Mike Belasus, 22 December 2017
with Kevin Martin and Bart Holterman
A Project on the Low German merchants’ trade from Bremen and Hamburg with the North Atlantic islands is not complete without the vessels that made trade across the ocean possible. The written sources tell us that many German ships were lost along the coast of Iceland, but until today, not a single mentioned shipwreck was found. This is not surprising, given the circumstance that underwater archaeology in Iceland is not even twenty years old.
Some facts about finding shipwrecks
The saying “Searching for a needle in a haystack” is fitting pretty well to the search for shipwrecks mentioned in historical documents, considering the abilities to determine a ship’s position in the past and the fact that 71 % of Earth’s surface is covered by water. Even today, many ships vanish without a trace. On top of that, the number of archaeologists looking for certain shipwrecks is extremely low. Therefore, most shipwrecks are found by accident and they remain more or less anonymous.
However, technology has improved and the potential for finding shipwrecks is much higher today than in the past. A number of devices can be used and combined for surveying the sea floor. A common combination is a side-scan sonar and a number of magnetometers, but this is still no guarantee for finding wrecks. When a ship is too decayed, or has settled into the sediment, the sound-rays of the side-scan sonar will not be able to produce a recognisable image. Moreover, magnetometers will not detect anomalies from the earth’s magnetic field when there is not enough metal on board the ship. Especially for medieval ships and early modern merchant vessels without guns, this can easily be the case. A sediment sonar could be a solution, which measures changes of density in the sediment with sound signals. The disadvantage is that waterlogged wood has the same density as waterlogged sediment. If a shipwreck has no denser cargo or ballast on board, the sonar will not detect density changes. Therefore such ships will remain hidden, even if an area is surveyed. Divers might be suggested as a final solution but a diver’s operating range is very limited, as he/she depends on depth, weather conditions, visibility and air supply.
The fact is that archaeologists rely in most cases on coincidences, which is how most of the known shipwrecks were found. The reason for this is that many other professions like builders, fishermen, geologists etc. spend much more time working on the sea floor than archaeologists, and sometimes stumble upon a wreck. One famous example for this is for example the “Bremen Cog” in the German Maritime Museum in Bremerhaven. A suction-excavator operator found it in the river Weser close to Bremen, when he was working on an extension of the riverbed. It became one of the most important finds in ship archaeology. Over the past decades, since underwater became an area of archaeological interest, hundreds of shipwrecks have been registered by the heritage departments of the German federal states but only a few of them could be identified by historical documents.
When we approach the shipwrecks from the historical documents, we have to realize that mentioned positions for lost ships are most of the time very vague. This situation can extend the search area dramatically without even knowing if the wrecks still exist as a coherent site. The conditions of the natural environment are vital to the survival of remains. Rugged coastlines, strong currents, micro-organisms, the chemical consistence of the water and the like can severely contribute to the decay of organic matter, metals and even certain types of stone. In the Mediterranean Sea, for example, ship hulls will not survive for a long time above the sea floor due to temperature, salinity and micro-organisms. Iron will soon vanish by the reaction with salt and oxygen and even marble and copper-alloys will decay. The opposite situation can be found in the Northern Baltic or the Sea Black Sea, for example. The Northern Baltic is deep, has a very low salt content and is cold. This prevents wood-decaying organisms to multiply and destruction from anchors or looting divers. The Black Sea water, on the other hand, has hardly any oxygen in a certain depth, which almost freezes time on the sea floor.
In addition, the circumstances of the sinking in connection to the natural environment play an important role. A German ship in Spanish service in 1588, the Gran Grifon, hardly left any structural remains. Another extreme example is the destruction of a united navy fleet of Lübeck and Danish ships during the Nordic Seven Years War on the west coast of Gotland in 1566. Out of thirty-seven ships, fifteen were lost and 6000 to 8000 sailors lost their lives in the so-called Visby-Disaster. Intensive archaeological diving surveys for several years could not reveal any wreck site. The ships literally vanished, most likely because they disintegrated in the storm on the rough sea floor and their remains were thrown up on the beach.
These examples shows us that a sunken ship mentioned in the historical documents does not always leave a proper wreck site. Sometimes scattered debris spread over square kilometres of sea floor can be all that is left of a ship.
The missing wrecks of Iceland
In theory, Iceland has a high potential for finding shipwrecks. Since the late 9th century, humans settle here and they had no other way than to cross the rough North Atlantic Ocean and make landfall on a rugged coast. Many ships, Norwegian, Danish, English, Low German and Dutch, evidently got lost on their route between continental Europe and Iceland. A considerable number of these found their end on the coast of Iceland, as we can read in the historical documents. Ragnar Edvardson has calculated a number of about 450 shipwrecks in Iceland, from the period from 1100 to 1900, most of which foundered on the West coast. Until today, no wreck from the period of the Low German trade has been found. The oldest ship found in Iceland is currently a Dutch merchantman, which sank in the harbour at Flatey in Breiðafjörður in 1659, and is currently under investigation by Kevin Martin. The scarcity of known shipwrecks in Iceland is not necessarily because they vanished completely, but because maritime archaeology is a very young discipline in this country and started only with sporadic projects from 1993 onwards. The relatively few inhabitants of Iceland also reduce the possibility of coincidental discoveries of wrecks due to the lack of building activities.
Vive la Coïncidence!
However, coincidences happen – even in Iceland. In 1998, an excavator hit timbers while digging a cable trench near Búðir on the Snæfellsnes peninsula in Western Iceland. They turned out to be fragments of a ship that was buried under the sand of an estuary. Some timbers and ballast stones were recovered by archaeologist Björn Stefánsson, and brought to the National Museum of Iceland in Reykjavik, were they were documented and put in storage. None of the recovered timbers was suitable for dendrochronological dating, but the ballast included stones that can possibly originate in Norway, Britain or Greenland. However, this does not give any indication were the ship was coming from, because ballast was taken from board and loaded in all harbours depending on the cargo and to adjust the angle of the ship in the water, to trim the ship. The recovered timbers give the impression of an old carvel-built ship. The material seems not to be of the finest quality, but this is hard to judge from only a handful of timbers. A promising fact is that Búðir was an important trading centre in Iceland for at least 200 years. It was already visited by Bremen merchants in the late 16th century, who called it “Bodenstede”. For a long time it was the main trading site for the Snæfellsness Peninsula, called first by cargo ships of the Low German merchants and later by Danish and Dutch ships. It was abandoned as a trading post in 1787.
For this reason, we have written sources on the activities on this site including the loss of several ships during these 200 years. The problem for the identification of the accidental found ship is the fact that there was not only one ship lost near the site. We know at least about eleven ships that are reported to have been lost at Búðir and its harbour Búðavik. Of those eleven ships, one was from Bremen. It belonged to the merchant Vasmer Bake and sank in or close to the harbour of Bodenstede in 1587. This was reported by Carsten Bake, his son, who himself was involved in the Iceland trade. In 1607, a Danish ship wrecked at Búðir. For 1666, another ship sank west of Búðir. More ships sank in this area in 1724, 1728 (Danish) and 1754, and lastly, Bjarni Sívertsen’s cutter, which got lost here in 1812.
Certainly, it is possible that the ship found in 1998 is Vasmer Bake’s lost vessel from 1587, but the timbers cannot give us any more detailed information. All we can say today about the few recovered timbers is that there are five floor timbers and two fragments with an unknown function. The floor timbers indicate that the cable trench hit the wreck most likely in the bow section, leaving most of the buried remains uncovered. Each plank was attached to each frame with at least two tree nails, and each frame was connected to the keel with one or two tree nails. The dimensions of the floor timbers vary a lot, from flat and wide to high and narrow. This might be an indication that whoever built the ship faced either a shortage of crooked compass timbers, or the building did not rely on an even framing in a shell based building method, or both. However, it can also not be excluded that we face the remains of Bjarni Sívertsen’s cutter from 1812. Only future investigations might give us an answer.
Edvardson R. and Grassel Ph., The Potential of Underwater Archaeology in the North Atlantic. In: N. Mehler, Travelling to Shetland, Faroe and Iceland during the 15th to 17th centuries (in press).
Stefànson, B., Skipsviðir úr Búðaósi. Rannsóknaskýrslur fornleifadeildar 1998. Fornleifadeild þjóðminjasafns Íslands, Reykjavik 1998
Bart Holterman, 7 September 2017
Bart Holterman presentated the project and his own research on August 5 in Haus der Wissenschaft in Bremen. Matthias Holthaus was there as well and wrote the following article about it in the Weser Kurier of August 21, which gives a nice overview of the situation German merchants had to deal with when trading in the North Atlantic in the late Middle Ages (in German).
Posted in: Press
Natascha Mehler, 27 April 2017
In the 16th century there were approximately 25 trading stations in Iceland which were regularly frequented by merchants from Bremen, Hamburg, Oldenburg or Lübeck. Most of these stations lay in the western part of the island, around the Snæfellsnes and Reykjanes peninsulas and in the Westfjords, because these were the areas where most fishing settlements were located and where cod, which was transformed into the main trading commodity stockfish, was abundant.
It is difficult to tell what these German trading stations looked like because none have survived until today. In 1604, the Danish king Christian IV ordered to tear down all German buildings in Iceland, as a consequence of his imposition of the Danish trade monopoly (1602–1787). During the early 17th century many of the German trading stations (such as Grundarfjörður in the image above) were taken over by Danish traders and subsequently developed into permanent and more substantial settlements, many of which still exist today. What we don’t know is whether the Danish traders really tore down the German buildings, as was the will of the king, or simply re-used the buildings and infrastructure. The latter would certainly have been more practical in a country where building material was scarce.
Luckily, a handful of German trading sites did not develop into modern settlements and remains have survived until today. One of them is situated on the small tidal island of Landey at the northern side of the Snæfellsnes peninsula, near the present farm of Bjarnarhöfn. Written sources mention a trading post visited by merchants from Oldenburg and Bremen in the 1590s, but the amount of information from the surviving documents is very low. In 1593, Carsten Bake from Bremen acquired a three-years license for the harbour, and afterwards the harbour was given to the count of Oldenburg. In 1599, a request by the count to renew the license for Landey was denied by the Danish king because it was already in use by someone else. Who this person was is not mentioned.
Today, Landey is a small island only used by sheep for grazing. The outlines of two buildings are clearly visible. They are situated on a small plain near a beach that was used as a landing site, to pull up small boats. The western side of the island faces a bay called Kumbaravogur, and on the other side of that bay lies Kaupstaðartangi, a headland where the remains of a trading station called Kummerwage have survived. This close proximity could be the reason why Landey started to be used as a trading post in the first place, as Kummerwage had been a harbour of Bremen merchants for a long time, but was given to Joachim Kolling from Oldenburg in 1580. Bremen tried hard to get the harbour back, but without success. The merchants of Bremen must have tricked the Danish king in requesting a license for the nearby Landey. When the king realised his mistake he included the clause that the license would go to Oldenburg after the Bremen one expired.
A trial excavation conducted with Fornleifastofnun Íslands at Landey in the summer of 2016 has shown that both buildings on Landey were indeed dwellings, although rather simple constructions. The walls were built in the Icelandic way, made of turf with stone linings. A fire place was discovered in the western building and no signs of furniture or other equipment were found. However, ten fragments of a ceramic cooking pot made of red earthenware were discovered. ICP-MS analysis of the sherds (conducted by Torbjörn Brorsson, KKS) has shown that the pot was made in Bremen, or in very close vicinity thereof (e.g. in Oldenburg).
The building to the east has two inner partition walls which indicates that it could have been used as a storage facility. Taking the scarce written and archaeological evidence together we can conclude that the buildings were indeed dwellings of some sort but only used for a very short time. This corresponds well with what we know about the Icelandic trading stations from other written sources: they were only used during the summers, when the foreign vessels were in Iceland, and their owners did not invest in solid infrastructure because they were not sure whether they would return the next summer. For Landey, records only exist for the 1590s.
Written sources indicate that merchants generally slept in the dwellings on land while the ship´s crew stayed on board. However, the excavations at Landey give food for thought. A Northern German merchant might have found it more comfortable to stay on board than to spend his nights in a damp and dark turf building…
D. Kohl, Der oldenburg-isländische Handel im 16. Jahrhundert. Oldenburger Jahrbuch 13, 1905, 34–53.
Mehler & M. Gardiner, On the Verge of Colonialism: English and Hanseatic Trade in the North Atlantic Islands, in Peter Pope and Shannon Lewis-Simpson (eds.), Exploring Atlantic Transitions: Archaeologies of Permanence and Transience in New Found Lands. Society of Post-Medieval Archaeology Monograph 8 (Woodbrige 2013) 1-15.
Oldenburg, Niedersächsisches Landesarchiv, Best. 20, -25, no. 6.
Copenhagen, Rigsarkivet, Tyske Kancelli D 11, Pakke 28 (Island og Færøer, Supplement II, no. 25)
Bart Holterman, 30 March 2017
Trade in the North Atlantic was neither simply an exchange of goods between ports in Iceland and in Germany, nor did the islanders stay at home and wait for German merchants to show up each spring. In some cases, more complex relations between islanders and Germans existed, and in what follows we will present an example from Iceland to highlight the complexity of the trade relations between Icelanders and Germans. Germans were not allowed to settle in Iceland, but Icelanders were free to move or travel to Germany, which some did, and established networks of their own. Sometimes these links can be reconstructed in considerable detail with the help of both historical and archaeological sources. In this post, we will focus on one such case, that of Eiríkur Árnason Brandssonar (c. 1530-1586), sýslumaður (bailiff) of Múlaþing district in the East of Iceland.
Eiríkur Árnason was a powerful man in Iceland, who repeatedly got into trouble with Guðbrandur Þorláksson, bishop at Hólar. Eiríkur was appointed klausturhaldari, the keeper of the property of Skriðuklaustur monastery, located in the valley of lake Lagarfljót about 50km inland from the southeastern coast of Iceland. The monastery had been secularized in 1554, and the now royal property was administered by the klausturhaldari. Eiríkur, whose grandfather Brandur Hrafnsson had been the last prior of the monastery, held this office between 1564 and 1578.
In 1584, it seems that Eiríkur settled in Hamburg. He embarked on a voyage to Hamburg on the ship of Jochim Warneke and shortly afterwards married Anna Korfemaker and bought a brewery. The next year he travelled back and forth between Iceland and Hamburg and it is most likely that he was a member of the fraternity of St Anne that united most merchants and sailors who travelled to Iceland. In 1586 he died in Hamburg and was buried there at the cemetery of St Jakob.
There are numerous sources which shed a light on Eiríkur’s contacts with German merchants in Iceland. In the fjord now known as Berufjörður, south of Skriðuklaustur, Bremen merchants (who called the fjord Ostforde) had been trading for a long time. In 1575, a conflict broke out between two of them, Bernd Losekanne and Christoffer Meyger, due to alleged mutiny. In the ensuing court case before the Bremen city council, one of the problems discussed was the ownership of a barrel of osemund (iron from Sweden) which Losekanne had sold to Eiríkur Arnason but which was left in the merchant´s booth, of which Eiríkur as bailiff had the keys. Also, a load of vaðmál he left in the booth was destined for a Hamburg merchant named Matthies.
It is clear that during his time as sýslumaður and klausturhaldari, Eiríkur had been trading actively with the German merchants in the area. This is also attested by contemporary objects. From the house of the klausturhaldari stems a complete Bartman jug of the 16th century, a common pouring vessel of that time, produced in one of the major stoneware production centers along the Rhine (such as Frechen, Raeren or Cologne). It is more than likely that this jug was used by Eiríkur who had bought it from the German merchants in Berufjörður. Another link is the gravestone of his mother Úlfhildur Þorsteinsdóttir which is thought to have been imported from Germany by Eiríkur.
Eiríkur’s connections with German merchants probably went further than that. In 1580 Bernd Losekanne and Christoffer Meyger (who had apparently reconciled) complained about interference of Hamburg merchant Matthias Eggers in Ostforde. Eggers, on the other hand, said that he had a general trading license for Iceland which obviously permitted him to use any harbour he liked. Losekanne and Meyger then replied to this that this was unfair because Eiríkur had acquired that license for himself when he had visited the Danish royal court, and had entered into a trade agreement with the Hamburg merchants, in return for a part of their ship. Eggers was also probably the man named Matthies for which Eiríkur had put aside the vaðmál in 1575.
Merchants from Bremen are known to have entered into trade relations with Icelanders as well, but that might be a topic for a future post.
Friederike Christiane Koch, Isländer in Hamburg 1520-1662. Beiträge zur Geschichte Hamburgs vol. 49 (Hamburg 1995) pp. 150-153.
Staatsarchiv Bremen, 2-R.11.ff.
Staatsarchiv Hamburg, 111-1, Cl. VII Lit. Kc. no. 11, Vol. 3.
Posted in: Stories
Bart Holterman, 22 December 2016
For most readers of our blog, probably the most mysterious part of our research is the work of Hans Christian Küchelmann, archaeozoologist, who uses archaeological finds of fish bones as traces of the late medieval North Atlantic (stock)fish trade. A fish bone found in the ground, however, does not say where the fish once came from or to which fish it belonged. So, how can one identify a stockfish merely by its bones? This blog post will shed a light on that mysterious procedure.
1. Stockfish production
In order to be able to identify a stockfish bone, it is essential to know how stockfish was produced. Luckily, stockfish is made in Iceland and Northern Norway to this day in ways that have hardly changed over the centuries, so in combination with historical sources, we can reliably reconstruct what a medieval stockfish must have been like.
Stockfish is made in polar regions from fish from the Gadidae family where they are hung to dry outside during the winter. Due to climatic conditions this is only possible in arctic regions as the cool weather prevents the fish from rotting before it dries. However, the entrails must be removed before the fish is hung to dry, and the heads are cut off. This is roughly done with two methods: a) the fish is gutted and beheaded, the rest of the body left intact (rundfisk), or b) the fish is beheaded and split in two, removing the entrails as well as most of the spinal column, and then hung to dry. The latter method is called råskjær in Norwegian (rotscher in the Low German medieval documents). As this leaves only a few caudal (tail) vertebrae in the stockfish, most bones of stockfish will come from rundfisk.
2. Species and distribution
As mentioned, stockfish was made from species of the Gadidae family. A trained eye will have no problem recognising a bone from a Gadidae fish in most cases. Due to long travel times and the absence of freezers in pre-industrial times, these fish could only be transported in preserved form, either salted or dried. Hence, a find of a Gadidae bone, especially on inland sites, hints at having belonged to a stockfish.
However, in coastal areas these fish were also eaten fresh, so how do you know a bone is from stockfish in that case? It is necessary to look at the distribution of the different Gadidae species. Three of the species that were used for producing stockfish, namely saithe (Pollachius virens), ling (Molva molva), and tusk (Brosme brosme) do not appear in the Southern North Sea and Baltic Sea. Bones from these three species found on the European mainland, especially in inland areas, are therefore a strong indicator for stockfish.
By far most of the stockfish, however, is and was made from cod (Gadus morhua), which does live in the Southern North Sea and the Baltic Sea. Bones of cod could therefore also belong to locally caught (and therefore not dried) fish. So we need another indicator to distinguish a stockfish from fresh local cod. The cod which live around the German shores are mostly smaller juveniles, older and larger fish live further North. By comparing the size of the bones to those of complete skeletons it is possible to estimate the size of the fish they belonged to. Bones belonging to fish larger than c. 75cm are less likely to have been local catch and were probably imported in dried form. Moreover, a high prevalence of fish in a specific size range is an indicator of stockfish, as these were sorted and sold according to size, whereas local catches will likely show a higher variety of fish in different sizes.
4. Bone composition
Because the head of the fish was cut off and remained at the production site one can expect head bones to be absent at consumption sites. Indeed, we find a clear overrepresentation of post-cranial (i.e. bones not belonging to the head, from cranium: head) bones on some sites in mainland Europe. On some archaeological sites in Iceland, for example, there are almost only cranial bones which is a clear sign that stockfish was produced there.
5. Cutting and hammering marks
The production and consumption of stockfish can also leave traces on the bones themselves. For example, the heads of fish were chopped of in a standardised way, leaving clear-cut chopping edges on the bones of the shoulder girdle directly behind the head. Also, the preparation of stockfish required hammering the fish for a while before soaking it, to make the flesh softer. As we have seen from our own experiences in preparing stockfish, this procedure can destroy or deform the vertebrae of the fish. Hence, deformed or broken vertebrae can be a sign of stockfish consumption in the archaeological record.
6. Isotope and aDNA analysis
Further, more advanced evidence from the North Atlantic stockfish trade can be acquired by applying methods such as aDNA analysis and isotope analysis, which can potentially retrace the remains of an animal to the area in which it lived. However, an explanation of these techniques might be a topic for a next post.
Barrett, James H. (2009): Cod bones and commerce: the medieval fishing revolution. – Current Arrchaeology 221, 20-25
Heinrich, Dirk (1986): Fishing and the Consumption of Cod (Gadus morhua Linnaeus, 1758) in the Middle Ages. in: Brinkhuizen, Dick Constantijne & Clason, Anneke T. (eds.): Fish and Archaeology.
Ólafsdottir, Gudbjörg Ásta / Westfall, Kristen M. / Edvardsson, Ragnar / Pálsson, Snæbjörn (2014): Historical DNA reveals the demographic history of Atlantic cod (Gadus morhua) in medieval and early modern Iceland. – Proceedings of the Royal Society of London B 281
Orton, David C. / Morris, James / Locker, Alison / Barrett, James H. (2014): Fish for the city: meta-analysis of archaeological cod remains and the growth of London’s northern trade. – Antiquity 88, 516-530
Posted in: General