Change and Resilience: The Occupation of Mediterranean Islands in Late Antiquity

Change and Resilience offers a view of the main Mediterranean islands from West to East in Late Antiquity because Mediterranean islands can contribute in fundamental ways to our understanding not only of earlier colonizations but also later periods. The volume explores specifically the time frame from the fall of the Roman empire to the Medieval period.



A first group of papers covers islands and island groups in the Central and Western Mediterranean, including the Balearic Islands, Corsica, Sardinia, Sicily, and the Adriatic islands. Together, these five papers highlight several common themes across the region: local or indigenous sites were often reoccupied in Late Antiquity, the rural countryside typically played a significant role in the contributions of islands to wider Mediterranean economic networks, and islands – big and small – often played significant roles in shifting political and religious power.



The second group focuses on the Eastern Mediterranean. Three papers cover a range of islands, including Crete, the Cyclades, and Cyprus. Together they emphasize the impacts external shifts in political power and economic ties in the Eastern Mediterranean had on island landscapes, as well as the connected relationship between sacred space and territorial occupation across many of these islands.



The final group of papers pivots on changing perceptions of island landscapes in Late Antiquity—or “island mindscapes.” Three papers focus on how communities adapted as they underwent Christianization in island contexts, emphasizing the diverse and varied ways that island landscapes became “Christianized,” as well as how other political and economic factors shaped the dynamics of change.

• Language: English
• Publisher: Oxbow Books
• Release date: Jun 30, 2019
• ISBN: 9781789251814

Book preview

— 1 —

The Occupation of Mallorca (Balearic Islands, Spain) in Late Antiquity: Tracing Change and Resilience

CATALINA MAS FLORIT AND MIGUEL ÁNGEL CAU ONTIVEROS

This chapter explores how the communities of the island of Mallorca adapted to a successive series of changes that occurred between the Roman period and the end of Antiquity. The available evidence shows a significant transformation in the cities and also in the countryside where the number of sites decreased in the third century A.D., and only a few large settlements remained occupied. This pattern changed abruptly at the end of the fifth or early sixth century with an increase in the number of rural settlements, including the reoccupation of old indigenous prehistoric sites and the construction of rural churches which are essential to understanding the process of Christianization of the countryside. Small villages or secondary agglomerations would also have played an important role in the configuration of the landscape.

The chapter briefly addresses the transformation of the city but is mainly focused on the transformation of the countryside. The phenomenon of the reoccupation of old indigenous sites, the fate of Roman rural sites, and the role of the early Christian churches are outlined to understand the transformation of this Mediterranean island between the fourth and the eighth centuries A.D.

Introduction

The Balearic Islands are the most remote archipelago of the Mediterranean sea. They lie off the coast of the Iberian Peninsula in a strategic position for navigation and trade routes in the western Mediterranean (Figure 1.1). Greek and Roman writers were fully aware of the archipelago and separated it into two groups of islands with substantial differences. On the one hand, Mallorca and Menorca formed the Baliarides (also called the Gymnēsiai by the Greeks), represented by the Talayotic culture. On the other hand, Ibiza and Formentera were considered the Pityussae, with a strong Phoenician-Punic character.

Figure 1.1. Location of the Balearic Islands in the western Mediterranean and map of Mallorca with the main sites cited in the text: 1. Palma; 2. Tumulus of Son Ferrer; 3. Rural Roman site of Sa Mesquida; 4. Basilica of Cas Frares; 5. Pollentia; 6. Puig de s’Escolà; 7. Talayotic village of Son Fornés; 8. Basilica of Son Fadrinet; 9. Es Fossar de Ses Salines; 10. Closos de Can Gaià; 11. Castell de Santueri; 12. Basilica of Sa Carrotja; 13. Basilica of Son Peretó; 14. Son Sard.

The society of the Baliarides experienced significant impacts from the Roman conquest in 123 B.C. Despite their society’s contacts with the Punic world and other influences, these islands were dominated by the Iron Age indigenous population at the time of the Roman military intervention. The Roman period brought with it the foundation of cities and the reorganization of the countryside. The islands were linked first to Hispania Citerior and later to Tarraconensis. The archipelago as a whole (i.e., the unified Balearides and Pityussae) did not become an independent province until the end of the fourth century. In A.D. 455, the Vandals conquered the islands and in A.D. 534 they fell into Byzantine hands. This latter dominance lasted, at least in theory, until the Islamic conquest of Isam-al-Jalawni in A.D. 902–903. The old Roman structures, transformed during Late Antiquity, probably vanished forever under the strong cultural influence of Islam.

Interest in the transformation of these islands in the period between the Roman military intervention and the Arab conquest of the tenth century is relatively recent. Written sources are in general very scarce, and much of the information has to come from archaeology. This chapter offers some insights into the occupation of Mallorca during Late Antiquity, trying to understand the changes witnessed and the strategies that the local population adopted in a period (or, better, subsequent periods) of crisis and upheaval.

The Transformation of the Cities

The Roman conquest in 123 B.C., led by Q. Cecilius Metellus, resulted in the foundation of two cities in Mallorca: Palma and Pollentia (Figure 1.1). Regarding the city of Palma, now under the current city of Palma, there are little data available to provide an outline of its transformation in Late Antiquity (Cau 2012). However, two ceramic deposits dated to the Vandal and the Byzantine periods show the occupation of the city (Cau et al. 2014). The evidence so far suggests that population at the time prior to the Muslim conquest of the city was both small and in decline (Gutiérrez 1987: 206). It seems that the Muslim city of Madîna Mayûrqa was placed over a former urban centre still in operation (Riera i Frau 1993: 27–29). Most of the available data for the study of the transformation of Mallorcan cities, therefore, comes from Pollentia.

Concerning Pollentia, archaeological data show continuous use of the urban space, albeit with various re-organizations, since the foundation levels dated around 70–60 B.C. The third century A.D. was a moment of serious disruption in the city. A major reorganization occurred either at the very end of the second century or the beginning of the third. The excavations in the forum have demonstrated a complete transformation of the socalled Insula I of tabernae to the West of the forum square. Many rooms were restructured, with walls rebuilt, changing the overall dimensions and building small buttresses in some of them. The streets were also profoundly modified, and many of the spaces between the columns of the porticoed streets were closed off (Cau 2012). At the end of the third century, a massive fire destroyed various parts of the city, as clearly attested in Insula I. This destruction has been dated with precision to around A.D. 270–280 (Arribas et al. 1973; Arribas and Tarradell 1987: 133; Equip d’Excavacions de Pollentia 1994: 142; Orfila 2000). The third century was, therefore, a time of strong disruption to the city. The construction of a wall in the so-called residential quarter of Sa Portella was probably done in the same century, as a response to this moment of upheaval (Figure 1.2) (Orfila et al. 2000; Riera i Rullan et al. 1999). Nevertheless, there is no doubt that inhabitation continued in the city after the third century. By the fourth and fifth centuries, there is apparently little building activity, but there exist some industrial structures in the forum of Pollentia built over the debris of the late third-century fire. Some of the tabernae seem to be re-occupied by squatters (Orfila et al. 1999: 112–113; Orfila 2000: 154). Public spaces were also occupied: in the fifth century, for example, a room was attached to the façade of the Macellum, invading the portico area on the east side of the forum. By this time the city was deeply transformed, and certain buildings, such as the Macellum, were now abandoned. Some tabernae and the forum square have provided materials that denote later reuse of these spaces.

Figure 1.2. View of the so-called residential quarter of Sa Portella in the Roman and Late Antique city of Pollentia (Alcúdia).

The most significant structure in Late Antiquity, however, was the fortification located to the north of the forum area. Its construction involved the use of spolia from other buildings, with many elements being reused in the inner filling of the wall. This defensive structure closed off at least some of the old forum that was partially in ruins and occupied parts of public spaces, such as the north street (Figure 1.3). A dating in the Byzantine period has been suggested for this (Orfila et al. 1999: 113–116; 2000). The fortification indicates the enclosure of a protected perimeter, such as a citadel, in the very heart of the city, coinciding with the location of the forum and reusing the wall of the Capitolium, Insula I and the Macellum. Outside this citadel, there was continued inhabitation of the city, and Sa Portella shows clear signs of occupation dating to the Late Roman, Vandal, and Byzantine periods, with significant material culture (e.g., Arribas et al. 1973; Gumà et al. 1997).

Figure 1.3. The Late Antique fortification of the forum of Pollentia occupying the porticoed street to the north of the forum with a wall dated to the end of the second century or beginning of the third century closing off the intercolumniation.

The Christianization of the topography of the city is virtually unknown. Some early discoveries were claimed to be related to the presence of Christian buildings in the city. For instance, in a necropolis area within the suburbia known as Can Fanals, a large building in a very poor state of preservation was interpreted as a possible basilica (Llabrés and Isasi 1934). Moreover, further north, in the so-called area of Santa Anna de Can Costa, some graves and the early Christian inscription of Arguta were tentatively linked to the presence of another Christian building. These were already inside the urban layout, suggesting the transformation of the city.

Over the ruins of the old forum, a large necropolis has been radiocarbondated to the Muslim period. The most interesting aspect is that the inhumations were still in supine position and not in a lateral position, as would have been the norm for Islamic populations; this might be a sign of non-Muslim communities being buried between the tenth and the twelfth centuries on top of the ruins of the Roman city (Cau et al. 2017). The only structure respected by the graves we have excavated in the forum is the Capitolium, so it seems that this building was still partially standing or at least respected, although parts would surely have been dismantled by then. Some authors have proposed the possible transformation of the Capitolium as a place of Christian worship (e.g., Arribas and Tarradell 1987).

The Transformation of the Rural Areas

The initial interest in the Roman rural occupation provided the first systematic data to understand the transformation of rural landscapes in Mallorca (Orfila 1988). Recent fieldwalking surveys of several areas across the island have contributed to new information that allows a first interpretation of the changes that occurred in the countryside from the Early Roman period to the Early Middle Ages (Mas Florit and Cau 2013). The general pattern of settlement on the island seems to be defined by the continuity of prehistoric sites and the creation of newly-founded Roman settlements, apparently rather few in number (Cardell et al. 1990). Some of these possible villae, hamlets, farms or farmsteads have been located by fieldwalking, although excavation has occurred on very few of them. Therefore, the phenomenon of the implantation of the villa system and its evolution in the countryside, as well as the role of the indigenous settlements in this wider framework, is still not sufficiently studied. As we will see later, the few data available suggest that these villae were implanted around the Augustan era or slightly earlier (Mas Florit et al. 2015). The landscape was clearly divided and distributed, as traces of centuriation have been identified (e.g., Cardell and Orfila 1991–1992), something that indeed facilitated the administration of property and tax collection.

Studies undertaken on the eastern side of the island show the abandonment or decay of many medium- and small-size rural sites during the late second and the third century A.D. Only a few sites, always with large dimensions and located in areas of fertile plains, have been documented. This decrease in the number of sites, and the fact that only large settlements seem to survive, has been interpreted as a sign of the concentration of property (Mas Florit and Cau 2013). Such a development has also been observed in other areas of the western Mediterranean, where medium and small properties were concentrated to form large landholdings (e.g., Chavarría 2006; Sfameni 2004; Raynaud 2001).

The villa of Sa Mesquida, located in the western part of the island (Calvià), is one of the few sites that contribute to an understanding of the fate of rural Roman sites. It was occupied from the first century B.C., with a floruit in the first and second centuries A.D. At the very end of the second century or the beginning of the third, the settlement suffered a traumatic event, as witnessed by the evidence of fire. This fire caused the abandonment of the structures excavated so far. To date, archaeological excavations have uncovered a series of rooms ranged around a courtyard with a well, an industrial deposit, and a ceramic kiln for the production of common ware. This kiln was abandoned around the end of the second century and later reused as a dump at the end of the fourth or beginning of the fifth century A.D. (Mas Florit et al. 2015).

During the mid-fifth century, coinciding with the beginning of the Vandal period on the island, a new reorganization of the countryside took place, possibly marking the end of the large estates and the villa system (Mas Florit and Cau 2013). This does not necessarily mean that the Vandal conquest was the cause of the end of the villas, but it is interesting to note that the two phenomena are roughly contemporary. Both old indigenous sites and Early Roman sites, normally located in fertile plains or their foothills, were occupied once again. This new pattern shows a change in the system of exploitation of the countryside, with clear differences from the Late Roman model of exploitation. This reorganization intensified at the end of the fifth century and the beginning of the sixth, which coincided with the Byzantine domination. Apart from old prehistoric settlements in the plains, probably related to agricultural activities, some former prehistoric sites located in marginal areas and with no continuity into the Roman period, were also reoccupied. Some of these sites could have served as defensive places; others may have specialized in activities other than agriculture. Interestingly, in some cases, these enclaves were in proximity to areas potentially suitable for sheep- and goat-herding. In addition, some camps related to the exploitation of salt, in the southern part of the island in the area of Colònia de Sant Jordi (Guerrero 1987), and some caves near the sea on the eastern side of the island, probably linked to the exploitation of marine resources, were now occupied (Cau and Mas Florit 2013). Areas near the sea, although scarce, also show signs of occupation in Late Antiquity.

In the seventh century, a new decline in the number of sites is witnessed in the rural areas. There is evidence of the arrival of imported material until the end of this century. The eighth and ninth centuries remain a problem. The lack of knowledge of the material culture of these two centuries, with the absence of African Red Slip Ware as an index fossil, complicates the dating of rural settlements, particularly when excavation results are not available. In the last decade, however, evidence of late occupation has been found at several sites. Such is the case at the basilica of Son Peretó (Manacor) which remained occupied at least until the end of the seventh century or the beginning of the eighth, with sporadic inhumations dated even in the ninth century, or the basilica of Son Fadrinet (Campos), where two Byzantine gold solidi of around A.D. 737/38 or 739 have been found (Ulbert 2003). Likewise, in the prehistoric site of Closos de Can Gaià there is evidence of occupation in the eighth and ninth centuries (Servera et al. 2004).

Forms of Occupation in the Countryside

The disintegration of the large estates in the mid-fifth century in some ways saw a return to previous stages of organization of the countryside. Old prehistoric sites that were not occupied during the Roman period were now reoccupied. In the southern and eastern parts of the island, most prehistoric sites with evidence of Late Antique occupation are interpreted as farms. Their location is near cultivable areas, visually controlled from the sites themselves, and in locations with favorable exposure to the sun in both summer and winter. At some sites, mills have been found, indicating farming activities (Mas Florit and Cau 2011). This could be the case of the indigenous village of Son Fornés (Montuïri), with clear evidence of reoccupation of some buildings during Late Antiquity (Lull et al. 2001; Fayas 2005). Spatial analysis shows that, in the areas of Manacor and San Llorenç des Cardassar, some of these sites are found in locations that are topographically prominent but with relatively poor visibility, suggesting a defensive role in times of upheaval (Mas Florit and Cau 2011). Other examples could be linked to the reuse of sites for surveillance or control, as might be the case with the settlement of Puig de s’Escolà (Llucmajor), where a rock shelter was reoccupied from the mid- or late fifth century. This site’s location is in a strategic spot that controls the only existing topographic passage through the immediate landscape. Also, at Closos of Can Gaià, one of the prehistoric buildings was reused, including a hearth dated by radiocarbon to the fifth or sixth centuries A.D.; this site also shows evidence of occupation between the eighth and the ninth centuries, as has already been mentioned (Servera et al. 2004).

New Late Antique foundations in pristine locations are in general very scarce across the island. These settlements appeared after the collapse of the villa system. In the eastern part of the island that has been studied in the most detail, only six of a total of 64 sites are thought to have been established in pristine locations. These new foundations are normally quite limited in extent (Mas Florit and Cau 2011), and some were placed in remote areas far from agricultural resources, which could be an indication of shepherding or defensive strategies (Mas Florit and Cau 2013). The presence of newly founded sites in pristine settings seems to be a marginal phenomenon in the context of the occupation of the Mallorcan Late Antique countryside.

From the available data, it is difficult to establish the role of the villas during these times. It seems clear that, after the collapse of the villa system, some parts of the villae were abandoned, but others remained occupied or were reused for other purposes. In the villa of Sa Mesquida, a pottery kiln and a cistern (Orfila and Cau 1994; Cau 2003) were reused as rubbish dumps in the fourth and fifth centuries respectively, and at least one of the productive areas was reused for graves during the sixth century. A Late Antique necropolis found in the village of Felanitx seems to be located over the ruins of an Early Roman ceramic workshop, possibly linked to a Roman villa or some form of rural settlement; the radiocarbon dates suggest a chronology between the sixth and the seventh century for these inhumations. Another case of reuse of Early Roman structures is the balnea recently discovered in Son Sard (Son Servera), associated with some form of rural settlement still to be determined. The structures were abandoned before A.D. 500 and reused for productive purposes, with two different phases; one is of uncertain chronology, maybe Vandal or early Byzantine, and the second dates to the Byzantine period (Palomar et al. 2013).

Figure 1.4. View of the Castell de Santueri, with the Serres de Llevant in the background.

Apart from the sites already mentioned, several defensives places in the mountains were clearly part of the Late Antique landscape of the island and probably fundamental for the last moments of Antiquity (Cau et al. 2005). Few Islamic written sources indicate that the population of the island prior to the Muslim conquest inhabited castles in naturally defended locations in the mountains (Kirchner 1998). The example of the castle of Santueri (Felanitx) (Figure 1.4), with numerous coins, lead seals (Illisch et al. 2005), and other remains of cultural material (e.g., Aguiló and Conde 2015; Gomila 2016), attributable to the Byzantine period, suggests the importance of the site and is reminiscent of the case of Castrum Perti in Liguria (e.g., Mannoni and Murialdo 2001). A relevant discovery comes from Menorca where a seal of an archon of Mallorca was found (Nicolás and Moll 2015). This suggest that Mallorca controlled Menorca and that the Balearics in the Byzantine period had a similar organization to that of neighboring