The Iberian Peninsula between 300 and 850: An Archaeological Perspective 9789048525744

Citation preview

The Iberian Peninsula between 300 and 850

Late Antique and Early Medieval Iberia Scholarship on the Iberian Peninsula in late antiquity and the early Middle Ages is burgeoning across a variety of disciplines and time periods, yet the publication profile of the field remains disjointed. ‘Late Antique and Early Medieval Iberia’ (LAEMI) provides a publication hub for high-quality research on Iberian Studies from the fields of history, archaeology, theology and religious studies, numismatics, palaeography, music, and cognate disciplines. Another key aim of the series is to break down barriers between the excellent scholarship that takes place in Iberia and Latin America and the Anglophone world. Series Editor Jamie Wood, University of Lincoln, UK Editorial Board Andrew Fear, University of Manchester, UK Nicola Clarke, Newcastle University, UK Iñaki Martín Viso, University of Salamanca, Spain Glaire Anderson, University of North Carolina, USA Eleonora Dell’Elicine, University of Buenos Aires, Argentina

The Iberian Peninsula between 300 and 850 An Archaeological Perspective

Javier Martínez Jiménez, Isaac Sastre de Diego, and Carlos Tejerizo García

Amsterdam University Press

Cover illustration: The palatium of Reccopolis on the Cerro de la Oliva near Zorita de los Canes (Guadalajara) Photograph by the author Cover design: Coördesign, Leiden Lay-out: Crius Group, Hulshout isbn 978 90 8964 777 1 e-isbn 978 90 4852 574 4 (pdf) doi 10.5117/9789089647771 nur 682 / 684 © Javier Martínez Jiménez, Isaac Sastre de Diego & Carlos Tejerizo García / Amsterdam University Press B.V., Amsterdam 2018 All rights reserved. Without limiting the rights under copyright reserved above, no part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the written permission of both the copyright owner and the author of the book. Every effort has been made to obtain permission to use all copyrighted illustrations reproduced in this book. Nonetheless, whosoever believes to have rights to this material is advised to contact the publisher.



Table of Contents

Acknowledgements 17 Preliminary notes

19

Preface 21 Introduction 23 An archaeological perspective on the Iberian peninsula between Rome and the Middle Ages

Part 1  The Late Roman period 1 The settings of late Roman Hispania Roman Spain on the eve of late Antiquity Roman internal transformations External factors The beginning of regionalization Changing tides in the economy of the late Empire

47 47 51 58 60 63

2 New townscapes in the late Empire 67 What constituted a late Roman city? 67 Fortifications 69 The old Roman public buildings 76 The fate of urban infrastructure 82 New late Roman monuments? 87 Suburbanization and de-urbanization 91 New domestic architecture 95 3 The economy and the rural world in the late Empire 101 Blurring lines between the urbs and the rus 102 The late Roman villa: redefinition, expansion and collapse 107 Industrial exploitations of the landscape 114 The rural societies: Towards a new settlement pattern 120

4 Christianization and Germanization: New evidence for current debates 125 Understanding Christianity through archaeology 125 Locals and barbarians 138

Part 2  The post-Roman period 5 Towns and cities under Christian prevalence The late Roman urban legacy in the post-Roman world The consolidation of a Christian monumentality Visigothic state formation and urban renewal State intervention in the Byzantine and Suevic territories Developments in the seventh and eighth centuries Trade and towns in the post-Roman period

153 153 161 168 180 186 188

6 The new rural landscape 193 Hillfort occupations 193 Farmstead and village networks and other lay rural settlements 202 Funerary rituals in the rural world 209 Churches, monasteries and ecclesiastical sites 218 Other rural sites 224 7 A new material culture: a new society, a new economy 229 Snapshots of the new daily life: pottery and glass 229 The solid foundations of society and state: building and architecture 236 Representing the self and the community: identity and display 247 Beyond pots: coins and slates in their economic context 255

Part 3  The Early Middle Ages 8 The formation of a new Medieval materiality The formation of new medieval polities Early material traces of the newcomers Archaeologies of power Archaeologies of religion and identity in al-Andalus Changing townscapes Transitions in the rural world

267 268 269 276 291 299 303

9 Conclusions

315

Appendix 1 Site reference table

323

Appendix 2 Maps

329

Appendix 3 Lists of rulers

335

From the collapse of the Roman Empire to a Brave New World

Abbreviations 339 Bibliography 341 Index 389

List of Figures Figure 0.1 Figure 0.2 Figure 0.3

Figure 0.4 Figure 0.5 Figure 0.6 Figure 1.1

Map of the Iberian peninsula, showing the main geographic elements referred to in the text 26 Graph outlining the span and overlap of the various chronological periods used in the literature dealing with the period 300-900 in the Iberian peninsula 28 The Hanging Crown of King Swinthila, a circlet of gold encrusted with precious stones and pearls, and a votive declaration in hanging letters (‘Suintilanus Rex offeret’) in gold and cloisonné, together with a hanging cross 32 Photograph of the Reccopolis excavations in the 1950s, with part of the ‘palace complex’ unearthed 34 Front cover of the III Reunió d’Arqueologia Cristiana hispànica, held in Menorca in 1988 and published in 1994 36 Photograph of one of the open-area excavations of the early 21st century: The site of Gózquez (Madrid) 38 Map of Roman Hispania, and its administration in the late Empire, with main administrative cities (provincial and conventual capitals): Mérida, Córdoba, Tarragona, Écija, Astorga, Braga, Zaragoza, Cartagena, Clunia, Cádiz, Seville, Lugo, Beja, and Santarem

49

Figure 1.2

Figure 1.3 Figure 1.4 Figure 1.5 Figure 1.6

Figure 1.7

Figure 2.1 Figure 2.2

Figure 2.3

Figure 2.4

Photograph of Mount Testaccio, in Rome. This artificial hill was created from the discarded and broken amphorae which brought Hispanic olive oil to Rome as part of the annona during the early Empire 51 The villa of Carranque, in Toledo, is a perfect example of one of the large palatial rural residences that can be linked to the late Roman Hispanic aristocracy 55 Late Antique (fourth-fifth century?) decorated plaque with Christian symbols, such as the Chi-Rho with the Alpha and Omega 57 ‘Line and arrow map’ showing traditional views of routes of invasion and migration of the barbarians during the Migration period or Völkerwanderungszeit 59 Photograph of the site of Bernardos, in Segovia, a late Roman hillfort which became a local centre of power once the nearby civitas lost its relevance as a regional node. Sites like Bernardos are very common across the north-western corner of the Iberian peninsula 62 Collection of African Red Slip (ARS) wares from Late Antique contexts in Cartagena. African produce flooded the late Roman Hispanic market and are a characteristic find in Late Antique archaeological contexts64 Walls as cities in the Late Antique imagination: Jerusalem is characteristically represented by its city walls in the Madaba mosaic in Jordan 70 Reinforcement of the walls of Mérida, built during the fifth century with ashlar blocks quarried from public buildings. This late Roman phase of the fortifications was done while preserving the original walls 73 Walls of Lugo, an example of fourth century refortification in a second-ranking Iberian town, in which parts of the early Roman town had to be left outside the new enclosure 75 Plan of Cartagena in Late Antiquity (fourth to seventh centuries), indicating the location of the early Roman and reduced, late Roman walled enclosures. Note the large area close to the forum which was left outside the new fortified area 77

Figure 2.5

Photograph of the excavated remains of the circus at Tarragona 81 Figure 2.6 Photograph of the collapsed remains of one of the aqueducts of Baelo Claudia, damaged by an earthquake during the fourth century and never repaired 84 Figure 2.7 Late Roman phase of the forum nymphaeum or monumental fountain of Valencia, at the site of l’Almoina. The fountain was restored in the fourth century and kept in used, together with the aqueduct into the fifth, and probably the sixth century as well, despite the post-Roman modifications to the fountains’ decoration 86 Figure 2.8 Reconstructed plan of the ‘palatine’ complex at Cercadilla, in Córdoba 89 Figure 2.9 Plan of the late Roman Francolí Christian complex, in Tarragona, with the funerary basilicas and the early episcopal complex 90 Figure 2.10 Plan of Tarragona, indicating the location of the main settled areas: The upper town, the lower town, and the harbor suburb 94 Figure 2.11 Plan of Alicante, with the location of the sites of Lucentum and the new, late Roman commercial hub at Benalúa 95 Figure 2.12 The domus of the Mithraeum in Mérida, with its cosmogonic polychrome mosaic representing the gods of the Heavens, Earth, and the Underworld 97 Figure 2.13 The ‘House of the columns’ (fourth to early sixth century), excavated at the Plaza de la Encarnación in Seville. An example of a late Roman aristocratic domus, linked to civic elites. The house has got a late Roman ‘oecus-type’ layout, is decorated with marbles, mosaics, and was directly linked to the functioning aqueduct by lead pipes 98 Figure 3.1 Figure 3.2

Late Roman industrial establishment built inside one of the old forum buildings in Valencia, with space for storage, food-processing, and a pressing vat Late Roman burials in the forum of Valencia: Inhumations covered with tiles which show a change from early Roman practices of cremation and specific burial areas by main roads

104

106

Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7

Photograph of a fourth century Christian sarcophagus depicting Christ as the Good Shepherd. Currently used as an altar in Écija 106 Photograph of one of the mosaic panels at the villa of Noheda 109 Reconstruction of the villa of La Olmeda, according to Abásolo and Martínez 2012 109 Plan of the excavations at El Pelícano, where the old villa has been replaced by a series of dispersed minor settlements, but which are still linked to the old estate 121 Table showing the percentage of furnished burials in late Roman rural cemeteries 123

Figure 4.1

Photograph of the Chi-Rho identified inside the cistern of the domus of the Sala Decumanus in Mérida, the earliest known depiction of Christian symbolism in the Iberian peninsula Figure 4.2 The patena from Castulo, a glass offering open form with the image of Christ in majesty, discovered during the excavations of 2014 Figure 4.3 Late sixth century Christian funerary inscription inscribed in a reused public document, currently in the Museum of Seville Figure 4.4 The altar slab from Casa Herrera, a sixth century sigmatic mensa with a commemorative inscription Figure 4.5 The Quiroga Chi-Rho, a Late Antique elite Christian liturgic dedication in marble Figure 4.6 Altar from the church of El Gatillo, as it stands today Figure 4.7 Two cloisonée Visigothic eagle brooches retrieved from the site of Alovera, and which are now at the National Archaeological Museum in Madrid Figure 5.1 Figure 5.2

Photograph of the walls of Veleia, showing the construction technique used, with large blocks of stone and reused column drums Plan of the Visigothic phase of the Domus of the Marbles, in Mérida. The old domus has been subdivided into various single-family dwellings around a communal courtyard

127 127 130 132 133 137 140

156

158

Figure 5.3 Figure 5.4

Figure 5.5 Figure 5.6 Figure 5.7a Figure 5.7b

Figure 5.8 Figure 5.9 Figure 5.10 Figure 5.11 Figure 5.12 Figure 5.13 Figure 5.14 Figure 5.15

Photograph of one of the pillars of the aqueduct of Los Milagros, in Mérida, apparently restored without success in the Visigothic period 160 Reconstructed plan of the episcopal complex of Barcelona, with its various phases, indicating the location of the episcopal buildings (hall, cathedral, baptistery, palace, baths) and the ‘count’s palace’ 165 Reconstructed plan of the episcopal complex of Valencia, at the site of La Almoina 166 Mosaic of the ‘Opposing Lions’, from the porticus of the suburban Christian complex of Mértola 168 Axonometric reconstruction of the building identified as the xenodochium of Masona in Mérida 169 Photo showing the existing remains of the xenodochium, with the reconstructed location of the marble pilasters and the disposition of the apsed hall and the open courtyards 169 Remains of the Late Antique public (?) structure built by the temple for the Imperial cult in the forum of Mérida 171 Schematic plan of the Visigothic civil administrative complex (palace and citadel) identified in Córdoba 172 General plan of the excavations at Reccopolis, highlighting the location of the palace complex (conjunto palatino) and the workshop area 174 Plan of the excavations at the site of El Tolmo de Minateda, identified with the Visigothic episcopal see of Eio 176 Plan of Toledo, indicating the location of the lower meadow (‘Vega baja’) suburb, with the location of the main Late Antique sites 177 Plan of one of the main buildings identified next to the basilica of Guarrazar 180 Plan of the Byzantine quarter excavated at the Roman theatre of Cartagena, which includes various dwelling units 182 Plan of the Suevic ‘acropolis’ of Falperra near Braga 184

Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 6.5 Figure 6.6 Figure 6.7

Figure 6.8 Figure 6.9 Figure 6.10 Figure 6.11 Figure 6.12 Figure 6.13 Figure 6.14

Figure 6.15

Figure 6.16 Figure 7.1

Figure 7.2

Extensions excavated in some hillfort sites of the northern part of the Iberian Peninsula 194 Air photograph of the excavations at Navasangil 197 Walls at the site of Castro Ventosa 198 Defensive system at Tedeja 201 Plan of the excavations at El Bovalar 203 Stone-footed building at El Pelícano 204 Table indicating the relative proportion between the number of sunken featured buildings excavated per site (first column) with the overall excavated area (second column). The result shows it is necessary to excavate quite large open areas in order to identify or find a relevant number of sunken featured buildings 205 Plan of the excavations at Gózquez 206 Aerial photography of the site of Zaballa 208 Pit burial at the site of La Mata del Palomar 211 Plan of the excavations at Duratón 213 Table depicting the different proportions of furnished burials and inhumations habillées in various excavated communal cemeteries 215 Aerial photography of Villanueva de Soportilla 217 Plan of the excavations of Casa Herrera, with its two phases: In dark grey the original AD 500 funerary basilica; in lighter grey the late sixth century expansion, linked to episcopal intervention 219 Front of the church Sain John in Baños. As it stands today, the building is not Visigothic, but it could be built with reused material from an earlier Visigothic church 221 Plan of the rural ‘Visigothic’ palace of Pla de Nadal (Valencia), as excavated in the 1970s 226 Rim and shoulder from a small cooking pot (‘olla’) from the excavations at Casa Herrera. Thrown on a slow wheel, and made with unrefined clay and large inclusions, it is a perfect example of pottery made in the sixth and seventh centuries 232 Glass bowls and glass typologies from Reccopolis 236

Figure 7.3

Fifth-century wall in Mérida at the site of Morería, where the ashlar reinforcement was made out of various reused blocks (tomb markers, column drums, theatre seats, etc.) 238 Figure 7.4 Opus signinum from the aqueduct of Reccopolis. Dated to the late sixth century, this opus signinum uses large crushed fragments of tile, rather than the ground and pulverized pot fragments that characterized early Roman signinum 239 Figure 7.5 Visigothic buildings from the Vega Baja suburb in Toledo, where most of the buildings were built with stone foundations (bound with clay, not mortar in this case) and the elevations would have been in mudbrick or trampled earth 240 Figure 7.6 Stone quarries located next to the aqueduct of Reccopolis. They were probably used to obtain blocks for the construction of the aqueduct itself 241 Figure 7.7 Sunken Feature Building (Grubenhaus) from the Visigothic period. It has the main characteristics that can be seen across Europe: It is dug into the ground and has post-holes for the vertical uprights 244 Figure 7.8 Various typologies of lyre-shaped (liriforme) brooches from the south east 249 Figure 7.9 Casa Herrera pilaster; a failed attempt to create a round column out of a prismatic block of marble without the necessary tools 253 Figure 7.10 Marble pilasters from Mérida, now in the water cistern of the Umayyad citadel. Carved out of reused blocks, these pilasters show the traditional geometric and vegetal motifs that characterize Visigothic sculpture254 Figure 7.11 Golden tremissis of Athanagild from Reccopolis now at the National Archaeology Museum, Madrid 256 Figure 7.12 Byzantine copper coin, minted in Cartagena: One of the few indicators for the circulation of newly-minted small currency 257 Figure 7.13 Graph depicting the evolution of gold content in Visigothic coins per reign. It is noticeable how the steady and standard production in the period of state formation contrasts sharply with the decline that follows it in the eighth century, with only a few minor examples 260

Figure 7.14 Reproduction of one of the Visigothic slates recovered from Diego Álvaro, now in the Museum in Guarrazar 261 Figure 8.1

Umayyad coin dated to the year 93H, with a Latin legend with the mint mark for al-Andalus (the star and the legend SPN, for Spania) 272 Figure 8.2 Arabic lead seal dated to 100-102H, bearing the legend ‘in the name of God, this is the treaty [sulh] of / Abd Allah ibn Mali’ 273 Figure 8.3 Pottery jug from the Islamic phases of Casa Herrera, Mérida. Note the characteristic high handle, but with a fabric still reminiscent from the earlier phase 274 Figure 8.4 Ceramic typologies from South-Eastern Spain in the late eighth century (according to Carvajal López 2009): a-c, j) cooking pots; d-f) ARS imitations; g) jar; h-i) storage; k-o) new types of cooking pots 275 Figure 8.5 Plan of the Umayyad ‘palaces’ excavated at the site of Morería, in Mérida, which were built on top of the domus of the Marbles 278 Figure 8.6a Photograph of the fortress giving access to the alcazaba of Mérida, including its dedicatory inscription. This Umayyad fortress was built to keep the peace in the city after various years of rebellion by the local muwalladun 280 Figure 8.6b Plan of the alcazaba of Mérida, showing the entrance fort, which controlled access to the whole city, and the location of the cistern 281 Figure 8.7 View of the settlement of Zorita from the hill of Reccopolis 283 Figure 8.8 Plan of early Umayyad Murcia, a new urban foundation to substitute the muwallad stronghold of Eio 284 Figure 8.9 Plan of the early remains at Oviedo (according to García de Castro and Ríos 2016, figure 15), indicating the location of the main early medieval monuments: 1) Monastery of Saint Pelagius; 2) Monastery of Saint Vincent; 3) Saint Mary and Royal Mausoleum; 4) Cathedral of the Holy Savior; 5) Church of Saint Tyrsus; 6) Episcopal complex and palace; 7) Ninthcentury castle 289

Figure 8.10 Santa María del Naranco church, near Oviedo. It was originally a royal chapel built by King Ramiro I in a completely new architectonic style, taking advantage of the vaulting techniques reintroduced into the Iberian peninsula by the Umayyads 290 Figure 8.11a Photograph of the interior of the Mosque of Córdoba, showing the forest of columns and the rows of superimposed arches. The mosque became not only the main religious and political centre of Islam in al-Andalus, but also the key dynastic monument of the Umayyads 293 Figure 8.11b Plan of the Mosque, as it was in its original phase. Originally built by Abd al-Rahman I, it went through major remodelations and expansion, including by Abd al-Rahman II in the mid-ninth century 294 Figure 8.12 Plan of the maqbara of Lucentum, located on top of the old foum and dated to between the seventh and tenth centuries, indicating the continuity of population in the area of Lucentum despite the abandonment of the old urban site 296 Figure 8.13 One of the Arabic graffiti inscribed on the columns of the basilica of Casa Herrera (ninth century) 298 Figure 8.14 Plan of the excavations at the suburb of Šaqunda in Córdoba, located on a bend of the river, south of the city and across the main bridge 304 Figure 8.15 Schematic plan of an early Umayyad rural settlement from the hillslopes in the Vega of Granada (by Miguel Jiménez Puertas, used with permission). A: Village with a dispersed pattern of settlement; N: Necropolis; C: Path with different possible destinations; F: Water source; P: Agricultural plots; CN: Contour lines 307 Figure 8.16 Site plan of Alegría-Dulantzi 310 Figure 8.17 Stratigraphy of the agrarian terrace of El Manso (Asturias) 314

Acknowledgements When I was first approached by Jamie Wood to contribute a volume on aqueducts to this great initiative of the Amsterdam University Press which is the series on Late Antique and Early Medieval Iberia, I thought it would be quite a better idea to involve Isaac in writing something a bit broader in scope. We thought that despite the irresistible appeal of Late Antique aqueducts we could write something which we thought was very necessary: a summarised and updated archaeological overview of Late Antique Spain. The idea for this book (which you are currently reading) was very welcome, and we soon embarked ourselves on drafting ideas, putting together sites, images, interpretative models, and so on and so forth. Little were we to know that four months into our writing project we both would find ourselves in very difficult personal and work situations, which kept us from finishing this book, into which we had to draw (or fool) Teje in order to be able to put everything together on time. We will never be thankful enough to Jamie Wood and Erin Dailey for their never-ending patience and encouragement in those months when we kept delaying and delaying our final draft. Even after that first fair-but-harsh evaluation report came back. Continuing with our topoi, there are many people we would like to thank for their help, support and suggestions, which have greatly improved the book (and caused us no shortage of headaches). In our list we necessarily have to mention Santiago Castellanos, Bryan Ward-Perkins, Alfonso Vigil-Escalera, Juan Antonio Quirós Castillo, Jamie Wood, Miguel Alba, Marcos García García, José Cristóbal Carvajal, Patricia González, José Carlos Sánchez Pardo, Guillermo García-Contreras, Óscar Bonilla, Rodrigo García-Velasco, and Robert Portass, who viewed different parts of early versions of the text. Without them, this book, which on more than one occasion we thought would be too big a project, would not have come in to being. Also we want to thank our friends and families for keeping us sane and on track with their encouragement. All the shortcomings, omissions, and sticks of which we may have got the wrong end of belong solely to us. Cambridge/Mérida/Catamarca, February 2018



Preliminary notes

All dates used in the book will be AD. Regarding the names of sites and towns, ancient names will be used only for a) those archaeological sites which can be identified (such as Reccopolis or Clunia), b) those that do not really correspond to their modern equivalent (like Lucentum with Alicante or Bilbilis with Calatayud), and c) sites mentioned in the sources which cannot be identified (such as Victoriacum). Modern Spanish/Portuguese names will be kept for other archaeological sites (such as the villa of Noheda or Gózquez) and for those Roman towns which are modern settlements, such as Mérida (instead of Augusta Emerita) or Barcelona (Barcino), with the only exceptions of Seville (for Sevilla) and Lisbon (for Lisboa), which are (by far) more common in the English language than, for instance, Saragossa or Cordova. Translations of texts and inscriptions are, unless mentioned, by the authors. Spanish conventions have been used to indicate the province to which every site belongs. This information has been tabulated in the reference appendix.

Preface This is an important book for two reasons. Firstly because, over the last thirty years or so, a considerable amount of sophisticated and important work has been done in Iberia on the archaeology of its late Roman and early medieval phases. But because most students, and indeed most medieval archaeologists, are poor linguists (at least in the anglophone world), and because the vast majority of Iberian research is written in Portuguese, Spanish or Catalan, it is almost completely ignored by university courses elsewhere in Europe, and passed over even in the scholarly literature. Here we have a book that provides a detailed overview of the late Roman and post-Roman archaeology of Iberia in the language that, for better or worse, is rapidly becoming the lingua franca of scholarship. Furthermore it is a serious book, with excellent illustrations, a full bibliography, and a confident awareness of archaeological developments elsewhere in Europe, written by three young scholars who are all actively involved in the archaeology of Spain, and who between them bring detailed first-hand knowledge on many of the central aspects of their subject, such as urbanism, rural settlement, water-supply, church architecture, and burial archaeology. With the publication of this book, Iberia in the crucial period AD 300-850 can no longer be ignored by archaeology courses, nor side-lined in general discussions of developments in post-Roman Europe. Secondly, understanding what happened in Iberia in this period from an archaeological perspective is peculiarly interesting and important, because, unlike most of the Roman Empire, the peninsula witnessed two major take-overs of power, and the formation of two new states, based on very different foundations: the Visigothic kingdom established in the fifth and sixth centuries, which of course paralleled developments in Gaul and Italy; but also a Muslim invasion from the Maghreb in 711 (that later evolved into an independent caliphate), which took most of Iberia in a very different direction from the other European former provinces of Rome. Did these momentous political changes, which are poorly documented in the written sources, have an impact on the ground, such that they show up in the archaeological record? Probably the most important conclusion of this book is that change was gradual and that, from an archaeological perspective, the 410s (when the Visigoths first became involved with Spain) and 711 (when the Muslims invaded) do not constitute sudden dramatic breaks – there is some evidence of violent destruction in both periods, but there is much more

22 

The Iberian Peninsul a bet ween 300 and 850

evidence of continuities across these supposed breaks. There was however very significant change within Iberia, though it happened over decades or centuries, rather than suddenly. In particular, the peninsula underwent transformations already familiar in much of Europe, but only recently well documented in Iberia: a simplification of the economy (with some imports and some craft skills disappearing, and others becoming severely restricted); a rise in regional difference; a definitive end of Roman patterns of rural settlement (with the disappearance of villas, some of which had been palatial in scale in the fourth century, and the emergence of villages and hilltop settlements); a shrinkage in urban life (with many minor towns disappearing and those that persisted losing their ‘classical’ form); and a gradual christianisation of both the urban and the rural landscapes. But within these changes, that broadly follow patterns documented in Italy and Gaul, there are also fascinating developments peculiar to the peninsula: for example, the evidence of powerful Visigothic monarchy in the sixth century, revealed by impressive new settlements like Eio and Reccopolis, the latter even provided with a newly built aqueduct, and the impact of Muslim rule on the principal cities of al-Andalus, with the building of urban fortresses and the introduction of an entirely new religious architectural form, the mosque. The authors are of course careful not to solve all the questions that the evidence poses, and they make it clear that the peninsula’s archaeology deserves our attention, not just for the answers it has already provided, but also for important questions that still remain – for instance, the real dating of the tiny, but impressively solid and well-decorated ‘Visigothic’ churches; and, perhaps most puzzling of all, the significance of burials with ‘Visigothic’ jewellery. What were the people buried with this jewellery trying to tell their contemporaries about their identity, and how can we, a thousand-five-hundred years later, and with only the burials themselves to go on, most sensibly understand them? Bryan Ward-Perkins University of Oxford, Trinity College

Introduction An archaeological perspective on the Iberian peninsula between Rome and the Middle Ages This book aims to put together in a single publication a comprehensive summary of the results of over 25 years of archaeological scholarship on Late Antique and Early Medieval Spain and Portugal, together with an overview of previous research on the topic. In recent years, there has been a remarkable increase in the quantity of available data covering these historical periods as a result of minor interventions, unexpected finds, long-term research projects, and the development of commercial archaeology.1 This huge amount of fresh data has enabled the development of new avenues of archaeological research for Late Antique and Early Medieval contexts. It has also opened the ground for wholly original, previously ignored or difficult to assess, areas (e.g. peasant societies or agricultural spaces), and also enabled the revisiting of old topics (e.g. the role of the Church or the transformation of Roman urban spaces). The long chronological scope of these processes prompted by the transformations of the late Roman world is the main narrative axis of this book. This publication is, to the best of our knowledge, the only general comprehensive archaeological approach to the material culture of the Iberian Peninsula for this period, not only in English, but in Spanish as well. Juan Antonio Quirós and Belén Bengoetxea have published recently (2010) a useful handbook of Medieval archaeology, which covers most of the material discussed here, but the scope of their publication is wider in terms of chronology and geography (as it includes Western Europe as well). Rose Walker has, similarly, published in this same series a book on Early Medieval art (Walker 2016), which partially addresses the Roman and Visigothic periods. Besides these, the only general and recent publications that cover this period in Spain have been produced by historians who, despite the various uses they give to the material record, 1 Development which has taken place in Spain as a consequence of the enactment of the Heritage Law (1985), which created an unstable commercial archaeology whose development caused the excavation of hundreds of sites (Demoule 2007; Parga-Dans 2010), including Early Medieval ones. In the case of Portugal, commercial archaeology has its origins in the late 1980s and early 1990s, related to some big projects on the river Alqueva, but we can consider 1997 to be the institutional starting point with the creation of the Instituto Portuguȇs de Arqueologia (IPA) (Bugalhão 2011).

24 

The Iberian Peninsul a be t ween 300 and 850

produce their publications from document based perspectives (e.g. Arce Martínez 2011; Collins 2005). As noted above, this material culture based perspective, elaborating a coherent narrative of the Iberian peninsula between the fourth and the ninth centuries, is the main novelty of this book. Recent finds in Spain and Portugal, and the scientific discussions that have developed from them, have only slowly made their way into wider academic circles. Some twenty years ago in these same circles the main archaeological references for Late Antique material were North Africa, Gaul and Italy, with the Iberian examples relegated to mere appendixes (e.g. Lavan 2001; Salzman 2002). Thankfully, due to the work of Spanish and Portuguese scholars, and the increasing interest of foreign academics, this situation has changed and the Iberian material currently features in most current discussions (e.g. Arbeiter 2014, 2015; Esmonde Cleary 2013; Gelichi and Hodges 2015; Wickham 2005). Despite this availability and interest, there is no comprehensive work in English that addresses the transition of Roman Hispania into the Middle Ages from an archaeological viewpoint (Diarte Blasco 2016).2 Furthermore, we ought to stress that this is, fundamentally, an archaeological account of the Late Antique and Early Medieval centuries in the Iberian peninsula. This material culture based perspective does not equate to a voluntary ignoring or leaving aside of written sources, which are fundamental to understand most of the topics discussed in this book. We believe that the stress given to material culture in this book enables the answering of new questions, which are difficult to address solely from a documentary perspective (Barceló Perelló 1988). This emphasis on the material culture, on an empirical basis, needs its own set of questions and perspectives. Due to this archaeological emphasis, the chronological framework of the book is not led by historical dates, but rather by cohesive materialities, whose chronologies and processes of evolution and consolidation move at different paces than historical events as recorded in the documentary sources do. That is why the period we cover stretches from the fourth up to the mid-ninth century. This challenges mainstream publications, which look for historical dates of military events to set their limits, such as the late Roman period (Arce Martínez 2009; Kulikowski 2004), the beginning 2 In this aspect, we have to thank Amsterdam University Press and Jamie Wood, the editor of the Late Antique and Early Medieval Iberia series (in which there are currently some works in preparation on this same topic), not only for inviting us to write this, but also for the initiative shown in setting it up.

Introduc tion

25

of the Visigothic kingdom after the defeat at Vouillé in 507 or of the Muslim period after 711 (e.g., Arce Martínez 2007, 2011; Collins 2005, 2012; Manzano Moreno 2006). Whereas the starting point is clearly defined in the fourth century (transformation of classical townscapes and reorganization of the villa economy as a result of political transformations in the Late Roman period), the closing date we propose is more controversial. We have included the eighth and the early ninth centuries in our book because these two centuries, especially following the Umayyad invasion, form the interphase with Early Medieval material culture. The attention here is to processes, and not specific episodes or historical landmarks. It is only as a result of the archaeological developments of the last two decades that it has been possible to move a step beyond these interpretations and focus on longue durée narratives more than on accounts and material representation of particular events. We propose an alternative to these views, one in which the five-century period covered in this book will be divided into three main processes of transformation that will be addressed globally for the Iberian peninsula (in rural and urban contexts): the end of the Roman system, the development of Late Antique regional responses, and the final political and social reorganisation that lead to the Early Medieval world. These three broad divisions better reflect, in our opinion, the technological and material changes seen through archaeology. Added to this, we will lay out the basic theoretical approach which nowadays we feel gives a more cohesive and comprehensive explanation of the period: the dialectical processes of regionalization and centralization which framed post-Roman Iberian societies. Lastly, we want to underline the fact that other than presenting the evidence and a summary of the available material culture, we are also presenting our own interpretations based on them.

I

Setting the frame: Geographical and chronological scope of the book

Before addressing the material evidence, it is perhaps necessary to lay out the geographical and chronological framework that we are dealing with in this book. Far from a central geologic unity, the Iberian peninsula has a very particular and varied geography. As a consequence, the peninsula is divided into various clearly-defined regions (figure 0.1), which play an important role in the processes of regionalization that characterize the period we discuss.

26 

The Iberian Peninsul a bet ween 300 and 850

Figure 0.1 Map of the Iberian peninsula, showing the main geographic elements referred to in the text

The main one is the central plateau, or central Meseta, which occupies most of central Spain. Old mountain ranges divide it into two halves (Northern and Southern Mesetas) along a south-west – north-east line. The Meseta itself is surrounded by high mountains on three of its sides (north and north-west, south, and east), the west being open towards the Portuguese lowlands and the Atlantic Ocean. Three main rivers cross this plateau from the mountains in the east into Portugal and the ocean: The Duero on the north (in fact, it is common to find the Northern Meseta referred to as the Duero basin) and the Tagus and the Guadiana in the south. These rivers form fertile valleys surrounded by drier high ground. These flat highlands are very rich in agricultural terms, but suffer from extreme climatic conditions. Furthermore, the high mountains that border it isolate the Meseta from the wealthier and better-connected coastal regions. The mountains bordering the Meseta descend very rapidly into the sea on its north face, creating a thin strip of land from the Pyrenees down to Portugal that is crossed by deep valleys. This area is the Cantabric region, also referred to as ‘Green Spain’ (la España verde), due to its abundant vegetation and high rainfalls. The Cantabric region, which includes the Galician massif and the western valleys of the Pyrenees, is very rugged and rough, and

Introduc tion

27

settlement patterns favour small disperse communities even to this day. In the south, the mountains separate the Meseta from the Guadalquivir valley, what is now Andalucía, a region of Atlantic orientation but Mediterranean in climate with very fertile soils. Similarly, the mountains that border the Meseta on the east are separated from the Pyrenees by the second main valley in the peninsula, that of the Ebro. In this book we will use these geographic names, but it is necessary to outline the political regions, historically and chronologically biased as they may be, because they are ever present in the literature. In the late and post-Roman periods, provinces and other administrative areas are the most common names. The north-west region is the Gallaecia, mostly separated from the Meseta by mountains and deep valleys. South of the Duero, and encompassing the western edge of the Meseta and modern Portugal is Lusitania, which was crossed by the Tagus and the Guadiana, and seat of the last Roman capital of Hispania, Mérida. The Guadalquivir valley roughly coincides with the Roman province of Baetica, a region which preserved its contacts with the rest of the Mediterranean area, even after the fall of Rome, whereas the Ebro valley mostly fits in Tarraconensis, the last province under Roman control. The large, irregular territory left out (the east half of the Meseta and the south-eastern coast) forms the Carthaginensis. In the Visigothic period the Eastern Meseta was divided into two smaller regions, Carpetania and Celtiberia, while the southern coast between the Straits of Gibraltar and the territory of Alicante became the ‘Byzantine province’ or Spania. This territorial distribution was abandoned after the Islamic invasion, and two main blocks emerged: al-Andalus (the Muslim territories) and the Christian north. The former was itself divided into three defensive marches (the Lower march to the west, the Middle march in the centre, and the Upper march in the north), whereas the latter was progressively divided into smaller political entities such as the Asturian Kingdom, Castile, Navarre, or the Hispanic March. The chronologies and period labels used in Late Antique and Early Medieval Iberia are varied and confusing as a direct consequence of the historiography on the topic. In some cases, the confusion is produced by the use of some generic terms (such as ‘Byzantine’ or ‘early Islamic’) which do not correspond to the chronologies usually linked to them in the wider Mediterranean context, neither to their cultural nor original meaning (figure 0.2). In other cases, it is a consequence of the use of regional names, widely accepted by specialists in this area, but potentially obscure to laymen in the subject. Moreover, it is quite common to use multiple terms, coming from different spheres (political, cultural and religious), when talking about

28 

The Iberian Peninsul a bet ween 300 and 850

Figure 0.2 Graph outlining the span and overlap of the various chronological periods used in the literature dealing with the period 300-900 in the Iberian peninsula

UMAYYAD ISLAMIC MOZARABIC

PALEOCHRISTIAN SUEVIC BYZANTINE state formation

VISIGOTHIC LATE ROMAN

EARLY MEDIEVAL

POST-ROMAN

EARLY MIDDLE AGES LATE ANTIQUITY AD 300

400

500

600

700

800

900

the same period and its material culture. This is particularly the case for sixth century evidence, which can be considered at the same time Visigothic, Late Antique, post-Roman, and early Christian – all of them terms which were used interchangeably in, for example, the prolific bibliography of Pedro de Palol. To start with, ‘Late Antiquity’ is a very broad term covering roughly the time between the third or fourth centuries up to the mid-eighth century. A chronology set used, for instance, by leading research clusters such as the Oxford Centre for Late Antiquity. It overlaps with the ‘Early Middle Ages’, which range from the end of Roman power in the fifth century to the tenth and eleventh centuries. The choice between these two concepts with regard to the chronological overlap is usually a reflection of the background of the authors, whether he or she believes there was rupture or continuity with the Roman past. In the Spanish case-study, it is not uncommon to find rural settlements and burial practices referred to as ‘Early Medieval’ (altomedieval), whereas urban contexts in the same chronology are ‘Late Antique’ (tardoantiguo).3 Traditionally, the term ‘early Christian’ (paleocristiano) 3 For instance, compare Fuentes Domínguez 2006 with Vigil-Escalera 2007 and 2009, wherethese two different labels are applied to the same territory and the same chronologies.

Introduc tion

29

has also been very prominent to refer to the material culture and monuments of the period (especially churches and other liturgical elements), as the name chosen for the first National Conferences of early Christian Archaeology – Reunió d’Arqueologia Paleocristiana Hispànica – show (the last took place in 2003). However, Christian contexts dated to the Islamic period are usually referred to as ‘Mozarabic’ (mozárabe). One way to avoid these terminologies is to use ‘late Roman’ (tardorromano) or ‘post-Roman’, the former referring usually to the fourth and fifth centuries, and the latter to the sixth century and beyond. This is a chronological and political division, using the presence or absence of the Roman state and its characteristic material culture, as a marker, can be put around 450. Other chrono-political labels are ‘Visigothic’ (visigodo), ‘Byzantine’ (bizantino), ‘Suevic’ (suevo) and ‘Umayyad’ (omeya), which currently (although this was not always the case) have no ethnic connotation. ‘(Visi)gothic’ refers to any given element which is geographically and chronologically inside the Visigothic kingdom or its area of influence. This ranges between historic dates such as 418 and 711, and although the Visigothic control over part of Iberia only really begins in the 450s, it is not until the sixth century that it can be considered to be effective (the ‘early Visigothic period’). Within the Visigothic period, the years 570-630 are usually referred to as the period of state formation. After 630, and into the eighth century, it is not unusual to see the terms ‘late Visigothic’ and ‘Hispano-Visigothic’ or ‘Hispano-Gothic’. Similarly, ‘Suevic’ refers to the kingdom established by the Sueves in northwestern Iberia between 411 and 580, so this term is not only chronological, but also geographical. In Spain, the ‘Byzantine period’ covers the seventy years between the 550s and the 620s, during which imperial troops controlled parts south of the peninsula. The presence of many lingering Roman cultural elements at various levels (monuments, material culture, law, etc.) makes post-Roman a generic and adequate term to cover these three periods. Finally, the ‘Islamic period’ starts in 711, and it only came to an end in 1492, although for the purpose of this book, only the Umayyad period will be taken into account. The years in between the Islamic conquest (711) and the establishment of the Umayyads (754) can be referred to as ‘early Islamic’ (paleoandalusí, although this would cover usually the whole eighth century) or ‘pre-Umayyad’, which is rare. The ‘Umayyad period’ covers the years between 754 and 1031 (whereas in the rest of the Mediterranean it finishes in 750); ‘early Umayyad’ or ‘emiral’ covers the period between 754 and 929 and ‘late Umayyad’ or ‘caliphal’ (califal) between 929 and 1031. In order to avoid this confusing palimpsest of terms and concepts, most of them not exempt from theoretical problems and substantial critiques,

30 

The Iberian Peninsul a bet ween 300 and 850

we will use general broad and general terms throughout the book. Thus, so as to establish a wider frame for the analysis, we will refer to the fourth and early f ifth centuries as the ‘late Roman’ period (covered in Part 1: Chapters 1-4), the late f ifth and into the early eight will be referred as the ‘post-Roman’ period (Part 2: Chapters 5-7), and the last phases, from the eighth into the ninth centuries, as the ‘early medieval’ period (Part 3: Chapter 8).

II 20th century approaches to the period As in the rest of Europe, scientific approaches to the medieval period began during the nineteenth century as a consequence of the development of Romanticism, nationalism and imperialism (Trigger 1995). In the Spanish case, the evolution of the understanding of the Late Antique and the Early Medieval archaeology has to be seen as a sequence of historiographic models with strong ideological components behind them, which have only been analysed very recently (Olmo Enciso 1991; Salvatierra Cuenca 2013). Long-held interpretations based on historical accounts or art history, such as the intrinsic ethnic meaning of ‘Visigothic’ and ‘Suevic’ grave goods, or the Byzantine nature of Late Antique Iberian Christianity, have weighed heavily on past researchers. Similarly, and completely dissociated from Late Antique studies, research on the Islamic and the ‘Christian’ areas during the Early Middle Ages have been heavily text based. These pivoted around the reach and impact of Islamization and of Feudalism, the origins of the Christian kingdoms and, above all, the spirit of the Reconquista, the conquest of the Islamic lands by the Christian kings of the north, who were to become the founders of ‘Spain’. It was only when archaeology emerged and developed as an independent science that these monolithic concepts could be questioned. In Portugal, the evolution of historical and archaeological research has followed similar paths, although with key differences, such as the search for the Suevic origins of Portugal, or its own characteristic individualities that marked it as distinct from Spain (as summarised in Díaz Martínez 2011: Introduction, or Ferreira Fernandes 2005). Focusing now on the evolution of the academic understanding of the late and post-Roman world, the first approaches to this topic in Spain developed as a result of three main events. The first of these was the discovery of the Treasure of Guarrazar (figure 0.3). This Treasure consists of a collection of golden votive hanging crowns, donated or dedicated by the Visigothic kings Swinthila (r. 621-631) and Recceswinth (r. 653-672), together with

Introduc tion

31

some other minor crowns, hanging crosses and other elements of liturgical jewellery (Perea Caveda 2001). It was first identified in the year 1858 at the church of Guarrazar, 13km outside Toledo, when torrential rains uncovered a sepulchre containing the objects, which were retrieved between 1858 and 1861.4 These finds were immediately linked to the legendary ‘hall of crowns’ of the Visigothic kings mentioned in the Islamic sources (Hernández Juberías 1996: 194-8; Ibn Habib, 44), soon leading to the first studies of Visigothic jewellery and Visigothic art (cf. Åberg 1922; Schlunk and Hauschild 1978; Zeiss 1934). These early approaches were still influenced by Romantic views on the Visigothic kingdom (Cortés Arrese 2012), but the sudden interest on the material culture was unprecedented. The second and third events are interconnected, as they both derive from the work of Amador de los Ríos (1818-1878): firstly, he coined term latino-bizantino (‘Latin-Byzantine’) and secondly, he carried out a first classification of Early Medieval sculpture. Latino-bizantino was the term used by Amador de los Ríos to describe the finds at Guarrazar (1861), and used it again in a later publication to describe the Late Antique monuments of Mérida (de los Ríos Serrano 1877). As for the classification of Early Medieval sculpture, his conclusions and typological classification set the grounds for later, art-historical interpretations while, at the same time, setting the architecture and sculpture of this period aside from later and earlier typologies. Following Amador de los Ríos’s work, many scholars gave (de Palol Salellas 1950; Palol and Ripoll 1988), and still give (Morín and Barroso 2010), a preeminent role to the Byzantine element in their descriptions of Late Antique Hispanic culture. These scholars define this period as a cultural ‘Renaissance’ in which, thanks to the Visigothic kingdom, a ‘new Constantinople’ was created on the other side of the Mediterranean. According to these authors, this transforming movement likely began at the second half of the sixth century, when the cultural presence of Byzantium would have been intensified in Spain, due to the military occupation of the south-eastern region under Justinian (555). As a result of the accumulation of new Byzantine features, Visigothic Art reached its aesthetic height in the second half of the seventh century. Mérida was considered to be the main centre at which Byzantine influences entered the peninsula, and from there it would have spread to the rest of the Visigothic kingdom. A notable number of Spanish scholars defend the idea that these ‘Byzantine influence and innovations’ found in the material culture from Mérida gave rise to a revolution in Visigothic society, culture and religion. 4

Currently under excavation by Juan Manuel Rojas.

32 

The Iberian Peninsul a bet ween 300 and 850

Figure 0.3 The Hanging Crown of King Swinthila, a circlet of gold encrusted with precious stones and pearls, and a votive declaration in hanging letters (‘Suintilanus Rex offeret’) in gold and cloisonné, together with a hanging cross

Retrieved from the excavations at Guarrazar (copy preserved at the Museum of Guarrazar; photograph by Dr. Patricia González, used with permission)

In Portugal, this early interest in Medieval archaeology came first as a by-product of other archaeological investigations (with the first catalogues of material culture by José Leite de Vasconcelos or Estácio da Vega) and as a result of the listing of Medieval monuments in the Comissões dos Monumentos Nacionais. Most of these were Later Medieval churches, castles and monasteries, but Islamic monuments were also included (Ferreira Fernandes 2005). The next great development in Late Antique archaeology was carried out by Manuel Gómez-Moreno (1870-1970), who first classified and categorized

Introduc tion

33

the pre-Romanesque Mozarabic churches, and the material culture associated with them (Gómez-Moreno 1919; 1966). He was also the first person who attempted to understand the Islamic material culture, beyond the standing monuments, from an archaeological point of view, during his time as director of the Instituto Valencia de Don Juan in Madrid. In his long research life Gómez-Moreno was always interested in the study of Late Antique and post-Roman material, such as the Visigothic slates and the corpus of early Christian epigraphy. It was, however, after the 1920s (and especially in the 1940s) that the first general interpretative paradigms were developed, partly based on Gómez-Moreno’s and de los Ríos’ work, but also partly due to the direct archaeological intervention in post-Roman sites. In this aspect, the creation of the Deutsches Archäologisches Institut (DAI) in Madrid in 1943 was of paramount importance, as it prompted the introduction of ‘modern’ archaeological methodologies into Spain as a part of the influence of German archaeology and the culture-historical approach (Olmo Enciso 1991). It is then that two main trends emerged around the archaeology of this period: the first trend, which was also promoted by most Spanish archaeologists, aimed to identify the Christian (or, as it was termed, paleocristiano) origins of the Spanish culture. The excavations of Serra i Rafòls at the early Christian sites of Casa Herrera, of Serra i Vilarò at the churches of the Francolí, and those of Juan Cabré at Reccopolis or El Bovalar (which would eventually be excavated by Pedro de Palol) are all good examples of this Spanish-Christian research (figure 0.4). Parallel to this, and very closely linked to the lines of research which were normative in Germany at that time, there was a great interest in identifying the ethnic Visigoths and to establish a direct racial and national link between Germany and the Goths (from whom allegedly derived the kingdom of Spain), as, for example, in the work of J. Werner (Werner 1946; 1948) or W. Reinhart (Reinhart 1945; 1952). On this issue, Julio Martínez Santa-Olalla was of capital importance in developing this archaeological theory, because of his direct links between both the Spanish fascist party (FET de las JONS) and Nazi Germany (Martínez Santa-Olalla 1934). These links were ultimately the cause of his fall and the demise of this archaeological paradigm after the end of World War II (Olmo and Castro 2011; Tejerizo García 2012a). As elsewhere in Europe, these theses were mostly based on grave goods and the ‘national typologies’ of brooches, belt buckles and other items recovered from burial sites like Duratón or Carpio del Tajo, understanding these objects as individual items, and works of art, outside their archaeological context.

34 

The Iberian Peninsul a bet ween 300 and 850

Figure 0.4 Photograph of the Reccopolis excavations in the 1950s, with part of the ‘palace complex’ unearthed

Olmo Enciso 2008b: figure 9

These two approaches to the material culture dominated archaeological research for most of the 20th century, especially as the political agenda of Franco’s regime encouraged the search for the racial and Christian origin of the Spanish Nation. Both research avenues worked hand in hand, and

Introduc tion

35

largely relegated the study of Islamic archaeology of this period to fine arts and architecture, disregarding archaeological finds, pottery typologies and even settlements, as they did not add to the approved political discourse. The main consequence of this was not just the conscious disposal and dismissal of post-Visigothic layers in many sites, but also the development of the concept of a lack of any truly datable material for the early Islamic period – an idea which was only dismissed at the turn of the century. In Portugal, the military dictatorship established in 1926 pursued similar objectives to its Spanish equivalent, mostly to create a ‘national discourse’ which would identify the Early Medieval origins of Portugal. As in Spain, the Visigoths and the Sueves provided archaeological material which fuelled these interpretations, and they were subject to the first monographic studies, both by archaeologists and art historians. The two main researchers were Abel Viana and Fernando de Almeida (de Almeida 1962; Viana 1959), who carried out the first excavations at ‘paleo-Christian’ sites (such as Idanhaa-velha) and carried out the first classifications of their material culture. In this context, from the late 1950s and until the 1980s, the main researcher who dealt with post-Roman archaeology in Spain was Pedro de Palol. His work focused mostly on church architecture and the archaeology of Christian objects, working on the typologies developed by Gómez-Moreno and expanding them in scope and chronology (de Palol Salellas 1958, 1966). Palol was also involved in the excavation of new sites (many of them located in Catalonia and the Balearic Islands), opening up a large new body of material later studied by his students and successors, such as Josep Maria Gurt and Gisela Ripoll. Perhaps Palol’s main contribution was to insert Hispanic-Visigothic materials and architecture into broader Mediterranean scholarly trends. For the first time, Spanish Late Antique archaeology was not studied in isolation, and the first points of comparison emerged. In fact, one of the main consequences of Palol’s research was the establishment of the Reunió d’Arqueología (Paleocristiana) Cristiana Hispànica (figure 0.5), the first scientific forum in which early Christian archaeology was discussed (Gurt and Tena 1995; 2000; Gurt and Ribera 2005). It was, however, in the 1980s that Late Antique and Early Medieval archaeology fully developed, especially after the concept of ‘Late Antiquity’, as established by Peter Brown, was introduced into the newly emerging university system. The end of the Franco’s dictatorship (1975) and the subsequent arrival of parliamentary democracy (1978) brought a new political system based on regional developments. As a consequence, several universities and other research institutions were founded. This context allowed the development of regional research projects focused on their own territories. Moreover, the

36 

The Iberian Peninsul a bet ween 300 and 850

Figure 0.5 Front cover of the III Reunió d’Arqueologia Cristiana hispànica, held in Menorca in 1988 and published in 1994

establishment of new autonomous regions in Spain prompted the creation and consolidation of regional research groups and technical teams, promoted and held by either local or territorial administrations. These groups had particular interests in post-Roman material culture, which seemed to be more politically profitable and easier to define specifically for each region than the homogeneous and universal Roman culture. Amongst these are ground-breaking theses and works produced by, amongst others, Manuel Acién, Agustín Azkarate, Rafael Azuar, Ramón Bohigas Roldán, Miquel Barceló, Pablo Díaz, José Avelino Gutiérrez González, Sonia Gutiérrez Lloret, Antonio Malpica, Lauro Olmo, Eduardo Manzano, Gisela Ripoll, or Juan Zozaya, which in many cases were directly linked to large new excavation projects on late Roman, post-Roman, and early Islamic sites such as El Tolmo de Minateda, Reccopolis, Segobriga, and other smaller sites in the Basque Country (Quirós Castillo 2009; Quirós and Bengoetxea 2010). Similarly, the

Introduc tion

37

transition to a democratic government in Portugal prompted the creation of research groups in Medieval archaeology in the main universities of Oporto and Lisbon, or in key sites such as in Mértola. Researchers such as Carlos de Almeida, Fernando Branco, Mário Barroca, Santiago Macías, Claudio Torres, Virgilio Lopes or Luis Fontes began their prolific careers in this context. These works set the basis for new archaeological interpretations which moved away from traditional historical approaches. This change in attitude was further accelerated in the 1990s when the Harris excavation methodology was definitively introduced into Spain, parallel to an exponential increase in the number of archaeological interventions linked to new construction projects, which has been the main motor of the Spanish economy in the last two decades (Parga-Dans 2010). The main result of this was not only that the volume of excavated material multiplied considerably, but also (and more importantly) that Early Medieval rural contexts were properly identified and excavated. New sampling techniques became widespread, so that for the first time bioarchaeology, microscopic and isotopic analyses became common, triggering the emergence of the first publications on archaeometric analyses of archaeological material. New local centres of archaeological research and heritage management emerged in this period, such as the Consorcio de Mérida (CCMM), the Servicio de Información Arqueológica Municipal (SIAM) in Valencia, or the Taller-Escuela de Arqueología (TEDA) in Tarragona, to mention some outstanding examples, which devoted a lot of work to Late Antique and Early Medieval phases. Their work was very important in improving our knowledge of the material culture, constantly refining the chronologies given by ceramic typologies (Aranda González 2014). It was at this time that Spanish archaeologists and historians first developed close academic links with foreign scholars with regards to the archaeology of this period; first with the Italians and later with the British. Derived from these contacts, Spanish archaeology first started to ask questions which differed from traditional documentary perspectives, whilst historians also began to introduce material culture into their investigations (cf. Quirós and Bengoetxea 2010).

III

New theories and new data: archaeology in the 21st century

As a result of the large body of evidence generated during the previous decade and based on the new theoretical perspectives developed by the earlier generation of scholars, in the early years of the 21 st century there

38 

The Iberian Peninsul a bet ween 300 and 850

Figure 0.6 Photograph of one of the open-area excavations of the early 21st century: The site of Gózquez (Madrid)

Photograph by AREA S.C., used with permission

were considerable changes in the study of the archaeology of the period between the fourth and ninth centuries. Furthermore, these years saw the consolidation of long-lasting ongoing projects and the discovery of new sites: Reccopolis, El Tolmo, el Castellar of Elche, Zaballa, Aistra, Morería in Mérida, La Almoina in Valencia, Gózquez de Arriba or Can Gambús-1 and the rest of the villages of Madrid and Catalonia, Madinat Ilbira in Granada, La Encarnación in Seville, Vega Baja in Toledo, Santa María in Vitoria, the pottery finds of Vigo and many others (figure 0.6). Also occurring at this time was a reassessment of old sites and monuments, such as the so-called Visigothic and Asturian churches, the monumental remains of Barcelona and the later phases of Conimbriga. These excavations and re-assessments have produced a large volume of new data, which have added to the even larger amount of large and small-scale rescue excavations which were carried out before the economic crisis and the collapse of large construction projects. This implies that, as of this date, there are an incalculable number of boxes with Late Antique finds stored in museums which are only very slowly (if at all) being processed. The corpus of literature produced in this

Introduc tion

39

period is vast (e.g. Ferreiro 2006; 2008; 2011; 2014), as are current trends and discussions (Wood and Martínez 2016). We will only highlight the main current archaeological debates. As a consequence of this lengthy period of ongoing research, new critical lines of argument began to be developed amongst archaeologists, as a result of the realisation that the large amount of new data did not correspond with the paradigms and systems which had been previously proposed. These new approaches aimed, necessarily, to reconsider traditional explanations. One of the main concerns of research in recent years has been the rural world and the peasant societies in this period, which have blurred some established paradigms (for example those who saw peasant societies as ‘simple’ and ‘miserable’) and opened new paths of research (Kirchner Granell 2010a; Quirós Castillo 2009; Tejerizo García 2017).5 Beyond old historic perspectives of collapse and decline (cf. Ward-Perkins 1997), the study of Late Antique urbanism has also involved a large group of researchers, mostly archaeologists, who have engaged with the subject of what towns were really like in the Late Antique period, and how a new type of urbanism emerged (mostly compiled in Olmo Enciso 2008a). Another clear example of this new trend in research on the Early Medieval period are the new chronologies assigned by Luis Caballero and his team (María Ángeles Utrero, Fernando Arce, José Ignacio Murillo, and Francisco Moreno) to a number of churches that Art History had traditionally assigned to the Visigothic period. Caballero’s research led him to investigate new questions about the monuments, questions that could only be answered by a consideration of the archaeology, and that directly collided with traditional interpretations when he pushed the chronology of some of these churches into the eighth-ninth centuries (Caballero Zoreda 2000; further discussed below, in Chapter 6). These interpretations have been very problematic and are not widely accepted amongst the Spanish academic community. However, beyond Spain and particularly amongst British scholars Caballero’s arguments are being more widely accepted (Collins 2012; Walker 2016; Wickham 2010). Despite these years of constant research and debate, and acknowledging the various contributions to our understanding of the period provided by the new archaeological research, there are still dividing lines. To provide some examples, the main divisions are those between scholars dealing with urban and rural contexts in the Visigothic period and those separating late and post-Roman from Islamic and Umayyad perspectives. Only in the last few years have these lines begun to be blurred, following various attempts to 5

Not without opposition and criticism, see Chavarría Arnau 2012.

40 

The Iberian Peninsul a bet ween 300 and 850

put together research groups in discussion fora such as Visigodos y Omeyas (organized by the National Research Council – CSIC), or the ‘711 congress’ held at Alcalá de Henares which commemorated the 1300th anniversary of the Islamic conquest of the Iberian peninsula.6 However, one of the residual ideas from earlier interpretative paradigms which still has considerable weight in current interpretations is the idea of a dark and unknown eighth century. AD 711, the year of the Islamic invasion of Spain, has for many decades been a dividing line that could hardly be crossed, either by historians or archaeologists, and many publications have defined it as an inflection point after which things really changed. This idea is only very slowly being revised, as new structures and finds are excavated that can be dated to these chronologies. It is only now that archaeological interpretations covering the long term can be given about the transition between the Visigothic and early Islamic periods, and the results show that, unsurprisingly, there was no real archaeological rupture in 711. It is this idea of a longer continuity of late and post-Roman structures into the first decades of the Islamic period that we want to underline in this publication, dismissing historicist approaches to the topic. Lastly, it should be noted that in recent years foreign scholars have begun to pay attention to the Late Antique and Early Medieval archaeology of the peninsula, not only from theoretical points of view, but also by taking active part in excavations in the Iberian peninsula. This demonstrates the internationalization of the topic and the engagement of post-Roman and Early Medieval Iberian material culture with broader historical questions.7

IV

The long and short of the evidence

As already noted, the body of available archaeological material is large, and it has expanded massively in recent years, but a short caveat should be introduced regarding the extent of the evidence available and the depth and range of interpretations. Whereas this is not the place in which to discuss the different approaches to understating the past from a material and textual perspective, it is important – this being a work based on the analysis of material culture – that we underline its shortcomings. 6 Caballero and Mateos 2000; Caballero et al. 2012a and b; VV. AA. 2012. 7 See, again, Esmonde Cleary 2013 and Wickham 2005, but also Bowes and Kulikowsi 2005, Reynolds and Quirós 2010.

Introduc tion

41

First and foremost, due to the nature of most excavated sites, the evidence is only partial. In only a very few examples we have access to a whole site, so conclusions and interpretations have to be made based on narrow windows into the stratigraphy. It is true that there are some extraordinary examples, such as Morería, Zaballa, Reccopolis, Eio, or Gózquez, in which the excavated area is quite extensive, but in most cases the dug areas do not offer a view of the whole site. Furthermore, in many cases modern truncations and earlier archaeological approaches to Late Antique levels have either destroyed or left these contexts unrecorded. Despite this, by putting together the small snippets of information within a large urban site (as it is the case in Mérida, Córdoba or Barcelona, where there are many small interventions with Late Antique phases), or by comparing and contrasting various similar sites (as can be done with the rural villages in the province of Madrid), it is possible to reach solidly-based conclusions and interpretations. Furthermore, the different amount of information available from various sites and regions causes some examples and case studies to appear overrepresented in our interpretations (e.g., Mérida). This sometimes causes problems with what the written sources may say, a point which historians usually make when addressing the material culture. This issue, however, when seen from an archaeological perspective, is not that important. The Aristotelian fallacy that ‘absence of evidence is not evidence for absence’ clearly applies to the archaeology of this period, and we can only make strong claims when dealing with fully excavated sites. This does not, however, prevent us from making general interpretations, comparisons and extrapolations based on the available evidence (or backed by theoretical explanations or comparisons from other regions), as long as these are indicated and highlighted. Another issue that the archaeology of this period faces is the lack of proper and well defined chronological sequences, and that the regional fragmentation of the material culture makes inter-regional chronological comparisons problematic. While this was a valid point perhaps ten or fifteen years ago, current understandings of pottery sequences and technologies have massively improved, not just in the Iberian peninsula, but in Europe as a whole.8 It is still difficult to spot-date contexts to particular chronologies, as can be done in the Roman period with the well-dated typologies of Samian wares, but we are dealing with a corpus of evidence that can be sequenced within 30-year margins, which is an incredible step forward compared to the situation just a decade ago. 8 Aranda González 2014; Caballero et al. 2003; Vigil-Escalera and Quirós 2016. Cf. Jervis et al. 2017.

42 

The Iberian Peninsul a bet ween 300 and 850

Finally, a more controversial issue that this book also considers is the standard and quality of the grey literature. Due to the combination of the urgent character of many of the archaeological interventions in the past decades, the period of crisis of 2008, administrative issues, or lack of contact between academics and professional archaeologists, most of the recent commercial excavations have been only partially (if at all) published. Sometimes, the technical reports are still to be finished years after the excavation or are so misleading that little useful information can be found in them. This problematic assessment of the primary evidence has been addressed very carefully in this book: as it is a problem with secondary literature, we have tried to focus only on the archaeological data, analysing the proposed interpretations with care.

V

The nature of the transformations: Regionalism vs centralization

One of the mainstream topics in recent research is the idea that centralization and regionalization are the two dialectical processes that explain the nature of Late Antiquity in the Iberian Peninsula. These interpretative paradigms of material culture have sprung from recent general studies, led mostly by British scholars (Ward-Perkins 2005; Esmonde Cleary 2013; Wickham 2005, 2008), soon followed by specif ic analyses on Iberian case-studies by Spanish scholars (Castellanos and Martín 2005; Escalona and Reynolds 2011). These two processes can be used to effectively answer the questions that arise from the new archaeological interpretations of the evidence, especially as they go beyond traditional teleological approaches. These traditional approaches def ined a linear sequence of different material cultures succeeding each other, a model that is applicable to the entire peninsula. Throughout most of the 20th century, the political unity of Spain and the Spanish nation was an essential part of the political agenda that backed this scholarship. Nevertheless, this explanation is too simplistic. It is impossible to give general, homogeneous, and all-encompassing views about the archaeology of the Iberian peninsula in either this or any other period, because there were many different regional and local realities. The same happened in many other parts of the Mediterranean, which were formed by a series of geographical and social regions. These regions needed to interact with each other in order to secure stability. As a result, as political relations become more complex, the relationships between regions and

Introduc tion

43

local communities increased (an issue raised for this period in Wickham 2005, but more broadly defined in Horden and Purcell 2000). For the Iberian peninsula these processes of regionalization became more acute after the crisis of the fifth century, closely related to the collapse of the Roman political system. The various geo-economic areas that Roman politics had merged into single larger units had to rely on local resources and alternatives when the imperial administration ceased to protect and promote these interactions (Esmonde Cleary 2013: 398-417; Kulikowski 2004). Up to a certain extent, it could be said that in some regions of the peninsula, inter-regional interactions and territorial models of the imperial period were Roman constructs that could not survive without the centralizing Roman infrastructure. Often the nature of these alternative systems of territorial control and regional interaction remind us of, and seem similar to, pre-Roman systems. The speed with which these were adopted may hint at some sort of return to previous patterns (Martínez and Tejerizo 2015). This may indicate how superficial the imperial system had been in peripheral areas such as the Cantabric coast or the Northern Meseta. In fact, many rural and fortified sites which had traditionally been dated to the Iron Age turned out to be, after careful study, post-Roman in date (e.g., Tejado Sebastián 2011). To counter these trends, there were also up to five different political entities, beyond the final attempts of the late Roman Empire, which carried out centralizing policies: the Visigothic kingdom, the Suevic monarchy, the Byzantine Empire, the Umayyad emirate, and the Asturian kingdom. Of these polities, only the Umayyads managed to fully achieve a strong centralizing force similar to the Roman Empire. Similarly, the Catholic Church managed to form a structured (not necessarily centralized) system that outlasted the other polities – perhaps because it was based on urban and local elites, rather than the imposition of a top-to-bottom centralized system. Throughout this book we will see how both centralizing models (political and religious) carried out their agendas, and how this can be seen through the archaeological record. The material culture of this period is a reflection of the various degrees of success and failure that these centralizing forces had on regional dynamics and the attempt to get a comparative picture of the different social regions will be one of the major aims of this book. With this in mind, as the starting point for our narrative we introduce the archaeological material and its related discussions by looking first at the situation in Roman Hispania by the beginning of the fourth century. The

44 

The Iberian Peninsul a bet ween 300 and 850

book is structured into three parts, each of them chronologically labelled according to the phases and nomenclature stated above: Part 1 (Chapters 1-4) will deal with the Late Roman period, Part 2 with the post-Roman period (Chapters 5-7), and Part 3 with the Early Medieval period (Chapter 8). Each part will be divided into thematic chapters, so Chapter 1 deals with the transformations of the Roman world, Chapter 2 with the transformations of the urban world, Chapter 3 with those of the rural world, and Chapter 4 with the Christianization and Germanization of the material culture. In Part 2 there is a chapter on towns (Chapter 5), the countryside (Chapter 6) and the characteristics of the material culture (Chapter 7). The last part contains only one specific Chapter (8) on the substitution of Late Antique material culture by newer, Early Medieval alternatives. All of this will be followed by a final chapter containing general conclusions, and the appendices with maps and lists of sites, as well as other useful reference tables.

Part 1 The Late Roman period

1

The settings of late Roman Hispania

The differences between the early and the late Roman Empires are both formal and essential. Following Peter Brown’s seminal works (especially Brown 1976), Late Antiquity can be seen as a completely different time period to the Greco-Roman, Classical Antiquity. As outlined before, the objective of this book is to focus on the responses in Late Antique period to the transformations of the Roman world which materialized from the fourth century onwards, and on how these evolved throughout the fifth to eighth centuries, and subsequently how by the ninth century they were later substituted by Medieval elements. In order to do this, it is necessary to understand the fundamental changes and their causes which characterized the period of Late Antiquity. These will be outlined in this chapter, although the consequences of these changes will be further explored in later ones.

Roman Spain on the eve of late Antiquity By the year 300, Iberia had been part of Rome for over five centuries, since the first arrival of the Romans during the Second Punic War. The slow conquest of the peninsula was paralleled to a slow but steady integration into the Roman system in all fields: from the economy to politics, culture and society. Iberia changed from a distant territory from which to extract raw materials into a heavily urbanized province – especially the Mediterranean coastal regions and Baetica, where the ‘Romanization’ (understanding this term in its broadest sense of ‘integration in the Roman culture and its administration’) was perhaps comparable to that of Southern Gaul and North Africa (at least for some parts). Furthermore, Hispania was the first place outside Italy where the Romans set up a colony, Italica, and the Hispanic landed aristocracy was important enough to provide Rome with its first non-Italian emperor: Trajan. From an archaeological point of view, this integration is first and foremost seen in the degree of urbanization. By the eve of Late Antiquity, towns in Hispania had been through at least three centuries of Roman monumentalization. The urban network of Roman Hispania was based on three types of settlement: pre-Roman settlements, Republican foundations and Augustan foundations. Many of the large monumental cities correspond to this last group, such as Mérida (Augusta Emerita), Zaragoza (Caesaraugusta), and Astorga (Asturica Augusta). Added to these, many pre-existing settlements

48 

The Iberian Peninsul a bet ween 300 and 850

were greatly modified or expanded by Augustus himself, as part of his process of integrating the provinces into his new imperial system, in order to control economic resources and taxes. These cities, around thirty, were usually equipped with elements and amenities traditionally considered to be indicators of Romanization (Houten 2016; cf. Wallace-Hadrill 2008: 149-96), that had a direct link with Roman and Italic architecture and urbanism, including such elements as fora with basilicas and temples, spectacle buildings, aqueducts, baths, and walls. This last element is particularly interesting because they have been traditionally interpreted as a reflection of urban decay after the third century. Nevertheless, walls are a key feature that characterizes a Roman town and their construction is not necessarily related to periods of political instability. Most of the walls of Hispanic towns were foundational, as the cases of the three Augustan capitals: Tarragona, Córdoba and Mérida. Reforms and reconstructions, usually considered a consequence of crisis in the third century, have since been excavated deeper, analysed, and re-dated to the fourth-fifth centuries (see Chapter 2). Augustan colonies were usually founded with the same administrative and social structures (brought by the settlers) as those found in contemporary Italy, whereas Republican foundations and pre-Roman towns only very slowly achieved and obtained all of these elements. It was only after the Flavian period, when Latin citizenship was extended to all Hispania, that the large monumental phases of most towns were constructed (Alfőldy 2000; Fear 1996: 131ff; Hernández Guerra 2008). Aqueducts, for instance, directly linked to bathing and Roman urban culture, and notoriously expensive to build, multiplied manifold throughout Roman Hispania during the Flavian period: from under ten in the Julio-Claudian period to over fifty by the end of the first century AD (Sánchez and Martínez 2016, appendix). HispanoRoman towns were organized and ranked in three main tiers, according to their function in the early imperial provincial system. There were three provincial capitals, which are, not surprisingly, the biggest and most heavily monumentalized cities in the peninsula: Mérida, Tarragona and Córdoba. Second to these were the capitals of the different conventūs iuridici, the intra-provincial fiscal and administrative subdivisions (figure 1.1). Under these were the remaining secondary towns. By the end of the second century, all the elements that characterized Roman urbanism and monumentality were well established in Hispania. Despite traditional interpretations of a third-century crisis (Arce Martínez 1978; Fernández Ubiña 1982; cf. the first critical analysis from an archaeological point of view in Cepas Palanca 1997), these were preserved into the fourth century, and many times expanded and improved (for an overview,

The se ttings of l ate Roman Hispania

49

Figure 1.1 Map of Roman Hispania, and its administration in the late Empire, with main administrative cities (provincial and conventual capitals): Mérida, Córdoba, Tarragona, Écija, Astorga, Braga, Zaragoza, Cartagena, Clunia, Cádiz, Seville, Lugo, Beja, and Santarem

see Kulikowski 2006). Beyond their monumentality, towns were also the key administrative centres, as all the territorial Imperial and provincial government was focused upon, and the whole landscape articulated around, them. Despite this overall urbanization, towns seem to cluster along the Mediterranean coast and up the Guadalquivir and Ebro valleys, the Meseta and the north being more sparsely dotted with urban centres. In this sense, the countryside in the towns’ immediate surroundings was equally Romanized, in as much as it displayed the same structures that could be identified in the rest of the Mediterranean. Large tracks of monoculture devoted to cereal, vines or olives, administered and exploited from villas belonging to the urban elites, were common place (Gorges 1979). The pre-eminence of the villa in the Roman countryside is such that these type of productive-residential complexes form the vast majority of Roman rural settlements found for the imperial period. These villas, if anything else, were a symbol of the control of the territory by urban elites, and integrated in the imperial economic system, as the large tracks of monoculture seem

50 

The Iberian Peninsul a bet ween 300 and 850

to have been directly aimed at exporting agricultural goods, such as oil and wine. The settled landscape of Roman Hispania was in many instances focused on these two pivotal points, and usually, settlement in lowland towns was a consequence of direct state enforcement. This was true especially in the areas of the western half of the peninsula, where military pacification was followed by the establishment of new settlements (Curchin 2004: 40-90). In this way, the Roman populated landscape was a reflection of the Roman administrative system, which broke with traditional, pre-Roman settlement and economic patterns. The Iberian peninsula became fully integrated in the Mediterranean trading network. Both as a result of private entrepreneurship and imperial investment it was thoroughly exploited for its abundant natural resources (further discussed in Chapter 3). Other productions closely related to statedirected trade, the annona, such as olive oil, wine, salted fish and sauces, were common, and Hispanic amphorae are to be found in many places in the Western Mediterranean, above all in the Mount Testaccio (figure 1.2) in Rome and the northern limes (Blázquez and Remesal 2000; Étienne and Mayet 2000; Lagóstena Barrios 2001; Revilla Calvo 1995). The Roman state was also very interested in the minerals present in the peninsula, and large areas were dug out to obtain iron, gold, copper, lead, and tin ores. The main mining centres were located in Las Médulas, Cartagena, and Río Tinto. Lastly, Hispania was also known for its marble and the lapis specularis. Marble quarries, such as those of Borba-Estremoz, in Portugal, or those at Trigaches, in Spain (Orejas and Sánchez-Palencia 2002; Rico 2005) were first exploited in the first century AD. In the southern Celtiberia, a region centred around the current province of Cuenca, lapis specularis (‘mirror stone’) was quarried under imperial supervision, and exported around the Empire, even to Italy, as a substitute for window glass due to its solid and see-through nature (Nat. Hist. 36.160-1). The exploitation of the landscape and the necessity to interconnect the various regional centres of Roman administration led to the establishment of a road and bridge network (Mañanes and Solana 1985; Sillières 1990). In many cases these roads were no more than the confirmation of pre-Roman routes, such as the so-called Vía de la Plata running north-south on the western part of the peninsula, or the Via Augusta, which runs all along the Mediterranean coast. These, together with the province-conventus system prompted the development and tightening of links between towns. Overall, it is widely accepted that Hispania was, in many aspects, a very Romanized and well-integrated region (Ozcáriz Gil 2013; Rodríguez Gutiérrez

The se ttings of l ate Roman Hispania

51

Figure 1.2 Photograph of Mount Testaccio, in Rome. This artificial hill was created from the discarded and broken amphorae which brought Hispanic olive oil to Rome as part of the annona during the early Empire

2011), perhaps with poorer soils, if compared to other provinces, and unevenly populated but, by and large, fully part of the Empire and playing an active role in its socio-cultural and economical system. This integration in the Roman system can be perfectly seen in the changes that Hispania went through after 300. Even if the Hispanic provinces played a minor role in the great events and civil wars of the fourth century, the transformation of the Empire had a direct impact on the Iberian provinces. The change, however, would eventually (as in most of the West) lead to the collapse of the Roman administrative and economic system.

Roman internal transformations In order to understand the transformations that characterize Late Antique Spain and how it drifted away from its ‘Classical’ past, it is necessary to look at the macro-historical processes that took place across the Empire. These are difficult to explain in archaeological terms, although following the

52 

The Iberian Peninsul a bet ween 300 and 850

theories put forward mostly by British scholars (Esmonde Cleary 2013; Halsall 2007; Heather 2005; Ward-Perkins 2005; Wickham 2005), and in conjunction with recent research, it may be possible to outline these circumstances. We will set the starting point for these transformations with the emergence of the Tetrarchy.9 The Tetrarchy was a response to the period of political instability during the third century, which in the Iberian case is clearly visible in the exponential increase in the number of miliaria (mile stones), which lined for miles the main Roman roads. These were dedicated to the many ephemeral emperors of the third century. Such proliferation of emperors and milestones is not enough evidence to claim that there was a third-century crisis in the Iberian peninsula, as has been traditionally proposed (Cepas Palanca 1997). The Tetrarchic system has traditionally been explained as the establishment of a way of running a heavily militarized empire which could not rely anymore on conquering campaigns to obtain funds (Jones, Arnold 1964; cf. Rees 2004). It also seems to have been the time in which the nature of Rome as a conquering empire changed into a defensive system for preserving the frontiers (cf. the classical approach to the topic, still largely valid: Luttwak 1976). It is after this, and not before it as has been traditionally proposed, that there is a general concern for the defence of towns, and the development of new town walls. The implantation of a renovated administration in the Tetrarchic period, together with its later adaptation by the Constantinian emperors, led to a series of root-and-branch political changes with far-reaching economic and social consequences. This is especially clear in Hispania, even if its role was secondary inside the Western Empire due to its rear-guard location, far away from the limes. The main change was the creation of a new administrative framework within the province, the civitas (city-territory), while the provinces were grouped into larger dioceses (dioeceses).10 In the Iberian case, this was the Dioecesis Hispaniarum. Such an administrative framework would become the basis for the ordination and division of the Hispanic Church and the Late Antique development of regional diversity. In fact, Tetrarchic reforms were as necessary for the survival of the imperial system during the fourth century, as they were inevitable accelerators of its end in the fifth. This 9 There has been much discussion about the origins of Late Antiquity and the development of the Tetrarchy; discussion which, even if it is of great interest, is at a tangent with regards to this work. 10 Notitia dignitatum occ. III.5-13: Provinciae Hispaniarum VII: Baetica, Lusitania, Gallaecia, Tarraconensis, Carthaginensis, Tingitana, Balearica.

The se ttings of l ate Roman Hispania

53

new administration was aimed at maximizing the revenue collected by the central state (mostly to pay the army) and to improve the defence of the provinces, should the necessity arise. Considering this, the creation of the civitas caused a gap between those towns which became primary nuclei and those which became secondary foci, usually those in areas which were not well communicated or which were not directly linked to state revenueextracting activities. Amongst the former stand out the seven capitals of the new Hispanic provinces: Pollentia (although this one perhaps did not stand out that much), Tangiers, Tarragona, Cartagena, Córdoba, Braga and, above the rest, Mérida (figure 1.1). Zaragoza, the most important city in the Ebro valley, could be added to this list. This prosperity also extended to those settlements directly linked to Mediterranean trade, as can be seen in the Mediterranean harbour of Portus Ilicitanus, where the increase of fourth century African imports (African Red Slips – ARS- and amphorae) is paralleled by an extended new harbour infrastructure. On the contrary, for many old towns of both primary and secondary rank, the new system meant the final blow to the mode of urban life which had begun in the first and second centuries, but which by the third had become stagnant. Some of them went into deep urban decay through the fourth and fifth centuries, including Munigua and Italica in the South, Calagurris, Bilbilis, Jaca and Lérida in the Ebro valley, Numantia and Termes in the northern plateau and Edeta, Sagunto, Baetulo and Emporiae in the East coast. Many others came to the end of their urban life, such as Contrebia Belaisca, Iuliobriga,11 Huesca, Mirobriga and two ancient capitals, Clunia, the most important capital in the northern plateau, and Cádiz. The total number of civitates is difficult to assess. Michael Kulikowski quotes 130 for the late Roman period (Kulikowski 2004: 287), but by the Visigothic period, when a civitas was linked to a bishop (a definition different to the concept of a late Roman civitas), the number of Visigothic civitates in Hispania was down to 71.12 This number does not include the seven bishoprics 11 Despite its mention in the Notitia Dignitatum (occ. LXII.24-32) as the place that, in the fourth century, hosted the Cohors I Celtiberorum from the castra of Ciadella (Lugo). 12 Carthaginensis (23): Acci, Basti, Beatia, Begastri, Carthago Spartaria, Castulo, Complutum, Dianium, Elo, Ercavica, Illici, Mentesa, Oretum, Palantia, Segobriga, Segontia, Segovia, Saetabis, Toletum, Urci, Uxama, Valentia, Valeria. Baetica (10): Assidona, Astigi, Corduba, Elepla, Hispalis, Iliberri, Italica, Malaca, Tucci. Lusitania (13): Abela, Augusta Emerita, Viseo, Caliabria, Caurium, Conimbriga, Ebora, Egitania, Lamecum, Olissipo, Ossonoba, Pax Julia, Salmantica. Gallaecia (10): Aquae Flaviae, Asturica, Auria, Bracara Augusta, Britonia, Dumium, Iria Flavia, Lucus Augusti, Magnetum, Tude. Tarraconensis (15): Auca, Ausona, Barcino, Calagurris, Caesaraugusta, Dertosa, Egara, Emporiae, Gerunda, Ilerda, Orgellum, Osca, Pompaelo, Tarraco, Turiaso. This list does

54 

The Iberian Peninsul a bet ween 300 and 850

of Septimania, and it has the added problem that not all of these bishoprics can archaeologically be classified as towns and that some of these episcopal sees were new Visigothic establishments (cf. Gutiérrez et al. 2005). However, this figure can give a rough indication of the territorial organization. In this new urge to extract more revenue for the state, civitates affected traditional urban civic culture. Traditional magistracies (but not the curiales) became redundant, as the new governing figures, the vicarii, the counts (comites) and their subordinates, were directly linked to the central administration, and not elected from the local elites. This meant that urban aristocrats were cut off their traditional route to civic power, and officeholding had to be achieved via other means. Furthermore, as more taxes were being collected from municipal territories and sent to the central administration, town councils were left without a steady income, making euergetic donations far more burdensome (Liebeschuetz 1992: 6-20, esp. 7-15; cf. Kulikowski 2005: 50, 56-7). In the long run the effects of these changes set the tracks on which the late Roman urban model developed. First, the new administrative framework of towns made old civic centres (such as fora and basilicas) functionally redundant and those public monuments funded or maintained by town councils economically unviable, unless there was a direct imperial intervention (such as Mérida and the Constantinian renovation of its spectacle buildings). All this, together with the need for a new type of architecture of power that represented the new authority, meant that early imperial monuments were abandoned, transformed, or even systematically pulled down. As part of this reorganization of local power and of the tax system, early imperial modes of land exploitation (i.e., the villas and their economic system) were also soon transformed. The villas were an extension into the countryside of curial elites, and were a fundamental basis of the economic system, as they mass produced basic elements (mostly grain, oil or wine) for generalized trade. This meant that they were as susceptible to change as the municipal curial system. By the end of the third and especially during the fourth centuries the Roman villa territory began to change: there was an uneven bipolarization of the villas (Chavarría Arnau 2007; Esmonde Cleary 2013: 250-6). This means that traditional early imperial villas (with their pars urbana and their pars rustica, according to Vitruvian nomenclature) gravitated towards two different points. The majority of villas decreased in importance, size and, especially, luxurious dwelling spaces, becoming not include, however, civitates without bishops, such as the new foundations of Reccopolis, Victoriacum, Amaya or Ologicus.

The se ttings of l ate Roman Hispania

55

Figure 1.3 The villa of Carranque, in Toledo, is a perfect example of one of the large palatial rural residences that can be linked to the late Roman Hispanic aristocracy

Photograph of the current excavations (by Dr. Patricia González, used with permission)

simpler land exploitation sites, linked to what the sources refer to as the ‘colonate system’ (Jones, Arnold 1958; Sirks 1993). On the opposite side we find that a small number of villas became larger and more elaborate, a few becoming truly palatial in their proportions and decoration. In these villas, the presence of productive areas is separated from their main building, often forming a separate complex, which matches the development of production-focused settlements as in the other type of late Roman villa. These archaeological transformations indicate not only changes in the economic focus of villas, but also the social dimension this implied. Some local aristocrats lost interest in luxurious country estates, and preferred to preserve them as simple economic centres. Conspicuous display and representations of power were now moved to the urban domestic sphere. It was in urban and suburban housing, not in countryside villas, that most of the urban elites now entertained their visitors. The grand palatial rural villas (figure 1.3) mentioned earlier (such as those preserved in the Castilian plateau: Carranque, Noheda, La Olmeda and Almenara-Puras) seem to be remarkable exceptions, perhaps even linked to imperial patronage or familial links to the imperial household (Gorges 1979: 51-7).

56 

The Iberian Peninsul a bet ween 300 and 850

Another consequence of the imposition of the Diocletianic-Constantinian administrative model is the militarization of the administration and of the elites. Whereas the military had always been a key element in the politics of Rome, during the late Empire the military was increasingly integrated into the government, especially following the period of the general-emperors of the military anarchy in the third century. As this permeated down to the provincial elites, it added another wedge to separate the curiales from their earlier municipal duties, offering a new career in which to obtain rank and status. Whereas in other provinces, closer to the frontier (such as Gaul and Germany) this is more evident (Esmonde Cleary 2013: 93-6), in the Iberian peninsula it is less so, but still not uncommon, especially after the accession to the throne of Theodosius, who was originally from Gallaecia, in the year 379. The emergence of a new type of military-like brooches (Ceprián et al. 2013), the change in the traditional motifs and scenes in late Roman mosaics, where representations of hunting and horse races were dominant (Blázquez Martínez 1986), and the development of small weapons (daggers) as indicators of prestige amongst rural communities (such as the Simancas-type knife) can be interpreted as signs of this slight change even in peaceful provinces. Lastly, the most important transformation which swept through late Roman society was Christianization (figure 1.4). The spread and imposition of Christianity has been listed as one of the causes that prompted the decline of Rome, most famously by Gibbon (1999), when, in fact, it had nothing to do with the militarization and reorganization of the administration and the economic system. Christianity had slowly spread through the West, and in Spain the earliest indication of Christian activity could be the possible domus ecclesia of Mérida (see Chapter 4). By the fourth century, Christianity was only loosely established in Hispania: archaeology shows that there is not a material culture that can indicate a thoroughly Christianized society until the end of the fourth century, when it seems to have permeated to a majority of the elites. The acceptance of the new religion, protected by the Constantinians, later embraced and promoted to the official faith of the Empire, had a very deep impact on Roman society, spreading slowly from urban to rural contexts. Archaeologically, this can be seen in four main aspects: The first point is the slow but steady Christianization of the material culture, as new, Christian decorative motifs become more common, providing a new iconography and a new pool of images in which art could flourish. In many cases this was simply the adaptation or promotion with a Christian meaning of panMediterranean (not necessarily pagan) elements. Second is the confirmation

The se ttings of l ate Roman Hispania

57

Figure 1.4 Late Antique (fourth-fifth century?) decorated plaque with Christian symbols, such as the Chi-Rho with the Alpha and Omega

Currently held at the Victoria and Albert Museum (photograph by Dr. Patricia González, used with permission)

of inhumation (as opposed to cremation) as the favoured way of burial. This burial practice was not introduced by Christianity, and it had in fact become popular from the second century onwards – as can be seen in the funerary areas excavated at Mérida (Bejarano Osorio 2004), Córdoba (Sánchez Ramos 2005) and Tarragona (del Amo Guinovart 1981), but with Christianity inhumation became an essential practice, as Christians would rise again from the dead on the Day of Judgement. Thirdly, Christianity, once officialized, formalized a highly organized and hierarchical Church, protected by the Roman state. This meant that the Church obtained economic and political power at a local level, as bishops and their underlings became the new local religious magistracies. By the end of the fourth century bishops were in such a prestigious position that

58 

The Iberian Peninsul a bet ween 300 and 850

they were able to become the sole urban civic and religious leaders after the collapse of the Roman system and its local magistracies (see Chapter 5). Lastly, and as direct consequence of the second and third points, Christianity prompted the development of new Christian urban foci, away from municipal civic and religious sites. This was directly linked, at these very early stages, to the fact that Christian patrons did not have enough power or support to substitute traditional centres of power (this would happen later, in the sixth century).

External factors Beyond the internal transformations of the Roman state and the Christianization of Roman society – although the Dark Ages Cold Period and the Justinianic Plague could also have had a role in these transformations – there were also some external factors linked to the transformations that characterize Late Antiquity. The most important change was the arrival and settlement of the Barbarian populations (Goths, Vandals, Alans, and Sueves; figure 1.5).13 The problem is that, chronologically, none of these seems to have had any direct impact until the fifth century – or after. This in turn highlights how important the internal transformations were during the fourth century for the development of a Late Antique materiality. In 409, invited by the usurper Gerontius, the Sueves, the Vandals, and the Alans invaded the Iberian peninsula. From 416 they were fought by the Visigoths, who did so on behalf of the Romans. The scant references in the chronicles to battles and sieges has led historians to assume that the Northern Meseta had been laid waste by these wars, and that it is there where the Visigoths later settled (ideologically linking Castile with the Visigothic kingdom and, therefore, with the true identity of Spain). This argument was usually based on the fact that cloisonné jewellery found in tombs normally came from this area. The destruction and abandonment of towns and villas, and any burnt layer found in any site abandoned in this period, was usually dated to one of the Germanic raids. These catastrophist views were later substituted by a more peaceful interpretation in the 1990s, in which the Germanic peoples were migrants who needed to be accommodated within 13 There is much debate in the critical literature about the terminology with which to refer to the groups of people who entered the Roman Empire during the fifth century. We will be using the term ‘barbarians’, which even if it can be interpreted in a negative light, is still the term that was used in contemporary Roman sources, who were in any case biased against them.

The se ttings of l ate Roman Hispania

59

Figure 1.5 ‘Line and arrow map’ showing traditional views of routes of invasion and migration of the barbarians during the Migration period or Völkerwanderungszeit

Wikimedia Commons

the Empire, casting a less violent and less destructive light upon their arrival. Current approaches to the topic have, however, backed away from this revised interpretation, acknowledging that the arrival of the Goths and the Sueves was usually violent but also that there is more to their arrival than an invasion and a wave of destruction (Ward-Perkins 1997; 2005). This fifth-century situation accelerated the disintegration of the Roman imperial system, its institutions and infrastructure in the Iberian provinces. This generated a power vacuum over which neither Rome nor the newly arrived Germanic peoples had any effective political control. Roman power broke down; however, much was retained in the main towns of the peninsula. The isolation of civitates as the Roman administration disappeared prompted the transformation of towns and their territories, which no longer had to function in a taxing state. Together with this separation from the Roman world, life became less secure and less complex: the over-specialized Roman populations had to solve new problems which had been for centuries taken for granted. The steep decline in industrial, constructive and trading activities, some of which had already begun in the fourth century, were greatly accelerated in the fifth. The invading Germanic peoples acted as a catalyst for these transformations.

60 

The Iberian Peninsul a bet ween 300 and 850

This situation would change later in the sixth century, as the Visigoths and the Sueves established and developed early states, but in the earlier phases interaction between the Germanic peoples and the local communities was quite limited. Despite claims that the peninsula was politically divided amongst the Sueves, the Alans, and the Vandals (and the later Visigothic occupation), this interaction has to be seen as a simple extraction of tributes, as will be discussed later in Chapter 4. Another external factor which framed this period of transformation was climate change. The Dark Ages Cold Period is a climatic low which followed the Roman Warm Period (or ‘Roman Climatic Optimum’), which in the north-western Iberian peninsula has been dated to roughly between 450 and 950 (Desprat et al. 2003). This means that almost all of the period covered by this book was affected by lower temperatures than in earlier and later periods, with exceptionally variable weather, and with especially cold summers (Büntgen et al. 2011).14 Therefore, even if this was not perhaps evident to them, the different environment and natural conditions added another element of change to which late antique populations had to adapt. The degree to which this affected agriculture (crop failures, different agricultural seasons), settlement relocation patterns, the expansion of forested areas, and the late spread of the Justinianic Plague (cf. Rosen 2007: 201-3) in the Iberian peninsula have not been studied, and this work is not the place to discuss it at length, but it is a factor that should be considered in later discussions.

The beginning of regionalization The various factors mentioned above, both internal and external, are key to understand the development of the various processes of regionalization in Late Antique Spain. For instance, it is not a coincidence that the old province of Tarraconensis, which was the only province with a military garrison (at León), was the last province to be lost to the Goths after the invasions of 409. However, Rome’s nominal control over the province did not imply a continuation of the Roman system throughout its territory, as the foci of Roman control were spread apart, and focused mostly along the Mediterranean coast of the province. The direct presence of the Roman 14 Weather portents in the written sources of this period are quite common, especially in Gregory of Tours, who mentions floods and heavy destructive rains (DLH IV.31; V.33; VI.14, 25) and unsually harsh winters (III.37), and cold, rainy and frosty summers (VI.44; VIII.23; IX.44; X.30). Similar portents are mentioned in the Lives of the Fathers of Mérida (=VSPE II.21.).

The se ttings of l ate Roman Hispania

61

state in certain areas of the peninsula (or rather, the more direct connection between some areas with the core of the Mediterranean economy) ensured tighter links to the general Roman system than in others where there was no direct connection. This can be clearly seen in the archaeological record. The unified Roman peninsula of the early fourth century (socially, politically, and economically) contrasts sharply with that of the mid-to-late fifth. As the Sueves took the north-western corner of the peninsula as their own, the ties that connected the Cantabric coast, Gallaecia, and the Northern Meseta with the rest of the peninsula were cut. They had been weakening for years as state intervention in mining had declined, but an increasing isolation of strong urban centres began, detrimental to those secondary settlements which could not play a main role in the new circumstances. The civitates of the Southern Meseta, which were poorer and had worse communications than its neighbouring regions, suffered similar patterns. In these areas, for instance, urban networks began to fade earlier than in coastal regions; some towns had already been in steep decline by the fourth century. The emergence in the Northern Meseta and the Cantabric region of alternative types of central settlements, especially on hillforts (figure 1.6), does not show a return to pre-Roman patterns, but instead shows that towns were highly artificial modes of settlement in these areas once the presence of the central state declined. In a similar fashion, continuing Roman townscapes and rural territories, such as Mérida, were connected to coastal areas or other territories which remained centres of power even after being cut away from their direct political links to Rome. The diversification and regionalization of the material culture is a very useful indicator for these changes, as late Roman regional patterns have far more distinctive elements than earlier Roman local variations. Pottery productions became increasingly regionalised and imports continued only in areas connected to Mediterranean trading links, whereas newly carved marbles and other specialized craftsmen only continued to be active in Mérida and its surroundings (Sastre de Diego 2015; Cruz Villalón 1982). The new Late Antique economic regions were often limited by natural boundaries, where similar climates and soils prompted similar responses to the post-Roman environment. Usual dichotomies in the evolution of neighbouring territories are coast v. inland, valley v. hills and mountains, and Mediterranean v. the interior and Atlantic. In this way, some clearly defined regions with particular archaeological records are inland and coastal Gallaecia, the Cantabric coast, inland and coastal Catalonia, the upper Duero, the Upper Tagus or the Upper Ebro valleys, the South-eastern Mediterranean coast, etc.

62 

The Iberian Peninsul a be t ween 300 and 850

Figure 1.6 Photograph of the site of Bernardos, in Segovia, a late Roman hillfort which became a local centre of power once the nearby civitas lost its relevance as a regional node. Sites like Bernardos are very common across the north-western corner of the Iberian peninsula

This process was, as expected, very gradual and irregular. Throughout the fourth century it is only in the more marginal areas on the northern coast or deep in the Central Meseta that we see the first signs of disruption and the decline of the Roman urban network. Oiasso, a small harbour town directly linked to state mining, ceased to have any real urban appearance by the time it had been relegated to a secondary fishing harbour (Sarasola and Moraza 2011; Urteaga Artigas 2008). This, however, does not mean that it was completely isolated, as shipwrecks indicate the continuation of trade well into the post-Roman period, but rather that the economic and habitation centre was not nucleated around the town (Martínez and Tejerizo 2015). Something similar happened in the early Roman nuclei of Termes, Ercavica and Valeria, towns in the Central Meseta which had lost all political importance during the previous centuries and slowly ceased to be real cities, despite some minor continuity in occupation (Cepas Palanca 1997). This contrasts sharply with the towns which were upgraded and the lavish villas that were built by the newly promoted elites around Córdoba or

The se ttings of l ate Roman Hispania

63

Mérida. The pace of change became clearly evident during the fifth century, but it is a process that continued all through the post-Roman period. By the end of the de iure Roman control over Hispania, the Roman political system, which had been already damaged beyond repair, began to slowly crumble. In its dissolution, solutions to new problems flourished in Late Antiquity.

Changing tides in the economy of the late Empire As the Iberian peninsula was becoming atomized into its internal regions, the whole Empire was equally being subdivided into increasingly autonomous macro-regions (Gaul, Italy, North Africa), which slowly but steadily began to develop new trading networks and links. Or rather, the over-representative routes linked to imperial trade faded away as the Roman state ceased to be able to keep them running. The collapse of the Roman system led to a simplification of the economy (cf. Loseby 2005; Ward-Perkins 2001). The decline of tax revenues prompted a reduction of state-funded trade (and its secondary cargoes, such as ARS), which limited the spread of mass-produced agricultural goods, forcing producers to invest in more diverse crops, which generated less revenue. The loss of provinces (and revenues) to barbarian groups, the need to fund the army (on many occasions by handing over landed revenue to barbarians), and the lack of interest of local elites in contributing to the economy accelerated these processes. As far as the Iberian peninsula is concerned, these trends may have started early in the late Roman period, as the demand (or competitivity) of Spanish produce began to decline. From the second half of the third century Hispanic agricultural products, which had had prevalence in the western Roman trade (especially wine, garum and olive oil from Baetica and Tarraconensis), were slowly replaced by North African products (figure 1.7), which by the fourth century had monopolized the market (Aquilué Abadías 2003). This new commercial reality was a hard blow for some southern towns, especially for the millenarian city of Cádiz, that could not recover its economic dominance. This period coincides with the change of the capital of Baetica, which moved to Seville from Córdoba. After this, the harbour of Seville became one of the most dynamic market places of Late Antique Hispania, included in the Mediterranean commercial networks. More than 2500 pieces of imported pottery, dated between 250 and 550, were found in the excavation of the site of La Encarnación, and 92% of them came from

64 

The Iberian Peninsul a bet ween 300 and 850

Figure 1.7 Collection of African Red Slip (ARS) wares from Late Antique contexts in Cartagena. African produce flooded the late Roman Hispanic market and are a characteristic find in Late Antique archaeological contexts

Courtesy of Dr. Jaime Vizcaíno

Africa Proconsularis (Amores et al. 2007). Despite this period of thriving North African export, the production of fish products seems to have kept its prevalence in regional scale trade during the fourth and the fifth centuries. The effects this had on Córdoba are evident, as most of the finds for this same period seem to be local productions and imitations. What is certain is that fourth century Hispania developed a new economic model that combined two realities: on the one hand there was the increased importation of new products which achieved an absolute dominance of the previous Roman market; on the other hand, there was the increase in production by some local craftsmen, achieved even by imitating and re-elaborating the foreign products which were dominating the market. The way in which both of these realities worked together and the gradual increase of the responses by local craftsmen over the following two centuries define Late Antique Hispania and were responsible for remarkable differences among its regions’ economic prospects. Overall the late Roman period saw a reduction in the demand for specialized productions as local alternatives became more prominent. In the long

The se ttings of l ate Roman Hispania

65

term this would eventually lead to a situation in the post-Roman period, where some specialized trades and produces disappeared altogether as there was no longer any demand for them. This became especially true for specialized or exclusive trades which could only really flourish under adequate economic circumstances (as we will see with the Visigothic foundation of Reccopolis). By the end of the fifth century elements such as mosaic paving (with the singular exception of the Balearic Islands), vaulted engineering, salted fish sauces, or marble quarrying seem to have been lost. The fall of the Roman Empire in 476 has been seen as one of the most important landmarks in the narratives of European history. In its mythical representation, the deposition of Romulus Augustulus has been considered the turning point of the Roman Empire itself and even the end of the civilization (Castellanos Martínez 2013). However, recent narrative accounts of this period have shown that 476, and hence the ‘end of the Roman Empire’ was merely the consequence of a long historical process which began nearly two centuries before, when the political and economic structures on which the Roman Empire was based showed important signs of change. The military conquest of the Iberian peninsula prompted a quick ‘Romanization’ of the Hispanic provinces. Between the first and the third centuries Hispania was transformed into an important economic centre within the Roman political economy, with the development of the main material characteristics of the imperial culture, such as traditional Roman townscapes, the appearance of villas as an extension of the political domination of urban elites, or the appearance of typical Roman pottery forms. This archaeological set of characteristics sees a period of major transformation in the fourth century as a consequence of the crisis in Roman macropolitics. This situation affected the Iberian peninsula, showing important archaeological transformations. Internal and external factors may also help to explain these changes. The decay of urban life and the development of rural villas may be some of the former and the increasing importance of African production and the appearance of new barbarian populations were evidence of the latter. As stated at the beginning of this chapter, these processes of long duration were the historical framework in which the substantial archaeological changes of the fifth century onwards are framed.

2

New townscapes in the late Empire

This chapter will analyse the formal and material changes that affected Iberian towns in the late Roman period. By the fourth century the urban landscapes were saturated with monuments. After three centuries of construction there was, understandably, an end to large building projects, because of the walls, aqueducts, baths, theatres, amphitheatres, etc., that were already built. Furthermore, Roman culture had slowly drifted away from old patterns. For instance, late Roman society preferred horse races in the circus and venationes over traditional theatre plays, while gladiatorial combats were becoming both too expensive and less popular; the lack of public interest causing the buildings linked to them to begin a period of decline. This, together with the political-administrative transformations outlined in the previous chapter, meant that there was a change in public priorities which, slowly, began to lay the foundations of the Late Antique town model, which would be in existence throughout the late and postRoman periods (cf. Diarte Blasco 2014).

What constituted a late Roman city? Theoretical approaches, based on historical analyses and inter-regional comparisons across the Empire of late Roman urbanism, can be used to define this concept of ‘late Roman city’, in sharp contrast with the ‘classical’ or ‘early Roman city’. In the last 20 years, the interest in Late Antique urbanism, especially after the proposals of Liebeschuetz (esp. 1992; cf. Dey 2014), has developed greatly, so that it is now almost a discipline in its own right. The definition of the ‘late Roman city’ model is, nevertheless, not straight forward and there are various possible interpretations (as we have defined and described elsewhere: Martínez and Tejerizo 2015: 83-5). In the Codex Iustinianus (I.3.35), a town was defined by the presence of a bishop (Brandes 1999: 27) but, archaeologically, there are other parameters. Bryan Ward-Perkins (1996:4-14) defined a town as having administrative and religious functions, together with some more quantitative characteristics such as size, status, urban planning and significant public and private buildings. This definition was largely kept by Gian Pietro Brogiolo (1999: 99), who defined a town as fulfilling legal, military, ecclesiastical, economic, and administrative functions. In his definition there are two main characteristics: a) towns are part of a wider

68 

The Iberian Peninsul a bet ween 300 and 850

political context (the Roman state) but, and at the same time, b) have some autonomy in terms of local administration. Guy Halsall (1996b: 236-7) lists similar characteristics expected from a town, such as being the permanent settlement of a large population, that fulfils an economic purpose, and with administrative functions, but also includes the concept that towns are places in which social differentiation is present. Similar terms were used by Simon Loseby (2006: 72, 85-8) when defining towns in Gaul as administrative centres where the local power was held by the urban elites, although he also expanded on this by mentioning the two different tiers of post-Roman administration (municipal and royal or central) in which the civitates had become tax units. More recently, Simon Esmonde Cleary (2013: 100-1) has defined towns in the late Roman period as dense, non-agricultural settlements which provided services (religious, educational, legal, etc.) to their territory and that were characterized by their monumentality and their economic roles. All of these definitions are quite flexible, as is necessary when attempting to define towns in a period when the apparent homogenization that characterized the early Roman period was thinning. The slow introduction of new late Roman social and political priorities (e.g., the end of curial administration and the consolidation of Christianity) was soon followed by the need for a new architecture and urbanism. However, it should be kept in mind that these changes had to be carried out within living, existing cities, where the past was present and standing in the form of older buildings, and it was not as simple as it would have been if towns were simply built anew. The integration and adaptation of old spaces for new necessities and priorities is what really characterizes late Roman urbanism. During the late Roman period, when the state was still fully operative, adaptations and transformations were not as destructive or negative as they would be in later centuries, and it is in many cases a matter of subtle change. Some of these changes were actually positive and beneficial, a seemingly sharp contrast with general preconceptions. For instance, the new economical context of the Empire is behind the topographical reorganization of the Mediterranean coast. While many old towns could not preserve their urban status they did develop as secondary economic nuclei for other urban settlements. This is the case of three seaports in the Valentian coast: Sagunto and its port, ‘Grau Vell’, where trade activity is recorded during the fourth-fifth centuries despite the demise of the old settlement; Cullera (the old Portus Sucronensis) that included a small island located close to the coast where a monastic centre was later developed during the sixth century (see Chapter 6); and Portus Illicitanus, which also become an Imperial naval station.

New townscapes in the l ate Empire

69

The process of change was, however, unstoppable, and slowly but steadily appeared in four main respects: increasing fortifications, the adaptation and decline of public monuments, the disrepair of public infrastructure, and a final redefinition of urban nuclei and of the urban space. These four avenues of change will each be analysed in depth in the following sections. A final caveat should be mentioned at this point: the political and administrative reorganization of the urban network prompted the promotion of a few towns to higher tiers and the consolidation of many as primary nuclei. However, the number of towns which were relegated to secondary or tertiary roles is significant. Throughout this chapter we will see how the examples of ‘active’ transformations (construction, repairs, maintenance, continuities) are confined mostly to those towns in the upper tiers, where town life was preserved. Those examples of ‘passive’ changes (not necessarily destructive, but rather slow decline and abandonment) are, nevertheless, common to both primary and secondary towns.

Fortifications The increasing military pressure on the frontiers and the militarization of the Roman administration soon prompted the development of new fortified enclosures which were more effective (or more easily defended) than the early imperial ones. By means of reinforcing walls and towers, increasing their height, blocking gates, and reducing the perimeters, the effective defensive capability of a town was greatly improved in the Late Antique period. It is interesting to see the correlation between those towns promoted by or benefitting from Diocletian’s reforms and the size and magnitude of the new walls. It is during this period that towns and town walls became very closely ideologically linked, with walls directly relating to prestige; so much so that in the seventh century Isidore of Seville mentioned that ‘the walls are the city itself’ (Etym. XV.2.1: urbs ipsa moenia sunt). This is not something that happened only in Spain and, to cite a couple of significant Late Antique examples, walls are equated with cities in the mosaics of Saint Apollinare in Ravenna, in the Peutinger map, and in the Madaba mosaic map in Palestine (figure 2.1). Walls were, furthermore, very expensive to build, and required large amounts of manpower and resources (for Mérida, it has been calculated that 200,000 ashlar blocks were needed)15 to carry out their construction. This reinforcement or refortifications of earlier walls cannot, 15 As calculated by Miguel Alba (Consorcio de Mérida, pers. comm.).

70 

The Iberian Peninsul a bet ween 300 and 850

Figure 2.1 Walls as cities in the Late Antique imagination: Jerusalem is characteristically represented by its city walls in the Madaba mosaic in Jordan

Judith S. McKenzie et al., Manar al-Athar Photo-Archive, Oxford 2013, available at http://www. manar-al-athar.ox.ac.uk

therefore, be seen as rushed construction in moments of chaos and panic; they are perfect examples of systematic planning by local authorities who managed their available resources in large projects. This is a pattern which is clearly seen (and has become paradigmatic in) Gaul and Italy (Christie 2006; Loseby 2006), but in Spanish towns it is also noticeable. Scholars such as Ángel Morillo, Carmen Fernández Ochoa (especially Fernández and Morillo 2002) and Adriaan de Man (2007a) have worked extensively on urban fortifications, especially those which are late in their dating. The chronology of these fortifications is quite problematic, as in many cases, dates were given on stylistic or comparative motives, without a clear archaeological confirmation. Traditional interpretations dated most of the late Roman walls to the third century, as a response to the ‘Frankish’ invasions of the 260s (cf. Nolla and Nieto 1979) – a proposal now largely discredited. It is only recently that the overall date of these later fortification phases has been moved to the fourth and fifth centuries. And in the same way that previous

New townscapes in the l ate Empire

71

scholarship doubted whether late town walls were third or fourth century in date, the question now is if to date them in the fourth or the fifth centuries. There is a large body of urban fortifications, but understanding the chronology in which they were built is important, because the historical circumstances of the fourth and the fifth centuries are different, and the interpretations that can be drawn from the two chronologies would be different. Furthermore, as there have been no general archaeological excavations or stratigraphic analyses of walls to determine general trends, it is difficult to give accurate chronologies – although some recent excavations at Ávila, for instance, have provided with more definite information (Estremera et al. 2006). However, it is safe to assume that each town developed its own late defended perimeter according to its own local circumstances, some during the fourth and others during the fifth century (cf. Fernández and Morillo 2002: 578-9). Most early Roman towns were built on flat areas, and despite the Pax Romana, they were still surrounded by walls. There are only a handful of examples, such as Lucentum or Ilici (Martínez and Moreno 2015: 278), where this is not the case (or where archaeology cannot confirm that walls existed). It could have been assumed that these towns would have acquired city walls during the fourth or fifth centuries, but this does not seem to have been the case. The construction of walls must have been too expensive for those already defenseless towns, especially since they were already secondary or tertiary nuclei. The refortification of walled enclosures is perhaps the best instance of these defensive efforts. The two best-known cases are those of Mérida and Barcelona, but there are other, less thoroughly studied examples, such as Gerona. For Gerona we know that the wall enclosure was preserved as it had been in the early Roman period (encompassing 7ha), but reinforced, although the chronology is still debatable (Nolla Bufrau 1999). In Barcelona and Mérida the old Augustan walls were enlarged, not in extension, but in thickness and height, adding or improving the old towers and making the gates more defendable, both by means of blocking minor or secondary gates and by refortifying the main ones. In the case of Barcelona, but especially for Mérida, the construction of these reinforced walls has to be seen as an indicator of the wealth and importance of these towns. The walls of Barcelona, set up in the Augustan period, surrounded the Roman town in an almond shape, covering a medium-sized area of 10ha. These were later expanded with new towers, doubling the thickness and the height, and reinforcing the gates, as can still be seen at the gate by the Plaza Nueva (Pallarés Salvador 1969; Puig and Rodà 2007: 628). The date first given to this was a fourth century one, due to the fact that it is reusing third-century

72 

The Iberian Peninsul a bet ween 300 and 850

inscriptions (mostly funerary) and other unidentified architectonic elements quarried from the city. Nevertheless, the presence of several fourth century coins in the filling of one of the wall towers has since allowed its excavator to date such works to the fifth century (Járrega Domínguez 1991). This reinforced wall was probably surrounded by a ditch, which has not been excavated, but which can be inferred from the mention of a ditch in the Vita Hludowici (13). This does not mean that the ditch encountered by the Franks in 801 was the same one, but it can be seen as a parallel of a new ditch dug (or re-dug) in the face of danger – as the fortification of the walls indicates. In Mérida we have a better-studied example. Its Augustan walls were improved in the early fifth century as a result of its political importance during the time of the invasions. The walls, originally some 3m thick, were doubled in thickness, although its height may have remained the same, at between 8 and 10m (Alba Calzado 1998). This was achieved through the addition of a new core of mortared rubble faced in reused granite blocks (figure 2.2), obtained from various public buildings of the city, such as the temple for the imperial cult (the ‘Temple of Diana’), the theatre, the amphitheatre, perhaps one of the aqueducts, and also many early imperial grave stones taken from the necropoleis surrounding the city. This large new granite facing of the wall included the blocking of some of the minor streets, which gave access to the secondary roads, and in one case (behind the amphitheatre), an old gate was turned into a turret. The systematic dismantling of the granite facing of the theatre and the removal of the granite seats of the amphitheatre are perfect examples of planned and directed quarrying; these buildings were deprived of purely ornamental elements (facings and seats) without structural purpose (the core of the buildings is opus caementicium). This means that even if the buildings themselves may have been quarried and defaced, they could have still been used. It is possible that the wall enclosures of the other main provincial capitals, Tarragona and Córdoba, were reinforced in this period. We know that the early Roman walled precincts were preserved throughout Late Antiquity, both in Córdoba and Tarragona, but despite the lack of positive evidence to prove late Roman repairs (cf. Molina and Valdivieso 2007: 44; Macias Solé 2013: 130) this does not mean that such improvements were not carried out. Both towns have been constantly inhabited, and their walled enclosures were reused in later Medieval periods, but none have undergone a thorough archaeological or architectonic study to reveal these phases (and it will not be surprising if they do appear). Despite there being well-studied cases, this type of intervention, where the original wall perimeter was preserved, was rare. In most towns there

New townscapes in the l ate Empire

73

Figure 2.2 Reinforcement of the walls of Mérida, built during the fifth century with ashlar blocks quarried from public buildings. This late Roman phase of the fortifications was done while preserving the original walls

seems to have been a real need to reduce the fortified enclosure in order to make it truly defendable. This not only demonstrates the true purpose of the original constructions, but also indicates a lack of manpower in order to man the walls. It would take up too much space here to list all the reduced walled enclosures of late Roman towns, as there are many, but it is still possible to see some emerging patterns. First of all, we see the emergence of reduced and improved fortified enclosures in those sites with direct links to the provincial administration or the army, as in Cartagena, Lugo, Veleia, León, Braga and Pollentia. In all these examples areas of the early Roman town were left outside the new walls, which were more robust and more easily defendable. In the case of Pollentia, this reduced enclosure used the back of earlier buildings (in this case, forum temples) within the walls.16

16 A pattern known for other cities in the late Empire, such as Nîmes and Arles, and perhaps in Termes.

74 

The Iberian Peninsul a bet ween 300 and 850

In the towns of the Northwest, the development of reduced, late Roman walled enclosures left relatively small areas of the town outside the new defences. The wall enclosure of early Roman Braga is not known, and across the literature it is assumed that the walls were built when the town’s status was elevated to being a provincial capital, between the late third and early fourth centuries (Lemos et al. 1998). It would be surprising to find an Augustan foundation without city walls, but they have yet to be found. Excavations have shown that the late Roman wall was built on top of earlier Roman structures, from which it is possible to presume that the urban area was reduced in the late Roman period (Ribeiro 2008: 273-81). The date for these fortifications is not certain, and the excavated remains of the walls show reconstructions and later phases of repair, perhaps dated to the post-Roman period (De Man 2011; Lemos et al. 1998). The walls of Braga are very similar in design and technique to other northern town-walls, such as those of Lugo, Astorga, León and Gijón, and can perhaps be dated to a similar date. Amongst them, the best excavated and analysed are the walls of Gijón, a double face-wall made on ashlar stones and opus caemeticium with circular towers every 18m, which have been dated to the beginning of the fourth century (García et al. 1992). The most impressive and best preserved walls are those from Lugo, a double-faced wall with 84 two-story circular towers that enclosed a perimeter of 28ha, including open areas, without buildings (figure 2.3). They have been dated to the beginning of the fourth century by the reuse of funerary inscriptions of the previous century in their core. Recent archeological works carried out close to one of the main gates, the ‘Puerta de Santiago’, seem to confirm that chronology and have shown how, in this area, the town walls were built over a third century wealthy house, the domus of the Mithraeum, whose peristyle was destroyed by the construction of the walls, reusing one of the dining-rooms of the house as a bread oven for the workers (Fernández and Rodríguez 2012; Rodríguez and Rodà 2007). In Astorga, León, and Conimbriga, the fortifications followed these same patterns (Fernández Ochoa 2006; Rodríguez and Rodà 2007; De Man 2011). Some other main towns in the peninsula, such as Zaragoza and probably Seville, also left only marginal areas outside reduced walled enclosures (Escudero et al. 2007: 43-6). However, in other cases, the reduction of the fortified space was far more dramatic. Such was the case of Cartagena. Cartagena had been promoted to provincial capital of the Carthaginensis. The wall was redefined to include the highest ground of the small peninsula on which the town is built, around the hill of El Molinete and the Castillo de la Concepción, leaving a substantial area of the old city outside. The area outside the new perimeter appears to have been

New townscapes in the l ate Empire

75

Figure 2.3 Walls of Lugo, an example of fourth century refortification in a secondranking Iberian town, in which parts of the early Roman town had to be left outside the new enclosure

largely depopulated, even if the new wall encompassed most of the public spaces, located towards the west and close to the harbour. The 40ha of the original enclosure was reduced to 15, most probably during the fifth century, although there is no direct evidence to confirm this, and the date is extrapolated from the presence of sixth-century burials in the eastern part of the old city (Ramallo et al. 2010: esp. p. 236). The example of Italica is also very clear. By the late Roman period, when the wall enclosure was reduced, much of the Hadrianic city expansion (which included most of the luxury houses, various sets of baths, and the amphitheatre) were left outside (from a total area of 60ha to an enclosure of only 20). The new wall has not been excavated (it has only been surveyed), but its course is clear, and it certainly seems to enclose only the original early Roman settlement and a small proportion of the second-century expansion (Sánchez Ramos 2010: 250). Such was the importance of wall renovations in this time, that in a number of small towns renovations are the only piece of evidence for the existence of urban life for the late Roman period. For example, Termes, where walls were built around the end of the third or the beginning of the fourth century. Despite this, Termes was abandoned at some point in the fourth century, the dating for this given by the last pottery used in the domus of the aqueduct (Argente and Díaz 1984).

76 

The Iberian Peninsul a bet ween 300 and 850

Considering then the importance of walls in this period, the extent of the works necessary to carry their renovations out, and the efforts made to make sure wall enclosures were manageable and functional, it should not be surprising that there were few efforts by city councils for investing in any other types of buildings.

The old Roman public buildings The idea of a Roman city as a centre for the public administration was still valid in the fourth century. What was clearly different then was the material scale of such public care. There is far less evidence for new public constructions in this period. A common explanation for such a decrease in public investment is that each town already had a set of public buildings, and so it was not necessary to build new fora, or new basilicas or theatres; the existing buildings simply needed maintenance and repair. The repertoire of public buildings inherited from the early Roman town included administrative buildings (like fora and basilicas) as well as leisure buildings (baths, theatres, etc.). They all, without exception, went through a period of disrepair and partial abandonment, except for a few noted examples (despite, in many cases, the existing imperial legislation: Codex Theodosianus XVI.10.15). Furthermore, the new administrative circumstances of the late Empire did not encourage private donations (euergetism), because there were hardly any political gains to be made by such tactics. Baths The only type of building which saw more constructive activity were bath complexes (Ward-Perkins 1984: 128), as they were ‘the most popular area of what remained of secular public buildings’ (Mango 2000: 934). This was partly because they needed not only constant work for repairs, maintenance and adaptation to new bathing patterns (smaller or individual hot tubs, which were more economic than large heated pools; Martínez Jiménez 2014: 30), but also because they had become one of the few places where local elites could still display their status as patrons (cf. Stirling 2012). Despite this, bath repairs were still quite limited. Similarly, the archaeological record shows the construction of some new public buildings in the fourth century, although this is limited to the main cities, such as Mérida, Cartagena, Córdoba, Braga and Barcelona. What is far more common, especially in secondary nuclei, is the slow but constant decline and dismantling of the old bathing areas.

New townscapes in the l ate Empire

77

Figure 2.4 Plan of Cartagena in Late Antiquity (fourth to seventh centuries), indicating the location of the early Roman and reduced, late Roman walled enclosures. Note the large area close to the forum which was left outside the new fortified area

Martínez and Moreno 2015, figure 4; re-drawn from Ramallo et al. 2010: figure 14

With regards to late baths, they are quite a common find even in sites where aqueducts had ceased to function, as new alternative ways of supplying water to baths were sought in order to preserve the bathing habit. In Cartagena, despite the reduction of the wall enclosure (figure 2.4), a new set of baths was built outside (which were in use into the fifth century), in the forum area by the harbour (Ramallo and Vizcaino 2007), together with a new public building (perhaps a macellum). In Barcelona, new public baths were built also near to the ancient forum although they stopped working in the following century (García-Entero 2005: 213), and in Tarragona the baths of the Calle San Miguel were kept in use throughout the late Roman period (Macias Solé 2004b: esp. 58-62, 83-4). The baths of Lisbon were also renovated in the fourth century according to an inscription (CIL II 191) that mentioned his patron, Numerius Albanus, praeses of Lusitania, which shows a hint of survival of bath euergetism (Mascarenhas et al. 2012). In Zaragoza, the

78 

The Iberian Peninsul a bet ween 300 and 850

two known public baths were equally preserved throughout this period (García-Entero 2005: 281), and a similar date for their abandonment has been given to those of Pamplona (Unzu et al. 2006: 432-5) and, perhaps, to the public baths under the Islamic fortress of Mérida (Alba and Mateos 2008: 268). Civic and religious buildings With regards to other main public civic buildings, especially those linked to the administration, such as fora, basilicas and temples, the degree of continuity is even less. The case that they had become redundant under the new Late Antique administration has already been put forward in a previous chapter, but besides that, it is clear that there were no great efforts to carry out any major works on these structures. It would rather seem that town councils were ready to let them fall into (orderly) disrepair. The very few cases of positive action on fora, basilicas and other civic buildings are seen in top-tier administrative cities, such as Mérida, Tarragona and Braga. In the case of Tarragona, the so-called provincial forum was partially privatized by dwellings, but the rest was still working and accessible as a public area until the second half of the fifth century (Aquilué et al. 1991; Bosch et al. 2005: 168, 171). Tarragona also preserves the latest example of imperial dedications in Hispania, an inscription mentioning the emperors Anthemius and Leo (467-472), which it has been suggested indicates the presence in the city, the last Hispanic capital under the imperial control, of the magister militum (Macias Solé 2013; CIL II4 109). The inscription has been linked to the dedication of a new fountain (Martínez Jiménez 2014). In Braga, the designated capital for the new province of Gallaecia, the forum improvements included a new fourth/fifth century pavement of opus signinum that replaced the previous pavement belonging to one of the largest Roman buildings found in the town close to the cathedral. During the fourth century this was paralleled by the construction of new porticated streets (Fontes et al. 2010). Additionally, at Valencia, which even if it was a smaller and secondary city, was linked to wider Mediterranean trading networks and had a powerful aristocracy, the forum seems to have been fully refurbished during the fourth century (forum as an open public space). Although its basilica (the remains of an obsolete urban administrative system) was turned into a food shop, where traces of hearths, fish bones, grape seeds and other fruit stones have been identified, together with a large press and several amphorae (Álvarez et al. 2005: 253-7).

New townscapes in the l ate Empire

79

This is, however, the total sum of constructions and maintenance of fora that can be seen in late Roman Spanish towns, as in most cases they were simply left abandoned. In prominent, top-rank cities, fora were kept as open spaces, indicating that they still fulfilled a social function within the urban space, but there does not seem to have been any new investment beyond some statues and inscriptions devoted by the highest administrative elite to tetrarchic and Constantinian emperors. Two possible examples from Mérida are the triumphal arch devoted to Maximilianus Herculeus (Arce Martínez 2002) and the colossus of Constantine by the theatre (Sastre de Diego 2015). The lack of building investment in fora is clear in main towns and cities across the peninsula, especially during the fourth and early fifth centuries, such as in Córdoba (Hidalgo Prieto 2005: 403), Segobriga (Abascal et al. 2008: 221), Braga (Fontes et al. 2010), Mérida (Mateos and Alba 2000: 148), and Barcelona (Beltrán de Heredia 2002a: 102). This could be because they lost most of their political roles. However, it can also be linked to the end of the habits of inscription or statue dedication across the West, after the Constantinian dynasty, which was especially evident in the Iberian peninsula (Witschel 2016) where there was no main imperial administrative centre (such as Trier, Arles, or Carthage). In most secondary towns, fora and basilicas were actively dismantled and already encroached upon by private buildings by the end of the fourth century, highlighting the decline of the secondary urban network (even in the highly urbanized regions of Baetica: Sánchez Ramos 2010). The site of Valeria is an extreme example of this, as it is clear there how the municipal authorities carefully dismantled and sold the building materials of the forum and the curia before houses were built on top (Fuentes Domínguez 2006: 205-6). This process even spread to primary cities, especially during the second half of the fifth century, as happened to the colonial fora of Mérida and Tarragona (Macias Solé 2013). Another element which was directly linked to early Roman public monumentality were temples. Both as part of Christianization and of the decline of traditional religious magistracies, old pagan temples ceased to be maintained throughout the fourth century, and by the end of the century they were even dismantled and quarried down, especially after the Theodosian law ordering the closing of the temples. There is no evidence for an active Christian destruction of any temples in Hispanic towns while they were in use. Rather they seem to have fallen into disuse, dismantling, and abandonment (Arce Martínez 2006a; López and Martínez 2006). Only after a period of abandonment do we see some of these buildings substituted by churches.

80 

The Iberian Peninsul a bet ween 300 and 850

Spectacula Regarding entertainment buildings (theatres, amphitheatres and circuses), the general pattern is again one of slow abandonment and eventual quarrying rather than violent episodes of destruction. This happened during the late fourth century, although it mostly occurred in the fifth century. Many of these are well-documented in the archaeology. Córdoba is perhaps the best example of this: the theatre was already abandoned during the second half of the fourth century, as one of the peristyle walls and the façade had collapsed and they were never repaired. Furthermore, the whole complex seems to have been turned into an intramural dump area for nearby workshops (Hidalgo Prieto 2005: 402-4). A similar process can be seen in most theatres in the Iberian peninsula although there are some rare examples of late continuity, where the use for which they were intended is unclear. In Zaragoza, for instance, the theatre’s orchestra seems to have already been quarried away by the fourth century, but this was then repaired and the theatre was again in use in the fifth century (Escudero and Galve 2007: 57). When it comes to amphitheatres, the situation is very similar: of the seventeen known amphitheatres in the peninsula, only six seem to last into the Late Antique period, most notably the amphitheatre of Tarragona (Jiménez Sánchez 2009-2010: figure 1), repaired during the reign of Constantine (Alfőldy 1975: 56-7). Overall, all but the amphitheatre of Mérida were abandoned by the end of the fourth century. In Mérida, the amphitheatre and the theatre were renovated under the Constantinian dynasty, but they were finally abandoned at the beginning of the fifth century (Ceballos and Ceballos 2003: 59; Sastre de Diego 2015). Regarding circuses, however, there seems to be a longer archaeological continuity for their existence as functional structures, especially considering how highly valued Spanish horses were in the late Empire.17 The circuses of Mérida, Toledo (Sánchez-Palencia and Sainz 2002) and Tarragona (Macias Solé 2008: 298) seem to have survived intact until the fifth century, when it seems that they began to be quarried (figure 2.5). The circus of Mérida had been, furthermore, repaired in 337-340 by the comes Hispaniarum, Tiberius Flacius Letus, and the praeses of the province Julius Saturninus, as commemorated in an inscription (AE 1927: 165), which mentions new columns, new decorations, and the possibility of filling it

17 Symmachus seems to have been very keen to provide these for the circus races; cf. Epist. 225, 226, 227, 232 and 234 (numbering according to Beltrán Rizo 2013).

New townscapes in the l ate Empire

81

Figure 2.5 Photograph of the excavated remains of the circus at Tarragona

with water.18 Furthermore, a funerary inscription at the nearby site of Casa Herrera dated to c. 500 mentions an auriga, a charioteer, a late indicator perhaps of continuing circus practices (Caballero and Ulbert 1976; Arce Martínez 2002).19

18 Floren[tissimo ac b]eatissimo s[ae]culo favente / feli[ci]tate [et clementia] dominorum Imperatorumque / nostror[[um Flav(ii) Cl(audii) Constantini]] maximi victoris / et Flav(ii) Iul(ii) Constanti et Flav(ii) Iul(ii) [[Constan]]tis victorum fortissi/morumque semper Augustorum circum vetustate conlapsum / Tiberius Flav(ius) Laetus v(ir) c(larissimus) comes columnis erigi novis ornamen/ torum fabricis cingi aquis inundari disposuit adque(!) / ita insistente v(iro) p(erfectissimo) Iulio Saturnino p(raeside) p(rovinciae) L(usitaniae) ita conpetenter / restituta eius facies(!) sp[l]endidissimae coloniae Emeriten/sium quam maximam tribuit voluptatem = In the most flourishing and happiest time, with the happy favour and clemency of our lords and emperors most victorious Flavius Claudius Constantine [II] and Flavius Julius Constantius, and Flavius Julius Constans the strongest and victorious, always augustus, count Tiberius Flavius Laetus, vir clarissimus, ordered the circus, collapsed by old age, to be reconstructed with new columns, surrounded by ornamental decorations, and flooded with water; thus, as insisted by the governor (praeses provinciae) of Lusitania Julius Saturninus, vir perfectissimus, was its aspect suitably reconstructed giving to the Colony of the Emeritenses the maximum happiness. 19 Unlike Gaul, Spain lacks references to later, Early Medieval circus events (cf. DLH V.17).

82 

The Iberian Peninsul a bet ween 300 and 850

This new attitude towards the spectacula seems to be a response to cultural (and religious) changes across the Empire, although it is difficult to pin-point the cause for this new attitude to public entertainment. It should be noted that throughout the late Empire the circus possessed a much more symbolic significance (as is famously known for Constantinople). This symbolic value may be the reason behind the preservation of circuses in the early centuries of Late Antiquity. This may not have been for horse racing, but perhaps for other, ceremonial, purposes. On the one hand there seems to be a move in taste away from plays, concerts and poetry (which probably continued in private environments, and not in public displays), in favor of animal hunts (rather than gladiatorial combats, which had probably already disappeared by the end of the third century: cf. Jiménez Sánchez 2009-2010) and, especially, horse and chariot racing. A very obscure reference in the Chronica Caesaraugustana mentions a circus race in the year 504.20 There is no known circus for the city, so this may indicate that there was some sort of racing event at a late date, even if the city lacked the relevant infrastructure (Jiménez Sánchez 2006). This would be comparable to what we know (archaeologically and historically) for Arles, where the circus was in use for horse racing as late as the sixth century (Heijmans 2004: 360-5; Loseby 1996: 53), and it also reflects traditions seen in the sixth-century charioteer mosaic from Gafsa (now in the Bardo museum). In the East this was also correlated with a decline in athletic competitions. In parallel with this, there was a general abandonment and disrepair of theatres and odea, as circuses (and to a lesser extent, amphitheatres) became more multifunctional buildings (cf. Saradi 2006: 295-7), a cheaper way of preserving public entertainment.

The fate of urban infrastructure Directly linked to public and civic buildings we find urban infrastructure, which suffered a similar development of a lack of repair and orderly dismantling. The overall slow decline of road and aqueduct maintenance can again be explained through the lack of interest of traditional city donors, but also as a change in the priorities of city councils. For instance: the alteration of street grids and the encroachment of public areas cannot really be interpreted as unorganized squatting but was probably being allowed (and charged for) by the local curiae on minor or secondary streets. 20 504; His co(n)s(ulibu)s Caesaraugustae circus spectatus est.

New townscapes in the l ate Empire

83

And again, as with other public buildings, the evolution of infrastructures throughout the late Roman period will underline the differences between first and second ranking towns, as two different trajectories will be seen. Aqueducts and sewers Regarding aqueducts, for instance, there is a degree of continuity in their use in major towns (probably related to the preservation of basic maintenance duties by town councils – Martínez Jiménez 2014: 174-6). It should be kept in mind that, when originally built in the early Empire, aqueducts where not urban necessities: Roman towns did not need aqueduct water in order to exist, as there were plenty of traditional alternative sources (wells, cisterns, water springs). They were part of the building program of Roman cities, just like theatres or amphitheatres, and especially so because they were connected to baths and bathing, which required large quantities of water, and were clear indicators of Roman urban living. Aqueducts can be seen, therefore, as indicators of Romanitas rather than urban needs.21 That is why they became quite popular: compare the 66 Roman aqueducts of Spain (Sánchez and Martínez 2016) with the mere seventeen amphitheatres listed above. Unsurprisingly the main capitals (Mérida, Tarragona, Córdoba, Braga) and other wealthy cities (Barcelona, Valencia, Seville, Lisbon, Zaragoza, Almuñécar) kept their aqueduct supply into the fifth century (Sánchez and Martínez 2016). Only a few cities had lost their aqueducts before the fourth (most remarkably, Cartagena and Segobriga: Egea Vivancos 2002; Almagro Basch 1978). Still, it is in the fourth century when there was a steep decline in the number of functioning aqueducts. In most cases the end of the aqueduct is dated to the actual end of the urban nucleus (a process which we will discuss in detail in the following section), but there is also a significant number of cases in which the town continued existing despite the end of the aqueduct. In the latter cases, aqueduct abandonment is archaeologically dated together with the end of the main water-consuming structures, such as baths or fish industries (cetariae), which are overall indicators of the economic prosperity of the city. In Baelo Claudia the link between urban aqueduct supply and the end of the fish cetariae in the fourth century is 21 Aqueducts became with time essential for urban living, as they enabled the development of a denser and more hygienic urban fabric, but in very few cases they were a necessity at the the time of the foundation. The end of aqueduct supplies after several centuries of constant use was the cause for major changes in the urbanism of those cities (Martínez Jiménez 2014).

84 

The Iberian Peninsul a bet ween 300 and 850

Figure 2.6 Photograph of the collapsed remains of one of the aqueducts of Baelo Claudia, damaged by an earthquake during the fourth century and never repaired

Photograph by Dr. Elena Sánchez, used with permission

reinforced by the fact that at least one of the aqueducts collapsed due to an earthquake (figure 2.6; Jiménez Martín 1974: 293). In this situation, with

New townscapes in the l ate Empire

85

the end of the economic hub of the city, it must have been difficult (if not impossible or, simply, unnecessary) to bring the aqueduct back into use. Lastly, it should be mentioned that aqueducts do seem to continue in those towns which, as mentioned above, appear to have preserved their baths. Directly related to aqueducts is the fate of sewers and drainage systems. As with aqueducts, and even though Roman sewers in Spain and Portugal have been the subject of various recent studies (Acero and Remolà 2011; cf. Dupré and Remolà 2000), the abandonment dates of drains are not usually clear, but overall they seem to be abandoned during the f ifth century. Despite this, in the main towns of the late Roman period, where there is a larger body of well-dated material evidence, there seems to be a slight continuity of sewer systems. These were slowly substituted by open-air gullies and drains, together with cess and rubbish pits. The end of the municipal sewer networks has to do in part with the end of aqueduct supply, as in many instances the aqueduct overflow (the aqua caduca) was meant to regularly flush the sewers clean. As aqueducts failed to supply water constantly, more and more rubbish accumulated in sewers, which could not always be unblocked or cleaned by rain water. This would eventually lead to a final clogging. During the fourth century, some towns still had the resources to re-design or enlarge their sewer network, as in the case of Barcelona (Beltrán de Heredia and Carreras 2011), Valencia (Ribera Lacomba 2005: 210-2), Cartagena (Egea Vivancos 2002), or Lugo (González and Carreño 2007: 259-60), but these are rare instances. In most places urban decline was aggravated by the demise and collapse of the sewer system. Surprisingly, this seems to have been the case of Córdoba, Seville and Mérida, although in these cases this only applied to the secondary branches; the main drains were in use and apparently maintained into the Islamic period (Marfil Ruiz 2000b: 119; Cf. Acero and Remolà 2011). Other public water services, such as fountains, seem to have had a period of brief splendor during the fourth century. Large nymphaea (monumental fountains) were rebuilt or redecorated during the fourth century, as happened in Valencia (figure 2.7; Ribera Lacomba 2005: 210-2) or Córdoba (Murillo et al. 2002: 265), perhaps along the lines of the rest of the Empire, where nymphaea were one of the few things local patrons still invested in (cf. Richard 2013: esp. 237-50). This period of bloom was short, because as aqueducts slowly ceased to work, so did nymphaea and other smaller, more functional, public fountains. In many instances, these were soon dismantled and their precious lead recycled, as is known to have happened in Mérida and Córdoba (Alba Calzado 2007: 163; Ventura Villanueva 1996: 95-8). As

86 

The Iberian Peninsul a bet ween 300 and 850

Figure 2.7 Late Roman phase of the forum nymphaeum or monumental fountain of Valencia, at the site of l’Almoina. The fountain was restored in the fourth century and kept in used, together with the aqueduct into the fifth, and probably the sixth century as well, despite the post-Roman modifications to the fountains’ decoration

From Albiach et al. 2000: figure 5

with many other urban elements, water supply came to rely more and more on domestic or private resources, such as wells and cisterns (Martínez Jiménez 2014). Street and road networks One of the most evident transformations of the public infrastructure is the change in the street grids, as the encroaching on and blocking of roads became more common. This process was Empire wide, and widely recognized as one of the main visual transformations of the late Roman city. In general terms, it can be seen that secondary roads and porticated areas were privatized and enclosed into houses, in a process which can be seen occurring as early as the first half of the fourth century. The addition of these spaces to large houses in Écija or Mérida (Alba Calzado 2005: 139), or even the partial blocking of smaller alleys and secondary streets by private constructions, has to be directly linked to a new organization of the urban

New townscapes in the l ate Empire

87

space. These transformations, especially at this early stage, cannot have taken place without the consent (and the adequate fine or fee) of the town council. This process of encroachment breaks directly with the ideal of the early Roman town and its orderly street grid. Even if this can be classified as an indicator of ‘decline’ from the early Roman standard, the encroachment on to the streets indicates the existence of a lively and active community, looking for new spaces in the building-saturated urban fabric. Directly related to the end of sewers is the abandonment of cobbled streets. Throughout the late Roman period it is evident that the accumulation of sediment on the streets led to the development of beaten earth surfaces, which seem to have been preferred to cobbled roads (especially for un-shoed horses). Besides this process, the general trend was to ‘invert’ the design of streets, so the central part of the road became a gulley, the sides of the road becoming higher, and forming a concave surface (which is the direct opposite to early Roman roads, where the rubbish was washed to the sides as the street humped in the centre).

New late Roman monuments? Despite this overall sense of dismantling and abandonment, where public investment seems to have been diverted to walls, the late Roman period saw the development of new forms of urban monumentality, although these were quite limited. It was only later, in the sixth century, when a brand new set of urban monuments were properly developed. Until then, the old urban administrative and political systems were still in place, and did not require the introduction of a new architecture of power; it was possible to evolve organically within the earlier Roman infrastructure. Because of this, the main efforts of late Roman munificence were focused on the restoration and renovation of the already existing public buildings, such as the theatre and the circus in Mérida, always under imperial supervision. That meant an increase of public investments, especially under the Tetrarchy and the Constantinian dynasty, which was proportionally related to the decrease of private donations. The two main types of new late Roman monuments are praetoria (or civil administrative palaces) and churches, although the latter were not developed in Hispania at least until the fifth century. Praetoria or other types of ‘administrative palaces’ developed in the western capitals (Cologne, Trier, Arles, Ravenna, Milan, etc.), where imperial intervention was evident. As a result, these are largely absent from the

88 

The Iberian Peninsul a bet ween 300 and 850

Iberian peninsula, as there were no seats of imperial power there. There are some hints that may point towards the existence of such new, late Roman administrative complexes in a few chosen towns in Spain. The obvious and early case would be the large villa/palace complex of Cercadilla, in Córdoba (figure 2.8), built in the late third century on the outskirts of the city, and equipped with all sort of amenities, including a reception room and an aqueduct (Hidalgo Prieto 2002; Ventura and Pizarro 2010: 196-9). Little is known about the functioning of this compound, as it was poorly excavated and later destroyed during the construction of the high-speed train tracks, but its size links it to direct imperial intervention (Esmonde Cleary 2013: 213-4). The overall plan of the complex makes it difficult to understand the site as a functional building, but it is possible that it served as an administrative centre during the late Roman period. It is generally assumed that in Mérida there must have been a palatine complex, as it was the seat of the vicarius of the Dioecesis Hispaniarum, although this assumption is yet to be proven. It is possible that the governor had a large residence, and it has been suggested that the domus of the Mithraeum was this residence, if the cosmogonic mosaic can indeed be linked to an eastern design and workmen brought by Constantinopolitan officials (as suggested by Arce Martínez 2002).22 A similar case has been put forward for the peri-urban villa of Centcelles, where a reassessment of the chronology, the archaeology, the sources, and the iconography of the mosaics have led researchers from the National Archaeology Museum in Tarragona (Remolà and Pérez 2013; cf. Arce Martínez 2006b) to suggest that this site and its surrounding complex were the military camp and praetorium for a high ranking officer, comes Asterius, in the early decades of the fifth century.23 The construction of new Christian buildings in the late Roman period has to be approached with caution; firstly because Christianity was mostly a private cult throughout most of the fourth century, even after it was legalized by Constantine (313), and secondly because the first monumental and independent buildings were not built inside towns. Christian monuments began as shrines dedicated to a martyrial cult, especially at the place of rest of the martyrs’ relics.24 As a result, these were always on the outskirts, in funerary areas. This is, again, something that can be seen throughout the 22 This is based on the nature and quality of the cosmogonic mosaic preserved at the oecus, but its dating is problematic – see below. 23 As opposed to an episcopal estate or a mausoleum. The debate is still ongoing (cf. Arbeiter 2004; Arce Martínez 2006b). 24 Which would lead to an important search, hunt, trade (and falsification) of relics. Cf. Castillo Maldonado 1999.

New townscapes in the l ate Empire

89

Figure 2.8 Reconstructed plan of the ‘palatine’ complex at Cercadilla, in Córdoba

From Hidalgo et al. 2008: figure 391

Empire, and is not characteristic of Hispania or the West. In the Iberian peninsula, the first and main suburban centres of the Christian cult were those of Saint Eulalia in Mérida (Mateos Cruz 1999), Saint Fructuosus in Tarragona (López Vilar 2006), Saints Justus and Pastor in Complutum (Rascón and Sánchez 2005) and Saint Acisclus in Córdoba (Sánchez Ramos 2007). These new Christian monuments were funded by private munificence. This was not really euergetism in the classical sense, but simply private donations by local members of the elites, or, in some cases, by the newly emerging leaders: bishops. These first martyrial shrines were little more than enlarged mausolea, which developed an apse for the altar, but were not, at this stage, true basilicas. Proper basilicas would eventually develop in the fifth century. Martyrial shrines were soon taken over by episcopal authority, and it is at these places where the first episcopal complexes developed. At this very early stage, episcopal complexes were quite small, especially if compared to their sixth century counterparts (cf. Miller 2000), which partly explains why so few are really known. In the late Roman period they compromised usually just a residence, a basilica, and a baptistery. All of these are elements that can be seen at the Christian complex by the Francolí river in Tarragona (figure 2.9; López Vilar 2006). Similarly, in Mérida, the suburban shrine and basilica of Saint Eulalia could have been under episcopal control. Its

90 

The Iberian Peninsul a bet ween 300 and 850

Figure 2.9 Plan of the late Roman Francolí Christian complex, in Tarragona, with the funerary basilicas and the early episcopal complex

Based on Macias Sole 2008: figure 2

earliest episcopal complex, devoted to Holy Jerusalem, is still archaeologically unknown and its original location under discussion, but it had to be contemporary with the great suburban development around the basilica of Saint Eulalia during the second half of the fifth century. One century later, the episcopal complex was already located inside the city walls, as we will see in Chapter 5 (Cruz Villalón 1985; Sastre de Diego 2015). In some

New townscapes in the l ate Empire

91

other examples, as seems to have been the case in Barcelona, the episcopal complex was built originally inside the city walls, occupying a couple of street blocks in the northeast corner. In this case, the Christian centre did not grow around a martyrial shrine, but probably developed out of an urban domus turned into the private residence of the bishop, on to which a small church was added (Beltrán de Heredia 2002a). The construction of these new public elements, together with the alterations, repairs and abandonment of other public structures mentioned above, give an overall view of the changes in the monumentality of cities in the late Roman period. But beyond the monuments, the urban fabric and the urban layout was also changing in this period.

Suburbanization and de-urbanization The alteration of the spatial organization in towns is one of the main characteristics of the Late Antique urban transformations. The concentrated, clustered, dense, and walled town of the early Roman period was slowly substituted by a wider, less dense, more spread out, and even poli-nucleated settlement. In most towns, these transformations have led archaeologists to propose a decline in the overall number of town inhabitants, which had to reduce the spread of the urbanized area. However, this explanation is perhaps too simplistic. Reduction of towns The most evident transformation of the urban layout is the reduction of the inhabited area inside the urban centre. This has already been mentioned earlier, when discussing the smaller wall enclosures, but in many cases dwelling areas were reduced to smaller spaces. This process usually went in parallel with the encroachment on to central public areas, like fora. In this way, urban settlements continued to be nucleated, while useless and redundant unoccupied central areas were built up. In parallel, those areas which were not occupied by dwellings were turned into dump yards, small horticultural gardens, or occupied by burials. In secondary towns, especially in the Northern Meseta and the Cantabric coast this process started already in the late third century, and continued in such a fashion that during the fifth century most ceased to be recognizable, or to function as, towns in the late Roman sense. In these territories, urbanization was a phenomenon directly linked to the Roman imperial system, and it is possible that as the fiscal and administrative system dissolved during the

92 

The Iberian Peninsul a bet ween 300 and 850

fifth century, towns ceased to have a purpose (Martínez and Tejerizo 2015). The preservation of the toponyms in many cases suggests the continuity of either a minor settlement, or the perpetuation of the memory about the sites. It is not until the late sixth century, as we will discuss in Chapter 5, that these towns seem to be partially recovered through the founding of bishoprics (as it is with Segovia, Salmantica or Segobriga). Some other minor towns, but still connected to Mediterranean trading networks, managed to preserve some degree of urban continuity during the fifth century, although on a much reduced scale. This is the case of Baetulo, where the Northeast sector was inhabited until the sixth century, but the rest of the town was largely abandoned, with similar processes happening in Ilici, Lucentum (Martínez and Moreno 2015), or various towns in Baetica such as Baelo, Carteia (Sánchez Ramos 2010), or even Cádiz, the oldest town in the Iberian peninsula and one of the main seaports at the beginning of the Roman Empire, as briefly mentioned before. The decline of this last example was even evident to contemporary writers, like Avienus, who mentions Cádiz as a ‘field of ruins’ (ruinarum ager - Ora Mar. 270-271). The example of Ampurias is, perhaps, the better known archaeologically. By the fourth century, all its public buildings (forum, amphitheatre) had been abandoned, and the settlement seems to have been shifted to the nearby hill of San Martí, while the old coastal settlement became a burial ground. Bulk finds suggest that Ampurias’s role as a trading node for its broad hinterland was lost with the decline of trade during the fifth century, and it turned into a small agricultural settlement. Fragments of North African amphorae and ARS suggest a continuation of trading patterns, even if at a much reduced scale. In major towns, including provincial capitals, this process is also evident, even if they managed to preserve town life through the late Roman centuries. In Córdoba, Tarragona, Seville (Tabales Rodríguez 2001: 420-2) and Zaragoza, for instance, the intramural population concentrated in a smaller area within the town walls, normally around the forum, the centre of power and administration, which acted as a focus around which town life still functioned. The only exception to this in the whole of the Iberian peninsula seems to be Mérida, where there was even an expansion of the built up area during the fourth century. The rise of suburbs Despite the possible decline in population numbers, it is true that in many cases the population did not just vanish, but simply shifted to another location in the surroundings of the city during the fourth century. The

New townscapes in the l ate Empire

93

development of new suburbs (understanding suburbs as extra-mural clustered settlements) is clear in the archaeological record. In some cases, suburbs emerged as a result of the reduction of the walled space, but in most cases it was the result of a new attractive focus, either economic or religious. After the reduction of the walled enclosures of Pollentia, Lugo, Zaragoza and, probably, of Braga, parts of the towns were left outside, but were still inhabited. These clusters of housing areas had not evolved as vici or suburbs, but ended up becoming similar sites. In coastal towns, fourth-century harbour suburbs seem to have emerged as a response to declining intramural urban centres, as suggested by the patchy archaeology of these areas. More vibrant and attractive, harbours were natural economic hubs that drew population around them. In Tarragona the suburb grew next to the city and functioned as a secondary nucleus of population (figure 2.10), as suggested by the various late dwelling areas excavated there (Macias Solé 2008: 296). In Lucentum, the old urban centre was largely depopulated, and in fact, it would seem as if the population had shifted to the new port area, at the site of Benalúa (figure 2.11). This site offered a better beaching area (the old harbour had silted up), even if it was over 3km away, although it has only been identified by the presence of rubbish deposits and large finds of pottery – no housing has been found (Reynolds, Paul 1987; Rosser Limiñana 1990). Excavations at Traducta follow similar patterns (Navarro et al. 2000: 226-7). In Vigo, in the Atlantic coast of Gallaecia, the large concentration of imported amphorae and fine wares (mostly African, but also Eastern) suggest the development of a new harbour settlement, which slowly substituted for the old town (Fernández Fernández 2014). Lastly, in Barcelona a new ‘economic hub’ suburb developed in the late Roman period, although rather than around a commercial harbour it instead grew around a workshop area (Gurt and Godoy 2000: 458). The main type of suburb that developed in the late Roman period was, however, the Christian vicus; even if they were quite small during the late Roman period, and would only really develop later, in the post-Roman era. The attraction offered by the holy ground that surrounded a martyr’s shrine made these areas appealing to the Christian population, who would then be living within closer range of the relics, and closer to the rest of their religious community. Tarragona and Mérida had perhaps the most explicitly Christian suburbs as they related directly to the early basilica complexes of Saint Fructuosus and Saint Eulalia, as mentioned above, but they were not the only ones. In the case of Complutum, for instance, it seems that the suburb around the shrine of the holy boys Justus and Pastor (the

94 

The Iberian Peninsul a bet ween 300 and 850

Figure 2.10 Plan of Tarragona, indicating the location of the main settled areas: The upper town, the lower town, and the harbor suburb

Adapted from Macias et al. 2007: Phase V

‘campus laudabilis’) became as important and inhabited as the rest of the old, diminished, town (Rascón and Sánchez 2008: 248). It has been suggested that a similar suburb developed outside Ávila, although the evidence is quite thin (Barraca de Ramos 1999: 187-8). An example which is more difficult to assess is that of the new northwestern suburb of Córdoba (the eastern suburb had disappeared by late Antiquity – Ventura Villanueva 1996: 99), which grew around the palatine complex of Cercadilla, as the population shifted from the old southwestern vicus (abandoned during the fourth century). It could be that population was attracted to the focus of imperial power. The location of the shrine of Saint Acisclus is debated (cf. Hidalgo Prieto 2002 and Marfil Ruiz 2000a), but it is certain that it was located in this western area, so it could have been a ‘Christian hub’ type of suburb. A more mundane explanation could be that this area still had an aqueduct supply (in fact, two different functional

New townscapes in the l ate Empire

95

Figure 2.11 Plan of Alicante, with the location of the sites of Lucentum and the new, late Roman commercial hub at Benalúa

conduits; Sánchez and Martínez 2016: 187-92), as opposed to the upper town inside the walls, which was slowly being depopulated, as the settlement shifted to the lower city (next to the bridge and closer to the water table). Easy access to ground water once aqueduct supply failed could also be an extra factor behind the development of suburbs (Martínez Jiménez 2014: 185-6). For instance, in Tarragona the harbour suburb was built on an area where wells can be dug, as opposed to the largely abandoned lower town (where the aqueduct supply had failed) and the still occupied upper town (where aqueduct water still flowed). Something similar could be behind the residential suburb of Toledo, built in the river meadow rather than on the granitic dome of the upper town at a time when the aqueduct was probably out of use (cf. Olmo Enciso 2009: 71-2).

New domestic architecture Despite of the increase of urban archaeology in the Spanish cities, the knowledge of late Roman private architecture is still quite partial. However, the majority of the archaeological remains of Roman houses belong to this period. These late constructive phases are the best preserved because, in

96 

The Iberian Peninsul a bet ween 300 and 850

many cases, they correspond to the last moment of house activity. And this is precisely the first point that has to be considered: late Roman houses were the last step in the evolution of older buildings which had been reformed and renovated many times. In fact, very few Roman houses continued in use after the fifth century, the moment in which they seem to have been replaced by new domestic models. Once again, the three main capitals provide us with the best known examples for elite housing. In the exceptional case of Mérida, archaeology has shown how many houses were expanded (around 600m 2 each) and monumentalized at the beginning of the fourth century. New mosaics (like the lavish cosmogonic mosaic of the domus of the Mithraeum, if it can be truly dated to this chronology; figure 2.12),25 baths, living rooms (the oecus), gardens and other amenities for the aristocracy had turned the old provincial town into a luxurious urbs. This extensive renovation coincides with the designation of Mérida as the capital of the Dioecesis Hispaniarum and the arrival of a new administrative elite. Córdoba offers a few examples of late Roman domestic architecture beyond the so-called tetrarchic complex of Cercadilla. As in Mérida, there is no evidence of multi-family buildings in Córdoba; they are mostly aristocratic domus, from earlier periods (Carrillo 1999). In Tarragona, aristocratic domus had been enlarged in the second century, but by the fourth they went through a series of more structural changes. The collapse of the sewer system in the fourth century had an impact on the way houses worked, as cesspits became more common. By the fifth century, the end of the sewer and aqueduct system had rendered many old baths useless, so they were reused as dwelling spaces. Despite the limited evidence, it is clear from excavations in other towns that the development of an enlarged urban domus, with new mosaics and private baths, were the two key elements of the new aristocratic houses between the third and fifth centuries. In Seville a number of these houses were unearthed at the site of La Encarnación, including the domus of the Columns (figure 2.13; Beltrán et al. 2005), and even in secondary towns such as Astorga (the domus of the Bear and Bird Mosaic; Sevillano and Vidal 2002) Complutum (domus of Bacchus and domus of the Cupids; Rascón and Sánchez 2005) or Conimbriga, with the famous domus of Cantaber (Correia and Reis 2000). The most recent survey of domestic housing of late Roman Braga (Martins et al. 2016) shows these very same patterns happening in the capital city of Gallaecia.

25 This is an early Roman domus with later monumentalization. For the controversy about the dating, see Sastre de Diego 2015: 62-3.

New townscapes in the l ate Empire

97

Figure 2.12 The domus of the Mithraeum in Mérida, with its cosmogonic polychrome mosaic representing the gods of the Heavens, Earth, and the Underworld

Beyond the big domus there is little evidence for urban housing, especially that of the non-elite classes. Excavations in Mérida suggest that artisans lived in the same industrial complexes where they worked, as a fourth-century fullonica excavated close to the walls suggests (Palma 1997; Alba Calzado 2004b; Sastre de Diego 2015). Many of the late Roman tabernae excavated in other towns of Hispania, where only the foundations of a flight of steps have been preserved, may suggest that artisans and craftsmen lived on top of their workshops. Other than these, those poorer dwellings built encroaching upon public spaces, such as the late fourth century houses of the forum of Córdoba seem to have been simple, single-roomed structures, built in perishable materials, such as timber posts and mud brick or trampled earth (Vaquerizo Gil 2004), which leave behind few archaeological remains. Similarly, the transformation of public spaces into dwellings (by means of subdividing the space into various living rooms) is common in Iberian cities. Baths were easily subdivided into living quarters, but the vaulted spaces of circuses, amphitheatres or porticated areas in the fora were also common (Macias Solé 2004a).

98 

The Iberian Peninsul a bet ween 300 and 850

Figure 2.13 The ‘House of the columns’ (fourth to early sixth century), excavated at the Plaza de la Encarnación in Seville. An example of a late Roman aristocratic domus, linked to civic elites. The house has got a late Roman ‘oecus-type’ layout, is decorated with marbles, mosaics, and was directly linked to the functioning aqueduct by lead pipes

Image by Benjamín Núñez, Wikimedia Commons

By the mid-fifth century, Roman townscapes had changed dramatically from what they had been at the beginning of the fourth century. The late Roman administration, with its new approach to civic munificence and building maintenance, had set the basis for the new late Roman townscape, where the old classical monuments ceased to have a real purpose. They were not replaced, and were slowly being set aside and dismantled or encroached upon. But even still there were urban elites who promoted and preserved town life. The slow spread of Christianity began to alter the outskirts of the cities, but the old civic centres remained largely devoid of new Christian monuments. Those towns which had been relegated to secondary administrative nuclei, or that through the fourth and early fifth centuries had been cut off from the main trading networks, slowly began to decline and to dissolve. They remained as centres of population, but they were not urban in nature. Still, until the early decades of the fifth century a precarious equilibrium was maintained within towns. The Tetrarchic and Constantinian investment in towns and monuments came to an end with the Theodosian dynasty, but once the Alans, Sueves, Vandals and Goths entered the peninsula, after 409, the Roman economic

New townscapes in the l ate Empire

99

and political system began its decline until its final end in the last decades of the century. Large areas of the Atlantic coast and the centre of Hispania were dissolved into a power vacuum, where neither the barbarians nor the Romans exercised any effective power. This caused, all across the peninsula (at different paces in different regions) an acceleration of these processes of urban disintegration. As we will see in the next chapter, the Roman countryside transformed at the same pace as the towns, for they were still intimately connected. It will only be in the post-Roman period that town and countryside saw new, positive, constructive changes, which firmly established the trajectories set forth in the late Roman period, marking the peak of Late Antique transformation.

3

The economy and the rural world in the late Empire

The late Roman administrative system, the civitas, was based on the centrality of the city (urbs), around which a rural hinterland, the territorium was articulated. This territory could be quite broad, and included a number of minor settlements and productive centres, such as villas. However, just as towns changed, so too did the countryside, and the transformations of the late Roman centuries show an intermediate point between the Roman and post-Roman patterns. Despite this, it should be noted that there was not an abrupt rupture with earlier Roman traditions, but rather a slow adaptation to the new political circumstances. This rupture only occurred after the Roman system collapsed, during and after the fifth century. The exploitation of the landscape was the main economic activity in Antiquity, mostly through agriculture but, particularly in Hispania, also through mining. The profit extracted from the countryside was funnelled towards the urban centres, from which it was turned into cash (either through trade or taxation), and thus benefited the economy. In this way, and as it can be inferred from late Roman Hispanic examples, the territory was organized in to three different levels, defined by the way the land was exploited and its main purpose. In the first level we find suburban and peri-urban agricultural and other forms of production sites. In the previous chapter we have described how harbour suburbs developed around towns throughout the fourth and fifth centuries, but these are just a type of suburban economic centre. Suburban and peri-urban economic activities have to be understood as those production sites within the immediate periphery of an urban settlement; i.e., those that could be reached within a day’s travel, and that therefore can be seen as direct suppliers of basic goods for their urban populations. The second level includes those sites devoted to large-scale production aimed at trade. These were mostly the larger agrarian exploitations, normally linked to large villas, with large tracks of monoculture used for grain, vines or olives, but also fish processing factories and large pottery kilns. These sites tended to be, but were not exclusively, located further away from urban nuclei, but still with strong communication links through roads, harbours, and rivers. They had flourished during the early imperial period, as Hispanic productions were traded throughout the Empire together with the annona, the imperial good taxation and redistribution system.

102 

The Iberian Peninsul a bet ween 300 and 850

The third level is less well-defined, and includes various economic activities and archaeological sites. In it are included all the other sites and economic centres which were either not directly linked to trade (such as large imperial mining complexes) or those isolated communities which were cut off from the main trading networks. The latter cases, even if rare in the early imperial period, became much more common in the late Roman centuries, especially as the fiscal pressure increased and the network of urban settlements declined.

Blurring lines between the urbs and the rus A Late Antique phenomenon usually mentioned in the literature is that of the ‘ruralization’ of towns, which explains this redefinition of the productive and residential areas (essential to our proposed understanding of what a city was: Esmonde Cleary 2013: 100-1; Martínez and Tejerizo 2015: 84-5). The presence of productive sites within walled areas could be seen as a sign of this, although it would perhaps be better to discuss more permeable barriers between ‘town’ and ‘country’ (or rather ‘residential/administrative’ and ‘productive’) in the late Roman centuries, rather than diminishing the urban nature of sites that contained production centres. They certainly show a change in the nature of the early Roman city, but they are still defined by a centralized economic hub. Another element that has usually been associated with ruralization is the blurring of the division between dwelling and burial areas in the late Roman period. In this section we will look at both processes, which continued throughout the late and post-Roman periods, and would only be turned around with the renovation of urban economies in the Early Medieval period in al-Andalus (see Chapter 8). Production areas In the fourth century there was a dense network of small production sites surrounding most Roman towns of Hispania. This was necessary in order to supply town dwellers with local market goods (such as clothing, cooking wares, glass, metalwares, etc.) and foodstuffs (fruits, vegetables, meats, etc.), and other elements which are not usually linked to bulk trade (like wine, oil, wheat, fish sauces or table wares).26 In most towns, the early fourth 26 This does not mean that there were no large-scale productions of cooking and metal wares aimed at regional trade, but these local workshops are the basis on which post-Roman production would develop (see Chapter 7).

The economy and the rur al world in the l ate Empire

103

century marked the peak for these small production sites. In the previous chapter we have already mentioned the existence of suburbs around towns as secondary and subsidiary nuclei of population, but beyond this role as dwelling areas, suburbs were mostly economic in nature. The archaeology of small-scale production sites is limited mostly to rescue and commercial interventions, as these are usually located outside or in the vicinity of Roman towns, are located usually by accident, and consequently do not draw the attention of research projects. As a result, our knowledge is quite sketchy when it comes to individual towns, but as a whole, there are general patterns that seem to apply to most of the peninsula. These sites range from individual farms and small factories to villas. For instance, small agricultural production centres immediately outside the city walls of late Roman date are known from Tarragona (López Vilar 2006: 171-82), Huelva (Vidal et al. 2010) and Toledo (cf. Olmo Enciso 2009: 71-2). Most of these cases can be interpreted as small villas, which were directly related to the town. In Mérida, the fertile meadow of the Albarregas was the location of many agricultural and industrial sites (Mateos et al. 2002: 39; Sánchez Barrero 2004), usually linked to small dwelling areas. The wider territory around Valencia (Hortelano Uceda 2008; Pérez Mínguez 2006; cf. Glick 1970: 190-3) and Córdoba (Lacort Navarro 1988), which has been studied mostly through the remains of their irrigation systems, show similar patterns for the early fourth century: dense with small settlements, clusters of farms or villas. In a few areas, such as the modern provinces of Seville and Toledo, there is a high number of fourth-century, non-villa rural settlements directly linked to the production and supply of the neighbouring towns (García and Vázquez 2013; Rodríguez and Barrio 2003). Other than these known agricultural centres, there were also minor industrial complexes. As we have noted in the previous chapter, there were industrial or trade-focused suburbs near Barcelona and Lucentum. Similarly, in Lugo urban kilns (Rodríguez Colmenero 2011: 87) and in Mérida smithing foundries (Bustamante Álvarez 2013: 128) were still active during the fourth century, and it not difficult to imagine that similar local production centres were present in most towns of Hispania. However, throughout the fourth century, the old distinction between productive and residential areas blurred. We have already examined how the urban space was being redefined, usually by means of the rearrangement of the walled circuit, but as part of the same process, suburban activities began to appear within the walls. The clustering of settlement around agricultural areas outside the walls was paralleled with the emergence of industrial activities inside them. In Valencia and Barcelona, for instance, fullers, dyers, and food processing sites have been identified inside the

104 

The Iberian Peninsul a bet ween 300 and 850

Figure 3.1 Late Roman industrial establishment built inside one of the old forum buildings in Valencia, with space for storage, food-processing, and a pressing vat

Based on Álvarez et al. 2005

walled areas (figure 3.1; Álvarez et al. 2005; Beltrán de Heredia 2002b). The presence of lime kilns in the theatre of Córdoba after its abandonment by the early fifth century, for instance, accounts both for the relocation of industrial activities and the alteration of urban priorities (Hidalgo Prieto 2005: 402). In Mérida animal pens and small workshops spread throughout the city from the fifth century onwards (Alba Calzado 2004a: 238-40). Beyond these relocations of industrial activities, the process of ‘ruralization’ is usually discussed as being linked to ‘black’ or ‘dark earths’, which are layers of dark soil with high organic matter linked to agricultural patches of land in urban contexts. In the Hispanic cases, this has been proposed for Córdoba (Márquez and Ventura 2005: 438), although it is more abundant in

The economy and the rur al world in the l ate Empire

105

the literature of towns in Western and Southern Europe, both in the south (Italy) and the north (England and France; Nicosia et al. 2013). However, as it has been rightly pointed out by Neil Christie (2006: 261-2), these soils do not necessarily mean that these were patches devoted to agriculture or animal pens, nor that they represent an exclusively new Late Antique phenomenon.27 Burial areas The areas surrounding towns had been, by law, the areas reserved for the burial of the dead. The emergence of necropoleis around Roman towns usually occurred around the main roads, developing what has been called the Grabenstraßen, or ‘tomb roads’. This practice, however, declined during the late Roman period. After the introduction of east-west inhumation as the main burial practice around the end of the third century (and not directly related to the spread of Christianity), burials seem to have transformed into a more ‘private’ affair. In the Iberian peninsula there was an end of Grabenstraßen with inscriptions calling out the names of the deceased, and these were replaced by burials in clusters and proper cemeteries within an enclosed and well-defined space (González Villaescusa 2001: 146-7; Sánchez Ramos 2014: 11-7). While the burials of the poor and humble were mostly tegula-covered inhumations (figure 3.2) in dug-out tombs, identified by perishable markers (possibly timber), richer burials were much more varied. Family mausolea ceased to be vertical and expanded horizontally and underground, with semi-buried crypts and chambers. These new mausolea preserved the intimacy and privacy of the family, rather than being aimed at the stranger walking by. Lavish sarcophagi (figure 3.3), mosaics and offering tables (mensae) were not visible except to those who were allowed to walk in. With regards to grave goods and funerary offerings, they became less frequent, and different in nature. Lamps and unguentaria were replaced with personal items and clothing. This change is essential, as it probably indicates a transition from embalming to shrouding. Furthermore, the few grave goods that keep appearing beyond these personal items are usually small jugs, signifying that burial offerings had moved from solid to mostly liquid offerings. These changes in burial practice are clear indicators of a new late Roman mentality and a new relationship between the dead and the living, which is essential to an understanding of changing 27 The ‘dark earths’ phenomenon is not exclusive of the late Roman Empire, being detected throughout the Early Medieval period. However, recent studies show that is precisely from this moment on when they appear exensively in urban contexts.

106 

The Iberian Peninsul a bet ween 300 and 850

Figure 3.2 Late Roman burials in the forum of Valencia: Inhumations covered with tiles which show a change from early Roman practices of cremation and specific burial areas by main roads

From Ribera Lacomba 2008

Figure 3.3 Photograph of a fourth century Christian sarcophagus depicting Christ as the Good Shepherd. Currently used as an altar in Écija

practices in the later part of the period, when so-called ‘Germanic’ grave goods begin to appear, and these practices were adopted by Christianity (see Chapter 4), as the Saint Menna ampullae from the Museum of Albacete, or the Saint Eulalia jug from El Gatillo would indicate.

The economy and the rur al world in the l ate Empire

107

Beyond these transformations of burial practice, it is clear that there was a redefinition of the burial space. Cemeteries tended to be moved away from roads and instead clustered around a focal element. In many cases these new foci were sites of Christian worship. Although this type of cemetery (the burial ad sanctos) emerged in the late Roman period, it only became widespread and established during the fifth century. These new late Roman cemeteries had a long continuity in time, usually lasting throughout the Late Antique period, and whereas some develop from early Roman Grabenstraßen, most seem to have emerged in separate areas close to the older road necropoleis. This is the case, for instance, with the various necropoleis of Tarragona (Francolí cemetery, Mas Rimbau necropolis; Remolá Valverdú 2004), Zaragoza (North and Santa Engracia necropoleis; Galve and Mostalac 2007), Cartagena (San Antón necropolis; Berrocal and Laiz 1995a; 1995b), or Mérida (Santa Eulalia; Bejarano Osorio 1996; Mateos Cruz 1999: 112-40). In some other cases, as in Valencia (necropoleis of Orriols and La Boatella; Machancoses López 2015: 634-42, 667-74), Córdoba (North and West necropoleis; Sánchez Ramos 2007: 191-6), or Málaga (Paseo de los Tilos; Vaquerizo Gil 2007: 394), they carried on in use through the mid and late Roman periods. In Baelo Claudia (Vaquerizo Gil 2010: 192-3), the East gate necropolis was an expansion of the early Roman burial area, but the tombs in it seem to have been arranged in clusters around ‘main’ or ‘monumental’ tombs.

The late Roman villa: redefinition, expansion and collapse Besides the small urban and suburban exploitations and workshops, most of the agricultural production was, in the late Roman period, still monopolized by the villa system, even if this overall control would collapse and implode along with the Roman administration in Hispania. Recent archaeological studies on Hispanic villas and their evolution into Late Antiquity (e.g., Chavarria Arnau 2007; Cordero and Franco 2012; Fornell Muñoz 2005) describe the fourth century as the apex of the splendor of the villa and the fifth century as its final point, with a much reduced, but still significant, continuity into the sixth century. This chronological correlation between the presence of a Roman state and the existence of the villa system is key to understanding the nature of the villa economy and the rural landscape that depended on it. Plus, the alternatives to villa sites that begin to emerge in the late Roman centuries need further attention, as they form the basis of the post-Roman and Medieval rural landscape.

108 

The Iberian Peninsul a bet ween 300 and 850

Villas for production; villas for pleasure One of the main characteristics that can be seen in the archaeological record, and that has not been stressed enough in the literature, is the radical polarization of early imperial villas. Although a large proportion of villas continued to function both as elite residences and production spaces, a significant number evolved into either a more residential or a more productive type of settlement. Whereas the continuity of third century villas into the fourth century is quite high, in most of them there is a noticeable change in their nature. The juridical transformations of the land in the third and fourth centuries (by which public land, usually allotted to colonists, could now be bought and privatized; Ariño and Díaz 2002) prompted the concentration of plots and the emergence of large landowners. These would slowly invest in their rural properties, enlarging them into sumptuous, luxury late Roman villas (Fornell Muñoz 2005: 630-3), whereas others would simply be transformed into agricultural production sites. This monumentalization is seen in the archaeological record in the addition of mosaics, marble decorations, baths, and new areas of representation, in a process that parallels that seen in other Mediterranean provinces (cf. Christie 2006: 430-7). This has been explained as a result of the new interests of the urban elites, which by this stage (and taking into account the transformations of municipal administration described earlier) preferred to display wealth and status in their private dwellings rather than investing in public monuments. Furthermore, in the territory around Mérida (Arce Martínez 2002), the transformation and expansion of the domestic parts of villas has been linked to the arrival of new administrative elites as Mérida was promoted to a diocesan capital. The list of villas that were monumentalized in the fourth century is extensive, and has been the subject of detailed publications (above all, Chavarría Arnau 2007, Fernández et al. 2008, and more recently Valero Tévar 2015), but it should be noted that this process varies from region to region. Monumentalized villas of the fourth and early fifth centuries appear in Mérida (Álvarez Martínez 2006-2007; Cordero and Franco 2012) and the southern Mediterranean coast (Fornell Muñoz 2005: figs. 70-3), but are not found in the broad hinterland of Seville (García and Vázquez 2013: 376-7). The most representative examples of monumental late Roman estates are the sites of Vilauba, Els Ametllers, Torre Llauder, the villa of Fortunatus, Arellano, Baños de la Reina, La Torrecilla, El Val, El Ruedo, La Daragoleja, Aguilafuente, El Saucedo, and that of Veranes. These all show the same patterns of expansion, embellishment and the transformation of the dwelling

The economy and the rur al world in the l ate Empire

109

Figure 3.4 Photograph of one of the mosaic panels at the villa of Noheda

Valero Tévar 2015: figure 118

Figure 3.5 Reconstruction of the villa of La Olmeda, according to Abásolo and Martínez 2012

space (cf. Mar and Verde 2008), above all with the construction of new apsidal rooms (oeci), the addition of mosaics and marble carvings, and the expansion of bathing complexes (figs. 3.4-5), in a similar fashion to what we have already discussed for urban housing (see Chapter 2). Architecturally, there seems to have been a principal development towards enclosed

110 

The Iberian Peninsul a bet ween 300 and 850

buildings, organized around a peristyle. The productive areas of these sites are kept aside, as independent sets of buildings, indicating a clear visual division between the areas reserved for the otium and those reserved for the negotium. In a few cases, the monumentalization of the residential quarters and extension of the pars urbana is paralleled with the development of the production and storage areas (as has been proposed for the territory of Mérida; Cordero and Franco 2012: 153),28 but this was not common. A similar and opposite trend is identifiable in most sites, as some early Roman villas were transformed into more functional production sites for oil and wine (see below). This is a process clearly identified on the Mediterranean coast (Chavarría Arnau 2007: 125-9, Peña Cervantes 2005-2006), and it is not unusual to find rooms decorated with mosaics or bath complexes lined in opus signinum reconverted into wine or oil presses.29 This transformation of some secondary villas into simple productive sites underlines an economic system still based on rural sites producing on a large scale, but breaking with earlier patterns as the elite dwellings appear to concentrate and be enlarged. There were larger and more luxurious rural palaces for fewer but more powerful aristocrats. New rural palaces Within the process of monumentalization of villas described in the previous section, a few examples went even further and became large rural palaces, dwarfing in size and decoration not only early Roman villas, but also most of their coeval villas. Some examples are those of Noheda, Torreáguila, La Cocosa, Carranque, La Olmeda, Centcelles, Els Munts, Liédena, Los Quintanares, Torre de Palma, Quinta das Longas and São Cucufate. In some cases, as in Carranque, Las Pizarras (Coca) and Cercadilla, it has been suggested, not without controversy, that their construction was the result of the intervention of elites linked to the imperial family (Hidalgo Prieto 2002; Pérez and Reyes 2003). For instance, the monumental villa of Carranque was built ex novo during the reign of Theodosius (379-395). It has a central domed building and a 60m-long basilica building decorated with several expensive Mediterranean 28 In fact, the example of Mérida may be exceptional due to the political importance of Emerita and its elites in the Late Antique period. 29 Opus signinum is water proof, so it makes an ideal surface to reuse in the production of oil or wine (Peña Cervantes 2005-2006).

The economy and the rur al world in the l ate Empire

111

marbles, some of them imported from Asia Minor, such as the columns of the central nave. According to Isabel Rodà (2001), the use of such luxurious materials is closely related to the highest echelons of late Roman society during the last quarter of the fourth century. The name of Maternus, probably the villa’s owner, was written in one of the many mosaic floors preserved in the residential complex. This has allowed some scholars to suppose that this Maternus was Theodosius’s friend Maternus Cinegius, consul and governor of Asia Minor although this has been dismissed by the archaeological team currently working on the site (García-Entero et al. 2008). Beside this, the existence of a Medieval church and of Visigothic burials over the remains of part of the basilica complex of Carranque is a weak basis to defend the proposed martyrial-cult purpose for this villa (Bowes 2001; FernándezGaliano 2001). The absence of a primary context for most of these pieces of evidence means that a prudent approach is necessary before attempting any interpretations. Carranque is one of the most impressive examples of an ex novo constructed villa, but others were the result of the expansion and embellishment of pre-existing Roman villas, as in Els Munts, Torre Llauder or Centcelles (Valero Tévar 2015: 78). The villa of Noheda (figure 3.4) is perhaps the most extraordinary among them. Discovered in the 1980s, it has only been fully excavated in the 2010s (Valero Tévar 2015). The main part is a 300m² rectangular room with three apses. In this room, the floors were all covered with exquisite mosaics, the bottom of the walls with marble opus sectile (28 different types of marble have been identified) and the rest of the walls in polychrome paintings. The mosaics shows various classical scenes (the life of Paris; the myth of Pelops, Oenomaus, and Hippodameia; a pantomime; etc.) in various panels, surrounded by smaller panels with geometric and plant motifs. A secondary octagonal room, adjacent to the main one, had geometrics mosaics too, and various other minor rooms were also decorated with wall paintings and, possibly, marble sculptures. The villa of La Olmeda (Abásolo and Martínez 2012), equally embellished with mosaics, painted walls and marble decorations, has one defining element, which should be perhaps taken into consideration when dealing with these large rural palaces: it gave the idea of being fortified (figure 3.5). The main building faced inwards, enclosed in its own perimeter walls, with ornamental towers at each of its four corners. The second floors had an open gallery, which undermines this possible interpretation of the villa’s defensive aspect. However, it nevertheless underlines both the monumentality of the building (the fortified aspect parallels the villa with a small town with its

112 

The Iberian Peninsul a bet ween 300 and 850

walls), and is perhaps an indicator of the militarization of the administrative elites (cf. Chavarría Arnau 2007: 108). The development of these large rural palaces goes in hand with the increased monumentality seen in some urban houses in the same period, and both clearly show the exponential growth in the wealth of some landed aristocrats. The concentration of estates into the hands of fewer members of the urban elites is seen in the monumental villas of the late Roman period. But these large palaces also show something else. In the fourth and fifth centuries, a powerful nobility of landowning aristocrats, linked to the imperial administration (senatorial and other office-holding families), concentrated even larger estates, achieving far greater power and influence than had members of the provincial elites in the early and mid-Roman centuries (cf. Wickham 2005: 156-8). This emergence of extremely wealthy and powerful elites can be seen across the Empire and, by the fifth century, resulted in the formation of powerful landed provincial ‘nobilities’. They would become the people with whom the Barbarians would be discussing terms of settlement and arrangements to rule and administer their new kingdoms (cf. Halsall 2007: 488-96). The end of the villa economy Villas existed not simply as luxury elite dwellings and places of personal representation, but were also, as explained earlier, an essential part of the Roman economic system. After all, most villas were directly linked to the large-scale production of tradable agricultural goods. Furthermore, taxation on agricultural estates generated a large part of the state’s income, either in cash or in kind. Thus, the end of the Roman villa in structural and architectonical terms can be also considered as the end of this ‘villa economy’. Consequently, this has to be seen as a sign of the profound changes in late Roman social structuring and in the way that elites built their hegemony. The end of the Roman villa has been a major object of study in western European historiography as one of the main elements that enabled the relation of the Roman world to Early Medieval times (Brogiolo et al. 2005; Ouzoulias et al. 2001; Wickham 2005: 473-495). Furthermore, there are many regional and particular case studies that deal with the issue of the end of the Roman villa in the whole peninsula (a recent compilation can be seen in Caballero et al. 2012a). In general, the archaeological record shows that the end of the Roman villa took place by the fifth century, although in territories of minor or secondary towns, as has been studied for Tarazona (Bonilla et al. 2011;

The economy and the rur al world in the l ate Empire

113

Bonilla and Serrano 2012), this could have taken place even earlier.30 The majority of villas which have been excavated seem to have been abandoned or deeply transformed before the end of that century. That is the case, for example, of important sites like La Olmeda (de Palol Salellas 1964), Els Munts (Tarrats et al. 1998), Veranes (Fernández et al. 2004) or Valdetorres de Jarama (Elvira et al. 1998). In field surveys and regional studies this seems to be the common period for the end of the Roman villa as a structural reality. In the case of the territorium of Seville, a decline of the number of functional occupied villas has been documented in the second half of the fifth century (García and Vázquez 2013: 105-7), similar to most parts of central Iberia (Ariño Gil 2006; 2013) or the northwestern regions (Sánchez Pardo 2013). However, this general situation cannot be generalized for the totality of sites. In some cases, hints of continued occupation in Roman villas after the fifth century can be detected, and indicates the possibility that in some territories the large proprieties and latifundia continued to form the basis of their local economies. For example, the existence of private cult spaces in some Roman villas built in the sixth or seventh centuries has been interpreted this way, as at La Cocosa, Fraga and Carranque (FernándezGaliano 1998; cf. Sastre de Diego 2013) or Navatejera (Benéitez and Miguel 1993-1994). In the written record, some large proprieties in Visigothic times have been interpreted as the continued usage of these old villas, transformed into smaller places called villulas (Isla Frez 2001). The sources describe the death of Recceswinth at his villula of Gerticos (HWR 3), commonly situated in the province of Salamanca (Ariño Gil 2013: 106). In many cases, this process of abandonment is archaeologically seen through the presence of a profound change to the functionality of the spaces. It is very common, for example, to find sixth/seventh century cemeteries or kilns obliterating old villa spaces, mostly on the pars urbana (Chavarría Arnau 2007: 125-141). That is the case of the examples already mentioned of Veranes and Els Munts, and also of Aguilafuente (Esteban Molina 2007), El Val (Sánchez and Rascón 2006) or El Tinto (Barroso et al. 2001). This can be interpreted as the use of ancient spaces by the communities that occupied the territory in post-Roman times (see Chapter 6). Another piece of archaeological evidence which has been detected in the later phases of the Roman villa, also documented in other European contexts, are the presence of hearths, rickety walls and postholes destroying 30 This is an ongoing research project, and we would like to thank Óscar Bonilla for his personal comments on the preliminary results.

114 

The Iberian Peninsul a bet ween 300 and 850

mosaics and floors. A very well-known example is the case of the ‘Hall of the Charioteer’ (named after the mosaic depicting an auriga) in the villa of El Val, where a significant number of postholes and some hearths have been unearthed (Rascón et al. 1991). Similar examples have been detected in La Olmeda, El Tinto or Vilauba, all of them interpreted as the consequence of a ‘squatter’ presence reusing villa spaces (Lewit 2002).31 However, other interpretations have related them with the reuse of old spaces by the local communities detected in the fifth century (see below; Tejerizo García 2016). There is obviously a continuity of the habitation of villa spaces during the fifth century, although this habitation had nothing to do with the socioeconomic system related to the villa. In the case of Vilauba, for example, recent excavations have unearthed a domestic complex outside the main residential building dated to the seventh century (Castanyer et al. 2013). Arguably, the community living in those residential spaces were using the ancient Roman villa for their own economic purposes, but that does not imply that the villa had the same function as a central articulator of production as it did in late Roman times. It was the same physical space, but with different structural functions embedded in different social contexts. In any case, the end of the Roman villa, in general terms, shows the important and profound changes operating in the Iberian peninsula during late Roman times, causing some of the landscape characteristics of postRoman centuries.

Industrial exploitations of the landscape Whereas in the late Empire there was a proliferation of fabricae (large, imperial-run, factories of military supplies such as weapons or clothes), these were located in garrisoned cities on the frontiers (Esmonde Cleary 2013: 92-3, 149). These large production sites certainly stimulated the local and regional economy, but in Hispania, they were absent. The Iberian peninsula was a place where the state traditionally obtained raw materials, mainly minerals, and where it promoted private entrepreneurs 31 The term ‘squatter’, even if it is usually found in the literature, is not really adequate, as it pre-supposes an illicit occupation, and there are no grounds to suggest that this was the case. The end of the villa economy was also directly linked to the end of villas as centres of political representation. The rearrangement of habitation patterns and building practices in the late phases of the villa could equally be the work of the rightful owners or workmen who belonged to the old villa system.

The economy and the rur al world in the l ate Empire

115

who produced goods such as fish products, oil, and wine, which could be distributed through the annona. These small and large scale industrial complexes which produced goods in order to trade in bulk will also be discussed in this section. The increasing demand of the army throughout the fourth century prompted the continuity in the production of goods demanded by the state, like oil and metals. However, as the spending capacity of the imperial administration declined, so too did investment and production. Furthermore, the new late Roman administrative framework, where taxes could be collected in kind, meant that it was no longer necessary to turn agricultural produce into cash (and as the taxation capability of the Empire decreased, so to did the production of marketable goods which could be taxed in kind). Finally, the political breakdown of the Empire during the fifth century and the increasing isolation of regions and provinces limited the scope for the production of marketable goods, causing production to respond by changing from specialized to maximized production. The simplification of the larger economy forced the transformation of the local economies, and this greatly disrupted the production patterns already existing in the Iberian peninsula. Mining Large-scale mining seems to have largely stopped in the late third century. It is generally assumed that the benefit obtained from these exploitations was not sufficient compensation for the necessary effort to obtain them, although it does not rule out that they continued at a greatly reduced level. For the production of iron, for instance, the process of smelting required vast quantities of charcoal, which in turn needed the constant felling of trees to provide fuel, and such a rate of deforestation could not have been sustainable or economic over a long time. It also seems to be the case that in the late Empire, mine exploitation appears to have been achieved through smaller, short-lived mines, of which it is difficult to obtain reliable archaeological data (Domergue 1990: 215; Edmondson 1989: 90-1). This is a pattern visible in the main mining districts of the Iberian peninsula. In the case of Las Médulas, gold was extracted by blowing up of whole mountains by means of pressurized water, the soil later sieved and panned (Orejas and Sánchez-Palencia 2002). This method was not just intensive, but also very costly, so the abandonment of the main mines should be linked both to the potentially reduced margin of profit obtainable at this site and the exhaustion of the auriferous deposits (cf. Domergue 1987: 308). The iron mines of the Basque region, exploited in a similar intensive

116 

The Iberian Peninsul a bet ween 300 and 850

fashion, seem to have been in use only during the first century (Urteaga Artigas 1997), which attests to this Roman pattern of short but intense exploitation. Similarly, large-scale exploitation of iron mines in the Iberian ranges does not seem to have continued beyond the second half of the second century.32 Despite this, a small number of copper and lead mines appear to have been active through the fourth century, especially in the southeast and southwest (Domergue 1990: 216-7). Overall, large-scale mining was only effective and intensive during the early imperial period (the mines in Dacia, where minerals are more easily accessible than in Iberia may have been a further cause). Mining at less intensive levels continued in the late Roman period, but evidently not promoted by the state. These local mining activities continued into the post-Roman period, when the development of the Visigothic state (see Chapter 5) seems to have prompted further mining activities, especially of iron, as identified in southern Spain (Martín Civantos 2001:327-328; Rivera and Pérez 2010). Fish industries One of the main Spanish productions during the early Empire was fish products; sauces and pastes such as garum, hallec, or liquamen, and salted fish, the salsamenta. These industries were produced in coastal locations all across the peninsula, from the Cantabric coast all along to Catalonia and the Balearic islands. Some of the major production sites were most probably aimed at the large-scale export of their products, shipped in amphorae across the Mediterranean. There is, however, a general decline of Hispanic fish production from the mid-third century onwards (cf. Wilson 2007), coming to a minimum in the fifth, and only showing signs of residual activity by the sixth century (Lagóstena Barrios 2001: 337). Similarly, the evidence for the last purple dye production site in the Iberian peninsula suggests a mid-fifth century abandonment (Bernal et al. 2009). This is usually evidenced by the transformation and abandonment of the fish factories and their vats into dumps and rubbish pits. These changes could have perhaps been expected in the case of small production centres such as Lisbon (Silva 2011: 205-6) or s’Argamassa (Gurrea and Fernández 1995), but it is surprising to f ind in the case of major production factories. Sites such as Almuñécar (Gómez Becerra 1995; Lagóstena Barrios 2001: 153), Baelo Claudia (Grützner et al. 2010; 32 Óscar Bonilla Santander (pers. comm.), regarding the still-unpublished results of his fieldwork.

The economy and the rur al world in the l ate Empire

117

Lagóstena Barrios 2001: 122) and Troia ceased production by the f ifth century. In the last example, the diversif ication of the amphora types linked to the diminished volume of production (Almeida et al. 2014) are clear indicators of the transformation of the nature of these sites: they ceased to supply the rest of the Mediterranean with bulk cargoes but preserved a smaller, local, distribution range (cf. Wickham 2005: 710). A similar case can be put forward for the area of Cartagena, which up until the mid-fifth century continued to produce local amphorae and imitations of North African forms for f ish products, although their distribution range was limited to the territory of Cartagena and Alicante (Berrocal Caparrós 2012). In Huelva, the same reduction of the distribution range is evident, although kilns seem to have been in use until the early sixth century (O’Kelly Sandrós 2012). Oil and wine Whereas the fourth century was still a period of economic prosperity, the decline of Imperial trading networks throughout the fifth century seems to have affected the production of wine and oil as well. Oil and wine were not only a main part of the diet, but were essential in the Christian liturgy and as fuel for lamps. This made them not a commodity but, for many people, a basic necessity. The problem of studying the consumption of wine and oil is that their consumption is mostly identified through the presence of amphorae, which demonstrate the existence of long distance trade, rather than regional production. Despite this, it is still possible to quantify the volume of production by the known number of functioning rural pressing sites (over 60 in total: Peña Cervantes 2005-2006). The production of wine and oil seems to have still been focused on villa sites during the late Roman period, both with continuing early Roman presses and newly built ones, as in Vilauba (Castanyer et al. 2013). But the majority of presses seem to come to an end by the end of the fifth century, with a steady decline of the surviving sites into the sixth and seventh centuries. Lastly, one problem that arises from the study of wine and oil production as derived from presses (torcularia) is that it is difficult to distinguish what these were used for: wine, oil, or both. Regarding wine, the production of late Hispanic amphorae, such as the Beltrán 68 and Keay types XXV, LXVIII and LXI (usually pear-shaped and with a conical end), seems to come to an end during the mid-fifth century (Bernal Casasola 2008b: 48-50; Sánchez Fuentes 2008: 66). Production was mostly in Mediterranean and Baetican types, which were the regions in

118 

The Iberian Peninsul a bet ween 300 and 850

which long-distance trade continued. As with the fish products, the distribution range also diminished during the fifth century, to such an extent that at points it can be argued that wine was only distributed at a regional and local level. As evidenced by the decline of productive villas, it seems that large-scale production of wine decreased. Added to these data, surveys around Tarragona, combined with pollen analysis at the Ebro delta, show that the extensive vine plantations of the Roman period were substituted for other crops; mostly pasture for animals (Pizzo 2008). Wine continued to be imported from Africa and (to a lesser extent) from the East, but this must reflect luxury trade rather than basic wine needs, the demand for which was probably supplied locally. Such local production sites are rare to find, but significant when identified. For example, in Tarragona wine production seems to have been concentrated in small plantations, and a large press complex has been excavated at the Francolí suburb (Sánchez Fuentes 2008). Otherwise, rural villas with identified wine or oil presses seem to continue mostly in western (Lusitania) and north-eastern (Tarraconensis) Iberia (Peña Cervantes 2005-2006: 105-6). Urban wine presses have also been identified in Barcelona and Valencia, both close to the forum areas, and both later dismantled by the construction of episcopal complexes on top of their sites (Peña Cervantes 2008). Olive oil had been the star produce of Baetica during the early Empire, as it was shipped to Rome and the German limes as part of the annona in the famous Dressel 20 amphora (the Testaccio amphora), so Hispanic landowners greatly benefited from this high and constant demand (Ponsich 1988). However, during the third century the Severan emperors changed this policy, so that it would be North African oil that would be shipped with the annona. This greatly affected Hispanic oil production, although it continued into the late Roman period. Olive oil amphorae were produced in the Iberian peninsula into the mid-fifth century. The main late Roman forms linked to oil were the Dressel 23, Tejarillo 1, and (perhaps) Keay XLI; the first two being from Baetica while the third was from Tarraconensis (Berni Millet 2012; Berni and Moros 2012; Berrocal Caparrós 2014). As for production sites, most oil producing villas seem to continue into the fifth century, as noted above. The excavations at the villa of Milreu, in southern Portugal, which was the largest single oil production site in the Iberian peninsula, suggest that the pressing areas may have been in use through to the end of the fifth century (Teichner 2001). Combining the information known about villas (production) and amphorae (distribution), it is clear that large-scale oil production continued into the fifth century, but steadily declined afterwards.

The economy and the rur al world in the l ate Empire

119

Fineware production Red Slip Spanish f inewares (Terra Sigillata Hispánica: TSH) were f irst produced on a large scale during the early Empire. There were two main production sites, Tritium Magallum, in the Ebro valley, and Andújar, in Baetica. These sites supplied table wares for not only the whole peninsula, but also nearby regions such Southern Gaul. These production sites seem to have come to an end in the third century, being replaced by later typologies. In northern Spain, this was the so-called Terra Sigillata Hispánica Tardía (TSHT), while Terra Sigillata Hispánica Meridional (TSHM) was produced in the south. Both where first produced from around the fourth century (Orfila Pons 1993, 2008; Paz Peralta 1991). The former were wares made from pinkish-orange clays, with a shiny but low-quality red-orange gloss. They were produced in moulds, and were usually decorated with simple geometric or plant decorations. The latter were also mould-made but were characterized by the use of local clays, which gave the pottery a pinkish surface, and also the presence of engraved decorations. The firing was not always homogeneous, and it may indicate the lack of specialized kilns or bad management of the gas influx (Juan et al. 2013). The quality of the decorations was poor, especially as moulds were used for long periods, so the quality of the print was never as sharp as Gaulish or Italic productions. During the fifth century Tritium Magallum seems to have continued (or restarted) producing TSHT, while other production sites have been located in the Duero valley (Pérez Rodríguez-Aragón 2014). The variety of typological forms is very wide, although they were predominantly open forms, such as plates, bowls, dishes and trays. As with wine, oil and fish products, bulk-produced finewares seem to have been only profitable and manageable inside the imperial economic system, and perhaps it could not really compete against African Red Slip wares, which continued to flood the Hispanic market. The late phases of the kiln sites linked to the production of TSHT have not been excavated, but the production and distribution of TSHT seems to come to an end during the mid-fifth century (Vigil-Escalera 2013a). The end of the TSHT productions and the interrupted flow of ARS from the coast to the interior prompted the development of local/regional production sites during the fifth century, especially in the north-west and the Central Meseta, although also further afield. These western Hispanic productions, referred to as Cerámicas de Imitación de Sigillata (CIS) (Juan Tovar 2012), are called Terra Sigillata Bracarense Tardía (TSBT) (Fernández and Morais 2012) in Portugal, demonstrating a tendency to productive regionalization. These are the Spanish equivalents of the more-famous and better-studied

120 

The Iberian Peninsul a bet ween 300 and 850

Gaulish Dérievées des Sigillées Paléochretiénnes (DSP), as they copy and imitate earlier Samian productions (Rigoir 1968; Raynaud 1993). These CIS were regionally produced (Portuguese coast, Segovia, Central Duero valley, South Meseta, Mérida, Upper Ebro valley, and Galicia), and do not seem to have spread beyond their main areas apart from some transitional areas where their distribution intersected. Technically, they are usually grey or dark red due to irregular firing; characteristically, they have no gloss as the walls of these pots are usually just polished. The decorations imitate those of TSHT, although as they are not mould made; the motives are usually incised with a stylus or else stamped. The CIS had a very short chronological span, and they do not seem to have continued beyond the mid-sixth century. The development of the CIS (and TSBT, etc.) wares as substitutes to better quality ARS and TSHT show not only a transition from empire-wide trade to regional production, but also substantial changes in technical skills, such as glossing the pottery. These imitations of other high-standard productions of late Roman chronologies in the Iberian peninsula were still integrated in regional distribution and showed partial traces of specialized production. They form, however, a small interim period, as none of these productions seem to have continued beyond the late fifth century. By the sixth century, this process of regionalization and simplification had become more evident (see Chapter 7).

The rural societies: Towards a new settlement pattern The end of the Roman villa and the processes of change in the urban spaces led to important transformations in the late Roman rural world. The most important characteristic was the rupture with the late Roman landscape through the emergence of new forms of occupation of the space. In these contexts, rural communities, mostly formed by peasants, generated new forms of settlement pattern during the fifth century in which dwelling areas were no longer tied to villa-centred estates (Vigil-Escalera 2015). They are mainly characterised by the use of sunken featured buildings and the presence of silos as the main form of storage. These sites have been mostly located in the central part of the Iberian peninsula, where this rupture with the late Roman landscape was deeper, although they seem to be widely spread over the peninsula. For example, in the territorium of Seville, rural settlement not only seems to have reduced in late Roman times, but their concentration also seems to have been around roads and other axes that linked them with operative towns, rather than being established in the immediate vicinity of villas (García and Vázquez 2013).

The economy and the rur al world in the l ate Empire

121

Figure 3.6 Plan of the excavations at El Pelícano, where the old villa has been replaced by a series of dispersed minor settlements, but which are still linked to the old estate

Vigil-Escalera and Strato 2013

These settlements were usually located in the surroundings of Roman villas, sometimes even directly above them, reusing old spaces and materials, which seems to have been a very common process during this period. For example, at El Pelícano (figure 3.6) recent excavations have documented a fifth century occupation in the surroundings of a Roman villa and an accumulation of Roman tiles coming from the villa, showing the intense reuse of spaces and building materials (Vigil-Escalera and Strato 2013: 177200). Another good example is the site of Carratejera at which a significant group of silos, obliterating kilns, and productive spaces from a nearby Roman villa have been documented (Marcos et al. 2010). Similarly, the presence of silos in the pars urbana of a villa has been documented in the lavish rural palace of Noheda (Valero Tévar 2015). One of their main distinguishing characteristics in archaeological terms were the (re)apparition in the archaeological record of sunken features, which were used with domestic functions from this point onwards. Most of them can be interpreted as huts and silos, with the presence of a significant number of unidentified features related perhaps to quarry works and water management as has been proposed in other similar contexts in France (Catteddu 2001). The

122 

The Iberian Peninsul a bet ween 300 and 850

really interesting fact is that, when compared to the domestic architecture of subsequent centuries (see Chapter 6), sunken featured buildings are especially irregular, lack structural homogeneity and very simple in the execution. The presence of silos is fundamental to understand some of the changes that took place in the rural world during the fifth century. Silos represent the resilience of these new peasant communities to a new world in which storing surplus had a coherent economic and social sense (Vigil-Escalera and Quirós 2013). When the centralized and extensive production of cash crops linked to the villa production system collapsed in most parts of the Iberian peninsula, these new forms of medium and large term storage took the place of old horrea (granaries) and dolia (large ceramic storage jars), showing both that towns ceased to provide services to the rural population and the end of the economic necessity of centralized production. As far as is currently known, the economy of these sites was basically focused on self-sufficient agriculture, combining animal husbandry and cultivation, as demonstrated by the presence of faunal and crop remains on the sites. Furthermore, bioarchaeological studies in the northern part of the Iberian peninsula show important changes during the fifth century regarding economic strategies in these communities once the Roman market economy was dismantled. For example, the size of some domestic species, such as pigs or cattle, seems to have decreased during this period (Grau Sologestoa 2009). Some palynological analyses carried out in central Iberia also show important transformations of the landscape management by local communities (Ariño Gil 2006; Blanco et al. 2014). However, other productive activities are detected in these sites, such as smelting and the repair of metalwares. During the fifth century it is relatively common to find the accumulation of metal instruments on the sites, showing not only the presence of metal workers in these communities but also the lack of new metal in circulation, which can be related to the end of the main mining activities. For example, at the site of Soto de Tovilla a deposit of 59 metal tools have been documented (Martín and San Gregorio 2008). In relation to the emergence of these new peasant contexts, one of the major changes traditionally considered in this period was the appearance of a particular new ritual in which the furnished burials acquired a significant importance in relation to late Roman burials (see above). The objects used in this ritual are mostly associated with late Roman material culture, like Samian wares, military brooches, a special type of knife known as ‘Simancas-type knife’ or bronze objects. San Miguel del Arroyo was the first of these cemeteries to be studied in depth and can be considered a good case-study (de Palol Salellas 1958). Other good examples are the northern

123

The economy and the rur al world in the l ate Empire

Figure 3.7 Table showing the percentage of furnished burials in late Roman rural cemeteries NAME Simancas Roda de Eresma La Olmeda Norte La Olmeda Sur Dehesa de la Oliva (Área 1000) San Miguel del Arroyo La Morterona Los Cuernos Suellacabras Cabriana Soto de Tovilla Las Merchanas El Soto/El Encadenado

EXCAVATED BURIALS

FURNISHED BURIALS

% OF FURNISHED BURIALS

145 33 111 526 33 30 30 28 ¿50? 71 26 36 9

52 28 75 214 8 15 9 – – 55 15 15 6

30 85 68 41 24.4 50 30 – – 78 58 41 67

cemetery of La Olmeda (Abásolo et al. 1997) or Simancas (Rivera Manescau 1939), after which this burial culture was named. These cemeteries present a significant number of burials in the same contexts and irregular plans, with several burial orientations prior to the Christian West-East imposition. Even if this ritual is one of the major particularities of these cemeteries, it is noteworthy that it was used for specific individuals within these communities (figure 3.7). This phenomenon, traditionally linked to the Duero basin and to the presence of a military frontier in this part of the Iberian peninsula – the so-called ‘necropoleis of the Duero’ (de Palol Salellas 1958) – has recently been reinterpreted by Alfonso Vigil-Escalera (Vigil-Escalera 2015). For Vigil-Escalera, these burials would be the first funerary expression of the first forms of postRoman peasant communities. Moreover, these cemeteries may be showing the inner social differences within communities. Vigil-Escalera distinguishes four social strata from his study of burial disposals. Therefore, these cemeteries may be the material expression of social tensions within the groups whose identities were in flux in a moment of structural changes (cf. Halsall 1995a). This settlement pattern did not seem to last very long. This model was mostly dismantled at the beginning of the sixth century, when the main transformation of the rural landscape took place. With the emergence of a new balance of power in the peninsula, the rural settlement pattern had to accommodate the final collapse of the Roman system.

124 

The Iberian Peninsul a bet ween 300 and 850

This chapter has dealt with the late Roman archaeology of the economy of towns and cities, focusing both on the surrounding landscapes and the rural world. Overall, the material record shows a significant complexity in the structure of the Roman hinterland. The presence of many productive sites directly related to the civitas, such as industrial or agricultural centres shows the articulation of urban economy beyond the limits of city walls in order to satisfy the demand of the urban population. Mining, wine and oil production or fish industries are some of the specialized late Roman economic productions identified in the archaeological record of the Iberian peninsula.33 Changes during this period also affected burial rites in urban contexts, with a visible shift from burials connected to the main roads (the Grabenstraßen) to the communal cemetery, stressing the increasing importance of urban political communities. However, the main process in the late Roman period, or at least the most visible in archaeological terms, is the development of a new conception of the villa. The presence of productive sites in the Roman rural world was highly common. But it was only during the fourth century that some these spaces were transformed into true rural palaces through the embellishment and the investment of large amounts of capital in them. This can be seen as an important shift of late Roman elites as a consequence of some key economic processes, like the concentration of rural estates or the stress in trans-territorial commerce. These villas articulated the rural world through a network of sites in close connection to urban economies. This settlement pattern collapsed in most parts of the Iberian peninsula during the fifth century. The end of an important number of villas, together with substantial changes in numerous urban contexts, shows a deep structural break in the rural world. This break is noticeable in important changes in some key productions like the pottery and also in the emergence of new settlements directly related to the dismantling of the villa network. New peasant realities emerged, characterized mainly by the presence of sunken featured buildings and silos as the main form of storage. Finally, the structural changes of the fifth century are also shown in the appearance of a new form of burial rite in which some grave goods like brooches or pottery artefacts played an important role, preceding what will be one of the most important funerary rituals of the sixth and seventh centuries.

33 Other productions, such as wild animals for races and games, cured meats and grain, or skilled workmen, are known to have come from Hispania through written sources (even in this late period), but there is very little that can be said about them archaeologically.

4

Christianization and Germanization: New evidence for current debates

Traditionally, both the conversion to Christianity and the arrival of the Barbarians have been considered the causes of change in the material culture of Late Antique Hispania. This was understood from the presence of new burial practices, new decorative motifs in architecture and jewellery, and the testimony of the written sources. Despite this, debate amongst scholars on the interpretation of these finds is still ongoing and, in many cases, highly controversial. This chapter intends to explain the way in which these two large transformations took place, how they can be observed through the material record, and what repercussions they had on the long-term changes of Late Antiquity.

Understanding Christianity through archaeology Christianity is one of the most defining factors of Late Antiquity, because its influence went beyond religion and belief: it influenced the material culture, cult spaces and practices; it was even used as an added element of people’s identities (Pohl 2013b). Beyond the impact of Christianity on Roman monumentality, urbanism, and burial practices (as discussed in Chapters 2 and 3), the archaeology of Christianity offers an extra dimension of analysis: how Christianity worked within late Roman society and how it functioned as a religion in its early centuries. The way it spread (geographically and socially) and how it appeared in the peninsula can be explained through the analysis of the material culture. Furthermore, the material culture can be studied from two perspectives: understanding it as the archaeology of Christianity (how the Christian cult worked physically) and the archaeology of Christians and their conversion (how Christianity became an extra layer of identity). It is in this context that the Archaeology of Liturgy, a recently developed method of analysis, has a key role in fully understanding the earliest traces of the Christian cult in the peninsula (Sastre de Diego 2013). The first Christians Archaeological evidence for Christianity and Christian material culture appeared in Hispania much later than the first literary mention of Christians

126 

The Iberian Peninsul a bet ween 300 and 850

in the Iberian peninsula. This is the famous Epistula 67 of Bishop Cyprian of Carthage, dated to 254, and addressed to the communities of Mérida and León-Astorga. Other than this, the first Hispanic recorded martyrdom (that of Bishop Fructuosus and his fellow bishops Augurius and Eulogius of Tarragona) also belongs to this same moment. However, it is not until (roughly) the first half of the fourth century, that Christian material evidence appears in the archaeological record, and it always appears in a minor proportion in comparison with non-Christian materials. These first elements are a few marble sarcophagi (that would usually be reused in later periods; Sotomayor Muro 1975; 1988) and domestic items (especially lamps and pottery, most of them originating from North Africa), which were decorated with Christian iconography. Both the early written sources and the material culture show that initially Christianity was urban in nature. The recent discovery of a Chi-Rho painted on the wall of a water cistern inside a Roman house in Merida (figure 4.1) has been interpreted as evidence of the development of private worship before the Constantinian legalization of Christianity with the Edict of Milan in 313. However, no liturgical features have been found inside the room, beyond a quadrangular mark that seems to be visible on the floor. This could be interpreted as the original location of an altar or a liturgical mensa. Besides its controversial potential function, this painting is proof of how Christianity was spreading through the upper echelons of the Roman society from the first decades of the fourth century and included wealthy owners of large domus (Sastre de Diego 2015). Another remarkable recent discovery, the so-called ‘patena de Cristo’ found at Castulo in 2014, confirms this urban character for the initial Hispanic Christianity, as well as the difficulty in interpreting the type of religious architecture where such a liturgical object worked. The ‘patena’, a circular glass tray, is Constantinian in date, according to the assemblage of finds (especially pottery and coinage) found in its archaeological context, and is one of the oldest representations of Christ in the Western Empire (figure 4.2).34 Even if most of such items with Christian iconography were consumed by the elite, Christian aristocratic families were still a minority during the Constantinian period. According to the limited statistical survey of the first Christian aristocrats carried out by Michele Salzman (2002: 90-92), which included those Roman citizens with senatorial range (illustris, spectabilis, 34 The archaeological team is lead by Marcelo Castro (Forum MMX Project), together with Bautista Cebrián and David Expósito; we would like to thank him and his team for facilitating access to their information.

Christianization and Germanization: New evidence for current debates

127

Figure 4.1 Photograph of the Chi-Rho identified inside the cistern of the domus of the Sala Decumanus in Mérida, the earliest known depiction of Christian symbolism in the Iberian peninsula

Photograph by the Consorcio de Mérida, used with permission

Figure 4.2 The patena from Castulo, a glass offering open form with the image of Christ in majesty, discovered during the excavations of 2014

128 

The Iberian Peninsul a bet ween 300 and 850

clarissimus), only a couple of Christians from Hispania are documented: Consul Acilius Severus (PLRE 1.834) and the poet Juvencus. The large-scale conversion of Hispanic Roman society took place between the end of the fourth century, especially under the reign of emperor Theodosius (who was originally from Hispania), and the middle of the fifth, a point at which the material culture, at least in urban contexts, appears to be completely Christianized. The archaeology of Christians Christianity was an identity marker in Late Antique society, introducing an element of alterity within Roman society (Christian v. non-Christian) which is reflected in the material record. The most abundant evidence for the presence of Christians is found in funerary epigraphy, but the presence of Christian iconography in daily objects underlines the spread of visual Christianity and the increasing number of people who would be able to recognise (and relate) to such symbols. Christian material culture Regarding iconography, it first appears in sarcophagi and pottery, as already noted. In both cases, this early Christian iconography is dominated by the idea of salvation and the power of God, usually through representations of Biblical scenes (e.g., the sacrifice of Isaac, Jonah and the whale, Daniel and the lions, Noah’s Ark, etc.). Nevertheless, many others fourth century fragments of either sarcophagi (Beltrán Fortes 1999; García 2012) or African pottery represent ambiguous motifs, which can be better understood as traditionally Mediterranean, since it is difficult to give to them a religious interpretation. Of course, pagans could have used plates and lamps with Christian decorations – in the same way that Christians (who lived in a Classical Roman culture) still read the classics and preserved pagan mythological iconography. But the Christian message was present, and it was understandable and identifiable. The case of sarcophagi is a little different, as these would not have been visible to the general public (according to the change in burial practices, see Chapter 3), and clearly show the relationship between Christianity and the family of the deceased. Besides the iconography, sarcophagi have traditionally been considered imports from the main western workshops, such as Rome and Carthage. However, marble analysis carried out on several remains found in the Southern peninsula show how some of them were made with local materials, which demonstrates that there was the creation (and understanding) of Christian art in Hispania

Christianization and Germanization: New evidence for current debates

129

from an early date, and there was not simply the consumption of imported art (García García 2012). Epigraphy in funerary contexts was used to underline the Christian identity of the deceased above all other layers (Roman, Barbarian, etc.). Although the first piece of evidence for Hispanic Christian epigraphy is dated to the second half of the fourth century, it is during the fifth century that Christian epigraphy became common place. Mérida and Tarragona, both the ancient provincial capital and main centre for martyrial cult (Eulalia and Fructuosus respectively), nowadays preserve the most relevant collections of Christian epigraphy in Spain: around 150 inscriptions in Tarragona, including its ager (according to the last survey, see Gorostidi 2013), and more than 200 inscriptions in the case of Mérida, just from within the city (Ramírez Sádaba 2008). Most of these Christian inscriptions (figure 4.3) are funerary in their character, and the majority of them very similar terms, with small differences depending on the region (cf. ICERV). However, they always provide the same information: firstly the name of the deceased (usually just one name in contrast with the traditional Roman tria nomina); secondly his/ her Christian denomination (servus Dei, famulus Dei); thirdly the age of the deceased (vixit annos plus minus); after this, the formula announcing the passing away into the peace of Christ (requievit in pace, quiescit); and finally the year of the death (according to the Consular date or the Hispanic era) including both the month and the day. This final date is perhaps the most important, as it marks not the departure of the person from this world, but rather the moment in which the person entered his/her new eternal life, the date when they were born again. Christianity and the elites Due to the nature of most existing remains, it is only possible to discuss in brief the adoption of Christianity by the elites. By the period of peak conversion (late fourth or early fifth centuries) they appear to have appropriated various elements of Christian iconography into traditional forms of elite display. For instance, in the large late Roman villas (see Chapter 3), Christian motifs and decorations are proudly displayed by their owners. Various mosaics depicting the Chi-Rho decorate the main rooms (the oecus) of these villas, such as Quinta das Longas (Portugal; Oliveira et al. 2011) and the villa of Fortunatus at Fraga. In both of these examples, the choice of the most representative room in the house in which to place the Christian symbol, and the wider iconographic display, clearly proclaim the Christian belief of the owners; even at Fraga, the name of Fortunatus, probably the owner, was put on the inscription together with the Chi-Rho. This villa was

130 

The Iberian Peninsul a bet ween 300 and 850

Figure 4.3 Late sixth century Christian funerary inscription inscribed in a reused public document, currently in the Museum of Seville

Photograph by Dr. Patricia González, used with permission). It reads: [Leontius]| famulus |dei, vixit| annos plus| minus LXX. Re|cessit in pa|ce sub die III k(alendas)| apr(ilis). Era DCXXXIIII – ‘Leontius, God’s servant, lived for 70 years. Rested in peace on the third day of the Kalends of April. In the Era 634 (= 30 March 595)’.

completely Christianized during the following centuries as a church was built, over several phases, reusing one of the sides of the residential complex of the villa; the liturgical features preserved there belong to the last phase (de Palol and Pladevall 1999; Sastre de Diego 2013; Utrero Agudo 2006). Another element of traditional elite display was public munificence. In later periods (especially later, in the sixth century – see Chapter 5), construction of churches became commonplace, but the very first examples of elite

Christianization and Germanization: New evidence for current debates

131

Christian donation are liturgic elements. These pious donations were slowly replacing public civic munificence as the principal way of displaying wealth and status by urban elites who, as seen in Chapter 1, were by that time losing political interest in funding new civic buildings. Although lay members could not enter into the sanctuary per se, social status was clearly reflected in the several features, both functional and decorative, which were present in churches, many of which were liturgical gifts given by members of the local elites. The first examples of elite liturgic donations are lamps, crosses, liturgical jars and trays; which were made in marble, precious metals, or glass. Examples of these have barely survived in the Iberian peninsula, the number of remains being still quite small: Mantius’ plate (Mérida) in marble, the already mentioned ‘patena de Cristo’ from Castulo, the patena in glass from Santa Pola, and the jar and the patena from El Gatillo (in bronze). Two other items of liturgical furniture that were donated by Christian aristocrats and that survive in larger numbers were mensae and labra. The mensae were tables, of different shapes and materials, where offerings were dedicated and liturgical rites developed. The labra were large, shallow round basins, used as cult objects in the fourth century as containers of sacred water (Ambrogi 2011). From the fifth century onwards mensae and labra began to be made in stone. Both from a morphologic and from semantic point of view, mensae and labra come from the pagan tradition. The dedication of mensae shows a continuation of the practice of symposia, which were firstly developed in domestic contexts (stibadium) and were quickly adopted in the funerary world. In this context, the mensae became tables for funerary banquets. The favourite shapes for this type of Christian mensa were the circular and the semi-circular or sigmatic (after one of the versions of the Greek letter sigma, which could be represented in capital letters as a C). These tables were created in Greek workshops and, until the last decade of the 20th century, this type of table had barely been documented in the Iberian peninsula. However, both new excavations (like the site of ‘La Encarnación’ in Seville) and new studies carried out in the last two decades have confirmed the presence of this type of mensa all along the Iberian coast. Consequently the Iberian peninsula is nowadays one of the Mediterranean areas with the greatest number of sigmatic tables (Sastre de Diego 2013). Most of the remains are fragments of tables imported from Greece between the fifth and the sixth centuries. However, there are also local productions, which imitate the Eastern examples while including original features. This is the case with the sigmatic slab found at the basilica of Casa Herrera, dated either to the end of the fifth century or the beginning of the sixth (fig 4.4).

132 

The Iberian Peninsul a bet ween 300 and 850

Figure 4.4 The altar slab from Casa Herrera, a sixth century sigmatic mensa with a commemorative inscription

From Caballero and Ulbert 1974

Its arch is framed in a bead string and flanked by two carvings, a decoration which does not appear in Eastern examples, and it has been interpreted as a local Lusitanian Roman motif (Silva Fernandes 2011). The same motif was used in other local productions, like chancel screens from the sixth-century workshop of Merida. The Graeco-Roman labra were also an element which was borrowed from pagan tradition for Christian liturgy (three labra found at the temple of Bona Dea in Trieste preserve their inscriptions dedicated to the goddess). With both sigmatic tables and labra the inscriptions which included the name of the offerent (usually a private donor) were placed on the edge of the slab. In the Iberian peninsula, several Late Antique objects made out of marble, with a rectangular design but with a circular carving inside appear to have copied the idea of classic labra. Most of the early Christian liturgical mensae found until now in the Iberian peninsula are circular as well. Depending on their dimensions (between 0.2 and 1m in diameter) and the material for their construction (stone, silver, bronze or pottery), they could have functioned as offering trays

Christianization and Germanization: New evidence for current debates

133

Figure 4.5 The Quiroga Chi-Rho, a Late Antique elite Christian liturgic dedication in marble

or as liturgical tables. An example of the former is the so-called Mantius’ dish, found in Mérida and dated to the end of the fourth century or the beginning of the fifth. Its surface is decorated with a portrait and an inscription of the donor, Mantius, together with a Chi-Rho. As for the latter case, the best example would be the Crismón de Quiroga (the ‘Quiroga Chi-Rho’; figure 4.5), from Orense, which is an exceptional masterpiece, perhaps dated to the fifth century. This item is a marble slab, almost 1 meter in diameter, decorated with a big Chi-Rho which occupies the whole surface, and an inscription on its edge: ‘[G]old is cheap for you, the weight of silver will pass; it is better that you shine through your own happiness’.35 This text mixes the words of authors such as Vergil with those of other, later, Christian authors (Fontaine 1972-74). Recently, a new dating to the second half of the sixth century has been proposed for this object, due to a possible literary influence from Venantius Fortunatus (Anguita Jaén 2011). Nevertheless, similar expressions alluding to metals such as gold and silver as metaphors for greed can be found in fifth-century Church Fathers such as Augustine (serm. 48.8) and Jerome. Furthermore, the meaning of the inscription fits better in the religious and cultural context of the fifth century, in the middle of the process of the Christianization of the aristocracy and continuity with traditional and 35 + aurum vile tibi est, argenti ponderacedant; plus est quod propia felicitate nites.

134 

The Iberian Peninsul a bet ween 300 and 850

pagan customs. The ‘Quiroga Chi-Rho’ was an expression of the donor’s charity, reinterpreting luxury products from a new Christian perspective. Similarly, in the funerary world, elites underlined in their grave markers both their identity as Christians and their position in society. For instance, in Baetica we find five clarissimae Christian women: Aurelia Proba (ICERV 539), Cervella (ICERV, 111) and Paula (ICERV 110) from Seville, Eusebia of Olontigi (CILA 2.4.1029) and Alexandria Nebrisensis (ICERV 131).36 The Christianization of the aristocracy saw the introduction of new types of burial practice. These elements include mosaic floors covering tombs, in which tessellated portraits and inscriptions show the high dignity of the owner. One of the most famous funerary Christian mosaics, from Tarragona, belongs to the Bishop Optimus (CIL II2 14.2085)37 and is dated to the beginning of the fifth century. The mosaic covering of Ampelius, with a remarkable decoration (CIL II2 14.2105) has the same origin and dating.38 Another mosaic tomb is that of Severina (IHC 410)39 in Denia, dated to the late fourth to early fifth century as well. Stone sarcophagi were still a choice amongst urban elites. However, sarcophagi with Christian iconography are in a minority and they tend to have been imported from Rome and other Mediterranean centres. Regarding these, it is quite difficult to establish when they were used because those that survive are usually in secondary locations, not in their original resting place. Overall, Christianity as an identity can be traced through the archaeology with relative ease, although not always with certainty. Whereas some displays of the Christian cult (such as votive offerings, or burial practice) can be clear indicators, the spread of Christian iconography is not as clear. Even so, by the end of the Roman period in Hispania, the overwhelming majority of the material culture can be understood in a Christian context. In many aspects it was an identity which was built upon the Romanitas of the people, but that still needed underlining in this period of transition away from the pagan past. The archaeology of Christianity Alongside the development of a Christian identity, and as a result of the growing number of converts, it became necessary to expand the number 36 This survey is currently being carried out by Dr. E. Sánchez Medina. 37 Optime magnarum [– – –] / cui maxima reru[m – – –] / divinas caeli quas promisera[t] / arces ecce dedit sancta Christi (!) / in sede quiescis. 38 / Ampeli / in pace / requiescas. 39 Severina / vixit an/[n]os XXXX / [r]ecessit in / pace terti/u(m) Idus Feb(ruarias).

Christianization and Germanization: New evidence for current debates

135

of centres for the Christian cult. Christianity, as a religion, had a set of specific ritual needs which determined the layout and nature of the cult areas, and as noted earlier, pagan temples were not readily adaptable as centres of Christian worship. Christianity is a religion for the community, and requires the faithful to be present during its mysteries. This is perhaps the key difference between earlier cults and the new religion, and why the basilica form was first adopted as a cult building. However, originally it just required a room with an altar. The earliest cases of centres for the Christian cult are the suburban martyrial shrines mentioned earlier (Chapter 2), such as those of Tarragona and Mérida (and the less well known examples of Complutum and Córdoba). These shrines were not more than small apsed rooms in which an altar was established, usually during the late fourth century. Christian building monumentality appeared later than the first appearances of Christian iconography in the material record. Although there is an exceptional fourth century precedent (the basilica of Torre de Palma), most of the early religious buildings were built from the fifth century onwards (and mostly only from the second half of the century). Although they are few, some interesting architectural inscriptions are also preserved from the earliest Christian buildings. Thanks to these inscriptions we know the name of some churches and basilicas and how they became sacred spaces (consecratio) by the inclusion (depositio) of relics in their altars. Most of them are urban in nature and belong to the sixth and seventh centuries (see Chapter 5), although the earliest example is a marble fragment with the fifth century inscription of Fructuosus and his fellow martyrs. The fragment, probably part of an altar table, was found in a secondary context, in the basilica of Francolí at Tarragona. Of the same date is the martyr-funerary basilica of Saint Eulalia at Mérida which developed from the original martyrial shrine. Remains of the earlier religious building were found during the excavations of the interior of the medieval church of Saint Eulalia (Caballero and Mateos, 1992), which was built with the same layout as the earlier building. Altars had to be placed in a secluded space (the sanctuary) which was limited only for priests, they had to be consecrated and, at least since the fifth century, they also had to contain relics. In this way, the altar and sanctuary compound clearly embodies strong symbolic and spiritual concepts, but with a material dimension as well. The centrality of the altar as cult element is such that in order to better understand the emergence and evolution of early Christian cult sites a new type of spatial analysis, the Archaeology of Liturgy, has been developed in recent years (Sastre de Diego 2013).

136 

The Iberian Peninsul a bet ween 300 and 850

The Archaeology of Liturgy is the stratigraphic analysis of altars within their architectonic and archaeological context, treating them as an archaeological f ind rather than as an isolated object – the way in which they had mostly been studied previously by art historians. This method is based on the same principles as the Archaeology of Architecture. Giving priority to the archaeological information, the liturgical elements of Late Antique churches can be divided into four groups (Sastre de Diego 2013): the remains found in situ belong to the first group; other archaeological evidence, even if not from a primary context, is in the second group. The third group is integrated by ancient liturgical features (mostly altars) that were reused in modern altars and religious installations, retaining their symbolic and religious function. Last and least valuable is the fourth group, which includes all the liturgical remains whose archaeological provenance is unknown. Even if there are remarkable masterpieces in this last group, most of them preserved in museums, it is not possible to obtain precise chronological and historical information from them. Only a comparison with the remains of the first two groups can make the pieces in the fourth group useful. Depending on the shape, material, dimensions, how the stone was carved, its decoration and other possible variations (such as the presence of epigraphy), liturgical sculpture can be differentiated into several types, more numerous than those established before. In addition, the groups are not as rigid as they have been traditionally considered. Another advantage of understanding these items as archaeological artefacts and not as works of art is that they can be read in their archaeological context and fitted into a sequence. For instance, marks which show where an altar or a chancel was located, as long as its chronological sequence is well known, is incredibly valuable information. Moreover, many altars and chancel fragments of abandoned or destroyed churches have been found in secondary contexts; they are thus records found in a secondary position but still belonging to an historical phase of the building and, in many cases, sharing the same material and constructive characteristics of the rest of architectonical elements. A piece of evidence found in these conditions is of far more value than a complete and profusely decorated piece kept in a museum whose original provenance is unknown. Furthermore, the altar as an archaeological item cannot be understood separately from, for instance, pilasters, columns, capitals or windows, because together they all belong to the building’s cultural context. For this same reason, altars can be held to the laws of stratigraphy, as they can also be integrated into the constructive cycle, and a refined dating can be obtained by the analysis of the technology with which an alter has been constructed.

Christianization and Germanization: New evidence for current debates

137

Figure 4.6 Altar from the church of El Gatillo, as it stands today

Thanks to this new archaeological perspective, the number of known Late Antique and Early Medieval altars has increased in the Iberian peninsula from less than the 60 identified, with varying degrees of consistency, in the traditional catalogues (de Palol Salellas 1967; Godoy Fernández 1995; Quevedo-Chigas 1995) to more than 550 (Sastre de Diego 2013). Two main factors explain this increase: on the one hand, considering the altar as a material object allows us to take into account every single piece, including the Roman elements reused, such as ara, pedestals and slabs. Indeed, reuse is one of the main characteristics of this period: one of the ancient Hispanic Christian altars is, in fact, a Roman ashlar block which was found in situ in the centre of the sanctuary of the rural church of El Gatillo, in Cáceres, dated to the fifth century (Caballero Zoreda 2009, figure 4.6). According to this research, the oldest remains of liturgical features cannot be dated to before the middle of the fifth century. For instance, several fragments of altar slabs from Catalonia and the Balearic Islands (Rosas, Ampurias, Saint Mary and Peter in Terrasa, Sanitja) are traditionally dated to a general period between the fourth-fifth centuries or under the generic epigraphy of ‘early Christian’ (de Palol Salellas 1967). However, all of these slabs have the same characteristics (rectangular shape, a ‘classical’ triple

138 

The Iberian Peninsul a bet ween 300 and 850

frame, thinness, made out of marble) as others which are stratigraphically dated more precisely from the fifth century onwards, especially in the sixth. For instance, the first remains of Terrasa belong to the construction of its episcopal complex, c. 450, according to the last analysis, but this type of slab is even used in the seventh century. Moreover, no liturgical furniture has been found in the late Roman buildings that have traditionally been considered the emblematic martyria of early Christian Hispania, such as Marialba in its first phase, La Alberca and Centcelles. 40 Without liturgical and epigraphic evidence, all of them would probably have been explained as late Roman mausolea. Until now, there are only two funerary buildings with liturgical furniture – more precisely slabs – and both of them are from a much later date. In Pueblanueva an altar slab was reused to cover one of the tombs located outside the Theodosian mausoleum, close to the apse which was built in a second phase, dated to the sixth century. The second example is the religious complex of the Roman villa of La Cocosa, which was built around an earlier privileged tetralobulated mausoleum, comprising a set of rooms that included a baptistery. Another altar slab, in this case belonging to the Panonian style, was found in a secondary position, and can also be dated to the sixth century.

Locals and barbarians The arrival of the Barbarians was in many cases violent and accompanied by war (cf. Ward-Perkins 2005) but this does not seem to have caused major destruction or changes to the urban or rural landscapes. These peoples were not nomads and were looking for places where to settle or to be settled by Rome. Despite this, and as an odd example, there are some traces of fifth century destruction layers that could be linked to barbarian activity and warfare, especially in the northern suburb of Mérida (Alba and Mateos 2008: 267; Mateos and Alba 2000: 148). However, this may be because Mérida was the main objective of barbarian attacks, as it was the capital of the diocese. Barbarians preferred to capture a city rather than level it to the ground, as there would be no long-term tax to obtain from burned ruins. In the countryside, however, the situation is slightly different, as there were no large walls to protect the inhabitants, and the wealthy villas were easy targets. There are some examples of late Roman villas with ‘destruction layers’ which 40 For Centcelles, this would further support its interpretation as a praetorium which was noted in Chapter 3.

Christianization and Germanization: New evidence for current debates

139

could be dated to the fifth century (e.g. the villa of Can Palau; Coll and Roig 2013), although they do not necessarily belong to violent sacking events by the Barbarians – an interpretation that would be quite risky to put forward. However, from an archaeological point of view, it is more interesting to see whether the arrival of the Visigoths and the Sueves (Vandal and Alan presence in the peninsula lasted only a very short time) had a real impact on the material culture or the structuring of urbanism and territory during the event of their first arrival in the fifth century. Understanding the settlement of the Barbarians and the new elements of material culture that appear in this period are essential to developing interpretations of the processes that modeled the last decades of the late Roman period, and that link it with the sixth century. The ‘Germanic’ grave goods Archaeologically, there is a series of objects found in graves (mostly jewellery and metalwork) which are commonly described as ‘Germanic’ or ‘Visigothic’. The use of this adjective is, in these cases, not a chronological indicator (i.e., ‘produced or dated to the Visigothic period’ – see the Introduction) but is rather used with ethnic connotations. Based on stylistic grounds, Visigothic fibulae and belt buckles are intrinsically considered to have been worn and fabricated by the Visigoths as identity markers. This ‘ethnic’ interpretation is quite problematic, as we will discuss below. Most of these finds correspond to items fabricated in the cloisonné technique (small bronze geometric cells in-filled with gems or glass paste), the most characteristic being the belt buckles and the eagle brooches (figure 4.7), but which also includes bracelets, rings and, sometimes, weapons (Ripoll López 1985; Zeiss 1934). This technique is, originally, from the Steppe world, but it spread quickly through the peoples living along the Danube and the Roman frontier (including the Goths) through the fourth century (Kim 2016: 164-73). However, and despite this very characteristic decorative pattern, most of the ‘Visigothic’ forms derive from Roman (especially military) typologies (Aurrecoechea Fernández 2001: 231-6; Mariné Isidro 2001: 274-6; cf. Swift 2000: 230-1). Finds of such items are common across the Iberian peninsula and Southern Gaul, although there are two areas of higher concentrations: Septimania and the Central Meseta north of the Tagus (see Ebel-Zepezauer 2000: figure 1), which has traditionally lead to the assumption that this was the ‘core’ of Visigothic settlement (further discussed below). The necropoleis of Daganzo/Complutum (Rascón Marqués 1999: 63) and of Argaray/Obietagarria (Pamplona; Mezquíriz de Irujo 2004: 47-64) are some of

140 

The Iberian Peninsul a bet ween 300 and 850

Figure 4.7 Two cloisonée Visigothic eagle brooches retrieved from the site of Alovera, and which are now at the National Archaeological Museum in Madrid

Photo by Larry Wentzel, Wikimedia Commons

the very few urban and peri-urban burial sites in which these type of grave goods appear, and both can be dated at least to the late fifth century. In these sites various of these ‘Germanic’ items were found: weapons, rings, fibulae, and buckles, in a few chosen burials which sharply contrasts with the grave goods found in the same necropoleis. Such finds in urban contexts are quite rare, and in many cases their relevance has perhaps been exaggerated, as with the ‘Vandalic’ grave goods found in a tomb in Málaga (Vaquerizo Gil 2010: 221-2) or the ‘Suevic’ one in Mérida (Heras and Olmedo 2015). Most of these finds have been retrieved from rural burial sites. Two of the most famous necropoleis linked to ‘Visigothic’ grave goods are those of El Carpio de Tajo (Toledo) and Duratón (Segovia), both dated mostly into the sixth centuries. Because both sites were excavated in the first half of the 20th century (Molinero Pérez 1948; Ripoll López 1985), there is not that much information about the gender and age of the bodies related to the finds (the relevance of this will be discussed later in this chapter). These necropoleis were both quite large (over 250 burials), and of the ‘communal cemetery’ type described in Chapter 6. The bodies were mostly orientated East-West, buried simply in dug-out graves, although some had stone linings,

Christianization and Germanization: New evidence for current debates

141

and the bodies seem to have been inhumed dressed and, possibly, shrouded. Finds-wise, both necropoleis follow the same pattern, in which most of the ‘Visigothic’ finds were located in a few burials, with the rest lacking any of these lavish grave goods. Also dated to the fifth century is a group of singular objects related to the ‘Germanic’ arrival and their influence in Hispanic culture. Particularly significant are the examples of the sword and belt buckles found in Beja, Baamorto, the necklace from Beiral, the knife fragment from Barcelona, or the necklace from Albaicín. Most of them are singular objects found without a clear archaeological context (López Quiroga 2004b), which makes their interpretation in social terms very difficult but they are clearly related to elite agencies during the late fifth century. Sors, tertiae, conquest, and settlement Far more important for the transformation of Late Antique Iberia than the invasion was the interaction between these new populations and the local communities and, ultimately the emergence of two independent polities within the Roman territory. These two factors are what mainly triggered the long-term social and political transformations that would define the post-Roman period, intersecting with other processes like the development of Christianity which has already been discussed. The patterns of settlement and the nature of accommodation are different between the Visigoths and the Sueves, but both developed a system by which the late Roman administration continued to function – with the main difference being that it was under new (Germanic) management (Kulikowski 2004: 256-7, 274) – without any evidence for major change until the mid-sixth century. This is an oversimplification of the overall process, about which little is known. It is not explicitly explained by the available sources, which makes all interpretations of the settlement a patchwork of educated guesses (Wood 1998). The Sueves, together with the Vandals and the Alans, entered the Iberian peninsula in 409 by invitation of the usurper Gerontius, after having been raiding in Gaul, and by 411 they had reached an agreement to divide the peninsula amongst them. This was soon complicated by the arrival of the Visigoths, who were sent by Ravenna to fight the usurper and his Barbarian supporters. After this, and for various other reasons (Heather 1998: 514-5; Wood 1998), the Visigoths were called back to Gaul (where they had been after the sack of Rome in 410). It is there that the settlement treaty was officially signed between the Romans and the Visigoths. This was the treaty or foedus of Wallia of 418.

142 

The Iberian Peninsul a bet ween 300 and 850

By this foedus, the Goths were allowed to settle in the region between Toulouse and the Atlantic, along the Garonne valley, according to the terms of Roman hospitalitas. Hospitalitas was a legal term which designated the hosting in Roman territory of federate troops who would receive housing and a third of the land of the host (Goffart 1980: 42, 45-50). After the original settlement, the hospitalitas turned into a subsidy paid to the Goths (Goffart 1980: 110, 116; Jiménez Garnica 1999: 97; cf. Heather 1998: 514-5). Over time the public revenues which were surrendered to the Goths were divided into thirds, two of which would become the Gothica sors (Burns 1992: 362-3; Halm 1998: 47-8) assigned to the Visigothic settlers as a substitute for the payment they received as Roman soldiers, while the remaining portion was the tertia, the share which went to the King (who represented the emperors’ interests as King of a federate gens; Goffart 1980: 123; cf. Wood 1998: 522-5). In this way, the Gothic settlement in Aquitaine was not linked to land ownership but rather to revenue control (Collins 2005: 29; Goffart 1980: 208; Jiménez Garnica 1999: 100-1). Regarding the Sueves, the settlement appears to have been even more complex than that of the Visigoths, as it seems that they took over Roman territory without reaching an agreement with the Imperial authorities. As a result of this, they found themselves in constant struggle against the Roman administration, the local aristocrats, and the Visigoths, who were constantly sent into Spain to fight against them (Ariño and Díaz 2014: 180-3; Wood 1998). The conflict between the Sueves and the locals led to a process of understanding and, ultimately, of collaboration, especially as the Roman central administration abandoned all Hispanic business to the Visigoths. In the face of this ‘betrayal’, the already settled Sueves seemed a better alternative for peace and order than the then invading Visigoths (Díaz Martínez 2011: 41). By this stage, it seems, from what can be interpreted from the sources, that the Sueves had already reached a settlement agreement with the local elites, probably including the allocation of sortes and tertiae to the Sueves, as if it were a foedus, but without the sanction of the Roman authorities or the necessary counterpart: the providing of foederati troops (Díaz Martínez 2011: 58-62). This established agreement may have later been approved by the Romans, but this did not occur until the 430s, and it may have at first involved the redistribution of taxes rather than of land, combined with billeting (Wood 1998: 523-4). In these early stages, it was only the Sueves who were really settled in the Iberian peninsula because the Vandals and Alans were expelled from Hispania by the Visigoths in the late 420s and because the Visigothic interests were above all centred on their Gaulish lands. Nothing suggests that the terms of the Visigothic foedus settlement were applied in Spain, even if we

Christianization and Germanization: New evidence for current debates

143

know from sixth-century laws that there were then sortes and tertiae. The Visigothic ‘conquest’ of Hispania in the fifth century was most probably just the occupation of the main towns after their successful campaigns against the Sueves (Kulikowski 2004: 205; Wolfram 2005: 153-4), with effective control only coming in the last decades of the century. This effective control was the result of the efforts of King Euric, who ultimately broke the foedus with the Romans and began to take lands and towns in Spain, prompting the resistance of the local nobles, who probably saw their lands and estates taken, as opposed to the Visigothic settlement of hospitalitas in Aquitaine (Wolfram 2005: 155). This late interest in Spain is mentioned in two notes to the Consularia or Chronica Caesaraugustana which read as follows for the years 494 and 497: Goti Hispanias ingressi sunt and Gotthi intra Hispanias sedes acceperunt. This has normally been taken as a sign of Goths migrating from Gaul in large numbers on the eve of Vouillé, but these entries simply describe a military conquest without necessarily substantial accompanying population movement (as the use of ingredior indicates in the rest of the Consularia;41 Kulikowski 2004: 208). The Visigoths had been trying to impose their control over the peninsula for the previous seventy years (Koch 2006: 88-97; Kulikowski 2004: 179-86), but by the 490s the Visigoths could still not exercise their rule over Iberian towns. The Visigoths may have taken over the administration through their military power after the definitive departure of the legions in 410 (Collins 2005: 19), 42 but this could only apply to those towns which were key to the Roman administration, like Mérida (Kulikowski 2004: 175). Most areas seem to have continued in a state of semi-independence from both Barbarians and the central imperial administration. Visigothic interest in Spain became a necessity after Alaric II was defeated at the battle of Vouillé in 507. In this battle Clovis not only killed the Visigoth king, but also took over a large portion of the kingdom, from the Loire to the Pyrenees, except for Mediterranean Gaul (Septimania) which remained a Visigothic province (DLH II.37; HG 36). At this stage, and after over two generations of settlement in Aquitaine, the Frankish conquest of the lands originally granted may not have caused a large population shift. By this time, the ‘Visigothic population’ in Spain, disregarding estimates of numbers, must have been limited to a few garrisons, augmented after Vouillé by the arrival of the royal retinue, the administrative elites, the army, and the large landowners who had their estates confiscated by the Franks (and their families; cf. Arce 41 For instance, s.a. 506: Dertosa a Gotthis ingressa est. 42 Cf. the persistence of imperial (non-provincial) armies according to the Epistula Honorii.

144 

The Iberian Peninsul a bet ween 300 and 850

Martínez 2010: 76-7). Most of the middle and lower classes may well have stayed in Gaul, and this might have underlined the identification of ‘Goth’ with soldier (and not with a specific ethnicity). This migration from Gaul, furthermore, must tell us that the terms of the hospitalitas only related to tax revenues, as it would make little sense for them to move after the battle if they actually owned the land. Identity, ethnicity, and archaeology This arrival and settlement of Visigoths and Sueves in the Iberian peninsula had an impact on the material culture. There is a general consensus amongst scholars that these peoples must have caused some sort of change in the archaeological record, and while it is safe to assume that something happened (the Visigoths and the Sueves existed, after all, even though it would be difficult to assess exactly what did it mean in terms of identity), it is far more complicated to explain what and why. As mentioned in the Introduction, the study of the Visigoths has been of paramount importance in traditional historiography. The search for the ‘first Spaniards’ and the ‘origins of Spain’ required a material culture that could reflect the arrival of the Visigoths, which signified the emergence of the first Spanish state. In traditional 20th-century nationalistic fashion, this was done mostly by looking at Late Antique and Early Medieval grave goods that could show ethnic markers of the Visigothic settlers (e.g., Zeiss 1934; cf. a British example: Clarke 1975). This tradition was well established amongst Spanish scholars and it has only been slowly de-constructed in recent years (Tejerizo García 2011; Quirós and VigilEscalera 2011). This ‘ethnic approach’ to these new types of Late Antique grave goods has many flaws and problems, and is the constant cause of scholarly argument (not only in Spain, but across Europe: Williams and Sayer 2009; Halsall 2011). As we have seen, it is more probable that the Visigoths settled themselves in the main cities and their surroundings in order to exercise their control over Hispania, but the large majority of ‘Visigothic’ grave goods have been found in rural burial contexts. The Visigoths were not a homogenous group of ‘Germanic warriors’ and nor were the Sueves (Díaz Martínez 2011). The Visigoths in particular were a mixed group formed by various groups of Goths (mostly Tervingi and Greuthungi), other barbarian tribes (not all of them Germanic), army deserters, and unhappy provincials which were ruled together by Alaric and ultimately settled in Gaul (Heather 1998). In these very early stages, it is doubtful that there was anything resembling a homogenous Visigothic material culture that could be identified in the archaeological record. In fact, the only traditionally identifiable

Christianization and Germanization: New evidence for current debates

145

‘Visigothic’ material culture corresponds to cloisonné jewellery found in tombs. Most of these grave finds are actually quite late in date (e.g. the famous cloisonné buckles and eagle brooches are usually sixth or seventh century in date; cf. Barroso et al. 2010), 43 and would not reflect the patterns of the ‘original’ fifth-century Visigoths. Furthermore, we know that the Visigoths did not usually bury themselves with grave goods, at least in Gaul, and it is strange that they should suddenly begin to do so in the sixth century in Spain (Koch 2006: 84-7; Halsall 2011: 16). Similarly, the overall geographical distribution of ‘Visigothic necropoleis’ in the Peninsula, where these grave goods have been found (Ebel-Zepezauer 2000; Kulikowski 2004: 268-9), overlaps the pre-existing ‘Duero valley necropoleis’ (Halsall 2007: 342-4). It can be argued that there is enough comparative ground to claim that these grave goods (and the material culture in general) are not ethnic markers, but rather symbols of rank and status within a community (Halsall 1995b; 2011; Tejerizo García 2011). This is especially true when considering that burial goods are ‘retrieved culture’: we get the impression of what those who buried their dead wanted to give, the intention of those burying and not of the buried one (Härke 1997: 25). Lastly, this argument is fallacious because it is based in at least two assumptions: firstly, that the Germanic migrant communities were necessarily concerned with a group identity; and secondly, that funerary practice was an appropriate context for asserting such a group identity (cf. Ucko 1969). Regardless of whether Late Antique grave goods show ‘Danubian’ or ‘Germanic’ art traditions or not, it is more interesting to understand the objects in their social context and what can we learn about the structure of the society which buried their dead with such ornaments than try to use them to find ethnic origins of the people buried with them. Cultural anthropology gives a different perspective on these grave goods: it may be possible to link the presence of certain grave goods to new ways of redefining status within a community. Indicators of high status could perhaps be used to explain ethnicity, although such an extrapolation from rank to ethnicity is not fully valid. Indicators of social structure? Considering the situation of towns and of the rural countryside in the fifth century, and the way in which the Visigoths and the Sueves settled and 43 Which corresponds to the period when ‘Visigothic’ as an ethnic identity was slowly blurring away:) and it was adopting a new meaning of ‘citizen of the Visigothic kingdom’ (Buchberger 2013).

146 

The Iberian Peninsul a bet ween 300 and 850

took over the provinces of Hispania, it is clear that there was a situation of political instability, a power vacuum in which the social fabric was being redefined. As shown in Chapters 2 and 3, this is more the case for the rural populations than the urban ones. Therefore, the redefinition of the fifth and sixth century rural social structures may be a better explanation for the new types of burial practices than the more simplistic ethnic approach. This does not mean that the Visigoths and Sueves must have had a role in this, but simply that they had become a new element to consider. Following Steuer’s theory (Steuer 1989; cf. Hodder 1983: 150 and LéviStrauss 1993: 278-9), social structure is defined as the group of rules that give meaning to social relationships, and can be understood together with other associated characteristics (i.e., not the social pyramid). These rules regulate the formal aspects of social interaction as it emerges from social relationships, being in this way cause and consequence. This means that the social structure controls the interaction of the individuals, which can be subject to rank and position. Furthermore, social structure can be seen at three different levels: kinship, social organization and social stratification (Lévi-Strauss 1993: 312). These rules are something that could hardly be represented in the archaeological record, unless we try to interpret them from rank. This social rank does not relate to the legal status (legal categories in law codes are not necessarily social classes: James 1988: 216-8), but to a wider range of possibilities, including age, gender, and economic position. Furthermore, for the Iberian case studies, and if we analyse the grave goods within their context (the body, the other grave finds and the cemeteries as a whole) we will be able to obtain a deeper analysis of the community linked to the burials than by examining the objects in isolation. These studies have been done in Gaulish and German examples, and there is no reason why they should not be valid for Hispania. As we have seen in the previous chapter, by the beginning of the sixth century burial practices in rural areas had already shifted to east-west row cemeteries, similar to what in other parts of the Empire are called Reiengräberfelder, equally of provincial Roman origin (cf. James 1979: 71-5; 1988: 45). One of the common elements of these burial areas is the concentration of the most luxurious grave goods in a few burials, the ones that can be considered to be those of the local elites, around which burials seem to cluster. The development of clusters can be linked to strong family groups or other forms of social grouping (retinues?), showing variations in rank, and these were essential parts of the newly forming community. Other indications of rank were age and gender, which can be identified through osteology, and then linked to the correlation between age, gender, and grave goods. The grave goods

Christianization and Germanization: New evidence for current debates

147

themselves have a symbolic value, which could derive from their nature as a gift, a ‘relic’, or an indicator of age or gender. And as has already been mentioned earlier, they were deposited with intentionality and display, which in the late Roman period has to be related to the redefinition of social ranks and status (Halsall 1995a: 66-7; 1996a: 14-6; Härke 1997: 25; James 1988: 224). Overall, the presence of new types of burials with so-called ‘Germanic’ grave goods in the transition between the fifth and sixth centuries in the cemeteries of the Iberian peninsula is most likely an indicator of a society in transition. The dismantlement of the Imperial system and the collapse of Rome were felt by local populations, and in the interim power vacuum there was a need to re-define the social structure. One of the ways in which society was redefined was by means of display during burial. Where are the Visigoths then? All things considered, it seems difficult to put together the arrival of the Barbarians and the apparent necessity of some historians to identify and label these people in the archaeological record, as if pin-pointing the Germanic populations to certain areas or items validated the fact that they ever existed. This is true not only in the context of early-20th century historiography, because it is a valid point to ask, especially when compared with the deep transformations noticeable in the early Islamic period (Chapter 8), where the arrival and settlement of the Muslim invaders can be seen in burial practices, technology, and settlement patterns. First of all, the terms of the settlement are very confusing, and do not clearly point at the distribution of land (at least during the fifth century), so it would be difficult to assign not just the presence of ‘Germanic’ grave goods systematically distributed across necropoleis of the Iberian peninsula, but also a specific type or area of settlement. Furthermore, the comparison between Islamic and Barbarian settlement is not valid because the social structure of both groups was very different. While the Islamic settlers had a tribal system (see Chapter 8), the Germanic peoples had a system which was far more open and fluid, based on individual prestige and patronage (Heather 1988). In fact, if the settlement rested on the basis of allocation of rent and revenues, it could even be that rather than establishing and settling in isolated and different (new) sites, smaller groups could have been distributed through various estates and settlements. Even in this situation, traditional markers such as grave goods or pottery technology cannot be used to identify Germanic settlers. In the former case, as explained earlier, because these became status and rank markers and the form of the

148 

The Iberian Peninsul a bet ween 300 and 850

objects and the burial practice derived ultimately from Roman military traditions. In the latter (cf. Chapter 7), it is impossible to tell if there were any Visigothic or Suevic pottery traditions, because most of their material culture appears to have been heavily Romanized. Perhaps the study of place names could be an indicator of Germanic settlement, but this is a problematic approach because it would be nearly impossible to differentiate a Late Antique Germanic original settlement from settlements in later (Visigothic and post-Visigothic) periods. While on the one hand it is difficult to see isolated groups of Germans settling on their own, on the other it is equally difficult to assume that the Barbarians simply took over as the new elite group. The settlement of the Visigoths and the Sueves was not necessarily the substitution of old elites, but the introduction of a new elite element, which could be identif ied through its heavily militarized Roman elite markers. Therefore, the search for the Visigoths has not to be one of new settlers with a distinctive material culture (which would be impossible to distinguish from the Roman one), but it should rather be a search for a new way of elite display and a redefinition of rank and status at a time (especially in the fifth century), when these were in constant negotiation between the old (Roman) and the new (Germanic) elites. After this point, Roman and Gothic ceased to be antagonistic terms in the local Hispanic context, because Roman as a political label no longer served any purpose. From then on all subjects within the Gothic kingdom could be seen as ‘Goths’: regardless of language or ethnicity. From the study of the sources (Buchberger 2013) it is clear that being a ‘Visigoth’ was to be a Christian inhabitant of the kingdom. The development of Christianity as a new ideological and political phenomenon and the ‘Germanization’ of society through the introduction of new sets of populations into the Iberian peninsula can be considered two of the major issues present during the fourth and fifth centuries. Each of them, as we have seen, had different effects in the materiality of Late Antique societies. For instance, Christianity, at least in its earliest phase, was above all related to elites and urban life, while the influence of Germanic traditions is mainly seen in the rural world. Christianity and Late Antique archaeology are intimately related, especially as the study of the late Roman period began, from an archaeological perspective, as an attempt to identify the first traces of Christianity. Christianity did not impose a cultural disruption, and as we have tried to explain, because Christian objects were formally (even if not semantically)

Christianization and Germanization: New evidence for current debates

149

Roman. Christianity has to be understood as part of a new late Roman identity, and as such it has to be studied beyond churches and burial practice formalities. Thus the materiality described which relates to the development of Christianity in the Iberian peninsula, like shrines, epigraphs or ritual elements, should be seen as the mediators of a context of deep ideological change. This was both as an expression of increasing top-to-bottom ideological imposition and also the bottom-to-top resilience by elites and local communities to this new set of habitus. In recent decades, there has been a very substantial shift in the understanding of the so-called ‘Barbarian invasions’, leading to a more complex scenario where these phenomena are seen more as a dialectal process and not as a deterministic and unilineal one. There is no doubt that the introduction of new sets of populations and, more importantly, of new elites and their related agencies, provoked major changes in the archaeological record. The appearance throughout the peninsula of ‘Germanic-style’ objects is one of the most important consequences of this process. However, the traditionally asserted direct connections between objects and rigid ethnic identities should be complicated as a more complex picture where these objects play a major role in defining identities and status in specific historical contexts. Christianity and the Barbarian invasions and the emergence of a new form of polity within the ancient Roman territory in the Iberian peninsula were two of the key elements that would define the constitution of post-Roman societies and their materialities during the following centuries. This is the subject we will tackle in Part 2.

Part 2 The post-Roman period

5

Towns and cities under Christian prevalence

The late Roman town, with its urbanism defined by the changes of the fourth century, continued to evolve along the same lines as described in previous chapters. However, following the barbarian invasions, there was an increase of new monumentality and of new constructions. This occurred in two different waves: firstly, during the late fifth century in those Roman towns which had become bishoprics (such as Mérida, Tarragona and Terrasa, among others); later, during the late sixth century, another input developed in the new capitals and urban foundations (above all Toledo, Reccopolis, El Tolmo and Cartagena), which emerged quickly together with other Roman centres of power. This was a consequence of both the development of new, stronger states (as happened in the Visigothic and Suevic kingdoms and in the Byzantine province) and of the consolidation of bishops as the new urban leaders in this new and fragmentary political framework. In archaeological terms, the development of towns in the post-Roman period is both the story of the creation of a properly Christian, Late Antique urbanism and (to a lesser extent) the creation of a ‘post-Roman’ type of town. Both processes of transformation from the early Roman pagan model were, however, clearly and evidently conditioned by the late Roman urbanism that had developed in earlier centuries. In this chapter we will explore the nature of this new post-Roman urbanism, from three perspectives: firstly, we will look briefly at the state of towns at the beginning of the post-Roman period, before moving on to explore the new ecclesiastical monumentality; finally we will look at the urban changes prompted by the political developments of the late sixth century.

The late Roman urban legacy in the post-Roman world Fifth-century urbanism is characterized by the appearance of the first episcopal complexes, which from the second half of the century had a great impact on towns. Nevertheless, there were no new architectural developments worth highlighting in these constructions, other than the continuation of the long-term processes of urban and technological change which had begun in the late Roman period – processes that would continue to develop through the sixth and seventh centuries. Urbanism slowly continued to

154 

The Iberian Peninsul a bet ween 300 and 850

evolve, but without the restrictions (or impetus) of the late Roman political system. The nominal control of the Visigoths and the Sueves of the peninsula which has been described earlier did nothing to palliate what was essentially a power vacuum in terms of the territorial administration. This preservation of Roman urbanism, understood as town life (and not simply as ‘life in towns’), varied greatly according to the relative importance each town had in this period of political uncertainty. Therefore, in those towns which still functioned as central places (Martínez and Tejerizo 2015), town life in the late Roman fashion is still noticeable whereas in the rest, town life finally came to an end – or was transformed in such a way that is difficult to interpret as such through the archaeological record (Sastre de Diego 2005). This dichotomy perhaps reflects the true nature of urbanization in Roman Hispania. The urbanization of large parts of the north and west of Iberia seems to have been a Roman construct, linked directly to Roman territorial administration, whereas on the Mediterranean coast urbanism pre-dated the arrival of the Romans, and territorial administration was organised around existing nuclei. End of Roman townscapes In terms of pure civil urbanism, there were no major constructions, infrastructures or remarkable additions to the late Roman urban pattern during the late fifth and early sixth centuries. On this issue it is perhaps noteworthy that it is during this period of power vacuum when many urban sites seem to have finally lost their urban character. For instance, the development of suburbs continued, to such extremes that in some cases this suburbanization led to the development of poli-nucleated settlements and, ultimately, to the dissolution of the urban fabric. This was usually accompanied by the ‘ruralization’ of former urban areas and the consolidation of Christian landscapes. Complutum, for instance, is a good example of the dissolution of the urban fabric, as by the sixth century the town centre had completely disappeared, and the three different population clusters (around the Christian shrine, around the villa of el Val and in the old town centre) barely resembled a single functional town, and there is no evidence for any hint of town life (Rascón and Sánchez 2008), even if it was still nominally a bishopric. In Tarragona (Carreté et al. 1995; Macias and Muñoz 2013; Panzram 2002: 111-20), the final abandonment of the larger part of the lower city (the circus, located in an intermediate level, was abandoned around the third quarter of the fifth century), led to the existence of two clear and separated urban clusters,

Towns and cities under Christian prevalence

155

one in the upper town, where the cathedral complex developed later in the sixth century, and the other clustered around the ancient port and the Francolí Christian area, a martyrial complex that arose over a Roman suburb (figure 2.10). In the case of Barcelona, the settlement was reduced to just one quarter of the walled enclosure, located close to a corner of the town walls. There, the cathedral became the new central point (Beltrán de Heredia 2002a). Such urban reduction and concentration around a cathedral complex seems to be one of the characteristics of Late Antique Hispanic towns. According to the archaeological works carried out at La Almoina something similar happened in Valencia (Ribera Lacomba 2005). Most secondary nuclei in the north-western half of the peninsula ceased to produce any identifiable urban remains. That is the case of Clunia, the ancient capital of the Conventus Cluniancesis which shows an incomplete and minimum of activity from the fourth century, far from the idea of urban life that, in any case, did not last beyond the fifth century. Its public buildings at the forum were abandoned at this point; the last coins found there belong to Theodosius. Nevertheless, a seventh-century necropolis shows that some kind of settlement survived on the site (de Palol Salellas 1994: 74; de Palol and Guitart 2000). The town of Veleia, in the Basque plain, is archaeologically well known, and shows a similar pattern. Veleia was a Roman fortified oppidum, linked to a military garrison and mentioned in the Notitia (figure 5.1), 44 which was fortified during the course of the fourth century. However, according to the early archaeologists (Elorza Guinea 1972: 185-91; Nieto Gallo 1958: 226; cf. Filloy and Gil 2007: 472), it appears to have been depopulated by the 450s, once the garrison left for good during the early fifth century. It had been thought that the local Basque population might have remained resident there once the garrison left, according to various ostraka written in proto-Basque, but these inscriptions (some in Basque, some of Christian content, even one which mentioned queen Nefertiti) were later proven to have been a hoax (Elkin 2009; Lakarra Andrinua 2008). The same process of depopulation could be said about other ancient Roman places located on the plateau (Martínez and Tejerizo 2015). One is Tiermes, where a nearby small church with Visigothic sculpture (Caballero Zoreda 1984) and some rock-cut tombs are the only evidence to suggest any sort of Late Antique habitation, possibly still inside the late Roman walls. In Uxama the late Roman walls had reduced the urban perimeter to almost a half (García Merino 2000). Salamanca and Cauca may have suffered similar evolutions (although the last two are mentioned in the Visigothic 44 Occ. XLII: In provincia Tarraconensi: Tribunus cohortis primae Gallicae, Veleia.

156 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.1 Photograph of the walls of Veleia, showing the construction technique used, with large blocks of stone and reused column drums

Wikimedia Commons

Councils, as the expansion of the Visigothic kingdom further prompted the establishment of bishoprics in old urban centres). Part of this northern-central region was the area which was most hotly contested during the wars of the fifth century, especially between the Sueves and the Visigoths. This insecurity, together with the superficial success in this area of towns as central places, provoked the rise of new elite settlements on higher ground in fortified hilltops, usually called castra – a phenomenon which will be further explored in Chapter 6. Apparently, something similar to what occurred in the rest of the northern plateau appears to happen in other towns such as Segovia and Astorga. Nevertheless, in these cases, even if the evidence is very partial, there is at least one Late Antique church, with its original liturgical elements, built in each town (Sastre de Diego 2013): Saint Martha’s in Astorga, placed close to the Medieval cathedral, and Saint John of the Knights in Segovia, a Romanesque church built over the architectural remains of at least two previous basilicas. Both Saint Martha’s and Saint John of the Knights were built inside the city walls. In the case of Segovia, this scarce evidence of a Late Antique town is

Towns and cities under Christian prevalence

157

supported by contemporary written sources, which mention the episcopal see (Sastre de Diego 2005). Ávila and Salamanca were also bishoprics in Visigothic times, but the archaeology is still too scant in evidence for this period. All this would show, even if it is still little evidenced, the continuation of some kind of urban activity and organization by the occupation of a well defended part of the Roman towns. It should also be acknowledged there have been far fewer excavations carried out in the current Castilian cities in contrast with Mediterranean sites, which have enjoyed an enviable level of archaeological activity in the last decade of the 20th century. Besides this, we must also consider that, overall, Roman town life was directly linked to political and economic centrality, and consequently towns in peripheral areas in this period suffered a final blow to their urban status. Furthermore, the preservation of town life and Roman urbanism in only a few sites is mostly what determined the development of a new urbanism and a new monumentality throughout the sixth century. Town continuity Four main archaeological indicators can be used to highlight the continuity of urban life into the sixth century: the presence of urban elites (not only the curiales, but also senators, inlustres, clarissimi and counts and dukes which appear in sources and inscriptions), the continuity of broad trading links, the existence and maintenance of public monuments, and the presence of a nucleated dense habitat. These indicators of late Roman urban life and urban elites can be found mostly in towns on the Mediterranean coast such as Valencia and Barcelona, and in the Guadalquivir valley, for example in Córdoba and Seville. Besides these, some exceptional inland sites such as Mérida or Mértola also show such indicators. Urban elites, which used to be identifiable through their large urban and suburban dwellings cease to be so easily recognisable, as there are very few examples of aristocratic housing from the late sixth century. The large domus of La Encarnación in Seville (García Vargas 2012: 900-5), the domūs of Bisbe Caçador and of Pati Llimona in Barcelona (Martín et al. 2000: 283; Miró and Puig 2000: 175; Oliver and Riu 1987) or the large house excavated at Hernán Cortés in Mérida (Sanabria et al. 2012) do not show signs of such late continuity. In fact, one of the transformations that can be observed in the large aristocratic houses of the late Roman period is that by the sixth century they seem to have been transformed into multi-family dwellings. The excavations of the domus of the Marbles in Mérida (Alba Calzado 2005: 131-42) have shown how the peristyle was turned into a communal courtyard,

158 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.2 Plan of the Visigothic phase of the Domus of the Marbles, in Mérida. The old domus has been subdivided into various single-family dwellings around a communal courtyard

Martínez Jiménez 2013: figure 4, based on Alba and Mateos 2008, figure 9

and the various rooms facing it arranged in groups of two or three so as to house various families (figure 5.2). As mentioned earlier, civic munificence, another clear indicator of the late Roman curiales, died out throughout the fourth century, and by the fifth it had mostly vanished. By the sixth century, private donations were rare enough as to be remarkable when identified, since donations were largely given by bishops, and, to a lesser extent, royal and civil officials. One of the very last recorded examples of such activity is the reconstruction of the bridge in Mérida. This took place in the year 483, according to a now-lost inscription of complex interpretation (ICERV 363 = IHC 23a): Solberat antiquas moles ruinosa vetustas, lapsum et senio ruptum pendebat opus. Perdierat usum suspense via p(er) amnem et liberum pontis casus negabat iter. Nunc tempore potentis Getarum Ervigi [sic: Eurici] regis, quo

Towns and cities under Christian prevalence

159

deditas sibi precepit excoli terras, studuit magnanimus factis extendere n(o)m(e)n, veterum et titulis addidit Salla suum. Nam postquam eximiis nobabit moenib(us) urbem, hoc magis miraculum patrare non destitit. Construxit arcos penitus fundabit in undis et mirum auctoris imitans vicit opus. Necnon patri(a)e tantum crare munimem sumi sacerdotis Zenonis suasit amor. Urbs Augusta [Emerita] felix mansura p(er) s(ae)c(u)la longa nobate studio ducis et pontificis. [A]era DXXI Old age had dissolved the old ruinous structures, and the fabric hung broken due to the passing of years. The suspended road over the river had lost its usefulness, and the fallen bridge prevented free transit. Now in the times of the powerful king of the Goths Euric, in which he dedicated himself to care for the lands [he had] received [deditas terras – Hispania/ Mérida?], the magnificent (Euric) was eager to expand his name through his works. And Salla added his illustrious [veterum] name to the inscriptions [titulis]. Then, once he renewed the city with remarkable walls he did not stop completing wonders [for the city]. He built the arches [of the bridge], dug the foundations into the waves, and though he was emulating [the original builder], he surpassed the miraculous work of him who designed it. Bishop Zeno’s love towards his country persuaded him [Salla] not the least to create such defences. The city of Mérida will remain happy for long centuries, with the renewal dedication of [its] duke and bishop. In the [Hispanic] era 521 (= 483 AD).

The inscription was compiled, together with other Visigothic inscriptions, in an early medieval codex (Códice de Azagra), and includes a mistake in the transcription of the name of the king, Ervigius instead of Euricus. This is the main reason why some scholars doubt its originality. However, the historical context described is coherent and matches the date and the political situation of Mérida in that period. Moreover, the two principal persons mentioned in the inscription, Duke Salla and Bishop Zeno, are two contemporary historically relevant people of the second half of the fifth century. The inscription also mentions the reconstruction of the walls, which was perhaps a partial repair of the expanded wall enclosure discussed above (Sastre de Diego 2015). Moreover, there is evidence for late repairs in the bridge, especially in its first section, although they are impossible to date. What has to be emphasized is that the bridge was repaired by the local civic leaders: the Bishop and the Duke (Arce Martínez 2008). Beyond this unique case of specific munificence, the regular maintenance of aqueducts is also a good indicator of the continuity of Roman elites and town life. As mentioned in Chapter 2, aqueduct continuity was linked to main

160 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.3 Photograph of one of the pillars of the aqueduct of Los Milagros, in Mérida, apparently restored without success in the Visigothic period

cities, and by the sixth century only Tarragona, Barcelona, Valencia, Córdoba, Seville and Mérida seem to have preserved their aqueducts. This continuity is most probably the consequence of constant maintenance rather than major reconstructions. In fact, in Mérida there was an attempt to repair one of the aqueducts, but there is no evidence that this was successful (figure 5.3). Beyond these main cities, smaller or minor towns, but with an evident economic relevance, such as Almuñécar, Lisbon or even Ceuta, also seem to have preserved their aqueduct supply at least into the sixth century (Martínez Jiménez 2014). Regarding non-elite housing and dwelling spaces, the post-Roman period saw the confirmation and further development of the patterns which had

Towns and cities under Christian prevalence

161

emerged in the late Roman period. Even if housing trends vary from site to site, it is possible to describe general patterns, based on the evolution of houses in sites such as La Encarnación in Seville, Reccopolis, Eio, the theatre quarter of Cartagena, the late Roman housing of Morería (in Mérida), the insula das Carvalheiras in Braga, the late phases of the domūs of Barcelona, etc. These are transformations that can be seen both in the evolution of earlier Roman houses and in newly built houses. Structurally, post-Roman urban housing tends to be smaller, both in newly built dwellings and in those resulting of the subdivision of earlier structures. The design is quite simple, with usually few (two or three) interconnected rooms, which are multi-purpose. Houses became smaller, and when possible arranged around a communal courtyard (usually with rubbish pits and wells) shared with other houses, thus creating smaller neighbour communities. Floor surfaces are usually of beaten earth, sometimes with lime mixed in for further consistency, but paved surfaces were largely abandoned. Hearths (used as indicators of a single family dwelling) on the clay floors, or on top of a set of bricks, became widespread elements.

The consolidation of a Christian monumentality Whereas in Rome, Constantinople and other main Christian centres, such as Jerusalem or Ephesus (Foss 1979; Krautheimer 1983), large public displays of Christian monumentality were common already in the late fourth and early f ifth century, this was not the case in the Iberian peninsula. In fact, in most of the West, large Christian urban monuments are rare until the late fifth century (Esmonde Cleary 2013: 150-81; Sastre de Diego 2013). We have already explored these early forms of urban Christianity in an earlier chapter, but it is important to understand why and how a new monumentality developed in the late f ifth and, especially, during the sixth century. This new monumentality was characterised by large Christian buildings in the central areas of urban nuclei, as opposed to suburban complexes. Archaeologically this can be seen in the construction of cathedrals and episcopal complexes, which are large built areas where these main buildings were concentrated. Lastly, it has to be noted that the development of these new complexes is a direct consequence of the elevation of bishops to local leaders, replacing old civic and political elites at an urban level (thanks in part to their alliance with the Visigothic elites during the period of state formation), and this new power required a new architecture.

162 

The Iberian Peninsul a bet ween 300 and 850

Most examples of these early intramural churches are known through inscriptions, such as those of Bishop Justinian in Valencia (IHC 409 = ICERV 279; CIL II2 14.90), Saint Mary’s in Mérida (Sastre de Diego 2011: 29) or the doubtful seventh century churches of Córdoba (CIL II2 7.640, 7.638; Hidalgo Prieto 2005: 407, 467; cf. Arce Sainz 2015). Archaeologically, the most famous one is perhaps the amphitheatre basilica of Tarragona (Dupré Raventós 1990: 206-34), which commemorated a local martyrdom in the arena by turning it into a site of Christian worship. 45 Nevertheless, this martyrial church did not belong to the episcopal complex of Tarragona, which was placed in the upper-town and became monumentalized at the beginning of the sixth century (Macias and Muñoz 2013). A little earlier, around 450, the monumental construction of the cathedral complex of Egara was built (García et al. 2009; Moro and Tuset 2001). These new churches were built reusing old monuments or public spaces, which should not be seen as an active way of ‘Christianizing’ the landscape or deleting the ‘pagan’ past, but rather as a reoccupation of redundant or disused areas. This occupation could also perhaps be linked to an idea of the occupation of a legitimizing location (i.e., the forum as the old centre of municipal power). In the West, episcopal complexes normally include a cathedral, a baptistery, and the bishop’s palace, but other additional buildings, such as chapels, martyrial shrines, bath complexes and even burial grounds can also be found (cf. Rapp 2005: 208-11). Architectonically (Caballero and Sastre 2014: esp. 264), churches were simple: unanimously basilical in shape, either single or double apsed, following the same late Roman design. The space for worship is usually arranged in a single axis with three parts, each separate by chancels: an aula for the people, an enclosed choir, and the eucharistic sanctuary were the altar (only one and separated from the wall) was located. This liturgical separation of spaces is confirmed by the canons of the Hispanic Visigothic Councils. The altars were usually a marble slab on a central ara (in imitation of pagan Roman arae) alone or/and four corner columns (Sastre de Diego 2013). Relics were placed into the ara -loculus- and covered by the table. This action, led by the bishop, meant that such space was sacred as a church and it could be used for worship. There are many typologies of baptistery: cross-shaped as in Torre de Palma, polygonal and stepped like in Mértola, Terrasa and Barcelona, or rectangular with steps like in Casa Herrera and San Pedro de Mérida; but they all fulfilled the same purpose, as baptism was achieved by immersion. Sometimes the baptistery was located in a small annexe basilica inside a larger complex, usually by 45 NB: The amphitheatre in Tarragona is not inside the walls though.

Towns and cities under Christian prevalence

163

the cathedral. Episcopal palaces were, from an archaeological point of view, the reflection of the power of bishops, and have been described as an evolution of aristocratic housing (Miller 2000: 13-20). The architectonic cohesion of episcopal complexes was loose, but this was tightened as time passed and bishops grew more powerful. In many cases, the lines between residences, spaces of worship and other administrative buildings were blurred, reinforcing the position of bishops within their towns as these complexes became new foci of power (Miller 2000: 50). Another urban innovation that developed as a result of the construction of episcopal complexes was the rise of intramural burials. Burials inside city walls are extraordinary (especially rare in the fourth century but more abundant in the fifth), breaking Roman laws that forbade burials inside the pomoerium or sacred boundary of the city. This was a result of the relocation of main Christian cult sites from the suburbs to the city centres and the wide-spread pattern of burials ad sanctos, but they also appear on old abandoned public structures in those sites which show a steep decline of their urban structure, as for instance in Málaga or Granada (Vaquerizo Gil 2010: 87, 221-2). At least two intramural graveyards developed in Mérida, clustered around two churches built near the ancient forum, but other well known cases are Valencia, Braga and Barcelona (Alba and Mateos 2008; Beltrán de Heredia 2008a; Ribera Lacomba 2008). A huge funerary area was also active in the ancient core of Écija between the fifth and the eighth centuries. In Córdoba, the eastern burial area expanded into the abandoned suburb (Sánchez Ramos 2007: 191). In parallel to these developments, there were also instances of the active dismantling of old necropolis. Sometimes this was done as a result of new suburban areas being built on top of burial grounds as in Tarragona (Remollà Valverdú 2004: 90); in the extreme case of Mérida, the grave markers were gathered and reused in the reinforcement of the city walls (Alba Calzado 2004a: 227-30). Eventually, episcopal complexes became the new foci of urban development, altering both the physical and the ideological landscape of the city. Unsurprisingly, the chronological development of these complexes coincides with the period in which bishops emerged as local urban leaders, which is to say in the late fifth and sixth centuries (Christie 2006: 89). In Gaul, for instance, various episcopal complexes fit this description, and mostly are dated to the late fifth century (Esmonde Cleary 2013:174-5). Beyond Gaul, few episcopal complexes are known in the Rhineland (as in Cologne: Ristow 2009), although more are known in North Africa (Leone 2007), all of which share the same chronologies and essential buildings as those found in Hispania.

164 

The Iberian Peninsul a bet ween 300 and 850

In the Iberian Peninsula we have several examples of episcopal complexes known through archaeology: Valencia, Barcelona, Terrasa, Eio, Idanha-avelha, and maybe also Dume. Beyond these, other towns had episcopal complexes, the most relevant are ecclesiastical capitals, such as Mérida, Tarragona, Toledo, Seville; unfortunately, nowadays all of them preserve much more written sources than archaeological evidence. Astorga could also be included as it has a few problematic archaeological remains beneath later Medieval cathedral phases (Sánchez Ramos 2014). The episcopal complex of Barcelona (figure 5.4) has been identified with the monumental post-Roman remains excavated in the north-eastern corner of the Roman city, under the current Plaza del Rey and the Gothic county palace (Beltrán de Heredia 2002a). In its final phase, the complex included not only the late Roman basilica and its fifth-century expansions (a baptistery and a large hall), but also a series of other buildings, including the remains of a possible palace, porches, and courtyards. In the immediate surroundings, a sixth century bath house has also been identified, which could be perhaps linked to the existing complex (García-Entero 2005: 213), as it is known from other examples in the West (Miller 2000: 19-20). The chronology of the buildings is problematic, as only a relative sequence can be proposed, although the excavators’ interpretation suggests a late-sixth and early-seventh century date for the palace extension (Beltrán de Heredia 2008a).46 This development of the episcopal complex would further fit with the period in which the bishops of Barcelona became the key political and religious figures in the region, eclipsing the Archbishop in Tarragona himself. The complex itself occupies several of the old street blocks of the Roman grid, indicating that by this period the street hierarchy had changed, and various minor streets could be privatised, blocked, and built upon. The location of the episcopal complex in this corner of the city is, furthermore, not coincidental, as it is located just at the point in which the aqueducts enter into the city. By building there, the bishops took over one of the prestigious legitimising elements of the Roman past (Martínez Jiménez 2014: 225-8). In Valencia, the remains excavated under the current cathedral at the Plaza de La Almoina point towards the existence of an episcopal complex on top of the old forum from the mid-sixth century onwards (figure 5.5). The 46 An alternative interpretation, however, suggests that the construction techniques and the layout of the episcopal palace could be early Umayyad in date, as it draws clear parallels with the early Umayyad buildings of Mérida (Alba Calzado 2001). This would also fill in the archaeological gap existing in Barcelona, where, despite its ninety years of Umayyad occupation, has yet to produce any Islamic archaeology – an interpretative gap which could be linked to certain historiographic traditions.

Towns and cities under Christian prevalence

165

Figure 5.4 Reconstructed plan of the episcopal complex of Barcelona, with its various phases, indicating the location of the episcopal buildings (hall, cathedral, baptistery, palace, baths) and the ‘count’s palace’

Based on Bonnet and Beltrán de Heredia 2002, figure 2

166 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.5 Reconstructed plan of the episcopal complex of Valencia, at the site of La Almoina

Based on Ribera Lacomba 2008, figure 8

Towns and cities under Christian prevalence

167

reoccupation of the forum has to be understood both as a way of reactivating and preserving the old civic centre and perhaps as an attempt to claim civic legitimacy, at a moment in history when Valencia was controlled directly neither by the Visigoths nor the Byzantines. The episcopal complex is only partly known, but it seems that its core was formed by a mid-sixth century basilica (the apse and various outer walls have been located), to which a cruciform baptistery or chapel and a small cruciform mausoleum were added later in the century (Ribera Lacomba 2005; 2008). The complex was enclosed within the walled-up forum perimeter, and a burial area developed inside. Inside the enclosure was the old public monumental fountain linked to the aqueduct, the nymphaeum, which was preserved in its usage, although now under the ‘protection’ and ‘patronage’ of the bishops (Martínez Jiménez 2011a; 2012). Archaeologically, the complex is only very partially understood, but due to the long stratigraphic sequence of the site it is known that the later Umayyad palace and the current cathedral were built in the immediate environment, so it is possible to claim with certainty that the remains formed the new civic centre of Valencia. Beyond purely episcopal constructions, there are in the Iberian peninsula a number of other urban and peri-urban Christian buildings of sixth century chronology, that perfectly exemplify the expansion of Christian monuments in the urban landscapes. One of these examples is that of Mértola (south Portugal). Mértola had been for many centuries the nexus that connected Mérida with the Mediterranean and Atlantic trading networks, as it was the last oceanic fluvial harbour up the Guadiana river. This economic importance was reflected in the post-Roman period in the development of an extensive Christian monument scheme (Lopes and Macias 2005; Lopes 2014; 2015). The most impressive of these is the double-apsed funerary basilica of Rossio do Carmo, originally from the late fifth century, but with a long chronological continuity, and the baptismal building. The baptismal building is more than a baptistery, as it is included in a large building with a portico (probably a reused bath house), by the city’s forum, and lavishly decorated with polychrome geometric and figurative mosaics, with remarkable hunting scenes, also dated to the late fifth century (figure 5.6). The site of Casa Herrera, in the territory of Mérida, has a very similar layout and function to that of Mértola, and can be dated to the same period, the end of the fifth or early sixth century. 47 These new Christian monuments were probably linked to a strong local Christian community rather than to a single individual. 47 Further discussed in Chapter 6.

168 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.6 Mosaic of the ‘Opposing Lions’, from the porticus of the suburban Christian complex of Mértola

From Lopes 2014, figure 219

A final example of Christian munificence that can be listed is the mysterious building located in the north suburb of Mérida, usually identified as the xenodochium (figure 5.7), or ‘pilgrim hospital’ built by bishop Masona in the second half of the sixth century (VSPE V.3.4-7). This building, built outside the town walls, is related to the Saint Eulalia martyrial complex, whose basilica was also renovated at the same period. The so-called xenodochium consists of a main apsed building, with a porticated wing at each sides and two towers flanking the apse. It is difficult to imagine how it worked as a hospital, and similar buildings excavated in the Mediterranean have been interpreted as monastic constructions (Sastre de Diego 2015) but, according to his excavator, it is clearly a second-half sixth century construction, probably public in nature, and decorated with marble pilasters (Mateos Cruz 1995).

Visigothic state formation and urban renewal The development of these large new episcopal complexes inside urban areas began in the mid-sixth century, but was given a definitive push by the reformed Visigothic monarchy of the late sixth. Throughout the sixth century, the Visigothic kings had attempted to establish their full control

Towns and cities under Christian prevalence

169

Figure 5.7a Axonometric reconstruction of the building identified as the xenodochium of Masona in Mérida

Based on Mateos Cruz 1995

Figure 5.7b Photo showing the existing remains of the xenodochium, with the reconstructed location of the marble pilasters and the disposition of the apsed hall and the open courtyards

170 

The Iberian Peninsul a be t ween 300 and 850

over the Iberian peninsula, taking steps towards a more centralized state, however this was only achieved during the reign of Liuvigild (r. 568-586). Liuvigild’s reforms (Fernández et al. 2013; García Moreno 2008; Koch 2008) were a consequence of the presence of Byzantine troops on the southern coast of the peninsula, which had arrived during the civil war between Agila and Athanagild in the 550s. The presence of imperial troops posed a direct political (even if not really military) threat to the Visigothic kingdom, so in order to deal with the Eastern Empire on equal terms, the Visigothic king had to firmly establish a state in the Iberian peninsula. This was achieved by strengthening the alliance between Visigothic and Hispano-Roman aristocrats (especially land owners and bishops), reorganising the central administration and creating a new tax system (cf. Martín Viso 2013b). Liuvigild and his immediate successors (Reccared, Sisebut, and Swinthila) developed these reforms and pursued further centralizing policies which culminated with the establishment of Toledo as a permanent capital, the conversion of the Visigoths from Arian to Nicene Christianity and with the unification of the peninsular territory after the conquest of the Suevic kingdoms (585) and of the Byzantine province (625). These historical transformations have a clear reflection in the archaeological record, for instance, in the emergence of a new gold coinage (see Chapter 7), a new court ceremonial and, above all, in the developments in urbanism.48 The new Visigothic regime required a new architecture of power and a new administrative infrastructure which, even if based on late Roman models, had its own forms. New administrative centres (‘palaces’), where taxes could be collected and justice executed, begin to appear in the main existing cities of the Visigothic kingdom, in parallel to new fortifications and, in those areas of key importance but no main central places, new urban foundations (Olmo Enciso 2007; 2008c; Martínez Jiménez 2013). The architecture of the new power In parallel with the emergence of episcopal complexes, which ultimately represent the political power of the local (ecclesiastical) elites, the new central administration was represented in a series of ‘civic palaces’, similar to the late Roman palatia. In the Iberian peninsula these are few and difficult to identify (cf. Morín et al. 2015), but are still very good indications of the development of civic administration in the late sixth century. 48 Dey (2014: 140-60) further highlights, from a historical perspective, the symbolic role of towns as part of royal displays of power.

Towns and cities under Christian prevalence

171

Figure 5.8 Remains of the Late Antique public (?) structure built by the temple for the Imperial cult in the forum of Mérida

In Barcelona (Beltrán de Heredia 2002a; 2008a: 280-1), a U-shaped masonry building, built in a corner of the episcopal palace has been identified as the palace of the comes, the count, although this interpretation is based more on its location (under the later medieval palace of the Counts of Barcelona) than on any material archaeological finds. Sources, such as the Chronica Caesaragustana (510), mention a palatium in Barcelona already in the early sixth century, of which sadly nothing else is known (Ripoll López 2000: 380). A similar structure has been identified in the forum of Mérida, by the so-called ‘Temple of Diana’, and although it has not been identified as such, is most certainly a public building (considering the effort taken to build it and its huge foundations). This structure (Mateos and Sastre 2004) is built in reused masonry and granite ashlar blocks and partially inserted in the portico and temenos of the old temple of the Imperial cult (figure 5.8), so certainly in a central and still politically significant site of the city. In Córdoba, however, the large remains found at the south end of the walled enclosure can be probably identified with the Visigothic administrative palace and the ‘citadel’ (Vaquerizo and Murillo 2010: 524-5; León and Murillo 2009). Even if the excavations have shown little with regards to its plan

172 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.9 Schematic plan of the Visigothic civil administrative complex (palace and citadel) identified in Córdoba

(figure 5.9), its location under the early Umayyad palace and the mention in the sources of the ‘Vallat Ludhriq’ (i.e., palatium Roderici) are solid indicators. In other cases, the presence of the central power can be seen in the construction of new fortifications. As with ‘palaces’, the number of walls repaired in this period is quite low, and most of these repairs are quite problematic, as they are mostly known about through written sources. Archaeologically, the fortif ications in the city of Valencia are the best known. These include not only the inclusion of the circus within a new walled enclosure (Ribera and Rosselló 2013), but also the construction of a small fort in the immediate vicinity, at the site of València la Vella (Ribera Lacomba 2005; Ribera and Rosselló 2000). This site had a double wall of masonry and mortar, enclosing a perimeter of 4ha, with square towers, and dated to the mid-sixth century through the remains of ARS (Hayes 91, 99, 101, 103-105). Similarly, in Conimbriga the bico da muralha or western spur has been identified as a late sixth-century citadel, probably related to the integration of that part of Lusitania in the Visigothic kingdom after the conquest of the Sueves (De Man 2007a; 2007b). Lastly, the city walls of Italica, Toledo (the capital), and Pamplona are allegedly repaired by various monarchs throughout the sixth and seventh centuries. The walls of Italica are said by John of Biclar to have been restored

Towns and cities under Christian prevalence

173

by Liuvigild during the rebellion of Hermengild, who was stationed at the well-defended and near-by city of Seville, in 579. The walls of Toledo were supposedly repaired by King Wamba (r. 672-680), according to a lost hexa­ meter inscription (Continuatio Hispana 46), although there is no archaeological trace of this. 49 Wamba is also described as allegedly restoring the walls of Pamplona in a much later source (Mezquíriz de Irujo 2004: 44). New urban foundations Besides these construction works, what perhaps characterizes the Visigothic monarchy most are its newly-founded towns, clearly imitating imperial practices in the east, and unparalleled in the post-Roman west (Martínez Jiménez 2017). The best examples are Reccopolis and Eio, together with the large suburban development of Toledo, although other there were other, less well known ‘foundations’, such as Victoriacum, Bigastrum, Amaya and Ologicus. Reccopolis, identified with the remains at the la Oliva hill in Zorita de los Canes (Guadalajara), was a medium-sized urban foundation, 16ha (excluding suburbs) in extension and surrounded by a wall (Olmo Enciso 2006; 2007; 2008b; 2008c; Olmo et al. 2008). John of Biclar mentions that it was built in the year 578 (coinciding with Liuvigild’s decennalia), with fortifications, and a suburb. Its inhabitants were, furthermore, granted privileges. The site itself can be securely dated to the late sixth century, as a real ex novo urban foundation, although earlier residual and reused material appears too. The upper town consists of a large ‘palace’ complex: two large, opposing, aisled, two-storied buildings, built in mortared rubble, with tiled roofs, and opus signinum floors. These buildings flank an open area, which is closed to the east by a smaller linking building, of similar characteristics, and a cruciform apsed basilica with a baptistery. This complex is accessed through a ‘monumental’ gate, which opens to the main street, unpaved, but flanked by a series of shops and workshops, all built in the same phase (figure 5.10). The ‘lower town’ which spreads outside the gate to the palace complex is organised in a regular fashion, with clear signs of a pre-arranged urbanism. Beyond the main street, excavations have revealed various domestic units and a public cistern. There does not seem to have been an orthogonal street system 49 Erexit, fautore Deo, rex inclytus urbemWamba, su(a)e celebrem protendens gentis honorem.Vos, sancti Domini, quorum hic praesentia fulget hanc urbem et plebem solito servate fabore. – With God as [his] patron, the famous king Wamba erected the city, to increase the honour (and) fame of his people. You, holy Masters (i.e. Saints), whose presence shines here, serve this city and [its] people with your customary favour.

174 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.10 General plan of the excavations at Reccopolis, highlighting the location of the palace complex (conjunto palatino) and the workshop area

Based on Olmo Enciso 2008a

Towns and cities under Christian prevalence

175

in Reccopolis, but there was an irregular grid with primary and secondary streets, with a main road linking the east and west gates. Outside the city walls, remains on the western slopes and near the river have been identified as the suburb, and to the east, the remains of a new aqueduct have also been identified (Martínez Jiménez 2015). The centrality of Reccopolis is not only evident because of its scale, monumentality, and architecture, it was probably a provincial capital (of Celtiberia, a new administrative region) and an economic hub: a mint (Castro Priego 2008), a glass production centre (Castro and Gómez 2008), and the recipient of Mediterranean imports (Bonifay and Bernal 2008).50 Slightly smaller, but also extensively excavated, is the site of El Tolmo de Minateda, identified with the frontier episcopal see of Eio or Elo. It is built upon the remains of an old early Roman oppidum (Gutiérrez et al. 2005: 351-7), but should nonetheless still be counted as a new foundation due to the lack of habitational continuity. It is located in a key place, near the Byzantine frontier, and established so as to claim the lands of the see of Ilici, which was then under Byzantine occupation (Gutiérrez et al. 2005: 345). The suggested dating is slightly later than Reccopolis, probably established during the reign of Reccared, as Reccopolis, Eio is built on a hilltop and well-fortified with walls and a main gate (figure 5.11). The main foci of the town (Abad et al. 2008; Gutiérrez and Sarabia 2013) are the fortified acropolis and the episcopal complex, which includes an apsed basilica, an episcopium, a baptistery and a cemetery. The rest of the site seems to be occupied with dwellings, organised around streets. The main characteristic of Eio, and which can be extrapolated to understand what happened in Reccopolis, is that the urban development was completed in one main phase, and according to a master plan. In fact, the outlines of the buildings seem to have been chiselled into the rock, presumably by the architects, for the builders to follow. In terms of domestic architecture, both Reccopolis and Eio show similar patterns of houses organized around a small courtyard, in a similar fashion to the domus of the Marbles of Mérida described above. A reoccupation and fortification of an old hilltop site, and in the same region, was that of Bigastrum, built to replace the see of Cartagena, although it is comparatively far less well known than Eio (González Blanco 2007). Despite these constructions, the one place where new constructions and monuments were to be expected was in the capital, Toledo, but until recently very little was known of the Visigothic phases (figure 5.12). Traditionally it 50 Although Reccopolis is impressive within its immediate context of post-Roman Gaul and Hispania, it is important also to keep in mind its wider context. After all, Reccopolis is coeval with Justiniana Prima, Hagia Sophia and the mosaic churches of Ravenna.

176 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.11 Plan of the excavations at the site of El Tolmo de Minateda, identified with the Visigothic episcopal see of Eio

Based on Gutiérrez and Sarabia 2013, figure 1

had been assumed that the ‘royal palace’ was in the upper town, possibly under the Umayyad palace (whose location was known), but there were no archaeological grounds for this claim. In the last ten years, excavations in the lower meadow, by the Tagus river (the site of Vega Baja) have uncovered a large suburban area with a very important late Roman and Visigothic phase. It was known from the acts of councils that a praetorium suburbanum and various royal basilicas existed in Toledo (Gurt and Diarte 2012; Teja and Acerbi 2010), but beyond a few samples of carved marble and the location of the suburban funerary basilica of Saint Leocadia, nothing was really known (e.g. Ripoll López 2000). Despite this, a palatial complex in the upper town should not be ruled out so easily, as it is simply a matter of archaeological visibility, and it would also follow the same pattern as Reccopolis and Eio, both located on defendable

Towns and cities under Christian prevalence

177

Figure 5.12 Plan of Toledo, indicating the location of the lower meadow (‘Vega baja’) suburb, with the location of the main Late Antique sites

hill tops. The alleged repair of the city walls by Wamba mentioned above would further support this statement. The remains in the lower meadow still require further excavation and interpretation, but the preliminary results show a large suburban complex with various units. Chronologically, it developed in its entirety in the last quarter of the sixth century (on top of a long-abandoned group of Roman suburban villas), and was set on a street grid. The remains of what might have been the ‘palace’ itself seem to lie beyond the excavation area, but impressive walls of masonry (quarried most probably from the nearby circus) and opus signinum floors have been identified under some of the nearby buildings (Olmo Enciso 2009; 2010). The material culture found in this area shows amphorae and ARS import patterns similar to those of Reccopolis, with glass wares and carved marbles from the Mérida workshop (de Juan et al. 2009). Other royal foundations known from the sources are more problematic, because they have not been identified archaeologically, so it is impossible to assess their urban nature. Two foundations have been located in the territory of the Vascones, west of Pamplona. The first, Victoriacum was founded by

178 

The Iberian Peninsul a bet ween 300 and 850

Liuvigild (John of Biclar, 581) and the second, Ologicus¸ by Swinthila (HG 63). Traditional identifications of Vitoria-Gasteiz (the capital of Álava) with Victoriacum have been proven to be wrong by archaeologists (Azkarate GaraiOraun 2004b: 34-5). Ologicus has traditionally been linked to Olite in Navarre, but on the basis of nothing more than the similarity of the names. Another site which is known through the sources to have been re-established by the Visigoths is Amaya, in northern Castile. This site has been excavated, and rather than a town, it seems to have been a large fortified castrum (Quintana López 2017), so it may be that both Victoriacum and Ologicus were similar hill-top fortified enclosures, their status as civitates perhaps being just an indication of their legal and administrative importance – not of their urban nature. Other examples of royal munificence? Other than these examples of large construction projects, directly linked to the monumental display of the new political power, the Visigothic monarchy took up the habits of public munificence as a way of legitimizing their position by imitating the Eastern emperors (rather than looking back at early Roman practices). However, the examples that we can mention are very few. The Visigothic monarchy of the late sixth and early seventh centuries was not the Ostrogothic monarchy of Theoderic, which was far more concerned with and connected with the Roman past and much more involved in munificence. By the time the Visigoths had firmly established and reorganised the kingdom in the last quarter of the sixth century, the Roman infrastructure had long disappeared. In fact, to the extent that Visigothic administration imitated late Roman practices, it had to be mostly reinvented (Martínez Jiménez 2014: 217-8). The involvement of kings in new building dedication is mostly known through inscriptions. Most relate to churches, as this type of patronage had been common practice since the late Roman period; in some cases they were built anew, in others re-consecrated (from Arianism). By promoting and continuing the late Roman tradition of religious patronage the Visigothic kings were legitimizing their position. Supporting the Church was a key factor in the structure and balance of power between the monarchy and the ecclesiastical elites. For instance, Reccared’s inscription from Toledo shows the dedication of the cathedral (IHC 155 = ICERV 302),51 whereas Recceswinth’s inscription at 51 In nomine d(o)m(i)n(i) consecra|ta ecclesia S(an)ct(a)e Mari(a)e| in cat(h)olico die pridie| idus aprilis anno feli|citer primo regni d(omi)ni| nostri gloriossisimi Fl(avii)| Reccaredi regis (a)era|

Towns and cities under Christian prevalence

179

the church of Saint John in Baños (IHC 143 = ICERV 314)52 is an example of the former.53 Chindaswinth’s inscription of Mérida (IHC 24 = ICERV 366),54 fragmentary as it is, could relate to a church, but not necessarily. Lastly, there is one inscription, preserved in the archaeological museum of Seville, which does not seem to belong to a church, and that could relate to an unknown public building. Hermengild’s inscription (IHC 76 = ICERV 364; cf. Fernández and Gómez 2001)55 is difficult to read and to interpret, but it seems to have been a lintel which shows, in any case, that Hermengild was developing an architecture of power for his own regime. The one example which is known through archaeology is the basilica/ monastic complex of Guarrazar in the province of Toledo, where the treasure presented in the Introduction was found. Recent excavations in 2013 and 2014 around the area have revealed a series of buildings which seem to point towards a religious complex. The finds suggest a seventh to eighth century date for the complex, which includes the remains of what appears to be an apsidal basilica, a large L-shaped building (interpreted as a monastic structure; figure 5.13), and a large, rectangular building, cut into the rock which encloses two natural water springs (Eger 2007; Rojas Rodríguez-Malo 2013; 2014). The reconstruction of the basilica is highly hypothetical, but the decorated friezes and reused marble column bases hint towards this interpretation. The complex itself may not necessarily have been dedicated by the Visigothic kings, but the fact that on the eve of the Islamic invasion of 711 the various gold votive objects were taken there and buried for safety suggests that it was most probably an important royal monastery. DCXXV – In the name of the Lord, the church of Saint Mary was consecrated on a Catholic holiday the day before the ides of April in the first joyful year of the reign of our lord most glorious king Flavius Reccared. (Given) in the era of 625 (12 April 587 AD – perhaps Good Friday). 52 Precursor D(omi)ni, martyr babtista Iohannes, posside constructam in eterno munere sede(m), quam devotus, ego, rex Reccesvinthus, amator nominis ipse tui, proprio de iure dicavi, tertii post dec(imu)m regni comes inclitus anno, sexcentum decies era nonagesima nobem – John the Baptist, Martyr, Predecessor of the Lord, receive this building built as an offering of eternal dedication which I, devout King Recceswinth, your follower, has dedicated at my own expense on my lands on the thirteenth year of the reign, in the [Hispanic] era of 699 (i.e. 661 AD). 53 Although neither the inscription nor the church, as preserved today, seem to be the original Visigothic inscriptions, as will be discussed in Chapter 6. 54 [Chi]ndasvintus rex pi[issi/mu]s in mperio/ [Ch]indasvintus [… 55 (Chi-rho) In nomine Domini ann[o f ]eliciter secundo regni Dom(i)/ni nostri erminigildi regis quem persequitur genetor (sic)/ su(u)s Dom(inus) Liuvigildus rex(.) in cibitate(m) ispa(lim) ducti aione – In the name of the Lord, in the second year of the happy reign of our lord King Hermengild, whom his father our lord King Liuvigild persecutes. (Hermenegild was) brought into the city of Seville for ever (aione).

180 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.13 Plan of one of the main buildings identified next to the basilica of Guarrazar

Based on Rojas 2014, figure 5

State intervention in the Byzantine and Suevic territories Outside the areas controlled by the Visigoths, the late fifth and early sixth centuries continued to develop as did the rest of the peninsula. However, in the late sixth century they were subject to processes of state formation (in the Suevic kingdom in Gallaecia) and to the establishment of a military administration (in the Byzantine province). The result in both cases was the same, as the new polities had to develop a new visual language of power and required new infrastructures to control the territories which had been in a virtual power vacuum since the mid-fifth century. Transformations in the Byzantine province The territories of the ‘Byzantine province’ can be loosely defined as the thin strip of coast between Cádiz and Cartagena, including also the Balearic Islands. They were swiftly conquered during the campaigns of 554 and

Towns and cities under Christian prevalence

181

555, and quickly introduced into Justinian’s imperial system. As opposed to Italy or Africa, the province of Spania (as it was termed) was not fully incorporated, due to military restrains, the plague, and renewed war efforts in the East. In order to control this territory with limited forces, a new system was developed based on small fortified harbour towns (‘beach-head model’; Martínez and Tejerizo 2015: 92-3; Martínez and Moreno 2015), with secondary nuclei, where even if they were not affected directly by state intervention, they indirectly benefited from the trading network which developed around the annona. Cartagena and Málaga, the two main cities of the province, have the best examples of ‘Byzantine’ archaeology in this period. In the case of Cartagena, archaeologists have identified a whole series of new harbour and docking structures; next to the harbour the theatre was turned into a dwelling area (figure 5.14), with various houses built in the cavea (Vizcaíno Sánchez 2008; 353-6). This reorganization of the harbour complex was completed with the refortification of the reduced late Roman walls or, at least, of one of the main gates, which was repaired by Comitiolus, the Magister Militum Spaniae – the master of the soldiers in Spania, i.e., the provincial governor. This inscription (CIL II 3420 = ICERV 362; Prego de Lis 2000)56 is the only epigraphic evidence for Byzantine munificence in the Iberian peninsula, and underline the importance of Cartagena. Málaga, the other main centre, has produced similar archaeological results (Navarro et al. 2000; 2001; Vizcaíno Sánchez 2009: 159-68, 424), with what are perhaps fortifications (reoccupation of the acropolis) and improved harbour facilities (new docking areas and a new harbour). Besides Málaga and Cartagena, Ceuta, across the straits of Gibraltar, was the main Byzantine enclave of the area, with improved fortifications (according to Procopius) and even a new basilica commissioned by Justinian (Bernal Casasola 2008a: 373-7). On the southern coast of the peninsula, within the area controlled by the Eastern Romans, small urban settlements, such as Algeciras, Carteia, 56 †Quisquis ardua turrium miraris culmina| vestibulumq(ue) urbis duplici porta firmatum| dextra levaq(ue) binos porticos arcos| quibus superum ponitur camera curia convexaq(ue)| Com[it] iolus sic haec iussit patricius| missus a Mauric(i)o Aug(usto) conra hostes barbaro[s]| magnus virtute magister mil(itum) Spaniae| sic semper spania tali rectore laetetur| dum poli rotantur dumq(ue) sol circuit orbem| ann(o) VIII Aug(usti) ind(ictio) VIII. †Whoever you may be, admire the formidable tops of the towers and the vestibule of the city, strengthened by a double gate. Left and right it has two arched porticoes, over which a curved convex chamber (i.e., vaulted) is placed. Patrician Comitiolus had this made, sent by Maurice Augustus against the barbarian enemies, great in virtue, master of the soldiers of Spania. May Hispania always rejoice in such a governor, as long as the poles turn and the sun orbits the world. (Given) in the eighth year of (Maurice) Augustus, eighth indiction

182 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.14 Plan of the Byzantine quarter excavated at the Roman theatre of Cartagena, which includes various dwelling units

Based on Ramallo and Ruiz 2005, figure 2

Baelo Claudia, Lucentum, and Ilici became small economic hubs for the redistribution of goods. This is evident not only in the disproportion between Eastern and African imports in these territories, which is similar to what can be found in Cartagena or Málaga, but very different to other Mediterranean towns such as Valencia or Barcelona (discussed in depth below). In these minor sites there are also indications of renovated or preserved harbour

Towns and cities under Christian prevalence

183

structures and docking areas – there are even new beaching and harbour facilities. But no traces of fortifications or other evident alterations of the urban fabric; not even new Christian buildings. In these cases, it can be assumed that the local urbanism continued with the long-term transformations that derive from the late Roman world, and were not greatly affected by the arrival of the Byzantine army other than by improved trading links. Besides these coastal territories, the Balearic Islands followed similar patterns. Unfortunately, there is little to say about the archaeology of towns in the islands, and the only general picture we have is that of Pollentia. During the late fifth century (or maybe even in the sixth) a large new wall, built partially with reused material, was erected through the town centre, reusing the back of the forum temples as part of the fortification (cf. Cau and Chávez 2003: 45; Orfila Pons 2000; Orfila et al. 2006; Riera et al. 1999: 337-9), in an effort which ahs been linked to either Vandalic or Byzantine intervention. Other towns in Minorca and Ibiza have provided less information, but there are ceramic sequences that suggest that they continued to be inhabited and active as trade centres through the brief Vandal occupation of the islands and the Byzantine conquest (Cardell and Cau 2005: 162). The Suevic monarchy In Gallaecia, the Suevic monarchy went through a very similar process of state formation (Díaz Martínez 2011), although the spur that triggered the transformations appears to have been the reorganization of the Visigothic kingdom, as it posed a direct threat to the existence of the Suevic one. Furthermore, the process of state formation was aided by the arrival of a man from Pannonia, Martin, who was one of the main figures to rearrange the Church hierarchy in the Suevic territories after his arrival around 550. This, together with the conversion of the Suevic king to the Nicene orthodoxy, finally ratified the allegiance of the local elites to the Suevic monarchs. However, Liuvigild conquered the Suevic kingdom in 585, and in this short period of time there does not seem to have been many construction programmes that could mirror those of the Visigoths. Even in Braga, the kingdom’s capital, are there no remains that could indicate main building activities. There are, however, two main archaeological sites around it that are relevant. The first site is Dume (or Dumio), a small monastic site, which was promoted to a bishopric by Saint Martin. The archaeological remains (Fontes 1995) range from an early Roman villa to a later Medieval church, but there is a sixth-century phase, in which a large basilica was built, dedicated

184 

The Iberian Peninsul a bet ween 300 and 850

Figure 5.15 Plan of the Suevic ‘acropolis’ of Falperra near Braga

Based on López Quiroga 2004a

with an inscription in verse and decorated with marble carvings. Even if it was a monastic community, the development of the site responds to royal patronage over Martin. The second site is the ‘acropolis’ of Falperra, which has been called ‘the Suevic Reccopolis’. The site itself (López Quiroga 2004a: 532-3) is a fortified hilltop enclosure, with a set of large masonry buildings dated generically between the fifth and sixth centuries (figure 5.15). Even if Falperra can with difficulty be described as a ‘royal residence’, or even as an administrative centre, it is large, fortified, and close to Braga, so there are grounds to link it to the Suevic monarchy – even if this is only at a military level. One of Martin of Braga’s main reforms was the promotion of various small nuclei to episcopal rank. With this promotion, various smaller towns became main administrative centres, although archaeologically there is not much to reflect this. One of these promoted secondary nuclei was Orense, although it does not seem to have been truly ‘urban’ in nature. It was created

Towns and cities under Christian prevalence

185

to substitute the disappeared bishopric of Aquae Flaviae, and the centre of the disperse settlement was a ‘rural’ basilical complex. Orense served as a political articulator of the surrounding rural population, in a region where towns as such had failed to function as central places. Furthermore, even in main cities, such as Lugo or Idanha, there are no traces of any transformations to their urbanism beyond those that can be traced back to the late Roman period. The development of a strong Suevic monarchy is better reflected in the establishment of several fortified central places near the Visigothic territories (Ariño and Díaz 2014), which shall be described in the next chapter. Conquest and integration in the Visigothic kingdom The military conquest of these two territories by the Visigoths during the late sixth and early seventh centuries put to an end both processes of state formation. Although in the Suevic case there are not many examples of ulterior urban transformations after the conquest, in the Byzantine territories it is clear that the Visigothic conquest put an abrupt end to urban developments (Martínez and Moreno 2015). This was not only the result of military action, but also because of the administrative reorganization of the territories during and after the conquest. From a military perspective, Cartagena and Málaga were destroyed (according to the sources) during the Visigothic conquest. Even if the degree of destruction is difficult to assess on archaeological grounds, in Cartagena the houses built in the theatre were destroyed by a fire which can be dated through pottery to the early seventh century (Vizcaíno Sánchez 2008: 356). In Málaga, a 0.5m thick destruction layer has been found in the harbour area and the lower town, suggesting perhaps that the settlement was reduced to the small citadel (Salado et al. 2002: 362-4). A side effect of this military conquest was, of course, the cessation of direct trading links with the wider Mediterranean through the military annona, which had a direct impact on the economy and prosperity of secondary coastal communities such as Almuñécar, Baelo Claudia, Algeciras, Elche or Lucentum. In terms of the administration, the way in which the Visigothic monarchy claimed back territories on the frontier with the Byzantine areas of influence was, as we have seen in the cases of Eio and Bigastrum, by founding new administrative sites (bishoprics), and even by creating a new centre of power, even closer to Elche, in Orihuela (Martínez and Moreno 2015: 287). From these it was possible not only to launch military campaigns but also to structure a hinterland which would have otherwise become a no-man’s land. This establishment

186 

The Iberian Peninsul a bet ween 300 and 850

was preserved after the fall of the Byzantine strongholds of Málaga and Cartagena, which declined to their royally-sponsored foundations. Both sites continued to function as secondary towns, preserving their bishoprics and functioning as trading harbours, but on a much reduced scale. Overall, and as far as it is known, while in the Suevic territories the conquest by the Visigoths did not inspire any forms of new urbanism, the Byzantine territories were greatly disrupted. In any case, the evolution of towns in the peninsula during and after the seventh century can be seen as a long steady stagnation.

Developments in the seventh and eighth centuries The military unification of the peninsula by Swinthila in the 630s put an end to the period of Visigothic state formation. The political foundations of the Visigothic monarchy had been established, and whereas it is debatable how sound and strong the kingdom was (cf. Poveda Arias 2015), from an archaeological point of view it is clear that the period of ‘urban renewal’ had come to an end. Mid- and late seventh century constructions are rare enough to be remarkable, and they are in most cases, ecclesiastical in nature, as the late royal constructions of Mérida (Chindaswinth), Saint John’s in Baños (Recceswinth) and Toledo (Wamba) show. The Continuatio Hispanica also mentions the construction (or probably just an expansion) of a palace by King Roderic (r. 710-711) in Córdoba. Beyond these, there are some new churches built by bishops, and a rare example of a lay foundation such as the churches of vir inlustris Gudiliuva (IHC 15 = ICERV 304).57 None of these 57 [In nomin]e d(e)i n(o)s(tr)i Ih(e)s(u) Chr(ist)i| consecrata est| [ec]clesia S(an)c(t)i Stefani primi martyris in locum Nativola a s(an)c(t)o Paulo Accitano pont(i)f(i)c(i)| d(ie) … an(no) … d(omi)ni n(o) s(tr)i gl(oriosi) Wittirici reg(i)s| er(a) DCXV| Item consecrata est ecles(s)ia| S(an)c(t)i Iohann(is) [bap] tist(a)e| Item consecrata est ecclesia S(an)c(t)i Vincentii| martyris Valentin(i) a s(an)c(t)o Lilliolo Accitano pont(i)f(i)c(i)| [die] XI kal(endas) febr(uarias) an(no) VIII gl(oriosi) d(omi)ni Reccaredi reg(i)s er(a) DCXXXII| h(a)ec s(an)c(t)a tria tabernacula in gloriam trinitatis …| cohoperantib(us) s(an)c(t)is aedificata sunt ab inl(ustre) Gudiliu[va]| cum operarios vernolos et sumptu proprio. In the name of God our Lord Jesus Christ, this church was consecrated at the site of Nativola to Saint Stephen, first martyr, by holy bishop Paul of Acci, in the […] day in the […] year of our glorious king Witteric, in the era 615 [sic, AD 577]. Likewise, the church to Saint John the Baptist was consecrated, likewise was the church of Saint Vincent martyr of Valencia by holy bishop Lilliolus of Acci, on the eleventh day [before] the calends of February of the eighth year of our glorious lord king Reccared, in the era of 632 [21st January AD 598]. These three holy shrines were built for the glory of the Trinity [and its?] holy cooperates by [vir] inlustris Gudiliuva with his own slave-workers and money.

Towns and cities under Christian prevalence

187

constructions, however, can be seen as marking any new trends or changing the urban landscape in any significant form (most of these churches were, furthermore, rural establishments), and rather return to late fourth and fifth century organic developments. In fact, despite the Islamic invasion of 711, and excluding a handful of exceptions (discussed in Chapter 8), these same trends continued to frame the urban development into the mid-eighth century. Towns entered into a period without mayor alterations to their urbanism. In fact, the late Roman organic processes of suburbanization, encroachment of public areas and dismantling of redundant and useless buildings which had characterized the late fourth and fifth centuries continued. In Málaga excavations have revealed how a street was narrowed to just over a metre wide in the seventh or eighth centuries (Bazzana 1991: 176). There is an apparent reduction of urban inhabited areas, which seem to cluster around the religious centres, although this may be simply a case of archaeological visibility. Seventh-century townscapes in Mérida (Mateos and Alba 2000: 152), Barcelona (Ripoll López 2002: 36-40) and Braga (Fontes et al. 2010) show a clustering of habitated areas around the cathedrals and a rearrangement of the main street axes to suit the new priorities. Even in newly founded sites such as Reccopolis and Eio there is disruption of the pre-arranged urbanism in parallel with what was seen in other urban settlements, where buildings were subdivided into smaller dwelling units and private buildings encroached upon the streets from the mid-seventh century on (Olmo et al. 2008: 69; Sanz Paratcha 2008). It is in this period that the aqueduct of Valencia seems to have finally gone out of use (Martínez Jiménez 2011a). In fact, the final phases of detectable Late Antique urban life in sites such as Málaga, Valencia, and Tarragona are dated to this time. Housing patterns did not change either in the post 711 period. The traditional Islamic urban house visible in the archaeological evidence is only visible from the ninth or tenth centuries (see Chapter 8). Having said this, it is also important to mention that the seventh and eighth centuries form a period of blurred archaeological chronologies. Many of these transformations or abandonments could well be due to archaeological invisibility or the lack of refined pottery sequences, excluding perhaps a few sites such as Mérida, Reccopolis, or El Tolmo. Overall, these final transformations of the Late Antique townscape do not necessarily mean that towns disappeared, or that they ceased to be seats of power. After all, most sixth century towns continued to be active mints and episcopal sees in the seventh century, and they were the places where the local elites negotiated with the invading Muslims. There had been a very clear visual

188 

The Iberian Peninsul a bet ween 300 and 850

transformation of the town, moving away from the Roman model, but they retained traces of their centrality and political importance. It was only after the mid-eighth century, when the Umayyads began their own process of state formation in the peninsula that a new active transformation of Hispanic townscapes took place. These (and not the Visigothic transformations) were what formed the basis of the Early Medieval townscape.

Trade and towns in the post-Roman period The collapse of western trade in the fifth century caused a reduction in the volume of imported goods to Hispanic towns, but trade continued into the sixth and seventh centuries. Towns were still the main trading points, but the range and distribution of imported goods was increasingly limited. As was argued in Chapter 3, large scale Hispanic industrial and agricultural productions collapsed with the fall of Rome, and the remaining productive sites seem to have focused mostly on the regional market. This does not mean that Spanish exports ceased in the post-Roman period, but it could be the case that they are too few to be relevant, or else archaeologically invisible. Gregory of Tours mentions ‘a ship from Spain with its usual cargo (negotio solito) arriv[ing] at its harbour [Marseilles]’58 in the sixth century, although there is no indication of what this ‘usual cargo’ was. Reccopolis and Mérida are two exceptional examples of sixth-century industrial production, in the former glass and jewellery, and in the latter carved marble, which were perhaps more widely distributed, although this may be related to their direct link with the Visigothic monarchy, and will be discussed at length in Chapter 7. When it comes to imports, the general evolution was a slow but steady decline in the volume of foreign imports, which were mostly wine (Eastern Mediterranean: LRA 1 = Kealy LIII and LRA 5/6; African: Spatheion 2 and 3, and Keay LXII) and fine wares (mostly ARS types Hayes 87, 88, 89, 91C, 99, 100, 101, 102, 103, 104A, 104B, 105, 106, 107, 108, and 109, but there were Phocaean Red Slip and DSP also present) and, to a lesser extent, African oil (Keay forms VIIIB, XXVII, LV, and LVI), and cooking wares (cf. figure 1.8). These imports largely ceased by the seventh century but, as is usual for this period, chronologies and volumes vary regionally. Towns in theory acted not only as recipients for these imports, but also as redistributive nuclei for the countryside, but in the late Roman period imports seem to have been 58 DLH IX.22: ‘(…) navis ab Hispania una cum negotio solito ad portum eius adpulsa est.’

Towns and cities under Christian prevalence

189

restricted to elite sites, which is to say towns: the penetration of imported goods even to their hinterlands by the sixth century was quite limited. On the Mediterranean coast, outside the Byzantine territories, most imports were from Africa, with a smaller proportion of Eastern trades. Finds from the excavations at La Almoina in Valencia, the Plaza del Rey in Barcelona and the harbour sites of Tarragona (and, to a lesser extent, in Zaragoza and other minor sites such as Iluro) all show similar importing patterns (Járrega Domínguez 2013; Macias and Remollà 2005; Rosselló Mesquida 2000). In these sites, up to 88% of the amphorae are African, and in all cases, the penetration of these finds into the hinterland is limited, if at all present. Gaulish and Sicilian wares also appear, with a minor proportion of Eastern amphorae. In general, these Mediterranean harbours continued to receive imports in considerable amounts up to the mid-seventh century, but this completely disappears by the eighth century. In the Byzantine territories (Vizcaíno Sánchez 2009: 605-30), including well-documented deposits from Algeciras, Almuñécar, Málaga, the theatre excavations of Cartagena, and the small harbour sites of Alicante (the harbour of Santa Pola, the island of Tabarca and the rubbish dumps from Benalúa) the proportions are quite different. In these areas, and during the Byzantine occupation, the benefits of the annona meant that Eastern imports were much more prominent – up to a 50% as compared to African imports (including amphorae types LRA 1/Keay LIII, and Keay LIV, LXI, LXII and LXV). These trading patterns, however, show little continuity after the Visigothic conquest, when these towns saw a sudden decline in the number of imports in the late seventh century and, as with the rest of the Mediterranean coast, an overwhelming majority of African wares. On the Atlantic coast the main port cities were Seville, Mértola, Vigo and Lisbon, which supplied the inland sites of Córdoba, Mérida, Braga and Santarem respectively. These harbours were connected to the Mediterranean networks, but were part of the longer routes that went up to the Atlantic coast of Gaul and Britain. The site of Vigo is particularly impressive (Fernández Fernández 2014), as it shows how a former secondary settlement (for which there is no clear evidence that it was fully urban) happens to have been one of the main trading nodes of the Iberian Atlantic coast. The large pottery deposits retrieved from the excavations at the villa of Toralla and Rua Rosalía de Castro show the continuity of long distance trade with the Mediterranean world during the post-Roman period. The stratigraphic sequence of the site of Vigo shows the chronological fluctuations of trade on the Atlantic coast. The early fifth century is characterized by a steady continuity of eastern and central Mediterranean imports, regardless of the Suevic invasion, where

190 

The Iberian Peninsul a bet ween 300 and 850

imported coarse and cooking wares were more abundant than fine wares and amphorae, parallel to local production of similar products. The central decades of the century, however, show a decline in the volume of imports, due to instability in the production centres, and the general imperial crisis, but then it blooms again in the later years of the fifth and into the sixth century. African table ware productions in this area continue to be more numerous than eastern or Gaulish imports (52% ARS, compared to 35% Phocaean RS and 6.6% DSP), but this compares sharply with wine and oil productions, where eastern amphorae represent the majority of the finds (62.4% Eastern compared to 17.8% African). As also happened with the Mediterranean examples, by the sixth century trade seems to be limited to coastal enclaves, with none of these imports reaching the interior. This high proportion of eastern wine amphorae is similar to the excavations at La Encarnación in Seville (García and Vázquez 2006); the evolution and proportion of fine wares is similar to what is found in Mértola (Lopes 2014: 198-200). It would not be fair to talk about a post-Roman urbanism for the Iberian peninsula, unless we want to underline both the development of a new state monumentality dissociated from the late Empire and the significant processes of regionalization that occurred in this period. Even so, it would be necessary to acknowledge that all the constructive and destructive elements which altered the Classical Roman townscape in this period are nothing more than the accelerated continuation of those patterns which began in the fourth and fifth centuries. Nevertheless, the transformations that took place in the centuries immediately following the fall of Rome are remarkable. Monument wise, episcopal and civil administrative complexes appeared in those cities which continued to have political and economic relevance in the successor kingdoms (Visigoths, Sueves, and even the Byzantine province). The renovated emphasis in fortifications and the new uses given to the old Roman areas of power display (such as fora, amphitheatres and circuses) underlie the different attitudes to the Roman urban fabric and its infrastructure. The relocation of Christian buildings for worship in central locations also highlights the triumph of Christian monumentality. And these are elements that appeared in new urban foundations as well like Reccopolis and Eio (a striking characteristic of the Visigothic kingdom in the post-Roman West), not just in old Roman towns, where space was limited and previous buildings partially dictated the layout and location of new ones. Furthermore, in Toledo, to counter the restrains of the old urban

Towns and cities under Christian prevalence

191

nuclei, the Visigothic monarchy chose to display its monumentality with a new palatine suburb. The vitality of towns and cities is seen not only in the new monuments linked to the development of new buildings and a new architecture of power, but also in their role as central trading nodes, with strong (even if limited and weakening) access to broader Mediterranean trading networks. Finally, it has to be stressed that the transformations of the sixth and seventh centuries continued along the same lines (begun in the fifth century) into the eighth century. As we will see in the next chapters, only the development of a new state with new monumental and religious necessities would bring an end to these Late Antique changes.

6

The new rural landscape

The dismantling of the Roman imperial political economy in the f ifth century, the necessary adaptation of both local communities and regional elites to the new political milieu and the development of new post-Roman Germanic polities (the Sueves and the Visigoths) led to a radical reorganization of rural settlement patterns. The new political context caused both the disappearance of previous realities like the late Roman villa (see Chapter 3) and the emergence of new settlement types such as hillforts, farmsteads and villages, and religious settlements under Church control. One of the characteristic main developments of this period is the consolidation of the village and its associated diversified economy, as introduced in Chapter 3. By the sixth century, the rural landscape was mostly formed by networks of farmsteads and villages whose regional development depended on the dialectical interaction of peasant agencies and the capacity of regional elites and the state to impose their power. Furthermore, this new peasant reality developed not only original means to construct rural landscapes and local economies, but also new funerary rituals, in which grave goods served both to represent social identities and to create a new social habitus within local communities. Another important phenomenon which took place during this period was the strengthening of the Church as an active power, directly related to the development of the Visigothic state. As we have already seen (in Chapter 4), Christian elites had the urban spaces as their main performing arena during the fourth and fifth centuries, where they developed a new pole of power which entailed major changes in the archaeological record. The main consequence of the expansion of Church influence into the countryside was the construction of shrines and monasteries, a process that was only fulfilled in later centuries. In this chapter we will analyse four main elements of the rural landscapes in the post-Roman Iberian peninsula: the occupation of hillforts, the networks of farmsteads and villages, the new burial rites, and the introduction of Christianity in the rural world.

Hillfort occupations A process of occupation of settlements located on elevated positions is archaeologically documented during the f ifth century all over Europe.

194 

The Iberian Peninsul a bet ween 300 and 850

Figure 6.1 Extensions excavated in some hillfort sites of the northern part of the Iberian Peninsula SITE Dehesa de la Oliva Castro Ventosa Bernardos El Castillón Cristo de San Esteban Cabeza de Navasangil Tedeja Los Castellares Monte Cildá Las Merchanas Cerro de la Virgen del Tormejón Peña Amaya

EXCAVATED AREA (m2)

TOTAL EXTENSION (m2)

% EXCAVATED

4000 45.43 130 ±300 1968 800 n.d n.d. ±660 n.d. 13 220

280000 53000 30000 30000 45000 16000 20000 19500 100000 54000 70000 420000

3.57 0.08 0.4 1 4.3 5 – – 0.6 – 0.02 0.05

This reoccupation was in direct relation to territorial administration, and these sites were in most cases fortified on hills from which a wide territory could be effectively controlled (Brogiolo 2000; Kobylinski 1990; Schneider 2001). This hillfort phenomenon is well observed in many regions of the Iberian peninsula but it is especially dense in its mountainous northern half, where the transformation and decay of the late Roman urban network was deeper (Quirós and Tejado 2012). Having said this, regional studies for the southern territories exist as well: e.g. the studies for the territory of Granada (Carvajal et al. 2015) or Guadix (Martín Civantos 2010). We can define them in general terms as sites characterized by a relative chronological stability (occupied for at least two or three generations), fortifications (normally walls and natural slopes) and, usually, an elevated location within the surrounding territory, although these are not exclusive (Vigil-Escalera and Tejerizo 2014). The appearance of hillforts was already identified as a Late Antique phenomenon in the late nineteenth century. Despite this, only few hilltop sites have been effectively excavated. Furthermore, these excavations have always been quite limited, covering less than 5% of the sites (see figure 6.1) and mostly focused on the defensive structures, so we have only a partial view of how their inhabitants lived. Moreover, most of these excavations are quite old, a situation which hinders current interpretations (Quirós Castillo 2012a: 18).

The new rur al l andscape

195

Traditionally, it has been assumed that the occupation of these hillfort sites was related to the invasions of either the third or the fifth century. Unfortunately, these biased historical interpretations were produced, and reinforced, by two related archaeological problems. First, there was an insufficient knowledge of the Late Antique pottery which, in any case, shows little formal change over several centuries (see Chapter 7); secondly, there was a lack of extensive excavations with absolute chronologies. As a result, there is still some confusion in the dating of these contexts, as casual finds found not in closed contexts from earlier periods in post-Roman phases, like steles or coins, have been interpreted wrongly. Because of this, some sites have been dated in very long chronological ranges between the first and eighth centuries. For example, in the case of Monte Cildá or El Cristo De San Esteban funerary steles reused in the walls have been taken as indicators for an occupation between the third and fourth centuries, even though there is no other piece of archaeological evidence to support such dating (Domínguez Bolaños 1993; García et al. 1968; 1973). Also, due to incorrect datings of the pottery, some scholars have proposed a long and continuous occupation for some sites, between the fifth and the eighth century, for example Bernardos, which has no contexts that are clearly datable beyond the sixth and seventh century (Gonzalo González 2007). Despite this, recent studies based on pottery analysis from closed contexts, together with their relative sequencing, have determined two different phases of hillfort occupations. Such phases are related to two different historic contexts, which Juan Antonio Quirós has called ‘first-’ and ‘second-generation hillforts’ (Quirós Castillo 2012a). First-generation hillforts (fifth-sixth centuries) This first-generation hillfort occupation is related to the moments of political, social and economic instability caused by the dismantling of the Roman Empire in most parts of the Iberian peninsula. Even though this process may appear homogeneous, the variability and complexity of situations in archaeological and in social terms was enormous. Understanding the hillforts Considering both the archaeology and the size of the sites, Juan Antonio Quirós has proposed three types of first-generation hillforts: small (less than 1ha); medium (2-10ha); and large (over 20ha), with a dense and complex internal organization (Quirós and Bengoetxea 2010: 159). Another classification proposed by Italian scholars (Brogiolo 2000; Valenti 2004) attempted

196 

The Iberian Peninsul a bet ween 300 and 850

to classify hillfort occupations based on the social agents behind their developments.59 When looking at the social agencies behind the development of these hillforts, we must look at the material culture found in them. First of all, one of the most important characteristics of these occupations is the presence of defensive structures. Many of them have been described as poorly made and rapidly executed, which have fueled arguments attempting to relate hillfort occupations with unstable groups and the defence against invasions (Ariño and Díaz 2014). But on the contrary, they are very similar to the wall structures of many cities in the northern part of the Iberian peninsula dated to the late fourth or early fifth century like Astorga, León or Gijón, which cannot be related to any specific invasions, and are instead related to the actions of late Roman urban elites. Furthermore, comparing the construction of these walls and the contemporary buildings in the rural world (see below), the presence in these sites of elite social groups is evident. Only elites had the capacity to control enough capital in order to generate the surplus that enabled the construction of these structures. It is difficult therefore to support the idea of peasant and local groups being responsible for the construction of, at least, the majority of these sites. The presence of luxury finds in some of these sites further supports this hypothesis. The weapons in domestic contexts in Navasangil (figure 6.2), El Castillón or El Cristo de San Esteban, some special types of fibula, like the Viskov type located in El Castillón, or luxurious personal objects like a comb in Castro Ventosa are all good examples. Also, the documentation of productive spaces within these sites indicates the presence of the division of productive tasks and, potentially, social inequality within these communities. At El Castillón two phases of a metal furnace were documented in recent excavations (Sastre and Fuentes 2011) and in La Virgen del Tormejón it is thought that a pottery kiln existed (Gozalo et al. 2013). By the fifth century, hillforts seem to be not just political, but also economic central places. In this sense, it is interesting to note the relationship between the decline of the late Roman urban network and the density of these sites. Considering their geographical distribution, their relative size, and the social agents behind their development, there appears to have been three different types of territorial and social structuring of hillforts. 59 A classification which perhaps stretches the interpretation of the evidence too much. This tripartite classification includes hillforts built by peasant initiatives (Gutiérrez González 2014), those developed by local elites (Ariño and Díaz 2014; Castellanos and Martín 2005), and those linked to direct state intervention (Brogiolo 2000).

The new rur al l andscape

197

Figure 6.2 Air photograph of the excavations at Navasangil

From Caballero and Peñas, 2012

In the first case, there are large hillforts, found in relative isolation, which may underline their link to political initiatives, if not to direct state action. The site of Castro Ventosa (f igure 6.3) is a perfect example of this f irst model. This hillfort is located in the northwestern part of the northern plateau, within the mountainous region of El Bierzo, occupying a high elevated plateau of 5,3ha, being one of the largest hillfort occupations known for this period. It seems to have been the reoccupation and fortification of an Iron Age enclosure which has been related to the site of Bergidum in the sources (Mañanes Pérez 1988). The site is surrounded by an impressive wall 1115m in length and it has been excavated several times since the 1980s. The archaeological works have focused on both the walls (related to a restoration project) and inside the enclosure (Díaz Álvarez 2003; Strato 2004), where some palynological studies have been carried out (Merino et al. 2008). These interventions, although very limited in extension, have documented some significant contexts like dump areas. In these dumps important data about the economy and waste management of the site have been retrieved. For instance, the impressive material

198 

The Iberian Peninsul a bet ween 300 and 850

Figure 6.3 Walls at the site of Castro Ventosa

From Strato, 2014

culture of the site includes not only luxury objects (like the comb already mentioned) and imported pottery, which is extremely rare in fifth century inland contexts, but also glass discards, which implies the presence of glass production within the site (see Chapter 7). In view of the historic and material significance of the site, some authors have related Castro Ventosa with the expansion of the Suevic monarchy in the fifth century (Díaz Martínez 2011). A second type would be those medium sites forming a lose network with other neighbouring medium hillfort sites. Such settlements seem to function as central places within their territory. Cabeza de Navasangil is a perfect example of this second model. It is located in the central part of the current province of Ávila, in the Amblés valley (Caballero and Peñas 2012). Excavations have identified a stone wall forming a 2ha enclosure, and unearthed three domestic units: two rectangular stone-based houses and also a possible granary outside the walls, where a significant amount of burnt seeds and a spatha (sword) were found. The hillfort seems to have been focused on the territorial control of the valley of the Amblés and the villages and agricultural resources of the region (Blanco et al. 2009).

The new rur al l andscape

199

The last model is that of small hillforts concentrated in the same territory. These seem to be related not only to urban decline but also to the gathering of most parts of local communities in to these hillforts. In the territory between the valleys of the rivers Voltoya and Eresma, in the northern part of the current province of Segovia, four (or even five) of these sites have been documented, all of them placed within 3km of each other (Tejerizo García 2014b). One of them is the ancient Roman town of Cauca, which seems to have been transformed into a hillfort during the fifth century, when its role as a central place within the territory was in decline (Martínez and Tejerizo 2015). All of these settlements are characterized by their small size, the building of walls and, as we have seen, the presence of productive spaces. The location of this network of hillfort occupations in direct relation to the Eresma River may show the existence of local communities and elites trying not only to centralize productive spaces but also to take the control of the routes along the Eresma. This could perhaps be linked with cattle paths in association with farming activities in the surrounding areas. Activities inside the hillforts With regards to the domestic architecture detected in first generation hillforts, the most important feature is that they are mostly constructed in stone directly on top of the geological level. Due to this, there is an evident absence of sunken featured buildings, which only appear to have been dug in soft soils. Houses among different sites have a similar conception even though there are slight differences. At the site of El Cristo de San Esteban two houses were excavated (Domínguez and Nuño 2014). The ‘southern house’ is a single-ailed rectangular construction, 31m 2 in extent, built in stone and excavated taking advantage of the slope. A posthole was documented in the centre of the structure, which probably means there was a pitched roof. In Navasangil, the common domestic architecture is a 1-to-3-roomed house, about 80m2 in area, built in stone and with an internal stone bench (Caballero and Peñas 2012). In Dehesa de la Oliva, on the contrary, it has been suggested that only the footing was made of stone, while the rest of the building would have been constructed in wattle and daub (Vigil-Escalera 2012). In this last site, the excavation of a cluster of four dwelling units enabled the analysis of the settlement organization, whose main characteristic was the practical absence of empty spaces in between the buildings. The same intense use of the space within the site has been confirmed in other hillforts, like El Castillón or El Cristo de San Esteban. Moreover, it has been suggested that, in the absence of empty spaces and of silos, storage

200 

The Iberian Peninsul a bet ween 300 and 850

was made in granaries, like the one excavated in El Castillón (Sastre and Catalán 2012),60 or perhaps in internal basements within the houses, which would imply two-floor houses (Vigil-Escalera 2012). Finally, excavations have provided important data about the economy and patterns of production of these sites, which seem to be not only complex, implying a strong division of labour, but they were also the main production centres for wide areas of the Iberian peninsula in the fifth to sixth centuries. Thus, in El Castillón several metallurgic furnaces have recently been excavated (Sastre and Fuentes 2011). We have already noted the potential presence of glass production in Castro Ventosa, and a possible pottery workshop in La Virgen del Tormejón (Gozalo et al. 2013). This rather specialized production coexisted with agricultural practices, as the burnt grain in Navasangil and also in El Cristo de San Esteban demonstrate. In Castro Ventosa a faunal study has revealed the existence of cattle, sheep, pig and deer, showing not only the presence of animal husbandry but also high status individuals (hunting) within the site (Díaz and Garín 1999). Second-generation hillforts (seventh-eighth centuries) Most of the first-generation hillforts were abandoned during the sixth century, probably as a result of several factors: the strengthening of the Visigothic monarchy during its process of state formation (see Chapter 5), the expansion of the network of farmsteads and villages (see below), or the introduction of regional elites into larger political scales (Castellanos and Martín 2005; Martínez and Tejerizo 2015). After a gap of approximately three or four generations, another phase of hillfort occupations appeared in the Iberian rural world. Even though there is much less information about them due to the absence of archaeological works, some case studies show that they were occupied from the late seventh and into the eighth centuries. This reactivation of hillforts as an effective mode of territorial control probably happened because of the new needs of the Visigothic monarchy and the regional elites (Quirós Castillo 2012a). This phenomenon is detected mostly in the northern and eastern parts of the Iberian peninsula. In the Ebro valley, a group of hillfort sites is documented in the seventh and eighth centuries, including the recently excavated Castillo de los Monjes, interpreted as a node within a military system whose function would have been to protect the Ebro valley towns (Tejado Sebastián 2012). Another important area to point out is the region of 60 Although it is very difficult, as far as we know, to confirm if thiswas a communal granary or a private one.

The new rur al l andscape

201

Figure 6.4 Defensive system at Tedeja

Based on Palomino et al. 2012

Trespaderne, in the Northern Meseta, where a significant hilltop site with a powerful wall, Tedeja (figure 6.4), is surrounded by a group of important churches, showing the presence of a powerful and religious elite in the region (Lecanda and Palomino 1999; Palomino et al. 2012). Even though the wall has been recently restored, unfortunately there is not much archaeological information about the site apart from some important radiocarbon dating, which shows the presence of, at least, two main phases within the site. Another important site in the northern part of the Iberian peninsula is the hillfort of Gauzón, which is the object of several archaeological interventions (Muñiz and García 2010; 2012). The site, which was firstly occupied in the fifth century, was radically transformed into a powerful defensive hillfort during the eighth century, probably related to the creation of the kingdom of Asturias. The ‘Victory Cross’, a famous Asturian work of art, was made there according to its inscription, demonstrating the symbolic importance of such places in the configuration of post-Roman polities. Finally, it is worth mentioning the presence of a group of hillfort sites on the Mediterranean coast of the Pyrenees. Two of the best known examples are the sites of Sant Julià de Ramis and Puig Rom, which may be related to a military frontier (the clausurae) due to the presence of a significant group of weaponry. They are dated to the seventh-eighth centuries (García and Vivó 2003; de Palol Salellas 2004).

202 

The Iberian Peninsul a bet ween 300 and 850

Farmstead and village networks and other lay rural settlements One of the most significant processes of the period between the sixth and the eighth centuries was the extension of a network of farmsteads and villages in most parts of Western and Central Europe (Hamerow 2002, 2012; Peytremann 2003; Quirós Castillo 2009; Schreg 2012). The Iberian peninsula was not isolated from this process, but it has only been archaeologically recognized and analysed in recent years (Quirós and Vigil-Escalera 2006). This network of farmsteads and villages, as we will see, had different regional developments, but it can be considered to be the expression and consequence of post-Roman peasant agencies in the construction of the landscape and its interaction with local and regional elites (Tejerizo García 2013; 2017). Although some post-Roman villages in the Iberian peninsula (figure 6.5) like the sites of Lancha de Trigo (Gutiérrez et al. 1958), El Castellar (García et al. 1963) or El Bovalar (Palol 1986), have been known about since the mid-20th century, it was not until very recently that they were recognized as a specific phenomenon of the post-Roman period (Azkarate and Quirós 2001; Fernández Mier 1999; Vigil-Escalera 2000). Although it has been believed that the Iberian peninsula is located at a far distance from other traditions, thanks to recent archaeological work, the way in which Early Medieval landscapes and peasant societies are now understood is changing (Kirchner Granell 2010a; Quirós Castillo 2009). Furthermore, it is necessary to stress that practically all our information comes from the northern part of the Iberian peninsula (in Madrid, Catalonia, the Basque Country, Castile and León and northern Portugal), where such studies are more developed. Considering this, we will describe here the main features of this phenomenon through some case studies, and we will also evaluate the current state of knowledge and the future agenda. The network of farmsteads and villages has five characteristic features which need to be pointed out. The f irst one is their geographical position, normally in flat areas in direct connection to water resources, whose management is essential to the development of agricultural economies. Traditionally, these villages have been associated with old Roman roads that would act as the poles of attraction for post-Roman settlement patterns (García Merino 1975; Mañanes and Solana 1985). However, new studies have demonstrated that such a relation is spurious; even if Roman roads were in use in later periods, they were not necessary the main element for the organisation of these sites, as we can find villages that were quite far from Roman roads (Tejerizo García 2013; Vigil-Escalera 2007).

The new rur al l andscape

203

Figure 6.5 Plan of the excavations at El Bovalar

Based on Palol 1986

Moreover, many of these Early Medieval villages can be found in direct stratigraphic superposition over pre-Historic settlements, mainly from the Chalcolithic, the Bronze Age and even the beginnings of the Iron Age. This superposition has been a main cause of confusion, and some contexts have been wrongly dated as prehistoric rather than sixth or seventh century. That is the case of Canto Blanco, where many of the silos were thought to be prehistoric (Strato 2010). Some authors have affirmed that this relationship

204 

The Iberian Peninsul a bet ween 300 and 850

Figure 6.6 Stone-footed building at El Pelícano

Photograph published in Vigil-Escalera and Strato, 2013

between Prehistoric and Early Medieval contexts is related to a kind of ‘rebirth’ of ancient societies within the context of the barbarian invasions (Lewit 2003; Rascón et al. 1991). We would rather side with those scholars who have proposed that such similarities are related more to the importance that peasant agencies had in both periods, prompting a similar settlement pattern (Tejerizo García 2013; 2015). The second characteristic is a common pattern of domestic architecture. It usually combines the use of both stone-footed buildings and sunken featured structures which formed each domestic unit in each village. Stone-footed buildings (figure 6.6) are usually rectangular, around 30m2 in internal space and constructed in daub or wattle-and-daub techniques as those seen in villages like Gózquez, El Pelícano or La Mata del Palomar (Vigil-Escalera and Strato 2013). They have usually been interpreted as the main dwellings for the domestic units, where the main functions of production and reproduction took place (Vigil-Escalera 2003). Such spaces were commonly organised with a cluster of ancillary buildings for complementary functions, like storage, weaving or food production (milk, beer, etc.) in the form of sunken featured buildings (Tipper 2004).

205

The new rur al l andscape

Figure 6.7 Table indicating the relative proportion between the number of sunken featured buildings excavated per site (first column) with the overall excavated area (second column). The result shows it is necessary to excavate quite large open areas in order to identify or find a relevant number of sunken featured buildings Site

Cárcava de la Peladera (Segovia) Mata del Palomar (Segovia) Navamboal (Valladolid) Ladera de los Prados (Valladolid) Canto Blanco (León) Congosto (Madrid) Gózquez (Madrid) La Huelga (Madrid) La Indiana (Madrid) El Pelícano (Madrid) Encadenado (Madrid) Can-Gambús 1 (Barcelona) Els Mallols Aiguacuit (Barcelona) Zaballa (Álava) Total

Sunken featured buildings

Area excavated (in m2)

Square metres excavated per sunken featured building

4 22 2 4 6 39 69 2 10 44 4 10 7 3 9 206

2013 5588 4000 7077 10557 13200 23900 5200 15415 54150 4000 17000 12000 10000 35000 Average

503 254 2000 1769 1759 338 349 2600 1541 1231 1000 1700 1714 3333 3889 1598.6

Sunken featured buildings and silos, hardly used in Roman times, became common domestic structures in farmsteads and villages across Europe (Chapelot and Fossier 1985; Hamerow 2002; 2012; Peytremann 2003). In contrast with the wider European context, the sunken featured buildings in the Iberian peninsula are characterized by their irregularity (figure 6.7) and lack of homogeneity (Tejerizo García 2012b). Even though most of the known European typologies (sub-rectangular features with 2, 4 or 6 postholes) can be found in the Iberian peninsula, the most common sunken featured buildings are sub-rounded and without postholes. This general absence of internal postholes may respond to its location outside and at a higher level, as seen in some examples in Catalonia (Roig Buxó 2009). Some other sunken featured buildings were present in these contexts, which are characterized by irregularity in their shapes. These can be mainly related to quarry works and water management, as has been proposed in other similar contexts in France (Catteddu 2001). Some other rare examples of specialized sunken featured buildings are those detected in Plaça Major

206 

The Iberian Peninsul a bet ween 300 and 850

Figure 6.8 Plan of the excavations at Gózquez

Based on Vigil-Escalera and Strato, 2013

del Castellar del Vallès, which have been interpreted as specialized areas for the production of bread (Roig and Coll 2010) or those that present inner division walls, like some examples in Gózquez (figure 6.8; Tejerizo García 2014a). Apart from stone-footed structures and sunken featured buildings, there are other structures documented in villages that should be mentioned. For example, the presence of wells, which although not very common is documented in sites like Canto Blanco, Gózquez or Senovilla. Also, the presence of stone boundaries, potentially used as markers for the domestic units, has been detected in contexts like La Mata del Palomar or La Cárcava de la Peladera. Very unusual is the structure observed in the village of Cañal. This context was studied in the 1980s before the construction of a dam – the village is currently under water – and it consists of around ten grouped domestic units, forming small clusters, without any hint of internal organization. This organization of the site would show the importance of the domestic unit in the development of the social organization of Early Medieval villages, as suggested by Alfonso Vigil-Escalera (2007). Surrounding the village, two long linear walls were documented and interpreted as a fence for cattle, serving to create a communal identity (Fabián et al. 1985). The third characteristic is their density. Different studies carried out in the province of Madrid have determined that these villages or farmsteads can be normally detected within ranges of 2 to 4km from each other, forming a complex network (Vigil-Escalera 2007). This high density has been detected in other regions like the Duero valley (Tejerizo García 2013) or Catalonia (Roig

The new rur al l andscape

207

Buxó 2009). However, what is difficult to assess is the presence of a hierarchy within these networks, as has been proposed for the Basque Country (Quirós Castillo 2011a). There are some hints of economic specialization that may show this hierarchy in local territories like Madrid (Vigil-Escalera 2007) but currently there is not enough data to further support this theory. A fourth element that should be underlined is regional variability. For instance, in the centre of the Iberian peninsula and Catalonia, there appears to exist a correlation between the high density of farmsteads and villages and the abandonment of Roman villas, which contrasts with the northwestern regions, where this correlation in the landscape is less evident for this period (Fernández et al. 2014; Sánchez Pardo 2013). In the case of the Basque Country, recent studies have also shown the presence of a very disperse pattern of rural villages up until the eighth century (Quirós Castillo 2011a). In the mountainous regions of Asturias, where important works are being carried out in the village of Vigaña (Fernández et al. 2014), villages show a highly dispersed settlement pattern, but with a complex social and economic articulation as indicated by the presence of local elites within the villages. In contrast, in the northern part of Portugal and the western part of the Meseta the irregularity of the rural settlement pattern and the absence in this period of clear articulated sites of power may demonstrate not only an absence of local and regional elites, but a high presence of peasant agency over the landscape (Martín Viso 2009a; Tente 2009). The last point to stress about the rural landscapes of the Iberian peninsula is the complexity of its economy and social structure. Traditionally, these communities have been seen as unstable and even ‘primitive’. However, recent studies at sites such as Zaballa have underlined the presence of complex economic strategies (Vigil-Escalera and Quirós 2013). In the case of Gózquez, bioarchaeological studies have shown not only a clear diversified economy and a social organization of the sites – with the presence, for example, of specialized areas of production and storage – but also the presence of olive oil and possibly even horse breeding (Vigil-Escalera et al. 2013). This economic complexity has to be understood in the anthropological context of peasant societies, in what Chris Wickham has conceptualized as the ‘peasant mode of production’ (Wickham 2005: 535), whose main aim is to minimize risks and diversify production within an apparent equality of the domestic units (Halstead and O’Shea, 1989). This is matched by the funerary evidence, which shows a very high social complexity. Gózquez is perhaps one of the most important villages that have been excavated, due to its size that covered near 2,3has (Vigil-Escalera and Strato 2013: 155-177). This site, located on a slight slope over a stream, was founded ex

208 

The Iberian Peninsul a bet ween 300 and 850

Figure 6.9 Aerial photography of the site of Zaballa

Photograph by Prof. Juan Antonio Quirós, used with permission

novo in the second quarter of the sixth century. Excavations have unearthed two areas of domestic units, divided by a cemetery, being one of the unique cases in which both spaces have been found together in the same site (see below). Each domestic unit was composed of a stone-footed building and a group of ancillary sunken featured buildings and even some specialized structures, like an olive press. These domestic units seem to move in generational cycles within each plot of land. The village was abandoned by the mid-eighth century, at a time of strong regional transformations. Another good example of an Early Medieval village is the site of Zaballa (figure 6.9), the first village to have a complete monographic publication in the Iberian peninsula (Quirós Castillo 2012b). The nearly 4,5has excavated have discovered a long lasting village, dated from the sixth century to present times. Phases 1 and 2 correspond to the post-Roman period. Between the sixth and seventh centuries, Zaballa is thought to have been a little farmstead composed of three sunken featured buildings made of daub in a context of scarce impact over the landscape, as pollen analysis has shown. By the eighth century, the little farmstead was transformed into a village consisting of a group of domestic units which incorporated stone-footed buildings and silos. The development of this community had a greater impact on the

The new rur al l andscape

209

landscape, with a reduction of the forest and the construction of agrarian terraces. The study of bioarchaeological remains (seeds and animal bone) have shown the presence of a diversified and hardly specialized economy which allowed the maximization of production which minimized the risks linked to specialized crops. Other relevant sites excavated over the last decades which show the same patterns as these are El Pelícano, Canto Blanco, La Mata del Palomar, Ladera de los Prados, Plaça Major de Castellar del Vallès, Can Gambús 1 or Berrocales (Roig Buxó 2009; Roig and Coll 2010; Tejerizo García 2015; Vigil-Escalera and Strato 2013). These villages have shown not only their importance in the constitution of post-Roman social landscapes, but also the outstanding possibilities of the archaeology of villages within Western European contexts (Quirós Castillo 2009).

Funerary rituals in the rural world As seen in Chapter 3, during the late Roman period there were some important changes in funerary rites. The so-called post-Imperial necropoleis, which were characterized by the use of particular grave goods related to a late Roman identity, dated to between the second half of the fifth century and the first decades of the sixth century, can be considered one of the major archaeological changes in the rural world. It was at this point, related to the expansion of the network of farmsteads and villages, that new forms of funerary ritual appeared all along the Iberian peninsula. Traditionally, studies on these new funerary rituals focused on what we called here ‘communal cemeteries’, traditionally characterized by the presence of ‘German-style’ grave goods and related to the Barbarian invasions of the fifth century and the presence of Visigothic peoples in the territory (de Palol Salellas 1966; Ripoll López 1989; 1991). However, due to the advance in archaeological studies and the implementation of new archaeological and anthropological methods of study, this interpretation has been questioned and reformulated from various points of view within a context of European debate on ethnicity and identity.61 61 It is impossible to summarize all the contributions to this debate. However, some of the most important would be Brather 1998; Brogiolo and Chavarría 2008; Curta 2013; Chavarría Arnau 2012; Geary 1983; Gillet 2002; Goffart 2002; Hakenbeck 2004; Halsall 2011; Härke 2007; Jones 1997; Pohl 1998, 2013a; Quirós and Vigil-Escalera 2011; Tejerizo García 2011; Valenti 2009; Von Rummel 2013.

210 

The Iberian Peninsul a bet ween 300 and 850

In this section we will highlight the most important characteristics of the funerary rituals in the post-Roman rural world of the Iberian peninsula focusing on three main ideas. Firstly, we will describe the complexity of funerary rituals and phenomena beyond the ‘communal cemeteries’, which have centred the debate but have also obscured other forms of post-Roman funerary rituality. Secondly, and consequently, we will underline the complexity of identity and the construction of ethnicity in post-Roman times. Ethnicity, therefore, has to be understood only as a part (a layer) within an intersectional identity (Halsall 2007, 2011; Pohl 1998). Thirdly, the importance of entangling the funerary rituals, the world of the dead, with other archaeological records regarding the world of the living needs to be properly understood. As M. Parker Pearson has proposed, the dead are ‘powerful’ and tend to distort the fundamental idea that the funerary world tells us about the living communities (Parker Pearson 1993). Overall, what characterizes the funerary practices in the Iberian peninsula in the post-Roman rural world is, precisely, its complexity. In recent decades a significant amount of funerary contexts have been excavated and they have shown a wide variety of forms and types of burial. In order to summarize all this information, we can divide them into five main types, which are not necessarily exclusive. Thus, we can find more than one type in the same archaeological context. Pit burials The first type is a particular kind of funerary context: the pit burial. This is characterized by the presence of one or two (but usually no more than three) individuals inside a pit, sometimes associated with other remains like animal bones (figure 6.10). These have been documented mainly in the central part of the Iberian peninsula and Catalonia, precisely in those areas where excavations have been larger in extension, suggesting that it could be a more widespread phenomenon, but not noticeable in the archaeological sample. Although it is a very common way of burying the deceased, it usually appears in low quantities in each site, in a number not usually higher than five. Exceptions are the Catalan sites of Can Gambús 1, (15 individuals), La Solana (10 individuals) or Els Mallols (8 individuals) (Roig and Coll 2011). In general they are detected in villages, but also in relation to churches, as the case of Sant Menna shows (Roig Buxó 2015). In a recent revision of this phenomenon, Alfonso Vigil-Escalera has proposed that these individuals, which are usually associated with animals and therefore treated like them, are being excluded from the communal cemeteries of the villages and, thus,

The new rur al l andscape

211

Figure 6.10 Pit burial at the site of La Mata del Palomar

Photograph published in Vigil-Escalera and Strato 2013

that it can be the reflection of their low social status. However, the fact that they are, at a basic level, following the same structural rite, may show rather their belonging to the community, being instead perhaps, slaves or subordinate groups of individuals, perhaps associated with the different domestic units (Vigil-Escalera 2013b). Isolated burials A second type of burial in rural sites is the isolated burial, documented all over Europe (Catteddu 2009; Fehring 2009; Hamerow 2002). These are usually presented in a dispersed pattern in clusters of one to five burials and usually geographically related with the domestic units. It is also noteworthy that, in the majority of cases, these burials are dated to later contexts, mainly the seventh or the eighth centuries. There are not many examples in rural settlements in the Iberian peninsula because they are usually documented through excavations in large areas, such as Gózquez, El Pelícano or Las Lagunillas. A typical case may be La Mata del Palomar. In this site a group of five to seven domestic units dated between the sixth and mid-eighth century have been documented. Within these domestic units there are seven burials divided in to two groups. It is interesting that in three of them grave goods have been documented: all common instruments such

212 

The Iberian Peninsul a bet ween 300 and 850

as a metal spoon, a sickle and a chisel. It is also noteworthy that in the case of Gózquez, for example, the isolated burials are contemporary to the development of the communal cemetery (see below) which led the excavator to propose that they may be the last manifestation of prestigious burials in a typical post-Roman tradition (Vigil-Escalera 2009: 11). However, other interpretations have proposed that these burials are the reflection of the capacity of the domestic units in a context of social tension to choose to place their funerary spaces within their properties, as a way of claiming, by means of their ancestors, their legitimacy over the land (Hamerow 2012; Martín Viso 2014). Church burials Another type of funerary context that can be found in the rural world are the cemeteries related to religious centres (which are not necessarily burials ad sanctos).62 As we will later discuss, the expansion of Christian institutions and buildings (churches, monasteries and shrines) in rural contexts is, in general, a process that began in the sixth but mainly developed in the seventh century. The introduction of these sacred spaces generated in many cases new areas of funerary rituality, even though the archaeological evidence is very scarce because of the existence of many ancient excavations dealing with this issue. However, we do know that many churches dated between the sixth and the ninth centuries had burials, both inside and outside the building. Some good examples of these are in the central part of the Iberian peninsula: Quintanilla de las Viñas (Camps y Cazorla 1939-1940), Saint John’s in Baños (de Palol et al. 1983), Santa María de Mijangos (Lecanda Esteban 2000) or Aistra (Reynolds and Quirós Castillo 2010). In Catalonia, other good examples are Santa María de Artés, Sant Menna and Santa Margarida de Martorell, characterized by its limited extension (in Sant Menna, 57 burials within a 400 year range) and the absence of furnished burials (Roig Buxó 2015). Also in the cases of recognized monasteries the presence of a funerary space has been documented, such as in Melque (Caballero and Moreno 2013) and, perhaps, in Dueñas (Fita 1902). These religious centres, once built in a given territory, acted as new poles of rituality, evidently and radically transforming, even though we are not capable of evaluating at what level, the topographies and identities of local communities. 62 Ad sanctos burials usually pre-date the construction of churches, as they are linked to smaller martyrial shrines in necropoleis.

The new rur al l andscape

213

Figure 6.11 Plan of the excavations at Duratón

Based on Molinero 1948

Communal cemeteries – Reihengräberfelder? The fourth type of cemetery is the communal cemetery (figure 6.11), which should not be exactly equated with the traditionally labelled ‘Visigothic cemeteries’ (Ripoll López 2007). The communal cemeteries are mainly characterized by a relatively high number of burials – even though there are few cemeteries fully excavated – with a regular organization (what has been called in continental literature Reihengräberfelder) and density of use of the space and the presence of furnished burials among them, even though they are variable and changes over time. As these cemeteries have been extensively studied, especially from ethnic perspectives and attention paid to the material culture (see Chapter 4), here we will highlight only a few questions related to the typology of the burials and their relationship with their settled environment. A regional analysis shows that these types of cemeteries are widely distributed and this often coincides with the areas where the network of

214 

The Iberian Peninsul a bet ween 300 and 850

farmsteads and villages was most developed. This can be clearly seen in the examples of Cataluña (Roig Buxó 2015) or Madrid (Quirós and VigilEscalera 2011) and, in particular, in cases such as Can Gambús-1 or Gózquez (Contreras and Fernández 2006), where we have both the cemetery and the dwelling spaces. Even though there are evident differences among them, what these cemeteries may be signaling is the presence of a village (or villages) in its surroundings. In other words, the distribution of these communal cemeteries coincides with the distribution of the network of farmsteads and villages in the post-Roman period. The differences in the number of burial, the proportion of furnished burials or the variability in the typology of the burials would be differences in the social structuring and the type of village they are serving as funerary space. Another important characteristic is that communal cemeteries usually show a long persistence over time, normally of two or three centuries. In line with what was just described, this persistence relates to the use of the domestic spaces of the village(s) and, consequently, with a high stability of territoriality and social organization of common spaces. Sites like Espirdo-Veladiez, Madrona, Gózquez or Castiltierra (Arias et al. 2000; Arias and Balmaseda 2015) show long chronological ranges (Contreras Martínez 2011; Jepure 2004, 2012). However, what has attracted more attention is the presence of a particular type of furnished burials with ‘German-style’ objects, the so-called inhumations habillées (Kazanski and Périn 2008), which appear in the archaeological record mainly between the fifth and seventh century, with some exceptions that continue into the ninth (figure 6.12). Regarding this, some points should be underlined in order to better comprehend the phenomenon. In the first place, not all furnished burials are inhumations habillées as there are many cemeteries, such as Piñel de Abajo (Arranz et al. 1989) or Las Quintanas (Velasco et al. 2003), which do not show the presence of these types of inhumations. In fact, if we place them in context with the number of burials overall, then we can consider these burials to be rather an exception than a rule. In second place, it is necessary to consider these objects in the chronological context in which they are present, that is, the development of the network of farmsteads and villages we have previously described. Comparing the type of material culture we can detect in the domestic spaces with these types of objects, it is evident that they are not only extremely rare and luxurious items but also that they are mainly conceived for the funerary ritual. As a result, it is extremely rare to find them in domestic spaces outside the cemetery. Furthermore, recent studies in Frankish brooches from France have determined an extremely complex technological chain of

215

The new rur al l andscape

Figure 6.12 Table depicting the different proportions of furnished burials and inhumations habillées in various excavated communal cemeteries SITE

Gózquez Daganzo de Arriba Cácera de las Ranas Tinto Juan de la Cruz La Indiana Duratón Herrera de Pisuerga Aguilafuente Espirdo-Veladiez Madrona San Miguel de Neguera- Sebulcor Ventosilla y Tejadilla Deza El Carpio de Tajo Segobriga Martels Boadilla – Alameda del Señorío Estagel TOTAL

NUMBER OF BURIALS

NUMBER OF FURNISHED BURIALS

NUMBER OF INHUMATIONS HABILLÉES

356 35 150 80 50 666 52 198 64 351 10 15 ¿100? 275 234 52 176 208 3072

121 (34%) 11 (31,43%) 64 (42,67%) 14 (17,5%) 6 (12%) 337 (50%) 36 (69,2%) 90 (45,45%) 16 (25%) 165 (47%) 2 (20%) 6 (40%) ¿30? (30%) 90 (32,72%) 40 (17,09%) 20 (39%) 46 (26%) 121 (58%) 1215 (39,5%)

12 (3,3%) 2 (5,7%) 6 (4%) 3 (3,7%) 0 (0%) 89 (13,36%) 13 (25%) 28? (14,14%) 4 (6,25%) 73 (20%) 1 (10%) 1 (6,7%) ¿5? (5%) 19 (7%) 0 (0%) 1 (2%) 2 (1,1%) 10 (4,8%) 269 (8,7%)

production which includes some raw materials brought from India (Calligaro et al. 2006-2007).63 Thus, the existence of furnished burials mainly indicates the existence of social differences and the various ways of identifying status and rank within these rural communities. Finally, it is interesting to note some regional differences which introduce an element of complexity to the phenomenon. On a regional scale it appears that the presence of furnished burials and inhumations habillés is widely spread in both Mesetas, but it is very rare in, for example, the northwestern part of the Iberian peninsula (Sánchez Pardo 2010) or in Catalonia, with only 73 furnished burials among 1585 known (Roig Buxó 2015). In this regional sense, what stands out is the presence of a particular burial rite in the northern part of Spain, with the presence of ‘Frankish-type’ objects in the 63 Similar long-distance trade of high-quality objects can be traced to Spain, for example, the Guarrazar Treasure.These rare objects show the presence of particular individuals who have the capacity of having them imported within the context of the post-Roman village society.

216 

The Iberian Peninsul a bet ween 300 and 850

so-called ‘Aldaieta-horizon’ related to the influence of the Frankish state in these regions (Azkarate Garai-Olaun 2004a). Rock-cut burials The fifth type of post-Roman burial type is the rock-cut (figure 6.13). Traditionally, these have been associated with the ‘repopulation’ of the northern and central part of the Iberian peninsula and, thus, dated from the ninth century onwards through some paradigmatic cases such as Cuyacabras or Villanueva de Soportilla (Castillo 1970, 1972). However, recent excavations in southern cemeteries, such as Nívar, Tózar or Martilla (Jiménez et al. 2011) and recent regional revisions of the northern and central parts of the Iberian peninsula and Portugal (Brooks et al. 2017; Martín Viso 2012; Quirós Castillo 2011b; Rubio Díez 2015; Tente 2009), have reinterpreted the chronologies, typologies and social implications of these cemeteries. The most recent chronological proposal for these cemeteries is positioned between the seventh and the eleventh, not discounting an earlier beginning (Martín Viso 2014). This chronology is interesting for two reasons. Firstly, it shows that these types of cemeteries are contemporary with the other types we have described here, introducing an element of complexity to the conception of funerary phenomena in the post-Roman period. Secondly, it shows a relative degree of continuity through the eighth century and the Islamic invasion, which is relevant to historical analysis of this transition (see Chapter 8). Iñaki Martín Viso (2012), based on his research in some territories in the Meseta, has recently proposed a typology divided into three main types of rock-cut burials depending on the number of burials and their concentration: 1) isolated burials or small groups of less than ten burials, normally subdivided in several groups of two-five burials; 2) ‘disorganized cemeteries’, with more than ten burials but not forming any coherent concentration; 3) ‘grouped and aligned cemeteries’, with the presence of several groups of burials. It is important to note that these three groups are commonly found in the same regional context. A fourth group would include those cemeteries cut in rock but closer in their characteristics to the communal cemeteries and located in the north-eastern part of the plateau, like Villanueva de Soportilla, recently re-excavated and dated from the seventh century onwards (Aratikos 2010). The interpretation of these cemeteries in social terms is not easy, when taking into account the scarcity of data and the taphonomy problems derived from these types of burial. However, it is important to set out some important questions. In the first place, we have already pointed out that

The new rur al l andscape

217

Figure 6.13 Aerial photography of Villanueva de Soportilla

Photograph by Aratikos, used with permission

these cemeteries are not usually very large in number and they are normally found only in specific geographical spaces, such as the mountainous regions of central Spain and Portugal. In the second place, even though they were not furnished they were still quite complex in architectural terms: rock-cut burials require not only specialized individuals but also a high investment of time in which to carve them. This investment has been calculated, through ethnographic parallels, to be between two days and three months (Rubio Díez 2015). In the third place, they seem to be associated with a dispersed settlement pattern composed mainly of only one or two domestic units which would make use of these cemeteries. Recent excavations at La Genestosa have related some of these cemeteries in rock with some stone-footed buildings dated between the sixth and seventh centuries (Martín and Rubio 2013). So, as we have already pointed out in the case of isolated burials within the villages, it appears that these burials are markers for different families and domestic units within a dispersed settlement pattern as a form to claiming their territory and their lands through the means of their ancestors (Martín Viso 2014; Williams 1998). This short overview of the different forms of funerary spaces in the postRoman Iberian peninsula demonstrates, above all, its enormous complexity.

218 

The Iberian Peninsul a bet ween 300 and 850

This complexity in the variability of forms in which local communities understood their funerary rituals and buried themselves is a reflection of the complexity of their social identities, irreducible to a partial element, be it their ethnicity or status. Social identities in a material context need to be seen as social constructs which depend on the interaction of multiple factors. Ethnicity, status, gender, profession, rurality and urbanity, religion, Romanitas, etc., are only some of the interacting factors of social identities which are expressed and reinforced in funerary rituals as a means to normalize and regulate the social habitus of a determined context (Bourdieu 2000). The profound changes of post-Roman times in the Iberian peninsula (as a part of the Empire) led to a deep regionalization of social processes which implied a variety of forms of resilience of local communities. It is this (stimulating) variety and complexity that archaeology attempts to explain.

Churches, monasteries and ecclesiastical sites The final transformation that characterized the rural world in the postRoman centuries is the spread of Christianity. The construction of churches, funerary basilicas, and the development of Christian monastic communities are signs not only of the conversion of the rural populations, but also of their integration within the networks of power of the Church – which as we saw in Chapter 5 was already fully present in urban contexts. Chronologically, there are traces of rural Christianity already found from the second half of the fourth century, especially through Christian mausolea and other burials in villa contexts (see Chapter 4). According to some scholars, the slow process of the visual Christianization of the countryside developed throughout the sixth century but it is not until the seventh that it is clearly noticeable through the construction of churches (cf. Chavarría Arnau 2010). Nevertheless, this process varies from region to region, and it can be seen as happening in those areas which were heavily urbanized and Christianized at an earlier stage, such as Southern-Western Lusitania and Baetica. Church typology and chronology Even if they were not numerous, early rural churches in the Iberian peninsula show interesting typologies and layouts. One of these types poses a chronological conundrum, which is the focus of a very hotly contested debate. Some Late Antique rural churches are characterized by the presence of two liturgical spaces: two faced apses that determine the liturgy of these

The new rur al l andscape

219

Figure 6.14 Plan of the excavations of Casa Herrera, with its two phases: In dark grey the original AD 500 funerary basilica; in lighter grey the late sixth century expansion, linked to episcopal intervention

Caballero and Ulbert 1976

buildings and their internal spatial organization. The function of the eastern apse is clear, as it was used as the sanctuary. In contrast the western apse has been linked to both a martyr cult and to funerary rites. In these basilicas, the area between the apses was open to lay members, and it was arranged in a classical way: that is, with three long naves separated by columns, with a wider central nave. Most of these churches are located in the south of the peninsula. The most representative examples are Torre de Palma, which was built inside a huge late Roman villa in the second half of the fourth century; San Pedro de Alcántara, dated to the middle sixth century; and El Germo, built around the year 600 according to the results of the old excavations that also unearthed remains from other buildings placed close to the church (Ulbert 1968). The best example of this typology is, however, the basilica of Casa Herrera, built in the late fifth century (Caballero and Ulbert 1976; Sastre de Diego 2013; figure 6.14). Soon after its construction, the interior was fully filled by burials apart the area of the apses, where altar remains were found: a rectangular table in the eastern apse (the sanctuary) and, in the western apse,64 a sigma-shaped table that preserved a fragmentary 64 The space around the centre of the western apse, where the sigmatic altar would be located, was occupied by burials exclusively for children.

220 

The Iberian Peninsul a bet ween 300 and 850

inscription perhaps related to martyrs’ relics. The apses were joined by a narrow corridor, a liturgical connexion that crossed the central nave. Another typology of rural church, similar to the double-apsed basilica is the contra-chorus church, which has also been recorded in urban contexts (such as El Tolmo de Minateda and Parc Central at Tarragona). This type is characterized by an internal separation at the western end of the central nave, delimiting a room by chancels and, sometimes, even by walls. Most of these buildings had a funerary use, and the western room at the opposite end from the apse was used to hold a privileged tomb. Examples of this kind of church are Gerena (c. 400), El Bovalar (from 450, de Palol Salellas 1989), and, in Mallorca, Son Peretó (from the fifth century, Riera et al. 2012) and Son Fadrinet (c. 600, Ulbert 2003). This type of church used to have a baptistery in the western part of the building, sometimes added to the original building. The presence of baptisteries became a common phenomenon from the late sixth century on, usually added to pre-existing basilicas as part of a secondary phase. One of the most interesting examples of the typology of contra-chorus church is the funerary church of El Bovalar. As in Casa Herrera, it has a central corridor which connected the so-called western contra-chorus with the eastern liturgical area, a triple space where the sanctuary, in the centre, stands out between the others. The church of El Bovalar, like the rest of the settlement where it was built, was in use until the middle of the eighth century, coinciding with the Islamic invasion. The case of Son Peretó in Mallorca is also quite interesting from both a cultural and a liturgical point of view (and in other Balearic churches as well, such as Es Cap des Port and Son Fradinet). Remains of the altar where found in situ inside the sanctuary. This altar is rather different from the common Late Antique Hispanic altars: the relics were put under the table under the paved floor, rather than in the box (loculus) opened inside the superior part of the ara as was common in the Peninsula (Sastre de Diego 2013). The liturgical change that shows Son Peretó, together with the presence of mosaic floors, could reflect a closer relationship with African and Eastern Mediterranean influences. There is a third type of rural church that stands out amongst the others for two main reasons (figure 6.15). One is its centred layout, mostly a cross shape inside a basilica plan, whose different spaces were covered by vaulted roofs. The second is the chronology. Due to their construction system, and more technological complexity, scholars do not agree about whether such buildings belong to Hispanic Visigothic Art (seventh century), or if they instead show the arrival of a new technology to the Iberian peninsula (ninth century). This last interpretation is closely related to the introduction of new

The new rur al l andscape

221

Figure 6.15 Front of the church Sain John in Baños. As it stands today, the building is not Visigothic, but it could be built with reused material from an earlier Visigothic church

construction techniques during the process of Umayyad state formation (Chapter 8), and they would demonstrate a response of the Christian communities, both inside and outside of al-Andalus (Caballero Zoreda 2000). Currently there is no consensus (Caballero and Utrero 2008; cf. Chavarría Arnau 2010; Wickham 2010; Sánchez et al. 2017; Sastre and Utrero 2015). Several open questions are still under discussion and feed one of the richest debates in current Spanish research. More excavations and research without aprioristic conclusions are still necessary. Besides the plan typology, these churches are characterized by the use of ashlar masonry and vaulted roofs supported by horse-shoe arches, as in Saint Mary’s in Melque (Caballero and Moreno 2013), Saint Lucia in El Trampal (Caballero and Sáenz 1999), Saint Mary’s in Quintanilla de las Viñas, Saint John’s in Baños, Saint Peter ‘de la Nave’, Saint Comba’s in Bande, etc. Beyond the chronological discussion, it is worth pointing out some obvious material features. Firstly, the design of these churches breaks with late Roman patterns. None of these churches are basilical in plan, as most of the known excavated churches have been dated to the post-Roman period. They are much smaller,

222 

The Iberian Peninsul a bet ween 300 and 850

and with a new distribution (including side apses and transepts). Secondly, they were vaulted or had horse-shoe arches. Similarly, they were built with ashlar blocks, which sometimes appear to be new, abandoning the traditional reuse of material that characterized Late Antique architecture. One of most controversial churches of this type is Saint John in Baños, famous because of its dedicatory inscription dated to the reign of King Recceswinth (dated to AD 661) mentioned in the previous chapter. Both the inscription slab and the decorated friezes, which on close examination are obviously not in their original sequence, seem out of place. The inscription is high up above the nave, hardly visible from the ground, and the carved friezes are re-used, and do not always fit or match. In fact, its lettering is more similar to later medieval inscriptions than to ‘contemporary’ Visigothic ones, as if it were a re-carving or a copy of an earlier original. There must have been a church dedicated by Recceswinth in the area, but it is not necessarily the church which is currently standing. On this issue, A. Chavarría has pointed out (Chavarría Arnau 2010) that the main problem with these churches is not the chronology of the buildings, but the interpretation that can be given to them: whether or not they can be inserted in post-Roman or Early Medieval social and economic contexts. This argument, however, could be turned around, because both chronology and the development of social processes are two sides of the same coin. It would perhaps be more advisable to use the archaeological chronology of the churches to study the society, rather than inferring the date from historical aprioristic paradigms. In this sense, unfortunately most of the known rural churches have only been partially excavated. The traditional archaeological projects focused only on the religious standing building, and it is not possible to know the type of settlement in which they existed. Cases such as the aforementioned Casa Herrera o El Bovalar are exceptional. Some of these churches seem to be in isolation, not linked to any settlement, so they cannot be interpreted as village parishes. This is, again, a general interpretation that needs to be confirmed by excavations. In a way, the seemingly isolated location of these churches has to be seen in relation with pre-Christian worship, which did not take place in settlements, but rather at ‘sacred’ spots in the landscape. The possible Christianization of such sites, or the translocation of cult sites to new Christian buildings should not, therefore, be ruled out as a possibility. The apparent isolated location of these early churches would imply that early rural Christianity was centralized upon these religious sites, and was perhaps an administrative measure by bishops attempting to control rural territories. This should not be surprising, considering that bishops were, in the Visigothic period, closely linked to the collection of taxes for the central

The new rur al l andscape

223

administration. Lastly, the construction of these churches can be seen as an indicator of the integration of rural elites (who were commissioning and building them) and the Visigothic state. Monasteries Until very recently, the archaeological remains for post-Roman monastic settlements was quite limited, and literary sources were the only evidence, especially the various monastic rules. Even if several Hispanic sources about monastic life talk about the first rules that were in use in Visigothic times (such as those written by Fructuosus and Isidore), the current archaeological evidence is not enough to establish either typologies or models of post-Roman Hispanic monasteries (Moreno Martín 2011). The general common element to these monastic rules is that the monks had to live in a community, ideally secluded from the outside world in order to focus on their religious life. Because of this, most Iberian post-Roman monasteries consisted largely of an enclosure, in which the main buildings that can be identified are living quarters and, of course, a church. Despite recent excavations, the number of confirmed monastic sites of the post-Roman period is still quite low. There is the possibility, however, that some of the known rural churches could have been linked to monastic communities, but the surrounding areas have not been excavated. The best known examples are perhaps Cullera, Dume, Cabrera, Melque, El Trampal, and the dubious case of Casa Herrera. Dume, as noted earlier (Chapter 5), was a royal foundation of the Suevic monarchy, located on the outskirts of Braga, and soon became the religious centre of the kingdom. Royal monastic foundations are not unusual; in fact, the sites of Guarrazar and the church of Recceswinth at Saint John’s in Baños could both perhaps be linked to royal monastic foundations. Casa Herrera may have been a monastic community because apart from the funerary basilica recent excavations have unearthed a series of two more rectangular buildings. Of these, one is 20m long, and is divided in to several rooms, built in parallel with the main axis of the basilica. This building is next to one of the aqueducts of the city, whose vault had been opened, perhaps to supply water to the inhabitants of the settlement (Sastre and Martínez 2012; 2013). The basilica was renovated at the end of the sixth century, when a baptistery was added to the north-east corner of the church and a nave was added to either side of the basilica. This construction seems to be related to the exploitation of the landscape under the control of the Church of Merida. This all merits the suggestion that this settlement could have held a monastic community (Sastre de Diego 2015).

224 

The Iberian Peninsul a bet ween 300 and 850

The remains excavated at Cullera, at the site of Punta de l’Illa, during the central decades of the 20th century have recently been interpreted by Miquel Roselló as a monastery (Rosselló Mesquida 2005; Rosselló and Juan 2003). The archaeological finds suggest a sixth century date. Its identification as a monastery is based on two main points. Firstly, the finds include various liturgical elements including a couple of crosses made in bronze. Second is its location, on top of a small island near to the coast, which made Punta de l’Illa not just a privileged place for developing an ascetic life, but also a strategic point for controlling trade. Besides, the layout and distribution of the buildings reminds of those at Casa Herrera. Amongst the finds, amphorae linked to North African olive oil and Palestinian wine were found stored in one of the buildings, which might have been a cellarium. Besides this, there is the presence of sixth century Vandal and Byzantine coins as well as Late Roman examples. Such variety of coinage seems to confirm the trade dynamism of Cullera. The settlement was destroyed and abandoned at the end of the sixth century. The monastery of Cabrera (Riera Rullán 2013; 2014), on a small island off the coast of Majorca, is a perfect example of a monastic community, because of its isolated location. Furthermore, this monastery had several small hermitages on smaller rocky islets around the island. The site is dated to between the fifth and the seventh centuries. Its necropolis is built cutting into an earlier (but still Late Antique in date) industrial facility for the production of purple dye and (probably) of wine too. Amongst the finds the excavators have highlighted the remains of marble liturgical furniture and glass lamps, together with imported amphorae. This monastery is, furthermore, known through the written sources in a letter from Pope Gregory in which he admonishes the monks for their licentious behaviour.

Other rural sites This last part of this chapter will focus on other rural sites which had a particular nature within the post-Roman landscape. These are the phenomenon of cave occupations (for dwelling and burial) and a couple of exceptional but relevant examples of high-status sites in the rural world. Cave occupations Although the occupation of caves has been identified as occurring before the fifth century, it is at this point that the phenomenon was especially prevalent.

The new rur al l andscape

225

We can divide the use of these spaces in to three different types. The first one would be the hermitages carved in rock or reusing different caves which are related to the expansion within post-Roman society of religious movements encouraging isolation and spirituality. They are usually dated to the seventh century, but nothing discredits the idea of prior occupations. Although they can be found in most parts of the Iberian peninsula, they are particularly numerous in the north, where they have been the object of particular studies (Azkarate Garai-Olaun 1988). Some of these sites are really impressive, with the presence of groups of dozens of hermits in the same space, constructing their dwellings and little chapels in the rock. Some good examples could be the site of Valdecanales, including three different caves constructed in the sandstone and covered with granite with a monumental entrance (de Navascués 1970). There is also the example of the small church of Saint Peter de Tartalés in Cilla and the adjacent ‘Cueva de los Portugueses’, a group of cavities in the lower part of a rocky outcrop. The second type of the occupation of caves is related to cattle paths and husbandry and, so, related to a temporary use of these spaces. In some cases, such as in the cave of ‘Los Husos’ this use has been present since prehistoric times, with continuation in the post-Roman period (Quirós and Alonso 2007-2008). A f inal type is a very special and particular phenomenon detected mostly on the northern coast. This is the use of caves as funerary spaces, characterized by the presence of burials inside the caves, sometimes in quite isolated and hidden places, and usually furnished with particularly luxurious brooches, as in in the case of La Garma or Las Penas. In a recent review of this ritual, it has been proposed that they were a response to epidemic episodes and the isolation of bodies from the communities (Hierro Gárate 2011). Although this cannot be dismissed, there are also other reasons to consider, such as questions of status within herding societies (Quirós and Bengoetxea 2010). Exceptional rural sites We will finish with the description of two particular sites. Their particularity can be related to a lack of archaeological data more than a real assertion of uniqueness. However, it is interesting to point them out, with the hope that more examples will be identified in the future.65 65 In 2017 there have been new excavations at the site of Los Hitos in Arisgotas (Toledo), where the excavators claim to have found a seventh-century rural palace, which they model on the

226 

The Iberian Peninsul a bet ween 300 and 850

Figure 6.16 Plan of the rural ‘Visigothic’ palace of Pla de Nadal (Valencia), as excavated in the 1970s

Based on Juan and Pastor 1989

The first site is Plà de Nadal (figure 6.16), which has been described as a luxurious villa of the late Visigothic period (Juan and Pastor 1989; Juan et al. 1992; Morín et al. 2015). This site is located on a slightly elevated position, 20km north of Valencia. Excavations in the 1980s unearthed an impressive building, the design of which seems to look back at old Roman villas. It consists of a principal body, formed by a central rectangular nave of nearly 17m in length and 5m high. South of this central nave there was an open space covered by a portico of similar dimensions. It is known that the original building had a second floor. What is also impressive is the collection of nearly 400 pieces of sculpture and decorated marbles, including one with the monogram of its potential owner. This site has been dated to the seventh century and, due to this lavish expenditure on symbolic capital, it has been interpreted as the residence of one of the Visigothic elites of Valencia. While it is a very individual site, recent excavations in Vega Baja and Guarrazar have unearthed similar buildings which may also have been constructed by Visigothic elites.66 Naranco monuments of Oviedo (see Chapter 8). As we finish this manuscript, there is no proper publication on the subject, beyond self-published generic texts and items in the news (cf. Morín and Sánchez 2015). Currently we find this interpretation too daring, and we remain awaiting for the publication of the results of the excavations. 66 The detailed results of these excavations are still unpublished, but as far as the excavators have told us, there are some very interesting and similar buildings.

The new rur al l andscape

227

The other site is the previously mentioned El Bovalar, situated on an elevated rock on the margins of the river Segre, which was identified in 1943 but excavated in the 1980s by Palol (1986). In this site some domestic units, a wine press and a church with its cemetery have been unearthed. The collection of Visigothic coins recovered in this site is remarkable, consisting of nearly 20 tremisses, which suggests a violent end to the site at either the beginning or the middle of the eighth century. A recent interpretation of the site has suggested that it may have been a tax centre for the surrounding territory in connection with the strengthening of the Visigothic state apparatus (Martín Viso 2013a). This chapter has dealt with the substantial changes that took place in the rural world between the fifth and eighth centuries, after the dismantlement of the Roman imperial economy. And it is only thanks to the important theoretical, interpretative, and methodological advances in post-Roman rural archaeology of the last two decades that these changes can be coherently summarized and explained. An overview of the available data and recent publications shows major shifts in the settlement pattern of the post-Roman rural world. There are four main phenomena which we have put forward. Firstly, the extensive occupation of hillforts as a new form of settlement pattern, mainly in those areas were urban decline was most acute, such as in the northern Meseta. The extension of this process of hillfort occupation should be read (in its complexity and social terms) with a variety of contextual situations in relation to the dialectics of resilient local communities and regional elites. Thus some of these hillforts may be a consequence of the agency of the state, such as Castro Ventosa, while others may be a product of the agencies of local and regional elites. Secondly, one of the most significant processes occurring from the sixth century onwards is the widespread appearance of different types of networks of farmsteads and villages. Far from being the material expression of unstable or poor communities, these networks show a high degree of complexity both socially and materially. The archaeological evidence shows the presence of peasant communities with a high variety of economic strategies responding to risk in the mid- and long term. Moreover, sites like Gózquez, Zaballa, La Mata del Palomar or Can Gambús-1, to cite some of the most outstanding examples, demonstrate significant regional tendencies which must be stressed in order to understand the complexity of the post-Roman Iberian rural world. These significant changes in the settlement patterns were related to changes in the funerary world. Traditionally, post-Roman funerary rituals

228 

The Iberian Peninsul a bet ween 300 and 850

were mainly associated with the so-called ‘Visigothic burials’. However, the evidence shows the presence of a high variety of funerary typologies, of which we have distinguished five types (pit burials, isolated burials, communal cemeteries, church cemeteries and rock-cut burials). Each of them refers to different social habitus which must be archaeologically contextualized in order to be correctly interpreted. Finally, another crucial process which took place in this period was the extension of Christianity through the rural world. As we have seen, this extension was progressive and it was not archaeologically significant until the seventh century. It was only once the Christian Church was established as the main religious ideology in the Iberian peninsula and its connection with the Visigothic state was secure (see Chapters 4 and 5), that its expansion through the rural world could begin. The presence of a significant number of churches in the rural world demonstrates this long term process of intervention of the Church in the rural world, which would not reach its end until the Middle Ages.

7

A new material culture: a new society, a new economy

The high degree of economic specialization which was achieved in the Roman world is evident in its material culture. The wide range of pottery typologies, the industrial scale of its productions and the long-distance trade of bulk commodities are witness to it. All this was slowly replaced, once the imperial system collapsed, by a much simpler world in which regional productions and interregional networks had to make do, and where the demand for specialized items plummeted. Only a small elite minority were still linked to Mediterranean trade through the importation of a few luxurious products, such as marble slabs from Greece, oil and wine from Palestine, or personal objects, most of them with a religious character. In previous chapters we have seen how the transformation and ultimate decline of Rome had a huge impact on the evolution of towns and their rural territories, on the economy and trade. In this chapter we will focus on the material culture and how this fits in the general narrative of the transformations of the Roman world in the Iberian peninsula. Glass, pottery, construction techniques, sculpture, jewellery and coins will be considered here. In all cases it will be seen how the materialities of the post-Roman period emerged as a consequence, as a response to the social, political, and economic changes that began in the late Roman period. Among these post-Roman material realities there are various which stand out; like the end of the specialized Roman production, meaning the abandonment of some techniques and materials, and, as a consequence of this, the loss of knowledge and specialized craftsmen. All of these form the essence of the Late Antique material culture, and it is this that will be replaced in the Early Medieval period by new alternatives once new socio-political dynamics arise.

Snapshots of the new daily life: pottery and glass One of the most evident transformations in the post-Roman period is how different some daily items were, such as pottery and glass. In the first centuries AD, pottery was produced on a large scale, with an incredible range of typologies, which were traded across the Empire. This system of production

230 

The Iberian Peninsul a bet ween 300 and 850

and supra-regional distribution collapsed in the majority of the Iberian peninsula with the decline of the Roman imperial economic system, and was replaced at first, during the fifth century, by regional imitations, which tried to replicate the old fine wares (see Chapter 3). However, these soon came to an end too and, by the eve of the sixth century, pottery productions were slowly moving away from the late Roman traditions both in form and decoration. These new replacement wares were local (sometimes hand-made) productions, which became absolutely predominant in the seventh century, directly continuing the style of Roman coarse and cooking wares. Something similar happened with glass; large scale production decreased and it was consequently replaced by regional and local solutions. In these local glass and pottery assemblages it is possible to see the simplification of the economy and the limits of trading networks for bulk goods. In post-Roman Hispanic glass and pottery we will see how, despite the end of specialised productions, the continuing demand for these daily goods meant that even if perhaps technically inferior to their Roman predecessors, smaller domestic workshops continued. Transformations in common pottery The formal transformation of late and post-Roman potteries is essential to understanding Late Antiquity, as they show a completely new reality in modes of production, available technologies, out-put expectations and, overall, of day to day life. Despite the lack of monographs for either specific case studies or general syntheses, the study of post-Roman pottery is currently achieving some important results due to the development of commercial archaeology and the effort of a considerable number of specialists (a recent overview of this process is in Vigil-Escalera and Quirós 2016). Here we will focus on a general overview rather than any specific case, perhaps losing specific particularities but still offering a coherent overview of the main processes of post-imperial pottery transformation. The collapse of local, Hispanic table Samian wares and their immediate imitations (TSHT, TSHM, CIS, etc.) and long-distance trade routes meant that local and regional production centres had to supply that demand with common (‘coarse’) wares. It is true that, in main sites (such as Tarragona, Seville, and Córdoba) African and Eastern productions were common through the fifth and sixth centuries (see Chapter 3), but these were increasingly rare, and evidently being replaced by local imitations. Distribution was similar even in rural contexts near the Mediterranean coast (such as ‘Cerro del Molino del Tercio’, in Granada).

A new material culture: a new socie t y, a new economy

231

One of these regional productions is the fifth century pottery of the Duero valley, which, despite being well known, has not been studied as a whole. These productions imitated the late Hispanic Samian productions (TSHT), but were remarkably different. These are usually dark in colour, with grey polished walls and stamped decorations which include geometric forms, rosettes, wheels, curves, tongues, triangles, crosses and planta pedis (foot shapes). One of these local furnaces has been documented in Cauca (Segovia), sited outside the city walls and close to the suburban villa of ‘Las Pizarras’, a residential complex that has been related to the emperor Theodosius, but long abandoned by this period (Blanco García 2002). The simplification of the available wares is evident, even in main sites such as Mérida and Seville. Whereas during the fifth century the range of late Roman options was usually still available (in fact, most local workshops continue to produce high-quality imitations until then; cf. Juan Tovar 2012), the sixth and the seventh century saw a contraction in the available pottery options. The pottery sequences show that the number of functional types declined. The available wares (cf. Maestre Borge 2012) seems to have been limited to: closed water container (cántaro), jugs for liquids ( jarro), multipurpose (i.e., eating, serving, preparing) bowls (cuencos, escudillas) and plates (plato), basins (lebrillo), closed-shape cooking pots (olla) and sauce pans with handles (cazuela), mortars (mortero) and storage vessels (orza and dolium). Other than these, bottles, drinking cups, flasks and lids also appear, but this simple variety contrasts sharply with the extensive list of wares in the early Roman period (figure 7.1). Beyond the typologies, the ceramics of the post-Roman period are visually very different from the earlier Roman ones. The shapes tend to be rounder, more globular, with very few examples showing angled shoulders or straight sections – excepting bowls, which tend to have a characteristic angled profile. The walls tend to be thicker, heavier because the pots are either hand-made or thrown on a slow wheel, and the clay is less refined and with more mineral inclusions. Bases tend to be thicker, wider and, surprisingly, not flat but convex, whereas handles are thick and solid. In the Mérida typologies, handles are also usually sloping down, never rising above the edge of the rim. The walls are rarely treated, but occasionally polished, whereas decorations are usually stamped or combed. All of these make the post-Roman forms look more robust than their Roman counterparts. The f iring of these pots is not usually done at high temperatures, and seems fairly irregular, so in colour they range from reddish brown to light grey. Another consequence of this is the irregular fracture of the pottery. In section, they are usually asymmetric,

232 

The Iberian Peninsul a bet ween 300 and 850

Figure 7.1 Rim and shoulder from a small cooking pot (‘olla’) from the excavations at Casa Herrera. Thrown on a slow wheel, and made with unrefined clay and large inclusions, it is a perfect example of pottery made in the sixth and seventh centuries

while the prof ile of the lip usually has a very tight S-shape. These are the descriptions of the pottery sequences of Mérida (Alba and Feijoo 2003), but are not that dissimilar from the ceramics from Seville (Maestre Borge 2012), Córdoba (Fuertes and Hidalgo 2003), El Tolmo (Gutiérrez et al. 2003), Reccopolis (Olmo and Castro 2008), the territory of Coca and Segovia (Blanco García 2002), and the rural sites of the province of Madrid (Vigil-Escalera 2003b). The technological implications of this new pottery are clear: local productions seem to have been done not at the large industrial sites or imperial fabricae (Peacock 1982), which characterized fine productions in the Roman world, but most probably in domestic environments, either hand-made or thrown on a slow wheel, and fired in hearths or ovens rather than in specialized kilns (Alba and Feijoo 2003). There was still a demand for new pottery, but imported mass-produced wares were no longer easily available everywhere. Still, while these ‘Visigothic’ productions might appear to be not as refined as their late Roman counterparts, they were still fully functional and, in a sense, sophisticated, as they responded successfully both to the necessities of post-Roman society and adapted to new structural

A new material culture: a new socie t y, a new economy

233

conditions. The irregularity and widening of the bases, for instance, can be linked to the wide-spread use of beaten earth as a flooring surface, perhaps improving in this way the stability of a pot on a ground which was not flat. The robustness and thickness are a result of the lack of fast potter’s wheels, but also a necessity when they are not fired at high temperatures in kilns. The inclusion of minerals strengthens the clay in these irregular firings. Similarly, when cooking, these minerals help to distribute the heat better across the pot when this is put by a hearth and not on top of a built-up kitchen (Kramer 1985; Van der Leeuw 1977). The diversity and complexity of the process of regionalization occurring during post-Roman times in the Iberian peninsula, clearly reflected in its pottery, should also be underlined. Thus, it is important to note the regional diversif ication of productions and the signif icant differences between territories. For example, the presence of imported pottery in the coast during the sixth and seventh centuries (as seen in the aforementioned case of Vigo) highly contrasts with its absence in the inner territories. For example, a relevant difference has been recently pointed out between the northern parts of the Peninsula, where the majority of productions were hand-made, with the centre and southern parts (and a few northern sites such as Astorga and Braga), where the use of the slow wheel was greater (Juan Tovar 2012; Vigil-Escalera and Quirós 2016). Glass Glass was not a luxury item in the Roman period: it was readily available and comparable in its distribution to finewares, although as a result of its production process (vessels had to be blown one at a time by a glassblower, as opposed to pottery where many can be fired in one kiln), must not have been as abundant (cf. Stern 2008: 541). This generalized spread of glass wares took place after the development of blown glass in the first century AD, which enabled the production of high-quality glass vessels. Despite the increase in the number of known production sites in the third and fourth centuries, they all came to an end by the fifth century (Sánchez and Cruz 2014), although there is evidence that suggests the presence of later glass production sites. This visible decline in the number of known production sites contrasts sharply on the one hand, with the apparent abundance of glass vessels in Visigothic contexts (even in rural sites) and, on the other hand, with the atomization of pottery production sites. The production of glass was divided in to two stages. First there was the making of the glass into large blocks out of silica sand and an alkaline

234 

The Iberian Peninsul a bet ween 300 and 850

agent (usually natron from the Wadi Natrum in Egypt, but alternatively the ash from some particular plants or from burnt bones), which could then be blown (Stern 2008). The production of raw glass involved large furnaces to mix the silica sand with an alkali solution, whereas the blowing of glass could be done in smaller furnaces and even pottery kilns or ovens to keep the glass at the right blowing temperature. Producing raw glass in Roman Hispania would have required the import of natron from the Levant, as European medieval plant ash-based glasses only begin to be produced in the ninth century (Shortland et al. 2006; Saguì 2007),67 but it is more probable that glass was simply imported in large blocks and then re-worked in the Hispanic workshops (de Juan and Schibille 2017a). Furthermore, and just as nowadays, glass could be easily recycled, which means that producing raw glass was not a constant necessity (Freestone 2015; 2006: 210). Itinerant glass blowers (aniversari vicini) collecting broken glass to recycle and blow it again may account for the lack of specific glass furnaces and for the continuity of glass finds in post-Roman sites. According to Ángel Fuentes Domínguez (2007) itinerant blowers covered wide territories in Late Antiquity, which they visited for long stays during the same periods every year. They collected broken fragments and reused them for making new products for local consumption, supplying both the site where the temporary workshop was located and the rural settlements around it. Added to this, there is further evidence that suggests the continuity of local glass production (or more probably, recycling) in post-Roman Iberia. The three glass furnaces of Reccopolis (Gómez and Castro 2008) form the largest known Late Antique glass workshop in the Iberian peninsula. They have been located behind the main street, in the lower town, and linked to a series of shops, actively in use into the first decades of the seventh century. There is evidence for crucibles and the working of raw glass, without dismissing the possibility of mixing this with recycled fragments. The chemical analyses of various glass fragments from late Roman contexts (fifth century) at the Plaza de la Encarnación indicate that Seville may have also been a production site in this period (Gómez et al. 2006). There is also evidence that suggests the production of glass at the site of Benalúa in Alicante, but this is evidenced by the presence of rejects in rubbish pits 67 It has been suggested that raw glass could have been produced at Reccopolis using plant ash (Castro and Gómez 2008: 119), but there is as yet no evidence to confirm this proposal. The analyses of tenth-century glass from the site of Vascos (de Juan and Schibille 2017b) show that even then, when ash-glass had been developed, it was made from Egyptiaun/Levantine natron glass, which would further undermine the Reccopolis hypothesis.

A new material culture: a new socie t y, a new economy

235

(Sánchez de Prado 2009). The large cluster of late Roman glass production sites of Gallaecia (da Cruz 2009) seem to have been out of use by then. Lastly, Zaragoza may have been a glass production centre as well in the post-Roman period (cf. Castro and Gómez 2008: 126). The homogenous colour and chemical composition of the glass f inds around Cartagena and the Byzantine province (Vizcaíno Sánchez 2009: 667-9) may also hint towards a single regional production site. All of these examples indicate the continuity of glass production in important and active towns, where perhaps the demand for glass continued, although it may have become more of a luxury than it had once used to be. Late Antique glass in Iberia is characterized by the usual presence of bubbles, formed due to the high temperatures. It is rarely coloured, so it is usually olive and blue-green, becoming darker green in the sixth and seventh centuries. Just as happened with pottery, the glass typologies of the post-Roman period are limited (f igure 7.2) to a few, generic, functional forms (Sánchez de Prado 2009; Castro and Gómez 2008) beyond the constructive materials (Castelo et al. 2011-2012): glasses, cups, and bowls and small plates. These are all typologically very similar, with thick ‘match-head’ rims, straight-cut edges and carved rosettes on the bases. Ribbed bowls appeared in the fourth century, and became the most common shape until the sixth century, when they became more open and with a sunken concave base. Among the bowls, the forms Isings 96 (more vertical) and, above all, Isings 116 (shallow bowl) and their derivations are the most common glass types in the Iberian peninsula. Cups with broad, thick-rimmed bases and long, thin flutes are typical of the later period, as also are bases with ‘umbilical’ profiles. Closed shapes such as unguentaria, bottles and flasks are rare, and only typical in funerary contexts; the forms vary from site to site, making overall typologies difficult to systematize. While the glass found on the northern plateau seems to imitate forms of the workshops from the Rhineland, especially from Cologne (Fernández Matallana 2002), southern finds follow Mediterranean patterns instead. The range in the Byzantine territories is wider, probably linked to the arrival of Mediterranean imports (Vizcaíno Sánchez 2009: 669-82), and it does not seem to have penetrated much beyond the immediate Byzantine economic area. Post-Roman glass types form a reduced but coherent set of table wares. Production sites suggest it was limited to a few, economically active, urban sites, being distributed perhaps further into their rural hinterland. Nevertheless, as was the case with pottery, these were productions that sufficed the local demand, and were not aimed at trade.

236 

The Iberian Peninsul a bet ween 300 and 850

Figure 7.2 Glass bowls and glass typologies from Reccopolis

Based on Castro and Gómez de la Torre 2008, figure 8

The solid foundations of society and state: building and architecture One of the most evident changes in the material culture of the post-Roman period, and one which also shows this adaptation in technical skills, can be seen in the changes in architecture and construction. Within a general paradigm of the simplification of the economy, the Roman construction model came to the end, including the stopping of stone quarrying and specialized workshops, such as carving and architects. After this, Hispania suffered a period of constructive stagnation between the early fifth and mid-sixth centuries. With this in mind, it is not surprising to see that by the point of Visigothic urban renewal construction techniques had changed greatly. In masonry buildings, the overwhelming presence of reused brick,

A new material culture: a new socie t y, a new economy

237

tile and stone is the main characteristic, together with the continuity of mud-brick and beaten earth (pisé) walls. The increasing visibility of timber structures is equally representative of the post-Roman centuries. Masonry buildings: the reuse of building material If we look at the building materials, it is evident that there was a change, and not only in monumental buildings. The reuse of ancient materials was a common feature, almost a characteristic, in Late Antique constructions. Beyond traditional interpretations of the ideological and symbolic meaning of old Roman stones being reused in Late Antique buildings, reuse had always had a role within the architectural productive cycle (Ward-Perkins 1971). In early Roman constructions this implied several activities and workers inside the construction chain (quarrying, transport, reworking and setting the material). Certainly, the wide-spread use of reused materials (usually and badly referred to as spolia) is something which is not unique to Late Antique constructions, but, as we said above, it is clearly representative of the period, as it is directly linked to the end of stone quarries and brick kilns.68 For instance, recent architectural analysis carried out on Late Antique buildings (Utrero and Sastre 2012) has concluded that there were no columns and other circular sectioned elements made ex novo (which had to be specifically cut in the quarry), thus implying an intensive plundering of Roman buildings between the fifth and tenth centuries. In many cases, when building materials were reused for a new construction they were seldom re-worked to fit their new purpose. This is clear, for instance, in defensive constructions, such as the late reinforcements of the city walls of Merida (figure 7.3). They were simply used whichever way they came: drums, carvings, ashlar blocks, etc. This contrasts with what happens with the reuse of building materials in Roman or Umayyad times, when the stone blocks were cut, carved, or modified so as to fit in the new construction (Mateos and Sastre 2004; see Chapter 8). The dismantling of old Roman monuments should not be seen as a conscious destruction of the past, but rather a transformation of urban priorities. In a period when there is evidently activity with new construction, the old redundant and disused monuments became the substitute quarry. For example, the premeditated dismantling of the basilica of Valeria, apparently to sell the building materials at the forum (Fuentes and Escobar 2013), must 68 Except, again, for Reccopolis, a fact which does nothing but to underline its uniqueness as a site.

238 

The Iberian Peninsul a bet ween 300 and 850

Figure 7.3 Fifth-century wall in Mérida at the site of Morería, where the ashlar reinforcement was made out of various reused blocks (tomb markers, column drums, theatre seats, etc.)

have been done by decision of the town council. Basilicas, as we have shown, had lost a political purpose in the new Late Antique administration. Similarly, the dismantling of the decorations of the nymphaeum of the forum of Valencia did not imply the abandonment of the fountain (Martínez Jiménez 2011a). The stripping of the theatre and amphitheatre granite ashlar facings in Mérida to reinforce the city walls was not damaging the infrastructure, because the ashlar was simply decorative: the building was kept in place because of its concrete core. In Córdoba, the construction of a kiln in the theatre is directly related to the burning of the marble decorations in order to obtain lime for mortar (Hidalgo Prieto 2005: 402). There are plenty of other examples like these. As a final note on the use of old building materials, it should be noted that this is a tradition which continued into the early Islamic period, and that characterized the early years of the conquest (see Chapter 8). Besides the occasional (sometimes systematic) reuse of stone blocks from old buildings, post-Roman buildings still used mortared rubble. Both public and private buildings in urban contexts archaeologically dated to the post-Roman period across the peninsula were largely built with mortar – in

A new material culture: a new socie t y, a new economy

239

Figure 7.4  Opus signinum from the aqueduct of Reccopolis. Dated to the late sixth century, this opus signinum uses large crushed fragments of tile, rather than the ground and pulverized pot fragments that characterized early Roman signinum

some cases, a very refined and good-quality mortar (Olmo Enciso 2008b), with some also appearing in rural sites. This late opus caementicium was, however, coarser, with a worse quality lime and larger blocks of caementa. This lack of refinement in the preparation of mortar is also visible in the Late Antique opus signinum, or water-proof mortar, which was originally a mix of mortar, rubble and finely grounded pottery (hence its usual pinkish colour) but which in the post-Roman centuries had been reduced to a lime-based mortar with fragments of pot and tile, smashed but not properly grounded (figure 7.4). These transitions are noticeable across the Roman world (Lamprecht 1987; cf. Porath 2002). This late opus signinum may not have been as water proof as its earlier counterpart (probably as a result of the larger fragments of pottery), which would explain why it was widely used as a high-status flooring surface, as in the palace of Reccopolis (Olmo Enciso 2008b), the basilica of Casa Herrera (Caballero and Ulbert 1976), the episcopal complexes of Barcelona and Tarragona (Macias Solé 2008) and the suburban complex of Toledo (Olmo Enciso 2009). Brick-faced mortar, opus

240 

The Iberian Peninsul a bet ween 300 and 850

Figure 7.5 Visigothic buildings from the Vega Baja suburb in Toledo, where most of the buildings were built with stone foundations (bound with clay, not mortar in this case) and the elevations would have been in mudbrick or trampled earth

testaceum, died out in the late Roman period, although a few examples can be seen, such as the basilica of Casa Herrera, always with reused bricks.69 In most cases mortared rubble was used only as part of the foundations, not for the whole height of the wall. Only in the larger buildings, such as churches, episcopal complexes or civil palaces, were the walls fully built of mortared rubble, in combination with reused masonry blocks. Most walls were built with mud-brick or, more often, pisé (trampled earth beaten and shaped inside timber coffering), as in Reccopolis, the adjacent buildings at Casa Herrera in Mérida, the Vega Baja suburb (figure 7.5), and the Byzantine houses of the theatre of Cartagena. This was a direct continuation of earlier Roman techniques in Hispania, where pisé and mud-brick walls on mortared rubble foundations were the most common building technique for houses (not in public buildings). In some cases, especially in larger buildings, vertical stone uprights (opus africanum) were used to create ‘cells’ to be filled in with mortared rubble or pisé. It is not uncommon either to find that where

69 Excepting, again, at Reccopolis where new roof tiles seem to have been fired.

A new material culture: a new socie t y, a new economy

241

Figure 7.6 Stone quarries located next to the aqueduct of Reccopolis. They were probably used to obtain blocks for the construction of the aqueduct itself

it was not possible to do coffering internal divider walls or external walls had been built with a mixed technique of wattle and daub. Overall, masonry in the post-Roman period seems to have continued earlier trends, without any major innovations, other than the increase in the use of reused stone, as a consequence of the closure of the main quarries. Wherever possible, old buildings became the new quarries, but in a few cases new stone was quarried, which only took place in those sites where obtaining reused material was difficult. This happened both in rural and urban contexts. The walls and buildings of Bernardos (cf. Chapter 6), dated to the fifth century (Gonzalo González 2007), were built practically entirely out of the local slate, without mortar. The fortification was built by rearranging the slate that could be obtained in situ. The only ashlar blocks used were probably obtained from the abandoned Roman villas found in the surroundings, which had been quarried away already. The same approach is seen at Reccopolis, where the aqueduct and most of the buildings excavated on site to the date are built using the soft and friable sandstone that can be quarried in the surrounding hills, resulting in some very rough and irregular blocks of stone (figure 7.6).

242 

The Iberian Peninsul a bet ween 300 and 850

A potential lack of engineers? In overall numbers, there was a decline in the volume of monumental constructions between the late and the post-Roman periods (even if including the buildings of the Visigothic period of urban renewal). Furthermore, the period of time separating the two peaks of construction (mid-fourth and late sixth centuries), with the near complete lack of main constructions in between these peaks beyond defensive structures, makes the continuity of highly-technical building skills even less probable. In fact, it may be possible that trained architects and engineers had completely died out (Martínez Jiménez 2014: 177-81). Without a constant demand for their skills, or any sort of institutional training (i.e., the army), it is not inconceivable (cf. Cuomo 2008; Lewis 2007; Mannoni 2008). This is clear from an examination of the archaeological evidence. The evolution and abandonment of aqueducts provides one case to support this theory, as it is clear that there had been no successful aqueduct reconstructions after the fourth century. The possible attempted repairs of Mérida could provide further evidence that there were no engineers who had the skills, tools (in particular the level, or chorobates) or knowledge to build (or re-build) a conduit that would fit the engineering of the old Roman conduit. The example of Reccopolis is again the exception which confirms the rule, for it was all part of a grand royal project, where foreign assistance could have been hired (Martínez Jiménez 2015). Another piece of evidence to support this proposal is the apparent loss of arching and vaulting techniques in post-Roman Iberia and the disproportionate thickness of most walls, which would indicate that there were no skilled architects who could calculate roofing loads upon walls either. Or at least this is what can be inferred from the archaeologically dated contexts.70 Building arches and vaults requires some basic knowledge of mathematics and engineering (Adam 2001: 164-77), but in more practical terms it requires two things which were not available in the post-Roman period because they are directly linked to the existence of quarries: firstly, special sets of tools, such as compasses, to outline the timber frames; and secondly, purpose-cut voussoirs that match and fit the vault/arch. Without either of these elements it is very difficult to set up an arch or a vault, and the evidence for the end of large quarries is very clear. Furthermore, the same skills which were necessary to build and align aqueducts were the 70 This does not just refer to the horseshoe arch (see Chapter 5), but also to the evidence for arches in Reccopolis, which has to be seen as an exception, not as an example.

A new material culture: a new socie t y, a new economy

243

ones that were needed to design and calculate load distribution in walls. When these calculations could not be made, it was simpler and safer to make thicker walls that could support the weight, as can be seen in the walls of Reccopolis or the Vega Baja in Toledo (Olmo Enciso 2009: 78). This lack of trained architects or engineers does not rule out the existence of skilled masons (‘vernacular builders’; Lavan 2008), as the large episcopal complexes and basilicas outlined in the previous chapter testify to their building capabilities. Furthermore, at the site of El Tolmo there is direct evidence for the planning of the urban layout according to a pre-conceived plan (the layout of the buildings was chiseled on the rock, as a guide for the builders who would later set up the buildings). However, there are very important differences between the construction of pitched-roof and timber-framed structures and vaulted buildings, with all the precisely-cut voussoirs that they require. This situation would be in line with the loss of simpler techniques and technologies mentioned earlier such as the fast potter’s wheel, the purification of clays or large firing kilns. Furthermore, this would help to explain the narrow typological variety of monumental buildings in the post-Roman period (apsidal basilicas, aisled hall/palaces, and fortified enclosures). The simplification of the economy in the post-Roman period, and the necessity to offer multi-purpose solutions to the problems that derived from the transformation of the Roman world may be one of the reasons behind this. Timber structures and sunken buildings The literature of Late Antique and Early Medieval sites has paid increasing attention to timber and sunken featured buildings, two elements which seem to suddenly appear from the fifth century onwards, and are usually labelled as indicators of late Antique/Early Medieval phases. The spread of structures built only in timber is evident in the northern provinces of the Empire (Hamerow 2002), whereas in Mediterranean regions there seems to be an increase in the use of timbers in mixed-construction buildings (cf. Valenti 2007). Similarly, ‘sunken featured buildings’ or SFBs (‘Grubenhäuser’ in traditional German scholarship) are a common feature in Roman Iron Age and Migration period communities north of the Empire, which then seem to appear in Roman contexts from the fifth century onwards (see Chapter 6). In the Iberian peninsula, this spread of post-Roman timber constructions has usually been seen as an indicator of the decline of construction techniques and specialized craftsmen, a step towards the ruralization of urban contexts and changes towards a regionalized pattern of construction (Azkarate and

244 

The Iberian Peninsul a bet ween 300 and 850

Figure 7.7 Sunken Feature Building (Grubenhaus) from the Visigothic period. It has the main characteristics that can be seen across Europe: It is dug into the ground and has post-holes for the vertical uprights

Photograph by STRATO, used with permission

Quirós 2001). Some scholars have seen the spread of SFBs as an indicator of the ‘Germanization’ of building practices (López Quiroga 2006) while others have interpreted them as a consequence of the expansion of peasant societies in post-Roman times (Tejerizo García 2014a). The contrast with other western European traditions is perhaps the main point to be remarked upon. Regarding timber constructions, there are no structures remaining from either the Roman or post-Roman periods due to their perishable nature. Besides, it would be very unlikely that there were post-Roman structures in Iberia built only in timber excepting in the northern coast, where some examples have been documented in sites like Zaballa, Gorliz or Aistra (Alfaro Suescun 2012; Campos et al. 2010) and there are some isolated and dubious examples like the timber building in the villa of El Val (Rascón et al. 1991). Furthermore, one of the main characteristic elements of European post-Roman constructions, the elongated, partly sunken, timber framed ‘longhouse’ (Hamerow 2002) is totally absent in the Iberian peninsula except for some particular and dubious examples

A new material culture: a new socie t y, a new economy

245

(Tejerizo García 2014a). It is far more probable that timber was used in conjunction with other building materials, as an additional supporting element, a role which timber beams already had in the Roman period (figure 7.7). The problem with post-Roman ‘timber structures’ (usually beam slots and post holes) is that they become more archaeologically visible in this period because they truncate and break through Roman layers (which are usually paved, so holes and cuts stand out). The general assumption is, nevertheless, that timber supporting elements (such as posts) were more common in the post-Roman period as an economic response to the decline in the production of traditional elements used for upright support, such as bricks or newly quarried stone for piers and columns. Similarly, although timber is a perishable material, this does not mean that these structures were ephemeral: they could easily have been intended to be permanent. The fact that these building alterations with timber supports are visible imply that there were secondary uses of old Roman buildings, although not usually on the monumental scale. Timber beams were also used in new constructions, just as they would have been used in the Roman period. Even more, perhaps, in monumental structures as a result of the loss of vaulting techniques. In newly built Late Antique monumental buildings this can be inferred from their plan: most have got a layout that would have required very broad spaces to be vaulted, while in those with narrower spans the walls are not thick enough to have been able to support the load of a vault. For instance, in the ‘palace complex’ of Reccopolis, the north building’s main nave has a narrow span,71 with various pillars and very thick walls, but this was probably due to the weight of the opus signinum floor from the upper level, which would have had to be sustained by timber beams, and not because it was vaulted (Olmo Enciso 2008b: 478). Vaults and arches have been identified in Reccopolis, at the ‘monumental’ gate that gave access to the public area, but Reccopolis has to be seen as an exception, not as the norm of Visigothic architecture. A similar argument can be made for the episcopal complex of Barcelona and its hall (Beltrán de Heredia 2002a). Even at a smaller scale, the domestic buildings at the site of Bernardos had all their covers and roofs supported by timber beams, and were probably thatched (Gonzalo González 2007: 36).

71 Usually not shown in plans, but visible on site, the north building had (at least) one further aisle to the north, as inferred from the opus signinum floors that continue towards the north. This part of the building has disappeared because of hillslides.

246 

The Iberian Peninsul a bet ween 300 and 850

Regarding the nature of SFBs, we have already discussed its significance mediating the emergence of the networks of post-Roman villages and farmsteads (see Chapter 6). We will now discuss some questions about their structural and constructive elements, related to the general overview of technological simplification we are dealing with in this chapter (Azkarate and Quirós 2001). As said, this was a type of structure found throughout Western Europe from the fifth century onwards, and found widespread in the Iberian peninsula, even if it is mostly recognized in northern Iberia due to the differences in archaeological methods when approaching this period in both regions. In a recent overview of this topic it has been established that the average size of these structures was 4,5 x 2,9m, which is very similar to those results in similar studies in England and France (Marshall and Marshall 1991; Tipper 2004; Peytremann 2012; Tejerizo García 2014a). However, postholes are quite different. For northern contexts, the most typical SFBs are those with two, four (perhaps one of the most common types) and six postholes, paired on each side of the structure. Even though all those typologies are found in the Iberian archaeological record, the most common feature is the one without any postholes in it. Perhaps this is due to the lack of recognition during excavation, but some case studies like the one in La Mata del Palomar show that one common architectural solution was to base the post over other surfaces, like slates or stones. It is noteworthy, however, that some regional trends can be discerned which can consequently produce regional typologies for this structure. In his seminal work on the topic, J. Tipper categorized some of the possible functions related to the SFBs including domestic residence, storage, loom workshops or places for bread preparation (Tipper 2004: 160). The functional advantage of this structure is generating some specific climatic conditions very suitable for processes like the fermentation of milk for cheese or grain for beer. Moreover, some geoarchaeological analyses in Iceland have determined the presence of multiple functions operating within the SFBs (Milek 2012). It is difficult to determine the specific function of these structures in the Iberian context, but it can be suggested that they were not used as a permanent residence – stone-footed buildings fulfilled that function – and, as an ancillary structure, they were commonly used as storage buildings and productive spaces for looming, due to the evidence of loom weights in a significant number of cases (Vigil-Escalera 2003a; Tejerizo García2014a). Finally, it has also been suggested that there is a potential relation between the SFBs and the work of women within peasant societies (Milek 2012).

A new material culture: a new socie t y, a new economy

247

Representing the self and the community: identity and display Identity is one of the key issues discussed in post-Roman studies. Traditional discourses which tried to identify ethnic markers and social distinctions based on ethnic classifications have, however, major argumentative flaws (see Chapter 4). In order to understand these ‘markers’ or elements of identity it is necessary to look at them in their archaeological context, trying to understand who were they made for, because such markers were mostly aimed at the community of viewers who would see them. In funerary displays, for instance, where the social rank and position of the deceased’s relatives was reaffirmed, elements of jewellery or weapons deposited with the body fulfilled that purpose. When looking at the broader picture, and when it is not the position of individual members within the community which needs defining but the status of larger elite groups (such as the church or the monarchy), sculpture and monumentality were a different way of approaching this issue. Jewellery, metal objects, and weapons Unlike construction techniques or pottery technology, jewellery and metalworks in post-Roman times experienced a remarkable development. This was due to the influence of prior traditions developed in Central Europe and imported to the Iberian peninsula.72 The presence of a significant number of metal ornaments throughout the territory and its diversity show a thriving craftwork which implied a complex network of production, distribution and consumption. Although late Roman metalwork was very specialized and diversified, as seen for example in the post-imperial cemeteries (de Palol Salellas 1958) or in objects like the ‘Simancas-type’ knives, the military brooches (Pérez Rodríguez-Aragón 1992) and the so-called osculatoria (interpreted as perfume stirrers; Regueras Grande 1990), between the second half of the fifth and the first half of the eighth century the level of specialization increased, creating some spectacular pieces usually related to post-imperial elite identities and social inequality within local societies (Pinar Gil 2012b; Tejerizo García 2011). It is interesting to note that the most impressive examples are found in funerary spaces while more common metalwork and jewellery are documented both in domestic and funerary areas. This spatial difference is a sign of the relation of this material culture with burial rites and the expression 72 Traditionally referred to in Ripoll López 2007.

248 

The Iberian Peninsul a bet ween 300 and 850

of identities within the local communities in which they are found. Recent studies have shown particular regional differences in the distribution of different types of ornaments (Pinar Gil 2012a; 2012b). Moreover, the level of specialization they require to be produced and the kind of materials they use (silver, gold, copper and precious stones, sometimes coming from far distances; Guerra et al. 2007) relate them with centralized forms of production and, probably, to state-sponsored workshops. Although these were not common, at least one such workshop has been identified at Reccopolis. This level of specialization led to a very complex typology of metal objects, which have been the object of some fundamental monographic studies (de Palol and Ripoll 1988; Ripoll López 1998; Sasse 2000). Here we will summarize the most relevant types of objects found in the post-Roman archaeological record. There is no doubt that one of the most important objects within postRoman material culture were personal objects in funerary contexts, which has consequently been labeled inhumations habillés (Kazanski and Périn 2008). Within this group of objects, the presence of a significant variety of belt buckles and brooches or fibulae clearly stands out. In the second half of the fifth century and all through the sixth century the most common type of buckle found in the archaeological record is the square-buckle with cloisonné decoration, a technique formed by first adding compartments (cloisons in French) to the metal by soldering glass or precious stones into it. During the sixth century the development of the cloisonné technique was extraordinary, reaching a high level of complexity in some cases such as the ones found in the cemeteries of Castiltierra or Carpio de Tajo (Ripoll and de Palol 1988: 239-240). Another type of buckle documented mainly in the second half of the sixth century is the rigid buckle, which in origin had Roman inspiration but developed differently afterwards (figure 4.6). Some of these buckles show pierced decorations, many of them with vegetal and animal motifs. Other types documented in Western Europe, like pisciform or cruciform buckle, also have their counterparts in the Iberian peninsula in contexts like Gózquez (Contreras and Fernández 2006) or Espirdo-Veladiez (Jepure 2004). From the end of the sixth and up to the eighth century a very characteristic type of buckle appears in the Iberian peninsula: the lyre- or tongue-shaped (liriforme) buckle (figure 7.8). The production and distribution of this type of buckle was widespread, especially in the southern part of the Iberian peninsula (Ripoll López 1998). Although the first types of this buckle may have been Eastern Mediterranean in origin, they were soon copied and produced within the frontiers of the Visigothic kingdom, with a high variety of forms and decorations which include vegetal, anthropomorphic and

A new material culture: a new socie t y, a new economy

249

Figure 7.8 Various typologies of lyre-shaped (liriforme) brooches from the south east

From Vizcaíno Sánchez 2008: figure 120

zoomorphic motifs. During the eighth century a particular type of lyreshaped buckle is found also in caves in the northern coast associated with complex funerary rituals, that have been interpreted as the consequence of epidemic episodes (Hierro Gárate 2011). Fibulae or brooches were another personal object with significant production, distribution and consumption amongst post-Roman individuals. There

250 

The Iberian Peninsul a bet ween 300 and 850

are two main types: the bow fibulae and the eagle fibulae. Bow (or cross-bow) fibulae are one of the most recognizable objects within the ‘Germanic’ art tradition across Europe since late pre-Historic times. In the post-Roman Iberian peninsula, the first documented types are made in three different parts clamped altogether, evolving afterwards into a single-piece solid fibula frequently with a radiated head (Ripoll López 1998). The eagle fibulae are a very rare and particular element of post-Roman material culture, which were also made in most cases in a cloisonné technique, and mainly found in the central and southern part of the territory (de Palol and Ripoll 1988). The consolidation of the Visigothic state and the need for identifying elements for the monarchy led also to generate a particular material culture in specialized workshops. That is the case of the unique objects recovered in the treasures of Guarrazar and Torredonjimeno, which include the wellknown votive crowns of Swinthila and Recceswinth and various crosses in gold and silver (de Palol and Ripoll 1988: 262-275). As we know from the documentary sources, these kind of objects were used in the institutional ceremonies of the monarchy as a means of propaganda and indicators of rank and office (Valverde Castro 2000: 191-195; Martínez Jiménez 2011b: 38-40). The latest interpretations, however, underline how it may be the case that the crowns themselves are, similarly, diplomatic gifts from Byzantium (Walker 2016: 114-5). In this way, only the hanging letters and crosses of the crowns would be of Visigothic manufacture, which are technically and stylistically far more similar to the fibulae noted above. Whereas these typologies are new elements which appear in the postRoman period, there are other elements which continued from Roman times into the sixth and seventh centuries. This is the case of bullae, earrings, the significant number of necklaces made of glass, rings or belt buttons, all of which continued to be produced with Roman techniques and motifs. These objects are found in the same funerary contexts as the ‘Germanic’ ones, showing the complexity of identities in post-Roman centuries and the difficulty of relating objects to one particular level of identity (Halsall 1995a; also see Chapter 6). Lastly weaponry should be mentioned. Although the presence of weapons in post-Roman Iberia is not as frequent as in other parts of post-Roman western Europe (Härke 1990; Kazanski and Périn 2008) they are found related to some particular types of burials, and in isolated contexts and finds all around the territory. Thus, we find examples of swords (longer spathae and shorter scramasaxes), arrows, axes or spears in sites like Herrera de Pisuerga, Castiltierra, Aguilafuente, Vigaña or Plà de l’Horta (Catalán Ramos 2015). Some territories in the north of the Peninsula deserve special attention,

A new material culture: a new socie t y, a new economy

251

where a group of cemeteries show the significant appearance of throwing axes related to the Frankish francisca. These cemeteries have been related by some scholars to the influence of the Frankish kingdom and its elites (Azkarate Garai-Olaún 2004a). Sculpture and the Mérida marble workshop Since the Greek and Roman periods sculpture (and in particular marble sculpture) had been an element of prestige in new buildings and constructions, and this was a trend which continued into the post-Roman period – increasingly so perhaps due to the lack of newly quarried marble. In the construction of new civil and religious buildings, newly carved friezes and other elements of architectonic sculpture were indicators of prestige, not only for whomever commissioned the building but, ultimately, to the whole local community who would have a new building decorated in such a fashion. As opposed to pottery and glass, which were still in general demand despite the lack of large-scale production sites, marble workshops, an expensive product which was non-essential for daily life, had to reduce their size dramatically in order to adjust themselves to the declining demand. This demand plummeted through the late fourth and into the fifth and sixth centuries. By the late sixth century, a new period of monumentality and urban renewal appeared in some towns, especially those that had a martyr patron and, closely linked to this, towns with strong bishoprics (see Chapter 5). Added to these Christian capitals, other towns and central places connected to the new Suevic and Visigothic authorities helped to re-activate an environment conducive to building. This new political and religious context meant a new demand for carved marbles. By this stage, only the Mérida marble workshop seems to have still been still active, all other production and carving centres having disappeared, along with the closure of the marble quarries. Furthermore, even in Mérida, the techniques, tools, and skills to work marble had been largely lost, resulting in its characteristic style, carried out on reused marble blocks (Utrero and Sastre 2012). Elsewhere, carved friezes and sculpture in other reused soft stones (such as limestone) had to be used. The distribution of the so-called ‘Visigothic’ carved marbles, most of which can be directly linked to the Mérida workshop, is limited to elite centres under control of the most affluent Hispanic religious elite (Mérida, Seville, Córdoba) or of a civil elite in the case of Toledo. Mérida is, obviously, the place where most carved marbles have been found, with currently over one thousand pieces preserved (Cruz Villalón 1985; Sastre de Diego 2015), followed by Beja, probably a minor but dynamic workshop. After its

252 

The Iberian Peninsul a be t ween 300 and 850

designation as the capital of the Visigoth kingdom, Toledo considerably increased its use of marble (Morín and Barroso 2007), but its sculpture is similar technically to Mérida. The royal foundation of Reccopolis also required marble decorations (Balmaseda Muncharaz 2008). Even if most of this sculpture was either found or preserved far from its primary context (as happens with the marble preserved in Córdoba; Sánchez Velasco 2006), they are usually linked to churches because of their religious iconography. This does not rule out that some of them could have appeared on civil buildings as well. Marble carvings in this period are very characteristic, and can be grouped into four types: columns and pilasters, friezes, chancel screens, and altar elements. The cross, and sometimes the Chi-Rho, is the most popular iconographic motif, especially in liturgical features, but it is not the only one. Besides this, many geometrical compositions, probably borrowed from late Roman mosaics, are rather common, decorating practically all kind of elements. Columns and pilasters form a very heterogeneous group of forms, but have a set of common markers such as, for instance, imitating the Roman Corinthian capitals, with very schematic acanthus leaves (many times carved simply as a protruding wave or a set of ripples), thick volutes, and knobby fleurons. Furthermore, these capitals tend to have pronounced corners, as opposed to their rounder Roman counterparts. Shafts are usually flat or decorated with spiral fluting (‘Solomonic’) and palm-tree trunk motifs, whereas bases try to follow the torus-scotia-torus pattern. A clear example of this loss of techniques and skills can be seen in the marble pilaster excavated at the site of Casa Herrera, in Mérida (Utrero and Sastre 2012). This pilaster (figure 7.9) was an attempt to create a column out of a (reused) prismatic block of marble, although the final result was not quite what the craftsmen had expected. The pilaster itself has all the elements of a column: plinth, base with torus and scotia, a smooth shaft, collar and a capital with acanthus leaves and fleurons under an abacus. However, it is evident at the base of the pilaster that the craftsman attempted to make a monolithic column out of the original block by chiseling and grinding the marble. The first torus was done by grinding off the corners of the block, using the square plinth as a guide, with an irregular result. The scotia was carved out, trying to use the irregular torus as a reference, resulting in a very uneven profile, so that when the second torus was ground to its shape, without a clear reference from the plinth or the scotia, it looked completely irregular and asymmetric. Perhaps in despair, the craftsman seems to have given up in his attempt as the rest of the pilaster is perfectly square; shaft, capital and all. Similarly, the pilasters found at the aljibe (well) of the Umayyad citadel

A new material culture: a new socie t y, a new economy

253

Figure 7.9 Casa Herrera pilaster; a failed attempt to create a round column out of a prismatic block of marble without the necessary tools

Currently in the Visigothic Collection of the National Museum of Roman Art (MNAR)

254 

The Iberian Peninsul a bet ween 300 and 850

Figure 7.10 Marble pilasters from Mérida, now in the water cistern of the Umayyad citadel. Carved out of reused blocks, these pilasters show the traditional geometric and vegetal motifs that characterize Visigothic sculpture

of Mérida (figure 7.10) try to represent engaged columns, although these are not truly carved as half-columns: they are deep bass-reliefs carved into the block. These two examples show how the tools (compasses and hydraulic saws) and knowledge to carve column drums (or other round elements) out of prismatic blocks of marble had been lost during the fifth and sixth centuries. Despite this loss of technique (probably due to the gap produced by the Barbarian crisis), the workshop of Mérida was active from the late Roman period through the whole of Late Antiquity, and active into the eighth century. This long continuity from late Roman times makes Mérida a key place to understand how both the continuity and change in technology worked, and how decorative motifs were developing, even after the Islamic invasion.

A new material culture: a new socie t y, a new economy

255

Beyond pots: coins and slates in their economic context The formation of the Visigothic and Suevic monarchies has been explained within the context of state formations in the aftermath of the fall of Rome. One of the key elements in the formation of such states was the establishment of taxation systems, in imitation of the late Roman model, to keep the state functioning. However, the Visigothic kingdom does not seem to have been an efficient tax-collecting polity, as can be seen through the analysis of coins (in combination, with written sources, especially the different sets of laws; cf. Liebeschuetz 2015: 170-4). However, Visigothic Spain has archaeological evidence for tax collecting in rural areas in the form of the difficult to interpret, but nonetheless striking, ‘Visigothic’ slates. Post-Roman coins One of the elements which drew attention to the post-Roman period from the very beginnings of historical research were coins, as they were golden, and helped to construct a chronological history of the Visigothic kingdom. As a result, a large proportion of the known Visigothic coins (of which there are no more than 10,000 in total) are the result of antiquarian finds, and outside archaeological contexts. Most studies on coins have tended to focus on their typologies and the information that can be obtained about them through the number, distribution and location of the mints; studies which have been followed later by general attempts to theorize about the role of coins in the post-Roman period, whether they are seen as indicators of economic or of fiscal activity (cf. Pliego Vázquez 2009). The study of Suevic and Byzantine coins has followed similar patterns, albeit with their much reduced number. Coins in the post-Roman world were, typologically, the same as those from the late Roman period. In fact, small late Roman coppers seem to have been the main form of day to day coin in the post-Roman centuries, and are usually found on post-Roman levels, although Visigothic bronze coins (nummi) also existed (Pliego Vázquez 2015-2016). The Germanic monarchies did not establish a new monetary system, as the Sueves and the Visigoths adapted the late Roman model, and simply continued to mint new gold coins (tremisses). The tremissis, which was a third of the solidus (the Imperial gold coin), and was usually 1.5g in weight (although this varied through time and from mint to mint) and its high value makes it an unlikely element of daily use (figure 7.11). In the Visigothic kingdom (Grierson 1953; Heiss 1872; Miles 1952; Pliego Vázquez 2009; Vico et al. 2006), coin typologies are characterized for their

256 

The Iberian Peninsul a bet ween 300 and 850

Figure 7.11 Golden tremissis of Athanagild from Reccopolis now at the National Archaeology Museum, Madrid

Photograph: Fernando Velasco, used with permission

particular legends and the images on the reverse and obverse, all of which are imitations of contemporaneous imperial coins. The legend is invariably a combination of cross, the king’s name, the title (rex or D[ominus] N[oster]) and an epithet ( felix, pius, iustus, inclitus, etc.), with the mint name or mark on the reverse. The images in obverse usually include the very schematic bust of the king (or busts of the king and his heir). The typology of the bust is different and characteristic in each province, and they evolved over time. The reverse also varies over time, but it usually includes a cross on steps, a winged victory, or a monogram with the name of the mint, all of which are elements directly copied from Byzantine coinage. The wide range and variety of types of coins indicates a large number of different minting dies, and as a result, of a constant process of minting (Pliego Vázquez 2009). Copper coins have also got a wide range of typologies, but include mostly crosses and mint monograms. Regarding the Suevic coins, they follow similar patterns (Cabral and Metcalf 1997; Díaz Martínez 2011), imitating Roman imperial coins, although from the mid-fifth century the Suevic kings were already minting coins with their own names (reverting back to imperial minting after reaching a treaty with Valentinian III). The most characteristic coins are the tremisses belonging to the latina munita (for moneta, which is ‘coin’) series, which belong

A new material culture: a new socie t y, a new economy

257

Figure 7.12 Byzantine copper coin, minted in Cartagena: One of the few indicators for the circulation of newly-minted small currency

Photo courtesy of Dr. Jaime Vizcaíno

to the period of expansion and consolidation of the kingdom in the sixth century, and which are of a better quality than earlier coins. Typologically, Suevic coins imitate Roman models of Honorius and Valentinian III (with the emperor in the obverse and a cross in a wreath in the reverse) – even the latina munita series. In the Byzantine area, the most recent survey by Jaime Vizcaíno (2009: 687-725) rightly underlines that this coinage was directly linked to the presence of the Byzantine military and its inclusion in the wider Byzantine trading network. The Byzantine territories seem to have had a more dynamic monetary economy, especially as there is not only an abundance of imperial gold coinage (which in the Visigothic territories is very limited), but there is also a newly-minted set of low-denomination copper coins (figure 7.12). Otherwise, coins from beyond the peninsula (Byzantine, Merovingian, or Vandal) are rare. Most of them can be found in hoards and votive deposits, such as the ‘foundational hoard’ of Reccopolis (Castro Priego 2008), in which the presence of various coins can be seen as an indicator of the reach of the power of Liuvigild and of gift exchange with foreign powers.73 It is true that on the Mediterranean coast such finds are more common (cf. the finds in Alicante: Martínez and Moreno 2015), and could perhaps be linked to trade, but the rarity of these finds, together with the lack of Visigothic and Suevic 73 This can be compared with the mix set of coins found, for instance, in Childeric’s burial (Chifflet 1655).

258 

The Iberian Peninsul a bet ween 300 and 850

coins in contexts beyond the Iberian peninsula underline how unlikely it is that these coins were used as elements of long-distance trade. Interpreting coins As noted, the gold tremissis is a coin of a very high denomination, which could not have been used in daily transactions, so its function has been proposed as either an indicator of taxation, as an element of (large-scale) trade or as a prestige element (Pliego Vázquez 2009: 219-30; Martín Viso 2013b). Their value was so high that individual isolated finds should be seen as hoards: perhaps not in the number of coins, but certainly attempts to hide away a lot of money. We are more inclined to view coins as part of the state apparatus, and coin minting as a way of monetizing the taxes which were collected in cash, rather than as indicators of trade. As Chris Wickham (2005: 702, n. 16) put it: ‘[I]t is often unclear how much coin distributions tell us about economics as opposed to the structures of public administration and of diplomatic gift exchange’. Despite this, there are more dimensions to the post-Roman coins that need to be addressed. From a political point of view, the minting of gold coins was a way for the Germanic monarchies had to reassert and legitimize their position – after all, coin minting was an imperial prerogative. By minting coins the monarchy was acknowledging its position within the Roman order, especially when minting in the name of the emperor or imitating such coins. However, during the process of state formation the Visigothic and Suevic kings, in order to assert their independence, began to mint with their own names and carry out their own ‘monetary reforms’. In the Visigothic case, this took place during the reign of Liuvigild. Liuvigild originally minted a series with the emperor’s name, but afterwards he had a set ‘transitional’ types made, with illegible legends (but certainly not the emperor’s name) to ‘test’ whether these new coins that diverted from the imperial standards were accepted. These were soon substituted by the first real independent royal coinage. After this, coins and their legends could be used as part of royal propaganda: Visigothic coins labeled Corduba[m] bis optinuit (‘conquered Córdoba twice’) or Emerita victoria, for instance, commemorate Liuvigild’s campaigns. Similarly, there is also the Suevic version with Gallica pax, ‘peace in Gallaecia’. Within an administrative context, the distribution and consistency of mints is an indicator of the reach and evolution of royal power. There seems to be a lack of minting consistency (Williams 2005, figure 5; Pliego Vázquez 2009), and whereas some mints are regular centres of coin production (like

A new material culture: a new socie t y, a new economy

259

Mérida, Córdoba, Seville, Narbonne, and Toledo) others are very rare (for instance Reccopolis, Valencia, and Tarazona). Some mints seem to have been active only during the reign of one monarch (e.g., Calahorra, Málaga, Orense). Overall, it seems that the location of mints was limited to a few chosen cities beyond the provincial capital. For example, in the south-east territories of the Visigothic kingdom, there is an increase in the number of mints (even the creation of new ones) and of coin types in the early decades of the seventh century, which has been directly related (Vizcaíno Sánchez 2009: 724) to the need to pay the troops during the final campaigns. Similarly, in Gallaecia, Visigothic mints were very abundant but equally very irregular in their production. These two models of production (few but consistent sites per province versus multiple and irregular mints in Gallaecia) indicate different types of provincial administration: a more centralized or restricted Visigothic model, and a more dispersed (perhaps due to the underlying lack of strong urban sites, perhaps due to dispersed nature of the centres of power in mountainous Gallaecia) Suevic model (Díaz Martínez 2004; cf. Pliego Vázquez 2012). Considering the whole of Visigothic coins, Ruth Pliego (2009) has underlined that there seems to have been a constant volume of gold in circulation, used by the state to carry out large payments (soldiers’s wages, peace treaties, etc.), but that it was incapable of extracting more cash out of regular taxation (cf. Castellanos Martínez 2003: 218). Thus, the monarchy was using the gold to pay out, but only in extraordinary circumstances was cashing gold in (fines, confiscations), as the recipients of the gold would have hoarded it and taxes were probably paid in kind. This interpretation can explain, for instance, why minting was so irregular in quantity and quality from reign to reign (figure 7.13). Therefore, during successful military campaigns (e.g., under Liuvigild, Reccared, or Swinthila), the extra revenues obtained from the campaigns could justify the high quality and volume of minted coins, whereas in periods of civil war or political instability, the only way to obtain more coin was to devalue the gold content, and make do with what was available. Lastly, it can be put forward that the economic and monetary stability which was achieved during the period of state formation (between the reigns of Liuvigild and Swinthila) coincides with the period of active military conquests. This could perhaps underline the fact that the Visigothic taxation system was not at all effective, and it was only with the extraordinary income obtained from looting and spoils of war that a state could be run. After all, the late Roman administrative reform had to be implemented because the state needed more money, which until the third century had been obtained

260 

The Iberian Peninsul a bet ween 300 and 850

Figure 7.13 Graph depicting the evolution of gold content in Visigothic coins per reign. It is noticeable how the steady and standard production in the period of state formation contrasts sharply with the decline that

RODERIC

WITIZA

EGICA

ERWIG

WAMBA

RECCESWINTH

CHINDASWINTH

TULGA

CHINTILA

SISENAND

SWINTHILA

RECCARED II

SISEBUT

GUNDEMAR

WITTERIC

LIUVA II

RECCARED

HERMENGILD

LIUVIGILD

follows it in the eighth century, with only a few minor examples

100 90

GOLD CONTENT IN VISIGOTHIC COINAGE

80 70 60 50 GOLD CONTENT IN VISIGOTHIC COINAGE

by expanding the Empire and obtaining extra cash in spoils, slaves and loot. The fact that most of the wealthy landowners were exempt from taxes, and that revenues usually came in large amounts through confiscations would further prove this hypothesis. The Visigothic slates One particular element of the material culture which has drawn the attention of scholars for their uniqueness and the difficulty in their interpretation are the Visigothic slates (figure 7.14) found in the western Meseta and (mostly) dated to the sixth and seventh centuries (Velázquez Soriano 1989; Martín Viso 2013a). These inscriptions appear in the late Roman period and seem to continue in use into the ninth century, but they seem to be widespread only in this limited geographical area (where slate is naturally occurring) in the post-Roman centuries, when they seem to become a means of record-keeping by local elites in private documents. In fact, it could be the case that the numerical and textual slates are not contemporary. There are over 160 of

A new material culture: a new socie t y, a new economy

261

Figure 7.14 Reproduction of one of the Visigothic slates recovered from Diego Álvaro, now in the Museum in Guarrazar

Photograph by Dr. Patricia González, used with permission

such slates known with fragments of text and over 1000 which are simply ‘numerical’ (i.e., which only have numbers listed, without text).74 74 It is widely known in Spanish scholarly circles, but little discussed, that a significant number of unstudied slates are currently being kept in a private collection.

262 

The Iberian Peninsul a bet ween 300 and 850

Leaving the chronological issue aside (the chronology is usually ascribed by paleographic style, although some are dated with regnal years), the most important issue is their content and context. The texts include school exercises, lawsuits, registers, and so on, although they have been placed in a direct relationship with accounting and, perhaps, tax or rent collecting. The fact that these slates have been found both in urban contexts (central places of local administration, like Ávila and Coca) and in rural elite sites (hill forts such as Bernardos or Navasangil) would support this interpretation. The largest concentration of such slates is in the site of Dehesa del Castillo in the village of Diego Álvaro, which has been interpreted as the private archive (due to the nature of the texts) of a local magnate (as can be inferred from the nature of the site). The transformation in the production and techniques of post-Roman materialities shows a major transformation from specialized Roman productions. However, what has to be underlined is not this decline in scale and quality, but rather the constant demand and resourcefulness of urban and rural dwellers in the fifth, sixth and seventh centuries. Faced with the lack of standardized offers and quality goods, there was the need to create new commodities and to keep up with the demand. This transformation is not only economic, but also deeply rooted in the social structures, as it was suddenly up to this people to provide for the needs of their local communities. Furthermore, in an environment of increasingly de-monetized exchange, new trades were perfectly valid means of activating the economy. The transformations in the techniques and typologies also reflect social changes, especially in domestic environments. This is especially visible, as seen, in glass and pottery. The characteristically different pastes and clays of the post-Roman cooking wares reflect the transition from Roman kitchens (where pots were kept over the fire on a purposely-built structure) into post-Roman domestic, multi-purpose hearths.75 The cooking pots had to be more resilient and had to distribute the heat more efficiently because they were now placed by the fire, on the ground. Similarly, the abundance of small, individual bowls (both in pottery and glass) suggests that there were new ways of eating: sharing food from common trays which was later served into individual bowls, rather than eating out of shared trays. Changes in building techniques are also a clear indicator of this general process of technological simplification and resilience. The reusing of bricks, tiles and stone in post-Roman buildings is a widespread phenomenon which 75 Which are what characterize post-Roman houses; cf. Chapter 5.

A new material culture: a new socie t y, a new economy

263

should be interpreted not as a lack of knowledge but of adaptive strategies in a context where some specific buildings (villas, fora, circuses, etc.) had lost their function and their place in the social habitus and urban priorities. This reusing of material can be the main explanation for the unequal documentation of timber buildings in the Iberian peninsula. Besides, what is reflected in the archaeological record is the continuation of many techniques through post-imperial times, such as mud-brick and pisé walls, although there were some important changes in their technological chain. The simplification of building techniques and the continued use of some late Roman buildings can be related, perhaps, with the absence of professional engineers, which can be hypothesized through the failure to repair some buildings. Similar conclusions can be drawn with the loss of technique in the sculpture workshops. Another issue that emerges from the archaeological record in relation is the use of sunken featured buildings in the Iberian peninsula as an adaptive strategy in specific contexts. Even though there are significant similarities with European contexts, the presence of these structures is very common in the archaeological record, showing at the same time a preference for simpler construction techniques but also a complex form for intertwining functionality and technical availabilities in post-Roman societies. Some of the technological changes of the period should be directly related with the emergence of the Visigothic state and the progressive imposition of a new ideological hegemony. That is the case of the jewellery and the personal adornments typically related to funerary contexts. These show not only the presence of very specialized workshops and techniques, but also the importance of material culture in mediating social relationships in many contexts where deep social tensions existed. Weapons, in this sense, can be seen as the reflection not only of a military ethos but also the importance of this identity in representing social status. Finally, coins and slates are two materialities related to the changes in taxation, economics and polities between the sixth and eighth centuries. The archaeological study of coins shows their importance not as an instrument for tax payment, but for social prestige and individual capital accumulation in specific social contexts. Slates, on the other hand, may be showing those adaptive strategies for taxation in rural contexts.

Part 3 The Early Middle Ages

8

The formation of a new Medieval materiality

During the eighth and ninth centuries, the political situation of the Iberian peninsula was radically transformed as the result of the Umayyad conquest of 711 and the later counter-attacks by Carolingian armies from beyond the Pyrenees. This reorganization of the power structures deeply disrupted the post-Roman political, social, and economic system which had developed during Late Antiquity and set the foundations for the developments of the Middle Ages. However, despite what historical accounts may say about this period, from an archaeological perspective there was no abrupt change from a Late Antique into a Medieval world and there was, as expected, a period of transition and adaptation by Iberian societies. As has been highlighted in the previous chapters, Late Antique material culture and its associated monumentality and settlement patterns were the results of the transformations of the late Roman world. The Christianization of the urban and rural landscapes, together with the simplification of the economy, the reorganization of the administration and the emergence of new material forms of social prestige are the key processes that explain Late Antique material culture. These evolving transformations broke down and were replaced during the eighth and ninth centuries. It will be evident that there is not a clear ‘closing’ date for our final section. As happens when dealing with material culture and archaeology, it is very difficult and usually pointless to highlight a single moment for the transition from one period to another. For this reason, this chapter covers a period roughly dated between 711 and 850. These two dates mark the beginning and the end of a period during which Late Antique materialities were still present, even if gradually being replaced by new developments. In this last chapter we will, therefore, look at the main areas of Late Antique material culture discussed in the previous chapters, and briefly explain the political and material circumstances of this period and their reflection in monuments and other archaeological elements. This will be achieved by f irst explaining the political developments and transformations of both the Umayyad and the Christian territories, leading to their respective periods of state formation. We will then look into the emergence of new political forms of representation, the development of religious materialities and, finally, the transformation of the urban and rural landscapes.

268 

The Iberian Peninsul a bet ween 300 and 850

The formation of new medieval polities The macro-historical developments of the eighth and ninth centuries are mainly determined by the existence of two heterogeneous and uneven sociopolitical blocks: the Christian north and the Islamic south, or al-Andalus. In order to understand the reasons behind the changes in the material culture it is necessary to briefly introduce the historical setting. The conquest is said to have occurred (Chalmeta Gendrón 1994: 68-72, 114-5) during the war between the elected King Roderic and Agila II (son of the previous king Wittiza), when the latter requested help from the Umayyad soldiers in North Africa.76 Tariq ibn Ziyad crossed into Spain with his Berber troops in July 711, prompting Roderic, who was then fighting in Pamplona, to march south with his army. Following the first pitched battle at the Guadalete river the invading troops captured the main cities of the peninsula over three yearly campaigns, during which more Berber and Arabic troops crossed into the peninsula. These attacks reached as far as Narbonne and Septimania, and can be seen as a continuation of the waves of conquest through North Africa in late seventh century (cf. Fenwick 2013). Apart from the few exceptions of open armed resistance like the famous cases of Córdoba, Mérida (which eventually surrendered), and the later ‘rebellion’ in Asturias, most parts of the Visigothic kingdom fell to the Umayyad troops. This is likely to be because the conquerors offered a set of conditions of surrender which barely disrupted local ways of life. The conditions of these early treaties of surrender (known as sulh) were very beneficial for urban elites, because they allowed previous structures of power and customs (including religious practice) to be maintained, as long as an extra tax, the Umma, was paid to the Islamic community, represented by the Umayyad caliphs (Chalmeta Gendrón 2013; 1994: 209, 213-5, 219-20). The only example of such a treaty that has been preserved is that of Theudimer, who appears in the Arab text as Tudmir, and who was allowed to keep his authority over seven cities (the territory thus known as the Kura of Tudmir, which included Alicante, Orihuela, Eio and Elche) as long as he paid a tribute in kind (Gutiérrez Lloret 1996; Llobregat Conesa 1973: 19, 73-5, 84). Only those towns which were conquered by force (like Córdoba – Arjona Castro 2001: 20-4) became fully controlled by the Muslims, 76 Islamic sources offer two perspectives on the conquest: one tradition underlines the connection and mutual aid between the Goths and the invading Islamic troops, whereas the other highlights the conquest over any collaboration with the Goths (Riess 2013: 222).

The formation of a new Medieval materialit y

269

who took possession of all public land and buildings (Halm 1998: 50) and settled in its territory.77 Soon after, a set of local nobles in the Cantabric coast began a rebellion against the Umayyad authorities, in what is traditionally considered the beginning of the Reconquista, the Christian ‘re-conquest’ of the Iberian peninsula from the Moors. Even if the so-called battle of Covadonga (the traditional beginning of the rebellion, dated to 722) may have been no more than a small skirmish, the consequence is that the Christian nobles north of the mountains, led by a nobleman named Pelayo in the sources (Pelagius in Latin, Belay in Arabic), ceased to be politically dependent on the Umayyad government. The military occupation of al-Andalus was brought to the edge of collapse in 741 when a major rebellion of troops in Africa spread into al-Andalus. The Berber Revolt began in Tangiers in 740 as a result of various converging causes (Chalmeta Gendrón 1994; Manzano Moreno 2006: 92-101), including ethnic feuds (Berbers vs. Arabs), unequal distribution of land (the Arabs had the wealthiest arable land) and power (Arabs ruling Berbers), but was finally triggered by the extra taxes and levies required by the local governor in order to raid Sicily. Eventually, the Syrian troops which had been sent to Africa from Damascus were requested in al-Andalus, where the rebellion had spread (Manzano Moreno 2006: 51; Chronica Rotensis, 13-14). They were organized in ajnad (sing. jund), which were territorial armies, and defeated the Berber revolt in 742 (Carvajal López 2013: 60). This political and military situation affected other regions in the Iberian peninsula. The Duero basin and the northern mountains were left, after the revolt, in a situation of a power vacuum (and not a thorough depopulation, as has been traditionally proposed; Martín Viso 2009a). This disrupted the pre-existing model of rural and urban settlement in a wide area between the Duero basin and the western Pyrenees and the upper Ebro valley, which from then on developed along lines different to those set in al-Andalus (see further below).

Early material traces of the newcomers One of the first consequences of the conquest of al-Andalus was the settlement of a set of newly arrived peoples, with a distinctive material culture 77 It should be noted that for most rural areas beyond towns we have no information about how they were conquered.

270 

The Iberian Peninsul a bet ween 300 and 850

and social system. As opposed to what happened with the arrival and settlement of the Germanic barbarians in the fifth century (see Chapter 4), the Islamic settlement had a deeper impact on the archaeological record. For one, it introduced in the Iberian peninsula a considerable (but unknown) number of an ethnically different population with a language and religion that was distinctively different. Settlement in the newly conquered Iberian territory was done largely on an ethnic basis since the Umayyad army was not a homogenous one, but a conglomeration of soldiers of different origins (Manzano Moreno 2006: 142). Ethnically, the troops were divided into Mauri (or ‘Moors’, inhabitants of the old Roman territories of Mauretania) or Berbers (a word which derives from ‘barbarian’ and refers to the inhabitants of the fringes of the Roman border in North Africa); a few of their leaders were actually Arabs. Their number is difficult to estimate, but they must not have been more than a few tens of thousands. These first conquerors/settlers attempted to link themselves with the local landed aristocracies through marriages, as a way of securing their landholdings and to further legitimize their political position (Manzano Moreno 2006: 239). However, in this case (and as opposed to what had happened with the Germanic invaders) this eventually led to a process of acculturation which went only in one direction.78 When it came to the distribution of lands, the Arabs settled mostly on the southern coast, and in the fertile valleys of the Guadalquivir and Ebro rivers, whereas the Berbers settled in the more barren lands of the Central Meseta and the Valencian coast. As had happened with the Barbarians, the alliance with the Hispanic elites was desirable although not wholly necessary, as they still held much of the land and wealth. Some of these noble Visigothic families converted to Islam, becoming muwalladun, or converts, in order to succeed in the new administration and avoid extra taxes, obtaining the benefits of being included within the tribal network of patronage. However, in the same way that not all of the baladiyun, as the early conquerors identified themselves (Cruz Hernández 1998: 54-5), sought to link themselves to muwalladun families, not all of the Christian nobles decided to mix. The powerful elites 78 Currently one of the main discussions on early Islamic Iberia is the alleged tribal system of the settlers, which has been presented as a closed system, opposed and antagonistic towards non-members, with a strong internal hierarchy and strengthened by patronage preserved through agnatic families. The tribal system would have introduced new, parallel vertical hierarchies in the Iberian peninsula which were opposed to the horizontal rank and status pre-Islamic social system. This is however debatable, and it goes beyond the reach of the material culture and the scope of this book (cf. Carvajal López 2013; Glick 2005: 147-9; Guichard 1976; 2002; 2007).

The formation of a new Medieval materialit y

271

of Mérida, Toledo and Zaragoza, for instance, remained largely Christian, but there were pockets of urban and rural Christian communities which never fully converted. The 722 rebellion in Asturias certainly did not inspire the creation of a new Christian kingdom trying to ‘re-conquer’ the Visigothic kingdom,79 but it did underline the lack of effective control of the early Umayyad administration over those mountainous zones and the lack of cooperation from some of the Christian elites. During the early years of the conquest, the Umayyad presence was mostly military in nature. Theoretically, the settlement of newcomers was restricted to garrisons which controlled a series of tax-collecting territories. These troops were paid for by the taxes (the diwan) that were extracted from the surrendered lands, according to the terms of the treaties. This situation changed after the Berber Revolt, as the Syrian troops of the ajnad took over and settled in lands which had previously been under the terms of the sulh, but which after the rebellion were considered by the Syrians to be conquered land. These lands taken by force were distributed in shares: a fifth to the caliph with the remainder allocated to the Umma (that is, they were probably distributed amongst the supporters of the Umayyads). The appropriation of lands and the settlement by the Syrians can be seen as an exchange in lieu of payment. This underlines the fact that, as Chris Wickham has pointed out (2005: 101), there was a lack of a fully-functional, pre-existing, taxation system in al-Andalus.80 Whereas in other regions of the caliphate, such as Persia and Egypt, the Umayyads found functional tax systems they could usurp, in Hispania the Visigothic tax system may not have been as effective; the only way to secure the establishment of a tax collecting system would be to distribute and settle the conquering army in the territory, rather than garrison it. Within the archaeological record, there is a range of evidence which, from 711 onwards, clearly indicates the presence of an Islamic population, specifically the conquering troops. Firstly, there is a series of gold coins, minted probably while troops were on campaign in order to pay them (Bates 1992; Cano Ávila 2005; Canto García 2011). These coins (figure 8.1) are the earliest series of Umayyad coins in al-Andalus, and are of great importance because they form three consecutive series, each with very narrow chronologies, which indicate a constant minting and distribution of these coins. What is perhaps most 79 The claim that it was is a later invention during the reign of Alfonso III in the late ninth century, as part of attempts to legitimie Alfonso’s rule. 80 Chris Wickham actually refers to the Visigothic system, but it can be extrapolated to the early Islamic period.

272 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.1 Umayyad coin dated to the year 93H, with a Latin legend with the mint mark for al-Andalus (the star and the legend SPN, for Spania)

Photograph by Tawfig Ibrahim, Tonegawa collection, used with permission

interesting are the legends they bare, for the earliest in the series is inscribed with the Latin ‘in nomine domini nisi deus sed deus solus non deus similis’ (‘in the name of the Lord, [there is] no god but God alone, there is no similar god’). These coins were a strong instrument of propaganda for both the local populations and the newcomers. They also had the Latin mint mark SPN or SPAN (for Spania), replaced by al-Andalus by 716-717, and a characteristic ‘star’ in the centre. The dissemination of this religious message, aimed at reinforcing the conversion of the local population, was preserved in the subsequent coin series, but with slight changes. The second series (715-717) was minted with the Latin text on one side and the Arabic text on the other, whereas the third series, minted only ten years after the first, was by then fully in Arabic.81 This early Umayyad coinage was, furthermore, completely different in its metrology and weight to earlier Visigothic counterparts. These first coins are few and unevenly distributed, which may suggest that minting was not strictly controlled by the central administration and that they were minted on an ad hoc basis. Furthermore, the earliest Umayyad coins return to a three-metal system, with gold dinars, silver dirhems and copper fels, perhaps in a direct attempt to revitalize monetary exchange, disrupting the Late Antique and Visigothic practice, although they were not regularly minted, nor were all three metals minted simultaneously. 81 It should be mentioned, however, that these early Andalusian coins follow not the patterns of Damascus, but those of Ifriqiya, on which al-Andalus was dependent.

The formation of a new Medieval materialit y

273

Figure 8.2 Arabic lead seal dated to 100-102H, bearing the legend ‘in the name of God, this is the treaty [sulh] of / Abd Allah ibn Mali’

Photograph by Tawfig Ibrahim, Tonegawa collection, used with permission

Directly related to the payment of the troops is a set of lead seals (figure 8.2) found at the hilltop fortified site of Ruscino, on the French slopes of the Pyrenees, in Septimania (Sénac 2013; Marichal and Sénac 2007). These seals are inscribed in Arabic with kufic writing, and have been linked to the sealing, prior to distribution, of chests of loot. The whole site can be dated very early in the eighth century and it is the earliest evidence for the presence of Umayyad troops beyond the Pyrenees. A last set of evidence of the very earliest settlement are the new suburb which emerged as a garrison camp in Córdoba south of the river, called Šaqunda (Murillo et al. 2004; Casal García 2008) along with the foundation of Algeciras and Tarifa, fortified harbours opposite Gibraltar and Ceuta, and the landing sites for all troops coming over from Africa (Torremocha Silva 2002). The transformations in pottery productions throughout the eighth century are very clear (Alba and Gutiérrez 2008), and may point towards the introduction of new techniques by the Islamic settlers: the depuration of clays, the increased homogeneity in the firing, higher firing temperatures, and thinner walls all point towards the appearance (again) of the fast, foot-powered wheel and of professional potters. Furthermore, this is something that is noticeable across the peninsula, despite the regionalization and lack of standardization of the various regional productions; this again highlights the homogeneous

274 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.3 Pottery jug from the Islamic phases of Casa Herrera, Mérida. Note the characteristic high handle, but with a fabric still reminiscent from the earlier phase

Photograph: MNAR, used with permission

foreign component in these technological innovations. These enabled the production of pots which were less bulky, taller, more refined; handles became more stylized, and protruded beyond the horizontal of the rim (figure 8.3). One specific example of production innovation is the way cooking pots are detached from the wheel after they have been thrown (by cutting of the base, leaving a convex mark, rather than simply separating the base from the wheel with a wire) as seen in examples from the region of Granada (Carvajal and Day 2013). A real innovation which appears in eighth-century pottery from Mérida are the first examples of Islamic medieval glazed wares (Alba Calzado 2001: 280-1). These early glazed wares (or rather vitrified) are few in number, and have a poorly finished, quite thick glazing (only in the inside, for purely functional rather than decorative purposes), which usually belongs to open forms. They typologically belong to Late Antique forms, but have a new imported finish. These transformations are similar to what is seen in

The formation of a new Medieval materialit y

275

Figure 8.4 Ceramic typologies from South-Eastern Spain in the late eighth century (according to Carvajal López 2009): a-c, j) cooking pots; d-f) ARS imitations; g) jar; h-i) storage; k-o) new types of cooking pots

the pottery sequences of El Tolmo, where cooking pots, for instance evolve towards flat bases and increasingly narrower mouths, with thinner walls, and with new red painted decorations appear (Gutiérrez Lloret 2011: 204). Similar developments can be seen from excavations in Córdoba (Casal García 2008). These early productions, however, still reflect local production and distribution patterns, similar to that which existed in the post-Roman period. Typology-wise (figure 8.4), new types appear in the Iberian peninsula after the Umayyad invasion, and they exist in parallel with pre-existing forms until the ninth century, when traditional Late Antique forms are replaced by the imported shapes, although there was a short period of hybridization (Carvajal López 2009: 391; Carvajal and Day 2013; cf. Gutiérrez Lloret 2011: 203). The first of these forms is the tannur or pottery bread oven, characterised by its conical section. The second form is the noria-jar, which was connected to water wheels to raise water from wells, usually for the purpose of irrigation. The third form is the drinking jar, a small, round-bodied vessel with a straight wide neck and a protruding lip with a small handle, which was used to drink from and replaced traditional, post-Roman drinking shouldered-bowls. A fourth form is the long-nozzled lamp, which obviously has a long nozzle, but is also characterized by its much smaller body and a tall-necked pouring hole, in

276 

The Iberian Peninsul a bet ween 300 and 850

contrast with the Roman and Late Antique models, which has a short nozzle and a large, rounded body (which allowed the top disc to be decorated). A final example would be the ataifor or serving tray, which despite evolving from late Roman paterae, in its Islamic form imitates Tang Chinese vessels. When they appeared in al-Andalus later in the ninth century they quickly replaced all previous forms of trays (Rosselló Bordoy 1986). These illustrative cases underline how the arrival of the Islamic settlers (mostly, but not exclusively troops) left clear evidence in the material record. In order to see consistent transformations in the material culture across the peninsula, we must look at longer processes of change.

Archaeologies of power The Umayyad state formation In the years following the Berber Revolt, until the 750s, the governor of alAndalus was Abd al-Rahman al-Fihri, who ruled virtually independent from Damascus. In 750 the entire Umayyad family was killed in Damascus, to be succeeded by the Abbasids as rulers of the Islamic world. However, a single Umayyad boy, Abd al-Rahman (born 731), managed to escape. Abd al-Rahman landed in Iberia in 754 and, after two years of civil war against al-Fihri, took full control of the peninsula and proclaimed himself independent emir with the support of the Syrian troops and other Umayyad clients (Manzano Moreno 2006: 105). This declaration of independence led to the reorganization of the Umayyad administration of al-Andalus, and it set in motion a process of state formation, not too dissimilar to the Visigothic one from two centuries earlier. The original reforms of Abd al-Rahman I (r. 756-788) were further developed by his great-grandson Abd al-Rahman II (r. 822-852). They included the development of a centralized administration, based around Córdoba, and the creation of three military frontiers, with capitals in Mérida (Lower March), Toledo (Middle March) and Zaragoza (Upper March), with the rest being subdivided into various districts. The emir was surrounded and advised by a series of palatine officers in imitation of Abbasid practices in the East (Chalmeta Gendrón 1998: 26-8; Lévi-Provençal 1957: 5; Souto Lasala 2001: 29). The Umayyads consolidated their support by promoting and protecting a few small circles of pro-Umayyad networks of patronage (which included Arab leaders and their clients, the ajnad, and some groups of pro-Umayyad Berbers). This was, in the long term, to the detriment of a large part of both the Berbers and the muwalladun. These reforms were

The formation of a new Medieval materialit y

277

backed with the reorganization of the monetary system introduced by the early governors (the ‘conquest solidi’ noted above). In this process of state formation, towns were an essential element and, paraphrasing Hugh Kennedy (1985) and Sonia Gutiérrez (1993), there was an administrative transition from the civitas to the madinah. Most Visigothic civitates, especially in the south, became mudun (plural of madinah) and they were soon ‘upgraded’ with monumental elements befitting the new Umayyad administration, such as the urban palaces of Mérida (see below). These towns were not organized around territories that could be traced to the Visigothic administrative territories, but rather amalgamated into different types of administrative and tax divisions (of a complex and confusing terminology), which changed Late Antique townscapes (cf. Navarro and Jiménez 2007).82 The local administration was based on appointed ‘magistrates’ (which marginalized older urban elites) and taxation collected for the central administration (Clarke 2012: 40-1). From a material and monumental perspective, this new administrative situation required new types of materialities that would legitimize the new polity. The transformations of urban monumentalities were carried out usually to the detriment of pre-existing buildings and urban layouts. Unlike the post-Roman curiales or the Visigothic monarchy, the Umayyads did not need to look back at the late Roman past to legitimize their position. The Roman past was not an element to preserve or to relate to, a clear indication of the change in the state mentality and of the beginning of Medieval processes (although there are some examples of the conscious reuse of Visigothic sculpture in new, demeaning contexts). For the Andalusian Umayyads, the mirror to see themselves in was in Damascus and Syria. With these clear points of reference to legitimize their position, the Roman past was little more than a background noise. This new political architecture can be seen in the construction of new urban military and administrative foci, which altered the Late Antique monumental townscape of the old Visigothic centres. These can be seen in two different types of building and the development of new settlements. On a minor scale, this can be seen in the construction of new ‘palaces’. The excavations at the site of Morería, in Mérida (Alba Calzado 2001: 288-95), have revealed a series of these ninth-century Umayyad ‘palaces’ (figure 8.5). These large buildings, built in reused masonry, are all different, but share an axial symmetry, with a central hall, on to which side rooms open. They must have had a second story, as suggested by the thick walls and the buttresses that are built against the outer walls, suggesting perhaps the presence of 82 Such as iqlim, kura and yuz, cf. Jiménez Puertas 1995: p. 77 (n. 35).

278 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.5 Plan of the Umayyad ‘palaces’ excavated at the site of Morería, in Mérida, which were built on top of the domus of the Marbles

From Alba Calzado 2001, figure 7

vaults or arches. Part of a similar building has been identified, also in Mérida, in front of the ‘Temple of Diana’, reoccupying the forum (Mateos and Sastre 2004). The layout, size and quality of these buildings have to be related to some important, probably public, function, so they have been interpreted

The formation of a new Medieval materialit y

279

as residences of the new urban elites. It is difficult to determine the relative chronology between these structures and the citadel, but it is not difficult to see, as suggested by Miguel Alba, a correlation between a military and a civil administration in this area of the city. Similar structures have been identified at La Almoina in Valencia (built on top of the old forum and supplied by an aqueduct; Martínez Jiménez 2011a) and perhaps in Barcelona (see Chapter 5). Similar examples are assumed to have existed in Córdoba (the balats), although these have not yet been identified (León and Murillo 2014). On a more monumental scale, this can be seen in the construction of new large urban citadels. The decision of the Umayyads to establish their seat of power in Córdoba was to the detriment of Toledo. The fact that there was a strong elite (ecclesiastical and administrative) and powerful Christian families resident in the old capital was perhaps one of the factors taken into account when this decision was taken. In Toledo (Almagro Gorbea 1987: 432; Delgado Valero 1999: 39-52; Pavón Maldonado 1992: 284), the local inhabitants repeatedly rebelled against the power of the emirs, at least six times between 743 and 797. This probably explains the abandonment of the royal suburb of Vega Baja (Izquierdo Benito 2009) in favour of the upper town (which is more easily defendable), and the construction of the 6ha almudayna or citadel, which included a small walled enclosure and the castle, probably on the same area where the Visigothic palace could have been. This was built in 806 and its main aim was to defend the garrison from the local population, rather than to defend the inhabitants from external attacks. It was independently connected to the bridge that gave access to the road to Córdoba, so reinforcements could be called in if necessary. Mérida, the second most important city in the Visigothic kingdom, and with an extensive architectural display of Visigothic and Christian power, suffered a similar fate to Toledo: a rebellious Christian city eventually put under the control of a military force stationed in an urban citadel (Alba Calzado 2001: 282-8). The alcazaba of Mérida (figs. 8.6a-b), built in 835, as known from its two monumental inscriptions (Barceló Torres 2002b),83 is a large square fortress, with square towers, and built out of reused material extracted from the fifth-century wall reinforcement (which was itself reused 83 Translation of the inscription preserved in Mérida, following Carmen Barceló’s reading: ‘In the name of God, the Clement, the Merciful. Blessings of God and His protection for those who obey God. The emir Abd al-Rahman [II], son of al-Hakam – may God give him glory – ordered this fortress (hisn) to be built as a place of refuge for the obedient [to the emir], by the means of his governor (amil) Abd Allah, son of Kulayb bin Talaba, and of his servant (mawla) and Master Builder (sahib al-bunyan) Hayqar b. Mukabbis, on the moon of the last rabi of the year 220 [= April 835]’.

280 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.6a Photograph of the fortress giving access to the alcazaba of Mérida, including its dedicatory inscription. This Umayyad fortress was built to keep the peace in the city after various years of rebellion by the local muwalladun

The formation of a new Medieval materialit y

281

Figure 8.6b Plan of the alcazaba of Mérida, showing the entrance fort, which controlled access to the whole city, and the location of the cistern

Creswell 1940, figure 171

Roman masonry). It was built inside the city to further reduce the risk of rebellion, and it was linked to a series of watch towers in its hinterland. The citadel has a small bastion which controls the access to the bridge, so that only through the citadel was it possible to access the city. The interior of the citadel was an open ground but for a single large tower which defended the large cistern decorated with reused Visigothic pilasters and that included a praying room with a qiblah wall and a mihrab. The citadel was a symbol of the Umayyad power over the people of Mérida (controlling the bridge and destroying the bridge gate, symbol of the city, together with the reuse

282 

The Iberian Peninsul a bet ween 300 and 850

of the Visigothic pilasters). Despite its scale, the inhabitants kept rebelling until 862, when the emirs decided to destroy the walls and abandon the city, moving all the administration of the territory to a new Umayyad foundation, Badajoz. In Córdoba, the Umayyads had the seat of their power in the old Visigothic palace, Vallat [= Palatium] Ruderici or ‘Ballat Ludhriq’, which sat on the southernmost corner of the walled city and directly opposite the Great Mosque (see below). The current building of the alcázar is a palimpsest of the various phases that the building went through, heavily modified during the caliphate and the later Medieval centuries. Very few elements remain from the early Umayyad phase, but it seems to have reused and improved most of the Visigothic civil complex and occupied the south west corner of the city walled enclosure (see Chapter 5). Throughout the emirate it would be further enlarged and enclosed in a wall, and equipped with a reception hall (León and Murillo 2014: 10-3). The palace and the mosque formed a large complex, which visually reflected the religious and civil power of the Umayyads. At some point in the ninth century, the alcázar was supplied with a brand new aqueduct, the qanat Amir (Pizarro Berengena 2012: 137-40), which as opposed to earlier (i.e., Roman) aqueducts, was only built to supply the palace complex, and not the town. Besides the construction of new centres of power in existing towns, the Umayyad state was clearly visible in the promotion and foundation of new sites, which in many cases replaced Visigothic administrative centres.84 In the cases of Reccopolis and Eio, for example, two new alternative sites were established to counter and substitute the old Visigothic nuclei. Reccopolis, which slowly declined as an urban nucleus during the early eighth century was countered by the establishment of a jund garrison on a hill opposite the city (figure 8.7), which became the town of Zorita (Olmo Enciso 2000: 393; Al Razi 20; Sanz Paratcha 2008: 166-73). After the 768 rebellion of Saqya 84 This is a process which had already begun in the mid-eighth century with the arrival and settlement of the ajnad, but which was later expanded and developed, especially as administrative centres were established around newly founded qila’ (pl. of qal’a), or castles. The qila’ were not meant to be new urban foundations; they were simply small fortified enclaves for the settlement of the newcomers (Acién Almansa 1995) that would, nevertheless, function as economic and political centres of a district created around them. However, either because of conscious policies against old nuclei or simply because the new sites attracted money and population, many of old Roman settlements in the vicinity of these sites were gradually abandoned in favour of the new, more dynamic, Umayyad settlements: Reccopolis by Zorita, Ocillis by Medinaceli, Mentesa by Jaen, Traducta by Algeciras, Complutum by Alcalá de Henares, Oretum by Calatrava, etc. (Espinar Moreno 1991: 215; Manzano Moreno 2006: 260-5; Torremocha Silva 2001: 305; Salado et al. 2002: 366-70).

The formation of a new Medieval materialit y

283

Figure 8.7 View of the settlement of Zorita from the hill of Reccopolis

al-Minkasi against the Umayyads the jund of Zorita became increasingly more important as a way of controlling the territory. In this second emiral phase, the ‘palace’, which had been abandoned, was turned into a small fortress, and the monumental arch was turned into a gate, and even sunken features appear in the upper complex, while the rest of the site largely ceased to function as a town. Before the turn of the century, the palatial complex/ fortress was burnt and Reccopolis finally abandoned. The development of Zorita is less known, because it is currently under the modern village, but from the remains which still stand it is clear that it included a fortified garrison on the hilltop and a walled lower town. Similarly, Eio (which was continuously occupied despite the 711 conquest, and by then known as Iyyuh) was destroyed, together with the site of Mula, by the emir in 826 in order to create the new city, Murcia, that would centralize the administration of the district (Abad et al. 2008: 334; Gutiérrez et al. 2005: 352; González et al. 2010). As an Umayyad new urban foundation, Murcia (figure 8.8) represented the strength of the state. It was settled with the displaced population of Iyyuh and with new troops. The town was equipped with a brick mosque, a stone alminar, an alcazaba for the governor, and a set of baths, all inside the walled enclosure. The internal layout of the settlement seems to have

284 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.8 Plan of early Umayyad Murcia, a new urban foundation to substitute the muwallad stronghold of Eio

Based on Navarro and Jiménez 2007, figure 5

been, in the very early period, quite lose and open, with large plots assigned to the first settlers, in which not just houses and workshops, but pens and gardens seem to have been in existence (Navarro and Jiménez 2007: 283). Similar to Murcia is the site of Ilbira (Malpica Cuello 2010), which substituted the old Roman town of Iliberris (which had ceased to function as a town, cf. Carvajal et al. 2015). The site of Ilbira is currently the object of a research project and offers perhaps the best example of an early Umayyad new urban foundation. The site includes a large fortified enclosure and an alcazaba, and it functioned as the main administrative centre of the region, so it has been directly linked to state intervention, or at least state encouragement, on a pre-existing smaller rural settlement. The town included besides the administrative/military centre a well-defined production quarter (pottery kilns) and dwelling areas towards the east, over the Late Antique and Roman settlement. Regarding construction techniques, various elements of Roman construction had been lost through the post-Roman period as a result of the simplification of the economy and the collapse of the demand for such specialized trades. This situation was reversed after the establishment of the Umayyad emirate, when construction techniques, and even builders’ tools (such as compasses and set squares) are evident again in the archaeological record (Caballero and Utrero 2008; Paz Peralta 2015: 42). The swift development of

The formation of a new Medieval materialit y

285

the horse-shoe arch and its spread across the peninsula during the Umayyad period is a clear example of this building revolution. The development of a new set of state-sponsored building programs not only prompted the reopening of quarries, but also underlined the need for specialized builders and architects. Furthermore, of the main construction projects that can be dated to the early Islamic period (before the foundation of the emirate) all are built with reused material, take advantage of pre-existing buildings, which are later modified, or simply adapt Roman buildings – there were no truly new constructions (Morales Martínez 1995: 43). This changed abruptly with the creation of the emirate of Córdoba and the new Umayyad propagandistic building program.85 The transformation of the ancient central places, the development of a new form of architectures of power, and the construction of new centres indicate the progressive formation of a new centralized state in a recently incorporated territory. During the early period, the Umayyad state relied on previous political institutions, thus re-appropriating ancient monumental elements of power, but it was also highly creative and innovative in constructing new material forms. The new Umayyad state progressively dissociated itself from previous models, ultimately developing a unique materiality, which ignored the Visigothic past and would itself be ignored by the Christian north. The Christian polities In the northern mountainous corners of the Iberian peninsula small armed groups emerged which were opposed to the Umayyad hegemony, and formed their own states in the eighth and ninth centuries. These areas were hardly urbanized in the Visigothic period, and the materialities of power were displayed differently. Despite this, the different processes of state formation of Asturias, Pamplona and Catalonia prompted the emergence of new forms of displaying power, which quickly substituted the old Late Antique models. Initially in Asturias, local elites (including the ‘kings’ of the early eighth century) had abandoned the Roman urban sites (such as Gijón, where the Umayyad authorities were based; Chronica Rotensis 11; Fernández and Gil 2007: 410) favouring smaller, non-urban settlements, in a similar fashion to the other local aristocracies of the north west which were mentioned in Chapter 6. It was at these small sites, such as Cangas or Pravia, that these 85 Which would fully develop under the caliphate of the tenth century (cf. Gurriarán Daza 2008).

286 

The Iberian Peninsul a bet ween 300 and 850

early local leaders had their central places. In the manner of other aristocrats in the Visigothic period, ‘kings’ such as Favila (r. 735-737) dedicated new churches (Rico Camps 2014; cf. ICERV 315 = IHC 149). However, as time passed the Asturian monarchy began its own process of state formation, attempting to legitimize itself by claiming direct links with the Visigothic monarchy, although this discourse seems to have been a later development, in the ninth and tenth centuries (Isla Frez 2011). In its early beginnings, it is conceivable that the Asturian elites had their foci of power on hillforts, such as Gauzón, following the patterns of post-Roman elites in this area (see Chapter 6). However, as Asturias grew as a political entity by aggregating the Christian elites in Galicia and Cantabria, it required the development of stronger and more economically active nuclei. This had to be done in a territory which lacked a previous (Late Antique) strong urban tradition. The kings of Asturias tried to exercise their power in the manner of the Visigothic monarchs, but there was not an urban network on which they could base their power, so instead they had to rely on previous settlement patterns and social configurations (Carvajal and Martín 2013). Their efforts and the architecture they used may have been deeply based on Visigothic patterns, but the whole concept of Late Antique city had vanished. Oviedo, Pravia, and Gijón can be seen as a new form of settlement: they were Medieval centres of power, and very different from the Late Antique towns which had disappeared from the area during the previous centuries. Traditionally, a first monumental attempt has been linked with the construction of a palatine church at Pravia (old Roman oppidum or fortified settlement) by Pelayo’s successor, self-proclaimed princeps Silo. This church to Saint John (allegedly part of a larger complex, a Palastkirche) is commemorated with a large labyrinth inscription, which is only preserved in fragments, in which Silo princeps fecit (‘Prince Silo made [it]’) can be read from the central S in every single direction in over 45,000 combinations (IHC 175; Rico Camps 2014). However, recent analyses of both the labyrinthian inscription and the church itself (Caballero and Rodríguez 2010; Gimeno and del Hoyo 2012) suggest that both are more likely to be tenth century in date, raising the issue of whether these are original eighth-century elements or, more likely, later legitimising attempts from a period when the Asturian monarchy was already firmly established. The first properly-dated urban attempt dates to the year 791, when King Alfonso II (r. 796-842) founded the city of Oviedo, which was to become (and still is) the capital of Asturias (García de Castro 1995; García de Castro and Ríos 2016). According to the Asturian chronicles written under King Alfonso III (r. 866-910), the reign of Alfonso II saw the first large urban

The formation of a new Medieval materialit y

287

renewal of the kingdom, as he was linked to the construction of several churches and a new palace. And not only in Oviedo. At the same time, the discovery (inventio) of Saint James the Apostle’s tomb took place in an old Roman burial site near Iria Flavia, in the current Santiago de Compostela where a new town and monastic complex were built (cf. López Alsina 1994). In Oviedo, Alfonso also built a new basilica ‘cum bis seno numero apostolorum altaris adiungens’.86 This, together with the complex at Santiago, was meant to replace the episcopal sees of Toledo and Mérida, the Visigothic political and religious capitals that were under the rule of the Emirs of Córdoba. As has already been noted, the town was further equipped with a royal palace and a palatine church, although most of these works were completed later than our period during the early ninth century. Surprisingly, Alfonso did not truly found Oviedo ex novo, as is evident from its foundation inscription (now lost): Quicumque cernis hoc templum dei honore dignum, (g)noscito hic| ante istum fuisse alterum, hoc eodem ordine situm, quod princeps| condidit Salvatori Dom(i)no supplex per omnia Froila, duodecim| Apostolis dedicans bissena altaria; por quo ad Deum sit vestra| cunctorum oratio pia, ut vobis det Dominus sine fine pra(e)mia dig|na. Praeteritum hic ante h(a)edificium fuit partem a gentilibuus di|rutum sordibusque contaminatum, quod denuo tuum a famulo| Dei Adefonso cognoscitur esse fundatum et omne in melius re|novatum. Sit merces illi pro tali, Christe, labore et laus hic iugis| sit sine fine tibi. Whoever looks at this honored temple worth of God, know that there was another before this one, placed in this very manner, which prince Fruela, supplicant of the Lord Savior, founded, offering twelve altars to the twelve Apostles. Because of this may your prayer be pious towards God, so that the Lord may give you worthy reward. This earlier building was partially destroyed by the pagans (gentilibus) and desecrated by the unclean (sordibus), so it was totally founded again by the serf of God Alfonso, and made better in everything. To him, oh Christ, may grace be conceded for such a work, and to You, here, never-ending praise.

The inscription suggests that the foundation of Oviedo was actually the urbanization of a pre-existing Christian settlement on the hill, where King Fruela (r. 757-768) had already built a basilica or monastery, of which there 86 Chronica Rotensis 21: ‘[W]ith twelve (bis seno) altars in accordance with the number of Apostles.’

288 

The Iberian Peninsul a bet ween 300 and 850

are several archaeological remains, and also a palace (García de Castro and Ríos 2016; García-Sampedro 2009). Oviedo may not have had a Roman or Visigothic past, but its landscape is dotted with late Roman villas, a fountain spring and various wells, and a postRoman hillfort. There is even a reference to a place of the Asturians called Ovetum in a Scottish sixth-century inscription. This early urban foundation, however, was a monumentalized, small centre of power, compromising a series of large royal and church buildings (the cathedral to the Holy Savior, chapel and royal mausoleum of Saint Mary, palace, episcopal complex, monastery of Saint Vincent, a monumental fountain, etc.) but without any trace of the kind of urban planning that would characterise Late Antique towns (figure 8.9). From the ninth century Oviedo had a wall, inside which the various royal buildings seem to have been located, not concentrated as in Reccopolis or Eio, without traces of a real urban life. In either case, the small fortified wall enclosure of Oviedo, together with its centre of religious and civil power, embodied the formation of the Asturian monarchy. This was a new form of urbanism that did not look back to previous Roman patterns. It was during the reign of Ramiro I (r. 842-850) that the Asturian kingdom was firmly established, in the middle of both internal and external conflicts (from palatial unrests to Viking attacks), and it is under this king that the main Asturian monuments were built (García de Castro 1995; Arias Páramo 2008; Caballero and Sastre 2014: 287; Walker 2016: 158-68). The church of Saint Michael de Lillo and the so-called ‘palace’ (its true original function is currently under discussion) of Saint Mary’s, built at the Mount Naranco (‘Mons Naurantius’), are the best examples of the new ‘Asturian’ art, which broke away from late and post-Roman models (f igure 8.10). This new architecture is characterized by the common use of newly quarried stone and large vaulted spaces, although the buildings are quite narrow and are supported with reinforcing buttresses (Sastre and Utrero 2015) and have to be seen as Ramiro’s display of monumentality, contrasting himself with what Abd al-Rahman II was building at the same time in Córdoba. In this respect, Ramiro’s monumentality is a clear indicator of the process of state formation of the Asturian monarchy, the need of justifying his position in the face of neighbouring leaders, the emir and the Carolingian emperor, by presenting himself as an equal, producing a set of monuments that would serve as a model for the future. During the eighth and ninth centuries other territories also went through processes of state formation, if without reaching Asturian standards of materialization and centralization of power. The territories of the Pyrenees

The formation of a new Medieval materialit y

289

Figure 8.9 Plan of the early remains at Oviedo (according to García de Castro and Ríos 2016, figure 15), indicating the location of the main early medieval monuments: 1) Monastery of Saint Pelagius; 2) Monastery of Saint Vincent; 3) Saint Mary and Royal Mausoleum; 4) Cathedral of the Holy Savior; 5) Church of Saint Tyrsus; 6) Episcopal complex and palace; 7) Ninth-century castle

became increasingly enclosed in the political orbit of the Carolingian Empire (the ‘Hispanic March’). At the eastern end, in Catalonia, this was achieved through direct military intervention (including the conquest of Barcelona in 801). The creation of the ‘Hispanic March’ involved the creation of new county capitals, which would eventually develop into urban settlements, but that

290 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.10 Santa María del Naranco church, near Oviedo. It was originally a royal chapel built by King Ramiro I in a completely new architectonic style, taking advantage of the vaulting techniques reintroduced into the Iberian peninsula by the Umayyads

in their formative stages in this period are very difficult to label as such (Banks 1999). Only Barcelona and Gerona were proper urban sites, as they preserved their structure and importance throughout the post-Roman period. In these two cities, it seems that the Late Antique town was largely unchanged, and the Carolingians simply managed with the pre-existing urban infrastructure, although Gerona had its wall enclosure expanded and improved (Beltrán de Heredia 2008b; Nolla and Sagrera 1999). Besides these, there were attempts to set up new central places at fortified locations which had urban characteristics, as it seems from the excavations at L’Esquerda (the town of Roda), or the foundation of Vich outside old Ausona (Vich derives from vicus Ausonae, or the suburb of Ausona; Ollich Castanyer 1999; Ollich et al. 2016). Old Roman sites, such as Ampurias, Jaca, and Urgel also became county capitals. In the western Pyrenees (Collins 1990; Faro et al. 2008), the territory of the Basques had been in the area of influence (if not directly controlled) of the dukes of Aquitaine, who were independent from the Carolingian rulers and formed a stable polity between the Visigothic and

The formation of a new Medieval materialit y

291

Frankish kingdoms. Pamplona was taken over by the invading Umayyad troops in the early eighth century, and due to its strategic location in the high Ebro valley it was in constant dispute between Asturian kings, the Carolingians, and the Banu Qasi (Lorenzo Jiménez 2010), all of which aimed at conquering Pamplona, as it was the only fortified city in the area and a strategically located site.87 Like Barcelona, it is only in the tenth century that the pre-Romanesque cathedral was built (Mezquíriz and Tabar 1993-1994), and there are no monumental constructions known for the interim period. As is also the case with the Catalan counties, it appears that the development of a new architecture of power (Romanesque) is only related to the (much) later process of state formation. The dismantling of the Visigothic kingdom created a power vacuum in the northern territories. The three new polities which emerged in the north during the eighth and ninth centuries (the Asturian Kingdom, the Hispanic March and Vasconia) developed different forms of materialities for public representation and the imposition of their power and ideology, positioning and comparing themselves against the Umayyad state.

Archaeologies of religion and identity in al-Andalus The arrival of a new faith in Iberia in 711 had a direct impact on the material culture (Zozaya Stabel-Hansen 1998). Islam slowly replaced Christianity as the main religion throughout most of the peninsula, becoming the majority religion only after the ninth century, following an increasing rate of conversion (Aillet 2010). This implies that in the Early Medieval period Muslims were in the minority, but their position of power is clearly reflected in the material record. Whereas in the Christian north we have already noted the continuity and evolution of Christian architecture in Asturias, in al-Andalus Christian monumentality, which had flourished in the Visigothic cities, had to give way. The monuments that best represent the establishment of the Umayyad emirate, and that are one of the key elements that could be expected to appear after the Islamic conquest, are mosques. This is not only because of their religious significance; the Umayyad rulers were, by definition, the leaders of the community of believers, the Umma, both in a religious and political sense (Crone 2005: 3; Santillana 1926: 2). By building 87 The Banu Qasi were a powerful muwallad dynasty who ruled semi-independently the Upper Ebro Valley for most of the ninth century; the name relates to the founder, count Cassius, who probably surrendered and converted during the early years of the conquest.

292 

The Iberian Peninsul a bet ween 300 and 850

mosques the Umayyads were facilitating the cult of the Umma and, in turn, the name of the Umayyad ruler would be proclaimed in all Friday prayers. In this sense, mosques were as much as a political statement as a religious one. Mosques (Insoll 1999: 26-33) are easily identifiable due to their particular orientation towards Mecca (qiblah), which in Umayyad al-Andalus is towards the south east. Technically, a simple wall with a qiblah orientation could be considered a mosque, although from the early eighth century they had to include a mihrab, a niche or apse in the qiblah wall which marks the true orientation of the prayer. Sometimes they include a courtyard with a fountain for the ritual ablutions and a tower from which to announce the five daily prayers. However, the small number of Muslims in the early period may have meant that mosques not an urgent necessity – especially because mosques, as a building, are not an essential requirement for praying. The written sources mention various mosques in the early years of the conquest (which should therefore be see more as political statements than the result of numeric needs), but there is no archaeological evidence for them (Calvo Capilla 2011). In Algeciras, Málaga and Seville such early mosques are mentioned, built by private initiative, and there is mention of a dismantled or re-converted church. There are mentions of possible shared temples, dedicated to Christian worship but used by Muslims at different days and times, as has been proposed for the alleged basilica of Saint Vincent basilica in Córdoba (e.g., Salcedo Hierro 2000: 17-8). However, the sources are quite late, and this could be seen as a literary topos mirroring what happened in Damascus (cf. Guidetti 2014). Even if there is evidence for an eighth-century early mosque in Zaragoza (Hernández Vera 2004), the first large public mosque known through the archaeology is directly linked to the development of the independent Umayyad emirate: the Great Mosque of Córdoba. The Great Mosque of Córdoba (figs. 8.11a-b) was built by Abd al-Rahman I between 784 and 786 (Arjona Castro 2001: 69, 81; Barrucand and Bednorz 2002: 39-46; Salcedo Hierro 2000: 17-8) on top of the alleged basilica of Saint Vincent (according to traditional interpretations), but the lack of solid archaeological evidence calls for a more cautious approach on this issue (cf. Arce Sainz 2015). It is located next to the old Visigothic palace, which had been taken over by the emirs as their residence (see below), so in this way al-Rahman retained the palace-shrine link originally created in the Visigothic period between the spaces of state and religious power (Salcedo Hierro 2000: 174). The reuse of Roman and Visigothic columns and capitals in the arched aisles was a way of indicating a transition between the ‘old’ and the ‘new’ regime, with an appropriation of the past giving it a completely new

The formation of a new Medieval materialit y

293

Figure 8.11a Photograph of the interior of the Mosque of Córdoba, showing the forest of columns and the rows of superimposed arches. The mosque became not only the main religious and political centre of Islam in alAndalus, but also the key dynastic monument of the Umayyads

meaning (Cressier 2001: 319, 323; cf. Van Leeuwen 1999: 8, 181). The design of the mosque indicates a Syrian tradition, as seen in the horse-shoe arches, the roofing techniques and the so-called ablaq technique (use of alternating colours in the masonry: Clarke 2012: 40-1; Creswell 1940: 156-7; Souto Lasala 2001: 30; Torres Balbás 1957: 245-8). All these architectural elements have Roman precedents, but the way in which they were used in al-Andalus was a fully Islamic one, which the Umayyads may have taken from Roman Syria. The mosque became a dynastic monument of the Umayyads, a symbol of their rule in al-Andalus, and so, as time passed by, various Umayyad emirs expanded and developed the mosque. Other than the mosque of Córdoba, there are only a handful of archaeologically known mosques from the early Umayyad period. The best known are perhaps those of Seville and Zaragoza. The construction of ‘main’ or ‘Friday’ mosques (masayid al-jami) are a reflection of the increasing importance of Muslim populations (either convert or immigrant) in the cities, which needed larger meeting places, and which in turn further legitimized

294 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.11b Plan of the Mosque, as it was in its original phase. Originally built by Abd al-Rahman I, it went through major remodelations and expansion, including by Abd al-Rahman II in the mid-ninth century

From Torres Balbás 1957, figure 143

the Umayyads’ power. These main mosques were not only used for ritual purposes but for other events linked to the civil power, and thus became a way of demonstrating their power (cf. Trillo San José 2011). Another archaeological indicator of religious conversion is the spread of maqabir or Islamic burial sites, very clear in the archaeological record as bodies are buried on their side and facing south-east, shrouded, and without grave goods (Insoll 1999: 168-70). The earliest and best known

The formation of a new Medieval materialit y

295

examples are those of Pamplona, Eio, Zaragoza, Écija, Lucentum (Olcina et al. 2008; figure 8.12) and Córdoba (from where we have the largest evidence for Islamic burials, even from an early date; Casal et al. 2006). These tend to be located outside city gates and they do not overlap with pre-existing Christian cemeteries, in some cases cohabiting. The maqbara of Pamplona is the most important of this early period (Faro et al. 2008: 236-47, 178; de Miguel Ibáñez 2007; 2016). It is not located at Obietagaria (the Visigothic necropolis), but directly outside the gate, on top of an old Roman bath. The earliest phase of this burial ground, which contains 190 graves, has been dated to the first half of the eighth century.88 Stable isotope and DNA analyses (together with some other cultural indicators, such as sharpened teeth and inscribed rings) have shown the possible presence of some North African individuals, and a significant number show signs of violent deaths (de Miguel et al. 2016).89 The maqbara of Eio (dated to 800) perhaps shows not the settlement of newcomers, but rather the conversion of the local population: in the same necropolis where the local Christians buried their dead, Islamic burials appear on top of these previous ones, without the stratigraphic disruption that would indicate a different phase (Gutiérrez Lloret 1996: 304). This could indicate that the same population, which had used the cemetery as Christians, also used it as Muslims. A similar pattern can be found at the Toledo Gate necropolis of Zaragoza (Galve and Mostalac 2007: 86-9), where the western necropolis shows uninterrupted continuity of burial grounds into the Muslim period. At some points the superimposition has provided layers five metres thick. Some Islamic burial areas began separated, but grew on top of earlier Christian sites (perhaps as a symbolic statement), as happened in Mérida and in Écija, a maqbara which expanded over 7000 m2. In some cases, such as El Soto/Encadenado in Madrid, dated between the eighth and the ninth centuries (Vigil-Escalera and Strato 2013), separate Christian and Muslim burial areas coexist within the same settlement. Burial practice and religious monumentality were not the only identity markers in al-Andalus. One of the main aprioristic differences between the Islamic newcomers and the local populations was the diet, because Islam has a set of dietary laws which sharply contrasted with Hispanic traditions (cf. Insoll 1999: 95). This has been termed the ‘no-pig horizon’, which is the decline of pig in faunal assemblages from archaeological contexts of the Islamic period, as compared to the pre-Islamic phase. Despite the name, 88 1330±40 BP, 2 Sigma: Cal AD 650-770 (Cal BP 1300-1180 95%). 89 It has been suggested that these burials are linked to the Umayyad garrison.

296 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.12 Plan of the maqbara of Lucentum, located on top of the old foum and dated to between the seventh and tenth centuries, indicating the continuity of population in the area of Lucentum despite the abandonment of the old urban site

Based on Olcina et al. 2008

which derives from the most evident change, this archaeological divide goes beyond pigs. Lawful animals sacrificed in the adequate way (halal) form the theoretical core of the Muslim animal diet. This excludes pigs, abundant otherwise in post-Roman contexts (Grau Sologestoa 2015a: 133-4), but also has to be linked to particular butchering patterns. The ‘no-pig horizon’, perhaps better labelled the ‘halal horizon’, could be expanded to include the prohibition of alcohol and drunkenness, and thus with the decline of particular vessel shapes linked with drinking and fermenting, and the abandonment of wine production sites (Insoll 1999: 99). However, it is evident through the written record that wine was widely consumed in al-Andalus, even if there is no precise archaeological evidence for its consumption or production – except, perhaps, the late phases of the wine presses at the villa of Torre de Palma in Portugal (Martínez Lillo 2008). For the Iberian case, zooarchaeological studies for the transitions between the post-Roman and the early Islamic periods are still incipient (García

The formation of a new Medieval materialit y

297

García 2016). Regarding pigs, however, there seems to have been a slight decline already from the late into the post-Roman period, parallel with an increase presence of sheep/goat (Grau Sologestoa 2015a: 125-50). This has been related to the need to diversify production, and to invest in animals of which secondary products (wool, milk, traction) could be derived, as pigs were only kept for their meat. Similarly, the mean size of sheep in the Islamic period seems to have been larger than it was in the post-Roman and post-Islamic periods (parallel with a decrease in the mean size of pigs and cattle in the same period), further indicating the specialization in sheep herding over pigs and cattle (Villar and García 2016). However, stable isotope analysis on small demographic samples dated to the High Middle Ages from Aragón (Mundee 2009; 2010) have shown that despite the expected difference between Christian and Muslim communities (Christians would have had higher 15N values due to their consumption of pigs, which are omnivores), the difference is minimal. Studies for the Late Medieval period have shown, however, that Christian and Muslim communities in the region of Valencia had different diets (Alexander et al. 2015). In that particular case study this cultural dietary divide was probably also influenced by the different social status of the two communities and cannot only be linked to the aftermath of the ‘no-pig horizon’. Another indicator of identity at first and of acculturation later is the spread of Arabic as a common language, which was used not only by the Muslim population but also increasingly by Christians. The spread of the Arabic language was, despite the monolingual Umayyad administration, slow, although most of the population seems to have been mostly Arabophone by the tenth century. The ‘conquest solidi’ are the first elements with Arabic writing in the Iberian peninsula (see above), but they do not indicate the widespread use of the language. One of the first non-formal (‘spontaneous’) pieces of evidence for the spread of Arabic is in a set of graffiti from the basilica of Casa Herrera (figure 8.13), which are also late ninth century in date. At this site, the old basilica and its monastery may have been put to a different use (apparently a prison). The prisoners scribbled complaints and desperate cries for freedom on the marble columns (Barceló Torres 2002a). Similar early Arabic graffiti, all of ninth century chronology, have been found on pot shards excavated at El Tolmo de Minateda and on columns at the villa of Milreu, in Estói (Martínez Núñez 2009). These have chronologies similar to the earliest public inscriptions, such as that of the alcazaba of Mérida. These early inscriptions suggest a widespread use of Arabic at various levels of society six generations after the initial conquest. This was partly as a result of the Arabization of the locals, and partly as a

298 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.13 One of the Arabic graffiti inscribed on the columns of the basilica of Casa Herrera (ninth century)

consequence of the increase in the number of settled Arab-speaking peoples, who did not learn Latin. Against the expanding force of Islam and Arabic, there is archaeological evidence in al-Andalus for the continuation of Latin and Christian practices into the twelfth and thirteenth centuries.90 Despite the steady conversion rate of urban populations during the Umayyad period, there was a surprisingly high degree of continuity of Christian communities. The evolution of these Christian Arab and Romance-speaking communities should perhaps be seen (to a certain degree) as the continuity of Late Antique populations and culture in the territories of al-Andalus.91 Archaeologically, these communities can be identif ied in various ways. Excavations at the site of Los Bodegones in Mérida show that the 90 As the preliminary results of the site at Tózar seem to show (Mattei et al. 2014). 91 Traditionally referred to as ‘mozárabes’. This term is controversial and it has been suggested that mustarab (an eleventh-century word) should apply only to the Arabized Christian communities of later periods, and not to the Romance speaking Christians of the eighth and ninth centuries (cf. Aillet 2010).

The formation of a new Medieval materialit y

299

Christian community used a post-Roman cemetery into the Umayyad period, abandoned only with the abandonment of the city (Delgado Molina 2006). Similar cases could be put forward in the rural settlements around Eio, the Christian necropolis of Pamplona, the episcopal burial ground of Barcelona, and many others (including the new Umayyad foundations such as Ilbira). These Christian urban communities do not, however, seem to have embarked themselves upon the construction of new Christian monuments, excepting, perhaps, in the case of the Saint Eulalia basilica in Mérida (renovated in the ninth century). Hardly any new churches were built in towns in this period (Caballero and Sastre 2014: 284). There are two explanations for this: firstly, bishops and other lay members of the Christian elites did not have any political incentives to set up new Christian monuments. And secondly, there was not a demand for new centres for worship, since Christian populations were not growing. Repairs and maintenance took place (as in Saint Eulalia), but there not much new building. It was a very different case in the countryside, where monasteries (built on private land) flourished, as at Melque or El Trampal (see below), and also in the suburbs, and in the Christian territories of the north. In the Umayyad period, Christian suburbs once again became relevant elements of urban development.

Changing townscapes The slow evolution of towns continued into the eighth century following the patterns set in the Late Antique period (see Chapter 5), although in some instances such processes were accelerated by the new political environment. The development of a new civic culture in the Islamic territories and the introduction of new laws and customs radically modified the way towns had been evolving from the late Roman period. This is evident when compared to the towns in the Christian north, which continued to evolve throughout the eighth and ninth centuries (until their later processes of state formation in the ninth and tenth centuries) along the same lines that they had in the late Visigothic period. Thus, the end of the Late Antique town overlaps with the creation of a new Islamic, Medieval, townscape, evident only from the ninth century onwards. Regarding the old urban nuclei of the Christian north there is, for the current state of our knowledge, little to say, especially when compared with the information about Roman towns in al-Andalus. There do not seem to have been major alterations to the urban fabric, beyond the slow

300 

The Iberian Peninsul a bet ween 300 and 850

continuation of seventh-century patterns (Chapter 5). In towns such as Pamplona (Faro et al. 2008: 233), Braga (López Quiroga 2004a: 77), León (Gutiérrez et al.2009) and Barcelona there was increasingly a reorganization of the urban foci, so that the street grid and the main roads led to and focused around the cathedral complexes. The consolidation of bishops as local leaders was key in this period, especially after the collapse of the Visigothic administration. Episcopal complexes had become the main centres of urban power so it is natural that town settlement gravitated towards them. Similarly, the thorough Christianization of townscapes and urban life further accentuated this trend. It could also be the case that the re-arrangement of street hierarchies and the increasing centrality of urban churches could be directly linked to urban religious processions (as has been suggested for Gaul; Esmonde Cleary 2013: 180). Without new excavations it is not possible to go into any further depth when describing the towns in the Christian north. It is only three centuries later, when the new Romanesque cathedrals are built (linked to the development of local processes of state formation), that new townscapes have been identified – essentially because these are still standing and visible, not necessarily because they have been unearthed during excavations. On the contrary, the numerous archaeological interventions in southern Iberia shows that the development of an Islamic state in Al-Andalus was linked to a moment of urban flourishing and transformation. Perhaps the most obvious transformation is the final abandonment of Roman street grids, or rather of the Roman system of street hierarchies, as new roads became more prominent. The occupation of streets and alteration of street widths had already begun in the late Roman period, but during the Islamic period the rearrangement of the street network was in many cases complete, so that in towns such as Toledo, Córdoba and Seville, it is only with difficulty and imagination that the Roman grid is visible. Similarly, with the apparent stability derived from the establishment of the emirate and the revitalization of trade, especially towards the Islamic east, towns flourished and grew beyond the limits of the Roman wall enclosures – which reverses the previous trend of urban reduction of the late Roman and Visigothic periods. This process of narrowing streets has traditionally been linked to the emergence of suqs (commercial and workshop areas: Crawford 1990; Foss 1979: 56, 82; Kennedy 1985: 13; Whitow 1990: 19). They are also the response to the lack of an open area, centralized market (Navarro and Jiménez 2007), but for al-Andalus, these do not seem to appear until the tenth century (as in Ilbira; Malpica Cuello 2011).

The formation of a new Medieval materialit y

301

The transformation of the urban grid is especially clear in Córdoba, and it is there that the two main reasons can be outlined: new patterns of housing and new urban foci. First of all, the main roads in the Islamic period evolve as the population was crammed inside the old wall enclosure, disregarding the Roman distribution. The pace at which these changes took place is difficult to assess, but it is probable that by the ninth century these new axes were already in use. They connect the main gates with each other and, ultimately, with the alcázar-mosque complex (Vaquerizo and Murillo 2010: 526-7; León and Murillo 2014), which form the new foci of the city. Out of these main streets, secondary roads sprang, and out of these, dead-end roads that became semi-private alleys inside street blocks. This last stage, the creation of dead-end alleys that could be shut, is the main transformation. This implies a complete change of urban mind-set, not only away from wide Roman roads, but also from the concept of public and private in the urban sphere (Navarro and Jiménez 2007). Roads formed a network, and they were clearly organized in a hierarchy, but it was very irregular, and not an orthogonal grid. In other towns, perhaps because the Umayyad occupation was shorter, the Roman street grid is still visible, as in Mérida or Zaragoza, although the streets are much narrower and winding. In fact, in some ninth-century chronicles Zaragoza (Souto and Bramón 1987: 11-3) is considered to be a foundation of the Goths (who lived in the time of Moses) indicating that the town had hardly changed during the eighth century. There was perhaps the strong presence of both muwallads and Christians, both of which might have favoured Late Antique approaches to housing: there was encroachment, but not necessarily the development of enclosed alleys. Old Roman monuments were usually still standing, even if in ruinous state, or partially quarried away, and in many cases entering a process of ‘conceptual’ abandonment, so that their original purpose and meaning was lost. Islamic ninth and tenth century authors still mention the presence of Roman ruins in their contemporary cities (Gozalbes Cravioto 2001: 650-5; Canto y De Gregorio 2001; Sarr 2010), but they do not always know what they were, and have very inventive interpretations. Amphitheatres, circuses and theatres, together with temples, had already lost their purpose in the sixth century; public baths had also largely come to an end in the Visigothic period; only the water supply and evacuation systems, together with bridges and roads, seem to have survived into the period of the Islamic invasion. These were used, but by then there was no maintenance or active repair work. There does not seem to have been any active attempt to preserve or maintain them by the central authorities (which can explain their final

302 

The Iberian Peninsul a bet ween 300 and 850

abandonment). This was perhaps a result of the change in the legal status of these monuments, the lack of political interest of the Umayyads in preserving Roman monuments, or else a consequence of their bad state of repair after so many centuries.92 Regarding aqueducts, public water supply f inished in the Umayyad period. As a result of their new legal situation, and of the lack of interest, only three of the old Roman aqueducts were in use by the ninth century, which included the new sixth-century aqueduct of Reccopolis (Sánchez and Martínez 2016). The abandonment of sewers was a consequence of both the end of aqueducts and the change in the nature of the streets. The transition from convex paved streets (which drained water off to the sides into drains) to concave earth streets (with a central open gulley to collect rain water) began in the post-Roman period, but was firmly established in the Umayyad period. In the Early Medieval period this was accelerated because of the increase in the use of cesspits and the abandonment of drains for the disposal of domestic waste (Rėklaitytė 2012; Vidal Castro 2000). There were notable exceptions, but always linked to very special circumstances. For instance, new urban foundations such as Murcia and Lérida had dug-out sewers, and in Zaragoza and Mérida, where large Roman sewers existed, the old drains seem to have still been in use (which is not to say that they were kept in a state of repair). This does not mean that hydraulic engineering was abandoned: after all, the Umayyads were responsible for the construction of a new aqueduct to feed the alcázar in Córdoba and to put the aqueduct of Valencia back in use to supply the alcázar (Martínez Jiménez 2011a), but these were built expressly to feed these Umayyad palaces. Moreover, the arrival of Islamic populations from the Levant and Africa prompted the importation of the qanat system in al-Andalus (Hodge 2000: 36; Rotolo 2014). Houses in the early Umayyad period slowly changed from their postRoman precedents, which as mentioned earlier (Chapters 2 and 5) evolved into multipurpose rooms (and even multi-family dwellings) located around a courtyard. This trend continued well into the Umayyad period. The ‘Islamic house’ was conceived around the centrality of the courtyard or wast al-dar (the ‘centre of the house’) (cf. Orihuela Uzal 2007). One innovative element which can be seen as characteristic of Islamic courtyard houses is that, 92 The change in their legal status was from public, municipal property to communal and belonging to the Umma (and under direct control of the Umayyad rulers), or sometimes granted as waqf (charitable private endowments): cf. Crone 2005: 3; Santillana 1926: 2, 134-6, 322; van Leeuwen 1999: 11-2, 68.

The formation of a new Medieval materialit y

303

despite its centrality, it was not directly accessible or visible from the street, which develops into a particular conceptualization of the public and private spaces in domestic architectures. Early Islamic urban houses have been excavated in Valencia, Mérida, Ilbira, Málaga and, of course, Córdoba, although these are very badly preserved (Orihuela Uzal 2007: 302-5; Malpica Cuello 2010). All excavations seem to agree on the presence of a central courtyard around which the dwellings with multi-purpose rooms are present, and built very similarly to post-Roman houses (beaten earth floors, rubble foundations with pisé walls, etc.). One new element that is noticeable and identified in Córdoba and Mérida is the use of new roofing tiles (Casal García 2008: 123). Otherwise, it is only after the ninth century (Bazzana 1991: 187-9) that a really identifiable Islamic house appears, where rooms are, again purpose-specific, and their number increases. It should be noted that it is in the Islamic period that there are found signs which indicate a growth in urban settlements. The early example of Šaqunda in Córdoba (figure 8.14) is perhaps the best excavated example, a suburb which grew on the south bank of the river, even if it was later destroyed after the troops living there rebelled in 818 (Manzano Moreno 2006: 125). This suburb grew organically, and was not part of any pre-conceived urban plan, which is why the streets were irregular and sinuous, and lacked any sort of drainage system (which caused considerable damage when the river flooded the suburb) and with various communal wells (Murillo et al. 2004; Casal García 2008). Other than suburbs, large suburban aristocratic estates (‘almunias’) also emerged in this period, forming a densely settled area around the city, directly linked to irrigation and agricultural production (León and Murillo 2014: 13-6). The most famous of these almunias was the Al-Rusafa estate, built by Abd al-Rahman I himself. In some cases, these almunias ended up developing into proper suburbs. Similar developments can be inferred from the growth of Seville and Tarazona and their expanded wall enclosures (Corral Lafuente 1991: 260). None of these suburban developments seem to have had any street grid either, simply organically evolving from the new, irregular street network beyond the original road, which becomes the main axis organising the new suburb.

Transitions in the rural world From the perspective of the rural world, it is more difficult to trace a linear evolution of the material culture as can be done in urban contexts (both

304 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.14 Plan of the excavations at the suburb of Šaqunda in Córdoba, located on a bend of the river, south of the city and across the main bridge

From Casal García 2004 figure 2

‘north’ and ‘south’). As far as the evidence can nowadays be understood, there is a rupture between the Late Antique/post-Roman rural settlement, social and economic patterns, and the properly Medieval world of the village and the castle (or the qura and the husun) which is evident by the end of the ninth century. The urban overlap of transformations is, in rural contexts, matched by the emergence of a set of hyper-regionalized set of responses to the Islamic invasion and the collapse of the Visigothic models which are difficult to assess, and which have little to do with the previous or subsequent models. But in most cases it works around a controlling hilltop site/lowland agricultural settlement dichotomy. Interpretative models for the rural world in southern and northern Iberia have different historiographical roots which have determined their

The formation of a new Medieval materialit y

305

subsequent development. In al-Andalus, the degree of continuity for Visigothic villages varied from region to region, but in all cases the old model seems to have been abandoned by the mid-eighth century. The new model which emerges in the ninth century (that of high-ground husun and lowland village-like qura) is well-defined and largely homogeneous across the various regions.93 In the Christian north the rural settlement was also based around lowland villages and hillfort settlements (which would eventually develop into castles), although they functioned differently to those of the Visigothic period. These processes would however only become fully developed after our period, in the tenth century.94 Hence we will focus on the major trends which relate to the emergence of the new polities. Again, it should be underlined that the scarce archaeology available for the eighth and ninth centuries shows a highly complex and hyper-regionalized set of models, all of which show the response of the local populations or of the new settlers to the socio-political and economic transformations arising from the Islamic invasion. Lastly, this complexity cannot be reduced to linear or determinist processes and has to be presented in general and cohesive regional responses. Qura and husun If we look first at al-Andalus, we find that the response to the Islamic invasion and the arrival of new settlers consists of four main patterns, paying attention to the location of the qura and the husun (and with a fifth response related to defensive towers). For the qura, the location of new settlements seems to have been directly linked to areas which could be irrigated. New, highly organized and mechanized (with the introduction of the noria water wheel) rural irrigation systems, which surpassed in extent the Roman 93 The study of these settlements and the interaction between husun and qura is still undergoing interpretative revision, which is necessary especially considering how underdeveloped Islamic rural archaeology is when compared to that of the Visigothic period (Eiroa Rodríguez 2012; Trillo San José 2006). The current debate still revolves around the nature of these hilltop sites (permanent settlements, fortified refuges, built by newcomers or planned by the state, etc.) and their degree of interaction, and the nature of this relationship with the lowland settlements. 94 We acknowledge the fact that this process is extremely controversial, and that there are many ongoing debates regarding Feudalism and incastellamento, especially after Pierre Toubert’s thesis (Toubert 1973), the birth of the Medieval village, the ‘revolution’ of the year 1000 (Quirós Castillo 2007), and the transition from Islamic husun to Christian castles (Glick 1995), but most of these processes take place after our chronological range, and they would lead away from our main archaeological focus in this chapter, which is concerned with the end of Late Antique rural structures.

306 

The Iberian Peninsul a bet ween 300 and 850

network of rural aqueducts (Glick 2005: 60-9, 258-61; Glick and Kirchner 2000), underlines the importance of such technology in the economy of the early settlers. This irrigation technique was introduced alongside a series of new crops. Although the general spread of such new crops (rice, oranges, sugar, cotton, etc.) took place in later centuries, some plants like the date palm were introduced in the eighth century (Glick 2005: 70-1). With these new crops and the increased productivity linked to irrigation, throughout the Umayyad period there was slow shift from the maximized and diversified agricultural economy of the post-Roman period, to a market focused and cash-crop model, similar to that of Roman times. In the territory of Ilbira for instance (Carvajal et al. 2015), there was a relocation of rural sites into new areas soon after the Islamic conquest, especially to places near the lowlands which could be irrigated (figure 8.15). In this case, the original settlement formation in the early Islamic period was later modified after the fitna, the ninth-century civil war. A parallel emergence of rural settlements in new areas where irrigation is possible are the lowlands of Alicante, as can be seen in the site of Cabezo Pardo, which could be a settlement of Umayyad date and using Islamic irrigation models, but used by Christians (mozarabs; López and Ximénez 2014). In these examples the settlement distribution and the land exploitation are completely different to pre-Islamic modes of rural occupation. In these two areas (the lowlands of Granada and Alicante), some hilltop sites remained in use, existing as ‘refuge husun’. Very early new irrigation systems (on relatively scarce data) have also been proposed for areas in Murcia (Puy Maeso 2014). In some known cases, like the hisn of El Molón (Valencia), the hilltop settlement was not simply an occasional refuge or a military establishment, but also had permanent dwelling places. Excavations at the site (Lorrio and Sánchez 2004) show an early Islamic hisn, dated between the mid-eighth and mid-tenth centuries, with a fortified enclosure on a pre-Roman hilltop site, which included an area to keep cattle safe, a densely built dwelling area and a mosque, plus a settled area outside the fortifications. This could be the second type of early Islamic settlement pattern. This situation contrasts with the countryside around Eio (Gutiérrez Lloret 2012), and the hinterland of Toledo-Reccopolis (the current province of Madrid; Vigil-Escalera 2007), and the territory of Mérida (Cordero and Franco 2012; Franco Moreno 2004). In these cases, the previous network of villages and other forms of rural settlement show a high degree of archaeological continuity until the late eighth century, when the newly established emirate reorganized the areas of Toledo and Murcia with new urban foundations

The formation of a new Medieval materialit y

307

Figure 8.15 Schematic plan of an early Umayyad rural settlement from the hillslopes in the Vega of Granada (by Miguel Jiménez Puertas, used with permission). A: Village with a dispersed pattern of settlement; N: Necropolis; C: Path with different possible destinations; F: Water source; P: Agricultural plots; CN: Contour lines

and rural fortifications (in Mérida, this change came later, in the ninth century). This exceptional situation of continuity could be linked to the still lingering past importance in these territories of the Visigothic system. There is a similar chronological continuity in the monasteries, which continued to exist during the early Umayyad period, probably as a result of the sulh treaties. Saint Mary’s in Melque and Saint Lucia’s in El Trampal were, furthermore, lavishly monumentalized in this period (Caballero and Moreno 2013; Caballero Zoreda 2000).95 In both cases, the still-standing monastic churches of Melque and El Trampal, with their stone vaults, their arched doorways, and their cruciform distribution, indicate a renovation with new building techniques. Furthermore, in the case of Melque, there 95 Albeit the chronologic debate noted in Chapter 6.

308 

The Iberian Peninsul a bet ween 300 and 850

is an extensive set of dams and irrigation channels. The continued existence of such a Christian community underlines the nature of the Islamic settlement in the first decades of al-Andalus. Monastic lands, belonging either to the Church or to landed aristocrats, would have been at least partially preserved by their original owners. Another possible example of Visigothic continuity are specialized productive sites, such as the fishing and salt-panning establishments in the coast around Cartagena (Martínez Rodríguez 2014) or the pre-hisn iron mining settlement at Jolopos (Bertrand and Sánchez 2000). The rural territory of Jaén (where there were no relevant urban nuclei in the Visigothic period) shows perhaps an intermediate experience, as there is evidence both for the continuity of pre-Islamic rural sites into the early Umayyad emirate and for the establishment of new settlements (Castillo Armenteros 1998). These included both hilltop husun and lowland nuclei (articulated along the river system). In southern Portugal, in the territory of Silves, it has also been suggested that there was a similar development of qura with the reoccupation of hillfort sites in this early period (Catarino 2000). Finally, regarding tower settlement, it has been proposed (although not yet on enough evidence) that already in the eighth century a series of rural establishments for the new settlers were created in the northern border areas of al-Andalus, particularly in Catalonia and the Duero valley around military signal towers, which formed a frontier network (Martí Castelló 2008; but read together with Martín Viso 2009b). Villages and castles Regarding the northern territories, the initial stages of conquest and the formation of al-Andalus were a moment of rupture in the settlement patterns in the territories north of the Central System. Traditionally, this rupture was considered to have been large enough to provoke a total depopulation of the Northern Meseta (Sánchez Albornoz 1966). However, studies based on place names and the social structuring of landscapes have demonstrated that total depopulation could not have been possible (for an overview, see Escalona Monge 2002 or Martín Viso 2009a). The archaeological recognition of what has been called the ‘long eighth century’ (Hansen and Wickham 2000) in the northern part of the Iberian Peninsula is still problematic because of the lack of excavations and ceramic studies which could provide an empirical chronological basis, but archaeology is slowly casting light on this process. Pottery assemblages show a simplification of types but also a diversification of production, distribution and consumption during the

The formation of a new Medieval materialit y

309

eighth century due to the structural regionalisation patterns for northern Iberia which were previously described (Vigil-Escalera and Quirós 2016). Large areas of the Northern Meseta show, during the first half of the eighth century, an important reduction in the number of rural settlements and the concentration of population in a set of specific nuclei (Vigil-Escalera 2007; Tejerizo et al. 2015; Ariño Gil 2013). Beyond depopulation theories, settlement seems to concentrate in three different types of site. The first type is the so-called ‘second generation of hillforts’ (Quirós Castillo 2012a) like Tedeja (Palomino et al. 2012), Peña Amaya (Quintana López 2017; Quirós Castillo 2015a) or Pancorbo (cf. Quirós Castillo 2015b). These hillforts (or castles) show the emergence of regional elites during the reorganization of political and social power in the eighth and ninth centuries. A second possibility suggested by the archaeological record is the concentration of people around a few, chosen, pre-existing rural sites, like Canto Blanco (Strato 2010), or perhaps those which continued into the Late Middle Ages, as has been suggested for the central part of France (Nissen-Jaubert 2012). The last type of site which can be seen as functional during the eighth century are those rural settlements which could be linked to small churches, like Villajimena (García et al. 1963) or the interesting site of Caleruega, a rural settlement on the top of a low hill in the Duero basin which developed around a church that was not eventually finished.96 On the Cantabric coast and the northern mountains, the situation was slightly different. Here, it is possible to affirm a profound change in the aftermath of the eighth century, when a dense network of villages and farmsteads is documented (Quirós Castillo 2011a). Moreover, in some of these villages, such as Zaballa, Alegría-Dulantzi (figure 8.16), Aistra, Bagoeta or Zornoztegi have shown (Azkarate et al. 2011; Loza and Niso 2011; Quirós Castillo 2012b) the presence of a complex hierarchy between them and also the presence of remarkable social inequalities within these communities, backed by the study of animal bone assemblages (Grau Sologestoa 2015b; 2013). This study of animal bones suggests that the patterns of consumption were significantly different between settlements, and biometrical analysis also suggests important changes in livestock management during this period. In this respect, the period after the eighth century shows, for the first time since Roman times, the effective and substantial transformation of the agrarian landscape in the Iberian peninsula by the action of rural communities, perhaps as a consequence of the reorganization of elites and their capacity to extract surplus from local societies. The study of terraces in 96 We want to thank ARATIKOS SL for this information, still unpublished.

310 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.16 Site plan of Alegría-Dulantzi

Based on Quirós et al. 2013

sites like Vigaña and Santo Adriano (in Asturias), Zaballa or Torrentejo (in the Basque Country) and also in many parts of Galicia (Kirchner Granell 2010b; Fernández et al. 2014; Quirós Castillo 2012b) have shown the complexity

The formation of a new Medieval materialit y

311

and the capacity of these Early Medieval rural societies for coordinating their productive spaces. Lastly, the third region we can focus on are the ‘grey areas’ of contact between the two main polities of the period. After the dismantling of the network of farmsteads and villages in the central part of the plateau during the first decades of the Islamic conquest, the settlement pattern seems to have concentrated into a chosen few nuclei acting as local and regional central places in the territory. The origin of these nuclei can be found in some cases in prior settlements and sometimes in sites created ex novo during this period, as seen in the examples of Talamanca or Calatalifa for the territory of Madrid (Vigil-Escalera 2007). The situation in the rural world in the southern part of the mountainous Central System could be extrapolated to the territory between the northern part of the mountains and the river Duero basin. Here, the absence of an organised network of settlements can be interpreted as a lack of an organized elite and a dispersed settlement pattern, which characterized the rural landscape up until the conquest of the Christian kingdoms. For other regions of the peninsula, the information is even scarcer. In northern Portugal, excavations carried out have also shown a dispersed settlement pattern with the presence of some particular sites that may be local references in the landscape, as happens on the Northern Meseta. For example, excavations at São Gens have unearthed a fortified village destroyed by a fire in the tenth century (Tente 2009). In the Upper Ebro valley, the rural settlement pattern is still largely unknown, as there have been no extensive excavations of ninth to tenth century material in this area, and there is little information about domestic architecture or settlement organization. However, there are many archaeological markers which show the complexity of the rural world during these centuries. Maybe the most important is the changes documented in the funerary rituals, with the appearance of the extensive rock-cut necropoleis that characterized the Early Medieval period in this region. These cemeteries were first studied by A. del Castillo in the early 1970s, who called them the ‘olerdolan cemeteries’ after one of the sites and related them to the process of re-population during the ninth and tenth century in the Ebro valley (del Castillo Yurrita 1970; 1972). The sites of Cuyacabras and Villanueva de Soportilla can be put forward as good examples of this kind of cemetery which are characterized, in contrast with the earlier, Visigothic rock-cut necropoleis discussed in Chapter 6, by the presence of large groups of burials in the same context with a very clear spatial organization. In some cases, as in Villanueva de Soportilla, the cemetery seems to be organized around a small building, possibly a

312 

The Iberian Peninsul a bet ween 300 and 850

shrine. However, recent excavations at this site have shown some hints of continuation between the seventh and the tenth centuries, showing the absence of a population vacuum during the eighth century (Aratikos 2010). In fact, this continuation can be seen in other sites in the territory, many of them linked to Christian places of worship which, possibly, attracted part of the rural population. Some sites already mentioned, like Peña del Mazo or the surroundings of Tedeja, have shown this continuity (Palomino and Negredo 2011; Palomino et al. 2012). Even though data for the eighth century remains scarce, it is clear in the archaeological record that the expansion of Christianity and the construction of religious buildings was fundamental in this period. The role of the church as one of the main axes that structured the settlement pattern in the rural world increased throughout the eighth century and after as seen in many examples in the Duero valley (Martín Viso 2014), and also in the territory of Galicia currently being studied (Sánchez Pardo 2015), and the already noted Asturian kingdom. However, the key importance of churches in this process needs to be revised. Many archaeological examples in the northern part of the Iberian peninsula have shown the preexistence of complex and organised communities before the construction of church buildings, for example, Sacramenia (Aratikos 2008) or Peña del Mazo. In these cases, the church was built over prior communal cemeteries, showing fundamental changes in the conception of community landscapes and relations of power within these communities. These examples show that the introduction of the Church and its elites into local communities, relations and landscapes was not a linear process but a complex and dialectical one. Nevertheless, these are the first examples of a long-lasting process of construction of the Medieval parishes which organized and nucleated rural societies from Medieval times until the present. The archaeology of the eighth (but especially of the ninth) century shows two very clearly identifiable processes which are the disruption of Late Antique (both late and post-Roman) rural and urban materialities and the appearance of new socio-economic circumstances linked to a new material culture. While the latter raises very interesting problems of the consolidation of feudal and Islamic patterns, these are processes that are fully visible in the late ninth century. These processes, furthermore, require the answering of questions that cannot be posed in a Roman or Late Antique interpretative framework because the circumstances are radically different. The Umayyad conquest was an historical rupture with the Visigothic past, which can be seen through the archaeology, with the slow but steady

The formation of a new Medieval materialit y

313

introduction of new archaeological markers such as new coinage, new civil and religious monuments, new burial practices, a new language, and new technologies. However, all these innovations were introduced into post-Roman society during a period of cohabitation, hybridation and slow replacement. In fact, the transformation of pottery techniques or the slow changes in urban patterns attest to this period of adaptation. It would only be after the Umayyad process of state formation in the ninth century that the Islamic material culture became the norm, even amongst the increasingly Arabized Christian communities. The Late Antique responses to the transformations of the Roman world (which were Roman in form and depth) were replaced. Similarly, the Christian polities which emerged in the north, faced with such a large and powerful political and cultural entity to their south had to respond, and in doing so they looked to the feudalizing world of the Carolingian Empire and to the alliance with the Church – only later fabricating claims of their continuity with the Visigothic kingdom. The urban architecture and the rural patterns that were developed in these northern enclaves were framed by their Late Antique past, but the policies and patterns of change behind the new constructions did not reflect a dialectical interaction with the Roman past or with late Roman problematics. Both north and south reflected the new Medieval world which was being developed in the wider Mediterranean.

314 

The Iberian Peninsul a bet ween 300 and 850

Figure 8.17 Stratigraphy of the agrarian terrace of El Manso (Asturias)

Based on Fernández et al. 2014

9 Conclusions From the collapse of the Roman Empire to a Brave New World It is not very extreme to assume that a person who lived around the year 300 was radically different from someone who lived in the same place in the year 800 (and not just in the Iberian peninsula). We can hypothetically picture their encounter in an imagined scenario. They would probably not be capable to communicate, not only because the language they would speak in would be different, but also because their mentalities, political economies, and perceptions of their circumstances would be strange to each other. If we could ask them which place was the most important in their world, we would obtain quite different answers. The first one would have quickly named Rome; Rome not just as a physical entity but also as a supra-idea of power and order. However, the answer to this question would not be so easy for our ninth century person; he could say Rome, or Jerusalem, if he was a Christian, or, if Muslim, Medina and Mecca. In political terms, the Early Medieval individual could give us several answers: Córdoba, the capital of the Emirate if he lived in al-Andalus; if he were a Christian then maybe Oviedo or Aachen, Charlemagne’s capital. Or even Constantinople. None of these places would mean anything to our fourth century friend. It is probable that they would even be physically different. Five centuries are a very long time when dealing with human cultures, something which is evident and reflected in the changes of the material culture; even more so if we consider the deep and long-lasting changes which shaped the Iberian peninsula over the chronological scope of this book. It is the complexity of these changes and the connection between those two worlds of our two imaginary friends which we aimed to engage with here from an archaeological perspective. A fourth century aristocrat, but also a freeman, even if already Christian, would have felt a familiarity with the numerous meanings of the multitude of mythic motives that decorated the mosaic floors found in houses and public buildings. Such common cultural understanding would had been impossible for a ninth-century individual, whether noble or poor, urban or rural. The same could well have happened with other cultural markers, such as personal belongings, funerary practices, or daily objects such as the clothing they wore. Trousers and cloaks (paenula) would have replaced the tunic and toga.

316 

The Iberian Peninsul a bet ween 300 and 850

Although the events covered in this book have attracted the attention of scholars since the end of the nineteenth century, it was not until recently that we were able to construct a coherent material narrative from them. Late Antique Iberian archaeology can now be described in its own terms, and that is thanks to the increasing scholarly interest in the period and the large volume of available data obtained through archaeology. Paradoxically, the period of heritage ‘destruction’ that took place over the last twenty years in the Iberian peninsula has enabled the development new perspectives in the archaeological study of Late Antiquity and early Middle Ages. If we were to briefly synthesize the main conclusions of the present work, it would run as follows: the materialities that emerge in the late Roman period are a result of the development of the dialectics between internal transformations and external constrictions of the Empire, and they partly endure, partly disappear and partly slowly evolve into the Middle Ages, when a new socio-political and economic milieu prompts new alternatives to similar problems. During this period of transitions (note the plural), there are many causes which underlie the changes in the material culture: to begin with, there are the internal transformations of Roman politics, but also the increasing processes of economic regionalization, the input of the Germanic successor states, and the re-definition of the elites and their social rankings. All of these elements contributed to modifying and redefining the material culture (from the monumental to the mundane) which had been present through the Roman period. And as we have tried to show throughout the book, this was far more complex than a slow but continuous decline of the Roman world into the ‘Dark Ages’ until a new Medieval world appears out of the blue in the ninth century.97 This complexity is also behind the various conclusions that can be drawn from the material we have put forward. We will try now to synthesize some of the key ideas which have been the results of our narrative. As stated in the introduction, even though the amount of archaeological data has significantly increased in recent years, it is not immune to problems and limitations. In this sense, our approach has tried to explain the existing and known remains from the Late Antique past. Whenever possible, and 97 A position which in any case since the earliest days of Late Antique archaeology could be sustained. The days when archaeology could only claim that the Visigoths prayed and died, because only churches (i.e., San Pedro de la Nave, etc.) and tombs (i.e., eagle-brooches and so on) were known – neither of which, by the way, can be really be linked to the Visigoths – are long gone.

Conclusions

317

taking into account general trends within Iberia and its various regions, we have tried to provide broader interpretations through comparisons and the occasional extrapolation. In some cases, we are in agreement with the most accepted current views on particular issues, while in others we have been taking sides in ongoing historiographical debates. None of these interpretations are final, and there is much archaeology and the re-assessment of excavated material that needs to be done, but we would like to believe that we have put forward a coherent and cohesive view on the transformation of Roman Hispania into the Middle Ages. It is for this reason that we proposed an archaeological periodization that avoided political and historical labels and absolute dates (like 418 or 711). The importance of Rome and its imperial system is, however, unavoidable, because the whole point of our conception of Late Antique material culture is to identify the materialities which emerged as a result of the transformations of the fourth century and the subsequent responses derived from the dismantling of the Roman Empire. The use of Roman past and the radical transformations of the late Roman world are therefore, in our understanding, the key elements that define Late Antiquity. From a ‘classical’ or ‘traditional’ archaeological perspective, we have identified a tri-partite chronology, with a ‘formative’, a ‘classical’ and a ‘late’ or ‘transitional’ phase, which is defined by the presence of a Roman infrastructure (late Roman), the lack thereof (post-Roman), and the collapse of this situation (Early Medieval). One of the main processes which takes place through this period is the development of a Christian materiality, in its widest conception. It is not possible to deny that Christianity had a huge effect on Roman culture, especially once it was adopted as an hegemonic ideology by the elites. Monument wise, the change from ‘paganism’ to Christianity is seen not only in the active construction of martyrial shrines (in the fourth century), suburban basilicas (in the fifth) and intra-mural episcopal complexes (in the sixth), but also in the partial quarrying and dismantling of old unused and closed temples. The spread of Christianity from towns into the countryside is noticeable as occurring perhaps slightly later than in urban contexts, but rural basilicas and monastic sites are clear examples of this. Besides these new ‘constructive’ elements of Christianity, the new religion gradually modified traditional burial practices. Even if inhumation was common in the imperial period, East-West, aligned and clustered burials (enclosed cemeteries and burials ad sanctos, as opposed to earlier Grabenstraße), and the evolution in the presence of grave goods can be linked to new burial practices. These new Christian sites, in towns, suburbs and in the countryside, eventually acted as poles of attraction, around which

318 

The Iberian Peninsul a bet ween 300 and 850

settlements clustered in a way pagan cult sites had not done. Even at a visual level, Christian images and iconography slowly made their way into wall paintings, inscriptions and pottery decorations. The material culture of Late Antiquity includes all of these Christian elements, but they remain however recognizably ‘Roman’ in many respects well into the post-Roman period, perhaps because they developed and continued to evolve within late and post-Roman contexts. It is only in the Early Medieval period, when new external influences (through Umayyad, Carolingian and Romanesque channels) that the early forms of Hispanic Christianity (both ritual and monumental) are fully replaced in the material record. Finally, the arrival and slow-but-steady diffusion of Islam introduced an element of religious alterity which is extremely noticeable in the material record, a result of its deep impact on social relations and its materialization. Christianity became a defining element, a layer of identity that was added to economic and legal categories such as Roman or senatorial. Whereas this incorporation of Christianity as an identity layer took place mostly in the late Roman period, other elements were added later in the post-Roman period as markers of social power and strategies of distinction. The arrival of the Visigoths and the Sueves added extra layers of identity (Germanic vs. Roman and Arian vs. Catholic), which were legal as much as they were religious or linguistic, and slowly faded away throughout the sixth and seventh centuries. The impact on the material culture of the other main historical agents of the Late Antique world (the Barbarians) is perhaps not as evident as that of Christianization, and certainly not linked to straightforward ethnic self-identification through objects: pots do not equal people. The arrival of the Visigoths and to a lesser extent, of the Sueves and their processes of state formation added another element of identity complexity to late Roman Hispania, especially within an elite which was already in the late Roman period, culturally (if not literally) militarized. The redefinition of the funerary habitus (especially through the nature of grave goods) reflects not only new cultural practices, but also the redefinition and consolidation of social hierarchies with new markers of rank – markers which were associated with the new (military) power. Because something that these Barbarian groups certainly did was to set up states by carving their own niche within the Roman provinces, taking advantage of the situation of the collapse of the Roman system in the Iberian peninsula from the mid-fifth century onwards. The consequence of the Barbarian Invasion was the Barbarian Kingdom, the key characteristic feature of any map of the Late Antique Mediterranean. The political replacement of Rome by the Germanic kingdoms is, however,

Conclusions

319

not that straight forward, and between the last identifiable remains of a functional Roman imperial system early in the fifth century and the processes of state formation of the late sixth century there is a large period of a de facto power vacuum. This can be seen in the social redefinitions mentioned in the previous paragraph, and less evidently in the decline (when there was not a total halt) of monumental constructive activity in town and countryside. Isolated examples of late fifth and early sixth century constructions do exist (Zeno and Salla’s repairs in Mérida or bishop Justinian’s works in Valencia), but they are the exceptions.98 The new wave of construction activity in this period reflects the needs of the Late Antique city, as it is focused on Christian monuments, new civil administrative palaces, and fortifications. These are the same set of monuments promoted by the late Roman administration, and the ones that can be found in the new Visigothic foundations. The Late Antique political and social powers did not need fora, or basilicas, or theatres and circuses: it needed churches and city gates. This was the materialization of a radical change in the constitution of an ideological hegemony. Old Roman monuments, redundant and anachronic, may have survived during the late Roman period, but they belonged to a completely different urban system, and they were more useful as quarries, and opened up new spaces in city centres for new, Late Antique constructions. Some few functional examples, such as aqueducts, bridges, or walls still offered a chance to invest in public munificence or in imitating Roman practices: it was perhaps one of the few ways in for the Visigothic and episcopal elites to claim the legitimizing Roman past. This situation of power vacuum and later reorganization is also the situation in which the first networks of village communities emerged across the peninsula, replacing the old villa sites, which had flourished under the Roman imperial system. The villa economy, where large states of monoculture dedicated to cash crops (olives and grapes) and cereal produced goods to export within the imperial tax network, was too niched to survived the collapse of the Roman system unchanged. The rural landscape in Late Antiquity saw therefore the abandonment of this old villa model and the emergence of two alternatives: one was the concentration of population in villages and the other was the development of exceptional large rural palaces. The latter was short-lived in time and it was far more related to the nature of elite display in the late phases of the Roman world than it was to a change in economic practices. The former, which was eventually successful and long-lasting, saw the emergence of smaller rural communities, 98 And they are both directly linked to episcopal activity.

320 

The Iberian Peninsul a bet ween 300 and 850

maximizing their agricultural output and occupying new areas not settled in the Roman period. In parallel with these developments (redefinition of social ranks, the reorganization of rural productive patterns, the transition from a power vacuum to Germanic state) we find the reoccupation of hillforts as a new form of central place. High-ground settlements, which had been unused since pre-Roman times, again became main centres from which to control large territories. This was especially true in those areas where towns had failed to become functional administrative units in the late Roman period, such as the Cantabric region or the Meseta. The occupation of these sites, however, fluctuated; they seem to be occupied only in those moments of Late Antiquity when there was not a strong state to interact with regional elites, such as the late fifth and the late seventh centuries. Both in rural and urban contexts there are changes in the material culture and in architecture, which underlines the simplification of the economy and productive processes. In terms of daily commodities such as pottery, the decline of imported finewares (in overall numbers and range of distribution) triggered the development of local and regional imitations, and ultimately their final substitution by ‘coarse wares’ with only a handful of types (i.e., ‘bowl’ or ‘cooking pot’) – a sharp contrast with the seemingly limitless variety of types within ‘plate’ or ‘bowl’ of the Roman period. Simpler methods seem to have been adopted as well in construction, not simply because of the increasing use of reused stone or brick (in many cases very selective and linked to the cessation of primary production), but also because of the increase in the archaeological visibility of timber and of sunken features. The constant reuse and quarrying of the built landscape underlines the change in constructive priorities as much as it highlights the difficulty in obtaining new materials. Beyond materials, building forms and designs both in the city and in the countryside are also simpler, limited to pitched roofs or flat ceilings, post- or pier-supported elevations and, in domestic contexts, multi-functional rooms. The characteristics listed above have to be seen through a prism of regional variation. Even if they are long-term processes of change, there were some areas in which late Roman continuities were stronger and more persistent than in others. This regionalization is nothing but the reality of different local reactions to the collapse of the Imperial system. And while it is easy to discuss Roman Hispania in general terms, when it comes down to the Late Antique period it is necessary to address the different geographic and political regions and the different paces at which the Late Antique transformations took place in them. The arrival of the barbarians,

Conclusions

321

the reorganization of the provincial administration, and the demise of long-distance trade networks all prompted the reversion of homogenizing and inclusive Roman connectivity; geographically-defined regions had to fall back to mostly relying on their own resources. Some areas, especially on the coast or along the major rivers, where connections and trade were still preserved to a degree, suffered less from this (the spread of Western Mediterranean amphorae and tablewares attest to this). Areas of the interior, or up in the mountainous areas of the north, were far more affected, and the pottery assemblages of these regions are much more local. In this respect, the Romanization of these areas (or rather, the degree of integration within the Imperial system) seems more superficial than other areas with longer historical Mediterranean contacts. All of the characteristics mentioned above (the spread of Christianity, the transformation of the Roman townscape, the development of village networks, regionalization of the economy and production, etc.) continued largely uninterrupted into the eighth century. As we have noted repeatedly, the Umayyad invasion did not imply a clean slate in terms of material culture. There was a period of transition during which Late Antique materialities coexisted, and at times hybridized, with the new ones, in a process which lasted into the ninth century. The new material culture introduced by the Islamic conquest (and especially after the Umayyad process of state formation) was far reaching and all encompassing. It included pottery typologies and techniques, burial practices and religious buildings linked to the new victorious religion (Islam), a new coinage system, a new language for the public administration (Arabic), as well as new groups of people with separate identities who, furthermore, transformed the settlement patterns to better serve their administrative model and to take advantage of new farming techniques. Even if many of the changes introduced in the Iberian peninsula had Roman precedents, these had been acquired by the Umayyads from the Eastern Empire, and did not reflect Western, post-Roman practices. In fact, the use and reuse of monuments and spaces varied greatly because the Roman period no longer served as a legitimising past, as was the case in Late Antiquity. Similar changes, at a slightly later date, appeared in the Christian north, were the new states mirrored and competed with Umayyad and Carolingian practices and only at a later date did they look to the Visigothic past for self-justification. In fact, the archaeology of the ninth, tenth and eleventh centuries, when the new forms of Medieval power and material culture are firmly established, is still underrepresented. The archaeology of the high Middle Ages is the next large period to tackle. Elements such as the emergence of the castle,

322 

The Iberian Peninsul a bet ween 300 and 850

the study of the Romanesque from an archaeological perspective, and the emergence of new forms of pottery and burial practice need to be addressed by archaeologists – but this is a whole new period, beyond the reach of this book. As a conclusion, in order to understand the transformations of the Early Middle ages, it is simply necessary to understand that the eighth, and in particular the ninth century, was not a transition. It was not a period for adaptation and change following trends from the past (as the fifth, sixth or seventh centuries had been). It was the beginning of a set of changes that would define the brave new world that was being formed; a world where Rome and Antiquity were not the primary material point of reference and comparison.



Appendix 1 Site reference table

NB: The chronological indicators of this table (R = Roman, LR = Late Roman, PR = Post-Roman, EM = Early Medieval) and the types of site (villa, village, town, hilltop, etc.) are indicators for generic reference. They are not sharply defined as both chronologies and site types overlap. Site Aguilafuente Aistra Alegría-Dulantzi Algeciras Almenara Almuñécar Amaya Andújar Arellano Astorga Ausona Ávila Baelo Claudia Baetulo Bagoeta Baños de la Reina Barcelona Benalúa Bernardos Berrocales Bigastrum Bilbilis Borba-Estremoz Braga Cabezo Pardo Cabrera monastery

Alternative Name

Traducta Sexi Peña Amaya Amaia

Asturica Augusta Osona Abela Bolonia Badalona

Barcino Antigons Cerro del Castillo Begastri Calatayud

Bracara Augusta

Map

Province (SP) Distrito (PT)

Type

Chronology

2 2 2 4 1 5 2

Segovia Álava Álava Cádiz Valladolid Granada Burgos

villa village village town villa town hilltop

R, LR, PR PR, EM LR, PR, EM R, LR, PR R, LR R, LR, PR, EM EM

5 2, 3 1

Jaén Navarre León

factory villa town

R R, LR R, LR, PR, EM

3 1, 5 4 3 2 5 2 5

Barcelona Ávila Cádiz Barcelona Álava Alicante Barcelona Alicante

R, PR, EM R, LR, PR, EM R, LR, PR R, LR EM LR R, LR, PR, EM LR, PR

2

Segovia

town town town town village villa city harbour settlement hilltop

5 5 3 4

Madrid Murcia Zaragoza Évora

PR PR R, EM R

1

Braga

village settlement town marble quarry town

5 3

Alicante Baleares

settlement EM church site PR

LR, PR

R, LR, PR, EM

324 

The Iberian Peninsul a bet ween 300 and 850

Site

Alternative Name

Cádiz Calagurris Calatalifa Caleruega Can Gambús Cangas Canto Blanco Cañal Carranque Carratejera Cartagena Carteia Casa Herrera Castellar del Vallés Castillo de los Monjes Castiltierra Castro Ventosa

Gades Calahorra

Centcelles Ceuta Clunia Complutum Conimbriga Contrebia Belaisca Córdoba Cullera Cuyacabras Daragoleja Dehesa de la Oliva Denia Dueñas Dume Duratón Écija Edeta Eio El Bovalar El Carpio de Tajo El Castillón

Map

Province (SP) Distrito (PT)

Type

Chronology

Cádiz La Rioja Madrid Burgos Barcelona Asturias León Salamanca Toledo Segovia Murcia Cádiz Badajoz Barcelona La Rioja

town town settlement church site village settlement village village villa settlement town town church site village hilltop

R, LR R EM EM PR EM PR, EM PR LR, PR LR, PR R, LR, PR, EM R, LR, PR PR, EM PR PR

2 1

Segovia León

cemetery hilltop

PR LR, PR

3 4 2

Tarragona Ceuta Burgos

villa town town

LR R, LR, PR, EM R

5

Madrid

town

R, LR, PR

4

Coimbra

town

R, LR, PR

3 4, 5 3, 5 2 5 2, 5 5 1, 2 1 2 4, 5 3, 5 5

Zaragoza Córdoba Valencia Burgos Granada Madrid Alicante Palencia Braga Segovia Seville Valencia Albacete

town city church site necropolis villa hilltop settlement church site church site cemetery town town town

R R, LR, PR, EM PR PR LR, PR PR R, LR EM PR, EM R, LR R, LR, EM R PR, EM

3 4, 5 1

Lérida Toledo Zamora

village cemetery hilltop

PR, EM PR PR

4 3 5 2 3 1 1 1 5 2 Carthago Nova 5 4 4 Plaça Major 3 2

Castrum Bergidum Septem Coruña del Conde Alcalá de Henares Condeixa-avelha Corduba Punta de l’Illa

Patones Dianium Dumio Confluentia Astigi Liria El Tolmo de Minateda

325

Appendix 1

Site El Cristo de San Esteban El Germo El Molón El Pelícano El Saucedo El Soto-Encadenado El Tinto El Trampal El Val Els Ametllers Els Munts Emporiae Ercavica Es Cap des Port Espirdo-Veladiez Falperra Fortunatus, villa of Gauzón Gerona Gerticos Gijón Gózquez Granada Guarrazar Huelva Idanha-a-velha Ilbira Ilici Iluro Italica Iuliobriga Jaca Jolopos La Alberca La Cárcava de la Peladera La Cocosa La Garma La Mata del Palomar La Olmeda

Alternative Name

Tinto Juan de la Cruz Santa Lucía del Trampal

Ampurias

Gerunda Gegia

Egitania Madinat Ilbira L’Alcudia Mataró

Iaca

Map

Province (SP) Distrito (PT)

Type

Chronology

1

Zamora

hilltop

PR

4, 5 3, 5 5 4, 5 5 5

Córdoba Valencia Madrid Toledo Madrid Madrid

church site hilltop village villa cemetery villa

PR EM LR, PR LR PR, EM R, LR

4

Cáceres

church site EM

5 3 3 3 5 3 2 1 3 1 3 1, 2 1 5 5 5 4 4 5 5 3 4 2 3 5 5 2

Madrid Gerona Tarragona Gerona Cuenca Balearics Segovia Braga Huesca Asturias Gerona Valladolid (?) Asturias Madrid Granada Toledo Huelva Castelo Branco Granada Alicante Barcelona Seville Cantabria Huesca Granada Murcia Segovia

villa villa villa town town church site cemetery settlement villa hilltop town villa town village town church site town town town town settlement city town town settlement villa village

R, LR, PR LR LR R, LR R, LR PR PR PR R, LR PR, EM R, LR, PR, EM PR R, LR, PR PR R, LR, EM PR R, LR R, LR, PR R, EM R, LR, PR R, LR R, LR, PR R, LR R, LR, EM PR, RM LR PR

4 2 2 1, 2

Badajoz Cantabria Segovia Palencia

villa cave village villa

LR PR PR LR

326 

The Iberian Peninsul a bet ween 300 and 850

Site

Alternative Name

Map

Province (SP) Distrito (PT)

Type

Chronology

La Virgen del Tormejón Ladera de los Prados Lancha de Trigo Las Médulas Las Penas Las Quintanas León Lérida L’Esquerda Liédena Lisbon Los Husos Los Quintanares Lucentum

Cerro del Tormejón

2

Segovia

hilltop

PR

1 1 1 2 2 2 3 3 2, 3 4 2 2 5

Valladolid Ávila León Cantabria Valladolid León Lérida Barcelona Navarre Lisbon Álava Soria Alicante

village settlement mine cave cemetery town town town villa town cave villa town

PR PR R PR PR R, LR, PR, EM R, LR EM LR R, LR, PR, EM PR LR R, LR, EM

1 2 5 1 5 4, 5 4

Lugo Segovia Málaga León Granada Toledo Badajoz

town cemetery town villa cemetery church site city

R, LR, PR, EM PR R, LR, PR, EM LR PR EM R, LR, PR, EM

4 4 4, 5 2 4 5 1 1 5 5 2 2 2 1 5 3 2 2 2

Beja Faro Badajoz Palencia Seville Murcia Ávila León Granada Cuenca Soria Guipúzcoa Navarre (?) Orense Alicante Huesca Asturias Navarra Burgos

town villa town hilltop town city hilltop villa cemetery villa town town settlement settlement settlement town town town hilltop

R, LR, PR, EM LR R, LR PR R EM PR LR, PR PR LR R R PR PR, EM LR, PR, EM R, EM EM R, LR, PR, EM EM

Lugo Madrona Málaga Marialba Martilla Melque Mérida Mértola Milreu Mirobriga Monte Cildá Munigua Murcia Navasangil Navatejera Nívar Noheda Numantia Oiasso Ologicus Orense Orihuela Osca Oviedo Pamplona Pancorbo

Diego Álvaro

Legio Ilerda

Olisipo

Tossal de Manises Lucus Augusti Malaca

Balatalmelc Emerita Augusta Myrtilis

Garray Irún

Auraiola Huesca Ovetum Pompaelo castillo Santa Marta

327

Appendix 1

Site Piñel de Abajo Plà de Nadal Pollentia Portus Ilicitanus Pravia Puig Rom Quinta das Longas Quintanilla de las Viñas Reccopolis Río Tinto Ruscino Sagunto Salamanca San Juan de Baños San Miguel del Arroyo San Pedro de Alcántara San Pedro de la Nave San Pedro de Mérida San Pedro de Tartalés Sant Julià de Ramis Sant Menna Santa Comba de Bande Santa María de Artés Santa María de Mijangos São Cucufate s’argamassa Segobriga Segovia Seville Simancas Son Fadrinet Son Peretó Talamanca Tangiers Tarazona Tarragona

Alternative Name

Alcudia Santa Pola

Cerro de la Oliva CastelRousillon Saguntum Salmantica

Saelices Hispalis

Tingis Turiaso Tarraco

Map

Province (SP) Distrito (PT)

Type

Chronology

2 3, 5 3 5 3 4 2

Valladolid Valencia Balearics Alicante Asturias Gerona Portalegre Burgos

cemetery settlement town town settlement hilltop villa church site

PR PR R, PR R, PR EM PR, EM LR EM

5

Guadalajara

city

PR, EM

4 3

mine hilltop

R R, LR, PR, EM

3, 5 1 2 2

Huelva Roussillon (France) Valencia Salamanca Palencia Valladolid

town town church site cemetery

R R, EM EM LR

4, 5

Málaga

church site PR

1 4 2

Zamora Badajoz Burgos

church site EM church site PR cave PR

3 3 1

Gerona Gerona Orense

hilltop PR church site PR church site EM

3 2

Barcelona Cantabria

church site PR church site EM

4 3 5 2, 5 4 1, 2 3 3 5 4 2 3

Beja Baleares Cuenca Segovia Seville Valladolid Balearics Balearics Madrid

villa factory town town city cemetery church site church site settlement town town city

Zaragoza Tarragona

LR R, LR R, LR, PR R, LR, PR, EM R, LR, PR, EM LR, PR PR PR EM R, LR, PR, EM R, LR, PR R, LR, PR, EM

328  Site

The Iberian Peninsul a bet ween 300 and 850

Map

Province (SP) Distrito (PT)

Type

Chronology

2 2 3 5 4 3 4 5 4

Burgos Soria Barcelona Toledo Portalegre Barcelona Badajoz Granada Beja

PR, EM R, LR R, LR, PR R, LR, PR, EM LR LR LR PR R

3 4 2 5 5 3, 5 3, 5

La Rioja Grândola Soria Jaén Madrid Valencia Valencia

Valeria Veleia Veranes Victoriacum Vigaña Vigo

5 2 1 2 1 1

Cuenca Álava Asturias Álava (?) Asturias Pontevedra

Vilauba Villajimena Villanueva de Soportilla Zaballa Zaragoza Zornoztegi

3 2 2

Gerona Valladolid Burgos

hilltop town town town villa villa villa cemetery marble quarry factory factory town cave villa town fortified site town town villa settlement village harbour settlement villa church site necropolis

2 Caesaraugusta 3 2

Álava Zaragoza Álava

village city village

Tedeja Termes Terrasa Toledo Torre de Palma Torre Llauder Torreáguila Tózar Trigaches Tritium Magallum Tróia Uxama Valdecanales Valdetorres Valencia València la Vella

Alternative Name Tiermes Egara Toletum

Tricio Osma

Valentia

R, LR R, LR R, LR PR R, LR R, LR, PR, EM PR R, LR R, LR R, LR PR PR, EM R, LR, PR LR EM PR EM R, LR, PR, EM LR, PR, EM



Appendix 2 Maps

330  Map 1

The Iberian Peninsul a bet ween 300 and 850

Appendix 2

Map 2

331

332  Map 3

The Iberian Peninsul a bet ween 300 and 850

Appendix 2

Map 4

333

Map 5



Appendix 3 Lists of rulers

NB: these lists are not complete. Roman emperors – Diocletian (284-305) with Maximian (286-305) – Constantine I (306-337), with Licinius (308-324), Maximinus Daia (308313), Maxentius (306-312) and Domitius Alexander (308-309) – Constantius II (337-361) with Constans (337-350) and Magnentius (350-353) – Julian (361-363) – Jovian (363-364) – Valentinian I (364-375) with Valens (364-378) – Gratian (375-383) with Valentinian II (375-392) – Theodosius I (379-395) Emperors in the West – Honorius (395-423) – Valentinian III (423-455) – Petronius Maximus (455) – Avitus (456-457) – Majorian (457-461) – Libius Severus (461-465) – Anthemius (467-472) – Glycerius (473-474) – Julius Nepos (474-475/480) – Romulus Augustus (475-476) Emperors in the East – Justinian I (527-565) – Justinus II, (565-578) – Maurice, (582-602) – Phocas (602-610) – Heraclius (610-641)

336 

Visigothic kings – Alaric I (395-410) – Athaulf (410-415) – Walia (415-418) – Theoderic I (418-451) – Theoderic II (453-466) – Euric (466-484) – Alaric II (484-507) – Agila or Achila I (549-551) – Athanagild (551-567) – Liuvigild (572-586) – Reccared (586-601) – Witteric (603-610) – Gundemar (610-612) – Sisebut (612-621) – Swinthila (621-631) – Chindaswinth (642-653) – Recceswinth (653-672) – Wamba (672-680) – Erwig (680-687) – Egica (687-700) – Witiza (700-710) – Roderic (710-711) – Agila or Achila II (711-714) Suevic kings – Hermeric (409-441) – Recchila I (441-448) – Rechiar I (448-456) – Theodemund (520-550) – Carriaric (550-559) – Ariamir (559-561) – Theodomir (561-570) – Miro (570-583) – Euric or Eboric (583-584) – Andeca (584-585) – Amalaric (585)

The Iberian Peninsul a bet ween 300 and 850

Appendix 3

Asturian kings – Pelayo (718-737) – Favila (737-739) – Alfonso I (739-757) – Fruela I (757-768) – Aurelius (768-774) – Silo (774-783) – Mauregatus (783-789) – Bermudo (789-791) – Alfonso II (791-842) – Ramiro I (842-850) – Ordoño I (850-866) Umayyad rulers Umayyad governors – Musa ibn Nusair al-Lakhmi (712-714) – Abd al-Aziz ibn Musa (714-716) – Abd al-Rahman ibn Abd Allah al-Ghafiqi (730-732) – Yusuf ibn Abd al-Rahman al-Fihri (747-756) Umayyad emirs – Abd al-Rahman I (754-788) – Hisham I (788-796) – al-Hakam I (796-822) – Abd al-Rahman II (822-852) – Muhammad I (852-886)

337

Abbreviations AAEspA AE AespA AyTM BAR IS CC SL CIL CILA CuPAUAM DLH EME Etym. HG HWR ICERV IHC JRA MGH  AA  SRM  SRG Nat. Hist. OJA PLRE RACH  IV RACH  V RACH  VI RACH VSPE

Anejos a Archivo Español de Arqueología L’Année épigraphique Archivo Español de Arqueología Aqueología y Territorio Medieval British Archaeological Reports, International Series Corpus Christianorum, Series Latina Corpus Inscriptionum Latinarum Corpus de Inscripciones Latinas de Andalucía Cuadernos de Prehistoria y Arqueología de la Universidad Autónoma de Madrid Gregory of Tours, Decem Libri Historiarum Early Medieval Europe Isidore of Seville, Etymologiae Isidore of Seville, Historia de Regibus Gothorum, Wandalorum et Sueborum Julian of Toledo, Historia Wambae regis Vives Castell 1942 Hübner 1871 Journal of Roman Archaeology Monumenta Germaniae Historica Auctores Antiquissimi Scriptores Rerum Merowingicarum Scriptores Rerum Germanicarum Pliny the Elder, Naturalis Historia Oxford Journal of Archaeology Prosopography of the Late Roman Empire Reunió d’Arquelogia Cristiana Hispànica Gurt and Tena (eds.) 1995 Gurt and Tena (eds.) 2000 Gurt and Ribera (eds.) 2005 Vitae Sanctorum Patrum Emeritensium

Bibliography Primary sources Al Razi, Ahmad ibn Muhammad, Ajbār mulūk Al-Andalus; in Diego Catalán and Miguel de Andrés (eds.). 1975. La crónica del moro Rasís (Madrid: Gredos). Astronomer, Vita Hludowici imperatoris, in MGH SRG 64; in Thomas Noble (ed. and transl.). 2009. Charlemagne and Louis the Pious: Lives by Einhard, Notker, Ermoldus, Thegan and the Astronomer (University Park: PennState University Press). Augustine, Sermones. Loeb Classical Library (Cambridge, Mass: Harvard University Press). Avienus, Ora Maritima; in Adolf Schulten (ed.). 1922. Fontes Hispaniae Antiquae vol. I (Barcelona: University of Barcelona). Chronica Caesaraugustana, in MGH AA 11, 221-22. Chronica Rotensis; in Gil Jiménez (ed.) 1985. Crónicas asturianas (Oviedo: Universidad de Oviedo); in Kenneth Wolf (ed. and transl.). 1999. Conquerors and Chroniclers of Early Medieval Spain, Second Edition (Liverpool: University of Liverpool Press). Codex Iustinianus; in Theodor Mommsen (ed.). 1915. Corpus Iuris Civilis, v. 2. (Berlin: Weidmann). Codex Theodosianus; in Clyde Pharr (ed.). 1952. The Theodosian Code and Novels and the Sirmondian Constitutions: A Translation with Commentary, Glossary, and Bibliography (Princeton: Princeton University Press). Continuatio Isidoriana Hispana a. DCCLIV, in MGH AA11, 323-69; translated in Kenneth Wolf (ed. and transl.). 1999. Conquerors and Chroniclers of Early Medieval Spain. Second Edition (Liverpool: University of Liverpool Press). Gregory of Tours, Decem Libri Historiarum, in MGM SRM 1.1; in Lewis Thorpe (transl.). 1974. The History of the Franks (London: Penguin). Ibn Abd al-Hakam, Futuh Ifriqiya wa-l-Andalus; in Eliseo Vidal Beltrán (ed. and transl.). 1966. Conquista de África del norte y de España (Valencia: Anubar). Ibn Habib, Abd Al-Malik. Kitab al-Tarikh; in Jorge Aguadé Bofill (ed. and transl.). 1991. Kitab al-Tarij’. La Historia. Fuentes hispano-arábicas 1 (Madrid: CSIC). Isidore of Seville, Etymologiae; in Wallace Lindsay (ed.). 1911. Etymologiarum libri viginti sive origines. Scriptorum Classicorum Bibliotheca Oxoniensis (Oxford: Clarendon). Isidore of Seville, Historia de Regibus Gothorum, Wandalorum et Sueborum, in MGH AA 11, 241-303; in Kenneth Wolf (ed. and transl.). 1999. Conquerors and Chroniclers of Early Medieval Spain, Second Edition (Liverpool: University of Liverpool Press). John of Biclar, Chronica a. DLXVII-DXC, in MGH AA 11, 207-220; in Kenneth Wolf (ed. and transl.). 1999. Conquerors and Chroniclers of Early Medieval Spain, Second Edition (Liverpool: University of Liverpool Press). Julian of Toledo, ‘Historia Wambae regis’, in CC SL 115. Notitia dignitatum, in Carmen Neira Faleiro (ed.). 2005. La Notitia Dignitatum. Nueva edición crítica y comentario histórico (Madrid: CSIC). Pliny the Elder, Naturalis Historia. Loeb Classical Library (Cambridge, Mass: Harvard University Press). Pliny the Younger, Episutlae. Loeb Classical Library (Cambridge, Mass: Harvard University Press). Vitae Sanctorum Patrum Emeritensium; as ‘The Lives of the Fathers of Mérida’, in Andrew Fear (ed. and transl.). 1997. Lives of the Visigothic Fathers (Liverpool: Liverpool University Press).

342 

The Iberian Peninsul a bet ween 300 and 850

Secondary sources Abad Casal, Lorenzo; Sonia Gutiérrez Lloret; Blanca Gamo Parras; and Pablo Cánovas Guillén. 2008. ‘Una ciudad en el camino: pasado y futuro de El Tolmo de Minateda (Hellín, Albacete)’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 323-36. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Abascal Palazón, José Manuel; Martín Almagro Gorbea; and Rosa Cebrián Fernández. 2008. ‘Segóbriga visigoda’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 221-41. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Abásolo Álvarez, José Antonio; Javier Cortes Álvarez de Miranda; and Fernando Rodríguez Aragón. 1997. La necrópolis Norte de la Olmeda. (Palencia: Diputación Provincial de Palencia). Abásolo Álvarez, José Antonio; and Rafael Martínez González. 2012. Villa romana de La Olmeda: guía arqueológica (Palencia: Diputación Provincial de Palencia). Åberg, Nils. 1922. Die Franken und Westgoten in der Völkerwanderungszeit (Uppsala: Akademiska bokhandeln i kommission). Acero Pérez, Jesús; and Josep Anton Remolà Vallverdú (eds.). 2011. La gestión de los residuos urbanos en Hispania. Xavier Dupré Raventós (1956-2006) in memoriam. AAEspA 60 (Mérida: CSIC). Acién Almansa, Manuel. 1995. ‘La fortificación en al-Andalus’ in Rafael López Guzmán (ed.). Arquitectura del Islam occidental, 29-42 (Granada: El Legado Andalusí). Adam, Jean-Pierre. 2001 [1989]. Roman Building: Materials and Techniques (London: Routledge). Aillet, Cyrille. 2010. Les mozarabes. Christianisme, islamisation et arabisation en péninsule Ibérique (Ixe-XIIe siècle). Bibliotheque de la Casa de Velázquez 45 (Madrid: Casa de Velázquez). Alba Calzado, Miguel. 1998. ‘Consideraciones arqueológicas en torno al siglo V en Mérida: Repercusiones en las viviendas y la muralla’. Mérida. Excavaciones Arqueológicas 2, 361-85. Alba Calzado, Miguel. 2001. ‘Mérida entre la tardoantigüedad y el Islam: datos documentados en el Área Arqueológica de Morería’ in Fernando Valdés Fernández (ed.). La islamización de la Extremadura romana, 265-308. Cuadernos Emeritenses 17 (Mérida: MNAR). Alba Calzado, Miguel. 2004a. ‘Evolución y final de los espacios romanos emeritenses a la luz de los datos arqueológicos (pautas de transformación de la ciudad Tardoantigua y Altomedieval)’ in Trinidad Nogales Basarrate (ed.). Augusta Emerita. Territorios, espacios, imágenes y gentes en Lusitania romana, 209-55 (Mérida: MNAR). Alba Calzado, Miguel. 2004b. ‘Arquitectura doméstica’ in Xavier Dupré Raventós (ed.). Mérida. Colonia Augusta Emerita. Las capitales provinciales de Hispania, 67-84 (Rome: L’erma di Bretschneider). Alba Calzado, Miguel. 2005. ‘La vivienda en Emerita durante la antigüedad tardía: propuesta de un modelo para Hispania’. VI RACH, 121-50. Alba Calzado, Miguel. 2007. ‘Contribuciones al estudio de las infraestructuras hidráulicas de Augusta Emerita’ in Julio Mangas Manjarrés and Santiago Martínez Caballero (eds.). El agua y las ciudades romanas, 147-82 (Madrid: Ediciones 2007). Alba Calzado, Miguel; and Santiago Feijoo Martínez. 2003. ‘Pautas evolutivas de la cerámica común de Mérida en épocas visigoda y emiral’ in Luis Caballero Zoreda, Pedro Mateos Cruz and Manuel Retuerce Velasco (eds.). Cerámicas tardorromanas y altomedievales en la Península Ibérica, 483-504. AAEspA 28 (Mérida: CSIC). Alba Calzado, Miguel; and Sonia Gutiérrez Lloret. 2008. ‘Las producciones de transición al Mundo Islámico: el problema de la cerámica paleoandalusí (siglos VIII y IX)’ in Darío Bernal Casasola and Albert Ribera Lacomba (eds.). Cerámicas hispanorromanas: un estado de la cuestión, 585-613 (Cádiz: Universidad de Cádiz).

Bibliogr aphy

343

Alba Calzado, Miguel; and Pedro Mateos Cruz. 2008. ‘El paisaje urbano de Emerita en época visigoda’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 261-73. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Alexander, Michelle; Christopher Gerrard; Alejandra Gutiérrez; and Andrew Millard. 2015. ‘Diet, society, and Economy in Late Medieval Spain: Stable Isotope Evidence from Muslims and Christians from Gandía, Valencia’. American Journal of Physical Anthropology 156, 263-73. Alfaro Suescun, Egoitz. 2012. ‘La arquitectura de Zaballa: materiales y técnicas constructivas’ in Juan Antonio Quirós Castillo (ed.). Arqueología del campesinado medieval. la aldea de Zaballa, 493-529 (Vitoria-Gasteiz: Universidad del País Vasco). Alfőldy, Géza. 1975. Die römischen Inschriften von Tarraco. Madrider Forschungen 10 (Berlin: W. de Gruyter). Alfőldy, Géza. 2000. ‘Spain’ in Alan Bowman, Peter Garnsey and Dominic Rathbone (eds.). The Cambridge Ancient History Volume XI: The High Empire, AD 70-192, Second Edition, 444-61 (Cambridge: Cambridge University Press). Almagro Basch, Martín. 1978. ‘Datos cronológicos para fechar el acueducto de Segobriga’. Revista de Archivos, Bibliotecas y Museos 81.1, 155-167. Almagro Gorbea, Antonio. 1987. ‘Planimetría de ciudades hispanomusulmanas’. Al-Qantara 8, 421-28. de Almeida, Rui Roberto; Inês Vaz Pinto; Ana Patrícia Magalhães; and Patrícia Brum. 2014. ‘Anforas piscicolas de Tróia: contextos de consumo versus contextos de produção’ in Rui Morais, Adolfo Fernández Fernández and Maria José Sousa (eds.). As produções cerámicas de imitação na Hispania, 405-23. Monografias Ex Officina Hispana II (Oporto: Faculdade de Letras, Univeristy of Porto). Álvarez García, Nuria; Carmina Ballester Martínez; Yolanda Carrión Marco; Elena Grau Almero; Guillermo Pascual Berlanga; Guillem Pérez Jordà; Albert Ribera Lacomba; and Carmen Rodríguez Santana. 2005. ‘L’area productiva d’un edifici del fórum de Valentia al baix imperi (segles IV-V)’. VI RACH, 251-60. Álvarez Martínez, José María. 2006-2007. ‘Aspectos de las élites emeritenses en el Bajo Imperio a través de los programas iconográficos’. Anas 19-20, 13-30. Ambrogi, Annarena. 2011. ‘Ricezione in ámbito periférico e provinciale dei modeli urbani: il caso dei labra marmorei’ in Trinidad Nogales Basarrate and Isabel Rodà de Llanza (eds.). Roma y las provincias: modelo y difusión (2 vols), vol. I, 473-83 (Rome-Mérida: L’Erma di Bretschneider). de Amores Carredano, Fernando; Daniel Gonzalez Acuña; and Enrique García Vargas. 2007. ‘Ánforas Tardoantiguas en Hispalis (Sevilla, España) y el Comercio Mediterráneo’ in Michel Bonifay and Jean-Cristophe Tréglia (eds.). LRCW 2. Late Roman Coarse Wares: Cooking Wares and Amphorae in the Mediterranean. Archaeology and Archaeometry, 133-46. BAR IS 1662 (Oxford: Archaeopress). Anguita Jaén, José María. 2011. ‘El disco de Quiroga y Venancio Fortunato’ in María José García Blanco, Teresa Amado Rodríguez, María José Martín Velasco, Amelia Pereiro Pardo and Manuel Vázquez Buján (eds.). Homenaje a Juan José Moralejo, 35-46 (Santiago de Compostela: Universidad de Santiago de Compostela). Aquilué Abadías, Xavier. 2003. ‘Estado actual de la investigación de la Terra Sigillata Africana en la Península Ibérica en los siglos VI-VII’ in Luis Caballero Zoreda, Pedro Mateos Cruz and Manuel Retuerce Velasco (eds.). Cerámicas tardorromanas y altomedievales en la Península Ibérica. ruptura y continuidad, 11-20. AAEspA 38 (Madrid: CSIC). Aquilué Abadías, Xavier; Xavier Dupré Raventós; Joan Massó Pau; and Joaquín Ruiz de Arbulo. 1991. ‘La cronología de les muralles de Tarraco’. Revista d’Arqueologia de Ponent 1, 271-301. Aranda González, Raúl. 2014. ‘Cerámica de época visigoda. Una historia de la cuestión’. Anales de Prehistoria y Arqueología 30, 107-31.

344 

The Iberian Peninsul a bet ween 300 and 850

Aratikos. 2008. Excavación y prospección arqueológica en la iglesia de San Miguel de Sacramenia (Segovia). Excavation report MS deposited at the Consejería de Cultura de Segovia. Aratikos. 2010. Excavación arqueológica en la necrópolis de Santa María de Tejuela. Villanueva Soportilla (Bozóo, Burgos). Excavation report MS deposited at the Servicio Territorial de Cultura de Burgos. Arbeiter, Achim. 2004. ‘Las túnicas no son clámides. Réplica a un enunciado sobre Centcelles de J. Arce’. Anas 17, 221-30. Arbeiter, Achim. 2014. ‘Zur Christianisierung der hispanischen Stadttopographien in der Spätantike’. Städte in Spanien. Moderne Urbanität seit 2000 Jahren, 103-14 (Mainz: Nünnerich). Arbeiter, Achim. 2015. ‘Die Sakralarchitektur Hispaniens “Reich von Toledo” aus der Sicht der aktuellen Forschung’. Antiquité tardive 23, 219-38. Arce Martínez, Javier. 1978. ‘La “crisis” del siglo III d.C en Hispania y las invasiones bárbaras’. Hispania 8, 257-74. Arce Martínez, Javier. 2002. Mérida tardorromana (300-580 d.C.). Cuadernos Emeritenses 22 (Mérida: MNAR). Arce Martínez, Javier. 2006a. ‘Fana, templa, delubra destrui praecipimus: el final de los templos de la Hispania romana’. AEspA 79, 115-24. Arce Martínez, Javier. 2006b. ‘Obispos, emperadores o propietarios enla cúpula de Centcelles’. Pyrenae 37.2, 131-41. Arce Martínez, Javier. 2007. Bárbaros y romanos en Hispania 400-507 A.D (Madrid: Marcial Pons). Arce Martínez, Javier. 2008. ‘La inscripción del puente de Mérida de época del rey Eurico (483 d.C.)’. Pyrenae 39.2, 121-26. Arce Martínez, Javier. 2009. El último siglo de la España romana (Madrid: Alianza). Arce Martínez, Javier. 2010. ‘El siglo V en Galia e Hispania’ in Jorge Morín de Pablos, Jorge López Quiroga and Artemio Martínez Tejera (eds.). El tiempo de los “Bárbaros”. Pervivencia y transformación en Galia e Hispania (ss. V-VI d. C.), 66-76. Zona Arqueológica 11 (Alcalá de Henares: Museo Arqueológico Regional). Arce Martínez, Javier. 2011. Esperando a los árabes. Los visigodos en Hispania (507-711) (Madrid: Marcial Pons). Arce Sainz, Fernando. 2015. ‘La supuesta basílica de San Vicente en Córdoba: de mito histórico a obsesión historiográfica’. Al-Qantara 36.1, 11-44. Argente Oliver, José Luis; and Adelia Díaz Díaz. 1984. ‘Casa del acueducto y acueducto romano’ in José Luis Argente Oliver (ed.). Tiermes II, 53-196. Excavaciones Arqueológicas de España 128 (Madrid: Ministerio de Cultura). Arias Sánchez, Isabel; and Luis Balmaseda Muncharaz (eds.). 2015. La necrópolis de época visigoda de Castiltierra (Segovia). Excavaciones dirigidas por E. Camps y J.M.ª Navascués, 1932-1935. Materiales conservados en el Museo Arqueológico Nacional (Madrid: Ministerio de Educación, Cultura y Deporte). Arias Sánchez, Isabel; Luis Balmaseda Muncharaz; Soledad Díaz Martínez; Ángela Franco Mata; Concepción Papí Rodes; Beatriz Robledo Sanz; and Gonzalo Trancho Gallo. 2000. ‘La necrópolis visigoda de Castiltierra: proyecto para el estudio de sus materiales’. Boletín del Museo Arqueológico Nacional 18.1-2, 187-196. Arias Páramo, Lorenzo. 2008. Geometría y proporción en la arquitectura asturiana prerrománica. AAEspA 99 (Madrid: CSIC). Ariño Gil, Enrique. 2006. ‘Modelos de poblamiento rural en la provincia de Salamanca (España) entre la antigüedad y la Alta Edad Media’. Zephyrus 59, 317-57. Ariño Gil, Enrique. 2013. ‘El hábitat rural en la Península Ibérica entre f inales del siglo IV y principios del VIII: un ensayo interpretativo’. Antiquité Tardive 21, 93-123.

Bibliogr aphy

345

Ariño Gil, Enrique; and Pablo Díaz Martínez. 2002. ‘El campo: propiedad y explotación de la tierra’ in Ramón Teja Casuso (ed.). La Hispania del siglo IV. Administración, economía, sociedad, cristianización, 59-96 (Bari: Edipuglia). Ariño Gil, Enrique; and Pablo Díaz Martínez. 2014. ‘La frontera suevo-visigoda: ensayo de lectura de un territorio en disputa’ in Raúl Catalán Ramos, Patricia Fuentes and José Carlos Sastre (eds.) Fortificaciones en la tardoantigüedad. Élites y articulación del territorio (siglos V-VIII d. C.), 179-90 (Madrid: La Ergástula). Arjona Castro, Antonio. 2001. Córdoba en la historia de a-Ándalus. Desarrollo, apogeo y ruina de la Córdoba Omeya (Córdoba: Real Academia de Córdoba). Arranz Mínguez, Jesús Álvaro; Santiago Carretero Vaquero; Salvado Repiso; and Luis Carlos San Miguel Mate. 1989. ‘Arqueología hispanovisigoda en Valladolid: el yacimiento de Piñel de Abajo’. Revista de Arqueología 104, 8-12. Aurrecoechea Fernández, Joaquín. 2001. Los cinturones romanos de la Hispania del bajo imperio. Monographies Instrumentum 19 (Montagnac: Monique Mergoil). Azkarate Garai-Olaun, Agustín. 1988. Arqueología cristiana de la Antigüedad Tardía en Álava, Guipúzcoa y Vizcaya (Vitoria: Diputación Foral de Álava). Azkarate Garai-Olaun, Agustín. 2004a. ‘¿Reihengräberfelder al sur de los Pirineos occidentales?’. Antigüedad y Cristianismo 21, 389-414. Azkarate Garai-Olaun, Agustín. 2004b. ‘El País Vasco en los siglos inmediatos a la desaparición del Imperio Romano’ in Pedro Baruso Barés and José Ángel Lema Pueyo (eds.). Historia del País Vasco. Edad Media (siglos V-XV), 23-50 (San Sebastián: Universidad del País Vasco). Azkarate Garai-Olaun, Agustín; and Juan Antonio Quirós Castillo. 2001. ‘Arquitectura doméstica altomedieval en la Península Ibérica. Reflexiones a partir de las excavaciones arqueológicas de la Catedral de Santa María de Vitoria-Gasteiz, País Vasco’. Archeologia Medievale 28, 25-60. Azkarate Garai-Olaun, Agustín; José Manuel Martínez Torrecilla; and José Luis Solaun Bustinza. 2011. ‘Metalurgia y hábitat en el País Vasco de época medieval: el asentamiento ferrón de Bagoeta, Álava (ss.VII-XIV d.C)’. AyTM 18, 71-89. Balmaseda Muncharaz, Luis. 2008. ‘La escultura de Recópolis’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en época visigoda, 14-57. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Banks, Philip. 1999. ‘Las ciudades y su papel’ in Jordi Camps (ed.). Cataluña en época carolingia. Arte y cultura antes del Románico (siglos IX y X), 65-70 (Barcelona: Museo Nacional de Arte de Cataluña). Barceló Perelló, Miquel. 1988. ‘Prólogo’ in Miquel Barceló Perelló (ed.). Arqueología medieval en las afueras del “medievalismo”, 9-17 (Barcelona: Crítica). Barceló Torres, Carmen. 2002a. ‘Escritos árabes en la basílica paleocristiana de Casa Herrera (Mérida)’. Madrider Mitteilungen 43, 299-315. Barceló Torres, Carmen. 2002b. ‘Las inscripciones omeyas de la alcazaba de Mérida’. AyTM 11.1, 59-78. Barraca de Ramos, Pilar. 1999. ‘Ávila en la antigüedad tardía’ in Luis García Moreno and Sebastián Rascón Marqués (eds.). Complutum y las ciudades hispanas en la antigüedad tardía, 181-192. Acta Antiqua Complutensia I (Alcalá de Henares: Universidad de Alcalá). Barroso Cabrera, Rafael; Salvador Jaque Ovejero; Manuele Major González; Jorge Morín De Pablos; Eduardo Penedo Cobo; and Pilar Oñate Baztán. 2001. ‘Los yacimientos de Tinto Juan de la Cruz, Pinto, Madrid (ss. I al VI d.C)’. Estudios de Prehistoria y Arqueología madrileñas 11, 128-204. Barroso Cabrera, Rafael; Jorge Morín de Pablos; and Jorge López Quiroga. 2010. ‘La presencia ‘bárbara’ en Hispania en las necrópolis del s. V d.C.’ in Jorge Morín de Pablos, Jorge López Quiroga and Artemio Martínez Tejera (eds.). El tiempo de los “Bárbaros”. Pervivencia y transformación en Galia e Hispania (ss. V-VI d. C.), 135-47. Zona Arqueológica 11 (Alcalá de Henares: Museo Arqueológico Regional).

346 

The Iberian Peninsul a bet ween 300 and 850

Barrucand, Marianne; and Achim Bednorz. 2002 [1992]. Moorish Architecture in Andalusia (Cologne: Taschen). Bates, Michael. 1992. ‘The Coinage of Spain Under the Umayyad Caliphs of the East, 711-750’ in Actas III Jarique de Numismática Hispano-árabe, Museo Arqueológico Nacional, Madrid, 13-16 diciembre 1990, 271-89 (Madrid: Museo Arqueológico Nacional). Bazzana, Andrea. 1991. Maisons d’al-Andalus. Habitat médievál et structures du peuplement dans l’Espagne orientale. Collection de la Casa de Velázquez 37 (Madrid: Casa Velázquez). Bejarano Osorio, Ana María. 1996. ‘Tipología de las sepulturas en las necrópolis cristianas de Mérida. Evolución de los espacios funerarios’. Mérida, excavaciones arqueológicas 2, 341-60. Bejarano Osorio, Ana María. 2004. El mausoleo del dintel de los ríos: los contextos funerarios tardíos de Augusta Emerita. Cuadernos Emeritenses 27 (Mérida: MNAR). Beltrán de Heredia, Julia. 2002a. ‘Continuity and Change in the Urban Topography: Archaeological Evidence of the North-East Quadrant of the City’ in Julia Beltrán de Heredia (ed.). From Barcino to Barchinona (1st to 7th centuries): The Archaeological Remains of the Plaça del Rei in Barcelona, 96-111 (Barcelona: Museo de Historia de la Ciudad). Beltrán de Heredia, Julia. 2002b. ‘A Garum and Salt Fish Factory in Barcino’ in Julia Beltrán de Heredia (ed.). From Barcino to Barchinona (1st to 7th centuries). The Archaeological Remains of the Plaça del Rei in Barcelona, 58-65 (Barcelona: Museo de Historia de la Ciudad). Beltrán de Heredia, Julia. 2008a. ‘Barcino durante la antigüedad tardía’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 274-91. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Beltrán de Heredia, Julia. 2008b. ‘De la ciutat tardoantiga a la ciutat medieval: Barcelona al segle XII’ in Manuel Castiñeiras and Jordi Camps (eds.). El romànic i la Mediterrània, 39-45 (Barcelona: Museo de Arte de Cataluña). Beltrán de Heredia, Julia; and Carlos Carreras Monfort. 2011. ‘Barcino’ in Josep Antón Remolà Vallverdú and Jesús Acero Pérez (eds.). La gestión de los residuos urbanos en Hispania, 233-54. AAEspA 60 (Mérida: CSIC). Beltrán Fortes, José. 1999. Los sarcófagos romanos de la Bética con decoración de tema pagano (Málaga: University of Málaga). Beltrán Fortes, José; Daniel González Acuña; and SalvadorOrdóñez Agulla. 2005. ‘Acerca del urbanismo de Hispalis. Estado de la cuestión y perspectivas’. Mainake 27, 61-88. Beltrán Rizo, Enric. 2013. La correspondencia de Símaco. Interpretación, cronología y cartas no conservadas (libros III-IV). PhD thesis (Universidad de Barcelona). Benéitez González, Carmen; and Fernando Miguel Hernández. 1993-1994. ‘Relectura arqueológica de la villa romana de Navatejera (León)’. Numantia 6, 103-26. Bernal Casasola, Darío. 2008a. ‘Ciudades del Fretum Gaditanum tardoantiguo’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 363-83. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Bernal Casasola, Darío. 2008b. ‘Ánforas y vino en la Antigüedad Tardía. El ejemplo de Hispania meridional’ in Juan Blázquez Miguel and Sebastián Celestino Pérez (eds.). El vino en época tardoantigua y medieval, 33-60. Serie Varia 8 (Madrid: Universidad Autónoma de Madrid). Bernal Casasola, Darío; Lourdes Roldán Gómez; Juan Blánquez Pérez; José Juan Díaz Rodríguez; and Fernando Prados Martínez. 2009. ‘Del marisqueo a la producción de púrpura. Estudio arqueológico del conchero tardorromano de Villa Victoria/Carteia (San Roque, Cádiz)’ in Darío Bernal Casasola (ed.). Arqueología de la pesca en el estrecho de Gibraltar. De la prehistoria al fin del Mundo Antiguo, 199-58 (Cádiz: Universidad de Cádiz). Berni Millet, Piero and Juan Moros Díaz. 2012. ‘Dressel 23 (Valle del Guadalquivir)’ in Amphorae ex Hispania. Paisajes de producción y de consumo (http://amphorae.icac.cat/tipol/view/2).

Bibliogr aphy

347

Berni Millet, Piero. 2012. Tejarillo I (Valle del Guadalquivir)», Amphorae ex Hispania. Paisajes de producción y de consumo (http://amphorae.icac.cat/tipol/view/51). Berrocal Caparrós, María del Carmen. 2012. ‘Producciones de ánforas en la costa meridional de Carthago Spartaria’ in Darío Bernal Casasola and Albert Ribera Lacomba (eds.). Cerámicas hispanorromanas II. Producciones regionales, 255-780. Monografías Historia y Arte (Cádiz: Universidad de Cádiz). Berrocal Caparrós, María del Carmen. 2014. ‘Keay XLI (Costa meridional de Tarraconensis)’ in Amphorae ex Hispania. Paisajes de producción y de consumo (http://amphorae.icac.cat/ tipol/view/86). Berrocal Caparrós, María del Carmen; and María Dolores Laiz Reverte. 1995a. ‘Elementos para la datación cronológica de la necrópolis paleocristiana de San Antón’ in IV RACH, 163-172. Berrocal Caparrós, María del Carmen; and María Dolores Laiz Reverte. 1995b. ‘Tipología de enterramientos en la necrópolis de San Antón, en Cartagena’ in IV RACH, 173-182. Bertrand, Maryelle; and José Ramón Sánchez Viciana. 2000. ‘Jolopos (La Peza, Grenade). Un hisn de la fitna’ in Isabel Ferreria Fernandes (ed.). Mil Anos de Fortificaçoes na Península Ibérica e no Magreb (500-1500). Simpósio Internacional sobre Castelos 2000 Castelos, 145-60 (Lisbon: Colibrí). Blanco García, Juan Francisco. 2002. ‘Coca. Cauca’ in Trinidad Mañanes Pérez (ed.). Arqueología del área central de la cuenca del río Duero: de Simancas a Coca, 127-73 (Valladolid: Junta de Castilla y León). Blanco González, Antonio; José Antonio López Sáez; Francisca Alba Sánchez; Daniel Abel Schaad; and Sebastián Pérez Díaz. 2014. ‘Medieval Landscapes in the Spanish Central System (450-1350): A Paleoenvironmental and Historial Perspective’. Journal of Medieval Iberian Studies 6, 1-17. Blanco González, Antonio; José Antonio López Sáez; and Lourdes López Merino. 2009. ‘Ocupación y uso del territorio en el sector centromeridional de la cuenca del Duero entre la Antigüedad y la Alta Edad Media (siglos I-XI d.C)’. AEspA 82, 275-300. Blázquez Martínez, José María. 1986. ‘Mosaicos hispanos de la época de las invasiones bárbaras. Problemas estéticos’. Antigüedad y Cristianismo 3, 463-90. Blázquez Martínez, José María; and Joaquín Remesal Rodríguez (eds.). 2000. Estudios sobre el Monte Testaccio (Roma) (Barcelona: Universidad de Barcelona). Bonifay, Michel; and Darío Bernal Casasola. 2008. ‘Recópolis, paradigma de las importaciones africanas en el visigothorum regnum’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 99-115. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Bonilla Santander, Óscar; and Begoña Serrano Arnáez. 2012. ‘El paisaje rural romano en torno al Municipium Turiaso (España)’ in Pilar Diarte Blasco and Alejandro Martín López (eds.). Arqueología de un Paisaje en Transición. Antigüedad Tardía y Alta Edad Media, 95-100 (Zaragoza: Universidad de Zaragoza). Bonilla Santander, Óscar; Begoña Serrano Arnáez; and Carlos García Benito. 2011. ‘Ager Turiasonense: los yacimientos de Torre don Antonio, Valoria y Valoria Soler (Tarazona, España)’. Turiaso 20, 33-54. Bosch Puche, Francisco; Josep Maria Macias Solé; Joan Menchon Bes; Andreu Muñoz Melgar; and Immaculada Teixell Navarro. 2005. ‘La transformació urbanística de l’acròpolis de Tarracona: avanç de les excavacions del pla director de la catedral de Tarragona (2000-2002)’ in VI RACH, 167-74. Bourdieu, Pierre. 2000 [1972]. Esquisse d’une théorie de la pratique (Paris: Éditions du Seuil). Bowes, Kim. 2001. ‘Nec sedere in villam: Villa Churches, Rural Piety and the Priscillianist Controversy’, in Thomas S. Burns and John William Eadie (eds.). Urban Centers and Rural contexts in Late Antiquity, 323-48 (Michigan: Michigan State University Press).

348 

The Iberian Peninsul a bet ween 300 and 850

Bowes, Kim; and Michael Kulikowski (eds.). 2005. Hispania in Late Antiquity: Current Perspectives (Leiden: Brill). Brandes, Wolfgang. 1999. ‘Byzantine Cities in the Seventh and Eighth Centuries: Different Sources, Different Stories?’ in Gian Pietro Brogiolo and Bryan Ward-Perkins (eds.). The Idea and Ideal of Town Between Late Antiquity and the Early Middle Ages, 25-58 (Leiden: Brill). Brather, Sebastian. 1998. ‘Ethnic Identities as Constructions of Archaeology: The Ccase of the Alamanni’ in Helmut Reimitz and Walter Pohl (eds.). Strategies of Distinction: The Construction of Ethnic Communities, 300-800, 149-75 (Leiden: Brill). Brogiolo, Gian Pietro. 1999. ‘Ideas of the Town in Italy During the Transition from Antiquity to the Middle Ages’ in Gian Pietro Brogiolo and Bryan Ward-Perkins (eds.) The Idea and Ideal of Town Between Late Antiquity and the Early Middle Ages, 99-126 (Leiden: Brill). Brogiolo, Gian Pietro. 2000. ‘Towns, Forts and the Countryside: Archaeological Models for Northern Italy in the Early Lombard Period (AD 568-650)’ in Towns and their Territories Between Late Antiquity and the Early Middle Ages, 299-323 (Leiden: Brill). Brogiolo, Gian Pietro; and Alexandra Chavarría Arnau. 2008. ‘Dei vandali ai longobardi: osservazioni sull´insediamento barbarico nelle campagne dell´occidente’ in Guido Berndt and Roland Steinacher (eds.). Das Reich der Vandalen und seine Vorgeschichte(n), 261-81. Forschungen zur Geschichte des Mittelalters 13 (Vienna: Institut für Mittelalterforschung der Österreichischen Akademie der Wissenschaften). Brogiolo, Gian Pietro; Alexandra Chavarría Arnau; and Marco Valenti. 2005. Dopo la fine delle ville. le campagne dal VI al IX secolo (Mantova: SAP). Brookes, Stuart; Catarina Tente and Sara Prata. 2017. ‘Interpreting rock-cut grave cemeteries: the early medieval necropolis and enclosure of São Gens, Portugal’. Medieval Archaeology 61.2, 216-23. Brown, Peter. 1976. The Making of Late Antiquity (Cambridge, Mass: Harvard University Press). Buchberger, Erica. 2013. From Romans to Goths and Franks: Ethnic Identities in Sixth- and SeventhCentury Spain and Gaul. DPhil Thesis (University of Oxford). Bugalhão, João. 2011. ‘A Arqueologia portuguesa nas últimas décadas’, Arqueologia & História 60-61, 19-43. Büntgen, Ulf; Willy Tegel; Kurt Nicolussi; Michael McCormick; David Frank; Valerie Trouet; Jed Kaplan; Franz Herzig; Karl-Uwe Heussner; Heinz Wanner; Jürg Luterbacher; and Jan Esper. 2011. ‘2500 Years of European Climate Variability and Human Susceptibility’. Science 331, 578-82. Burns, Vincent. 1992. ‘The Visigothic Settlement in Aquitania: Imperial Motives’ in Hisotria: Zeitschrift für Alte Geshichte 41.3, 362-73. Bustamante Álvarez, Macarena. 2013. ‘El trabajo artesanal en Augusta Emerita’. Zephyrus 72, 113-38. Caballero Arribas, Jesús; and David Peñas Pedrero. 2012. ‘Un castrum de época visigoda en el Valle Amblés: La Cabeza de Navasangil (Solosancho, Ávila)’ in Juan Antonio Quirós Castillo (ed.). Arqueología e historia de los castillos altomedievales en el cuadrante noroccidental de la Península Ibérica, 213-38 (Bilbao: University of the Basque Country). Caballero Zoreda, Luis. 1984. ‘Arqueología tardorromana y visigoda en la provincia de Soria’ in I Symposio de arqueología soriana, 435-457 (Soria: Museo Numantino). Caballero Zoreda, Luis. 2000. ‘La arquitectura denominada de época visigoda, ¿es realmente tardorromana o prerrománica?’ in Luis Caballero Zoreda and Pedro Mateos Cruz (eds.). Visigodos y Omeyas. Un debate entre la Antigüedad tardía y la alta Edad Media, 207-48. AAEspA 23 (Madrid: CSIC). Caballero Zoreda, Luis. 2009. ‘La iglesia de El Gatillo de Arriba (Cáceres). Apuntes sobre una iglesia rural en los siglos VI al VIII’ in Luis Caballero Zoreda, Pedro Mateos Cruz, and María Ángeles Utrero Agudo (eds.). El siglo VII frente al siglo VII. Arquitectura, 155-85. AAEspA 51 (Madrid: CSIC).

Bibliogr aphy

349

Caballero Zoreda, Luis; and Pedro Mateos Cruz (eds.). 2000. Visigodos y Omeyas. Un debate entre la Antigüedad tardía y la alta Edad Media (Mérida, abril de 1999). AAEspA 23 (Madrid: CSIC). Caballero Zoreda, Luis; and Pedro Mateos Cruz. 1992. ‘Excavaciones en Santa Eulalia de Mérida’. I Jornadas de Prehistoria y Arqueología en Extremadura (1986-1990), 525-46. Extremadura Arqueológica 2 (Mérida: Asamblea de Extremadura). Caballero Zoreda, Luis; and Fransico José Moreno Martín. 2013. ‘Balatalmelc, Santa María de Melque. Un monasterio del siglo VIII en territorio toledano’ in Xavier Ballestín Navarro and Ernesto Pastor Díaz (eds.). Lo que vino de Oriente. Horizontes, praxis y dimensión material de los sistemas de dominación fiscal en Al-Andalus (ss. VII-IX), 182-204. BAR IS 2525 (Oxford: Archaeopress). Caballero Zoreda, Luis; and Eduardo Rodríguez Trobajo. 2010. Las iglesias asturianas de Pravia y Tuñón. Arqueología de la Arquitectura. AAEspA 54 (Madrid: CSIC). Caballero Zoreda, Luis; and Isaac Sastre de Diego. 2014. ‘Espacios de la liturgia hispana de los siglos V-X. Según la arqueología’ in Ismael Fernández de la Cuesta, Rosario Álvarez Martínez and Ana Llorens Martín (eds.). El canto mozárabe y su entorno. Estudios sobre la música de la liturgia viejo hispánica, 259-91 (Madrid: Sociedad española de musicología). Caballero Zoreda, Luis; and Fernando Sáenz Lara. 1999. La iglesia mozárabe de Santa Lucía del Trampal, Alcuéscar (Cáceres). Arqueología y arquitectura (Mérida: Consejería de Cultura). Caballero Zoreda, Luis; and Thilo Ulbert. 1976. La basílica paleocristiana de Casa Herrera en las cercanías de Mérida (Badajoz) (Madrid: Dirección General del Patrimonio). Caballero Zoreda, Luis; and María Ángeles Utrero Agudo. 2008. ‘Una aproximación a las técnicas constructivas de la alta edad media peninsular. Entre visigodos y omeyas’ Arqueología de la Arquitectura 4, 169-92. Caballero Zoreda, Luis; Pedro Mateos Cruz; and Tomás Cordero Ruiz (eds.). 2012a. Visigodos y omeyas. El territorio. AAEspA 61 (Mérida: CSIC). Caballero Zoreda, Luis; Pedro Mateos Cruz; and César García de Castro. Asturias entre visigodos y mozárabes. Visigodos y omeyas VI. AAEspA 63 (Madrid: CSIC). Caballero Zoreda, Luis; Pedro Mateos Cruz; and Manuel Retuerce Velasco (eds.). 2003. Cerámicas tardorromanas y altomedievales en la Península Ibérica. ruptura y continuidad. AAEspA 28 (Madrid: CSIC). Cabral, João Manuel Peixoto; and David Metcalf. 1997. A moeda Sueva = Suevic Coinage (Porto: Sociedade Portuguesa de Numismática). Calligaro, Thomas; Patrick Périn; Françoise Vallet; and Jean-Paul Poirot. 2006-2007. ‘Contribution à l’étude des grenats mérovingiens (Basilique de Saint-Denis et autres collections du musée d’Archéologie nationale, diverses collections publiques et objets de fouilles récentes’. Antiquités nationales 38, 111-44. Calvo Capilla, Susana. 2011. ‘Les premières mosquées et la transformation des sanctuaires wisigothiques (92H/711-170H/785)’. Mélanges de la casa Velázquez 41.2, 131-63. Campos López, Teresa; David Martínez Izquierdo; and Silvia Cajigas Panera. 2010. ‘Los orígenes altomedievales de Górliz. Una primera aproximación’ in Juan Antonio Quirós Castillo (ed.). The Archaeology of Early Medieval Villages in Europe, 457-61 (Vitoria-Gasteiz: University of the Basque Country). Camps y Cazorla, Emilio. 1939-1940. ‘El visigotismo de Quintanilla de las Viñas’. Boletín del Seminario de Estudios de Arte y Arqueología 6, 125-34. Cano Ávila, Pedro. 2005. ‘Numismática omeya y oriental en al-Andalus’. Phililogia Hispalensis 19, 23-31. Canto García, Alberto. 2011. ‘Las monedas y la conquista’ in 711. Arqueología e Historia entre dos mundos, 135-46. Zona Arqueológica 15 (Alcalá de Henares: Museo Arqueológico Regional).

350 

The Iberian Peninsul a bet ween 300 and 850

Canto y De Gregorio, Alicia. 2001. ‘Fuentes árabes para la Mérida Romana’ in Fernando Valdés and Agustín Velázquez (eds.) La islamización de la Extremadura romana, 9-86. Cuadernos Emeritenses 17 (Mérida: Museo Nacional de Arte Romano). Cardell Perelló, Jaume; and Miguel Ángel Cau Ontiveros. 2005. ‘Las Baleares bizantinas a partir de la documentación arqueológica’ in Rafael Durán Tapia (ed.). Mallorca y Bizancio, 157-88 (Palma de Mallorca: Asociación de Amigos del Castillo de San Carlos). Carreté, Josep Maria; Simon Keay; and Martin Millet. 1995. A Roman Provincial Capital and its Hinterland: The Survey of the Territory of Tarragona 1985-1990. JRA Supplement Series 15 (Ann Harbour: JRA). Carrillo, José Ramón. 1999. ‘Evolución de la arquitectura doméstica en Colonia Patricia Corduba’ in Francisco García Verdugo and Francisco Acosta Ramírez (eds.). Actas del Coloquio Córdoba en la Historia. la construcción de la Urbe, 75-87 (Córdoba: Universidad de Córdoba). Carvajal Castro, Álvaro; and Iñaki Martín Viso. 2013. ‘Historias regionales de la repoblación: los reyes asturleoneses y las ‘políticas de la tierra’ en el oeste de la Meseta del Duero’ in Pablo Díaz Martínez, Luis Corral Lafuente and Iñaki Martín Viso (eds.). El historiador y la sociedad, 39-52 (Salamanca: Universidad de Salamanca). Carvajal López, José Cristóbal. 2009. ‘Pottery Production and Islam in South East Spain: A Social Model’. Antiquity 83, 388-98. Carvajal López, José Cristóbal. 2013. ‘Islamicisation or Islamicisations? Islamic Expansion and Social Practice in the Vega of Granada (South East Spain)’. World Archaeology 45.1, 57- 7. Carvajal López, José Cristobal; and Peter Day. 2013. ‘Cooking Pots and Islamicization in the Early Medieval Vega of Granada (Al-Andalus, Sixth to Twelfth Centuries)’. OJA 32.4, 433-51. Carvajal López, José Cristobal; Julio Román Punzón; Miguel Jiménez Puertas; and Javier Martínez Jiménez. 2015. ‘When the East Came to the West: The Seventh Century in the Vega of Granada (South East Spain). Visigoths, Byzantines and Muslims’ in Alessandro Gnasso, Emanuele Intagliata, Thomas MacMaster, and Bethan Morris (eds.), The Long Seventh Century: Continuity and Discontinuity in an Age of Transition, 135-62 (Bern: Peter Lang). Casal García, María Teresa. 2008. ‘Características generales del urbanismo cordobés de la primera etapa emiral: el arrabal de Saqunda’. Anales de Arqueología Cordobesa 1, 109-34. Casal García, María Teresa; Alberto León Muñoz; Rosa López Guerrero; Ana Valdivieso Ramos; and PatricioSoriano Castro. 2006. ‘Espacio y usos funerarios en la Qurtuba islámica’. Anales de Arqueología Cordobesa 17.2, 257-90. Castanyer Masoliver, Pere; Joaquim Tremoleda Trilla; and Rafael Dehesa Carreira. 2013. ‘El establecimiento rural de época visigoda de Vilauba. Algunas reflexiones sobre el final de las villas romanas en el nordeste de la tarraconense’ in Jean-Luc Fiches, Rosa Plana-Mallart and Victor Revilla Calvo (eds.). Paisajes rurales y territorios en las ciudades del occidente romano, Gallia e Hispania, 313-27 (Montpellier: Presses Universitaires de la Méditerranée). Castellanos Martínez, Santiago. 2003. ‘The Political Nature of Taxation in Visigothic Spain’. EME 12.3, 201-28. Castellanos Martínez, Santiago. 2013. En el final de Roma (ca. 455-480). La solución intelectual (Madrid: Marcial Pons). Castellanos Martínez, Santiago; and Iñaki Martín Viso. 2005. ‘The Local Articulation of Central Power in the North of the Iberian Peninsula’. EME 13, 1-42. Castelo, Raquel; Carolina Gutiérrez Neira; JoaquínBarrio Martín. 2011-2012. ‘Estudio arqueohistórico y analítico de un conjunto de vidrios de la villa romana de El Saucedo (Talavera La Nueva, Toledo)’. CuPAUAM 37-38, 687-703. Castillo Armenteros, Juan Carlos. 1998. La campiña de Jaén en época emiral (s. VIII-X) (Jaén: University of Jaén).

Bibliogr aphy

351

del Castillo Yurrita, Alberto. 1970. ‘Cronología de las tumbas llamadas olerdolanas’ in Gratiniano Nieto (ed.) XI Congreso Nacional de Arqueología, 835-45 (Zaragoza: Secretaría General de los Congresos Arqueológicos Nacionales). del Castillo Yurrita, Alberto. 1972. Excavaciones altomedievales en las provincias de Soria, Logroño y Burgos (Madrid: Comisaría General de Excavaciones Arqueológicas). Castillo Maldonado, Pedro. 1999. Los mártires hispanorromanos y su culto en la Hispania de la Antigüedad Tardía (Granada: Universidad de Granada). Castro Priego, Manuel. 2008. ‘Los hallazgos numismáticos de Recópolis: aspectos singulars de su integración en la secuencia histórica del yacimiento’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 130-41. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Castro Priego, Manuel; and Amaya Gómez de la Torre-Verdejo. 2008. ‘La actividad artesanal en Recópolis: la producción de vidrio’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 117-28. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Catalán Ramos, Raúl. 2015. ‘Elementos de armamento y mundo funerario en Hispania (siglos V-VIII)’ in Juan Antonio Quirós Castillo and Santiago Castellanos (eds.). Identidad y etnicidad en Hispania: propuestas teóricas y cultura material en los siglos V-VIII, 291-312 (Bilbao: Universidad del País Vasco). Catarino, Helena. 2000. ‘Castelos e território omíada na Kura de Ocsonoba’ in Isabel Ferreria Fernandes (ed.). Mil Anos de Fortificaçoes na Península Ibérica e no Magreb (500-1500). Simpósio Internacional sobre Castelos 2000 Castelos, 29-44 (Lisbon: Colibrí). Catteddu, Isabelle. 2001. Les habitats carolingiens de Montours et La Chapelle-Saint-Aubert (Ile-et-Vilaine) (Paris: Éditions de la Maison des sciences de l’Homme). Catteddu, Isabelle. 2009. Archéologie médiévale en France (Paris: La Découverte). Cau Ontiveros, Miguel Ángel; and María Esther Chávez Álvarez. 2003. ‘El fenómeno urbano en Mallorca en época romana: los ejemplos de Pollentia y Palma’. Mayurqa 29, 27-49. Ceballos Hornero, Alberto; and David Ceballos Hornero. 2003. ‘Los espectáculos del anfiteatro en Hispania’. Iberia 6, 57-70. Cepas Palanca, Adela. 1997. Crisis y continuidad en la Hispania del siglo III. AAEspA 17 (Madrid: CSIC). Ceprián del Castillo, Bautista; Carmen Repullo Roldán; and Jorge Alarcón Fernández. 2013. ‘Broches de cinturón de tipo hispano en Cástulo: una contribución al ejército bajo imperial en Hispania’. Gladius 33, 81-112. Chalmeta Gendrón, Pedro. 1994. Invasión e islamización. La sumisión de Hispania y la formación de al-Andalus (Jaén: MAPFRE). Chalmeta Gendrón, Pedro. 1998. ‘Al-Andalus: La implantación de una nueva superestructura’ in Pedro Chalmeta Gendrón (ed.). Ruptura o Continiudad. Pervivencias preislámicas en al-Ándalus, 9-28. Cuadernos Emeritenses 15 (Mérida: MNAR). Chalmeta Gendrón, Pedro. 2013. ‘Derecho y práctica fiscal musulamana: el primer siglo y medio’, in Xavier Ballestín and Ernesto Pastor (eds.). Lo que vino de Oriente. Horizontes, praxis y dimensión material de los sistemas de dominación fiscal en Al-Andalus (ss. VII-IX), 1-17. BAR IS 2525 (Oxford: Archaeopress). Chapelot, Jean; and Robert Fossier. 1985. Le village et la maison au Moyen Age. (London: Batsford). Chavarría Arnau, Alexandra. 2007. El final de las villae en Hispania (siglos IV-VII d.C.). Biliotèque de l’Antiquité Tardive 7 (Turnhaut: Brepols). Chavarría Arnau, Alexandra. 2010. ‘Churches and Aristocracies in Seventh-Century Spain: Some Thoughts on the Debate on Visigothic Churches’. EME 18.2, 160-74.

352 

The Iberian Peninsul a bet ween 300 and 850

Chavarría Arnau, Alexandra. 2012. ‘¿Castillos en el aire? Paradigmas interpretativos “de moda” en la arqueología medieval española’ in De Mahoma a Carlomagno. Los primeros tiempos (siglos VII-IX). Actas XXXIX Semana de estudios medievales, 131-66 (Pamplona: Gobierno de Navarra). Chifflet, Jean-Jacques. 1655. Anastasis Childerici I Regis sive thesaurus sepulchralis Tornaci Nerviorum effossus, et commentario illustratus (Antwerp: Balthasar Moretus). Christie, Neil. 2006. From Constantine to Charlemagne: An Archaeology of Italy AD 300-800 (Aldershot: Ashgate). Clarke, Giles. 1975. ‘Population Movements and Late Roman Cemeteries’. WorldArchaeology 7.1, 46-56. Clarke, Nicola. 2012. The Muslim Conquest of Iberia: Medieval Arabic Narratives (London: Routledge). Coll Riera, Joan Manuel; and Jordi Roig Buxó. 2013. ‘La fi de les vil·les romanes baiximperials a la depressió prelitoral (segles IV-V): contextos estratigràfics i registrematerial per datar-los’ in Isabel Fernández, Joan Menchon and Josep María Vila (eds.). Actes del IV congrés d’arqueologia medieval i moderna a Cataunya, 161-72 (Tarragona: Ayuntamiento de Tarragona). Collins, Roger. 1990. The Basques (Oxford: Blackwell). Collins, Roger. 2005 [2004]. La España Visigoda 409-711 (Barcelona: Crítica). Collins, Roger. 2012. Caliphs and Kings: Spain 796-1031. A History of Spain (Oxford: Blackwell). Contreras Martínez, Miguel. 2011. La necrópolis altomedieval de Gózquez de Arriba (San Martín de la Vega). Excavation Report MS deposited at the Dirección General de Patrimonio de la Comunidad de Madrid. Contreras Martínez, Miguel; and Antonio Fernández Ugalde. 2006. ‘El espacio funerario en el poblado de época visigoda de Gózquez de Arriba’ in Jorge Morín De Pablos (ed.). La investigación arqueológica de la época visigoda en la Comunidad de Madrid (3 vols), vol. II, 517-34. Zona Arqueológica 8 (Alcalá de Henares: Museo Arqueológico Regional). Cordero Ruiz, Tomás; and Bruno Franco Moreno. 2012. ‘El territorio emeritense durante la Antigüedad Tardía y la Alta Edad Media’ in Luis Caballero Zoreda, Pedro Mateos Cruz and Tomás Cordero Ruiz (eds.). Visigodo y Omeyas: El territorio, 147-69. AAEspA 61 (Madrid: CSIC). Corral Lafuente, José Luis. 1991. ‘Las ciudades de la Marca Superior de Al-Ándalus’ in Mikel de Epalza (ed.). Simposio internacional sobre la ciudad islámica, 253-90 (Zaragoza: Institución Fernando el Católico). Correia, Virgilio Hipólito; and Maria Pilar Reis. 2000. ‘As termas de Conimbriga: tipologías arquitectónicas e integração urbana’ in Carmen Fernández Ochoa and Virginia García-Entero (eds.). Termas romanas en el occidente del imperio. II Coloquio internacional de arqueología en Gijón, 271-80 (Gijón: Ayuntamiento de Gijón). Cortés Arrese, Miguel. 2012. Los visigodos de los románticos. Madrid: Catarata. Crawford, John. 1990. The Byzantine Shops at Sardis (Cambridge, Mass: Harvard University Press). Cressier, Patrice. 2001. ‘El acarreo de obras antiguas en la arquitectura islámica de primera época’ in Fernando Valdés and Agustín Velázquez (eds.). La islamización de la Extremadura romana, 309-34. Cuadernos Emeritenses 17 (Mérida: MNAR). Creswell, Keppel. 1940. Early Islamic Architecture: Umayyads, Early Abbasids and Tulumds. Part Two: Early Abbasids, Umayyads of Cordova, Aghlabids, Tulunids and Samanids AD 751-905 (Oxford: Clarendon). Crone, Patricia. 2005 [2004]. Medieval Islamic Political Thought (Edinburgh: Edinburgh University Press). Cruz Hernández, Miguel. 1998 [1979]. ‘The Social Structure of Al-Andalus During the Muslim Occupation (711-755) and the Founding of the Umayyad Monarchy’ in Manuela Martín (ed.). The Formation of Al-Andalus, Part 1: History and Society, 51-85 (Aldershot: Ashgate).

Bibliogr aphy

353

Cruz Villalón, María. 1982. ‘Los materiales de la escultura visigoda de Mérida’. Norba. Revista de arte, geografía e historia 3, 7-14. Cruz Villalón, María. 1985. Mérida visigoda. La escultura arquitectónica y litúrgica (Badajoz: Diputación Provincial de Badajoz). da Cruz, Mário. 2009. ‘Fragmentos de vidro. Fragmentos da memória. Aproximação a actividade vidreira em Bracara Augusta’. Cultura, Espaço e Memória 2, 83-94. Cuomo, Seraf ina. 2008. ‘Ancient Written Sources for Engineering and Technology’ in John Peter Oleson (ed.). Engineering and Technology in the Classical World, 15-34 (Oxford: Oxford University Press). Curchin, Leonard. 2004. The Romanization of Central Spain: Complexity, Diversity and Change in a Provincial Hinterland (London: Routledge). Curta, Florin. 2013. ‘The Elephant in the Room: A Reply to Sebastian Brather’. Ephemeris Napocensis 23, 163-174. de Almeida, Fernando. 1962. Arte Visigótica em Portugal (Lisbon: Alfa). de Juan Ares, Jorge; María del Mar Gallego García; and Javier García Gómez. 2009. ‘La cultura material de la Vega Baja’ in Alfonso García and Diego Perís (eds.). La Vega Baja de Toledo, 115-47 (Toledo: Toletum Visigodo). de Juan Ares, Jorge; and Nadine Schibille. 2017a. ‘La Hispania antigua y medieval a través del vidrio: la aportación de la arqueometría’. Boletín de la Sociedad Española de Cerámica y Vidrio 56.5, 195-204. de Juan Ares, Jorge; and Nadine Schibille. 2017b. ‘Glass Import and Production in Hispania During the Early Medieval Period: The Glass From Ciudad de Vascos (Toledo)’. Plos One 12.7, 1-19. de los Ríos Serrano, José Amador. 1861. El arte latino-bizantino en España y las coronas visigodas de Guarrazar: Ensayo histórico crítico (Madrid: Imprenta Nacional). de los Ríos Serrano, José Amador. 1877. Monumentos latino-bizantinos de Mérida (Madrid: José Gil Dorregaray). De Man, Adriaan. 2007a. ‘A muralha tardia de Conimbriga’ in Antonio Rodríguez Colmenero and Isabel Rodá de Llanza (eds.). Murallas de ciudades romanas en el occidente del imperio. Lucus Augusti como paradigma, 699-714 (Lugo: Diputación Provincial de Lugo). De Man, Adriaan. 2007b. ‘The late urban fort of the Bico da Muralha, Conimbriga (Portugal)’. Mediterranean Archaeology and Archaeometry 7, 3-14. De Man, Adriaan. 2011. Defesas Urbanas Tardias da Lusitânia. Stvdia Lusitania 6 (Mérida: MNAR). de Miguel Ibáñez, María Paz. 2007. ‘La maqbara de la Plaza del Castillo (Pamplona, Navarra): avance del estudio ostoearqueológico’ in Philippe Sénac (ed.). Villes et Campagnes de Tarraconaise et d’al-Andalus (VIe-XIe siècle). La transition, 183-97. Études Médiévales Ibériques (Toulouse: CNRS). de Miguel Ibáñez, María Paz. 2016. La maqbara de Pamplona (s. VIII). Aportes de la osteoarqueología al conocimiento de la islamización de la Marca Superior. PhD thesis (University of Alicante). de Miguel Ibáñez, María Paz; Lara Fontecha Martínez; Neskuts Izagirre Arribalzaga; and Concepción de la Rúa Vaca. 2016. ‘Paleopatología, ADN y diferenciación social en la “maqbara” de Pamplona. Límites y posibilidades’ in Juan Antonio Quirós Castillo (ed.). Demografía, paleopatologías y desigualdad social en el noroeste peninsular en época medieval, 163-82. Documentos de arqueología medieval 10 (Bilbao: Universidad del País Vasco) del Amo Guinovart, María Dolores. 1981. Estudio crítico de la necrópolis paleocristiana de Tarragona (Tarragona: Institut d’Estudis Tarraconenses Ramón Berenguer IV). Delgado Molina, Pedro Ángel. 2006. ‘Excavación de un área cristiana en época andalusí. Intervención arqueológica llevada a cabo en la calle B. Arias Montano c.v. a Avda. Reina Sofía (Mérida)’. Mérida. Excavaciones arqueológicas 2003, 9, 285-312.

354 

The Iberian Peninsul a bet ween 300 and 850

Delgado Valero, Clara. 1999. ‘La estructura urbana de Toledo en época islámica’ in Regreso a Tulaytula. Guía del Toledo Islámico: siglos VIII-XI, 11-160 (Toledo: Junta de Comunidades de Castilla-La Mancha). DeMoule, Jean-Paul. 2007. ‘Introduction’ in Jean-Paul DeMoule (ed.). L’archéologie préventive dans la monde, 7-13 (Paris: La Découverte). Desprat, Sebastién; María Fernanda Sánchez Goñi; and Michel-Françoise Loutre. 2003. ‘Revealing Climatic Variability of the Last Three Millennia in Northwestern Ibera using Pollen Influx Data’. Earth and Planetary Science Letters 213, 63-78. Dey, Hendrik. 2014. The Afterlife of the Roman City: Architecture and Ceremony in Late Antiquity and the Early Middle Ages (Cambridge: Cambridge University Press). Diarte Blasco, Pilar. 2014. ‘Un camino sin retorno: la desarticulación de la ciudad clásica en la antigüedad tardía (ss. IV-V d. C)’ in Manuel Martín-Bueno and Carlos Sáenz Preciado (eds.). Modelos edilicios y prototipos en la monumentalización de las ciudades de Hispania, 97-106. Monografías Arqueológicas 49 (Zaragoza: Universidad de Zaragoza). Diarte Blasco, Pilar. 2016. ‘New Thinking in Old Landscapes: Discoveries, Research and Approaches for Late Antique and Early Medieval Iberia’. Medieval Settlement Research 31, 1-14. Díaz Álvarez, Inés. 2003. ‘Intervención arqueológica en Castro Ventosa: limpieza de las murallas de 1988’ in José Antonio Balboa de Paz, Inés Díaz Álvarez and Vicente Fernández Vázquez (eds.). Actas de las jornadas sobre Castro Ventosa, 35-48 (Cacabelos: Ayuntamiento de Cacabelos). Díaz Álvarez, Inés; and Alberto Garín. 1999. ‘Estudio de los materiales arqueológicos de Castro Ventosa’. Estudios Bercianos 25, 74-95. Díaz Martínez, Pablo. 2004. ‘Añación monetaria y organización administrativa en la Gallaecia’. Zephyrus, 367-375. Díaz Martínez, Pablo. 2011. El Reino Suevo (411-585) (Madrid: Akal). Díaz Martínez, Pablo; Celia Martínez Maza; and Francisco Javier Sanz Huesma. 2007. Hispania tardoantigua y visigoda (Madrid: Istmo). Domergue, Claude. 1987. Catalogue des mines et des fonderies antiques de la Péninsule Ibérique. Serie Archeologie 7 (Madrid: Casa de Velázquez). Domergue, Claude. 1990. Les mines de la péninsule Ibérique dans l’Antiquité Romaine. Collection de l’École Française de Rome 127 (Rome: École Française de Rome). Domínguez Bolaños, Alonso. 1993. ‘Avance de las excavaciones arqueológicas en el Castro de San Esteban, Muelas del Pan’. Anuario del Instituto de Estudios Zamoranos Florián de Ocampo, 201-209. Domínguez Bolaños, Alonso; and Jaime Nuño González. 2014. ‘Dos viviendas del siglo VI sin noticias de élites locales en el Cristo de San Esteban (Muelas del Pan, Zamora)’ in Raúl Catalán Ramos, Patricia Fuentes Melgar and José Carlos Sastre Blanco (eds.). Las fortificaciones en la tardoantigüedad. Élites y articulación del territorio (siglos V-VIII d.C), 275-96 (Madrid: La Ergástula). Dupré Raventós, Xavier. 1990. L’anfiteatre romà de Tarragona, la basílica visigótica i l’església románica (Tarragona: TED’A) Dupré Raventós, Xavier; and Josep Anton Remolà Vallverdú (eds.). 2000. Sordes urbis: la eliminación de residuos en la ciudad romana: actas de la Reunión de Roma, 15-16 de noviembre de 1996 (Rome: L’erma di Bretschneider). Ebel-Zepezauer, Wolfgang. 2000. Studien zur Archäologie der Westgoten vom 5.-7. Jh. N. Chr. Iberia Archaeologica 2 (Mainz: Philipp von Zabern). Edmondson, Jonathan. 1989. ‘Mining in the later Roman Empire and beyond: Continuity or disruption?’. Journal of Roman Studies 79, 84-102. Egea Vivancos, Alejandro. 2002. ‘Características principales del sistema de captación, abastecimiento, distribución y evacuación de aguas de Carthago Nova’. Empúries 53, 13-28.

Bibliogr aphy

355

Eger, Christoph. 2007. ‘Guarrazar (Provinz Toledo). Bericht zu den Unterschungen 2002 bis 2005’. Madrider Mitteilungen 48, 267-305. Eiroa Rodríguez, Jorge. 2012. ‘Past and Present of the Archaeology of the Alquerías’. Imago Temporis. Medium Aevum 6, 49-72. Elkin, Mike. 2009. ‘The Veleia affair’. Archaeology 62.5. Elorza Guinea, Juan Carlos. 1972. ‘A propósito de la muralla de Iruña’. Estudios de Arqueología Alavesa 5, 183-94. Elvira Barba, Miguel Ángel; Luis Caballero Zoreda; and Javier Arce Martínez. 1998. ‘El edificio octogonal de Valdetorres de Jarama (Madrid)’ in Ramón Teja Casuso and Cesáreo Pérez González (eds.). Congreso Internacional la Hispania de Teodosio (2 vols),vol. 2, 321-38 (Valladolid: Junta de Castilla y León). Escalona Monge, Juio; and Andrew Reynolds. 2011. Scale and Scale Change in the Early Middle Ages: Exploring Landscape, Local Society, and the World Beyond (Turnhout: Brepols). Escalona Monge, Julio. 2002. Sociedad y territorio en la Alta Edad Media Castellana: La Formación del Alfoz de Lara. BAR IS 1079 (Oxford: Archaeopress). Escudero Escudero, Francisco; and María Pilar Galve Izquierdo. 2007. ‘Edificios de espectáculos’ in Francisco Beltrán Lloris (ed.). Zaragoza. Colonia Caesar Augusta, 57-70 (Rome: L’erma di Bretschneider). Escudero, Francisco; José Antonio Hernández Vera and Julio Núñez Marcén. 2007. ‘Arquitectura oficial’ in Francisco Beltrán Lloris (ed.). Zaragoza. Colonia Caesar Augusta, 43-56 (Rome: L’erma di Bretschneider). Esmonde Cleary, Simon. 2013. The Roman West, AD 200-500: An Archaeological Study (Cambridge: Cambridge University Press). Espinar Moreno, Manuel. 1991. ‘Del urbanismo musulmán al urbanismo cristiano II: Andalucía oriental’ in Mikel de Epalza (coord.). Simposio internacional sobre la ciudad islámica, 203-52 (Zaragoza: Institución Fernando el Católico). Esteban Molina, Jorge. 2007. La villa romana y la necrópolis visigoda de Santa Lucía, Aguilafuente (Segovia). Nuevas aportaciones para su estudio (Segovia: Ayuntamiento de Aguilafuente). Estremera Portela, Maríá, Inés Centeno Cea, and Javier Quintana López (eds.). 2006. Arqueología urbana en Ávila: La intervención en los solares del Palacio de Don Gaspar del Águila y Bracamonte (Valladolid: Junta de Castilla y León). Étienne, Robert; and Mayet, Françoise. 2000. Le vin hispanique, Trois clés de l’économie de l’Hispanie romaine, 1 (Paris: Boccard). Étienne, Robert; and Françoise Mayet. 2004. L’huile hispanique (Paris: Boccard). Fabián García, José Francisco; Manuel Santonja Gómez; Antonio Fernández Moyano, and Nicolás Benet. 1985. ‘Los poblados hispano-visigodos de Cañal, Pelayos (Salamanca)’. Congreso de Arqueología Medieval Española (2 vols), vol. I, 187-201 (Huesca: Diputación General de Aragón). Faro Carballa, José Antonio; María García-Berbena Unzu; and Mercedes Unzu Urmeneta. 2008. ‘Pamplona y el Islam. Nuevos testimonios arqueológicos’. Trabajos de Arqueología de Navarra 20, 229-84. Fear, Andrew. 1996. Rome and Baetica: Urbanization in Southern Spain c. 50 BC-AD 150 (Oxford: Oxford University Press). Fehring, Günter. 2009. Die Archäologie des Mittelalters. Eine Einführung. (Darmstadt: WBG). Fenwick, Corisande. 2013. ‘From Africa to Ifriqiya: Settlement and Society in Early Medieval North Africa (650-800)’. Al-Masaq 25.1, 9-33. Fernández Colmenero, Alicia; and Celso Rodríguez Cao. 2012. ‘Anastilosis virtual de A domus do Mitreo do Lvcvs Avgvsti’. Virtual Archaeology Review 3.5, 104-08.

356 

The Iberian Peninsul a bet ween 300 and 850

Fernández Delgado, Aitor; Javier Martínez Jiménez; and Carlos Tejerizo García. 2013 ‘New and Old Elites in the Visigothic kingdom (550-650 AD)’ in Elsbeth Van der Wilt and Javier Martínez Jiménez (eds.). Tough Times: The Archaeology of Crisis and Recovery: Proceedings of the GAO Annual Conferences 2010 and 2011, 161-70. BAR IS 2478 (Oxford: Archaeporess). Fernández Fernández, Adolfo. 2014. El comercio tardoantiguo (ss. IV-VII) en el Noroeste peninsular a través del registro cerámico de la ría de Vigo. Roman and Late Antique Mediterranean Pottery 5 (Oxford: Archaeopress). Fernández Fernández, Adolfo; and Rui Morais. 2012. ‘Terra Sigillata Bracarense Tardía (TSBT)’ in Darío Bernal Casasola and Albert Ribera Lacomba (eds.). Cerámicas hispanorromanas II. Producciones regionales, 131-75. Monografías Historia y Arte (Cádiz: Universidad de Cádiz). Fernández Martínez, Concepción; and Joan Gómez Pallarés. 2001. ‘Hermenegildo, ¿para siempre en Sevilla? Una nueva interpretación de IHC n. 76 = ILCV n.50’. Gerión 19, 629-58. Fernández Matallana, Francisco. 2002. ‘Vidrios procedentes de la excavación de urgencia de Begastri (Cuesta del Río). Diciembre 1995’. Memorias de Arqueología 10, 421-427. Fernández Mier, Margarita. 1999. Génesis del territorio en la Edad Media: arqueología del paisaje y evolución histórica en la montaña asturiana. (Oviedo: Universidad de Oviedo). Fernández Mier, Margarita; Jesús Fernández Fernández; Pablo Alonso González; José Antonio López Sáez; Sebastián Pérez Díaz; and Begoña Hernández Beloqui. 2014. ‘The Investigation of Currently Inhabited Villages of Medieval Origin: Agrarian Archaeology in Asturias (Spain)’. Quaternary International 346, 41-55. Fernández Mier, Margarita; Carlos Tejerizo García; and Patricia Aparicio Martínez. 2014. ‘El territorio de Cea entre la tardo-romanidad y la Alta Edad Media’ in Raúl Catalán Ramos, Patrcia Fuentes Melgar and José Carlos Sastre Blanco (eds.). Las fortificaciones en la tardoantigüedad. Élites y articulación del territorio (siglos V-VIII d.C), 159-77 (Madrid: La Ergástula). Fernández Ochoa, Carmen. 2006. ‘The Army and the Urban Walls in Late Roman Spain’ in Ángel Morillo Cebrián and Joaquín Aurrecoechea Fernández (eds.). The Roman Army in Hispania: An Archaeological Guide, 189-210 (León: Universidad de León). Fernández Ochoa, Carmen; and Fernando Gil Sendino. 2007. ‘El recinto amurallado de Gijón. Origen y permanencia hasta la Edad Media’ in Antonio Rodríguez Colmenero and Isabel Rodá de Llanza (eds.). Murallas de ciudades romanas en el occidente del imperio. Lucus Augusti como paradigma, 401-16 (Lugo: Diputación Provincial de Lugo). Fernández Ochoa, Carmen; and Ángel Morillo Cebrián. 2002. ‘Entre el prestigio y la defensa: la problemática estratégico-defensiva de las murallas tardorromanas en Hispania’ in Ángel Morillo Cebrián (ed.). Arqueología militar romana en Hispania, 577-90. Anejos a Gladius 5 (Madrid: CSIC). Fernández Ochoa, Carmen; Virginia García-Entero, and Fernando Gil Sendino (eds.). 2008. Las villae tardorromanas en el occidente del Imperio: arquitectura y función. IV Coloqio Internacional de Arqueología en Gijón (Gijón: Trea). Fernández Ochoa, Carmen; Fernando Gil Sendino; and Almudena Orejas Saco. 2004. ‘La villa de Veranes. El complejo rural tardorromano y propuesta de estudio del territorio’. AEspA 77, 197-219. Fernández Ubiña, José. 1982. La crisis del siglo III y el fin del mundo antiguo (Madrid: Akal). Fernández-Galiano, Dimas. 1998. ‘La villa romana de Carranque’ in En el año de Trajano. Hispania. El legado de Roma. Catálogo de la exposición, 437-440 (Zaragoza: La Lonja). Fernández-Galiano, Dimas. 2001. Carranque. Esplendor de la Hispania de Teodosio. (Barcelona: Museo de Arqueología de Cataluña). Ferreira Fernandes, Isabel Cristina. 2005. ‘Arqueologia Medieval em Portugal: 25 anos de investigação’. Portugalia 26, 149-73. Ferreiro, Alberto. 2006. The Visigoths in Gaul and Iberia: A Supplemental Bibliography 1983-2004 (Leiden: Brill).

Bibliogr aphy

357

Ferreiro, Alberto. 2008. The Visigoths in Gaul and Iberia (Update): A Supplemental Bibliography 2004-2006 (Leiden: Brill). Ferreiro, Alberto. 2011. The Visigoths in Gaul and Iberia (Update): A Supplemental Bibliography 2007-2009 (Leiden: Brill). Ferreiro, Alberto. 2014. The Visigoths in Gaul and Iberia (Update): A Supplemental Bibliography 2010-2012 (Leiden: Brill). Filloy Nieva, Idoia; and Eliseo Gil Zubillaga. 2007. ‘Vida cotidiana al abrigo de las murallas. Novedades de la investigación sobre el recinto amurallado tardorromano de Veleia (Iruña de Oca, Álava, País Vasco)’ in Antonio Rodríguez Colmenero and Isabel Rodá de Llanza (eds.). Murallas de ciudades romanas en el occidente del imperio. Lucus Augusti como paradigma, 467-82 (Lugo: Diputación Provincial de Lugo). Fita, Fidel. 1902. ‘Inscripciones visigóticas y suévicas de Dueñas, Baños de Cerrato, Vairáom, Baños de Bande y San Pedro de Rocas’. Boletín de la Real Academia de la Historia 41, 476-511. Fontaine, Jacques. 1972-74. ‘Le distique du Chrismon de Quiroga. Sources littéraires et contexte spirituel’ AEspA 45-46, 557-585. Fontes, Luis. 1995. ‘A igreja sueva de Dume (Braga)’ in IV RACH, 415-28. Fontes, Luis; Manuela Martins; Maria do Carmo Ribeiro; and Helena Paula Carvalho. 2010. ‘A cidade de Braga o seu território nos séculos V-VIII’ in Alfonso García, Ricardo Izquierdo, Lauro Olmo and Diego Perís (eds.). Espacios urbanos en el occidente mediterráneo, ss. VI-VIII, 255-62 (Toledo: Toletum Visigodo). Fornell Muñoz, Alejandro. 2005. Las villae romanas en la Andalucía Mediterránea y del Estrecho (Jaén: Universidad de Jaén). Foss, Clive. 1979. Ephesus after Antiquity: A Late Antique, Byzantine and Turkish City (Cambridge: Cambridge University Press). Franco Moreno, Bruno. 2004. ‘Territorio y poblamiento en la Kura de Marida durante el emirato Omeya (siglos VIII-X/II-IV)’. Espacio, tiempo y forma (Series III) 17, 167-84. Freestone, Ian. 2006. ‘Glass Production in Late Antiquity and the Early Islamic Period: a Geochemical Perspective’ in Marino Maggetti and Bruno Messiga (eds.). Geomaterials in Cultural Heritage, 201-16. Special Publications 257 (London: Geological Society). Freestone, Ian. 2015. ‘The Recycling and Reuse of Roman Glass: Analytical Approaches’. Journal of Glass Studies 51, 29-40. Fuentes Domínguez, Ángel. 1989. La Necrópolis tardorromana de Albalate de las Nogueras (Cuenca) y el problema de las denominadas ‘Necrópolis del Duero’. Arqueología Conquense 10 (Cuenca: Diputación Provincial de Cuenca). Fuentes Domínguez, Ángel. 2006. ‘La antigüedad tardía en Castilla-La Mancha’ in Ángel Fuentes Domínguez (coord.). Castilla-La Mancha en época romana y antigüedad tardía, 176-213 (Ciudad Real: Universidad de Castilla-La Mancha). Fuentes Domínguez, Ángel. 2007. ‘Vidrio de la Antigüedad tardía (ss.V-X). Cuestiones de fabricación y comercialización. Problemas de identificación’ in Vidrio islámico en Al-Andalus, 13-36 (Madrid: Fundación Centro Nacional del Vidrio). Fuentes Domínguez, Ángel; and Rosario Escobar Fernández. 2013. ‘El edif icio de la curia en el foro de Valeria’ in Pedro Mateos Cruz, José Miguel Noguera Celdrán, and Joaquín Ruiz de Arbulo (eds.). Las sedes de los ordines decurionum en Hispania. Análisis arquitectónico y modelo tipológico, 215-32. AAEspA 67 (Mérida: CSIC). Fuertes Santos, María del Camino; and Rafael Hidalgo Prieto. 2003. ‘Cerámicas tardorromanas y altomedievales de Córdoba’ in Luis Caballero, Pedro Mateos and Manuel Retuerce (eds.). Cerámicas tardorromanas y altomedievales en la Península Ibérica, 505-40. AAEspA 28 (Mérida: CSIC).

358 

The Iberian Peninsul a bet ween 300 and 850

Galve Izquierdo, María Pilar; and Antonio Mostalac Carrillo. 2007. ‘Las necrópolis’ in Xavier Dupré Raventós (ed.). Zaragoza. Colonia Caesar Augusta, 85-96. Capitales provinciales de Hispania 4 (Rome: L’erma di Bretschneider). García de Castro, César. 1995. Arqueología Cristiana de la alta Edad Media en Asturias (Oviedo: Real Instituto de Estudios Asturianos). García de Castro, César; and Sergio Ríos González. 2016. ‘El origen de Oviedo’. Anejos de Nailos 3, 31-119. García Díaz, Paloma; Carmen Fernández Ochoa; Alexandra Uscatescu. 1992. ‘Gijón en el período tardoantiguo. Cerámicas importadas de las excavaciones de Cimadevilla’. AEspA 65, 105-49. García García, Marcos. 2016. ‘Some Remarks on the Provisioning of Animal Products to Urban Centres in Medieval Islamic Iberia: The Cases of Madinat Ilbirah (Granada) and Cercadilla (Cordoba)’. Quaternary International, online edition. García García, Miguel Ángel. 2012. ‘Sarcófagos romanos decorados del siglo IV en el territorio andaluz: enfoques y problemática vigente’. Spal 21, 183-93. García Guinea, Miguel Ángel; Joaquín González Echegaray; and Benito Madariaga de la Campa. 1963. ’El Castellar. Villajimena (Palencia)’. Publicaciones de la Institución Tello Téllez de Meneses 23, 124-58. García Guinea, Miguel Ángel; Joaquín González Echegaray; and José Antonio San Martín Ruiz. 1968. ’Excavaciones en Monte Cildá, Olleros de Pisuerga (Palencia): campañas de 1963-1965: memoria redactada’. Publicaciones de la Institución Tello Téllez de Meneses 26, 1-68. García Guinea, Miguel Ángel; Joaquín González Echegaray; and José Antonio San Martín Ruiz. 1973. ’Excavaciones de Monte Cildá, Olleros de Pisuerga (Palencia): campaña de 1966 a 1969’. Publicaciones de la Institución Tello Téllez de Meneses 34, 1-95. García Jiménez, Gustavo; and David Vivó i Codina. 2003. ‘Sant Julià de Ramis y Puig Rom: dos ejemplos de yacimientos con armamento y equipamiento militar visigodo en el noroeste peninsular’. Gladius 23, 161-190. García Merino, Carmen. 1975. Población y poblamiento en Hispania romana. El Conventus Cluniensis (Valladolid: Universidad de Valladolid). García Merino, Carmen. 2000. ‘Las raices históricas de la sede episcopal oxomense: aproximación a la etapa tardoantigua de Uxama’ in Teófilo Portillo Capilla (ed.). I Semana de estudios históricos de la Diócesis de Osma-Soria: 15-17 de septiembre de 1997, 179-96 (Soria: Diputación Provincial de Soria). García Moreno, Luis. 2008. Leovigildo, unidad y diversidad de un reinado (Madrid: Real Academia de la Historia). García Vargas, Enrique. 2012. ‘La Sevilla tardoantigua. Diez años después (2000-2010)’ in José Beltrán Fortes and Oliva Rodríguez Gutiérrez (eds.). Hispaniae urbes. Investigaciones arqueológicas en ciudades históricas, 881-926 (Seville: Universidad de Sevilla). García Vargas, Enrique; and Jacobo Vázquez Paz. 2006. ‘Sevilla y el comercio transmarino en el bajo imperio y en la antigüedad tardía: el testimonio de la arqueología’, in Álvaro Jiménez Sancho (ed.). La catedral en la ciudad II. De Isidoro a Abd Ar Rahman, 44-99 (Seville: Catedral de Sevilla). García Vargas, Entique; and Jacobo Vázquez Paz. 2013. ‘Rural Population of Farmlands South of the Guadalquivir valley in Late Antiquity (Fourth-Sixth Century AD)’ in Rosa García-Gasco, Sergio González Sánchez,and David Hernández de la Fuente (eds.). The Theodosian Age (A.D. 379-455): Power, Place, Belief and Learning at the End of the Western Empire, 99-122. BAR IS 2493 (Oxford: Archaeopress). García Llinares, Gemma; Antonio Moro García; and Francesc Tuset Bertrán. 2009. La seu episcopal d’Ègara. Arqueologia d’un conjunt cristià del segle IV al IX. Documenta 8 (Tarragona: ICAC). García-Entero, Virginia. 2005. Los balnea domésticos – ámbito rural y urbano- en la Hispania romana. AAEspA 38 (Madrid: CSIC).

Bibliogr aphy

359

García-Entero, Virginia; María del Mar Salán Asensio; and Sergio Vidal Álvarez. 2008. ‘El marmor en el yacimiento de Carranque (Toledo). Algunas consideraciones sobre las marcas de herramientas’, in Trinidad Nogales Basarrate, Julia Beltrán de Heredia (eds.). Marmora Hispana: explotación y uso de los materiales pétreos en la Hispania romana, 197-211 (Rome: L’erma di Bretschneider). García-Sampedro, Cristina. 2009. ‘Origen y fundación de la ciudad de Oviedo’ Gallaecia 28, 167-85. Geary, Patrick. 1983. ‘Ethnic Identity as Situational Construct in the Early Middle Ages’. Mitteilungen der Anthropologischen Gesellschaft in Wien 113, 15-26. Gelichi, Sauro; and Richard Hodges (eds.). 2015. New Directions in Early Medieval European Archaeology: Spain and Italy Compared. Essays for Riccardo Francovich (Turnhout: Brepols). Gibbon, Edward. 1999 [1776-1789]. The Decline and Fall of the Roman Empire (London: Wordsworth). Gillet, Andrew. 2002. ‘Introduction: Ethnicity, History and Methodology’ in Andrew Gillet (ed.). On Barbarian Identity: Critical Approaches to Ethnicity in the Early Medieval Ages, 1-18 (Turnhout: Brepols). Gimeno Pascual, Helena; and Javierdel Hoyo Calleja. 2012. ‘La epigrafía medieval asturiana. Los epígrafes de la iglesia de San Juan de Santianes de Pravia’ in Luis Caballero Zoreda, Pedro Mateos Cruz and César García de Castro (eds.). Asturias entre visigodos y mozárabes. Visigodos y omeyas VI, 13-25. AAEspA 63 (Madrid: CSIC). Glick, Thomas. 1970. Irrigation and Society in Medieval Valencia (Cambridge, Mass: Harvard). Glick, Thomas. 1995. From Muslim Fortress to Christian Castle: Social and Cultural Change in Medieval Spain (Manchester: Manchester University Press). Glick, Thomas. 2005. Islamic and Christian Spain in the Early Middle Ages. The Medieval and Early Modern Iberian World 27 (Leiden: Brill). Glick, Thomas; and Helena Kirchner Granell. 2000. ‘Hydraulic Systems and Technologies in Islamic Spain: History and Archaeology’ in Paolo Squatriti (ed.). Working with Water in Medieval Europe. Technology and Resource-use, 267-330 (Leiden: Brill). Godoy Fernández, Cristina. 1995. Arqueología y liturgia. Iglesias Hispánicas (siglos IV al VIII) (Barcelona: Universidad de Barcelona). Goffart, Walter. 1980. Barbarians and Romans: The Techniques of Accomodation (Princeton: Princeton University Press). Goffart, Walter. 2002. ‘Does the Distant Past Impinge on the Invasion Age Germans?’ in Andrew Gillet (ed.). On Barbarian Identity: Critical Approaches to Ethnicity in the Early Middle Ages, 21-38 (Turnhout: Brepols). Gómez Becerra, Antonio. 1995. ‘Almuñécar en el tránsito de la Antigüedad a la Edad Media’. Florentia Iliberritana 6, 175-201. Gómez Tubío, Blanca; María Ángeles Ontalba Salamanca; Inés Ortega Feliu; María ÁngelesRespaldiza Galilsteo; Fernando Amores Carredano; and Daniel González Acuña. 2006. ‘PIXE-PIGE Analysis of Late Roman Glass Fragments’. Nuclear Instruments and Methods in Physics Research B 249, 616-21. Gómez-Moreno, Manuel. 1919. Iglesias mozárabes. Arte español de los siglos IX a XI (Madrid: Centro de Estudios Históricos). Gómez-Moreno, Manuel. 1966. Primicias de arte cristiano español (Madrid: Archivo Español de Arte). González Blanco, Antonio. 2007. ‘Las murallas de Begastri (Cehegín, Murcia)’ in Antonio Rodríguez Colmenero and Isabel Rodá de Llanza (eds.). Murallas de ciudades romanas en el occidente del imperio. Lucus Augusti como paradigma, 549-66 (Lugo: Diputación Provincial de Lugo). González Fernández, Enrique; and María Covadonga Carreño Gascón. 2007. ‘Las puertas romanas de la muralla de Lugo’ in Antonio Rodríguez Colmenero and Isabel Rodá de Llanza (eds.). Murallas de ciudades romanas en el occidente del imperio. Lucus Augusti como paradigma, 283-312 (Lugo: Diputación Provincial de Lugo).

360 

The Iberian Peninsul a bet ween 300 and 850

González Villaescusa, Ricardo. 2001. El mundo funerario romano en el País Valenciano: monumentos funerarios y sepulturas entre los siglos I a. de C. – VII d. de C. (Madrid: Casa de Velázquez). González Villaescusa, Ricardo; Rafael González Fernández; and Francisco Fernández Matallana. 2010. ‘Mula: el final de una ciudad de la cora de Tudmîr’. Pyrenae 41.2, 81-119. Gonzalo González, José María. 2007. El Cerro del Castillo, Bernardos (Segovia). Un yacimiento arqueológico singular en la provincia de Segovia durante la Antigüedad Tardía (Segovia: Obra Social y Cultural de Caja Segovia). Gorges, Jéan Gérard. 1979. Les villas hispano-romaines. Inventaire et problématique archéologiques (Paris: Baccard). Gorostidi, Diana. 2013. ‘L’epigrafia paleocristiana de Tarraco. Característiques generals i estat de la qüestió’ in José María Macias and Andreu Muñoz (eds.). Tarraco Christiana Ciuitas, 43-68 (Tarragona: ICAC). Gozalbes Cravioto, Enrique. 2001. ‘De la civitas hispano-romana a la madina andalusí’ in Antonio Torremocha Silva and Virgilio Martínez Enamorado (eds.). II Congreso internacional de la ciudad en al-Ándalus y el Magreb, 641-56 (Cádiz: El Legado Andalusí). Gozalo Viejo, Francisco; José María Gonzalo González; and Juan Francisco Blanco García. 2013. ‘El Cerro Tormejón (Armuña, Segovia). Análisis de sus materiales cerámicos tardoantiguos’. CuPAUAM 39, 151-182. Grau Sologestoa, Idoia. 2009. ‘Ganadería en la Alta Edad Media. Estudio comparativo de los yacimientos alaveses de Zornoztegi, Zaballa y Salvatierra-Agurain’. Munibe 60, 253-280. Grau Sologestoa, Idoia. 2013. ‘Zooarqueología de las aldeas y villas en el País Vasco: actividad ganadera y usos animales en la Edad media (siglos V-XV)’. Debates de Arqueología Medieval 3, 99-116. Grau Sologestoa, Idoia. 2015a. The Zooarchaeology of Medieval Álava in its Iberian Context. BAR IS 2769 (Oxford: Archaeopress). Grau Sologestoa, Idoia. 2015b. ‘Livestock Management in Spain from Roman to Post-Medieval Times: A Biometrical Analysis of Cattle, Sheep/Goat and Pig’. Journal of Archaeological Science 54, 123-134. Grierson, Philip. 1953. ‘Visigothic Metrology’. The Numismatic Chronicle 13.43, 74-87. Grützner, Christoph; Klaus Reichter; and Pablo Silva. 2010. ‘Comparing Semi-Quantitative Logic Trees for Archaeoseismology and Palaeoseismology: The Baelo Claudia (Southern Spain) Case Study’ in Manuel Sintubin, Iain Stewart, Tina Niemi and Erhan Altunel (eds.). Ancient Earthquakes, 129-43. The Geological Society of America Special Paper 471 (Boulder: Geological Society of America). Guerra, María Filomena; Thomas Calligaro; and AliciaPerea Caveda. 2007. ‘The Treasure of Guarrazar: Tracing the Gold Supplies in the Visigothic Iberian Peninsula’. Archaeometry 49, 53-74. Guichard, Pierre. 1976. Al-Andalus. Estructura antropológica de una sociedad islámica en Occidente (Barcelona: Barral). Guichard, Pierre. 2002. De la expansión árabe a la reconquista: Esplendor y fragilidad de al-Andalus (Granada: El Legado Andalusí). Guichard, Pierre. 2007. ‘Villes et campagnes aux premieres siècles de l’Islam dans le Sharq al-Andalus’ in Philippe Sénac (ed.). Villes et Campagnes de Tarraconaise et d’al-Andalus (VIe-XIe siècle) : la transition, 263-73. Études Médiévales Ibériques (Toulouse: CNRS). Guidetti, Mattia. 2014. ‘Churches Attracting Mosques: Religious Architecture in Early Islamic Syria’, in Mohammad Gharipour (ed.). Sacred Precincts: The Religious Architecture of NonMuslim Communities in Islamic Societies, 11-27 (Leiden: Brill). Gurrea, Ramon; and Jordi Fernández. 1995. ‘S’Argamassa’ in Enciclopèdia d’Eivissa I Formentera, 210-1 (Ibiza: Consell d’Eivissa). Gurriarán Daza, Pedro. 2008. ‘Una arquitectura para el califato: poder y construcción en AlAndalus durante el siglo X’. Anales de Arqueología Cordobesa 19, 261-76.

Bibliogr aphy

361

Gurt Esparraguera, Josep Maria; and Pilar Diarte Blasco. 2012. ‘La basílica de santa Leocadia y el final de uso del circo romano de Toledo: Una nueva interpretación’. Zephyrus 69, 149-63. Gurt Esparraguera, Josep Maria; and Cristina Godoy Fernández. 2000. ‘Barcino, de sede imperial a urbs regia en época visigoda’ in Gisela Ripoll López and Josep Maria Gurt Esparraguera (eds.). Sedes Regiae (ann. 400-800), 425-66 (Barcelona: Real Academia de las Buenas Letras). Gurt Esparraguera, Josep Maria; and Albert Ribera Lacomba, (eds.). 2005. VI Reunció d’Arqueologia Cristiana Hispànica. Les ciutats tardoantiques d’Hispania: Cristianització i topografia (Barcelona: Institut d’Estudis Catalans). Gurt Esparraguera, Josep María; and Nuria Tena, (eds.). 1995. IV Reunció d’Arqueologia Cristiana Hispànica (Barcelona: Institut d’Estudis Catalans). Gurt Esparraguera, Josep Maria; and Nuria Tena, (eds.). 2000. V Reunció d’Arqueologia Cristiana Hispànica (Barcelona: Institut d’Estudis Catalans). Gutiérrez González, José Avelino. 2013. ‘Poblamiento de los siglos VII-VIII y conquista musulmana del antiguo Conventus Asturum’ in Xavier Ballestín Navarro and Ernesto Pastor Díaz (eds.).Lo que vino de Oriente. Horizontes, praxis y dimensión material de los sistemas de dominación fiscal en Al-Andalus (ss. VII-IX), 102-21. BAR IS 2525 (Oxford: Archaeopress). Gutiérrez González, José Avelino. 2014. ‘Fortificaciones tardoantiguas y visigodas en el norte peninsular (ss. V-VIII)’ in Raúl Catalán Ramos, Patricia Fuentes Melgar and José Carlos Sastre Blanco (eds.). Las fortificaciones en la tardoantigüedad. Élites y articulación del territorio (siglos V-VIII d.C.), 191-214 (Madrid: La Ergástula). Gutiérrez González, José Avelino; and Fernando Miguel Hernández. 2009. ‘La cerámica altomedieval en León: producciones locales y andalusíes de Puerta Obispo’ in Juan Zozaya Stabel-Hansen, Manuel Retuerce Velasco, Miguel Ángel Hervás Herrera and Antonio de Juan García (eds.).Actas del VIII Congreso Internacional de Cerámica Medieval, 443-62 (Ciudad Real: Universidad de Castilla-La Mancha). Gutiérrez Lloret, Sonia. 1993. ‘De la civitas a la madīna: destrucción y formación de la ciudad en el sureste de Al-Andalus. El debate arqueológico’ in Sociedades en transición: actas / IV Congreso de Arqueología Medieval Española, Alicante, 4-9 de octubre, 13-35 (Madrid: Asociación Española de Arqueología Medieval). Gutiérrez Lloret, Sonia. 1996. La cora de Tudmir. De la antigüedad tardía al mundo islámico (Alicante: Universidad de Alicante). Gutiérrez Lloret, Sonia. 2011. ‘El reconocimiento arqueológico de la islamización: una mirada desde al-Andalus’ in 711. Arqueología e Historia entre dos mundos, 189-210. Zona Arqueológica 15 (Alcalá de Henares: Museo Arqueológico Regional). Gutiérrez Lloret, Sonia. 2012. ‘La arqueología en la historia del temprano al-Andalus: espacios sociales, cerámica e islamización’ in Philippe Sénac (ed.). Histoire et archéologie de l’Occident musulman (VIIe-XVe). Al-Andalus, Maghreb, Sicile, 33-66. Villa 4 (Toulouse: CNRS). Gutiérrez Lloret, Sonia; Lorenzo Abad Casal; and Blanca Gamo Parras. 2005. ‘“Eio”, “Iyyuh” y el Tolmo de Minateda (Hellín, Albacete). De sede episcopal a “madîna” islámica’ in VI RACH, 345-70. Gutiérrez Lloret, Sonia; and Julia Sarabia Bautista. 2013. ‘The Episcopal Complex of Eio-El Tolmo de Minateda (Hellín, Albacete, Spain): Architecture and Spatial Organization, 7 th to 8th centuries AD’. Hortus Artium Medeivalium 19, 267-300. Gutiérrez Lloret, Sonia; Blanca Gamo Parras, and Victoria Amorós Ruiz. 2003. ‘Los contextos cerámicos altomedievales de El Tolmo de Minateda y la cerámica altomedieval del Sudeste de la Península’ in Luis Caballero Zoreda, Pedro Mateos Cruz and Manuel Retuerce Velasco (eds.). Cerámicas tardorromanas y altomedievales en la Península Ibérica, 119-68. AAEspA 28 (Mérida: CSIC). Gutiérrez Palacios, Arsenio; Manuel Díaz Díaz; and Juan Maluquer de Motes. 1958. ‘Excavaciones en la Lancha del Trigo, Diego Álvaro (Ávila)’. Zephyrus 9, 59-78.

362 

The Iberian Peninsul a bet ween 300 and 850

Hakenbeck, Susanne. 2004. ‘Reconsidering Ethnicity: An Introduction’. Archaeological Review from Cambridge 19.2, 1-6. Halm, Heinz. 1998 [1989]. ‘Al-Andalus and the Gothic sors’ in Manuela Marín Niño (ed.). The Formation of Al-Andalus, Part 1: History and Society, 39-50 (Aldershot: Ashgate). Halsall, Guy. 1995a. Early Medieval Cemeteries: An Introduction to Burial Archaeology in the Post-Roman West (Skelmorlie: Cruithne Press). Halsall, Guy. 1995b. Settlement and Social Organization: The Merovingian Region of Metz (Cambridge: Cambridge University Press). Halsall, Guy. 1996a. ‘Female Status and Power in Early Merovingian Central Austrasia: The Burial Evidence’. EME 5.1, 1-24. Halsall, Guy. 1996b. ‘Towns, Societies and Ideas: The Not-So-Strange Case of Late Roman and Early Merovingian Metz’ in Neil Christie and Simon Loseby (eds.). Towns in Transition: Urban Evolution in Late Antiquity and the Early Middle Ages, 235-61 (Aldershot: Scolar Press). Halsall, Guy. 2007. Barbarian Migrations and the Roman West, 376-568 (Cambridge: Cambridge University Press). Halsall, Guy. 2011. ‘Ethnicity and Early Medieval Cemeteries’. AyTM 18, 15-28. Halstead, Paul; and John O’shea. 1989. ‘Introduction: Cultural Responses to Risk and Uncertainty’ in Paul Halstead and John O’shea (eds.). Bad Year Economics. Cultural Responses to Risk and Uncertainty, 1-7 (Cambridge: Cambridge University Press). Hamerow, Helena. 2002. Early Medieval Settlements: The Archaeology of Rural Communities in North-West Europe 400-900 (Oxford: Oxford University Press). Hamerow, Helena. 2012. Rural Settlements and Society in Anglo-Saxon England (Oxford: Oxford University Press). Hansen, Inge; and Chris Wickham. 2000. The Long Eighth Century (Leiden: Brill). Härke, Heinrich. 1990. ‘“Warrior graves”? The Background of the Anglo-Saxon Weapon Burial Rite’. Past and Present 126, 22-43. Härke, Heinrich. 2007. ‘Ethnicity, ‘Race’ and Migration in Mortuary Archaeology: An Attempt at a Short Answer’. Anglo-Saxon Studies in Archaeology and History 14, 12-18. Härke, Heinrich. 1997. ‘The Nature of Burial Data’ in Claus Jensen and Karen Nielsen (eds.). Burials and Society: The Chronological and Social Analysis of Archaeology and Burial Data, 19-29 (Aarhus: Aarhus University Press). Heather, Peter. 1998. ‘Goths and Huns’ in Averil Cameron and Peter Garnsey (eds.). Cambridge Ancient History, Volume XIII: The Late Empire, AD 337-425, 487-515 (Cambridge: Cambridge University Press). Heather, Peter. 2005. The Fall of the Roman Empire: A New History of Rome and the Barbarians (Oxford: Oxford University Press). Heijmans, Marc. 2004. Arles durant l’Antiquité tardive. De la duplex Arelas à l’urbs Genesii. Collection de l’École Française a Rome 324 (Rome: École Française de Rome). Heiss, Aloiss. 1872. Descripción general de las monedas visigodas (Madrid: Cayón). Heras Mora, Francisco Javier; and Ana Belén Olmedo Gragera. 2015. ‘Identidad y contexto en la necrópolis tardorromana de Mérida’ in Juan Antonio Quirós Castillo y Santiago Castellanos Martínez (eds.). Identidad y etnicidad en Hispania. Propuestas teóricas y cultura material en los siglos V-VIII, 275-290 (Bilbao: Universidad del País Vasco). Hernández Guerra, Liborio. 2008. ‘Algunas consideraciones sobre municipios flavios en la Meseta septentrional’. Gerión 26.1, 407-38. Hernández Juberías, Julia. 1996. La península imaginaria: mitos y leyendas sobre Al-Ándalus (Madrid: CSIC).

Bibliogr aphy

363

Hernández Vera. José Antonio. 2004. ‘La mezquita aljama de Zaragoza a la luz de la información arqueológica’. ‘Ilu. Revista de ciencias de las religiones 10, 65-91. Hidalgo Prieto, Rafael. 2002. ‘De edificio imperial a complejo de culto: la ocupación cristiana del palacio de Cercadilla’ in Desiderio Vaquerizo Gil (ed.). Espacio y usos funerarios en el occidente romano (2 vols), vol. II, 343-72 (Córdoba: Universidad de Córdoba). Hidalgo Prieto, Rafael. 2005. ‘Algunas cuestiones sobre la Corduba de la antigüedad tardía’ in VI RACH, 401-14. Hierro Gárate, José Ángel. 2011. ‘La utilización sepulcral de las cuevas en época visigoda: los casos de Las Penas, La Garma y el Portillo del Arenal (Cantabria)’. Munibe 62, 351-402. Hodder, Ian. 1983. ‘The Identification and Interpretation of Ranking in Prehistory: A Contextual Approach’ in Collin Renfrew and Steven Shennan (eds.). Ranking, Resource and Exchange, 150-4 (Cambridge: Cambridge University Press). Hodge, Trevor. 2000. ‘Qanats’ in Örjan Wikander (ed.). Handbook of Ancient Water Technology, 35-8 (Leiden: Brill). Horden, Peregrin; and Nicholas Purcell. 2000. The Corrupting Sea: A Study of Mediterranean History (Oxford: Blackwell). Hortelano Uceda, Ignacio. 2008. ‘La red de acueductos de la Valentia romana. Canales de abastecimiento rural al sur del Turia’. Lucentum 27, 69-85. Houten, Pieter. 2016. ‘Monumentality in Hispanoroman Cities: a Social Network Approach’ Cuadernos de Arqueoloía de la Universidad de Navarra 14, 49-80. Hübner, Emil. 1871. Inscriptiones Hispaniae Christianae (Berlin: Reimer). Insoll, Timothy. 1999. The Archaeology of Islam (Oxford: Blackwell). Isla Frez, Amancio. 2001. ‘Villa, villula, castellum. Problemas de terminología rural en época visigoda’. AyTM 8, 9-19. Isla Frez, Amancio. 2011. ‘Identidades y goticismo en época de Alfonso III: las propuestas de la Albeldense’. Territorio, Sociedad y Poder 6, 11-21 Izquierdo Benito, Ricardo. 2009. ‘¿De complejo palatino visigodo a arrabal islámico?’ in Alfonso García and Diego Perís (eds.). La Vega Baja de Toledo, 95-113 (Toledo: Toletum Visigodo). James, Edward. 1979. ‘Cemeteries and the Problem of Frankish Settlement in Gaul’, in Peter Sawyer (ed.). Names, Words and Graves: Early Medeieval Settlement, 55-90 (Leeds: University of Leeds). James, Edward. 1988. The Franks (Oxford: Blackwell). Járrega Domínguez, Ramón. 1991. ‘Consideraciones sobre la cronología de las murallas tardorromanas de Barcelona. ¿Una fortificación del siglo V?’. AEspA 64, 326-35. Járrega Domínguez, Ramón. 2013. ‘Las últimas importaciones romanas de cerámica en el este de Hispania Tarraconensis: una aproximación’. Spal 22, 143-72. Jepure, Antonel. 2004. La necrópolis de época visigoda de Espirdo-Veladiez (Segovia: Junta de Castilla y León). Jepure, Antonel. 2012. Das westgotenzeitliche Gräberfeld von Madrona (Segovia, Spanien) (Würzburg: Universität Würzburg). Jervis, Ben; Luke Barber; Lyn Blackmore; John Cotter; Chris Jarrett; Phil Jones; Lorraine Mepham; and Berni Seddon. 2017. ‘Early Anglo-Saxon Pottery in South East England: Recent Work and a Research Framework for the Future’. Medieval Ceramics 36. Jiménez Garnica, Ana María. 1999. ‘Settlement of the Visigoths in the Fifth Century’ in Peter Heather (ed.). The Visigoths: From the Migration Period to the Seventh Century, 93-115 (Woodbrige: Boydell). Jiménez Martín, Alfonso. 1974. ‘Los acueductos de Bellone Claudia (Bolonia, Cádiz)’ Habis 5, 273-93.

364 

The Iberian Peninsul a bet ween 300 and 850

Jiménez Puertas, Miguel. 1995. ‘El poblamiento rural de la tierra de Loja a f ines de la Edad Media’. AyTM 2, 63-82. Jiménez Puertas, Miguel; Luca Mattei; and Ana Ruiz Jiménez. 2011. ‘Rituales y espacios funerarios en la Alta Edad Media: las necrópolis excavadas en la roca de Martilla y Tózar (Granada)’ in Miguel Jiménez Puertas and Guillermo García-Contreras Ruiz (eds.). Paisajes históricos y Arqueología medieval, 140-175 (Granada: Alhulia). Jiménez Sánchez, Juan Antonio. 2006. ‘Los últimos ludi circenses realizados en Hispania en época visigoda’. Faventia 28.1, 99-113. Jiménez Sánchez, Juan Antonio. 2009-2010. La desaparición de los espectáculos de gladiadores en Hispania’. Hispania Antiqua 33-34, 273-94. Jones, Arnold Hughes Martin. 1958. ‘The Roman Colonate’. Past and Present 13.1, 1-13. Jones, Arnold Hughes Martin. 1964. The Later Roman Empire, 284-602: A Social, Economic and Administrative Survey (Norman: University of Oklahoma Press). Jones, Siân. 1997. The Archaeology of Ethnicity: Constructing Identities in the Past and Present (London: Routledge). Juan Navarro, Empar; and Ignacio Pastor Cubillo. 1989. ‘El yacimiento de época visigótica de Pla de Nadal’. Archivo de Prehistoria Levantina 19, 357-373. Juan Navarro, Empar; Josep Vicent Lerma; and Ignacio Pastor Cubillo. 1992. ‘Pla de Nadal. Una villa nobiliaria de época visigoda’. Revista de Arqueología 131, 22-32. Juan Tovar, Luis Carlos. 2012. ‘Las cerámicas de imitación de sigillata en el occidente de la Península Ibérica durante el siglo V d.C.’ in Darío Bernal Casasola and Albert Ribera Lacomba (eds.).Cerámicas hispanorromanas II. Producciones regionales, 97-130. Monografías Historia y Arte (Cádiz: Universidad de Cádiz). Juan Tovar, Luis Carlos; Juan Sanguino Vázquez; Pilar Oñate Baztán; and Eduardo Penedo Cobo. 2013. ‘Hornos cerámicos bajoimperiales y tardoantiguos en el sur de la Comunidad de Madrid: presentación preliminar’. Actas del I Congreso Internacional de la SECAH, 319-335 (Madrid: SECAH). Kazanski, Michel; and Patrick Périn. 2008. ‘Identité ethnique en Gaule à l’époque des Grandes Migrations et des Royaumes barbares: étude de cas archéologiques’. Antiquités nationales 39, 181-216. Kennedy, Hugh. 1985. ‘From Polis to Madina: Urban Change in Late Antique to Early Islamic Syria’. English Historical Review 106, 3-27. Kim, Hyun Jin. 2016. The Huns (London: Routledge). Kirchner Granell, Helena. 2010a. ‘Sobre la Arqueología de las aldeas altomedievales’.Studia Histórica: Historia Medieval 28, 243-253. Kirchner Granell, Helena. 2010b. Por una arqueología agraria: perspectivas de investigación sobre espacios de cultivo en las sociedades medievales hispánicas. BAR IS 2062 (Oxford: Archaeopress). Kobylinski, Zbigniew. 1990. ‘Early Medieval Hillforts in Polish Lands in the 6th to the 8th Centuries: Problems of Origins, Function, and Spatial Organization’ in David Austin and Leslie Alcock (eds.). From the Baltic to the Black Sea: Studies in Medieval Archaeology, 147-56 (London: Unwin Hyman). Koch, Manuel. 2006. ‘Gotthi intra Hispanias sedes acceperunt. Consideraciones sobre la supuesta inmigración visigoda en la península Ibérica’. Pyrenae 37.2, 83-104. Koch, Manuel. 2008. ‘La imperialización del reino visigodo bajo Leovigildo. ¿Es la imitatio imperii de Leovigildo la manifestación de un momento de cambio en la pretensión del poder y la ideología visigodas?’. Pyrenae 39.2, 101-17. Kramer, Carol. 1985. ‘Ceramic Ethnoarchaeology’. Annual Review of Anthropology 14, 77-102. Krautheimer, Richard. 1983. Three Christian Capitals: Topography and Politics (Berkeley: University of California). Kulikowski, Michael. 2004. Late Roman Spain and its Cities (London: Johns Hopkins University Press).

Bibliogr aphy

365

Kulikowski, Michael. 2005. ‘Cities and Government in Late Antique Hispania: Recent Advances and Future Research’ in Kim Bowes and Michael Kulikowski (eds.). Hispania in Late Antiquity: Current Perspectives, 31-76 (Leiden: Brill). Kulikowski, Michael. 2006. ‘The Late Roman City in Spain’, in Jens-Uwe Krause and Christian Witschel (eds.). Die Stadt in der Spätantike- Niedergang oder Wandel?, 129-40 (Stuttgart: Franz Steiner Verlag). Lacort Navarro, Pedro José. 1988. ‘Infraestructura hidráulica rural en la campiña de Córdoba’. Memoria de Historia Antigua 9, 511-82. Lagóstena Barrios, Lázaro. 2001. La producción de salsas y conservas de pescado en la Hispania Romana (II a.C. – VI d.C.) (Barcelona: Universidad de Barcelona). Lakarra Andrinua, Joseba. 2008. Informe sobre las supuestas inscripciones eusquéricas antiguas de Veleia. MS report presented at the Diputación Foral de Álava http://www.araba.eus/ publicar/Informes/Veleia_Inf_07.pdf Lamprecht, Heinz. 1987. Opus Caementitium. Bautechnik der Römer. (Düsseldorf: Beton-Verlag). Lavan, Luke. 2001. ‘The Late-Antique City: A Bibliographic Essay’ in Luke Lavan (ed.). Recent Research in Late Antique Urbanism, 9-26. JRA SS 42 (Portsmouth: JRA). Lavan, Luke. 2008. ‘Explaining Technological Change: Innovation, Stagnation, Recession and Replacement’ in Luke Lavan, Enrico Zanini and Alexander Sarantis (eds.). Technology in Transition: AD 300-650, xv-xl. Late Antique Archaeology 4 (Leiden: Brill). Lecanda Esteban, José Ángel. 2000. ‘Mijangos: la aportación de la epigrafía y el análisis arqueológico al conocimiento de la transición a la Alta Edad Media en Castilla’ in Luis Caballero Zoreda and Pedro Mateos Cruz (eds.). Visigodos y Omeyas: un debate entre la Antigüedad Tardía y la Alta Edad Media, 181-206. AAEspA 23 (Madrid: CSIC). Lecanda Esteban, José Ángel; and Ángel Luis Palomino Lázaro. 1999. ‘Dos modelos de ocupación del territorio en época visigoda en la provincia de Burgos: la región montañesa septentrional y la comarca ribereña del Duero’ in V Congreso de Arqueología Medieval Española (2 vols), vol. I, 38-56 (Valladolid: Junta de Castilla y León). Lemos, Francisco; Manuela Martins; Luis Fontes; Jose Manuel Leite; and Armandinho Cunha. 1998. ‘A redescoberta da muralha romana e suevica-visigotica de Braga’. Forum 24, 11-26. León Muñoz, Alberto; and Juan Francisco Murillo Redondo. 2009. ‘El complejo civil tardoantiguo de Córdoba y su continuidad en el alcázar omeya’. Madrider Mitteilungen 50, 399-432. León Muñoz, Alberto; and Juan Francisco Murillo Redondo. 2014. ‘Advances in Research on Islamic Córdoba’. Journal of Islamic Archaeology 1.1, 5-36. Leone, Anna. 2007. Changing Townscapes in North Africa from Late Antiquity to the Arab Conquest (Bari: Edipuglia). Lévi-Provençal, Évariste. 1957. España musulmana hasta la caída del califato de Córdoba (711-1031 de J. C.). Instituciones y vida social e intelectual. Historia de España de Ramón Menéndez Pidal 5 (Madrid: Espasa-Calpe). Lévi-Strauss, Claude. 1993 [1963]. Structural Anthropology (London: Penguin). Lewis, Michael. 2007. ‘Antique Engineering in the Byzantine world’ in Luke Lavan, Enrico Zanini and Alexander Sarantis (eds.). Technology in Transition A.D. 300-650, 367-78. Late Antique Archaeology 4 (Leiden: Brill). Lewit, Tamara. 2002. ‘Bones in the Bathhouse: Re-Evaluating the Notion of “Squatter Occupation” in 5th-7th Century Villas’ in Gian Pietro Brogiolo, Alexandra Chavarría Arnau and Marco Valenti (eds.). Dopo la fine delle ville: le campagne dal VI al IX secolo, 251-61 (Mantua: SAP). Lewit, Tamara. 2003. ‘Vanishing Villa: What Happened to Elite Rural Habitation in the West in the 5th-6th c.?’. JRA 16, 260-74.

366 

The Iberian Peninsul a bet ween 300 and 850

Liebeschuetz, Wolfgang. 1992. ‘The End of the Ancient city’ in John Rich (ed.). The City in Late Antiquity, 1-50 (London: Ashgate). Liebeshuetz, Wolfgang. 2015. East and West in Late Antiquity: Invasions, Settlement, Ethnogenesis and Conflict of Religion. Impact of Empire 20 (Leiden: Brill). Llobregat, Enrique. 1973. Teodomiro de Oriola, su vida y su obra (Alicante: Caja de Ahorros Provincial de Alicante). Lopes, Virgilio. 2014. Mértola e o seu território na antiguidade tardia (séculos IV-VIII). PhD Thesis (Universidad de Huelva). Lopes, Virgilio. 2015. ‘La antigüedad tardía en Mértola (Portugal)’. Onoba 3, 105-28. Lopes, Virgilio; and Macias, Santiago. 2005. ‘Mértola na antiguidade tardia’ in VI RACH, 449-64. López Alsina, Fernando. 1994. ‘La invención del sepulcro de Santiago y la difusión del culto jacobeo’ in El Camino de Santiago y la articulación del espacio hispanico, 59-83 (Pamplona: Gobierno de Navarra). López Padilla, Juan Antonio; and María Teresa Ximénez de Embún Sánchez. 2014. ‘Excavaciones arqueológicas en Cabezo Pardo (San Isidro/Granja de Rocamora, Alicante). Últimas actuaciones’ in Manuel Olcina Doménech and Jorge Soler Díaz (eds.). II Jornadas de Arqueología y Patrimonio Alicantino. Arqueología en Alicante en la primera década del siglo XXI, 215-221 (Alicante: MARQ). López Quiroga, Jorge. 2004a. El final de la antigüedad tardía en Gallaecia. La transformación de las estructuras de poblamiento entre Miño y Duero (siglos V al X) (Santiago de Compostela: Fundación Pedro Barrié de la Maz). López Quiroga, Jorge. 2004b. ‘La presencia “Germánica” en Hispania en el siglo V d.C’. CuPAUAM 30, 213-223. López Quiroga, Jorge. 2006. ‘¿Dónde vivían los “germanos”?’ in Jorge Morín De Pablos (ed.). La investigación arqueológica de la época visigoda en la comunidad de Madrid (3 vols), vol. I, 310-64. Zona Arqueológica 8 (Alcalá de Henares: Museo Arqueológico Regional). López Quiroga, Jorge; and Artemio Martínez Tejera. 2006. ‘El destino de los templos paganos en Hispania durante la Antigüedad Tardía’. AEspA 79, 125-53. López Vilar, Jordi. 2006. Les basíliques paleocristianes del suburbi occidental de Tarraco. El temple septentrional i el complex martirial de Sant Fructuós, Tarragona (Tarragona: ICAC). Lorenzo Jiménez, Jesús. 2010. La Dawla de los Banu Qasi. Origen, auge y caída de una dinastía muladí en la frontera superior de Al-Andalus (Madrid: CSIC). Lorrio, Alberto; and María Dolores Sánchez de Prado. 2004. ‘La mezquita y el hisn de El Molón (Camporrobles, Valencia)’ in Javier Jover Maestre and Concepción Navaro Poveda (eds.). De la medina a la vila. II jornadas de Arqueología Medieval, 139-66 (Alicante: MARQ). Loseby, Simon. 1996. ‘Arles in Late Antiquity: Gallula Roma Arelas and Urbs Genesii’ in Neil Christie and Simon Loseby (eds.). Towns in Transition: Urban Evolution in Late Antiquity and the Early Middle Ages, 45-70 (Aldershot: Scholar Press). Loseby, Simon. 2005. ‘The Mediterranean economy’ in Paul Fouracre (ed.). The New Cambridge Medieval History I (c.500-c.700), 605-38 (Cambridge: Cambridge University Press). Loseby, Simon. 2006 ‘Decline and Change in the Cities of Late Antique Gaul’, in Jens-Uwe Krause and Christian Witschel (eds.). Die Stadt in der Spätantike – Niedergang oder Wandel?, 67-104 (Stuttgart: Franz Steiner Verlag). Loza Uriarte, Miguel; and Javier Niso Lorenzo. 2011. ‘Resultados preliminares de la intervención arqueológica de San Martín de Dulantzi (Alegría-Dulantzi, Álava)’ in Juan Antonio Quirós Castillo (ed.). Vasconia en la Alta Edad Media, 450-1000, 235-245 (Vitoria-Gasteiz: Universidad del País Vasco). Luttwak, Edward. 1976. The Grand Strategy of the Roman Empire: From the First Century A.D. to the Third (Baltimore: John Hopkins University Press).

Bibliogr aphy

367

Machancoses López, Mirella. 2015. Topografía urbana de la Valentia romana altoimperial: ciudad y suburbio. PhD Thesis (University of Valencia). Macias Solé, Josep Maria. 2004a. ‘Arquitectura doméstica’ in Xavier Dupré (ed.). Tarragona. Colonia Iulia Urbs Triumphalis Tarraco, 73-82 (Rome: L’erma di Bretschneider). Macias Solé, Josep Maria. 2004b. Les termes publiques de l’area portuaria de Tàrraco. Carrer de Sant Miquel de Tarragona (Tarragona: ICAC). Macias Solé, Josep Maria. 2008. ‘Tarracona visigoda. ¿Una ciudad en declive?’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en época visigoda, 292-301. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Macias Solé, Josep Maria. 2013. ‘La medievalización de la ciudad romana’ in Josep Maria Macias Solé and Andreu Muñoz Melgar (eds.). Tarraco christiana ciuitas, 123-48. Documenta 24 (Tarragona: ICAC). Macias Solé, Josep Maria; and Andreu Muñoz Melgar, (eds.). 2013. Tarraco christiana ciuitas. Documenta 24 (Tarragona: ICAC) Macias Solé, Josep Maria; and Josep Anton Remolà Vallverdú. 2005. ‘El port de Tarraco a l’antiguitat tardana’, in RACH VI, 175-87. Maestre Borge, Cinta. 2012. ‘La cerámica común tardía en Sevilla (siglos IV-VI d.C.)’ in Darío Bernal Casasola and Albert Ribera Lacomba (eds.). Cerámcias hispanorromanas II. Producciones regionales, 487-510. Monografías Historia y Arte (Cádiz: University of Cádiz). Malpica Cuello, Antonio. 2010. ‘El surgimiento de la ciudad de Ilbira y su relación con el territorio de la vega de Granada en la edad media temprana’ in Paola Galletti (ed.). Paesaggi, comunità, villaggi medievali. Atti del Convegno internazionale di studio, 91-106 (Spoleto: Centro italiano di studi sull’alto medioevo). Malpica Cuello, Antonio. 2011. ‘La ciudad andalusí de Ilbira. Su formación y desarrollo’ in Cristãos e muçulmanos na Idade Média peninsular. Encontros e desencontros, 27-49 (Lisbon: Instituto de Arqueologia e Palèociencias). Mango, Marlia. 2000. ‘Building and Architecture’, in Averil Cameron, Bryan Ward-Perkins, and Michael Whitby (eds.). The Cambridge Ancient History, vol. XIV. Late Antiquity: Empire and Successors, A.D. 425-600, 918-71 (Cambridge: Cambridge University Press). Mannoni, Tizino. 2008. ‘The Transmission of Craft Techniques According to the Principles of Material Culture: Continuity and Rupture’ in Luke Lavan, Enrico Zanini and Alexander Sarantis (eds.). Technology in Transitio:. AD 300-650, xli-lx. Late Antique Archaeology 4 (Leiden: Brill). Manzano Moreno, Eduardo. 2006. Conquistadores, emires y califas. Los Omeyas y la formación de Al-Ándalus (Barcelona: Crítica). Mañanes Pérez, Tomás. 1988. Arqueología de la cuenca leonesa del río Sil (Laceana, Bierzo, Cabrera) (Valladolid: Junta de Castilla y León). Mañanes Pérez, Tomás; and Jose María Solana Sainz. 1985. Ciudades y vías romanas en la cuenca del Duero (Castilla y León) (Valladolid: Universidad de Valladolid). Mar, Ricardo; and Guiseppe Verde. 2008. ‘Las villas romanas tardoantiguas: cuestiones de tipología arquitectónica’ in Carmen Fernández Ochoa, Virginia García-Entero and Fernando Gil Sendino (eds.). Las villae tardorromanas en el occidente del Imperio: arquitectura y función. IV Coloqio Internacional de Arqueología en Gijón, 49-83 (Gijón: Trea). Marcos Contreras, Gregorio José; Francisco Javier Sanz García; Jesús Carlos Misiego Tejeda; Miguel Ángel Martín Carbajo; and Luis Ángel del Caño García. 2010. ‘La ocupación tardorromana en el yacimiento de Carratejera, en Navalmanzano (Segovia)’ in Santiago Martínez Caballero, Jorge Santiago Pardo and Alonso Zamora Canellada (eds.). Segovia Romana II. Gentes y territorios, 379-392 (Segovia: Caja Segovia).

368 

The Iberian Peninsul a bet ween 300 and 850

Marfil Ruiz, Pedro. 2000a. ‘La sede episcopal cordobesa en época bizantina: evidencia arqueológica’ in V RACH, 157-75. Marfil Ruiz, Pedro. 2000b. ‘Córdoba de Teodosio a Abd al-Rahmán III’ in Luis Caballero Zoreda and Pedro Mateos Cruz (eds.). Visigodos y Omeyas. Un debate entre la Antigüedad tardía y la alta Edad Media (Mérida, abril de 1999), 117-42. AAEspA 23 (Madrid: CSIC). Marichale, Rémi; and Philippe Sénac. 2007. ‘Ruscino : un établissement musulman du VIIIe siècle’ in Philippe Sénac (ed.). Villes et Campagnes de Tarraconaise et d’al-Andalus (VIe-XIe siècle) : la transition, 67-94. Études Médiévales Ibériques (Toulouse: CNRS). Mariné Isidro, María. 2001. Fíbulas romanas en Hispania: la meseta. AAEspA 24 (Madrid: CSIC). Márquez Moreno, Carlos; and Ángel Ventura Villanueva. 2005. ‘Corduba tras las guerras civiles’ in Enrique Melchor Gil, Joaquín Mellado Rodríguez and Juan Francisco Rodríguez-Neila (eds.). Julio César y Corduba: tiempo y espacio en la campaña de Munda (49-45 a.C.), 429-66 (Córdoba: Universidad de Córdoba). Marshall, Anne; and Garry Marshal. 1991. ‘A Survey and Analysis of the Buildings of Early and Middle Anglo-Saxon England’. Medieval Archaeology 35, 29-43. Martí Castelló, Ramón (ed.). 2008. Fars de l’islam. Antigues alimares d’al-Andalus (Barcelona: EDAR). Martín Civantos, José María. 2001. ‘Alquife, un castillo con vocación minera en el Zenete’ AyTM 8, 325-345. Martín Civantos, José María. 2010. ‘La formación de la tierra de Guadix (Granada)’ in Antonio Malpica, Rafael G. Peinado and Adela Fábregas (eds.). Historia de Andalucía. VII Coloquio, 49-64 (Granada: Universidad de Granada). Martín Rodríguez, Eva María; and San Gregorio Hernández, Diego. 2008. ‘La necrópolis tardorromana del Soto de Tovilla (Tudela de Duero, Valladolid)’. Estudios del Patrimonio Cultural 1, 19-32. Martín Viso, Iñaki. 2009a. ‘Espacios sin Estado: los territorios occidentales entre el Duero y el Sistema Central (S. VIII-IX)’ in Iñaki Martín Viso (ed.). ¿Tiempos oscuros? Territorios y sociedad en el centro de la Península Ibérica (Siglos VII-X), 107-135 (Madrid: Sílex Universidad). Martín Viso, Iñaki. 2009b. ‘MARTÍ, Ramón (ed.) Fars de l’islam. Antigues alimares d’al-Andalus. EDAR. Barcelona, 2008, 350 pp’. Studia Historica 27, 214-7. Martín Viso, Iñaki. 2012. ‘Enterramientos, memoria social y paisaje en la Alta Edad Media: propuestas para un análisis de las tumbas excavadas en roca en el centro-oeste de la Península Ibérica’. Zephyrus 69, 165-187. Martín Viso, Iñaki. 2013a. ‘The “Visigothic” Slates and their Archaeological Contexts’. Journal of Medieval Iberian Studies 5.2, 145-68. Martín Viso, Iñaki. 2013b. ‘Prácticas locales de la f iscalidad en el Reino de Toledo’ in Xavier Ballestín Navarro and Ernesto Pastor Díaz (eds.). Lo que vino de Oriente. Horizontes, praxis y dimensión material de los sistemas de dominación fiscal en Al-Andalus (ss. VII-IX), 72-85. BAR IS 2525 (Oxford: Archaeopress) Martín Viso, Iñaki. 2014. ‘El espacio del más acá: las geografías funerarias entre la Alta y la Plena Edad Media’ in Esther López Ojeda (ed.). De la tierra al cielo. Ubi sunt qui ante nos in hoc mundo fuere. XXIV Semana de estudios medievales, 75-140 (Logroño: Gobierno de la Rioja). Martín Viso, Iñaki; and Rubén Rubio Díez. 2013. Informe técnico: primera campaña de excavación arqueológica en el yacimiento de La Genestosa (Casillas de Flores, Salamanca). Excavation report MS deposited at the Servicio Territorial de Cultura de Salamanca. Martín Menéndez, Albert; Núria Miró Alaix; and Emili Revilla Cubero. 2000. ‘El complejo termal privado de la domus de la calle Bisbe Caçador de Barcelona’ in Carmen Fernández Ochoa and Virginia García-Entero (eds.). Termas romanas en el occidente del imperio. II Coloquio internacional de arqueología en Gijón, 283-7 (Gijón: Ayuntamiento de Gijón).

Bibliogr aphy

369

Martínez Jiménez, Javier. 2011a. ‘The Continuity of Roman Water Supply Systems in Post-Roman Spain: The Case of Valentia, a Reliable Example?’. Arkeogazte 1, 125-44. Martínez Jiménez, Javier. 2011b. ‘Monte Barro: An Ostrogothic Fortified Site in the Alps’. Assemblage 11, 34-46. Martínez Jiménez, Javier. 2012. ‘Reuse, Repair and Reconstruction: Functioning Aqueducts in Post-Roman Spain’ in Ben Jervis and Alison Kyle (eds.). Make-do and Mend: the Archaeologies of Compromise, 26-44. BAR IS 2408 (Oxford: Archaeopress). Martínez Jiménez, Javier. 2013. ‘Crisis or Crises? The End of Roman Towns in Iberia, Between the Late Roman and the Early Umayyad Periods’ in Elsbeth Van der Wilt and Javier Martínez Jiménez (eds.). Tough Times: The Archaeology of Crisis and Recovery, Proceedings of the GAO Annual Conferences 2010 and 2011, 77-90. BAR IS 2478 (Oxford: Archaeopress). Martínez Jiménez, Javier. 2014. Aqueducts and Water Supply in the Towns of Post-Roman Spain AD 400-1000. DPhil Thesis (University of Oxford). Martínez Jiménez, Javier. 2015. ‘A Preliminary Study of the Aqueduct of Reccopolis’. OJA 34.4, 301-20. Martínez Jiménez, Javier. 2017. ‘Water Supply in the Visigothic Urban Foundations of Eio (El Tolmo de Minateda) and Reccopolis’ in Efthymios Rizos (ed.). New Cities in Late Antiquity: Documents and Archaeology, 233-45. Bibliothèque de l’Antiquité Tardive 35 (Turnhout: Brepols). Martínez Jiménez, Javier; and José María Moreno Narganes. 2015. ‘Nunc autem a Gothis subversa: The Province of Alicante and the Spanish Mediterranean Towns Between the Byzantine and Visigothic Periods’. EME 23.3, 263-89. Martínez Jiménez, Javier; and Carlos Tejerizo García. 2015. ‘Central Places in the Post-Roman Mediterranean: Regional Models for the Iberian Peninsula’. Journal of Mediterranean Archaeology 28.1, 81-104. Martínez Lillo, Sergio. 2008. ‘El vino consumido en al-Andalus’ in Juan Blázquez Miguel and Sebastián Celestino Pérez (eds.).El vino en época tardoantigua y medieval, 253-80. Serie Varia 8 (Madrid: Universidad Autónoma de Madrid). Martínez Núñez, María Antonia. 2009. ‘Epigrafía árabe e historia de al-Andalus: nuevos hallazgos y datos’ in Actas del 6º Encontro de Arqueologia do Algarve. O Gharb no al-Andalus: síntesis e perspectivas de estudo. Homenagem a José Luis Matos (Silves, 23, 24 e 25 de outubro 2008), 39-54. Xelb, 9 (Silves: Camara Municipal de Silves). Martínez Rodríguez, Antonio. 2014. ‘Territorio y poblamiento del litoral murciano en el periodo andalusí’. @rqueología y Territorio 11, 123-36. Martínez Santa-Olalla, Julio. 1934. ‘Notas para un ensayo de sistematización de la arqueología visigoda en España’. Archivo Español de Arte y Arqueología 10.29, 139-76. Martins, Manuela; Fernanda Magalhães; Raquel Peñin; and Jorge Ribeiro. 2016. ‘The Housing Evolution of Braga Between Late Antiquity and the Early Middle Ages’ in Folcel Sabaté and Jesús Brufal (eds.). Arqueologia Medieval: Hàbitats Medieval, 35-52 (Lérida: Pages). Mascarenhas, José Manuel, Francisco Bilous; and Nuno Sousa Neves. 2012. ‘O aqueduto romano de Olisipo: viabilidade ou utopia? Ensaio de traçado apoiado em modelação geográf ica’. Revista Portuguesa de História 43, 239-64. Mateos Cruz, Pedro. 1995. ‘La identificación del xenodochium fundado por Masona en Mérida’ in IV RACH, 309-16. Mateos Cruz, Pedro. 1999. La basílica de Santa Eulalia de Mérida. Arqueología y urbanismo. AAEspA 19 (Madrid: CSIC). Mateos Cruz, Pedro; and Miguel Alba Calzado. 2000. ‘De Emerita Augusta a Marida’ in Luis Caballero Zoreda and Pedro Mateos Cruz (eds.). Visigodos y Omeyas. Un debate entre la Antigüedad tardía y la alta Edad Media (Mérida, abril de 1999), 143-68. AAEspA 23 (Madrid: CSIC).

370 

The Iberian Peninsul a bet ween 300 and 850

Mateos Cruz, Pedro; and Isaac Sastre de Diego. 2004. ‘Elementos arquitectónicos tardoantiguos y altomedievales en el Templo de Diana (Mérida, España)’. Mérida. Excavaciones arqueológicas 7, 397-415. Mateos Cruz, Pedro; Rocío Ayerbe Vélez; Teresa Barrientos Vera, and Santiago Feijoo Martínez. 2002. ‘La gestión del agua en Augusta Emerita’. Empúries 53, 67-88. Mattei, Luca; Marcos García García; Cristina Martínez Álvarez; and Manuel Linares Losa. 2014. ‘Iluminando sepulturas rupestres. Primeros datos arqueológicos sobre la necrópolis y el poblado altomedieval de Tózar (Granada)’. Revista del CEHGR 26, 3-32. Merino López, Lourdes; José Antonio López Sáez; Daniel Abel Schaad; Francisco Javier SánchezPalencia, and Guillermo Reher Díez. 2008. ‘Dinámica antrópica en El Bierzo (León) desde época romana: estudio palinológico de Castro Ventosa’. Polen 18, 25-36. Mezquíriz de Irujo, María Ángeles. 2004 [1965]. ‘Necrópolis visigoda de Pamplona’. Trabajos de Arqueología de Navarra 17, 43-90. Mezquíriz Irujo, María Ángeles, and María Inés Tabar Sarría. 1993-1994. ‘Excavaciones arqueológicas en la catedral de Pamplona’. Trabajos de arqueología Navarra 11-12, 310-11. Milek, Karen. 2012. ‘The Roles of Pit Houses and Gendered Spaces on Viking-Age Farmsteads in Iceland’. Medieval Archaeology 56, 85-130. Miles, George. 1952. The Coinage of the Visigoths of Spain: Leovigild to Achilla II (New York: ANE). Miller, Maureen. 2000. The Bishop’s Palace: Architecture and Authority in Medieval Italy (London: Cornell University Press). Miró Alaix, Carme; and Ferrán Puig Verdaguer. 2000. ‘Edificios termales públicos y privados en Barcino’ in Carmen Fernández Ochoa and Virginia García-Entero (eds.). Termas romanas en el occidente del imperio. Coloquio internacional Gijón 2000, 171-78 (Gijón: Ayuntamiento de Gijón). Molina Mahedero, Juan; and Ana Valdivieso Ramo. 2007. ‘Aportaciones sobre la evolución de las murallas de la Córdoba romana partir de los datos arqueológicos’. Rómula 6, 29-50. Molinero Pérez, Antonio. 1948. La necrópolis visigoda de Duratón (Segovia). Excavaciones del plan nacional de 1942 y 1943. Acta Arqueológica Hispánica 4 (Madrid: Ministerio de Educación Nacional). Morales Martínez, Alfredo. 1995. ‘A la sombra de Damasco. Arquitectura en al-Andalus hasta la proclamación del califato de Córdoba’ in Rafael López Guzmán (ed.). La arquitectura del Islam occidental, 43-51 (Barcelona: Lunwerg). Moreno Martín, Francisco. 2011. La arquitectura monástica hispana entre la Tardoantigüedad y la alta Edad Media. BAR IS 2287 (Oxford: Archaeopress). Morín de Pablos, Jorge; and Rafael Barroso Cabrera. 2007. Regia Sedes Toletana. El Toledo visigodo a través de su escultura monumental. (Toledo: Real Fundación de Toledo). Morín de Pablos, Jorge; and Rafael Barroso Cabrera. 2010. ‘El mundo funerario de las necrópolis tardorromanas a los cementerios hispanovisigodos en el oeste peninsular’ in Jorge Morín de Pablos, Jorge López Quiroga and Artemio Martínez Tejera (eds.) El tiempo de los “Bárbaros”. Pervivencia y transformación en Galia e Hispania (ss. V-VI d. C.), 148-80. Zona Arqueológica 11 (Alcalá de Henares: Museo Arqueológico Regional). Morín de Pablos, Jorge; Albert Ribera Lacomba; and Isabel Sánchez Ramos, (eds.). 2015. Preactas. Jornadas sobre Pla de Nadal y los palacios y espacios de representación en época visigoda. Homenaje a Empar Juan. Ribarroja Del Turia (Valencia) 25-28 Febrero 2015 (Madrid: Audema). Morín de Pablos, Jorge; and Isabel Sánchez Ramos (eds.). 2015. Los Hitos. Arisgotas – Orgaz, Toledo. De palacio a panteón visigodo (self-published). Moro García, Antonio; and FrancescTuset Bertrán. 2001. ‘Les darreres excavacions arqueològiques: 1995-2001’ in Les esglésies de Sant pere de Terrassa: de seu episcopal a conjunt monumental, 51-65 (Barcelona: Institut d’Estudis Catalans).

Bibliogr aphy

371

Mundee, Michelle. 2009. ‘An Isotopic Approach to Diet in Medieval Spain’ in Sera Baker, Annie Gray, Kay Lakin, Richard Madgwick, Kristopher Poole and Michela Sandias (eds.). Food and Drink in Archaeology 2, 64-72 (Totnes: Prospect). Mundee, Michelle. 2010. Exploring Diet and Society in Medieval Spain: New Approaches Using Stable Isotope Analysis. PhD Thesis (Durham University). Muñiz López, Iván; and Alejandro García Álvarez-Busto. 2010. ‘El castillo de Gauzón (Asturias, España) y el proceso europeo de feudalización entre la Antigüedad Tardía y la Edad Media a través de las fortificaciones’. Munibe 61, 289-328. Muñiz López, Iván; and Alejandro García Álvarez-Busto. 2012. ‘En los orígenes de las fortificaciones altomedievales: el castillo de Gauzón (Asturias). De asentamiento tardoantiguo a fortaleza de los reyes de Asturias (siglos VII-X)’ in Juan Antonio Quirós Castillo and José María Tejado Sebastián (eds.). Los castillos altomedievales en el noroeste de la Península Ibérica, 77-98 (Bilbao: Universidad del País Vasco). Murillo Redondo, Juan Francisco; María TeresaCasal García, and Elena Castro del Río. 2004. ‘Madinat Qurtuba. Aproximación al proceso de formación de la ciudad emiral y califal a partir de la información arqueológica’. Cuadernos de Madinat al-Zahra 5, 257-90. Murillo Redondo, Juan Francisco; Maudilio Moreno-Almenara; María Dolores Ruiz Lara; Sonia Vargas Cantos; and José Ramón Carrillo Díaz-Pines. 2002. ‘Los monumentos funerarios de Puerta de Gallegos- Colonia Patricia Corduba’ in Desiderio Vaquerizo Gil (ed.). Espacio y usos funerarios en el occidente romano (2 vols), vol. II, 247-74 (Córdoba: Universidad de Córdoba). Navarro Luengo, Ildefonso, Antonio Torremocha Silva; and Juan Bautista Salado Escaño. 2000. ‘Primeros testimonios arqueológicos sobre Algeciras en época bizantina’ in V RACH, 223-7. Navarro Luengo, Ildefonso; Luis Efrén Fernández Rodríguez; José Suárez Padilla; José Antonio Rambla Torralvo; and José Francisco Mayorga Mayorga. 2000. ‘Malaca bizantina: primeros datos arqueológicos’ in V RACH, 271-8. Navarro Luengo, Ildefonso; José Suárez Padilla; Ana Arancibia Román; María Isabel Cisneros García; María del Mar Escalante Aguilar; Luis Efrén Fernández Rodríguez; José Francisco Mayorga Mayorga; José Antonio Rambla Torralbo; and Juan Bautista Salado Escaño. 2001. ‘Comercio y comerciantes en la Málaga bizantina’ in Fernando Wulff Alonso, Gonzalo Cruz Andreotti and Celia Martínez Maza (eds.). II Congreso de Historia Antigua de Málaga: Comercio y comerciantes en la Historia Antigua de Málaga (siglo VII a.C. – 711 d.C.), 681-700 (Málaga: Diputación Provincial de Málaga). Navarro Palazón, Julio; and Pedro Jiménez Castillo. 2007. ‘Algunas reflexiones sobre el urbanismo islámico’. Artigrama 22, 259-98. de Navascués, Joaquín María. 1970. ‘El oratorio de Valdecanales (Jaén)’. Academia. Boletín de la Real Academia de Bellas Artes de San Fernando 31, 82-84. Nicosia, Cristiano; Yiannos Devos; and Quentin Borderie. 2013. ‘The Contribution of Geosciences to the Study of European Dark Earths: AReview’. Post-Classical Archaeologies 3, 145-70. Nieto Gallo, Germán.1958. El oppidum de Iruña (Vitoria: Diputación Foral de Álava). Nissen-Jaubert, Anne. 2012. ‘Ruptures et continuités dans l’habitat rural du Moyen Âge en Pays de la Loire’ in Alain Valais (ed.). L’habitat rural au Moyen Âge dans le Nord-Ouest de la Franc, 295-314 (Rennes: Presses Universitaires de Rennes). Nolla Brufau, José María. 1999. ‘El urbanismo de la ciudad de Gerona en la Alta Edad Media. Una primera aproximación’ in El urbanismo de los estados cristianos peninsulares, 205-26. Codex Aovilarensis 15 (Aguilar de Campoo: Fundación Santa María la Real). Nolla Brufau, José María; and Francisco Javier Nieto Prieto. 1979. ‘Acerca de la cronología de la muralla romana tardía de Gerunda: la terra sigillata clara de “Casa Pastors”‘. Faventia 1.2, 263-83.

372 

The Iberian Peninsul a bet ween 300 and 850

Nolla Brufau, José María; and JordiSagrera Arandilla. 1999. ‘Girona. La ciudad carolingia’ in Jordi Camps (ed.). Cataluña en época carolingia. Arte y cultura antes del Románico (siglos IX y X), 77-9 (Barcelona: Museo Nacional de Arte de Cataluña). O’Kelly Sandrós, Jessica. 2012. ‘Las ánforas onubenses en época tardorromana’ in Darío Bernal Casasola and Albert Ribera Lacomba (eds.). Cerámicas hispanorromanas II. Producciones regionales, 179-97. Monografías Historia y Arte (Cádiz: Universidad de Cádiz). Olcina Doménech, Manuel; Eva Tendero Porras; and Antonio Guilabert Mas. 2008. ‘La maqbara del Tossal de Manises (Alicante)’. Lucentum 28, 213-27. Oliveira, Cristina; António Carvalho; and Maria José Almeida. 2011. ‘La villa de Quinta das Longas (Elvas-Portugal): Les mosaïques du Bas-Empire’ in Virgilio Correia (ed.) O Mosaico Romano nos Centros e nas Periferias. O originalidades, influências e identidades (Actas do X Colóquio Internacional da Associação Internacional para o Estudo do Mosaico Antigo (AIEMA), 903-14 (Lisbon: Instituto dos Museus e da Conservação). Oliver Castaños, Ana; and Eduard Riu Barrera. 1987. ‘Pati d’en Llimona (Barcelona, Barcelonés)’. Arqueología 84-85, 79. Ollich Castanyer, Imma. 1999. ‘Vic. La ciudad en época carolingia’ in Jordi Camps (ed.) Cataluña en época carolingia. Arte y cultura antes del Románico (siglos IX y X), 89-94 (Barcelona: Museo Nacional de Arte de Cataluña). Ollich Castanyer, Inma; Montserrat Rocaf iguera Espona; and María Ocaña Subirana. 2016. ‘The Southern Carolingian Frontier in Marca Hispanica Along the River Ter: Roda Civitas and the Archaeological Site of L’Esquerda (Catalonia)’ in Neil Christie and Hajnalka Herold (eds.). Fortified Settlements in Early Medieval Europe: Defended Communities of the 8th-10th Centuries, 205-20 (Oxford: Oxbow). Olmo Enciso, Lauro. 1991. ‘Ideología y Arqueología: los estudios sobre el período visigodo en la primera mitad del siglo XX’ in Javier Arce Martínez and Ricardo Olmos Romera (eds.). Historiografía de la Arqueología y de la Historia Antigua en España (siglos XVIII-XX), 157-60 (Madrid: Ministerio de Cultura). Olmo Enciso, Lauro. 2000. ‘Ciudad y procesos de transformación social entre los siglos VI y IX: de Recópolis a Racupel’ in Luis Caballero Zoreda and Pedro Mateos Cruz (eds.). Visigodos y Omeyas. Un debate entre la Antigüedad tardía y la alta Edad Media, 385-400. AAEspA 23 (Madrid: CSIC). Olmo Enciso, Lauro. 2006. Recópolis. Un paseo por la ciudad visigoda (Alcalá de Henares: Museo Arqueológico Regional). Olmo Enciso, Lauro. 2007. ‘The Royal Foundation of Recópolis and the Urban Renewal in Iberia During the Second Half of the Sixth Century’ in Joachim Henning (ed.). Post-Roman Towns and Trade in Europe and Byzantium, 181-99 (Berlin: DeGruyter). Olmo Enciso, Lauro (ed.). 2008a. Recópolis y la ciudad en la época visigoda. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Olmo Enciso, Lauro. 2008b. ‘Recópolis: una ciudad en época de transformaciones’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 40-63. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Olmo Enciso, Lauro. 2008c. ‘Fuentes escritas y primeras investigaciones sobre Recópolis’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 22-39. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Olmo Enciso, Lauro. 2009. ‘La Vega Baja en época visigoda: una investigación arqueológica en construcción’ in Alfonso García and Diego Perís (eds.). La Vega Baja de Toledo, 69-94 (Toledo: Toletum Visigodo).

Bibliogr aphy

373

Olmo Enciso, Lauro. 2010. ‘Ciudad y Estado en época Visigoda: la construcción de un nuevo paisaje urbano’ in Alberto García, Ricardo Izquierdo Benito, Lauro Olmo Enciso and Diego Perís López (eds.). Espacios urbanos en el occidente mediterráneo, ss. VI-VIII, 87-112 (Toledo: Toletum Visigodo). Olmo Enciso, Lauro and Manuel Castro Priego. 2008. ‘La cerámica de época visigoda de Recópolis: apuntes tipológicos desde un análisis estratigráfico’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 89-96. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Olmo Enciso, Lauro; and Manuel Castro Priego. 2011. ‘La época visigoda a través de la arqueología’, in 711 Arqueología e historia entre dos mundos [Exhibition catalogue], 47-78 (Alcalá de Henares: Museo Arqueológico Regional). Olmo Enciso, Lauro; Manuel Castro Priego; Amaya Gómez de la Torre-Verdejo; and Álvaro Sanz Paratcha. 2008. ‘Recópolis y su justificación científica: la secuencia estratigráfica’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 64-75. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Orejas Saco, Almudena; and Francisco Javier Sánchez-Palencia. 2002. ‘Mines, Territorial Organization, and Social Structure in Roman Iberia: The Examples of Carthago Noua and the Peninsular Northwest’. American Journal of Archaeology 106.4, 581-99. Orfila Pons, Margarita. 1993. ‘Terra Sigillata Hispánica Meridional’. AEspA 66, 125-47. Orfila Pons, Margarita (ed.). 2000. El Fòrum de Pollentia. Memòria de les campanyes d’excavacions realitzades entre els anys 1996 i 1999 (Palma de Mallorca: Ajuntament d’Alcúdia). Orfila Pons, Margarita. 2008. ‘La vajilla Terra Sigillata Hispánica Tardía Meridional’ in Darío Bernal Casasola and Albert Ribera Lacomba (eds.). Cerámicas hispanorromanas: un estado de la cuestión, 401-11 (Cádiz: Universidad de Cádiz). Orfila Pons, Margarita; María Esther Chávez Álvarez; and Miguel Ángel Cau Ontiveros. 2006. ‘Pollentia and the Cities of the Balearic Islands’ in Lorenzo Abad Casal, Simon Keay, and Sebastián Ramallo Asensio (eds.). Early Roman Towns in Hispania Tarraconensis, 133-145. JRA, Supplementary Series 62 (Ann Harbour: JRA). Orihuela Uzal, Antonio. 2007. ‘La casa andalusí. Un recorrido a través de su evolución’ Atigrama 22, 299-335. Ouzoulias, Pierre; Christophe Pellecuer; Claude Raynaud; Paul Van Osse, and Pierre Garmy (eds.). (2001). Les campagnes de la Gaule à la fin de l’Antiquité (Antibes: Éditions APDCA). Ozcáriz Gil, Pablo. 2013. La administración de la provincia Hispania Citerior durante el alto imperio romano. Organización territorial, cargos administrativos y fiscalidad (Barcelona: Universidad de Barcelona). Pallarés Salvador, Francisco. 1969. ‘Las excavaciones en la plaza de San Miguel y la topografía romana de Barcino’. Cuadernos de Arqueología e Historia 13, 5-42. Palma, Félix. 1997. ‘Las casas romanas intramuros en Mérida. Estado de la cuestión’. Mérida. Excavaciones arqueológicas 1997. Memoria 3, 347-66. de Palol Salellas, Pedro. 1950. ‘Romanocristianos y visigodos (ensayo de síntesis históricoarqueológica)’. Ampurias 12, 239-41. de Palol Salellas, Pedro. 1958. ‘Las excavaciones de San Miguel del Arroyo. Un conjunto de necrópolis tardorromanas del valle del Duero’. Boletín del Seminario de Estudios de Arte y Arqueología 24, 209-17. de Palol Salellas, Pedro. 1964. ‘Cuchillo hispanorromano del siglo IV de J.C.’ in Boletín del Seminario de Estudios de Arte y Arqueología 30, 67-102.

374 

The Iberian Peninsul a bet ween 300 and 850

de Palol Salellas, Pedro. 1966. ‘Demografía y Arqueología hispánicas de los siglos IV-VIII’. Boletín del Seminario de Estudios de Arte y Arqueología 32, 5-66. de Palol Salellas, Pedro. 1967. Arqueología cristiana de la España Romana, siglos IV-VI (Valladolid: CSIC). de Palol Salellas, Pedro. 1986. ‘Las excavaciones del conjunto de “El Bovalar”, Seros (Segria, Lérida) y el reino de Akhila’. Antigüedad y Cristianismo 3, 513-26. de Palol Salellas, Pedro. 1989. El Bovalar. Conjunt d’època paleocristiana i visigótica (Lérida: Diputación de Lérida). de Palol Salellas, Pedro. 1994. Clunia. Historia de la ciudad y guía de las excavaciones (Burgos: Diputación Provincial de Burgos). de Palol Salellas, Pedro. 2004. El castrum del Puig de les muralles de Puig Rom (Roses, Alt Empordà) (Gerona: Museo Arqueológico de Cataluña). de Palol Salellas, Pedro; and Josep Guitart. 2000. Los grandes conjuntos públicos. El foro colonial de Clunia, Clunia VIII. 1 (Burgos: Diputación Provincial de Burgos) de Palol Salellas, Pedro; and Antoni Pladevall (eds.). 1999. Del romà al romanic. Història, art i cultura de la Tarraconense Mediterránea entre els segles IV i X. (Barcelona: Enciclopèdia Catalana). de Palol Salellas, Pedro; and Gisela Ripoll López. 1988. Los godos en el occidente europeo (Madrid: Ediciones Encuentro). de Palol Salellas, Pedro; Francesc Tuset Bertrán, and Javier Cortes Álvarez de Miranda. 1983. ‘Excavaciones en la iglesia visigoda de San Juan de Baños, Palencia’. Publicaciones de la Institución Tello Téllez de Meneses 49, 241-264. Palomino Lázaro, Ángel Luis; and MaríaNegredo García. 2011. ‘Arqueología de la transición en la Castilla del Ebro. El yacimiento “Peña del Mazo” en Pajares, Valle de Tobalina (Burgos)’ in Juan Antonio Quirós Castillo (ed.). Vasconia en la Alta Edad Media 450-1000. Poderes y comunidades rurales en el Norte Peninsular, 193-218 (Vitoria-Gasteiz: Universidad del País Vasco). Palomino Lázaro, Ángel Luis; María Negredo García; and Ramón Bohigas Roldán. 2012. ‘La fortaleza de Tedeja en Trespaderne y el castillo de Poza de la Sal (Burgos). Variables arqueológicas para el análisis de la articulación del poder local en el tránsito de la tardoantigüedad a la Alta Edad media en la Castilla del Ebro’ in Juan Antonio Quirós Castillo and José María Tejado Sebastián (eds.). Los castillos altomedievales en el noroeste de la Península Ibérica, 263-90 (Bilbao: Universidad del País Vasco). Panzram, Sabine. 2002. Stadtbild und Elite: Tarraco, Corduba und Augusta Emerita zwischen Republik und spätantike (Stutgart: University of Münster). Parga-Dans, Eva. 2010. ‘Commercial Archaeology in Spain: Its Growth, Development, and the Impact of the Global Economic Crisis’ in Nathan Schlanger, and Kenneth Aitchison (eds.). Archaeology and the Global Economic Crisis: Multiple Impacts, Possible Solutions, 45-54 (Tervuren: Culture Lab Editions). Parker Pearson, Mike. 1993. ‘The Powerful Dead: Archaeological Relationships Between the Living and the Dead’. Cambridge Archaeological Journal 3.2, 203-29. Pavón Maldonado, Basilio. 1992. Ciudades hispanomusulmanas (Madrid: MAPFRE). Paz Peralta, Juan Ángel. 1991. Cerámica de mesa romana de los siglos III al VI d.C. en la provincia de Zaragoza: (terra sigillata hispánica tardía, african red slip ware, sigillata gálica tardía y phocaean red slip ware) (Zaragoza: Institución Fernando el Católico). Peacock, David. 1982. Pottery in the Roman World: An Ethnoarchaeological Approach (London: Longman). Peña Cervantes, Yolanda. 2005-2006. ‘Producción de vino y aceite en los asentamientos rurales de Hispania durante la antigüedad tardía (s. IV-VII d.C.)’. CuPAUAM 31-32, 103-116.

Bibliogr aphy

375

Peña Cervantes, Yolanda. 2008. ‘La producción de vino en contextos eclesiásticos tardoantiguos hispanos’ in Juan Blázquez Miguel and Sebastián Celestino Pérez (eds.). El vino en época tardoantigua y medieval, 343-56. Serie Varia 8 (Madrid: Universidad Autónoma de Madrid). Perea Caveda, Alicia (ed.). 2001. El tesoro Visigodo de Guarrazar (Madrid: CSIC). Pérez González, Carmen; and Óscar Reyes Hernando. 2003. ‘Cauca: el pago de la tierra de Las Pizarras. Avance de la campaña de excavaciones del año 2000’. Sautuola 9, 213-27. Pérez Mínguez, Rafael. 2006. Aspectos del mundo rural romano en el territorio comprendido entre los ríos Turia y Palancia (Valencia: Diputación Provincial de Valencia). Pérez Rodríguez-Aragón, Fernando. 1992. ‘Los cingula militae tardorromanos de la Península Ibérica’. Boletín del Seminario de Estudios de Arte y Arqueología 58, 239-62. Pérez Rodríguez-Aragón, Fernando. 2014. ‘Los centros de producción de la Terra Sigillata Hispánica Tardía. Antiguos y nuevos centros, hornos, estructuras asociadas’. Oppidum 10, 147-76. Peytremann, Édith. 2003. Archéologie de l´habitat rural dans le nord de la France du IVe au XIIe siècle(Saint-Germain-en-Laye: AFAM). Peytremann, Édith. 2012. ‘The Archaeology of Early Medieval (6th-12th) Rural Settlements in France’. Arqueología de la Arquitectura 9, 213-30. Pinar Gil, Joan. 2012a. Accesorios de indumentaria del regnum visigodo temprano (siglos V-VI) (Bologna: University of Bologna). Pinar Gil, Joan. 2012b. ‘A Crossroads of Cultures in a Mosaic of Regions? The Early Visigothic Regnum from the Perspective of Small Finds.’ Archaeologia Baltica 18, 103-16. Pizarro Berengena, Guadalupe. 2012. El abastecimiento de agua a Córdoba. Arqueología e historia. PhD thesis (University of Córdoba). Pizzo, Antonio. 2008. ‘La fase final de la producción de vino en “Hispania” romana. Análisis de los principales conjuntos arquitectónicos’ in Juan Blázquez Miguel and Sebastián Celestino Pérez (eds.). El vino en época tardoantigua y medieval, 325-42. Serie Varia 8 (Madrid: Universidad Autónoma de Madrid). Pliego Vázquez, Ruth. 2009. La moneda visigoda I. Historia monetaria del Reino visigodo de Toledo (c. 569-711) (Seville: Universidad deSevilla). Pliego Vázquez, Ruth. 2012. ‘Gallaecia en tiempos del Reino Visigodo de Toledo: sus emisiones monetarias’ in Francisco Cebreiro Ares (ed.). Introducción a la Historia monetaria de Galicia (ss. II a.C.-XVII d.C.), 65-104 (Santiago de Compostela: Labirinto). Pliego Vázquez, Ruth. 2015-2016. ‘The Circulation of Copper Coins in the Iberian Peninsula During the Visigothic Period: New Approaches’. Journal of Archaeological Nummismatics 5/6, 125-60. Pohl, Walter. 1998. ‘Introduction: Strategies of Distinction’ in Walter Pohl and Helmut Reimitz (eds.). Strategies of Distinction: The Construction of Ethnic Communities, 300-800, 1-15 (Leiden: Brill). Pohl, Walter. 2013a. ‘Introduction: Strategies of Identif ication: A Methodological Prof ile’ in Walter Pohl and Gerda Heydemann (eds.). Strategies of Identification. Ethnicity and Religion in Early Medieval Europe, 1-64 (Turnhout: Brepols). Pohl, Walter. 2013b. ‘Christian and Barbarian Identities in the Early Medieval West: Introduction’ in Walter Pohl and Gerda. Heydemann (eds.). Post-Roman Transitions: Christian and Barbarian Identities in the Early Medieval West, 1-46 (Turhnout: Brepols). Ponsich, Michel. 1988. Aceite de oliva y salazones de pescado. Factores geo-económicos de Betica y Tingitania (Madrid: Complutense University). Porath, Yosef. 2002. ‘Hydraulic Plaster in Aqueduts as a Chronological Indicator’ in David Amit, Joseph Patrich and Yizhar Hirschfeld (eds.). The Aqueducts of Israel, 25-36. JRA Supplement Series 46 (Portsmouth: JRA).

376 

The Iberian Peninsul a bet ween 300 and 850

Poveda Arias, Pablo. 2015. ‘Relectura de la supuesta crisis del fin del reino visigodo de Toledo: una aproximación al reinado de Egica a través de sus fuentes legales’. Anuario de Historia del Derecho Español 85, 13-46. Prego de Lis, Augusto. 2000. ‘La inscripción de Comitiolus del Museo Municipal de Arqueología de Cartagena’ in V RACH, 383-92. Puig Verdaguer, Francisco; and Isabel Rodà de Llanza. 2007. ‘Las murallas de Barcino. Nuevas aportaciones al conocimiento de la evolución de sus sistemas de fortificación’ in Antonio Rodríguez Colmenero and Isabel Rodá de Llanza (eds.). Murallas de ciudades romanas en el occidente del imperio. Lucus Augusti como paradigma, 595-632 (Lugo: Diputación Provincial de Lugo). Puy Maeso, Arnald. 2014. ‘Land Selection for Irrigation in Al-Andalus, Spain (8th Century AD.)’. Journald of Field Archaeology 39.1, 84-100. Quevedo-Chigas, Elena. 1995. Early Medieval Iberian Architecture and the Hispanic Liturgy: A Study of the Development of Church Planning from the Fifth to the Tenth Centuries. PhD Thesis (New York University). Quintana López, Javier. 2017. El castro de Peña Amaya (Amaya, Burgos): del nacimiento de Cantabria al de Castilla. Documentos de Arqueología Cántabra 1 (Santander: Instituto Sautola). Quirós Castillo, Juan Antonio. 2007. ‘Las aldeas de los historiadores y de los arqueólogos en la Alta Edad Media en el norte peninsular’. Territorio, sociedad, poder. Revista de Estudios Medievales 2, 63-87. Quirós Castillo, Juan Antonio (ed.). 2009. Archaeology of Early Medieval Villages in Europe (Vitoria-Gasteiz: Universidad del País Vasco). Quirós Castillo, Juan Antonio. 2011a. ‘Los paisajes altomedievales en el País Vasco, 500-900. De la desarticulación territorial a la emergencia de los condados’ in Juan Antonio Quirós Castillo (ed.). Vasconia en la Alta Edad Media 450-1000, 29-54 (Vitoria-Gasteiz: Universidad del País Vasco). Quirós Castillo, Juan Antonio (ed.). 2011b. Vasconia en la Alta Edad Media 450-1000. Poderes y comunidades rurales en el Norte Peninsular (Vitoria-Gasteiz: Universidad del País Vasco). Quirós Castillo, Juan Antonio. 2012a. ‘Introducción’ in Juan Antonio Quirós Castillo and José María Tejado Sebastián (eds.). Los castillos altomedievales del cuadrante noroccidental de la Península Ibérica, 17-27 (Bilbao: Universidad del País Vasco). Quirós Castillo, Juan Antonio (ed.). 2012b. Arqueología del campesinado medieval: la aldea de Zaballa (Bilbao: Universidad del País Vasco). Quirós Castillo, Juan Antonio. 2013. ‘Identidades y ajuares en las necrópolis altomedievales. Estudios isotópicos del cementerio de San Martín de Dulantzi, Álava (siglos VI-X)’. AEspA 86, 215-232. Quirós Castillo, Juan Antonio. 2015a. Memoria 2015 del proyecto arqueológico de Pancorbo, Burgos. Excavation report MS deposited at the Servicio Territorial de Cultura de Burgos. Quirós Castillo, Juan Antonio. 2015b. Informe de la prospección geofísica en Peña Amaya. Excavation report MS deposited at the Servicio Territorial de Cultura de Burgos. Quirós Castillo, Juan Antonio; and Alberto Alonso Martín. 2007-2008. ‘Las ocupaciones rupestres en el fin de la Antigüedad. Los materiales cerámicos de Los Husos (Laguardia, Álava)’. Veleia 24-25, 1123-1142. Quirós Castillo, Juan Antonio; and Belén Bengoetxea Rementería. 2010 Arqueologia III. Arqueología medieval y postmedieval (Madrid: UNED). Quirós Castillo, Juan Antonio; and José María Tejado Sebastián (eds.). 2012. Los castillos altomedievales en el noroeste de la Península Ibérica. (Bilbao: Universidad del País Vasco). Quirós Castillo, Juan Antonio; and Alfonso Vigil-Escalera. 2006. ‘Networks of peasant villages between Toledo and Velegia Alabense, Northwestern Spain (V-Xth centuries)’. Archeologia Medievale 33, 79-128.

Bibliogr aphy

377

Quirós Castillo, Juan Antonio; and Alfonso Vigil-Escalera. 2011. ‘Dove sono i visigoti? Cimitieri e villaggi nella Spagna settentrionale nei secoli VI e VII’ in Carlo Ebanista and Marcelo Rotili (eds.). Archeologia e storia delle migrazioni: Europa, Italia, Mediterraneo fra tarda età romana e alto medioevo. Atti del Convegno internazionale di studi, Cimitile-Santa Maria Capua Vetere, 17-18 giugno 2010, 219-41 (Cimitile: Tavolario Edizioni). Ramallo Asensio, Sebastián; and Jaime Vizcaino Sánchez. 2007. ‘Evolución del sistema defensivo de Cartagena durante la Antigüedad’ in Antonio Rodríguez Colmenero and Isabel Rodà de Llanza (eds.). Murallas de ciudades romanas en el occidente del Imperio, 483-524 (Lugo: Museo Provincial de Lugo). Ramallo Asensio, Sebastián; Antonio Javier Murcia Muñoz; and Jaime Vizcaíno Sánchez. 2010. ‘Carthago Nova y su espacio suburbano. Dinámicas de ocupación en la periferia de la urbs’ in Desiderio Vaquerizo Gil (ed.). Las áreas suburbanas en la ciudad histórica. Topografía, usos, función, 211-54 (Córdoba: Universidad deCórdoba). Ramírez Sádaba, José Luis. 2008. ‘La primera epigrafía cristiana de Mérida’, in Antonio González, Agustín Velázquez (eds.). Los orígenes del cristianismo en Lusitania, 101-122. Cuadernos Emeritenses 34 (Mérida: MNAR). Rapp, Claudia. 2005. Holy Bishops in Late Antiquity: The Nature of Christian Leadership in an Age of Transition. The Transformation of the Classical Heritage 37 (Berkeley: University of California Press). Rascón Marqués, Sebastián. 1999. ‘La ciudad de Complutum en la tardoantigüedad: restauración y renovación’ in Luis García Moreno and Sebastián Rascón Marqués (eds.). Acta Antiqua Complutensia I: Complutim y las ciudades hispanas en la antigüedad tardía, 51-70 (Alcalá de Henares: Universidad deAlcalá). Rascón Marqués, Sebastián; and Ana Lucía Sánchez Montes. 2005. ‘Realidades cambiantes: Complutum tardoantiguo’ in VI RACH, 499-517. Rascón Marqués, Sebastián; and Ana Lucía Sánchez Montes. 2008. ‘Urbanismo de la ciudad de Complutum los siglos VI-VII’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 243-58. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Rascón Marqués, Sebastián; Antonio Méndez Madariaga; and Pedro Díaz del Río. 1991. La reocupación del mosaico del Auriga Victorioso (Villa romana de El Val). Un estudio de microespacio’. Arqueología, Paleontología y Etnografía 1, 181-200. Raynaud, Claude. 1993. ‘Céramique Estampée grise et orangée dite “derivée de sigillée paléochrétienne”‘. Lattara 6, 410-418. Rees, Roger. 2004. Diocletian and the Tetrarchy (Edinburgh: Edinburgh University Press). Regueras Grande, Fernando. 1990. ‘¿Osculatorios, removedores de perfumes, “ruecas votivas”? Sobre una nueva pieza hallada en Villafuerte (Valladolid) y algunas reflexiones en torno a este tipo de útiles’. Numantia 3, 175-194. Reinhart Wilhelm. 1945. ‘Sobre el asentamiento de los visigodos en la Península’. AEspA 18, 124-139. Reinhart, Wilhelm. 1952. La necrópolis de Duratón’. AEspA 25, 193-195. Rėklaitytė, Ieva. 2012. Vivir en una ciudad de al-Ándalus. Hidráulica, saneamiento y condiciones de vida. Monografías arqueológicas 47 (Zaragoza: Universidad deZaragoza). Remolá Vallverdú, Josep Antón. 2004. ‘Arquitectura funeraria’ in Xavier Dupré Raventós (ed.). Tarragona. Colonia Iulia Urbs Triumphalis Tarraco, 83-96. Capitales provinciales de Hispania 3 (Rome: L’erma di Bretschneider). Remolà Valverdú, Josep Anton; and Meritxell Pérez Martínez. 2013. ‘Centcelles y el praetorium del comes Hispaniarum Asterius en Tarraco’. AEspA 86, 161-86.

378 

The Iberian Peninsul a bet ween 300 and 850

Revilla Calvo, Víctor. 1995. ‘Producción artesanal, viticultura y propiedad rural en la Hispania Tarraconense’. Gerión 13, 305-38. Reynolds, Andrew; and Juan Antonio Quirós Castillo. 2010. Memoria final del proyecto arqueológico de Aistra (Zalduondo, Álava). Excavation report MS deposited at the Diputción Provincial de Ávila. Reynolds, Paul. 1987. El yacimiento tardorromano de Lucentum (Benalúa-Alicante): las cerámicas finas (Alicante: Universidad deAlicante). Ribeiro, Maria do Carmo Franco. 2008. Braga entre a época romana e a Idade Moderna. Uma metodologia de análise para a leitura da evolução da paisagem urbana. PhD thesis (University of Minho). Ribera Lacomba, Albert. 2005. ‘Origen i desenvolupament del nucli episcopal de València’, in VI RACH, 207-243. Ribera Lacomba, Albert. 2008. ‘La ciudad de Valencia durante el periodo visigodo’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 303-20. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Ribera Lacomba, Albert; and Miquel Rosselló Mesquida. 2000. La ciudad de Valentia en época visigoda’ in Albert Ribera Lacomba (ed.). Los orígenes del cristianismo en Valencia y su entorno, 151-64 (Valencia: Ayuntamiento de Valencia). Ribera Lacomba, Albert; and Miquel Rosselló Mesquida. 2013. ‘La ocupación tardoantigua del circo romano de Valentia’ in Olof Brandt (ed.). Acta XV Congressus internationalis archaeologiae christianae Toleti, 8-12.9.2008: episcopus, civitas, territorium, 47-62. Studi di antichità cristiana 65 (Vatican City: Pontificio Istituto di Archeologia Cristiana). Richard, Julian. 2013. Water for the City, Fountains for the People: Monumental Fountains in the Roman East. Studies in Eastern Mediterranean Archaeology 9 (Turnhout: Brepols). Rico Camps, Daniel. 2014. ‘Inscripciones monumentales del siglo VIII (de Cangas a Pravia)’. Territorio, Sociedad y Poder 9, 67-98. Rico, Christian. 2005. ‘Vingt ans de recherches sur les mines et les metallurgies romaines en peninsule Iberique (1985-2004)’. Pallas 67, 217-39. Riera Rullán, Mateu. 2013. ‘El monasterio de la isla de Cabrera (Islas Baleares. Siglos V-VIII d.C.). Testimonios arqueológicos de los monjes reprobados por el papa Gregorio Magno’. Hortus Artium Medievalium 19, 47-61. Riera Rullán, Mateu. 2014. El monestir de Cabrera (Segles V-VIII dC) (Palma de Mallorca: Ajuntament de Palma). Riera Rullán, Mateu; Miguel Ángel Cau Ontiveros, and Magdalena Salas Burguera (eds.). 2012. Cent anys de Son Peretó: descobrint el passat cristià (Palma de Mallorca: Consell de Mallorca). Riera Rullán, Mateu; Margarita Orfila Pons; and Miguel Ángel Cau Ontiveros. 1999. ‘Els últims segles de Pollentia’. Bolletí de la Societat Arqueològica Lul·liana 55, 335-46. Riess, Frank. 2013. Narbonne and its Territory in Late Antiquity: From the Visigoths to the Arabs (London: Ashgate). Rigoir, Jacqueline. 1968. ‘Les sigillées paléochrétiennes grises et orangées’. Gallia 26, 177-244. Ripoll López, Gisela. 1985. La necrópolis visigoda de El Carpio de Tajo (Toledo). Excavaciones Arqueológicas de España 142. (Madrid: Ministerio de Cultura). Ripoll López, Gisela. 1989. ‘Características generales del poblamiento y la Arqueología funeraria visigoda de Hispania’. Espacio, Tiempo y Forma 2, 389-418. Ripoll López, Gisela. 1991. La ocupación visigoda en época romana a través de sus necrópolis (Hispania) (Barcelona: University of Barcelona). Ripoll López, Gisela. 1998. Toréutica de la Bética (siglos VI y VII d.C) (Barcelona: Real Academia de las Buenas Letras).

Bibliogr aphy

379

Ripoll López, Gisela. 2007. ‘Las necrópolis visigodas’ in Hispania Gothorum: San Ildefonso y el reino visigodo de Toledo, 59-74 (Toledo: Museo de Santa Cruz). Ripoll López, Gisela. 2000. ‘Sedes regiae en la Hispania de la Antigüedad Tardía’ in Gisela Ripoll López and Josep María Gurt Esparraguera (eds.). Sedes Regiae (ann. 400-800), 371-402 (Barcelona: Real Academia de las Buenas Letras). Ripoll López, Gisela. 2002. ‘The Transformations of the City of Barcino During Late Antiquity’ in Julia Beltrán de Heredia (ed.). From Barcino to Barchinona (1st to 7th Centuries: The Archaeological Remains of the Plaça del Rei in Barcelona, 34-47 (Barcelona: Museo de Historia de la Ciduad). Ristow, Sebastian. 2009. ‘Das Kölner Baptisterium am Dom und die frühchristlichen Tauforte nördlich der Alpen’ in Ulrich Krings and Reiner Will (eds.). Das Baptisterium am Dom. Kölns erster Taufort, 22-44 (Cologne: Grewen Verkig). Rivera Jiménez, Timoteo; and Juan Aurelio Pérez Macías. 2010. ‘Metalurgia de época visigoda en la mina Zarina (Cala, Huelva)’ in Juan Aurelio Pérez Macías and Juan Luis Carriazo Cenago (eds.). Estudios de minería medieval en Andalucía, 31-8 (Huelva: Universidad de Huelva). Rivera Manescau, Saturnino. 1939. ‘La necrópolis visigoda de Simancas’. Boletín del Seminario de Estudios de Arte y Arqueología 5, 7-20. Rodríguez Colmenero, Antonio. 2011. Lucus Augusti. La ciudad romano-germánica del Finisterre ibérico. Génesis y evolución histórica (14 a.C. – 711 d.C.) (Lugo: Concello de Lugo). Rodríguez Colmenero, Antonio; and Isabel Rodà de Llanza (eds.). 2007. Murallas de ciudades romanas en el occidente del Imperio (Lugo: Museo Provincial de Lugo). Rodríguez Gutiérrez, Oliva. 2011. Hispania Arqueológica. Panorama de la Cultura Material de las Provincias Hispanorromanas (Seville: Universidad de Sevilla). Rodríguez Untoria, Santiago; and CarlosBarrio Aldea. 2003. ‘Poblamiento tardorromano en el valle medio del río Guadarrama (Toledo)’. Bolskan 20, 267-75. Roig Buxó, Jordi. 2009. ‘Asentamientos rurales y poblados tardoantiguos y altomedievales en Cataluña (siglos VI al X)’ in Juan Antonio Quirós Castillo (ed.). Archaeology of Early Medieval Villages in Europe, 207-51 (Vitoria-Gasteiz: Universidad del País Vasco). Roig Buxó, Jordi. 2015. ‘Necrópolis de época visigoda, ajuares funerarios y depósitos humanos anómalos de los s. V-VIII en la Tarraconense oriental (Cataluña): ¿indicadores de etnicidad y/o nivel económico? e indicios arqueológicos de desigualdad y exclusión social’ in Juan Antonio Quirós Castillo and Santiago Castellanos Martínez (eds.). Identidad y etnicidad en Hispania. Propuestas teóricas y cultura material en los siglos V-VIII, 333-394 (Bilbao: Universidad del País Vasco). Roig Buxó, Jordi; and Jaume Coll Conesa. 2011. ‘Esquelets humans en sitges, pous i abocadors als assentaments rurals i vilatges de l’antiguitat tardana de Catalunya (segles V-VIII)’ in Isabel Fernández del Moral, Joan Menchón Fortí and Josep Vila Carabasa (eds.). Actes del IV Congrès d’Arqueologia Medieval i Moderna a Catalunya (Tarragona 2010) (2 vols), vol. I, 75-82 (Tarragona: Ayuntamiento de Tarragona) Roig Buxó, Jordi; and Joan-Manel Coll Riera. 2010. ‘El jaciment de la Plaça Major de Castellar del Vallès: de l’assentament del Neolític al vilatge de l’Antiguitat Tardana. 5000 anys d’evolució histórica’. Recerca 7, 77-108. Rojas Rodríguez-Malo, Juan Manuel. 2013. ‘El primer año de trabajos en Guarrazar. La confirmación de un importante yacimiento arqueológico’ in VII Jornadas de Cultura Visigoda, 37-66 (Guadamur: Ayuntamiento de Guadamur). Rojas Rodríguez-Malo, Juan Manuel. 2014. ‘Guarrazar: Arqueología y nuevos recursos. Un proyecto con un siglo y medio de retraso’ in VI Jornadas de Cultura Visigoda, 31-52 (Guadamur: Ayuntamiento de Guadamur).

380 

The Iberian Peninsul a bet ween 300 and 850

Rosen, William. 2007. Justinian’s Flea: The First Great Plague and the Fall of the Roman Empire (London: Penguin). Rosselló Bordoy, Guillermo. 1986. ‘Algunas puntualizaciones sobre el ataifor andadusi: tipología y cronología’. CuPAUAM 13, 281-9. Rosselló Mesquida, Miguel. 2005. ‘El territorium de Valentia a l’antiguitat tardana’ in VI RACH, 279-304. Rosselló Mesquida, Miquel. 2000. ‘Economía y comercio en época visigoda’, in Albert Ribera Lacomba (ed.). Los orígenes del cristianismo en Valencia y su entorno, 207-17 (Valencia: Ayuntamiento de Valencia). Rosselló Mesquida, Miquel; and Empar Juan Navarro. 2003. ‘Yacimientos no urbanos en época visigoda (ss. VI-VII)’ in Helena Bonet Rosado, Rosa Albiach Descals, and Manuel Gozalbes Fernández (eds.). Romanos y visigodos en tierras valencianas, 175-84 (Valencia: Diputación Provincial de Valencia). Rosser, Pablo. 1990. ‘Nuevos descubrimientos arqueológicos en el término municipal de Alicante’, in Enrique Llobregat Conesa and Lorenzo Abad Casal (eds.). Historia de la ciudad de Alicante (4 vols), vol. I, 189-285 (Alicante: Ayuntamiento de Alicante). Rosser Limiñana, Pablo. 1993. ‘La ciudad de Alicante y la arqueología del poblamiento en época medieval islámica’. Lqnt 2, 11-34. Rotolo, Antonio. 2014. ‘Drainage Galleries in the Iberian Peninsula During the Islamic Period’. Water History 6.3, 191-2010. Rubio Díez, Rubén. 2015. Arqueología, paisaje y territorio post-romano. Las tumbas excavadas en roca en el occidente del campo de Ciudad Rodrigo (Salamanca) (Ciudad Rodrigo: Centro de Estudios Mirobrigenses). Saguì, Lucia. 2007. ‘Glass in Late Antiquity: The Continuity of Technologi and Sources of Supply’ in Luke Lavan, Enrico Zanini and Alexander Sarantis (eds.). Technology in Transition A.D. 300-650, 367-78. Late Antique Archaeology 4 (Leiden: Brill). Salado Escaño, Juan Bautista; Ildefonso Navarro Luengo; Ana Arancibia Román; José Francisco Mayorga Mayorga; and José Antonio Rambla Torralvo. 2002. ‘Evolución urbana de la Málaga islámica. Siglos VIII-XV’ in Antonio Torremocha Silva and Virgilio Martínez Enamorado (eds.). II Congreso internacional de la ciudad en a-Ándalus y el Magreb, 361-90 (Cádiz: El Legado Andalusí). Salcedo Hierro, Miguel. 2000. La mezquita, catedral de Córdoba (Córdoba: Cajasur). Salvatierra Cuenca, Vicente. 2013. ‘La primera arqueología medieval española. Análisis de un proceso frustrado (1844-1925)’. Studia histórica. Historia Medieval 31, 183-210. Salzman, Michele Renee. 2002. The Making of an Aristocracy: Social and Religious Change in the Western Roman Empire (Cambridge, Mass: Harvard University Press). Sanabria Murillo, Diego; Fernando Sánchez Hidalgo; Víctor Gibello Bravo; and Andrea Menéndez Menéndez. 2012. ‘La domus suburbana del solar del antigüo cuartel de Hernán Cortés (Mérida, Badajoz). Un ejemplo de arquitectura residencial en la tardoantigüedad’ unpublished presentation given at Iº Congresso Internacional sobre Arqueologia de Transição: Entre o Mundo Romano e a Idade Média (Évora: Universidade de Évora). Sánchez Albornoz, Claudio. 1966. Despoblación y repoblación del valle del Duero (Buenos Aires: Instituto de Historia de España). Sánchez Barrero, Pedro Dámaso. 2004. ‘El territorio’ in Xavier Dupré Raventós (ed.). Mérida. Colonia Augusta Emerita, 101-12. Capitales provinciales de Hispania 2 (Rome: L’erma di Bretschneider). Sánchez de Prado, María Dolores and Márioda Cruz. 2014. ‘La producción de vidrio en Hispania. Una versión actualizada’ in José María Álvarez Martínez, Trinidad Nogales Basarrate, and

Bibliogr aphy

381

Isabel Rodà de Llanza (eds.). Actas del XVIII Congreso Internacional de Arqueología Clásica, 1393-7 (Mérida: MNAR). Sánchez de Prado, María Dolores. 2009. ‘La vajilla de vidrio durante la Antigüedad Tardía en el conventus carthaginensis’. Boletín del Seminario de Estudios de Arte y Arqueología 75, 159-200. Sánchez López, Elena; and Javier Martínez Jiménez. 2016. Los acueductos de Hispania. Construcción y abandono (Madrid: Fundación Juanelo Turriano). Sánchez Montes, Ana Lucía; and Sebastián Rascón Marqués. 2006. ‘La villa del Val y la necrópolis del Camino de los Afligidos (Alcalá de Henares)’ in Jorge Morín de Pablos (ed.). La investigación arqueológica de la época visigoda en la comunidad de Madrid, 293-305. Zona arqueológica 8, II (Alcalá de Henares: Museo Arqueológico Regional). Sánchez Pardo, José Carlos. 2010. ‘Poblamiento rural tardorromano y altomedieval en Galicia (ss.V-X). Una revisión arqueológica’. Archeologia Medievale 37, 285-306. Sánchez Pardo, José Carlos. 2013. ‘Power and Rural Landscapes in Early Medieval Galicia (400-900 AD): Towards a Re-Incorporation of the Archaeology into the Historical Narrative’. EME 21.2, 140-68. Sánchez Pardo, José Carlos. 2015. ‘Power Strategies in the Early Medieval Churches of Galicia (711-910 AD)’ in José Carlos Sánchez Pardo and Mary Shapland (eds.). Churches and Social Power in Early Medieval Europe: Integrating Archaeological and Historical Approaches, 227-68 (Turnhout: Brepols). Sánchez Pardo, José Carlos; Rebeca Blanco Rotea; Jorge Sanjurjo Sánchez. 2017. ‘The Church of Santa Comba de Bande and Early Medieval Iberian Architecture: New Chronological Results’. Antiquity 91.358, 1011-102. Sánchez Ramos, Isabel. 2005. ‘Las necrópolis de Corduba durante la Antigüedad tardía’. Anales de Prehistoria y Arqueología de la Universidad de Murcia 22, 165-177. Sánchez Ramos, Isabel. 2007. ‘La cristianización de las necrópolis de Corduba. Fuentes escritas y testimonios arqueológicos’. AEspA 80, 191-206. Sánchez Ramos, Isabel. 2010. ‘Las ciudades de la Bética en la antigüedad tardía’. Antiquité Tardive 18, 243-73. Sánchez Ramos, Isabel. 2014. Topografía cristiana de las ciudades hispanas durante la Antigüedad tardía. BAR IS 2606 (Oxford: Archaeopress). Sánchez Velasco, Jerónimo. 2006. Elementos arquitectónicos de época visigoda en el Museo Arqueológico de Córdoba (Córdoba: Junta de Andalucía). Sánchez Fuentes, José Javier. 2008. ‘Vite praetexeris tarraconensis. Aproximación a la producción y consumo de vino en Tarraco durante la Antigüedad Tardía’ in Juan Blázquez Miguel and Sebastián Celestino Pérez (eds.). El vino en época tardoantigua y medieval, 61-74. Serie Varia 8 (Madrid: Universidad Autónoma de Madrid). Sánchez-Palencia, Jaime; and María Jesús Sainz. 2002. ‘El circo romano de Toletum’, in Trinidad Nogales Basarrate and Jaime Sánchez-Palencia (eds.). El circo en la Hispania romana, 97-115 (Mérida: MNAR). Santillana, David. 1926. Istituzioni di diritto musulmano malichita con riguardo anche al sistema sciafiita (Rome: ARE). Sanz Paratcha, Álvaro. 2008. ‘Vida después de la muerte: los contextos cerámicos de Recópolis en época emiral’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 165-79. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Saradi, Hélène. 2006. The Byzantine City in the Sixth Century: Literary Images and Historical Reality (Athens: Society of Messenian Archaeological Studies). Sarasola Etxegoien, Nerea; and Alfredo Moraza Barea. 2011. Erdi Aroko Arkeologia Gipuzkoa. Arqueología Medieval en Gipuzkoa (San Sebastián: Diputación Foral de Gipuzkoa).

382 

The Iberian Peninsul a bet ween 300 and 850

Sarr, Bilal. 2010. ‘La Granada prezirí (siglos VIII-XI): ¿qal’a, ḥiṣn o madīna?’ in Philipp Sénac (ed.). Villa 3: Histoire et archéologie des sociétés de la vallée de l’Ebre (VIIe-XIe), 381-400 (Toulouse: CNRS) Sasse, Barbara. 2000. “Westgotische” gräberfelder auf der Iberischen Halbinsel (Mainz am Rhein: von Zabern). Sastre Blanco, José Carlos; and Raúl Catalán Ramos. 2012. ‘Un asentamiento fortificado en la tardoantigüedad: el castro de El Castillón (Santa Eulalia de Tábara, Zamora)’ in Juan Antonio Quirós Castillo and José María Tejado Sebastián (eds.). Los castillos altomedievales en el noroeste de la Península Ibérica, 193-211 (Bilbao: Universidad del País Vasco). Sastre Blanco, José Carlos; and Patricia Fuentes Melgar. 2011. ‘Late Roman Metallurgy in Castro of El Castillón (Santa Eulalia de Tábara, Zamora)’ in David Hernández de la Fuente (ed.). New Perspectives on Late Antiquity, 229-44 (Cambridge: Cambridge Scholars Publishing). Sastre de Diego, Isaac. 2005. ‘Arquitectura tardoantigua en Mérida: el mobiliario litúrgico contextualizado en los edificios cristianos’ in VI RACH, 465-74. Sastre de Diego, Isaac. 2011. Los primeros edificios cristianos de Extremadura. Sus espacios y elementos litúrgicos. Serie Ataecina 5 (Mérida: Asamblea de Extremadura). Sastre de Diego, Isaac. 2013. Los altares de las iglesias hispanas tardoantiguas y altomedievales. BAR IS 2503 (Oxford: Archaeopress). Sastre de Diego, Isaac. 2015. Mérida capital cristiana. De Roma a al-Andalus. Cuadernos Emeritenses 41 (Mérida: MNAR). Sastre de Diego, Isaac; and Javier Martínez Jiménez. 2012. Informe de las excavaciones en Casa Herrera. Campaña del XVI Curso Internacional de Arqueología de Mérida (2012). Excavation report MS deposited at the Consorcio de Mérida. Sastre de Diego, Isaac; and Javier Martínez Jiménez. 2013. Informe de las excavaciones en Casa Herrera. Campaña del XVII Curso Internacional de Arqueología de Mérida (2013). Excavation report MS deposited at the Consorcio de Mérida. Sastre de Diego, Isaac; and María Ángeles Utrero Agudo. 2015. ‘Tracing Influences in Mozarabic Material Culture: Building Technology in 8-10th Century Hispanic Churches’ in Robet Hoyland (ed.). The Late Antique World of Early Islam: Muslims Among Jews and Christians in the East Mediterranean, 195-229. Studies in Late Antiquity and Early Islam 25 (Princeton: Darwin Press). Schlunk, Helmut; and Theodor Hauschild. 1978. Die Denkmäler der frühchristlichen und westgotischen Zeit (Mainz: Philipp von Zabern). Schneider, Laurent. 2001. ‘Oppida et castra tardo-antiques à propos des établissements de hauteur de la Gaule Méditerranéenne’ in Pierre Ouzoulias, Christophe Pellecuer, Claude Raynaud, Paul Van Ossel and Pierre Garmy (eds.). Les campagnes de la Gaule à la fin de l’Antiquité, 433-48 (Antibes: Éditions APDCA). Schreg, Rainer. 2012. ‘Farmsteads in Early Medieval Germany: Architecture and Organisation’. Arqueología de la arquitectura 9, 249-67. Sénac, Philippe. 2013. ‘Au nord d’al-Andalus. La conquête de la Narbonnaise et de la Tarraconise (VIIIe siècle)’ in Xavier Ballestín Navarro and Ernesto Pastor Díaz (eds.). Lo que vino de Oriente. Horizontes, praxis y dimensión material de los sistemas de dominación fiscal en Al-Andalus (ss. VII-IX), 122-32. BAR IS 2525 (Oxford: Archaeopress). Sevillano Fuertes, María Ángeles; and Julio Vidal Encinas. 2002. Urbs Magnifica. Una aproximación a la Arqueología de Asturica Augusta (Astorga, León) (León: Museo Romano). Shortland, Andrew; Lukas Schachner; Ian Freestone; and Michael Tite. 2006. ‘Natron as Flux in the Early Vitreous Material Industry: Sources, Beginnings and Reasons for Decline’. Journal of Archaeological Science 33, 211-31.

Bibliogr aphy

383

Sillières, Philipp. 1990. Les Voies de comunication de l’Hispanie méridionele (Paris: Boccard). da Silva Fernandes, Luis. 2011. ‘Placas funerarias decoradas del conventus Emeritensis. Rutas de difusión de un modelo’, in Trinidad Nogales Basarrate and Isabel Rodà de Llanza (eds.). Roma y las provincias: modelo y difusión (2 vols), vol. I, 671-80 (Roma-Mérida: L’erma di Bretschneider). da Silva, Rodrigo Banha. 2011. ‘Olisipo’ in Jesús Acero Pérez and Josep Anton Remolà Vallverdú (eds.). La gestión de los residuos urbanos en Hispania. Xavier Dupré Raventós (1956-2006) in memoriam, 203-15. AAEspA 60 (Mérida: CSIC). Sirks, Boudewijn. 1993. ‘Reconsidering the Roman Colonate’. Zeitschrift der Savigny-Stiftung für Rechtsgeschichte. Romanistische Abteilung 110.1, 331-69. Sotomayor Muro, Manuel. 1975. Sarcófagos romano-cristianos de España. Estudio iconográfico (Granada: Universidad deGranada). Sotomayor Muro, Manuel. 1988. ‘Sarcófagos palecristianos en Murcia y zonas limítrofes’. Antigüedad y Cristianismo 5, 165-83 (Murcia: University of Murcia). Souto Lasala, Juan Antonio. 2001. ‘Arquitectura y estado omeya’ in Antonio Rodríguez Jiménez (ed.). El esplendor de los Omeyas cordobeses, 29-31 (Córdoba: El Legado Andalusí). Souto Lasala, Juan Antonio; and Dolors Bramón Planas. 1987. Las maravillas de Zaragoza’ Aragón en la Edad Media 7, 7-26. Stern, Marianne. 2008. ‘Glass Production’ in John Peter Oleson (ed.). The Oxford Handbook of Engineering and Technology in the Ancient World, 520-50 (Oxford: Oxford University Press). Steuer, Heiko. 1989. ‘Archaeology and History: Proposals on the Social Structure of the Merovingian Kingdom’ in Klaus Randsborg (ed.). The Birth of Europe: Archaeology and Social Development in the First Millennium A.D., 100-22 (Rome: L’erma di Bretschneider). Stirling, Lea. 2012. ‘Patrons, Viewers, and Statues in Late Antique Baths’, in Stine Birk and Birte Poulsen (eds.). Patrons and Viewers in Late Antiquity. Aarhus Studies in Mediterranean Antiquity 10 (Aarhus: Aarhus University Press). Strato. 2004. ‘Trabajos arqueológicos en el yacimiento de Castroventosa vinculados a las tareas de restauración de la muralla, en Cacabelos (León)’. Excavation report MS deposited at the Servicio Territorial de Cultura de León. Strato. 2010. Excavación en Área en el yacimiento “Canto Blanco”. Excavation report MS deposited at the Servicio Territorial de Cultura de León. Swift, Ellen. 2000. Regionality and Dress Accessories in the Late Roman West. Monographies Instrumentum 11 (Mentagnac: Monique Mergiol). Tabales Rodríguez, Miguel Ángel. 2001. ‘Algunas aportaciones arqueológicas para el conocimiento urbano de Hispalis’. Habis 32, 387-423. Tarrats Bou, Francesc; Josep Anton Remolà Vallverdú, and Jacinto Sánchez Gil de Montes. 1998. ‘La vil·la romana dels Munts (Altafulla, Tarragonès) i Tarraco’. Empuries 51, 197-225. Teichner, Felix. 2001. ‘Uma nova interpretação da área 21, a partir da planta elaborada por Sebastião Philippe Martins Estácio da Veiga, sobre a villa romana de Milreu (Estói, Algarve)’. O Arqueólogo Português Serie IV, 19, 187-98. Teja Casuso, Ramón; and Silvia Acerbi. 2010. ‘El palacio visigodo y el circo de Toledo: hipótesis de localización’. Reti Medievali 11, 81-6. Tejado Sebastián, José María. 2011. Arqueología y gestión del territorio en el alto Valle del Iregua: El castro de “El Castillo de los Monjes” (Lumbreras, La Rioja). PhD thesis (University of La Rioja). Tejado Sebastián, José María. 2012. ‘Castros altomedievales en el alto Iregua (La Rioja): el caso de El Castillo de los Monjes’ in Juan Antonio Quirós Castillo and José María Tejado Sebastián

384 

The Iberian Peninsul a bet ween 300 and 850

(eds.). Los castillos altomedievales en el noroeste de la Península Ibérica,163-192 (Bilbao: Universidad del País Vasco). Tejerizo García, Carlos. 2011. ‘Ethnicity in Early Middle Age Cemeteries: The Case of the “Visigothic” Burials’. AyTM 18, 29-43. Tejerizo García, Carlos. 2012a. ‘Identidad nacional y arqueología en el primer franquismo. Julio Martínez Santa-Olalla y la arqueología de época visigoda’ in José Manuel Aldea Celada, Paula Ortega Martínez, Iván Pérez Miranda, María Reyes de Soto García (eds.). Historia, identidad y alteridad: Actas del III Congreso Interdisciplinar de Jóvenes Historiadores, 479-502 (Salamanca: Antema). Tejerizo García, Carlos. 2012b. ‘Early Medieval Household Archaeology in Northwest Iberia (6th-11th centuries)’. Arqueología de la arquitectura 9, 183-96. Tejerizo García, Carlos. 2013. ‘La arqueología de las aldeas altomedievales en la cuenca del Duero (ss. V-VIII): problemas y perspectivas’. Debates de Arqueología Medieval 3, 289-315. Tejerizo García, Carlos. 2014a. ‘Estructuras de fondo rehundido altomedievales en la Península Ibérica’. Munibe 65, 215-237. Tejerizo García, Carlos. 2014b. Informe de la segunda campaña de prospección en el entorno de la aldea altomedieval de La Mata del Palomar (Nieva, Segovia). Excavation report MS deposited at Servicio Territorial de Cultura de Segovia. Tejerizo García, Carlos. 2015. Arqueología del campesinado medieval en la cuenca del Duero (ss.V-VIII d.C.). PhD thesis (University of the Basque Country). Tejerizo García, Carlos. 2016. ‘The End of the World as We Know It: Post-Imperial Social Lanscapes in North-Central Iberia (5th-6th centuries)’. Archeologia Medievale 43, 383-97. Tejerizo García, Carlos. 2017. Arqueología de las sociedades campesinas en la cuenca del Duero durante la Primera Alta Edad Media (Bilbao: Universidad del País Vasco). Tejerizo García, Carlos; Álvaro Carvajal Castro; Carlos Marín Suárez; Cristina Martínez Álvarez, and Rafael Mansilla Hortigüela. 2015. ‘La construcción histórica de los paisajes en el sector central de la cuenca del Duero. Primeros resultados de una prospección intensiva’. Territorio, Sociedad y Poder 10, 39-62. Tente, Catarina. 2009. ‘Viver em autarcia. A organização do alto Mondego (Portugal) entre os séculos V a X’ in Iñaki Martín Viso (ed.). ¿Tiempos oscuros? Territorios y sociedad en el centro de la Península Ibérica (siglos VII-X), 137-57 (Salamanca: Universidad de Salamanca). Tipper, Jess. 2004. The Grubenhaus in Anglo-Saxon England (Yedingham: Landscape Research Centre). Torremocha Silva, Antonio. 2001. ‘El fenómeno urbano portuario en el Estrecho’ in Antonio Torremocha Silva and Virgilio Martínez Enamorado (eds.). II Congreso internacional de la ciudad en al-Ándalus y el Magreb, 293-326 (Cádiz: Universidad deCádiz). Torremocha Silva, Antonio. 2002. ‘Algeciras y el Estrecho de Gibraltar en época islámica’ in Ceuta en el Medievo: la ciudad en el universo árabe, 199-252 (Ceuta: Instituto de Estudios Ceutíes). Torres Balbás, Leopoldo. 1957. ‘Arte hispanomusulmán hasta la caída del califato de Córdoba’ in Évariste Lévi-Provençal (ed.) España musulmana hasta la caída del califato de Córdoba (711-1031 de J. C.). Instituciones y vida social e intelectual. Historia de España de Ramón Menéndez Pidal V (Madrid: Espasa-Calpe). Toubert, Pierre. 1973. Le Latium méridional et la Sabine du IXe à la fin du XIIe siècles (Rome: Bibliothèque de l’École Française de Rome). Trigger, Bruce. 1995. ‘Romanticism, Nationalism and Archaeology’ in Philip Kohl and Clare Fawcett (eds.). Nationalism, Politics, and the Practice of Archaeology, 263-79 (Cambridge: Cambridge University Press).

Bibliogr aphy

385

Trillo San José, Carmen. 2006. ‘La alquería y su territorio en al-Andalus: estrategias sociales de organización y conservación’ in Almudena Orejas Saco (ed.). Arqueología espacial: espacios agrarios, 243-62. Arqueología Espacial 26 (Teruel: Seminario de Arqueología y Etnología Turolense). Trillo San José, Carmen. 2011. ‘Mezquitas en al-Andalus: un espacio entre las comunidades y el poder’. Studia Historica. Historia Medieval 29, 73-98. Ucko, Peter. 1969. ‘Ethnography and Archaeological Interpretation of Funerary Remains’. World Archaeology 1.2, 262-80. Ulbert, Thilo. 1968. ‘El Germo. Kirche und Profanbau aus dem frühen 7. Jahrhundert’. Madrider Mitteilungen 9, 329-98. Ulbert, Thilo. 2003. ‘El yacimiento paleocristiano de Son Fadrinet (Campos, Mallorca)’. Mayurqa 29, 173-87. Unzu Urmeneta, Mercedes; Paz Prieto; María Jesús Pérex Agorreta; and Amparo Hernando Grande. 2006. ‘Roman Baths in Pamplona (Navarra, Spain)’ in Gilbert Wiplinger (ed.). Cura aquarum in Ephesus (2 vols), vol. 2, 431-36 (Leuven: Peeters). Urteaga Artigas, Mercedes. 1997. ‘Minería Romana en Gipúzcoa’. Isturitz 8, 491-515. Urteaga Artigas, Mercedes. 2008. ‘El Vasconum saltus y Oiasso’. Boletín Arkeolan 15, 171-84. Utrero Agudo, María Ángeles. 2006. Iglesias tardoantiguas y altomedievales en la Península Ibérica. Análisis arqueológico y sistemas de abovedamiento. AAEspA 40 (Madrid: CSIC). Utrero Agudo, María Ángeles; and Isaac Sastre de Diego. 2012. ‘Reutilizando materiales en las construcciones de los siglos VII-X. ¿Una posibilidad o una necesidad?’. Anales de historia del arte 22, 309-23. Valenti, Marco. 2004. L’insediamento altomedievale nelle campagne toscane: paesaggi, popolamento e villaggi tra VI e X secolo (Florencia: All’Insegna del Giglio). Valenti, Marco. 2007. ‘Architecture and Infraestructure in the Early Medieval Village: The Case of Tuscany’ in Luke Lavan, Enrico Zanini and Alexander Sarantis (eds.). Technology in Transition. A.D: 300-650, 451-489 (Leiden: Brill). Valenti, Marco. 2009. ‘Ma i “barbari” sono veramente arrivati in Italia?’ in Gulia Volpe (ed.). V Congresso Nazionale di Archeologia Medievale, 25-30 (Firenze: All’Insegna del Giglio). Valero Tévar, Miguel Ángel. 2015. La villa romana de Noheda: la sala triclinar y los mosaicos. PhD thesis (University of Castilla-La Mancha). Valverde Castro, María del Rosario. 2000. Ideología, simbolismo y ejercicio del poder real en la monarquía visigoda: un proceso de cambio (Salamanca: Universidad de Salamanca). Van der Leeuw, Sander. 1977. ‘Towards a Study of the Economics of Pottery Making’ in B. L. van Beek, R. W. Brandt, and W. Groenman-van Waateringe (eds.). Ex Horreo, 68-85 (Amsterdam: Universiteit van Amsterdam). van Leeuwen, Richard. 1999. Waqfs and Urban Structures: The Case of Ottoman Damascus (Leiden: Brill). Vaquerizo Gil, Desiderio. 2004. ‘Arquitectura doméstica y privada’, in Xavier Dupré (ed.). Las capitales provinciales de Hispania. Córdoba. Colonia Patricia Corduba, 81-94 (Rome: L’erma di Bretschneider). Vaquerizo Gil, Desiderio. 2007. ‘El mundo funerario en la Malaca romana. Estado de la cuestión’. Mainake 29, 377-99. Vaquerizo Gil, Desiderio. 2010. Necrópolis urbanas en Baetica. Documenta 15 (Tarragona: ICAC). Vaquerizo Gil, Desiderio; and Juan Francisco Murillo Redondo (eds.). 2010. El anfiteatro romano de Córdoba y su entorno urbano. Análisis arqueológico (ss. I-XIII d.C.). Monografías de arqueología cordobesa 19 (Córdoba: Universidad deCórdoba).

386 

The Iberian Peninsul a bet ween 300 and 850

Velasco Vázquez, Javier; Carlos Sanz Mínguez; and Inés Centeno Cea. 2003. ‘La necrópolis tardoantigua e hispanovisigoda de Las Quintanas’ in Carlos Sanz Mínguez and Javier Velasco Vázquez (eds.). Pintia. Un oppidum en los confines orientales de la región vaccea, 221-250 (Valladolid: Universidad deValladolid). Velázquez Soriano, Isabel. 1989. Las pizarras visigodas. Edición crítica y estudio. Antigüedad y Cristianismo 6 [special issue] (Murcia: Universidad deMurcia). Ventura Villanueva, Ángel. 1996. El abastecimiento de agua a la Córdoba romana II. Acueductos, ciclos de distribución y urbanismo (Córdoba: Universidad de Córdoba). Ventura Villanueva, Ángel; and Guadalupe Pizarro Berengena. 2010. ‘El Aqua Augusta (acueducto de Valdepuentes) y el abastecimiento de agua a la Colonia Patricia Corduba: investigaciones recientes (2000-2010)’ in Las técnicas y las construcciones en la ingeniería romana, 175-204 (Madrid: Fundación de la Ingeniería Técnica de Obras Públicas). Viana, Abel. 1959. Suevos e Visigodos no Baixo Alentejo (Braga: Bracara Augusta). Vico Monteolivo, Jesús; María Cruz Cores Gomendio; and Gonzalo Cores Uria. 2006. Corpus Nummorum Visigothorum ca. 575-714. Leovigildus – Achila (Madrid: Self Published). Vidal Castro, Francisco. 2000. ‘Agua y urbanismo: evacuación de aguas en fatwà-s de al-Andalus y el Norte de África’ in Patrice Cressier, Maribel Fierro Bello and Jean-Pierre van Staëuel (eds.). L’urbanisme dans l’occident musulmana u Moyen Âge. Aspects juridiques, 101-23 (Madrid: Casa Velázquez). Vidal Teruel, Nuria; José Manuel Campos Carrasco; and Ángel Gómez Rodríguez. 2010. ‘La ocupación del entorno rural de Onoba en época romana: la villa de “la Almagra” (Huelva)’. Huelva en su Historia 13, 31-50. Vigil-Escalera, Alfonso. 2000. ‘Cabañas de época visigoda: evidencias arqueológicas del sur de Madrid. Tipología, elementos de datación y discusión’. AEspA 73, 223-252. Vigil-Escalera, Alfonso. 2003a. ‘Arquitectura de tierra, piedra y madera en Madrid (ss.V-IX d.C.). Variables materiales, consideraciones sociales’. Arqueología de la arquitectura 2, 287-91. Vigil-Escalera, Alfonso. 2003b. ‘Cerámicas tardorromanas y altomedievales de Madrid’ in Luis Caballero, Pedro Mateos and Manuel Retuerce (eds.). Cerámicas tardorromanas y altomedievales en la Península Ibérica, 371-88. AAEspA 28 (Mérida: CSIC). Vigil-Escalera, Alfonso. 2007. ‘Granjas y aldeas altomedievales al norte de Toledo (450-800 d.C)’. AEspA 80, 239-284. Vigil-Escalera, Alfonso. 2009. ‘Apuntes sobre la geneología política de aldeas y granjas altomedievales’ in Iñaki Martín Viso (ed.). ¿Tiempos oscuros? Territorios y sociedad en el centro de la Península Ibérica (siglos VII-X), 31-44 (Madrid: Sílex). Vigil-Escalera, Alfonso. 2012. ‘El asentamiento encastillado altomedieval de la Dehesa de la Oliva (Patones, Madrid)’ in Juan Antonio Quirós Castillo and José María Tejado Sebastián (eds.). Arqueología e historia de los castillos altomedievales en el cuadrante noroccidental de la Península Ibérica, 239-62 (Bilbao: Universidad del País Vasco). Vigil-Escalera, Alfonso. 2013a. ‘Comunidad política aldeana y exclusión. Una revisión de las formas de inhumación altomedievales (ss. V-VIII d.C.)’ Reti Medievali Rivista 14, 1-42. Vigil-Escalera, Alfonso. 2013b. ‘Las últimas producciones de TSHT en el interior peninsular’. Ex officina Hispania. Cuadernos de la SECAH 1, 11-24. Vigil-Escalera, Alfonso. 2015. Los primeros paisajes altomedievales en el interior de Hispania. Registros campesinos del siglo quinto d.C. (Bilbao: Universidad del País Vasco). Vigil-Escalera, Alfonso; and Juan Antonio Quirós Castillo (eds.) 2016. La cerámica de la Alta Edad Media en el cuadrante noroeste de la Península Ibérica (siglos V-X) (Bilbao: Universidad del País Vasco). Vigil-Escalera, Alfonso; and Juan Antonio Quirós Castillo. 2013. ‘Un ensayo de interpretación del registro arqueológico’ in Juan Antonio Quirós Castillo (ed.). El poblamiento rural de época

Bibliogr aphy

387

visigoda en Hispania. Arqueología del campesinado en el interior peninsular, 347-400 (Bilbao: Universidad del País Vasco). Vigil-Escalera, Alfonso; and Strato. 2013. ‘El registro arqueológico del campesinado del interior peninsular en época altomedieval’ in Juan Antonio Quirós Castillo (ed.). El poblamiento rural de época visigoda en Hispania. Arqueología del campesinado en el interior peninsular, 289-328 (Bilbao: Universidad del País Vasco). Vigil-Escalera, Alfonso; and Carlos Tejerizo García. 2014. ‘Asentamientos fortificados altomedievales en la Meseta: algunas distorsiones historiográficas’ in Raúl Catalán Ramos, Patricia Fuentes Melgar and José Carlos Sastre Blanco (eds.). Las fortificaciones en la tardoantigüedad. Élites y articulación del territorio (siglos V-VIII d.C.), 229-45 (Madrid: La Ergástula). Vigil-Escalera, Alfonso; Marta Moreno García; Leonor Peña-Chocarro; Arturo Morales Muñiz; Laura Llorente Rodríguez; Diego Sabato; and Mariano Ucchesu. 2013. ‘Productive Strategies and Consumption Patterns in the Early Medieval Village of Gózquez (Madrid, Spain)’. Quaternary International, 7-19. Villar Mañas, Sonia; and García García, Marcos. 2016. ‘Propuestas para el estudio de la ganadería Andalusí. Aproximaciones desde el análisis de los paisajes históricos’ in Antonio Malpica and Guillermo García-Contreras (eds.). El registro arqueológico y la Arqueología Medieval, 257-97 (Granada: Alhulia). Vives Castell, José. 1942. Inscripciones cristianas de la España romana y visigoda (Barcelona: CSIC). Vizcaíno Sánchez, Jaime. 2008. ‘Carthago Spartaria, una ciudad hispana bajo el dominio de los milites romani’ in Lauro Olmo Enciso (ed.). Recópolis y la ciudad en la época visigoda, 338-60. Zona Arqueológica 9 (Alcalá de Henares: Museo Arqueológico Regional). Vizcaíno Sánchez, Jaime. 2009. La presencia bizantina en Hispania (siglos VII-VIII). La documentación arqueológica. Second Edition. Antigüedad y Cristianismo 24 [special issue] (Murcia: Universidad de Murcia). von Rummel, Philipp. 2013. ‘The Fading Power of Images: Romans, Barbarians, and the Uses of a Dichotomy in Early Medieval Archaeology’ in Walter Pohl and Gerda Heydemann (eds.). Post-Roman Traditions: Christian and Barbarian Identities in the Early Medieval West, 365-406 (Turnhout: Brepols). VV. AA. 2011. 711. Arqueología e Historia entre dos mundos. Zona Arqueológica 15 (Alcalá de Henares: Museo Arqueológico Regional). Walker, Rose. 2016. Art in Spain and Portugal from the Romans to the Early Middle Ages. Late Antique and Early Medieval Iberia 1 (Amsterdam: Amsterdam University Press). Wallace-Hadrill, Andrew. 2008. Rome’s Cultural Revolution (Cambridge: Cambridge University Press). Ward-Perkins, Bryan. 1984. From Classical Antiquity to the Middle Ages: Urban Public Building in Northern and Central Italy AD 300-850 (Oxford: Oxford University Press). Ward-Perkins, Bryan. 1996. ‘Urban continuity?’ in Neil Christie and Simon Loseby (eds.). Towns in Transition: Urban Evolution in Late Antiquity and the Early Middle Ages, 4-17 (Aldershot: Scholar Press). Ward-Perkins, Bryan. 1997. ‘Continuitists, Catastrophists and the Towns of Post-Roman Northern Italy’. Papers of the British School at Rome 65, 157-76. Ward-Perkins, Bryan. 2001. ‘Specialized Production and Exchange’ in Averil Cameron, Bryan Ward-Perkins and Michael Whitby (eds.). Cambridge Ancient History XIV, 346-91 (Cambridge: Cambridge University Press).

388 

The Iberian Peninsul a bet ween 300 and 850

Ward-Perkins, Bryan. 2005. The Fall of Rome and the End of Civilization (Oxford: Oxford University Press). Ward-Perkins, John Bryan. 1971. ‘Quarries and Stoneworking in the Early Middle Ages: The Heritage of the Ancient World’. XVIII Settimane di Studio del Centro Italiano di Studi sull’Altomedioevo: Artigianato e tecnica nella Società dell’Altomedioevo Occidentale. Spoleto 1970, vol. II, 525-44 (Spoleto: CISAM). Werner, Joachim. 1946. ‘Las excavaciones del Seminario de historia primitiva del hombre, en 1941, en el cementerio visigodo de Castiltierra (Segovia)’. Cuadernos de Historia Primitiva 1, 46-50. Werner, Joachim. 1948. ‘Hallazgos de origen bizantino en España’. Cuadernos de Historia Primitiva 3.2, 107-12. Whitow, Mark. 1990. ‘Ruling the Late Roman and Early Byzantine city: A Continuous History’. Past and Present 129, 3-29. Wickham, Chris. 2005. Framing the Early Middle Ages: Europe and the Mediterranean 400-800 (Oxford: Oxford University Press). Wickham, Chris. 2008. The Inheritance of Rome: A History of Europe from 400 to 1000 (London: Penguin). Wickham, Chris. 2010. ‘Asturias entre visigodos y mozárabes. Observaciones finales’ in Luis Caballero Zoreda, Pedro Mateos Cruz and César García de Castro (eds.). Asturias entre visigodos y mozárabes (Visigodos y Omeyas VI), 475-90. AAEspA 63 (Madrid: CSIC). Williams, Howard. 1998. ‘The Past in the Past: The Reuse of Ancient Monuments’. World Archaeology 30.1, 90-108. Williams, Howard; and Duncan Sayer. 2009. ‘Halls of Mirrors: Death and Identity in Medieval Archaeology’ in Duncan Sayer and Howard Williams (eds.). Mortuary Practices and Social identities in the Middle Ages: Essays in Burial Archaeology in Honour of Heinrich Härke, 1-23 (Exeter: Exeter University Press). Williams, Javier. 2005. ‘Las acuñaciones reales visigodas’ in Carmen Alfaro Asins, Carmen Marcos Alonso, Paloma Otero Morán (eds.). XIII Congreso Internacional de Numismática, Madrid, 2003: actas-proceedings-actes, 1269-79 (Madrid: Ministerio de Cultura). Wilson, Andrew. 2007. ‘Quantif ication of Fish-Salting Infrastructure Capacity in the Roman Rorld’ OXREP working paper. http://oxrep.classics.ox.ac.uk/working%20papers/ quantification_fishsalting_infrastructure_capacity_roman_world/ Witschel, Christian. 2016. ‘Hispania, Gallia and Raetia’ in Bert Smith and Bryan Ward-Perkins (eds.). The Last Statues of Antiquity, 69-79 (Oxford: Oxford University Press). Wolfram, Herwig. 2005 [1990]. The Roman Empire and its Germanic Peoples (Berkeley: University of California Press). Wood, Ian. 1998. ‘The Barbarian Invasions and the First Settlements’ in Averil Cameron and Peter Garnsey (eds.). Cambridge Ancient History, Volume XIII: The Late Empire, AD 337-425, 516-36 (Cambridge: Cambridge University Press). Wood, Jamie; and Javier Martínez Jiménez. 2016. ‘New Directions in the Study of Visigothic Spain’. History Compass 14, 29-38. Zeiss, Hans. 1934. Die Grabfunde as dem Spanischen Westgotenreich. Germanische Denkmäler der Völkerwanderungszeit 2 (Berlin: Walter de Gruyter). Zozaya Stabel-Hansen, Juan. 1998. ‘The Islamic Consolidation in Al-Andalus’ in Roland-Pierre Gayraud (ed.). Colloque international d’archéologie islamique, 245-58. Textes et Études Islamiques 36 (Cairo: Institut Française d’Archéologie Orientale).

Index Abd al-Rahman I, Emir of Córdoba 276, 292, 294, 303 Abd al-Rahman II, Emir of Córdoba 276, 279, 288, 294 agrarian growth 101, 112, 309 Aistra 38, 212, 244, 309 Al-Andalus 22, 27, 102, 221, 268-9, 271-2, 276-7, 291-3, 295-308, 315 Alans 58, 60, 98, 141-2 Albarregas 103 Alcalá de Henares see Complutum Alcázar 282, 301-2 Aldaieta 216 Alegría-Dulantzi 309-10 Alfonso II, King of Asturias 286-7 Algeciras 181, 185, 189, 273, 282, 292 Alicante see Lucentum Almeida, Fernando de, 35 Almenara 55 Almudayna 279 Almunias 303 Almuñécar 83, 116, 160, 185, 189 Amaya 54, 173, 178, 194, 309 amphitheatres 67, 72, 75, 80-3, 92, 97, 162, 190, 238, 301 amphorae 50-1, 53, 78, 92-3, 116-8, 177, 189-90, 224, 321 Ampurias 53, 92, 137, 290 Annona 50-1, 101, 115, 118, 181, 185, 189 aqueducts 22, 48, 67, 72, 75, 77, 82-6, 88, 94-6, 98, 159-60, 164, 167, 175, 187, 223, 239, 241-2, 279, 282, 302, 306, 319 Aquitaine 142-3, 290 Arabs Arabic language 272-3, 297, 313, 321 Arabic epigraphy 297-8 Arabic seals 273 people 268-9, 271, 276, 298 Aragón 297 archaeology bioarchaeology 37, 122, 207, 209 commercial 23, 42, 103, 230 funerary see burials history of 24, 30-40 geoarchaeology 246 of architecture 136 of liturgy 136-8 zooarchaeology 296 arches and vaulting 65, 97, 159, 181, 220-3, 242-3, 245, 278, 288, 290, 293, 307 horseshoe arches 221-2, 242, 293 architecture architects and engineers 175, 236, 242-3, 263, 285 church architecture 35, 126, 161-8

domestic architecture 95-7, 121-2, 175, 198-9, 204-8, 214, 227, 245, 262, 303, 311, 320 mosque architecture 291-4 Arianism 170, 178 ARS see pottery art history 30-1, 35, 39, 136 Astorga 47, 74, 96, 126, 156, 164, 196, 233 Asturias region 201, 207, 268, 271, 285-6, 291 Asturian art 38, 201 Asturian monarchy 286, 291, 312 Asturian state formation 43, 286-8 Ausona 53, 290 Ávila 71, 94, 157, 198, 262, 271 Badajoz 282 Baelo Claudia 83, 107, 116, 182, 185 Baetica 27, 47, 52-3, 63, 79, 92, 117-9, 134, 218 Baetulo 53, 92 Bagoeta 309 Baladiyun and muwalladun 270, 276 Balearic Islands 35, 52, 65, 116, 137, 180, 183 Balearic basilicas 65, 220 Banu Qasi 291 Barbarian invasions 58-9, 63, 65, 99, 125, 138-9, 141, 143-4, 147-9, 153, 204, 209, 254, 270, 318, 320 Barcelona 41, 76-7, 79, 85, 118, 141, 155, 182, 187, 189, 289-90, 300 aqueduct 83, 160, 279 civil palace 171 domus 157, 161 episcopal complex 38, 91, 155, 162-5, 239, 245, 291, 299 suburb 93, 103 walls 71 basilicas (Christian) 89, 93, 135, 156, 162, 164, 167, 181, 185, 218-21, 223, 243, 287, 292, 317; see also churches basilicas (civil) 48, 54, 76, 78-9, 110-1, 237-8, 319 Basque Country 36, 115, 155, 202, 207, 290, 310 baths 48, 67, 75, 76-8, 83, 85, 96-7, 108, 283, 301 Benalúa 93, 95, 189, 234 Berber Revolt 269, 276 Berbers 268, 270-1, 276 Bernardos 62, 194-5, 241, 245, 262 Berrocales, Los 209 Bigastrum 53, 173, 175, 185 Bilbilis 53 bishops 53-4, 57, 67, 89, 91, 153-4, 156-8, 161-4, 167, 171, 185-6, 222, 251, 299-300 Fructuosus, Augurius and Eulogius (Tarragona martyrs) 126 Justinian (Valencia) 162, 319

390  Martin (Dume) 183 Masona (Mérida) 168 Optimus (Tarragona) 134 Zeno (Mérida) 159, 319 Braga 49, 53, 73-4, 76, 78-9, 83, 93, 96, 161, 163, 183-4, 187, 189, 223, 233, 300 Brown, Peter 35, 47 burials, cemeteries, necropoleis ad sanctos necropolis 107, 162-3, 212, 317 burial practices 28, 56-7, 102, 105, 124-5, 128, 134, 146-8, 193, 209-10, 247, 317 communal cemeteries 33, 146, 213-6, 311 furnished burials 122-3, 140-1, 144-5 Grabenstraßen 105-7, 124 inhumations habillées 214-5 intramural burials 75, 91-2, 163 isolated burials 211 maqbara 294-6 mausolea 88-9, 105, 138, 167, 218, 288 pit burials 210-1 rock-cut burials 155, 216-8 Byzantines and the Byzantine province 27, 153, 170, 175, 180-3, 185-6, 189-90, 235, 240, 257 Caballero Zoreda, Luis 39, 137, 155, 221, 307 Cabezo Pardo 306 Cabrera monastery 223-4 Cádiz 49, 53, 63, 92, 180 Calahorra 53, 259 Calatalifa 311 Caleruega 309 caliphate 29, 268, 271, 282, 285 Can Gambús 38, 205, 209-10, 214, 227 Cangas [de Onís] 285 Cantabric region 26, 43, 61, 91, 116, 269, 286, 309, 320 Canto Blanco 203, 205-6, 209, 309 Cañal 206 Carolingians see France Carpetania 27 Carranque 55, 110-1, 113 Carratejera 121 Cartagena 49-50, 53, 64, 76-7, 83, 85, 107, 117, 153, 175, 181, 185-6, 235, 257, 308 Comitiolus inscription 180 Theatre excavation 161, 182, 185, 189, 240 walls 73-5 Carteia 92, 181 Carthaginensis 27, 52-3, 74 Casa Herrera 33, 81, 131, 222, 232 basilica and baptistery 162, 167, 219-20, 223-4, 239-40, 274 graffiti 297-8 pilaster 252-3 Castile 27, 58, 178, 202 Castillo de los Monjes 200 Castiltierra 214, 248, 250 castles 32, 279, 282, 304-5, 308-9, 321 Castra see hillforts

The Iberian Peninsul a be t ween 300 and 850

Castro Ventosa 194, 196-8, 200, 227 Castulo 53, 126-7, 131 Catalonia 35, 38, 61, 116, 137, 202, 205-7, 210, 212, 215, 285, 289, 308 Cauca 155, 199, 231 cave habitation 224-5 Celtiberia 27, 50, 175 Centcelles 88, 110-1, 138 Central March 27, 276 centralization 25, 42-3, 102, 122, 170, 259, 276, 283, 285, 288 Cercadilla see Córdoba Ceuta 160, 181, 273 Chindaswinth, King of the Visigoths Chindaswinth’s inscription 179, 186 christianity 123-38, 148, 317 christianization 56, 58, 79, 88, 98, 105-6, 222, 300, 318 conversions 170, 183 iconography 56, 88, 126, 128-9, 134-5, 252, 318 identity 128-30, 148, 298, 308, 313 liturgy 117, 132, 135-8, 218, 298 origins in Iberia 56, 123-8 Church, Arian see Arianism Church, Catholic (institution) 43, 57, 178, 183, 193, 212, 218, 223, 228, 247, 308, 312 churches (buildings) see basilicas church dating controversy 39, 220-3 circus 67, 80-2, 87, 97, 154, 172, 177, 190, 263, 301, 319 CIS see pottery civitates 52-4, 59, 61, 68, 101, 124, 178, 277 climate 27, 60-1 cloisonné 139, 145, 248, 250 Clunia 49, 53, 155 coinage 255 and taxation 258-60 Byzantine coins 224, 255-7 Suevic coins 256-7 Umayyad coins 271-2, 313 Visigothic coins 170, 255-6 Colonate 55 Complutum 53, 89, 93, 96, 154, 282 Daganzo necropolis 139-40 martyr cult 89, 135 Villa of El Val 108, 113-4, 154, 244 Conimbriga 38, 53, 74, 96, 172 Constantine, Roman Emperor 79-80, 88, 126 Constantinian dynasty 52, 54, 56, 79, 80-1, 87, 98 Constantinople 31, 82, 161, 315 construction materials and techniques ashlar 69, 74, 171, 221-2, 238, 241 brick and tile 106, 121, 161, 173, 236-7, 239-40, 262, 283, 303, 320 mortared rubble 72, 173, 238-40, 303 opus africanum 240 opus caementicium 72, 74, 239 opus sectile 111

Index

opus signinum 78, 110, 173, 177, 239, 245 pisé and mud brick 97, 237, 240, 263, 303 reuse and recycle 72, 74, 126, 130, 136-8, 156, 163, 171, 173, 179, 183, 195, 221-2, 237-41, 251-2, 254, 277, 279, 281, 285, 292, 320 timber building 97, 240, 242-5, 263, 320 wattle and daub 199, 204, 241 Contrebia Belaisca 53 Córdoba 41, 48-9, 53, 57, 62-4, 92, 103, 107, 135, 157, 162-3, 189, 230, 232, 251-2, 258-9, 268, 273, 275-6, 279, 292, 300-1, 303, 315 aqueducts 83, 85, 160, 302 baths 76 Cercadilla 86, 89, 96 Forum 79, 97, 104 Great Mosque 292-3 maqbara 295 palace 282 suburb (Roman) 94 suburb (Umayyad) see Šaqunda theatre 80, 104, 238 walls 72, 171-2 Covadonga 269 Crismón de Quiroga 133-4 Cristo de San Esteban 194-6, 199-200 Cullera, monastery 68, 223-4 curiales see urban elites Cuyacabras 216, 311 Damascus 269, 272, 276-7, 292 Dark Ages Cold Period 58, 60 dark earths 104-5 De los Ríos, Amador 31, 33 Dehesa de la Oliva 123, 194, 199 Diego Álvaro (Lancha de Trigo) 202, 261-2 DSP see pottery Dueñas 212 Duero 26, 119, 206 Duero basin depopulation 269, 308, 311 Duero valley necropoleis 123, 145 pottery production 119-20, 231 Dume 164, 183, 223 Duratón 33, 140, 213, 215 Early Middle Ages 28, 30, 316 Ebro 27, 49, 53, 61, 118-20, 200, 269-70, 291, 311 Écija 49, 86, 106, 163, 295 Edeta 53 Egypt 234, 271 Eio (El Tolmo de Minateda) 41, 161, 173, 175-6, 185, 187, 190, 268, 282, 288, 299, 306 acropolis 175 destruction 283 episcopal complex 164, 175 grafiti 297 maqbara 295 El Bovalar 33, 202-3, 220, 222, 227 El Carpio de Tajo 33, 140, 215, 248 El Castellar 202

391 El Castillón 194, 196, 199-200 El Gatillo 106, 131, 137 El Germo 219 El Molón 306 El Pelícano 121, 204-5, 209, 211 El Saucedo 108 El Soto 123, 295 El Tinto [Juan de la Cruz] 113-4, 215 El Tolmo de Minateda see Eio El Trampal 221, 223, 299, 307 Elche see Ilici Els Ametllers 108 Els Mallols 205, 210 Els Munts 110-1, 113 Emirate of Córdoba 43, 282-5, 291-2, 300, 306, 308, 315 end of town life 69, 92, 154, 157, 159 episcopal complexes 89-91, 118, 138, 153, 161-8, 170-1, 175-6, 190, 239-40, 243, 245, 288, 300, 317 Ercavica 53, 62 Es Cap des Port 220 Espirdo-Veladiez 214-5, 248 ethnicity 30, 33, 139, 144-6, 148, 209-10, 213, 218, 270, 318 Euric, King of the Visigoths 143, 158-9 fabricae 114, 232 Falperra 184 Favila, King of Asturias 286 feudalism 30, 305, 312-3 fortifications Asturian and Catalan 289-90 Byzantine 181, 183 Late Roman 69-75, 155-6 castra see hillforts rural 111, 310-1 Suevic 184-5 Visigothic 170, 172-3, 175, 190, 243 Umayyad 273, 283-4, 305-6 fora 48, 54, 76-9, 86, 91-2, 97, 118, 162-4, 167, 171, 183, 190, 237-8, 263, 278-9, 319 France, Gaul, and the Franks 22, 24, 47, 56, 63, 68, 70, 81, 105, 119-21, 139, 141-6, 163, 175, 189-90, 205, 214, 246, 251, 300, 309 Carolingians 72, 267, 288-91, 313 Merovingians 143, 214-6, 251, 257 Francolí complex see Tarragona Fruela, King of Asturias 287 Galician massif 26 Gallaecia 27, 52-3, 56, 61, 78, 93, 96, 180, 183, 235, 258-9 Garum and fish industries 63, 116 Gaul see France Gauzón 201, 286 gender 140, 146-7, 218 Germany and the Germans 33, 56, 58-60, 106, 118, 139-40, 144-8, 209, 214, 243, 250, 270, 318

392  Gerona 58, 71, 290 Gerticos 113 Gijón 74, 196, 285-6 gladiatorial games 67, 82 glass 50, 102, 126, 131, 139, 177, 224, 229-30, 233-6, 248, 250-1, 262 production 175, 188, 198, 200, 233-5 typologies 235-6, 262 Gómez-Moreno, Manuel 32-3, 35 Gorliz 244 Gothic (language) 148 Gózquez de Arriba 38, 41, 204-7, 211-2, 214-5, 227, 248 grain production 54, 101-2, 124, 200 Granada 38, 163, 194, 230, 274, 306-7 grave goods 30, 33, 105-6, 124, 139-41, 144-7, 193, 209, 211, 294, 317 Guadalquivir 27, 49, 157, 270 Guadiana 26-7, 167 Guarrazar site 179-80, 223, 226 treasure 30-2, 215, 250 harbours 53, 62-3, 75, 77, 93, 95, 101, 167, 181-2, 185-6, 188-9, 273 hillforts 61, 193-201, 227, 262, 286, 288, 305, 308-9, 320 first-generation 195-200 second-generation 200-1 Hispanic March 27, 289-91 Hospitalitas 142-4 housing see architecture Huesca 53 Husun 279, 304-8 Idanha-a-velha 35, 53, 164, 185 identity 22, 58, 123, 125, 128-9, 134, 144-5, 149, 193, 206, 209-10, 212, 218, 247-8, 250, 263, 291, 295, 297, 318 Ilbira see Madinat Ilbira Ilici 71, 92, 175, 182 Incastellamento 305 inscriptions 72, 74, 76, 78-81, 105, 129, 132-5, 157-9, 162, 173, 178-9, 181, 184, 201, 220, 222, 260, 279-80, 286-8, 297, 318 Iria Flavia 53, 287 irrigation 103, 275, 303, 305-6, 308 Islamic conquest 29, 40, 267-9, 271, 283, 291-2, 297, 306, 311-2, 321 Islamic settlement 267, 269-70, 273, 282, 284, 295, 306, 308 Islamization: Arabization and conversion 30, 272, 291, 294-5, 297-8 Italica 47, 53, 75, 172 Italy 22, 24, 47-8, 50, 63, 70, 105, 181 Iuliobriga 53 Jaca 53, 290 Jaén 282, 308

The Iberian Peninsul a bet ween 300 and 850

jewellery 22, 31, 58, 125, 139, 145, 188, 229, 247, 263 Jund (ajnad) 269, 271, 276, 282-3 Justinian, East Roman Emperor 31, 58, 60, 181 Justinianic Plague 58, 60, 181 La Alberca 138 La Cárcava de la Peladera 205-6 La Cocosa 110, 113, 138 La Garma 225 La Genestosa 217 La Mata del Palomar 204-6, 209, 211, 227, 246 La Olmeda 55, 109-10, 111, 113-4, 123 La Virgen del Tormejón 194, 196, 200 Labra 131-2 Ladera de los Prados 205, 209 Lancha de Trigo see Diego Álvaro Lapis specularis 50 Las Lagunillas 211 Las Médulas 50, 115 Las Penas 225 Las Quintanas 214 late Antiquity 28, 35, 42, 47, 58, 63, 72, 82, 107, 125, 230, 234, 254, 316-9 late Roman urbanism 67-8, 153 Latin (language) 298 lead seals 273 León 60, 73-4, 126, 196, 202, 300 Lérida 53, 302 L’Esquerda 290 Liédena 110 lyre-shaped brooch 248-9 limes 50, 52, 118 Lisbon 37, 77, 83, 116, 160, 189 Liuvigild, King of the Visigoths 170, 173, 178-9, 183, 257-9 Los Husos 225 Los Quintanares 110 Lower March 27, 276 Lucentum 71, 92-3, 95, 103, 182, 186, 295-6 maqbara 296 Lugo 49, 73-5, 85, 93, 103, 185 Lusitania 27, 52-3, 77, 81, 118, 132, 172, 218 macellum 77 Madinat Ilbira 38 Madrid, rural sites 38, 41, 202, 205-7, 214, 232, 295, 306 Madrona 213, 215 Málaga 107, 140, 163, 181-2, 185-7, 189, 259, 292, 303 Mantius’s plate 131, 133 marble 50, 61, 65, 98, 108-9, 111, 126, 128, 131-3, 135, 138, 157-8, 162, 168-9, 175-7, 179, 184, 188, 224, 226, 229, 238, 251-3, 254, 278, 297 decoration 108, 111, 168, 184, 252 Mérida workshop 61, 157, 175, 188, 251-4, 278 production and quarrying 50, 65, 109, 251-2 sarcophagi 126

393

Index

recycling 179, 238 Marialba 138 Marseilles 188 Martilla 216 Martínez Santa-Olalla, Julio 33 martyrial cult 88, 111, 129 Maternus 111 Mauri 270 Mecca 292, 315 Medina 315 Mediterranean trade 50, 53, 61, 63, 78, 92, 116-7, 167, 175, 188-91, 229, 235, 321 Melque 212, 221, 223, 299, 307 mensae 105, 131-2 Mentesa 53, 282 Mérida 27, 31, 37-8, 41, 47-9, 53-4, 56-7, 60-1, 63, 69, 71-3, 76, 78-80, 83, 85-9, 92-3, 96-7, 103-4, 107-8, 110, 120, 126-7, 129, 131-3, 135, 138, 140, 143, 153, 157-64, 167-9, 171, 175, 179, 186-9, 223, 231-2, 237-8, 240, 242, 251-2, 254, 259, 271, 274, 276-81, 287, 295, 297-9, 301-3, 306-7, 319 alcazaba 279-81, 297 amphitheatre 80, 238 aqueducts 83, 160 bridge/bridge inscription 158-9, 281 chi-rho 126-7 domus of the Marbles 157-8, 175, 278 domus of the Mithraeum 74, 88, 96-7 forum 171, 278 Los Bodegones 298 Morería 38, 41, 161, 238, 277-8 suburb 89-90, 93, 135, 138, 168 Temple of Diana 72, 171, 278 Xenodochium 168-9 Mértola 37, 157, 162, 167-8, 189-90 Meseta see also Duero valley metal working 102, 122, 139, 196, 200, 247-8 militarization of the elites 56, 112 Milreu 118, 297 mining 50, 61-2, 102, 115-6, 122, 124, 308 Mirobriga 53 monasteries 32, 179, 193, 212, 218, 223-4, 287-9, 297, 299, 307 monoculture 49, 101, 319 Monte Cildá 194-5 mosques 22, 282-3, 291-4, 301, 306 Mount Testaccio 50-1 Mozarabic 29, 33 Munigua 53 Murcia 283-4, 302, 306 muwalladun 270, 276, 280 nationalism 30 Portuguese 32-3 Spanish 30-1 Navarre 27, 178 Navasangil 194, 196-200, 262 Navatejera 113 Nívar 216

Noheda 55, 109-11, 121 North Africa 24, 47, 63-4, 92, 117-8, 126, 163, 224, 268, 270, 295 Numantia 53 nymphaea 85, 167, 238 Ocillis 282 Oiasso 62 olerdolan cemeteries 311 olives and olive oil 49-1, 63, 101, 118, 207-8, 224, 319 Ologicus 54, 173, 178 Oporto 37 opus africanum 240 opus caementicium 72, 74, 239 opus signinum 78, 110, 173, 177, 239, 245 Orense 133, 184, 259 Orihuela 185, 268, 302-3 osculatoria 247 Oviedo 226, 286-290, 315 pagan temples 79, 135 palaces 34, 87-8, 110-2, 121, 124, 162-4, 167, 170-4, 176-7, 186, 226, 239-40, 243, 245, 277, 279, 282-3, 287-9, 292, 302, 319 Asturian 287-9 late Roman praetoria 87 Visigothic 170-4, 176-7, 226, 245, 279, 282, 292 Umayyad 167, 277-8, 282-3, 302, 319 Palol, Pedro de 28, 33, 35 Pamplona 139, 172-3, 177, 268, 285, 291, 295, 299-300 Pancorbo 309 peasants 23, 39, 120, 122-4, 193, 196, 202, 204, 207, 227, 244, 246 peasant societies 23, 39, 202, 207, 244, 246 Pelayo, King of Asturias 269, 286 Peña del Mazo 312 Piñel de Abajo 214 Pisé see construction pit burials 210-1, 228 Plà de Nadal 226 Plaça Major de Castellar del Vallès 206, 209 Pollentia 53, 73, 93, 183 Portugal 23-4, 26-7, 30, 32, 35, 37, 50, 85, 118-9, 129, 167, 202, 207, 216-7, 296, 308, 311 Portus Ilicitanus 53 pottery 35, 38, 41, 61, 63-5, 75, 93, 101, 119-20, 124, 126, 128, 132, 147, 185, 187, 189, 195-6, 198, 200, 229-35, 239, 247, 251, 262, 273-5, 284, 308, 313, 318, 320-2 African Red Slip (ARS) 53, 64, 119 Cerámicas de Imitacion de Sigillata (CIS) 119-120 coarse wares 190, 230, 239, 320 cooking wares 102, 188, 190, 230-2, 262, 274-5, 320 Derivés de Sigillée Paléochrétienne (DSP) 120, 188, 190

394  fine wares 119-20, 187 islamic innovations 275-6 imported 63, 177, 198, 232-3, 320 pottery technology 101, 147, 196, 200, 231-4, 247, 284, 313 Terra Sigillata Bracarense Tardía (TSBT) 119-20 Terra Sigillata Hispánica Tardía (TSHT) 119-20, 230-1 Terra Sigillata Meridional Tardía (TSHM) 119-20 Pravia 285-6 prehistoric settlements 203-4, 225 pre-Romanesque 33, 291 Pueblanueva 138 Puig Rom 201 Pyrenees 26-7, 143, 201, 267, 269, 273, 288, 290 qiblah 281, 292 qila’ 282 quarries 50, 237, 241-2, 251, 285, 319 Quinta das Longas 110, 129 Quintanilla de las Viñas 212, 221 Quirós Castillo, Juan Antonio 23, 195 Ramiro I, King of Asturias 288, 290 Reccared, King of the Visigoths 170, 175, 178-9, 186, 259-60 Recceswinth, King of the Visigoths 30, 113, 178-9, 186, 222-3, 250, 260 Reccopolis 22, 33-4, 36, 38, 41, 54, 65, 153, 161, 173-7, 187-8, 190, 232, 234, 236-7, 239-43, 245, 248, 256-7, 259, 282, 283, 288, 302, 306 aqueduct 22, 239, 241 basilica 173-5 construction techniques 240-2 domestic architecture 174-5 glass production 175, 177 palace complex 239, 242, 245 pottery 232 Reconquista 30, 269 regionalization 25, 42-3, 60-1, 119-20, 190, 218, 233, 273, 316, 320-1 Reihengräberfelder 213 Relics 88, 93, 135, 147, 162, 220 Roderic, King of the Visigoths 172, 186, 268 Rome and the Romans 23, 27, 47, 50-2, 56, 58-61, 99, 118, 128, 134, 138, 141-3, 147, 154, 161, 181, 188, 190, 229, 255, 315, 317-8, 322 administration 36, 43, 47, 49-50, 52-4, 56, 59, 68, 73, 76, 78, 92, 98, 107-8, 112, 115, 141-3, 154, 238 citizenship 48 colonization 47-8, 79, 81 roads 50, 52, 72, 82, 86-7, 101, 105-7, 120, 124, 159, 175, 202, 279, 300-1, 303 Roman Empire 43, 47, 58, 65, 92, 105, 195, 315, 317 Romanitas 83, 134, 218

The Iberian Peninsul a bet ween 300 and 850

romanization 47-8, 65, 321 urban culture 48 royal patronage 147, 178-9, 184, 270 ruralization 102, 104, 154, 243 Ruscino 273 Sacramenia 312 Sagunto 53, 68 Salamanca 113, 155, 157 Salla, Duke 159, 319 San Juan de Baños 179, 186, 212, 221-3 San Julià 201 San Miguel de Lillo 288 San Miguel del Arroyo 122 San Pedro de Alcántara 219 San Pedro de la Nave 316 San Pedro de Tartalés de Cilla 225 Sant Menna 106, 210, 212 Santa Comba de Bande 221 Santa Lucía del Trampal 221, 223, 299, 307 Santa Margarida de Martorell 212 Santa María de Artés 212 Santa María de Melque 212, 221, 223, 299, 307 Santa María de Mijangos 212 Santa María de Naranco 226, 288, 290 Santiago de Compostela 287 Santo Adriano 310 São Cucufate 110 São Gens 311 Šaqunda 273, 303-4 sarcophagi 105, 126, 128, 134 sculpture 31, 111, 136, 155, 226, 229, 247, 251-2, 254, 263, 277 Segobriga 36, 53, 79, 83, 92, 215 Segovia 53, 62, 92, 120, 140, 156, 199, 205, 231-2 Senovilla 206 Septimania 54, 139, 143, 268, 273 Seville 38, 49, 74, 83, 85, 92, 96, 98, 103, 108, 113, 120, 130-1, 134, 157, 160-1, 164, 173, 179, 189-90, 230-2, 234, 251, 259, 292, 293, 300, 303 La Encarnación 38, 98, 131, 157, 161, 190 pottery 230-2, 234 sewers 83, 85, 87, 96, 302 Sicily 269 Silo, King of Asturias 286 silos 120-2, 124, 199, 203, 205, 208 Simancas-type knife 56, 122, 247 simplification of the economy 22, 63, 115, 120, 230-1, 236, 243, 267, 284, 320 slates 33, 241, 246, 255, 260-3 social habitus 149, 193, 218, 228, 263, 318 social structure 48, 145-7, 207, 262 Son Fadrinet 220 Son Peretó 220 Soto de Tovilla 122-3 Southern Gaul see Septimania Spania see Byzantine province spolia see construction squatter 114

395

Index

state formation 29, 161, 168-9, 183-6, 188, 200, 221, 255, 258-60, 267, 276-7, 285-6, 288, 291, 299-300, 313, 318-9, 321 Asturian 285-6, 288 Carolingian 288 Suevic 183-4, 255, 258, 318 Umayyad 188, 221, 267, 276-7, 313, 321 Visigothic 29, 161, 168-9, 185-6, 200, 255, 258-9, 318 streets 72, 78, 82, 86-7, 164, 175, 187, 300-3 suburbs and suburbanization 55, 89-95, 101, 103, 107, 118, 135, 138, 154-5, 157, 161, 163, 168, 173, 175-7, 187, 191, 231, 239-40, 273, 279, 290, 299, 303-4, 317 Sueves 28-30, 35, 43, 58-61, 98, 139-46, 148, 153-4, 156, 170, 172, 180, 183-6, 189-90, 193, 198, 223, 251, 255-9, 318 Kingdom 183-5, 198, 223, 258 state formation 183-4, 255, 258, 318 sulh 268, 271, 273, 307 Sunken Feature buildings (Grubenhaus) 120-2, 124, 199, 204-6, 208, 243-6, 263, 283, 320 suqs 300 Swinthila, King of the Visigoths 30, 32, 170, 178, 186, 250, 259-60 Syrians 269, 271, 276-7, 293 Tagus 26-7, 61, 139, 176 Talamanca 311 Tangiers 53, 269 Tarazona 112, 259, 303 Tariq, Berber military leader 268 Tarraconensis 27, 52-3, 60, 63, 118 Tarragona 37, 48-9, 53, 57, 72, 77-81, 83, 88-9, 92-6, 103, 107, 118, 126, 129, 134-5, 153-4, 160, 162-4, 187, 189, 220, 230, 239 amphitheatre 162-3 Francolí complex 89-90 walls 103 taxes and taxation 48, 54, 59, 63, 68, 101, 112, 115, 138, 142, 144, 170, 222, 227, 255, 258-60, 262-3, 268-71, 277, 319 Tedeja 194, 201, 309, 312 Terrasa 137-8, 153, 162, 164 Tetrarchy 52, 87 theatres 67, 72, 76, 79-80, 82-3, 87, 104, 161, 181-2, 185, 189, 238, 240, 301, 319 Theodosius, Roman Emperor 56, 110-1, 128, 155, 231 Tiermes 155 timber building see construction Toledo 31, 55, 80, 95, 103, 140, 153, 164, 170, 172-3, 175-9, 186, 190, 225, 239-40, 243, 251-2, 259, 271, 276, 279, 287, 295, 300, 306 Vega Baja 38, 176-7, 226, 240, 243, 279 torcularia 117 Torre de Palma 110, 135, 162, 219, 296 Torre Llauder 108, 111 Torreáguila 110

Torredonjimeno treasure 250 Torrentejo 310 towns see civitates and urbanism town councils 54, 76, 78, 82-3, 176 Tózar 216, 298 trade 50, 53-4, 62-5, 68, 88, 92, 101-3, 115, 117-8, 120, 183, 188-91, 215, 224, 229-30, 235, 257-8, 262, 284, 300, 321 African 53, 63-5, 92-3, 117, 119, 182, 188-90, 224, 230 Atlantic 167, 189 Mediterranean 50, 53, 61, 63, 78, 92, 116-7, 167, 175, 188-91, 229, 235, 321 Traducta 93, 282 Trajan, Roma Emperor 47 Tritium Magallum 119 Triumphal arch 79 TSBT see pottery TSH see pottery TSHM see pottery TSHT see pottery Tudmir (Theudimer), Visigothic Duke 268 Umayyads 29, 43, 188, 271, 276-7, 279, 282-3, 290, 292-4, 302, 321 Umma 268, 271, 291-2, 302 Upper Ebro valley 61, 120, 269, 291, 311 Upper March 27, 276 urbanism 39, 48, 67-8, 83, 125, 139, 153-4, 157, 170, 173, 183, 185-7, 190, 288 clustering 103, 187 encroachment 82, 87, 91, 187, 301 fortification 69-77, 170, 172-3, 175, 181, 183-5, 190, 284, 290-1 urban elites 49, 55, 65, 68, 98, 108, 112, 131, 134, 157, 196, 268, 277, 279 Uxama 53, 155 Valdetorres de Jarama 113 Valencia 78, 85, 103, 107, 157, 162-3, 182, 259, 303, 319 aqueduct 83, 160, 187, 302 fortifications 172 forum 103-4, 118 la Almoina 38, 155, 164-7, 189, 279 nymphaeum 85-6, 167, 238 Valeria 53, 62, 79, 237 Vallat Luhdriq 172, 282 Vandals 58, 60, 98, 141-2 Vasconcelos, José de 32 vaulting see arches Vega Baja see Toledo Veleia 73, 155-6 Veranes 108, 113 Vich 290 Victoriacum 54, 173, 177-8 Vigaña 207, 250, 310 Vigo 38, 93, 189, 233 Vilauba 108, 114, 117

396  villas 22, 25, 49, 54-5, 58, 62, 65, 88, 101, 103, 107-14, 117-8, 120-2, 124, 129-30, 138, 154, 177, 183, 189, 193, 207, 218-9, 226-7, 241, 244, 263, 288, 296-7, 319 abandonment 22, 58, 112-4, 120, 207, 241 villa economy 25, 107-10, 112, 114, 122, 319 Villa of Fortunatus 108, 129, 133 villages 22, 38, 41, 193, 198, 200, 202-10, 214-5, 217, 222, 227, 246, 262, 283, 304-6, 308-9, 311, 319, 321 Villanueva de Soportilla 216-7, 311 Villula 113 Visigoths 33, 35, 58, 60, 139, 141-9, 154, 167, 180, 183, 190, 193, 209, 270, 316-7 art 31, 155, 220, 222, 245, 251, 277, 281-2 councils 155-6, 162, 176 identity 139, 144-5 jewellery 31, 139, 144 kingdom and monarchy 25, 29, 31, 43, 116, 153, 156, 173, 178-9, 185, 186, 190-1, 200, 248, 250, 252, 255, 258-9, 268, 271, 277, 279, 286, 291, 313 settlement 58, 139, 141-4

The Iberian Peninsul a bet ween 300 and 850

state formation 29, 161, 170-2, 186, 200, 255, 258-9, 318-9 urban renewal 170-8, 186, 236, 242, 251 Vitoria-Gasteiz 38, 178 Vouillé, Battle 25, 143 walls 48, 52, 67, 69-76, 87, 90-2, 95, 103, 124, 138, 155-6, 159, 163, 167-8, 172-3, 175, 177, 181, 194-5, 198-9, 237-8, 241-3, 282, 319 Wamba, King of the Visigoths 173, 177, 186 wast al-dar 302 weapons 56, 114, 139-40, 196, 201, 247, 250, 263 wine 50, 54, 63, 102, 110, 115, 117-9, 124, 188, 190, 224, 227, 229, 296 xenodochium see Mérida Zaballa 38, 41, 205, 207-8, 227, 244, 309-10 Zaragoza 47, 49, 53, 74, 77-8, 80, 83, 92-3, 107, 189, 235, 271, 276, 301-2 maqbara 295 mosque 292-3 Zornoztegi 309