Think of Vikings and you see longships, swords and raids. The part of their world that lived on fish hardly shows up in the pictures, and the written sources are strikingly quiet about it too. Archaeology can say more than the sagas: a handful of cod bones from the harbour mud of Hedeby has been matched by ancient DNA to the cod stock off the Lofoten Islands, isotope measurements sort fish by catch region, and around AD 1000 sea fishing changes visibly in parts of northern Europe. This article sets the studies side by side, says openly where the evidence is thin, and keeps findings apart from interpretation.
Ask people what Vikings ate and the answers are mead, roast meat and bread. The article What Did the Vikings Eat? clears up several of these images. Fish belongs in that picture but rarely gets taken seriously, for a simple reason: it barely appears in the famous stories. The Icelandic sagas, written down from the 13th century onward, tell of feuds, lawsuits and honour. Someone gutting cod on a beach is scenery there, not a hero.
The picture is therefore partly a problem of sources. What people do every day and what makes no story is rarely recorded. A second problem concerns preservation and is discussed below: fish bones, nets and lines survive only under favourable conditions. That we see little does not mean little happened.
Two terms first. Viking Age in research roughly means the years between about AD 800 and 1050 in Scandinavia and the regions reached from there. Zooarchaeology is the study of animal remains from excavations – bones, fish bones, scales and shells – and it is the main source of this article.
Before turning to finds, an honest look at method. Fish bones are small and fragile. If soil is removed with a trowel alone, they are missed; only washing and sieving through fine mesh brings them out. Many older excavations worked in ways that lost small herring bones, while large cattle bones are over-represented in the record. That is a known bias and has to be kept in mind with every statement about "much" or "little" fish.
Soil chemistry matters as well. Bone dissolves in acid soils and survives in lime-rich, wet or oxygen-poor deposits. The harbour mud of Hedeby, from which the cod bones described later come, is such an environment. A cemetery on sandy ground often yields nothing at all.
For this article I consulted only a limited number of sources in this round, mostly summaries and press texts about the studies rather than the full papers. Where it matters, that is stated. Figures I have not seen myself in the authors' own words are either omitted or marked as reported.
Cod, Norwegian "torsk", is a predatory fish of cold seas and belongs to the cod family, which also includes saithe, ling and haddock. Zooarchaeologists therefore often write "gadid" when a single bone cannot be assigned to a species. Herring is a small shoaling fish that can be caught in large numbers at once and takes salting or smoking well. Both groups play the leading role in the studies discussed below.
One counter-example shows how careful generalisations must be. According to a summary on the Archaeology Orkney site, herring is absent from Viking Age and Late Norse sites on Orkney, although it was popular elsewhere in the Norse North Atlantic. On that island group, large fish of the cod family shaped the finds. Herring, which turns up so naturally in pictures of the Middle Ages, is therefore not documented everywhere for the Viking Age. What holds for Orkney cannot be transferred automatically to Hedeby, Birka or the Norwegian coast.
Freshwater fish and fishing in rivers and lakes are left out here. I saw no analysis of them in this round. The article concentrates on the sea and so covers only part of Viking fishing.
James H. Barrett, Alison M. Locker and Callum M. Roberts evaluated fish bones from England for the years 600 to 1600 in a 2004 paper in the Proceedings of the Royal Society of London. According to the summary I saw, the finds show that marine fishing there changed rapidly around AD 1000: cod and herring become much more common. The authors link this to the rise of towns and to human impacts on freshwater habitats. They find it striking that the change falls in the Medieval Warm Period, when North Sea herring and cod productivity was probably low. The summary gives no figures, so I give none.
Researchers have settled on the name Fish Event Horizon for this shift. A horizon in archaeology is a narrowly dated layer at which something changes clearly. A page from Archaeology Orkney explicitly calls the Fish Event Horizon around AD 1000 well recognised for the Northern Isles of Scotland. According to that summary, intensive use of large cod-family fish peaks there in the 11th to 13th centuries, and isotope values from bone show that the share of marine food in human diet was highest in that period. By the 15th century, large cod bones are missing from ordinary sites.
The classification matters. The Fish Event Horizon lies at the end of the Viking Age and reaches far beyond it. It does not describe an invention of the Norse, and the studies I saw do not pin it to one people. What triggered it is debated. Urban growth, trade and changes in waters are named; no single cause is secure. Christian fasting is also discussed in the literature, but it does not appear in the summaries I read here. I leave it as an open question rather than asserting it as an explanation.
A fish bone does not reveal its origin at first glance. This is where isotope analysis comes in. Isotopes are variants of a chemical element with different mass. In bone collagen, the protein framework of bone, the ratios of carbon-13 and nitrogen-15 can be measured. These values depend on what an animal ate and on the food chain of the water it lived in. A cod from the Baltic carries a different signature from one from the Norwegian Sea.
James Barrett and a large group of co-authors presented this approach in 2008 in the Journal of Archaeological Science (volume 35, issue 4, pages 850–861). According to the abstract I saw, they measured the values in medieval reference samples from Arctic Norway, the North Sea, the Kattegat and the Baltic. Regional differences were clear enough to identify bones of cod caught in distant waters. As a first test they analysed three bone collections thought to derive from imported dried cod. The authors describe the result cautiously: it tentatively suggests that cod were transported or traded over very long distances from the end of the first millennium. Which collection came from which site is not in the abstract I saw, so I assign none to any place.
This is a good example of source criticism. The method allows a statement such as "this fish probably did not come from local waters". It says little about how the fish reached its find spot – as trade goods, as provisions or as a gift.
The thread is picked up in 2017 by a study in the Proceedings of the National Academy of Sciences (PNAS). First author is Bastiaan Star of the University of Oslo, senior author James Barrett of the University of Cambridge; the Centre for Baltic and Scandinavian Archaeology in Schleswig also took part. I have not read the paper in full, only the University of Cambridge press release and a report on it. Everything below is therefore reported.
The study as a whole covered 15 cod specimens from sites in Germany, Norway and Britain, from the Viking Age and the early Middle Ages (according to the dataset record); the Arctic origin concerns the Viking Age Hedeby bones. The team analysed ancient DNA from five cod bones found in the mud at the former wharves of Hedeby, a trading port on the Schlei, then within the sphere of Danish kings and now in Schleswig-Holstein; more is in the article Hedeby. The bones are dated between AD 800 and 1066. Compared with modern cod from different areas, their genetic profile matched the Arctic stock that spawns off the Lofoten Islands in northern Norway. According to the press release that implies a journey of more than a thousand miles.
The press release calls them the earliest known remains of northern Norwegian fish on mainland Europe and says that, before the study, there was no evidence of a European stockfish trade before the 12th century. I cannot check the first claim independently; whether earlier evidence exists is a question for the specialist literature.
The authors themselves name limits:
One point is easily overlooked: Lofoten in the study designates the spawning ground of the Arctic stock, not necessarily where the fish was caught or dried. Cod migrate there to spawn; where boats pulled them from the water does not follow from the genetics alone.
Stockfish is cod dried in the air without salt, traditionally on wooden racks in cold, dry air. The method works only when temperature and wind are right, as they are in northern Norway in winter and spring. The Lofoten Islands lie off that coast, and Arctic cod migrate to spawn in their waters. Dried fish is light, keeps for a long time and can be moved over long distances. Those are the properties of a trade good.
The stress lies on "properties". No document shows a Viking Age merchant selling stockfish. Even the Hedeby bones do not prove that the fish was dried; the press release speaks of stockfish, but a cod from the north turning up in a southern port first shows origin, not preservation. That it travelled dried is the obvious interpretation, because fresh fish would hardly have survived such a route. It remains an interpretation.
The press release hints that the stockfish trade later became important when it mentions the lack of evidence "before the 12th century". I did not check details of that later phase in this round and leave it at this brief statement.
The press release links the find to a 9th-century travel account. Ohthere (Old Norse probably Óttarr) was a man from Hålogaland in northern Norway who told King Alfred of Wessex about his voyages; the account stands in the Old English Orosius. The press release notes that the text covers a journey from Hålogaland to Hedeby and does not mention dried fish. A fuller account is in Ohthere of Hålogaland.
How Ohthere describes his wealth is revealing. He lists payments he received from the Finnas, as the text calls the people of the north (usually interpreted as Sami):
Þæt gafol bið on deora fellum, and on fugela feðerum, and hwales bane, and on þæm sciprapum þe beoð of hwæles hyde geworht, and of seoles.
"This tribute consists of animal skins, of birds' feathers and whale bone, and of the ship ropes made from the hide of whales and of seals."Old English Orosius, Book I (Ohthere account); wording checked against Bright's Anglo-Saxon Reader (Wikisource, public domain), macrons omitted; translation from this site's article on Ohthere
Fish is not on this list, and that should not be overrated. The account does know fishing: according to the Old English text the Finnas live by fishing at the sea in summer, and the northern land is empty except for hunters, fishers and fowlers. That is fishing for own use, not trade fish. These are payments from third parties, not everything a northern farm produced or sold. An account meant to impress a king lists furs and whale bone, not what every household had. Silence implies neither that fish trade did not exist nor that it did. The voyage from Hålogaland to Hedeby shows that such journeys took place; whether fish was on board, the text does not say.
Fishing needs gear, and the evidence is awkward. Nets and lines were made of plant fibre or animal hair, floats of bark or wood. These materials vanish in most soils entirely. When an excavation report lists no net, that means very little. Metal and stone last better: hooks of iron or bone, sinkers of stone, lead or clay, netting needles of bone or wood. That finds of this kind occur at Viking Age sites matches the general state of knowledge. I did not check specific find numbers or sites in this round and therefore give none.
What can be read from gear is limited anyway. A hook shows that lines were used, not how many hooks hung on one line or how far out people fished. Between fishing from the shore, from a boat in a fjord and from the open sea lie large differences, and gear alone does not separate them cleanly. Fish size helps. According to the Archaeology Orkney summary, the large cod of the 11th to 13th centuries came from deeper, open water. That points to trips beyond the immediate coast. The text does not say with what gear this was done, and I do not guess.
Sea fishing needs a boat. The large ships described in Viking Longship were vessels of war and trade and are no model for the fisher. Which boat types served for fishing I did not check in this round; that they were smaller craft is an assumption, not a statement from a source.
The dispute over the Hedeby cod turns on a simple question: was the fish sold, or did someone eat it aboard? The bones lay in harbour mud, where ships docked. Who put them there, the find does not say. According to the report, Barrett suggests that trade may have involved staple commodities as well as luxury goods. That is a researcher's suggestion, not a result.
How much fish actually flowed through trade is unresolved. Viking Trade describes the better-known routes of the north, such as silver and furs. That fish rarely shows in written sources on trade may reflect the character of those sources; it is not certain. For Birka I saw nothing in this round that documents fish trade.
A thought experiment shows what a find like this is worth. Five bones show that fish from the far north arrived in Hedeby. They do not show whether that happened ten times or ten thousand times. That is a task for future research, and the authors say more fish remains are to be studied.
For an article resting on a few good studies it is honest to list the gaps. First, larger samples are missing: five bones from Hedeby are a beginning. Second, dating is coarse, because the span 800 to 1066 covers more than two centuries. Third, we know little about everyday fish use on farms in Norway, Iceland or Denmark; the sources evaluated here concern ports and islands, not the average farm. Fourth, the link between the Fish Event Horizon in England and Scotland and fishing in the Norse heartlands is unresolved. I saw no study in this round that connects the two systematically.
The finds from Hedeby are a reminder of how much everyday life stands behind the familiar symbols of the Viking Age – a thought we also keep in mind in the family workshop Glanz & Gravur when working on our Nordic motifs.
Did the Vikings eat a lot of fish? For Orkney and parts of England it can be shown that sea fish gained strongly in importance around AD 1000 and after. For Viking Age Scandinavia as a whole I saw no summary figure and give none.
What is stockfish? Unsalted cod dried in the air. When exactly and on what scale the Norse produced it is not conclusively documented; the Hedeby bones point to fish transport from the north.
How do we know the Hedeby cod came from the Lofoten? From ancient DNA of five bones that matches the Arctic cod stock. This is a genetic comparison, not proof of the catch location, and the sample is small.
What does "Fish Event Horizon" mean? An archaeological term for the rapid change in sea fishing around AD 1000, visible in bone finds from England and Scotland.
Are there Viking fishing hooks? Hooks and sinkers are regarded as typical finds; I did not check find numbers. Nets and lines hardly survive.
What Did the Vikings Eat? – Hedeby – Ohthere of Hålogaland – Viking Trade – Viking Longship – Birka – Jarlshof and the Shetlanders – Viking Daily Life
Viking food & drink · Ohthere of Hålogaland · The longships · Viking Trade · Jarlshof and the Shetlanders.
All History & Finds articles →
Between the Broads and the North Sea, a cluster of Norfolk villages carry names that sound as if they came from …
Read more →A man takes three ravens to sea, lets them fly and follows the third. He winters in a fjord in north-western Iceland …
Read more →The Fosi were a Germanic people known from exactly one sentence. Around AD 98 Tacitus names them in the Germania as …
Read more →Step out of our workshop in Alsdorf and look west, and you are looking at the heartland of a power that grew out of a …
Read more →