Where is the eighth century in the towns of the Meuse valley?

Millennium-Studien zu Kultur und Geschichte des ersten Jahrtausends n. Chr. Millennium Studies in the culture and history of the first millennium C.E. Herausgegeben von / Edited by Wolfram Brandes, Alexander Demandt, Helmut Krasser, Hartmut Leppin, Peter von Möllendorff Volume 5/1 Walter de Gruyter · Berlin · New York Post-Roman Towns, Trade and Settlement in Europe and Byzantium Vol. 1 The Heirs of the Roman West Edited by Joachim Henning Walter de Gruyter · Berlin · New York 앝 Printed on acid-free paper which falls within the guidelines of the ANSI 앪 to ensure permanence and durability. ISBN 978-3-11-018356-6 ISSN 1862-1139 Library of Congress Cataloging-in-Publication Data A CIP catalogue record for this book is available from the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de. 쑔 Copyright 2007 by Walter de Gruyter GmbH & Co. KG, D-10785 Berlin All rights reserved, including those of translation into foreign languages. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording or any information storage and retrieval system, without permission in writing from the publisher. Printed in Germany Cover design: Christopher Schneider, Berlin Where is the eighth century in the towns of the Meuse valley? FRANS THEUWS The early medieval centres of the Meuse valley – Dinant, Namur, Huy and Maastricht – enjoy a long and distinguished historiography.1 The results of recent archaeological research will be added to this historiography in the near future. The excavations in Namur and Maastricht are of special importance due to their scale and distribution.2 These centres played an important role in the creation of grand narratives on the transformation of the Roman world. They were for instance constantly in the back of the mind of Henri Pirenne, the great Belgian historian.3 They played a role in the formation of ideas about the decline of the Mediterranean world and the rise of the Carolingian Empire and the shift of the economic centre of gravity from the western Mediterranean to the north. They also played an important role in the development of ideas on long distance trade as the major incentive to economic growth and especially the rise of the towns.4 Moreover these centres are situated in the heartlands and original home country of the Pippinids and Carolingians.5 If we are looking for barometers of the transformation of the Roman world, these centres are good candidates for research. For Henri Pirenne the real emergence of towns was a feature of the eleventh century as a result of a new increase in commercial activities. This occurred after a period of decline in trade from the early eighth century on, culminating in a almost total eclipse of commercial trade in the ninth century.6 In his opinion the decline of the eighth century was, as we all know, a result of the cutting off of the external markets from the Carolingian Empire as a consequence of the invasion of the Mediterranean by the Arabs. In 1930, however, Felix Rousseau contested this image of Carolingian discontinuity and decline and indicated that the rise of the towns in the Meuse valley in the eleventh century 1 2 3 4 5 6 Verhulst 1999. Maastricht: Theuws 2001; Panhuysen/De La Haye/Gauthier, 2002; Theuws 2004; idem 2005. Namen: Plumier-Torfs/Plumier 1996; eadem 1997; Plumier 1996; Vanmechelen/ Mees/Robinet/Plumier 2001. Pirenne 1925 [1980]. Rousseau 1930. Werner 1980. Pirenne 1925 [1980]. 154 Frans Theuws Fig. 1. Plan of early medieval Namur with the Merovingian vicus at the confluence of the Meuse and Sambre and the castrum above, as well as major excavation sites was a logical outcome of a continuous organic development of these since Merovingian times.7 Although a number of Felix Rousseau’s basic ideas has been contested since, the idea of continuity has since never been a matter of debate again. It is taken as a given. The debate concentrated on the structural characteristics of the growth of these centres in Merovingian and Carolingian times. For Felix Rousseau this was clear: They were ‘étapes de batellerie’, landing stations at regular intervals along the river. In his view the early medieval towns owed their existence to the long distance trade along the river Meuse, part of extensive trade networks from the Mediterranean to the Baltic. In his view this system was mainly run by the well-known Frisian traders. Research by Georges Despy showed that there was no evidence in the written sources to substantiate either the role of the Frisians in the Meuse valley in the seventh to ninth centuries or the existence of long distance trade routes along the Meuse river.8 He pointed out that the local and regional economy and exchange should be considered as important. Further research by Jean-Pierre Devroey substantiated this model, and he stressed the 7 8 Rousseau 1930. Despy 1968. Where is the eighth century in the towns of the Meuse valley? 155 role of the Middle Meuse centres in a limited region, with interregional trade mainly directed to the north rather than to the south across the Ardennes forest.9 And following an idea by Stephan Lebecq, he argued (I think succesfully) for the existence of two separate trade systems along the rivers Meuse and Rhine, which were mainly or even only connected to each other at places where they meet just before they reached the sea in the middle-Dutch river area.10 One important connection between the two may however have existed: the old Roman road from Cologne to Northern France. Where it crosses the river Meuse we find the town of Maastricht. But before we go into the archaeology of this place and others let me stress that since Felix Rousseau in the 1930s no one seriously contested the idea that the Meuse towns came into existence as a result of a continuous and, above all, organic development since early Merovingian or even late Roman times.11 However, as Adriaan Verhulst also pointed out, they were at best local and regional trade centres until they once again became internationally oriented in the tenth century.12 However, Maastricht may have been an exception to this, having been an international trade centre at all times. How does such an image of continuity come into existence? Probably because of the connection of individual pieces of evidence scattered through time and space into a single narrative. At first sight such points in time are a perfect illustration of a continuous development of economic and bureaucratic centres. However, a series of major transformations hidden below the surface of scattered historical evidence should be part of this constructed narrative of continuity and economic growth. And that is the moment when archaeology comes in. Recent excavations in the town of Namur, especially in that part of the town where the early medieval vicus is situated, exposed an interesting sequence of layers related to habitation, artisanal and commercial activities and the creation of waterfront constructions.13 The Merovingian vicus occupies the triangular terrain at the confluence of the Rivers Meuse and Sambre (Fig. 1). On the left bank of the Sambre Merovingian burials and some indications of habitation have been found. Several excavations took place at the confluence: one in the court and cellars of the Saint-Gilles hospital, one at the site of the ancient chapel of Saint-Hilaire and one on the site Grognon.14 The comprehensive report on the excavations is about to appear but the preliminary reports presented by Jean Plumier, Raphael Vanmechelen and others already allow us to establish the longterm history of the site. In Roman times a terrace, which was occupied from the early 9 10 11 12 13 14 Devroey 1991. Lebecq 1983. Verhulst 1999, 43. Ibid., 47-51. Vanmechelen/Mees/Robinet/Plumier 2001. Saint-Gilles: several contributions by Plumier-Torfs and Plumier in: Plumier 1996a. SaintHilaire: idem 1996b; idem 1997. Grognon: Vanmechelen/Mees/Robinet/Plumier 2001. 156 Frans Theuws Fig. 2. Excavation locations in the town centre of Maastricht Roman period into the Middle Ages, was created. This terrace slopes down to the River Meuse. The Grognon excavations showed that already in Roman times, limited efforts were made to improve the waterfront. In the sixth century artisans occupied the terrace. Clear indications of bronze, bone and antler working have been found at several places, suggesting that artisanal production took place in a special artisanal quarter at a scale surpassing local demand. At the same time the waterfront was improved at a much greater scale than in Roman times. Yet these waterfront constructions changed in nature in the seventh century, when the first port-like constructions were made in order to create a landing place for river boats. The excavators suggest that a change in the character of the site from artisanal production to trading is responsible for this. These observations are in accordance with the traditional opinion that Namur is an important regional trading place in the seventh century. It is one of the centres where gold tremisses are minted at that time.15 The excavations further show that around 700 a major transformation in the use of the site took place. No artisanal activities have been found from the eighth and early 15 I am not suggesting here that the minting of coins is exclusively related to commercial activities. See Devroey 2003, 158-169; Theuws 2004. Where is the eighth century in the towns of the Meuse valley? 157 Fig. 3. Plan of Maastricht in 1830 on which the location of the Roman fortress and other later topographical elements are indicated. 1. Saint-Servatius church; 2. church of Our Lady; 3. Saint Martin’s church (Merovingian?); 4. Late Roman fortress; 5. Roman bridge; 6. Saint-Servatius bridge (thirteenth century). Market places: a. Vissersmaas; b. Houtmaas; c. Graanmarkt; d. Kersenmarkt; e. Moesmarkt; f. macellum (meat-hall); g. Varkensmarkt; h. Zaterdagmarkt and Houtmarkt ninth century yet. The waterfront constructions decayed and were not renewed or repaired. No structures indicating habitation have been discovered. Instead a thick black layer, in which a number of human burials are found, was formed. They had various orientations and some graves showed an unfamiliar burial ritual. The irregularity of the burials as well as their location in the mud of a decaying waterfront is not what we expect to find in a thriving Carolingian vicus. The triangular area at the confluence of the Meuse and Sambre, however, has not been entirely excavated yet. It is possible that the location of commercial activities was restricted to the (as yet unexcavated) right bank of the River Sambre. This situation existed until the middle of the ninth century, when new waterfront constructions were made, a chapel was built on the terrace, and a parcelling of the terrace and domestic activities seem to have taken place. The chapel blocked the old Roman road that was probably in use in the early Middle Ages as well; this shows that a change in the use of the site had occured since Merovingian times. In the beginning of the tenth century, a completely new waterfront was created. It appears that two major transformations in the use of the site took place, one around AD 700 and one around AD 850. What happened in between these dates is an 158 Frans Theuws Fig. 4. Maastricht, draft of a map with excavation locations where structural elements (buildings, ditches etc.) (closed dots), finds (open dots), artisanal activities (Û) and burials (+) are found from the Merovingian period (until ca. AD 700) enigma. There seem to be hardly any indications of artisanal or commercial activities at this site in this period. Were these activities moved to another location – for instance, to the bank of the River Sambre? Did these activities stop altogether at Namur in the Carolingian period? There must have been continuity of habitation to some extent, as the burials and the finds from the black layer indicate this. We have to await the final report to evaluate these finds. One conclusion can already be drawn: There is no such thing as a continuous organic development from the Merovingian vicus with its artisanal activities and coin production to the tenth century portus. Is this just a local phenomenon in Namur? The evidence from Maastricht points to a more general development. In Maastricht a large number of excavations has been carried out, but not many of them have been published in great detail (Fig. 2).16 Nevertheless, a general picture seems to emerge. Part of this picture is a daring hypothesis about the location of a Carolingian fortress. 16 Theuws 2005. Where is the eighth century in the towns of the Meuse valley? 159 Fig. 5. Maastricht, draft of a map with excavation locations where structural elements (buildings, ditches etc.) (closed dots), finds (open dots), artisanal activities (€) and burials (+) are found from the Carolingian period (until ca. AD 900) Maastricht is the location where the Romans built a bridge that was kept in use until the thirteenth century. The bridge was, since late Roman times, protected by a small fortress on the left bank of the Meuse (Fig. 3). This fortress and the bridge determine to a large extent Maastricht’s early medieval history, together with two major churches: Saint-Servatius on a low hill to the west of the fortress and Our Lady inside of the fortress. Both churches date to the Merovingian period and may have been built in the context of the elevation of Maastricht as a bishop’s seat. This was probably in the first half of the sixth century, after a period when no bishops were present in the civitas of which the town of Tongres was the capital. Maastricht like Namur developed into a thriving vicus in the sixth and seventh centuries. Excavations show that all over the modern town centre there was habitation (Fig. 4). Significant remains of pottery production, bronze working, leather working, glass making, bone working and antler working have been found. Maastricht is also an important minting place; thirteen monetarii from the seventh century are known by name.17 Unfortunately it has up till 17 Pol 1995. 160 Frans Theuws now not been possible to investigate the Merovingian waterfront, and we have to fear that it has for the most part been destroyed. Maastricht must have been of great importance to the Pippinids and Carolingians.18 In Carolingian times a splendid new basilica of Saint Servatius was built and the cult centre was transformed into an an abbey.19 One would expect to find many archaeological indications of a flourishing trading and artisanal settlement at that time, the existence of which is indicated around 830 by Einhard, Charlemagne’s biographer and abbot of Saint Servatius.20 However, no indications of any artisanal activity are found up till now. What remains, as far as we can make a proper judgment on the basis of the present state of knowledge, is a narrow strip of habitation on both sides of the Roman road west of the river Meuse (Fig. 5). The number of sites with remains from the eighth and early ninth century is very small in relation to the Merovingian period. No structural remains of habitation of any significance at the site of the old Merovingian vicus have been found. Again, layers of black earth developed in some locations. Their meaning is highly enigmatic. Maastricht, too, went through a major transformation after ca. AD 700. Again we have to ask: Were artisan and commercial activities moved to another place? (For instance, to the other side of the river where today a quarter is situated named Wyck, a name that might have been derived from vicus.) Or did these activities stop? This is unlikely in view of the written evidence. In 779 Maastricht was mentioned as one of five important toll stations in the northern Frankish Empire.21 This information however is not in itself proof of a continuous, organic development. It was concluded that in Namur a settlement phase started around the middle of the ninth century, which may have marked the beginnings of the development of Namur as a town. In Maastricht it seems clear that a fortress was built that included the abbey of Saint-Servatius. Probably it was a royal fortress.22 Later in the ninth century, Maastricht was indicated as a municipium (a fortified settlement) in the capitularies.23 This fortress has not been discovered up till now, but I suggest that it is to be found to the west of the location of the old Merovingian vicus (Fig. 6). We have to realize that before the fortress was built, the settlement on the west bank of the river seems to have been reduced in size and importance to a considerable degree in Carolingian times. Now we have arrived at the unexpected conclusion that two major centres in the Middle Meuse valley show serious transformations shortly after AD 700. There seems to be no continuous organic development of these centres from the Merovingian period to the tenth century. Surprisingly, the chronology seems to fit Pirenne’s model in some 18 19 20 21 22 23 Panhuysen/Leupen 1990. Theuws 2001. Panhuysen 1991; Panhuysen/De La Haye/Gauthier 2002. Einhard, Translatio, c. 4,13. Lebecq 1983, I, 116; II, 418-419. Leupen 1996. Van Ommeren 1991, 35 (854), nr 67; 36 (871), nr 71; 41 (908), nr 90. Where is the eighth century in the towns of the Meuse valley? 161 Fig. 6. Maastricht, possible location of a castrum and areas built up in Carolingian times as well as the different possible locations of a royal palace (small polygons) suggested in the literature since the nineteenth century, and the location of a royal estate centre (large polygon) way better than that of modern historiography.24 This however does not force us to accept Pirenne’s explanation for the decline of the centres (Arab invasions) and the development of an autarkic Carolingian Empire.25 Historians and archaeologists alike refer to the mushroom towns like Quentowic and Dorestad to illustrate that intensive commercial activities continued uninterrupted in this period. These centres developed at the same time that Maastricht and Namur were transformed and seem to have declined. However, it might be too hasty to draw the conclusion that a new type of trade centres (emporia) was simply replacing an older one. Quentowic and Dorestat were totally different centres and their rise cannot be regarded unproblematically as a continuation of the trade that made centres such as Maastricht, Namur and others flourish in the seventh century. Moreover Dorestad and Quentovic cannot be regarded as exemplary of the development of Carolingian exchange in general. They are exceptional features in a liminal position related to cross-cultural trade; different systems of ideas, norms and values were articulated in these centres.26 They do not seem to have formed part of the 24 It must, however, be stressed that Verhulst classified them as regional centres rather than international trade centres. He also emphasises their continuous development. 25 See also the comments on this complex field of problems by Devroey 2003, 148-175. 26 See for the problems involved in cross-cultural trade in this context: Theuws 2004. 162 Frans Theuws normal system of Christian exchange of the Frankish kingdom, as I recently explained. The intimate relationship between exchange and religion was absent in the emporia where no major cult places are found. One point of critique on the image of emporia dominating the exhange of the Carolingian age is that the inland exchange centres such as Maastricht have no place in this model. But the critics’ image of inland centres that develop continuously and organically also does not seem to hold in a number of important cases. That is why I indicated before the eighth and early ninth centuries as ‘an age of experiments’: We hardly understand the complexities of the whole sphere of Carolingian production and exchange, in which formal exchange based on emerging market principles coexisted with informal modes of exchange based on social relations, religious belief and cultural attitudes. The intimate relation between religion, exchange and production that existed in Merovingian times may have been given form in a different way in Carolingian times – for instance, by moving production to monastic sites.27 But were commercial activities also moved there? Stephane Lebecq indicates that they did, especially to the large monasteries outside major centres such as Paris and Trier.28 We have to make a clear distinction between what happened to exchange activities as well as artisanal and agrarian production. And finally we have to realize that centres such as Namur and Maastricht lay in the heartlands of Carolingian power and that they were transformed at exactly the same time the Pippinids and Carolingians gained control of the kingdom. At the same time a process of ruralisation of the elite took place. But this ruralisation does not automatically imply autarky, as Henri Pirenne suggested, although in general the material culture of rural settlements in the southern Netherlands (to some extent the hinterland of Maastricht) was relatively poor in comparison to that of the Merovingian period.29 Farmyards were less complex, indicating a changed (and reduced level of) household production as well. Could it be that centres such as Maastricht and Namur not only were transformed as a result of economic and social changes, but also because of important changes in aristocratic mentalities related to this ruralisation, to the extent that they had no need for the old centres? There is still a long road ahead before we really understand the articulation of Carolingian social, economic and religious transformations. 27 Hodges 2000; Lebecq 2000. 28 Ibid., 148. One wonders what happened to these central places themselves in Carolingian times. 29 This information is based on my own observations. Unfortunately not many of the excavated settlements have been published in sufficient detail to deal with this problem extensively. Imported ceramics from the Rhineland do, however, turn up at almost every site. But most of the time that is all. Where is the eighth century in the towns of the Meuse valley? 163 Bibliography Despy 1968: Georges Despy, “Villes et campagnes aux IXe et Xe siècles: l’exemple du pays mosan”, in: Revue du Nord 50, 1968, pp. 145-168. Devroey 1991: Jean-Pierre Devroey, ,,Villes, campagnes, croissance agraire dans le pays Mosan avant l’an mil: vingt ans après”, in: Jean-Marie Duvosquel/Alain Dierkens (Eds), Villes et campagnes au moyen âge. Mélanges Georges Despy, Liège 1991, pp. 223-260 (reprinted in: Jean-Pierre Devroey, Études sur le grand domaine carolingien, Aldershot 1993). Devroey 2003: Jean-Pierre Devroey, Economie rurale et société dans l’Europe franque (VIe-IXe siècles) 1: Fondements matériels, échanges et lien social, Paris 2003. Einhard: Georg Waitz, “Einhard, Translatio et Miracula S.S. Marcellini et Petri”, in: Monumenta Germaniae Historica, Scriptores 15, Hannover 1887, pp. 238-264. Hodges 2000: Richard Hodges, Towns and Trade in the Age of Charlemagne, London 2000. Lebecq 1983: Stéphane Lebecq, Marchands et Navigateurs frisons du haut moyen âge, I en II, Lille 1983. Lebecq 2000: Stéphane Lebecq, ,,The role of monasteries in the systems of production and exchange in the Frankish world between the seventh and the beginning of the ninth centuries”, in: Inge Lyse Hansen/Chris Wickham, The long eighth century, Leiden 2000, pp. 121-148. Leupen 1996: Piet Leupen, ,,Viking-age Raids and Urban Settlement on the Rivers Rhine and Meuse in the 9th century”, in: Lars Nilsson/Sven Lilja (Eds), The Emergence of Towns: Archaeology and Early Urbanization in Non-Roman, North-West Europe, Stockholm 1996, pp. 79-94. Ommeren 1991: H.R. van Ommeren, ,,Bronnen voor de geschiedenis van Maastricht (3591204) 1: bronnen betreffende het tijdvak vanaf het jaar 359 tot en met 923”, in: Publications de la Société Historique et Archéologique dans le Limbourg 127, 1991, pp. 5-48. Panhuysen/Leupen 1990: T. A. S. M. Panhuysen/P. H. D. Leupen, ,,Maastricht in het eerste millenium: de vroegste stadsontwikkeling in Nederland”, in: Ontstaan en vroegste Geschiedenis van de middeleeuwse steden in de zuidelijke Nederlanden: een archeologisch en historisch probleem, Brussels 1990, pp. 411-455. Panhuysen 1991: T. A. S. M. Panhuysen, ,,De Sint-Servaaskerk te Maastricht in de vroege middeleeuwen: voorlopig eindverslag van de opgravingen door de dienst Stadsontwikkeling Maastricht in de periode 1981-1989”,in: Bulletin KNOB, 1991, pp. 15-24. Panhuysen/De La Haye/Gauthier 2002: T. A. S. M. Panhuysen/R. De La Haye/Nancy Gauthier, ,,Maastricht”, in: Nancy Gauthier/Jean-Charles Picard, Topographie chrétienne des cites de la Gaule des origines au milieu du VIIIe siècle, 12, Province ecclesiatique de Cologne, Paris 2002, pp. 91-115. Pirenne 1925: Henri Pirenne, Medieval Cities: their Origins and the Revival of Trade, Princeton 1925 [1980]. Plumier 1996a: Jean Plumier (Ed.), Cinq années d’archéologie en province de Namur 1990-1995, Namur 1996. Plumier 1996b: Jean Plumier, ,,La chapelle Saint Hilaire”, in: Jean Plumier (Ed.), Cinq années d’archéologie en province de Namur 1990-1995, Namur 1996, pp. 95-98. Plumier 1996c: Jean Plumier, ,,Le port et son quartier”, in: Jean Plumier (Ed.), Cinq années d’archéologie en province de Namur 1990-1995, Namur 1996, pp. 81-86. Plumier 1997 : Jean Plumier, ,,Namur : La chapelle Saint Hilaire”, in: Marie-Hélène Corbiau (Ed.), Le patrimoine archéologique de Wallonie, Namur 1997, pp. 484-486. 164 Frans Theuws Plumier-Torfs/Plumier 1996: Sophie Plumier-Torfs/Jean Plumier, ,,L’époque merovingienne”, in: Jean Plumier (Ed.), Cinq années d’archéologie en province de Namur 1990-1995, Namur 1996, pp. 91-94. Plumier-Torfs/Plumier 1997 : Jean Plumier/Sophie Plumier-Torfs, ,,Namur: L’artisanat mérovingien au <Grognon>, in: Marie-Hélène Corbiau, (Ed.), Le patrimoine archéologique de Wallonie, Namur 1997, pp. 422-423. Pol 1995: A.Pol, ,,Les monetaires à Huy et Maastricht: Production et distribution des monnaies mérovingiens mosanes”, in: Bulletin de l’Instituut Archéologique Liègois 107, 1995, pp. 185-200. Rousseau 1930: Félix Rousseau, La Meuse et le pays mosan en Belgique: leur importance historique avant le XIIIe siècle, Brussels 1930. Theuws 2001: Frans Theuws, ,,Maastricht as a centre of Power”, in: Mayke de Jong/Frans Theuws/Carine van Rhijn (Eds), Topographies of Power in the Early Middle Ages, LeidenBoston-Köln 2001, pp. 155-216. Theuws 2004: Frans Theuws, ,,Exchange, religion, identity and central places in the Early Middle Ages”, in: Archaeological Dialogues 10, 2004, pp. 121-138, 149-159 (with comments by Richard Hodges , pp. 138-144 and John Moreland, pp. 144-149). Theuws 2005: Frans Theuws, ,,Drie modellen voor de ontwikkeling van het middeleeuwse Maastricht”, in: Reinout Rutte/Hildo van Engen (Eds), Stadswording in de Nederlanden: op zoek naar overzicht, Hilversum 2005, pp. 87-122. Vanmechelen/Mees/Robinet/Plumier 2001: Raphaël Vanmechelen/Nathalie Mees/Caroline Robinet/Jean Plumier, ,,Namur/Namur: évolution du bord de Meuse au Grognon (Ive-Xie siècle)”, in: Chronique de l‘Archéologie wallonne 9, 2001, pp. 217-220. Verhulst 1999: Adriaan Verhulst, The rise of Cities in North-west Europe, Cambridge 1999. Werner 1980: Matthias Werner, Der Lütticher Raum in frühkarolingischer Zeit: Untersuchungen zur Geschichte einer karolingischen Stammlandschaft, Göttingen 1980.