CONQUEST AND ROMANIZATION OF THE UPPER VALLEY OF GUADALQUIVIR RIVER

Marcelo Castro López

Luis Gutiérrez Soler

This study analyzes archaeological evidence for the romanization of the upper Guadalquivir valley. This key region, which corresponds to the present-day province of Jaén (Andalucía, Spain), straddled the boundaries of the Roman provinces of Baetica and Tarraconensis. Our research attempts to understand developments following the Roman conquest of this region, by focusing upon two neighbouring territories. The results highlight apparent contradictions and divergences in the archaeological record for the region as a whole (Castro 1993; Gutiérrez 1998). In particular, it analyzes the archaeological evidence from the two contrasting oppida of Giribaile and Atalayuelas, of which the former represents a break with earlier settlement patterns, while the latter points to their continuation. It is suggested that these differences can be explained by Rome’s adaptation of a particular strategy of domination and exploitation to the distinctive conditions of the region, as well as by its development of administrative policies through time (Castro 1989). Previous research has already underlined the distinctive organization of the Sierra Morena mining district, the occupation of new land in the San Juan river valley (Castro et al. 1993), the foundation of colonies under Augustus and the appearance of the first terra sigillata production centres (Roca 1976; Sotomayor 1977; Fernández García, 1998). Such diversity in the techniques of the economic exploitation of the region demands an overall historical framework of interpretation to explain the romanization of the region between the Roman conquest and the concession of ius latii to communities in the region by the Emperor Vespasian (Pliny, N.H. III, 30), between AD 70 and 74 (according to Segura, 1988; Cortijo 1993:201).

GEOGRAPHICAL LOCATION

The study area is situated in the modern province of Jaén, in upper Andalucía. It has a distinctive topography on account of the river Guadalquivir which rises in the Sierra de Cazorla in the east of the province. This has created the Great Central, or Baetic, Depression which widens out as the river makes its way westwards. In the upper reaches of the river, the landscape of the valley is determined by the confluence of two very different morphogenetic units. To the north, lies the Sierra Morena, which belongs to the underlying ancient massif of the Iberian Peninsula, while to lay the newer relief of the Subbética range. Both of these mountain ranges formed a depression, or central basin, within which lay much of the agricultural resources of the upper Guadalquivir valley.

In the centre of this basin stands "La Loma de Úbeda" (the Hillock of Úbeda), which separates the course of the Guadalquivir from one of its main tributaries, the Guadalimar. One of the survey projects which forms the subject of this paper, that focusing upon the oppidum of Giribaile and the middle valley of the river Guadalimar (Royo et al. 1995; Gutiérrez et al. 1995) lay in this region. Between the Guadalquivir and the southern mountain rages (Subbética range) in the western area of the upper valley, lay a flat area of cultivable land, known as the Campiña de Jaén. This was the site of the second survey project, which focused upon the Roman landscape around the settlement of Atalayuelas.

RESEARCH INTO THE ROMAN PERIOD

Archaeological research in Jaén has concentrated upon the study of Iberian settlement patterns, particularly those centred in the Campiña de Jaén. These have suggested that a cereal economy was based on the existence of a network of oppida (Ruiz Rodríguez and Molinos 1993), which articulated the control of political and economic territories in the region. There is little doubt that the best-known and most productive contribution in this field has been the systematic excavation of the oppidum of Plaza de Armas de Puente Tablas (Ruiz Rodríguez and Molinos 1987). With this model as a point of reference,  an alternative pattern of settlement has been proposed for the easternmost area of Jaén province. This was centred upon an analysis of the Guadalquivir basin, and revealed alignments of oppida following the direction of the rivers.

Analyses of the landscape are our principal source of information about the process of romanization in the upper Gudalquivir valley. For some scholars, this represented a continuity of the Iberian pattern until at least the Republican period, to the point that the period between the end of the Second Punic War and the reign of Augustus has been labelled as the “late Iberian horizon” (Bendala 1981; Ruiz Rodríguez et al. 1985). Aside from landscape studies, however, there are other important sources of archaeological data which can contribute to this debate. In the last few years, there has been a great increase in urban archaeology in the city of Jaén, the site of ancient Aurgi. This has uncovered much evidence for the presence of peasant rural sites involved in irrigated agriculture (Hornos et al. 1998) during the early empire. This can be paralleled by the recent discovery of similar settlements on the industrial estate of Martos, the ancient Colonia Augusta Gemella Tucci (Serrano et al. 1997) and elsewhere in the upper Guadalquivir region. In addition, there have also been important new results from excavations at ancient Cástulo (Linares) and Obulco (Porcuna).

Romanization and Urban Development

Traditional studies have reduced the study of towns to a simple architectural entity emerging from native communities in response to stimuli of Roman civilization (Fernández Castro 1982; Arteaga and Blech 1987). Consequently, it has been assumed that this culture was little more than a facade and that it was both hierarchical and the result of cultural impulses emanating from Rome alone. These assumptions have underpinned a persistent view of the Roman period in Iberia which is still maintained in some quarters. This sees the need to isolate agents of romanization and recognizes differing degrees of maturity, in an almost biological explanation of cultural change.

The line of argument that romanization was an undeniable and inevitable evolutionary process is still sustained by many scholars, and only a few suggest that it may have been a contradictory or unequal process. This was supported by the established historical picture. Traditional historiography (Cortadella 1988) has established an impenetrable net of arguments and facts. More recently, however, the position has changed somewhat, with the admission that certain issues, such as the motivation of Roman expansion or seeing romanization as simply a cultural phenomenon, being dropped (González 1981).  Consequently, the archaeological evidence has been interpreted to suggest that native communities underwent the same kind of linear evolution, with no room for any alternative hypotheses. This is supported by the continued appearance of some distinctive elements of the material culture of the pre-Roman communities in contexts dating to the Roman period. Other important evidence, however, is omitted since it is considered to be residual and, therefore, best considered in the pre-Roman context.

When materialist and other approaches have approached romanization from the social and economic perspective, there has been a genuine attempt to provide alternative explanations. However, the perspective adopted has been one which sees the Iberian peninsula as a socially backward periphery (Barbero and Vigil 1974), and in which the south is understood to have undergone a rapid process of romanization. This view is furthered by assumptions that have been made about the awareness that local communities may have had about city-based organization prior to the arrival of Rome in the region (Cortijo 1993; León and Rodríguez Oliva 1993). In any event, the traditional view that romanization was the consequence of Roman military and political activity still dominates. As a result, the cultural aspect of the romanization of indigenous communities has been reduced to such superficial aspects as the latin language, the recruitment of natives into the Roman army, the adoption of certain classes of Roman style pottery (Roca 1976) or the adoption of well-defined architectural types (Arteaga and Blech 1987).

It has been stated that the introduction of the Roman town was inextricably linked with the imposition of a way of production. In effect, from the traditional standpoint which has emphasized the inter-dependence of the economy and political institutions of the city, this means that the political-institutional determines, and is part of, social and economic transformation (González 1981:96). Consequently, once the constitution and new political framework of a city is known, then we are in a position to understand how the means of production were transformed. Our perspective is to look at this from the other way round. Namely, that one can only be sure that a new political order has been established if the means of production have been transformed.

In fact, romanization cannot be understood without reference to the imposition of the social and economic organisation as represented by the Roman town. Yet, the Roman Republic was not interested in the indiscriminate promotion of urban institutions (Gros 1988), nor had the development of their own native communities necessarily resulted in the constitution of Roman towns (Ruiz Rodríguez 1987). The introduction of the Roman town would thus have been the result of contradictory interaction among different interest groups. On the one hand, there were those involved in the political and social strategies who would have retained local power after the Roman conquest, while on the other there were those who came directly from Rome and were able to play a leading role in provincial politics.

In contrast to the more traditional and generalist approach to romanization, ours attempts to recognize the historical nature of the process. Consequently, if the town is understood to be an immediate consequence of the Roman conquest, it is not surprising that it has been often been seen as a catalyst in the political, social and economic transformation of native peoples, or romanization. In this scenario, the town almost behaved as an independent entity, hastening the crisis of native social relationships brought about through the extension of Roman concept property and possession (González 1981). In time, this would lead inexorably to the establishment of the slave mode of production in the agricultural economy which, in turn developed within a framework of growing crop specialisation and an increase in regional market exchange (Funari 1986).

If Roman towns had not developed within the context of the social and political transformation of the native communities, they would have been little more than an empty archaeological setting (Gros 1988). The key to understanding their role does not lie in any single morphological or judicial definition, but rather in analysing their spatial organization and the way in which it reproduced the new Roman urban society. The form that this took cannot be reduced simply to their orthogonal layout or the sanctity imposed by their walled enclosures - since these have both been attested on indigenous sites long before the Roman conquest (Ruiz Rodríguez and Molinos 1987; Castro et al. 1989; Cunliffe and Fernández, 1990). The most obvious contrast between Roman towns and the oppida which preceded them, lies in the way in which former projected itself upon its surrounding territory. The physical presence of the town cannot be divorced from its role in structuring the administrative and economic management of the surrounding population or, indeed, their agricultural exploitation.

Finally, it must be emphasized that the immediate cause of this new agrarian way of production is to be explained by a change in the way that the Roman State administered its conquered territories. They came to be governed through municipalities, which were able to regulate a steady source of oncome from agrarian production and also to bring about an increase in profits. This transformation came about under the Flavians with the concession of ius latii to settled communities in the province and represent a significant change in the politics of Roman political administration. However, it also raises important questions at the local level of our research. The new socio-political implications entails disruption of land-ownership, as can be seen from the creation of new rural settlements all over the region, and from the introduction of new land exploitation. These involve technological and managerial changes in farming and production, such as the spread of rotating mills, the permanent stay of families in rural areas, and the establishment of home storage. All in all, the foundations of a new agrarian economy are laid that is apart from the ownership relations and from the organization of pre-roman farming.

Conquest and Romanization

If we restrict the focus of the romanization debate to the establishment of Roman towns and their territories, it should be stated that it was initially strictly limited to the colonial settlements of the early Imperial period, like the Colonia Salaria (Úbeda) and the Colonia Augusta Gemella Tucci, and only became generalized with the municipalisation of the Flavian period. However, some changes clearly did begin to take place amongst the native communities of the Guadalquivir valley prior to the early imperial period and even though they do not fall within the concept of romanization as we have defined it here, should be explained.

For example, the territory of the oppidum of Giribaile provides an example of the traumatic consequences that the Roman conquest had for some native communities (Gutiérrez 1998). By contrast, that of the oppidum of Atalayuelas illustrates the way in which the military and political organization of the region under the Republic ensured the fossilization of the pre-existing territorial order (Castro and Choclán 1988; Ruiz Rodríguez et al.1992). Between these two extremes are a range of other post-conquest situations elsewhere in the upper Guadalquivir valley. Thus, some Iberian oppida, such as the Plaza de Armas de Puente Tablas (Ruiz Rodríguez and Molinos 1987) were simply abandoned after the conquest, whereas others, such as the Cabeza Baja de Encina Hermosa in Castillo de Locubin (Castro et al. 1993), were completely rebuilt. These contrasting responses to the Roman conquest, and indeed others at the key native centres of Cástulo and Obulco, should perhaps warn us against searching for a common Roman policy against native communities in the region, in the aftermath of the Second Punic War. The same is true for the subsequent early imperial period. The politics of the colonial foundations of Salaria and Augusta Gemella Tucci in the present-day province of Jaén, or the establishment of the first production centres of terra sigillata hispánica, such as at Los Villares de Andújar (Roca 1976), in the Tiberian and Claudian periods shows that romanization can be manifested in many different ways. Moreover, analysis of the recent archaeological evidence for the Roman conquest and subsequent romanization makes it clear that there were a variety of different ‘responses’ and that a single cause of explanation is not sufficient.

The Roman Conquest

The Roman conquest of the Iberian Peninsula can be understood as an indirect consequence of conflicts between Rome and Carthage that had begun towards the middle of the 3rd century BC. While military activity was clearly key to Rome’s ultimate success, it should be remembered that diplomatic activity was also important. In other words, both the Carthaginians and the Romans searched for an understanding with local communities. Consequently, warfare seemed to occur in the context of a fragile network of alliances, the continuity of which was dependent upon the success or failure of the Romans or Carthaginians. After disembarkation pf Cnaeus Cornelius Scipio at Empúries in 218 BC, the Romans consolidated their positions to the north of the river Ebro before reaching the upper Guadalquivir valley at around 215 BC. In this same year, the Romans ratified a pact with Iliturgi (Cerro Maquiz, near Mengíbar) and shortly afterwards enlisted the support of Cástulo and were able to win battles at Biguerra and Auringis. This first stage in the Roman occupation of the region ended with the disaster of the year 211 B.C., when the region was again seized by the Carthaginian army and the Romans withdrew to the north of the river Ebro.

The final stages of Rome’s conflict with Carthage in Iberia began with the arrival of Scipio Africanus. In 209 BC he re-established Roman control over the upper Guadalquivir valley, after defeating Hasdrubal at the battle of Baecula (Livy XXVIII.13). The Romans subsequently sacked Orongis (possibly a mistranslation of Auringis) and then directed their attention to the conquest of the middle and lower Guadalquivir valley. However, Roman domination of the upper valley was not completed until 206 BC. It was in this year that Scipio Africanus destroyed Iliturgi and Roman diplomacy brought about the surrender of the Punic garrison at Cástulo after signing a pact with a representative of the town’s aristocracy (Cerdubeles: Livy XXVIII.19). This, therefore, was the way that Roman control of the upper Guadalquivir was achieved during the last years of the 3rd century BC, even though continued military operations and diplomacy were needed to sustain for the greater part of the following century. In the meantime, however, the systematic exploitation of the conquered native communities had begun.

Friends and Allies

The payment of tribute to the Roman State and their agents by provincials, necessarily required the co-operation of the local aristocracy, who gained the gratitude of Rome as a result. This may sometimes have been made explicit in the granting Roman citizenship to certain individuals. This meant that pre-existing systems of production and taxation were essentially compatible with the mechanisms of the Roman administrative system and, therefore, must have been largely retained. In this way, the development of the native aristocracy under Roman domination was linked to their role in being subordinate to the broader interests of Rome (Ruiz Rodríguez et al. 1992). In economic terms, this would have involved managing key stockpiles in the region. In political terms, it meant that key features of native society at the end of the 3rd century BC would have become fossilized, with crucial social relationships being underpinned by a combination of taxation and military control.

The spoils of war were the main economic “fruits” of the military occupation of the region, and may well have accounted for an important part of Roman activity during the Second Punic War and the subsequent years of military control in the region. Moreover, this kind of intervention was to have a small influence throughout the 2nd century BC, while the social and political conditions that would sustain a more regular and systematic exploitation were being established. Key members of the regional native aristocracy, the regulus and princeps, would have acted as interlocutors between Rome and the indigenous communities and would have to have been created de novo (Ruiz Rodríguez et al. 1992). This arrangement would have ensured the continued pre-eminence of the oppida in the region during the Roman period, in terms of its role as a fortified centre for the residence of elites and the surrounding population engaged in a range of non-agricultural activities.

The colonization of the upper Guadalquivir valley during the Republican period can be viewed from the perspective of a double strategy of Roman exploitation. The abandonment or destruction of some oppida would establish the scope of the military exploitation of the region, whereas the continuation of others suggests that some kind of regular taxation was introduced immediately after the conquest. It seems likely that the latter would have involved the active participation of the native aristocracy, which implies that they were to some degree integrated into the new order established by the conquerors. The imposition of regular taxation only had repercussions on certain settlements, such as Giribaile, in whose territory a dispersed model of colonization was developed. Elsewhere, on the other hand, local communities occupied stretches of land which hitherto had been neglected; the San Juan River valley is one example. Furthermore, other areas which had been affected by military action at an early stage of the conquest were also reorganised. While these developments clearly represent change, it is difficult to support the idea that these territories were immediately romanized with the development of a Roman civic system, since native elites remained at their head of their social hierarchies with Roman approval (Ruiz Rodríguez 1989).

THE OPPIDUM AND THE TERRITORY OF GIRIBAILE

Location and Background

Systematic surface survey in the region of the oppidum of Giribaile in the middle Guadalimar valley (Gutiérrez et al. 1995; Royo et al. 1995), allows us to develop a model of romanization for the initial early years of the Roman occupation in the innermost lands of the upper Guadalquivir. The survey defined a succession of archaeological horizons in the middle course of the Guadalimar dating to between the first half of the 4th century BC (the foundation date of Giribaile), and the Ist century AD (the first distribution of land to Roman citizens). This long timescale from the Iberian to the early imperial periods was important in providing the necessary chronological perspective in understanding the development of settlement patterns in the region.

Republican Occupation

The survey has revealed a dispersed settlement pattern consisting of approximately one hundred new sites located on flat ground in the Guadalimar valley. Many were located on the first river terrace, and yielded materials dateable to the 2nd century BC. The detailed analysis of the sites by means of transects suggested revealed a fairly standarized ceramic assemblage, which included amphorae. Distribution of the sites, however, suggests that only the best quality agricultural lands were occupied and that, prior to the imposition of a Roman centuriated landscape, this was a largely agrarian mode of exploitation. Only a few sites were located to near rich silver seams or to have taken advantage of them to complement their agricultural economies.

It seems that this new pattern of exploitation, which is probably to began immediately after the conclusion of the Second Punic War, can be understood as a response to the introduction of Roman power at Cástulo, a short distance to the north. The appearance of rural settlements around Giribaile is perhaps to be explained by the need of the community to pay taxes to Rome. Achieving this, however, could only be managed by breaking with more traditional means of exploitation. A proportion of the population would have left the security of the oppida’s fortifications and been settled directly on the land so that it could be worked more intensively and generate the necessary surplus. In this way, the community was effectively paying tribute in return for being able to continue being ruled by its own elite. In some senses, therefore, the advent of the Roman period represented a rupture with the preceding period. At the same time, however, there were important elements of continuity. For example, Rome was clearly respecting pre-existing social links between oppida, which had been articulated through the practice known as fides ibérica (Rodríguez 1948). Moreover, throughout the period, traditional burial places are maintained and there is no evidence for cemeteries in the valley. However, it should be noted that transformations in the landscape similar to those taking place at Giribaile were comparatively rare elsewhere in the upper Guadalquivir valley, and have only been attested at sites like Toya in the easternmost part of Jaén province (Ruiz Rodríguez and Molinos 1996).

Territorial Reorganization during the Sertorian Period

This model of settlement had a relatively brief duration. Recent work has revealed many sites being abandoned in the first half of the 1st century BC, while Giribaile itself was violently destroyed and later replaced by La Monaria, a fortified lowland settlement (Royo et al. 1997). In addition, two towers were constructed as part of a strategy to control the main route of communication running through the Guadalimar valley. At the same time, metalworking refuses at some sites attests to the continued exploitation of some of the richer metal seams in the region. This is part of a large-scale reorgonization of settlement in the region, which involved the re-commencement of mining in the Sierra Morena, on the north side of the Guadalquivir valley, which began towards the end of the 2nd century BC (Domergue 1987). Sites like El Centenillo, Los Escoriales, Palazuelos and Salas de Galiarda were major fortified mining settlements involved in this activity.

However, this new phase of economic exploitation in the valley enjoyed an even briefer period of life than its predecessor. It seems to have come to a violent end, with traces of burning being found at the great majority of settlements, such as Castellones de Céal (Chapa et al. 1998). The discovery of coin hoards in the region also points to general instability in the region, which may perhaps be explained as the result of factional struggles during the Civil War of the middle 1st century BC (García 1991).

The Flavian Period

The archaeological record in the region is now silent until the middle of the 1st century AD. At some time around this date, however, there is evidence that the territory previously belonging to Giribaile was centuriated. Clearly this belonged to a new settlement focus, and it seems likely that this would have been the Flavian municipality of Baesucci, possibly to be identified with present day Vilches (González and Mangas 1991). The associated patterning of rural settlement was very clearly different from that of the preceding periods. It suggests that the land was subdivided into regular lots, each of which corresponded to divisions of centuria, which would then have been distributed to settlers at Baesucci as private property.

THE OPPIDUM AND THE TERRITORY OF ATALAYUELAS

Historical Antecedents and Territorial Definition

Research in this part of Jaén province has attempted to reconstruct the agricultural landscape of a small town between the 1st and 2nd centuries AD. It lies to the south-west of Giribaile and the Guadalimar valley, in what was the easternmost limit of the province of Baetica. Its focus is the oppidum of Atalayuelas (Fuerte del Rey, Jaén), a settlement which was occupied between the Copper Age and the middle of the 2nd century AD.

During the early Imperial period, this was a relatively small site which covered between 5 and 7 hectares. The absence of monumental architecture and epigraphic evidence for possible municipal status initially suggested that it may have been a village (vicus) that might have been dependant upon the Flavian municipality of Aurgi (Jaén), a short distance to the south. Recent archaeological research, however, suggests that this view should be revised - or at least qualified. The regularity of the settlement pattern, as well as the discovery of commemorative inscriptions during the 1987 excavations at Atalayuelas (Castro et al. 1989), suggests that the settlement was in fact a town and possibly yet another Flavian municipality. At present, however, no further corroborative epigraphic evidence has come to light in recently published compilations (González and Mangas 1991), while the inscription from the 1987 excavation is at most, little more than suggestive.

The oppidum of Atalayuelas was probably established in around the 7th or 6th centuries BC. The boundaries of its territory can be traced with some confidence and seem to have followed the line of secondary water courses, usually streams of the third and fourth order, which only coincided with the interfluvial line in areas close to the river headwaters. During the late Iberian period and prior to the Roman conquest in the late 3rd century BC, the oppidum raised a number of towers (turres) to watch over this territory. These were rectangular in plan and constructed from large blocks of stone. They were established at regular distances of between 1 and 2 km to form a fairly impenetrable system of control, similar to that at nearby oppida. Early research has suggested that amongst other purposes, these towers served to defend the territory (Fortea and Bernier 1970; Ruiz Rodríguez et al. 1983; Ruiz Rodríguez et al. 1985; Cortijo 1985), even though it is not clear how this defence would have been articulated. Whatever their function may have been, it is clear that nearly all these towers continued to be used into the Roman period, providing a fairly clear impression of the limits of the territory of the Roman town of Atalayuelas.

Survey work has shown that before the Roman conquest of the Campiña de Jaén, the landscape in which Atalayuelas lies, Iberian oppida were distributed fairly regularly. The first Roman intervention in the region, which was intended to strengthen military control of the region and establish the regular taxation of conquered communities, led to the abandonment of some oppida. This has been verified by excavations at the Plaza de Armas de Puente Tablas (Ruiz  Rodríguez and Molinos 1987) and is suggested by surface survey at other sites, such as the Cerro Villargordo and Torrejón (Ruiz Rodríguez et al. 1992). However, most settlements, such as Atalayuelas, continued to be occupied. Later, after the conclusion of the Cantabrian Wars in the later 1st century BC, a contingent of legionary veterans was settled on the site of an Iberian oppidum to found the Colonia Augusta Gemella Tucci (Serrano 1987). This coincided with the appearance of rural settlements, that was almost unique in the region. It was a dramatic innovation because hitherto there were no small permanent settlements in the upper Guadalquivir valley. Even though little is known about farming in the Iberian period, it seems that exploitation of the land was managed from the oppida, and may have been organized on a collective basis.

Indeed, it is not until the late 1st century AD that one sees a similar process happening elsewhere in the region and seems to have coincided with the promotion of native communities to municipal status. This coincided with important changes in the character of the oppida themselves. There is evidence for the advent of true urban-style planning and the private ownership if land, which gave rise to a distinctive agrarian landscape.

The Agrarian Landscape in the Flavian Period

Analysis of the survey evidence suggested that archaeological sites tended to concentrate in the southern area of the territory of Atalayuelas. The reason for this was not immediately apparent. However, given that the settlement might have been a Flavian municipality, it seemed possible that it reflects the way in which the land around it may have been assigned and distributed. In an attempt to verify this hypothesis, the distribution of rural sites were analyzed to establish whether or not they suggested possible Roman land divisions, rather than adducing the rural settlements to confirm the existence of pre-determined land-divisions. As it happens, analyses of aerial photographs and 1: 10,000 maps did suggest the existence of reticulate divisions of the landscapes. By themselves alone they do not prove that the landscape was regularly divided up. However, when taken in conjunction with the survey, they do suggest the existence of a Roman system of land-division.

Consequently, there does seem to have been some kind land division in the territory of Atalayuelas in the late 1st century AD. A synthesis of the consecutive steps which led to the configuration of this particular model of land-division was prepared (Fig.000). This is an attempt to rationalize the decisions that led to the division of the land, rather than a dissection of the sequence followed in their gestation. The divided lands form a compact block whose centre is situated at the highest point of the territory, and near the Roman town. From this point, the main axes of the land-division would have been made to coincide with the longer axes of the territory, which deviate approximately 15.3º in relation to the geographical north-south axis. The smallest unit of the division has been the square centuria of 706 m x 706m. A group of four centuriae, named saltus by the Roman surveyors (Ariño 1990), is formed by three assigned centuriae and one centuria free of occupation. The latter may have been exploited from the nearby assigned plots, possibly as common grazing land belonging to the local saltus). The best justification for the distribution of the settlements was the second division of the land inside the centuria boundaries, which should have consisted of its division into three lots or plots of the same size. The first of these was parallel to the decumanus maximus and occupied the northern third of the centuria, while the second and third plots were in the two remaining thirds and their main boundaries would have been transverse to the first. The assigned plot had a surface area of 16.6 hectares, while the overall group of arable lands and grazings that were created by this division covered an area of 3143 hectares,  which is approximately half the territory of Atalayuelas.

The remaining half of the territory of the town was organized in a different way. The land where the Iberian turres were sited, formed an almost continuous belt around the periphery of the territory, and was only interrupted at points where centuriated land lay adjacent to the borders of neighbouring territory. This unassigned land was used to support longer-term agricultural strategies. On the other hand, unassigned land lying away from the edges of the territory of Atalayuelas, particularly to the north, remained unoccupied. This was presumably given over to forests, which presumably would have been developed on a collective basis.

One exception to this situation in the north, consists of an important rural settlement which is located close to where the unoccupied land abutted assigned lands. Indeed, it is the only example from the territory of 1st and 2nd century Atalayuelas of a settlement which may be identified as a Roman villa; all other rural sites are perhaps better identified as farms. This site, called Buenaharina, is located beneath the modern village of Fuerte del Rey and covers a slightly larger area than most Flavian sites recorded to date. The site has yielded columns, wallplaster, tesserae and other evidence which suggests that it boasted a magnificent residence, while the presence of large millstones and dolia indicates that it was also involved in large-scale agricultural production.

CONCLUSIONS

The traditional conception of romanization in southern Hispania has been based on the idea of an early appearance of Roman towns. However, this does not help explain the pattern and development of rural settlement in the upper Guadalquivir during the Republican and early Imperial periods. Indeed, this part of Iberia is distinctive in that the territorial changes contingent upon the emergence of Roman towns do not appear until comparatively late. They either coincide with the first Augustan colonies or the with extension of municipal status to native communities under the Flavians. When the Roman town eventually appears as an extended and generalized phenomenon of territorial organization, it would appear it was based on an agrarian economy supported by a predominance of the small and medium peasant property-owners. This conclusion challenges the traditional conception of Baetica as a province where specialized agriculture and the slave mode of production enjoyed wide currency.

The late establishment of Roman towns in the upper Guadalquivir does not imply that the Roman conquest had no immediate consequences for native communities. Indeed, at least two alternative scenarios can be outlined: the appearance of new settlement patterns around oppida like Giribaile, and the continuity of the Iberian settlement pattern, as in the case of Atalayuelas. These different cases can be understood as symptomatic of the different responses to the presence of Rome that are conventionally labelled as “romanization”. Alternatively, they can be interpreted as Rome’s adaptation of strategies of control and domination to the distinctive social and political conditions of each native community. This hypothesis seems to be the most reasonable approach to understanding the divergences found in the archaeological evidence, and is the main interpretative framework for our current research into the romanization of the upper Guadalquivir.

BIBLIOGRAPHY

Ariño, E., 1990. Catastros Romanos en el Convento Jurídico Caesaraugustano. La Región Aragonesa, Departament de Ciencias de la Antigüedad-Universidad de Zaragoza, Zaragoza.

Arteaga, O. and Blech, M., 1987. "La romanización en las zonas de Porcuna y Mengibar (Jaén)", in VVAA, Coloquio Los Asentamientos Ibéricos ante la Romanización (27-28 febrero 1986), Ministerio de Cultura-Casa de Velazquez, Madrid, 89 - 101.

Barbero, A. and Vigil, M., 1974. Sobre los Orígenes Sociales de la Reconquista, Ariel, Barcelona.       

Bendala, M., 1981. "La etapa final de la cultura ibero-turdetana y el impacto romanizador", in VVAA, La Baja Época de la Cultura Ibérica, Asociación Española de Amigos de la Arqueología, Madrid, 30 - 48.

Castro, M., 1989. 'De César a Teodosio (49 a.C. - 395 d.C.)'. in VV.AA., Jaén, tomo II, Editorial Andalucía, Granada, 423-441.

Castro, M., 1993. 'Reconstruyendo un paisaje agrario: La Campiña de Jaén en los siglos I-II'. in Actas de las III Jornadas Históricas del Alto Guadalquivir (Quesada, mayo de 1993), i.p.

Castro, M. and Choclán, C., 1988. 'La Campiña del alto Guadalquivir en los siglo I-II d.C.: Asentamientos, estructura agraria y mercado', Arqueología Espacial, 12, 205 - 221.

Castro, M., Hornos, F. and Choclán, C., 1993. 'Cabeza Baja de Encina Hermosa (Castillo de Locubín, Jaén): Una reflexión sobre el desarrollo del territorio ciudadano en la Campiña'. in Actas del I Coloquio de Historia Antigua de Andalucía (Córdoba, 1988), Publicaciones del Monte de Piedad y Caja de Ahorros de Córdoba, Córdoba, 451 - 467.

Castro, M., López, J., Zafra, N., Crespo, J. and Choclán, C., 1989. 'Prospección con sondeo estratigráfico en el yacimiento de Atalayuelas, Fuerte del Rey (Jaén)'. in VV.AA., Anuario Arqueológico de Andalucía II/1987, Consejería de Cultura y Medio Ambiente, Sevilla, 207 - 215.

Chapa, T., Pereira, J., Madrigal, A. and Mayoral, V., 1998. La Necrópolis Ibérica de Los Castellones de Céal (Hinojares, Jaén), Junta de Andalucía-Universidad de Jaén, Jaén.

Cortadella, J., 1988 "M. Almagro Basch y la idea de la unidad de España", in Studia Historica-H.ª Antigua, vol. VI.

Cortijo, M.ªL., 1993. La Administración Territorial de la Betica Romana, Caja Provincial de Ahorros de Córdoba, Córdoba.

Cunliffe B.W. and Fernández, M.ªC., 1990. "Torreparedones (Castro del Río-Baena, Córdoba). Informe preliminar. Campaña de 1987. Prospección arqueológica con sondeo estratigráfico", in VV.AA., Anuario Arqueológico de Andalucía II/1987, Consejería de Cultura de la Junta de Andalucía, Sevilla, 193 - 200.

Domergue, C., 1987. Catalogue des Mines et des Fonderies Antiques de la Péninsule Ibérique, Diffusion de Boccard, Madrid.

Fernández Castro, M.ª C., 1982. Villas Romanas en España, Ministerio de Cultura, Madrid.

Fernández García, M.ªI., 1998. Terra Sigillata Hispánica. Estado Actual de la Investigación, Universidad de Jaén, Jaén.

Fortea, J. and Bernier, J., 1970. Recintos y Fortificaciones Ibéricos en la Bética, Universidad de Salamanca, Salamanca.

Funari, P.P.A., 1986. "As estrategias de exploraçao de recursos do Vale do Guadalquivir em época romana", Revista Brasileira de História.

García, F., 1991. Quinto Sertorio. Roma, Universidad de Granada, Granada.

González, C., 1981. Imperialismo y Romanización en la Provincia Hispania Ulterior, Universidad de Granada, Granada.

González, C. and Mangas, J., 1991. Corpus de Inscripciones Latinas de Andalucía, volumen III: Jaén, tomo I, Consejería de Cultura de la Junta de Andalucía, Sevilla. 

Gros, P., 1988. 'L'età imperiale'. in P. Gros and M. Torelli, Storia dell'Urbanistica: Il mondo romano, Laterza, Bari, 165 - 426.

Gutiérrez, L.M., 1998.  El poblamiento ibérico en el curso medio del río Guadalimar. Tesis doctoral. Microfichas, Servicio de Publicaciones de la Universidad de Jaén, Jaén.

Gutiérrez, L.M., Royo, M.A., Barba, V. and Bellón, J.P., 1995. 'Informe sobre la primera campaña de prospección superficial en el Guadalimar Medio-Hinterland de Cástulo'. in VV.AA., Anuario Arqueológico de Andalucía II/1992, Consejería de Cultura y Medio Ambiente, Sevilla, 249 - 256.

Hornos, F., Zafra, N. and Castro, M., 1998. 'La gestión de una zona arqueológica urbana: La experiencia de investigación arqueológica aplicada en Marroquíes Bajos (Jaén)', Boletín del Instituto Andaluz del Patrimonio Histórico, 22, 82 - 91.

León, P. and Rodríguez, P., 1993. 'La ciudad hispanorromana en Andalucía'. in Catálogo de la Exposición: La Ciudad Hispanorromana, Ministerio de Cultura, Madrid, 13 - 53.

Roca, M., 1976. Sigillata Hispánica producida en Andújar, Instituto de Estudios Giennenses, Jaén.

Rodríguez, F., 1948. "La fides ibérica", Emérita, XIV.

Royo, M.A., Gutiérrez, L.M., Bellón, J.P. and Barba, V., 1995. 'Prospección arqueológica superficial de urgencia en la Presa de Giribaile (Jaén)'. in VV.AA., Anuario Arqueológico de Andalucía III/1992, Consejería de Cultura y Medio Ambiente, Sevilla, 408 - 414.

Royo, M.A., Gutiérrez, L.M., Bellón, J.P. and Barba, V., 1997. 'Documentación gráfica del yacimiento romano de La Monaria'. in VV.AA., Anuario Arqueológico de Andalucía III/1993, Consejería de Cultura y Medio Ambiente, Sevilla, 386 - 390.

Ruiz, A., 1987. "Ciudad y territorio en el poblamiento ibérico del Alto Guadalquivir", in VVAA, Coloquio Los Asentamientos Ibéricos ante la Romanización (27-28 febrero 1986), Ministerio de Cultura-Casa de Velazquez, Madrid, 9 - 21.

Ruiz, A., 1989. 'La Protohistoria: El 1.º milenio a.n.e.'. in VV.AA., Jaén, tomo II, Editorial Andalucía, Granada, 401 - 422.

Ruiz, A., Castro, M. and Choclán, C., 1992. 'Aurgi-Tucci: La formación de la ciudad romana en la Campiña de Jaén', Dialoghi di Archeologia, 3.ª Serie, 1-2, 211 - 229.

Ruiz, A. and Molinos, M., 1987. 'Excavación arqueológica sistemática en Puente Tablas (Jaén)'. in VV.AA., Anuario Arqueológico de Andalucía II/1986, Consejería de Cultura y Medio Ambiente, Sevilla, 401 - 407.

Ruiz, A. and Molinos, M., 1993. Los iberos. Análisis arqueológico de un proceso histórico, Crítica, Barcelona.

Ruiz, A. and Molinos, M., 1996. "Desde los cazadores-recolectores a los agricultores", in VV.AA., Historia de Jaén y su provincia, Ideal, Granada.

Ruiz, A., Molinos, M., Hornos, F. and Choclán, C., 1985. "El poblamiento ibérico en el Alto Guadalquivir", in VV.AA., Iberos. Actas de las I Jornadas sobre el Mundo Ibérico/Jaén 1985, Ayuntamiento de Jaén-Junta de Andalucía, Jaén, 239 - 256.

Ruiz, A., Molinos, M., López, J., Crespo, J.M.ª, Choclán, C. and Hornos, F., 1983. "El horizonte ibérico antiguo del Cerro de la Coronilla (Cazalilla, Jaén). Cortes A y F", Cuadernos de Prehistoria de la Universidad de Granada, 8, 251 - 299.

Segura, M.ªL., 1988. La ciudad Ibero-romana de Igabrum (Cabra, Córdoba), Diputación de Córdoba, Córdoba.

Serrano, J.L., Zafra, J., Sánchez, M.ªC. and Chica, M.ªP., 1997. 'Intervención arqueológica de urgencia en el polideportivo de Martos (Jaén) y terrenos aledaños-1993'. in VV.AA., Anuario Arqueológico de Andalucía II/1993, Consejería de Cultura, Sevilla, 367 - 374.

Serrano, J.M., 1987. La Colonia Romana de Tucci, Asociación Artística Cultural Tucci, Jaén.

Sotomayor, M., 1977. Marcas y estilos en la Sigillata decorada de Andújar, Instituto de Estudios Giennenses, Jaén.