Approaching Interaction in Iron Age Sardinia: Multi-Scalar Survey Evidence from the Sinis Archaeological Project and the Progetto S’Urachi

2.1 Interaction, Connectivity, and Scale

A multi-scalar approach to first millennium BCE interactions in Sardinia requires both a conceptual reframing of the data and a clear methodological strategy. “Interaction” is a slippery term, encompassing all manner of social and material interconnections in archaeological scholarship, so we first begin with our definition. By “interaction,” we mean both direct communication between people from different places (or social interaction) and the exchange of ideas and goods (immaterial and material interactions, respectively) that can accompany this direct communication or happen even without it through indirect processes like trade. Interaction therefore encompasses both human mobility, the physical movement of people from one place to another, and “connectivity,” or the links across space and time created through the movement of objects, ideas, and people (Gosner & Hayne, in press). Therefore, interactions are “the processes that create ‘connectivity’” (Dietler, 2022, p. 239).

In recent years, after Horden and Purcell’s (2000) push to consider the Mediterranean as a singular unit of study in The Corrupting Sea, scholarship has focused on revealing the increasing “connectedness” of the Mediterranean in the first millennium BCE. One prominent framework for conceptualizing this connectedness is to apply modern theories of globalization to antiquity (Hodos, 2010, 2017, 2020; Pitts & Versluys, 2015; Versluys, 2014). Globalization is characterized not only by the widespread flow of ideas, objects, and customs through increasing connectivity, but also by the growing social awareness of increased connection (Hodos, 2017, p. 4). Although some applications of globalization emphasize the spread of shared practices and values (De Angelis, 2013; LaBianca & Arnold Scham, 2006; Pitts & Versluys, 2015), this increasing sense of one-placeness can elicit a resurgence of local identity in an effort to draw deliberate contrasts, creating greater distinctions between the local and the global (Hodos, 2022, p. 22).

In considerations of both connectivity and globalization, rural localities and landscapes have been understudied because they have been believed to exist outside of the impact of intensified interconnection. Recent considerations of “rural globalization,” however, show that these non-urban regions bred specific rural identities that also saw change as a result of increased interaction in ways that are different from their urban counterparts (Nowlin, in press; van Dommelen, 2019; van Dommelen & Gómez Bellard, 2008; Woods, 2007). Such an investigation of rural groups in their regional context can illuminate the value that non-coastal groups placed on foreign goods, ideas, peoples, and practices and the ways in which their lives both shaped and were shaped by greater connectivity. Efforts should therefore be made to see the extent to which Mediterranean networks extended beyond the coast and into inland locales (e.g., Kearns, 2022a,b) as well as the ways in which shifting connections affected the relationship between local groups and their landscapes, as can be observed in changes in settlement patterns, resource utilization, territorial control, and the formation and reinforcement of boundaries.

The overt modern implications of globalization as a theoretical framework have led some scholars to opt for a seemingly less-laden concept of “connectivity” in its place (e.g., Leidwanger & Knappett, 2019). By focusing on connectivity in an abstract sense rather than categorizing the interaction through specific historical processes, the framework of connectivity has been applied to outline the specific nature of interaction between different regions. As Dietler has warned, rather than simply identifying the presence of connectivity and quantifying its intensity and degree, greater emphasis needs to be placed on explaining the “nature and logic of interaction” (Dietler, 2022, p. 239). While the flattening nature of connectivity has been part of its broad appeal, it also risks obfuscating or ignoring the messy and often violent results of the interactions that come from greater connectivity. As Morris (2003, p. 43) has stated, a globalization approach is “more aware that history has winners and losers” than narratives crafted around connectivity. This has been a recent emphasis in post-colonial approaches to interactions in the Iron Age and Roman periods (Dietler, 2010; Fernández-Götz, Maschek, & Roymans, 2020; Osanna, 2017; Padilla Peralta, 2020; van Dommelen & Rowlands, 2012; for additional bibliography: Ramgopal, 2022), highlighting the conflicts that can arise from the power differentials that are created or exacerbated by greater connectivity.

In light of the recent scholarly emphasis on connectivity in the first millennium BCE Mediterranean, it is essential to remember that connectivity in this period was not inevitable and some communities may have deliberately chosen not to engage within the wider Mediterranean world based on local decision-making (Riva & Grau Mira, 2022, p. 25). Such universalizing assumptions are the risk of top-down approaches that apply large-scale models without making allowances for human agency and the specificity of local context. Recently, there have been greater calls for employing both a microhistorical and multi-scalar approach as a means of building historical explanations from the ground up (e.g., Beck, 2020; Dietler, 2022; Gosner & Hayne, in press; Riva & Grau Mira, 2022). It is argued that either by enlarging or shrinking the scale at which certain phenomena are studied – both geographically and temporally – or by tacking back and forth between scales, contradictions in grand historical narratives can be revealed and investigated further (Dietler, 2022, pp. 235–236; Riva & Grau Mira, 2022, p. 5).

Multi-scalar approaches offer the ability to build a new view of connectivity that accounts for the experiences of the diverse indigenous groups that inhabited the Mediterranean in the first millennium BCE. This is especially true when such approaches incorporate analysis of rural areas that have been brought into processes of rural globalization and fully acknowledge that not all groups were impacted by connectivity in equal ways. Our conceptual approach to interaction, therefore, explores mobility and connectivity in rural landscapes from multi-scalar perspectives and seeks to explore how interactions impacted such communities in unique or specific ways.

2.2 Multi-Scalar Archaeological Survey Methodologies for Assessing Interaction

The implementation of a multi-scalar conceptual framework for interaction demands rigorous integration of field methodologies that target different scales of analysis, as the conceptual frameworks outlined above can fall short without practical attention to how they relate to the archaeological record. Investigations of connectivity have tended to utilize data from individual objects or sites which in turn point to the long-distance migration of peoples and movement of objects or ideas. They use an analysis of the micro-scale to illuminate macro-scale processes. These approaches typically rely on long-term excavations at single sites (often coastal and urban) and object biography approaches to the artifacts recovered from them. What is often lost when employing such methodologies is the regional scale, which is critical for considering both the impacts of interaction within regional landscapes and the effects that intra- and inter-regional exchange, not simply long-distance exchange, had on local groups. A regional scale analysis is necessary for accessing the impacts of connectivity on rural landscapes, which are often omitted from the larger narratives of first millennium BCE Mediterranean connectivity, as discussed above. This is a task that survey methodologies are well suited to address (Caraher & Pettegrew, 2016).

We suggest, therefore, that a multi-scalar approach to landscape survey can provide a solution to these dual challenges, both the lack of regional-scale data and the underrepresentation of rural landscapes in narratives of interaction and connectivity. This approach can be particularly clarifying when its results are integrated with data produced through other methodologies such as excavation and genetic analyses. Landscape survey across the Mediterranean basin has provided unprecedented meso- or intermediate-scale data over the past 50 years (Alcock & Cherry, 2004; Attema et al., 2020; Barker, 1996; Cherry, 1983; Knodell, Wilkinson, Leppard, & Orengo, 2023 with substantial bibliography; for examples of central Mediterranean survey projects, see Foxhall et al., 2007; Holt, Aguilar, & Schirru, 2022; Kolb, Vecchio, & Giglio, 2021; Murphy, Leppard, Roppa, Madrigali, & Esposito, 2019; Sevara et al., 2020; see FastiOnline for a database of archaeological surveys in the Mediterranean). Mediterranean landscape survey techniques are diverse but encompass a common set of field methods grounded in the systematic documentation of surface remains using multitiered, multidisciplinary research strategies, often including pedestrian transect or grid-based sampling, as well as geomorphological, geophysical, architectural, historical, and ethnographic components (Knodell et al., 2023, p. 266).

For accessing the archaeology of interaction, we have employed a three-tiered, multi-scalar approach implementing different methodologies at each scale of analysis:

1. Small Scale: Site-based survey and targeted excavation around known sites of interaction. Site-based survey can be conducted using “gridded collection,” which involves laying out a systematic grid across a known site and then collecting artifacts from the surface to analyze their spatial distributions (Attema et al., 2020, pp. 3–4, 47; Knodell et al., 2023, p. 266). Alternatively, “point sampling,” the method we employed and detail below, involves the collection of artifacts at delineated points on a grid, usually by excavating small test trenches or clearing the immediate surface to uncover artifacts (van de Velde, 2001). This method can be more effective than gridded collection at sites that have never been plowed or previously excavated, or those where heavy vegetation impedes visibility (for application of gridded collection and point sampling in combination, see Balco, Kirk, & Kolb, 2020). Site survey can be combined with targeted excavation of small test trenches at points of interest revealed through gridded collection or point sampling, which help clarify and contextualize the results of the site survey.

2. Intermediate Scale: Pedestrian survey in landscapes that fall in between known sites of interaction. Pedestrian survey or “intensive survey” involves systematically exploring the landscape with teams of field walkers who traverse transects (typically 10–20 m apart) within survey units and collect and/or count artifacts along their path. Survey units are bound by existing field boundaries or artificial grids imposed on the terrain (Attema et al., 2020, pp. 3–4, 47; Knodell et al., 2023, p. 266). Once analyzed and mapped, the distribution of artifacts can reveal information about the chronologies and wider patterns of landscape use. This method works best in plowed agricultural fields where artifacts are periodically brought to the surface and are therefore more likely to be visible to field walkers. Its utility in agrarian landscapes is particularly important for assessing interaction in rural locales outside of the nodes, or sites, that are the focus of most studies. Such an approach is sometimes referred to as “siteless survey” because analysis is based on the presence of artifacts rather than focusing on the presence (or absence) of settlements (Caraher, Nakassis, & Pettegrew, 2006; Caraher & Pettegrew, 2016, p. 137). Point sampling, the method discussed above, is also effective across large non-agricultural or marginal landscapes in which traditional intensive survey provides limited results because of the lack of visibility (Attema et al., 2020, pp. 6–7; for regional-scale implementation in the Riu Mannu Survey, see Annis, van Dommelen, & van de Velde, 1995; van de Velde, 2001; van Dommelen, 1998).

3. Large Scale: Extensive survey across wider landscapes of interaction. Extensive survey involves regional reconnaissance, where activity focuses on finding and identifying sites visible in the landscape, such as standing architecture, and/or ground-truthing, or verifying, the location of sites predicted to exist through other remote sensing methods (Knodell et al., 2023, p. 266).

Landscape survey, and particularly intermediate-scale pedestrian survey, has often been treated as a stand-alone methodology in its own right, rather than a means to an end (i.e., to find a new place to excavate). Consequently, survey results are not always re-integrated with sources of data at other scales in ways that can inform our view of interaction and connectivity. This is, in large part, because survey data can answer questions about landscape use and chronologies of habitation that have been more central concerns of survey archaeologists’ research in recent decades. Additionally, the lack of integration of survey and excavation data can sometimes be attributed to the separate permitting process required for excavation and survey (the procedure in Greece and now also in Italy, for instance). Nevertheless, for questions of interaction, the data produced through this tripartite approach to survey must then be integrated with other available sources. In the chapter discussed above, Dietler (2022, p. 236) suggests that any multi-scalar conceptual approach is best “accomplished by reorienting research practices towards teams of scholars working together in multi-sited projects with collaborative strategies and publications.” The multi-scalar survey methodology outlined here, therefore, will be most fruitful when its results are re-integrated with existing narratives of connectivity produced through site-level excavation, genetic data, and other methodologies in the archaeologists’ toolkit.

3.1 Small Scale: Site Survey at S’Urachi and Su Padrigheddu

S’Urachi is among the largest nuraghi in the Alto Campidano region of west-central Sardinia (Tore & Stiglitz, 1987a,b). This monumental basalt stone tower complex was constructed in the Bronze Age and long-term habitation is known outside the tower walls lasting through the early Roman period, with some sporadic evidence through the Late Roman period. Su Padrigheddu, an adjoining settlement once thought to be a cemetery, has provided additional evidence for Iron Age habitation near the nuraghe (Roppa, 2012). Archaeological work at S’Urachi in the 1980s focused primarily on excavations of the tower itself (Tore, 1984a,b), and recent work has focused on the occupation immediately abutting the tower walls, providing a deep but spatially constrained view of interaction at the site. These recent excavations have shown that S’Urachi was a particularly important Nuragic settlement in the region during the Iron Age, where local people living in this inland location became increasingly intertwined with colonial networks (Pérez-Jordà, Hurley, Ramis, & van Dommelen, 2020; Stiglitz, Díes Cusí, Ramis, Roppa, & van Dommelen, 2015; van Dommelen et al., 2018).

Our small-scale survey aimed to better understand the occupation phases and land use around S’Urachi using point sampling and targeted excavation. We used an existing 20 m × 20 m grid, collecting material at each of the 63 points on the grid within the archaeological site (Figure 2). These collection points were each roughly 2 m² in size, laid out in circles with an 81 cm radius for expediency. Because the site is enclosed by walls and not used for agriculture, surface visibility was low at all points. We therefore removed the topsoil, clearing each point down to c. 10 cm and collecting all archaeological material recovered through screening. Using this method of clearing 2 m² for every 20 m × 20 m in the grid, we achieved 0.48% coverage of the site (Gosner & Smith, 2018, pp. 8–10). Once collected, we analyzed the ceramic material, grouping it into chronological/cultural categories and then mapping the proportions of material across the site. The categories were as follows: Late Bronze Age-Iron Age Nuragic (twelfth–eighth centuries BCE), Iron Age Phoenician (ninth–sixth centuries BCE), Punic-Early Roman (sixth–first centuries BCE), and Imperial Roman (first–fifth centuries CE) (Gosner & Smith, 2018, p. 11).

Figure 2

Map of the site-based survey at S’Urachi and Su Padrigheddu, including the collection points and test trench locations.

We found 63 sherds from the earliest chronological category (twelfth–eighth centuries BCE), which came primarily from S’Urachi’s northern border near the modern road and from Su Padrigheddu, southeast of the nuraghe (Gosner & Smith, 2018, pp. 13–14). Iron Age material that could be categorized as Phoenician was particularly scarce, with only one Phoenician sherd coming from Su Padrigheddu. This may be due to the fact that most Iron Age material was produced in local clays and styles and therefore categorized as Nuragic within our scheme. Additionally, some sherds categorized as Punic-Early Roman may have been produced earlier, but the continuity of form across the Phoenician and Punic-Early Roman chronological groupings meant it was difficult to pinpoint the date of surface materials recovered from survey without stratified contexts (Gosner & Smith, 2018, pp. 15–16). Finally, Punic-Early Roman ceramic material was pervasive, with 1,772 sherds recovered from point sampling, with particularly high quantities found north of the nuraghe. Ceramics included Punic amphorae and other storage vessels, commonware in local fabrics, and imported and locally produced tableware. Imports came from North Africa, peninsular Italy, and Attica (Gosner & Smith, 2018, pp. 16–17). These patterns suggest that the earliest material interactions with the outside world were limited to the import of a small number of objects from elsewhere, and perhaps social interaction with the people who brought them. As time progressed, material connectivity indicated more intensified interactions of all types.

To explore these patterns, we augmented the survey by excavating a series of test trenches at points of particular interest revealed through point sampling in 2016 and 2017 (Gosner et al., 2021). In Su Padrigheddu, we excavated a 1.5 m × 1.5 m trench (Test Trench 1) at the location where Iron Age Nuragic and Phoenician material had been recovered in survey and from previous unsystematic collection after the area had been deep plowed to plant eucalyptus trees (on Su Padrigheddu: Roppa, 2012; on Test Trench 1: Gosner et al., 2021, pp. 5–8; Gosner, Hayne, Madrigali, & Nowlin, 2020, p. 1708; Madrigali, Gosner, Hayne, Ramis, & Nowlin, 2020, p. 110). Although there was no intact stratigraphy from the settlement, the ceramic and faunal materials from the trench helped to fill in the chronological gap in the site survey for the Early Iron Age as well as illuminate additional aspects of interaction between local Nuragic inhabitants and Phoenicians. Most of the ceramic was locally produced in the Nuragic tradition, including vessels for food preparation and cooking as well as a small collection of fineware, which dated the assemblage to around the eighth century BCE (Figure 3, nos. 1–5). However, these were often finished on a slow wheel, an innovation adopted by Nuragic potters as a result of external interactions: either exposure to imported pottery types or contact with foreign artisans. The assemblage also included Phoenician material, attesting at least to material connectivity. Phoenician ceramics included primarily tableware and fineware classes of the mid-seventh–sixth centuries BCE. Several Phoenician amphorae were also recovered dating to the late seventh century BCE, imported from as far as the Iberian Peninsula (Figure 3, nos. 6–11; Gosner et al., 2020, pp. 1708–1711). The faunal remains recovered alongside this ceramic assemblage consisted of bovines (39%), wild deer (25%), wild boar (21%), and sheep and goats (15%) (Gosner et al., 2020, p. 1711; Madrigali et al., 2020, pp. 114–115). The inhabitants of S’Urachi were therefore consuming locally available wild animals and raising some livestock, particularly cattle, and cooking them in vessels consistent with local traditions. At least by the mid-seventh century BCE, however, Phoenician trade networks brought in vessels and consumables from outside of Sardinia. These networks also seem to have brought Phoenician ways of dining to S’Urachi – even though traditional foods were being prepared and eaten, they were served in Phoenician styles of tableware. “Interaction” was therefore much deeper than sporadic trade and extended to transformations in the quotidian practices of food preparation and dining.

Figure 3

Pottery classified as Nuragic (nos. 1–5) and Phoenician (nos. 6–11) from Test Trench 1 of the site-based survey at Su Padrigheddu (Gosner et al., 2020, Figure 3).

In sum, the results from this site-based survey illuminated interactions at S’Urachi over the course of the first millennium BCE at the smallest scale. The earliest Iron Age ceramics were produced primarily in local traditions, with relatively few imported wares, suggesting cultural continuity with the Late Bronze Age and relative isolation of the inland Nuragic community at S’Urachi and Su Padrigheddu. Over time interaction intensified, with imported goods increasing at the site. Beyond trade, cultural and technological influences were also visible in the ceramic record. This suggests interaction among artisans in the production sphere as well as social interactions that bled into practices of everyday life related to consumption. By the later Iron Age imports were pervasive, showing that the local community had access to goods from elsewhere and, by extension, more sustained interaction with outsiders via trade.

3.2 Intermediate Scale: Pedestrian Survey in Zone A

At the intermediate scale, SAP conducted a pedestrian survey in the wider landscape surrounding S’Urachi within an area termed Zone A. This zone was surveyed between 2018 and 2022 and subsequent work has focused on Zone B located to the west of S’Urachi closer to the coast. Zone A is at the northernmost point of Sardinia’s Alto Campidano, a large plain used today and in antiquity for agriculture and pastoralism. Zones A and B are within the boundaries of the Comune di San Vero Milis because of permitting considerations. Because of the modern agricultural land use for crops including cereals, vineyards, vegetables, and olives, the soil is regularly plowed or cleared. These conditions mean that terrain is ideally suited to the intermediate-scale pedestrian survey techniques described above. We focused on surveying fields with good visibility (i.e., above 60%) in the proximity of S’Urachi to situate the site-based, small-scale data in its regional context. Fields were identified visually as well as through daily NDVI (normalized difference vegetation index) satellite imagery, which assisted in the selection of recently plowed fields ideal for survey (Plekhov, Gosner, Smith, & Nowlin, 2020).

In Zone A, we surveyed 473 survey units, totaling 3.1 km². Although we collected and counted material from all periods, we focus solely on the first millennium BCE ceramic assemblage in this article. Ceramics identified as Nuragic and Phoenician were exceptionally rare in the overall assemblage (Figure 4). These included 171 Nuragic and 22 Phoenician sherds. The Phoenician pottery comprised both imported amphorae and painted tableware as well as locally produced wares in Phoenician styles, hinting that patterns observed at S’Urachi were mirrored in the more rural territories outside of the settlement. Sherds dating to the Punic-Early Roman period were more pervasive across the landscape than were those of the Early Iron Age, with 1,204 sherds of this later period recovered across Zone A (Figure 5).

Figure 4

Survey units in Zone A showing the concentration of Nuragic and Phoenician ceramic materials across the rural landscape around S’Urachi.

Figure 5

Survey units in Zone A showing the concentration of Punic-Early Roman ceramic materials across the rural landscape around S’Urachi.

In general, ancient ceramics clustered in areas of higher elevation, close to the locations of ancient waterways but beyond the reach of their flood zones. Areas with a higher density of ceramic finds were often near known nuraghi located within Zone A. Most of those locations with Iron Age material were therefore reoccupations of earlier Bronze Age settlements rather than new foundations. This pattern of long-term occupation and increasing interaction seems to have been true not only for the inhabitants of S’Urachi and Su Padrigheddu but also for those of the smaller settlements associated with single tower and trilobate nuraghi throughout the zone.

The frequent recovery of Punic-Early Roman ceramics beyond such known sites, however, suggests that the rural land was beginning to be used more intensively by the later Iron Age for agricultural purposes. Material connectivity was not only coming in the form of trade and some contact with outsiders, as seems to have been the case for the Early Iron Age. Instead, agricultural productivity was increasing among local communities, possibly in response to the demand for external trade contacts, first with the Carthaginians and then the Romans. These same patterns of intensified interaction in the rural landscape are mirrored in other central Mediterranean landscapes of the fourth and third centuries BCE. Survey archaeology in parts of peninsular Italy shows “a dynamic infilling” of rural areas during this period consistent with an increase in population and intensification of agriculture (Padilla Peralta & Bernard, 2022, p. 14, with substantial bibliography). This same phenomenon can be seen in the landscapes around Carthage (Fentress & Docter, 2008), which also provides a logical point of comparison given that similar strategies of Punic landscape intervention may have been at play in west-central Sardinia.

The intensive survey, then, provides a regional complement to the small-scale patterns observed at the site of S’Urachi itself and gives an indication of how interactions were affecting those who lived outside of the largest indigenous settlement of the region. In general, the patterns at S’Urachi are mirrored at smaller Nuragic sites, where inhabitants continued to dwell in places of Bronze Age foundation but came into increasing contact with foreign materials and ideas. The data also suggest that these more limited interactions eventually gave way to widespread changes in land use for agriculture by the later Iron Age, which would have changed local ways of life as rural communities focused on producing surplus goods for external consumption. These changes would have brought such communities into more frequent and more complex social and economic relationships with the outside world.

3.3 Large Scale: Extensive Survey in Zone A

We complemented the intermediate-scale pedestrian survey with regional reconnaissance (or extensive survey) across Zone A, allowing for the investigation of points of interest beyond S’Urachi and the more restricted swath of terrain targeted in the pedestrian survey (Figure 6). These points were identified either through the pedestrian survey itself or through previously published archaeological work (e.g., a shapefile of known archaeological sites available at Sardegna Geoportale: https://www.sardegnageoportale.it/). Using this method, we recorded 34 points of interest in the wider landscape, including – for ancient periods – settlements, nuraghi, cemeteries, roads, bridges, fountains, wells, and quarries.

Figure 6

Extensive survey territory in Zone A marking the locations of points of interest.

The pedestrian survey yielded many more Punic-Early Roman ceramics than those from any other period. By contrast, the points of interest recorded through regional reconnaissance that dated to this period made up a small proportion of the overall number. This seems to have been because many sites were reoccupied rather than newly constructed during this period. Many of the 12 nuraghi recorded in Zone A have evidence of Punic-Early Roman period reoccupation, not just in the form of ceramics, but sometimes also in architectural modification. Nuraghe Zoddias is one such structure where later occupation was evident in the presence of ceramic scatters on the ground as well as the construction of later walls for field boundaries that reused basalt blocks from the original Nuragic complex (Maisola, 2012, pp. 740–747; Taramelli, 1935, p. 167; Tore & Stiglitz, 1987a, pp. 97, 100, 105; Usai, 2005, pp. 28, 41–42) (Figure 7). Although this nuraghe has not been excavated, comparing this superficial evidence with the detailed excavations at S’Urachi as well as our small-scale interventions described above (Section 3.1) suggests that similar developments were at play: the local community maintained a pre-existing site but came into more and more frequent interaction with the outside world.

Figure 7

Nuraghe Zoddias.

Many of the recorded points from the Punic-Early Roman phase have functions related to new transport infrastructure that facilitated interaction across west-central Sardinia, including roads and bridges (Dyson & Rowland, 2007, p. 149; Tronchetti, 1986, p. 20). This monumentalization of routes across the landscape in the later Iron Age illustrates how interactions intensified by this period. Some roadways were paved with basalt around this time and stone bridges were built to facilitate access across waterways. One example of this is the roadway and bridge near Tramatza that traverses the Riu di Mare Foghe, allowing penetration deeper into the interior of the island towards Forum Traiani (Fordongianus). The standing remains are thought to date between the first century BCE and the first century CE with repairs in the Medieval period and in 1826 (Unione dei Comuni del Montiferru e Alto Campidano) (Figure 8). This new infrastructure meant that routes that brought people and goods from the coast into inland territories were easier and faster to traverse, thereby facilitating interaction between the coastal and inland territories. This corresponds well with the evidence for agriculture from the pedestrian survey: while outside goods, people, and ideas were penetrating these rural landscapes, these same routes that brought interaction from the outside also facilitated the export of agricultural produce towards the coast and eventually to other Mediterranean contexts. Connectivity, therefore, grew increasingly multi-directional, multi-scalar, and more frequent and intense over the course of the Iron Age.

Figure 8

Roman bridge and road at Tramatza.

3.4 Future Work of the SAP: Collaboration and Multi-Scalar Contextualization

While this analysis has focused on the inland site of S’Urachi and Zone A, our future work will facilitate a comparison between this zone and the wider territory of west-central Sardinia of interest to SAP. Work in 2022 and 2023 is focusing on our Zone B, the coastal territory to the west of Zone A, in which there is ample evidence for limestone quarrying as well as the exploitation of maritime resources and those materials associated with the seasonal salt flats and lagoons, such as the Stagno di Sal’e Porcus, with its salt, fish, birds, and wild plants used for basketry. This zone is also interesting for questions of regional connectivity because it lies between two major Phoenician coastal sites, Cornus and Tharros, but it does not contain a large colonial settlement of its own.

We are particularly interested in exploring how the uneven distribution of resources across this rural but ecologically diverse landscape shaped Phoenician, Punic, and Roman interventions in the region, and by extension how this impacted the scale and intensity of interaction in different locales. For instance, the exploitation of iron in Monte Ferru to the north might have brought more intense interaction with the outside world centered around metal export from the Phoenician site of Cornus (Blasetti Fantauzzi & de Vincenzo, 2013, 2016; Sanna, 2006a,b; Spano, 1864; Taramelli, 1918; Zucca, 1988). A need for the exchange of coastal resources with those of the inland agricultural zones (and vice versa) might have brought increasingly frequent interregional interaction among local communities in each territory. This survey will thus be able to move beyond simply showing that interaction between local communities and foreign colonizers and traders happened to also address the “nature and logic of interaction” (Dietler, 2022, p. 239), or how resources and local communities shaped connections with the outside world and how this connectivity varied in type and intensity across time and space.

In keeping with our methodological framework outlined above, our multi-scalar survey will take advantage of collaborative strategies and publications, re-integrating our own findings with those produced from other active projects and with other types of data. To complement our survey-based approach, we are collaborating with excavations within our survey region. The case study already highlighted some of the ways that we integrated our site-based survey work with the excavation team working at the inland, indigenous site of S’Urachi. Additionally, we have plans to collaborate with the University of Cincinnati team currently working at the coastal, colonial site of Tharros – a Phoenician settlement with occupation through the Late Roman period (Ellis et al., 2021). Our future work will involve comparing the ceramic assemblages from the survey with those of S’Urachi and Tharros, with shared team members carrying out fabric analyses on both sets of materials. This will facilitate the chronological alignment of assemblages while also enabling comparisons of material between an inland settlement, a coastal settlement, and the surrounding rural areas. Such collaboration will aid further exploration of whether rural communities were differently impacted by Iron Age interaction than their urban counterparts in both coastal and inland locales.

Finally, SAP is also a collaborating field project with Migration and the Making of the Ancient Greek World Project (MIGMAG), a European Research Council-funded initiative directed by Naoíse MacSweeney (University of Vienna). This project “combines archaeological and historical evidence to explore the human mobilities c. 1200–550 BCE that led to the creation of the ancient Greek world as we know it” (MIGMAG, 2023). SAP offers survey evidence from west-central Sardinia, a landscape much more impacted by Phoenician migration than Greek, to complement the work of surveys in other parts of the Mediterranean where Greek migration was more prominent. Other collaborating partners include the Piano di Gioia Taura Project in the inland plains of Reggio Calabria, the Klazomenai Survey Project in the Urla-Çeşme peninsula in western Turkey, and the Taşeli-Karaman Archaeological Project in southern Turkey. Ultimately, cooperation with these projects that are exploring similar questions with similar data from the same chronological period will help us contextualize our Sardinian case and embed it into the largest Mediterranean scale.