Next stop Iceland?

This post is about pre-Viking settlement in the North Atlantic, and here specifically the Faroe Islands. I originally stubbed this in September 2013, when I came across a relevant web report on Melissa Snell’s medieval history blog at, but since that time it seems that they’ve stopped keeping archives over there, so that post is now gone. Actually this is good news, because it forced me to web-search out other reports, and the obvious one, on Nature World News, also includes a link to the actual article behind it all, so now I can write you something informed.1

View from the south-eastern extremity of Sandoy in the Faroe Islands, looking south down the chain

View from the south-eastern extremity of Sandoy in the Faroe Islands, looking south down the chain, and shamelessly plundered from a Faroes tourism site, linked through

The story is that new research proves that the Vikings were not the first colonisers of the Faroes. This, and the fact that this was being published in an geoscience rather than a specifically archæological or historical journal, gave me some faint worries about sensationalisation of research meant to prove other things, as we have seen here many times before, but in this case, as far as I can see, it’s the real deal. The team in question were doing a low-urgency rescue dig on a coastal site in danger of erosion at Á Sondum on the island of Sandoy in 2002 and 2006. The site had been dug once before in 1994 and had then shown a Viking longhouse at the bottom of its occupation sequence (which subsequently went all the way up to the 19th century, always on the same spot, not least, as the article says, because there just aren’t that many sensible places to put settlements on the Faroes).2 This time they located the long-house’s hearth, and then they kept going and found more beneath. It wasn’t much more, just peat ash with carbonised barley grains in it, but this is significant for two reasons: firstly, barley isn’t native to the Faroes, and secondly, peat-burning requires usually months of cutting, stacking and drying, and thus implies prolonged settlement, not just drop-in visits. And they duly had radio-carbon analysis done on the barley grains and got the figures you see below.

Calibration graph and stratigraphical context of the radio-carbon samples from Á Sondum, Sandoy

From Church, Arge, Edwards, Ascough, Bond, Cook, Dockrill, Dougmore, McGovern, Nesbitt & Simpson, “Vikings were not the first colonizers”, p. 230, larger version linked through

There are some things I don’t fully understand about this, mainly because to get these dates they used not individual samples but a combined sample, whose results were mathematically combined using a ‘weighted mean’. Well, weighted how? I feel we should be told more about this.3 Still, unless it was actually weighted outside the sample dates, which would be some strange new definition of ‘mean’ of which I was not previously aware, the results still seem fairly inarguable: two straigraphically separate episodes older than the Viking site by between half a century and three centuries. I feel that a mean critic might be able to handwave away the more recent result, given it was right under the hearth and the dates could just about overlap, but the older one is much harder to get around, which gives one that bit more confidence in the younger one.

A forbidding shoreline in South-Central Iceland

An even more forbidding shoreline in South-Central Iceland, apparently being slowly washed away by glacial outwash, as presented in Albert C. Hine, Jon Boothroyd & Dag Nummedal, “Glacial Outwash Plain Shoreline, South-Central Iceland”, Beach of the Month, Coastal Care 1st August 2013, linked through

So, OK, what does this all mean? Well, most obviously it means that someone was on Sandoy burning peat and scattering its ash on the shoreline well before the Vikings got there (and the study is extremely careful to emphasise that we have no idea who). What our authors are most excited about is that there were, previously, very small signs of pre-Viking occupation in the palaeobotanical record (which is to say, barley pollen had been found before in a hard-to-date but potentially early context4), and that having been able to pin it down with actual digging like this has big implications for other island chains where such out-of-place floral data has turned up but been dismissed. But it also made them think of what it makes me think of, which is an old question about whether anyone got to Iceland before the Vikings. There are stories about this in the texts, firstly bewildering seafaring fantasies like the Irish Navigatio Sanctio Brendani, in which Brendan and his monks find a series of fantastic islands in the Western Ocean while in search of Paradise, which they duly locate and are allowed to stop in for a bit before turning for home to spread the good word.5 Much effort has been spent on pinning actual Atlantic islands into his tale, but, to be cynical again, one might as well try the same exercise with The Voyage of the Dawn Treader, the islands are equally allegorical in both texts.6 But there is also the Íslendingabók of Ari Thorgillson, ‘The Book of the Icelanders’, in which Ari reports that the first Viking settlers on Iceland found white-clad men they called ‘papar’ there who fled before the Norsemen, leaving behind croziers and bells.7 That looks so like the kind of people St Brendan actually led that it’s usually been assumed these were Irish religious taking the spiritual exile thing unusually seriously. But there is no archæological evidence to show that Ari was doing anything more than making this up by extrapolation from stories about the Western Isles of Scotland or similar. The archæologists here are suitably cautious about fitting these older debates around their findings:

“Who were these earlier settlers, how many of them were there and where did they come from? Were they single ecclesiastical anchorites as described by Dicuil from Ireland, Late Iron Age groups of colonists from Atlantic Scotland or pre-Viking explorers from Scandinavia? Did they continue on to Iceland, or did the Faroes represent a temporary diasporal bottle-neck, requiring later Viking seafaring technology to be able to sail to Iceland and Greenland, and importantly to sail back? To answer these questions, more and better-preserved pre-Viking Faroese archaeological sites need to be identified, excavated and analysed.”8

To that, I would only say that the Vikings’ seafaring technology was perfected well before the Viking disapora started, so that might not be the significant limitation, but the Faroes radio-carbon dates are still very early and the authors’ caution here is absolutely commendable.9 What they do do, however, is establish an excellent idea of where to look for such remains in island contexts, and that gives me hope that Iceland might, indeed, be the next stop, just because it would be fun finally to have some kind of archæology to set against Ari’s stories. I’ll let the authors of the article have the last word, though, because they have bigger things in mind than my antiquarianism, and will myself just finish by saying that I wish all science publications with historical impact could manage it as well as these writers have:

“This detailed examination at the base of coastal erosion archaeological sections was successfully used to identify the first archaeological sites of Mesolithic date in the Western Isles of
Scotland. A similar palynological argument for early human settlement was proposed for the ‘invisible Mesolithic’ of the Western Isles, where small-scale clearance episodes evident in pre-Neolithic pollen sequences, coupled with a rise in microscopic charcoal, led researchers to suggest that hunter-gatherer fire ecology accounted for these disturbances, despite the lack of any Mesolithic archaeological sites in the island chain. In the Faroes and the Western Isles of Scotland, small-scale perturbations in pollen sequences were interpreted as ephemeral human occupation events in periods prior to the orthodox landnám [‘land-taking’] events attested by the substantial settlement record of structures, dating to the Viking and Neolithic periods respectively. In both cases, the palaeoecological interpretations were eventually proved to be correct on discovery of ephemeral archaeological remains of pre-landnám date at the base of large coastal erosion archaeological sections and this raises intriguing issues about perturbations in palaeoenvironmental sequences prior to orthodox landnám chronological horizons in other island systems elsewhere in the North Atlantic and across the world, as in the Pacific."10

1. Mike J. Church, Símun V. Arge, Kevin J. Edwards, Philippa L. Ascough, Julie M. Bond, Gordon T. Cook, Steve J. Dockrill, Andrew J. Dugmore, Thomas H. McGovern, Claire Nesbitt, Ian A. Simpson, “The Vikings were not the first colonizers of the Faroe Islands” in Quartnerary Science Reviews Vol. 77 (Amsterdam 2013), pp. 228-232, DOI: 10.1016/j.quascirev.2013.06.011. Assigning author order on that must have taken nearly as long as the project itself…

2. Ibid. pp. 228-230; the previous dig, led by the second author here, seems not to have been published.

3. Rather than tell us about it, they refer to two other studies, P. J. Ashmore, “Radiocarbon dating: avoiding errors by avoiding mixed samples” in Antiquity Vol. 73 (Edinburgh 1999), pp. 124-130, which you’ll notice is a vote against doing this, and Ascough, Cook & Dugmore, “North Atlantic marine 14C reservoir effects: Implications for late-Holocene chronological studies” in Quarternary Geochronology Vol. 4 (Amsterdam 2009), pp. 171-180, DOI: 10.1016/j.quageo.2008.12.002, which presumably isn’t.

4. Here they cite a whole rook of stuff, the most recent of which is K. J. Edwards & D. M. Borthwick, “The pollen content of so-called ‘ancient’ field systems in Suethuroy, Faroe Islands, and the question of cereal cultivation” in S.-A. Bengtson, P. C. Buckland, P. H. Enckell & A. M. Fosaa (edd.), Dorete – Her Book – Being a Tribute to Dorete Bloch and to Faroese Nature, Annales Societatis Scientiarum Færoensis, Supplementum LII (Tørshavn 2010), pp. 96-116.

5. I’m not sure what the preferred edition of the Navigatio is these days but in trying to find out I have found Guy Vincent (ed.), “Navigation de Saint Brendan : texte latin et traduction”,, last modified 24th October 2005 as of 26th August 2014, which will probably do. There is a good English translation in J. F. Webb (transl.), Lives of the Saints (Harmondsworth 1965), pp. 31-68, repr. in D. H. Farmer (ed./transl.) & Webb (transl.), The Age of Bede (Harmondsworth 1983), pp. 231-267 even though the text is likely from a century after Bede.

6. There are various studies on the Atlantic world implied by the text in John W. De Courcy & David Sheehy (edd.), Atlantic Visions (Dun Laoghaire 2009), which I will admit straight up I have never seen but just now pulled out of a database; any other suggestions?

7. There seems to be an edition of Íslendingabók by Guðni Jónsson online here but I have no idea how good it is; again, any better references would be appreciated. The standard translation, Siân Grønlie (transl.), Íslendingabók; Kristni Saga. The Book of the Icelanders; The Story of the Conversion (London 2006), pp. 3-34, is online here in PDF.

8. Church, Arge, Edwards, Ascough, Bond, Cook, Dockrill, Dugmore, McGovern, Nesbitt & Simpson, “Vikings were not the first colonizers”, p. 231.

9. Jan Bill, “Ships and Seamanship” in Peter Sawyer (ed.), The Oxford Illustrated History of the Vikings (Oxford 1997), pp. 182-201.

10. Church, Arge, Edwards, Ascough, Bond, Cook, Dockrill, Dugmore, McGovern, Nesbitt & Simpson, “Vikings were not the first colonizers”, p. 231.

27 responses to “Next stop Iceland?

  1. Dicuil’s De mensura orbs terrae — cited in one of your quotes — is probably a better initial textual reference point than the Navigatio, even if it’s simply another monastic text beset with problems. *If* Dicuil was referring to the Faroes (which is not certain), then his words would indicate that Insular anchorites had dwelt there for an extended period, and had apparently only recently been driven out by Scandinavians, sometime before the second quarter of the ninth century. This in itself would suggest an initial period of Norse activity half a century or more before the conventional dating of the settlement of Iceland. If anchorites were indeed known to have been going to the Faroes for some time in the decades before Dicuil wrote, then there would have to have been a certain amount of traffic, simply for a colony of celibate men to maintain itself… In which case it would be reasonable to expect less ascetic types to show an interest too. I really only mention it though because this is the only text in the history of writing that captures the rare pleasure of the lonely island-dweller, able to pick the lice from his garment under the bright sky of a northern summer night.

    • Ah good, I was hoping this might draw you in, JPG. I don’t know Dicuil as more than a name so I didn’t want to risk saying more about it, but you’ll probably be pleased to know that the article does also reference the lice!

  2. From whom were the brochs in Scotland meant to defend the inhabitants? My guess is that the easterlies that blow in May and June would have brought Norsemen long before the Vikings. I dare say that easterlies blow in the Faroes too.

    • I have to admit, I’ve tended to assume that, what with there being so many brochs, there must have also been quite a number of small-scale lords setting them up. In which case, the primary threat to them was probably each other… But your point is a good one, because it’s also been one of the counters to genetics studies trying to look at Viking settlement that the British Isles have been seeing settlement from Scandinavia pretty much throughout history.

  3. Excavations in Orkney back in the 90s (summarized in _A Persona for the northern Picts_ (1997)) showed reindeer antlers in a context below a 6th-century paved floor. Reindeer not so much with the native to Orkney.

  4. The Orkney combs are rather problematic.

    Already in the 1990s it was pointed out by several scholars that no sufficient account of the methodology for identifying the material as Scandinavian reindeer antler had been supplied, leaving open the possibility that they were made from red deer antler. More recent work by Steven Ashby has confirmed that the material is probably reindeer antler; but he also shows that none of the combs can be shown to have come from contexts predating the C9th.

  5. The standard edition of Dicuil’s De mensura orbis terrae is by J. J. Tierney and L. Bieler — published by the Dublin Institute for Advanced Studies, 1967.

  6. I too wonder how much scientists are under pressure to present every small new result as a game-changer. But I confess I do not fully understand the science, especially all the bits that rely on maths… :-), so all I can do is try to interpret what they present as their results.

    Anyway, if you’re interested, here is a snippet from a forthcoming work by yours truly, in which I try to put these new Faroese results in context (the ‘more recent paper’ referred to below is the one by Church et al):

    “There has been some debate among scientists about possible evidence in the form of early field systems and radiocarbon-dated pollen which seem to suggest cereal cultivation, and therefore some sort of human habitation, before the Viking Age. Some of the results of recent research using high-resolution pollen sampling and small-sample radiocarbon dating are ambiguous, but suggest more than one stage of human impact on the environment. There is evidence of pollen from a barley-type of cereal (Hordeum) at c. 730, and similar grains were perhaps present at 540. A more recent paper presents new radiocarbon dates from one archaeological site that place the earliest human settlement in the Faroes in the fourth to sixth centuries. Underneath a ninth-century Viking Age longhouse (dated on the basis of carbonised barley grains) on the island of Sandoy, there appear to have been two periods of small-scale human settlement, during the fourth to sixth centuries, and sixth to eighth centuries, dated on the basis of burnt patches of peat ash of anthropogenic origin, including carbonised barley grains. Previously, the earliest radiocarbon dates from Faroese archaeological sites were estimated at 690-890 and 690-970. These ranges cannot be distinguished statistically and are both consistent with a date of arrival of Scandinavians, suggested on historical grounds, of about 800.

    The new dates are now declared by the scientists to be ‘firm archaeological evidence for the human colonisation of the Faroes by people of unknown geographical and ethnic origin some 300-500 years before the large scale Viking colonisation of the 9th century’, but the actual evidence remains rather small and the authors of the recent study resort to various explanations as to why there is not more of it. While they keep looking for further evidence, and continue to refine the answers to the question of when the Faroes were first inhabited by humans, it remains clear for the moment that the major impact on the landscape came during the Viking Age. Other pollen data (and of course the archaeological, historical and linguistic evidence) support the likelihood of major human impact from about 800, on a scale different from any earlier impact.”

  7. P.S. Ari doesn’t say anything about the clothing of the papar….

  8. Allan McKinley

    We could probably do with a gradiation of terms for debates like this, as people can know of, visit, discover (which implies recording) or colonise somewhere like Faroe, so treating all human presence as the same is rather pointless. Unless you are asking the equally pointless ‘who was here first’ (so nineteenth century darling…) question.

  9. I don’t see that attempting to unpick the chronology and background of prehistorical and historical colonisation in a particular environment is asinine in the least. Indeed, I’d say it’s fundamental to understanding and challenging comfortable myths.

    • I think what I mean to indicate is my cynicism that this tells the inhabitants of that environment who they are in any sense that they don’t already carry around with them. Of course you are right that challenging that ubiquitous idea requires knowledge of the real state of affairs, but even then the fact that people have been coming to what are now the British Isles for centuries doesn’t really enable us to see our culture as any more Scandinavian or indeed less specifically Viking than before. Of course it makes a difference, in some ways, to how we see, for example, the tenth and eleventh centuries! But that’s not really what ventures like Blood of the Vikings were about, is it?

  10. Well quite. But that’s a commonplace problem for academic historians in general, isn’t it? When it comes to contemporary identity politics and perceptions of the past, it’s particularly clear that public interest is not really determined (and in many cases is not affected to any degree at all) by the activities of academics: which is precisely why they should keep on sticking their noses in to clarify (or sometimes, de-clarify) the parameters of debate, and (even better) contest one another’s ideas in public a bit. It’s not a question of plugging a particular interpretation, so much as plugging honest ways of thinking and talking about complex stuff. Isn’t it?

    What’s I think’s so interesting about the Faroes is that you have, on the face of it, a comparatively simple context into which new evidence can be slotted, and yet even here the earliest data takes us well over the lines of what’s knowable or identifiable, right under the feet of contemporary Faroese farmers. It’s rather invigorating.

    • It’s not a question of plugging a particular interpretation, so much as plugging honest ways of thinking and talking about complex stuff. Isn’t it?

      Well, if I didn’t agree with that I’d hardly be running a blog, would I? I’m all for some public contesting. The best way we can combat such things is to write comprehensible stuff that comes up well in websearches! But I don’t think that makes the logic of we-are-somehow-our-ancestors-(by-which-we-mean-whoever-lived-here-long-ago) any less empty, even as (as originally) I recognise its power to move people…

      • Sorry, but there’s something wrong in this position. I fully agree that we-are-somehow-our-ancestors-(by-which-we-mean-whoever-lived-here-long-ago) is an empty notion, but the complementary: we-are-not-our-ancestors, is also empty; so, there’s a contradiction going on here.

        The problem, imo, is that ‘true’ or ‘false’ are not relevant categories when talking about memory (the inner capability behind human identification process). Let’s not be fooled by social taboos.

        • Well, when I mock the we-are-our-ancestors position I don’t wish to deny that that which has gone before us has some part in how things are now! That would be a very nihilistic position for a historian to take. What annoys me is the idea that only things that can be seen in this way are interesting and worthy of study. So there may be something in both ancestry and non-ancestry arguments: some things that people did before us remain forces upon us, some do not and we have some agency in our own actions and choices, especially when what went before is not or cannot be known.

          I run very close to truism there, but I think, in fact, that one of the most productive debates historians of any area and time could be having at the moment, and perhaps at any time, is one on inheritance. What do we bring forward from those cultures and people who precede us? The full range from deliberate affectation through to genetic determinism could be causative but may not be. What can we choose to inherit, or imitate, and what can we not avoid doing? Who are the agents in this process? (Is it, for example, historians? If not, why not?) If we could achieve laws of history at all, this is probably the place where they would be most useful. And yes, I agree, memory is one of the factors: how far can it be accurately informed, how widely, what rôle does it play? I suspect there will be more thoughts from me on this as I bed further into this job, as one of the reasons it’s on my mind just now is that it is a big theme of the first exhibition I have to set up!

        • Allan McKinley

          That is pretty much why I was suggesting gradiated language. We can’t and probably shouldn’t try and debate these feelings of origin as history (certainly not if, like most of those of us here I guess, we tend to see labels as assumed not inherited). We should however provide a clear understanding of what the evidence says in terms outside the simpler debates about original and first.

  11. Yep, I was rather conflating that with the more general issue of contemporary ‘relevance’… I’ll come back when I’ve got something to disagree with you about.

  12. Michael Newton

    It is disappointing that people know rather more about Norse history and wayfaring than they do about Celtic (inc. Pictish) equivalents, and while the possibility of Norse navigation is automatically afforded credibility, Celtic navigation is not.

    Just because the Irish Navigatio Sanctio Brendani is a narrative that functions as a parable doesn’t mean that it isn’t also grounded in real maritime experiences. The Life of St Columba also mentions saints going out north to sea, and hoping for the protection of the Pictish king of Orkney. Here’s a quote from the (outdated) online translation by Rees: “AT another time a soldier of Christ, named Cormac, about whom we have related a few brief particulars in the first part of this book, made even a second attempt to discover a desert in the ocean. After he had gone far from the land over the boundless ocean at full sail, St. Columba, who was then staying beyond the Dorsal Ridge of Britain (Drumalban), recommended him in the following terms to King Brude, in the presence of the ruler of the Orcades (Orkneys): ‘Some of our brethren have lately set sail, and are anxious to discover a desert in the pathless sea; should they happen, after many wanderings, to come to the Orcadian islands, do thou carefully instruct this chief, whose hostages are in thy hand, that no evil befall them within his dominions.’ The saint took care to give this direction, because he knew that after a few months Cormac would arrive at the Orcades.”

    We know from contemporary Gaelic annals that Aedan mac Gabrain was conducting warfare on the Orkneys from his base in Dal Riada in c.580. Gaelic monks did operate in Orkney before the arrival of Vikings, and both Orkney and Shetland have names given in Celtic languages.

    Furthermore, there are placenames in Iceland that indicate the existence of Gaelic monks there before Viking settlement.

    So, why isn’t it a viable hypothesis for Gaels and Picts to be going to the Faroes before the Vikings?

    • I don’t think anyone has denied its possibility! To your cites I would add the Senchús fer nAlbán‘s list of ships and crews on which the king of Dál Riata could call in the mid-seventh century, for example. Brendan’s boat was very clearly imagined as being bigger than a coracle as we understand them, too, for all that it was supposedly built of hides over withies. I think that is one of the reasons the supposition of Gaelic ocean exploration is not more often made, the apparent fragility of the ships Irish hagiography describes! But probably a far more important one is teleology: by the time the historiography got going, the Faroes and Iceland were Norse, not Irish. Thus, the explanations started there…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.