic WHGs of Denmark

The many samples dating to the Danish Mesolithic indicate that the genetic landscape of the Mesolithic in the region was a homogenous one, with individuals showing autosomal ancestry almost exclusively derived from a Western-Hunter-Gatherer (WHG) source that most likely entered Scandinavia with the Ahrensburg culture. Interestingly, and unsurprisingly, all Danish Mesolithic males carried the Y-DNA haplogroup I2, which is rare among present-day Scandinavians (less than 3% among Danes, Swedes and Norwegians). This will not come as a shock to anyone who has kept up with archaeogenetics lately, of course. Instead, present-day Scandinavians predominantly carry Y-haplogroups I1(M253), R1b-U106 and R1a-Z284, in that order of frequency. As such, it is clear that there is no real paternal continuity from the Mesolithic (and not from the Middle Neolithic farmers either) in Scandinavia, but we will get into that later in this article.

In summary, the Mesolithic Danish foragers display mostly unmixed WHG ancestry and carried I2 paternal haplogroups at a frequency of 100%. They seem to have suffered a demographic decline by the end of the Mesolithic and, judging by Neolithic samples from Denmark, not made any significant contributions to Neolithic Scandinavians or to later Scandinavian populations. Exactly what happened to them is unclear, but some samples such as sample NEO962 (also known as the “Dragsholm man”) tell an interesting story. Namely, that some Ertebølle culture individuals with a fully Mesolithic WHG autosomal profile adopted a Neolithic farming lifestyle later on, and some of them were likely integrated into the later Funnelbeaker culture.

It is worth noting that despite living in close proximity to the EHG-shifted Scandinavian Hunter-Gatherers (SHGs), the Mesolithic Danish hunter-gatherers did not mix with SHGs to any significant extent.

Swedish and Norwegian Mesolithic and Neolithic hunter-gatherers

We have written about the Scandinavian-Hunter-Gatherer (SHG) population and the material cultures associated with them here and here before. With the arrival of this new study, some questions previously pondered have been answered. In previous genetic studies, the SHG population has been described as a simple two-way mix of WHGs and EHGs, with the WHG population entering Scandinavia from the south and the EHGs from the north. However, this simple two-way model has never made much sense archaeologically. For example, Manninen et al. (2021) describes as much as six Mesolithic migration events into Scandinavia. As such, the two single-event migrations that formed the SHG population as described in Günther et al. 2018 never felt very realistic.

Now, ancient DNA has once again aligned with archaeology, and Allentoft et al. (2022) found that a third population (described as Ukraine_10000BP_4000BP in the study) from Mesolithic Ukraine also contributed to the hunter-gatherers in Sweden and Norway. Thus, the two-way model for the formation of the SHG genetic profile can be discarded. The Mesolithic Ukrainian cluster, much like SHGs themselves, was a mix of EHG and WHG ancestries. What this means is that the SHG population may have been formed through the admixture between one southern WHG population, one Mesolithic Ukrainian population, as well as one northern EHG population that entered Scandinavia from what is now Finland and Russia. As for the Mesolithic Ukrainian population, it likely entered Scandinavia by crossing the Baltic Sea.

The multiple migration events that formed the SHG population is not the only hypothesis brought forth by archaeology that has now been corroborated by the new genomic data. In this article, we wrote about the Neolithic Pitted Ware hunter-gatherers from Sweden and their incursions into Denmark as suggested by the Pitted Ware arrowheads found at various Danish archaeological sites. Their exploits have now been confirmed by DNA, as the roughly 5000-year old Danish samples NEO33 and NEO898 were found to cluster with Swedish Pitted Ware individuals and derived most of their ancestry from SHGs, unlike the other Danish hunter-gatherers whose ancestry was almost exclusively derived from WHGs.

In terms of Y-DNA, the SHG genomes published in Allentoft et al. (2022) do not differ much from those published in previous studies. The Scandinavian hunter-gatherers carried almost exclusively I2 haplogroups (>98%), with a small minority of R1b-Y13202 (one sample, sample VK531 from Troms) and I-Z2699*/pre-I1 (one sample, sample SF11 from Gotland). The I2 lineages did not survive the arrival of steppe-derived populations, and neither did the EHG-derived R1b-Y13202. Those lineages were replaced by steppe-derived R1b-M269 and R1a-M417. However, the single I1/pre-I1 sample found on Mesolithic Gotland may or may not have been the progenitor of later I1 lineages (which all descend from a common ancestor (TMRCA 4600 ybp) in LN and Bronze Age Scandinavians (and by extension that found among later Germanic samples such as Anglo-Saxons, Goths, Lombards, Gepids, Vikings, et cetera) emerging from a severe bottleneck and exploding in frequency during the Nordic Bronze Age. We will get to that later in this article.

But for now, let us move on to the next group of people in the Scandinavian archaeological timeline.

The Funnelbeaker culture: not as relevant as previously thought?

We have discussed the Neolithic Funnelbeaker culture (FBC) in other articles quite extensively before. Up until Allentoft et al. (2022) was published, the genomic data from the Funnelbeaker culture was relatively incomplete, as samples were sparse. Now, however, a massive number of FBC samples have been published. With the new samples came new revelations. So let’s get to it.

Present-day Scandinavians have low Anatolian-farmer-related ancestry compared to most European populations. In Europe, only the Balts, the Finns and the Estonians carry less EEF-related ancestry than Scandinavians. Previously, this was thought be the result of the naturally harsh conditions of prehistoric Scandinavia, conditions that halted the expansion of the FBC and eventually led to the aggressive expansion of the hunter-gatherer Pitted Ware culture into FBC territory. The picture that now emerges in light of the new data is a little different. The EEF-related ancestry in present-day Scandinavians is derived from a source rich in hunter-gatherer ancestry, so the Funnelbeaker culture seemed like a natural and reasonable source. Well, until now. Allentoft et al. (2022) states the following:

“The contribution of local Neolithic farmer ancestry to the incoming groups was high in eastern, western and southern Europe, reaching >50% on the Iberian Peninsula (“postNeol” set; Extended Data Fig. 4, 6B, C)34. Scandinavia, however, portrays a dramatically different picture, with a near-complete replacement of the local Neolithic farmer population inferred across all sampled individuals (Extended Data Fig. 7B, C). “

“Following the second transition, Neolithic Anatolian-related farmer ancestry remains in Scandinavia, but the source is now different. It can be modelled as deriving almost exclusively from a genetic cluster associated with the Late Neolithic Globular Amphora Culture (GAC) (Poland_5000BP_4700BP; Extended Data Fig. 4)”

These findings are very significant, especially since the researchers used IBD (identity-by-descent)-sharing detection methods to pinpoint the EEF signal in later Scandinavian samples. As can be seen in the excerpts above, their analysis shows that the Funnelbeaker signal disappears (probably as a result of the incoming warlike steppe-derived Battle Axe culture). So, were the Funnelbeaker folks completely replaced by the Battle Axe culture (BAC)? If the HG-rich EEF ancestry in later Scandinavian populations is not from the FBC, where does it come from? Well, if we are to believe Allentoft and his colleagues, it is derived almost exclusively from the Globular Amphora culture in Poland, and was picked up by the early steppe-derived Corded Ware people there before they entered Scandinavia.

Thus, in light of these new results, it looks like there is a strong possibility that the people of the Funnelbeaker culture met with a grim fate. With hostile and aggressive hunter-gatherer populations to the north of them, and eventually the ever-expanding, warlike steppe-derived Corded Ware Indo-Europeans pressing in on them from the south, the people of the FBC truly seem to have found themselves in an unfortunate situation.

Whereas it was previously believed that they were largely outcompeted but still contributed to the genome of modern Scandinavians through female-mediated admixture, that no longer seems to be the most likely scenario. Rather, the IBD-based clustering analysis in Allentoft’s new study very much makes it look like they may have gone extinct as a population. Interestingly, Allentoft and his team used Bayesian modelling to to determine how long the typical Anatolian farmer-related ancestry remained the dominant type of ancestry in Denmark and estimated that it took approximately 800-1000 years between the first appearance of Anatolian farmer-related ancestry and the appearance of steppe-related ancestry in Denmark, and that the Anatolian-related ancestry remained dominant for less than 50 generations. This, of course, signals just how much of an abrupt ending the EEF-derived culture in Denmark faced.

If we look at the Y-DNA haplogroups, the Funnelbeaker culture is now very well-sampled and shows almost no variety. The Funnelbeaker males in Scandinavia carried almost exclusively Y-DNA haplogroup I2, with a small minority of haplogroup Q1b (two samples) and R1b (one sample, likely subclade R1b-V88 or R1b-Y13200). As mentioned earlier in this article, haplogroup I2 is very rare among present-day Scandinavians, and has not been found among Bronze Age Scandinavians, either.

Contrary to what one might think about the Q1b and R1b Funnelbeaker samples, they are not autosomal outliers, nor do they derive from SHGs. They most likely represent the diversity of hunter-gatherer lineages present in Germany before the FBC expanded into southern Scandinavia. This should not come as a shock to anyone familiar with Y-DNA diversity in Middle Neolithic Germany, as multiple HG-rich Neolithic farmer samples with R1b have already been found among them. It is important to keep in mind, however, that none of those pre-steppe R1b lineages survived the expansion of the Corded Ware and Bell Beaker cultures, from which the steppe-derived R1b-M269 that is the most common haplogroup in Western Europe today comes from.

What we can gather from the Y-DNA results of the Funnelbeaker culture samples is that the Funnelbeaker males largely disappeared, which makes all the more sense when you consider that the FBC autosomal signal also appears to have disappeared. Courtesies of the Battle Axe and Pitted Ware cultures, most likely.

Despite the fact that Allentoft et al. 2022 states that there is no FBC ancestry in later Scandinavian samples, some more data is still needed before we can say with full certainty that there was no contribution at all from the people of that culture into the gene pool of the later inhabitants of Scandinavia. Even so, any potential contribution is likely to be fairly small.

Scandinavian Y-DNA haplogroups

Since we will be talking more about the most common Scandinavian patrilineages later in this article, let us first rehearse some basic knowledge about them. The three most common Y-DNA haplogroups among present-day Scandinavians are the following:

• I1-M253 (TMRCA 4600 YBP): Haplogroup I1 is present at a frequency of 37% among Norwegians (Dupuy et al. 2006, Genet et al. 2004), 37% among Swedes (Lappalainen et al. 2008, Genet et al. 2004), and 35% among Danes (Genet et al. 2004). I1 is also present at high frequencies among Icelanders, Frisians in the Netherlands, in Western Finland, the eastern parts of England, and in Northern Germany.
• R1b-U106 (TMRCA 4600 YBP): Haplogroup R1b is present at a frequency of 31.3% among Norwegians (Dupuy et al. 2006), 23.6% among Swedes (Karlsson et al. 2006), and 36.1% among Danes (Sanchez et al. 2004). The plurality of R1b lineages among Scandinavians belong to the R1b-U106 sublineage, with the second most common sublineage being R1b-P312. In Europe, R1b-U106 reaches its peak in frequency among Frisians in the Netherlands.
• R1a-Z284 (TMRCA 4200 YBP): Haplogroup R1a is present at a frequency of 26.3% among Norwegians (Dupuy et al. 2006), 24.4% among Swedes (Lappalainen et al. 2008), and 16.5% among Danes (Sanchez et al. 2004). Most of the R1a in Scandinavia is of the Z284 branch, descended from the Battle Axe culture. In Europe, R1a-Z284 reaches its peak in the northern, northeastern and northwestern parts of Norway and Sweden.

Three Nordic clusters become one

Allentoft and colleagues investigated the fine-scale genetic structure in southern Scandinavia after the arrival of steppe-related ancestry into the region. Broadly speaking, the genetic structure remained roughly the same (with some minor shifts in ancestry) which indicates genetic stability, but three different clusters with their own fine-scale genetic signatures and dominant Y-DNA lineages were identified. The paternal lineages found among them are the same Y-DNA lineages that are found among present-day Scandinavians, as well as in ancient DNA from Bronze Age, Iron Age and Viking Age Scandinavians.

Scandinavia_4600BP_3800BP cluster

As we know, the steppe-derived Corded Ware culture entered Scandinavia around 2900 BC in the form of the Battle Axe culture and introduced steppe ancestry and Y-DNA haplogroup R1a. More specifically, they carried subclades of R1a that are ancestral to the typically Scandinavian R1a-Z284 branch that came to become the most common subclade of R1a in Scandinavia.

While Battle Axe groups rich in R1a-Z284 dominated Scandinavia shortly after their arrival during the Scandinavian Middle Neolithic phase III-V, the Battle Axe people eventually became more peripheral, likely as a result of the expansion and dominance of groups rich in I1 and R1b-U106. Nowadays, R1a-Z284 is the third most common haplogroup in Scandinavia, but its distribution is shifted towards North-Central and Northern Sweden and, in particular, North-Central and Northern Norway.

In the new study, Allentoft calls the R1a-rich Battle Axe-associated cluster “Scandinavia_4600BP_3800BP”.

Scandinavia_4200BP_3200BP cluster

Moving on, the steppe-derived Single Grave culture prospered in Denmark starting around 2800 BC, and the males of the SGC carried Y-DNA haplogroup R1b-M269 at a very high frequency. They likely belonged to subclades of R1b-U106 and R1b-P312 that are common among modern Scandinavians, as well as some lineages that ended up having less success, such as R1b-V1636. However, it does appear that the Scandinavian males with R1b-U106 differed from the ones with P312 in terms of autosomal ancestry, the former being more Northern-shifted and belonging to what the study describes as the Scandinavia_4200BP_3200BP cluster. See the following excerpt from the study:

“Interestingly, more fine-scale sub-haplogroup placements of those individuals revealed that Y chromosome lineages distinguished samples from distinct genetic clusters inferred from autosomal IBD sharing (Fig. S3b.6, S3b.7). In particular, individuals associated with the Scandinavian cluster Scandinavia_4200BP_3200BP were all placed within the sub-haplogroup R1b1a1b1a1a1 (R1b-U106), whereas the two Scandinavian males associated with the Western European cluster Europe_4500BP_2000BP were placed within R1b1a1b1a1a2 (R1b-P312) (Fig. 928 S3b.7)”.

The P312 males on the other hand, seemingly belong to a group that clusters with roughly contemporaneous Western European samples. The P312 males more closely resembled Western European Bell Beaker individuals. Thus, it is not certain that the R1b-U106 and R1b-P312 lineages in Scandinavia were initially associated with the same type of autosomal profile, or that the two lineages first arrived in Scandinavia with the same culture. More samples spanning a longer duration of time are needed before we can truly know with certainty.

For now, the evidence available suggests that the Late Neolithic and Bronze Age Scandinavian R1b-U106 samples found in this study and older studies were associated with the cluster dubbed “Scandinavia_4200BP_3200BP”. R1b-U106 is the most common subclade of R1b in Scandinavia, and the second most common haplogroup overall. The bulk of it most likely arrived from Denmark to Sweden and Norway with Single Grave-derived Late Neolithic population movements, who in turn may have picked up some additional hunter-gatherer and EEF ancestry while passing through Zealand. While ancient DNA shows that R1b-U106 in Scandinavia had started increasing in frequency as early as the Late Neolithic, it is possible that much like I1, it experienced a more significant founder effect in the Nordic Bronze Age. It is very likely that both lineages became associated with prominent elite clans of the Nordic Bronze Age.

Scandinavia_4000BP_3000BP

The discovery of this third and final Scandinavian cluster is hugely significant and brings us one step closer to truly understanding the population dynamics of Scandinavia at the time when the Bronze Age was right around the corner. As can be seen in the PC2 in figure 8 of the extended data in the study, this cluster was deeply divergent and represents a type of genetic drift unique to Scandinavians and, by extension, later Germanic tribes. The Y-chromosome haplogroups of the men belonging to this group associated with the 4000BP_3000BP cluster that were published in Allentoft et al. 2022 were mainly haplogroup I1 (in fact, all of the 4000BP_3000BP-cluster samples from this study carried I1 except for one, who carried R1a). The following excerpt from the study is of great interest:

“a final stage from c. 3,800 BP onwards, where a distinct cluster of Scandinavian individuals dominated by males with I1 Y-haplogroups appears (Extended Data Fig. 8E). Using individuals associated with this cluster (Scandinavia_4000BP_3000BP) as sources in supervised ancestry modelling (see “postBA”, Extended Data Fig. 4), we find that it forms the predominant source for later Iron- and Viking Age Scandinavians, as well as ancient European groups outside Scandinavia who have a documented Scandinavian or Germanic association (e.g., Anglo-Saxons, Goths; Extended Data Fig. 4).”

“Y-chromosome haplogroup I1 is one of the dominant haplogroups in present-day Scandinavians, and we document its earliest occurrence in a ~4,000-

year-old individual from Falköping in southern Sweden (NEO220). The rapid expansion of this haplogroup and associated genome-wide ancestry in the early Nordic Bronze Age indicates a considerable reproductive advantage of individuals associated with this cluster over the preceding groups across large parts of Scandinavia.

Of course, this data is highly relevant and helpful when one wants to understand later Germanic tribes and their expansions.

It is likely also from a population similar to the Scandinavia_4000BP_3000BP cluster that modern Finnish people get most of their Y-DNA I1 from. Finns carry I1 at a frequency of roughly 28% nationally, reaching up to 50% in Satakunta in Western Finland. Aside from the Y-chromosome, it is probably also from a Scandinavia_4000BP_3000BP-like source that Finns get most of their Germanic autosomal ancestry.

It is worth mentioning that, while the 4000BP_3000BP cluster samples from Falköping had a very high frequency of I1, autosomally similar populations were likely distributed around Scania in Sweden and Zealand in Denmark. While these populations probably also had a high frequency of I1, they would also have carried other haplogroups.

Middle Don River foragers and Western-Steppe-Herder ancestry

Allentoft’s new study has brought us new samples from the Middle Don region, dated to about 7300 years ago. Interestingly, the Middle Don hunter-gatherers were found to derive the bulk of their ancestry from a local EHG:WHG-mixed hunter-gatherer population, but they were also found to have 20-30% of their ancestry derived from a CHG (Caucasus-Hunter-Gatherer) source. Furthermore, the Don river HGs were found to have contributed a significant amount of ancestry to later steppe individuals. In addition to that, the researchers model Western-Steppe-Herders (WSH) as deriving approximately 65% of their ancestry from Middle Don hunter-gatherers, and the remaining 35% from additional CHG-related admixture.

So, this mix of Middle Don foragers and Caucasus foragers is how Allentoft and colleagues model Steppe ancestry, and it is more or less the same as how steppe ancestry has been modelled in earlier studies. But what does this have to do with Scandinavia, again? Well, we know that Scandinavians have a very substantial amount of steppe ancestry. Allentoft et al. 2022 is largely in line with earlier studies in this regard, and the study provides no surprises in that sense. See the excerpt from the study below:

“Neolithic Anatolian-related farmer ancestry is concentrated around the Mediterranean basin, with high levels in southern Europe, the Near East, and North Africa, including the Horn of Africa, but is less frequent in Northern Europe. This is in direct contrast to the Steppe-related ancestry, which is found in high levels in northern Europe, peaking in Ireland, Iceland, Norway, and Sweden, but decreases further south. “

It is likely that a source population similar to the Middle Don foragers contributed to populations in the forest steppe, thus potentially being relevant for the formation of the early Corded Ware culture genetic profile, which in turn is highly relevant in regards to the linguistic spread of Late-Proto-Indo-European (LPIE).

The giants of the Nordic Late Neolithic

As we know, Scandinavia had seen a coexistence of the agropastoralist people of the Battle Axe culture and the hunter-fisher people of the Pitted Ware culture. After the final phase of the Middle Neolithic in Scandinavia, the Nordic Late Neolithic (roughly 2400BC-1800BC) followed and significant changes took place in the region. This period is characterized by a merging of old customs into what became a more unified, dominant Scandinavian culture. This seemingly dramatic period brought an influx of flint daggers, often left as grave goods, and one of the things that makes it so interesting is that it also brought with it new Bell Beaker-related customs from Northern Jutland in what is now Denmark, that merged with the existing customs and resulted in a fusion of previous Scandinavian cultural expressions and practices into something far more unified. Not to mention, some skeletons of unusually tall stature.

The transition from the Middle Neolithic (Battle Axe culture, late Funnelbeaker, Pitted Ware culture) to the Late Neolithic in Scandinavia was a rapid, dynamic process. In Danish and Swedish archaeology, the period is often called the Dagger Period, named after the flint daggers that spread along the movement of people during said period. There were two main areas of production, Jutland in what is now Denmark, as well as Scania/Skåne in what is now southern Sweden.

The Nordic Late Neolithic is sometimes also called the Stone Cist Period, named after the stone cist burials that became very common during the Late Neoltithic. Interestingly, older megalithic graves from the Middle Neolithic era were sometimes destroyed, only to be converted into stone cists. Other times, the megalith graves would only be slightly modified and then re-used. It is unknown why the Late Neolithic people of Scandinavia did this, but it may have been a way to mark their dominance against the preceding Middle Neolithic FBC people. Or, it could have served as a way to construct continuity, even where there was none.

During this period in Scandinavian archaeology, the inland-based agropastoralism of the Battle Axe culture was combined with the more hierarchical social elements of the Single Grave culture. Coastal traditions became increasingly important in the Late Neolithic, some of them strikingly similar to those already practiced by the hunter-gatherers of the Pitted Ware culture. The flint weapons that are often found in the male Stone Cist burials may have been connected to an emerging warrior identity that would continue into the Nordic Bronze Age and well beyond it.

While places like Zealand in Denmark and Gotland in Sweden had remained somewhat culturally isolated from the rest of Scandinavia during the Middle Neolithic, they were incorporated into a shared Nordic cultural sphere during the Scandinavian Late Neolithic. The importance of maritime networks during the transition from MN to LN must be emphasized, and some academics have even suggested that it was the maritime network linking the different regions of Scandinavia together that was also responsible for the formation of a shared language. Likewise, the remarkably unified Scandinavian cultural sphere during the LN cannot be stressed enough. It may have been the case that the Late Neolithic people finally achieved order in a region that had previously been quite chaotic, one in which Europe’s northernmost border of agriculture ended and the last bastion of hunter-gathering began, not to mention one in which warlike pastoralists roamed the inlands. Under the new Late Neolithic order, the old customs were all integrated and unified.

Getting back to the tall stature of Nordic LN skeletal remains that was briefly mentioned earlier, they can tell us quite a bit about the lifestyle, health and diet of the LN people of Scandinavia. Anna Tornberg at the University of Lund in Sweden writes elaborately on this in her 2015 report “A tale of the tall:A short report on stature in Late Neolithic–Early Bronze Age southern Scandinavia” in which she not only reviews older data from Bennike 1985, Petersen 2005 and Gejvall 1963 but also goes through some more recently gathered data from her own 2013 study.

The skeletal remains from the Nordic LN in Southern Scandinavia are generally fairly well-preserved and great in numbers, allowing for a representative sampling of their stature and overall health. For example, Gejvall 1963 measured the male skeletons at a Late Neolithic site in Sweden and found an average height of 181.4 centimeters. That is roughly the same as the modern average male height in Scandinavian countries, but bearing in mind that the average male height in LN and Bronze Age Europe was ∼167 cm (L. Cox et al. (2019), the Nordic Late Neolithic peoples must have appeared nothing less than gigantic. Additionally, Bennike 1985 and Tornberg’s earlier studies found a slightly lower, but still very high average height, roughly 10-12 centimeters above the average male height in the rest of Europe during that time.

It is worth noting that many of the sites and skeletons featured in the studies about stature, diet and health in the Late Neolithic such as Abbekås, Öllsjö, Ängamöllan and others have now also been featured in studies about ancient DNA, most recently Allentoft et al 2022 but also older studies. Fascinatingly, Allentoft et al. 2022 predicted the polygenic height scores of a large number of ancient Scandinavian samples, and they align remarkably well with what the skeletal material tells us.

In Tornberg 2013, the extraordinary stature and excellent health of the people of the Nordic Late Neolithic is explained by good dietary conditions and good living conditions. However, the study also remarks that the Late Neolithic folks consumed more products derived from agriculture and had a higher rate of dental caries, which should according to some theories be associated with a lower stature and less physical robustness. Tornberg 2015 gives a satisfying explanation, and concludes that the high lactase persistence among Late Neolithic and Bronze Age Scandinavians and by extension, the ability to digest milk, may have been what made them so tall. That is quite a logical conclusion as the protein, Vitamin D and calcium provided by the dairy products they consumed would most likely have been more than enough to combat the suboptimal Neolithic grain-rich diet which was often low in protein and animal fats.

Interestingly, Tornberg 2013 also highlights the fact that many of the Late Neolithic Scandinavian skeletal remains show signs of serious skeletal trauma as a result of violence. Intrapersonal violence appears to have been common – this was not a peaceful time.

But enough about their height and dental health. Why are the Late Neolithic Scandinavians worth talking about, and what happened to them?

Simply put, to understand the Nordic Bronze Age, we must understand the Nordic Late Neolithic. That is because the people of the former culture were predominantly descended from the latter. The Nordic Bronze Age, in turn, is massively important to the formation of what became the Germanic tribes. In terms of DNA, the most striking thing about the Nordic LN is that it is during that period that we find the earliest combined presence of haplogroups I1 and R1b-U106 (samples oll009 and RISE98, more recently also samples NEO220 and NEO875) at roughly the same time, in roughly the same place. Both lineages are later found amongst practically all Migration Period Germanic samples. The Nordic LN also happens to be where we find the earliest I1-DF29, a branch that essentially all (>99%) later I1 men descend directly from. So strictly from a Y-chromosomal point of view, it is a period of huge significance.

Additionally, the Nordic_LN autosomal profile contributed most of the ancestry to the people of the Nordic Bronze Age.

An interesting tidbit about the Nordic_LN and Nordic_BA samples is that some very substantial WHG ancestry is observable in their genomes. Much of that WHG ancestry is unlikely to be local to Sweden and Norway, but was rather probably picked up somewhere in Denmark by Single Grave-derived groups before they went further north. Zealand is one such area where the Single Grave people may have picked up a lot of WHG ancestry from local groups.

Below is a simple ancestry model for three Nordic Late Neolithic samples. Two males (RISE179 with Y-DNA I1 and RISE98 with Y-DNA R1b-U106) and one female (RISE97).

In their genomes, the hunter-gatherer-rich ancestry that may have been derived in part from somewhere in Denmark but also from somewhere in Sweden makes itself known.

Present-day Scandinavians, much like their Late Neolithic ancestors, also have high levels of WHG-related ancestry, as can be seen in the model below:

The origin of haplogroup I1-M253

Haplogroup I1, also known as I-M253, has puzzled experts and amateurs alike for a long time. Why was it so rare prior to the Nordic Bronze Age? How did it become the dominant Y-DNA haplogroup in Scandinavia? Is it native to Scandinavia, or did it get there far later than most would assume? These are questions that, with the help of old and new studies, are answerable. At least some of them.

I1 has an estimated TMRCA (Time To Most Recent Common Ancestor) of 4600 ybp, or 2600 BC. What this means is that all living I1 men descend from just one man who lived roughly 4600 years ago. The haplogroup has a dramatic, shallow phylogenetic structure, which strongly suggests that it was heavily bottlenecked before it expanded rapidly. Furthermore, it has an exceptional >310 unique mutations, a telltale sign of it being in a severely bottlenecked state before its explosive expansion during the Bronze Age, Iron Age and Viking Age. The phylogenetic tree below further illustrates the shallow phylogeny of the I1 haplogroup, which is a strong indicator of a rapid, recent growth:

While I1 itself is 27,000 years old, it underwent a crash in terms of frequency and subclade diversity to the point where it was almost extinct. This long bottleneck lasted approximately between 25,000 BC and 2600 BC. During the bottleneck event, I1 must have been limited to a remarkably small group of men, perhaps at times even a single man. This is visible in the phylogenetic structure of I1, and has led to much speculation about where I1 managed to survive for so long before finally making a “comeback” and becoming a common Y-DNA haplogroup.

Sometimes, when an ancient sample is derived for some of the SNPs that define I1, but also ancestral for many other SNPs that define the haplogroup, that sample is labelled I-Z2699*/pre-I1. What this means is that the sample represents a position in the phylogenetic tree that can be described as being between I and I1, essentially on its way to developing into what we now know as I1. It was from one of those pre-I1 samples that the progenitor of all later I1 descended before that man ended the genetic bottleneck.

So far, the only pre-I1/I-Z2699* found anywhere in Northern Europe is the Mesolithic Scandinavian-Hunter-Gatherer sample SF11 (Stora Förvar 11) from what is now Gotland in Sweden. SF11 is derived for some of the SNPs that define I1, but also ancestral for some of them. In order to know with certainty that the progenitor of later I1 men had his deep origins in a SHG population, more pre-I1 samples from such a population is needed. Until then, we can only guess.

After SF11 who is dated to around 6500 BC, there is a large gap in time where no I1 samples are found in Scandinavia. Almost all of the Mesolithic samples from Scandinavia belong to haplogroup I2, and samples from Middle Neolithic Scandinavia (Funnelbeaker, Battle Axe, Pitted Ware) have yielded zero samples belonging to haplogroup I1. It is only in Late Neolithic Scandinavia, right on the verge of the Bronze Age, that the first I1 samples are found and the bottleneck appears to have ended. At the point where we find these early I1 carriers in the ancient DNA record, near Falköping in Sweden, they are already heavily admixed with Corded Ware-derived populations and derive the bulk of their ancestry from the steppe. Therefore, it is hard to figure out how and when I1-carriers first encountered and mixed with Corded Ware peoples. Future samples may help in solving this mystery. Until then, based on the data, there are three possible and realistic scenarios:

• I1 was present as a very rare haplogroup among Pitted Ware hunter-gatherers in Scandinavia and encountered the Battle Axe people and mixed with them (or mixed with the FBC people before encountering the BAC)
• I1 was picked up and assimilated by Single Grave people somewhere in Zealand or Scania
• I1 was brought by the Corded Ware-derived groups from somewhere in Poland where it originally got into the Corded Ware population from a Globular Amphora source

The first scenario has some support (although evidence is still weak) if we take into account the single pre-I1 SHG sample. The Globular Amphora culture, however, is very well-sampled, and has only yielded haplogroup I2. Similarly, we now have plenty of samples from the local people in Zealand during the Middle Neolithic, thanks to Allentoft’s new study, with no samples of I1 found among them.

Upcoming samples from Mesolithic and Middle Neolithic Scandinavia may be able to tell us more about the ultimate origin of haplogroup I1. In the meantime, thanks to Allentoft’s new study, we know that I1 underwent a sharp, dramatic rise in frequency during the Nordic Bronze Age and spread when Germanic tribes migrated out of Scandinavia into Continental Europe. Until we get samples from hunter-gatherers from more remote regions of Mesolithic and Neolithic Scandinavia, the deep origins of I1 will likely remain shrouded in mystery.

Interestingly, studies like Patterson et al. 2022, Papac et al. 2021 and Rohrlach et al. 2021 found zero samples of haplogroup I1 in Bronze Age Continental Europe, despite sequencing many hundreds of samples from a very wide range of sites and countries all across Continental Europe. This, of course, is indicative of a rapid Late Bronze Age-Early Iron Age dispersal from southern Scandinavia into central Europe.

Sun, swords and bronze: new Nordic Bronze Age samples

Prior to this new data, only a handful of samples from the Nordic Bronze Age had been published. Despite that, they were seen by many as a sign of things to come. Now that far more samples are available, it does indeed seem like the NBA samples from Allentoft et al. 2015 were quite indicative of those that would become available in Allentoft et al. 2022.

In terms of uniparentals, the new samples make it fairly clear that the people of the NBA had direct paternal descent from the Nordic Late Neolithic. The new Bronze Age samples from Denmark yielded five samples of I1, and 2 samples of R1b-U106. If we add the old samples (Allentoft et al. 2015) from the Nordic Bronze Age (three samples of R1b and three samples of I1) that gives us a total of eight I1 samples and five R1b samples for the Nordic Bronze Age period. So far, Y-DNA R1a appears to be absent from the Nordic Bronze Age samples, supporting our view that it became more peripheral during the Nordic Late Neolithic and, during the Bronze Age, likely was not present in significant frequencies further south than Götaland in Sweden.

Autosomally, the Nordic Bronze Age samples display a genetic profile that by all means looks like it provided the bulk of ancestry not only to Iron Age Scandinavians, but also many other Germanic tribes, as stated by the authors of the study. That is not a shocking revelation by any means, but it is still a refreshingly relevant piece to the puzzle that is the formation of Proto-Germanic peoples. In the Nordic Bronze Age, Scandinavia_4000BP_3000BP-related genome-wide ancestry spread rapidly.

“Each of these countries was like a mighty hive, which, by the vigour of propagation and health of climate, growing too full of people, threw out some new swarm at certain periods of time, that took wing, and sought out some new abode, expelling or subduing the old inhabitants, and seating themselves in their rooms”

–  Sir William Temple, 1st Baronet, about the migration of Germanic tribes out of Scandinavia

During the periods following the Nordic Bronze Age, people from Scandinavia migrated south in great numbers. This is very visible in the aDNA record, in particular when looking at Y-DNA lineages, in which we see that haplogroups like I1 were completely absent in Continental Bronze Age Europe, only for a large amount of I1 samples to be found in the Iron Age in Germanic burial contexts. Needless to say, the I1 samples are usually accompanied by R1b-U106 and, in the case of Lombards and Goths, also by the Scandinavian R1a-Z284 lineage. Gretzinger et al. (2022) also found R1a-Z284 in samples from Anglo-Saxon England at the sites of Sedgeford and Polhill, as well as in Migration Period Germany at the site of Drantum.

These Germanic tribes brought with them a genetic profile that was more shifted towards steppe and forager ancestry than, for example, the more EEF-rich Celtic and pre-Celtic inhabitants of Central Europe. It is worth mentioning that they did not differ significantly from the Bell Beaker-derived pre-Germanic inhabitants of the Netherlands from an autosomal point of view. The uniparentals, however, are an important part of the equation and tells a different story. One of people expanding from the north to the south, and a clear case of a gradual shift towards a higher frequency of Scandinavian-derived patrilineages and genome-wide ancestry in conjunction with the arrival of various Germanic groups.

Moving on, East Germanics migrated to what is now Poland and settled there, laying the foundations for what would become the Wielbark culture. Y-DNA haplogroups and autosomal results from the Wielbark culture (as well as from later Goths) strongly indicates that East Germanic tribes originated in Scandinavia, likely in the eastern parts of Götaland, as well as Gotland in what is now Sweden. It is possible that they also had a significant presence on the island of Bornholm in Denmark.

In the Wielbark culture, stone circles typical of LBA and early Iron Age Scandinavia start to appear in conjunction with the arrival of Scandinavian-like ancestry. In the case of samples from Kowalewko in Poland, we can also observe groups rich in Scandinavian paternal lineages. Genomes from a cemetery associated with early Goths and Gepids at the site of Weklice in Poland were sequenced in Antonio et al. (2022) and, much like the Goths from Kowalewko, the majority of them were genetically Scandinavian-like. In the PCA below, it can be observed how the Goth samples cluster around present-day Scandinavians. Generally, they are the closest to present-day Swedes, but also have a general affinity towards the rest of the Scandinavians.

The unmistakingly Scandinavian genetic profile of the Goths offers some serious vindication to the writings of the 6^th century historian Jordanes, who himself was of Gothic origin. Jordanes wrote that the Goths migrated from their homeland, known as Scandza, to Gothiscandza. The former is often associated with Scandinavia, whereas the latter is usually associated with the Wielbark culture in Poland.

Furthermore, haplogroup I1-A11380 being found among the Weklice Goths (sample R10636) is a fascinating discovery, as it confirms what many have suspected: that the A11380 subclade found in moderate frequencies in the Balkans in modern times is a legacy of the East Germanic tribes that settled there. More samples will allow us to elaborate on that, as currently the only Gothic samples from Serbia have yielded Y-DNA haplogroup I1-Z140 (Olalde et al.2021),which is a different subclade.

Getting back to the Nordic Bronze Age, one of the most significant findings of Allentoft et al. 2022 is that the transition from the Late Neolithic to the Bronze Age was not accompanied by a massive gene flow from the south. Additionally, the current data indicates that the transition to the Early Bronze Age did not introduce a new elite. Instead, the elites were seemingly locals of Late Neolithic origin.

An example of this is if we compare the Late Neolithic elite/high-status sample RISE179 and Nordic Bronze Age high-status sample RISE175, it is evident that their ancestry is derived mainly from the same sources. It should be mentioned, however, that some high-status Late Bronze Age samples from Scandinavia may be more southern-shifted, as it was common for them to marry women from the Urnfield culture.

RISE179 was buried in a gallery grave covered by a large mound in what is now southern Sweden. His grave goods included a flint spearhead and a bronze dagger. RISE179 is dated to 2010-1776 BC. If we compare him to NBA sample RISE175, dated to 1395-1132 BC, we can see some very significant similarities. RISE175 was buried in a mound too, at the same site, no less (but hundreds of years later), with valuable grave goods, including a horse-head razor made of bronze. Interestingly, both men carried haplogroup I1, suggesting no discontinuity in the paternal lines of the ruling clans in southern Scandinavia during the Nordic Bronze Age.

With all that said, more samples are still needed (as always). This is especially true for Norway, since Allentoft’s new paper only included a single sample from Norway, and that sample is from the Mesolithic. Ideally, we would get a decent sample size from Late Neolithic and Bronze Age Norway, so that we can fully understand the population dynamics in all parts of LN-BA Scandinavia. However, the data from Allentoft 2022 is highly illuminating and goes a long way.

Summary

Scandinavia underwent substantial genetic turnovers during the transition from late Mesolithic to early Neolithic periods as well as during the introduction of steppe-related ancestry with the arrival of the Corded Ware-derived Battle Axe culture. Danish Mesolithic ancestry saw an almost complete replacement, and so did the Neolithic Funnelbeaker-related ancestry later on.

During the Late Neolithic, the core population that would form the basis for later Scandinavian populations had emerged, with its own genetic drift and the three major Scandinavian Y-DNA haplogroups already present at high frequencies. The rest of Europe would later see substantial migrations and invasions from the Nordic Bronze Age-derived Scandinavian populations during the many Germanic expansions.

The new data has filled many of the previously existing gaps in data that made the population dynamics of the region more difficult to understand. However, more data is required, especially when taking into account the relative lack of samples from Norway. With at least four more Scandinavian aDNA studies currently underway, the odds of getting an even more clear picture of things are looking excellent. The most undersampled regions and periods of significance include Early-to-Late Neolithic and Bronze Age Norway as well as Mesolithic Scania in Sweden. In order to truly get a representative sampling, Bronze Age samples from further to the north in Sweden could also be helpful. Furthermore, more samples from high-status burials in the Scania-Zealand region would not hurt. Hopefully, we will see all of this, and more, in the near future.

References

Allentoft et al. (2022): Population Genomics of Stone Age Eurasia https://www.biorxiv.org/content/10.1101/2022.05.04.490594v2

Patterson et al. (2021): Large-scale migration into Britain during the Middle to Late Bronze Age https://www.nature.com/articles/s41586-021-04287-4

Papac et al. (2021): Dynamic changes in genomic and social structures in third millennium BCE central Europe https://www.science.org/doi/10.1126/sciadv.abi6941

Blank et al. (2018): New Perspectives on the Late Neolithic of South-Western Sweden. An Interdisciplinary Investigation of the Gallery Grave Falköping Stad 5 https://www.researchgate.net/publication/323559918_New_Perspectives_on_the_Late_Neolithic_of_South-Western_Sweden_An_Interdisciplinary_Investigation_of_the_Gallery_Grave_Falkoping_Stad_5

Manninen et al. (2021): First encounters in the north: cultural diversity and gene flow in Early Mesolithic Scandinavia https://www.researchgate.net/publication/348790878_First_encounters_in_the_north_cultural_diversity_and_gene_flow_in_Early_Mesolithic_Scandinavia

Tornberg, Anna, Stockholm University (2013): Diet, Health and Agriculture: The Late Neolithic-Early Bronze Age Example of Abbekås https://www.researchgate.net/publication/329611949_Diet_Health_and_Agriculture_The_Late_Neolithic-Early_Bronze_Age_Example_of_Abbekas

Rohrlach et al. (2021): Using Y-chromosome capture enrichment to resolve haplogroup H2 shows new evidence for a two-path Neolithic expansion to Western Europe https://www.nature.com/articles/s41598-021-94491-z

L. Cox et al. (2019): Genetic contributions to variation in human stature in prehistoric Europe https://www.pnas.org/doi/10.1073/pnas.1910606116

Tornberg, Anna, Lund University, (2015): A tale of the tall : A short report on stature in Late Neolithic–Early Bronze Age southern Scandinavia https://www.lunduniversity.lu.se/lup/publication/b3f18a1b-5cbc-458d-978c-16dddf9a8710

Olalde et al. (2021): Cosmopolitanism at the Roman Danubian Frontier, Slavic Migrations, and the Genomic Formation of Modern Balkan Peoples https://www.biorxiv.org/content/10.1101/2021.08.30.458211v1

Bergerbrant et al. (2015): New Perspectives on the Bronze Age Proceedings of the 13th Nordic Bronze Age Symposium held in Gothenburg 9th to 13th June 2015 https://gup.ub.gu.se/file/207192

Dupuy et al. (2006): Geographical heterogeneity of Y-chromosomal lineages in Norway https://pubmed.ncbi.nlm.nih.gov/16337760/

Lappalainen et al. (2008): Migration Waves to the Baltic Sea Region https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1469-1809.2007.00429.x

Genet et al. (2004): Phylogeography of Y-Chromosome Haplogroup I Reveals Distinct Domains of Prehistoric Gene Flow in Europe https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1181996/

Karlsson et al. (2006):Y-chromosome diversity in Sweden – A long-time perspective https://www.nature.com/articles/5201651

Sanchez et al. (2004):Y chromosome SNP haplogroups in Danes, Greenlanders and Somalis https://www.sciencedirect.com/science/article/abs/pii/S0531513103016352

Zenczak et al. (2017): Y-chromosome haplogroup assignment through next generation sequencing of enriched ancient DNA libraries https://www.academia.edu/33791135

Jordanes (551 AD): Getica: The Origin and Deeds of the Goths.

Batini et al. (2015): Large-scale recent expansion of European patrilineages shown by population resequencing https://www.nature.com/articles/ncomms8152

Egfjord et al. (2021): Genomic Steppe ancestry in skeletons from the Neolithic Single Grave Culture in Denmark https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0244872

Gretzinger et al. (2022): The Anglo-Saxon migration and the formation of the early English gene pool https://www.nature.com/articles/s41586-022-05247-2

© Genomic Atlas 2022