Journal of Archaeology in the Low Countries 2-1 (May 2010)W.A. van Es; W.J.H. Verwers: Early Medieval settlements along the Rhine: precursors and contemporaries of Dorestad

5 Conclusions

1. When Dorestad emerged in the mid-seventh century, there were already settlements of various types and origins along the Rhine.

In the first place, there were the former Roman castella. These combined the prestige of their Roman origin with the practical advantage of their location directly on the international waterway and were sites eminently suitable for the development of settlements of supra-local importance. The old (harbour) facilities, however dilapidated, were still present.

Second, there were rural settlements consisting of several farmsteads clustered together in an orderly arrangement. So far, no isolated farmsteads are known.

Castella and nearby farming settlements may have operated in a complementary fashion. This is what we presume for Levefanum and the Risuuic villa, which together seem to have constituted one of the precusors to Dorestad. Such a symbiosis may also have existed at Roomburg. Other known farming settlements lay too far from a castellum to make close relations likely. De Geer, another precursor to Dorestad, is an example of such ‘autonomous’ settlements, and maybe also Lote/Leut.

A third type is the smaller ‘trading settlement’ situated on the bank of the main stream, of which so far Valkenburg-De Woerd is the only excavated example. We imagine that it operated like a minor version of Dorestad. In this case too, there may have been operational links with a castellum. These smaller Early Medieval vici – Dorestad being a vicus famosus (‘of renown’) – probably incorporated an agrarian component, as did their illustrious example, but may also have maintained economic relations with surrounding farming communities.

2. In many cases, several settlements of similar or different types might be in the hands of a single landowner. Apart from the king and the church, whose landownership steadily increased, there were also large private landowners. At Rijnsburg we caught a glimpse of one of these families who lived on one of their estates in the region. Whether this was the usual state of affairs, we do not know. It is likely to have varied through time.

The castella are regarded as royal estates. If Levefanum belonged to the (Merovingian) king, might De Geer, separated from the castellum site by the river Lek, then have been in the possession of one or several regional potentates? As Dorestad emerged, two kings reigned in the Rhine region: a Merovingian king and a Frisian one. This situation goes back to the sixth century, following a fifth-century power vacuum in the Rhine delta. In the context of Frisian-Frankish antagonism, landed properties probably changed hands quite frequently.

3. The beginnings of the settlements that we know of in the seventh century are also quite varied. Some, like De Geer, went back to the (Late) Roman period. Others, such as Koudekerk, did not emerge until the Merovingian period. For the castella, continuity from Roman times is usually hard to prove, yet often quite plausible.

Upstream from Utrecht there seems to be more continuity of occupation than downstream in the delta, with the exception of the Rhine mouth, where the castella of Roomburg, Valkenburg and Brittenburg (now submerged) are sites of potential continuity.

East of Utrecht, we find evidence of Frankish colonisation in the Late Roman period which reversed at least part of the depopulation that marked the third century. Theoretically, in the Late Roman period the area around the Rhine mouth could have accommodated a similar repopulation by Frisians, but so far archaeological evidence of this is rare. Any remaining population in the fifth century might have been outnumbered by an adventus Frisionum similar in time and process to the adventus Saxonum in Brittannia, an influx of groups of newcomers under their own leaders. This could have laid the basis for the subsequent Frisian expansion in the Rhine delta.

4. In the central Netherlands, from the fifth century onwards imported wheel-thrown pottery wholly or largely replaced the native handmade pottery. As yet, little is known about the provenance of these Merovingian products but the German Rhineland must have played an important part even then. A reorganisation of production methods in the Eifel mountains and the Vorgebirge resulted in an increased stream of exports, which through the facilities offered by Dorestad was mainly directed towards the Rhine delta and the Dutch North Sea coast. The hinterland in the north-eastern Netherlands obtained these wares mostly through the port of Deventer.