The Early-Medieval harbour of Dorestad

Topography

The Early-Medieval site of Dorestad was situated to the north of the Late-Medieval town of Wijk bij Duurstede, in the province of Utrecht (the Netherlands). At Wijk bij Duurstede, the northernmost branch of the Rhine (the Lower Rhine) splits into the Lek and the Kromme Rijn (see fig.1).

Due to its situation at this `fork', the settlement was included in the network of Northwest-European shipping routes: the Lower Rhine formed a direct link with the German Rhine area, the Kromme Rijn with (sea)routes to England, the North of France, the Northern Netherlands, North Germany and Scandinavia (fig.2). The Lek possibly gave access to the lower reaches of the Meuse and the Scheldt.

History of research

The most extensive excavations of Dorestad took place from 1967-1977, during which approximately 30 ha were exposed. The area of the Early-Medieval settlement begins approximately 1 km north of the Late-Medieval town centre of Wijk bij Duurstede and disappears almost 2 km south under the Lek dike. The settlement may have extended at least 1 km further south, representing a total length of about 3 km.
Dorestad must have been made up of three parts: the northern harbour district, a central part and a castellum-quarter to the south (the remnants of the former Roman castellum Levefanum).
The northern harbour district was divided in three parts: the actual harbour in the Carolingian Rhine bed, adjoining on the left bank the trading settlement - the vicus in the strict sense of the word - and behind it an area of agricultural character with scattered farmbuildings (see fig.1). The analysis of the data has so far concentrated on the harbour, the main subject of this homepage.
 

The port installations

During the excavations in the harbour, remains were found of an extensive complex of wooden constructions, interpreted as platforms (see fig.3). These were about 6 to 8 m wide, situated at right angles to the river bank and firmly anchored in the subsoil with long vertical piles. The construction of the platforms is connected with the development of the Rhine meander here: they bridged the in-creasing distance - caused by the gradual retreat of the river towards the east - between the bank and the stream.
It seems that the platforms had a more or less closed front in the various phases and that the ships were drawn to the beach in front of the platforms in order to be unloaded. This would imply that the platforms formed a sort of link between the relatively high bank and the river beach, on which the ships were drawn. The ships then had to be unloaded on the beach and the cargo afterwards had to be hoisted on the platforms. This seems rather impractical, unloading directly on the platforms - in fact long platforms or jetties - seems more reasonable. The ships then had to come alongside, or against the platforms, which possibility is also stated by the excavators.

The harbour parcels adjoined a road, approximately 3 m wide, which ran along the edge of the em-bankment from north to south, with the vicus situated behind it. Since little of the vicus has been excavated, the following outline cannot be more than a hypothesis.

The harbour parcels were continued into the vicus, which means that it was also divided into plots, running from east to west. These were approximately 20 m wide and corresponded each with a couple of platforms in the riverbed. Houses on these plots stood with their longitudinal axes towards the river. These were rectangular, 6 m-wide wooden buildings, of which no further details are known. Lengthwise, there was room for at least three houses, one behind the other, with an average length of 20-25 m each. It is possible that the houses had enclosed yards. The vicus might then have consisted of three rows of houses parallel to the river, situated behind each ther.
 
 

The causeways

The plan (see fig. 3) shows a bewildering and apparently inextricable tangle of posts and postholes.
However, long and straight rows of closely set posts and postholes stand out. The posts vary in size, the bigger ones having a diameter of roughly 20 cm. The post rows include a great number of coupled posts, at right angles to the axis of the rows. Where posts do not stand exactly opposite each other, they are placed in a more or less regular zigzag line. The post must have held a wooden revetment, probably of planks, but a revetment of wickerwork occurred as well, but more rarely. These revetments were used to strengthen the edges of narrow strips of land, between 6 and 8 m wide. The strips are divided in a number of segments of varying length by transverse connections. The total length of the strips is about 200 m.
Construction of the platforms took place in several phases, the sections representing the individual construction periods. The growth of the platforms was mainly in length, but the occurrence of multiple side revetments, show that the strips of land could also be slightly widened.

Sections of platforms completely free of internal features are rare. In most cases, the inner space is marked by the presence of rows of vertical posts - with roughly a diameter of 15 cm - parallel to the long axis of the platforms. They might have been connected by planks or beams, fastened to the top, which probably served as foundations for the road pavement. The construction of the pavements itself remains obscure.

The complex of the platforms is the result of a long building process. The development started on the west bank of the river in approximately AD 675/700, from which the building activities spread to-wards the east, gradually penetrating into the river bed. The total complex of the harbour is made up of two parts: a western (Part 1) and an eastern (Part 2), characterised by a significant change in building-style. The division runs across the whole complex from north to south, at about the transition from squares A/K-8 to A/K-9 (see fig.3 and fig.4). To the west of this division, the sections are relatively narrow and often rather vaguely outlined, with mostly short compartments; well-defined rows of inner posts are rare. To the east, the platforms tend to be slightly wider, with clear outlines, many long compart-ments and well-defined inner post rows.

Part 1 is subdivided into two zones, the Sectors 1A and 1B, of which the transition corresponds roughly with the squares A/K 5.

Sector 1A is characterised by by several isolated, short but remarkably regular ground plans (see fig.4). These substantially built platforms (see the reconstructed platforms in fig.5 and fig.6) - some 10 or 12 m long and 6 to 7 m wide - represent the starting points in the development of the whole complex. Most of them were extended afterwards by the addition of new compartments.
 
 
 
 
 
 

Sector 1B is characterised by its `vagueness' (fig.7). The not easily definable platform sections probably represent comparatively light constructions.
The Sectors 1A and 1B are dated to approximately AD 675-700/725.
 
 
 
 
 

In Part 2 the complex acquires its definite form, with a relatively regular system of well defined, substantially built platforms. The length of the compartments and the regularity of the general layout suggest a more rapid growth than in Part 1.

Part 2 also shows a subdivision into successive zones, but of minor importance, as the difference between the individual sectors is far less pronounced than in Part 1. Three zones are distinguishable: Sector 2A (division line with 2B mainly through squares A/K-16), Sector 2B (division line with 2C mainly through squares A/L-18/19) and Sector 2C.
 
 

Sector 2A (see fig.8) shows unusually long compartments (see the reconstructed cause-ways on fig.9 and fig.10), extending nearly as far as square 17, with their eastern ends forming an almost straight line, from north to south. Sector 2A is dated to approximately AD 700/725-750/775.
 
 
 
 
 
 
 
 
 
 

Sector 2B (see fig.11) has compartments rather shorter than those in Sector 2A, but the difference is not great. This sector is dated to approximately AD 750-775.
 
 
 
 
 

Sector 2C (see fig.12) represents the end of the development of the harbour. It shows a re-markable change, since the compartments are now extremely short. The eastern edge presents a frayed and ragged line, caused by substantial erosion. The excavators presume however, that the eastern edge in reality represent the original end of the platforms.
 
 

Period 2C is to be subdivided in:
a.  completion of harbour works: circa AD 775-800/825;
b.  continued occupation with minor additions and repairs in a time of growing economical stagna-tion: circa
     AD 825-850/875;
c.  end of function as international trading site: circa AD 850/875.
 

Literature

This text is based on:
W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
W.A.van Es, Wijk bij Duurstede: Dorestad, een handelsplaats uit de tijd van Karel de Grote, in: W.A.van Es, H.Sarfatij & P.J.Woltering (eds), Archeologie in Nederland, Amersfoort/Amsterdam, 1988, pags. 170-173.
W.A.van ES, Dorestad centred, in: J.C.Besteman, J.M.Bos & H.A.Heidinga (eds), Medieval Archaeology in the Netherlands, Assen/Maastricht 1990.

fig.1: after fig.2 in: W.A.van ES, `Dorestad centred', in: J.C.Besteman, J.M.Bos & H.A.Heidinga, Medieval Archaeology in the Netherlands, Assen/Maastricht 1990.
fig.2: after fig. on page 171 in: W.A.,van ES, H.Sarfaty & P.J.Woltering, Archeologie in Nederland, Amsterdam 1988.
fig.3: fig.8 in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig.4: after fig.21 in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig. 5: fig.11c in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig. 6: fig.10c in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig.7: after fig.21 in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig.8: after fig.21 in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig.9: fig.15c in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig.10: fig.16c in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig.11: after fig.21 in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.
fig.12: after fig.21 in: W.A.van Es & W.J.H.Verwers, Excavations at Dorestad 1. The Harbour: Hoogstraat I. Nederlandse Oudheden 9, Amersfoort 1980.