1. Introduction

Traditionally, archaeological studies dedicated to Late Antiquity and/or the Early Middle Ages in Western Europe have tended to look back to the Roman classical past and to look for the diverse evolutions that society, economy and culture followed after Roman rule. [1] This starting point has been crucial for understanding the transformations that towns and cities witnessed - including the progressive abandonment of some sites - and the reconfiguration of the rural areas, in which the former villae disappeared as the major points of production and as a controlling element, just as a new settlement pattern started to appear, comprising hamlets, villages and farms. In Hispania (now modern Spain and Portugal) these transformations have long been studied [2] and so we have a good, albeit rough, picture of the Iberian Peninsula during the final phases of the Empire and the creation and the consolidation of the Suevic and Visigothic Kingdoms in the 5th to 6th centuries A.D. The key is recognizing that heterogeneity is probably one of the main characteristics of this process of transition. Indeed, landscape diversity was behind most of the great differences in the responses.

The end of the Empire and the breakdown of Roman imperial control - above all, the cessation of its vast administrative and economic system - assured a redefinition of both the rural and urban spheres. However, in this paper, we want to show a different case, one characterized by a landscape only partially occupied during the Roman period, but which saw the construction of a new city under the Visigothic period. Our focus is the site and context of Reccopolis, the city founded by King Leovigild in AD 576, which can be considered as an unicum, because, although we have textual information about other royal Visigothic urban foundations (cited in the contemporary writings of John Biclar and Isidore of Seville), such as Victoriacum, also founded by Leovigild, and Oligicus established under King Swinthila, no other city is known archaeologically. Indeed, the modern excavations and survey work at Reccopolis enable us to see the planning and the development of a city in the later 6th century and, above all, to see how this huge building project was not only a major undertaking but marked also an intensive reshaping of a territory, one away from the main communication routes. As a result, the new dynamism of this landscape relates specifically to the creation and investment in a new and royal central place.

2. Geographical and Historical Context

Reccopolis was founded in the Celtiberia region - a historic appellation that revived in Visigothic times the pre-Roman name of one of the most numerous groups of gentes that had occupied part of the interior of the Iberian Peninsula, coinciding with parts of the modern regions of Aragon, Castilla y León and Castilla-La Mancha. With the consolidation of the Visigothic Kingdom, Celtiberia corresponded with the province of Guadalajara and Cuenca, in the southern part of the Meseta Central, bordering to the west with the Carpetania region, where the capital of the Visigothic Kingdom, Toledo, was located.

Reccopolis occupies a hill plateau (a site and area scarcely inhabited prior to the site’s foundation) (Fig. 1), close to the fertile Tagus River, and at a location which permitted control of an extensive area of arable lands. Why this specific place was chosen for the new city is, however, still difficult to establish. Perhaps one factor was that the large Celtiberia region featured a scarce number of episcopal sees (Segobriga, Valeria and Ercavica) to act as economic foci and bases for taxation. Moreover, it is important to bear in mind that these towns were characterized, at least from the 5th century, by a de-structuring of the classical Roman urban fabric and the rise of a more scattered habitat, hence the necessity to create a major (new) central place, where the elite presence could be more evident, and territorial control could be better structured. The direct line of communication, via the left bank of the Tagus River, with the capital Toledo was also crucial, connecting the heart of the Peninsula with the Mediterranean coast.

Apart from these strategic reasons, what seems clear is that the project coincided with the consolidation of the Visigothic Kingdom under the rule of Leovigild [3] and this king’s will to create a political-ideological programme similar to the Byzantine Empire. In this process of aemulatio imperii, the founding of a city with the name of his son, Reccared, saw Leovigild seeking to be on a par with the Byzantine emperor Justinian (527-565), who had given the name of Iustiniana to diverse urban centres, best evident in the imperial province of North Africa. [4]

It is striking that, in a period when most of the building programmes developed in the cities were managed by the episcopal powers, the construction of Reccopolis and the project in the Vega Baja of Toledo are the only known state/royal building initiatives. The archaeological work at the capital, outside the city walls, shows intensive construction activity between the mid-6th and the mid-7th centuries, [5] marked by a palatine complex, a set of houses and a commercial zone of shops and workshops, occupying a wide space of c. 85 hectares (the area is partially excavated to c. 20 ha). Previously, until the substantial transformation of the Visigothic period, this area had featured the Roman circus, some suburban villae and burial zones. [6]

As outlined below, together with the Toledo’s Vega Baja, Reccopolis was one of the few cases in early medieval Hispania, where the king and his court were able, with significant economic effort, to show their building capacity, both to display their power and to reaffirm cultural identities. The foundation of Reccopolis was in reality a major act, a declaration of intent by a king who wanted to establish and solidify a nascent, but strong kingdom.

3. Reccopolis: a brief archaeological description

The city of Reccopolis was projected as an ex-novo foundation and as an explicit political power statement. For two and a half centuries, from the Visigothic period until the early Islamic period in the mid-9th century, this city was a dynamic production and consumption centre. The multiple archaeological campaigns (since 1944) undertaken at the site have revealed a hierarchical urban plan, with the topography dividing the urban space into two areas: one formed by the palatine complex, and the second, lower one, mainly occupied by houses and workshop units. A monumental gateway and street connected the areas; the partially-excavated road was lined by at least two large buildings of clear commercial and productive character, with spaces for glassmakers, jewelers and probably also for merchants. [7]

Reccopolis was contained within impressive city walls, built with masonry blocks and lime mortar and with a maximum thickness of 2 m and a preserved maximum height of 5 m. The circuit’s towers are mostly quadrangular but there is at least one semicircular example, with the spacing between the towers ranging from 20 to 40 m. [8]

What stands out prominently in the city plan (as so far excavated) is the palatine area, which is located on the highest ground in the north-eastern area of the site. This occupied almost 1.5 ha and was formed by three buildings: the palatine church and the so-called palatium, which consisted of two buildings (the largest being a full 140 m long) each of two floors, probably with the lower one used as storage, for economic and fiscal activities, while the second floor, mainly paved with opus signinum and with stucco decoration, was the ceremonial space. [9] In the last two years, we were able to pin down the extent of the palatial complex based on the results obtained from an extensive georadar survey.

Significant data come from the study of the palace’s demise and collapse which came before the end of the 8th century, as a consequence of a destructive fire. [10] Thanks to the combination of the DSM (Digital Surface Model) with 3D digital photography obtained by Laser Scanning [11] (Fig. 2), we have developed the volumetric study of the building’s end, which confirms that the collapse of the second floor occurred in a single moment, although, potentially part of the roof and upper floor was dismantled immediately following this fire destruction.

Analysis of the charred remains of wood from the beams supporting the opus signinum floor showed a predominance of oak wood (70%), compared to 30% pine. Radiocarbon dating of the wood prompts some interesting discussion: along with samples that connect to the original construction of the building [12] (AD 582-660), others pre-date the foundation (AD 405-550), some quite considerably (AD 137-334). This surprising set of dates surely points to the reuse of construction materials, gathered together along with newly-cut materials.

The destruction of the palatine complex coincided with a phase of profound change in the physiognomy of early medieval Reccopolis, affecting much of the original urban layout. Over the remains of the complex has been detected an articulation of a set of domestic spaces, which reused the still visible structures and were attached to the monumental gate to the Palatine complex; these houses were built with perishable materials and their chronology is suggested by the discovery within one house of a dírhem issued during the government of al-Ḥakam I (197/812 H), emir of al-Andalus. [13] By the second half of the 9th century, most of the architectural materials from the palatine building were looted and removed for the construction of the emiral citadel in nearby Zorita de los Canes.

4. Methodology and understanding of the landscape organization

Landscape Archaeology and its paradigms have been core to the new research phase developed over the last decade at Reccopolis, in which one of the main objectives has been to identify and register the formation and evolution of the social structures and their reflection in the surrounding context. Indeed, recognizing that the landscape is a social construction, so the study of culture and biology [14] is key to understanding the meaning of a previously de-urbanized landscape and, during the Early Middle Ages, of the foundation of a new city and its impact on its landscape. Thus, although systematic archaeological excavation has continued within the project, the incorporation of Landscape Archaeology has permitted us to add a multidimensional view to our archaeological analysis of Reccopolis. In that sense, our research activity could be divided into two related spheres: a palaeoenvironmental analysis and a geo-topographical study.

The palaeoenvironmental component, discussed also in other publications [15], is based on the analysis of natural sedimentary deposits dated by calibrated radiocarbon methods, and cultural deposits from archaeological contexts. The evidence offered by the palaeoenvironmental record from all across the Iberian Peninsula [16] suggests that the so-called Early Medieval Cold Episode (A.D. 450-950), which brought deep climatic disturbances to the Mediterranean and the European West, [17] was characterized by lower temperatures and greater aridity, especially in the inland areas. The palynological record of Reccopolis shows evidence of intensive deforestation, creating an open landscape, as revealed by the scarce quantity of tree pollen (less than 40%), these mainly comprising Pinus and Juniperus, followed by Olea (Fig. 3). Interestingly, the recorded plants and herbs could be included within the group of steppe taxa, being also characteristic of pasture. Indeed, the dry Mediterranean climate of this area and the intense anthrophic activity seen between the 6th and 8th centuries fits perfectly with the archaeobiological results of this analysis. [18] Moreover, although the faunal evidence is still limited, the data overall show a predominance of grazing activity by sheep and domesticated goats which, considering also the age of the specimens identified, seems to relate to a mixed exploitation regime, in which wool and milk, more than meat, were the key products.

Apart from these analyses, the geo-topographical study included intensive archaeological survey - such as the micro-spatial survey carried out in the valley of Arroyo Badujo - geophysical prospection, and the use of remote sensing techniques (LIDAR, thermography), along with more traditional aerial photography analysis, to interrogate more fully the conformation of the new landscape. All of these survey types have generated very useful indicators of spatial hierarchy in the territory and of the types of settlements that make up the suburbia (Fig. 4).

In the territory close to the city, the surveys, together with the development of a Digital Terrain Model (DTM) using LiDAR Technology (Fig. 5) has revealed evidence for the emergence of a new agro-ecosystem that we can use to generate a new model of land parcel or division, one intimately linked to the foundation of Reccopolis between the 6th and 7th centuries, [19] combined with an alteration to and reorganization of the road network, and the creation of specific infrastructural elements.

Without doubt, identification of parts of the aqueduct supplying the new city has been one of the most interesting discoveries. Traces have been located 1.5 km south east of the city. Although its existence has been known since the 1970s when a section of 100 m length was documented, the generation of a specific DTM, combined with an intensive survey campaign, has expanded data regarding its layout, enabling us to propose an extension of between 2.7-3.0 km until it reached/entered the city. The bulk was cut underground or on the surface, although some sectors are elevated (250 m). Its structure was formed by a specus made in opus signinum, with a width of between 60-75 cm. The canal was built in limestone masonry, bonded with lime mortar. The surveyed section is characterized by a chiefly South-North direction, one adapting to the unevenness of the terrain and the slope down to the Tagus River. Most likely, the disappearance of part of this line of aqueduct came in the second half of the 20th century, when olive cultivation within the immediate area became more intensive.

Meanwhile, we are starting to accumulate data, via the combination of non-invasive techniques of subsoil documentation (Georadar and Magnetometer) and the use of Remote Sensing, related to the city’s suburbia and associated rural settlements [20] (Figs. 5 and 6). The territory also reflects a new rural landscape of similar characteristics to other areas within the centre of the Hispanic peninsula, which marks a change from the previous late antique one. It comprises different types of settlements, mainly villages and farms, which were situated within a maximum radius of 3 km, showing their direct relationship with the city. Noteworthy is the presence in these sites of ceramic products comparable to those found at Reccopolis, and even (perhaps surprisingly) imported goods, such as North African amphorae (including the widespread Keay 61).

Of these nearby settlements, the site of La Paeriza (Zorita de los Canes) (Fig. 6) is of particular interest, covering an area of just over half a hectare and located 1.0 km east of the city. There are good surface traces of well-crafted masonry walls, and ceramic material from the Visigothic period, which allow us to establish a period of occupation between the end of the 6th century and the 8th. The rural settlement seemingly combines exploitation of a wide agricultural area, along with extraction of materials from nearby quarries and perhaps also oversight/maintenance of the aqueduct supplying Reccopolis. Two other sites of the 7th and 9th centuries have been identified: Loma del Badujo (Zorita de los Canes) and Los Arroyuelos (Albalate de Zorita), both located close to the main communication routes. Although concrete data are currently lacking, both these rural communities likely remained active throughout the 8th century and the Arab take-over, pointing to landscape and urban sites retaining roles and activity after Visigothic control ended.

Another aspect of significant change prompted by the foundation of the city in the later 6th century is the alteration in routeways. The organization of the territory around the city followed the same radial pattern that is so characteristic of other regions in the High Middle Ages: the network of roads was structured like a star shape. [21] Associated with this re-articulation, our analysis has yielded valuable information on the organization of the new agro-ecosystem and the distribution of crops. We can observe a re-structuring of the system on two levels: firstly, on a local scale, the new centrality of Reccopolis allowed for the consolidation of a direct land communication route between the city and Toledo, following the left bank of the Tagus River. This route was still well-known in 19th century cartography, labelled then as ‘Via Salinera’, which extended onto Illana (Guadalajara), and from there onto the present-day Madrid region. [22] This road system gave scope for Reccopolis to become a central communication hub between the centre of the Iberian Peninsula and the eastern peninsular coast.

In this sense, the analysis of the roads and related communications was carried out at the micro-spatial level using cartography at the scales of 1:5,000 and 1:10,000 of the National Geographic Institute (IGN) dated to between 1883 and 1910. Such map evidence was then orthorectified with selected points of topographic support, and subsequently transformed into digital form to be integrated into a GIS platform, via which several analysis models were established, based on the road.

The 19th century cartography also gave some important information about land-use, crop types, their extension and their relationship with water resources and routeways. From all of this, topologies of land use have been developed for the period 1890-1990, which can then be considered retrogressively, comparing these data with those environmental indicators gathered from palynological samples, and the textual references for the area, which can often be very specific with regard to medieval irrigated space and land-usage.

5. Conclusions and a future proposal

The urban reality and the modified rural dynamics created as a consequence of the foundation of Visigothic Reccopolis, show a hierarchical landscape organized around what was a prominent and royal central place. Noticeably, the evidence from our landscape archaeological work shows the city as the ‘magnet’ to a number of satellite rural groups (villages, farms, etc.). This situation calls into question the current suggested ‘hamlet model’, in which peasants themselves organized and exploited the landscape. [23] Instead, what we see in the case of Reccopolis is a post-classical city which was able to initiate a specific agro-ecosystem, connected by a much modified road network and a new rural settlement pattern, which would continue to define the landscape of this territory until the consolidation of the renewed social structure during the Emiral period.

On many levels, the foundation and construction of Reccopolis constituted a route to disciplining the environment, understood as a social space [24] and also a clear demonstration of the exercising of power through the control of resources. This planning entailed the concept of disciplining not only the urban environment but also the territory and through this the familiarization of the population with those authorities and elites providing supervision and order. In this regard, we should not forget how the spaces promoted by the dominant ideologies also helped to ensure the cohesion of these elites. [25] We can argue that the extraction of surpluses and the organization of a fiscal model that continued growing until the first third of the 7th century played a major role here. Reccopolis, therefore, generated a landscape of power that expressed the visibility of an ideology and the manifestation of this in the surrounding landscape. [26]

Our field evidence now indicates the continuity of this new, royal urban centre beyond Visigothic rule and throughout the 8th century [27], and so confirms the ongoing importance of the city as a fiscal centre, until the end of this century. Our dataset, supported by the Remote Sensing technologies, point not only to the intensive transformation of the classical and late antique landscape in the early medieval period, but also to an enduring model of social and economic implementation into the early 9th century. This multiplicity of evidence and of evidence types and the extension of a project that moves away from the excavation of urban space and/or basic archaeological survey as an interpretive unicum, rooted in the principles of Archeographie [28] to one that is strongly interdisciplinary and enables a fuller and clearer contextualisation of the site in space, environment, ecology and time. The development in detailed and systematic micro-spatial analyses such as have been undertaken in Reccopolis, and their exploitation, mapping and interpretation through diverse information sources and platforms, is one of the key challenges that must be confronted in exploring and questioning settlement models in the centre of the Iberian Peninsula during the Early Middle Ages, building on fieldwork undertaken over the last two decades. Moreover, the need to incorporate better the human dimension in the archaeological material record will remain a challenge in the years to come at Reccopolis and more broadly: looking for and understanding the full range of people in here - the citizens, artisans, workers, farmers, quarrymen, etc. - are as important as exploring the elite’s residences and their relationship with the urban structure and the landscape. Deploying an array of techniques and interpretative approaches to tackle people, places, and practice in both city and countryside across the early medieval centuries will be the essential next steps in the Reccopolis Project.

Image gallery

Reference list


1. This paper has been written as part of the research line Cambio Climático y Construcción del Paisaje Medieval: Dinámicas de Variabilidad en un Periodo de Transformaciones (HAR2017-84144-P); and as part of the Juan de la Cierva-Incorporación (JCI-2015-24930) based at the Universidad de Alcalá and funded by the Ministry of Science, Innovation and Universities of the Spanish Government. Many thanks go to Prof Neil Christie of the University of Leicester for his assistance in checking the English of this paper.

2. For a recent overview, see Diarte-Blasco 2018, 34-80.

3. Olmo-Enciso 1992; Olmo-Enciso 2007, 183ff.

4. Claude 1965, 176; Haldon 1999, 10; Pringle 2001; Olmo-Enciso 2008,

5. For a detailed collection of papers on this area, see García et al. 2010.

6. Olmo-Enciso 2010, 88.

7. Gómez de la Torre-Verdejo 2011, 273; Castro-Priego and Gómez de la Torre-Verdejo 2008.

8. Gómez de la Torre-Verdejo 2008, 77-86.

9. Olmo-Enciso 2008, 47-49.

10. Olmo-Enciso 1995, 216.

11. Olmo Enciso et al. 2017, 79-88.

12. Beta 394657 (AD 405-550, 95.4%); Beta 394658 (AD 240-395, 95.4%); Beta 394659 (AD 405-550, 95.4%) Beta 480494 (582-660, AD 95.4% probability); Beta 480495 (AD 137-334, 95.4% probability). Beta 480496 (AD 582-660, 95.4% probability), Beta Analytic-London.

13. Olmo Enciso 2002; 2011:54-55; Castro-Priego 2008: 139.

14. McGuire 1992, 129-131; Mrozowsky 2006

15. Leveau et. al. 2002; Mrozowski 2006; Rotherham 2013; Rottoli 2014; Verhulst 1995.

16. Riera-Mora 2006, 19-26; 2008, 29, 30-35 and 37; Kaal et al. 2011, 172-173; López-Sáez et al. 2014, 113 and 117; Varón-Hernández et al. 2012, 300-301; Hernández-Beloqui et al. 2015, 83-4.

17. McCormick et al. 2012; Delogu 2012.

18. Olmo-Enciso 2015, 34-35; Olmo-Enciso et al. forthcoming.

19. Olmo-Enciso et al. forthcoming; Olmo Enciso et al. 2018, 159; Olmo Enciso et al. 2017, 97.

20. Olmo Enciso et al. 2017, 97-98: La Paeriza (Zorita de los Canes), Cabanillas (Albalate de Zorita), Campillo de Abajo (Zorita de los Canes) and Los Arroyuelos (Albalate de Zorita).

21. Olmo-Enciso 2015.

22. Olmo Enciso et al. 2018, 156.

23. Quirós-Castillo (ed.) 2009; Vigil-Escalera Guirado 2007.

24. Olmo-Enciso 2015

25. Paynter and McGuire 1991, 10.

26. Olmo Enciso 2008 and 2011; Olmo Enciso 2015, 40-41.

27. Castro-Priego and Olmo Enciso 2016; Olmo Enciso et al. 2017.

28. Chouquer 2003; Watteaux 2011.