Roman Frontier Studies 2009

ROMAN FRONTIER STUDIES 2009 Proceedings of the XXI International Congress of Roman Frontier Studies (Limes Congress) held at Newcastle upon Tyne in August 2009 Access edited by

Nick Hodgson, PaulOpen Bidwell and Judith Schachtmann

Archaeopress

Archaeopress Roman Archaeology 25

ISBN 978 1 78491 590 2 ISBN 978 1 78491 591 9 (e-Pdf) © Archaeopress and the authors 2017

Front cover illustration: The reconstructed south-west gate, South Shields Roman Fort, Great Britain Back cover illustrations: The site of the Roman bridge (Pons Aelius) Accessacross the river Tyne at Newcastle; inscription (RIB 1322) found in the Tyne at Newcastle, recording the transfer of legionary detachments between Britain and Germany, c. 158 Open

All rights reserved. No part of this book may be reproduced, in any form or by any means, electronic, mechanical, photocopying or otherwise, Archaeopresswithout the prior written permission of the copyright owners.

Printed in England by Oxuniprint, Oxford This book is available direct from Archaeopress or from our website www.archaeopress.com

Foreword ��ix Introduction ��xi A Record of the Congress ��xii Acknowledgements ��xv Attendees ��xvi

Women and Families in the Roman Army Session organisers: Carol van Driel-Murray, Martina Meyr, Colin Wells

Women, the military and patria potestas in Roman Britain �� 3 Lindsay Allason-Jones

Beyond von Petrikovits – artefact distribution and socio-spatial practices in the Roman military �� 9 Penelope Allison

Some thoughts about the archaeological legacy of soldiers’ families in the countryside of the civitas Batavorum ��Access 16 Harry van Enckevort

The families of Roman auxiliary soldiers in the military diplomas �� 23 Elizabeth M. Greene Open British families in the Roman army: living on the fringes of the Roman world �� 26 Tatiana Ivleva

Women and children in military inscriptions from northern Germania Superior �� 34 Michael J. Klein

The empress and her relationship to the Roman army �� 42 Kai M. Töpfer

Women and children at the Saxon Shore fort of Oudenburg (Belgium) �� 48 S. Vanhoutte and A. VerbruggeArchaeopress

Roman Roads Decem Pagi at the end of antiquity and the fate of the Roman road system in eastern Gaul �� 55 Joachim Henning, Michael McCormick and Thomas Fischer

The planning of Roman Dere Street, Hadrian’s Wall, and the Antonine Wall in Scotland �� 62 John Poulter

Some notes on the development of the military road network of the Roman Empire �� 69 Zsolt Visy

i © Archaeopress and the authors, 2016. The Roman Frontier in Wales Session organisers: Barry Burnham, Jeffrey Davies

Rewriting The Roman Frontier in Wales: an introduction �� 81 Barry C. Burnham and Jeffrey L. Davies

Recent work on the site of the legionary fortress at Caerleon �� 85 Peter Guest and Tim Young

Roman roads in Wales �� 93 R. J. Silvester

The Cadw-grant-aided ‘Roman Fort Environs Project’ – the contribution of geophysics �� 99 David Hopewell

Roman frontiers in Wales: 40 years on �� 104 Jeffrey L. Davies

The military ‘vici’ of Wales – progress since Jarrett 1969 �� 111 Barry C. Burnham

The Eastern and North African Frontiers Access Session organisers: James Crow, Eberhard Sauer

Transformation patterns of Roman Forts in the Limes Arabicus from Severan to Tetrarchic and Justinianic periods �� 121 Ignacio Arce Open

Recent research on the Anastasian Wall in Thrace and late antique linear barriers around the Black Sea ��131 James Crow

New research on the Roman frontier in Arabia �� 139 S. Thomas Parker

The archaeology of Sasanian frontier troops: recent fieldwork on frontier walls in Northern Iran �� 145 Hamid Omrani Rekavandi, Eberhard Sauer, Tony Wilkinson and Jebrael Nokandeh

Soldiers or tribesmen: whoArchaeopress guarded the frontiers of late Roman Africa? �� 151 Alan Rushworth

Roman-Armenian borders, part I: the Upper Euphrates frontier �� 160 Everett L. Wheeler

Smaller Structures: Towers and Fortlets Session organisers: Bill Hanson, Matt Symonds

Bauliche und funktionale Gliederung des Obergermanisch-Raetischen Limes anhand der Turmgrundrisse �� 171 Thomas Becker

A Roman road station on the Pannonian limes �� 180 Szilvia Bíró

ii © Archaeopress and the authors, 2016. River frontiers or fortified corridors?. �� 186 Erik Graafstal

A battle of wills: manoeuvre warfare and the Roman defence of the North Yorkshire coast in the late C4th ��194 Alistair McCluskey

The Castelinho dos Mouros (Alcoutim) and the ‘casas fuertes’ of southern Portugal �� 200 Thomas Schierl, Felix Teichner, Gerald Grabherr, Alexandra Gradim

Smaller structures on Hadrian’s coastal frontier �� 207 Matthew F. A. Symonds

Roman towers �� 213 David Woolliscroft

Recognising Differences in Lifestyles through Material Culture Session organisers: Stefanie Hoss, Sonja Jilek, Eckhard Deschler-Erb

La céramique « militaire » dans le Nord de la Gaule de la Conquête au début du IIe siècle après J.-C.: Faciès et particularités �� 221 Cyrille Chaidron, Raphaël Clotuche et Sonja Willems Access Auxiliaries and their forts: expression of identity? �� 229 Julia Chorus

Military versus civilian and legionary versus auxiliary: the case of Germania Inferior �� 236 Stefanie Hoss Open

Die zivile Nutzung militärischen Baumaterials – Kontexte und Interpretation �� 241 Thomas Schmidts

Barbaricum Session organiser: Thomas Grane

Barbaricum: an introduction to the session �� 251 Thomas Grane Archaeopress An imported bronze casket from the Przeworsk culture cemetery in Lachmirowice, distr. Inowrocław ��260 Katarzyna Czarnecka

Multifunctional coins – a study of Roman coins from the Zealandic isles in eastern Denmark �� 267 Mads Drevs Dyhrfjeld–Johnsen

Medical instruments, tools and excavation locations – ‘The reason why…’ �� 273 Annette Frölich

Patterns in cross-frontier relations �� 278 Marjan C. Galestin

Bemerkungen zu den Formen des Zustroms der Importgüter in das germanische Siedlungsmilieu während der Römischen Kaiserzeit im mittleren Donauraum �� 284 Balázs Komoróczy

iii © Archaeopress and the authors, 2016. Römische Bronzegefäßgarnituren Römischer und germanischer Fundkontext im Vergleich und deren jeweilige Aussage: Wo, wann, wie, warum, wer, für was? Südskandinavien und die römischen Provinzen ��292 Ulla Lund Hansen

The C3rd AD Romano-Germanic battlefield at Harzhorn near Kalefeld, Landkreis Northeim �� 298 Michael Meyer, Felix Bittmann, Michael Geschwinde, Henning Haßmann, Petra Lönne and Günther Moosbauer

Hacksilber inside and outside the late Roman world: a view from Traprain Law �� 304 Kenneth Painter and Fraser Hunter

Why are the South Scandinavian weapon deposits relevant for limes research? An update of research progress �� 311 Xenia Pauli Jensen

Corpus der römischen Funde im europäischen Barbaricum – Rückblick und Ausblick �� 318 Hans-Ulrich Voss und Claus-Michael Hüssen

Britain Hadrian’s Wall and the Mommsen thesis �� 329 David J. Breeze

Continuing the search for an ‘Antonine gap’ on Hadrian’s Wall ��Access 331 R. J. Brickstock

A late Roman military command in Britain reinstated �� 336 Roger White Open A new Roman fort at Staxton in the Vale of Pickering, Yorkshire, England �� 342 Pete Wilson

Roman Frontiers in a Globalised World Session organisers: Richard Hingley, Divya Tolia-Kelly, Rob Witcher

Does history repeat itself?- �� 349 The Roman frontiers from the viewpoint of a European archaeologist of today �� 349 Eduard Nemeth Archaeopress

The attraction of opposites: Owen Lattimore and studies of the Inner Asian frontiers of China �� 357 Naomi Standen

Changing presents interpret the past, AD 1500-2010: the frontier on the Limes and the upper Danube �� 365 Peter S. Wells

Civil Settlements Session organisers: Edward Dąbrowa, Pete Wilson

Military colonization in the Near East and Mesopotamia under the Severi �� 373 Edward Dąbrowa

iv © Archaeopress and the authors, 2016. The Canabae Legionis of Carnuntum: modelling a Roman urban landscape from systematic, non-destructive prospection and excavation �� 379 Christian Gugl, Michael Doneus and Nives Doneus

Neues vom Vicus der Saalburg �� 386 Cecilia Moneta

Viminacium – Roman city and legionary camp: topography, evolution and urbanism �� 393 Nemanja Mrđić and Bebina Milovanović

The veterans’ colony Aequum, the legionary fortress Tilurium and the Sinj field. Re-examining old problems �� 401 Mirjana Sanader

Death and Commemoration Session organiser: Maureen Carroll

Some aspects of death, ritual and commemoration in the Lower Rhineland (Germany) �� 409 Clive Bridger

Dress, self and identity in Roman funerary commemoration on the Rhine and Danube frontiers �� 415 Maureen Carroll Access The Funerary commemoration of veterans and soldiers at the colony of Augusta Emerita (Mérida, Spain), 25 BC – AD 235 �� 421 Jonathan Edmondson Open The Roman cemetery at Pottenbrunn. Structural analysis of a rural necropolis �� 430 Eva Hoelbling

The Roman cemetery at Mannersdorf am Leithagebirge, Lower Austria: The late antique inhumations as an information source of the population of the C4th and C5th �� 435 René Ployer

Ein neue Grabinschrift für einen Soldaten der legio VI Victrix in Novaesium/Neuss �� 442 Marcus Reuter

The Decoration of some EarlyArchaeopress Imperial Tombs of Primi Pili �� 447 Kai M. Töpfer

Danubian and Balkan Provinces Stories and facts about the function of Dacia’s south-eastern frontier: forty years of research �� 455 I. Bogdan Cătăniciu

Maximinus Thrax in Novae �� 461 Piotr Dyczek and Jerzy Kolendo

A contribution to the study of the Roman Limes in the Croatian Danube region �� 466 Mato Ilkić and Daška Osonjački

Overlapping phases in the defensive systems of the Roman forts and the archaeological experience: the case of Roman Dacia �� 470 Dan Isac

v © Archaeopress and the authors, 2016. Detail eines römischen Kellers aus dem Vicus von Aquincum – Víziváros (Budapest) �� 478 Katalin H. Kérdő

Vindobona fortress – barracks, fabrica and intervallum �� 483 Martin Mosser

The Late Roman Principia in Tarsatica, part of Claustra Alpium Iuliarum �� 488 Josip Višnjić and Luka Bekić

Romuliana – Gamzigrad in der Provinz Dacia ripensis. Kaiserpalast und Militärstation �� 495 Gerda von Bülow

The Army in the Hinterland – a case study of Pons Aeni/Pfaffenhofen �� 502 Meike B Weber

The Scythian section of Notitia Dignitatum: a structural and chronological analysis �� 509 Mihail Zahariade

Camps Session organiser: Rebecca Jones

What is a Roman camp? ��Access 521 Rebecca H. Jones

The marching camp at Deer’s Den, Aberdeenshire: a précis of the excavations �� 531 Murray Cook Open Römische Feldlager aus der Zeit der Markomannenkriege in der Slowakei �� 537 Ján Rajtár und Claus-Michael Hüssen

GIS application in Roman military invasion survey within barbarian territories during the Marcomannic wars – introduction into problems and perspectives �� 545 Balázs Komoróczy and Marek Vlach

The Roman Republican battlefield at Pedrosillo (Casas de Reina, Badajoz, Spain): new research (2007) �� 552 Ángel Morillo, Germán Rodríguez Martín and Esperanza Martín Hernández

Remains of the Roman baggageArchaeopress train at the battlefield of Kalkriese �� 559 Achim Rost

The function of temporary camps along Hadrian’s Wall �� 565 Humphrey Welfare

The battlefield of Kalkriese: The rampart at the site ‘Oberesch’ during and after the battle �� 571 Susanne Wilbers-Rost

Logistics and Supply Session organisers: Bill Hanson, Valerie Maxfield

Voorburg-Arentsburg: a Roman harbour with a British connection in the hinterland of the Limes �� 579 Mark Driessen

vi © Archaeopress and the authors, 2016. The grain supply for the Roman army in Hispania during the Republican period �� 586 Javier Salido Domínguez

Die römischen Steinbruchinschriften des Brohltals �� 593 Markus Scholz (unter Mitarbeit von Holger Schaaff)

A sustainable frontier? Timber supply for the Roman army in the Lower-Rhine delta, AD 40-150 �� 603 Pauline van Rijn

The Germanies and Augustan and Tiberian Germany Session organiser: C. Sebastian Sommer

Lahnau – Waldgirmes. Die Ausgrabungen 2007 – 2009 �� 611 Armin Becker

Neue Luftbilder zu den Militärlagern und den canabae legionum von Vetera castra I (Xanten) �� 616 Norbert Hanel und Baoquan Song

The Augustan legionary camp on the Hunerberg in Nijmegen (NL) revised New information and re-interpretation of old data of the defence system �� 619 Elly N. A. Heirbaut Access New thoughts on the so-called temple of Mars in the legionary camp of Vindonissa �� 625 Andrew Lawrence

Iupiter im Brunnen – Neues zur siedlungsgeschichtlichen Entwicklung im Nordvicus von Heidelberg �� 631 Petra Mayer-Reppert Open

The Roman military presence in the Rhine delta in the pre-Flavian period �� 636 Marinus Polak

Quarries on the Raetian Limes, the height and construction of the wall �� 642 C. Sebastian Sommer

Spain Archaeopress The Roman fort in El Real (Campo de Criptana, Ciudad Real, Spain) �� 653 Antxoka Martínez Velasco

The Cantabrian Wars (26-25 BC campaigns): contesting old interpretations �� 659 Ángel Morillo

Frontier Fleets Session organisers: Boris Rankov, Jorit Wintjes

Antiqua ... Arte Cilix (Lucan., Phars. 4. 449) �� 669 Siniša Bilić-Dujmušić

Ultro Citroque Discurrere – operational patterns and tactics of Late Roman frontier fleets on rivers �� 674 Florian Himmler

vii © Archaeopress and the authors, 2016. Project Exploratio Danubiae – new insights into troop transport on the river Danube in the Late Roman Period �� 679 Heinrich Konen

The frontier fleets: what were they and what did they do? �� 687 Boris Rankov

The northern fleets in the Principate �� 691 Christoph Rummel

Did the Romans have a fleet on the Red Sea? �� 696 Denis B. Saddington

The ghost fleet of Seleucia Pieria �� 699 Jorit Wintjes

Miscellaneous Contributions Wells and ritual deposition at the Newstead roman military complex �� 705 Simon Clarke

A cost-control model for Imperial frontiers? �� 711 Raphael M. J. Isserlin Access

Der Soldat und die Götter – wie privat war Religion? �� 719 Nina Willburger Open

Archaeopress

viii © Archaeopress and the authors, 2016. Auxiliaries and their forts: expression of identity?

Julia Chorus

Abstract Little is known about the origin of the pre-Flavian troops in the auxiliary forts in the Lower Rhine delta (The Netherlands). For the greater part this also applies to the transition phase just after the Batavian Revolt, in the early seventies AD. At that time the forts were all built in timber, so hardly any epigraphic evidence shows up during excavations. Can research on building techniques and their possible background fill a gap in the knowledge on the military occupation in the Rhine delta? When comparing the defences and internal buildings of the forts in the research area and also outside that region, the variety of building techniques is striking. Instead of environmental circumstances this must above all be due to the differences between the soldiers. Comparison with late Iron Age building techniques (fortifications, houses) shows some interesting relations between forts, soldiers, their possible recruitment area and building traditions. The ramparts of the forts, for instance, provide information on their builders and seem to show the origin of the builders/soldiers in the fort. They must have built their forts according to the tradition of their ancestors and meanwhile expressed their identity.

Introduction Access During the early forties AD the Roman army built several timber auxiliary forts on the southern bank of the Lower Rhine. Remains of nine timber forts and a fortlet, sometimes well preserved, have been found in the West- Netherlands, in the Rhine delta (Figure 1). There are also indications of a military presence further upstream, but there are hardly any traces to be found. The Rhine has destroyedOpen most of the evidence, except for dredge finds and some structures in Arnhem-Meinerswijk (Bechert and Willems 1995, 65, 77-8; Willems 1986, 329-356). Wood was available in the surroundings and was used for building the forts (Van Rijn in prep.). Only in the last quarter of the C2nd AD were most of these auxiliary forts rebuilt in stone. In this paper I will deal with the timber constructions during the C1st and early C2nd AD.

Archaeopress

Figure 1. Roman forts and fortresses in the research area, the Lower Rhine Area in the Early and/or Middle Roman period. 1: legionary base. 2: auxiliary fort. 3: possible auxiliary fort (Polak 2009, fig. 1).

229 © Archaeopress and the authors, 2016.

Roman Frontier Studies 2009

When comparing the forts in the Rhine delta the variety of building techniques is striking. In this paper I would like to show that this variety in building must above all be due to the differences in ethnicity of the soldiers.

Little is known about the origin of the pre-Flavian troops in the auxiliary forts in the Lower Rhine delta (Alföldy 1968, 4-6; Polak in this volume). Hardly any epigraphic evidence showed up during excavations. In my opinion, research on building techniques and their possible background in terms of building traditions can help to fill a gap in the knowledge on the military occupation of the Rhine delta and also abroad. Building techniques might even tell more and can contribute to discussions on identity. The main question in this paper is: did the auxiliary soldiers express their identity in their forts?

Similarities and differences

In the research area, the Rhine delta, an auxiliary fort was built every 12 to 15 kilometres – and later on even closer. One might expect that the forts, being built in a small area and in a rather short period of time, more or less would have looked alike. There are some similarities, like their shape and their position in the landscape, extremely close to the river and at a crossing of the Rhine and a riverbranch. The internal division in only two parts instead of three, the praetentura and the latera praetorii, is very characteristic for these forts in the West Netherlands. Much more common, especially in a later stage, is a division in three parts like almost all other forts for example in the Rhine and Danube region and in Britain show (Johnson 1987, 264).

Besides other similarities the differences are even more interesting. Comparing the structures of the forts, most striking is the variety of design and building techniques of the ramparts (Figure 2).1 In Utrecht and in the earliest fort in Valkenburg timber corduroy was used to form the foundation for a rampart body of sods with a timber framework to keep the sods together. In Valkenburg the sods formed cheeksAccess at the front and rear side, while in Utrecht the whole body existed of sods. Probably a similar kind of rampart was built in Woerden, Alphen aan den Rijn and in the fortlet of Valkenburg-Marktveld. In these fortifications only the lowest part of the wall, the foundation or – in the case of Alphen aan den Rijn some 30cm of a rampart body was preserved. In some cases, like rebuilding phases in Utrecht, Woerden and Valkenburg, a timber revetment was put at the front side. The rampart body behind the front revetment consisted of sods and a timberOpen framework.

A completely different  Before 40 ± 40 - ± 47 - 70 After 70 kind of rampart was found

Vechten 1a Vechten 1b-e Valkenburg 2 Vechten 2 in Vechten, Bodegraven, Bodegraven Zwammerdam 2 Alphen a/d Rijn 2 a rebuilding phase in Leiden-Roomburg 2? Alphen aan den Rijn and in the rebuilding phase in Valkenburg. Here, the remains belonged to a box    rampart type: the foundations Valkenburg 1 Utrecht 3/4 of these box ramparts varied  ArchaeopressAlphen a/d Rijn 1 Woerden 3?  Woerden 1 Valkenburg-Marktveld substantially: in Vechten two  Utrecht 1 trenches were visible with   postholes and sometimes  timber remains still inside.  The same technique was used

Utrecht 2 Valkenburg 4, 5 in Zwammerdam and possibly Woerden 2 Woerden 3? also in Leiden-Roomburg. In Valkenburg 3 Bodegraven the rampart was built with beams and posts with mortice-and-tenon  joints at the front and rear side. Of a second phase of the Figure 2. Reconstructions of the ramparts of the Roman forts in the Rhine delta fort in Alphen aan den Rijn (drawings R. Reijnen based on author’s data). (heavy posts were found in

1 Vechten: Braat 1939; Van Giffen 1950; Polak in Reddé et al. 2006; Utrecht: Ozinga et al. 1989; Van Giffen, Vollgraff and Van Hoorn 1934-1938; Woerden: Blom and Vos 2008; Bogaers and Haalebos 1983; Bodegraven: Van der Kooij et al. 2005; Zwammerdam: Haalebos 1977; Alphen aan den Rijn: Polak et al. 2004a; Leiden-Roomburg: Polak et al. 2004b; Valkenburg-Marktveld: drawings; Valkenburg: Van Giffen 1948, 1955. Besides these publications new interpretations of the archaeological data of ramparts in Chorus 2008; Chorus in prep.

postholes at the front side and the rear side of the rampart. Also in its second phase Valkenburg showed a front revetment with heavy oak beams and posts with the mortice-and-tenon technique. At the rear side the posts were put either in postholes or in a trench. All these box ramparts were probably filled up with sand and clay. Except for the somewhat smaller and narrower rampart of the fortlet at Valkenburg-Marktveld, all ramparts were between 3 and 3.6m wide, that is 10 to 12 Roman feet (pes monetalis).

Figure 2 shows reconstructions of the ramparts in the Roman forts in the Rhine delta: the vertical ones, the horizontal and the mixture. These ramparts show rather varying building techniques in a short period of time and in a small area. It is also striking that in Valkenburg the ramparts often were rebuilt and always in a different way. Why were there so many different kinds of ramparts? One might think of environmental conditions: maybe the close position of the forts to the river caused problems with water and therefore all those experiments were necessary in order to find the right rampart. But why then was not one type of rampart the best way of defence and why did the soldiers not stick to that type in the whole area? Even after the Batavian Revolt, when all forts were rebuilt, there were still several types of rampart construction in use.

Another reason that one could think of is the availability of wood. Did the builders need more timber for the one type of rampart than for another? It is not plausible that they did, because when counting the necessary amount of wood per type, similar quantities are needed. The types of trees, however, do differ. However, there seems no logic as regards the use of timber, geographically nor chronologically.

Comparing the forts in the Rhine delta with timber forts abroad, further eastward in Germania Inferior, in Germania Superior and Raetia and – to the west – in Britannia, a very similar picture appears. The variety of rampart types is not only limited to the West Netherlands research area. Also abroad there are places in which in each new building phase different types succeeded each other. Access Building traditions

The question on this variety therefore needs a different approach. I think that each kind of rampart indicates a different occupation within the forts. Assuming that in the pre-FlavianOpen period the auxiliaries were led by their own regional leaders (e.g. Alföldy 1968, 111-116) and that the auxiliaries built their own forts at that time (e.g. Campbell 2009, 37; Sommer 1984, 59-60), every new garrison with its own commander built or adapted the forts according to their own practices. The same applies to the ramparts.

Apparently the rules of ‘how to build a fort’ were not very strict. Anne Johnson mentions the freedom in the way of building in her book on Roman forts, when she speaks about the enormous diversity of fort plans in general (Johnson 1987, 319). The diversity of building techniques must above all be due to the differences between the soldiers, differences in origin, in background, in tradition of building. If auxiliary troops could build in their own way within certain limits, where did the knowledge come from?

Until now the type of building construction of forts was only connected with earlier Republican camps, for instance those in Italy and Spain (JohnsonArchaeopress 1987, 245). Most of the excavated camps in Spain seem to have been built in stone (e.g. Morillo and Aurrecoechea 2006, 211-358). Stone was available in that area. The temporary camps that could provide information on timber constructions have hardly survived. The ramparts that have been preserved often consisted of an agger of earth and pebbles. Ramparts and city walls of the Republican Italian peninsula were built in stone (Johnson 1987, 245). The use of wood for defences in Roman contexts only seems to occur in the Northern and North-western provinces of the Roman Empire.

In the search for further possible building influences I looked for late Iron Age building techniques in Europe. Some interesting relations between forts, soldiers, their possible recruitment area and building traditions occurred.

The diversity of rampart types of the Roman forts show a link with ramparts of Late Iron Age fortifications, the oppida, and possibly also with walled enclosures in other regions of that time. Audouze and Büchsenschütz (1992, 90ff), Fichtl (2005, 47ff) and others (most recently Ralston 2006, 48-57) demonstrate that various timber rampart constructions were in use in the oppida of Iron Age Europe. Two groups are distinguished (Figure 3): a horizontal construction with horizontal timbers on the one hand and ramparts with vertical timbers, set in the ground, on the other. The most famous type with horizontal timbers is the murus gallicus. In this rampart stones were kept together by a timber framework at several levels. The vertical ones consisted of posts set in the ground and a revetment at either both front and rear sides or only at the front: the Altkönig-Preist rampart and the Kelheim type. The insides of these so-called Pfostenschlitzmauern were filled up with stones.

Roman Frontier Studies 2009

Vertical Vertical HorizontalHorizontal The distribution of the different types of these ramparts is interesting (Figure 4).

The murus gallicus construction was mainly in use west of the Rhine. A concentration of

the vertical ones can be found in Germany,

the Czech Republic and Slovakia and box ramparts up north in the direction of Poland. In my opinion these concentrations

of different rampart constructions in the

late Iron Age represent different cultural traditions. Altkönig-PreistAltkönig-Preist construction construction Ehrang constructionEhrang construction

Iron Age influences on Roman forts

Elaborating on this point I believe that these traditions are also reflected in the Roman forts. As shown in Figure 2 the reconstructions of the ramparts in the

Roman forts in the Rhine delta can roughly be divided in two traditions: a vertical and a horizontal one.

The various rampart constructions from Kelheim constructionKelheim construction Murus gallicusMurus construction gallicus constructiontheAccess late Iron Age seem to have been decisive

for the specific rampart constructions in

Figure 3. Development of the two types of Iron Age ramparts: horizontal and the Roman forts in the Rhine delta. The vertical construction knowledge of how to build a rampart was (based on Collis 1984 and Audouze and Büchsenschütz 1992). combined with Roman elements like the Openunits of measurement (Roman feet) and

Archaeopress

Figure 4. Distribution of main types of rampart in continental Europe during late La Tène D, 100-20 BC (after Audouze and Büchsenschütz 1992, fig. 50).

the use of Roman building equipment and instruments like the groma. It seems that the forts, the soldiers, their recruitment area and their building tradition are related. On the one hand one could point out Gaul as a recruitment area with the murus gallicus, leaving traces in the forts in Utrecht and Valkenburg (phase 1 and perhaps 3) and probably in earliest forts in Woerden, Alphen aan den Rijn and Valkenburg-Marktveld.

On the other hand Germany and the regions further east could be seen as a recruitment area where the vertical ramparts were used to be built. Maybe the Batavians with their former homelands from this region, where the Chatti lived, are part of this cluster. Possible traces of these recruits can be found in the forts in Vechten, Bodegraven, a second phase in Alphen aan den Rijn and Valkenburg (phase 2). The soldiers in these forts seem to have built according to the tradition of their ancestors.

In addition to this hypothesis a small case-study of Valkenburg can be interesting. As already mentioned, there is hardly any epigraphic evidence for the occupation in the forts during the pre-Flavian period. The only information on that matter was provided by two writing tablets, found in Valkenburg, from an earliest phase context. The tablets show the presence of the third cohors Gallorum, recruited in Gaul and fit in the hypothesis that is just discussed. On one of writing tablets is written, in translation: ‘To Tigernilus, soldier of the cohors III Gallorum equitata’ (Bogaers in Glasbergen 1972, 70-74). A second fragment tells the address: ‘C III G ALLORUM’ (Glasbergen 1972, 74-75). The murus gallicus type of rampart in this earliest phase suits very well with the information on these writing tablets. The troops of this third cohors Gallorum must have build their fort at Valkenburg.

Besides the ramparts other parts of the defences, like angle towers and gates show differences in plan or foundation methods. Also foundations and walls of buildings inside the defences of the forts show varying building techniques. In buildings like barracks, the headquarters and granaries timber uprights in foundation trenches occur on the one hand. On the other hand we see uprights in individual postholes or into sleeperAccess beams that were let into the ground: the mortice-and-tenon joint. These jointing techniques have predecessors in the Iron Age and further back in time. The foundation of posts in postholes and the raised floors of the Roman horrea are known from Iron Age granaries. These often had smaller dimensions but also raised floors.

Another difference between buildings in forts is pointed outOpen by remains of wattle-and-daub walls. Mostly, they show a vertical pattern while sometimes a horizontal way of wattle appears (cf. Glasbergen and Groenman-van Waateringe 1974, 19). Horizontal wattling was also applied to walls of Iron Age and Roman farmhouses.

Again, those differences are recognizable within the research area as well as abroad and, probably, they are indications for diverse building traditions too.

Until now I have particularly focussed on auxiliary forts. However, legionary fortresses turn out to be very diverse in building style and technique in the comparison with the forts outside the Rhine delta as well. The differences between legionary fortresses are as evident as in the case of the auxiliary forts.

In the C1st AD recruits not only came from the Italic peninsula but also from the provinces like parts of Spain, Gaul and Noricum (Campbell 1994,Archaeopress 9; Le Bohec 2005, 85). The hypothesis of building traditions, discussed in this paper, might also be applicable to the legionaries and their fortresses. In the variety in building style and technique one might – also in this kind of military camps – read the variety of homelands of legionaries.

Expression of identity?

Assuming that the various building constructions and techniques that are striking in the auxiliary forts indeed indicate different building traditions of recruited auxiliary soldiers, then many new questions arise. Did these soldiers actually want to make a statement by building according to their ancestral traditions? Did they express their identity by doing so and were they at all aware of their collective identity?

If the soldiers from similar homelands had a collective awareness, did they only use their building knowledge because they didn’t know otherwise? Did they just build out of habit and ‘passively repeat’ their own and their ancestors’ way of building (Wells 1999, 170)? Maybe, they wanted to express their identity, as a group, and maybe we can see the building process as an ‘active re-creation’ of old customs. The Roman army apparently gave that space. Explicitly or not: auxiliaries seem to have expressed their ethnic identity by building their forts.

Acknowledgements

The research project ‘A sustainable frontier? The establishment of the Roman frontier in the Rhine delta’ was funded by the Netherlands Organization for Scientific Research, Biax Consult, Radboud University Nijmegen, the City of Utrecht and the Foundation for Provincial Roman Archaeology. The project team consists of M. Erdrich, M. Polak and the author J. P. Chorus (Radboud University Nijmegen), L. I. Kooistra and P. van Rijn (Biax Consult), C. Cavallo (University of Amsterdam), M. van Dinter (ADC Archeoprojecten), M. Dütting (Hazenberg Archeologie Leiden) and E. P. Graafstal (City of Utrecht).

Bibliography

Alföldy, G. 1968. Die Hilfstruppen der römischen Provinz Germania inferior. Epigraphische Studien, 6. Düsseldorf, Rheinland-Verlag. Audouze, F. and Büchsenschütz, O. 1992. Towns, villages and countryside of Celtic Europe. London, Book Club Associates/B. T. Batsford. Bechert, T. and W. J. H. Willems (eds) 1995: Die römische Reichsgrenze zwischen Mosel und Nordseeküste. Stuttgart, Theiss. (Dutch edition: De Romeinse Rijksgrens tussen Moezel en Noordzeekust. Utrecht, Matrijs). Blom, E., Lanzing, J. J. and Vos, W. K. 2004. Woerden-Kerkplein, deelgebied B en C en het schip de ‘Woerden 7’. ADC-Rapport, 254. Amersfoort, ADC-ArcheoProjecten. Blom, E. and Vos, W. K. (eds) 2008. Woerden-Hoochwoert. De opgravingen 2002-2004 in het Romeinse Castellum Laurium, de vicus en van het schip de ‘Woerden 7’. ADC Monografie 2, ADC Rapport 910. Amersfoort, ADC-ArcheoProjecten. Bogaers, J. E. and Haalebos, J. K. 1983. Op zoek naar een castellum in Woerden. Spiegel Historiael 18 (6): 302-9. Braat, W. C. 1939. Opgravingen te Vechten 1931-32 en 1936-37. Oudheidkundige mededeelingen uit het Rijksmuseum van Oudheden te Leiden, Nieuwe Reeks 20: 47-65. Access Campbell, B. 1994. The Roman army, 31 BC-AD 337: a sourcebook. London/New York, Routledge. Campbell, D. B. 2009. Roman auxiliary forts 27 BC-AD 378. Oxford, Osprey Publishing. Chorus, J. P. 2008. Romeinse forten in de Rijndelta. Over stammetjes en staanders. In M. Bergmans et al. (eds) SOJA- Bundel 2007, 7-16. Groningen: Symposium voor Onderzoek door Jonge Archeologen. Chorus, J. P. (in prep.) Timber forts along the Lower Rhine in the WestOpen Netherlands in the first and early second centuries AD. PhD thesis, Radboud University Nijmegen. Fichtl, S. 2005. La ville Celtique. Les oppida de 150 av. J.-C. à 15 ap. J.-C. Paris, Editions Errance. Giffen, A. E. van, Vollgraff, C. W. and Van Hoorn, G. 1934-1938. Opgravingen op het Domplein te Utrecht. Wetenschappelijke verslagen I-IV. Haarlem, Tjeenk Willink. Giffen, A. E. van 1948. De Romeinsche castella in den dorpsheuvel te Valkenburg aan den Rijn (Z. H.) (Praetorium Agrippinae) I. De opgravingen in 1941. Jaarverslag van de Vereeniging voor Terpenonderzoek, 25-28, 1-316. Groningen, Wolters. Giffen, A. E. van 1950. Inheemse en Romeinse terpen, Opgravingen in de dorpswierde te Ezinge en de Romeinse terpen van Utrecht, Valkenburg Z.H. en Vechten. Jaarverslag van de Vereeniging voor Terpenonderzoek, 29-32, 1-58. Groningen, Wolters. Giffen, A. E. van 1955. De Romeinse castella in de dorpsheuvel te Valkenburg aan de Rijn (Z.H.), II. Jaarverslag van de Vereeniging voor TerpenonderzoekArchaeopress, 33-37, 1-209. Groningen , Wolters. Glasbergen, W. 1972. De Romeinse castella te Valkenburg Z.H. De opgravingen in de dorpsheuvel in 1962. Cingula 1. Groningen, Wolters-Noordhoff. Glasbergen, W. and Groenman-van Waateringe, W. 1974. The pre-Flavian garrisons of Valkenburg Z.H. Fabriculae and bipartite barracks. Cingula 2. Amsterdam/London, North-Holland Publishing Company. Haynes, I. 1999. Military service and cultural identity in the auxilia, in A. Goldsworthy and I. Haynes (eds) The Roman army as a community. Journal of Roman Archaeology, Supplement 34, 165-174. Portsmouth, Journal of Roman Archaeology. Johnson, A. 1987. Römische Kastelle des 1. und 2. Jahrhunderts n. Chr. in Britannien und in den germanischen Provinzen des Römerreiches. Kulturgeschichte der antiken Welt, 37. Mainz am Rhein, Philipp von Zabern. Le Bohec, Y. 2005. L’armée romaine sous le Haut-Empire. Paris, Picard. Morillo, A. and Aurrecoechea, J. 2006. The Roman Army in Hispania, an archaeological guide. León, University of León. Ozinga, L. R. P. et al. 1989. Het Romeinse castellum te Utrecht. De opgravingen in 1936, 1938, 1943/44 en 1949 uitgevoerd onder leiding van A.E. van Giffen met medewerking van H. Brunsting, aangevuld met latere waarnemingen. Studies in prae- en protohistorie, 3. Utrecht, Broese Kemink. Polak, M. et al. 2004a. Alphen aan den Rijn - Albaniana 2001-2002. Opgravingen tussen de Castellumstraat, het Omloopkanaal en de Oude Rijn. Libelli Noviomagenses, 7. Nijmegen, Radboud Universiteit Nijmegen.

Polak, M. et al. 2004b. Op zoek naar het castellum Matilo en het St. Margarethaklooster te Leiden-Roomburg. Het archeologisch onderzoek in 1999-2000. Rapportage Archeologische Monumentenzorg 109. Amersfoort, Rĳksdienst voor het Oudheidkundig Bodemonderzoek. Polak, M. 2009. The Roman military presence in the Rhine delta in the period c. AD 40-140. In A. Morillo, N. Hanel and E. Martín (eds), Limes XX: XX Congreso Internacional de Estudios sobre la Frontera Romana, XXth International Congress of Roman Frontier Studies, Léon (España), Septiembre, 2006, 945-953. Madrid, Polifemo (Anejos de Gladius, 13, 2). Ralston, I. 2006. Celtic fortifications. Stroud, Tempus. Reddé, M., Brulet, R., Fellmann, R., Haalebos, J. K. and von Schnurbein, S. (eds) 2006. Les fortifications militaires. L’architecture de la Gaule romaine, 1. Bordeaux/Paris, Ausonius/MSH. Sommer, C. S. 1984. The military vici in Roman Britain: Aspects of their origins, their location and layout, administration, function and end. British Archaeological Reports 129. British Series. van der Kooij, D., Sprey, S., Dijkstra, M. and Postma, H. 2005. Romeinen in Bodegraven. AWN-opgravingen in de periode van 1995 tot 2002. Westerheem 54, 275-306. van Rijn, P. (in prep.). Wood supply for the Roman army and reconstruction of the woodlands from c. AD 40-140 in the Lower Rhine Delta of the Netherlands. In P. Scherrer (hrgs.), Lignum. Holz in der Antike. Akten des interdisziplinären Symposiums Holz in der Antike (Graz, 5.-7. November 2009) 29-40. Graz (Keryx 1). Wells, P. S. 1999. The barbarians speak. How the conquered peoples shaped Roman Europe. Princeton, Princeton University Press. Willems, W. J. H. 1986. Romans and Batavians. A regional study in the Dutch eastern river area. Amersfoort.

Access

Open

Archaeopress