Journal of the North Altantic
    JONA Home
    Aim and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Other Eagle Hill Science Journals
    Northeastern Naturalist
    Southeastern Naturalist
    Caribbean Naturalist
    Neotropical Naturalist
    Urban Naturalist
    Prairie Naturalist
    Eastern Paleontologist

Eagle Hill Institute Home

Central Places in a Rural Archaeological Landscape

Michelle Comber*

*Archaeology, School of Geography and Archaeology, National University of Ireland Galway.

Journal of the North Atlantic, No. 36 (2018)

Archaeological survey in western Ireland has identified the existence of clusters of activity within the mapped landscapes of the 5th to 12th centuries A.D. Exploring this further, it is possible to identify elements characteristic of such clusters, and discuss the possible significance of such places. The basics of German geographer Walter Christaller’s Central Place Theory provide an interesting analytical tool in this regard. Although a spatial theory developed in the study of urban geography, some elements of Christaller’s work have been applied to urban archaeological landscapes in recent times. Their application in the rural ringfort landscapes of western Ireland proves an interesting exercise, one that suggests that Central Places also existed in more dispersed, rural communities in Early Medieval Ireland.

pdf iconDownload Full-text pdf (Accessible only to subscribers. To subscribe click here.)



Site by Bennett Web & Design Co.
Central Places in a Rural Archaeological Landscape Michelle Comber* Abstract – Archaeological survey in western Ireland has identified the existence of clusters of activity within the mapped landscapes of the 5th to 12th centuries A.D. Exploring this further, it is possible to identify elements characteristic of such clusters, and discuss the possible significance of such places. The basics of German geographer Walter Christaller’s Central Place Theory provide an interesting analytical tool in this regard. Although a spatial theory developed in the study of urban geography, some elements of Christaller’s work have been applied to urban archaeological landscapes in recent times. Their application in the rural ringfort landscapes of western Ireland proves an interesting exercise, one that suggests that Central Places also existed in more dispersed, rural communities in Early Medieval Ireland. Introduction In the early 1930s, German geographer Walter Christaller (1933) developed Central Place Theory, a spatial theory in urban geography designed to explain the distribution, size and number of cities and towns in a region. Since then, his ideas have occasionally been employed in an archaeological context, exploring the remains of nucleated urban settlements, their distribution patterns and territorial reach. This study explores whether or not it is possible to apply some of the ideas of Central Place Theory to dispersed rural settlements. Do rural Central Places exist in the archaeological record and, if so, how can they be recognised? For Christaller, a Central Place was one that provided goods and/or services to the surrounding population (1933:27). He further divided such places into low- and high-order Central Places, with the former offering low-order (frequently required, less expensive) items such as foodstuffs, and the latter providing more specialized goods and services, those required less often. Low-order Central places were plentiful, being required by the entire population on a regular basis, while high-order Central Places were fewer in number and more widely distributed—in order to serve a larger population less frequently. More recently, five different levels of Central Place were identified by Nakoinz (2012); hamlet, village, town, city and regional capital—all nucleated settlements. Christaller came to the conclusion that, in a uniform landscape, the spatial distribution of settlements would be perfectly regular (Smith 1974:168). He assumed a flat topography with no barriers to movement, evenly distributed natural resources and population, and that the “consumers” would always patronize the nearest Central Place. For Christaller, all Central Places developed for purely economic reasons. What, then, might be the expected characteristics of a traditional Central Place? At its core, it is a settlement providing goods and/or services to a surrounding population. To do this, it must be accessible by that population—physically, financially, and politically. It will be equidistant from other similar-scale Central Places, with a hierarchy of lower-scale Central Places in its hinterland or territory. These territories will form regular patterns on the landscape, influenced by the economics of supply and demand. Although primarily a tool for model building in economic geography, Central Place Theory has become increasingly popular in anthropological and archaeological studies, facilitating exploration of “underlying tendencies towards spatial order in pre-industrial societies” (Bintliff 2002:212). Despite the inherent difficulties in applying such an ideal-type model to the more complex reality of varied past landscapes and societies, scholars such as Nakoinz (2012) and Mulligan et al. (2012) have made clear the validity of the principles that underlie Central Place Theory. The application, and influence, of Central Place Theory can be seen in several German, Scandinavian, Roman, and English studies (e.g., Bintliff 2002, Hirschel and Ludowici 2010, Hodder and Orton 1976, Nakoinz 2010). Central Place Theory assumes that market exchange is key to the economy, and that centres exist to facilitate that exchange (Smith 1974:168–169). These centres are located so as to minimise distance/travel and transport costs, in a landscape that can be easily traversed from all directions by an evenly distributed population. Central Place Theory also assumes that suppliers of goods and services desire maximum profit, and that competition exists amongst suppliers (ibid.). Of course, such ideal conditions are not often found in the real world, especially in the varied landscapes of the past. This paper uses a specific archaeological landscape, the Early Medieval (5th–12th century A.D.) Burren region of western Ireland, to explore some of the underlying principles of Central Place Theory in a setting very different to that traditionally used. Key to the economy in Early Medieval Ireland was social exchange, not market exchange. A system of clientship saw varying grades of nobility, free farmers, and unfree classes tied together in a social framework that involved the regular movement of goods between settlements, and the relatively widespread provision of both specialist and more common services. In addition to this different economic base, the Early Medieval Irish landscape was characterised by dispersed rather than nucleated settlement. A final major difference between this study and the more traditional applications of Central Place Theory is its scale. The Burren study area is tiny compared to some Central-Place-Theory targets, with scholars such as de Haas (2017) assessing evidence from all of Roman Italy. For these reasons, this examination does not involve a standard application of Central Place Theory, rather it takes some of the underlying basics and tests whether or not such criteria can be usefully applied in a very different context—in this case, a small-scale, non-urban, rural context. What criteria, then, might realistically be expected of a Central Place in such a landscape? Basics such as a local settlement hierarchy, provision of goods and services, accessibility, local natural resources, and access to exotic or less-common goods all seem plausible. In the specific context of Early Medieval Irish society, physical links with the ancestors and with Christianity might also be expected at a place providing ceremonial as well as economic and social services. While issues such as the contemporaneity of settlements and biased preservation of evidence are concerns in every study of this type, the nature of the evidence in the current study minimizes these issues. Whilst all cashels (the primary settlement form in question here) were not necessarily built and used at exactly the same time, they can be broadly attributed to the Early Medieval period. As such, they reflect settlement activity in a given place, often over many centuries, and may in fact support the idea explored below that continuity of use/a link to the “ancestors” (Early Medieval or older) constitutes an additional Central-Place attribute in an Early Medieval Irish context. Another problem often encountered in the application of Central Place Theory is the uneven preservation of remains in an archaeological landscape. While no landscape boasts perfect preservation of the past, the Burren region of western Ireland ranks highly in this regard. The vast tracts of exposed limestone bedrock in the area provided the main source of building material for prehistoric ritual monuments and the later Early Medieval cashels. Due to its largely upland nature and difficult topography, this region saw little modern mechanised farming. This, in conjunction with the sturdy building material, has led to the widespread preservation of standing monuments, particularly those of relevance to this study. Acknowledging the many differences with more traditional Central Place studies in terms of scale, economy, landscape, and society, the current study explores whether the criteria presented above can be used to identify Central Places in Early Medieval western Ireland and, if they can be identified, whether or not the term “Central Place” is truly applicable. As such, this paper deviates from traditional studies/applications of Central Place Theory, yet is inspired by them. Research in the Burren study area has been underway since 2005. An overview of the relevant archaeology is presented below, moving from large-scale to small-scale study, from landscape analysis to site excavation, searching for the criteria mentioned above, and building a model for rural Central-Place activity in the process. The Burren Archaeological Project The distinctive karst landscape of the Burren, County Clare in western Ireland provides a valuable opportunity to examine the well-preserved remains of past settlement, in turn facilitating an exploration of the factors influencing its distribution. Occupation sites and associated field systems can be found throughout this area, reflecting human activity from the Mesolithic to modern periods. The most visible and plentiful settlements from the past date to the second half of the first millennium A.D., the Early Medieval period. These settlement enclosures mostly comprise drystone ringforts (called cashels), with a much smaller number of earthen ringforts (raths). Their numbers reflect relatively dense occupation of the Burren in the Early Medieval period, and many are surrounded by preserved field systems, some quite extensive. A research project, the Burren Archaeological Project, has been established to investigate these landscapes. Initial work involved three stages of survey followed by an ongoing excavation stage. Much of the Burren lacks continuous soil cover, indeed some areas are completely devoid of soils. The terraced uplands of crumbled limestone do host a thin soil cover, perfect for upland pasturage, with drinking water provided by springs. These upland soils (rendzinas) are well-drained all year round, and are warmed by the heat-retentive limestones beneath, providing excellent winter grazing (Finch 1971:34). The deeper soils of the valleys often comprise brown earths and grey-brown podzols, more suited to crop cultivation (ibid.). Survey work commenced with an examination of the cashels and associated remains in a study area extending south-west from Caherconnell to Kilfenora, east to Carran and Cahercommaun, and south to Leamaneh (Comber 2005) (Fig. 1). The first stage saw the analysis of approximately 300 relevant monuments within this area (mostly ringforts, enclosures and ecclesiastical remains). This revealed that many of these settlements were deliberately sited to best exploit the most fertile farmland in the area. It also suggested that some may have been strategically positioned with regard to communication strategies and territorial politics. Stage 2 saw the detailed digital survey and mapping of a preserved archaeological landscape within the larger Stage 1 study area (Comber 2006). Extensive field systems and enclosures were recorded in this area (Fig. 2). Elements from various periods of the past were identified, reflecting the continued use of this zone throughout prehistory, the Early Medieval period, and the Medieval period. Work in Stage 2 identified several small settlement clusters, each containing at least two settlement enclosures/cashels. Questions arose regarding the chronological and functional relationship of various recorded features in these clusters, questions that could really only be addressed by excavation. With this in mind, a cluster was selected for excavation in Caherconnell townland on the northern edge of the original study area. Stage 3 of the survey, then, focused on the townland of Caherconnell (Comber 2008). A wealth of archaeological remains were visible in this area (including a tight cluster of 3 cashels, resembling the settlement clusters identified in the previous stage of the project), an area that borders both the fertile green valley of Kilcorney to the south and the exposed karst uplands of the west and north. Another important factor in selecting this area, at this stage of the project, was the opportunity to undertake excavations across this landscape, with the strong support of the landowner. This allows the incorporation of excavated functional and chronological evidence into the analysis of the digitally mapped remains, and facilitates extrapolation of data elsewhere. The Stage 4 excavation in Caherconnell townland commenced with small-scale volunteer excavations, the results of which highlighted the archaeological potential of this area and led to the establishment of larger-scale research excavations in 2010. This work is ongoing and is investigating various elements of the mapped archaeological landscape, in an attempt to explore relationships between different elements and with the landscape itself. It is also facilitating this first step in the development of a spatial model for rural landscapes. The Caherconnell cluster Caherconnell townland shares its name with the largest monument within its boundaries. Caherconnell cashel is a well-preserved example of a large cashel located on the southern border of the Burren uplands. Its 3 grouped cashels are strategically positioned near the intersection of two natural routeways through the uplands, both of which are marked by modern roads (Fig. 3). The use of these natural routeways stretches back to at least Early Medieval times, and probably earlier. The remains of an old “green road” (a grassed-over road or track no longer in use) skirt Caherconnell cashel to the east and south, with the road running off to the south-west for a short distance. Local tradition records this road leading all the way to Kilcorney parish church just over 1 km to the west. Unfortunately, however, this length of road cannot be traced on the ground today, though the tradition does provide a link between the cashel and ecclesiastical site. Although Caherconnell cashel does not occupy the highest land in the area, it does have a commanding view of its immediate landscape. The site occupies the edge of a kilometre-wide terrace overlooking a shallow valley to the south and southeast. This basin contains some of the most fertile soils in the Burren, suitable for both grazing and crop cultivation. Prehistoric remains are scattered across Caherconnell townland. They include an excavated Bronze Age house (Hull and Comber 2011), a couple of Bronze Age boulder burials and a Late Bronze Age Iron Age barrow. The construction of the later cashels amongst these features may reflect a deliberate desire on behalf of the cashel-builders to be associated with the important places of the ancestors. In the surveyed area, the four larger enclosures in the townland appear to be drystone cashels of the Early Medieval period (Fig. 4). Three of the cashels are roughly circular in plan, with Caherconnell cashel clearly the most imposing of these. It is larger than the others and far better preserved. The circular cashel to the south presents as a typical stone ringfort of the Early Medieval period. The sub-square cashel to the south of this is an example of a cashel sub-type found in relatively small quantities in the Burren. Its shape sets it apart from the standard circular cashel, yet its walls are constructed in much the same manner (see below). The only other cashel in the surveyed area is another typical example, located to the west on the eastern edge of an inland cliff. The sub-square enclosure (Fig. 5) The sub-square drystone enclosure in Caherconnell townland lies approximately 100 m south of the main cashel of Caherconnell. Features uncovered in the interior included 3 sub-circular structures/houses and a number of walls sub-dividing the space between the structures (Comber 2010–2012). The enclosure entrance, in the south wall, comprised an entrance passage with a roughly paved surface, one stone of which concealed a deliberately placed prehistoric stone axe. Associated occupation deposits were rich in faunal remains, and a quantity of artefacts was also recovered. The artefacts included everyday items, and a small collection of slightly higher-status personal ornaments. Radiocarbon dates and recovered artefacts indicate that the enclosure was used during the Early Medieval period, from the 7th to the 9th centuries A.D. The material culture and relatively large size of the enclosure (38 m across, compared to the typical Burren cashel diameter of 20–30 m) reflect an above-average status for the occupants of the site. It is distinctly different to the vast majority of contemporary cashels in the Burren, being sub-square in shape, larger in size, and orientated to the south (rather than the east). The reason behind these differences, the reason why this enclosure was set apart, might be explained by a closer examination of the excavated evidence and the landscape setting. The enclosure is located just below the highest point in the immediate locality (now occupied by Caherconnell cashel), a point that saw the burial of a small group of people in a low mound in the 6th/7th century A.D. (the same date as teeth and skull fragments recovered from the sub-square enclosure)—not far from what appears to be a Late Bronze Age or Iron Age barrow. An ancestral link could, therefore, be claimed by the builders of the sub-square enclosure. The downslope siting also afforded the settlement some shelter, whilst its southern aspect and orientation were conducive to comfortable living, easy access to the good farmland of the Kilcorney valley, and control of a crossing-point of two adjacent routeways. The distinctive enclosure could be easily identified, and accessed, by anyone travelling along either routeway. The material culture from the site reflects the use of these routeways in the acquisition of items from outside the immediate area. Objects such as two bronze dress-pins and a small number of glass beads were imported to the site, reflecting contact with the outside world and the slightly raised status of the occupants. The inhabitants of the site were self-sufficient, producing their own foodstuffs and everyday implements. Their settlement was larger than most contemporary enclosures, and had plenty of usable space within its walls (the structures occupying a relatively small percentage of the internal area). It was physically distinctive and in a strategic location, suggesting the possibility of its marking a Central Place, a place people came to for a specific reason. The slightly raised status, plentiful storage space, easily identifiable physical presence, link to the ancestors, and location along an important routeway are not incompatible with functions such as tribute collection, something Early Medieval documents reveal were often organised by a rechtaire or steward, on behalf of an overlord or local king (Kelly 1988:65). Caherconnell cashel (Fig. 6) Caherconnell cashel has a diameter of 42 m, and is defined by drystone walls 3 m wide and high (though higher originally). Its entrance, like that of most ringforts, faces east. Its location on the northern slopes of the Kilcorney valley provided the settlement with a commanding view of the surrounding landscape, and easy control of the adjacent routeways. Its agricultural needs were well met by the surrounding pasturelands and fertile valleys, while its imposing morphology and connection with the past (ancestral burials and activity at the sub-square enclosure) contributed to the statement it made on the landscape. Excavation (Comber 2010–2017; Comber and Hull 2010) has uncovered several phases of activity within Caherconnell cashel, the earliest of which (late 6th/early 7th century A.D. burial mound containing the remains of two infants and an elderly woman, and a fire-pit from the second half of the 7th century A.D.) pre-date the construction of the enclosure (possibly linked to the early use of the sub-square site). Cashel construction followed in the late 10th century A.D. The cashel wall was built directly on the limestone bedrock, except where the builders chose to deliberately incorporate the burial mound into their new settlement by constructing the cashel wall over the top of it. The adoption of ancestors as part of a territorial claim is something seen at some other high-status Early Medieval settlements, perhaps most dramatically at Knowth in Co. Meath where the “royal” Uí Neill settlement was constructed on top of the Neolithic passage-tomb mound. At Caherconnell, the date of cashel construction coincides with the date of two written references to a high-status figure/figures called “Conghal” in the area (Caherconnell: the cashel or caher of Conall or Conghal). One records Conghal as the son of a local king, Anruadan (d. 936) (Annals of the Four Masters M987.7, O’Donovan 1848–1851), the other as his cousin (genealogy in Gibson 2012:289). Whether or not the entries refer to two different individuals is uncertain but, either way, a member of the ruling family is reflected in the placename. Such an individual might well be expected to claim an association—and display a physical connection—with the local territorial ancestors. Four phases of high-status occupation followed, each with their own domestic and ancillary structures. Artefacts were plentiful from most phases, including regular domestic, agricultural and craft-working implements, higher-status personal ornaments, weapons, and imported items. The local environment provided occasional fish and shellfish, hazelnuts, possibly iron and lead ores, hazel, ash, birch and yew wood, and supported the growing of free-threshing wheat, barley and oats, and the grazing of cattle, sheep, pigs and red deer. The artefacts reflect something of the activities that took place within the cashel, and the status of its occupants. Both ironworking and non-ferrous metalworking occurred within the cashel, as did woodworking, stone-working, bone- and antler-working, textile production/clothes manufacture, and the processing of grain. Several, if not all, of these activities may have produced a surplus. Less “domestic” high-status activities are represented by armour-piercing arrowheads, harp-pegs, and gaming pieces. Trade/the use of the adjacent routeways is evident in the presence of English coins, a German jetton, bronze, silver, gold, glass (including a Venetian bead) and amber at the site. Caherconnell discussion Both survey and excavation suggest that agriculture formed the mainstay of the economy, with the karst uplands of the Burren providing excellent all-year grazing for animals, and the deeper pockets of soil in its sheltered valleys ideal for cereal cultivation. That the early farmers of the Burren exploited these landscapes in an organised and systematic manner is evident in the extensive field systems still surviving across the region. Two local pollen diagrams, at Cappanawalla to the north and Gortaclare a short distance to the northeast (Feeser and O’Connell 2009), show significant clearance of woodland in the region ca. 500 A.D. This coincides with Early Medieval agricultural expansion and population growth, seen archaeologically in the appearance of several hundred ringfort settlements and associated field systems in the region. The typical Burren ringfort comprised a circular drystone cashel, averaging 20 m to 30 m in diameter, with walls up to 2 m thick and 2 m wide, and with an eastern entrance. The sub-square cashel at Caherconnell stands apart from these. It is morphologically distinct, of slightly above-average status, and in a strategic location. It may mark a Central Place in the Early Medieval landscape, a place people came to for a specific purpose, possibilities including the paying of tribute or ”rent” to a superior, seeking medical or health assistance, education or fosterage services etc. Access to good farmland, control of natural routeways, links to the ancestors, and the established “Central Place” idea all contributed to the construction of a high-status settlement at Caherconnell in the late 10th century. Two written references to a high-status figure (or figures) named “Conghal” occur at this time, supporting the archaeological evidence of imposing morphology and excavated material culture. At least some of this status and wealth derived from the surrounding landscape—control of communications, agriculture and other natural resources. The inhabitants of Caherconnell cashel were inextricably tied to the resources and advantages provided by the local landscape, both natural and cultural. Discussion It is clear from the outline above that settlement in this area was unevenly distributed across the landscape and had a primarily agricultural focus (Stage 1 survey, Fig. 1). Both Stage 1 and Stage 2 surveys highlight the clustering of settlements, though a clustering that never reached fully nucleated/urban status. Nonetheless, clusters emerge as places where people gathered, travelled to/from for a number of reasons (Fig. 2). It is clear from the Stage 3 survey and the Stage 4 excavation that Caherconnell functioned as one such cluster or Central Place during the Early Medieval period, and the detailed characteristics of this Early Medieval western Irish Central Place can now be identified. Both of the excavated Early Medieval settlements were of elevated status, evident in their morphology and excavated material culture, and supported by placename and documentary references. Other Early Medieval settlements in the immediate vicinity are physically smaller and, although unexcavated, probably of lower status, placing the Caherconnell settlements at the top of a local settlement hierarchy. The hierarchical nature of society in Early Medieval Ireland is well documented (e.g., Kelly 1988:29–33), comprising a social system—clientship—that saw the movement of people and goods between related social groups and their settlements. Kings and lords provided seed stock, specialist goods, and legal and military protection to their socially inferior clients. In return, the latter paid an annual tribute in foodstuffs (mostly), and provided labour, hospitality, and military service. All of this required that the high-status settlements be accessible places where people could gather—and this ease of access is clearly visible at Caherconnell. It has already been suggested that the occupants of the sub-square cashel at Caherconnell were of above-average status, and that they provided a specific service to the surrounding population, something not available elsewhere. As the dwelling of a lord or king, the slightly later Caherconnell cashel would have provided some or all of the clientship services mentioned above. The excavated evidence also suggests that the settlement may have produced surplus items and foodstuffs, in addition to providing access to foreign/imported goods. All of these elements—provision of goods and services, a local settlement hierarchy, accessibility, and access to “exotic” or less-common items—identify Caherconnell as a high-order Central Place. Three other features are notable at Caherconnell—a respect for/concern with the ancestors in the presence of, and interaction with, pre- and early-Christian monuments, proximity to good farmland (a valuable resource), and an association with Christian religious activity. Therefore, in this part of Early Medieval Ireland, a rural high-order Central Place has a number of distinct attributes: high-status settlement, a concentration of lower-status settlements, a distinctive sub-square or rectangular settlement enclosure (possibly reflecting the provision of a specific service), good communications/routeways, prehistoric ritual monuments, an ecclesiastical link, and proximity to good farmland and/or other natural resources. Using these criteria, identified at Caherconnell, it is possible to locate other high-order Central Places in the wider, Stage 1, study area (approximately 75 km2 in extent). Five are proposed (Fig. 7); CP1 (Central Place 1) Ballykinvarga/Kilfenora, CP2 Noughaval/Ballyganner North, CP3 Caherconnell/Kilcorney, CP4 the Poulacarran Valley, and CP5 Cahercommaun/Tullycommon. A much smaller, less convincing example occurs in Iskancullin townland. This comprises a high-status settlement (morphologically identified as a settlement enclosure of 40 m+ diameter), a rectangular enclosure, one low-status settlement, and a megalithic tomb. It does not, however, have a concentration of settlements, the best farmland, or ease of access. As such, it might be better described as a low-order Central Place, i.e., a place lacking some of the high-order criteria identified above. In the southwest of the study area, CP1 comprises a settlement hierarchy, most likely headed by the imposing cashel at Ballykinvarga. It also contains a rectangular enclosure, the monastery of Kilfenora, routeways, a number of prehistoric sites, and is located in the fertile lowlands along the southern limit of the Burren plateau. To its northeast lies CP2 and its hierarchy of settlement enclosures. This includes at least one obviously high-status site in Noughaval townland, and up to 5 rectangular cashels. The ecclesiastical remains at Noughaval have Early Medieval origins, and appear to lie on an ecclesiastical routeway between Kilfenora to its south-west and Kilcorney to its northeast. Prehistoric sites include cairns and megalithic tombs and, although somewhat higher than CP1, this Central Place also enjoys good farmland. Caherconnell, or CP3, lies to the north-east of this. It, too, has a hierarchy of settlement forms, headed by Caherconnell cashel, and a rectangular enclosure. Its ecclesiastical connection occurs at Kilcorney, a site with Early Medieval origins. The routeways, prehistoric activity, and good farmland have already been discussed. South-east of Caherconnell, and quite close to it, lies CP4, centring on the Poulacarran Valley. Numerous settlement sites are located in this area, including up to 5 high-status cashels and 5 rectangular enclosures. Carran church is situated on the south-western edge of this Central Place, a place that takes advantage of the same “crossroads” as Caherconnell, albeit on the opposite side to Caherconnell. Cairns, fulachtai fia (Bronze Age cooking places), megalithic tombs and a stone row represent ancestral activity, and the soils of this deep valley are particularly fertile. The final Central Place identified in the study area, CP5, lies to the east. Its settlement hierarchy is clearly headed by the impressive trivallate cashel of Cahercommaun (Hencken 1936), by far the largest settlement in the area. Up to 4 rectangular sites occur, as do prehistoric cairns, a cist, and megalithic tombs. A modern road runs through this area, and may mark an earlier routeway. Of all the high-order Central Places identified, this plateau is least suited to arable farming, though does provide very rich grazing lands. Elsewhere in the study area there are a few scattered, more isolated, high-status sites and rectangular enclosures. Like those identified in Stage 2 of the Burren Project, small clusters of low- to medium-status settlements also occur, often located near lesser routeways and decent farmland, and occasionally incorporating prehistoric remains. All of these do, however, lack some of the criteria for high-order Central Places, but may have functioned as low-order Central Places, places where common goods and services could be obtained on a regular basis. The remaining widely dispersed, most isolated, settlements represent non-central places, places normally only frequented by their immediate inhabitants. Conclusion It is evident, then, that some basic elements of traditional Central Place Theory are applicable to some rural landscapes lacking nucleated or urban settlements. In both traditional Central Place Theory and in the current study, Central Places can be identified as places providing goods and services to a surrounding community, fulfilling an economic role. In both cases, these are also places that incorporate a hierarchy of settlement, and are relatively easy to access from outside. However, in rural Early Medieval Ireland deviations from Central Place Theory also occur, primarily the identification of a collection of individual settlement sites as a Central Place, rather than the traditional single urban settlement. Aside from this, perhaps the most obvious difference from the original Central Place Theory evidenced in the current study is in the distribution of identified Central Places. Christaller’s work suggested that Central Places of equal standing (i.e., high-order or low-order) would be equidistant from one another, and surrounded by regularly shaped and sized territories. This is clearly not the case in Early Medieval western Ireland, where several factors influence distribution and size. Here, access to natural resources (good farmland being perhaps the most valued of these in agricultural Early Medieval Ireland), and a connection with the past and the ancestors (mostly represented by standing ritual monuments) appear to have played a significant role in the siting of high-order Central Places. Perhaps even more influential, however, were the local political systems and associated territorial organisation. Members of the nobility ruled over individual territories in Early Medieval Ireland. These territories varied in size, but each probably contained at least one high-order Central Place. The exact boundaries of these early territories are unknown, however it is suggested that later medieval parishes were based on existing secular territories (e.g., MacCotter 2008, Ní Ghabhláin 1995, Nugent 2007). When a map of parish boundaries is added to the study area with its archaeologically identified Central Places, an intriguing correlation emerges (Fig. 8). Parts of five different parishes comprise the bulk of the study area, and each one contains one of the five identified high-order Central Places. Kilfenora Parish contains CP1, Noughaval Parish CP2, Kilcorney Parish CP3, Carran Parish CP4, and Killinaboy Parish CP5. With the possible exception of CP 2 in Noughaval, none of these are in central locations within their parishes, instead they are located closer to the parish/territorial boundary, perhaps to facilitate interaction (of varied form) between territories and groups, not just within them. The current evidence suggests that high-order Central Places were sited with reference to political territory and, within that, to accessibility, good farmland, and ancestral presence. Focal Place might, therefore, be a more accurate term than Central Place, to allow for non-central locations within territories. In Early Medieval Ireland, and perhaps elsewhere, the location of high-order Central or Focal Places was dictated less by economics, and more by topography, natural resources, and politics. Clearly, it is possible, using some of the traditional elements of Central Place Theory and additional place- and time-specific factors, to identify Central or Focal Places in well-preserved rural archaeological landscapes. The model developed here, using detailed landscape survey and excavation, can facilitate the identification of rural Focal or Central Places, inspired by Christaller’s work of the 1930s.